Compare commits

...

169 Commits

Author SHA1 Message Date
Kyle McMartin 825b8ce73a changelog 2010-12-20 10:15:05 -05:00
Kyle McMartin f8d75b80e6 backport cs checker fixes for r600 2010-12-20 10:13:41 -05:00
Kyle McMartin 3dbb45b665 temporarily revert v4l rebase due to issues reported in previous update 2010-12-19 15:27:51 -05:00
Kyle McMartin 68b623121d copy nhorman's AF_PACKET vmalloc patch from f13 (#637619) 2010-12-18 11:05:38 -05:00
Kyle McMartin 558463d9fb fs-call-security_d_instantiate-in-d_obtain_alias (#662344) 2010-12-18 10:54:08 -05:00
Jarod Wilson 703515438b A few additional mceusb driver fixups
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-12-16 16:24:48 -05:00
Jarod Wilson fbfed34fec damn you, CONFIG_VIDEO_SH_MOBILE_CSI2...
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-12-15 17:23:49 -05:00
Jarod Wilson 92d37ba1d3 v4l/dvb/rc: rebase to latest upstream
- Rebase v4l/dvb/rc code to latest upstream, should fix a fair
  number of ir/rc-related issues, including bugzilla #662071
- Drop the spaghetti mess of patches we had for v4l/dvb/rc

Note: new patches can be trivially created using this:

http://git.linuxtv.org/media_build.git
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-12-15 17:14:26 -05:00
Chuck Ebbert 8f007394cd Linux 2.6.35.10
Remove merged patches and fix up conflicts:
   drm-polling-fixes.patch
   linux-2.6-v4l-dvb-hdpvr-updates.patch
   kvm-fix-fs-gs-reload-oops-with-invalid-ldt.patch
Drop merged patches:
   linux-2.6-rcu-sched-warning.patch
   pnpacpi-cope-with-invalid-device-ids.patch
   ipc-zero-struct-memory-for-compat-fns.patch
   ipc-shm-fix-information-leak-to-user.patch
   r8169-01-fix-rx-checksum-offload.patch
   r8169-02-_re_init-phy-on-resume.patch
   r8169-03-fix-broken-checksum-for-invalid-sctp_igmp-packets.patch
   hda_realtek-handle-unset-external-amp-bits.patch
2010-12-15 15:12:09 -05:00
Kyle McMartin f045c5648e fix typo pci_clear_aspm -> pcie_clear_aspm 2010-12-10 14:55:52 -05:00
Kyle McMartin f2ed8cf30b PCI: Disable ASPM if BIOS asks us to 2010-12-10 11:13:54 -05:00
Kyle McMartin 5a95077aa5 fix patches
don't do a make prep /after/ changing branches... cough.
2010-12-10 10:32:57 -05:00
Kyle McMartin 54f1e03004 fix some issues mounting btrfs devices with subvolumes (#656465) 2010-12-10 10:17:40 -05:00
Kyle McMartin 9d6e5dc425 fix jbd2 warnings when using quotas (#578674) 2010-12-10 09:47:05 -05:00
Kyle McMartin 4bd3ae5daa orinoco: initialise priv->hw before assigning the interrupt 2010-12-09 18:29:25 -05:00
Kyle McMartin bd61adb09d Copy tpm-fix-stall-on-boot.patch from rawhide tree. (#530393) 2010-12-09 17:06:27 -05:00
Chuck Ebbert 983a2bb5c5 Require newt-devel for building perf, to enable the perf TUI (#661180) 2010-12-09 11:20:43 -05:00
Kyle McMartin bed92c4e50 ameliorate the load average account issues with dynticks (rhbz#650934) 2010-12-08 17:20:15 -05:00
Kyle McMartin 20b0493429 enable cplus_demangle in perf 2010-12-04 12:18:59 -05:00
Kyle McMartin 65c06ec080 enable hpilo on x86_64 2010-12-03 07:03:54 -05:00
Kyle McMartin 76a7ebfb42 kswapd fixes from mmotm 2010-12-02 11:21:28 -05:00
Kyle McMartin d40233bb75 forgot to uncomment... 2010-11-30 15:31:20 -05:00
Kyle McMartin 592f4fb443 cherry picked and updated 2010-11-30 15:20:21 -05:00
Kyle McMartin 48eb5c319b add a patch to log pnp resources (from rawhide) 2010-11-29 20:17:45 -05:00
Kyle McMartin 26ea0f2631 drm/ttm: Fix two race conditions + fix busy codepaths [1df6a2eb] (#615505) 2010-11-29 19:10:59 -05:00
Kyle McMartin ca2a17f441 quiet a build warning the inet_diag fix caused 2010-11-26 15:20:25 -05:00
Kyle McMartin 6abfea1c89 plug various tty/serial stack leaks 2010-11-26 15:01:31 -05:00
kyle 7a80ec3e7d hda/realtek and r8169 fixes 2010-11-26 11:20:50 -05:00
kyle e1c14200b1 copy nodebug/nodebuginfo from master 2010-11-25 12:47:00 -05:00
John W. Linville dd99fe005c rtl8180: improve signal reporting for rtl8185 hardware
rtl8180: improve signal reporting for actual rtl8180 hardware
2010-11-24 10:52:09 -05:00
Kyle McMartin 003654d6ca posix-cpu-timers: workaround to suppress the problems with mt exec (rhbz#656264) 2010-11-23 11:40:33 -05:00
Kyle McMartin c86d4707a6 fix logic error in INET_DIAG bytecode auditing (CVE-2010-3880) 2010-11-23 11:12:31 -05:00
Kyle McMartin de964e3639 zero struct memory in ipc shm (CVE-2010-4072) 2010-11-23 10:58:12 -05:00
Kyle McMartin 55415ebd4a initialize struct memory to zero in ipc compat (CVE-2010-4073) 2010-11-23 10:08:40 -05:00
Kyle McMartin cc7fb594a9 fix i8k inline asm to workaround miscompilation with newer gcc 2010-11-23 09:47:06 -05:00
Jarod Wilson 51157289c8 Update to 2.6.35.9 final, tack on some IR fixes
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-11-22 16:13:38 -05:00
Chuck Ebbert 0081a7139c Linux 2.6.35.9-rc1
Comment out upstreamed patches:
kvm-fix-regression-with-cmpxchg8b-on-i386-hosts.patch
2010-11-20 07:09:04 -05:00
Ben Skeggs 466450dbeb nouveau: a couple of bug fixes 2010-11-19 19:04:49 +10:00
Kyle McMartin b2e700d12a bump baserelease to 59 2010-11-15 22:02:29 -05:00
Kyle McMartin 4c76cc7f94 pull in support for MBA3 2010-11-15 22:00:06 -05:00
Dave Airlie d5b83e3877 drm: fix stupid in warning 2010-11-12 14:33:52 +10:00
Dave Airlie 5f72d41789 drm: fix EDID issues 2010-11-11 20:49:55 -05:00
Dave Airlie 171d7cfb89 drm: fix invalid edid issues 2010-11-11 20:49:15 -05:00
Justin M. Forbes adfff05db3 fix regression with cmpxchg8b on i386 hosts (rhbz#650215) 2010-11-11 08:49:27 -06:00
Jarod Wilson 28683a07fb 2.6.35.8, ir-core, lirc and hdpvr updates
* Wed Nov 10 2010 Jarod Wilson <jarod@redhat.com> 2.6.35.8-53
- Linux 2.6.35.8
- Drop patches upstreamed in 2.6.35.8
- More ir-core and lirc updates
- HD-PVR driver updates

Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-11-10 12:42:45 -05:00
Dave Airlie 5ddd90a5d8 - add i915 polling s/r patch 2010-11-09 14:40:50 +10:00
Dave Airlie a474625d92 - Backport polling fixes + radeon hang fixes from upstream 2010-11-08 20:14:05 -05:00
Dave Airlie 05493c4e57 drm: apply polling + radeon fixes from upstream
remove the old polling stuff
2010-11-08 20:12:47 -05:00
Ben Skeggs 383a70142a bump baserelease 2010-11-02 13:37:05 +10:00
Ben Skeggs c7b0072805 nouveau: nv86 workaround + dual-head fix 2010-11-02 13:36:15 +10:00
Ben Skeggs df57a85a75 nouveau: add workaround for nv86 hw bug 2010-11-02 13:09:33 +10:00
Jarod Wilson 030d3b3b4b Fix brown paper bag hardlock bug in imon driver
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-10-23 16:03:57 -04:00
Chuck Ebbert 643d353feb More security fixes
drm-i915-sanity-check-pread-pwrite.patch;
 fix CVE-2010-2962, arbitrary kernel memory write via i915 GEM ioctl
kvm-fix-fs-gs-reload-oops-with-invalid-ldt.patch;
 fix CVE-2010-3698, kvm: invalid selector in fs/gs causes kernel panic
v4l1-fix-32-bit-compat-microcode-loading-translation.patch;
 fixes CVE-2010-2963, v4l: VIDIOCSMICROCODE arbitrary write
2010-10-22 11:02:18 -04:00
Kyle McMartin cbdb312a9e last minute f14 kernel fixes 2010-10-22 10:34:02 -04:00
Kyle McMartin 8311d5168d Revert "ir-core, lirc and hdpvr updates"
This reverts commit 59a83bd584
and 61d7a69fe6.
2010-10-22 10:26:15 -04:00
Jarod Wilson 61d7a69fe6 add pointers to the trees these patches come from
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-10-20 11:16:02 -04:00
Jarod Wilson 59a83bd584 ir-core, lirc and hdpvr updates
* Tue Oct 19 2010 Jarod Wilson <jarod@redhat.com> 2.6.35.6-47
- More ir-core and lirc updates
- HD-PVR driver updates

Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-10-19 17:24:50 -04:00
Kyle McMartin c4e830cb35 build 2.6.35.6-46 2010-10-18 22:44:57 -04:00
Kyle McMartin e38a760c9a dmar: add dev->device print like i had intended... 2010-10-18 22:38:42 -04:00
Kyle McMartin f1b58c7ca8 ricoh e822 sdhci support 2010-10-18 22:24:45 -04:00
Kyle McMartin a1b8c4ed84 bomb out of edp bringout that's going to be fail anyway 2010-10-18 19:28:12 -04:00
Kyle McMartin 6268215f01 fix ima patch 2010-10-18 13:35:50 -04:00
Kyle McMartin 6aca938822 ima: Default it to off
Pass ima=on to enable. Reduce impact of the option
when disabled.
2010-10-18 12:38:23 -04:00
kyle c76455ec66 copy the right patch this time 2010-10-18 11:52:57 -04:00
kyle 915df7f6de Quirk to disable DMAR with Ricoh card reader/firewire 2010-10-18 11:39:49 -04:00
Kyle McMartin 7d350a94ee Two networking fixes (skge, r8169) from sgruska 2010-10-18 10:59:42 -04:00
Kyle McMartin 9ee136bcf7 pnpacpi: cope with invalid device IDs. (rhbz#641468) 2010-10-15 15:04:48 -04:00
Kyle McMartin 97e19f7196 build for f14 2010-10-15 13:34:08 -04:00
Peter Jones 7810142e6c Fix missing dm_put().
There's an error path that fails to call dm_put(), which means bad
things happen.  This patch adds the dm_put() call.
2010-10-15 13:16:48 -04:00
Dave Jones c13cbf3b7d bump build 2010-10-13 17:08:13 -04:00
Kyle McMartin c19187e334 disable xhci by default
see spec %changelog for rationale.
2010-10-13 00:01:19 -04:00
Kyle McMartin a2d889c53a "i build tested it" are famous last words 2010-10-12 14:53:56 -04:00
Kyle McMartin 7140c9dd2d dm-allow-setting-of-uuid-via-rename-if-not-already-set.patch
Fix devicemapper UUID field cannot be assigned after map creation
  (rhbz#641476) thanks pjones@.
2010-10-12 13:40:04 -04:00
Jarod Wilson 167ee12e60 Grab a few more little IR fixes and document commits
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-10-11 16:49:13 -04:00
Jarod Wilson 870acad08e drop cruft from patch
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-10-11 14:07:36 -04:00
Jarod Wilson 866bf4f059 IR stack update (matches 2.6.37-rc1-to-be code)
- update imon driver to fix issues with key releases and properly
  auto-configure another 0xffdc device (VFD + MCE IR)
- add new nuvoton-cir driver (for integrated IR in ASRock ION 330HT)
- add lirc compat ioctl portability fixups

Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-10-11 11:37:53 -04:00
Ben Skeggs bd31d54eac fix ttm bug 2010-10-11 16:21:22 +10:00
Kyle McMartin f7f1f5ef4a bump to f14 2010-10-08 11:52:32 -04:00
Dave Jones a09b817bc1 Another day, another rcu_dereference warning. (#640673) 2010-10-06 14:09:52 -04:00
Kyle McMartin 2fe1e8e397 build intel_idle into the kernel 2010-10-04 16:13:30 -04:00
Kyle McMartin eaab8be2d4 enable printk.time by default 2010-10-04 10:37:57 -04:00
Kyle McMartin bc9e072d9b efifb updates from upstream 2010-10-03 17:32:27 -04:00
Ben Skeggs 2e4b75da82 nouveau: more fixes 2010-10-01 15:52:28 +10:00
Dave Jones db140fac23 silence another rcu_reference warning 2010-09-30 16:01:42 -04:00
Ben Skeggs 7043d42b5a nouveau: fix potential dma race 2010-09-30 13:19:07 +10:00
Dave Jones 452369b936 build fix, again. I suck. 2010-09-29 18:50:05 -04:00
Dave Jones 8f3d6a8c3f build fix 2010-09-29 18:31:36 -04:00
Dave Jones fbce23b0ff bump for build 2010-09-29 18:11:07 -04:00
Dave Jones bbaea70ee5 Add back an old hack to make an SDV e1000e variant work. 2010-09-29 18:07:59 -04:00
Dave Jones 04093b926b Enable IB700 watchdog (used by qemu/kvm). (#637152) 2010-09-29 12:34:17 -04:00
Chuck Ebbert 4141bfc9b1 clean up patches merged in -stable 2010-09-27 13:44:47 -04:00
Kyle McMartin 05f615c82e Linux 2.6.35.6 2010-09-26 22:43:03 -04:00
Chuck Ebbert b999dfdb6a Add xen-fix-typo-in-xen-irq-fix.patch, fixes typo in 2.6.35.5 patch. 2010-09-25 08:38:14 -04:00
Chuck Ebbert b49b116bb0 Add sched-35-increment-cache_nice_tries-only-on-periodic-lb.patch 2010-09-25 08:29:41 -04:00
Chuck Ebbert eda054e8bd Linux 2.6.35.6-rc1
Comment out merged patches:
 aio-check-for-multiplication-overflow-in-do_io_submit.patch
 linux-2.6.35.4-virtio_console-fix-poll.patch
 fix-unprotected-access-to-task-credentials-in-whatid.patch
 dell-wmi-add-support-for-eject-key-studio-1555.patch
 irda-correctly-clean-up-self-ias_obj-on-irda_bind-failure.patch
 keys-fix-bug-in-keyctl_session_to_parent-if-parent-has-no-session-keyring.patch
 keys-fix-rcu-no-lock-warning-in-keyctl_session_to_parent.patch
 sched-00-fix-user-time-incorrectly-accounted-as-system-time-on-32-bit.patch
Revert: "drm/nv50: initialize ramht_refs list for faked 0 channel"
(our DRM update removes ramht_refs entirely.)
2010-09-25 08:18:19 -04:00
Kyle McMartin c2aa9135d2 Bump baserelease 2010-09-23 23:02:02 -04:00
Kyle McMartin 2348a607b7 Serialize mandocs/htmldocs build 2010-09-23 23:01:13 -04:00
Kyle McMartin b13dd03f62 Slay another rcu_dereference_check warning pointed out by rmcgrath 2010-09-23 20:59:39 -04:00
Kyle McMartin 6289a5fca9 Enable -debug flavours and switch default image to release builds
- Bump NR_CPUS on i686 to 64.
2010-09-23 20:39:02 -04:00
Ben Skeggs 31ce2e5e16 nouveau: IGP and DP fixes 2010-09-23 08:54:13 +10:00
Chuck Ebbert b2398b3ef3 Dump stack on failed ATA commands in pata_it821x driver (#632753) 2010-09-22 11:54:31 -04:00
Chuck Ebbert 69a476baa9 Fix possible lockup with new scheduler idle balance code. 2010-09-22 08:40:20 -04:00
Kyle McMartin 027325ab53 Add new btusb ids for MacBookPro from wwoods@ 2010-09-21 11:30:59 -04:00
Chuck Ebbert b218718b2b Scheduler fixes for Bugzilla #635813 and #633037 2010-09-21 09:27:25 -04:00
Chuck Ebbert 7a68cb0b55 Linux 2.6.35.5
Drop merged patches:
01-compat-make-compat_alloc_user_space-incorporate-the-access_ok-check.patch
02-compat-test-rax-for-the-system-call-number-not-eax.patch
03-compat-retruncate-rax-after-ia32-syscall-entry-tracing.patch
direct-io-move-aio_complete-into-end_io.patch
ext4-move-aio-completion-after-unwritten-extent-conversion.patch
xfs-move-aio-completion-after-unwritten-extent-conversion.patch
alsa-seq-oss-fix-double-free-at-error-path-of-snd_seq_oss_open.patch
2010-09-20 21:04:51 -04:00
Dennis Gilmore 42589b1be2 add changelog entry for sparc fixes 2010-09-16 19:29:12 +00:00
Dennis Gilmore 6daeb4142a setup kernel headers to be buit on sparcv9 for 32 bit sparc 2010-09-16 19:26:54 +00:00
Dennis Gilmore c549b813ae disable some modules to get the kerenl building again on sparc 2010-09-16 19:26:53 +00:00
Hans de Goede 7193965656 - Small fix to virtio_console poll fix from upstream review 2010-09-16 10:53:51 +02:00
Dave Jones 291833d307 Fix another RCU lockdep warning (cgroups). 2010-09-15 16:12:01 -04:00
Hans de Goede cb6f3ab735 - virtio_console: Fix poll/select blocking even though there is data to read 2010-09-15 17:33:15 +02:00
Chuck Ebbert db056c27c3 Fix 3 CVEs
/dev/sequencer open failure is not handled correctly (CVE-2010-3080)
NULL deref and panic in irda (CVE-2010-2954)
keyctl_session_to_parent NULL deref system crash (CVE-2010-2960)
2010-09-14 21:31:20 -04:00
Chuck Ebbert 14525abb6d Fix DOS with large argument lists. 2010-09-14 21:05:09 -04:00
Kyle McMartin a1b0ec05b6 security fixes from git head 2010-09-14 20:51:27 -04:00
Chuck Ebbert 9bd241a830 Add support for perl and python scripting to perf (#632942) 2010-09-14 00:35:08 -04:00
Ben Skeggs 4badda4370 nouveau: fix oops in acpi edid support 2010-09-13 09:11:50 +10:00
Jarod Wilson 17a76a95c4 ir-core rebase to current upstream
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-09-10 11:57:42 -04:00
Bastien Nocera 9ea075ebba Fix patch application 2010-09-10 16:42:02 +01:00
Bastien Nocera ea6d873cc2 - Update AppleIR patch to work, and support the enter key on
newer remotes
2010-09-10 16:11:56 +01:00
Kyle McMartin 88a50a0b1d Revert 102e2d67
This reverts commit 102e2d6718 which
accidentally snuck into the wrong branch.
2010-09-10 10:33:28 -04:00
Kyle McMartin 102e2d6718 bz620313 2010-09-10 10:28:50 -04:00
Chuck Ebbert 63fcde2459 Disable asynchronous suspend by default. 2010-09-10 10:17:06 -04:00
Ben Skeggs fbc4283e12 nouveau: disable accel on nva3/nva5/nva8 2010-09-10 15:22:18 +10:00
Kyle McMartin 3ceeb77a79 enable GPIO_SYSFS 2010-09-08 20:52:22 -04:00
Kyle McMartin 72b3bf972e linux-2.6-defaults-pci_use_crs 2010-09-08 15:09:01 -04:00
Ben Skeggs 0fe99e9810 nouveau: fix compile error due to upstream differences 2010-09-08 16:49:21 +10:00
Ben Skeggs 215a71b03d nouveau: more updates 2010-09-08 16:23:30 +10:00
Dave Jones 2eba9aa0ec Disable hung task checker, it only ever causes false positives. (#630777) 2010-09-07 12:40:07 -04:00
Jarod Wilson f4da561c3d Enhance HID layer to fully support TiVo Slide remote and dongle
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-09-07 12:07:10 -04:00
Kyle McMartin 8b695bdd15 flexcop: fix registering braindead stupid names (#575873) 2010-09-06 16:05:54 -04:00
Kyle McMartin 426c30b2f1 add in f13 patch to fix eject key on dell studio 1555 2010-09-06 15:17:29 -04:00
Kyle McMartin cfdae029cc fix rcu_dereference_check warning 2010-09-06 14:22:48 -04:00
Jarod Wilson 8ccfe4ce6a Restore the rest of the appleir driver patch
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-09-06 02:10:02 -04:00
Ben Skeggs 1001ab34f7 nouveau: misc fixes from upstream + NVAF support
This looks far far worse than it is... The reason:
 - NVAF support came after nv50_grctx.c was de-magiced (ie. lots of
   comments explaining what all the hexified numbers are for)
 - nv50_grctx.c de-magic depended on an earlier commit which changed
   all gpu object accessors from 32-bit indices to byte offsets

All in all, it was easier to pull in those big scary-looking (but
harmless) commits than to risk messing up a backport.
2010-09-06 15:55:25 +10:00
Kyle McMartin d3a72eaedc restore appleir driver that got lost from f13 2010-09-03 13:25:36 -04:00
Chuck Ebbert 44c4f071d9 Re-enable I2O, but only for 32-bit x86 (#629676) 2010-09-03 05:23:58 -04:00
Dave Jones 7def7eaed3 Scatter-gather on via-velocity is hopelessly broken.
Just switch it off for now.
2010-09-02 17:23:32 -04:00
Dave Jones 35cc504508 Simplify the VIA Velocity changes. The last round of fixes introduced some bugs. 2010-09-02 13:15:02 -04:00
Dave Jones 705539efb5 One more velocity fix. 2010-09-01 21:45:56 -04:00
Dave Jones c348adbe39 Improved version of the VIA Velocity DMA unmapping fix. 2010-09-01 16:45:48 -04:00
Chuck Ebbert 0c8a5559c3 Don't build UIO platform drivers: they require additional platform code. 2010-09-01 08:00:00 -04:00
Kyle McMartin 3c5d4de07c add in patch from lmacken to support more mac models with efifb 2010-08-31 17:23:33 -04:00
Dave Jones 90303f9462 bump release 2010-08-31 11:45:25 -04:00
Dave Jones 10931f1112 Fix incorrect DMA size freeing error in via-velocity. 2010-08-31 11:44:00 -04:00
Chuck Ebbert 27e8302707 Drop obsolete ssb patch 2010-08-27 02:11:04 -04:00
Chuck Ebbert 8f132e348c kprobes-x86-fix-kprobes-to-skip-prefixes-correctly.patch (#610941) 2010-08-27 02:07:13 -04:00
Chuck Ebbert b30c150da1 Linux 2.6.35.4 2010-08-27 01:59:01 -04:00
Chuck Ebbert 3e8e94e0cd Linux 2.6.35.4-rc1
Fix up linux-2.6-i386-nx-emulation.patch for 2.6.35.4
2010-08-25 08:17:36 -04:00
Chuck Ebbert 74e18cdf9e Linux 2.6.35.3
Drop merged patches:
   mm-fix-page-table-unmap-for-stack-guard-page-properly.patch
   mm-fix-up-some-user-visible-effects-of-the-stack-guard-page.patch
2010-08-21 00:29:58 -04:00
Dave Jones 4135368fb0 systemd is dependant upon autofs, so build it in instead of modular. 2010-08-18 19:41:03 -04:00
Chuck Ebbert 13dcde2426 fix duplicate patch numbers 2010-08-17 18:27:57 -04:00
Chuck Ebbert bfc8e84a86 Fix fallout from the stack guard page fixes.
(mm-fix-page-table-unmap-for-stack-guard-page-properly.patch,
 mm-fix-up-some-user-visible-effects-of-the-stack-guard-page.patch)
2010-08-17 18:26:23 -04:00
Kyle McMartin e2302e0109 create a /sys/fs/cgroup mountpoint for cgroupfs 2010-08-17 10:58:11 -04:00
Kyle McMartin 4cfceabb6c fix make local, see %changelog for description 2010-08-17 10:32:29 -04:00
Jarod Wilson b9878dcfec Add ir-core streamzap driver, drop lirc_streamzap
Signed-off-by: Jarod Wilson <jarod@redhat.com>
2010-08-16 11:57:58 -04:00
Chuck Ebbert e1151597d5 Linux 2.6.35.2 2010-08-15 04:26:48 -04:00
Chuck Ebbert d6d51a1a42 - Linux 2.6.35.2-rc1
- Comment out patches merged in -stable:
    linux-2.6-tip.git-396e894d289d69bacf5acd983c97cd6e21a14c08.patch
    linux-2.6-ext4-fix-freeze-deadlock.patch
- New config option:
    CONFIG_CRYPTO_MANAGER_TESTS=y
2010-08-13 07:47:42 -04:00
Chuck Ebbert 4ed348073c - Linux 2.6.35.1 2010-08-10 16:35:19 -04:00
Ben Skeggs 71e9b6763f nouveau: bring in misc fixes + fermi kms support 2010-08-10 11:43:52 +10:00
Chuck Ebbert 74b9092d38 Linux 2.6.35.1-rc1 2010-08-07 12:55:01 -04:00
Chuck Ebbert 9a98a7ebe3 Require newer linux-firmware package 2010-08-07 12:36:54 -04:00
Chuck Ebbert 6c58ce89ec Copy fix for bug #617699 from f13. 2010-08-06 15:34:00 -04:00
Chuck Ebbert 85a693ce78 Disable CONFIG_MULTICORE_RAID456 2010-08-05 14:22:55 -04:00
Dave Jones 4326aa56d8 sched: Revert nohz_ratelimit() which causes a lot of extra wakeups burning CPU, and my legs. 2010-08-04 15:06:07 -04:00
Chuck Ebbert 4d5b945030 Port shared source tree updates from f13. 2010-08-02 10:21:13 -04:00
Chuck Ebbert b6805461ca Merge branch 'master' into f14 2010-08-02 08:25:07 -04:00
Chuck Ebbert 557b0fb7c4 Merge branch 'f14/master' of ssh://pkgs.fedoraproject.org/kernel into f14 2010-08-02 08:14:45 -04:00
Dave Jones f8200febc9 Bump baserelease 2010-08-01 23:25:18 -04:00
Dave Jones 37cd8b7c4c .35 is a released kernel, so set it as such. 2010-08-01 20:57:07 -04:00
101 changed files with 193293 additions and 22199 deletions

View File

@ -88,6 +88,7 @@ debug:
@perl -pi -e 's/# CONFIG_KDB_KEYBOARD is not set/CONFIG_KDB_KEYBOARD=y/' config-nodebug
@perl -pi -e 's/# CONFIG_CPU_NOTIFIER_ERROR_INJECT is not set/CONFIG_CPU_NOTIFIER_ERROR_INJECT=m/' config-nodebug
@perl -pi -e 's/# CONFIG_DEBUG_PER_CPU_MAPS is not set/CONFIG_DEBUG_PER_CPU_MAPS=y/' config-nodebug
#@perl -pi -e 's/# CONFIG_PCI_DEFAULT_USE_CRS is not set/CONFIG_PCI_DEFAULT_USE_CRS=y/' config-nodebug
@# just in case we're going from extremedebug -> debug
@perl -pi -e 's/CONFIG_DEBUG_PAGEALLOC=y/# CONFIG_DEBUG_PAGEALLOC is not set/' config-nodebug
@ -97,6 +98,10 @@ debug:
@perl -pi -e 's/^%define debugbuildsenabled 1/%define debugbuildsenabled 0/' kernel.spec
@perl -pi -e 's/^%define rawhide_skip_docs 0/%define rawhide_skip_docs 1/' kernel.spec
nodebuginfo:
@perl -pi -e 's/^%define with_debuginfo %\{\?_without_debuginfo: 0\} %\{\?\!_without_debuginfo: 1\}/%define with_debuginfo %\{\?_without_debuginfo: 0\} %\{\?\!_without_debuginfo: 0\}/' kernel.spec
nodebug: release
@perl -pi -e 's/^%define debugbuildsenabled 1/%define debugbuildsenabled 0/' kernel.spec
release:
@perl -pi -e 's/CONFIG_SLUB_DEBUG_ON=y/# CONFIG_SLUB_DEBUG_ON is not set/' config-nodebug
@perl -pi -e 's/CONFIG_LOCK_STAT=y/# CONFIG_LOCK_STAT is not set/' config-nodebug
@ -156,6 +161,7 @@ release:
#@perl -pi -e 's/CONFIG_KGDB_KDB=y/# CONFIG_KGDB_KDB is not set/' config-nodebug
#@perl -pi -e 's/CONFIG_KDB_KEYBOARD=y/# CONFIG_KDB_KEYBOARD is not set/' config-nodebug
@perl -pi -e 's/CONFIG_DEBUG_PER_CPU_MAPS=y/# CONFIG_DEBUG_PER_CPU_MAPS is not set/' config-nodebug
#@perl -pi -e 's/CONFIG_PCI_DEFAULT_USE_CRS=y/# CONFIG_PCI_DEFAULT_USE_CRS is not set/' config-nodebug
@perl -pi -e 's/CONFIG_DEBUG_PAGEALLOC=y/# CONFIG_DEBUG_PAGEALLOC is not set/' config-debug
@perl -pi -e 's/CONFIG_DEBUG_PAGEALLOC=y/# CONFIG_DEBUG_PAGEALLOC is not set/' config-nodebug

View File

@ -0,0 +1,702 @@
From 46fadae732d825141f45bf2fbd6381451da26ad7 Mon Sep 17 00:00:00 2001
From: Bastien Nocera <hadess@hadess.net>
Date: Fri, 10 Sep 2010 16:40:46 +0100
Subject: [PATCH] Input: add appleir USB driver
This driver was originally written by James McKenzie, updated by
Greg Kroah-Hartman, further updated by myself, with suspend support
added.
More recent versions of the IR receiver are also supported through
a patch by Alex Karpenko. The patch also adds support for the 2nd
and 5th generation of the controller, and the menu key on newer
brushed metal remotes.
Tested on a MacbookAir1,1
Signed-off-by: Bastien Nocera <hadess@hadess.net>
---
Documentation/input/appleir.txt | 46 ++++
drivers/hid/hid-apple.c | 4 -
drivers/hid/hid-core.c | 7 +-
drivers/hid/hid-ids.h | 5 +-
drivers/input/misc/Kconfig | 13 +
drivers/input/misc/Makefile | 1 +
drivers/input/misc/appleir.c | 519 +++++++++++++++++++++++++++++++++++++++
7 files changed, 588 insertions(+), 7 deletions(-)
create mode 100644 Documentation/input/appleir.txt
create mode 100644 drivers/input/misc/appleir.c
diff --git a/Documentation/input/appleir.txt b/Documentation/input/appleir.txt
new file mode 100644
index 0000000..db637fb
--- /dev/null
+++ b/Documentation/input/appleir.txt
@@ -0,0 +1,46 @@
+Apple IR receiver Driver (appleir)
+----------------------------------
+ Copyright (C) 2009 Bastien Nocera <hadess@hadess.net>
+
+The appleir driver is a kernel input driver to handle Apple's IR
+receivers (and associated remotes) in the kernel.
+
+The driver is an input driver which only handles "official" remotes
+as built and sold by Apple.
+
+Authors
+-------
+
+James McKenzie (original driver)
+Alex Karpenko (05ac:8242 support)
+Greg Kroah-Hartman (cleanups and original submission)
+Bastien Nocera (further cleanups, brushed metal "enter"
+button support and suspend support)
+
+Supported hardware
+------------------
+
+- All Apple laptops and desktops from 2005 onwards, except:
+ - the unibody Macbook (2009)
+ - Mac Pro (all versions)
+- Apple TV (all revisions prior to September 2010)
+
+The remote will only support the 6 (old white) or 7 (brushed metal) buttons
+of the remotes as sold by Apple. See the next section if you want to use
+other remotes or want to use lirc with the device instead of the kernel driver.
+
+Using lirc (native) instead of the kernel driver
+------------------------------------------------
+
+First, you will need to disable the kernel driver for the receiver.
+
+This can be achieved by passing quirks to the usbhid driver.
+The quirk line would be:
+usbhid.quirks=0x05ac:0x8242:0x40000010
+
+With 0x05ac being the vendor ID (Apple, you shouldn't need to change this)
+With 0x8242 being the product ID (check the output of lsusb for your hardware)
+And 0x10 being "HID_QUIRK_HIDDEV_FORCE" and 0x40000000 being "HID_QUIRK_NO_IGNORE"
+
+This should force the creation of a hiddev device for the receiver, and
+make it usable under lirc.
diff --git a/drivers/hid/hid-apple.c b/drivers/hid/hid-apple.c
index bba05d0..0059d5a 100644
--- a/drivers/hid/hid-apple.c
+++ b/drivers/hid/hid-apple.c
@@ -361,10 +361,6 @@ static void apple_remove(struct hid_device *hdev)
}
static const struct hid_device_id apple_devices[] = {
- { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ATV_IRCONTROL),
- .driver_data = APPLE_HIDDEV | APPLE_IGNORE_HIDINPUT },
- { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL4),
- .driver_data = APPLE_HIDDEV | APPLE_IGNORE_HIDINPUT },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MIGHTYMOUSE),
.driver_data = APPLE_MIGHTYMOUSE | APPLE_INVERT_HWHEEL },
diff --git a/drivers/hid/hid-core.c b/drivers/hid/hid-core.c
index 866e54e..1d5e284 100644
--- a/drivers/hid/hid-core.c
+++ b/drivers/hid/hid-core.c
@@ -1239,8 +1239,6 @@ static const struct hid_device_id hid_blacklist[] = {
{ HID_USB_DEVICE(USB_VENDOR_ID_3M, USB_DEVICE_ID_3M2256) },
{ HID_USB_DEVICE(USB_VENDOR_ID_A4TECH, USB_DEVICE_ID_A4TECH_WCP32PU) },
{ HID_USB_DEVICE(USB_VENDOR_ID_A4TECH, USB_DEVICE_ID_A4TECH_X5_005D) },
- { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ATV_IRCONTROL) },
- { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL4) },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MIGHTYMOUSE) },
{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_MAGICMOUSE) },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_ANSI) },
@@ -1571,6 +1569,11 @@ static const struct hid_device_id hid_ignore_list[] = {
{ HID_USB_DEVICE(USB_VENDOR_ID_AIPTEK, USB_DEVICE_ID_AIPTEK_24) },
{ HID_USB_DEVICE(USB_VENDOR_ID_AIRCABLE, USB_DEVICE_ID_AIRCABLE1) },
{ HID_USB_DEVICE(USB_VENDOR_ID_ALCOR, USB_DEVICE_ID_ALCOR_USBRS232) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL2) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL3) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL4) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL5) },
{ HID_USB_DEVICE(USB_VENDOR_ID_ASUS, USB_DEVICE_ID_ASUS_T91MT)},
{ HID_USB_DEVICE(USB_VENDOR_ID_ASUSTEK, USB_DEVICE_ID_ASUSTEK_LCM)},
{ HID_USB_DEVICE(USB_VENDOR_ID_ASUSTEK, USB_DEVICE_ID_ASUSTEK_LCM2)},
diff --git a/drivers/hid/hid-ids.h b/drivers/hid/hid-ids.h
index 31601ee..9afe3bc 100644
--- a/drivers/hid/hid-ids.h
+++ b/drivers/hid/hid-ids.h
@@ -98,8 +98,11 @@
#define USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_JIS 0x023b
#define USB_DEVICE_ID_APPLE_FOUNTAIN_TP_ONLY 0x030a
#define USB_DEVICE_ID_APPLE_GEYSER1_TP_ONLY 0x030b
-#define USB_DEVICE_ID_APPLE_ATV_IRCONTROL 0x8241
+#define USB_DEVICE_ID_APPLE_IRCONTROL 0x8240
+#define USB_DEVICE_ID_APPLE_IRCONTROL2 0x1440
+#define USB_DEVICE_ID_APPLE_IRCONTROL3 0x8241
#define USB_DEVICE_ID_APPLE_IRCONTROL4 0x8242
+#define USB_DEVICE_ID_APPLE_IRCONTROL5 0x8243
#define USB_VENDOR_ID_ASUS 0x0486
#define USB_DEVICE_ID_ASUS_T91MT 0x0185
diff --git a/drivers/input/misc/Kconfig b/drivers/input/misc/Kconfig
index c44b9ea..76a12b7 100644
--- a/drivers/input/misc/Kconfig
+++ b/drivers/input/misc/Kconfig
@@ -199,6 +199,19 @@ config INPUT_KEYSPAN_REMOTE
To compile this driver as a module, choose M here: the module will
be called keyspan_remote.
+config INPUT_APPLEIR
+ tristate "Apple infrared receiver (built in)"
+ depends on USB_ARCH_HAS_HCD
+ select USB
+ help
+ Say Y here if you want to use a Apple infrared remote control. All
+ the Apple computers from 2005 onwards include such a port, except
+ the unibody Macbook (2009), and Mac Pros. This receiver is also
+ used in the Apple TV set-top box prior to the 2010 model.
+
+ To compile this driver as a module, choose M here: the module will
+ be called appleir.
+
config INPUT_POWERMATE
tristate "Griffin PowerMate and Contour Jog support"
depends on USB_ARCH_HAS_HCD
diff --git a/drivers/input/misc/Makefile b/drivers/input/misc/Makefile
index 71fe57d..62a5c60 100644
--- a/drivers/input/misc/Makefile
+++ b/drivers/input/misc/Makefile
@@ -9,6 +9,7 @@ obj-$(CONFIG_INPUT_AD714X) += ad714x.o
obj-$(CONFIG_INPUT_AD714X_I2C) += ad714x-i2c.o
obj-$(CONFIG_INPUT_AD714X_SPI) += ad714x-spi.o
obj-$(CONFIG_INPUT_APANEL) += apanel.o
+obj-$(CONFIG_INPUT_APPLEIR) += appleir.o
obj-$(CONFIG_INPUT_ATI_REMOTE) += ati_remote.o
obj-$(CONFIG_INPUT_ATI_REMOTE2) += ati_remote2.o
obj-$(CONFIG_INPUT_ATLAS_BTNS) += atlas_btns.o
diff --git a/drivers/input/misc/appleir.c b/drivers/input/misc/appleir.c
new file mode 100644
index 0000000..3817a3c
--- /dev/null
+++ b/drivers/input/misc/appleir.c
@@ -0,0 +1,519 @@
+/*
+ * appleir: USB driver for the apple ir device
+ *
+ * Original driver written by James McKenzie
+ * Ported to recent 2.6 kernel versions by Greg Kroah-Hartman <gregkh@suse.de>
+ *
+ * Copyright (C) 2006 James McKenzie
+ * Copyright (C) 2008 Greg Kroah-Hartman <greg@kroah.com>
+ * Copyright (C) 2008 Novell Inc.
+ *
+ * This program is free software; you can redistribute it and/or modify it
+ * under the terms of the GNU General Public License as published by the Free
+ * Software Foundation, version 2.
+ *
+ */
+
+#include <linux/kernel.h>
+#include <linux/slab.h>
+#include <linux/input.h>
+#include <linux/usb/input.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/usb.h>
+#include <linux/usb/input.h>
+#include <asm/unaligned.h>
+#include <asm/byteorder.h>
+
+#define DRIVER_VERSION "v1.2"
+#define DRIVER_AUTHOR "James McKenzie"
+#define DRIVER_DESC "Apple infrared receiver driver"
+#define DRIVER_LICENSE "GPL"
+
+MODULE_AUTHOR(DRIVER_AUTHOR);
+MODULE_DESCRIPTION(DRIVER_DESC);
+MODULE_LICENSE(DRIVER_LICENSE);
+
+#define USB_VENDOR_ID_APPLE 0x05ac
+#define USB_DEVICE_ID_APPLE_IRCONTROL 0x8240
+#define USB_DEVICE_ID_APPLE_IRCONTROL2 0x1440
+#define USB_DEVICE_ID_APPLE_IRCONTROL3 0x8241
+#define USB_DEVICE_ID_APPLE_IRCONTROL4 0x8242
+#define USB_DEVICE_ID_APPLE_IRCONTROL5 0x8243
+
+#define URB_SIZE 32
+
+#define MAX_KEYS 9
+#define MAX_KEYS_MASK (MAX_KEYS - 1)
+
+#define dbginfo(dev, format, arg...) do { if (debug) dev_info(dev , format , ## arg); } while (0)
+
+static int debug;
+module_param(debug, int, 0644);
+MODULE_PARM_DESC(debug, "Enable extra debug messages and information");
+
+/* I have two devices both of which report the following */
+/* 25 87 ee 83 0a + */
+/* 25 87 ee 83 0c - */
+/* 25 87 ee 83 09 << */
+/* 25 87 ee 83 06 >> */
+/* 25 87 ee 83 05 >" */
+/* 25 87 ee 83 03 menu */
+/* 26 00 00 00 00 for key repeat*/
+
+/* Thomas Glanzmann reports the following responses */
+/* 25 87 ee ca 0b + */
+/* 25 87 ee ca 0d - */
+/* 25 87 ee ca 08 << */
+/* 25 87 ee ca 07 >> */
+/* 25 87 ee ca 04 >" */
+/* 25 87 ee ca 02 menu */
+/* 26 00 00 00 00 for key repeat*/
+/* He also observes the following event sometimes */
+/* sent after a key is release, which I interpret */
+/* as a flat battery message */
+/* 25 87 e0 ca 06 flat battery */
+
+/* Alexandre Karpenko reports the following responses for Device ID 0x8242 */
+/* 25 87 ee 47 0b + */
+/* 25 87 ee 47 0d - */
+/* 25 87 ee 47 08 << */
+/* 25 87 ee 47 07 >> */
+/* 25 87 ee 47 04 >" */
+/* 25 87 ee 47 02 menu */
+/* 26 87 ee 47 ** for key repeat (** is the code of the key being held) */
+
+/* Bastien Nocera's "new" remote */
+/* 25 87 ee 91 5f followed by
+ * 25 87 ee 91 05 gives you >"
+ *
+ * 25 87 ee 91 5c followed by
+ * 25 87 ee 91 05 gives you the middle button */
+
+static const unsigned short appleir_key_table[] = {
+ KEY_RESERVED,
+ KEY_MENU,
+ KEY_PLAYPAUSE,
+ KEY_FORWARD,
+ KEY_BACK,
+ KEY_VOLUMEUP,
+ KEY_VOLUMEDOWN,
+ KEY_ENTER,
+ KEY_RESERVED,
+};
+
+struct appleir {
+ struct input_dev *input_dev;
+ unsigned short keymap[ARRAY_SIZE(appleir_key_table)];
+ u8 *data;
+ dma_addr_t dma_buf;
+ struct usb_device *usbdev;
+ unsigned int flags;
+ struct urb *urb;
+ struct timer_list key_up_timer;
+ int current_key;
+ int prev_key_idx;
+ char phys[32];
+};
+
+static DEFINE_MUTEX(appleir_mutex);
+
+enum {
+ APPLEIR_OPENED = 0x1,
+ APPLEIR_SUSPENDED = 0x2,
+};
+
+static struct usb_device_id appleir_ids[] = {
+ { USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL) },
+ { USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL2) },
+ { USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL3) },
+ { USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL4) },
+ { USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_IRCONTROL5) },
+ {}
+};
+MODULE_DEVICE_TABLE(usb, appleir_ids);
+
+static void dump_packet(struct appleir *appleir, char *msg, u8 *data, int len)
+{
+ int i;
+
+ printk(KERN_ERR "appleir: %s (%d bytes)", msg, len);
+
+ for (i = 0; i < len; ++i)
+ printk(" %02x", data[i]);
+ printk(" (should be command %d)\n", (data[4] >> 1) & MAX_KEYS_MASK);
+}
+
+static int get_key(int data)
+{
+ switch (data) {
+ case 0x02:
+ case 0x03:
+ /* menu */
+ return 1;
+ case 0x04:
+ case 0x05:
+ /* >" */
+ return 2;
+ case 0x06:
+ case 0x07:
+ /* >> */
+ return 3;
+ case 0x08:
+ case 0x09:
+ /* << */
+ return 4;
+ case 0x0a:
+ case 0x0b:
+ /* + */
+ return 5;
+ case 0x0c:
+ case 0x0d:
+ /* - */
+ return 6;
+ case 0x5c:
+ /* Middle button, on newer remotes,
+ * part of a 2 packet-command */
+ return -7;
+ default:
+ return -1;
+ }
+}
+
+static void key_up(struct appleir *appleir, int key)
+{
+ dbginfo(&appleir->input_dev->dev, "key %d up\n", key);
+ input_report_key(appleir->input_dev, key, 0);
+ input_sync(appleir->input_dev);
+}
+
+static void key_down(struct appleir *appleir, int key)
+{
+ dbginfo(&appleir->input_dev->dev, "key %d down\n", key);
+ input_report_key(appleir->input_dev, key, 1);
+ input_sync(appleir->input_dev);
+}
+
+static void battery_flat(struct appleir *appleir)
+{
+ dev_err(&appleir->input_dev->dev, "possible flat battery?\n");
+}
+
+static void key_up_tick(unsigned long data)
+{
+ struct appleir *appleir = (struct appleir *)data;
+
+ if (appleir->current_key) {
+ key_up(appleir, appleir->current_key);
+ appleir->current_key = 0;
+ }
+}
+
+static void new_data(struct appleir *appleir, u8 *data, int len)
+{
+ static const u8 keydown[] = { 0x25, 0x87, 0xee };
+ static const u8 keyrepeat[] = { 0x26, };
+ static const u8 flatbattery[] = { 0x25, 0x87, 0xe0 };
+
+ if (debug)
+ dump_packet(appleir, "received", data, len);
+
+ if (len != 5)
+ return;
+
+ if (!memcmp(data, keydown, sizeof(keydown))) {
+ int index;
+
+ /* If we already have a key down, take it up before marking
+ this one down */
+ if (appleir->current_key)
+ key_up(appleir, appleir->current_key);
+
+ /* Handle dual packet commands */
+ if (appleir->prev_key_idx > 0)
+ index = appleir->prev_key_idx;
+ else
+ index = get_key(data[4]);
+
+ if (index > 0) {
+ appleir->current_key = appleir->keymap[index];
+
+ key_down(appleir, appleir->current_key);
+ /* Remote doesn't do key up, either pull them up, in the test
+ above, or here set a timer which pulls them up after 1/8 s */
+ mod_timer(&appleir->key_up_timer, jiffies + HZ / 8);
+ appleir->prev_key_idx = 0;
+ return;
+ } else if (index == -7) {
+ /* Remember key for next packet */
+ appleir->prev_key_idx = 0 - index;
+ return;
+ }
+ }
+
+ appleir->prev_key_idx = 0;
+
+ if (!memcmp(data, keyrepeat, sizeof(keyrepeat))) {
+ key_down(appleir, appleir->current_key);
+ /* Remote doesn't do key up, either pull them up, in the test
+ above, or here set a timer which pulls them up after 1/8 s */
+ mod_timer(&appleir->key_up_timer, jiffies + HZ / 8);
+ return;
+ }
+
+ if (!memcmp(data, flatbattery, sizeof(flatbattery))) {
+ battery_flat(appleir);
+ /* Fall through */
+ }
+
+ dump_packet(appleir, "unknown packet", data, len);
+}
+
+static void appleir_urb(struct urb *urb)
+{
+ struct appleir *appleir = urb->context;
+ int status = urb->status;
+ int retval;
+
+ switch (status) {
+ case 0:
+ new_data(appleir, urb->transfer_buffer, urb->actual_length);
+ break;
+ case -ECONNRESET:
+ case -ENOENT:
+ case -ESHUTDOWN:
+ /* This urb is terminated, clean up */
+ dbginfo(&appleir->input_dev->dev, "%s - urb shutting down with status: %d", __func__,
+ urb->status);
+ return;
+ default:
+ dbginfo(&appleir->input_dev->dev, "%s - nonzero urb status received: %d", __func__,
+ urb->status);
+ }
+
+ retval = usb_submit_urb(urb, GFP_ATOMIC);
+ if (retval)
+ err("%s - usb_submit_urb failed with result %d", __func__,
+ retval);
+}
+
+static int appleir_open(struct input_dev *dev)
+{
+ struct appleir *appleir = input_get_drvdata(dev);
+ struct usb_interface *intf = usb_ifnum_to_if(appleir->usbdev, 0);
+ int r;
+
+ r = usb_autopm_get_interface(intf);
+ if (r) {
+ dev_err(&intf->dev,
+ "%s(): usb_autopm_get_interface() = %d\n", __func__, r);
+ return r;
+ }
+
+ mutex_lock(&appleir_mutex);
+
+ if (usb_submit_urb(appleir->urb, GFP_ATOMIC)) {
+ r = -EIO;
+ goto fail;
+ }
+
+ appleir->flags |= APPLEIR_OPENED;
+
+ mutex_unlock(&appleir_mutex);
+
+ usb_autopm_put_interface(intf);
+
+ return 0;
+fail:
+ mutex_unlock(&appleir_mutex);
+ usb_autopm_put_interface(intf);
+ return r;
+}
+
+static void appleir_close(struct input_dev *dev)
+{
+ struct appleir *appleir = input_get_drvdata(dev);
+
+ mutex_lock(&appleir_mutex);
+
+ if (!(appleir->flags & APPLEIR_SUSPENDED)) {
+ usb_kill_urb(appleir->urb);
+ del_timer_sync(&appleir->key_up_timer);
+ }
+
+ appleir->flags &= ~APPLEIR_OPENED;
+
+ mutex_unlock(&appleir_mutex);
+}
+
+static int appleir_probe(struct usb_interface *intf,
+ const struct usb_device_id *id)
+{
+ struct usb_device *dev = interface_to_usbdev(intf);
+ struct usb_endpoint_descriptor *endpoint;
+ struct appleir *appleir = NULL;
+ struct input_dev *input_dev;
+ int retval = -ENOMEM;
+ int i;
+
+ appleir = kzalloc(sizeof(struct appleir), GFP_KERNEL);
+ if (!appleir)
+ goto allocfail;
+
+ appleir->data = usb_alloc_coherent(dev, URB_SIZE, GFP_KERNEL,
+ &appleir->dma_buf);
+ if (!appleir->data)
+ goto usbfail;
+
+ appleir->urb = usb_alloc_urb(0, GFP_KERNEL);
+ if (!appleir->urb)
+ goto urbfail;
+
+ appleir->usbdev = dev;
+
+ input_dev = input_allocate_device();
+ if (!input_dev)
+ goto inputfail;
+
+ appleir->input_dev = input_dev;
+
+ usb_make_path(dev, appleir->phys, sizeof(appleir->phys));
+ strlcpy(appleir->phys, "/input0", sizeof(appleir->phys));
+
+ input_dev->name = "Apple Infrared Remote Controller";
+ input_dev->phys = appleir->phys;
+ usb_to_input_id(dev, &input_dev->id);
+ input_dev->dev.parent = &intf->dev;
+ input_dev->keycode = appleir->keymap;
+ input_dev->keycodesize = sizeof(unsigned short);
+ input_dev->keycodemax = ARRAY_SIZE(appleir->keymap);
+
+ input_dev->evbit[0] = BIT(EV_KEY) | BIT(EV_REP);
+
+ memcpy(appleir->keymap, appleir_key_table, sizeof(appleir->keymap));
+ for (i = 0; i < ARRAY_SIZE(appleir_key_table); i++)
+ set_bit(appleir->keymap[i], input_dev->keybit);
+ clear_bit(KEY_RESERVED, input_dev->keybit);
+
+ input_set_drvdata(input_dev, appleir);
+ input_dev->open = appleir_open;
+ input_dev->close = appleir_close;
+
+ endpoint = &intf->cur_altsetting->endpoint[0].desc;
+
+ usb_fill_int_urb(appleir->urb, dev,
+ usb_rcvintpipe(dev, endpoint->bEndpointAddress),
+ appleir->data, 8,
+ appleir_urb, appleir, endpoint->bInterval);
+
+ appleir->urb->transfer_dma = appleir->dma_buf;
+ appleir->urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+
+ setup_timer(&appleir->key_up_timer,
+ key_up_tick, (unsigned long) appleir);
+
+ retval = input_register_device(appleir->input_dev);
+ if (retval)
+ goto inputfail;
+
+ usb_set_intfdata(intf, appleir);
+
+ return 0;
+
+inputfail:
+ input_free_device(appleir->input_dev);
+
+urbfail:
+ usb_free_urb(appleir->urb);
+
+usbfail:
+ usb_free_coherent(dev, URB_SIZE, appleir->data,
+ appleir->dma_buf);
+
+allocfail:
+ kfree(appleir);
+
+ return retval;
+}
+
+static void appleir_disconnect(struct usb_interface *intf)
+{
+ struct appleir *appleir = usb_get_intfdata(intf);
+
+ usb_set_intfdata(intf, NULL);
+ input_unregister_device(appleir->input_dev);
+ usb_free_urb(appleir->urb);
+ usb_free_coherent(interface_to_usbdev(intf), URB_SIZE,
+ appleir->data, appleir->dma_buf);
+ kfree(appleir);
+}
+
+static int appleir_suspend(struct usb_interface *interface,
+ pm_message_t message)
+{
+ struct appleir *appleir = usb_get_intfdata(interface);
+
+ mutex_lock(&appleir_mutex);
+ if (appleir->flags & APPLEIR_OPENED)
+ usb_kill_urb(appleir->urb);
+
+ appleir->flags |= APPLEIR_SUSPENDED;
+
+ mutex_unlock(&appleir_mutex);
+
+ return 0;
+}
+
+static int appleir_resume(struct usb_interface *interface)
+{
+ struct appleir *appleir;
+ int r = 0;
+
+ appleir = usb_get_intfdata(interface);
+
+ mutex_lock(&appleir_mutex);
+ if (appleir->flags & APPLEIR_OPENED) {
+ struct usb_endpoint_descriptor *endpoint;
+
+ endpoint = &interface->cur_altsetting->endpoint[0].desc;
+ usb_fill_int_urb(appleir->urb, appleir->usbdev,
+ usb_rcvintpipe(appleir->usbdev, endpoint->bEndpointAddress),
+ appleir->data, 8,
+ appleir_urb, appleir, endpoint->bInterval);
+ appleir->urb->transfer_dma = appleir->dma_buf;
+ appleir->urb->transfer_flags |= URB_NO_TRANSFER_DMA_MAP;
+
+ /* And reset the USB device */
+ if (usb_submit_urb(appleir->urb, GFP_ATOMIC))
+ r = -EIO;
+ }
+
+ appleir->flags &= ~APPLEIR_SUSPENDED;
+
+ mutex_unlock(&appleir_mutex);
+
+ return r;
+}
+
+static struct usb_driver appleir_driver = {
+ .name = "appleir",
+ .probe = appleir_probe,
+ .disconnect = appleir_disconnect,
+ .suspend = appleir_suspend,
+ .resume = appleir_resume,
+ .reset_resume = appleir_resume,
+ .id_table = appleir_ids,
+};
+
+static int __init appleir_init(void)
+{
+ return usb_register(&appleir_driver);
+}
+
+static void __exit appleir_exit(void)
+{
+ usb_deregister(&appleir_driver);
+}
+
+module_init(appleir_init);
+module_exit(appleir_exit);
--
1.7.2.2

263
add-macbookair3-ids.patch Normal file
View File

@ -0,0 +1,263 @@
diff -uNrp kernel-2.6.35.fc14.orig/drivers/bluetooth/btusb.c kernel-2.6.35.fc14.new/drivers/bluetooth/btusb.c
--- kernel-2.6.35.fc14.orig/drivers/bluetooth/btusb.c 2010-11-12 12:35:49.390791080 +0100
+++ kernel-2.6.35.fc14.new/drivers/bluetooth/btusb.c 2010-11-12 12:48:22.090611963 +0100
@@ -68,6 +68,9 @@ static struct usb_device_id btusb_table[
/* Apple MacBookPro6,2 */
{ USB_DEVICE(0x05ac, 0x8218) },
+ /* Apple MacBookAir3,1, MacBookAir3,2 */
+ { USB_DEVICE(0x05ac, 0x821b) },
+
/* AVM BlueFRITZ! USB v2.0 */
{ USB_DEVICE(0x057c, 0x3800) },
diff -uNrp kernel-2.6.35.fc14.orig/drivers/hid/hid-apple.c kernel-2.6.35.fc14.new/drivers/hid/hid-apple.c
--- kernel-2.6.35.fc14.orig/drivers/hid/hid-apple.c 2010-11-12 12:35:49.153805968 +0100
+++ kernel-2.6.35.fc14.new/drivers/hid/hid-apple.c 2010-11-12 12:48:35.689816431 +0100
@@ -59,6 +59,27 @@ struct apple_key_translation {
u8 flags;
};
+static const struct apple_key_translation macbookair_fn_keys[] = {
+ { KEY_BACKSPACE, KEY_DELETE },
+ { KEY_ENTER, KEY_INSERT },
+ { KEY_F1, KEY_BRIGHTNESSDOWN, APPLE_FLAG_FKEY },
+ { KEY_F2, KEY_BRIGHTNESSUP, APPLE_FLAG_FKEY },
+ { KEY_F3, KEY_SCALE, APPLE_FLAG_FKEY },
+ { KEY_F4, KEY_DASHBOARD, APPLE_FLAG_FKEY },
+ { KEY_F6, KEY_PREVIOUSSONG, APPLE_FLAG_FKEY },
+ { KEY_F7, KEY_PLAYPAUSE, APPLE_FLAG_FKEY },
+ { KEY_F8, KEY_NEXTSONG, APPLE_FLAG_FKEY },
+ { KEY_F9, KEY_MUTE, APPLE_FLAG_FKEY },
+ { KEY_F10, KEY_VOLUMEDOWN, APPLE_FLAG_FKEY },
+ { KEY_F11, KEY_VOLUMEUP, APPLE_FLAG_FKEY },
+ { KEY_F12, KEY_EJECTCD, APPLE_FLAG_FKEY },
+ { KEY_UP, KEY_PAGEUP },
+ { KEY_DOWN, KEY_PAGEDOWN },
+ { KEY_LEFT, KEY_HOME },
+ { KEY_RIGHT, KEY_END },
+ { }
+};
+
static const struct apple_key_translation apple_fn_keys[] = {
{ KEY_BACKSPACE, KEY_DELETE },
{ KEY_ENTER, KEY_INSERT },
@@ -157,10 +178,14 @@ static int hidinput_apple_event(struct h
if (fnmode) {
int do_translate;
- trans = apple_find_translation((hid->product < 0x21d ||
+ if(hid->product >= 0x023f && hid->product <= 0x0244 ) {
+ trans = apple_find_translation(macbookair_fn_keys, usage->code);
+ } else {
+ trans = apple_find_translation((hid->product < 0x21d ||
hid->product >= 0x300) ?
powerbook_fn_keys : apple_fn_keys,
usage->code);
+ }
if (trans) {
if (test_bit(usage->code, asc->pressed_fn))
do_translate = 1;
@@ -435,6 +460,18 @@ static const struct hid_device_id apple_
.driver_data = APPLE_HAS_FN | APPLE_ISO_KEYBOARD },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING3_JIS),
.driver_data = APPLE_HAS_FN | APPLE_RDESC_JIS },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4_ANSI),
+ .driver_data = APPLE_HAS_FN },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4_ISO),
+ .driver_data = APPLE_HAS_FN | APPLE_ISO_KEYBOARD },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4_JIS),
+ .driver_data = APPLE_HAS_FN | APPLE_RDESC_JIS },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4A_ANSI),
+ .driver_data = APPLE_HAS_FN },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4A_ISO),
+ .driver_data = APPLE_HAS_FN | APPLE_ISO_KEYBOARD },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4A_JIS),
+ .driver_data = APPLE_HAS_FN | APPLE_RDESC_JIS },
{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ANSI),
.driver_data = APPLE_NUMLOCK_EMULATION | APPLE_HAS_FN },
{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ISO),
diff -uNrp kernel-2.6.35.fc14.orig/drivers/hid/hid-core.c kernel-2.6.35.fc14.new/drivers/hid/hid-core.c
--- kernel-2.6.35.fc14.orig/drivers/hid/hid-core.c 2010-11-12 12:35:49.153805968 +0100
+++ kernel-2.6.35.fc14.new/drivers/hid/hid-core.c 2010-11-12 12:48:35.690816373 +0100
@@ -1273,6 +1273,12 @@ static const struct hid_device_id hid_bl
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING3_ANSI) },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING3_ISO) },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING3_JIS) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4_ANSI) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4_ISO) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4_JIS) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4A_ANSI) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4A_ISO) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4A_JIS) },
{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ANSI) },
{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ISO) },
{ HID_BLUETOOTH_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_JIS) },
@@ -1738,6 +1744,12 @@ static const struct hid_device_id hid_mo
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING3_ANSI) },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING3_ISO) },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING3_JIS) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4_ANSI) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4_ISO) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4_JIS) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4A_ANSI) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4A_ISO) },
+ { HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_WELLSPRING4A_JIS) },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_FOUNTAIN_TP_ONLY) },
{ HID_USB_DEVICE(USB_VENDOR_ID_APPLE, USB_DEVICE_ID_APPLE_GEYSER1_TP_ONLY) },
{ }
diff -uNrp kernel-2.6.35.fc14.orig/drivers/hid/hid-ids.h kernel-2.6.35.fc14.new/drivers/hid/hid-ids.h
--- kernel-2.6.35.fc14.orig/drivers/hid/hid-ids.h 2010-11-12 12:35:49.153805968 +0100
+++ kernel-2.6.35.fc14.new/drivers/hid/hid-ids.h 2010-11-12 12:48:35.691816314 +0100
@@ -93,6 +93,12 @@
#define USB_DEVICE_ID_APPLE_WELLSPRING3_ANSI 0x0236
#define USB_DEVICE_ID_APPLE_WELLSPRING3_ISO 0x0237
#define USB_DEVICE_ID_APPLE_WELLSPRING3_JIS 0x0238
+#define USB_DEVICE_ID_APPLE_WELLSPRING4_ANSI 0x023f
+#define USB_DEVICE_ID_APPLE_WELLSPRING4_ISO 0x0240
+#define USB_DEVICE_ID_APPLE_WELLSPRING4_JIS 0x0241
+#define USB_DEVICE_ID_APPLE_WELLSPRING4A_ANSI 0x0242
+#define USB_DEVICE_ID_APPLE_WELLSPRING4A_ISO 0x0243
+#define USB_DEVICE_ID_APPLE_WELLSPRING4A_JIS 0x0244
#define USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ANSI 0x0239
#define USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_ISO 0x023a
#define USB_DEVICE_ID_APPLE_ALU_WIRELESS_2009_JIS 0x023b
diff -uNrp kernel-2.6.35.fc14.orig/drivers/hwmon/applesmc.c kernel-2.6.35.fc14.new/drivers/hwmon/applesmc.c
--- kernel-2.6.35.fc14.orig/drivers/hwmon/applesmc.c 2010-11-12 12:35:49.618776754 +0100
+++ kernel-2.6.35.fc14.new/drivers/hwmon/applesmc.c 2010-11-13 12:25:05.810472278 +0100
@@ -162,6 +162,10 @@ static const char *temperature_sensors_s
/* Set 22: MacBook Pro 7,1 */
{ "TB0T", "TB1T", "TB2T", "TC0D", "TC0P", "TN0D", "TN0P", "TN0S",
"TN1D", "TN1F", "TN1G", "TN1S", "Th1H", "Ts0P", "Ts0S", NULL },
+/* Set 23: MacBook Air 3,1 */
+ { "TB0T", "TB1T", "TB2T", "TC0D", "TC0E", "TC0P", "TC1E", "TCZ3",
+ "TCZ4", "TCZ5", "TG0E", "TG1E", "TG2E", "TGZ3", "TGZ4", "TGZ5",
+ "TH0F", "TH0O", "TM0P" },
};
/* List of keys used to read/write fan speeds */
@@ -1524,11 +1528,21 @@ static __initdata struct dmi_match_data
{ .accelerometer = 1, .light = 1, .temperature_set = 21 },
/* MacBook Pro 7,1: accelerometer, backlight and temperature set 22 */
{ .accelerometer = 1, .light = 1, .temperature_set = 22 },
+/* MacBook Air 3,1: accelerometer, backlight and temperature set 15 */
+ { .accelerometer = 0, .light = 0, .temperature_set = 23 },
};
/* Note that DMI_MATCH(...,"MacBook") will match "MacBookPro1,1".
* So we need to put "Apple MacBook Pro" before "Apple MacBook". */
static __initdata struct dmi_system_id applesmc_whitelist[] = {
+ { applesmc_dmi_match, "Apple MacBook Air 3", {
+ DMI_MATCH(DMI_BOARD_VENDOR, "Apple"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "MacBookAir3,2") },
+ &applesmc_dmi_data[23]},
+ { applesmc_dmi_match, "Apple MacBook Air 3", {
+ DMI_MATCH(DMI_BOARD_VENDOR, "Apple"),
+ DMI_MATCH(DMI_PRODUCT_NAME, "MacBookAir3,1") },
+ &applesmc_dmi_data[23]},
{ applesmc_dmi_match, "Apple MacBook Air 2", {
DMI_MATCH(DMI_BOARD_VENDOR, "Apple"),
DMI_MATCH(DMI_PRODUCT_NAME, "MacBookAir2") },
diff -uNrp kernel-2.6.35.fc14.orig/drivers/input/mouse/bcm5974.c kernel-2.6.35.fc14.new/drivers/input/mouse/bcm5974.c
--- kernel-2.6.35.fc14.orig/drivers/input/mouse/bcm5974.c 2010-11-12 12:35:50.004752503 +0100
+++ kernel-2.6.35.fc14.new/drivers/input/mouse/bcm5974.c 2010-11-12 12:48:13.140136374 +0100
@@ -55,6 +55,14 @@
#define USB_DEVICE_ID_APPLE_WELLSPRING3_ANSI 0x0236
#define USB_DEVICE_ID_APPLE_WELLSPRING3_ISO 0x0237
#define USB_DEVICE_ID_APPLE_WELLSPRING3_JIS 0x0238
+/* MacbookAir3,2 (unibody), aka wellspring5 */
+#define USB_DEVICE_ID_APPLE_WELLSPRING4_ANSI 0x023f
+#define USB_DEVICE_ID_APPLE_WELLSPRING4_ISO 0x0240
+#define USB_DEVICE_ID_APPLE_WELLSPRING4_JIS 0x0241
+/* MacbookAir3,1 (unibody), aka wellspring4 */
+#define USB_DEVICE_ID_APPLE_WELLSPRING4A_ANSI 0x0242
+#define USB_DEVICE_ID_APPLE_WELLSPRING4A_ISO 0x0243
+#define USB_DEVICE_ID_APPLE_WELLSPRING4A_JIS 0x0244
#define BCM5974_DEVICE(prod) { \
.match_flags = (USB_DEVICE_ID_MATCH_DEVICE | \
@@ -80,6 +88,14 @@ static const struct usb_device_id bcm597
BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING3_ANSI),
BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING3_ISO),
BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING3_JIS),
+ /* MacbookAir3,2 */
+ BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING4_ANSI),
+ BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING4_ISO),
+ BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING4_JIS),
+ /* MacbookAir3,1 */
+ BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING4A_ANSI),
+ BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING4A_ISO),
+ BCM5974_DEVICE(USB_DEVICE_ID_APPLE_WELLSPRING4A_JIS),
/* Terminating entry */
{}
};
@@ -234,6 +250,30 @@ static const struct bcm5974_config bcm59
{ DIM_X, DIM_X / SN_COORD, -4460, 5166 },
{ DIM_Y, DIM_Y / SN_COORD, -75, 6700 }
},
+ {
+ USB_DEVICE_ID_APPLE_WELLSPRING4_ANSI,
+ USB_DEVICE_ID_APPLE_WELLSPRING4_ISO,
+ USB_DEVICE_ID_APPLE_WELLSPRING4_JIS,
+ HAS_INTEGRATED_BUTTON,
+ 0x84, sizeof(struct bt_data),
+ 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS,
+ { DIM_PRESSURE, DIM_PRESSURE / SN_PRESSURE, 0, 300 },
+ { DIM_WIDTH, DIM_WIDTH / SN_WIDTH, 0, 2048 },
+ { DIM_X, DIM_X / SN_COORD, -4460, 5166 },
+ { DIM_Y, DIM_Y / SN_COORD, -75, 6700 }
+ },
+ {
+ USB_DEVICE_ID_APPLE_WELLSPRING4A_ANSI,
+ USB_DEVICE_ID_APPLE_WELLSPRING4A_ISO,
+ USB_DEVICE_ID_APPLE_WELLSPRING4A_JIS,
+ HAS_INTEGRATED_BUTTON,
+ 0x84, sizeof(struct bt_data),
+ 0x81, TYPE2, FINGER_TYPE2, FINGER_TYPE2 + SIZEOF_ALL_FINGERS,
+ { DIM_PRESSURE, DIM_PRESSURE / SN_PRESSURE, 0, 300 },
+ { DIM_WIDTH, DIM_WIDTH / SN_WIDTH, 0, 2048 },
+ { DIM_X, DIM_X / SN_COORD, -4460, 5166 },
+ { DIM_Y, DIM_Y / SN_COORD, -75, 6700 }
+ },
{}
};
diff -uNrp kernel-2.6.35.fc14.orig/drivers/video/backlight/mbp_nvidia_bl.c kernel-2.6.35.fc14.new/drivers/video/backlight/mbp_nvidia_bl.c
--- kernel-2.6.35.fc14.orig/drivers/video/backlight/mbp_nvidia_bl.c 2010-11-12 12:35:49.159805591 +0100
+++ kernel-2.6.35.fc14.new/drivers/video/backlight/mbp_nvidia_bl.c 2010-11-12 12:48:47.412131884 +0100
@@ -335,6 +335,24 @@ static const struct dmi_system_id __init
},
.driver_data = (void *)&nvidia_chipset_data,
},
+ {
+ .callback = mbp_dmi_match,
+ .ident = "MacBookAir 3,1",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "MacBookAir3,1"),
+ },
+ .driver_data = (void *)&nvidia_chipset_data,
+ },
+ {
+ .callback = mbp_dmi_match,
+ .ident = "MacBookAir 3,2",
+ .matches = {
+ DMI_MATCH(DMI_SYS_VENDOR, "Apple Inc."),
+ DMI_MATCH(DMI_PRODUCT_NAME, "MacBookAir3,2"),
+ },
+ .driver_data = (void *)&nvidia_chipset_data,
+ },
{ }
};
diff -uNrp kernel-2.6.35.fc14.orig/sound/pci/hda/patch_cirrus.c kernel-2.6.35.fc14.new/sound/pci/hda/patch_cirrus.c
--- kernel-2.6.35.fc14.orig/sound/pci/hda/patch_cirrus.c 2010-11-12 12:35:49.005815268 +0100
+++ kernel-2.6.35.fc14.new/sound/pci/hda/patch_cirrus.c 2010-11-12 12:48:40.379542432 +0100
@@ -1139,6 +1139,7 @@ static const char *cs420x_models[CS420X_
static struct snd_pci_quirk cs420x_cfg_tbl[] = {
SND_PCI_QUIRK(0x10de, 0xcb79, "MacBookPro 5,5", CS420X_MBP55),
SND_PCI_QUIRK(0x8086, 0x7270, "IMac 27 Inch", CS420X_IMAC27),
+ SND_PCI_QUIRK(0x10de, 0x0d94, "MacBookAir 3,1(2)", CS420X_MBP55),
{} /* terminator */
};

View File

@ -0,0 +1,65 @@
From 8f172904b45a6b530eaa345b23956682629bddee Mon Sep 17 00:00:00 2001
From: Josef Bacik <josef@redhat.com>
Date: Fri, 22 Oct 2010 15:26:53 -0400
Subject: Btrfs: fix error handling in btrfs_get_sb
If we failed to find the root subvol id, or the subvol=<name>, we would
deactivate the locked super and close the devices. The problem is at this point
we have gotten the SB all setup, which includes setting super_operations, so
when we'd deactiveate the super, we'd do a close_ctree() which closes the
devices, so we'd end up closing the devices twice. So if you do something like
this
mount /dev/sda1 /mnt/test1
mount /dev/sda1 /mnt/test2 -o subvol=xxx
umount /mnt/test1
it would blow up (if subvol xxx doesn't exist). This patch fixes that problem.
Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
---
fs/btrfs/super.c | 7 +++----
1 files changed, 3 insertions(+), 4 deletions(-)
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index f2393b3..c246f25 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -629,7 +629,7 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
if (IS_ERR(root)) {
error = PTR_ERR(root);
deactivate_locked_super(s);
- goto error;
+ goto error_free_subvol_name;
}
/* if they gave us a subvolume name bind mount into that */
if (strcmp(subvol_name, ".")) {
@@ -643,14 +643,14 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
deactivate_locked_super(s);
error = PTR_ERR(new_root);
dput(root);
- goto error_close_devices;
+ goto error_free_subvol_name;
}
if (!new_root->d_inode) {
dput(root);
dput(new_root);
deactivate_locked_super(s);
error = -ENXIO;
- goto error_close_devices;
+ goto error_free_subvol_name;
}
dput(root);
root = new_root;
@@ -668,7 +668,6 @@ error_close_devices:
btrfs_close_devices(fs_devices);
error_free_subvol_name:
kfree(subvol_name);
-error:
return error;
}
--
1.7.3.3

View File

@ -0,0 +1,41 @@
From 3d07b06c5d62e98b46ef21980e5c2a904990149f Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@mcmartin.ca>
Date: Fri, 10 Dec 2010 10:32:29 -0500
Subject: [PATCH] Btrfs - fix race between btrfs_get_sb() and umount
When mounting a btrfs file system btrfs_test_super() may attempt to
use sb->s_fs_info, the btrfs root, of a super block that is going away
and that has had the btrfs root set to NULL in its ->put_super(). But
if the super block is going away it cannot be an existing super block
so we can return false in this case.
Signed-off-by: Ian Kent <raven@themaw.net>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Conflicts:
fs/btrfs/super.c
---
fs/btrfs/super.c | 6 ++++++
1 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index 89e299f..f4a4dd3 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -551,6 +551,12 @@ static int btrfs_test_super(struct super_block *s, void *data)
struct btrfs_root *test_root = data;
struct btrfs_root *root = btrfs_sb(s);
+ /*
+ * If this super block is going away, return false as it
+ * can't match as an existing super block.
+ */
+ if (!atomic_read(&s->s_active))
+ return 0;
return root->fs_info->fs_devices == test_root->fs_info->fs_devices;
}
--
1.7.3.3

View File

@ -0,0 +1,117 @@
From 2049a8887f699650cd66c3da220e84d5c140c546 Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@mcmartin.ca>
Date: Fri, 10 Dec 2010 10:09:15 -0500
Subject: [PATCH] Btrfs: setup blank root and fs_info for mount time
There is a problem with how we use sget, it searches through the list of supers
attached to the fs_type looking for a super with the same fs_devices as what
we're trying to mount. This depends on sb->s_fs_info being filled, but we don't
fill that in until we get to btrfs_fill_super, so we could hit supers on the
fs_type super list that have a null s_fs_info. In order to fix that we need to
go ahead and setup a blank root with a blank fs_info to hold fs_devices, that
way our test will work out right and then we can set s_fs_info in
btrfs_set_super, and then open_ctree will simply use our pre-allocated root and
fs_info when setting everything up. Thanks,
Signed-off-by: Josef Bacik <josef@redhat.com>
Signed-off-by: Chris Mason <chris.mason@oracle.com>
Conflicts:
fs/btrfs/super.c
---
fs/btrfs/disk-io.c | 6 ++----
fs/btrfs/super.c | 34 +++++++++++++++++++++++++++++++---
2 files changed, 33 insertions(+), 7 deletions(-)
diff --git a/fs/btrfs/disk-io.c b/fs/btrfs/disk-io.c
index 34f7c37..b6c3dad 100644
--- a/fs/btrfs/disk-io.c
+++ b/fs/btrfs/disk-io.c
@@ -1539,10 +1539,8 @@ struct btrfs_root *open_ctree(struct super_block *sb,
GFP_NOFS);
struct btrfs_root *csum_root = kzalloc(sizeof(struct btrfs_root),
GFP_NOFS);
- struct btrfs_root *tree_root = kzalloc(sizeof(struct btrfs_root),
- GFP_NOFS);
- struct btrfs_fs_info *fs_info = kzalloc(sizeof(*fs_info),
- GFP_NOFS);
+ struct btrfs_root *tree_root = btrfs_sb(sb);
+ struct btrfs_fs_info *fs_info = tree_root->fs_info;
struct btrfs_root *chunk_root = kzalloc(sizeof(struct btrfs_root),
GFP_NOFS);
struct btrfs_root *dev_root = kzalloc(sizeof(struct btrfs_root),
diff --git a/fs/btrfs/super.c b/fs/btrfs/super.c
index f2393b3..89e299f 100644
--- a/fs/btrfs/super.c
+++ b/fs/btrfs/super.c
@@ -548,12 +548,20 @@ static int btrfs_show_options(struct seq_file *seq, struct vfsmount *vfs)
static int btrfs_test_super(struct super_block *s, void *data)
{
- struct btrfs_fs_devices *test_fs_devices = data;
+ struct btrfs_root *test_root = data;
struct btrfs_root *root = btrfs_sb(s);
- return root->fs_info->fs_devices == test_fs_devices;
+ return root->fs_info->fs_devices == test_root->fs_info->fs_devices;
}
+static int btrfs_set_super(struct super_block *s, void *data)
+{
+ s->s_fs_info = data;
+
+ return set_anon_super(s, data);
+}
+
+
/*
* Find a superblock for the given device / mount point.
*
@@ -567,6 +575,8 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
struct super_block *s;
struct dentry *root;
struct btrfs_fs_devices *fs_devices = NULL;
+ struct btrfs_root *tree_root = NULL;
+ struct btrfs_fs_info *fs_info = NULL;
fmode_t mode = FMODE_READ;
char *subvol_name = NULL;
u64 subvol_objectid = 0;
@@ -595,8 +605,24 @@ static int btrfs_get_sb(struct file_system_type *fs_type, int flags,
goto error_close_devices;
}
+ /*
+ * Setup a dummy root and fs_info for test/set super. This is because
+ * we don't actually fill this stuff out until open_ctree, but we need
+ * it for searching for existing supers, so this lets us do that and
+ * then open_ctree will properly initialize everything later.
+ */
+ fs_info = kzalloc(sizeof(struct btrfs_fs_info), GFP_NOFS);
+ tree_root = kzalloc(sizeof(struct btrfs_root), GFP_NOFS);
+ if (!fs_info || !tree_root) {
+ error = -ENOMEM;
+ goto error_close_devices;
+ }
+ fs_info->tree_root = tree_root;
+ fs_info->fs_devices = fs_devices;
+ tree_root->fs_info = fs_info;
+
bdev = fs_devices->latest_bdev;
- s = sget(fs_type, btrfs_test_super, set_anon_super, fs_devices);
+ s = sget(fs_type, btrfs_test_super, btrfs_set_super, tree_root);
if (IS_ERR(s))
goto error_s;
@@ -666,6 +692,8 @@ error_s:
error = PTR_ERR(s);
error_close_devices:
btrfs_close_devices(fs_devices);
+ kfree(fs_info);
+ kfree(tree_root);
error_free_subvol_name:
kfree(subvol_name);
error:
--
1.7.3.3

View File

@ -0,0 +1,41 @@
From kernel-bounces@lists.fedoraproject.org Fri Sep 17 17:09:15 2010
From: Will Woods <wwoods@redhat.com>
To: Marcel Holtmann <marcel@holtmann.org>
Subject: [PATCH 2/2] bluetooth: add support for controller in MacBookPro6,2
Date: Fri, 17 Sep 2010 17:09:21 -0400
Once again the device class is ff(vend.) instead of e0(wlcon).
output from 'usb-devices':
T: Bus=01 Lev=03 Prnt=03 Port=02 Cnt=03 Dev#= 8 Spd=12 MxCh= 0
D: Ver= 2.00 Cls=ff(vend.) Sub=01 Prot=01 MxPS=64 #Cfgs= 1
P: Vendor=05ac ProdID=8218 Rev=00.22
S: Manufacturer=Apple Inc.
S: Product=Bluetooth USB Host Controller
C: #Ifs= 4 Cfg#= 1 Atr=e0 MxPwr=0mA
I: If#= 0 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=01 Prot=01 Driver=(none)
I: If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=(none)
I: If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
I: If#= 3 Alt= 0 #EPs= 0 Cls=fe(app. ) Sub=01 Prot=01 Driver=(none)
Signed-off-by: Will Woods <wwoods@redhat.com>
---
drivers/bluetooth/btusb.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index eac44e4..320e798 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -65,6 +65,9 @@ static struct usb_device_id btusb_table[] = {
/* Apple iMac11,1 */
{ USB_DEVICE(0x05ac, 0x8215) },
+ /* Apple MacBookPro6,2 */
+ { USB_DEVICE(0x05ac, 0x8218) },
+
/* AVM BlueFRITZ! USB v2.0 */
{ USB_DEVICE(0x057c, 0x3800) },
--
1.7.2.3

View File

@ -0,0 +1,42 @@
From kernel-bounces@lists.fedoraproject.org Fri Sep 17 17:09:18 2010
From: Will Woods <wwoods@redhat.com>
To: Marcel Holtmann <marcel@holtmann.org>
Subject: [PATCH 1/2] bluetooth: add support for controller in MacBookPro7,1
Date: Fri, 17 Sep 2010 17:09:20 -0400
As with iMac11,1 the device class is ff(vend.) instead of e0(wlcon).
output from 'usb-devices':
T: Bus=04 Lev=02 Prnt=04 Port=00 Cnt=01 Dev#= 5 Spd=12 MxCh= 0
D: Ver= 2.00 Cls=ff(vend.) Sub=01 Prot=01 MxPS=64 #Cfgs= 1
P: Vendor=05ac ProdID=8213 Rev=01.86
S: Manufacturer=Apple Inc.
S: Product=Bluetooth USB Host Controller
S: SerialNumber=58B0359C28ED
C: #Ifs= 4 Cfg#= 1 Atr=e0 MxPwr=0mA
I: If#= 0 Alt= 0 #EPs= 3 Cls=ff(vend.) Sub=01 Prot=01 Driver=btusb
I: If#= 1 Alt= 0 #EPs= 2 Cls=e0(wlcon) Sub=01 Prot=01 Driver=btusb
I: If#= 2 Alt= 0 #EPs= 2 Cls=ff(vend.) Sub=ff Prot=ff Driver=(none)
I: If#= 3 Alt= 0 #EPs= 0 Cls=fe(app. ) Sub=01 Prot=00 Driver=(none)
Signed-off-by: Will Woods <wwoods@redhat.com>
---
drivers/bluetooth/btusb.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/drivers/bluetooth/btusb.c b/drivers/bluetooth/btusb.c
index d22ce3c..eac44e4 100644
--- a/drivers/bluetooth/btusb.c
+++ b/drivers/bluetooth/btusb.c
@@ -59,6 +59,9 @@ static struct usb_device_id btusb_table[] = {
/* Generic Bluetooth USB device */
{ USB_DEVICE_INFO(0xe0, 0x01, 0x01) },
+ /* Apple MacBookPro7,1 */
+ { USB_DEVICE(0x05ac, 0x8213) },
+
/* Apple iMac11,1 */
{ USB_DEVICE(0x05ac, 0x8215) },
--
1.7.2.3

View File

@ -87,5 +87,7 @@ CONFIG_PM_ADVANCED_DEBUG=y
CONFIG_CEPH_FS_PRETTYDEBUG=y
CONFIG_QUOTA_DEBUG=y
# CONFIG_PCI_DEFAULT_USE_CRS is not set
CONFIG_KGDB_KDB=y
CONFIG_KDB_KEYBOARD=y

View File

@ -684,7 +684,7 @@ CONFIG_MD_RAID0=m
CONFIG_MD_RAID1=m
CONFIG_MD_RAID10=m
CONFIG_MD_RAID456=m
CONFIG_MULTICORE_RAID456=y
# CONFIG_MULTICORE_RAID456 is not set
CONFIG_ASYNC_RAID6_TEST=m
CONFIG_BLK_DEV_DM=y
CONFIG_DM_CRYPT=m
@ -1423,11 +1423,11 @@ CONFIG_ATMEL=m
CONFIG_B43=m
CONFIG_B43_PCMCIA=y
CONFIG_B43_SDIO=y
CONFIG_B43_DEBUG=y
# CONFIG_B43_DEBUG is not set
CONFIG_B43_PHY_LP=y
# CONFIG_B43_FORCE_PIO is not set
CONFIG_B43LEGACY=m
CONFIG_B43LEGACY_DEBUG=y
# CONFIG_B43LEGACY_DEBUG is not set
CONFIG_B43LEGACY_DMA=y
CONFIG_B43LEGACY_PIO=y
CONFIG_B43LEGACY_DMA_AND_PIO_MODE=y
@ -2193,7 +2193,7 @@ CONFIG_WDTPCI=m
# CONFIG_ACQUIRE_WDT is not set
# CONFIG_ADVANTECH_WDT is not set
# CONFIG_EUROTECH_WDT is not set
# CONFIG_IB700_WDT is not set
CONFIG_IB700_WDT=m
# CONFIG_MIXCOMWD is not set
# CONFIG_SCx200_WDT is not set
# CONFIG_60XX_WDT is not set
@ -2308,6 +2308,7 @@ CONFIG_DRM_NOUVEAU=m
CONFIG_DRM_NOUVEAU_BACKLIGHT=y
CONFIG_DRM_NOUVEAU_DEBUG=y
CONFIG_DRM_I2C_CH7006=m
CONFIG_DRM_I2C_SIL164=m
CONFIG_DRM_VMWGFX=m
#
@ -2366,6 +2367,7 @@ CONFIG_VIDEO_EM28XX_DVB=m
CONFIG_VIDEO_CX231XX=m
CONFIG_VIDEO_CX231XX_ALSA=m
CONFIG_VIDEO_CX231XX_DVB=m
CONFIG_VIDEO_CX231XX_RC=y
CONFIG_VIDEO_HEXIUM_ORION=m
CONFIG_VIDEO_HEXIUM_GEMINI=m
CONFIG_VIDEO_IVTV=m
@ -2380,6 +2382,7 @@ CONFIG_VIDEO_SAA6588=m
CONFIG_VIDEO_SAA7134=m
CONFIG_VIDEO_SAA7134_ALSA=m
CONFIG_VIDEO_SAA7134_DVB=m
CONFIG_VIDEO_SAA7134_RC=y
CONFIG_VIDEO_STRADIS=m
CONFIG_VIDEO_USBVISION=m
CONFIG_VIDEO_W9966=m
@ -2394,6 +2397,10 @@ CONFIG_VIDEO_ZORAN_ZR36060=m
CONFIG_VIDEO_FB_IVTV=m
CONFIG_VIDEO_SAA7164=m
CONFIG_VIDEO_TLG2300=m
# CONFIG_VIDEO_TIMBERDALE is not set
CONFIG_VIDEO_SR030PC30=m
# Doesn't build on 2.6.35
# CONFIG_VIDEO_VIA_CAMERA is not set
CONFIG_USB_VIDEO_CLASS=m
CONFIG_USB_VIDEO_CLASS_INPUT_EVDEV=y
@ -2409,6 +2416,7 @@ CONFIG_MEDIA_ATTACH=y
CONFIG_MEDIA_TUNER_CUSTOMISE=y
CONFIG_MEDIA_TUNER_SIMPLE=m
CONFIG_MEDIA_TUNER_TDA8290=m
CONFIG_MEDIA_TUNER_TDA18218=m
CONFIG_MEDIA_TUNER_TEA5761=m
CONFIG_MEDIA_TUNER_TEA5767=m
CONFIG_MEDIA_TUNER_MT20XX=m
@ -2491,6 +2499,10 @@ CONFIG_DVB_ATBM8830=m
CONFIG_DVB_TDA665x=m
CONFIG_DVB_STV0299=m
CONFIG_DVB_MB86A16=m
CONFIG_DVB_USB_LME2510=m
CONFIG_DVB_S5H1432=m
CONFIG_DVB_MB86A20S=m
CONFIG_DVB_IX2505V=m
#
# Supported Frontend Modules
@ -2581,14 +2593,21 @@ CONFIG_VIDEO_PVRUSB2_SYSFS=y
# CONFIG_VIDEO_PVRUSB2_DEBUGIFC is not set
CONFIG_RC_MAP=m
CONFIG_RC_CORE=m
CONFIG_IR_NEC_DECODER=m
CONFIG_IR_RC5_DECODER=m
CONFIG_IR_RC6_DECODER=m
CONFIG_IR_JVC_DECODER=m
CONFIG_IR_SONY_DECODER=m
CONFIG_IR_RC5_SZ_DECODER=m
CONFIG_IR_LIRC_CODEC=m
CONFIG_IR_ENE=m
CONFIG_IR_IMON=m
CONFIG_IR_MCEUSB=m
CONFIG_IR_NUVOTON=m
CONFIG_IR_STREAMZAP=m
CONFIG_IR_WINBOND_CIR=m
CONFIG_RC_LOOPBACK=m
CONFIG_V4L_MEM2MEM_DRIVERS=y
# CONFIG_VIDEO_MEM2MEM_TESTDEV is not set
@ -3031,6 +3050,8 @@ CONFIG_USB_GL860=m
CONFIG_USB_GSPCA_JEILINJ=m
CONFIG_USB_GSPCA_SPCA1528=m
CONFIG_USB_GSPCA_SQ930X=m
CONFIG_USB_GSPCA_KONICA=m
CONFIG_USB_GSPCA_XIRLINK_CIT=m
CONFIG_USB_IBMCAM=m
CONFIG_USB_KONICAWC=m
@ -3038,6 +3059,7 @@ CONFIG_USB_KONICAWC=m
CONFIG_USB_S2255=m
CONFIG_USB_SE401=m
# CONFIG_VIDEO_SH_MOBILE_CEU is not set
# CONFIG_VIDEO_SH_MOBILE_CSI2 is not set
# CONFIG_USB_STV680 is not set
# CONFIG_USB_SN9C102 is not set
CONFIG_USB_ZR364XX=m
@ -3052,6 +3074,8 @@ CONFIG_SOC_CAMERA_OV772X=m
CONFIG_SOC_CAMERA_MT9T112=m
CONFIG_SOC_CAMERA_RJ54N1=m
CONFIG_SOC_CAMERA_OV9640=m
CONFIG_SOC_CAMERA_IMX074=m
CONFIG_SOC_CAMERA_OV6650=m
#
# USB Network adaptors
@ -3321,7 +3345,8 @@ CONFIG_QUOTACTL=y
CONFIG_DNOTIFY=y
# Autofsv3 is obsolete.
# CONFIG_AUTOFS_FS is not set
CONFIG_AUTOFS4_FS=m
# systemd is dependant upon AUTOFS, so build it in.
CONFIG_AUTOFS4_FS=y
CONFIG_EXOFS_FS=m
# CONFIG_EXOFS_DEBUG is not set
CONFIG_NILFS2_FS=m
@ -3601,6 +3626,7 @@ CONFIG_CRYPTO_FIPS=y
CONFIG_CRYPTO_HW=y
CONFIG_CRYPTO_BLKCIPHER=y
CONFIG_CRYPTO_MANAGER=m
CONFIG_CRYPTO_MANAGER_TESTS=y
# CONFIG_CRYPTO_CRYPTD is not set
CONFIG_CRYPTO_AES=m
CONFIG_CRYPTO_ARC4=m
@ -3722,7 +3748,7 @@ CONFIG_BLK_CGROUP=y
# CONFIG_SYSFS_DEPRECATED_V2 is not set
CONFIG_RELAY=y
# CONFIG_PRINTK_TIME is not set
CONFIG_PRINTK_TIME=y
CONFIG_ENABLE_MUST_CHECK=y
# CONFIG_ENABLE_WARN_DEPRECATED is not set
@ -3742,7 +3768,7 @@ CONFIG_IBMASR=m
CONFIG_PM_DEBUG=y
CONFIG_PM_TRACE=y
# CONFIG_PM_VERBOSE is not set
CONFIG_PM_TEST_SUSPEND=y
# CONFIG_PM_TEST_SUSPEND is not set
CONFIG_PM_RUNTIME=y
## BEGIN ISA Junk.
@ -3987,8 +4013,8 @@ CONFIG_AUXDISPLAY=y
CONFIG_UIO=m
CONFIG_UIO_CIF=m
CONFIG_UIO_SMX=m
CONFIG_UIO_PDRV=m
CONFIG_UIO_PDRV_GENIRQ=m
# CONFIG_UIO_PDRV is not set
# CONFIG_UIO_PDRV_GENIRQ is not set
CONFIG_UIO_AEC=m
CONFIG_UIO_SERCOS3=m
CONFIG_UIO_PCI_GENERIC=m
@ -4013,7 +4039,7 @@ CONFIG_LIRC_SERIAL=m
CONFIG_LIRC_SERIAL_TRANSMITTER=y
CONFIG_LIRC_SASEM=m
CONFIG_LIRC_SIR=m
CONFIG_LIRC_STREAMZAP=m
# CONFIG_LIRC_STREAMZAP is not set
CONFIG_LIRC_TTUSBIR=m
# CONFIG_SAMPLES is not set
@ -4175,7 +4201,7 @@ CONFIG_USB_ATMEL=m
# CONFIG_RCU_CPU_STALL_DETECTOR is not set
# CONFIG_DEBUG_BLOCK_EXT_DEVT is not set
# CONFIG_FUNCTION_GRAPH_TRACER is not set
CONFIG_BOOT_TRACER=y
# CONFIG_BOOT_TRACER is not set
CONFIG_EARLY_PRINTK_DBGP=y
CONFIG_SECURITYFS=y
@ -4253,7 +4279,7 @@ CONFIG_DEBUG_NX_TEST=m
CONFIG_DEBUG_BOOT_PARAMS=y
CONFIG_DETECT_SOFTLOCKUP=y
# CONFIG_BOOTPARAM_SOFTLOCKUP_PANIC is not set
CONFIG_DETECT_HUNG_TASK=y
# CONFIG_DETECT_HUNG_TASK is not set
# CONFIG_BOOTPARAM_HUNG_TASK_PANIC is not set
CONFIG_ATOMIC64_SELFTEST=y
@ -4269,7 +4295,7 @@ CONFIG_BLK_DEV_DRBD=m
# CONFIG_DEBUG_GPIO is not set
# CONFIG_W1_MASTER_GPIO is not set
# CONFIG_LEDS_GPIO is not set
# CONFIG_GPIO_SYSFS is not set
CONFIG_GPIO_SYSFS=y
# CONFIG_GPIO_MAX732X is not set
# CONFIG_GPIO_PCA953X is not set
# CONFIG_GPIO_PCF857X is not set

View File

@ -3,3 +3,6 @@ CONFIG_HIGHMEM64G=y
CONFIG_XEN_DEV_EVTCHN=m
CONFIG_XEN_SYS_HYPERVISOR=y
# I2O only works on non-PAE 32-bit x86
# CONFIG_I2O is not set

View File

@ -2,90 +2,92 @@ CONFIG_SND_VERBOSE_PRINTK=y
CONFIG_SND_DEBUG=y
CONFIG_SND_PCM_XRUN_DEBUG=y
CONFIG_DEBUG_MUTEXES=y
CONFIG_DEBUG_RT_MUTEXES=y
CONFIG_DEBUG_LOCK_ALLOC=y
CONFIG_PROVE_LOCKING=y
CONFIG_DEBUG_VM=y
CONFIG_DEBUG_SPINLOCK=y
CONFIG_PROVE_RCU=y
# CONFIG_DEBUG_MUTEXES is not set
# CONFIG_DEBUG_RT_MUTEXES is not set
# CONFIG_DEBUG_LOCK_ALLOC is not set
# CONFIG_PROVE_LOCKING is not set
# CONFIG_DEBUG_VM is not set
# CONFIG_DEBUG_SPINLOCK is not set
# CONFIG_PROVE_RCU is not set
# CONFIG_PROVE_RCU_REPEATEDLY is not set
CONFIG_DEBUG_PER_CPU_MAPS=y
# CONFIG_DEBUG_PER_CPU_MAPS is not set
CONFIG_CPUMASK_OFFSTACK=y
CONFIG_CPU_NOTIFIER_ERROR_INJECT=m
# CONFIG_CPU_NOTIFIER_ERROR_INJECT is not set
CONFIG_FAULT_INJECTION=y
CONFIG_FAILSLAB=y
CONFIG_FAIL_PAGE_ALLOC=y
CONFIG_FAIL_MAKE_REQUEST=y
CONFIG_FAULT_INJECTION_DEBUG_FS=y
CONFIG_FAULT_INJECTION_STACKTRACE_FILTER=y
CONFIG_FAIL_IO_TIMEOUT=y
# CONFIG_FAULT_INJECTION is not set
# CONFIG_FAILSLAB is not set
# CONFIG_FAIL_PAGE_ALLOC is not set
# CONFIG_FAIL_MAKE_REQUEST is not set
# CONFIG_FAULT_INJECTION_DEBUG_FS is not set
# CONFIG_FAULT_INJECTION_STACKTRACE_FILTER is not set
# CONFIG_FAIL_IO_TIMEOUT is not set
CONFIG_SLUB_DEBUG_ON=y
# CONFIG_SLUB_DEBUG_ON is not set
CONFIG_LOCK_STAT=y
# CONFIG_LOCK_STAT is not set
CONFIG_DEBUG_STACK_USAGE=y
# CONFIG_DEBUG_STACK_USAGE is not set
CONFIG_ACPI_DEBUG=y
# CONFIG_ACPI_DEBUG is not set
# CONFIG_ACPI_DEBUG_FUNC_TRACE is not set
CONFIG_DEBUG_SG=y
# CONFIG_DEBUG_SG is not set
# CONFIG_DEBUG_PAGEALLOC is not set
CONFIG_DEBUG_WRITECOUNT=y
CONFIG_DEBUG_OBJECTS=y
# CONFIG_DEBUG_WRITECOUNT is not set
# CONFIG_DEBUG_OBJECTS is not set
# CONFIG_DEBUG_OBJECTS_SELFTEST is not set
CONFIG_DEBUG_OBJECTS_FREE=y
CONFIG_DEBUG_OBJECTS_TIMERS=y
# CONFIG_DEBUG_OBJECTS_FREE is not set
# CONFIG_DEBUG_OBJECTS_TIMERS is not set
CONFIG_DEBUG_OBJECTS_ENABLE_DEFAULT=1
CONFIG_X86_PTDUMP=y
# CONFIG_X86_PTDUMP is not set
CONFIG_CAN_DEBUG_DEVICES=y
# CONFIG_CAN_DEBUG_DEVICES is not set
CONFIG_MODULE_FORCE_UNLOAD=y
# CONFIG_MODULE_FORCE_UNLOAD is not set
CONFIG_SYSCTL_SYSCALL_CHECK=y
# CONFIG_SYSCTL_SYSCALL_CHECK is not set
CONFIG_DEBUG_NOTIFIERS=y
# CONFIG_DEBUG_NOTIFIERS is not set
CONFIG_DMA_API_DEBUG=y
# CONFIG_DMA_API_DEBUG is not set
CONFIG_MMIOTRACE=y
# CONFIG_MMIOTRACE is not set
CONFIG_DEBUG_CREDENTIALS=y
# CONFIG_DEBUG_CREDENTIALS is not set
# off in both production debug and nodebug builds,
# on in rawhide nodebug builds
CONFIG_DEBUG_FORCE_WEAK_PER_CPU=y
# CONFIG_DEBUG_FORCE_WEAK_PER_CPU is not set
CONFIG_EXT4_DEBUG=y
# CONFIG_EXT4_DEBUG is not set
CONFIG_DEBUG_PERF_USE_VMALLOC=y
# CONFIG_DEBUG_PERF_USE_VMALLOC is not set
CONFIG_JBD2_DEBUG=y
# CONFIG_JBD2_DEBUG is not set
CONFIG_DEBUG_CFQ_IOSCHED=y
# CONFIG_DEBUG_CFQ_IOSCHED is not set
CONFIG_DRBD_FAULT_INJECTION=y
# CONFIG_DRBD_FAULT_INJECTION is not set
CONFIG_ATH_DEBUG=y
CONFIG_IWLWIFI_DEVICE_TRACING=y
# CONFIG_ATH_DEBUG is not set
# CONFIG_IWLWIFI_DEVICE_TRACING is not set
CONFIG_DEBUG_OBJECTS_WORK=y
CONFIG_DEBUG_STRICT_USER_COPY_CHECKS=y
# CONFIG_DEBUG_OBJECTS_WORK is not set
# CONFIG_DEBUG_STRICT_USER_COPY_CHECKS is not set
CONFIG_DMADEVICES_DEBUG=y
CONFIG_DMADEVICES_VDEBUG=y
# CONFIG_DMADEVICES_DEBUG is not set
# CONFIG_DMADEVICES_VDEBUG is not set
CONFIG_PM_ADVANCED_DEBUG=y
CONFIG_CEPH_FS_PRETTYDEBUG=y
CONFIG_QUOTA_DEBUG=y
# CONFIG_CEPH_FS_PRETTYDEBUG is not set
# CONFIG_QUOTA_DEBUG is not set
# CONFIG_PCI_DEFAULT_USE_CRS is not set
CONFIG_KGDB_KDB=y
CONFIG_KDB_KEYBOARD=y

View File

@ -200,4 +200,9 @@ CONFIG_SERIAL_GRLIB_GAISLER_APBUART=m
CONFIG_GRETH=m
CONFIG_FB_XVR1000=y
CONFIG_CRYPTO_DEV_NIAGARA2=y
CONFIG_CRYPTO_DEV_NIAGARA2=m
# Bellow is changes made to get the kernel building on sparc again, they need to have upstream fixes
# CONFIG_DEBUG_STRICT_USER_COPY_CHECKS is not set
# CONFIG_ATA_OVER_ETH is not set
# CONFIG_INFINIBAND is not set

View File

@ -37,7 +37,7 @@ CONFIG_M686=y
# CONFIG_MCYRIXIII is not set
# CONFIG_MVIAC3_2 is not set
CONFIG_SMP=y
CONFIG_NR_CPUS=32
CONFIG_NR_CPUS=64
CONFIG_X86_GENERIC=y
# CONFIG_X86_PPRO_FENCE is not set
CONFIG_HPET=y
@ -100,6 +100,16 @@ CONFIG_SECCOMP=y
CONFIG_CAPI_EICON=y
# I2O enabled only for 32-bit x86, disabled for PAE kernel
CONFIG_I2O=m
CONFIG_I2O_BLOCK=m
CONFIG_I2O_SCSI=m
CONFIG_I2O_PROC=m
CONFIG_I2O_CONFIG=y
CONFIG_I2O_EXT_ADAPTEC=y
CONFIG_I2O_CONFIG_OLD_IOCTL=y
CONFIG_I2O_BUS=m
#
# APM (Advanced Power Management) BIOS Support
#
@ -479,6 +489,6 @@ CONFIG_TOSHIBA_BT_RFKILL=m
CONFIG_VGA_SWITCHEROO=y
CONFIG_LPC_SCH=m
CONFIG_INTEL_IDLE=m
CONFIG_INTEL_IDLE=y
CONFIG_PCI_CNB20LE_QUIRK=y

View File

@ -15,7 +15,7 @@ CONFIG_NUMA=y
CONFIG_K8_NUMA=y
CONFIG_X86_64_ACPI_NUMA=y
# CONFIG_NUMA_EMU is not set
CONFIG_NR_CPUS=512
CONFIG_NR_CPUS=256
CONFIG_X86_POWERNOW_K8=m
CONFIG_X86_P4_CLOCKMOD=m
CONFIG_IA32_EMULATION=y
@ -403,7 +403,9 @@ CONFIG_CRYPTO_GHASH_CLMUL_NI_INTEL=m
CONFIG_VGA_SWITCHEROO=y
CONFIG_LPC_SCH=m
CONFIG_INTEL_IDLE=m
CONFIG_INTEL_IDLE=y
CONFIG_I7300_IDLE=m
CONFIG_PCI_CNB20LE_QUIRK=y
CONFIG_HP_ILO=m

View File

@ -0,0 +1,56 @@
From: Greg KH <gregkh@suse.de>
Date: Thu, 5 Aug 2010 20:53:35 +0000 (-0700)
Subject: cgroupfs: create /sys/fs/cgroup to mount cgroupfs on
X-Git-Tag: v2.6.36-rc1~521^2~7
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=676db4af043014e852f67ba0349dae0071bd11f3
cgroupfs: create /sys/fs/cgroup to mount cgroupfs on
We really shouldn't be asking userspace to create new root filesystems.
So follow along with all of the other in-kernel filesystems, and provide
a mount point in sysfs.
For cgroupfs, this should be in /sys/fs/cgroup/ This change provides
that mount point when the cgroup filesystem is registered in the kernel.
Acked-by: Paul Menage <menage@google.com>
Acked-by: Dhaval Giani <dhaval.giani@gmail.com>
Cc: Li Zefan <lizf@cn.fujitsu.com>
Cc: Lennart Poettering <lennart@poettering.net>
Cc: Kay Sievers <kay.sievers@vrfy.org>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
diff --git a/kernel/cgroup.c b/kernel/cgroup.c
index a8ce099..d83cab0 100644
--- a/kernel/cgroup.c
+++ b/kernel/cgroup.c
@@ -1623,6 +1623,8 @@ static struct file_system_type cgroup_fs_type = {
.kill_sb = cgroup_kill_sb,
};
+static struct kobject *cgroup_kobj;
+
static inline struct cgroup *__d_cgrp(struct dentry *dentry)
{
return dentry->d_fsdata;
@@ -3894,9 +3896,18 @@ int __init cgroup_init(void)
hhead = css_set_hash(init_css_set.subsys);
hlist_add_head(&init_css_set.hlist, hhead);
BUG_ON(!init_root_id(&rootnode));
+
+ cgroup_kobj = kobject_create_and_add("cgroup", fs_kobj);
+ if (!cgroup_kobj) {
+ err = -ENOMEM;
+ goto out;
+ }
+
err = register_filesystem(&cgroup_fs_type);
- if (err < 0)
+ if (err < 0) {
+ kobject_put(cgroup_kobj);
goto out;
+ }
proc_create("cgroups", 0, NULL, &proc_cgroupstats_operations);

View File

@ -0,0 +1,78 @@
From 799c10559d60f159ab2232203f222f18fa3c4a5f Mon Sep 17 00:00:00 2001
From: Linus Torvalds <torvalds@linux-foundation.org>
Date: Fri, 15 Oct 2010 11:09:28 -0700
Subject: [PATCH] De-pessimize rds_page_copy_user
Don't try to "optimize" rds_page_copy_user() by using kmap_atomic() and
the unsafe atomic user mode accessor functions. It's actually slower
than the straightforward code on any reasonable modern CPU.
Back when the code was written (although probably not by the time it was
actually merged, though), 32-bit x86 may have been the dominant
architecture. And there kmap_atomic() can be a lot faster than kmap()
(unless you have very good locality, in which case the virtual address
caching by kmap() can overcome all the downsides).
But these days, x86-64 may not be more populous, but it's getting there
(and if you care about performance, it's definitely already there -
you'd have upgraded your CPU's already in the last few years). And on
x86-64, the non-kmap_atomic() version is faster, simply because the code
is simpler and doesn't have the "re-try page fault" case.
People with old hardware are not likely to care about RDS anyway, and
the optimization for the 32-bit case is simply buggy, since it doesn't
verify the user addresses properly.
Reported-by: Dan Rosenberg <drosenberg@vsecurity.com>
Acked-by: Andrew Morton <akpm@linux-foundation.org>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
net/rds/page.c | 27 +++++++--------------------
1 files changed, 7 insertions(+), 20 deletions(-)
diff --git a/net/rds/page.c b/net/rds/page.c
index 595a952..1dfbfea 100644
--- a/net/rds/page.c
+++ b/net/rds/page.c
@@ -57,30 +57,17 @@ int rds_page_copy_user(struct page *page, unsigned long offset,
unsigned long ret;
void *addr;
- if (to_user)
+ addr = kmap(page);
+ if (to_user) {
rds_stats_add(s_copy_to_user, bytes);
- else
+ ret = copy_to_user(ptr, addr + offset, bytes);
+ } else {
rds_stats_add(s_copy_from_user, bytes);
-
- addr = kmap_atomic(page, KM_USER0);
- if (to_user)
- ret = __copy_to_user_inatomic(ptr, addr + offset, bytes);
- else
- ret = __copy_from_user_inatomic(addr + offset, ptr, bytes);
- kunmap_atomic(addr, KM_USER0);
-
- if (ret) {
- addr = kmap(page);
- if (to_user)
- ret = copy_to_user(ptr, addr + offset, bytes);
- else
- ret = copy_from_user(addr + offset, ptr, bytes);
- kunmap(page);
- if (ret)
- return -EFAULT;
+ ret = copy_from_user(addr + offset, ptr, bytes);
}
+ kunmap(page);
- return 0;
+ return ret ? -EFAULT : 0;
}
EXPORT_SYMBOL_GPL(rds_page_copy_user);
--
1.7.3.2

View File

@ -0,0 +1,31 @@
diff --git a/drivers/usb/host/xhci-pci.c b/drivers/usb/host/xhci-pci.c
index 11482b6..b05ff9c 100644
--- a/drivers/usb/host/xhci-pci.c
+++ b/drivers/usb/host/xhci-pci.c
@@ -179,9 +179,25 @@ static struct pci_driver xhci_pci_driver = {
.shutdown = usb_hcd_pci_shutdown,
};
+
+static int enable;
+module_param(enable, int, S_IRUGO);
+MODULE_PARM_DESC(enable, "Enable XHCI host controller");
+
int xhci_register_pci(void)
{
- return pci_register_driver(&xhci_pci_driver);
+ /* xhci will prevent suspend/resume if it's loaded.
+ * force user to pass xhci.enable=1 to the kernel in order
+ * to get usb3.0 support for the time being.
+ *
+ * ugly yes, but there's few enough users out there using
+ * usb3.0, and a lot who just have the hardware breaking
+ * their suspend.
+ */
+ if (enable)
+ return pci_register_driver(&xhci_pci_driver);
+ else
+ return 0;
}
void xhci_unregister_pci(void)

View File

@ -0,0 +1,205 @@
From ce5fa9851090cc5f3de4139bf0f343eb78d1c568 Mon Sep 17 00:00:00 2001
From: Peter Jones <pjones@redhat.com>
Date: Mon, 11 Oct 2010 15:49:28 -0400
Subject: [PATCH] device-mapper: Allow setting of UUID via rename if not already set
This makes it possible to use DM_DEV_RENAME to add a uuid to a device so
long as one has not been previously set either with DM_DEV_CREATE or
with DM_DEV_RENAME. This is needed because sometimes in it's necessary
to create the device before the uuid is known, and in such cases the
uuid must be filled in after the creation.
Also bump the minor number to 19.
Signed-off-by: Peter Jones <pjones@redhat.com>
---
drivers/md/dm-ioctl.c | 95 ++++++++++++++++++++++++++++++++--------------
include/linux/dm-ioctl.h | 11 ++++-
2 files changed, 74 insertions(+), 32 deletions(-)
diff --git a/drivers/md/dm-ioctl.c b/drivers/md/dm-ioctl.c
index bb6bdc8..d102269 100644
--- a/drivers/md/dm-ioctl.c
+++ b/drivers/md/dm-ioctl.c
@@ -298,15 +298,15 @@ retry:
static int dm_hash_rename(uint32_t cookie, uint32_t *flags, const char *old,
const char *new)
{
- char *new_name, *old_name;
+ char *new_data, *old_data;
struct hash_cell *hc;
struct dm_table *table;
/*
* duplicate new.
*/
- new_name = kstrdup(new, GFP_KERNEL);
- if (!new_name)
+ new_data = kstrdup(new, GFP_KERNEL);
+ if (!new_data)
return -ENOMEM;
down_write(&_hash_lock);
@@ -314,13 +314,18 @@ static int dm_hash_rename(uint32_t cookie, uint32_t *flags, const char *old,
/*
* Is new free ?
*/
- hc = __get_name_cell(new);
+ if (*flags & DM_NEW_UUID_FLAG)
+ hc = __get_uuid_cell(new);
+ else
+ hc = __get_name_cell(new);
if (hc) {
- DMWARN("asked to rename to an already existing name %s -> %s",
+ DMWARN("Unable to change %s on device, %s to one that "
+ "already exists: %s",
+ (*flags & DM_NEW_UUID_FLAG) ? "uuid" : "name",
old, new);
dm_put(hc->md);
up_write(&_hash_lock);
- kfree(new_name);
+ kfree(new_data);
return -EBUSY;
}
@@ -329,22 +334,46 @@ static int dm_hash_rename(uint32_t cookie, uint32_t *flags, const char *old,
*/
hc = __get_name_cell(old);
if (!hc) {
- DMWARN("asked to rename a non existent device %s -> %s",
+ DMWARN("Unable to rename non-existent device, %s to %s",
old, new);
up_write(&_hash_lock);
- kfree(new_name);
+ kfree(new_data);
return -ENXIO;
}
- /*
- * rename and move the name cell.
- */
- list_del(&hc->name_list);
- old_name = hc->name;
- mutex_lock(&dm_hash_cells_mutex);
- hc->name = new_name;
- mutex_unlock(&dm_hash_cells_mutex);
- list_add(&hc->name_list, _name_buckets + hash_str(new_name));
+ if (*flags & DM_NEW_UUID_FLAG) {
+ /*
+ * Does this device already have a uuid?
+ */
+ if (hc->uuid) {
+ DMWARN("Unable to change uuid of device, %s because "
+ "uuid is already set to %s",
+ old, hc->uuid);
+ dm_put(hc->md);
+ up_write(&_hash_lock);
+ kfree(new_data);
+ return -EINVAL;
+ }
+ /*
+ * change uuid and move the uuid cell.
+ */
+ list_del(&hc->uuid_list);
+ old_data = hc->uuid;
+ mutex_lock(&dm_hash_cells_mutex);
+ hc->uuid = new_data;
+ mutex_unlock(&dm_hash_cells_mutex);
+ list_add(&hc->uuid_list, _uuid_buckets + hash_str(new_data));
+ } else {
+ /*
+ * rename and move the name cell.
+ */
+ list_del(&hc->name_list);
+ old_data = hc->name;
+ mutex_lock(&dm_hash_cells_mutex);
+ hc->name = new_data;
+ mutex_unlock(&dm_hash_cells_mutex);
+ list_add(&hc->name_list, _name_buckets + hash_str(new_data));
+ }
/*
* Wake up any dm event waiters.
@@ -360,7 +388,7 @@ static int dm_hash_rename(uint32_t cookie, uint32_t *flags, const char *old,
dm_put(hc->md);
up_write(&_hash_lock);
- kfree(old_name);
+ kfree(old_data);
return 0;
}
@@ -773,23 +801,32 @@ static int invalid_str(char *str, void *end)
static int dev_rename(struct dm_ioctl *param, size_t param_size)
{
int r;
- char *new_name = (char *) param + param->data_start;
+ char *new_data = (char *) param + param->data_start;
- if (new_name < param->data ||
- invalid_str(new_name, (void *) param + param_size) ||
- strlen(new_name) > DM_NAME_LEN - 1) {
- DMWARN("Invalid new logical volume name supplied.");
- return -EINVAL;
- }
+ if (param->flags & DM_NEW_UUID_FLAG) {
+ if (new_data < param->data ||
+ invalid_str(new_data, (void *) param + param_size) ||
+ strlen(new_data) > DM_UUID_LEN - 1) {
+ DMWARN("Invalid new device uuid supplied.");
+ return -EINVAL;
+ }
+ } else {
+ if (new_data < param->data ||
+ invalid_str(new_data, (void *) param + param_size) ||
+ strlen(new_data) > DM_NAME_LEN - 1) {
+ DMWARN("Invalid new device name supplied.");
+ return -EINVAL;
+ }
- r = check_name(new_name);
- if (r)
- return r;
+ r = check_name(new_data);
+ if (r)
+ return r;
+ }
param->data_size = 0;
return dm_hash_rename(param->event_nr, &param->flags, param->name,
- new_name);
+ new_data);
}
static int dev_set_geometry(struct dm_ioctl *param, size_t param_size)
diff --git a/include/linux/dm-ioctl.h b/include/linux/dm-ioctl.h
index 2c445e1..3bbcb3a 100644
--- a/include/linux/dm-ioctl.h
+++ b/include/linux/dm-ioctl.h
@@ -266,9 +266,9 @@ enum {
#define DM_DEV_SET_GEOMETRY _IOWR(DM_IOCTL, DM_DEV_SET_GEOMETRY_CMD, struct dm_ioctl)
#define DM_VERSION_MAJOR 4
-#define DM_VERSION_MINOR 17
-#define DM_VERSION_PATCHLEVEL 0
-#define DM_VERSION_EXTRA "-ioctl (2010-03-05)"
+#define DM_VERSION_MINOR 19
+#define DM_VERSION_PATCHLEVEL 1
+#define DM_VERSION_EXTRA "-ioctl (2010-10-12)"
/* Status bits */
#define DM_READONLY_FLAG (1 << 0) /* In/Out */
@@ -321,4 +321,9 @@ enum {
*/
#define DM_UEVENT_GENERATED_FLAG (1 << 13) /* Out */
+/*
+ * If set, rename operates on uuid, not name.
+ */
+#define DM_NEW_UUID_FLAG (1 << 14) /* In */
+
#endif /* _LINUX_DM_IOCTL_H */
--
1.7.2.3

View File

@ -0,0 +1,35 @@
From 8d2f6746f7f82e1aee2dc40a937b5954cfc73414 Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@mcmartin.ca>
Date: Sun, 17 Oct 2010 15:55:32 -0400
Subject: [PATCH] dmar: disable if ricoh multifunction detected
---
drivers/pci/intel-iommu.c | 10 ++++++++++
1 files changed, 10 insertions(+), 0 deletions(-)
diff --git a/drivers/pci/intel-iommu.c b/drivers/pci/intel-iommu.c
index 8e499e8..076c5de 100644
--- a/drivers/pci/intel-iommu.c
+++ b/drivers/pci/intel-iommu.c
@@ -3755,6 +3755,18 @@ static void __devinit quirk_iommu_rwbf(struct pci_dev *dev)
DECLARE_PCI_FIXUP_HEADER(PCI_VENDOR_ID_INTEL, 0x2a40, quirk_iommu_rwbf);
+/* https://bugzilla.redhat.com/show_bug.cgi?id=605888 */
+static void __devinit quirk_ricoh_multifunction(struct pci_dev *dev)
+{
+ printk(KERN_INFO "intel_iommu: broken Ricoh device %04X detected, disabling...\n",
+ dev->device);
+ dmar_disabled = 1;
+}
+DECLARE_PCI_FIXUP_HEADER(0x1180, 0xe822, quirk_ricoh_multifunction);
+DECLARE_PCI_FIXUP_HEADER(0x1180, 0xe230, quirk_ricoh_multifunction);
+DECLARE_PCI_FIXUP_HEADER(0x1180, 0xe832, quirk_ricoh_multifunction);
+DECLARE_PCI_FIXUP_HEADER(0x1180, 0xe476, quirk_ricoh_multifunction);
+
/* On Tylersburg chipsets, some BIOSes have been known to enable the
ISOCH DMAR unit for the Azalia sound device, but not give it any
TLB entries, which causes it to deadlock. Check for that. We do
--
1.7.3.1

54
drm-edid-invalid.patch Normal file
View File

@ -0,0 +1,54 @@
diff -up linux-2.6.35.x86_64/drivers/gpu/drm/drm_edid.c.da linux-2.6.35.x86_64/drivers/gpu/drm/drm_edid.c
--- linux-2.6.35.x86_64/drivers/gpu/drm/drm_edid.c.da 2010-08-01 18:11:14.000000000 -0400
+++ linux-2.6.35.x86_64/drivers/gpu/drm/drm_edid.c 2010-11-11 20:46:10.000000000 -0500
@@ -229,7 +229,7 @@ drm_do_probe_ddc_edid(struct i2c_adapter
.addr = DDC_ADDR,
.flags = I2C_M_RD,
.len = len,
- .buf = buf + start,
+ .buf = buf,
}
};
@@ -242,7 +242,7 @@ drm_do_probe_ddc_edid(struct i2c_adapter
static u8 *
drm_do_get_edid(struct drm_connector *connector, struct i2c_adapter *adapter)
{
- int i, j = 0;
+ int i, j = 0, valid_extensions = 0;
u8 *block, *new;
if ((block = kmalloc(EDID_LENGTH, GFP_KERNEL)) == NULL)
@@ -269,14 +269,28 @@ drm_do_get_edid(struct drm_connector *co
for (j = 1; j <= block[0x7e]; j++) {
for (i = 0; i < 4; i++) {
- if (drm_do_probe_ddc_edid(adapter, block, j,
- EDID_LENGTH))
+ if (drm_do_probe_ddc_edid(adapter,
+ block + (valid_extensions + 1) * EDID_LENGTH,
+ j, EDID_LENGTH))
goto out;
- if (drm_edid_block_valid(block + j * EDID_LENGTH))
+ if (drm_edid_block_valid(block + (valid_extensions + 1) * EDID_LENGTH)) {
+ valid_extensions++;
break;
+ }
}
if (i == 4)
- goto carp;
+ printk(KERN_WARNING
+ "%s: Ignoring invalid EDID block %d.\n",
+ drm_get_connector_name(connector), j);
+ }
+
+ if (valid_extensions != block[0x7e]) {
+ block[EDID_LENGTH-1] += block[0x7e] - valid_extensions;
+ block[0x7e] = valid_extensions;
+ new = krealloc(block, (valid_extensions + 1) * EDID_LENGTH, GFP_KERNEL);
+ if (!new)
+ goto out;
+ block = new;
}
return block;

57
drm-i2c-ch7006-fix.patch Normal file
View File

@ -0,0 +1,57 @@
From 9fa9e790eb301bade8fe4ea0fd9ecb72617f0928 Mon Sep 17 00:00:00 2001
From: Francisco Jerez <currojerez@riseup.net>
Date: Thu, 5 Aug 2010 22:57:08 +0200
Subject: [PATCH 3/5] drm-i2c-ch7006-fix
drm/i2c/ch7006: Don't use POWER_LEVEL_FULL_POWER_OFF on early chip versions.
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
---
drivers/gpu/drm/i2c/ch7006_drv.c | 1 +
drivers/gpu/drm/i2c/ch7006_mode.c | 5 ++++-
drivers/gpu/drm/i2c/ch7006_priv.h | 1 +
3 files changed, 6 insertions(+), 1 deletions(-)
diff --git a/drivers/gpu/drm/i2c/ch7006_drv.c b/drivers/gpu/drm/i2c/ch7006_drv.c
index 81681a0..8c760c7 100644
--- a/drivers/gpu/drm/i2c/ch7006_drv.c
+++ b/drivers/gpu/drm/i2c/ch7006_drv.c
@@ -454,6 +454,7 @@ static int ch7006_encoder_init(struct i2c_client *client,
priv->hmargin = 50;
priv->vmargin = 50;
priv->last_dpms = -1;
+ priv->chip_version = ch7006_read(client, CH7006_VERSION_ID);
if (ch7006_tv_norm) {
for (i = 0; i < NUM_TV_NORMS; i++) {
diff --git a/drivers/gpu/drm/i2c/ch7006_mode.c b/drivers/gpu/drm/i2c/ch7006_mode.c
index e447dfb..c860f24 100644
--- a/drivers/gpu/drm/i2c/ch7006_mode.c
+++ b/drivers/gpu/drm/i2c/ch7006_mode.c
@@ -316,7 +316,10 @@ void ch7006_setup_power_state(struct drm_encoder *encoder)
}
} else {
- *power |= bitfs(CH7006_POWER_LEVEL, FULL_POWER_OFF);
+ if (priv->chip_version >= 0x20)
+ *power |= bitfs(CH7006_POWER_LEVEL, FULL_POWER_OFF);
+ else
+ *power |= bitfs(CH7006_POWER_LEVEL, POWER_OFF);
}
}
diff --git a/drivers/gpu/drm/i2c/ch7006_priv.h b/drivers/gpu/drm/i2c/ch7006_priv.h
index b06d3d9..9487123 100644
--- a/drivers/gpu/drm/i2c/ch7006_priv.h
+++ b/drivers/gpu/drm/i2c/ch7006_priv.h
@@ -95,6 +95,7 @@ struct ch7006_priv {
int flicker;
int scale;
+ int chip_version;
int last_dpms;
};
--
1.7.2

View File

@ -0,0 +1,50 @@
https://bugzilla.redhat.com/show_bug.cgi?id=639146
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index ce8ff0e..3bbd2c4 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1353,6 +1353,27 @@ static bool i915_switcheroo_can_switch(struct pci_dev *pdev)
return can_switch;
}
+/* check the VBT to see whether the eDP is on DP-D port */
+static bool intel_dpd_is_edp(struct drm_device *dev)
+{
+ struct drm_i915_private *dev_priv = dev->dev_private;
+ struct child_device_config *p_child;
+ int i;
+
+ if (!dev_priv->child_dev_num)
+ return false;
+
+ for (i = 0; i < dev_priv->child_dev_num; i++) {
+ p_child = dev_priv->child_dev + i;
+
+ if (p_child->dvo_port == PORT_IDPD &&
+ p_child->device_type == DEVICE_TYPE_eDP)
+ return true;
+ }
+
+ return false;
+}
+
static int i915_load_modeset_init(struct drm_device *dev,
unsigned long prealloc_start,
unsigned long prealloc_size,
@@ -1409,6 +1430,15 @@ static int i915_load_modeset_init(struct drm_device *dev,
if (ret)
DRM_INFO("failed to find VBIOS tables\n");
+ /* XXX: eDP doesn't even work in git HEAD,
+ * bail out and pray that text mode still works...
+ */
+ if (intel_dpd_is_edp(dev)) {
+ DRM_ERROR("eDP support is currently non-functional, please boot with \"nomodeset xdriver=vesa\"\n");
+ ret = -EINVAL;
+ goto cleanup_ringbuffer;
+ }
+
/* if we have > 1 VGA cards, then disable the radeon VGA resources */
ret = vga_client_register(dev->pdev, dev, NULL, i915_vga_set_decode);
if (ret)

View File

@ -0,0 +1,70 @@
From ab7959dd389be36c0bc63e3e883b7891d2c1bfc4 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Wed, 8 Sep 2010 09:45:11 +0100
Subject: [PATCH] drm/i915: Disable output polling across suspend & resume
Suspending (especially hibernating) may take a finite amount of time,
during which a hotplug event may trigger and we will attempt to handle
it with inconsistent state. Disable hotplug polling around suspend and
resume.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=30070
Reported-by: Rui Tiago Matos <tiagomatos@gmail.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
---
drivers/gpu/drm/i915/i915_dma.c | 2 --
drivers/gpu/drm/i915/i915_drv.c | 11 ++++++++++-
2 files changed, 10 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index ce8ff0e..c569617 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1334,10 +1334,8 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev, enum vga_switcheroo_
/* i915 resume handler doesn't set to D0 */
pci_set_power_state(dev->pdev, PCI_D0);
i915_resume(dev);
- drm_kms_helper_poll_enable(dev);
} else {
printk(KERN_ERR "i915: switched off\n");
- drm_kms_helper_poll_disable(dev);
i915_suspend(dev, pmm);
}
}
diff --git a/drivers/gpu/drm/i915/i915_drv.c b/drivers/gpu/drm/i915/i915_drv.c
index 194e0c4..3ad3ebe 100644
--- a/drivers/gpu/drm/i915/i915_drv.c
+++ b/drivers/gpu/drm/i915/i915_drv.c
@@ -263,6 +263,8 @@ int i915_suspend(struct drm_device *dev, pm_message_t state)
if (state.event == PM_EVENT_PRETHAW)
return 0;
+ drm_kms_helper_poll_disable(dev);
+
error = i915_drm_freeze(dev);
if (error)
return error;
@@ -306,12 +308,19 @@ static int i915_drm_thaw(struct drm_device *dev)
int i915_resume(struct drm_device *dev)
{
+ int ret;
+
if (pci_enable_device(dev->pdev))
return -EIO;
pci_set_master(dev->pdev);
- return i915_drm_thaw(dev);
+ ret = i915_drm_thaw(dev);
+ if (ret)
+ return ret;
+
+ drm_kms_helper_poll_enable(dev);
+ return 0;
}
/**
--
1.7.2.3

View File

@ -0,0 +1,89 @@
From ce9d419dbecc292cc3e06e8b1d6d123d3fa813a4 Mon Sep 17 00:00:00 2001
From: Chris Wilson <chris@chris-wilson.co.uk>
Date: Sun, 26 Sep 2010 20:50:05 +0100
Subject: drm/i915: Sanity check pread/pwrite
From: Chris Wilson <chris@chris-wilson.co.uk>
commit ce9d419dbecc292cc3e06e8b1d6d123d3fa813a4 upstream.
Move the access control up from the fast paths, which are no longer
universally taken first, up into the caller. This then duplicates some
sanity checking along the slow paths, but is much simpler.
Tracked as CVE-2010-2962.
Reported-by: Kees Cook <kees@ubuntu.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
drivers/gpu/drm/i915/i915_gem.c | 28 ++++++++++++++++++++--------
1 file changed, 20 insertions(+), 8 deletions(-)
--- a/drivers/gpu/drm/i915/i915_gem.c
+++ b/drivers/gpu/drm/i915/i915_gem.c
@@ -465,8 +465,15 @@ i915_gem_pread_ioctl(struct drm_device *
*/
if (args->offset > obj->size || args->size > obj->size ||
args->offset + args->size > obj->size) {
- drm_gem_object_unreference_unlocked(obj);
- return -EINVAL;
+ ret = -EINVAL;
+ goto err;
+ }
+
+ if (!access_ok(VERIFY_WRITE,
+ (char __user *)(uintptr_t)args->data_ptr,
+ args->size)) {
+ ret = -EFAULT;
+ goto err;
}
if (i915_gem_object_needs_bit17_swizzle(obj)) {
@@ -478,8 +485,8 @@ i915_gem_pread_ioctl(struct drm_device *
file_priv);
}
+err:
drm_gem_object_unreference_unlocked(obj);
-
return ret;
}
@@ -568,8 +575,6 @@ i915_gem_gtt_pwrite_fast(struct drm_devi
user_data = (char __user *) (uintptr_t) args->data_ptr;
remain = args->size;
- if (!access_ok(VERIFY_READ, user_data, remain))
- return -EFAULT;
mutex_lock(&dev->struct_mutex);
@@ -928,8 +933,15 @@ i915_gem_pwrite_ioctl(struct drm_device
*/
if (args->offset > obj->size || args->size > obj->size ||
args->offset + args->size > obj->size) {
- drm_gem_object_unreference_unlocked(obj);
- return -EINVAL;
+ ret = -EINVAL;
+ goto err;
+ }
+
+ if (!access_ok(VERIFY_READ,
+ (char __user *)(uintptr_t)args->data_ptr,
+ args->size)) {
+ ret = -EFAULT;
+ goto err;
}
/* We can only do the GTT pwrite on untiled buffers, as otherwise
@@ -963,8 +975,8 @@ i915_gem_pwrite_ioctl(struct drm_device
DRM_INFO("pwrite failed %d\n", ret);
#endif
+err:
drm_gem_object_unreference_unlocked(obj);
-
return ret;
}

View File

@ -0,0 +1,21 @@
diff -up linux-2.6.35.x86_64/drivers/gpu/drm/nouveau/nouveau_connector.c.da linux-2.6.35.x86_64/drivers/gpu/drm/nouveau/nouveau_connector.c
--- linux-2.6.35.x86_64/drivers/gpu/drm/nouveau/nouveau_connector.c.da 2010-11-08 19:55:42.000000000 -0500
+++ linux-2.6.35.x86_64/drivers/gpu/drm/nouveau/nouveau_connector.c 2010-11-08 19:55:49.000000000 -0500
@@ -298,7 +298,7 @@ detect_analog:
}
static enum drm_connector_status
-nouveau_connector_detect_lvds(struct drm_connector *connector)
+nouveau_connector_detect_lvds(struct drm_connector *connector, bool force)
{
struct drm_device *dev = connector->dev;
struct drm_nouveau_private *dev_priv = dev->dev_private;
@@ -319,7 +319,7 @@ nouveau_connector_detect_lvds(struct drm
/* Try retrieving EDID via DDC */
if (!dev_priv->vbios.fp_no_ddc) {
- status = nouveau_connector_detect(connector);
+ status = nouveau_connector_detect(connector, force);
if (status == connector_status_connected)
goto out;
}

View File

@ -0,0 +1,32 @@
From d0301ece9e093c484f880893dc86d97848360892 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Fri, 19 Nov 2010 18:50:57 +1000
Subject: [PATCH 2/2] drm-nouveau-evo-hang
On some GF8+ boards, the display engine will stop processing its push
buffer if a wrap-around occurs at a certain point. The exact cause
is not known.
This patch by David Dillow (rhbz#537065) is a safe enough work-around
until it can be solved properly.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
---
drivers/gpu/drm/nouveau/nv50_display.c | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index 11d366a..4e5402c 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -364,6 +364,7 @@ nv50_display_init(struct drm_device *dev)
nv_wr32(dev, 0x610300, nv_rd32(dev, 0x610300) & ~1);
evo->dma.max = (4096/4) - 2;
+ evo->dma.max &= ~7;
evo->dma.put = 0;
evo->dma.cur = evo->dma.put;
evo->dma.free = evo->dma.max - evo->dma.cur;
--
1.7.3.2

167
drm-nouveau-imac-g4.patch Normal file
View File

@ -0,0 +1,167 @@
From b4166db6d1f951a460e5e7f52bb4b4d96f27f55a Mon Sep 17 00:00:00 2001
From: Francisco Jerez <currojerez@riseup.net>
Date: Fri, 19 Nov 2010 18:08:47 +1000
Subject: [PATCH 1/2] drm-nouveau-imac-g4
drm/nouveau: fabricate DCB encoder table for iMac G4
In typical Apple fashion there's no standard information about what
encoders are present on this machine, this patch adds a quirk to
provide it.
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
---
drivers/gpu/drm/nouveau/nouveau_bios.c | 102 ++++++++++++--------------------
1 files changed, 38 insertions(+), 64 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nouveau_bios.c b/drivers/gpu/drm/nouveau/nouveau_bios.c
index 72905c9..8c287f8 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bios.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bios.c
@@ -5985,52 +5985,17 @@ static struct dcb_entry *new_dcb_entry(struct dcb_table *dcb)
return entry;
}
-static void fabricate_vga_output(struct dcb_table *dcb, int i2c, int heads)
+static void fabricate_dcb_output(struct dcb_table *dcb, int type, int i2c,
+ int heads, int or)
{
struct dcb_entry *entry = new_dcb_entry(dcb);
- entry->type = 0;
+ entry->type = type;
entry->i2c_index = i2c;
entry->heads = heads;
- entry->location = DCB_LOC_ON_CHIP;
- entry->or = 1;
-}
-
-static void fabricate_dvi_i_output(struct dcb_table *dcb, bool twoHeads)
-{
- struct dcb_entry *entry = new_dcb_entry(dcb);
-
- entry->type = 2;
- entry->i2c_index = LEGACY_I2C_PANEL;
- entry->heads = twoHeads ? 3 : 1;
- entry->location = !DCB_LOC_ON_CHIP; /* ie OFF CHIP */
- entry->or = 1; /* means |0x10 gets set on CRE_LCD__INDEX */
- entry->duallink_possible = false; /* SiI164 and co. are single link */
-
-#if 0
- /*
- * For dvi-a either crtc probably works, but my card appears to only
- * support dvi-d. "nvidia" still attempts to program it for dvi-a,
- * doing the full fp output setup (program 0x6808.. fp dimension regs,
- * setting 0x680848 to 0x10000111 to enable, maybe setting 0x680880);
- * the monitor picks up the mode res ok and lights up, but no pixel
- * data appears, so the board manufacturer probably connected up the
- * sync lines, but missed the video traces / components
- *
- * with this introduction, dvi-a left as an exercise for the reader.
- */
- fabricate_vga_output(dcb, LEGACY_I2C_PANEL, entry->heads);
-#endif
-}
-
-static void fabricate_tv_output(struct dcb_table *dcb, bool twoHeads)
-{
- struct dcb_entry *entry = new_dcb_entry(dcb);
-
- entry->type = 1;
- entry->i2c_index = LEGACY_I2C_TV;
- entry->heads = twoHeads ? 3 : 1;
- entry->location = !DCB_LOC_ON_CHIP; /* ie OFF CHIP */
+ if (type != OUTPUT_ANALOG)
+ entry->location = !DCB_LOC_ON_CHIP; /* ie OFF CHIP */
+ entry->or = or;
}
static bool
@@ -6297,8 +6262,36 @@ apply_dcb_encoder_quirks(struct drm_device *dev, int idx, u32 *conn, u32 *conf)
return true;
}
+static void
+fabricate_dcb_encoder_table(struct drm_device *dev, struct nvbios *bios)
+{
+ struct dcb_table *dcb = &bios->dcb;
+ int all_heads = (nv_two_heads(dev) ? 3 : 1);
+
+#ifdef __powerpc__
+ /* Apple iMac G4 NV17 */
+ if (of_machine_is_compatible("PowerMac4,5")) {
+ fabricate_dcb_output(dcb, OUTPUT_TMDS, 0, all_heads, 1);
+ fabricate_dcb_output(dcb, OUTPUT_ANALOG, 1, all_heads, 2);
+ return;
+ }
+#endif
+
+ /* Make up some sane defaults */
+ fabricate_dcb_output(dcb, OUTPUT_ANALOG, LEGACY_I2C_CRT, 1, 1);
+
+ if (nv04_tv_identify(dev, bios->legacy.i2c_indices.tv) >= 0)
+ fabricate_dcb_output(dcb, OUTPUT_TV, LEGACY_I2C_TV,
+ all_heads, 0);
+
+ else if (bios->tmds.output0_script_ptr ||
+ bios->tmds.output1_script_ptr)
+ fabricate_dcb_output(dcb, OUTPUT_TMDS, LEGACY_I2C_PANEL,
+ all_heads, 1);
+}
+
static int
-parse_dcb_table(struct drm_device *dev, struct nvbios *bios, bool twoHeads)
+parse_dcb_table(struct drm_device *dev, struct nvbios *bios)
{
struct drm_nouveau_private *dev_priv = dev->dev_private;
struct dcb_table *dcb = &bios->dcb;
@@ -6318,12 +6311,7 @@ parse_dcb_table(struct drm_device *dev, struct nvbios *bios, bool twoHeads)
/* this situation likely means a really old card, pre DCB */
if (dcbptr == 0x0) {
- NV_INFO(dev, "Assuming a CRT output exists\n");
- fabricate_vga_output(dcb, LEGACY_I2C_CRT, 1);
-
- if (nv04_tv_identify(dev, bios->legacy.i2c_indices.tv) >= 0)
- fabricate_tv_output(dcb, twoHeads);
-
+ fabricate_dcb_encoder_table(dev, bios);
return 0;
}
@@ -6383,21 +6371,7 @@ parse_dcb_table(struct drm_device *dev, struct nvbios *bios, bool twoHeads)
*/
NV_TRACEWARN(dev, "No useful information in BIOS output table; "
"adding all possible outputs\n");
- fabricate_vga_output(dcb, LEGACY_I2C_CRT, 1);
-
- /*
- * Attempt to detect TV before DVI because the test
- * for the former is more accurate and it rules the
- * latter out.
- */
- if (nv04_tv_identify(dev,
- bios->legacy.i2c_indices.tv) >= 0)
- fabricate_tv_output(dcb, twoHeads);
-
- else if (bios->tmds.output0_script_ptr ||
- bios->tmds.output1_script_ptr)
- fabricate_dvi_i_output(dcb, twoHeads);
-
+ fabricate_dcb_encoder_table(dev, bios);
return 0;
}
@@ -6787,7 +6761,7 @@ nouveau_bios_init(struct drm_device *dev)
if (ret)
return ret;
- ret = parse_dcb_table(dev, bios, nv_two_heads(dev));
+ ret = parse_dcb_table(dev, bios);
if (ret)
return ret;
--
1.7.3.2

View File

@ -0,0 +1,33 @@
From 8f51b07995c55369bfd27921ef10b76c2b203fe8 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Tue, 17 Aug 2010 14:20:29 +1000
Subject: [PATCH] drm-nouveau-nv50-crtc-update-delay
rhbz#614552
Adds a short delay before doing framebuffer-only CRTC updates. Fixes
an issue that effects some cards (Quadro NVS295/FX580) where two of
these updates in quick succession hangs the display engine.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
---
drivers/gpu/drm/nouveau/nv50_crtc.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nv50_crtc.c b/drivers/gpu/drm/nouveau/nv50_crtc.c
index 2423c92..5c2aa1e 100644
--- a/drivers/gpu/drm/nouveau/nv50_crtc.c
+++ b/drivers/gpu/drm/nouveau/nv50_crtc.c
@@ -540,6 +540,9 @@ nv50_crtc_do_mode_set_base(struct drm_crtc *crtc, int x, int y,
nouveau_bo_unpin(ofb->nvbo);
}
+ if (update)
+ mdelay(1);
+
nv_crtc->fb.offset = fb->nvbo->bo.offset - dev_priv->vm_vram_base;
nv_crtc->fb.tile_flags = fb->nvbo->tile_flags;
nv_crtc->fb.cpp = drm_fb->bits_per_pixel / 8;
--
1.7.3.1

231
drm-nouveau-nv86-bug.patch Normal file
View File

@ -0,0 +1,231 @@
From e7e3837f6395e44b5b5fb7cdbcccaba5baf76803 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Fri, 22 Oct 2010 10:26:24 +1000
Subject: [PATCH] drm-nouveau-nv86-bug
drm/nv50: implement possible workaround for NV86 PGRAPH TLB flush hang
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
---
drivers/gpu/drm/nouveau/nouveau_drv.h | 5 +++
drivers/gpu/drm/nouveau/nouveau_mem.c | 14 +++-----
drivers/gpu/drm/nouveau/nouveau_sgdma.c | 8 ++--
drivers/gpu/drm/nouveau/nouveau_state.c | 10 ++++++
drivers/gpu/drm/nouveau/nv50_fifo.c | 5 +++
drivers/gpu/drm/nouveau/nv50_graph.c | 52 +++++++++++++++++++++++++++++++
drivers/gpu/drm/nouveau/nv50_instmem.c | 1 -
7 files changed, 82 insertions(+), 13 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index be53e92..4273390 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -304,6 +304,7 @@ struct nouveau_fifo_engine {
void (*destroy_context)(struct nouveau_channel *);
int (*load_context)(struct nouveau_channel *);
int (*unload_context)(struct drm_device *);
+ void (*tlb_flush)(struct drm_device *dev);
};
struct nouveau_pgraph_object_method {
@@ -336,6 +337,7 @@ struct nouveau_pgraph_engine {
void (*destroy_context)(struct nouveau_channel *);
int (*load_context)(struct nouveau_channel *);
int (*unload_context)(struct drm_device *);
+ void (*tlb_flush)(struct drm_device *dev);
void (*set_region_tiling)(struct drm_device *dev, int i, uint32_t addr,
uint32_t size, uint32_t pitch);
@@ -944,6 +946,7 @@ extern int nv50_fifo_create_context(struct nouveau_channel *);
extern void nv50_fifo_destroy_context(struct nouveau_channel *);
extern int nv50_fifo_load_context(struct nouveau_channel *);
extern int nv50_fifo_unload_context(struct drm_device *);
+extern void nv50_fifo_tlb_flush(struct drm_device *dev);
/* nvc0_fifo.c */
extern int nvc0_fifo_init(struct drm_device *);
@@ -1021,6 +1024,8 @@ extern int nv50_graph_load_context(struct nouveau_channel *);
extern int nv50_graph_unload_context(struct drm_device *);
extern void nv50_graph_context_switch(struct drm_device *);
extern int nv50_grctx_init(struct nouveau_grctx *);
+extern void nv50_graph_tlb_flush(struct drm_device *dev);
+extern void nv86_graph_tlb_flush(struct drm_device *dev);
/* nvc0_graph.c */
extern int nvc0_graph_init(struct drm_device *);
diff --git a/drivers/gpu/drm/nouveau/nouveau_mem.c b/drivers/gpu/drm/nouveau/nouveau_mem.c
index 4f0ae39..514ad9f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_mem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_mem.c
@@ -173,11 +173,10 @@ nv50_mem_vm_bind_linear(struct drm_device *dev, uint64_t virt, uint32_t size,
}
}
}
- dev_priv->engine.instmem.flush(dev);
- nv50_vm_flush(dev, 5);
- nv50_vm_flush(dev, 0);
- nv50_vm_flush(dev, 4);
+ dev_priv->engine.instmem.flush(dev);
+ dev_priv->engine.fifo.tlb_flush(dev);
+ dev_priv->engine.graph.tlb_flush(dev);
nv50_vm_flush(dev, 6);
return 0;
}
@@ -207,11 +206,10 @@ nv50_mem_vm_unbind(struct drm_device *dev, uint64_t virt, uint32_t size)
pte++;
}
}
- dev_priv->engine.instmem.flush(dev);
- nv50_vm_flush(dev, 5);
- nv50_vm_flush(dev, 0);
- nv50_vm_flush(dev, 4);
+ dev_priv->engine.instmem.flush(dev);
+ dev_priv->engine.fifo.tlb_flush(dev);
+ dev_priv->engine.graph.tlb_flush(dev);
nv50_vm_flush(dev, 6);
}
diff --git a/drivers/gpu/drm/nouveau/nouveau_sgdma.c b/drivers/gpu/drm/nouveau/nouveau_sgdma.c
index 7f028fe..d55a583 100644
--- a/drivers/gpu/drm/nouveau/nouveau_sgdma.c
+++ b/drivers/gpu/drm/nouveau/nouveau_sgdma.c
@@ -120,8 +120,8 @@ nouveau_sgdma_bind(struct ttm_backend *be, struct ttm_mem_reg *mem)
dev_priv->engine.instmem.flush(nvbe->dev);
if (dev_priv->card_type == NV_50) {
- nv50_vm_flush(dev, 5); /* PGRAPH */
- nv50_vm_flush(dev, 0); /* PFIFO */
+ dev_priv->engine.fifo.tlb_flush(dev);
+ dev_priv->engine.graph.tlb_flush(dev);
}
nvbe->bound = true;
@@ -162,8 +162,8 @@ nouveau_sgdma_unbind(struct ttm_backend *be)
dev_priv->engine.instmem.flush(nvbe->dev);
if (dev_priv->card_type == NV_50) {
- nv50_vm_flush(dev, 5);
- nv50_vm_flush(dev, 0);
+ dev_priv->engine.fifo.tlb_flush(dev);
+ dev_priv->engine.graph.tlb_flush(dev);
}
nvbe->bound = false;
diff --git a/drivers/gpu/drm/nouveau/nouveau_state.c b/drivers/gpu/drm/nouveau/nouveau_state.c
index be85960..47c353c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_state.c
+++ b/drivers/gpu/drm/nouveau/nouveau_state.c
@@ -333,6 +333,15 @@ static int nouveau_init_engine_ptrs(struct drm_device *dev)
engine->graph.destroy_context = nv50_graph_destroy_context;
engine->graph.load_context = nv50_graph_load_context;
engine->graph.unload_context = nv50_graph_unload_context;
+ if (dev_priv->chipset != 0x86)
+ engine->graph.tlb_flush = nv50_graph_tlb_flush;
+ else {
+ /* from what i can see nvidia do this on every
+ * pre-NVA3 board except NVAC, but, we've only
+ * ever seen problems on NV86
+ */
+ engine->graph.tlb_flush = nv86_graph_tlb_flush;
+ }
engine->fifo.channels = 128;
engine->fifo.init = nv50_fifo_init;
engine->fifo.takedown = nv50_fifo_takedown;
@@ -344,6 +353,7 @@ static int nouveau_init_engine_ptrs(struct drm_device *dev)
engine->fifo.destroy_context = nv50_fifo_destroy_context;
engine->fifo.load_context = nv50_fifo_load_context;
engine->fifo.unload_context = nv50_fifo_unload_context;
+ engine->fifo.tlb_flush = nv50_fifo_tlb_flush;
engine->display.early_init = nv50_display_early_init;
engine->display.late_takedown = nv50_display_late_takedown;
engine->display.create = nv50_display_create;
diff --git a/drivers/gpu/drm/nouveau/nv50_fifo.c b/drivers/gpu/drm/nouveau/nv50_fifo.c
index a46a961..1da65bd 100644
--- a/drivers/gpu/drm/nouveau/nv50_fifo.c
+++ b/drivers/gpu/drm/nouveau/nv50_fifo.c
@@ -464,3 +464,8 @@ nv50_fifo_unload_context(struct drm_device *dev)
return 0;
}
+void
+nv50_fifo_tlb_flush(struct drm_device *dev)
+{
+ nv50_vm_flush(dev, 5);
+}
diff --git a/drivers/gpu/drm/nouveau/nv50_graph.c b/drivers/gpu/drm/nouveau/nv50_graph.c
index cbf5ae2..8b669d0 100644
--- a/drivers/gpu/drm/nouveau/nv50_graph.c
+++ b/drivers/gpu/drm/nouveau/nv50_graph.c
@@ -402,3 +402,55 @@ struct nouveau_pgraph_object_class nv50_graph_grclass[] = {
{ 0x8597, false, NULL }, /* tesla (nva3, nva5, nva8) */
{}
};
+
+void
+nv50_graph_tlb_flush(struct drm_device *dev)
+{
+ nv50_vm_flush(dev, 0);
+}
+
+void
+nv86_graph_tlb_flush(struct drm_device *dev)
+{
+ struct drm_nouveau_private *dev_priv = dev->dev_private;
+ struct nouveau_timer_engine *ptimer = &dev_priv->engine.timer;
+ bool idle, timeout = false;
+ unsigned long flags;
+ u64 start;
+ u32 tmp;
+
+ spin_lock_irqsave(&dev_priv->context_switch_lock, flags);
+ nv_mask(dev, 0x400500, 0x00000001, 0x00000000);
+
+ start = ptimer->read(dev);
+ do {
+ idle = true;
+
+ for (tmp = nv_rd32(dev, 0x400380); tmp && idle; tmp >>= 3) {
+ if ((tmp & 7) == 1)
+ idle = false;
+ }
+
+ for (tmp = nv_rd32(dev, 0x400384); tmp && idle; tmp >>= 3) {
+ if ((tmp & 7) == 1)
+ idle = false;
+ }
+
+ for (tmp = nv_rd32(dev, 0x400388); tmp && idle; tmp >>= 3) {
+ if ((tmp & 7) == 1)
+ idle = false;
+ }
+ } while (!idle && !(timeout = ptimer->read(dev) - start > 2000000000));
+
+ if (timeout) {
+ NV_ERROR(dev, "PGRAPH TLB flush idle timeout fail: "
+ "0x%08x 0x%08x 0x%08x 0x%08x\n",
+ nv_rd32(dev, 0x400700), nv_rd32(dev, 0x400380),
+ nv_rd32(dev, 0x400384), nv_rd32(dev, 0x400388));
+ }
+
+ nv50_vm_flush(dev, 0);
+
+ nv_mask(dev, 0x400500, 0x00000001, 0x00000001);
+ spin_unlock_irqrestore(&dev_priv->context_switch_lock, flags);
+}
diff --git a/drivers/gpu/drm/nouveau/nv50_instmem.c b/drivers/gpu/drm/nouveau/nv50_instmem.c
index ac3de05..c836ddc 100644
--- a/drivers/gpu/drm/nouveau/nv50_instmem.c
+++ b/drivers/gpu/drm/nouveau/nv50_instmem.c
@@ -402,7 +402,6 @@ nv50_instmem_bind(struct drm_device *dev, struct nouveau_gpuobj *gpuobj)
}
dev_priv->engine.instmem.flush(dev);
- nv50_vm_flush(dev, 4);
nv50_vm_flush(dev, 6);
gpuobj->im_bound = 1;
--
1.7.3.1

View File

@ -0,0 +1,106 @@
From 07a51882863d9e45b0715dcffbb66491adf2fb4e Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Wed, 30 Jun 2010 13:34:05 +1000
Subject: [PATCH] drm/nouveau: disable acceleration on NVA3/NVA5/NVA8 by default
There's an GPU lockup problem for which the cause is currently unknown
on these chipsets.
Until it's resolved, it's better to leave the user with a working system
without acceleration than to have random lockups.
With this patch, acceleration will be off by default if a known problem
chipset is detected, but can be re-enabled with nouveau.noaccel=0 on
the kernel commandline.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
---
drivers/gpu/drm/nouveau/nouveau_drv.c | 2 +-
drivers/gpu/drm/nouveau/nouveau_drv.h | 1 +
drivers/gpu/drm/nouveau/nouveau_state.c | 23 +++++++++++++++++++----
3 files changed, 21 insertions(+), 5 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.c b/drivers/gpu/drm/nouveau/nouveau_drv.c
index 946748a..9b69328 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.c
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.c
@@ -72,7 +72,7 @@ int nouveau_ignorelid = 0;
module_param_named(ignorelid, nouveau_ignorelid, int, 0400);
MODULE_PARM_DESC(noaccel, "Disable all acceleration");
-int nouveau_noaccel = 0;
+int nouveau_noaccel = -1;
module_param_named(noaccel, nouveau_noaccel, int, 0400);
MODULE_PARM_DESC(nofbaccel, "Disable fbcon acceleration");
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index 24b3d03..0cf1bee 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -504,6 +504,7 @@ enum nouveau_card_type {
struct drm_nouveau_private {
struct drm_device *dev;
+ bool noaccel;
/* the card type, takes NV_* as values */
enum nouveau_card_type card_type;
diff --git a/drivers/gpu/drm/nouveau/nouveau_state.c b/drivers/gpu/drm/nouveau/nouveau_state.c
index be85960..896f6ae 100644
--- a/drivers/gpu/drm/nouveau/nouveau_state.c
+++ b/drivers/gpu/drm/nouveau/nouveau_state.c
@@ -563,7 +563,7 @@ nouveau_card_init(struct drm_device *dev)
if (ret)
goto out_timer;
- if (nouveau_noaccel)
+ if (dev_priv->noaccel)
engine->graph.accel_blocked = true;
else {
/* PGRAPH */
@@ -613,10 +613,10 @@ out_irq:
out_display:
engine->display.destroy(dev);
out_fifo:
- if (!nouveau_noaccel)
+ if (!dev_priv->noaccel)
engine->fifo.takedown(dev);
out_graph:
- if (!nouveau_noaccel)
+ if (!dev_priv->noaccel)
engine->graph.takedown(dev);
out_fb:
engine->fb.takedown(dev);
@@ -655,7 +655,7 @@ static void nouveau_card_takedown(struct drm_device *dev)
dev_priv->channel = NULL;
}
- if (!nouveau_noaccel) {
+ if (!dev_priv->noaccel) {
engine->fifo.takedown(dev);
engine->graph.takedown(dev);
}
@@ -861,6 +861,21 @@ int nouveau_load(struct drm_device *dev, unsigned long flags)
if (ret)
goto err_mmio;
+ if (nouveau_noaccel == -1) {
+ switch (dev_priv->chipset) {
+ case 0xa3:
+ case 0xa5:
+ case 0xa8:
+ dev_priv->noaccel = true;
+ break;
+ default:
+ dev_priv->noaccel = false;
+ break;
+ }
+ } else {
+ dev_priv->noaccel = (nouveau_noaccel != 0);
+ }
+
/* Map PRAMIN BAR, or on older cards, the aperture withing BAR0 */
if (dev_priv->card_type >= NV_40) {
int ramin_bar = 2;
--
1.7.2.2

View File

@ -0,0 +1,25 @@
From 7abba51e3fc9e3fdd43c63eeb1a680a2e258a833 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Fri, 19 Nov 2010 18:59:15 +1000
Subject: [PATCH] drm-nouveau-nvaf-grclass
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
---
drivers/gpu/drm/nouveau/nv50_graph.c | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nv50_graph.c b/drivers/gpu/drm/nouveau/nv50_graph.c
index 8b669d0..235be5f 100644
--- a/drivers/gpu/drm/nouveau/nv50_graph.c
+++ b/drivers/gpu/drm/nouveau/nv50_graph.c
@@ -400,6 +400,7 @@ struct nouveau_pgraph_object_class nv50_graph_grclass[] = {
{ 0x8297, false, NULL }, /* tesla (nv8x/nv9x) */
{ 0x8397, false, NULL }, /* tesla (nva0, nvaa, nvac) */
{ 0x8597, false, NULL }, /* tesla (nva3, nva5, nva8) */
+ { 0x8697, false, NULL }, /* tesla (nvaf) */
{}
};
--
1.7.3.2

141
drm-nouveau-race-fix.patch Normal file
View File

@ -0,0 +1,141 @@
From 4733f633c4bfb0672d5bd88a8d19a03e27a3c1d0 Mon Sep 17 00:00:00 2001
From: Ben Skeggs <bskeggs@redhat.com>
Date: Fri, 23 Jul 2010 09:06:52 +1000
Subject: [PATCH 2/2] drm-nouveau-race-fix
drm/nouveau: fix race condition when under memory pressure
rhbz#602663
When VRAM is running out it's possible that the client's push buffers get
evicted to main memory. When they're validated back in, the GPU may
be used for the copy back to VRAM, but the existing synchronisation code
only deals with inter-channel sync, not sync between PFIFO and PGRAPH on
the same channel. This leads to PFIFO fetching from command buffers that
haven't quite been copied by PGRAPH yet.
This patch marks push buffers as so, and forces any GPU-assisted buffer
moves to be done on a different channel, which triggers the correct
synchronisation to happen before we submit them.
After discussion with another nouveau developer, it was agreed that while
this patch is fine in itself, that we'd prefer to work out a nicer, but
likely much more invasive, fix upstream.
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
---
drivers/gpu/drm/nouveau/nouveau_bo.c | 15 +++++++++++++
drivers/gpu/drm/nouveau/nouveau_drv.h | 1 +
drivers/gpu/drm/nouveau/nouveau_gem.c | 36 +++++++++++++++++++++++---------
3 files changed, 42 insertions(+), 10 deletions(-)
diff --git a/drivers/gpu/drm/nouveau/nouveau_bo.c b/drivers/gpu/drm/nouveau/nouveau_bo.c
index 553a01d..5e62d1b 100644
--- a/drivers/gpu/drm/nouveau/nouveau_bo.c
+++ b/drivers/gpu/drm/nouveau/nouveau_bo.c
@@ -36,6 +36,21 @@
#include <linux/log2.h>
#include <linux/slab.h>
+int
+nouveau_bo_sync_gpu(struct nouveau_bo *nvbo, struct nouveau_channel *chan)
+{
+ struct nouveau_fence *prev_fence = nvbo->bo.sync_obj;
+ int ret;
+
+ if (!prev_fence || nouveau_fence_channel(prev_fence) == chan)
+ return 0;
+
+ spin_lock(&nvbo->bo.lock);
+ ret = ttm_bo_wait(&nvbo->bo, false, false, false);
+ spin_unlock(&nvbo->bo.lock);
+ return ret;
+}
+
static void
nouveau_bo_del_ttm(struct ttm_buffer_object *bo)
{
diff --git a/drivers/gpu/drm/nouveau/nouveau_drv.h b/drivers/gpu/drm/nouveau/nouveau_drv.h
index 2eb622b..70a16f3 100644
--- a/drivers/gpu/drm/nouveau/nouveau_drv.h
+++ b/drivers/gpu/drm/nouveau/nouveau_drv.h
@@ -1167,6 +1167,7 @@ extern u16 nouveau_bo_rd16(struct nouveau_bo *nvbo, unsigned index);
extern void nouveau_bo_wr16(struct nouveau_bo *nvbo, unsigned index, u16 val);
extern u32 nouveau_bo_rd32(struct nouveau_bo *nvbo, unsigned index);
extern void nouveau_bo_wr32(struct nouveau_bo *nvbo, unsigned index, u32 val);
+extern int nouveau_bo_sync_gpu(struct nouveau_bo *, struct nouveau_channel *);
/* nouveau_fence.c */
struct nouveau_fence;
diff --git a/drivers/gpu/drm/nouveau/nouveau_gem.c b/drivers/gpu/drm/nouveau/nouveau_gem.c
index 62ac673..613f878 100644
--- a/drivers/gpu/drm/nouveau/nouveau_gem.c
+++ b/drivers/gpu/drm/nouveau/nouveau_gem.c
@@ -361,16 +361,11 @@ validate_list(struct nouveau_channel *chan, struct list_head *list,
list_for_each_entry(nvbo, list, entry) {
struct drm_nouveau_gem_pushbuf_bo *b = &pbbo[nvbo->pbbo_index];
- struct nouveau_fence *prev_fence = nvbo->bo.sync_obj;
- if (prev_fence && nouveau_fence_channel(prev_fence) != chan) {
- spin_lock(&nvbo->bo.lock);
- ret = ttm_bo_wait(&nvbo->bo, false, false, false);
- spin_unlock(&nvbo->bo.lock);
- if (unlikely(ret)) {
- NV_ERROR(dev, "fail wait other chan\n");
- return ret;
- }
+ ret = nouveau_bo_sync_gpu(nvbo, chan);
+ if (unlikely(ret)) {
+ NV_ERROR(dev, "fail pre-validate sync\n");
+ return ret;
}
ret = nouveau_gem_set_domain(nvbo->gem, b->read_domains,
@@ -381,7 +376,7 @@ validate_list(struct nouveau_channel *chan, struct list_head *list,
return ret;
}
- nvbo->channel = chan;
+ nvbo->channel = (b->read_domains & (1 << 31)) ? NULL : chan;
ret = ttm_bo_validate(&nvbo->bo, &nvbo->placement,
false, false, false);
nvbo->channel = NULL;
@@ -390,6 +385,12 @@ validate_list(struct nouveau_channel *chan, struct list_head *list,
return ret;
}
+ ret = nouveau_bo_sync_gpu(nvbo, chan);
+ if (unlikely(ret)) {
+ NV_ERROR(dev, "fail post-validate sync\n");
+ return ret;
+ }
+
if (nvbo->bo.offset == b->presumed.offset &&
((nvbo->bo.mem.mem_type == TTM_PL_VRAM &&
b->presumed.domain & NOUVEAU_GEM_DOMAIN_VRAM) ||
@@ -615,6 +616,21 @@ nouveau_gem_ioctl_pushbuf(struct drm_device *dev, void *data,
mutex_lock(&dev->struct_mutex);
+ /* Mark push buffers as being used on PFIFO, the validation code
+ * will then make sure that if the pushbuf bo moves, that they
+ * happen on the kernel channel, which will in turn cause a sync
+ * to happen before we try and submit the push buffer.
+ */
+ for (i = 0; i < req->nr_push; i++) {
+ if (push[i].bo_index >= req->nr_buffers) {
+ NV_ERROR(dev, "push %d buffer not in list\n", i);
+ ret = -EINVAL;
+ goto out;
+ }
+
+ bo[push[i].bo_index].read_domains |= (1 << 31);
+ }
+
/* Validate buffer list */
ret = nouveau_gem_pushbuf_validate(chan, file_priv, bo, req->buffers,
req->nr_buffers, &op, &do_reloc);
--
1.7.2.2

File diff suppressed because it is too large Load Diff

601
drm-polling-fixes.patch Normal file
View File

@ -0,0 +1,601 @@
commit 879d56da9b89b52de2109cadf1369967522428e8
Author: Dave Airlie <airlied@redhat.com>
Date: Tue Oct 26 12:55:52 2010 +1000
drm/radeon/kms: don't poll dac load detect.
This is slightly destructive, cpu intensive and can cause lockups.
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit 2563a90cdda1fe490947dc032ce17bf1118afc39
Author: Dave Airlie <airlied@redhat.com>
Date: Tue Nov 9 10:26:00 2010 +1000
drm: Use a nondestructive mode for output detect when polling (v2)
v2: Julien Cristau pointed out that @nondestructive results in
double-negatives and confusion when trying to interpret the parameter,
so use @force instead. Much easier to type as well. ;-)
And fix the miscompilation of vmgfx reported by Sedat Dilek.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
Conflicts:
drivers/gpu/drm/i915/intel_tv.c
drivers/gpu/drm/nouveau/nouveau_connector.c
commit deb557bbbcaa7fa8281d51490c73b51f579bc84d
Author: Dave Airlie <airlied@redhat.com>
Date: Tue Nov 9 10:24:06 2010 +1000
drm: Use a nondestructive mode for output detect when polling
Destructive load-detection is very expensive and due to failings
elsewhere can trigger system wide stalls of up to 600ms. A simple
first step to correcting this is not to invoke such an expensive
and destructive load-detection operation automatically.
Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=29536
Bugzilla: https://bugzilla.kernel.org/show_bug.cgi?id=16265
Reported-by: Bruno Prémont <bonbons@linux-vserver.org>
Tested-by: Sitsofe Wheeler <sitsofe@yahoo.com>
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: stable@kernel.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
Conflicts:
drivers/gpu/drm/i915/intel_tv.c
drivers/gpu/drm/nouveau/nouveau_connector.c
commit 8d5d3cd3612618a3c2214048788f0b2cc463ce0f
Author: Chris Wilson <chris@chris-wilson.co.uk>
Date: Mon Sep 6 23:53:47 2010 +0100
drm: Fix regression in disable polling e58f637
I broke out my trusty i845 and found a new boot failure, which upon
inspection turned out to be a recursion within:
drm_helper_probe_single_connector_modes() -> drm_helper_hpd_irq_event()
-> intel_crt_detect() -> drm_helper_probe_single_connector_modes()
Calling drm_kms_helper_poll_enable() instead performs the desired
re-initialisation of the polling should the user have toggled the
parameter, without the recursive side-effect.
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Dave Airlie <airlied@redhat.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit ec0becade1eae9b0f40b12ea4fcc3ea36e993ba3
Author: Dave Airlie <airlied@redhat.com>
Date: Tue Nov 9 09:48:34 2010 +1000
drm/kms: Add a module parameter to disable polling
Polling for a VGA device on an old system can be quite expensive,
causing latencies on the order of 600ms. As we hold the mode mutex for
this time and also need the same mutex to move the cursor, we trigger a
user-visible stall.
The real solution would involve improving the granulatity of the
locking and so perhaps performing some of the probing not under the lock
or some other updates can be done under different locks. Also reducing the
cost of probing for a non-existent monitor would be worthwhile. However,
exposing a parameter to disable polling is a simple workaround in the
meantime.
In order to accommodate users turning polling on and off at runtime, the
polling is potentially re-enabled on every probe. This is coupled to
the user calling xrandr, which seems to be a vaild time to reset the
polling timeout since the information on the connection has just been
updated. (The presumption being that all connections are probed in a
single xrandr pass, which is currently valid.)
References:
Bug 29536 - 2.6.35 causes ~600ms latency every 10s
https://bugs.freedesktop.org/show_bug.cgi?id=29536
Bug 16265 - Why is kslowd accumulating so much CPU time?
https://bugzilla.kernel.org/show_bug.cgi?id=16265
Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>
Reported-and-tested-by: Bruno Prémont <bonbons@linux-vserver.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Conflicts:
drivers/gpu/drm/drm_crtc_helper.c
commit 77e90256db2f7e6424717f7cc984b22d2832243f
Author: Dan Carpenter <error27@gmail.com>
Date: Thu Aug 19 11:46:29 2010 +0200
drm: move dereference below check
"fb_helper_conn" is dereferenced before the check for NULL. It's never
actually NULL here, so this is mostly to keep the static checkers happy.
Signed-off-by: Dan Carpenter <error27@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit 86e25290d9df7a84d185dfc037851d72d270a6c0
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Tue Jul 20 03:24:11 2010 -0400
drm/radeon/kms: make sure HPD is set to NONE on analog-only connectors
HPD is digital only.
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index 25d70d6..d537039 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -34,6 +34,9 @@
#include "drm_crtc_helper.h"
#include "drm_fb_helper.h"
+static bool drm_kms_helper_poll = true;
+module_param_named(poll, drm_kms_helper_poll, bool, 0600);
+
static void drm_mode_validate_flag(struct drm_connector *connector,
int flags)
{
@@ -98,8 +101,10 @@ int drm_helper_probe_single_connector_modes(struct drm_connector *connector,
connector->status = connector_status_disconnected;
if (connector->funcs->force)
connector->funcs->force(connector);
- } else
- connector->status = connector->funcs->detect(connector);
+ } else {
+ connector->status = connector->funcs->detect(connector, true);
+ drm_kms_helper_poll_enable(dev);
+ }
if (connector->status == connector_status_disconnected) {
DRM_DEBUG_KMS("%s is disconnected\n",
@@ -820,6 +825,9 @@ static void output_poll_execute(struct slow_work *work)
bool repoll = false, changed = false;
int ret;
+ if (!drm_kms_helper_poll)
+ return;
+
mutex_lock(&dev->mode_config.mutex);
list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
@@ -839,7 +847,7 @@ static void output_poll_execute(struct slow_work *work)
!(connector->polled & DRM_CONNECTOR_POLL_HPD))
continue;
- status = connector->funcs->detect(connector);
+ status = connector->funcs->detect(connector, false);
if (old_status != status)
changed = true;
}
@@ -874,6 +882,9 @@ void drm_kms_helper_poll_enable(struct drm_device *dev)
struct drm_connector *connector;
int ret;
+ if (!dev->mode_config.poll_enabled || !drm_kms_helper_poll)
+ return;
+
list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
if (connector->polled)
poll = true;
@@ -909,9 +920,12 @@ void drm_helper_hpd_irq_event(struct drm_device *dev)
{
if (!dev->mode_config.poll_enabled)
return;
+
delayed_slow_work_cancel(&dev->mode_config.output_poll_slow_work);
/* schedule a slow work asap */
- delayed_slow_work_enqueue(&dev->mode_config.output_poll_slow_work, 0);
+ if (drm_kms_helper_poll)
+ delayed_slow_work_enqueue(&dev->mode_config.output_poll_slow_work, 0);
+
}
EXPORT_SYMBOL(drm_helper_hpd_irq_event);
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 7196620..cef8d8d 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -94,10 +94,11 @@ static bool drm_fb_helper_connector_parse_command_line(struct drm_fb_helper_conn
int i;
enum drm_connector_force force = DRM_FORCE_UNSPECIFIED;
struct drm_fb_helper_cmdline_mode *cmdline_mode;
- struct drm_connector *connector = fb_helper_conn->connector;
+ struct drm_connector *connector;
if (!fb_helper_conn)
return false;
+ connector = fb_helper_conn->connector;
cmdline_mode = &fb_helper_conn->cmdline_mode;
if (!mode_option)
diff --git a/drivers/gpu/drm/drm_sysfs.c b/drivers/gpu/drm/drm_sysfs.c
index 101d381..9500af1 100644
--- a/drivers/gpu/drm/drm_sysfs.c
+++ b/drivers/gpu/drm/drm_sysfs.c
@@ -159,7 +159,7 @@ static ssize_t status_show(struct device *device,
struct drm_connector *connector = to_drm_connector(device);
enum drm_connector_status status;
- status = connector->funcs->detect(connector);
+ status = connector->funcs->detect(connector, true);
return snprintf(buf, PAGE_SIZE, "%s\n",
drm_get_connector_status_name(status));
}
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index ee0732b..3886b47 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -402,7 +402,8 @@ intel_crt_load_detect(struct drm_crtc *crtc, struct intel_encoder *intel_encoder
return status;
}
-static enum drm_connector_status intel_crt_detect(struct drm_connector *connector)
+static enum drm_connector_status
+intel_crt_detect(struct drm_connector *connector, bool force)
{
struct drm_device *dev = connector->dev;
struct drm_encoder *encoder = intel_attached_encoder(connector);
@@ -421,6 +422,9 @@ static enum drm_connector_status intel_crt_detect(struct drm_connector *connecto
if (intel_crt_detect_ddc(encoder))
return connector_status_connected;
+ if (!force)
+ return connector->status;
+
/* for pre-945g platforms use load detect */
if (encoder->crtc && encoder->crtc->enabled) {
status = intel_crt_load_detect(encoder->crtc, intel_encoder);
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index d9de8f1..b58249d 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1287,7 +1287,7 @@ ironlake_dp_detect(struct drm_connector *connector)
* \return false if DP port is disconnected.
*/
static enum drm_connector_status
-intel_dp_detect(struct drm_connector *connector)
+intel_dp_detect(struct drm_connector *connector, bool force)
{
struct drm_encoder *encoder = intel_attached_encoder(connector);
struct intel_encoder *intel_encoder = enc_to_intel_encoder(encoder);
diff --git a/drivers/gpu/drm/i915/intel_dvo.c b/drivers/gpu/drm/i915/intel_dvo.c
index 227feca..48a1889 100644
--- a/drivers/gpu/drm/i915/intel_dvo.c
+++ b/drivers/gpu/drm/i915/intel_dvo.c
@@ -211,7 +211,8 @@ static void intel_dvo_mode_set(struct drm_encoder *encoder,
*
* Unimplemented.
*/
-static enum drm_connector_status intel_dvo_detect(struct drm_connector *connector)
+static enum drm_connector_status
+intel_dvo_detect(struct drm_connector *connector, bool force)
{
struct drm_encoder *encoder = intel_attached_encoder(connector);
struct intel_encoder *intel_encoder = enc_to_intel_encoder(encoder);
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 83bd764..d1decfc 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -134,7 +134,7 @@ static bool intel_hdmi_mode_fixup(struct drm_encoder *encoder,
}
static enum drm_connector_status
-intel_hdmi_detect(struct drm_connector *connector)
+intel_hdmi_detect(struct drm_connector *connector, bool force)
{
struct drm_encoder *encoder = intel_attached_encoder(connector);
struct intel_encoder *intel_encoder = enc_to_intel_encoder(encoder);
diff --git a/drivers/gpu/drm/i915/intel_lvds.c b/drivers/gpu/drm/i915/intel_lvds.c
index 7d42ff1..9ad3425 100644
--- a/drivers/gpu/drm/i915/intel_lvds.c
+++ b/drivers/gpu/drm/i915/intel_lvds.c
@@ -546,7 +546,8 @@ static void intel_lvds_mode_set(struct drm_encoder *encoder,
* connected and closed means disconnected. We also send hotplug events as
* needed, using lid status notification from the input layer.
*/
-static enum drm_connector_status intel_lvds_detect(struct drm_connector *connector)
+static enum drm_connector_status
+intel_lvds_detect(struct drm_connector *connector, bool force)
{
struct drm_device *dev = connector->dev;
enum drm_connector_status status = connector_status_connected;
@@ -641,7 +642,9 @@ static int intel_lid_notify(struct notifier_block *nb, unsigned long val,
* the LID nofication event.
*/
if (connector)
- connector->status = connector->funcs->detect(connector);
+ connector->status = connector->funcs->detect(connector,
+ false);
+
/* Don't force modeset on machines where it causes a GPU lockup */
if (dmi_check_system(intel_no_modeset_on_lid))
return NOTIFY_OK;
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 76993ac..76c9b3d 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -1496,7 +1496,7 @@ intel_analog_is_connected(struct drm_device *dev)
if (!analog_connector)
return false;
- if (analog_connector->funcs->detect(analog_connector) ==
+ if (analog_connector->funcs->detect(analog_connector, false) ==
connector_status_disconnected)
return false;
@@ -1567,7 +1567,8 @@ intel_sdvo_hdmi_sink_detect(struct drm_connector *connector)
return status;
}
-static enum drm_connector_status intel_sdvo_detect(struct drm_connector *connector)
+static enum drm_connector_status
+intel_sdvo_detect(struct drm_connector *connector, bool force)
{
uint16_t response;
u8 status;
diff --git a/drivers/gpu/drm/i915/intel_tv.c b/drivers/gpu/drm/i915/intel_tv.c
index 6d553c2..ad40f1b 100644
--- a/drivers/gpu/drm/i915/intel_tv.c
+++ b/drivers/gpu/drm/i915/intel_tv.c
@@ -1336,7 +1336,7 @@ static void intel_tv_find_better_format(struct drm_connector *connector)
* we have a pipe programmed in order to probe the TV.
*/
static enum drm_connector_status
-intel_tv_detect(struct drm_connector *connector)
+intel_tv_detect(struct drm_connector *connector, bool force)
{
struct drm_crtc *crtc;
struct drm_display_mode mode;
@@ -1351,7 +1351,7 @@ intel_tv_detect(struct drm_connector *connector)
if (encoder->crtc && encoder->crtc->enabled) {
type = intel_tv_detect_type(encoder->crtc, intel_encoder);
- } else {
+ } else if (force) {
crtc = intel_get_load_detect_pipe(intel_encoder, connector,
&mode, &dpms_mode);
if (crtc) {
@@ -1359,8 +1359,9 @@ intel_tv_detect(struct drm_connector *connector)
intel_release_load_detect_pipe(intel_encoder, connector,
dpms_mode);
} else
- type = -1;
- }
+ return connector_status_unknown;
+ } else
+ return connector->status;
tv_priv->type = type;
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index 149ed22..1085376 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -228,7 +228,7 @@ nouveau_connector_set_encoder(struct drm_connector *connector,
}
static enum drm_connector_status
-nouveau_connector_detect(struct drm_connector *connector)
+nouveau_connector_detect(struct drm_connector *connector, bool force)
{
struct drm_device *dev = connector->dev;
struct nouveau_connector *nv_connector = nouveau_connector(connector);
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index adccbc2..1680600 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -467,7 +467,8 @@ static int radeon_lvds_mode_valid(struct drm_connector *connector,
return MODE_OK;
}
-static enum drm_connector_status radeon_lvds_detect(struct drm_connector *connector)
+static enum drm_connector_status
+radeon_lvds_detect(struct drm_connector *connector, bool force)
{
struct radeon_connector *radeon_connector = to_radeon_connector(connector);
struct drm_encoder *encoder = radeon_best_single_encoder(connector);
@@ -582,7 +583,8 @@ static int radeon_vga_mode_valid(struct drm_connector *connector,
return MODE_OK;
}
-static enum drm_connector_status radeon_vga_detect(struct drm_connector *connector)
+static enum drm_connector_status
+radeon_vga_detect(struct drm_connector *connector, bool force)
{
struct radeon_connector *radeon_connector = to_radeon_connector(connector);
struct drm_encoder *encoder;
@@ -621,6 +623,11 @@ static enum drm_connector_status radeon_vga_detect(struct drm_connector *connect
ret = connector_status_connected;
}
} else {
+
+ /* if we aren't forcing don't do destructive polling */
+ if (!force)
+ return connector->status;
+
if (radeon_connector->dac_load_detect && encoder) {
encoder_funcs = encoder->helper_private;
ret = encoder_funcs->detect(encoder, connector);
@@ -679,7 +686,8 @@ static int radeon_tv_mode_valid(struct drm_connector *connector,
return MODE_OK;
}
-static enum drm_connector_status radeon_tv_detect(struct drm_connector *connector)
+static enum drm_connector_status
+radeon_tv_detect(struct drm_connector *connector, bool force)
{
struct drm_encoder *encoder;
struct drm_encoder_helper_funcs *encoder_funcs;
@@ -736,7 +744,8 @@ static int radeon_dvi_get_modes(struct drm_connector *connector)
* we have to check if this analog encoder is shared with anyone else (TV)
* if its shared we have to set the other connector to disconnected.
*/
-static enum drm_connector_status radeon_dvi_detect(struct drm_connector *connector)
+static enum drm_connector_status
+radeon_dvi_detect(struct drm_connector *connector, bool force)
{
struct radeon_connector *radeon_connector = to_radeon_connector(connector);
struct drm_encoder *encoder = NULL;
@@ -806,6 +815,11 @@ static enum drm_connector_status radeon_dvi_detect(struct drm_connector *connect
if ((ret == connector_status_connected) && (radeon_connector->use_digital == true))
goto out;
+ if (!force) {
+ ret = connector->status;
+ goto out;
+ }
+
/* find analog encoder */
if (radeon_connector->dac_load_detect) {
for (i = 0; i < DRM_CONNECTOR_MAX_ENCODER; i++) {
@@ -962,7 +976,8 @@ static int radeon_dp_get_modes(struct drm_connector *connector)
return ret;
}
-static enum drm_connector_status radeon_dp_detect(struct drm_connector *connector)
+static enum drm_connector_status
+radeon_dp_detect(struct drm_connector *connector, bool force)
{
struct radeon_connector *radeon_connector = to_radeon_connector(connector);
enum drm_connector_status ret = connector_status_disconnected;
@@ -1082,6 +1097,8 @@ radeon_add_atom_connector(struct drm_device *dev,
drm_connector_attach_property(&radeon_connector->base,
rdev->mode_info.load_detect_property,
1);
+ /* no HPD on analog connectors */
+ radeon_connector->hpd.hpd = RADEON_HPD_NONE;
connector->polled = DRM_CONNECTOR_POLL_CONNECT;
connector->interlace_allowed = true;
connector->doublescan_allowed = true;
@@ -1096,6 +1113,8 @@ radeon_add_atom_connector(struct drm_device *dev,
drm_connector_attach_property(&radeon_connector->base,
rdev->mode_info.load_detect_property,
1);
+ /* no HPD on analog connectors */
+ radeon_connector->hpd.hpd = RADEON_HPD_NONE;
connector->interlace_allowed = true;
connector->doublescan_allowed = true;
break;
@@ -1186,6 +1205,8 @@ radeon_add_atom_connector(struct drm_device *dev,
drm_connector_attach_property(&radeon_connector->base,
rdev->mode_info.tv_std_property,
radeon_atombios_get_tv_info(rdev));
+ /* no HPD on analog connectors */
+ radeon_connector->hpd.hpd = RADEON_HPD_NONE;
}
connector->interlace_allowed = false;
connector->doublescan_allowed = false;
@@ -1209,7 +1230,7 @@ radeon_add_atom_connector(struct drm_device *dev,
break;
}
- if (hpd->hpd == RADEON_HPD_NONE) {
+ if (radeon_connector->hpd.hpd == RADEON_HPD_NONE) {
if (i2c_bus->valid)
connector->polled = DRM_CONNECTOR_POLL_CONNECT;
} else
@@ -1276,6 +1297,8 @@ radeon_add_legacy_connector(struct drm_device *dev,
drm_connector_attach_property(&radeon_connector->base,
rdev->mode_info.load_detect_property,
1);
+ /* no HPD on analog connectors */
+ radeon_connector->hpd.hpd = RADEON_HPD_NONE;
connector->polled = DRM_CONNECTOR_POLL_CONNECT;
connector->interlace_allowed = true;
connector->doublescan_allowed = true;
@@ -1290,6 +1313,8 @@ radeon_add_legacy_connector(struct drm_device *dev,
drm_connector_attach_property(&radeon_connector->base,
rdev->mode_info.load_detect_property,
1);
+ /* no HPD on analog connectors */
+ radeon_connector->hpd.hpd = RADEON_HPD_NONE;
connector->interlace_allowed = true;
connector->doublescan_allowed = true;
break;
@@ -1328,6 +1353,8 @@ radeon_add_legacy_connector(struct drm_device *dev,
drm_connector_attach_property(&radeon_connector->base,
rdev->mode_info.tv_std_property,
radeon_combios_get_tv_info(rdev));
+ /* no HPD on analog connectors */
+ radeon_connector->hpd.hpd = RADEON_HPD_NONE;
}
connector->interlace_allowed = false;
connector->doublescan_allowed = false;
@@ -1345,7 +1372,7 @@ radeon_add_legacy_connector(struct drm_device *dev,
break;
}
- if (hpd->hpd == RADEON_HPD_NONE) {
+ if (radeon_connector->hpd.hpd == RADEON_HPD_NONE) {
if (i2c_bus->valid)
connector->polled = DRM_CONNECTOR_POLL_CONNECT;
} else
diff --git a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
index cfaf690..5b638cb 100644
--- a/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
+++ b/drivers/gpu/drm/vmwgfx/vmwgfx_ldu.c
@@ -335,7 +335,8 @@ static void vmw_ldu_connector_restore(struct drm_connector *connector)
}
static enum drm_connector_status
- vmw_ldu_connector_detect(struct drm_connector *connector)
+ vmw_ldu_connector_detect(struct drm_connector *connector,
+ bool force)
{
if (vmw_connector_to_ldu(connector)->pref_active)
return connector_status_connected;
@@ -516,7 +517,7 @@ static int vmw_ldu_init(struct vmw_private *dev_priv, unsigned unit)
drm_connector_init(dev, connector, &vmw_legacy_connector_funcs,
DRM_MODE_CONNECTOR_LVDS);
- connector->status = vmw_ldu_connector_detect(connector);
+ connector->status = vmw_ldu_connector_detect(connector, true);
drm_encoder_init(dev, encoder, &vmw_legacy_encoder_funcs,
DRM_MODE_ENCODER_LVDS);
@@ -610,7 +611,7 @@ int vmw_kms_ldu_update_layout(struct vmw_private *dev_priv, unsigned num,
ldu->pref_height = 600;
ldu->pref_active = false;
}
- con->status = vmw_ldu_connector_detect(con);
+ con->status = vmw_ldu_connector_detect(con, true);
}
mutex_unlock(&dev->mode_config.mutex);
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 93a1a31..afbb578 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -420,7 +420,15 @@ struct drm_connector_funcs {
void (*dpms)(struct drm_connector *connector, int mode);
void (*save)(struct drm_connector *connector);
void (*restore)(struct drm_connector *connector);
- enum drm_connector_status (*detect)(struct drm_connector *connector);
+
+ /* Check to see if anything is attached to the connector.
+ * @force is set to false whilst polling, true when checking the
+ * connector due to user request. @force can be used by the driver
+ * to avoid expensive, destructive operations during automated
+ * probing.
+ */
+ enum drm_connector_status (*detect)(struct drm_connector *connector,
+ bool force);
int (*fill_modes)(struct drm_connector *connector, uint32_t max_width, uint32_t max_height);
int (*set_property)(struct drm_connector *connector, struct drm_property *property,
uint64_t val);

View File

@ -0,0 +1,854 @@
commit 4adf332cc24ee2d46064aaafd8216169d29566d5
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Sun Nov 14 20:24:35 2010 -0500
drm/radeon/kms: fix and unify tiled buffer alignment checking for r6xx/7xx
Tiled buffers have the same alignment requirements regardless of
whether the surface is for db, cb, or textures. Previously, the
calculations where inconsistent for each buffer type.
- Unify the alignment calculations in a common function
- Standardize the alignment units (pixels for pitch/height/depth,
bytes for base)
- properly check the buffer base alignments
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit c37cb9e61dce7437f63280d9347a9ffdf4ec34e7
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Wed Oct 27 01:44:35 2010 -0400
drm/radeon/kms: fix tiled db height calculation on 6xx/7xx
Calculate height based on the slice bitfield rather than the size.
Same as Dave's CB fix.
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit b80c8fdc2fadd8182b958e91a10f2fa287f993e4
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Tue Oct 26 20:22:42 2010 -0400
drm/radeon/kms: fix r6xx/7xx 1D tiling CS checker v2
broken by:
drm/radeon/r600: fix tiling issues in CS checker.
v2: only apply it to 1D tiling case.
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit 4336ac5c0a4e5dfbb51631ad680d6a5d0b295cd3
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Mon Oct 18 23:45:39 2010 -0400
drm/radeon/kms: fix 2D tile height alignment in the r600 CS checker
macro tile heights are aligned to num channels, not num banks.
Noticed by Dave Airlie.
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Cc: stable@kernel.org
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit 5c76976a1419a633f9f33c6547bae00348b855d2
Author: Dave Airlie <airlied@redhat.com>
Date: Thu Oct 21 13:55:40 2010 +1000
drm/radeon/r600: fix tiling issues in CS checker.
The CS checker had some incorrect alignment requirements for 2D surfaces,
this made rendering to mipmap levels that were 2D broken.
Also the CB height was being worked out from the BO size, this doesn't work
at all when rendering mipmap levels, instead we work out what height userspace
wanted from slice max and use that to check it fits inside the BO, however
the DDX send the wrong slice max for an unaligned buffer so we have to workaround
for that even though its a userspace bug.
Reviewed-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit fa479e9df4558af6f091c45be37f713e64b836a1
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Tue Sep 14 10:10:47 2010 -0400
drm/radeon/kms: only warn on mipmap size checks in r600 cs checker (v2)
The texture base address registers are in units of 256 bytes.
The original CS checker treated these offsets as bytes, so the
original check was wrong. I fixed the units in a patch during
the 2.6.36 cycle, but this ended up breaking some existing
userspace (probably due to a bug in either userspace texture allocation
or the drm texture mipmap checker). So for now, until we come
up with a better fix, just warn if the mipmap size it too large.
This will keep existing userspace working and it should be just
as safe as before when we were checking the wrong units. These
are GPU MC addresses, so if they fall outside of the VRAM or
GART apertures, they end up at the GPU default page, so this should
be safe from a security perspective.
v2: Just disable the warning. It just spams the log and there's
nothing the user can do about it.
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Cc: Jerome Glisse <glisse@freedesktop.org>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit 6e8df81d43d5c95fe37db7f0ef55332de1a4b698
Author: Dave Airlie <airlied@redhat.com>
Date: Thu Aug 12 09:40:05 2010 +1000
drm/radeon: drop old and broken mesa warning
This never really got fixed in mesa, and the kernel deals with the problem
just fine, so don't got reporting things that confuse people.
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit 23f012fb9a0633f2f8901440e314d6276255b1c0
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Wed Aug 11 11:54:25 2010 -0400
drm/radeon/kms: another r6xx/r7xx CS checker fix
add default case for buffer formats
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Cc: Andre Maasikas <amaasikas@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit 2c7e76decda2d437f0ca064fef1a2d5d8892288e
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Fri Aug 6 02:54:05 2010 -0400
drm/radeon/kms: r600 CS parser fixes
- buffer offsets in the base regs are 256b aligned so
shift properly when comparing, fixed by Andre Maasikas
- mipmap size was calculated wrong when nlevel=0
- texture bo offsets were used after the bo base address was added
- vertex resource size register is size - 1, not size
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Cc: Andre Maasikas <amaasikas@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit 85d1363c9f15b5d4303b635142cee0ba9d1473fc
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Fri Jun 4 18:41:42 2010 -0400
drm/radeon/kms: fix CS alignment checking for tiling (v2)
Covers depth, cb, and textures. Hopefully I got this right.
v2: - fix bugs:
https://bugs.freedesktop.org/show_bug.cgi?id=28327
https://bugs.freedesktop.org/show_bug.cgi?id=28381
- use ALIGNED(), IS_ALIGNED() macros
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
commit b956b6e7c7fb207daf32520c0a72c8c06ef1d5f5
Author: Alex Deucher <alexdeucher@gmail.com>
Date: Thu May 20 12:43:52 2010 -0400
drm/radeon/kms: add tiling support to the cs checker for r6xx/r7xx
Check for relocs for DB_DEPTH_INFO, CB_COLOR*_INFO, and texture
resources.
Signed-off-by: Alex Deucher <alexdeucher@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
drivers/gpu/drm/radeon/r600_cs.c | 391 ++++++++++++++++++++++++++++++--------
drivers/gpu/drm/radeon/r600d.h | 12 ++
2 files changed, 324 insertions(+), 79 deletions(-)
diff --git a/drivers/gpu/drm/radeon/r600_cs.c b/drivers/gpu/drm/radeon/r600_cs.c
index 144c32d..0f90fc3 100644
--- a/drivers/gpu/drm/radeon/r600_cs.c
+++ b/drivers/gpu/drm/radeon/r600_cs.c
@@ -25,6 +25,7 @@
* Alex Deucher
* Jerome Glisse
*/
+#include <linux/kernel.h>
#include "drmP.h"
#include "radeon.h"
#include "r600d.h"
@@ -49,6 +50,7 @@ struct r600_cs_track {
u32 nsamples;
u32 cb_color_base_last[8];
struct radeon_bo *cb_color_bo[8];
+ u64 cb_color_bo_mc[8];
u32 cb_color_bo_offset[8];
struct radeon_bo *cb_color_frag_bo[8];
struct radeon_bo *cb_color_tile_bo[8];
@@ -66,6 +68,7 @@ struct r600_cs_track {
u32 db_depth_size;
u32 db_offset;
struct radeon_bo *db_bo;
+ u64 db_bo_mc;
};
static inline int r600_bpe_from_format(u32 *bpe, u32 format)
@@ -132,12 +135,75 @@ static inline int r600_bpe_from_format(u32 *bpe, u32 format)
case V_038004_FMT_GB_GR:
case V_038004_FMT_BG_RG:
case V_038004_COLOR_INVALID:
+ default:
*bpe = 16;
return -EINVAL;
}
return 0;
}
+struct array_mode_checker {
+ int array_mode;
+ u32 group_size;
+ u32 nbanks;
+ u32 npipes;
+ u32 nsamples;
+ u32 bpe;
+};
+
+/* returns alignment in pixels for pitch/height/depth and bytes for base */
+static inline int r600_get_array_mode_alignment(struct array_mode_checker *values,
+ u32 *pitch_align,
+ u32 *height_align,
+ u32 *depth_align,
+ u64 *base_align)
+{
+ u32 tile_width = 8;
+ u32 tile_height = 8;
+ u32 macro_tile_width = values->nbanks;
+ u32 macro_tile_height = values->npipes;
+ u32 tile_bytes = tile_width * tile_height * values->bpe * values->nsamples;
+ u32 macro_tile_bytes = macro_tile_width * macro_tile_height * tile_bytes;
+
+ switch (values->array_mode) {
+ case ARRAY_LINEAR_GENERAL:
+ /* technically tile_width/_height for pitch/height */
+ *pitch_align = 1; /* tile_width */
+ *height_align = 1; /* tile_height */
+ *depth_align = 1;
+ *base_align = 1;
+ break;
+ case ARRAY_LINEAR_ALIGNED:
+ *pitch_align = max((u32)64, (u32)(values->group_size / values->bpe));
+ *height_align = tile_height;
+ *depth_align = 1;
+ *base_align = values->group_size;
+ break;
+ case ARRAY_1D_TILED_THIN1:
+ *pitch_align = max((u32)tile_width,
+ (u32)(values->group_size /
+ (tile_height * values->bpe * values->nsamples)));
+ *height_align = tile_height;
+ *depth_align = 1;
+ *base_align = values->group_size;
+ break;
+ case ARRAY_2D_TILED_THIN1:
+ *pitch_align = max((u32)macro_tile_width,
+ (u32)(((values->group_size / tile_height) /
+ (values->bpe * values->nsamples)) *
+ values->nbanks)) * tile_width;
+ *height_align = macro_tile_height * tile_height;
+ *depth_align = 1;
+ *base_align = max(macro_tile_bytes,
+ (*pitch_align) * values->bpe * (*height_align) * values->nsamples);
+ break;
+ default:
+ return -EINVAL;
+ }
+
+ return 0;
+}
+
static void r600_cs_track_init(struct r600_cs_track *track)
{
int i;
@@ -151,10 +217,12 @@ static void r600_cs_track_init(struct r600_cs_track *track)
track->cb_color_info[i] = 0;
track->cb_color_bo[i] = NULL;
track->cb_color_bo_offset[i] = 0xFFFFFFFF;
+ track->cb_color_bo_mc[i] = 0xFFFFFFFF;
}
track->cb_target_mask = 0xFFFFFFFF;
track->cb_shader_mask = 0xFFFFFFFF;
track->db_bo = NULL;
+ track->db_bo_mc = 0xFFFFFFFF;
/* assume the biggest format and that htile is enabled */
track->db_depth_info = 7 | (1 << 25);
track->db_depth_view = 0xFFFFC000;
@@ -166,70 +234,58 @@ static void r600_cs_track_init(struct r600_cs_track *track)
static inline int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
{
struct r600_cs_track *track = p->track;
- u32 bpe = 0, pitch, slice_tile_max, size, tmp, height;
+ u32 bpe = 0, slice_tile_max, size, tmp;
+ u32 height, height_align, pitch, pitch_align, depth_align;
+ u64 base_offset, base_align;
+ struct array_mode_checker array_check;
volatile u32 *ib = p->ib->ptr;
+ unsigned array_mode;
if (G_0280A0_TILE_MODE(track->cb_color_info[i])) {
dev_warn(p->dev, "FMASK or CMASK buffer are not supported by this kernel\n");
return -EINVAL;
}
- size = radeon_bo_size(track->cb_color_bo[i]);
+ size = radeon_bo_size(track->cb_color_bo[i]) - track->cb_color_bo_offset[i];
if (r600_bpe_from_format(&bpe, G_0280A0_FORMAT(track->cb_color_info[i]))) {
dev_warn(p->dev, "%s:%d cb invalid format %d for %d (0x%08X)\n",
__func__, __LINE__, G_0280A0_FORMAT(track->cb_color_info[i]),
i, track->cb_color_info[i]);
return -EINVAL;
}
- pitch = (G_028060_PITCH_TILE_MAX(track->cb_color_size[i]) + 1) << 3;
+ /* pitch in pixels */
+ pitch = (G_028060_PITCH_TILE_MAX(track->cb_color_size[i]) + 1) * 8;
slice_tile_max = G_028060_SLICE_TILE_MAX(track->cb_color_size[i]) + 1;
- if (!pitch) {
- dev_warn(p->dev, "%s:%d cb pitch (%d) for %d invalid (0x%08X)\n",
- __func__, __LINE__, pitch, i, track->cb_color_size[i]);
- return -EINVAL;
- }
- height = size / (pitch * bpe);
+ slice_tile_max *= 64;
+ height = slice_tile_max / pitch;
if (height > 8192)
height = 8192;
- switch (G_0280A0_ARRAY_MODE(track->cb_color_info[i])) {
+ array_mode = G_0280A0_ARRAY_MODE(track->cb_color_info[i]);
+
+ base_offset = track->cb_color_bo_mc[i] + track->cb_color_bo_offset[i];
+ array_check.array_mode = array_mode;
+ array_check.group_size = track->group_size;
+ array_check.nbanks = track->nbanks;
+ array_check.npipes = track->npipes;
+ array_check.nsamples = track->nsamples;
+ array_check.bpe = bpe;
+ if (r600_get_array_mode_alignment(&array_check,
+ &pitch_align, &height_align, &depth_align, &base_align)) {
+ dev_warn(p->dev, "%s invalid tiling %d for %d (0x%08X)\n", __func__,
+ G_0280A0_ARRAY_MODE(track->cb_color_info[i]), i,
+ track->cb_color_info[i]);
+ return -EINVAL;
+ }
+ switch (array_mode) {
case V_0280A0_ARRAY_LINEAR_GENERAL:
+ break;
case V_0280A0_ARRAY_LINEAR_ALIGNED:
- if (pitch & 0x3f) {
- dev_warn(p->dev, "%s:%d cb pitch (%d x %d = %d) invalid\n",
- __func__, __LINE__, pitch, bpe, pitch * bpe);
- return -EINVAL;
- }
- if ((pitch * bpe) & (track->group_size - 1)) {
- dev_warn(p->dev, "%s:%d cb pitch (%d) invalid\n",
- __func__, __LINE__, pitch);
- return -EINVAL;
- }
break;
case V_0280A0_ARRAY_1D_TILED_THIN1:
- if ((pitch * 8 * bpe * track->nsamples) & (track->group_size - 1)) {
- dev_warn(p->dev, "%s:%d cb pitch (%d) invalid\n",
- __func__, __LINE__, pitch);
- return -EINVAL;
- }
- height &= ~0x7;
- if (!height)
- height = 8;
+ /* avoid breaking userspace */
+ if (height > 7)
+ height &= ~0x7;
break;
case V_0280A0_ARRAY_2D_TILED_THIN1:
- if (pitch & ((8 * track->nbanks) - 1)) {
- dev_warn(p->dev, "%s:%d cb pitch (%d) invalid\n",
- __func__, __LINE__, pitch);
- return -EINVAL;
- }
- tmp = pitch * 8 * bpe * track->nsamples;
- tmp = tmp / track->nbanks;
- if (tmp & (track->group_size - 1)) {
- dev_warn(p->dev, "%s:%d cb pitch (%d) invalid\n",
- __func__, __LINE__, pitch);
- return -EINVAL;
- }
- height &= ~((16 * track->npipes) - 1);
- if (!height)
- height = 16 * track->npipes;
break;
default:
dev_warn(p->dev, "%s invalid tiling %d for %d (0x%08X)\n", __func__,
@@ -237,17 +293,43 @@ static inline int r600_cs_track_validate_cb(struct radeon_cs_parser *p, int i)
track->cb_color_info[i]);
return -EINVAL;
}
+
+ if (!IS_ALIGNED(pitch, pitch_align)) {
+ dev_warn(p->dev, "%s:%d cb pitch (%d) invalid\n",
+ __func__, __LINE__, pitch);
+ return -EINVAL;
+ }
+ if (!IS_ALIGNED(height, height_align)) {
+ dev_warn(p->dev, "%s:%d cb height (%d) invalid\n",
+ __func__, __LINE__, height);
+ return -EINVAL;
+ }
+ if (!IS_ALIGNED(base_offset, base_align)) {
+ dev_warn(p->dev, "%s offset[%d] 0x%llx not aligned\n", __func__, i, base_offset);
+ return -EINVAL;
+ }
+
/* check offset */
- tmp = height * pitch;
+ tmp = height * pitch * bpe;
if ((tmp + track->cb_color_bo_offset[i]) > radeon_bo_size(track->cb_color_bo[i])) {
- dev_warn(p->dev, "%s offset[%d] %d to big\n", __func__, i, track->cb_color_bo_offset[i]);
- return -EINVAL;
+ if (array_mode == V_0280A0_ARRAY_LINEAR_GENERAL) {
+ /* the initial DDX does bad things with the CB size occasionally */
+ /* it rounds up height too far for slice tile max but the BO is smaller */
+ tmp = (height - 7) * 8 * bpe;
+ if ((tmp + track->cb_color_bo_offset[i]) > radeon_bo_size(track->cb_color_bo[i])) {
+ dev_warn(p->dev, "%s offset[%d] %d %d %lu too big\n", __func__, i, track->cb_color_bo_offset[i], tmp, radeon_bo_size(track->cb_color_bo[i]));
+ return -EINVAL;
+ }
+ } else {
+ dev_warn(p->dev, "%s offset[%d] %d %d %lu too big\n", __func__, i, track->cb_color_bo_offset[i], tmp, radeon_bo_size(track->cb_color_bo[i]));
+ return -EINVAL;
+ }
}
/* limit max tile */
tmp = (height * pitch) >> 6;
if (tmp < slice_tile_max)
slice_tile_max = tmp;
- tmp = S_028060_PITCH_TILE_MAX((pitch >> 3) - 1) |
+ tmp = S_028060_PITCH_TILE_MAX((pitch / 8) - 1) |
S_028060_SLICE_TILE_MAX(slice_tile_max - 1);
ib[track->cb_color_size_idx[i]] = tmp;
return 0;
@@ -289,7 +371,12 @@ static int r600_cs_track_check(struct radeon_cs_parser *p)
/* Check depth buffer */
if (G_028800_STENCIL_ENABLE(track->db_depth_control) ||
G_028800_Z_ENABLE(track->db_depth_control)) {
- u32 nviews, bpe, ntiles;
+ u32 nviews, bpe, ntiles, size, slice_tile_max;
+ u32 height, height_align, pitch, pitch_align, depth_align;
+ u64 base_offset, base_align;
+ struct array_mode_checker array_check;
+ int array_mode;
+
if (track->db_bo == NULL) {
dev_warn(p->dev, "z/stencil with no depth buffer\n");
return -EINVAL;
@@ -321,7 +408,6 @@ static int r600_cs_track_check(struct radeon_cs_parser *p)
dev_warn(p->dev, "z/stencil buffer size not set\n");
return -EINVAL;
}
- printk_once(KERN_WARNING "You have old & broken userspace please consider updating mesa\n");
tmp = radeon_bo_size(track->db_bo) - track->db_offset;
tmp = (tmp / bpe) >> 6;
if (!tmp) {
@@ -332,11 +418,63 @@ static int r600_cs_track_check(struct radeon_cs_parser *p)
}
ib[track->db_depth_size_idx] = S_028000_SLICE_TILE_MAX(tmp - 1) | (track->db_depth_size & 0x3FF);
} else {
+ size = radeon_bo_size(track->db_bo);
+ /* pitch in pixels */
+ pitch = (G_028000_PITCH_TILE_MAX(track->db_depth_size) + 1) * 8;
+ slice_tile_max = G_028000_SLICE_TILE_MAX(track->db_depth_size) + 1;
+ slice_tile_max *= 64;
+ height = slice_tile_max / pitch;
+ if (height > 8192)
+ height = 8192;
+ base_offset = track->db_bo_mc + track->db_offset;
+ array_mode = G_028010_ARRAY_MODE(track->db_depth_info);
+ array_check.array_mode = array_mode;
+ array_check.group_size = track->group_size;
+ array_check.nbanks = track->nbanks;
+ array_check.npipes = track->npipes;
+ array_check.nsamples = track->nsamples;
+ array_check.bpe = bpe;
+ if (r600_get_array_mode_alignment(&array_check,
+ &pitch_align, &height_align, &depth_align, &base_align)) {
+ dev_warn(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__,
+ G_028010_ARRAY_MODE(track->db_depth_info),
+ track->db_depth_info);
+ return -EINVAL;
+ }
+ switch (array_mode) {
+ case V_028010_ARRAY_1D_TILED_THIN1:
+ /* don't break userspace */
+ height &= ~0x7;
+ break;
+ case V_028010_ARRAY_2D_TILED_THIN1:
+ break;
+ default:
+ dev_warn(p->dev, "%s invalid tiling %d (0x%08X)\n", __func__,
+ G_028010_ARRAY_MODE(track->db_depth_info),
+ track->db_depth_info);
+ return -EINVAL;
+ }
+
+ if (!IS_ALIGNED(pitch, pitch_align)) {
+ dev_warn(p->dev, "%s:%d db pitch (%d) invalid\n",
+ __func__, __LINE__, pitch);
+ return -EINVAL;
+ }
+ if (!IS_ALIGNED(height, height_align)) {
+ dev_warn(p->dev, "%s:%d db height (%d) invalid\n",
+ __func__, __LINE__, height);
+ return -EINVAL;
+ }
+ if (!IS_ALIGNED(base_offset, base_align)) {
+ dev_warn(p->dev, "%s offset[%d] 0x%llx not aligned\n", __func__, i, base_offset);
+ return -EINVAL;
+ }
+
ntiles = G_028000_SLICE_TILE_MAX(track->db_depth_size) + 1;
nviews = G_028004_SLICE_MAX(track->db_depth_view) + 1;
tmp = ntiles * bpe * 64 * nviews;
if ((tmp + track->db_offset) > radeon_bo_size(track->db_bo)) {
- dev_warn(p->dev, "z/stencil buffer too small (0x%08X %d %d %d -> %d have %ld)\n",
+ dev_warn(p->dev, "z/stencil buffer too small (0x%08X %d %d %d -> %u have %lu)\n",
track->db_depth_size, ntiles, nviews, bpe, tmp + track->db_offset,
radeon_bo_size(track->db_bo));
return -EINVAL;
@@ -724,7 +862,25 @@ static inline int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx
track->db_depth_control = radeon_get_ib_value(p, idx);
break;
case R_028010_DB_DEPTH_INFO:
- track->db_depth_info = radeon_get_ib_value(p, idx);
+ if (r600_cs_packet_next_is_pkt3_nop(p)) {
+ r = r600_cs_packet_next_reloc(p, &reloc);
+ if (r) {
+ dev_warn(p->dev, "bad SET_CONTEXT_REG "
+ "0x%04X\n", reg);
+ return -EINVAL;
+ }
+ track->db_depth_info = radeon_get_ib_value(p, idx);
+ ib[idx] &= C_028010_ARRAY_MODE;
+ track->db_depth_info &= C_028010_ARRAY_MODE;
+ if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
+ ib[idx] |= S_028010_ARRAY_MODE(V_028010_ARRAY_2D_TILED_THIN1);
+ track->db_depth_info |= S_028010_ARRAY_MODE(V_028010_ARRAY_2D_TILED_THIN1);
+ } else {
+ ib[idx] |= S_028010_ARRAY_MODE(V_028010_ARRAY_1D_TILED_THIN1);
+ track->db_depth_info |= S_028010_ARRAY_MODE(V_028010_ARRAY_1D_TILED_THIN1);
+ }
+ } else
+ track->db_depth_info = radeon_get_ib_value(p, idx);
break;
case R_028004_DB_DEPTH_VIEW:
track->db_depth_view = radeon_get_ib_value(p, idx);
@@ -757,8 +913,25 @@ static inline int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx
case R_0280B4_CB_COLOR5_INFO:
case R_0280B8_CB_COLOR6_INFO:
case R_0280BC_CB_COLOR7_INFO:
- tmp = (reg - R_0280A0_CB_COLOR0_INFO) / 4;
- track->cb_color_info[tmp] = radeon_get_ib_value(p, idx);
+ if (r600_cs_packet_next_is_pkt3_nop(p)) {
+ r = r600_cs_packet_next_reloc(p, &reloc);
+ if (r) {
+ dev_err(p->dev, "bad SET_CONTEXT_REG 0x%04X\n", reg);
+ return -EINVAL;
+ }
+ tmp = (reg - R_0280A0_CB_COLOR0_INFO) / 4;
+ track->cb_color_info[tmp] = radeon_get_ib_value(p, idx);
+ if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO) {
+ ib[idx] |= S_0280A0_ARRAY_MODE(V_0280A0_ARRAY_2D_TILED_THIN1);
+ track->cb_color_info[tmp] |= S_0280A0_ARRAY_MODE(V_0280A0_ARRAY_2D_TILED_THIN1);
+ } else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO) {
+ ib[idx] |= S_0280A0_ARRAY_MODE(V_0280A0_ARRAY_1D_TILED_THIN1);
+ track->cb_color_info[tmp] |= S_0280A0_ARRAY_MODE(V_0280A0_ARRAY_1D_TILED_THIN1);
+ }
+ } else {
+ tmp = (reg - R_0280A0_CB_COLOR0_INFO) / 4;
+ track->cb_color_info[tmp] = radeon_get_ib_value(p, idx);
+ }
break;
case R_028060_CB_COLOR0_SIZE:
case R_028064_CB_COLOR1_SIZE:
@@ -796,8 +969,6 @@ static inline int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx
return -EINVAL;
}
ib[idx] = track->cb_color_base_last[tmp];
- printk_once(KERN_WARNING "You have old & broken userspace "
- "please consider updating mesa & xf86-video-ati\n");
track->cb_color_frag_bo[tmp] = track->cb_color_bo[tmp];
} else {
r = r600_cs_packet_next_reloc(p, &reloc);
@@ -824,8 +995,6 @@ static inline int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx
return -EINVAL;
}
ib[idx] = track->cb_color_base_last[tmp];
- printk_once(KERN_WARNING "You have old & broken userspace "
- "please consider updating mesa & xf86-video-ati\n");
track->cb_color_tile_bo[tmp] = track->cb_color_bo[tmp];
} else {
r = r600_cs_packet_next_reloc(p, &reloc);
@@ -852,10 +1021,11 @@ static inline int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx
return -EINVAL;
}
tmp = (reg - CB_COLOR0_BASE) / 4;
- track->cb_color_bo_offset[tmp] = radeon_get_ib_value(p, idx);
+ track->cb_color_bo_offset[tmp] = radeon_get_ib_value(p, idx) << 8;
ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
track->cb_color_base_last[tmp] = ib[idx];
track->cb_color_bo[tmp] = reloc->robj;
+ track->cb_color_bo_mc[tmp] = reloc->lobj.gpu_offset;
break;
case DB_DEPTH_BASE:
r = r600_cs_packet_next_reloc(p, &reloc);
@@ -864,9 +1034,10 @@ static inline int r600_cs_check_reg(struct radeon_cs_parser *p, u32 reg, u32 idx
"0x%04X\n", reg);
return -EINVAL;
}
- track->db_offset = radeon_get_ib_value(p, idx);
+ track->db_offset = radeon_get_ib_value(p, idx) << 8;
ib[idx] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
track->db_bo = reloc->robj;
+ track->db_bo_mc = reloc->lobj.gpu_offset;
break;
case DB_HTILE_DATA_BASE:
case SQ_PGM_START_FS:
@@ -946,8 +1117,9 @@ static inline unsigned minify(unsigned size, unsigned levels)
}
static void r600_texture_size(unsigned nfaces, unsigned blevel, unsigned nlevels,
- unsigned w0, unsigned h0, unsigned d0, unsigned bpe,
- unsigned *l0_size, unsigned *mipmap_size)
+ unsigned w0, unsigned h0, unsigned d0, unsigned bpe,
+ unsigned pitch_align,
+ unsigned *l0_size, unsigned *mipmap_size)
{
unsigned offset, i, level, face;
unsigned width, height, depth, rowstride, size;
@@ -960,18 +1132,18 @@ static void r600_texture_size(unsigned nfaces, unsigned blevel, unsigned nlevels
height = minify(h0, i);
depth = minify(d0, i);
for(face = 0; face < nfaces; face++) {
- rowstride = ((width * bpe) + 255) & ~255;
+ rowstride = ALIGN((width * bpe), pitch_align);
size = height * rowstride * depth;
offset += size;
offset = (offset + 0x1f) & ~0x1f;
}
}
- *l0_size = (((w0 * bpe) + 255) & ~255) * h0 * d0;
+ *l0_size = ALIGN((w0 * bpe), pitch_align) * h0 * d0;
*mipmap_size = offset;
- if (!blevel)
- *mipmap_size -= *l0_size;
if (!nlevels)
*mipmap_size = *l0_size;
+ if (!blevel)
+ *mipmap_size -= *l0_size;
}
/**
@@ -985,16 +1157,32 @@ static void r600_texture_size(unsigned nfaces, unsigned blevel, unsigned nlevels
* the texture and mipmap bo object are big enough to cover this resource.
*/
static inline int r600_check_texture_resource(struct radeon_cs_parser *p, u32 idx,
- struct radeon_bo *texture,
- struct radeon_bo *mipmap)
+ struct radeon_bo *texture,
+ struct radeon_bo *mipmap,
+ u64 base_offset,
+ u64 mip_offset,
+ u32 tiling_flags)
{
+ struct r600_cs_track *track = p->track;
u32 nfaces, nlevels, blevel, w0, h0, d0, bpe = 0;
u32 word0, word1, l0_size, mipmap_size;
+ u32 height_align, pitch, pitch_align, depth_align;
+ u64 base_align;
+ struct array_mode_checker array_check;
/* on legacy kernel we don't perform advanced check */
if (p->rdev == NULL)
return 0;
+
+ /* convert to bytes */
+ base_offset <<= 8;
+ mip_offset <<= 8;
+
word0 = radeon_get_ib_value(p, idx + 0);
+ if (tiling_flags & RADEON_TILING_MACRO)
+ word0 |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
+ else if (tiling_flags & RADEON_TILING_MICRO)
+ word0 |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
word1 = radeon_get_ib_value(p, idx + 1);
w0 = G_038000_TEX_WIDTH(word0) + 1;
h0 = G_038004_TEX_HEIGHT(word1) + 1;
@@ -1021,24 +1209,59 @@ static inline int r600_check_texture_resource(struct radeon_cs_parser *p, u32 i
__func__, __LINE__, G_038004_DATA_FORMAT(word1));
return -EINVAL;
}
+
+ /* pitch in texels */
+ pitch = (G_038000_PITCH(word0) + 1) * 8;
+ array_check.array_mode = G_038000_TILE_MODE(word0);
+ array_check.group_size = track->group_size;
+ array_check.nbanks = track->nbanks;
+ array_check.npipes = track->npipes;
+ array_check.nsamples = 1;
+ array_check.bpe = bpe;
+ if (r600_get_array_mode_alignment(&array_check,
+ &pitch_align, &height_align, &depth_align, &base_align)) {
+ dev_warn(p->dev, "%s:%d tex array mode (%d) invalid\n",
+ __func__, __LINE__, G_038000_TILE_MODE(word0));
+ return -EINVAL;
+ }
+
+ /* XXX check height as well... */
+
+ if (!IS_ALIGNED(pitch, pitch_align)) {
+ dev_warn(p->dev, "%s:%d tex pitch (%d) invalid\n",
+ __func__, __LINE__, pitch);
+ return -EINVAL;
+ }
+ if (!IS_ALIGNED(base_offset, base_align)) {
+ dev_warn(p->dev, "%s:%d tex base offset (0x%llx) invalid\n",
+ __func__, __LINE__, base_offset);
+ return -EINVAL;
+ }
+ if (!IS_ALIGNED(mip_offset, base_align)) {
+ dev_warn(p->dev, "%s:%d tex mip offset (0x%llx) invalid\n",
+ __func__, __LINE__, mip_offset);
+ return -EINVAL;
+ }
+
word0 = radeon_get_ib_value(p, idx + 4);
word1 = radeon_get_ib_value(p, idx + 5);
blevel = G_038010_BASE_LEVEL(word0);
nlevels = G_038014_LAST_LEVEL(word1);
- r600_texture_size(nfaces, blevel, nlevels, w0, h0, d0, bpe, &l0_size, &mipmap_size);
+ r600_texture_size(nfaces, blevel, nlevels, w0, h0, d0, bpe,
+ (pitch_align * bpe),
+ &l0_size, &mipmap_size);
/* using get ib will give us the offset into the texture bo */
- word0 = radeon_get_ib_value(p, idx + 2);
+ word0 = radeon_get_ib_value(p, idx + 2) << 8;
if ((l0_size + word0) > radeon_bo_size(texture)) {
dev_warn(p->dev, "texture bo too small (%d %d %d %d -> %d have %ld)\n",
w0, h0, bpe, word0, l0_size, radeon_bo_size(texture));
return -EINVAL;
}
/* using get ib will give us the offset into the mipmap bo */
- word0 = radeon_get_ib_value(p, idx + 3);
+ word0 = radeon_get_ib_value(p, idx + 3) << 8;
if ((mipmap_size + word0) > radeon_bo_size(mipmap)) {
- dev_warn(p->dev, "mipmap bo too small (%d %d %d %d %d %d -> %d have %ld)\n",
- w0, h0, bpe, blevel, nlevels, word0, mipmap_size, radeon_bo_size(texture));
- return -EINVAL;
+ /*dev_warn(p->dev, "mipmap bo too small (%d %d %d %d %d %d -> %d have %ld)\n",
+ w0, h0, bpe, blevel, nlevels, word0, mipmap_size, radeon_bo_size(texture));*/
}
return 0;
}
@@ -1228,7 +1451,7 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
}
for (i = 0; i < (pkt->count / 7); i++) {
struct radeon_bo *texture, *mipmap;
- u32 size, offset;
+ u32 size, offset, base_offset, mip_offset;
switch (G__SQ_VTX_CONSTANT_TYPE(radeon_get_ib_value(p, idx+(i*7)+6+1))) {
case SQ_TEX_VTX_VALID_TEXTURE:
@@ -1238,7 +1461,11 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
DRM_ERROR("bad SET_RESOURCE\n");
return -EINVAL;
}
- ib[idx+1+(i*7)+2] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
+ base_offset = (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
+ if (reloc->lobj.tiling_flags & RADEON_TILING_MACRO)
+ ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_2D_TILED_THIN1);
+ else if (reloc->lobj.tiling_flags & RADEON_TILING_MICRO)
+ ib[idx+1+(i*7)+0] |= S_038000_TILE_MODE(V_038000_ARRAY_1D_TILED_THIN1);
texture = reloc->robj;
/* tex mip base */
r = r600_cs_packet_next_reloc(p, &reloc);
@@ -1246,12 +1473,17 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
DRM_ERROR("bad SET_RESOURCE\n");
return -EINVAL;
}
- ib[idx+1+(i*7)+3] += (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
+ mip_offset = (u32)((reloc->lobj.gpu_offset >> 8) & 0xffffffff);
mipmap = reloc->robj;
r = r600_check_texture_resource(p, idx+(i*7)+1,
- texture, mipmap);
+ texture, mipmap,
+ base_offset + radeon_get_ib_value(p, idx+1+(i*7)+2),
+ mip_offset + radeon_get_ib_value(p, idx+1+(i*7)+3),
+ reloc->lobj.tiling_flags);
if (r)
return r;
+ ib[idx+1+(i*7)+2] += base_offset;
+ ib[idx+1+(i*7)+3] += mip_offset;
break;
case SQ_TEX_VTX_VALID_BUFFER:
/* vtx base */
@@ -1261,10 +1493,11 @@ static int r600_packet3_check(struct radeon_cs_parser *p,
return -EINVAL;
}
offset = radeon_get_ib_value(p, idx+1+(i*7)+0);
- size = radeon_get_ib_value(p, idx+1+(i*7)+1);
+ size = radeon_get_ib_value(p, idx+1+(i*7)+1) + 1;
if (p->rdev && (size + offset) > radeon_bo_size(reloc->robj)) {
/* force size to size of the buffer */
- dev_warn(p->dev, "vbo resource seems too big for the bo\n");
+ dev_warn(p->dev, "vbo resource seems too big (%d) for the bo (%ld)\n",
+ size + offset, radeon_bo_size(reloc->robj));
ib[idx+1+(i*7)+1] = radeon_bo_size(reloc->robj);
}
ib[idx+1+(i*7)+0] += (u32)((reloc->lobj.gpu_offset) & 0xffffffff);
diff --git a/drivers/gpu/drm/radeon/r600d.h b/drivers/gpu/drm/radeon/r600d.h
index 84bc28e..9577945 100644
--- a/drivers/gpu/drm/radeon/r600d.h
+++ b/drivers/gpu/drm/radeon/r600d.h
@@ -51,6 +51,12 @@
#define PTE_READABLE (1 << 5)
#define PTE_WRITEABLE (1 << 6)
+/* tiling bits */
+#define ARRAY_LINEAR_GENERAL 0x00000000
+#define ARRAY_LINEAR_ALIGNED 0x00000001
+#define ARRAY_1D_TILED_THIN1 0x00000002
+#define ARRAY_2D_TILED_THIN1 0x00000004
+
/* Registers */
#define ARB_POP 0x2418
#define ENABLE_TC128 (1 << 30)
@@ -1155,6 +1161,10 @@
#define S_038000_TILE_MODE(x) (((x) & 0xF) << 3)
#define G_038000_TILE_MODE(x) (((x) >> 3) & 0xF)
#define C_038000_TILE_MODE 0xFFFFFF87
+#define V_038000_ARRAY_LINEAR_GENERAL 0x00000000
+#define V_038000_ARRAY_LINEAR_ALIGNED 0x00000001
+#define V_038000_ARRAY_1D_TILED_THIN1 0x00000002
+#define V_038000_ARRAY_2D_TILED_THIN1 0x00000004
#define S_038000_TILE_TYPE(x) (((x) & 0x1) << 7)
#define G_038000_TILE_TYPE(x) (((x) >> 7) & 0x1)
#define C_038000_TILE_TYPE 0xFFFFFF7F
@@ -1358,6 +1368,8 @@
#define S_028010_ARRAY_MODE(x) (((x) & 0xF) << 15)
#define G_028010_ARRAY_MODE(x) (((x) >> 15) & 0xF)
#define C_028010_ARRAY_MODE 0xFFF87FFF
+#define V_028010_ARRAY_1D_TILED_THIN1 0x00000002
+#define V_028010_ARRAY_2D_TILED_THIN1 0x00000004
#define S_028010_TILE_SURFACE_ENABLE(x) (((x) & 0x1) << 25)
#define G_028010_TILE_SURFACE_ENABLE(x) (((x) >> 25) & 0x1)
#define C_028010_TILE_SURFACE_ENABLE 0xFDFFFFFF

View File

@ -1,958 +0,0 @@
From 5b904034b0ab5195d971b139d0c0b67ab21b063c Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@dreadnought.i.jkkm.org>
Date: Mon, 21 Jun 2010 20:33:16 +0100
Subject: Revert "drm/fbdev: rework output polling to be back in the core. (v4)"
This reverts commit eb1f8e4f3be898df808e2dfc131099f5831d491d.
Conflicts:
drivers/gpu/drm/drm_crtc_helper.c
drivers/gpu/drm/i915/i915_dma.c
drivers/gpu/drm/i915/intel_fb.c
drivers/gpu/drm/nouveau/nouveau_fbcon.c
drivers/gpu/drm/radeon/radeon_fb.c
include/drm/drm_crtc_helper.h
---
drivers/gpu/drm/Kconfig | 2 +-
drivers/gpu/drm/drm_crtc_helper.c | 111 ------------------------
drivers/gpu/drm/drm_fb_helper.c | 123 +++++++++++++++++++++++----
drivers/gpu/drm/i915/i915_dma.c | 1 -
drivers/gpu/drm/i915/i915_irq.c | 3 +-
drivers/gpu/drm/i915/intel_crt.c | 5 -
drivers/gpu/drm/i915/intel_display.c | 2 -
drivers/gpu/drm/i915/intel_dp.c | 2 -
drivers/gpu/drm/i915/intel_drv.h | 2 +-
drivers/gpu/drm/i915/intel_fb.c | 14 ++--
drivers/gpu/drm/i915/intel_hdmi.c | 1 -
drivers/gpu/drm/i915/intel_sdvo.c | 2 -
drivers/gpu/drm/nouveau/nouveau_connector.c | 12 ---
drivers/gpu/drm/nouveau/nouveau_display.c | 1 -
drivers/gpu/drm/nouveau/nouveau_fbcon.c | 13 ++-
drivers/gpu/drm/nouveau/nouveau_fbcon.h | 2 +-
drivers/gpu/drm/nouveau/nouveau_state.c | 5 +-
drivers/gpu/drm/nouveau/nv50_display.c | 2 +-
drivers/gpu/drm/radeon/radeon_connectors.c | 13 ---
drivers/gpu/drm/radeon/radeon_display.c | 10 --
drivers/gpu/drm/radeon/radeon_fb.c | 15 +++-
drivers/gpu/drm/radeon/radeon_irq_kms.c | 5 +-
drivers/gpu/drm/radeon/radeon_mode.h | 3 +-
include/drm/drm_crtc.h | 17 ----
include/drm/drm_crtc_helper.h | 6 --
include/drm/drm_fb_helper.h | 13 +++-
26 files changed, 155 insertions(+), 230 deletions(-)
diff --git a/drivers/gpu/drm/Kconfig b/drivers/gpu/drm/Kconfig
index c2711c6..a51a1e4 100644
--- a/drivers/gpu/drm/Kconfig
+++ b/drivers/gpu/drm/Kconfig
@@ -9,7 +9,6 @@ menuconfig DRM
depends on (AGP || AGP=n) && PCI && !EMULATED_CMPXCHG && MMU
select I2C
select I2C_ALGOBIT
- select SLOW_WORK
help
Kernel-level support for the Direct Rendering Infrastructure (DRI)
introduced in XFree86 4.0. If you say Y here, you need to select
@@ -24,6 +23,7 @@ config DRM_KMS_HELPER
depends on DRM
select FB
select FRAMEBUFFER_CONSOLE if !EMBEDDED
+ select SLOW_WORK
help
FB and CRTC helpers for KMS drivers.
diff --git a/drivers/gpu/drm/drm_crtc_helper.c b/drivers/gpu/drm/drm_crtc_helper.c
index 9b2a541..b142ac2 100644
--- a/drivers/gpu/drm/drm_crtc_helper.c
+++ b/drivers/gpu/drm/drm_crtc_helper.c
@@ -807,114 +807,3 @@ int drm_helper_resume_force_mode(struct drm_device *dev)
return 0;
}
EXPORT_SYMBOL(drm_helper_resume_force_mode);
-
-static struct slow_work_ops output_poll_ops;
-
-#define DRM_OUTPUT_POLL_PERIOD (10*HZ)
-static void output_poll_execute(struct slow_work *work)
-{
- struct delayed_slow_work *delayed_work = container_of(work, struct delayed_slow_work, work);
- struct drm_device *dev = container_of(delayed_work, struct drm_device, mode_config.output_poll_slow_work);
- struct drm_connector *connector;
- enum drm_connector_status old_status, status;
- bool repoll = false, changed = false;
- int ret;
-
- mutex_lock(&dev->mode_config.mutex);
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
-
- /* if this is HPD or polled don't check it -
- TV out for instance */
- if (!connector->polled)
- continue;
-
- else if (connector->polled & (DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT))
- repoll = true;
-
- old_status = connector->status;
- /* if we are connected and don't want to poll for disconnect
- skip it */
- if (old_status == connector_status_connected &&
- !(connector->polled & DRM_CONNECTOR_POLL_DISCONNECT) &&
- !(connector->polled & DRM_CONNECTOR_POLL_HPD))
- continue;
-
- status = connector->funcs->detect(connector);
- if (old_status != status)
- changed = true;
- }
-
- mutex_unlock(&dev->mode_config.mutex);
-
- if (changed) {
- /* send a uevent + call fbdev */
- drm_sysfs_hotplug_event(dev);
- if (dev->mode_config.funcs->output_poll_changed)
- dev->mode_config.funcs->output_poll_changed(dev);
- }
-
- if (repoll) {
- ret = delayed_slow_work_enqueue(delayed_work, DRM_OUTPUT_POLL_PERIOD);
- if (ret)
- DRM_ERROR("delayed enqueue failed %d\n", ret);
- }
-}
-
-void drm_kms_helper_poll_disable(struct drm_device *dev)
-{
- if (!dev->mode_config.poll_enabled)
- return;
- delayed_slow_work_cancel(&dev->mode_config.output_poll_slow_work);
-}
-EXPORT_SYMBOL(drm_kms_helper_poll_disable);
-
-void drm_kms_helper_poll_enable(struct drm_device *dev)
-{
- bool poll = false;
- struct drm_connector *connector;
- int ret;
-
- list_for_each_entry(connector, &dev->mode_config.connector_list, head) {
- if (connector->polled)
- poll = true;
- }
-
- if (poll) {
- ret = delayed_slow_work_enqueue(&dev->mode_config.output_poll_slow_work, DRM_OUTPUT_POLL_PERIOD);
- if (ret)
- DRM_ERROR("delayed enqueue failed %d\n", ret);
- }
-}
-EXPORT_SYMBOL(drm_kms_helper_poll_enable);
-
-void drm_kms_helper_poll_init(struct drm_device *dev)
-{
- slow_work_register_user(THIS_MODULE);
- delayed_slow_work_init(&dev->mode_config.output_poll_slow_work,
- &output_poll_ops);
- dev->mode_config.poll_enabled = true;
-
- drm_kms_helper_poll_enable(dev);
-}
-EXPORT_SYMBOL(drm_kms_helper_poll_init);
-
-void drm_kms_helper_poll_fini(struct drm_device *dev)
-{
- drm_kms_helper_poll_disable(dev);
- slow_work_unregister_user(THIS_MODULE);
-}
-EXPORT_SYMBOL(drm_kms_helper_poll_fini);
-
-void drm_helper_hpd_irq_event(struct drm_device *dev)
-{
- if (!dev->mode_config.poll_enabled)
- return;
- delayed_slow_work_cancel(&dev->mode_config.output_poll_slow_work);
- /* schedule a slow work asap */
- delayed_slow_work_enqueue(&dev->mode_config.output_poll_slow_work, 0);
-}
-EXPORT_SYMBOL(drm_helper_hpd_irq_event);
-
-static struct slow_work_ops output_poll_ops = {
- .execute = output_poll_execute,
-};
diff --git a/drivers/gpu/drm/drm_fb_helper.c b/drivers/gpu/drm/drm_fb_helper.c
index 08c4c92..dcc6601 100644
--- a/drivers/gpu/drm/drm_fb_helper.c
+++ b/drivers/gpu/drm/drm_fb_helper.c
@@ -42,6 +42,8 @@ MODULE_LICENSE("GPL and additional rights");
static LIST_HEAD(kernel_fb_helper_list);
+static struct slow_work_ops output_status_change_ops;
+
/* simple single crtc case helper function */
int drm_fb_helper_single_add_all_connectors(struct drm_fb_helper *fb_helper)
{
@@ -423,13 +425,19 @@ static void drm_fb_helper_crtc_free(struct drm_fb_helper *helper)
int drm_fb_helper_init(struct drm_device *dev,
struct drm_fb_helper *fb_helper,
- int crtc_count, int max_conn_count)
+ int crtc_count, int max_conn_count,
+ bool polled)
{
struct drm_crtc *crtc;
int ret = 0;
int i;
fb_helper->dev = dev;
+ fb_helper->poll_enabled = polled;
+
+ slow_work_register_user(THIS_MODULE);
+ delayed_slow_work_init(&fb_helper->output_status_change_slow_work,
+ &output_status_change_ops);
INIT_LIST_HEAD(&fb_helper->kernel_fb_list);
@@ -486,6 +494,8 @@ void drm_fb_helper_fini(struct drm_fb_helper *fb_helper)
drm_fb_helper_crtc_free(fb_helper);
+ delayed_slow_work_cancel(&fb_helper->output_status_change_slow_work);
+ slow_work_unregister_user(THIS_MODULE);
}
EXPORT_SYMBOL(drm_fb_helper_fini);
@@ -703,7 +713,7 @@ int drm_fb_helper_set_par(struct fb_info *info)
if (fb_helper->delayed_hotplug) {
fb_helper->delayed_hotplug = false;
- drm_fb_helper_hotplug_event(fb_helper);
+ delayed_slow_work_enqueue(&fb_helper->output_status_change_slow_work, 0);
}
return 0;
}
@@ -816,7 +826,7 @@ int drm_fb_helper_single_fb_probe(struct drm_fb_helper *fb_helper,
if (crtc_count == 0 || sizes.fb_width == -1 || sizes.fb_height == -1) {
/* hmm everyone went away - assume VGA cable just fell out
and will come back later. */
- DRM_INFO("Cannot find any crtc or sizes - going 1024x768\n");
+ DRM_ERROR("Cannot find any crtc or sizes - going 1024x768\n");
sizes.fb_width = sizes.surface_width = 1024;
sizes.fb_height = sizes.surface_height = 768;
}
@@ -1362,7 +1372,12 @@ bool drm_fb_helper_initial_config(struct drm_fb_helper *fb_helper, int bpp_sel)
* we shouldn't end up with no modes here.
*/
if (count == 0) {
- printk(KERN_INFO "No connectors reported connected with modes\n");
+ if (fb_helper->poll_enabled) {
+ delayed_slow_work_enqueue(&fb_helper->output_status_change_slow_work,
+ 5*HZ);
+ printk(KERN_INFO "No connectors reported connected with modes - started polling\n");
+ } else
+ printk(KERN_INFO "No connectors reported connected with modes\n");
}
drm_setup_crtcs(fb_helper);
@@ -1370,16 +1385,71 @@ bool drm_fb_helper_initial_config(struct drm_fb_helper *fb_helper, int bpp_sel)
}
EXPORT_SYMBOL(drm_fb_helper_initial_config);
-bool drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper)
+/* we got a hotplug irq - need to update fbcon */
+void drm_helper_fb_hpd_irq_event(struct drm_fb_helper *fb_helper)
+{
+ /* if we don't have the fbdev registered yet do nothing */
+ if (!fb_helper->fbdev)
+ return;
+
+ /* schedule a slow work asap */
+ delayed_slow_work_enqueue(&fb_helper->output_status_change_slow_work, 0);
+}
+EXPORT_SYMBOL(drm_helper_fb_hpd_irq_event);
+
+bool drm_helper_fb_hotplug_event(struct drm_fb_helper *fb_helper, bool polled)
{
int count = 0;
+ int ret;
u32 max_width, max_height, bpp_sel;
- bool bound = false, crtcs_bound = false;
- struct drm_crtc *crtc;
if (!fb_helper->fb)
return false;
+ DRM_DEBUG_KMS("\n");
+
+ max_width = fb_helper->fb->width;
+ max_height = fb_helper->fb->height;
+ bpp_sel = fb_helper->fb->bits_per_pixel;
+
+ count = drm_fb_helper_probe_connector_modes(fb_helper, max_width,
+ max_height);
+ if (fb_helper->poll_enabled && !polled) {
+ if (count) {
+ delayed_slow_work_cancel(&fb_helper->output_status_change_slow_work);
+ } else {
+ ret = delayed_slow_work_enqueue(&fb_helper->output_status_change_slow_work, 5*HZ);
+ }
+ }
+ drm_setup_crtcs(fb_helper);
+
+ return drm_fb_helper_single_fb_probe(fb_helper, bpp_sel);
+}
+EXPORT_SYMBOL(drm_helper_fb_hotplug_event);
+
+/*
+ * delayed work queue execution function
+ * - check if fbdev is actually in use on the gpu
+ * - if not set delayed flag and repoll if necessary
+ * - check for connector status change
+ * - repoll if 0 modes found
+ *- call driver output status changed notifier
+ */
+static void output_status_change_execute(struct slow_work *work)
+{
+ struct delayed_slow_work *delayed_work = container_of(work, struct delayed_slow_work, work);
+ struct drm_fb_helper *fb_helper = container_of(delayed_work, struct drm_fb_helper, output_status_change_slow_work);
+ struct drm_connector *connector;
+ enum drm_connector_status old_status, status;
+ bool repoll, changed = false;
+ int ret;
+ int i;
+ bool bound = false, crtcs_bound = false;
+ struct drm_crtc *crtc;
+ repoll = fb_helper->poll_enabled;
+
+ /* first of all check the fbcon framebuffer is actually bound to any crtc */
+ /* take into account that no crtc at all maybe bound */
list_for_each_entry(crtc, &fb_helper->dev->mode_config.crtc_list, head) {
if (crtc->fb)
crtcs_bound = true;
@@ -1387,21 +1457,38 @@ bool drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper)
bound = true;
}
- if (!bound && crtcs_bound) {
+ if (bound == false && crtcs_bound) {
fb_helper->delayed_hotplug = true;
- return false;
+ goto requeue;
}
- DRM_DEBUG_KMS("\n");
- max_width = fb_helper->fb->width;
- max_height = fb_helper->fb->height;
- bpp_sel = fb_helper->fb->bits_per_pixel;
+ for (i = 0; i < fb_helper->connector_count; i++) {
+ connector = fb_helper->connector_info[i]->connector;
+ old_status = connector->status;
+ status = connector->funcs->detect(connector);
+ if (old_status != status) {
+ changed = true;
+ }
+ if (status == connector_status_connected && repoll) {
+ DRM_DEBUG("%s is connected - stop polling\n", drm_get_connector_name(connector));
+ repoll = false;
+ }
+ }
- count = drm_fb_helper_probe_connector_modes(fb_helper, max_width,
- max_height);
- drm_setup_crtcs(fb_helper);
+ if (changed) {
+ if (fb_helper->funcs->fb_output_status_changed)
+ fb_helper->funcs->fb_output_status_changed(fb_helper);
+ }
- return drm_fb_helper_single_fb_probe(fb_helper, bpp_sel);
+requeue:
+ if (repoll) {
+ ret = delayed_slow_work_enqueue(delayed_work, 5*HZ);
+ if (ret)
+ DRM_ERROR("delayed enqueue failed %d\n", ret);
+ }
}
-EXPORT_SYMBOL(drm_fb_helper_hotplug_event);
+
+static struct slow_work_ops output_status_change_ops = {
+ .execute = output_status_change_execute,
+};
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 59a2bf8..76ace2d 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1430,7 +1430,6 @@ static int i915_load_modeset_init(struct drm_device *dev,
if (ret)
goto cleanup_irq;
- drm_kms_helper_poll_init(dev);
return 0;
cleanup_irq:
diff --git a/drivers/gpu/drm/i915/i915_irq.c b/drivers/gpu/drm/i915/i915_irq.c
index 2479be0..6350bd3 100644
--- a/drivers/gpu/drm/i915/i915_irq.c
+++ b/drivers/gpu/drm/i915/i915_irq.c
@@ -271,7 +271,8 @@ static void i915_hotplug_work_func(struct work_struct *work)
}
}
/* Just fire off a uevent and let userspace tell us what to do */
- drm_helper_hpd_irq_event(dev);
+ intelfb_hotplug(dev, false);
+ drm_sysfs_hotplug_event(dev);
}
static void i915_handle_rps_change(struct drm_device *dev)
diff --git a/drivers/gpu/drm/i915/intel_crt.c b/drivers/gpu/drm/i915/intel_crt.c
index 22ff384..125eded 100644
--- a/drivers/gpu/drm/i915/intel_crt.c
+++ b/drivers/gpu/drm/i915/intel_crt.c
@@ -584,10 +584,5 @@ void intel_crt_init(struct drm_device *dev)
drm_sysfs_connector_add(connector);
- if (I915_HAS_HOTPLUG(dev))
- connector->polled = DRM_CONNECTOR_POLL_HPD;
- else
- connector->polled = DRM_CONNECTOR_POLL_CONNECT;
-
dev_priv->hotplug_supported_mask |= CRT_HOTPLUG_INT_STATUS;
}
diff --git a/drivers/gpu/drm/i915/intel_display.c b/drivers/gpu/drm/i915/intel_display.c
index d753257..70537cf 100644
--- a/drivers/gpu/drm/i915/intel_display.c
+++ b/drivers/gpu/drm/i915/intel_display.c
@@ -5036,7 +5036,6 @@ intel_user_framebuffer_create(struct drm_device *dev,
static const struct drm_mode_config_funcs intel_mode_funcs = {
.fb_create = intel_user_framebuffer_create,
- .output_poll_changed = intel_fb_output_poll_changed,
};
static struct drm_gem_object *
@@ -5538,7 +5537,6 @@ void intel_modeset_cleanup(struct drm_device *dev)
mutex_lock(&dev->struct_mutex);
- drm_kms_helper_poll_fini(dev);
intel_fbdev_fini(dev);
list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
diff --git a/drivers/gpu/drm/i915/intel_dp.c b/drivers/gpu/drm/i915/intel_dp.c
index 49b54f0..1815df5 100644
--- a/drivers/gpu/drm/i915/intel_dp.c
+++ b/drivers/gpu/drm/i915/intel_dp.c
@@ -1393,8 +1393,6 @@ intel_dp_init(struct drm_device *dev, int output_reg)
DRM_MODE_CONNECTOR_DisplayPort);
drm_connector_helper_add(connector, &intel_dp_connector_helper_funcs);
- connector->polled = DRM_CONNECTOR_POLL_HPD;
-
if (output_reg == DP_A)
intel_encoder->type = INTEL_OUTPUT_EDP;
else
diff --git a/drivers/gpu/drm/i915/intel_drv.h b/drivers/gpu/drm/i915/intel_drv.h
index df931f7..3230e8d 100644
--- a/drivers/gpu/drm/i915/intel_drv.h
+++ b/drivers/gpu/drm/i915/intel_drv.h
@@ -235,5 +235,5 @@ extern int intel_overlay_put_image(struct drm_device *dev, void *data,
extern int intel_overlay_attrs(struct drm_device *dev, void *data,
struct drm_file *file_priv);
-extern void intel_fb_output_poll_changed(struct drm_device *dev);
+void intelfb_hotplug(struct drm_device *dev, bool polled);
#endif /* __INTEL_DRV_H__ */
diff --git a/drivers/gpu/drm/i915/intel_fb.c b/drivers/gpu/drm/i915/intel_fb.c
index c3c5052..79098b3 100644
--- a/drivers/gpu/drm/i915/intel_fb.c
+++ b/drivers/gpu/drm/i915/intel_fb.c
@@ -211,6 +211,12 @@ static int intel_fb_find_or_create_single(struct drm_fb_helper *helper,
return new_fb;
}
+void intelfb_hotplug(struct drm_device *dev, bool polled)
+{
+ drm_i915_private_t *dev_priv = dev->dev_private;
+ drm_helper_fb_hpd_irq_event(&dev_priv->fbdev->helper);
+}
+
static struct drm_fb_helper_funcs intel_fb_helper_funcs = {
.gamma_set = intel_crtc_fb_gamma_set,
.gamma_get = intel_crtc_fb_gamma_get,
@@ -256,7 +262,7 @@ int intel_fbdev_init(struct drm_device *dev)
ret = drm_fb_helper_init(dev, &ifbdev->helper,
dev_priv->num_pipe,
- INTELFB_CONN_LIMIT);
+ INTELFB_CONN_LIMIT, false);
if (ret) {
kfree(ifbdev);
return ret;
@@ -278,9 +284,3 @@ void intel_fbdev_fini(struct drm_device *dev)
dev_priv->fbdev = NULL;
}
MODULE_LICENSE("GPL and additional rights");
-
-void intel_fb_output_poll_changed(struct drm_device *dev)
-{
- drm_i915_private_t *dev_priv = dev->dev_private;
- drm_fb_helper_hotplug_event(&dev_priv->fbdev->helper);
-}
diff --git a/drivers/gpu/drm/i915/intel_hdmi.c b/drivers/gpu/drm/i915/intel_hdmi.c
index 83bd764..acaca07 100644
--- a/drivers/gpu/drm/i915/intel_hdmi.c
+++ b/drivers/gpu/drm/i915/intel_hdmi.c
@@ -240,7 +240,6 @@ void intel_hdmi_init(struct drm_device *dev, int sdvox_reg)
intel_encoder->type = INTEL_OUTPUT_HDMI;
- connector->polled = DRM_CONNECTOR_POLL_HPD;
connector->interlace_allowed = 0;
connector->doublescan_allowed = 0;
intel_encoder->crtc_mask = (1 << 0) | (1 << 1);
diff --git a/drivers/gpu/drm/i915/intel_sdvo.c b/drivers/gpu/drm/i915/intel_sdvo.c
index 76993ac..1c716b5 100644
--- a/drivers/gpu/drm/i915/intel_sdvo.c
+++ b/drivers/gpu/drm/i915/intel_sdvo.c
@@ -2218,7 +2218,6 @@ intel_sdvo_dvi_init(struct intel_encoder *intel_encoder, int device)
}
connector = &intel_connector->base;
- connector->polled = DRM_CONNECTOR_POLL_CONNECT | DRM_CONNECTOR_POLL_DISCONNECT;
encoder->encoder_type = DRM_MODE_ENCODER_TMDS;
connector->connector_type = DRM_MODE_CONNECTOR_DVID;
@@ -2285,7 +2284,6 @@ intel_sdvo_analog_init(struct intel_encoder *intel_encoder, int device)
return false;
connector = &intel_connector->base;
- connector->polled = DRM_CONNECTOR_POLL_CONNECT;
encoder->encoder_type = DRM_MODE_ENCODER_DAC;
connector->connector_type = DRM_MODE_CONNECTOR_VGA;
sdvo_connector = intel_connector->dev_priv;
diff --git a/drivers/gpu/drm/nouveau/nouveau_connector.c b/drivers/gpu/drm/nouveau/nouveau_connector.c
index 149ed22..9a61f3c 100644
--- a/drivers/gpu/drm/nouveau/nouveau_connector.c
+++ b/drivers/gpu/drm/nouveau/nouveau_connector.c
@@ -846,7 +846,6 @@ nouveau_connector_create(struct drm_device *dev,
switch (dcb->type) {
case DCB_CONNECTOR_VGA:
- connector->polled = DRM_CONNECTOR_POLL_CONNECT;
if (dev_priv->card_type >= NV_50) {
drm_connector_attach_property(connector,
dev->mode_config.scaling_mode_property,
@@ -858,17 +857,6 @@ nouveau_connector_create(struct drm_device *dev,
case DCB_CONNECTOR_TV_3:
nv_connector->scaling_mode = DRM_MODE_SCALE_NONE;
break;
- case DCB_CONNECTOR_DP:
- case DCB_CONNECTOR_eDP:
- case DCB_CONNECTOR_HDMI_0:
- case DCB_CONNECTOR_HDMI_1:
- case DCB_CONNECTOR_DVI_I:
- case DCB_CONNECTOR_DVI_D:
- if (dev_priv->card_type >= NV_50)
- connector->polled = DRM_CONNECTOR_POLL_HPD;
- else
- connector->polled = DRM_CONNECTOR_POLL_CONNECT;
- /* fall-through */
default:
nv_connector->scaling_mode = DRM_MODE_SCALE_FULLSCREEN;
diff --git a/drivers/gpu/drm/nouveau/nouveau_display.c b/drivers/gpu/drm/nouveau/nouveau_display.c
index 74e6b4e..9d7928f 100644
--- a/drivers/gpu/drm/nouveau/nouveau_display.c
+++ b/drivers/gpu/drm/nouveau/nouveau_display.c
@@ -101,6 +101,5 @@ nouveau_user_framebuffer_create(struct drm_device *dev,
const struct drm_mode_config_funcs nouveau_mode_config_funcs = {
.fb_create = nouveau_user_framebuffer_create,
- .output_poll_changed = nouveau_fbcon_output_poll_changed,
};
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.c b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
index c9a4a0d..0a59f96 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.c
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.c
@@ -326,11 +326,15 @@ nouveau_fbcon_find_or_create_single(struct drm_fb_helper *helper,
return new_fb;
}
-void
-nouveau_fbcon_output_poll_changed(struct drm_device *dev)
+void nouveau_fbcon_hotplug(struct drm_device *dev)
{
struct drm_nouveau_private *dev_priv = dev->dev_private;
- drm_fb_helper_hotplug_event(&dev_priv->nfbdev->helper);
+ drm_helper_fb_hpd_irq_event(&dev_priv->nfbdev->helper);
+}
+
+static void nouveau_fbcon_output_status_changed(struct drm_fb_helper *fb_helper)
+{
+ drm_helper_fb_hotplug_event(fb_helper, true);
}
int
@@ -370,6 +374,7 @@ static struct drm_fb_helper_funcs nouveau_fbcon_helper_funcs = {
.gamma_set = nouveau_fbcon_gamma_set,
.gamma_get = nouveau_fbcon_gamma_get,
.fb_probe = nouveau_fbcon_find_or_create_single,
+ .fb_output_status_changed = nouveau_fbcon_output_status_changed,
};
@@ -387,7 +392,7 @@ int nouveau_fbcon_init(struct drm_device *dev)
dev_priv->nfbdev = nfbdev;
nfbdev->helper.funcs = &nouveau_fbcon_helper_funcs;
- ret = drm_fb_helper_init(dev, &nfbdev->helper, 2, 4);
+ ret = drm_fb_helper_init(dev, &nfbdev->helper, 2, 4, true);
if (ret) {
kfree(nfbdev);
return ret;
diff --git a/drivers/gpu/drm/nouveau/nouveau_fbcon.h b/drivers/gpu/drm/nouveau/nouveau_fbcon.h
index e7e1268..bf8e00d 100644
--- a/drivers/gpu/drm/nouveau/nouveau_fbcon.h
+++ b/drivers/gpu/drm/nouveau/nouveau_fbcon.h
@@ -58,6 +58,6 @@ void nouveau_fbcon_zfill_all(struct drm_device *dev);
void nouveau_fbcon_save_disable_accel(struct drm_device *dev);
void nouveau_fbcon_restore_accel(struct drm_device *dev);
-void nouveau_fbcon_output_poll_changed(struct drm_device *dev);
+void nouveau_fbcon_hotplug(struct drm_device *dev);
#endif /* __NV50_FBCON_H__ */
diff --git a/drivers/gpu/drm/nouveau/nouveau_state.c b/drivers/gpu/drm/nouveau/nouveau_state.c
index b02a231..4dcb976 100644
--- a/drivers/gpu/drm/nouveau/nouveau_state.c
+++ b/drivers/gpu/drm/nouveau/nouveau_state.c
@@ -519,10 +519,8 @@ nouveau_card_init(struct drm_device *dev)
dev_priv->init_state = NOUVEAU_CARD_INIT_DONE;
- if (drm_core_check_feature(dev, DRIVER_MODESET)) {
+ if (drm_core_check_feature(dev, DRIVER_MODESET))
nouveau_fbcon_init(dev);
- drm_kms_helper_poll_init(dev);
- }
return 0;
@@ -844,7 +842,6 @@ int nouveau_unload(struct drm_device *dev)
struct drm_nouveau_private *dev_priv = dev->dev_private;
if (drm_core_check_feature(dev, DRIVER_MODESET)) {
- drm_kms_helper_poll_fini(dev);
nouveau_fbcon_fini(dev);
if (dev_priv->card_type >= NV_50)
nv50_display_destroy(dev);
diff --git a/drivers/gpu/drm/nouveau/nv50_display.c b/drivers/gpu/drm/nouveau/nv50_display.c
index 580a5d1..e6a44af 100644
--- a/drivers/gpu/drm/nouveau/nv50_display.c
+++ b/drivers/gpu/drm/nouveau/nv50_display.c
@@ -980,7 +980,7 @@ nv50_display_irq_hotplug_bh(struct work_struct *work)
if (dev_priv->chipset >= 0x90)
nv_wr32(dev, 0xe074, nv_rd32(dev, 0xe074));
- drm_helper_hpd_irq_event(dev);
+ nouveau_fbcon_hotplug(dev);
}
void
diff --git a/drivers/gpu/drm/radeon/radeon_connectors.c b/drivers/gpu/drm/radeon/radeon_connectors.c
index 0c7ccc6..40a24c9 100644
--- a/drivers/gpu/drm/radeon/radeon_connectors.c
+++ b/drivers/gpu/drm/radeon/radeon_connectors.c
@@ -1085,7 +1085,6 @@ radeon_add_atom_connector(struct drm_device *dev,
drm_connector_attach_property(&radeon_connector->base,
rdev->mode_info.load_detect_property,
1);
- connector->polled = DRM_CONNECTOR_POLL_CONNECT;
break;
case DRM_MODE_CONNECTOR_DVIA:
drm_connector_init(dev, &radeon_connector->base, &radeon_vga_connector_funcs, connector_type);
@@ -1212,12 +1211,6 @@ radeon_add_atom_connector(struct drm_device *dev,
break;
}
- if (hpd->hpd == RADEON_HPD_NONE) {
- if (i2c_bus->valid)
- connector->polled = DRM_CONNECTOR_POLL_CONNECT;
- } else
- connector->polled = DRM_CONNECTOR_POLL_HPD;
-
connector->display_info.subpixel_order = subpixel_order;
drm_sysfs_connector_add(connector);
return;
@@ -1279,7 +1272,6 @@ radeon_add_legacy_connector(struct drm_device *dev,
drm_connector_attach_property(&radeon_connector->base,
rdev->mode_info.load_detect_property,
1);
- connector->polled = DRM_CONNECTOR_POLL_CONNECT;
break;
case DRM_MODE_CONNECTOR_DVIA:
drm_connector_init(dev, &radeon_connector->base, &radeon_vga_connector_funcs, connector_type);
@@ -1348,11 +1340,6 @@ radeon_add_legacy_connector(struct drm_device *dev,
break;
}
- if (hpd->hpd == RADEON_HPD_NONE) {
- if (i2c_bus->valid)
- connector->polled = DRM_CONNECTOR_POLL_CONNECT;
- } else
- connector->polled = DRM_CONNECTOR_POLL_HPD;
connector->display_info.subpixel_order = subpixel_order;
drm_sysfs_connector_add(connector);
return;
diff --git a/drivers/gpu/drm/radeon/radeon_display.c b/drivers/gpu/drm/radeon/radeon_display.c
index c73444a..ed756be 100644
--- a/drivers/gpu/drm/radeon/radeon_display.c
+++ b/drivers/gpu/drm/radeon/radeon_display.c
@@ -887,15 +887,8 @@ radeon_user_framebuffer_create(struct drm_device *dev,
return &radeon_fb->base;
}
-static void radeon_output_poll_changed(struct drm_device *dev)
-{
- struct radeon_device *rdev = dev->dev_private;
- radeon_fb_output_poll_changed(rdev);
-}
-
static const struct drm_mode_config_funcs radeon_mode_funcs = {
.fb_create = radeon_user_framebuffer_create,
- .output_poll_changed = radeon_output_poll_changed
};
struct drm_prop_enum_list {
@@ -1044,8 +1037,6 @@ int radeon_modeset_init(struct radeon_device *rdev)
radeon_pm_init(rdev);
radeon_fbdev_init(rdev);
- drm_kms_helper_poll_init(rdev->ddev);
-
return 0;
}
@@ -1058,7 +1049,6 @@ void radeon_modeset_fini(struct radeon_device *rdev)
radeon_pm_fini(rdev);
if (rdev->mode_info.mode_config_initialized) {
- drm_kms_helper_poll_fini(rdev->ddev);
radeon_hpd_fini(rdev);
drm_mode_config_cleanup(rdev->ddev);
rdev->mode_info.mode_config_initialized = false;
diff --git a/drivers/gpu/drm/radeon/radeon_fb.c b/drivers/gpu/drm/radeon/radeon_fb.c
index dc1634b..7dc38f6 100644
--- a/drivers/gpu/drm/radeon/radeon_fb.c
+++ b/drivers/gpu/drm/radeon/radeon_fb.c
@@ -316,9 +316,16 @@ int radeon_parse_options(char *options)
return 0;
}
-void radeon_fb_output_poll_changed(struct radeon_device *rdev)
+void radeonfb_hotplug(struct drm_device *dev, bool polled)
{
- drm_fb_helper_hotplug_event(&rdev->mode_info.rfbdev->helper);
+ struct radeon_device *rdev = dev->dev_private;
+
+ drm_helper_fb_hpd_irq_event(&rdev->mode_info.rfbdev->helper);
+}
+
+static void radeon_fb_output_status_changed(struct drm_fb_helper *fb_helper)
+{
+ drm_helper_fb_hotplug_event(fb_helper, true);
}
static int radeon_fbdev_destroy(struct drm_device *dev, struct radeon_fbdev *rfbdev)
@@ -357,6 +364,7 @@ static struct drm_fb_helper_funcs radeon_fb_helper_funcs = {
.gamma_set = radeon_crtc_fb_gamma_set,
.gamma_get = radeon_crtc_fb_gamma_get,
.fb_probe = radeon_fb_find_or_create_single,
+ .fb_output_status_changed = radeon_fb_output_status_changed,
};
int radeon_fbdev_init(struct radeon_device *rdev)
@@ -379,7 +387,7 @@ int radeon_fbdev_init(struct radeon_device *rdev)
ret = drm_fb_helper_init(rdev->ddev, &rfbdev->helper,
rdev->num_crtc,
- RADEONFB_CONN_LIMIT);
+ RADEONFB_CONN_LIMIT, true);
if (ret) {
kfree(rfbdev);
return ret;
@@ -388,6 +396,7 @@ int radeon_fbdev_init(struct radeon_device *rdev)
drm_fb_helper_single_add_all_connectors(&rfbdev->helper);
drm_fb_helper_initial_config(&rfbdev->helper, bpp_sel);
return 0;
+
}
void radeon_fbdev_fini(struct radeon_device *rdev)
diff --git a/drivers/gpu/drm/radeon/radeon_irq_kms.c b/drivers/gpu/drm/radeon/radeon_irq_kms.c
index 059bfa4..b0178de 100644
--- a/drivers/gpu/drm/radeon/radeon_irq_kms.c
+++ b/drivers/gpu/drm/radeon/radeon_irq_kms.c
@@ -26,7 +26,6 @@
* Jerome Glisse
*/
#include "drmP.h"
-#include "drm_crtc_helper.h"
#include "radeon_drm.h"
#include "radeon_reg.h"
#include "radeon.h"
@@ -56,7 +55,9 @@ static void radeon_hotplug_work_func(struct work_struct *work)
radeon_connector_hotplug(connector);
}
/* Just fire off a uevent and let userspace tell us what to do */
- drm_helper_hpd_irq_event(dev);
+ radeonfb_hotplug(dev, false);
+
+ drm_sysfs_hotplug_event(dev);
}
void radeon_driver_irq_preinstall_kms(struct drm_device *dev)
diff --git a/drivers/gpu/drm/radeon/radeon_mode.h b/drivers/gpu/drm/radeon/radeon_mode.h
index 67358ba..fdd1611 100644
--- a/drivers/gpu/drm/radeon/radeon_mode.h
+++ b/drivers/gpu/drm/radeon/radeon_mode.h
@@ -588,6 +588,5 @@ void radeon_fbdev_fini(struct radeon_device *rdev);
void radeon_fbdev_set_suspend(struct radeon_device *rdev, int state);
int radeon_fbdev_total_size(struct radeon_device *rdev);
bool radeon_fbdev_robj_is_fb(struct radeon_device *rdev, struct radeon_bo *robj);
-
-void radeon_fb_output_poll_changed(struct radeon_device *rdev);
+void radeonfb_hotplug(struct drm_device *dev, bool polled);
#endif
diff --git a/include/drm/drm_crtc.h b/include/drm/drm_crtc.h
index 93a1a31..a7148d2 100644
--- a/include/drm/drm_crtc.h
+++ b/include/drm/drm_crtc.h
@@ -31,7 +31,6 @@
#include <linux/idr.h>
#include <linux/fb.h>
-#include <linux/slow-work.h>
struct drm_device;
struct drm_mode_set;
@@ -461,15 +460,6 @@ enum drm_connector_force {
DRM_FORCE_ON_DIGITAL, /* for DVI-I use digital connector */
};
-/* should we poll this connector for connects and disconnects */
-/* hot plug detectable */
-#define DRM_CONNECTOR_POLL_HPD (1 << 0)
-/* poll for connections */
-#define DRM_CONNECTOR_POLL_CONNECT (1 << 1)
-/* can cleanly poll for disconnections without flickering the screen */
-/* DACs should rarely do this without a lot of testing */
-#define DRM_CONNECTOR_POLL_DISCONNECT (1 << 2)
-
/**
* drm_connector - central DRM connector control structure
* @crtc: CRTC this connector is currently connected to, NULL if none
@@ -514,8 +504,6 @@ struct drm_connector {
u32 property_ids[DRM_CONNECTOR_MAX_PROPERTY];
uint64_t property_values[DRM_CONNECTOR_MAX_PROPERTY];
- uint8_t polled; /* DRM_CONNECTOR_POLL_* */
-
/* requested DPMS state */
int dpms;
@@ -555,7 +543,6 @@ struct drm_mode_set {
*/
struct drm_mode_config_funcs {
struct drm_framebuffer *(*fb_create)(struct drm_device *dev, struct drm_file *file_priv, struct drm_mode_fb_cmd *mode_cmd);
- void (*output_poll_changed)(struct drm_device *dev);
};
struct drm_mode_group {
@@ -593,10 +580,6 @@ struct drm_mode_config {
struct drm_mode_config_funcs *funcs;
resource_size_t fb_base;
- /* output poll support */
- bool poll_enabled;
- struct delayed_slow_work output_poll_slow_work;
-
/* pointers to standard properties */
struct list_head property_blob_list;
struct drm_property *edid_property;
diff --git a/include/drm/drm_crtc_helper.h b/include/drm/drm_crtc_helper.h
index 1121f77..b1fa0f8 100644
--- a/include/drm/drm_crtc_helper.h
+++ b/include/drm/drm_crtc_helper.h
@@ -127,10 +127,4 @@ static inline void drm_connector_helper_add(struct drm_connector *connector,
}
extern int drm_helper_resume_force_mode(struct drm_device *dev);
-extern void drm_kms_helper_poll_init(struct drm_device *dev);
-extern void drm_kms_helper_poll_fini(struct drm_device *dev);
-extern void drm_helper_hpd_irq_event(struct drm_device *dev);
-
-extern void drm_kms_helper_poll_disable(struct drm_device *dev);
-extern void drm_kms_helper_poll_enable(struct drm_device *dev);
#endif
diff --git a/include/drm/drm_fb_helper.h b/include/drm/drm_fb_helper.h
index f0a6afc..9b55a94 100644
--- a/include/drm/drm_fb_helper.h
+++ b/include/drm/drm_fb_helper.h
@@ -30,6 +30,8 @@
#ifndef DRM_FB_HELPER_H
#define DRM_FB_HELPER_H
+#include <linux/slow-work.h>
+
struct drm_fb_helper;
struct drm_fb_helper_crtc {
@@ -69,6 +71,9 @@ struct drm_fb_helper_funcs {
int (*fb_probe)(struct drm_fb_helper *helper,
struct drm_fb_helper_surface_size *sizes);
+
+ void (*fb_output_status_changed)(struct drm_fb_helper *helper);
+
};
struct drm_fb_helper_connector {
@@ -90,6 +95,8 @@ struct drm_fb_helper {
u32 pseudo_palette[17];
struct list_head kernel_fb_list;
+ struct delayed_slow_work output_status_change_slow_work;
+ bool poll_enabled;
/* we got a hotplug but fbdev wasn't running the console
delay until next set_par */
bool delayed_hotplug;
@@ -100,7 +107,7 @@ int drm_fb_helper_single_fb_probe(struct drm_fb_helper *helper,
int drm_fb_helper_init(struct drm_device *dev,
struct drm_fb_helper *helper, int crtc_count,
- int max_conn);
+ int max_conn, bool polled);
void drm_fb_helper_fini(struct drm_fb_helper *helper);
int drm_fb_helper_blank(int blank, struct fb_info *info);
int drm_fb_helper_pan_display(struct fb_var_screeninfo *var,
@@ -123,8 +130,10 @@ void drm_fb_helper_fill_fix(struct fb_info *info, uint32_t pitch,
int drm_fb_helper_setcmap(struct fb_cmap *cmap, struct fb_info *info);
-bool drm_fb_helper_hotplug_event(struct drm_fb_helper *fb_helper);
+bool drm_helper_fb_hotplug_event(struct drm_fb_helper *fb_helper,
+ bool polled);
bool drm_fb_helper_initial_config(struct drm_fb_helper *fb_helper, int bpp_sel);
int drm_fb_helper_single_add_all_connectors(struct drm_fb_helper *fb_helper);
+void drm_helper_fb_hpd_irq_event(struct drm_fb_helper *fb_helper);
#endif
--
1.7.0.1

592
drm-sil164-module.patch Normal file
View File

@ -0,0 +1,592 @@
From f1719f0dcd68ca4de42c7b00ef2b37658007dda7 Mon Sep 17 00:00:00 2001
From: Francisco Jerez <currojerez@riseup.net>
Date: Thu, 22 Jul 2010 17:06:18 +0200
Subject: [PATCH 2/5] drm-sil164-module
drm: Import driver for the sil164 I2C TMDS transmitter.
sil164 transmitters are used for DVI outputs on Intel/nvidia and ATI setups.
So far only nouveau can use this driver.
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Tested-by: Patrice Mandin <patmandin@gmail.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
drivers/gpu/drm/i2c/Makefile | 3 +
drivers/gpu/drm/i2c/sil164_drv.c | 462 ++++++++++++++++++++++++++++++++++++++
drivers/gpu/drm/nouveau/Kconfig | 9 +
include/drm/i2c/sil164.h | 63 +++++
4 files changed, 537 insertions(+), 0 deletions(-)
create mode 100644 drivers/gpu/drm/i2c/sil164_drv.c
create mode 100644 include/drm/i2c/sil164.h
diff --git a/drivers/gpu/drm/i2c/Makefile b/drivers/gpu/drm/i2c/Makefile
index 6d2abaf..9286256 100644
--- a/drivers/gpu/drm/i2c/Makefile
+++ b/drivers/gpu/drm/i2c/Makefile
@@ -2,3 +2,6 @@ ccflags-y := -Iinclude/drm
ch7006-y := ch7006_drv.o ch7006_mode.o
obj-$(CONFIG_DRM_I2C_CH7006) += ch7006.o
+
+sil164-y := sil164_drv.o
+obj-$(CONFIG_DRM_I2C_SIL164) += sil164.o
diff --git a/drivers/gpu/drm/i2c/sil164_drv.c b/drivers/gpu/drm/i2c/sil164_drv.c
new file mode 100644
index 0000000..0b67732
--- /dev/null
+++ b/drivers/gpu/drm/i2c/sil164_drv.c
@@ -0,0 +1,462 @@
+/*
+ * Copyright (C) 2010 Francisco Jerez.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining
+ * a copy of this software and associated documentation files (the
+ * "Software"), to deal in the Software without restriction, including
+ * without limitation the rights to use, copy, modify, merge, publish,
+ * distribute, sublicense, and/or sell copies of the Software, and to
+ * permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the
+ * next paragraph) shall be included in all copies or substantial
+ * portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+ * IN NO EVENT SHALL THE COPYRIGHT OWNER(S) AND/OR ITS SUPPLIERS BE
+ * LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+ * OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+ * WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#include "drmP.h"
+#include "drm_crtc_helper.h"
+#include "drm_encoder_slave.h"
+#include "i2c/sil164.h"
+
+struct sil164_priv {
+ struct sil164_encoder_params config;
+ struct i2c_client *duallink_slave;
+
+ uint8_t saved_state[0x10];
+ uint8_t saved_slave_state[0x10];
+};
+
+#define to_sil164_priv(x) \
+ ((struct sil164_priv *)to_encoder_slave(x)->slave_priv)
+
+#define sil164_dbg(client, format, ...) do { \
+ if (drm_debug & DRM_UT_KMS) \
+ dev_printk(KERN_DEBUG, &client->dev, \
+ "%s: " format, __func__, ## __VA_ARGS__); \
+ } while (0)
+#define sil164_info(client, format, ...) \
+ dev_info(&client->dev, format, __VA_ARGS__)
+#define sil164_err(client, format, ...) \
+ dev_err(&client->dev, format, __VA_ARGS__)
+
+#define SIL164_I2C_ADDR_MASTER 0x38
+#define SIL164_I2C_ADDR_SLAVE 0x39
+
+/* HW register definitions */
+
+#define SIL164_VENDOR_LO 0x0
+#define SIL164_VENDOR_HI 0x1
+#define SIL164_DEVICE_LO 0x2
+#define SIL164_DEVICE_HI 0x3
+#define SIL164_REVISION 0x4
+#define SIL164_FREQ_MIN 0x6
+#define SIL164_FREQ_MAX 0x7
+#define SIL164_CONTROL0 0x8
+# define SIL164_CONTROL0_POWER_ON 0x01
+# define SIL164_CONTROL0_EDGE_RISING 0x02
+# define SIL164_CONTROL0_INPUT_24BIT 0x04
+# define SIL164_CONTROL0_DUAL_EDGE 0x08
+# define SIL164_CONTROL0_HSYNC_ON 0x10
+# define SIL164_CONTROL0_VSYNC_ON 0x20
+#define SIL164_DETECT 0x9
+# define SIL164_DETECT_INTR_STAT 0x01
+# define SIL164_DETECT_HOTPLUG_STAT 0x02
+# define SIL164_DETECT_RECEIVER_STAT 0x04
+# define SIL164_DETECT_INTR_MODE_RECEIVER 0x00
+# define SIL164_DETECT_INTR_MODE_HOTPLUG 0x08
+# define SIL164_DETECT_OUT_MODE_HIGH 0x00
+# define SIL164_DETECT_OUT_MODE_INTR 0x10
+# define SIL164_DETECT_OUT_MODE_RECEIVER 0x20
+# define SIL164_DETECT_OUT_MODE_HOTPLUG 0x30
+# define SIL164_DETECT_VSWING_STAT 0x80
+#define SIL164_CONTROL1 0xa
+# define SIL164_CONTROL1_DESKEW_ENABLE 0x10
+# define SIL164_CONTROL1_DESKEW_INCR_SHIFT 5
+#define SIL164_GPIO 0xb
+#define SIL164_CONTROL2 0xc
+# define SIL164_CONTROL2_FILTER_ENABLE 0x01
+# define SIL164_CONTROL2_FILTER_SETTING_SHIFT 1
+# define SIL164_CONTROL2_DUALLINK_MASTER 0x40
+# define SIL164_CONTROL2_SYNC_CONT 0x80
+#define SIL164_DUALLINK 0xd
+# define SIL164_DUALLINK_ENABLE 0x10
+# define SIL164_DUALLINK_SKEW_SHIFT 5
+#define SIL164_PLLZONE 0xe
+# define SIL164_PLLZONE_STAT 0x08
+# define SIL164_PLLZONE_FORCE_ON 0x10
+# define SIL164_PLLZONE_FORCE_HIGH 0x20
+
+/* HW access functions */
+
+static void
+sil164_write(struct i2c_client *client, uint8_t addr, uint8_t val)
+{
+ uint8_t buf[] = {addr, val};
+ int ret;
+
+ ret = i2c_master_send(client, buf, ARRAY_SIZE(buf));
+ if (ret < 0)
+ sil164_err(client, "Error %d writing to subaddress 0x%x\n",
+ ret, addr);
+}
+
+static uint8_t
+sil164_read(struct i2c_client *client, uint8_t addr)
+{
+ uint8_t val;
+ int ret;
+
+ ret = i2c_master_send(client, &addr, sizeof(addr));
+ if (ret < 0)
+ goto fail;
+
+ ret = i2c_master_recv(client, &val, sizeof(val));
+ if (ret < 0)
+ goto fail;
+
+ return val;
+
+fail:
+ sil164_err(client, "Error %d reading from subaddress 0x%x\n",
+ ret, addr);
+ return 0;
+}
+
+static void
+sil164_save_state(struct i2c_client *client, uint8_t *state)
+{
+ int i;
+
+ for (i = 0x8; i <= 0xe; i++)
+ state[i] = sil164_read(client, i);
+}
+
+static void
+sil164_restore_state(struct i2c_client *client, uint8_t *state)
+{
+ int i;
+
+ for (i = 0x8; i <= 0xe; i++)
+ sil164_write(client, i, state[i]);
+}
+
+static void
+sil164_set_power_state(struct i2c_client *client, bool on)
+{
+ uint8_t control0 = sil164_read(client, SIL164_CONTROL0);
+
+ if (on)
+ control0 |= SIL164_CONTROL0_POWER_ON;
+ else
+ control0 &= ~SIL164_CONTROL0_POWER_ON;
+
+ sil164_write(client, SIL164_CONTROL0, control0);
+}
+
+static void
+sil164_init_state(struct i2c_client *client,
+ struct sil164_encoder_params *config,
+ bool duallink)
+{
+ sil164_write(client, SIL164_CONTROL0,
+ SIL164_CONTROL0_HSYNC_ON |
+ SIL164_CONTROL0_VSYNC_ON |
+ (config->input_edge ? SIL164_CONTROL0_EDGE_RISING : 0) |
+ (config->input_width ? SIL164_CONTROL0_INPUT_24BIT : 0) |
+ (config->input_dual ? SIL164_CONTROL0_DUAL_EDGE : 0));
+
+ sil164_write(client, SIL164_DETECT,
+ SIL164_DETECT_INTR_STAT |
+ SIL164_DETECT_OUT_MODE_RECEIVER);
+
+ sil164_write(client, SIL164_CONTROL1,
+ (config->input_skew ? SIL164_CONTROL1_DESKEW_ENABLE : 0) |
+ (((config->input_skew + 4) & 0x7)
+ << SIL164_CONTROL1_DESKEW_INCR_SHIFT));
+
+ sil164_write(client, SIL164_CONTROL2,
+ SIL164_CONTROL2_SYNC_CONT |
+ (config->pll_filter ? 0 : SIL164_CONTROL2_FILTER_ENABLE) |
+ (4 << SIL164_CONTROL2_FILTER_SETTING_SHIFT));
+
+ sil164_write(client, SIL164_PLLZONE, 0);
+
+ if (duallink)
+ sil164_write(client, SIL164_DUALLINK,
+ SIL164_DUALLINK_ENABLE |
+ (((config->duallink_skew + 4) & 0x7)
+ << SIL164_DUALLINK_SKEW_SHIFT));
+ else
+ sil164_write(client, SIL164_DUALLINK, 0);
+}
+
+/* DRM encoder functions */
+
+static void
+sil164_encoder_set_config(struct drm_encoder *encoder, void *params)
+{
+ struct sil164_priv *priv = to_sil164_priv(encoder);
+
+ priv->config = *(struct sil164_encoder_params *)params;
+}
+
+static void
+sil164_encoder_dpms(struct drm_encoder *encoder, int mode)
+{
+ struct sil164_priv *priv = to_sil164_priv(encoder);
+ bool on = (mode == DRM_MODE_DPMS_ON);
+ bool duallink = (on && encoder->crtc->mode.clock > 165000);
+
+ sil164_set_power_state(drm_i2c_encoder_get_client(encoder), on);
+
+ if (priv->duallink_slave)
+ sil164_set_power_state(priv->duallink_slave, duallink);
+}
+
+static void
+sil164_encoder_save(struct drm_encoder *encoder)
+{
+ struct sil164_priv *priv = to_sil164_priv(encoder);
+
+ sil164_save_state(drm_i2c_encoder_get_client(encoder),
+ priv->saved_state);
+
+ if (priv->duallink_slave)
+ sil164_save_state(priv->duallink_slave,
+ priv->saved_slave_state);
+}
+
+static void
+sil164_encoder_restore(struct drm_encoder *encoder)
+{
+ struct sil164_priv *priv = to_sil164_priv(encoder);
+
+ sil164_restore_state(drm_i2c_encoder_get_client(encoder),
+ priv->saved_state);
+
+ if (priv->duallink_slave)
+ sil164_restore_state(priv->duallink_slave,
+ priv->saved_slave_state);
+}
+
+static bool
+sil164_encoder_mode_fixup(struct drm_encoder *encoder,
+ struct drm_display_mode *mode,
+ struct drm_display_mode *adjusted_mode)
+{
+ return true;
+}
+
+static int
+sil164_encoder_mode_valid(struct drm_encoder *encoder,
+ struct drm_display_mode *mode)
+{
+ struct sil164_priv *priv = to_sil164_priv(encoder);
+
+ if (mode->clock < 32000)
+ return MODE_CLOCK_LOW;
+
+ if (mode->clock > 330000 ||
+ (mode->clock > 165000 && !priv->duallink_slave))
+ return MODE_CLOCK_HIGH;
+
+ return MODE_OK;
+}
+
+static void
+sil164_encoder_mode_set(struct drm_encoder *encoder,
+ struct drm_display_mode *mode,
+ struct drm_display_mode *adjusted_mode)
+{
+ struct sil164_priv *priv = to_sil164_priv(encoder);
+ bool duallink = adjusted_mode->clock > 165000;
+
+ sil164_init_state(drm_i2c_encoder_get_client(encoder),
+ &priv->config, duallink);
+
+ if (priv->duallink_slave)
+ sil164_init_state(priv->duallink_slave,
+ &priv->config, duallink);
+
+ sil164_encoder_dpms(encoder, DRM_MODE_DPMS_ON);
+}
+
+static enum drm_connector_status
+sil164_encoder_detect(struct drm_encoder *encoder,
+ struct drm_connector *connector)
+{
+ struct i2c_client *client = drm_i2c_encoder_get_client(encoder);
+
+ if (sil164_read(client, SIL164_DETECT) & SIL164_DETECT_HOTPLUG_STAT)
+ return connector_status_connected;
+ else
+ return connector_status_disconnected;
+}
+
+static int
+sil164_encoder_get_modes(struct drm_encoder *encoder,
+ struct drm_connector *connector)
+{
+ return 0;
+}
+
+static int
+sil164_encoder_create_resources(struct drm_encoder *encoder,
+ struct drm_connector *connector)
+{
+ return 0;
+}
+
+static int
+sil164_encoder_set_property(struct drm_encoder *encoder,
+ struct drm_connector *connector,
+ struct drm_property *property,
+ uint64_t val)
+{
+ return 0;
+}
+
+static void
+sil164_encoder_destroy(struct drm_encoder *encoder)
+{
+ struct sil164_priv *priv = to_sil164_priv(encoder);
+
+ if (priv->duallink_slave)
+ i2c_unregister_device(priv->duallink_slave);
+
+ kfree(priv);
+ drm_i2c_encoder_destroy(encoder);
+}
+
+static struct drm_encoder_slave_funcs sil164_encoder_funcs = {
+ .set_config = sil164_encoder_set_config,
+ .destroy = sil164_encoder_destroy,
+ .dpms = sil164_encoder_dpms,
+ .save = sil164_encoder_save,
+ .restore = sil164_encoder_restore,
+ .mode_fixup = sil164_encoder_mode_fixup,
+ .mode_valid = sil164_encoder_mode_valid,
+ .mode_set = sil164_encoder_mode_set,
+ .detect = sil164_encoder_detect,
+ .get_modes = sil164_encoder_get_modes,
+ .create_resources = sil164_encoder_create_resources,
+ .set_property = sil164_encoder_set_property,
+};
+
+/* I2C driver functions */
+
+static int
+sil164_probe(struct i2c_client *client, const struct i2c_device_id *id)
+{
+ int vendor = sil164_read(client, SIL164_VENDOR_HI) << 8 |
+ sil164_read(client, SIL164_VENDOR_LO);
+ int device = sil164_read(client, SIL164_DEVICE_HI) << 8 |
+ sil164_read(client, SIL164_DEVICE_LO);
+ int rev = sil164_read(client, SIL164_REVISION);
+
+ if (vendor != 0x1 || device != 0x6) {
+ sil164_dbg(client, "Unknown device %x:%x.%x\n",
+ vendor, device, rev);
+ return -ENODEV;
+ }
+
+ sil164_info(client, "Detected device %x:%x.%x\n",
+ vendor, device, rev);
+
+ return 0;
+}
+
+static int
+sil164_remove(struct i2c_client *client)
+{
+ return 0;
+}
+
+static struct i2c_client *
+sil164_detect_slave(struct i2c_client *client)
+{
+ struct i2c_adapter *adap = client->adapter;
+ struct i2c_msg msg = {
+ .addr = SIL164_I2C_ADDR_SLAVE,
+ .len = 0,
+ };
+ const struct i2c_board_info info = {
+ I2C_BOARD_INFO("sil164", SIL164_I2C_ADDR_SLAVE)
+ };
+
+ if (i2c_transfer(adap, &msg, 1) != 1) {
+ sil164_dbg(adap, "No dual-link slave found.");
+ return NULL;
+ }
+
+ return i2c_new_device(adap, &info);
+}
+
+static int
+sil164_encoder_init(struct i2c_client *client,
+ struct drm_device *dev,
+ struct drm_encoder_slave *encoder)
+{
+ struct sil164_priv *priv;
+
+ priv = kzalloc(sizeof(*priv), GFP_KERNEL);
+ if (!priv)
+ return -ENOMEM;
+
+ encoder->slave_priv = priv;
+ encoder->slave_funcs = &sil164_encoder_funcs;
+
+ priv->duallink_slave = sil164_detect_slave(client);
+
+ return 0;
+}
+
+static struct i2c_device_id sil164_ids[] = {
+ { "sil164", 0 },
+ { }
+};
+MODULE_DEVICE_TABLE(i2c, sil164_ids);
+
+static struct drm_i2c_encoder_driver sil164_driver = {
+ .i2c_driver = {
+ .probe = sil164_probe,
+ .remove = sil164_remove,
+ .driver = {
+ .name = "sil164",
+ },
+ .id_table = sil164_ids,
+ },
+ .encoder_init = sil164_encoder_init,
+};
+
+/* Module initialization */
+
+static int __init
+sil164_init(void)
+{
+ return drm_i2c_encoder_register(THIS_MODULE, &sil164_driver);
+}
+
+static void __exit
+sil164_exit(void)
+{
+ drm_i2c_encoder_unregister(&sil164_driver);
+}
+
+MODULE_AUTHOR("Francisco Jerez <currojerez@riseup.net>");
+MODULE_DESCRIPTION("Silicon Image sil164 TMDS transmitter driver");
+MODULE_LICENSE("GPL and additional rights");
+
+module_init(sil164_init);
+module_exit(sil164_exit);
diff --git a/drivers/gpu/drm/nouveau/Kconfig b/drivers/gpu/drm/nouveau/Kconfig
index 1175429..6b8967a 100644
--- a/drivers/gpu/drm/nouveau/Kconfig
+++ b/drivers/gpu/drm/nouveau/Kconfig
@@ -41,4 +41,13 @@ config DRM_I2C_CH7006
This driver is currently only useful if you're also using
the nouveau driver.
+
+config DRM_I2C_SIL164
+ tristate "Silicon Image sil164 TMDS transmitter"
+ default m if DRM_NOUVEAU
+ help
+ Support for sil164 and similar single-link (or dual-link
+ when used in pairs) TMDS transmitters, used in some nVidia
+ video cards.
+
endmenu
diff --git a/include/drm/i2c/sil164.h b/include/drm/i2c/sil164.h
new file mode 100644
index 0000000..205e273
--- /dev/null
+++ b/include/drm/i2c/sil164.h
@@ -0,0 +1,63 @@
+/*
+ * Copyright (C) 2010 Francisco Jerez.
+ * All Rights Reserved.
+ *
+ * Permission is hereby granted, free of charge, to any person obtaining
+ * a copy of this software and associated documentation files (the
+ * "Software"), to deal in the Software without restriction, including
+ * without limitation the rights to use, copy, modify, merge, publish,
+ * distribute, sublicense, and/or sell copies of the Software, and to
+ * permit persons to whom the Software is furnished to do so, subject to
+ * the following conditions:
+ *
+ * The above copyright notice and this permission notice (including the
+ * next paragraph) shall be included in all copies or substantial
+ * portions of the Software.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT.
+ * IN NO EVENT SHALL THE COPYRIGHT OWNER(S) AND/OR ITS SUPPLIERS BE
+ * LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN ACTION
+ * OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN CONNECTION
+ * WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE SOFTWARE.
+ *
+ */
+
+#ifndef __DRM_I2C_SIL164_H__
+#define __DRM_I2C_SIL164_H__
+
+/**
+ * struct sil164_encoder_params
+ *
+ * Describes how the sil164 is connected to the GPU. It should be used
+ * as the @params parameter of its @set_config method.
+ *
+ * See "http://www.siliconimage.com/docs/SiI-DS-0021-E-164.pdf".
+ */
+struct sil164_encoder_params {
+ enum {
+ SIL164_INPUT_EDGE_FALLING = 0,
+ SIL164_INPUT_EDGE_RISING
+ } input_edge;
+
+ enum {
+ SIL164_INPUT_WIDTH_12BIT = 0,
+ SIL164_INPUT_WIDTH_24BIT
+ } input_width;
+
+ enum {
+ SIL164_INPUT_SINGLE_EDGE = 0,
+ SIL164_INPUT_DUAL_EDGE
+ } input_dual;
+
+ enum {
+ SIL164_PLL_FILTER_ON = 0,
+ SIL164_PLL_FILTER_OFF,
+ } pll_filter;
+
+ int input_skew; /** < Allowed range [-4, 3], use 0 for no de-skew. */
+ int duallink_skew; /** < Allowed range [-4, 3]. */
+};
+
+#endif
--
1.7.2

View File

@ -0,0 +1,45 @@
From 74ef65374ae6d0eead4a631aea3aca80d016ff0f Mon Sep 17 00:00:00 2001
From: Francisco Jerez <currojerez@riseup.net>
Date: Thu, 22 Jul 2010 17:07:38 +0200
Subject: [PATCH 1/5] drm-simplify-i2c-config
drm/kms: Simplify setup of the initial I2C encoder config.
In most use cases the driver will be using the same static config all
the time: interpreting i2c_board_info::platform_data as the default
config we can can save the GPU driver a redundant set_config() call.
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
drivers/gpu/drm/drm_encoder_slave.c | 7 +++++++
1 files changed, 7 insertions(+), 0 deletions(-)
diff --git a/drivers/gpu/drm/drm_encoder_slave.c b/drivers/gpu/drm/drm_encoder_slave.c
index f018469..d62c064 100644
--- a/drivers/gpu/drm/drm_encoder_slave.c
+++ b/drivers/gpu/drm/drm_encoder_slave.c
@@ -41,6 +41,9 @@
* &drm_encoder_slave. The @slave_funcs field will be initialized with
* the hooks provided by the slave driver.
*
+ * If @info->platform_data is non-NULL it will be used as the initial
+ * slave config.
+ *
* Returns 0 on success or a negative errno on failure, in particular,
* -ENODEV is returned when no matching driver is found.
*/
@@ -85,6 +88,10 @@ int drm_i2c_encoder_init(struct drm_device *dev,
if (err)
goto fail_unregister;
+ if (info->platform_data)
+ encoder->slave_funcs->set_config(&encoder->base,
+ info->platform_data);
+
return 0;
fail_unregister:
--
1.7.2

View File

@ -0,0 +1,171 @@
From 08ae078a33245bc01dcf895bd886f30103cc6178 Mon Sep 17 00:00:00 2001
From: Thomas Hellstrom <thellstrom@vmware.com>
Date: Thu, 30 Sep 2010 12:36:45 +0200
Subject: drm/ttm: Fix two race conditions + fix busy codepaths
This fixes a race pointed out by Dave Airlie where we don't take a buffer
object about to be destroyed off the LRU lists properly. It also fixes a rare
case where a buffer object could be destroyed in the middle of an
accelerated eviction.
The patch also adds a utility function that can be used to prematurely
release GPU memory space usage of an object waiting to be destroyed.
For example during eviction or swapout.
The above mentioned commit didn't queue the buffer on the delayed destroy
list under some rare circumstances. It also didn't completely honor the
remove_all parameter.
Fixes:
https://bugzilla.redhat.com/show_bug.cgi?id=615505
http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=591061
Signed-off-by: Thomas Hellstrom <thellstrom@vmware.com>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
drivers/gpu/drm/ttm/ttm_bo.c | 84 +++++++++++++++++++++++++++++++++++------
include/drm/ttm/ttm_bo_api.h | 4 +-
2 files changed, 74 insertions(+), 14 deletions(-)
diff --git a/drivers/gpu/drm/ttm/ttm_bo.c b/drivers/gpu/drm/ttm/ttm_bo.c
index 555ebb1..77f22ba 100644
--- a/drivers/gpu/drm/ttm/ttm_bo.c
+++ b/drivers/gpu/drm/ttm/ttm_bo.c
@@ -442,6 +442,43 @@ out_err:
}
/**
+ * Call bo::reserved and with the lru lock held.
+ * Will release GPU memory type usage on destruction.
+ * This is the place to put in driver specific hooks.
+ * Will release the bo::reserved lock and the
+ * lru lock on exit.
+ */
+
+static void ttm_bo_cleanup_memtype_use(struct ttm_buffer_object *bo)
+{
+ struct ttm_bo_global *glob = bo->glob;
+
+ if (bo->ttm) {
+
+ /**
+ * Release the lru_lock, since we don't want to have
+ * an atomic requirement on ttm_tt[unbind|destroy].
+ */
+
+ spin_unlock(&glob->lru_lock);
+ ttm_tt_unbind(bo->ttm);
+ ttm_tt_destroy(bo->ttm);
+ bo->ttm = NULL;
+ spin_lock(&glob->lru_lock);
+ }
+
+ if (bo->mem.mm_node) {
+ drm_mm_put_block(bo->mem.mm_node);
+ bo->mem.mm_node = NULL;
+ }
+
+ atomic_set(&bo->reserved, 0);
+ wake_up_all(&bo->event_queue);
+ spin_unlock(&glob->lru_lock);
+}
+
+
+/**
* If bo idle, remove from delayed- and lru lists, and unref.
* If not idle, and already on delayed list, do nothing.
* If not idle, and not on delayed list, put on delayed list,
@@ -456,6 +493,7 @@ static int ttm_bo_cleanup_refs(struct ttm_buffer_object *bo, bool remove_all)
int ret;
spin_lock(&bo->lock);
+retry:
(void) ttm_bo_wait(bo, false, false, !remove_all);
if (!bo->sync_obj) {
@@ -464,32 +502,52 @@ static int ttm_bo_cleanup_refs(struct ttm_buffer_object *bo, bool remove_all)
spin_unlock(&bo->lock);
spin_lock(&glob->lru_lock);
- put_count = ttm_bo_del_from_lru(bo);
+ ret = ttm_bo_reserve_locked(bo, false, !remove_all, false, 0);
+
+ /**
+ * Someone else has the object reserved. Bail and retry.
+ */
- ret = ttm_bo_reserve_locked(bo, false, false, false, 0);
- BUG_ON(ret);
- if (bo->ttm)
- ttm_tt_unbind(bo->ttm);
+ if (unlikely(ret == -EBUSY)) {
+ spin_unlock(&glob->lru_lock);
+ spin_lock(&bo->lock);
+ goto requeue;
+ }
+
+ /**
+ * We can re-check for sync object without taking
+ * the bo::lock since setting the sync object requires
+ * also bo::reserved. A busy object at this point may
+ * be caused by another thread starting an accelerated
+ * eviction.
+ */
+
+ if (unlikely(bo->sync_obj)) {
+ atomic_set(&bo->reserved, 0);
+ wake_up_all(&bo->event_queue);
+ spin_unlock(&glob->lru_lock);
+ spin_lock(&bo->lock);
+ if (remove_all)
+ goto retry;
+ else
+ goto requeue;
+ }
+
+ put_count = ttm_bo_del_from_lru(bo);
if (!list_empty(&bo->ddestroy)) {
list_del_init(&bo->ddestroy);
++put_count;
}
- if (bo->mem.mm_node) {
- bo->mem.mm_node->private = NULL;
- drm_mm_put_block(bo->mem.mm_node);
- bo->mem.mm_node = NULL;
- }
- spin_unlock(&glob->lru_lock);
- atomic_set(&bo->reserved, 0);
+ ttm_bo_cleanup_memtype_use(bo);
while (put_count--)
kref_put(&bo->list_kref, ttm_bo_ref_bug);
return 0;
}
-
+requeue:
spin_lock(&glob->lru_lock);
if (list_empty(&bo->ddestroy)) {
void *sync_obj = bo->sync_obj;
diff --git a/include/drm/ttm/ttm_bo_api.h b/include/drm/ttm/ttm_bo_api.h
index 267a86c..2040e6c 100644
--- a/include/drm/ttm/ttm_bo_api.h
+++ b/include/drm/ttm/ttm_bo_api.h
@@ -246,9 +246,11 @@ struct ttm_buffer_object {
atomic_t reserved;
-
/**
* Members protected by the bo::lock
+ * In addition, setting sync_obj to anything else
+ * than NULL requires bo::reserved to be held. This allows for
+ * checking NULL while reserved but not holding bo::lock.
*/
void *sync_obj_arg;
--
1.7.3.2

39
drm-ttm-fix.patch Normal file
View File

@ -0,0 +1,39 @@
From 0fbecd400dd0a82d465b3086f209681e8c54cb0f Mon Sep 17 00:00:00 2001
From: Francisco Jerez <currojerez@riseup.net>
Date: Tue, 21 Sep 2010 02:15:15 +0200
Subject: [PATCH] drm/ttm: Clear the ghost cpu_writers flag on ttm_buffer_object_transfer.
It makes sense for a BO to move after a process has requested
exclusive RW access on it (e.g. because the BO used to be located in
unmappable VRAM and we intercepted the CPU access from the fault
handler).
If we let the ghost object inherit cpu_writers from the original
object, ttm_bo_release_list() will raise a kernel BUG when the ghost
object is destroyed. This can be reproduced with the nouveau driver on
nv5x.
Reported-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Reviewed-by: Jerome Glisse <jglisse@redhat.com>
Tested-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: Francisco Jerez <currojerez@riseup.net>
Signed-off-by: Dave Airlie <airlied@redhat.com>
---
drivers/gpu/drm/ttm/ttm_bo_util.c | 1 +
1 files changed, 1 insertions(+), 0 deletions(-)
diff --git a/drivers/gpu/drm/ttm/ttm_bo_util.c b/drivers/gpu/drm/ttm/ttm_bo_util.c
index 7cffb3e..3451a82 100644
--- a/drivers/gpu/drm/ttm/ttm_bo_util.c
+++ b/drivers/gpu/drm/ttm/ttm_bo_util.c
@@ -351,6 +351,7 @@ static int ttm_buffer_object_transfer(struct ttm_buffer_object *bo,
INIT_LIST_HEAD(&fbo->lru);
INIT_LIST_HEAD(&fbo->swap);
fbo->vm_node = NULL;
+ atomic_set(&fbo->cpu_writers, 0);
fbo->sync_obj = driver->sync_obj_ref(bo->sync_obj);
kref_init(&fbo->list_kref);
--
1.7.3.1

124
efifb-add-more-models.patch Normal file
View File

@ -0,0 +1,124 @@
commit a5757c2a474a15f87e5baa9a4caacc31cde2bae6
Author: Luke Macken <lmacken@redhat.com>
Date: Wed Sep 22 13:05:04 2010 -0700
efifb: support the EFI framebuffer on more Apple hardware
Enable the EFI framebuffer on 14 more Macs, including the iMac11,1
iMac10,1 iMac8,1 Macmini3,1 Macmini4,1 MacBook5,1 MacBook6,1 MacBook7,1
MacBookPro2,2 MacBookPro5,2 MacBookPro5,3 MacBookPro6,1 MacBookPro6,2 and
MacBookPro7,1
Information gathered from various user submissions.
https://bugzilla.redhat.com/show_bug.cgi?id=528232
http://ubuntuforums.org/showthread.php?t=1557326
[akpm@linux-foundation.org: coding-style fixes]
Signed-off-by: Luke Macken <lmacken@redhat.com>
Signed-off-by: Peter Jones <pjones@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/drivers/video/efifb.c b/drivers/video/efifb.c
index c082b61..70477c2 100644
--- a/drivers/video/efifb.c
+++ b/drivers/video/efifb.c
@@ -39,17 +39,31 @@ enum {
M_I20, /* 20-Inch iMac */
M_I20_SR, /* 20-Inch iMac (Santa Rosa) */
M_I24, /* 24-Inch iMac */
+ M_I24_8_1, /* 24-Inch iMac, 8,1th gen */
+ M_I24_10_1, /* 24-Inch iMac, 10,1th gen */
+ M_I27_11_1, /* 27-Inch iMac, 11,1th gen */
M_MINI, /* Mac Mini */
+ M_MINI_3_1, /* Mac Mini, 3,1th gen */
+ M_MINI_4_1, /* Mac Mini, 4,1th gen */
M_MB, /* MacBook */
M_MB_2, /* MacBook, 2nd rev. */
M_MB_3, /* MacBook, 3rd rev. */
+ M_MB_5_1, /* MacBook, 5th rev. */
+ M_MB_6_1, /* MacBook, 6th rev. */
+ M_MB_7_1, /* MacBook, 7th rev. */
M_MB_SR, /* MacBook, 2nd gen, (Santa Rosa) */
M_MBA, /* MacBook Air */
M_MBP, /* MacBook Pro */
M_MBP_2, /* MacBook Pro 2nd gen */
+ M_MBP_2_2, /* MacBook Pro 2,2nd gen */
M_MBP_SR, /* MacBook Pro (Santa Rosa) */
M_MBP_4, /* MacBook Pro, 4th gen */
M_MBP_5_1, /* MacBook Pro, 5,1th gen */
+ M_MBP_5_2, /* MacBook Pro, 5,2th gen */
+ M_MBP_5_3, /* MacBook Pro, 5,3rd gen */
+ M_MBP_6_1, /* MacBook Pro, 6,1th gen */
+ M_MBP_6_2, /* MacBook Pro, 6,2th gen */
+ M_MBP_7_1, /* MacBook Pro, 7,1th gen */
M_UNKNOWN /* placeholder */
};
@@ -64,14 +78,28 @@ static struct efifb_dmi_info {
[M_I20] = { "i20", 0x80010000, 1728 * 4, 1680, 1050 }, /* guess */
[M_I20_SR] = { "imac7", 0x40010000, 1728 * 4, 1680, 1050 },
[M_I24] = { "i24", 0x80010000, 2048 * 4, 1920, 1200 }, /* guess */
+ [M_I24_8_1] = { "imac8", 0xc0060000, 2048 * 4, 1920, 1200 },
+ [M_I24_10_1] = { "imac10", 0xc0010000, 2048 * 4, 1920, 1080 },
+ [M_I27_11_1] = { "imac11", 0xc0010000, 2560 * 4, 2560, 1440 },
[M_MINI]= { "mini", 0x80000000, 2048 * 4, 1024, 768 },
+ [M_MINI_3_1] = { "mini31", 0x40010000, 1024 * 4, 1024, 768 },
+ [M_MINI_4_1] = { "mini41", 0xc0010000, 2048 * 4, 1920, 1200 },
[M_MB] = { "macbook", 0x80000000, 2048 * 4, 1280, 800 },
+ [M_MB_5_1] = { "macbook51", 0x80010000, 2048 * 4, 1280, 800 },
+ [M_MB_6_1] = { "macbook61", 0x80010000, 2048 * 4, 1280, 800 },
+ [M_MB_7_1] = { "macbook71", 0x80010000, 2048 * 4, 1280, 800 },
[M_MBA] = { "mba", 0x80000000, 2048 * 4, 1280, 800 },
[M_MBP] = { "mbp", 0x80010000, 1472 * 4, 1440, 900 },
[M_MBP_2] = { "mbp2", 0, 0, 0, 0 }, /* placeholder */
+ [M_MBP_2_2] = { "mbp22", 0x80010000, 1472 * 4, 1440, 900 },
[M_MBP_SR] = { "mbp3", 0x80030000, 2048 * 4, 1440, 900 },
[M_MBP_4] = { "mbp4", 0xc0060000, 2048 * 4, 1920, 1200 },
[M_MBP_5_1] = { "mbp51", 0xc0010000, 2048 * 4, 1440, 900 },
+ [M_MBP_5_2] = { "mbp52", 0xc0010000, 2048 * 4, 1920, 1200 },
+ [M_MBP_5_3] = { "mbp53", 0xd0010000, 2048 * 4, 1440, 900 },
+ [M_MBP_6_1] = { "mbp61", 0x90030000, 2048 * 4, 1920, 1200 },
+ [M_MBP_6_2] = { "mbp62", 0x90030000, 2048 * 4, 1680, 1050 },
+ [M_MBP_7_1] = { "mbp71", 0xc0010000, 2048 * 4, 1280, 800 },
[M_UNKNOWN] = { NULL, 0, 0, 0, 0 }
};
@@ -92,7 +120,12 @@ static const struct dmi_system_id dmi_system_table[] __initconst = {
EFIFB_DMI_SYSTEM_ID("Apple Computer, Inc.", "iMac6,1", M_I24),
EFIFB_DMI_SYSTEM_ID("Apple Inc.", "iMac6,1", M_I24),
EFIFB_DMI_SYSTEM_ID("Apple Inc.", "iMac7,1", M_I20_SR),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "iMac8,1", M_I24_8_1),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "iMac10,1", M_I24_10_1),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "iMac11,1", M_I27_11_1),
EFIFB_DMI_SYSTEM_ID("Apple Computer, Inc.", "Macmini1,1", M_MINI),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "Macmini3,1", M_MINI_3_1),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "Macmini4,1", M_MINI_4_1),
EFIFB_DMI_SYSTEM_ID("Apple Computer, Inc.", "MacBook1,1", M_MB),
/* At least one of these two will be right; maybe both? */
EFIFB_DMI_SYSTEM_ID("Apple Computer, Inc.", "MacBook2,1", M_MB),
@@ -101,14 +134,23 @@ static const struct dmi_system_id dmi_system_table[] __initconst = {
EFIFB_DMI_SYSTEM_ID("Apple Computer, Inc.", "MacBook3,1", M_MB),
EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBook3,1", M_MB),
EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBook4,1", M_MB),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBook5,1", M_MB_5_1),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBook6,1", M_MB_6_1),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBook7,1", M_MB_7_1),
EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookAir1,1", M_MBA),
EFIFB_DMI_SYSTEM_ID("Apple Computer, Inc.", "MacBookPro1,1", M_MBP),
EFIFB_DMI_SYSTEM_ID("Apple Computer, Inc.", "MacBookPro2,1", M_MBP_2),
+ EFIFB_DMI_SYSTEM_ID("Apple Computer, Inc.", "MacBookPro2,2", M_MBP_2_2),
EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookPro2,1", M_MBP_2),
EFIFB_DMI_SYSTEM_ID("Apple Computer, Inc.", "MacBookPro3,1", M_MBP_SR),
EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookPro3,1", M_MBP_SR),
EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookPro4,1", M_MBP_4),
EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookPro5,1", M_MBP_5_1),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookPro5,2", M_MBP_5_2),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookPro5,3", M_MBP_5_3),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookPro6,1", M_MBP_6_1),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookPro6,2", M_MBP_6_2),
+ EFIFB_DMI_SYSTEM_ID("Apple Inc.", "MacBookPro7,1", M_MBP_7_1),
{},
};

View File

@ -0,0 +1,135 @@
commit 85a00d9bbfb4704fbf368944b1cb9fed8f1598c5
Author: Peter Jones <pjones@redhat.com>
Date: Wed Sep 22 13:05:04 2010 -0700
efifb: check that the base address is plausible on pci systems
Some Apple machines have identical DMI data but different memory
configurations for the video. Given that, check that the address in our
table is actually within the range of a PCI BAR on a VGA device in the
machine.
This also fixes up the return value from set_system(), which has always
been wrong, but never resulted in bad behavior since there's only ever
been one matching entry in the dmi table.
The patch
1) stops people's machines from crashing when we get their display wrong,
which seems to be unfortunately inevitable,
2) allows us to support identical dmi data with differing video memory
configurations
This also adds me as the efifb maintainer, since I've effectively been
acting as such for quite some time.
Signed-off-by: Peter Jones <pjones@redhat.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
diff --git a/MAINTAINERS b/MAINTAINERS
index 726433a..4d4881d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -2199,6 +2199,12 @@ W: http://acpi4asus.sf.net
S: Maintained
F: drivers/platform/x86/eeepc-laptop.c
+EFIFB FRAMEBUFFER DRIVER
+L: linux-fbdev@vger.kernel.org
+M: Peter Jones <pjones@redhat.com>
+S: Maintained
+F: drivers/video/efifb.c
+
EFS FILESYSTEM
W: http://aeschi.ch.eu.org/efs/
S: Orphan
diff --git a/drivers/video/efifb.c b/drivers/video/efifb.c
index 815f84b..c082b61 100644
--- a/drivers/video/efifb.c
+++ b/drivers/video/efifb.c
@@ -13,7 +13,7 @@
#include <linux/platform_device.h>
#include <linux/screen_info.h>
#include <linux/dmi.h>
-
+#include <linux/pci.h>
#include <video/vga.h>
static struct fb_var_screeninfo efifb_defined __devinitdata = {
@@ -116,7 +116,7 @@ static int set_system(const struct dmi_system_id *id)
{
struct efifb_dmi_info *info = id->driver_data;
if (info->base == 0)
- return -ENODEV;
+ return 0;
printk(KERN_INFO "efifb: dmi detected %s - framebuffer at %p "
"(%dx%d, stride %d)\n", id->ident,
@@ -124,18 +124,55 @@ static int set_system(const struct dmi_system_id *id)
info->stride);
/* Trust the bootloader over the DMI tables */
- if (screen_info.lfb_base == 0)
+ if (screen_info.lfb_base == 0) {
+#if defined(CONFIG_PCI)
+ struct pci_dev *dev = NULL;
+ int found_bar = 0;
+#endif
screen_info.lfb_base = info->base;
- if (screen_info.lfb_linelength == 0)
- screen_info.lfb_linelength = info->stride;
- if (screen_info.lfb_width == 0)
- screen_info.lfb_width = info->width;
- if (screen_info.lfb_height == 0)
- screen_info.lfb_height = info->height;
- if (screen_info.orig_video_isVGA == 0)
- screen_info.orig_video_isVGA = VIDEO_TYPE_EFI;
- return 0;
+#if defined(CONFIG_PCI)
+ /* make sure that the address in the table is actually on a
+ * VGA device's PCI BAR */
+
+ for_each_pci_dev(dev) {
+ int i;
+ if ((dev->class >> 8) != PCI_CLASS_DISPLAY_VGA)
+ continue;
+ for (i = 0; i < DEVICE_COUNT_RESOURCE; i++) {
+ resource_size_t start, end;
+
+ start = pci_resource_start(dev, i);
+ if (start == 0)
+ break;
+ end = pci_resource_end(dev, i);
+ if (screen_info.lfb_base >= start &&
+ screen_info.lfb_base < end) {
+ found_bar = 1;
+ }
+ }
+ }
+ if (!found_bar)
+ screen_info.lfb_base = 0;
+#endif
+ }
+ if (screen_info.lfb_base) {
+ if (screen_info.lfb_linelength == 0)
+ screen_info.lfb_linelength = info->stride;
+ if (screen_info.lfb_width == 0)
+ screen_info.lfb_width = info->width;
+ if (screen_info.lfb_height == 0)
+ screen_info.lfb_height = info->height;
+ if (screen_info.orig_video_isVGA == 0)
+ screen_info.orig_video_isVGA = VIDEO_TYPE_EFI;
+ } else {
+ screen_info.lfb_linelength = 0;
+ screen_info.lfb_width = 0;
+ screen_info.lfb_height = 0;
+ screen_info.orig_video_isVGA = 0;
+ return 0;
+ }
+ return 1;
}
static int efifb_setcolreg(unsigned regno, unsigned red, unsigned green,

View File

@ -0,0 +1,63 @@
From be31c919e6ff27f1a08bc3e0725c51935313b002 Mon Sep 17 00:00:00 2001
From: Jan Kara <jack@suse.cz>
Date: Tue, 27 Jul 2010 11:56:07 -0400
Subject: ext4: Always journal quota file modifications
When journaled quota options are not specified, we do writes
to quota files just in data=ordered mode. This actually causes
warnings from JBD2 about dirty journaled buffer because ext4_getblk
unconditionally treats a block allocated by it as metadata. Since
quota actually is filesystem metadata, the easiest way to get rid
of the warning is to always treat quota writes as metadata...
Signed-off-by: Jan Kara <jack@suse.cz>
Signed-off-by: "Theodore Ts'o" <tytso@mit.edu>
---
fs/ext4/super.c | 19 +++++--------------
1 files changed, 5 insertions(+), 14 deletions(-)
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index a45ced9..f12daa7 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -4030,7 +4030,6 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type,
ext4_lblk_t blk = off >> EXT4_BLOCK_SIZE_BITS(sb);
int err = 0;
int offset = off & (sb->s_blocksize - 1);
- int journal_quota = EXT4_SB(sb)->s_qf_names[type] != NULL;
struct buffer_head *bh;
handle_t *handle = journal_current_handle();
@@ -4055,24 +4054,16 @@ static ssize_t ext4_quota_write(struct super_block *sb, int type,
bh = ext4_bread(handle, inode, blk, 1, &err);
if (!bh)
goto out;
- if (journal_quota) {
- err = ext4_journal_get_write_access(handle, bh);
- if (err) {
- brelse(bh);
- goto out;
- }
+ err = ext4_journal_get_write_access(handle, bh);
+ if (err) {
+ brelse(bh);
+ goto out;
}
lock_buffer(bh);
memcpy(bh->b_data+offset, data, len);
flush_dcache_page(bh->b_page);
unlock_buffer(bh);
- if (journal_quota)
- err = ext4_handle_dirty_metadata(handle, NULL, bh);
- else {
- /* Always do at least ordered writes for quotas */
- err = ext4_jbd2_file_inode(handle, inode);
- mark_buffer_dirty(bh);
- }
+ err = ext4_handle_dirty_metadata(handle, NULL, bh);
brelse(bh);
out:
if (err) {
--
1.7.3.3

84
fix-i8k-inline-asm.patch Normal file
View File

@ -0,0 +1,84 @@
commit 22d3243de86bc92d874abb7c5b185d5c47aba323
Author: Jim Bos <jim876@xs4all.nl>
Date: Mon Nov 15 21:22:37 2010 +0100
Fix gcc 4.5.1 miscompiling drivers/char/i8k.c (again)
The fix in commit 6b4e81db2552 ("i8k: Tell gcc that *regs gets
clobbered") to work around the gcc miscompiling i8k.c to add "+m
(*regs)" caused register pressure problems and a build failure.
Changing the 'asm' statement to 'asm volatile' instead should prevent
that and works around the gcc bug as well, so we can remove the "+m".
[ Background on the gcc bug: a memory clobber fails to mark the function
the asm resides in as non-pure (aka "__attribute__((const))"), so if
the function does nothing else that triggers the non-pure logic, gcc
will think that that function has no side effects at all. As a result,
callers will be mis-compiled.
Adding the "+m" made gcc see that it's not a pure function, and so
does "asm volatile". The problem was never really the need to mark
"*regs" as changed, since the memory clobber did that part - the
problem was just a bug in the gcc "pure" function analysis - Linus ]
Signed-off-by: Jim Bos <jim876@xs4all.nl>
Acked-by: Jakub Jelinek <jakub@redhat.com>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Andreas Schwab <schwab@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
commit 6b4e81db2552bad04100e7d5ddeed7e848f53b48
Author: Jim Bos <jim876@xs4all.nl>
Date: Sat Nov 13 12:13:53 2010 +0100
i8k: Tell gcc that *regs gets clobbered
More recent GCC caused the i8k driver to stop working, on Slackware
compiler was upgraded from gcc-4.4.4 to gcc-4.5.1 after which it didn't
work anymore, meaning the driver didn't load or gave total nonsensical
output.
As it turned out the asm(..) statement forgot to mention it modifies the
*regs variable.
Credits to Andi Kleen and Andreas Schwab for providing the fix.
Signed-off-by: Jim Bos <jim876@xs4all.nl>
Cc: Andi Kleen <andi@firstfloor.org>
Cc: Andreas Schwab <schwab@linux-m68k.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
diff --git a/drivers/char/i8k.c b/drivers/char/i8k.c
index 3bc0eef..d72433f 100644
--- a/drivers/char/i8k.c
+++ b/drivers/char/i8k.c
@@ -120,7 +120,7 @@ static int i8k_smm(struct smm_regs *regs)
int eax = regs->eax;
#if defined(CONFIG_X86_64)
- asm("pushq %%rax\n\t"
+ asm volatile("pushq %%rax\n\t"
"movl 0(%%rax),%%edx\n\t"
"pushq %%rdx\n\t"
"movl 4(%%rax),%%ebx\n\t"
@@ -146,7 +146,7 @@ static int i8k_smm(struct smm_regs *regs)
: "a"(regs)
: "%ebx", "%ecx", "%edx", "%esi", "%edi", "memory");
#else
- asm("pushl %%eax\n\t"
+ asm volatile("pushl %%eax\n\t"
"movl 0(%%eax),%%edx\n\t"
"push %%edx\n\t"
"movl 4(%%eax),%%ebx\n\t"
@@ -167,7 +167,8 @@ static int i8k_smm(struct smm_regs *regs)
"movl %%edx,0(%%eax)\n\t"
"lahf\n\t"
"shrl $8,%%eax\n\t"
- "andl $1,%%eax\n":"=a"(rc)
+ "andl $1,%%eax\n"
+ :"=a"(rc)
: "a"(regs)
: "%ebx", "%ecx", "%edx", "%esi", "%edi", "memory");
#endif

View File

@ -0,0 +1,27 @@
diff --git a/kernel/sched.c b/kernel/sched.c
index 6d0dbeb..3640c20 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -5155,9 +5155,11 @@ void __cpuinit init_idle_bootup_task(struct task_struct *idle)
void __cpuinit init_idle(struct task_struct *idle, int cpu)
{
struct rq *rq = cpu_rq(cpu);
+ struct rq *oldrq = task_rq(idle);
unsigned long flags;
- raw_spin_lock_irqsave(&rq->lock, flags);
+ local_irq_save(flags);
+ double_rq_lock(oldrq, rq);
__sched_fork(idle);
idle->state = TASK_RUNNING;
@@ -5170,7 +5172,8 @@ void __cpuinit init_idle(struct task_struct *idle, int cpu)
#if defined(CONFIG_SMP) && defined(__ARCH_WANT_UNLOCKED_CTXSW)
idle->oncpu = 1;
#endif
- raw_spin_unlock_irqrestore(&rq->lock, flags);
+ double_rq_unlock(oldrq, rq);
+ local_irq_restore(flags);
/* Set the preempt count _outside_ the spinlocks! */
#if defined(CONFIG_PREEMPT)

View File

@ -0,0 +1,30 @@
diff --git a/drivers/media/dvb/b2c2/flexcop-pci.c b/drivers/media/dvb/b2c2/flexcop-pci.c
index 227c020..7465308 100644
--- a/drivers/media/dvb/b2c2/flexcop-pci.c
+++ b/drivers/media/dvb/b2c2/flexcop-pci.c
@@ -39,6 +39,7 @@ MODULE_PARM_DESC(debug,
#define DRIVER_VERSION "0.1"
#define DRIVER_NAME "Technisat/B2C2 FlexCop II/IIb/III Digital TV PCI Driver"
+#define FLEXCOP_MODULE_NAME "b2c2-flexcop"
#define DRIVER_AUTHOR "Patrick Boettcher <patrick.boettcher@desy.de>"
struct flexcop_pci {
@@ -299,7 +300,7 @@ static int flexcop_pci_init(struct flexcop_pci *fc_pci)
return ret;
pci_set_master(fc_pci->pdev);
- if ((ret = pci_request_regions(fc_pci->pdev, DRIVER_NAME)) != 0)
+ if ((ret = pci_request_regions(fc_pci->pdev, FLEXCOP_MODULE_NAME)) != 0)
goto err_pci_disable_device;
fc_pci->io_mem = pci_iomap(fc_pci->pdev, 0, 0x800);
@@ -313,7 +314,7 @@ static int flexcop_pci_init(struct flexcop_pci *fc_pci)
pci_set_drvdata(fc_pci->pdev, fc_pci);
spin_lock_init(&fc_pci->irq_lock);
if ((ret = request_irq(fc_pci->pdev->irq, flexcop_pci_isr,
- IRQF_SHARED, DRIVER_NAME, fc_pci)) != 0)
+ IRQF_SHARED, FLEXCOP_MODULE_NAME, fc_pci)) != 0)
goto err_pci_iounmap;
fc_pci->init_state |= FC_PCI_INIT;

View File

@ -0,0 +1,65 @@
From linux-fsdevel-owner@vger.kernel.org Thu Nov 18 21:03:11 2010
From: Josef Bacik <josef@redhat.com>
To: linux-fsdevel@vger.kernel.org, eparis@redhat.com,
linux-kernel@vger.kernel.org, sds@tycho.nsa.gov,
selinux@tycho.nsa.gov, bfields@fieldses.org
Subject: [PATCH] fs: call security_d_instantiate in d_obtain_alias V2
Date: Thu, 18 Nov 2010 20:52:55 -0500
Message-Id: <1290131575-2489-1-git-send-email-josef@redhat.com>
X-Mailing-List: linux-fsdevel@vger.kernel.org
While trying to track down some NFS problems with BTRFS, I kept noticing I was
getting -EACCESS for no apparent reason. Eric Paris and printk() helped me
figure out that it was SELinux that was giving me grief, with the following
denial
type=AVC msg=audit(1290013638.413:95): avc: denied { 0x800000 } for pid=1772
comm="nfsd" name="" dev=sda1 ino=256 scontext=system_u:system_r:kernel_t:s0
tcontext=system_u:object_r:unlabeled_t:s0 tclass=file
Turns out this is because in d_obtain_alias if we can't find an alias we create
one and do all the normal instantiation stuff, but we don't do the
security_d_instantiate.
Usually we are protected from getting a hashed dentry that hasn't yet run
security_d_instantiate() by the parent's i_mutex, but obviously this isn't an
option there, so in order to deal with the case that a second thread comes in
and finds our new dentry before we get to run security_d_instantiate(), we go
ahead and call it if we find a dentry already. Eric assures me that this is ok
as the code checks to see if the dentry has been initialized already so calling
security_d_instantiate() against the same dentry multiple times is ok. With
this patch I'm no longer getting errant -EACCESS values.
Signed-off-by: Josef Bacik <josef@redhat.com>
---
V1->V2:
-added second security_d_instantiate() call
fs/dcache.c | 3 +++
1 files changed, 3 insertions(+), 0 deletions(-)
diff --git a/fs/dcache.c b/fs/dcache.c
index 23702a9..119d489 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -1201,9 +1201,12 @@ struct dentry *d_obtain_alias(struct inode *inode)
spin_unlock(&tmp->d_lock);
spin_unlock(&dcache_lock);
+ security_d_instantiate(tmp, inode);
return tmp;
out_iput:
+ if (res && !IS_ERR(res))
+ security_d_instantiate(res, inode);
iput(inode);
return res;
}
--
1.6.6.1
--
To unsubscribe from this list: send the line "unsubscribe linux-fsdevel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html

View File

@ -0,0 +1,163 @@
[PATCH] hid: support for bluetooth tivo slide remote and usb dongle
This patch adds full support for the TiVo Slide Remote, primarily by way
of extending the existing generic HID support. Only four keys are not
usages within a standard usage page, but they're easily handled by the
addition of a HID_UP_TIVOVENDOR usage page. Note that the UP is 0xffff,
which matches the mask, but its also a valid vendor-specific UP, according
to the spec.
What's actually connected to the computer is a Broadcom-made usb dongle,
which has an embedded hub, bluetooth adapter, mouse and keyboard devices.
You pair with the dongle, then the remote sends data that its converted
into HID on the keyboard interface (the mouse interface doesn't do anything
interesting right now, so far as I can tell).
lsusb for this device:
Bus 004 Device 005: ID 0a5c:2190 Broadcom Corp.
Bus 004 Device 004: ID 0a5c:4503 Broadcom Corp.
Bus 004 Device 003: ID 150a:1201
Bus 004 Device 002: ID 0a5c:4500 Broadcom Corp. BCM2046B1 USB 2.0 Hub (part of BCM2046 Bluetooth)
Speaking of the keyboard interface, the remote actually does contain a
keyboard as well. The top slides away, revealing a reasonably functional
qwerty keyboard (not unlike many slide cell phones), thus the product
name.
Now for some caveats... This device seems to report 0xc (consumer usage
page) 0x20 ("+10") after most key presses. At the moment, this will simply
be ignored, but if a mapping for that usage is added, the remote behaves
very badly (we end up w/a repeating/stuck key until another key is pressed).
Not quite sure what to do about that one, if that usage does get mapped. I
guess a device-specific quirk to just ignore it would work.
Anyway, the thing is working 100% as expected with this patch right now.
Three more things to note... This patch fixes an incorrect mapping of 0xc 0x45,
which was mapped to KEY_RADIO, which is definitely wrong, but may cause some
existing device to now report KEY_RIGHT there. Second, there's also an
unrelated fix for a redundant KERN_DEBUG in a dbg_hid call. Third, the
additions to HID_UP_GENDESK aren't strictly needed by the remote, but the
dongle does try to register those, and they're all technically correct, so
I've included them for the benefit of a device that comes along and
actually does try to use them.
Applies cleanly to hid master, tested w/a 2.6.35.4-based Fedora kernel.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
---
drivers/hid/hid-input.c | 45 ++++++++++++++++++++++++++++++++++++++++-----
include/linux/hid.h | 1 +
2 files changed, 41 insertions(+), 5 deletions(-)
diff --git a/drivers/hid/hid-input.c b/drivers/hid/hid-input.c
index 6c03dcc..bd1479e 100644
--- a/drivers/hid/hid-input.c
+++ b/drivers/hid/hid-input.c
@@ -44,11 +44,11 @@ static const unsigned char hid_keyboard[256] = {
72, 73, 82, 83, 86,127,116,117,183,184,185,186,187,188,189,190,
191,192,193,194,134,138,130,132,128,129,131,137,133,135,136,113,
115,114,unk,unk,unk,121,unk, 89, 93,124, 92, 94, 95,unk,unk,unk,
- 122,123, 90, 91, 85,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,
+ 122,123, 90, 91, 85,unk,unk,unk,unk,unk,unk,unk,111,unk,unk,unk,
unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,
unk,unk,unk,unk,unk,unk,179,180,unk,unk,unk,unk,unk,unk,unk,unk,
unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,
- unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,unk,
+ unk,unk,unk,unk,unk,unk,unk,unk,111,unk,unk,unk,unk,unk,unk,unk,
29, 42, 56,125, 97, 54,100,126,164,166,165,163,161,115,114,113,
150,158,159,128,136,177,178,176,142,152,173,140,unk,unk,unk,unk
};
@@ -136,7 +136,7 @@ static int hidinput_setkeycode(struct input_dev *dev,
clear_bit(old_keycode, dev->keybit);
set_bit(usage->code, dev->keybit);
- dbg_hid(KERN_DEBUG "Assigned keycode %d to HID usage code %x\n", keycode, scancode);
+ dbg_hid("Assigned keycode %d to HID usage code %x\n", keycode, scancode);
/* Set the keybit for the old keycode if the old keycode is used
* by another key */
if (hidinput_find_key (hid, 0, old_keycode))
@@ -235,6 +235,18 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
case 0x1: map_key_clear(KEY_POWER); break;
case 0x2: map_key_clear(KEY_SLEEP); break;
case 0x3: map_key_clear(KEY_WAKEUP); break;
+ case 0x4: map_key_clear(KEY_CONTEXT_MENU); break;
+ case 0x5: map_key_clear(KEY_MENU); break;
+ case 0x6: map_key_clear(KEY_PROG1); break;
+ case 0x7: map_key_clear(KEY_HELP); break;
+ case 0x8: map_key_clear(KEY_EXIT); break;
+ case 0x9: map_key_clear(KEY_SELECT); break;
+ case 0xa: map_key_clear(KEY_RIGHT); break;
+ case 0xb: map_key_clear(KEY_LEFT); break;
+ case 0xc: map_key_clear(KEY_UP); break;
+ case 0xd: map_key_clear(KEY_DOWN); break;
+ case 0xe: map_key_clear(KEY_POWER2); break;
+ case 0xf: map_key_clear(KEY_RESTART); break;
default: goto unknown;
}
break;
@@ -343,12 +355,24 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
case HID_UP_CONSUMER: /* USB HUT v1.1, pages 56-62 */
switch (usage->hid & HID_USAGE) {
case 0x000: goto ignore;
+ case 0x030: map_key_clear(KEY_POWER); break;
case 0x034: map_key_clear(KEY_SLEEP); break;
case 0x036: map_key_clear(BTN_MISC); break;
case 0x040: map_key_clear(KEY_MENU); break;
- case 0x045: map_key_clear(KEY_RADIO); break;
-
+ case 0x041: map_key_clear(KEY_SELECT); break;
+ case 0x042: map_key_clear(KEY_UP); break;
+ case 0x043: map_key_clear(KEY_DOWN); break;
+ case 0x044: map_key_clear(KEY_LEFT); break;
+ case 0x045: map_key_clear(KEY_RIGHT); break;
+
+ case 0x069: map_key_clear(KEY_RED); break;
+ case 0x06a: map_key_clear(KEY_GREEN); break;
+ case 0x06b: map_key_clear(KEY_BLUE); break;
+ case 0x06c: map_key_clear(KEY_YELLOW); break;
+ case 0x06d: map_key_clear(KEY_ZOOM); break;
+
+ case 0x082: map_key_clear(KEY_VIDEO_NEXT); break;
case 0x083: map_key_clear(KEY_LAST); break;
case 0x088: map_key_clear(KEY_PC); break;
case 0x089: map_key_clear(KEY_TV); break;
@@ -390,6 +414,7 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
case 0x0e5: map_key_clear(KEY_BASSBOOST); break;
case 0x0e9: map_key_clear(KEY_VOLUMEUP); break;
case 0x0ea: map_key_clear(KEY_VOLUMEDOWN); break;
+ case 0x0f5: map_key_clear(KEY_SLOW); break;
case 0x182: map_key_clear(KEY_BOOKMARKS); break;
case 0x183: map_key_clear(KEY_CONFIG); break;
@@ -491,6 +516,16 @@ static void hidinput_configure_usage(struct hid_input *hidinput, struct hid_fiel
}
break;
+ case HID_UP_TIVOVENDOR:
+ switch (usage->hid & HID_USAGE) {
+ case 0x3d: map_key_clear(KEY_PROG1); break;
+ case 0x3e: map_key_clear(KEY_TV); break;
+ case 0x41: map_key_clear(KEY_PAGEDOWN); break;
+ case 0x42: map_key_clear(KEY_PAGEUP); break;
+ default: goto unknown;
+ }
+ break;
+
default:
unknown:
if (field->report_size == 1) {
diff --git a/include/linux/hid.h b/include/linux/hid.h
index 42a0f1d..083cfb2 100644
--- a/include/linux/hid.h
+++ b/include/linux/hid.h
@@ -200,6 +200,7 @@ struct hid_item {
#define HID_UP_MSVENDOR 0xff000000
#define HID_UP_CUSTOM 0x00ff0000
#define HID_UP_LOGIVENDOR 0xffbc0000
+#define HID_UP_TIVOVENDOR 0xffff0000
#define HID_USAGE 0x0000ffff

View File

@ -0,0 +1,145 @@
From 785465d9cffd65b5a69dd2f465d2f7c917713220 Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@mcmartin.ca>
Date: Mon, 18 Oct 2010 13:30:39 -0400
Subject: [PATCH] ima: provide a toggle to disable it entirely
Signed-off-by: Kyle McMartin <kyle@redhat.com>
---
security/integrity/ima/ima.h | 1 +
security/integrity/ima/ima_iint.c | 9 +++++++++
security/integrity/ima/ima_main.c | 24 +++++++++++++++++++++---
3 files changed, 31 insertions(+), 3 deletions(-)
diff --git a/security/integrity/ima/ima.h b/security/integrity/ima/ima.h
index 3fbcd1d..65c3977 100644
--- a/security/integrity/ima/ima.h
+++ b/security/integrity/ima/ima.h
@@ -37,6 +37,7 @@ enum tpm_pcrs { TPM_PCR0 = 0, TPM_PCR8 = 8 };
/* set during initialization */
extern int iint_initialized;
extern int ima_initialized;
+extern int ima_enabled;
extern int ima_used_chip;
extern char *ima_hash;
diff --git a/security/integrity/ima/ima_iint.c b/security/integrity/ima/ima_iint.c
index afba4ae..3d191ef 100644
--- a/security/integrity/ima/ima_iint.c
+++ b/security/integrity/ima/ima_iint.c
@@ -54,6 +54,9 @@ int ima_inode_alloc(struct inode *inode)
struct ima_iint_cache *iint = NULL;
int rc = 0;
+ if (!ima_enabled)
+ return 0;
+
iint = kmem_cache_alloc(iint_cache, GFP_NOFS);
if (!iint)
return -ENOMEM;
@@ -116,6 +119,9 @@ void ima_inode_free(struct inode *inode)
{
struct ima_iint_cache *iint;
+ if (!ima_enabled)
+ return;
+
spin_lock(&ima_iint_lock);
iint = radix_tree_delete(&ima_iint_store, (unsigned long)inode);
spin_unlock(&ima_iint_lock);
@@ -139,6 +145,9 @@ static void init_once(void *foo)
static int __init ima_iintcache_init(void)
{
+ if (!ima_enabled)
+ return 0;
+
iint_cache =
kmem_cache_create("iint_cache", sizeof(struct ima_iint_cache), 0,
SLAB_PANIC, init_once);
diff --git a/security/integrity/ima/ima_main.c b/security/integrity/ima/ima_main.c
index e662b89..6e91905 100644
--- a/security/integrity/ima/ima_main.c
+++ b/security/integrity/ima/ima_main.c
@@ -26,6 +26,7 @@
#include "ima.h"
int ima_initialized;
+int ima_enabled;
char *ima_hash = "sha1";
static int __init hash_setup(char *str)
@@ -36,6 +37,14 @@ static int __init hash_setup(char *str)
}
__setup("ima_hash=", hash_setup);
+static int __init ima_enable(char *str)
+{
+ if (strncmp(str, "on", 2) == 0)
+ ima_enabled = 1;
+ return 1;
+}
+__setup("ima=", ima_enable);
+
struct ima_imbalance {
struct hlist_node node;
unsigned long fsmagic;
@@ -148,7 +157,7 @@ void ima_counts_get(struct file *file)
struct ima_iint_cache *iint;
int rc;
- if (!iint_initialized || !S_ISREG(inode->i_mode))
+ if (!ima_enabled || !iint_initialized || !S_ISREG(inode->i_mode))
return;
iint = ima_iint_find_get(inode);
if (!iint)
@@ -215,7 +224,7 @@ void ima_file_free(struct file *file)
struct inode *inode = file->f_dentry->d_inode;
struct ima_iint_cache *iint;
- if (!iint_initialized || !S_ISREG(inode->i_mode))
+ if (!ima_enabled || !iint_initialized || !S_ISREG(inode->i_mode))
return;
iint = ima_iint_find_get(inode);
if (!iint)
@@ -269,7 +278,7 @@ int ima_file_mmap(struct file *file, unsigned long prot)
{
int rc;
- if (!file)
+ if (!ima_enabled || !file)
return 0;
if (prot & PROT_EXEC)
rc = process_measurement(file, file->f_dentry->d_name.name,
@@ -294,6 +303,9 @@ int ima_bprm_check(struct linux_binprm *bprm)
{
int rc;
+ if (!ima_enabled)
+ return 0;
+
rc = process_measurement(bprm->file, bprm->filename,
MAY_EXEC, BPRM_CHECK);
return 0;
@@ -313,6 +325,9 @@ int ima_file_check(struct file *file, int mask)
{
int rc;
+ if (!ima_enabled)
+ return 0;
+
rc = process_measurement(file, file->f_dentry->d_name.name,
mask & (MAY_READ | MAY_WRITE | MAY_EXEC),
FILE_CHECK);
@@ -324,6 +339,9 @@ static int __init init_ima(void)
{
int error;
+ if (!ima_enabled)
+ return 0;
+
error = ima_init();
ima_initialized = 1;
return error;
--
1.7.3.1

View File

@ -0,0 +1,105 @@
From 56e0d4414e5eeacd9eaf68ce93dcb80f9c62bfb4 Mon Sep 17 00:00:00 2001
From: Nelson Elhage <nelhage@ksplice.com>
Date: Wed, 3 Nov 2010 16:35:41 +0000
Subject: inet_diag: Make sure we actually run the same bytecode we audited.
We were using nlmsg_find_attr() to look up the bytecode by attribute when
auditing, but then just using the first attribute when actually running
bytecode. So, if we received a message with two attribute elements, where only
the second had type INET_DIAG_REQ_BYTECODE, we would validate and run different
bytecode strings.
Fix this by consistently using nlmsg_find_attr everywhere.
Signed-off-by: Nelson Elhage <nelhage@ksplice.com>
Signed-off-by: Thomas Graf <tgraf@infradead.org>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
net/ipv4/inet_diag.c | 27 ++++++++++++++++-----------
1 files changed, 16 insertions(+), 11 deletions(-)
diff --git a/net/ipv4/inet_diag.c b/net/ipv4/inet_diag.c
index e5fa2dd..7403b9b 100644
--- a/net/ipv4/inet_diag.c
+++ b/net/ipv4/inet_diag.c
@@ -490,9 +490,11 @@ static int inet_csk_diag_dump(struct sock *sk,
{
struct inet_diag_req *r = NLMSG_DATA(cb->nlh);
- if (cb->nlh->nlmsg_len > 4 + NLMSG_SPACE(sizeof(*r))) {
+ if (nlmsg_attrlen(cb->nlh, sizeof(*r))) {
struct inet_diag_entry entry;
- struct rtattr *bc = (struct rtattr *)(r + 1);
+ const struct nlattr *bc = nlmsg_find_attr(cb->nlh,
+ sizeof(*r),
+ INET_DIAG_REQ_BYTECODE);
struct inet_sock *inet = inet_sk(sk);
entry.family = sk->sk_family;
@@ -512,7 +514,7 @@ static int inet_csk_diag_dump(struct sock *sk,
entry.dport = ntohs(inet->inet_dport);
entry.userlocks = sk->sk_userlocks;
- if (!inet_diag_bc_run(RTA_DATA(bc), RTA_PAYLOAD(bc), &entry))
+ if (!inet_diag_bc_run(nla_data(bc), nla_len(bc), &entry))
return 0;
}
@@ -527,9 +529,11 @@ static int inet_twsk_diag_dump(struct inet_timewait_sock *tw,
{
struct inet_diag_req *r = NLMSG_DATA(cb->nlh);
- if (cb->nlh->nlmsg_len > 4 + NLMSG_SPACE(sizeof(*r))) {
+ if (nlmsg_attrlen(cb->nlh, sizeof(*r))) {
struct inet_diag_entry entry;
- struct rtattr *bc = (struct rtattr *)(r + 1);
+ const struct nlattr *bc = nlmsg_find_attr(cb->nlh,
+ sizeof(*r),
+ INET_DIAG_REQ_BYTECODE);
entry.family = tw->tw_family;
#if defined(CONFIG_IPV6) || defined (CONFIG_IPV6_MODULE)
@@ -548,7 +552,7 @@ static int inet_twsk_diag_dump(struct inet_timewait_sock *tw,
entry.dport = ntohs(tw->tw_dport);
entry.userlocks = 0;
- if (!inet_diag_bc_run(RTA_DATA(bc), RTA_PAYLOAD(bc), &entry))
+ if (!inet_diag_bc_run(nla_data(bc), nla_len(bc), &entry))
return 0;
}
@@ -618,7 +622,7 @@ static int inet_diag_dump_reqs(struct sk_buff *skb, struct sock *sk,
struct inet_diag_req *r = NLMSG_DATA(cb->nlh);
struct inet_connection_sock *icsk = inet_csk(sk);
struct listen_sock *lopt;
- struct rtattr *bc = NULL;
+ const struct nlattr *bc = NULL;
struct inet_sock *inet = inet_sk(sk);
int j, s_j;
int reqnum, s_reqnum;
@@ -638,8 +642,9 @@ static int inet_diag_dump_reqs(struct sk_buff *skb, struct sock *sk,
if (!lopt || !lopt->qlen)
goto out;
- if (cb->nlh->nlmsg_len > 4 + NLMSG_SPACE(sizeof(*r))) {
- bc = (struct rtattr *)(r + 1);
+ if (nlmsg_attrlen(cb->nlh, sizeof(*r))) {
+ bc = nlmsg_find_attr(cb->nlh, sizeof(*r),
+ INET_DIAG_REQ_BYTECODE);
entry.sport = inet->inet_num;
entry.userlocks = sk->sk_userlocks;
}
@@ -672,8 +677,8 @@ static int inet_diag_dump_reqs(struct sk_buff *skb, struct sock *sk,
&ireq->rmt_addr;
entry.dport = ntohs(ireq->rmt_port);
- if (!inet_diag_bc_run(RTA_DATA(bc),
- RTA_PAYLOAD(bc), &entry))
+ if (!inet_diag_bc_run(nla_data(bc),
+ nla_len(bc), &entry))
continue;
}
--
1.7.3.2

File diff suppressed because it is too large Load Diff

View File

@ -0,0 +1,98 @@
From: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Date: Tue, 29 Jun 2010 05:53:50 +0000 (+0900)
Subject: kprobes/x86: Fix kprobes to skip prefixes correctly
X-Git-Tag: v2.6.36-rc1~41^2~53
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=567a9fd86735ccdc897768ed2dacdd5e83a13509
kprobes/x86: Fix kprobes to skip prefixes correctly
Fix resume_execution() and is_IF_modifier() to skip x86
instruction prefixes correctly by using x86 instruction
attribute.
Without this fix, resume_execution() can't handle instructions
which have non-REX prefixes (REX prefixes are skipped). This
will cause unexpected kernel panic by hitting bad address when a
kprobe hits on two-byte ret (e.g. "repz ret" generated for
Athlon/K8 optimization), because it just checks "repz" and can't
recognize the "ret" instruction.
These prefixes can be found easily with x86 instruction
attribute. This patch introduces skip_prefixes() and uses it in
resume_execution() and is_IF_modifier() to skip prefixes.
Signed-off-by: Masami Hiramatsu <masami.hiramatsu.pt@hitachi.com>
Cc: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
LKML-Reference: <4C298A6E.8070609@hitachi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
diff --git a/arch/x86/kernel/kprobes.c b/arch/x86/kernel/kprobes.c
index 345a4b1..175f85c 100644
--- a/arch/x86/kernel/kprobes.c
+++ b/arch/x86/kernel/kprobes.c
@@ -126,16 +126,22 @@ static void __kprobes synthesize_reljump(void *from, void *to)
}
/*
- * Check for the REX prefix which can only exist on X86_64
- * X86_32 always returns 0
+ * Skip the prefixes of the instruction.
*/
-static int __kprobes is_REX_prefix(kprobe_opcode_t *insn)
+static kprobe_opcode_t *__kprobes skip_prefixes(kprobe_opcode_t *insn)
{
+ insn_attr_t attr;
+
+ attr = inat_get_opcode_attribute((insn_byte_t)*insn);
+ while (inat_is_legacy_prefix(attr)) {
+ insn++;
+ attr = inat_get_opcode_attribute((insn_byte_t)*insn);
+ }
#ifdef CONFIG_X86_64
- if ((*insn & 0xf0) == 0x40)
- return 1;
+ if (inat_is_rex_prefix(attr))
+ insn++;
#endif
- return 0;
+ return insn;
}
/*
@@ -272,6 +278,9 @@ static int __kprobes can_probe(unsigned long paddr)
*/
static int __kprobes is_IF_modifier(kprobe_opcode_t *insn)
{
+ /* Skip prefixes */
+ insn = skip_prefixes(insn);
+
switch (*insn) {
case 0xfa: /* cli */
case 0xfb: /* sti */
@@ -280,13 +289,6 @@ static int __kprobes is_IF_modifier(kprobe_opcode_t *insn)
return 1;
}
- /*
- * on X86_64, 0x40-0x4f are REX prefixes so we need to look
- * at the next byte instead.. but of course not recurse infinitely
- */
- if (is_REX_prefix(insn))
- return is_IF_modifier(++insn);
-
return 0;
}
@@ -803,9 +805,8 @@ static void __kprobes resume_execution(struct kprobe *p,
unsigned long orig_ip = (unsigned long)p->addr;
kprobe_opcode_t *insn = p->ainsn.insn;
- /*skip the REX prefix*/
- if (is_REX_prefix(insn))
- insn++;
+ /* Skip prefixes */
+ insn = skip_prefixes(insn);
regs->flags &= ~X86_EFLAGS_TF;
switch (*insn) {

View File

@ -0,0 +1,161 @@
From: Avi Kivity <avi@redhat.com>
Date: Tue, 19 Oct 2010 14:46:55 +0000 (+0200)
Subject: KVM: Fix fs/gs reload oops with invalid ldt
X-Git-Tag: v2.6.36~4^2
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=9581d442b9058d3699b4be568b6e5eae38a41493
KVM: Fix fs/gs reload oops with invalid ldt
kvm reloads the host's fs and gs blindly, however the underlying segment
descriptors may be invalid due to the user modifying the ldt after loading
them.
Fix by using the safe accessors (loadsegment() and load_gs_index()) instead
of home grown unsafe versions.
This is CVE-2010-3698.
KVM-Stable-Tag.
Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Marcelo Tosatti <mtosatti@redhat.com>
---
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index 502e53f..c52e2eb 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -652,20 +652,6 @@ static inline struct kvm_mmu_page *page_header(hpa_t shadow_page)
return (struct kvm_mmu_page *)page_private(page);
}
-static inline u16 kvm_read_fs(void)
-{
- u16 seg;
- asm("mov %%fs, %0" : "=g"(seg));
- return seg;
-}
-
-static inline u16 kvm_read_gs(void)
-{
- u16 seg;
- asm("mov %%gs, %0" : "=g"(seg));
- return seg;
-}
-
static inline u16 kvm_read_ldt(void)
{
u16 ldt;
@@ -673,16 +659,6 @@ static inline u16 kvm_read_ldt(void)
return ldt;
}
-static inline void kvm_load_fs(u16 sel)
-{
- asm("mov %0, %%fs" : : "rm"(sel));
-}
-
-static inline void kvm_load_gs(u16 sel)
-{
- asm("mov %0, %%gs" : : "rm"(sel));
-}
-
static inline void kvm_load_ldt(u16 sel)
{
asm("lldt %0" : : "rm"(sel));
diff --git a/arch/x86/kvm/svm.c b/arch/x86/kvm/svm.c
index 81ed28c..8a3f9f6 100644
--- a/arch/x86/kvm/svm.c
+++ b/arch/x86/kvm/svm.c
@@ -3163,8 +3163,8 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu)
sync_lapic_to_cr8(vcpu);
save_host_msrs(vcpu);
- fs_selector = kvm_read_fs();
- gs_selector = kvm_read_gs();
+ savesegment(fs, fs_selector);
+ savesegment(gs, gs_selector);
ldt_selector = kvm_read_ldt();
svm->vmcb->save.cr2 = vcpu->arch.cr2;
/* required for live migration with NPT */
@@ -3251,10 +3251,15 @@ static void svm_vcpu_run(struct kvm_vcpu *vcpu)
vcpu->arch.regs[VCPU_REGS_RSP] = svm->vmcb->save.rsp;
vcpu->arch.regs[VCPU_REGS_RIP] = svm->vmcb->save.rip;
- kvm_load_ldt(ldt_selector);
- kvm_load_fs(fs_selector);
- kvm_load_gs(gs_selector);
load_host_msrs(vcpu);
+ kvm_load_ldt(ldt_selector);
+ loadsegment(fs, fs_selector);
+#ifdef CONFIG_X86_64
+ load_gs_index(gs_selector);
+ wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs);
+#else
+ loadsegment(gs, gs_selector);
+#endif
reload_tss(vcpu);
diff --git a/arch/x86/kvm/vmx.c b/arch/x86/kvm/vmx.c
index 49b25ee..7bddfab 100644
--- a/arch/x86/kvm/vmx.c
+++ b/arch/x86/kvm/vmx.c
@@ -803,7 +803,7 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu)
*/
vmx->host_state.ldt_sel = kvm_read_ldt();
vmx->host_state.gs_ldt_reload_needed = vmx->host_state.ldt_sel;
- vmx->host_state.fs_sel = kvm_read_fs();
+ savesegment(fs, vmx->host_state.fs_sel);
if (!(vmx->host_state.fs_sel & 7)) {
vmcs_write16(HOST_FS_SELECTOR, vmx->host_state.fs_sel);
vmx->host_state.fs_reload_needed = 0;
@@ -811,7 +811,7 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu)
vmcs_write16(HOST_FS_SELECTOR, 0);
vmx->host_state.fs_reload_needed = 1;
}
- vmx->host_state.gs_sel = kvm_read_gs();
+ savesegment(gs, vmx->host_state.gs_sel);
if (!(vmx->host_state.gs_sel & 7))
vmcs_write16(HOST_GS_SELECTOR, vmx->host_state.gs_sel);
else {
@@ -841,25 +841,19 @@ static void vmx_save_host_state(struct kvm_vcpu *vcpu)
static void __vmx_load_host_state(struct vcpu_vmx *vmx)
{
- unsigned long flags;
-
if (!vmx->host_state.loaded)
return;
++vmx->vcpu.stat.host_state_reload;
vmx->host_state.loaded = 0;
if (vmx->host_state.gs_ldt_reload_needed) {
kvm_load_ldt(vmx->host_state.ldt_sel);
- /*
- * If we have to reload gs, we must take care to
- * preserve our gs base.
- */
- local_irq_save(flags);
- kvm_load_gs(vmx->host_state.gs_sel);
#ifdef CONFIG_X86_64
- wrmsrl(MSR_GS_BASE, vmcs_readl(HOST_GS_BASE));
+ load_gs_index(vmx->host_state.gs_sel);
+ wrmsrl(MSR_KERNEL_GS_BASE, current->thread.gs);
+#else
+ loadsegment(gs, vmx->host_state.gs_sel);
#endif
- local_irq_restore(flags);
}
if (vmx->host_state.fs_reload_needed)
loadsegment(fs, vmx->host_state.fs_sel);
@@ -2589,8 +2583,8 @@ static int vmx_vcpu_setup(struct vcpu_vmx *vmx)
vmcs_write16(HOST_CS_SELECTOR, __KERNEL_CS); /* 22.2.4 */
vmcs_write16(HOST_DS_SELECTOR, __KERNEL_DS); /* 22.2.4 */
vmcs_write16(HOST_ES_SELECTOR, __KERNEL_DS); /* 22.2.4 */
- vmcs_write16(HOST_FS_SELECTOR, kvm_read_fs()); /* 22.2.4 */
- vmcs_write16(HOST_GS_SELECTOR, kvm_read_gs()); /* 22.2.4 */
+ vmcs_write16(HOST_FS_SELECTOR, 0); /* 22.2.4 */
+ vmcs_write16(HOST_GS_SELECTOR, 0); /* 22.2.4 */
vmcs_write16(HOST_SS_SELECTOR, __KERNEL_DS); /* 22.2.4 */
#ifdef CONFIG_X86_64
rdmsrl(MSR_FS_BASE, a);

View File

@ -0,0 +1,30 @@
Dump stack once on unsupported commands to see who is submitting them.
(#632753)
--- linux-2.6.34.noarch.orig/drivers/ata/pata_it821x.c
+++ linux-2.6.34.noarch/drivers/ata/pata_it821x.c
@@ -399,6 +399,16 @@ static void it821x_passthru_dev_select(s
ata_sff_dev_select(ap, device);
}
+static void it821x_dump_stack_once(void)
+{
+ static int dumped = 0;
+
+ if (!dumped) {
+ dump_stack();
+ dumped = 1;
+ }
+}
+
/**
* it821x_smart_qc_issue - wrap qc issue prot
* @qc: command
@@ -433,6 +443,7 @@ static unsigned int it821x_smart_qc_issu
return ata_sff_qc_issue(qc);
}
printk(KERN_DEBUG "it821x: can't process command 0x%02X\n", qc->tf.command);
+ it821x_dump_stack_once();
return AC_ERR_DEV;
}

View File

@ -0,0 +1,47 @@
Message-ID: <4C805293.1020305@cn.fujitsu.com>
Date: Fri, 03 Sep 2010 09:42:43 +0800
From: Li Zefan <lizf@cn.fujitsu.com>
To: David Miller <davem@davemloft.net>
CC: Herbert Xu <herbert@gondor.hengli.com.au>, Dave Jones <davej@redhat.com>,
netdev <netdev@vger.kernel.org>, LKML <linux-kernel@vger.kernel.org>,
Peter Zijlstra <peterz@infradead.org>, Paul Menage <menage@google.com>
Subject: [PATCH v2] cls_cgroup: Fix rcu lockdep warning
Dave reported an rcu lockdep warning on 2.6.35.4 kernel
task->cgroups and task->cgroups->subsys[i] are protected by RCU.
So we avoid accessing invalid pointers here. This might happen,
for example, when you are deref-ing those pointers while someone
move @task from one cgroup to another.
Reported-by: Dave Jones <davej@redhat.com>
Signed-off-by: Li Zefan <lizf@cn.fujitsu.com>
---
include/net/cls_cgroup.h | 10 ++++++++--
1 files changed, 8 insertions(+), 2 deletions(-)
diff --git a/include/net/cls_cgroup.h b/include/net/cls_cgroup.h
index dd1fdb8..a4dc5b0 100644
--- a/include/net/cls_cgroup.h
+++ b/include/net/cls_cgroup.h
@@ -27,11 +27,17 @@ struct cgroup_cls_state
#ifdef CONFIG_NET_CLS_CGROUP
static inline u32 task_cls_classid(struct task_struct *p)
{
+ int classid;
+
if (in_interrupt())
return 0;
- return container_of(task_subsys_state(p, net_cls_subsys_id),
- struct cgroup_cls_state, css)->classid;
+ rcu_read_lock();
+ classid = container_of(task_subsys_state(p, net_cls_subsys_id),
+ struct cgroup_cls_state, css)->classid;
+ rcu_read_unlock();
+
+ return classid;
}
#else
extern int net_cls_subsys_id;

View File

@ -0,0 +1,11 @@
--- linux-2.6.34.noarch.orig/kernel/power/main.c
+++ linux-2.6.34.noarch/kernel/power/main.c
@@ -45,7 +45,7 @@ int pm_notifier_call_chain(unsigned long
}
/* If set, devices may be suspended and resumed asynchronously. */
-int pm_async_enabled = 1;
+int pm_async_enabled;
static ssize_t pm_async_show(struct kobject *kobj, struct kobj_attribute *attr,
char *buf)

View File

@ -0,0 +1,29 @@
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index cea0cd9..c326065 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -2142,3 +2142,8 @@ source "crypto/Kconfig"
source "arch/x86/kvm/Kconfig"
source "lib/Kconfig"
+
+config PCI_DEFAULT_USE_CRS
+ def_bool y
+ prompt "Use PCI Host Bridge Windows from ACPI by default?"
+ depends on ACPI
diff --git a/arch/x86/pci/acpi.c b/arch/x86/pci/acpi.c
index 15466c0..3099406 100644
--- a/arch/x86/pci/acpi.c
+++ b/arch/x86/pci/acpi.c
@@ -16,7 +16,11 @@ struct pci_root_info {
int busnum;
};
+#ifdef CONFIG_PCI_DEFAULT_USE_CRS
static bool pci_use_crs = true;
+#else
+static bool pci_use_crs = false;
+#endif
static int __init set_use_crs(const struct dmi_system_id *id)
{

View File

@ -0,0 +1,30 @@
This only showed up in one SDV (Montevina).
The PCIE slots don't seem to like network cards, so this is the only hope
to get networking working. It's never going upstream, but it's low impact
enough to carry just to keep those SDVs working.
--- linux-2.6.35.noarch/drivers/net/e1000e/ich8lan.c~ 2010-09-29 17:53:13.000000000 -0400
+++ linux-2.6.35.noarch/drivers/net/e1000e/ich8lan.c 2010-09-29 17:54:00.000000000 -0400
@@ -424,6 +424,12 @@ static s32 e1000_init_phy_params_ich8lan
/* Verify phy id */
switch (phy->id) {
+ case 0:
+ if (hw->adapter->pdev->device == 0x10be)
+ e_dbg("got 0 phy id, trying anyway");
+ /* Fall through to IGP03E1000 case below */
+ else
+ return -E1000_ERR_PHY;
case IGP03E1000_E_PHY_ID:
phy->type = e1000_phy_igp_3;
phy->autoneg_mask = AUTONEG_ADVERTISE_SPEED_DEFAULT;
--- linux-2.6.35.noarch/drivers/net/e1000e/netdev.c~ 2010-09-29 17:54:07.000000000 -0400
+++ linux-2.6.35.noarch/drivers/net/e1000e/netdev.c 2010-09-29 17:54:29.000000000 -0400
@@ -5994,6 +5994,7 @@ static DEFINE_PCI_DEVICE_TABLE(e1000_pci
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_AMT), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH9_IGP_M_V), board_ich9lan },
+ { PCI_VDEVICE(INTEL, 0x10be), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_LM), board_ich9lan },
{ PCI_VDEVICE(INTEL, E1000_DEV_ID_ICH10_R_BM_LF), board_ich9lan },

View File

@ -1,46 +0,0 @@
[PATCH] ext4: fix freeze deadlock under IO
Commit 6b0310fbf087ad6 caused a regression resulting in deadlocks
when freezing a filesystem which had active IO; the vfs_check_frozen
level (SB_FREEZE_WRITE) did not let the freeze-related IO syncing
through. Duh.
Changing the test to FREEZE_TRANS should let the normal freeze
syncing get through the fs, but still block any transactions from
starting once the fs is completely frozen.
I tested this by running fsstress in the background while periodically
snapshotting the fs and running fsck on the result. I ran into
occasional deadlocks, but different ones. I think this is a
fine fix for the problem at hand, and the other deadlocky things
will need more investigation.
Reported-by: Phillip Susi <psusi@cfl.rr.com>
Signed-off-by: Eric Sandeen <sandeen@redhat.com>
---
diff --git a/fs/ext4/super.c b/fs/ext4/super.c
index 4e8983a..a45ced9 100644
--- a/fs/ext4/super.c
+++ b/fs/ext4/super.c
@@ -241,7 +241,7 @@ handle_t *ext4_journal_start_sb(struct super_block *sb, int nblocks)
if (sb->s_flags & MS_RDONLY)
return ERR_PTR(-EROFS);
- vfs_check_frozen(sb, SB_FREEZE_WRITE);
+ vfs_check_frozen(sb, SB_FREEZE_TRANS);
/* Special case here: if the journal has aborted behind our
* backs (eg. EIO in the commit thread), then we still need to
* take the FS itself readonly cleanly. */
@@ -3491,7 +3491,7 @@ int ext4_force_commit(struct super_block *sb)
journal = EXT4_SB(sb)->s_journal;
if (journal) {
- vfs_check_frozen(sb, SB_FREEZE_WRITE);
+ vfs_check_frozen(sb, SB_FREEZE_TRANS);
ret = ext4_journal_force_commit(journal);
}

View File

@ -538,15 +538,16 @@
static void unmap_region(struct mm_struct *mm,
struct vm_area_struct *vma, struct vm_area_struct *prev,
unsigned long start, unsigned long end);
@@ -388,6 +401,8 @@
__vma_link_list(struct mm_struct *mm, struct vm_area_struct *vma,
struct vm_area_struct *prev, struct rb_node *rb_parent)
@@ -388,6 +401,9 @@
{
struct vm_area_struct *next;
+ if (vma->vm_flags & VM_EXEC)
+ arch_add_exec_range(mm, vma->vm_end);
+
vma->vm_prev = prev;
if (prev) {
vma->vm_next = prev->vm_next;
prev->vm_next = vma;
next = prev->vm_next;
@@ -489,6 +504,8 @@
rb_erase(&vma->vm_rb, &mm->mm_rb);
if (mm->mmap_cache == vma)

View File

@ -0,0 +1,16 @@
diff --git a/include/linux/rcupdate.h b/include/linux/rcupdate.h
index 9fbc54a..435c502 100644
--- a/include/linux/rcupdate.h
+++ b/include/linux/rcupdate.h
@@ -454,7 +454,7 @@ static inline notrace void rcu_read_unlock_sched_notrace(void)
* Makes rcu_dereference_check() do the dirty work.
*/
#define rcu_dereference_bh(p) \
- rcu_dereference_check(p, rcu_read_lock_bh_held())
+ rcu_dereference_check(p, rcu_read_lock_bh_held() || irqs_disabled())
/**
* rcu_dereference_sched - fetch RCU-protected pointer, checking for RCU-sched

View File

@ -1 +1,33 @@
nil
From 615661f3948a066fd22a36fe8ea0c528b75ee373 Mon Sep 17 00:00:00 2001
From: Marcin Slusarz <marcin.slusarz@gmail.com>
Date: Sun, 22 Aug 2010 20:54:08 +0200
Subject: drm/nv50: initialize ramht_refs list for faked 0 channel
From: Marcin Slusarz <marcin.slusarz@gmail.com>
commit 615661f3948a066fd22a36fe8ea0c528b75ee373 upstream.
We need it for PFIFO_INTR_CACHE_ERROR interrupt handling,
because nouveau_fifo_swmthd looks for matching gpuobj in
ramht_refs list.
It fixes kernel panic in nouveau_gpuobj_ref_find.
Signed-off-by: Marcin Slusarz <marcin.slusarz@gmail.com>
Signed-off-by: Ben Skeggs <bskeggs@redhat.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
drivers/gpu/drm/nouveau/nv50_instmem.c | 2 ++
1 file changed, 2 insertions(+)
--- a/drivers/gpu/drm/nouveau/nv50_instmem.c
+++ b/drivers/gpu/drm/nouveau/nv50_instmem.c
@@ -141,6 +141,8 @@ nv50_instmem_init(struct drm_device *dev
chan->file_priv = (struct drm_file *)-2;
dev_priv->fifos[0] = dev_priv->fifos[127] = chan;
+ INIT_LIST_HEAD(&chan->ramht_refs);
+
/* Channel's PRAMIN object + heap */
ret = nouveau_gpuobj_new_fake(dev, 0, c_offset, c_size, 0,
NULL, &chan->ramin);

View File

@ -1,165 +0,0 @@
commit 611225f5e7f9d11c4b119fac224f1bd6903b0150
Author: Jarod Wilson <jarod@redhat.com>
Date: Sun Mar 7 17:55:43 2010 -0300
V4L/DVB: dvb: add support for kworld 340u and ub435-q to em28xx-dvb
This adds support for the KWorld PlusTV 340U and KWorld UB345-Q ATSC
sticks, which are really the same device. The sticks have an eMPIA
em2870 usb bridge chipset, an LG Electronics LGDT3304 ATSC/QAM
demodulator and an NXP TDA18271HD tuner -- early versions of the 340U
have a a TDA18271HD/C1, later models and the UB435-Q have a C2.
The stick has been tested succesfully with both VSB_8 and QAM_256 signals.
Its using lgdt3304 support added to the lgdt3305 driver by a prior patch,
rather than the current lgdt3304 driver, as its severely lacking in
functionality by comparison (see said patch for details).
Signed-off-by: Jarod Wilson <jarod@redhat.com>
Signed-off-by: Michael Krufky <mkrufky@kernellabs.com>
Signed-off-by: Mauro Carvalho Chehab <mchehab@redhat.com>
diff --git a/Documentation/video4linux/CARDLIST.em28xx b/Documentation/video4linux/CARDLIST.em28xx
index 3a623aa..5c56875 100644
--- a/Documentation/video4linux/CARDLIST.em28xx
+++ b/Documentation/video4linux/CARDLIST.em28xx
@@ -72,3 +72,4 @@
73 -> Reddo DVB-C USB TV Box (em2870)
74 -> Actionmaster/LinXcel/Digitus VC211A (em2800)
75 -> Dikom DK300 (em2882)
+ 76 -> KWorld PlusTV 340U or UB435-Q (ATSC) (em2870) [1b80:a340]
diff --git a/drivers/media/video/em28xx/em28xx-cards.c b/drivers/media/video/em28xx/em28xx-cards.c
index 3a4fd85..ffbe544 100644
--- a/drivers/media/video/em28xx/em28xx-cards.c
+++ b/drivers/media/video/em28xx/em28xx-cards.c
@@ -158,6 +158,22 @@ static struct em28xx_reg_seq evga_indtube_digital[] = {
{ -1, -1, -1, -1},
};
+/*
+ * KWorld PlusTV 340U and UB435-Q (ATSC) GPIOs map:
+ * EM_GPIO_0 - currently unknown
+ * EM_GPIO_1 - LED disable/enable (1 = off, 0 = on)
+ * EM_GPIO_2 - currently unknown
+ * EM_GPIO_3 - currently unknown
+ * EM_GPIO_4 - TDA18271HD/C1 tuner (1 = active, 0 = in reset)
+ * EM_GPIO_5 - LGDT3304 ATSC/QAM demod (1 = active, 0 = in reset)
+ * EM_GPIO_6 - currently unknown
+ * EM_GPIO_7 - currently unknown
+ */
+static struct em28xx_reg_seq kworld_a340_digital[] = {
+ {EM28XX_R08_GPIO, 0x6d, ~EM_GPIO_4, 10},
+ { -1, -1, -1, -1},
+};
+
/* Pinnacle Hybrid Pro eb1a:2881 */
static struct em28xx_reg_seq pinnacle_hybrid_pro_analog[] = {
{EM28XX_R08_GPIO, 0xfd, ~EM_GPIO_4, 10},
@@ -1667,6 +1683,16 @@ struct em28xx_board em28xx_boards[] = {
.tuner_gpio = reddo_dvb_c_usb_box,
.has_dvb = 1,
},
+ /* 1b80:a340 - Empia EM2870, NXP TDA18271HD and LG DT3304, sold
+ * initially as the KWorld PlusTV 340U, then as the UB435-Q.
+ * Early variants have a TDA18271HD/C1, later ones a TDA18271HD/C2 */
+ [EM2870_BOARD_KWORLD_A340] = {
+ .name = "KWorld PlusTV 340U or UB435-Q (ATSC)",
+ .tuner_type = TUNER_ABSENT, /* Digital-only TDA18271HD */
+ .has_dvb = 1,
+ .dvb_gpio = kworld_a340_digital,
+ .tuner_gpio = default_tuner_gpio,
+ },
};
const unsigned int em28xx_bcount = ARRAY_SIZE(em28xx_boards);
@@ -1788,6 +1814,8 @@ struct usb_device_id em28xx_id_table[] = {
.driver_info = EM2820_BOARD_IODATA_GVMVP_SZ },
{ USB_DEVICE(0xeb1a, 0x50a6),
.driver_info = EM2860_BOARD_GADMEI_UTV330 },
+ { USB_DEVICE(0x1b80, 0xa340),
+ .driver_info = EM2870_BOARD_KWORLD_A340 },
{ },
};
MODULE_DEVICE_TABLE(usb, em28xx_id_table);
diff --git a/drivers/media/video/em28xx/em28xx-dvb.c b/drivers/media/video/em28xx/em28xx-dvb.c
index cf1d8c3..3ac8d30 100644
--- a/drivers/media/video/em28xx/em28xx-dvb.c
+++ b/drivers/media/video/em28xx/em28xx-dvb.c
@@ -30,11 +30,13 @@
#include "tuner-simple.h"
#include "lgdt330x.h"
+#include "lgdt3305.h"
#include "zl10353.h"
#include "s5h1409.h"
#include "mt352.h"
#include "mt352_priv.h" /* FIXME */
#include "tda1002x.h"
+#include "tda18271.h"
MODULE_DESCRIPTION("driver for em28xx based DVB cards");
MODULE_AUTHOR("Mauro Carvalho Chehab <mchehab@infradead.org>");
@@ -231,6 +233,18 @@ static struct lgdt330x_config em2880_lgdt3303_dev = {
.demod_chip = LGDT3303,
};
+static struct lgdt3305_config em2870_lgdt3304_dev = {
+ .i2c_addr = 0x0e,
+ .demod_chip = LGDT3304,
+ .spectral_inversion = 1,
+ .deny_i2c_rptr = 1,
+ .mpeg_mode = LGDT3305_MPEG_PARALLEL,
+ .tpclk_edge = LGDT3305_TPCLK_FALLING_EDGE,
+ .tpvalid_polarity = LGDT3305_TP_VALID_HIGH,
+ .vsb_if_khz = 3250,
+ .qam_if_khz = 4000,
+};
+
static struct zl10353_config em28xx_zl10353_with_xc3028 = {
.demod_address = (0x1e >> 1),
.no_tuner = 1,
@@ -247,6 +261,17 @@ static struct s5h1409_config em28xx_s5h1409_with_xc3028 = {
.mpeg_timing = S5H1409_MPEGTIMING_CONTINOUS_NONINVERTING_CLOCK
};
+static struct tda18271_std_map kworld_a340_std_map = {
+ .atsc_6 = { .if_freq = 3250, .agc_mode = 3, .std = 0,
+ .if_lvl = 1, .rfagc_top = 0x37, },
+ .qam_6 = { .if_freq = 4000, .agc_mode = 3, .std = 1,
+ .if_lvl = 1, .rfagc_top = 0x37, },
+};
+
+static struct tda18271_config kworld_a340_config = {
+ .std_map = &kworld_a340_std_map,
+};
+
static struct zl10353_config em28xx_zl10353_xc3028_no_i2c_gate = {
.demod_address = (0x1e >> 1),
.no_tuner = 1,
@@ -572,6 +597,14 @@ static int dvb_init(struct em28xx *dev)
}
}
break;
+ case EM2870_BOARD_KWORLD_A340:
+ dvb->frontend = dvb_attach(lgdt3305_attach,
+ &em2870_lgdt3304_dev,
+ &dev->i2c_adap);
+ if (dvb->frontend != NULL)
+ dvb_attach(tda18271_attach, dvb->frontend, 0x60,
+ &dev->i2c_adap, &kworld_a340_config);
+ break;
default:
em28xx_errdev("/2: The frontend of your DVB/ATSC card"
" isn't supported yet\n");
diff --git a/drivers/media/video/em28xx/em28xx.h b/drivers/media/video/em28xx/em28xx.h
index 6216786..1c61a6b 100644
--- a/drivers/media/video/em28xx/em28xx.h
+++ b/drivers/media/video/em28xx/em28xx.h
@@ -114,6 +114,7 @@
#define EM2870_BOARD_REDDO_DVB_C_USB_BOX 73
#define EM2800_BOARD_VC211A 74
#define EM2882_BOARD_DIKOM_DK300 75
+#define EM2870_BOARD_KWORLD_A340 76
/* Limits minimum and default number of buffers */
#define EM28XX_MIN_BUF 4

View File

@ -1,350 +0,0 @@
From b71e18093e2e7f240797875c50c49552722f8825 Mon Sep 17 00:00:00 2001
From: Jarod Wilson <jarod@redhat.com>
Date: Mon, 15 Feb 2010 17:13:25 -0500
Subject: [PATCH 1/2] dvb: add lgdt3304 support to lgdt3305 driver
There's a currently-unused lgdt3304 demod driver, which leaves a lot to
be desired as far as functionality. The 3304 is unsurprisingly quite
similar to the 3305, and empirical testing yeilds far better results
and more complete functionality by merging 3304 support into the 3305
driver. (For example, the current lgdt3304 driver lacks support for
signal strength, snr, ucblocks, etc., which we get w/the lgdt3305).
For the moment, not dropping the lgdt3304 driver, and its still up to
a given device's config setup to choose which demod driver to use, but
I'd suggest dropping the 3304 driver entirely.
As a follow-up to this patch, I've got another patch that adds support
for the KWorld PlusTV 340U (ATSC) em2870-based tuner stick, driving
its lgdt3304 demod via this lgdt3305 driver, which is what I used to
successfully test this patch with both VSB_8 and QAM_256 signals.
A few pieces are still a touch crude, but I think its a solid start,
as well as much cleaner and more feature-complete than the existing
lgdt3304 driver.
Signed-off-by: Jarod Wilson <jarod@redhat.com>
---
drivers/media/dvb/frontends/lgdt3305.c | 206 ++++++++++++++++++++++++++++++--
drivers/media/dvb/frontends/lgdt3305.h | 6 +
2 files changed, 203 insertions(+), 9 deletions(-)
diff --git a/drivers/media/dvb/frontends/lgdt3305.c b/drivers/media/dvb/frontends/lgdt3305.c
index fde8c59..40695e6 100644
--- a/drivers/media/dvb/frontends/lgdt3305.c
+++ b/drivers/media/dvb/frontends/lgdt3305.c
@@ -1,5 +1,5 @@
/*
- * Support for LGDT3305 - VSB/QAM
+ * Support for LG Electronics LGDT3304 and LGDT3305 - VSB/QAM
*
* Copyright (C) 2008, 2009 Michael Krufky <mkrufky@linuxtv.org>
*
@@ -357,7 +357,10 @@ static int lgdt3305_rfagc_loop(struct lgdt3305_state *state,
case QAM_256:
agcdelay = 0x046b;
rfbw = 0x8889;
- ifbw = 0x8888;
+ if (state->cfg->demod_chip == LGDT3305)
+ ifbw = 0x8888;
+ else
+ ifbw = 0x6666;
break;
default:
return -EINVAL;
@@ -409,8 +412,18 @@ static int lgdt3305_agc_setup(struct lgdt3305_state *state,
lg_dbg("lockdten = %d, acqen = %d\n", lockdten, acqen);
/* control agc function */
- lgdt3305_write_reg(state, LGDT3305_AGC_CTRL_4, 0xe1 | lockdten << 1);
- lgdt3305_set_reg_bit(state, LGDT3305_AGC_CTRL_1, 2, acqen);
+ switch (state->cfg->demod_chip) {
+ case LGDT3304:
+ lgdt3305_write_reg(state, 0x0314, 0xe1 | lockdten << 1);
+ lgdt3305_set_reg_bit(state, 0x030e, 2, acqen);
+ break;
+ case LGDT3305:
+ lgdt3305_write_reg(state, LGDT3305_AGC_CTRL_4, 0xe1 | lockdten << 1);
+ lgdt3305_set_reg_bit(state, LGDT3305_AGC_CTRL_1, 2, acqen);
+ break;
+ default:
+ return -EINVAL;
+ }
return lgdt3305_rfagc_loop(state, param);
}
@@ -543,6 +556,11 @@ static int lgdt3305_i2c_gate_ctrl(struct dvb_frontend *fe, int enable)
enable ? 0 : 1);
}
+static int lgdt3304_sleep(struct dvb_frontend *fe)
+{
+ return 0;
+}
+
static int lgdt3305_sleep(struct dvb_frontend *fe)
{
struct lgdt3305_state *state = fe->demodulator_priv;
@@ -571,6 +589,55 @@ static int lgdt3305_sleep(struct dvb_frontend *fe)
return 0;
}
+static int lgdt3304_init(struct dvb_frontend *fe)
+{
+ struct lgdt3305_state *state = fe->demodulator_priv;
+ int ret;
+
+ static struct lgdt3305_reg lgdt3304_init_data[] = {
+ { .reg = LGDT3305_GEN_CTRL_1, .val = 0x03, },
+ { .reg = 0x000d, .val = 0x02, },
+ { .reg = 0x000e, .val = 0x02, },
+ { .reg = LGDT3305_DGTL_AGC_REF_1, .val = 0x32, },
+ { .reg = LGDT3305_DGTL_AGC_REF_2, .val = 0xc4, },
+ { .reg = LGDT3305_CR_CTR_FREQ_1, .val = 0x00, },
+ { .reg = LGDT3305_CR_CTR_FREQ_2, .val = 0x00, },
+ { .reg = LGDT3305_CR_CTR_FREQ_3, .val = 0x00, },
+ { .reg = LGDT3305_CR_CTR_FREQ_4, .val = 0x00, },
+ { .reg = LGDT3305_CR_CTRL_7, .val = 0xf9, },
+ { .reg = 0x0112, .val = 0x17, },
+ { .reg = 0x0113, .val = 0x15, },
+ { .reg = 0x0114, .val = 0x18, },
+ { .reg = 0x0115, .val = 0xff, },
+ { .reg = 0x0116, .val = 0x3c, },
+ { .reg = 0x0214, .val = 0x67, },
+ { .reg = 0x0424, .val = 0x8d, },
+ { .reg = 0x0427, .val = 0x12, },
+ { .reg = 0x0428, .val = 0x4f, },
+ { .reg = LGDT3305_IFBW_1, .val = 0x80, },
+ { .reg = LGDT3305_IFBW_2, .val = 0x00, },
+ { .reg = 0x030a, .val = 0x08, },
+ { .reg = 0x030b, .val = 0x9b, },
+ { .reg = 0x030d, .val = 0x00, },
+ { .reg = 0x030e, .val = 0x1c, },
+ { .reg = 0x0314, .val = 0xe1, },
+ { .reg = 0x000d, .val = 0x82, },
+ { .reg = LGDT3305_TP_CTRL_1, .val = 0x5b, },
+ { .reg = LGDT3305_TP_CTRL_1, .val = 0x5b, },
+ };
+
+ lg_dbg("\n");
+
+ ret = lgdt3305_write_regs(state, lgdt3304_init_data,
+ ARRAY_SIZE(lgdt3304_init_data));
+ if (lg_fail(ret))
+ goto fail;
+
+ ret = lgdt3305_soft_reset(state);
+fail:
+ return ret;
+}
+
static int lgdt3305_init(struct dvb_frontend *fe)
{
struct lgdt3305_state *state = fe->demodulator_priv;
@@ -639,6 +706,88 @@ fail:
return ret;
}
+static int lgdt3304_set_parameters(struct dvb_frontend *fe,
+ struct dvb_frontend_parameters *param)
+{
+ struct lgdt3305_state *state = fe->demodulator_priv;
+ int ret;
+
+ lg_dbg("(%d, %d)\n", param->frequency, param->u.vsb.modulation);
+
+ if (fe->ops.tuner_ops.set_params) {
+ ret = fe->ops.tuner_ops.set_params(fe, param);
+ if (fe->ops.i2c_gate_ctrl)
+ fe->ops.i2c_gate_ctrl(fe, 0);
+ if (lg_fail(ret))
+ goto fail;
+ state->current_frequency = param->frequency;
+ }
+
+ ret = lgdt3305_set_modulation(state, param);
+ if (lg_fail(ret))
+ goto fail;
+
+ ret = lgdt3305_passband_digital_agc(state, param);
+ if (lg_fail(ret))
+ goto fail;
+
+ ret = lgdt3305_agc_setup(state, param);
+ if (lg_fail(ret))
+ goto fail;
+
+ /* reg 0x030d is 3304-only... seen in vsb and qam usbsnoops... */
+ switch (param->u.vsb.modulation) {
+ case VSB_8:
+ lgdt3305_write_reg(state, 0x030d, 0x00);
+#if 1
+ lgdt3305_write_reg(state, LGDT3305_CR_CTR_FREQ_1, 0x4f);
+ lgdt3305_write_reg(state, LGDT3305_CR_CTR_FREQ_2, 0x0c);
+ lgdt3305_write_reg(state, LGDT3305_CR_CTR_FREQ_3, 0xac);
+ lgdt3305_write_reg(state, LGDT3305_CR_CTR_FREQ_4, 0xba);
+#endif
+ break;
+ case QAM_64:
+ case QAM_256:
+ lgdt3305_write_reg(state, 0x030d, 0x14);
+#if 1
+ ret = lgdt3305_set_if(state, param);
+ if (lg_fail(ret))
+ goto fail;
+#endif
+ break;
+ default:
+ return -EINVAL;
+ }
+
+#if 0
+ /* the set_if vsb formula doesn't work for the 3304, we end up sending
+ * 0x40851e07 instead of 0x4f0cacba (which works back to 94050, rather
+ * than 3250, in the case of the kworld 340u) */
+ ret = lgdt3305_set_if(state, param);
+ if (lg_fail(ret))
+ goto fail;
+#endif
+
+ ret = lgdt3305_spectral_inversion(state, param,
+ state->cfg->spectral_inversion
+ ? 1 : 0);
+ if (lg_fail(ret))
+ goto fail;
+
+ state->current_modulation = param->u.vsb.modulation;
+
+ ret = lgdt3305_mpeg_mode(state, state->cfg->mpeg_mode);
+ if (lg_fail(ret))
+ goto fail;
+
+ /* lgdt3305_mpeg_mode_polarity calls lgdt3305_soft_reset */
+ ret = lgdt3305_mpeg_mode_polarity(state,
+ state->cfg->tpclk_edge,
+ state->cfg->tpvalid_polarity);
+fail:
+ return ret;
+}
+
static int lgdt3305_set_parameters(struct dvb_frontend *fe,
struct dvb_frontend_parameters *param)
{
@@ -847,6 +996,10 @@ static int lgdt3305_read_status(struct dvb_frontend *fe, fe_status_t *status)
switch (state->current_modulation) {
case QAM_256:
case QAM_64:
+#if 0 /* needed w/3304 to set FE_HAS_SIGNAL */
+ if (cr_lock)
+ *status |= FE_HAS_SIGNAL;
+#endif
ret = lgdt3305_read_fec_lock_status(state, &fec_lock);
if (lg_fail(ret))
goto fail;
@@ -992,6 +1145,7 @@ static void lgdt3305_release(struct dvb_frontend *fe)
kfree(state);
}
+static struct dvb_frontend_ops lgdt3304_ops;
static struct dvb_frontend_ops lgdt3305_ops;
struct dvb_frontend *lgdt3305_attach(const struct lgdt3305_config *config,
@@ -1012,11 +1166,21 @@ struct dvb_frontend *lgdt3305_attach(const struct lgdt3305_config *config,
state->cfg = config;
state->i2c_adap = i2c_adap;
- memcpy(&state->frontend.ops, &lgdt3305_ops,
- sizeof(struct dvb_frontend_ops));
+ switch (config->demod_chip) {
+ case LGDT3304:
+ memcpy(&state->frontend.ops, &lgdt3304_ops,
+ sizeof(struct dvb_frontend_ops));
+ break;
+ case LGDT3305:
+ memcpy(&state->frontend.ops, &lgdt3305_ops,
+ sizeof(struct dvb_frontend_ops));
+ break;
+ default:
+ goto fail;
+ }
state->frontend.demodulator_priv = state;
- /* verify that we're talking to a lg dt3305 */
+ /* verify that we're talking to a lg dt3304/5 */
ret = lgdt3305_read_reg(state, LGDT3305_GEN_CTRL_2, &val);
if ((lg_fail(ret)) | (val == 0))
goto fail;
@@ -1035,12 +1199,36 @@ struct dvb_frontend *lgdt3305_attach(const struct lgdt3305_config *config,
return &state->frontend;
fail:
- lg_warn("unable to detect LGDT3305 hardware\n");
+ lg_warn("unable to detect %s hardware\n",
+ config->demod_chip ? "LGDT3304" : "LGDT3305");
kfree(state);
return NULL;
}
EXPORT_SYMBOL(lgdt3305_attach);
+static struct dvb_frontend_ops lgdt3304_ops = {
+ .info = {
+ .name = "LG Electronics LGDT3304 VSB/QAM Frontend",
+ .type = FE_ATSC,
+ .frequency_min = 54000000,
+ .frequency_max = 858000000,
+ .frequency_stepsize = 62500,
+ .caps = FE_CAN_QAM_64 | FE_CAN_QAM_256 | FE_CAN_8VSB
+ },
+ .i2c_gate_ctrl = lgdt3305_i2c_gate_ctrl,
+ .init = lgdt3304_init,
+ .sleep = lgdt3304_sleep,
+ .set_frontend = lgdt3304_set_parameters,
+ .get_frontend = lgdt3305_get_frontend,
+ .get_tune_settings = lgdt3305_get_tune_settings,
+ .read_status = lgdt3305_read_status,
+ .read_ber = lgdt3305_read_ber,
+ .read_signal_strength = lgdt3305_read_signal_strength,
+ .read_snr = lgdt3305_read_snr,
+ .read_ucblocks = lgdt3305_read_ucblocks,
+ .release = lgdt3305_release,
+};
+
static struct dvb_frontend_ops lgdt3305_ops = {
.info = {
.name = "LG Electronics LGDT3305 VSB/QAM Frontend",
@@ -1064,7 +1252,7 @@ static struct dvb_frontend_ops lgdt3305_ops = {
.release = lgdt3305_release,
};
-MODULE_DESCRIPTION("LG Electronics LGDT3305 ATSC/QAM-B Demodulator Driver");
+MODULE_DESCRIPTION("LG Electronics LGDT3304/5 ATSC/QAM-B Demodulator Driver");
MODULE_AUTHOR("Michael Krufky <mkrufky@linuxtv.org>");
MODULE_LICENSE("GPL");
MODULE_VERSION("0.1");
diff --git a/drivers/media/dvb/frontends/lgdt3305.h b/drivers/media/dvb/frontends/lgdt3305.h
index 9cb11c9..a7f30c2 100644
--- a/drivers/media/dvb/frontends/lgdt3305.h
+++ b/drivers/media/dvb/frontends/lgdt3305.h
@@ -41,6 +41,11 @@ enum lgdt3305_tp_valid_polarity {
LGDT3305_TP_VALID_HIGH = 1,
};
+enum lgdt_demod_chip_type {
+ LGDT3305 = 0,
+ LGDT3304 = 1,
+};
+
struct lgdt3305_config {
u8 i2c_addr;
@@ -65,6 +70,7 @@ struct lgdt3305_config {
enum lgdt3305_mpeg_mode mpeg_mode;
enum lgdt3305_tp_clock_edge tpclk_edge;
enum lgdt3305_tp_valid_polarity tpvalid_polarity;
+ enum lgdt_demod_chip_type demod_chip;
};
#if defined(CONFIG_DVB_LGDT3305) || (defined(CONFIG_DVB_LGDT3305_MODULE) && \
--
1.6.6

View File

@ -1,142 +0,0 @@
drivers/media/IR/imon.c | 20 +-------------------
drivers/media/IR/mceusb.c | 15 +--------------
2 files changed, 2 insertions(+), 33 deletions(-)
diff --git a/drivers/media/IR/imon.c b/drivers/media/IR/imon.c
index 65c125e..c185422 100644
--- a/drivers/media/IR/imon.c
+++ b/drivers/media/IR/imon.c
@@ -87,7 +87,6 @@ static ssize_t lcd_write(struct file *file, const char *buf,
struct imon_context {
struct device *dev;
struct ir_dev_props *props;
- struct ir_input_dev *ir;
/* Newer devices have two interfaces */
struct usb_device *usbdev_intf0;
struct usb_device *usbdev_intf1;
@@ -1656,7 +1655,6 @@ static struct input_dev *imon_init_idev(struct imon_context *ictx)
{
struct input_dev *idev;
struct ir_dev_props *props;
- struct ir_input_dev *ir;
int ret, i;
idev = input_allocate_device();
@@ -1671,12 +1669,6 @@ static struct input_dev *imon_init_idev(struct imon_context *ictx)
goto props_alloc_failed;
}
- ir = kzalloc(sizeof(struct ir_input_dev), GFP_KERNEL);
- if (!ir) {
- dev_err(ictx->dev, "remote ir input dev allocation failed\n");
- goto ir_dev_alloc_failed;
- }
-
snprintf(ictx->name_idev, sizeof(ictx->name_idev),
"iMON Remote (%04x:%04x)", ictx->vendor, ictx->product);
idev->name = ictx->name_idev;
@@ -1706,14 +1698,9 @@ static struct input_dev *imon_init_idev(struct imon_context *ictx)
props->change_protocol = imon_ir_change_protocol;
ictx->props = props;
- ictx->ir = ir;
- memcpy(&ir->dev, ictx->dev, sizeof(struct device));
-
usb_to_input_id(ictx->usbdev_intf0, &idev->id);
idev->dev.parent = ictx->dev;
- input_set_drvdata(idev, ir);
-
ret = ir_input_register(idev, RC_MAP_IMON_PAD, props, MOD_NAME);
if (ret < 0) {
dev_err(ictx->dev, "remote input dev register failed\n");
@@ -1723,8 +1710,6 @@ static struct input_dev *imon_init_idev(struct imon_context *ictx)
return idev;
idev_register_failed:
- kfree(ir);
-ir_dev_alloc_failed:
kfree(props);
props_alloc_failed:
input_free_device(idev);
@@ -1944,7 +1929,6 @@ static struct imon_context *imon_init_intf0(struct usb_interface *intf)
urb_submit_failed:
ir_input_unregister(ictx->idev);
- input_free_device(ictx->idev);
idev_setup_failed:
find_endpoint_failed:
mutex_unlock(&ictx->lock);
@@ -2014,10 +1998,8 @@ static struct imon_context *imon_init_intf1(struct usb_interface *intf,
return ictx;
urb_submit_failed:
- if (ictx->touch) {
+ if (ictx->touch)
input_unregister_device(ictx->touch);
- input_free_device(ictx->touch);
- }
touch_setup_failed:
find_endpoint_failed:
mutex_unlock(&ictx->lock);
diff --git a/drivers/media/IR/mceusb.c b/drivers/media/IR/mceusb.c
index 78bf7f7..65b0738 100644
--- a/drivers/media/IR/mceusb.c
+++ b/drivers/media/IR/mceusb.c
@@ -228,7 +228,6 @@ static struct usb_device_id std_tx_mask_list[] = {
/* data structure for each usb transceiver */
struct mceusb_dev {
/* ir-core bits */
- struct ir_input_dev *irdev;
struct ir_dev_props *props;
struct ir_raw_event rawir;
@@ -739,7 +738,7 @@ static void mceusb_dev_recv(struct urb *urb, struct pt_regs *regs)
if (ir->send_flags == RECV_FLAG_IN_PROGRESS) {
ir->send_flags = SEND_FLAG_COMPLETE;
- dev_dbg(&ir->irdev->dev, "setup answer received %d bytes\n",
+ dev_dbg(ir->dev, "setup answer received %d bytes\n",
buf_len);
}
@@ -861,7 +860,6 @@ static struct input_dev *mceusb_init_input_dev(struct mceusb_dev *ir)
{
struct input_dev *idev;
struct ir_dev_props *props;
- struct ir_input_dev *irdev;
struct device *dev = ir->dev;
int ret = -ENODEV;
@@ -878,12 +876,6 @@ static struct input_dev *mceusb_init_input_dev(struct mceusb_dev *ir)
goto props_alloc_failed;
}
- irdev = kzalloc(sizeof(struct ir_input_dev), GFP_KERNEL);
- if (!irdev) {
- dev_err(dev, "remote ir input dev allocation failed\n");
- goto ir_dev_alloc_failed;
- }
-
snprintf(ir->name, sizeof(ir->name), "Media Center Ed. eHome "
"Infrared Remote Transceiver (%04x:%04x)",
le16_to_cpu(ir->usbdev->descriptor.idVendor),
@@ -902,9 +894,6 @@ static struct input_dev *mceusb_init_input_dev(struct mceusb_dev *ir)
props->tx_ir = mceusb_tx_ir;
ir->props = props;
- ir->irdev = irdev;
-
- input_set_drvdata(idev, irdev);
ret = ir_input_register(idev, RC_MAP_RC6_MCE, props, DRIVER_NAME);
if (ret < 0) {
@@ -915,8 +904,6 @@ static struct input_dev *mceusb_init_input_dev(struct mceusb_dev *ir)
return idev;
irdev_failed:
- kfree(irdev);
-ir_dev_alloc_failed:
kfree(props);
props_alloc_failed:
input_free_device(idev);

File diff suppressed because it is too large Load Diff

File diff suppressed because it is too large Load Diff

View File

@ -1,362 +0,0 @@
From: Martin Rubli <martin_rubli@logitech.com>
Date: Wed, 19 May 2010 22:51:56 +0000 (+0200)
Subject: uvcvideo: Add support for absolute pan/tilt controls
X-Git-Url: http://git.linuxtv.org/pinchartl/uvcvideo.git?a=commitdiff_plain;h=d3c2f664ec76aff14c3841c99e84cd78d7227f79
uvcvideo: Add support for absolute pan/tilt controls
Signed-off-by: Martin Rubli <martin_rubli@logitech.com>
---
diff --git a/drivers/media/video/uvc/uvc_ctrl.c b/drivers/media/video/uvc/uvc_ctrl.c
index aa0720a..5ec2f4a 100644
--- a/drivers/media/video/uvc/uvc_ctrl.c
+++ b/drivers/media/video/uvc/uvc_ctrl.c
@@ -606,6 +606,26 @@ static struct uvc_control_mapping uvc_ctrl_mappings[] = {
.set = uvc_ctrl_set_zoom,
},
{
+ .id = V4L2_CID_PAN_ABSOLUTE,
+ .name = "Pan (Absolute)",
+ .entity = UVC_GUID_UVC_CAMERA,
+ .selector = UVC_CT_PANTILT_ABSOLUTE_CONTROL,
+ .size = 32,
+ .offset = 0,
+ .v4l2_type = V4L2_CTRL_TYPE_INTEGER,
+ .data_type = UVC_CTRL_DATA_TYPE_UNSIGNED,
+ },
+ {
+ .id = V4L2_CID_TILT_ABSOLUTE,
+ .name = "Tilt (Absolute)",
+ .entity = UVC_GUID_UVC_CAMERA,
+ .selector = UVC_CT_PANTILT_ABSOLUTE_CONTROL,
+ .size = 32,
+ .offset = 32,
+ .v4l2_type = V4L2_CTRL_TYPE_INTEGER,
+ .data_type = UVC_CTRL_DATA_TYPE_UNSIGNED,
+ },
+ {
.id = V4L2_CID_PRIVACY,
.name = "Privacy",
.entity = UVC_GUID_UVC_CAMERA,
From: Hans de Goede <hdegoede@redhat.com>
Date: Wed, 19 May 2010 23:15:00 +0000 (+0200)
Subject: uvcvideo: Make button controls work properly
X-Git-Url: http://git.linuxtv.org/pinchartl/uvcvideo.git?a=commitdiff_plain;h=2bd47ad4894bfaf1a97660b821cbc46439a614d6
uvcvideo: Make button controls work properly
According to the v4l2 spec, writing any value to a button control should
result in the action belonging to the button control being triggered.
UVC cams however want to see a 1 written, this patch fixes this by
overriding whatever value user space passed in with -1 (0xffffffff) when
the control is a button control.
Signed-off-by: Hans de Goede <hdegoede@redhat.com>
---
diff --git a/drivers/media/video/uvc/uvc_ctrl.c b/drivers/media/video/uvc/uvc_ctrl.c
index 5ec2f4a..8bb825d 100644
--- a/drivers/media/video/uvc/uvc_ctrl.c
+++ b/drivers/media/video/uvc/uvc_ctrl.c
@@ -698,6 +698,14 @@ static void uvc_set_le_value(struct uvc_control_mapping *mapping,
int offset = mapping->offset;
__u8 mask;
+ /* According to the v4l2 spec, writing any value to a button control
+ * should result in the action belonging to the button control being
+ * triggered. UVC devices however want to see a 1 written -> override
+ * value.
+ */
+ if (mapping->v4l2_type == V4L2_CTRL_TYPE_BUTTON)
+ value = -1;
+
data += offset / 8;
offset &= 7;
From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Date: Thu, 18 Feb 2010 19:38:52 +0000 (+0100)
Subject: uvcvideo: Support menu controls in the control mapping API
X-Git-Url: http://git.linuxtv.org/pinchartl/uvcvideo.git?a=commitdiff_plain;h=4930f2662e47d33e5baedac620da401a225bc3a8
uvcvideo: Support menu controls in the control mapping API
The UVCIOC_CTRL_MAP ioctl doesn't support menu entries for menu
controls. As the uvc_xu_control_mapping structure has no reserved
fields, this can't be fixed while keeping ABI compatibility.
Modify the UVCIOC_CTRL_MAP ioctl to add menu entries support, and define
UVCIOC_CTRL_MAP_OLD that supports the old ABI without any ability to add
menu controls.
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
diff --git a/drivers/media/video/uvc/uvc_ctrl.c b/drivers/media/video/uvc/uvc_ctrl.c
index 8bb825d..c88d72e 100644
--- a/drivers/media/video/uvc/uvc_ctrl.c
+++ b/drivers/media/video/uvc/uvc_ctrl.c
@@ -1606,6 +1606,28 @@ void uvc_ctrl_cleanup_device(struct uvc_device *dev)
}
}
+void uvc_ctrl_cleanup(void)
+{
+ struct uvc_control_info *info;
+ struct uvc_control_info *ni;
+ struct uvc_control_mapping *mapping;
+ struct uvc_control_mapping *nm;
+
+ list_for_each_entry_safe(info, ni, &uvc_driver.controls, list) {
+ if (!(info->flags & UVC_CONTROL_EXTENSION))
+ continue;
+
+ list_for_each_entry_safe(mapping, nm, &info->mappings, list) {
+ list_del(&mapping->list);
+ kfree(mapping->menu_info);
+ kfree(mapping);
+ }
+
+ list_del(&info->list);
+ kfree(info);
+ }
+}
+
void uvc_ctrl_init(void)
{
struct uvc_control_info *ctrl = uvc_ctrls;
diff --git a/drivers/media/video/uvc/uvc_driver.c b/drivers/media/video/uvc/uvc_driver.c
index 838b56f..34818c1 100644
--- a/drivers/media/video/uvc/uvc_driver.c
+++ b/drivers/media/video/uvc/uvc_driver.c
@@ -2261,6 +2261,7 @@ static int __init uvc_init(void)
static void __exit uvc_cleanup(void)
{
usb_deregister(&uvc_driver.driver);
+ uvc_ctrl_cleanup();
}
module_init(uvc_init);
diff --git a/drivers/media/video/uvc/uvc_v4l2.c b/drivers/media/video/uvc/uvc_v4l2.c
index 7c9ab29..485a899 100644
--- a/drivers/media/video/uvc/uvc_v4l2.c
+++ b/drivers/media/video/uvc/uvc_v4l2.c
@@ -29,6 +29,71 @@
#include "uvcvideo.h"
/* ------------------------------------------------------------------------
+ * UVC ioctls
+ */
+static int uvc_ioctl_ctrl_map(struct uvc_xu_control_mapping *xmap, int old)
+{
+ struct uvc_control_mapping *map;
+ unsigned int size;
+ int ret;
+
+ map = kzalloc(sizeof *map, GFP_KERNEL);
+ if (map == NULL)
+ return -ENOMEM;
+
+ map->id = xmap->id;
+ memcpy(map->name, xmap->name, sizeof map->name);
+ memcpy(map->entity, xmap->entity, sizeof map->entity);
+ map->selector = xmap->selector;
+ map->size = xmap->size;
+ map->offset = xmap->offset;
+ map->v4l2_type = xmap->v4l2_type;
+ map->data_type = xmap->data_type;
+
+ switch (xmap->v4l2_type) {
+ case V4L2_CTRL_TYPE_INTEGER:
+ case V4L2_CTRL_TYPE_BOOLEAN:
+ case V4L2_CTRL_TYPE_BUTTON:
+ break;
+
+ case V4L2_CTRL_TYPE_MENU:
+ if (old) {
+ ret = -EINVAL;
+ goto done;
+ }
+
+ size = xmap->menu_count * sizeof(*map->menu_info);
+ map->menu_info = kmalloc(size, GFP_KERNEL);
+ if (map->menu_info == NULL) {
+ ret = -ENOMEM;
+ goto done;
+ }
+
+ if (copy_from_user(map->menu_info, xmap->menu_info, size)) {
+ ret = -EFAULT;
+ goto done;
+ }
+
+ map->menu_count = xmap->menu_count;
+ break;
+
+ default:
+ ret = -EINVAL;
+ goto done;
+ }
+
+ ret = uvc_ctrl_add_mapping(map);
+
+done:
+ if (ret < 0) {
+ kfree(map->menu_info);
+ kfree(map);
+ }
+
+ return ret;
+}
+
+/* ------------------------------------------------------------------------
* V4L2 interface
*/
@@ -974,7 +1039,8 @@ static long uvc_v4l2_do_ioctl(struct file *file, unsigned int cmd, void *arg)
info->flags = xinfo->flags;
info->flags |= UVC_CONTROL_GET_MIN | UVC_CONTROL_GET_MAX |
- UVC_CONTROL_GET_RES | UVC_CONTROL_GET_DEF;
+ UVC_CONTROL_GET_RES | UVC_CONTROL_GET_DEF |
+ UVC_CONTROL_EXTENSION;
ret = uvc_ctrl_add_info(info);
if (ret < 0)
@@ -982,32 +1048,12 @@ static long uvc_v4l2_do_ioctl(struct file *file, unsigned int cmd, void *arg)
break;
}
+ case UVCIOC_CTRL_MAP_OLD:
case UVCIOC_CTRL_MAP:
- {
- struct uvc_xu_control_mapping *xmap = arg;
- struct uvc_control_mapping *map;
-
if (!capable(CAP_SYS_ADMIN))
return -EPERM;
- map = kzalloc(sizeof *map, GFP_KERNEL);
- if (map == NULL)
- return -ENOMEM;
-
- map->id = xmap->id;
- memcpy(map->name, xmap->name, sizeof map->name);
- memcpy(map->entity, xmap->entity, sizeof map->entity);
- map->selector = xmap->selector;
- map->size = xmap->size;
- map->offset = xmap->offset;
- map->v4l2_type = xmap->v4l2_type;
- map->data_type = xmap->data_type;
-
- ret = uvc_ctrl_add_mapping(map);
- if (ret < 0)
- kfree(map);
- break;
- }
+ return uvc_ioctl_ctrl_map(arg, cmd == UVCIOC_CTRL_MAP_OLD);
case UVCIOC_CTRL_GET:
return uvc_xu_ctrl_query(chain, arg, 0);
diff --git a/drivers/media/video/uvc/uvcvideo.h b/drivers/media/video/uvc/uvcvideo.h
index d1f8840..14f77e4 100644
--- a/drivers/media/video/uvc/uvcvideo.h
+++ b/drivers/media/video/uvc/uvcvideo.h
@@ -27,6 +27,8 @@
#define UVC_CONTROL_RESTORE (1 << 6)
/* Control can be updated by the camera. */
#define UVC_CONTROL_AUTO_UPDATE (1 << 7)
+/* Control is an extension unit control. */
+#define UVC_CONTROL_EXTENSION (1 << 8)
#define UVC_CONTROL_GET_RANGE (UVC_CONTROL_GET_CUR | UVC_CONTROL_GET_MIN | \
UVC_CONTROL_GET_MAX | UVC_CONTROL_GET_RES | \
@@ -40,6 +42,15 @@ struct uvc_xu_control_info {
__u32 flags;
};
+struct uvc_menu_info {
+ __u32 value;
+ __u8 name[32];
+};
+
+struct uvc_xu_control_mapping_old {
+ __u8 reserved[64];
+};
+
struct uvc_xu_control_mapping {
__u32 id;
__u8 name[32];
@@ -50,6 +61,11 @@ struct uvc_xu_control_mapping {
__u8 offset;
enum v4l2_ctrl_type v4l2_type;
__u32 data_type;
+
+ struct uvc_menu_info __user *menu_info;
+ __u32 menu_count;
+
+ __u32 reserved[4];
};
struct uvc_xu_control {
@@ -60,6 +76,7 @@ struct uvc_xu_control {
};
#define UVCIOC_CTRL_ADD _IOW('U', 1, struct uvc_xu_control_info)
+#define UVCIOC_CTRL_MAP_OLD _IOWR('U', 2, struct uvc_xu_control_mapping_old)
#define UVCIOC_CTRL_MAP _IOWR('U', 2, struct uvc_xu_control_mapping)
#define UVCIOC_CTRL_GET _IOWR('U', 3, struct uvc_xu_control)
#define UVCIOC_CTRL_SET _IOW('U', 4, struct uvc_xu_control)
@@ -198,11 +215,6 @@ struct uvc_streaming_control {
__u8 bMaxVersion;
};
-struct uvc_menu_info {
- __u32 value;
- __u8 name[32];
-};
-
struct uvc_control_info {
struct list_head list;
struct list_head mappings;
@@ -625,6 +637,7 @@ extern int uvc_ctrl_init_device(struct uvc_device *dev);
extern void uvc_ctrl_cleanup_device(struct uvc_device *dev);
extern int uvc_ctrl_resume_device(struct uvc_device *dev);
extern void uvc_ctrl_init(void);
+extern void uvc_ctrl_cleanup(void);
extern int uvc_ctrl_begin(struct uvc_video_chain *chain);
extern int __uvc_ctrl_commit(struct uvc_video_chain *chain, int rollback);
From: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
Date: Fri, 25 Jun 2010 07:58:43 +0000 (+0200)
Subject: uvcvideo: Add support for Manta MM-353 Plako
X-Git-Url: http://git.linuxtv.org/pinchartl/uvcvideo.git?a=commitdiff_plain;h=352e661e1f347390a86cf34bc5e41adbdd1caa41
uvcvideo: Add support for Manta MM-353 Plako
The camera requires the PROBE_MINMAX quirk. Add a corresponding entry
in the device IDs list
Signed-off-by: Laurent Pinchart <laurent.pinchart@ideasonboard.com>
---
diff --git a/drivers/media/video/uvc/uvc_driver.c b/drivers/media/video/uvc/uvc_driver.c
index 34818c1..1a89384 100644
--- a/drivers/media/video/uvc/uvc_driver.c
+++ b/drivers/media/video/uvc/uvc_driver.c
@@ -2174,6 +2174,15 @@ static struct usb_device_id uvc_ids[] = {
.bInterfaceSubClass = 1,
.bInterfaceProtocol = 0,
.driver_info = UVC_QUIRK_PROBE_EXTRAFIELDS },
+ /* Manta MM-353 Plako */
+ { .match_flags = USB_DEVICE_ID_MATCH_DEVICE
+ | USB_DEVICE_ID_MATCH_INT_INFO,
+ .idVendor = 0x18ec,
+ .idProduct = 0x3188,
+ .bInterfaceClass = USB_CLASS_VIDEO,
+ .bInterfaceSubClass = 1,
+ .bInterfaceProtocol = 0,
+ .driver_info = UVC_QUIRK_PROBE_MINMAX },
/* FSC WebCam V30S */
{ .match_flags = USB_DEVICE_ID_MATCH_DEVICE
| USB_DEVICE_ID_MATCH_INT_INFO,

View File

@ -0,0 +1,28 @@
Date: Wed, 01 Sep 2010 13:37:33 -0700 (PDT)
Message-Id: <20100901.133733.223467599.davem@davemloft.net>
To: davej@redhat.com
Cc: simon.kagstrom@netinsight.net, netdev@vger.kernel.org
Subject: Re: via-velocity dma-debug warnings again. (2.6.35.2)
From: David Miller <davem@davemloft.net>
In-Reply-To: <20100901.133547.236248297.davem@davemloft.net>
References: <20100901200555.GA30689@redhat.com>
<20100901.133414.24593005.davem@davemloft.net>
<20100901.133547.236248297.davem@davemloft.net>
Ok, this is becomming hopeless. Let's just try turning off SG support
for now, the length calculations are correct in those cases.
diff --git a/drivers/net/via-velocity.c b/drivers/net/via-velocity.c
index fd69095..f534123 100644
--- a/drivers/net/via-velocity.c
+++ b/drivers/net/via-velocity.c
@@ -2824,7 +2824,7 @@ static int __devinit velocity_found1(struct pci_dev *pdev, const struct pci_devi
netif_napi_add(dev, &vptr->napi, velocity_poll, VELOCITY_NAPI_WEIGHT);
dev->features |= NETIF_F_HW_VLAN_TX | NETIF_F_HW_VLAN_FILTER |
- NETIF_F_HW_VLAN_RX | NETIF_F_IP_CSUM | NETIF_F_SG;
+ NETIF_F_HW_VLAN_RX | NETIF_F_IP_CSUM;
ret = register_netdev(dev);
if (ret < 0)

File diff suppressed because it is too large Load Diff

View File

@ -0,0 +1,390 @@
From 9ced9810f0450a7f05eccb40dce4f9e4616c0fb6 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 24 Nov 2010 22:18:23 -0500
Subject: [PATCH 1/2] mm: page allocator: Adjust the per-cpu counter threshold when memory is low
Commit aa45484 ("calculate a better estimate of NR_FREE_PAGES when memory
is low") noted that watermarks were based on the vmstat NR_FREE_PAGES. To
avoid synchronization overhead, these counters are maintained on a per-cpu
basis and drained both periodically and when a threshold is above a
threshold. On large CPU systems, the difference between the estimate and
real value of NR_FREE_PAGES can be very high. The system can get into a
case where pages are allocated far below the min watermark potentially
causing livelock issues. The commit solved the problem by taking a better
reading of NR_FREE_PAGES when memory was low.
Unfortately, as reported by Shaohua Li this accurate reading can consume a
large amount of CPU time on systems with many sockets due to cache line
bouncing. This patch takes a different approach. For large machines
where counter drift might be unsafe and while kswapd is awake, the per-cpu
thresholds for the target pgdat are reduced to limit the level of drift to
what should be a safe level. This incurs a performance penalty in heavy
memory pressure by a factor that depends on the workload and the machine
but the machine should function correctly without accidentally exhausting
all memory on a node. There is an additional cost when kswapd wakes and
sleeps but the event is not expected to be frequent - in Shaohua's test
case, there was one recorded sleep and wake event at least.
To ensure that kswapd wakes up, a safe version of zone_watermark_ok() is
introduced that takes a more accurate reading of NR_FREE_PAGES when called
from wakeup_kswapd, when deciding whether it is really safe to go back to
sleep in sleeping_prematurely() and when deciding if a zone is really
balanced or not in balance_pgdat(). We are still using an expensive
function but limiting how often it is called.
When the test case is reproduced, the time spent in the watermark
functions is reduced. The following report is on the percentage of time
spent cumulatively spent in the functions zone_nr_free_pages(),
zone_watermark_ok(), __zone_watermark_ok(), zone_watermark_ok_safe(),
zone_page_state_snapshot(), zone_page_state().
vanilla 11.6615%
disable-threshold 0.2584%
Reported-by: Shaohua Li <shaohua.li@intel.com>
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: Christoph Lameter <cl@linux.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
include/linux/mmzone.h | 10 ++-----
include/linux/vmstat.h | 5 +++
mm/mmzone.c | 21 ---------------
mm/page_alloc.c | 35 +++++++++++++++++++-----
mm/vmscan.c | 26 ++++++++++--------
mm/vmstat.c | 68 +++++++++++++++++++++++++++++++++++++++++++++++-
6 files changed, 117 insertions(+), 48 deletions(-)
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index 8b2db3d..1e3d0b4 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -463,12 +463,6 @@ static inline int zone_is_oom_locked(const struct zone *zone)
return test_bit(ZONE_OOM_LOCKED, &zone->flags);
}
-#ifdef CONFIG_SMP
-unsigned long zone_nr_free_pages(struct zone *zone);
-#else
-#define zone_nr_free_pages(zone) zone_page_state(zone, NR_FREE_PAGES)
-#endif /* CONFIG_SMP */
-
/*
* The "priority" of VM scanning is how much of the queues we will scan in one
* go. A value of 12 for DEF_PRIORITY implies that we will scan 1/4096th of the
@@ -668,7 +662,9 @@ void get_zone_counts(unsigned long *active, unsigned long *inactive,
unsigned long *free);
void build_all_zonelists(void *data);
void wakeup_kswapd(struct zone *zone, int order);
-int zone_watermark_ok(struct zone *z, int order, unsigned long mark,
+bool zone_watermark_ok(struct zone *z, int order, unsigned long mark,
+ int classzone_idx, int alloc_flags);
+bool zone_watermark_ok_safe(struct zone *z, int order, unsigned long mark,
int classzone_idx, int alloc_flags);
enum memmap_context {
MEMMAP_EARLY,
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index eaaea37..e4cc21c 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -254,6 +254,8 @@ extern void dec_zone_state(struct zone *, enum zone_stat_item);
extern void __dec_zone_state(struct zone *, enum zone_stat_item);
void refresh_cpu_vm_stats(int);
+void reduce_pgdat_percpu_threshold(pg_data_t *pgdat);
+void restore_pgdat_percpu_threshold(pg_data_t *pgdat);
#else /* CONFIG_SMP */
/*
@@ -298,6 +300,9 @@ static inline void __dec_zone_page_state(struct page *page,
#define dec_zone_page_state __dec_zone_page_state
#define mod_zone_page_state __mod_zone_page_state
+static inline void reduce_pgdat_percpu_threshold(pg_data_t *pgdat) { }
+static inline void restore_pgdat_percpu_threshold(pg_data_t *pgdat) { }
+
static inline void refresh_cpu_vm_stats(int cpu) { }
#endif
diff --git a/mm/mmzone.c b/mm/mmzone.c
index e35bfb8..f5b7d17 100644
--- a/mm/mmzone.c
+++ b/mm/mmzone.c
@@ -87,24 +87,3 @@ int memmap_valid_within(unsigned long pfn,
return 1;
}
#endif /* CONFIG_ARCH_HAS_HOLES_MEMORYMODEL */
-
-#ifdef CONFIG_SMP
-/* Called when a more accurate view of NR_FREE_PAGES is needed */
-unsigned long zone_nr_free_pages(struct zone *zone)
-{
- unsigned long nr_free_pages = zone_page_state(zone, NR_FREE_PAGES);
-
- /*
- * While kswapd is awake, it is considered the zone is under some
- * memory pressure. Under pressure, there is a risk that
- * per-cpu-counter-drift will allow the min watermark to be breached
- * potentially causing a live-lock. While kswapd is awake and
- * free pages are low, get a better estimate for free pages
- */
- if (nr_free_pages < zone->percpu_drift_mark &&
- !waitqueue_active(&zone->zone_pgdat->kswapd_wait))
- return zone_page_state_snapshot(zone, NR_FREE_PAGES);
-
- return nr_free_pages;
-}
-#endif /* CONFIG_SMP */
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index f7cc624..cf5d4c0 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1454,24 +1454,24 @@ static inline int should_fail_alloc_page(gfp_t gfp_mask, unsigned int order)
#endif /* CONFIG_FAIL_PAGE_ALLOC */
/*
- * Return 1 if free pages are above 'mark'. This takes into account the order
+ * Return true if free pages are above 'mark'. This takes into account the order
* of the allocation.
*/
-int zone_watermark_ok(struct zone *z, int order, unsigned long mark,
- int classzone_idx, int alloc_flags)
+static bool __zone_watermark_ok(struct zone *z, int order, unsigned long mark,
+ int classzone_idx, int alloc_flags, long free_pages)
{
/* free_pages my go negative - that's OK */
long min = mark;
- long free_pages = zone_nr_free_pages(z) - (1 << order) + 1;
int o;
+ free_pages -= (1 << order) + 1;
if (alloc_flags & ALLOC_HIGH)
min -= min / 2;
if (alloc_flags & ALLOC_HARDER)
min -= min / 4;
if (free_pages <= min + z->lowmem_reserve[classzone_idx])
- return 0;
+ return false;
for (o = 0; o < order; o++) {
/* At the next order, this order's pages become unavailable */
free_pages -= z->free_area[o].nr_free << o;
@@ -1480,9 +1480,28 @@ int zone_watermark_ok(struct zone *z, int order, unsigned long mark,
min >>= 1;
if (free_pages <= min)
- return 0;
+ return false;
}
- return 1;
+ return true;
+}
+
+bool zone_watermark_ok(struct zone *z, int order, unsigned long mark,
+ int classzone_idx, int alloc_flags)
+{
+ return __zone_watermark_ok(z, order, mark, classzone_idx, alloc_flags,
+ zone_page_state(z, NR_FREE_PAGES));
+}
+
+bool zone_watermark_ok_safe(struct zone *z, int order, unsigned long mark,
+ int classzone_idx, int alloc_flags)
+{
+ long free_pages = zone_page_state(z, NR_FREE_PAGES);
+
+ if (z->percpu_drift_mark && free_pages < z->percpu_drift_mark)
+ free_pages = zone_page_state_snapshot(z, NR_FREE_PAGES);
+
+ return __zone_watermark_ok(z, order, mark, classzone_idx, alloc_flags,
+ free_pages);
}
#ifdef CONFIG_NUMA
@@ -2425,7 +2444,7 @@ void show_free_areas(void)
" all_unreclaimable? %s"
"\n",
zone->name,
- K(zone_nr_free_pages(zone)),
+ K(zone_page_state(zone, NR_FREE_PAGES)),
K(min_wmark_pages(zone)),
K(low_wmark_pages(zone)),
K(high_wmark_pages(zone)),
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 9753626..18f4038 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2007,7 +2007,7 @@ static int sleeping_prematurely(pg_data_t *pgdat, int order, long remaining)
if (zone->all_unreclaimable)
continue;
- if (!zone_watermark_ok(zone, order, high_wmark_pages(zone),
+ if (!zone_watermark_ok_safe(zone, order, high_wmark_pages(zone),
0, 0))
return 1;
}
@@ -2104,7 +2104,7 @@ loop_again:
shrink_active_list(SWAP_CLUSTER_MAX, zone,
&sc, priority, 0);
- if (!zone_watermark_ok(zone, order,
+ if (!zone_watermark_ok_safe(zone, order,
high_wmark_pages(zone), 0, 0)) {
end_zone = i;
break;
@@ -2155,7 +2155,7 @@ loop_again:
* We put equal pressure on every zone, unless one
* zone has way too many pages free already.
*/
- if (!zone_watermark_ok(zone, order,
+ if (!zone_watermark_ok_safe(zone, order,
8*high_wmark_pages(zone), end_zone, 0))
shrink_zone(priority, zone, &sc);
reclaim_state->reclaimed_slab = 0;
@@ -2176,7 +2176,7 @@ loop_again:
total_scanned > sc.nr_reclaimed + sc.nr_reclaimed / 2)
sc.may_writepage = 1;
- if (!zone_watermark_ok(zone, order,
+ if (!zone_watermark_ok_safe(zone, order,
high_wmark_pages(zone), end_zone, 0)) {
all_zones_ok = 0;
/*
@@ -2184,7 +2184,7 @@ loop_again:
* means that we have a GFP_ATOMIC allocation
* failure risk. Hurry up!
*/
- if (!zone_watermark_ok(zone, order,
+ if (!zone_watermark_ok_safe(zone, order,
min_wmark_pages(zone), end_zone, 0))
has_under_min_watermark_zone = 1;
}
@@ -2326,9 +2326,11 @@ static int kswapd(void *p)
* premature sleep. If not, then go fully
* to sleep until explicitly woken up
*/
- if (!sleeping_prematurely(pgdat, order, remaining))
+ if (!sleeping_prematurely(pgdat, order, remaining)) {
+ restore_pgdat_percpu_threshold(pgdat);
schedule();
- else {
+ reduce_pgdat_percpu_threshold(pgdat);
+ } else {
if (remaining)
count_vm_event(KSWAPD_LOW_WMARK_HIT_QUICKLY);
else
@@ -2364,15 +2366,16 @@ void wakeup_kswapd(struct zone *zone, int order)
if (!populated_zone(zone))
return;
- pgdat = zone->zone_pgdat;
- if (zone_watermark_ok(zone, order, low_wmark_pages(zone), 0, 0))
+ if (!cpuset_zone_allowed_hardwall(zone, GFP_KERNEL))
return;
+ pgdat = zone->zone_pgdat;
if (pgdat->kswapd_max_order < order)
pgdat->kswapd_max_order = order;
- if (!cpuset_zone_allowed_hardwall(zone, GFP_KERNEL))
- return;
if (!waitqueue_active(&pgdat->kswapd_wait))
return;
+ if (zone_watermark_ok_safe(zone, order, low_wmark_pages(zone), 0, 0))
+ return;
+
wake_up_interruptible(&pgdat->kswapd_wait);
}
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 26d5716..41dc8cd 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -81,6 +81,30 @@ EXPORT_SYMBOL(vm_stat);
#ifdef CONFIG_SMP
+static int calculate_pressure_threshold(struct zone *zone)
+{
+ int threshold;
+ int watermark_distance;
+
+ /*
+ * As vmstats are not up to date, there is drift between the estimated
+ * and real values. For high thresholds and a high number of CPUs, it
+ * is possible for the min watermark to be breached while the estimated
+ * value looks fine. The pressure threshold is a reduced value such
+ * that even the maximum amount of drift will not accidentally breach
+ * the min watermark
+ */
+ watermark_distance = low_wmark_pages(zone) - min_wmark_pages(zone);
+ threshold = max(1, (int)(watermark_distance / num_online_cpus()));
+
+ /*
+ * Maximum threshold is 125
+ */
+ threshold = min(125, threshold);
+
+ return threshold;
+}
+
static int calculate_threshold(struct zone *zone)
{
int threshold;
@@ -159,6 +183,48 @@ static void refresh_zone_stat_thresholds(void)
}
}
+void reduce_pgdat_percpu_threshold(pg_data_t *pgdat)
+{
+ struct zone *zone;
+ int cpu;
+ int threshold;
+ int i;
+
+ get_online_cpus();
+ for (i = 0; i < pgdat->nr_zones; i++) {
+ zone = &pgdat->node_zones[i];
+ if (!zone->percpu_drift_mark)
+ continue;
+
+ threshold = calculate_pressure_threshold(zone);
+ for_each_online_cpu(cpu)
+ per_cpu_ptr(zone->pageset, cpu)->stat_threshold
+ = threshold;
+ }
+ put_online_cpus();
+}
+
+void restore_pgdat_percpu_threshold(pg_data_t *pgdat)
+{
+ struct zone *zone;
+ int cpu;
+ int threshold;
+ int i;
+
+ get_online_cpus();
+ for (i = 0; i < pgdat->nr_zones; i++) {
+ zone = &pgdat->node_zones[i];
+ if (!zone->percpu_drift_mark)
+ continue;
+
+ threshold = calculate_threshold(zone);
+ for_each_online_cpu(cpu)
+ per_cpu_ptr(zone->pageset, cpu)->stat_threshold
+ = threshold;
+ }
+ put_online_cpus();
+}
+
/*
* For use when we know that interrupts are disabled.
*/
@@ -826,7 +892,7 @@ static void zoneinfo_show_print(struct seq_file *m, pg_data_t *pgdat,
"\n scanned %lu"
"\n spanned %lu"
"\n present %lu",
- zone_nr_free_pages(zone),
+ zone_page_state(zone, NR_FREE_PAGES),
min_wmark_pages(zone),
low_wmark_pages(zone),
high_wmark_pages(zone),
--
1.7.3.2

View File

@ -0,0 +1,165 @@
From 235d4a7af803e65d8fabdf9cd6d1445c11e2fd45 Mon Sep 17 00:00:00 2001
From: Mel Gorman <mel@csn.ul.ie>
Date: Wed, 24 Nov 2010 22:24:24 -0500
Subject: [PATCH 2/2] mm: vmstat: Use a single setter function and callback for adjusting percpu thresholds
reduce_pgdat_percpu_threshold() and restore_pgdat_percpu_threshold() exist
to adjust the per-cpu vmstat thresholds while kswapd is awake to avoid
errors due to counter drift. The functions duplicate some code so this
patch replaces them with a single set_pgdat_percpu_threshold() that takes
a callback function to calculate the desired threshold as a parameter.
Signed-off-by: Mel Gorman <mel@csn.ul.ie>
Reviewed-by: Christoph Lameter <cl@linux.com>
Reviewed-by: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
[the various mmotm patches updating this were rolled up. --kyle]
---
include/linux/vmstat.h | 10 ++++++----
mm/vmscan.c | 18 ++++++++++++++++--
mm/vmstat.c | 36 +++++++-----------------------------
3 files changed, 29 insertions(+), 35 deletions(-)
diff --git a/include/linux/vmstat.h b/include/linux/vmstat.h
index e4cc21c..833e676 100644
--- a/include/linux/vmstat.h
+++ b/include/linux/vmstat.h
@@ -254,8 +254,11 @@ extern void dec_zone_state(struct zone *, enum zone_stat_item);
extern void __dec_zone_state(struct zone *, enum zone_stat_item);
void refresh_cpu_vm_stats(int);
-void reduce_pgdat_percpu_threshold(pg_data_t *pgdat);
-void restore_pgdat_percpu_threshold(pg_data_t *pgdat);
+
+int calculate_pressure_threshold(struct zone *zone);
+int calculate_normal_threshold(struct zone *zone);
+void set_pgdat_percpu_threshold(pg_data_t *pgdat,
+ int (*calculate_pressure)(struct zone *));
#else /* CONFIG_SMP */
/*
@@ -300,8 +303,7 @@ static inline void __dec_zone_page_state(struct page *page,
#define dec_zone_page_state __dec_zone_page_state
#define mod_zone_page_state __mod_zone_page_state
-static inline void reduce_pgdat_percpu_threshold(pg_data_t *pgdat) { }
-static inline void restore_pgdat_percpu_threshold(pg_data_t *pgdat) { }
+#define set_pgdat_percpu_threshold(pgdat, callback) { }
static inline void refresh_cpu_vm_stats(int cpu) { }
#endif
diff --git a/mm/vmscan.c b/mm/vmscan.c
index 18f4038..e8df269 100644
--- a/mm/vmscan.c
+++ b/mm/vmscan.c
@@ -2327,9 +2327,23 @@ static int kswapd(void *p)
* to sleep until explicitly woken up
*/
if (!sleeping_prematurely(pgdat, order, remaining)) {
- restore_pgdat_percpu_threshold(pgdat);
+ /*
+ * vmstat counters are not perfectly
+ * accurate and the estimated value
+ * for counters such as NR_FREE_PAGES
+ * can deviate from the true value by
+ * nr_online_cpus * threshold. To
+ * avoid the zone watermarks being
+ * breached while under pressure, we
+ * reduce the per-cpu vmstat threshold
+ * while kswapd is awake and restore
+ * them before going back to sleep.
+ */
+ set_pgdat_percpu_threshold(pgdat,
+ calculate_normal_threshold);
schedule();
- reduce_pgdat_percpu_threshold(pgdat);
+ set_pgdat_percpu_threshold(pgdat,
+ calculate_pressure_threshold);
} else {
if (remaining)
count_vm_event(KSWAPD_LOW_WMARK_HIT_QUICKLY);
diff --git a/mm/vmstat.c b/mm/vmstat.c
index 41dc8cd..5b44b82 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -81,7 +81,7 @@ EXPORT_SYMBOL(vm_stat);
#ifdef CONFIG_SMP
-static int calculate_pressure_threshold(struct zone *zone)
+int calculate_pressure_threshold(struct zone *zone)
{
int threshold;
int watermark_distance;
@@ -105,7 +105,7 @@ static int calculate_pressure_threshold(struct zone *zone)
return threshold;
}
-static int calculate_threshold(struct zone *zone)
+int calculate_normal_threshold(struct zone *zone)
{
int threshold;
int mem; /* memory in 128 MB units */
@@ -164,7 +164,7 @@ static void refresh_zone_stat_thresholds(void)
for_each_populated_zone(zone) {
unsigned long max_drift, tolerate_drift;
- threshold = calculate_threshold(zone);
+ threshold = calculate_normal_threshold(zone);
for_each_online_cpu(cpu)
per_cpu_ptr(zone->pageset, cpu)->stat_threshold
@@ -183,46 +183,24 @@ static void refresh_zone_stat_thresholds(void)
}
}
-void reduce_pgdat_percpu_threshold(pg_data_t *pgdat)
+void set_pgdat_percpu_threshold(pg_data_t *pgdat,
+ int (*calculate_pressure)(struct zone *))
{
struct zone *zone;
int cpu;
int threshold;
int i;
- get_online_cpus();
- for (i = 0; i < pgdat->nr_zones; i++) {
- zone = &pgdat->node_zones[i];
- if (!zone->percpu_drift_mark)
- continue;
-
- threshold = calculate_pressure_threshold(zone);
- for_each_online_cpu(cpu)
- per_cpu_ptr(zone->pageset, cpu)->stat_threshold
- = threshold;
- }
- put_online_cpus();
-}
-
-void restore_pgdat_percpu_threshold(pg_data_t *pgdat)
-{
- struct zone *zone;
- int cpu;
- int threshold;
- int i;
-
- get_online_cpus();
for (i = 0; i < pgdat->nr_zones; i++) {
zone = &pgdat->node_zones[i];
if (!zone->percpu_drift_mark)
continue;
- threshold = calculate_threshold(zone);
- for_each_online_cpu(cpu)
+ threshold = (*calculate_pressure)(zone);
+ for_each_possible_cpu(cpu)
per_cpu_ptr(zone->pageset, cpu)->stat_threshold
= threshold;
}
- put_online_cpus();
}
/*
--
1.7.3.2

View File

@ -0,0 +1,46 @@
From kernel-bounces@lists.fedoraproject.org Thu Sep 9 09:24:46 2010
From: Stanislaw Gruszka <sgruszka@redhat.com>
Subject: [PATCH 3/3] mmc: add ricoh e822 pci id
Date: Thu, 9 Sep 2010 15:24:12 +0200
From: Pablo Castillo <CyberCastle@gmail.com>
Upstream 568133ebda39f7507759a744fa9cf4d5097bad2f commit.
Signed-off-by: Pablo Castillo <CyberCastle@gmail.com>
Signed-off-by: Stanislaw Gruszka <sgruszka@redhat.com>
Tested-by: Gregg Lebovitz <gregg@lebovitz.net>
Cc: Maxim Levitsky <maximlevitsky@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
drivers/mmc/host/sdhci-pci.c | 8 ++++++++
1 files changed, 8 insertions(+), 0 deletions(-)
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index e021431..e8aa99d 100644
--- a/drivers/mmc/host/sdhci-pci.c
+++ b/drivers/mmc/host/sdhci-pci.c
@@ -415,6 +415,14 @@ static const struct pci_device_id pci_ids[] __devinitdata = {
},
{
+ .vendor = PCI_VENDOR_ID_RICOH,
+ .device = 0xe822,
+ .subvendor = PCI_ANY_ID,
+ .subdevice = PCI_ANY_ID,
+ .driver_data = (kernel_ulong_t)&sdhci_ricoh_mmc,
+ },
+
+ {
.vendor = PCI_VENDOR_ID_ENE,
.device = PCI_DEVICE_ID_ENE_CB712_SD,
.subvendor = PCI_ANY_ID,
--
1.7.1
_______________________________________________
kernel mailing list
kernel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/kernel

View File

@ -0,0 +1,150 @@
From kernel-bounces@lists.fedoraproject.org Thu Sep 9 09:24:33 2010
From: Stanislaw Gruszka <sgruszka@redhat.com>
Subject: [PATCH 2/3] mmc: make sdhci work with ricoh mmc controller
Date: Thu, 9 Sep 2010 15:24:11 +0200
From: Maxim Levitsky <maximlevitsky@gmail.com>
Upstream ccc92c23240cdf952ef7cc39ba563910dcbc9cbe commit.
The current way of disabling it is not well tested by vendor and has all
kinds of bugs that show up on resume from ram/disk. A very good example
is a dead SDHCI controller.
Old way of disabling is still supported by continuing to use
CONFIG_MMC_RICOH_MMC.
Based on 'http://list.drzeus.cx/pipermail/sdhci-devel/2007-December/002085.html'
Therefore most of the credit for this goes to Andrew de Quincey
Signed-off-by: Maxim Levitsky <maximlevitsky@gmail.com>
Cc: Andrew de Quincey <adq_dvb@lidskialf.net>
Acked-by: Philip Langdale <philipl@overt.org>
Cc: "Rafael J. Wysocki" <rjw@sisk.pl>
Cc: <linux-mmc@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
drivers/mmc/host/sdhci-pci.c | 41 +++++++++++++++++++++++++++++++++++++++++
drivers/mmc/host/sdhci.c | 3 ++-
drivers/mmc/host/sdhci.h | 4 ++++
3 files changed, 47 insertions(+), 1 deletions(-)
diff --git a/drivers/mmc/host/sdhci-pci.c b/drivers/mmc/host/sdhci-pci.c
index 65483fd..e021431 100644
--- a/drivers/mmc/host/sdhci-pci.c
+++ b/drivers/mmc/host/sdhci-pci.c
@@ -17,6 +17,7 @@
#include <linux/pci.h>
#include <linux/dma-mapping.h>
#include <linux/slab.h>
+#include <linux/device.h>
#include <linux/mmc/host.h>
@@ -84,7 +85,30 @@ static int ricoh_probe(struct sdhci_pci_chip *chip)
if (chip->pdev->subsystem_vendor == PCI_VENDOR_ID_SAMSUNG ||
chip->pdev->subsystem_vendor == PCI_VENDOR_ID_SONY)
chip->quirks |= SDHCI_QUIRK_NO_CARD_NO_RESET;
+ return 0;
+}
+
+static int ricoh_mmc_probe_slot(struct sdhci_pci_slot *slot)
+{
+ slot->host->caps =
+ ((0x21 << SDHCI_TIMEOUT_CLK_SHIFT)
+ & SDHCI_TIMEOUT_CLK_MASK) |
+
+ ((0x21 << SDHCI_CLOCK_BASE_SHIFT)
+ & SDHCI_CLOCK_BASE_MASK) |
+ SDHCI_TIMEOUT_CLK_UNIT |
+ SDHCI_CAN_VDD_330 |
+ SDHCI_CAN_DO_SDMA;
+ return 0;
+}
+
+static int ricoh_mmc_resume(struct sdhci_pci_chip *chip)
+{
+ /* Apply a delay to allow controller to settle */
+ /* Otherwise it becomes confused if card state changed
+ during suspend */
+ msleep(500);
return 0;
}
@@ -95,6 +119,15 @@ static const struct sdhci_pci_fixes sdhci_ricoh = {
SDHCI_QUIRK_CLOCK_BEFORE_RESET,
};
+static const struct sdhci_pci_fixes sdhci_ricoh_mmc = {
+ .probe_slot = ricoh_mmc_probe_slot,
+ .resume = ricoh_mmc_resume,
+ .quirks = SDHCI_QUIRK_32BIT_DMA_ADDR |
+ SDHCI_QUIRK_CLOCK_BEFORE_RESET |
+ SDHCI_QUIRK_NO_CARD_NO_RESET |
+ SDHCI_QUIRK_MISSING_CAPS
+};
+
static const struct sdhci_pci_fixes sdhci_ene_712 = {
.quirks = SDHCI_QUIRK_SINGLE_POWER_WRITE |
SDHCI_QUIRK_BROKEN_DMA,
@@ -374,6 +407,14 @@ static const struct pci_device_id pci_ids[] __devinitdata = {
},
{
+ .vendor = PCI_VENDOR_ID_RICOH,
+ .device = 0x843,
+ .subvendor = PCI_ANY_ID,
+ .subdevice = PCI_ANY_ID,
+ .driver_data = (kernel_ulong_t)&sdhci_ricoh_mmc,
+ },
+
+ {
.vendor = PCI_VENDOR_ID_ENE,
.device = PCI_DEVICE_ID_ENE_CB712_SD,
.subvendor = PCI_ANY_ID,
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index 955cad9..8e1dda3 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -1692,7 +1692,8 @@ int sdhci_add_host(struct sdhci_host *host)
host->version);
}
- caps = sdhci_readl(host, SDHCI_CAPABILITIES);
+ caps = (host->quirks & SDHCI_QUIRK_MISSING_CAPS) ? host->caps :
+ sdhci_readl(host, SDHCI_CAPABILITIES);
if (host->quirks & SDHCI_QUIRK_FORCE_DMA)
host->flags |= SDHCI_USE_SDMA;
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index eb5efe0..030de49 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -241,6 +241,8 @@ struct sdhci_host {
#define SDHCI_QUIRK_CAP_CLOCK_BASE_BROKEN (1<<25)
/* Controller cannot support End Attribute in NOP ADMA descriptor */
#define SDHCI_QUIRK_NO_ENDATTR_IN_NOPDESC (1<<26)
+/* Controller is missing device caps. Use caps provided by host */
+#define SDHCI_QUIRK_MISSING_CAPS (1<<27)
int irq; /* Device IRQ */
void __iomem * ioaddr; /* Mapped address */
@@ -293,6 +295,8 @@ struct sdhci_host {
struct timer_list timer; /* Timer for timeouts */
+ unsigned int caps; /* Alternative capabilities */
+
unsigned long private[0] ____cacheline_aligned;
};
--
1.7.1
_______________________________________________
kernel mailing list
kernel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/kernel

254
net-AF_PACKET-vmalloc.patch Normal file
View File

@ -0,0 +1,254 @@
Author: Neil Horman <nhorman@tuxdriver.com>
Date: Fri Dec 17 13:35:36 2010 -0500
Enhance AF_PACKET to support using non-contiguous memory when allocating ring
buffer space. This is a combined backport of the following commits from
net-next-2.6:
0e3125c755445664f00ad036e4fc2cd32fd52877
bbce5a59e4e0e6e1dbc85492caaf310ff6611309
0af55bb58f8fa7865004ac48d16affe125ac1b7f
920b8d913bd3d963d5c88bca160a272b71e0c95a
diff -up linux-2.6.34.x86_64/net/packet/af_packet.c.orig linux-2.6.34.x86_64/net/packet/af_packet.c
--- linux-2.6.34.x86_64/net/packet/af_packet.c.orig 2010-12-17 12:16:58.000000000 -0500
+++ linux-2.6.34.x86_64/net/packet/af_packet.c 2010-12-17 12:30:14.000000000 -0500
@@ -61,6 +61,7 @@
#include <linux/kernel.h>
#include <linux/kmod.h>
#include <linux/slab.h>
+#include <linux/vmalloc.h>
#include <net/net_namespace.h>
#include <net/ip.h>
#include <net/protocol.h>
@@ -161,8 +162,14 @@ struct packet_mreq_max {
static int packet_set_ring(struct sock *sk, struct tpacket_req *req,
int closing, int tx_ring);
+#define PGV_FROM_VMALLOC 1
+struct pgv {
+ char *buffer;
+ unsigned char flags;
+};
+
struct packet_ring_buffer {
- char **pg_vec;
+ struct pgv *pg_vec;
unsigned int head;
unsigned int frames_per_block;
unsigned int frame_size;
@@ -214,6 +221,13 @@ struct packet_skb_cb {
#define PACKET_SKB_CB(__skb) ((struct packet_skb_cb *)((__skb)->cb))
+static inline struct page *pgv_to_page(void *addr)
+{
+ if (is_vmalloc_addr(addr))
+ return vmalloc_to_page(addr);
+ return virt_to_page(addr);
+}
+
static void __packet_set_status(struct packet_sock *po, void *frame, int status)
{
union {
@@ -226,11 +240,11 @@ static void __packet_set_status(struct p
switch (po->tp_version) {
case TPACKET_V1:
h.h1->tp_status = status;
- flush_dcache_page(virt_to_page(&h.h1->tp_status));
+ flush_dcache_page(pgv_to_page(&h.h1->tp_status));
break;
case TPACKET_V2:
h.h2->tp_status = status;
- flush_dcache_page(virt_to_page(&h.h2->tp_status));
+ flush_dcache_page(pgv_to_page(&h.h2->tp_status));
break;
default:
pr_err("TPACKET version not supported\n");
@@ -253,10 +267,10 @@ static int __packet_get_status(struct pa
h.raw = frame;
switch (po->tp_version) {
case TPACKET_V1:
- flush_dcache_page(virt_to_page(&h.h1->tp_status));
+ flush_dcache_page(pgv_to_page(&h.h1->tp_status));
return h.h1->tp_status;
case TPACKET_V2:
- flush_dcache_page(virt_to_page(&h.h2->tp_status));
+ flush_dcache_page(pgv_to_page(&h.h2->tp_status));
return h.h2->tp_status;
default:
pr_err("TPACKET version not supported\n");
@@ -280,7 +294,8 @@ static void *packet_lookup_frame(struct
pg_vec_pos = position / rb->frames_per_block;
frame_offset = position % rb->frames_per_block;
- h.raw = rb->pg_vec[pg_vec_pos] + (frame_offset * rb->frame_size);
+ h.raw = rb->pg_vec[pg_vec_pos].buffer +
+ (frame_offset * rb->frame_size);
if (status != __packet_get_status(po, h.raw))
return NULL;
@@ -771,15 +786,11 @@ static int tpacket_rcv(struct sk_buff *s
__packet_set_status(po, h.raw, status);
smp_mb();
{
- struct page *p_start, *p_end;
- u8 *h_end = h.raw + macoff + snaplen - 1;
+ u8 *start, *end;
- p_start = virt_to_page(h.raw);
- p_end = virt_to_page(h_end);
- while (p_start <= p_end) {
- flush_dcache_page(p_start);
- p_start++;
- }
+ end = (u8 *)PAGE_ALIGN((unsigned long)h.raw + macoff + snaplen);
+ for (start = h.raw; start < end; start += PAGE_SIZE)
+ flush_dcache_page(pgv_to_page(start));
}
sk->sk_data_ready(sk, 0);
@@ -886,7 +897,6 @@ static int tpacket_fill_skb(struct packe
}
err = -EFAULT;
- page = virt_to_page(data);
offset = offset_in_page(data);
len_max = PAGE_SIZE - offset;
len = ((to_write > len_max) ? len_max : to_write);
@@ -905,11 +915,11 @@ static int tpacket_fill_skb(struct packe
return -EFAULT;
}
+ page = pgv_to_page(data);
+ data += len;
flush_dcache_page(page);
get_page(page);
- skb_fill_page_desc(skb,
- nr_frags,
- page++, offset, len);
+ skb_fill_page_desc(skb, nr_frags, page, offset, len);
to_write -= len;
offset = 0;
len_max = PAGE_SIZE;
@@ -2230,37 +2240,76 @@ static const struct vm_operations_struct
.close = packet_mm_close,
};
-static void free_pg_vec(char **pg_vec, unsigned int order, unsigned int len)
+static void free_pg_vec(struct pgv *pg_vec, unsigned int order,
+ unsigned int len)
{
int i;
for (i = 0; i < len; i++) {
- if (likely(pg_vec[i]))
- free_pages((unsigned long) pg_vec[i], order);
+ if (likely(pg_vec[i].buffer)) {
+ if (pg_vec[i].flags & PGV_FROM_VMALLOC)
+ vfree(pg_vec[i].buffer);
+ else
+ free_pages((unsigned long)pg_vec[i].buffer,
+ order);
+ pg_vec[i].buffer = NULL;
+ }
}
kfree(pg_vec);
}
-static inline char *alloc_one_pg_vec_page(unsigned long order)
+static inline char *alloc_one_pg_vec_page(unsigned long order,
+ unsigned char *flags)
{
- gfp_t gfp_flags = GFP_KERNEL | __GFP_COMP | __GFP_ZERO | __GFP_NOWARN;
+ char *buffer = NULL;
+ gfp_t gfp_flags = GFP_KERNEL | __GFP_COMP |
+ __GFP_ZERO | __GFP_NOWARN | __GFP_NORETRY;
+
+ buffer = (char *) __get_free_pages(gfp_flags, order);
- return (char *) __get_free_pages(gfp_flags, order);
+ if (buffer)
+ return buffer;
+
+ /*
+ * __get_free_pages failed, fall back to vmalloc
+ */
+ *flags |= PGV_FROM_VMALLOC;
+ buffer = vmalloc((1 << order) * PAGE_SIZE);
+
+ if (buffer) {
+ memset(buffer, 0, (1 << order) * PAGE_SIZE);
+ return buffer;
+ }
+
+ /*
+ * vmalloc failed, lets dig into swap here
+ */
+ *flags = 0;
+ gfp_flags &= ~__GFP_NORETRY;
+ buffer = (char *)__get_free_pages(gfp_flags, order);
+ if (buffer)
+ return buffer;
+
+ /*
+ * complete and utter failure
+ */
+ return NULL;
}
-static char **alloc_pg_vec(struct tpacket_req *req, int order)
+static struct pgv *alloc_pg_vec(struct tpacket_req *req, int order)
{
unsigned int block_nr = req->tp_block_nr;
- char **pg_vec;
+ struct pgv *pg_vec;
int i;
- pg_vec = kzalloc(block_nr * sizeof(char *), GFP_KERNEL);
+ pg_vec = kcalloc(block_nr, sizeof(struct pgv), GFP_KERNEL);
if (unlikely(!pg_vec))
goto out;
for (i = 0; i < block_nr; i++) {
- pg_vec[i] = alloc_one_pg_vec_page(order);
- if (unlikely(!pg_vec[i]))
+ pg_vec[i].buffer = alloc_one_pg_vec_page(order,
+ &pg_vec[i].flags);
+ if (unlikely(!pg_vec[i].buffer))
goto out_free_pgvec;
}
@@ -2276,7 +2325,7 @@ out_free_pgvec:
static int packet_set_ring(struct sock *sk, struct tpacket_req *req,
int closing, int tx_ring)
{
- char **pg_vec = NULL;
+ struct pgv *pg_vec = NULL;
struct packet_sock *po = pkt_sk(sk);
int was_running, order = 0;
struct packet_ring_buffer *rb;
@@ -2438,15 +2487,22 @@ static int packet_mmap(struct file *file
continue;
for (i = 0; i < rb->pg_vec_len; i++) {
- struct page *page = virt_to_page(rb->pg_vec[i]);
+ struct page *page;
+ void *kaddr = rb->pg_vec[i].buffer;
int pg_num;
for (pg_num = 0; pg_num < rb->pg_vec_pages;
- pg_num++, page++) {
+ pg_num++) {
+ if (rb->pg_vec[i].flags & PGV_FROM_VMALLOC)
+ page = vmalloc_to_page(kaddr);
+ else
+ page = virt_to_page(kaddr);
+
err = vm_insert_page(vma, start, page);
if (unlikely(err))
goto out;
start += PAGE_SIZE;
+ kaddr += PAGE_SIZE;
}
}
}

View File

@ -0,0 +1,29 @@
From 38f1f0db010ac5b981ae06f1fe2fd64095ebb171 Mon Sep 17 00:00:00 2001
From: Nelson Elhage <nelhage@ksplice.com>
Date: Wed, 3 Nov 2010 16:35:40 +0000
Subject: [PATCH] netlink: Make nlmsg_find_attr take a const nlmsghdr*.
This will let us use it on a nlmsghdr stored inside a netlink_callback.
Signed-off-by: Nelson Elhage <nelhage@ksplice.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
---
include/net/netlink.h | 2 +-
1 files changed, 1 insertions(+), 1 deletions(-)
diff --git a/include/net/netlink.h b/include/net/netlink.h
index a63b219..c344646 100644
--- a/include/net/netlink.h
+++ b/include/net/netlink.h
@@ -384,7 +384,7 @@ static inline int nlmsg_parse(const struct nlmsghdr *nlh, int hdrlen,
*
* Returns the first attribute which matches the specified type.
*/
-static inline struct nlattr *nlmsg_find_attr(struct nlmsghdr *nlh,
+static inline struct nlattr *nlmsg_find_attr(const struct nlmsghdr *nlh,
int hdrlen, int attrtype)
{
return nla_find(nlmsg_attrdata(nlh, hdrlen),
--
1.7.3.2

View File

@ -0,0 +1,86 @@
From 1b3df4f489345b0fd6e83645ad5d464aee55f7de Mon Sep 17 00:00:00 2001
From: David Kilroy <kilroyd@googlemail.com>
Date: Sat, 4 Dec 2010 18:36:30 +0000
Subject: [PATCH] orinoco: initialise priv->hw before assigning the interrupt
The interrupt handler takes a lock - but since commit bcad6e80f3f this
lock goes through an indirection specified in the hermes_t structure.
We must therefore initialise the structure before setting up the
interrupt handler.
<https://bugzilla.kernel.org/show_bug.cgi?id=23932>
Fix both orinoco_cs and spectrum_cs
Bisected by: Matt Domsch <Matt_Domsch@dell.com>
Tested by: Matt Domsch <Matt_Domsch@dell.com>
Signed-off by: David Kilroy <kilroyd@googlemail.com>
---
drivers/net/wireless/orinoco/orinoco_cs.c | 12 ++++++------
drivers/net/wireless/orinoco/spectrum_cs.c | 12 ++++++------
2 files changed, 12 insertions(+), 12 deletions(-)
diff --git a/drivers/net/wireless/orinoco/orinoco_cs.c b/drivers/net/wireless/orinoco/orinoco_cs.c
index b16d5db..66c7bcc 100644
--- a/drivers/net/wireless/orinoco/orinoco_cs.c
+++ b/drivers/net/wireless/orinoco/orinoco_cs.c
@@ -251,19 +251,19 @@ orinoco_cs_config(struct pcmcia_device *link)
goto failed;
}
- ret = pcmcia_request_irq(link, orinoco_interrupt);
- if (ret)
+ mem = ioport_map(link->io.BasePort1, link->io.NumPorts1);
+ if (!mem)
goto failed;
/* We initialize the hermes structure before completing PCMCIA
* configuration just in case the interrupt handler gets
* called. */
- mem = ioport_map(link->io.BasePort1, link->io.NumPorts1);
- if (!mem)
- goto failed;
-
hermes_struct_init(hw, mem, HERMES_16BIT_REGSPACING);
+ ret = pcmcia_request_irq(link, orinoco_interrupt);
+ if (ret)
+ goto failed;
+
/*
* This actually configures the PCMCIA socket -- setting up
* the I/O windows and the interrupt mapping, and putting the
diff --git a/drivers/net/wireless/orinoco/spectrum_cs.c b/drivers/net/wireless/orinoco/spectrum_cs.c
index b51a9ad..0148763 100644
--- a/drivers/net/wireless/orinoco/spectrum_cs.c
+++ b/drivers/net/wireless/orinoco/spectrum_cs.c
@@ -325,20 +325,20 @@ spectrum_cs_config(struct pcmcia_device *link)
goto failed;
}
- ret = pcmcia_request_irq(link, orinoco_interrupt);
- if (ret)
+ mem = ioport_map(link->io.BasePort1, link->io.NumPorts1);
+ if (!mem)
goto failed;
/* We initialize the hermes structure before completing PCMCIA
* configuration just in case the interrupt handler gets
* called. */
- mem = ioport_map(link->io.BasePort1, link->io.NumPorts1);
- if (!mem)
- goto failed;
-
hermes_struct_init(hw, mem, HERMES_16BIT_REGSPACING);
hw->eeprom_pda = true;
+ ret = pcmcia_request_irq(link, orinoco_interrupt);
+ if (ret)
+ goto failed;
+
/*
* This actually configures the PCMCIA socket -- setting up
* the I/O windows and the interrupt mapping, and putting the
--
1.7.2.2

View File

@ -0,0 +1,144 @@
From linux-kernel-owner@vger.kernel.org Mon Dec 6 14:01:17 2010
From: Matthew Garrett <mjg@redhat.com>
To: linux-pci@vger.kernel.org
Cc: linux-kernel@vger.kernel.org, jbarnes@virtuousgeek.org,
Matthew Garrett <mjg@redhat.com>
Subject: [PATCH v2] PCI: Disable ASPM if BIOS asks us to
Date: Mon, 6 Dec 2010 14:00:56 -0500
Message-Id: <1291662056-6055-1-git-send-email-mjg@redhat.com>
We currently refuse to touch the ASPM registers if the BIOS tells us that
ASPM isn't supported. This can cause problems if the BIOS has (for any
reason) enabled ASPM on some devices anyway. Change the code such that we
explicitly clear ASPM if the FADT indicates that ASPM isn't supported,
and make sure we tidy up appropriately on device removal in order to deal
with the hotplug case. If ASPM is disabled because the BIOS doesn't hand
over control then we won't touch the registers.
Signed-off-by: Matthew Garrett <mjg@redhat.com>
---
Implement Rafael's suggestion to use two separate functions, and also
ensure that we clear the clkpm bit as well as the ASPM bits.
drivers/pci/pci-acpi.c | 1 +
drivers/pci/pcie/aspm.c | 21 +++++++++++++++++----
include/linux/pci-aspm.h | 5 ++++-
3 files changed, 22 insertions(+), 5 deletions(-)
diff --git a/drivers/pci/pci-acpi.c b/drivers/pci/pci-acpi.c
index 24e19c5..d7ea699 100644
--- a/drivers/pci/pci-acpi.c
+++ b/drivers/pci/pci-acpi.c
@@ -399,6 +399,7 @@ static int __init acpi_pci_init(void)
if (acpi_gbl_FADT.boot_flags & ACPI_FADT_NO_ASPM) {
printk(KERN_INFO"ACPI FADT declares the system doesn't support PCIe ASPM, so disable it\n");
+ pcie_clear_aspm();
pcie_no_aspm();
}
diff --git a/drivers/pci/pcie/aspm.c b/drivers/pci/pcie/aspm.c
index 7122281..8112415 100644
--- a/drivers/pci/pcie/aspm.c
+++ b/drivers/pci/pcie/aspm.c
@@ -68,7 +68,7 @@ struct pcie_link_state {
struct aspm_latency acceptable[8];
};
-static int aspm_disabled, aspm_force;
+static int aspm_disabled, aspm_force, aspm_clear_state;
static DEFINE_MUTEX(aspm_lock);
static LIST_HEAD(link_list);
@@ -139,7 +139,7 @@ static void pcie_set_clkpm(struct pcie_link_state *link, int enable)
{
/* Don't enable Clock PM if the link is not Clock PM capable */
if (!link->clkpm_capable && enable)
- return;
+ enable = 0;
/* Need nothing if the specified equals to current state */
if (link->clkpm_enabled == enable)
return;
@@ -498,6 +498,10 @@ static int pcie_aspm_sanity_check(struct pci_dev *pdev)
struct pci_dev *child;
int pos;
u32 reg32;
+
+ if (aspm_clear_state)
+ return -EINVAL;
+
/*
* Some functions in a slot might not all be PCIe functions,
* very strange. Disable ASPM for the whole slot
@@ -563,12 +567,15 @@ void pcie_aspm_init_link_state(struct pci_dev *pdev)
struct pcie_link_state *link;
int blacklist = !!pcie_aspm_sanity_check(pdev);
- if (aspm_disabled || !pci_is_pcie(pdev) || pdev->link_state)
+ if (!pci_is_pcie(pdev) || pdev->link_state)
return;
if (pdev->pcie_type != PCI_EXP_TYPE_ROOT_PORT &&
pdev->pcie_type != PCI_EXP_TYPE_DOWNSTREAM)
return;
+ if (aspm_disabled && !aspm_clear_state)
+ return;
+
/* VIA has a strange chipset, root port is under a bridge */
if (pdev->pcie_type == PCI_EXP_TYPE_ROOT_PORT &&
pdev->bus->self)
@@ -641,7 +648,7 @@ void pcie_aspm_exit_link_state(struct pci_dev *pdev)
struct pci_dev *parent = pdev->bus->self;
struct pcie_link_state *link, *root, *parent_link;
- if (aspm_disabled || !pci_is_pcie(pdev) ||
+ if ((aspm_disabled && !aspm_clear_state) || !pci_is_pcie(pdev) ||
!parent || !parent->link_state)
return;
if ((parent->pcie_type != PCI_EXP_TYPE_ROOT_PORT) &&
@@ -899,6 +906,12 @@ static int __init pcie_aspm_disable(char *str)
__setup("pcie_aspm=", pcie_aspm_disable);
+void pcie_clear_aspm(void)
+{
+ if (!aspm_force)
+ aspm_clear_state = 1;
+}
+
void pcie_no_aspm(void)
{
if (!aspm_force)
diff --git a/include/linux/pci-aspm.h b/include/linux/pci-aspm.h
index 91ba0b3..ce68105 100644
--- a/include/linux/pci-aspm.h
+++ b/include/linux/pci-aspm.h
@@ -27,6 +27,7 @@ extern void pcie_aspm_init_link_state(struct pci_dev *pdev);
extern void pcie_aspm_exit_link_state(struct pci_dev *pdev);
extern void pcie_aspm_pm_state_change(struct pci_dev *pdev);
extern void pci_disable_link_state(struct pci_dev *pdev, int state);
+extern void pcie_clear_aspm(void);
extern void pcie_no_aspm(void);
#else
static inline void pcie_aspm_init_link_state(struct pci_dev *pdev)
@@ -41,7 +42,9 @@ static inline void pcie_aspm_pm_state_change(struct pci_dev *pdev)
static inline void pci_disable_link_state(struct pci_dev *pdev, int state)
{
}
-
+static inline void pcie_clear_aspm(void)
+{
+}
static inline void pcie_no_aspm(void)
{
}
--
1.7.3.2
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/

View File

@ -0,0 +1,84 @@
From 8ba192510be47fc327c59449a4b5d347e9669329 Mon Sep 17 00:00:00 2001
From: Bjorn Helgaas <bjorn.helgaas@hp.com>
Date: Wed, 29 Sep 2010 12:24:23 -0600
Subject: PNP: log PNP resources, as we do for PCI
ACPI devices are often involved in address space conflicts with PCI devices,
so I think it's worth logging the resources they use. Otherwise we have to
depend on lspnp or groping around in sysfs to find them.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
---
drivers/pnp/core.c | 5 +++--
drivers/pnp/resource.c | 10 +++++-----
2 files changed, 8 insertions(+), 7 deletions(-)
diff --git a/drivers/pnp/core.c b/drivers/pnp/core.c
index 5dba909..e457f99 100644
--- a/drivers/pnp/core.c
+++ b/drivers/pnp/core.c
@@ -191,8 +191,9 @@ int pnp_add_device(struct pnp_dev *dev)
for (id = dev->id; id; id = id->next)
len += scnprintf(buf + len, sizeof(buf) - len, " %s", id->id);
- pnp_dbg(&dev->dev, "%s device, IDs%s (%s)\n",
- dev->protocol->name, buf, dev->active ? "active" : "disabled");
+ dev_printk(KERN_DEBUG, &dev->dev, "%s device, IDs%s (%s)\n",
+ dev->protocol->name, buf,
+ dev->active ? "active" : "disabled");
return 0;
}
diff --git a/drivers/pnp/resource.c b/drivers/pnp/resource.c
index e3446ab..a925e6b 100644
--- a/drivers/pnp/resource.c
+++ b/drivers/pnp/resource.c
@@ -523,7 +523,7 @@ struct pnp_resource *pnp_add_irq_resource(struct pnp_dev *dev, int irq,
res->start = irq;
res->end = irq;
- pnp_dbg(&dev->dev, " add %pr\n", res);
+ dev_printk(KERN_DEBUG, &dev->dev, "%pR\n", res);
return pnp_res;
}
@@ -544,7 +544,7 @@ struct pnp_resource *pnp_add_dma_resource(struct pnp_dev *dev, int dma,
res->start = dma;
res->end = dma;
- pnp_dbg(&dev->dev, " add %pr\n", res);
+ dev_printk(KERN_DEBUG, &dev->dev, "%pR\n", res);
return pnp_res;
}
@@ -568,7 +568,7 @@ struct pnp_resource *pnp_add_io_resource(struct pnp_dev *dev,
res->start = start;
res->end = end;
- pnp_dbg(&dev->dev, " add %pr\n", res);
+ dev_printk(KERN_DEBUG, &dev->dev, "%pR\n", res);
return pnp_res;
}
@@ -592,7 +592,7 @@ struct pnp_resource *pnp_add_mem_resource(struct pnp_dev *dev,
res->start = start;
res->end = end;
- pnp_dbg(&dev->dev, " add %pr\n", res);
+ dev_printk(KERN_DEBUG, &dev->dev, "%pR\n", res);
return pnp_res;
}
@@ -616,7 +616,7 @@ struct pnp_resource *pnp_add_bus_resource(struct pnp_dev *dev,
res->start = start;
res->end = end;
- pnp_dbg(&dev->dev, " add %pr\n", res);
+ dev_printk(KERN_DEBUG, &dev->dev, "%pR\n", res);
return pnp_res;
}
--
1.7.3.2

View File

@ -0,0 +1,60 @@
From 2876b1571839c25ce5e7485ead8417506d720c73 Mon Sep 17 00:00:00 2001
From: Oleg Nesterov <oleg@redhat.com>
Date: Fri, 5 Nov 2010 16:53:42 +0100
Subject: posix-cpu-timers: workaround to suppress the problems with mt exec
posix-cpu-timers.c correctly assumes that the dying process does
posix_cpu_timers_exit_group() and removes all !CPUCLOCK_PERTHREAD
timers from signal->cpu_timers list.
But, it also assumes that timer->it.cpu.task is always the group
leader, and thus the dead ->task means the dead thread group.
This is obviously not true after de_thread() changes the leader.
After that almost every posix_cpu_timer_ method has problems.
It is not simple to fix this bug correctly. First of all, I think
that timer->it.cpu should use struct pid instead of task_struct.
Also, the locking should be reworked completely. In particular,
tasklist_lock should not be used at all. This all needs a lot of
nontrivial and hard-to-test changes.
Change __exit_signal() to do posix_cpu_timers_exit_group() when
the old leader dies during exec. This is not the fix, just the
temporary hack to hide the problem for 2.6.37 and stable. IOW,
this is obviously wrong but this is what we currently have anyway:
cpu timers do not work after mt exec.
In theory this change adds another race. The exiting leader can
detach the timers which were attached to the new leader. However,
the window between de_thread() and release_task() is small, we
can pretend that sys_timer_create() was called before de_thread().
Signed-off-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
kernel/exit.c | 8 ++++++++
1 files changed, 8 insertions(+), 0 deletions(-)
diff --git a/kernel/exit.c b/kernel/exit.c
index ac90425..85daf1d 100644
--- a/kernel/exit.c
+++ b/kernel/exit.c
@@ -95,6 +95,14 @@ static void __exit_signal(struct task_struct *tsk)
sig->tty = NULL;
} else {
/*
+ * This can only happen if the caller is de_thread().
+ * FIXME: this is the temporary hack, we should teach
+ * posix-cpu-timers to handle this case correctly.
+ */
+ if (unlikely(has_group_leader_pid(tsk)))
+ posix_cpu_timers_exit_group(tsk);
+
+ /*
* If there is any task waiting for the group exit
* then notify it:
*/
--
1.7.3.2

View File

@ -1,65 +0,0 @@
From 69b711c0c5e3d9cb3a5b9f741fb4cdc96b5739cb Mon Sep 17 00:00:00 2001
From: Kyle McMartin <kyle@dreadnought.i.jkkm.org>
Subject: Revert "drm/kms: disable/enable poll around switcheroo on/off"
This reverts commit fbf81762e385d3d45acad057b654d56972acf58c, mostly.
---
drivers/gpu/drm/i915/i915_dma.c | 4 +---
drivers/gpu/drm/nouveau/nouveau_state.c | 3 ---
drivers/gpu/drm/radeon/radeon_device.c | 2 --
diff --git a/drivers/gpu/drm/i915/i915_dma.c b/drivers/gpu/drm/i915/i915_dma.c
index 59a2bf8..2df3286 100644
--- a/drivers/gpu/drm/i915/i915_dma.c
+++ b/drivers/gpu/drm/i915/i915_dma.c
@@ -1320,14 +1320,12 @@ static void i915_switcheroo_set_state(struct pci_dev *pdev, enum vga_switcheroo_
struct drm_device *dev = pci_get_drvdata(pdev);
pm_message_t pmm = { .event = PM_EVENT_SUSPEND };
if (state == VGA_SWITCHEROO_ON) {
- printk(KERN_INFO "i915: switched on\n");
+ printk(KERN_INFO "i915: switched off\n");
/* i915 resume handler doesn't set to D0 */
pci_set_power_state(dev->pdev, PCI_D0);
i915_resume(dev);
- drm_kms_helper_poll_enable(dev);
} else {
printk(KERN_ERR "i915: switched off\n");
- drm_kms_helper_poll_disable(dev);
i915_suspend(dev, pmm);
}
}
diff --git a/drivers/gpu/drm/nouveau/nouveau_state.c b/drivers/gpu/drm/nouveau/nouveau_state.c
index b02a231..0c28266 100644
--- a/drivers/gpu/drm/nouveau/nouveau_state.c
+++ b/drivers/gpu/drm/nouveau/nouveau_state.c
@@ -376,15 +376,12 @@ out_err:
static void nouveau_switcheroo_set_state(struct pci_dev *pdev,
enum vga_switcheroo_state state)
{
- struct drm_device *dev = pci_get_drvdata(pdev);
pm_message_t pmm = { .event = PM_EVENT_SUSPEND };
if (state == VGA_SWITCHEROO_ON) {
printk(KERN_ERR "VGA switcheroo: switched nouveau on\n");
nouveau_pci_resume(pdev);
- drm_kms_helper_poll_enable(dev);
} else {
printk(KERN_ERR "VGA switcheroo: switched nouveau off\n");
- drm_kms_helper_poll_disable(dev);
nouveau_pci_suspend(pdev, pmm);
}
}
diff --git a/drivers/gpu/drm/radeon/radeon_device.c b/drivers/gpu/drm/radeon/radeon_device.c
index f10faed..225a9f2 100644
--- a/drivers/gpu/drm/radeon/radeon_device.c
+++ b/drivers/gpu/drm/radeon/radeon_device.c
@@ -546,10 +546,8 @@ static void radeon_switcheroo_set_state(struct pci_dev *pdev, enum vga_switchero
/* don't suspend or resume card normally */
rdev->powered_down = false;
radeon_resume_kms(dev);
- drm_kms_helper_poll_enable(dev);
} else {
printk(KERN_INFO "radeon: switched off\n");
- drm_kms_helper_poll_disable(dev);
radeon_suspend_kms(dev, pmm);
/* don't suspend or resume card normally */
rdev->powered_down = true;

View File

@ -0,0 +1,166 @@
commit 8b73fb8e29e9ae0458d36cc0dc25e2717587dfd4
Author: John W. Linville <linville@tuxdriver.com>
Date: Wed Jul 21 16:26:40 2010 -0400
rtl8180: improve signal reporting for actual rtl8180 hardware
Adapted from Realtek-provided driver...
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Tested-by: Pauli Nieminen <suokkos@gmail.com>
diff --git a/drivers/net/wireless/rtl818x/rtl8180_dev.c b/drivers/net/wireless/rtl818x/rtl8180_dev.c
index 31808f9..d8b186a 100644
--- a/drivers/net/wireless/rtl818x/rtl8180_dev.c
+++ b/drivers/net/wireless/rtl818x/rtl8180_dev.c
@@ -103,7 +103,7 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev)
{
struct rtl8180_priv *priv = dev->priv;
unsigned int count = 32;
- u8 signal;
+ u8 signal, agc, sq;
while (count--) {
struct rtl8180_rx_desc *entry = &priv->rx_ring[priv->rx_idx];
@@ -132,12 +132,16 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev)
rx_status.antenna = (flags2 >> 15) & 1;
rx_status.rate_idx = (flags >> 20) & 0xF;
- /* TODO: improve signal/rssi reporting for !rtl8185 */
- signal = (flags2 >> 17) & 0x7F;
- if (rx_status.rate_idx > 3)
- signal = 90 - clamp_t(u8, signal, 25, 90);
- else
- signal = 95 - clamp_t(u8, signal, 30, 95);
+ agc = (flags2 >> 17) & 0x7F;
+ if (priv->r8185) {
+ if (rx_status.rate_idx > 3)
+ signal = 90 - clamp_t(u8, agc, 25, 90);
+ else
+ signal = 95 - clamp_t(u8, agc, 30, 95);
+ } else {
+ sq = flags2 & 0xff;
+ signal = priv->rf->calc_rssi(agc, sq);
+ }
rx_status.signal = signal;
rx_status.freq = dev->conf.channel->center_freq;
rx_status.band = dev->conf.channel->band;
diff --git a/drivers/net/wireless/rtl818x/rtl8180_grf5101.c b/drivers/net/wireless/rtl818x/rtl8180_grf5101.c
index 947ee55..5cab9df 100644
--- a/drivers/net/wireless/rtl818x/rtl8180_grf5101.c
+++ b/drivers/net/wireless/rtl818x/rtl8180_grf5101.c
@@ -69,6 +69,15 @@ static void grf5101_write_phy_antenna(struct ieee80211_hw *dev, short chan)
rtl8180_write_phy(dev, 0x10, ant);
}
+static u8 grf5101_rf_calc_rssi(u8 agc, u8 sq)
+{
+ if (agc > 60)
+ return 65;
+
+ /* TODO(?): just return agc (or agc + 5) to avoid mult / div */
+ return 65 * agc / 60;
+}
+
static void grf5101_rf_set_channel(struct ieee80211_hw *dev,
struct ieee80211_conf *conf)
{
@@ -176,5 +185,6 @@ const struct rtl818x_rf_ops grf5101_rf_ops = {
.name = "GCT",
.init = grf5101_rf_init,
.stop = grf5101_rf_stop,
- .set_chan = grf5101_rf_set_channel
+ .set_chan = grf5101_rf_set_channel,
+ .calc_rssi = grf5101_rf_calc_rssi,
};
diff --git a/drivers/net/wireless/rtl818x/rtl8180_max2820.c b/drivers/net/wireless/rtl818x/rtl8180_max2820.c
index 6c825fd..16c4655 100644
--- a/drivers/net/wireless/rtl818x/rtl8180_max2820.c
+++ b/drivers/net/wireless/rtl818x/rtl8180_max2820.c
@@ -74,6 +74,22 @@ static void max2820_write_phy_antenna(struct ieee80211_hw *dev, short chan)
rtl8180_write_phy(dev, 0x10, ant);
}
+static u8 max2820_rf_calc_rssi(u8 agc, u8 sq)
+{
+ bool odd;
+
+ odd = !!(agc & 1);
+
+ agc >>= 1;
+ if (odd)
+ agc += 76;
+ else
+ agc += 66;
+
+ /* TODO: change addends above to avoid mult / div below */
+ return 65 * agc / 100;
+}
+
static void max2820_rf_set_channel(struct ieee80211_hw *dev,
struct ieee80211_conf *conf)
{
@@ -148,5 +164,6 @@ const struct rtl818x_rf_ops max2820_rf_ops = {
.name = "Maxim",
.init = max2820_rf_init,
.stop = max2820_rf_stop,
- .set_chan = max2820_rf_set_channel
+ .set_chan = max2820_rf_set_channel,
+ .calc_rssi = max2820_rf_calc_rssi,
};
diff --git a/drivers/net/wireless/rtl818x/rtl8180_sa2400.c b/drivers/net/wireless/rtl818x/rtl8180_sa2400.c
index cea4e0c..d064fcc 100644
--- a/drivers/net/wireless/rtl818x/rtl8180_sa2400.c
+++ b/drivers/net/wireless/rtl818x/rtl8180_sa2400.c
@@ -76,6 +76,31 @@ static void sa2400_write_phy_antenna(struct ieee80211_hw *dev, short chan)
}
+static u8 sa2400_rf_rssi_map[] = {
+ 0x64, 0x64, 0x63, 0x62, 0x61, 0x60, 0x5f, 0x5e,
+ 0x5d, 0x5c, 0x5b, 0x5a, 0x57, 0x54, 0x52, 0x50,
+ 0x4e, 0x4c, 0x4a, 0x48, 0x46, 0x44, 0x41, 0x3f,
+ 0x3c, 0x3a, 0x37, 0x36, 0x36, 0x1c, 0x1c, 0x1b,
+ 0x1b, 0x1a, 0x1a, 0x19, 0x19, 0x18, 0x18, 0x17,
+ 0x17, 0x16, 0x16, 0x15, 0x15, 0x14, 0x14, 0x13,
+ 0x13, 0x12, 0x12, 0x11, 0x11, 0x10, 0x10, 0x0f,
+ 0x0f, 0x0e, 0x0e, 0x0d, 0x0d, 0x0c, 0x0c, 0x0b,
+ 0x0b, 0x0a, 0x0a, 0x09, 0x09, 0x08, 0x08, 0x07,
+ 0x07, 0x06, 0x06, 0x05, 0x04, 0x03, 0x02,
+};
+
+static u8 sa2400_rf_calc_rssi(u8 agc, u8 sq)
+{
+ if (sq == 0x80)
+ return 1;
+
+ if (sq > 78)
+ return 32;
+
+ /* TODO: recalc sa2400_rf_rssi_map to avoid mult / div */
+ return 65 * sa2400_rf_rssi_map[sq] / 100;
+}
+
static void sa2400_rf_set_channel(struct ieee80211_hw *dev,
struct ieee80211_conf *conf)
{
@@ -198,5 +223,6 @@ const struct rtl818x_rf_ops sa2400_rf_ops = {
.name = "Philips",
.init = sa2400_rf_init,
.stop = sa2400_rf_stop,
- .set_chan = sa2400_rf_set_channel
+ .set_chan = sa2400_rf_set_channel,
+ .calc_rssi = sa2400_rf_calc_rssi,
};
diff --git a/drivers/net/wireless/rtl818x/rtl818x.h b/drivers/net/wireless/rtl818x/rtl818x.h
index 8522490..22d9384 100644
--- a/drivers/net/wireless/rtl818x/rtl818x.h
+++ b/drivers/net/wireless/rtl818x/rtl818x.h
@@ -193,6 +193,7 @@ struct rtl818x_rf_ops {
void (*stop)(struct ieee80211_hw *);
void (*set_chan)(struct ieee80211_hw *, struct ieee80211_conf *);
void (*conf_erp)(struct ieee80211_hw *, struct ieee80211_bss_conf *);
+ u8 (*calc_rssi)(u8 agc, u8 sq);
};
/**

View File

@ -0,0 +1,43 @@
commit 8b74964c73ca9eed7078388d871cc7fae973cb63
Author: John W. Linville <linville@tuxdriver.com>
Date: Mon Jul 19 16:35:20 2010 -0400
rtl8180: improve signal reporting for rtl8185 hardware
The existing code seemed to be somewhat based on the datasheet, but
varied substantially from the vendor-provided driver. This mirrors the
handling of the rtl8185 case from that driver, but still neglects the
specifics for the rtl8180 hardware. Those details are a bit muddled...
Signed-off-by: John W. Linville <linville@tuxdriver.com>
diff --git a/drivers/net/wireless/rtl818x/rtl8180_dev.c b/drivers/net/wireless/rtl818x/rtl8180_dev.c
index 4270502..31808f9 100644
--- a/drivers/net/wireless/rtl818x/rtl8180_dev.c
+++ b/drivers/net/wireless/rtl818x/rtl8180_dev.c
@@ -103,6 +103,7 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev)
{
struct rtl8180_priv *priv = dev->priv;
unsigned int count = 32;
+ u8 signal;
while (count--) {
struct rtl8180_rx_desc *entry = &priv->rx_ring[priv->rx_idx];
@@ -130,10 +131,14 @@ static void rtl8180_handle_rx(struct ieee80211_hw *dev)
skb_put(skb, flags & 0xFFF);
rx_status.antenna = (flags2 >> 15) & 1;
- /* TODO: improve signal/rssi reporting */
- rx_status.signal = (flags2 >> 8) & 0x7F;
- /* XXX: is this correct? */
rx_status.rate_idx = (flags >> 20) & 0xF;
+ /* TODO: improve signal/rssi reporting for !rtl8185 */
+ signal = (flags2 >> 17) & 0x7F;
+ if (rx_status.rate_idx > 3)
+ signal = 90 - clamp_t(u8, signal, 25, 90);
+ else
+ signal = 95 - clamp_t(u8, signal, 30, 95);
+ rx_status.signal = signal;
rx_status.freq = dev->conf.channel->center_freq;
rx_status.band = dev->conf.channel->band;
rx_status.mactime = le64_to_cpu(entry->tsft);

View File

@ -0,0 +1,276 @@
From: Venkatesh Pallipadi <venki@google.com>
Date: Tue, 18 May 2010 01:14:43 +0000 (-0700)
Subject: sched: Avoid side-effect of tickless idle on update_cpu_load
X-Git-Tag: v2.6.36-rc1~531^2~22
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=fdf3e95d3916f18bf8703fb065499fdbc4dfe34c
sched: Avoid side-effect of tickless idle on update_cpu_load
tickless idle has a negative side effect on update_cpu_load(), which
in turn can affect load balancing behavior.
update_cpu_load() is supposed to be called every tick, to keep track
of various load indicies. With tickless idle, there are no scheduler
ticks called on the idle CPUs. Idle CPUs may still do load balancing
(with idle_load_balance CPU) using the stale cpu_load. It will also
cause problems when all CPUs go idle for a while and become active
again. In this case loads would not degrade as expected.
This is how rq->nr_load_updates change looks like under different
conditions:
<cpu_num> <nr_load_updates change>
All CPUS idle for 10 seconds (HZ=1000)
0 1621
10 496
11 139
12 875
13 1672
14 12
15 21
1 1472
2 2426
3 1161
4 2108
5 1525
6 701
7 249
8 766
9 1967
One CPU busy rest idle for 10 seconds
0 10003
10 601
11 95
12 966
13 1597
14 114
15 98
1 3457
2 93
3 6679
4 1425
5 1479
6 595
7 193
8 633
9 1687
All CPUs busy for 10 seconds
0 10026
10 10026
11 10026
12 10026
13 10025
14 10025
15 10025
1 10026
2 10026
3 10026
4 10026
5 10026
6 10026
7 10026
8 10026
9 10026
That is update_cpu_load works properly only when all CPUs are busy.
If all are idle, all the CPUs get way lower updates. And when few
CPUs are busy and rest are idle, only busy and ilb CPU does proper
updates and rest of the idle CPUs will do lower updates.
The patch keeps track of when a last update was done and fixes up
the load avg based on current time.
On one of my test system SPECjbb with warehouse 1..numcpus, patch
improves throughput numbers by ~1% (average of 6 runs). On another
test system (with different domain hierarchy) there is no noticable
change in perf.
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <AANLkTilLtDWQsAUrIxJ6s04WTgmw9GuOODc5AOrYsaR5@mail.gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
diff --git a/kernel/sched.c b/kernel/sched.c
index f37a961..a757f6b 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -457,6 +457,7 @@ struct rq {
unsigned long nr_running;
#define CPU_LOAD_IDX_MAX 5
unsigned long cpu_load[CPU_LOAD_IDX_MAX];
+ unsigned long last_load_update_tick;
#ifdef CONFIG_NO_HZ
u64 nohz_stamp;
unsigned char in_nohz_recently;
@@ -1803,6 +1804,7 @@ static void cfs_rq_set_shares(struct cfs_rq *cfs_rq, unsigned long shares)
static void calc_load_account_idle(struct rq *this_rq);
static void update_sysctl(void);
static int get_update_sysctl_factor(void);
+static void update_cpu_load(struct rq *this_rq);
static inline void __set_task_cpu(struct task_struct *p, unsigned int cpu)
{
@@ -3050,23 +3052,102 @@ static void calc_load_account_active(struct rq *this_rq)
}
/*
+ * The exact cpuload at various idx values, calculated at every tick would be
+ * load = (2^idx - 1) / 2^idx * load + 1 / 2^idx * cur_load
+ *
+ * If a cpu misses updates for n-1 ticks (as it was idle) and update gets called
+ * on nth tick when cpu may be busy, then we have:
+ * load = ((2^idx - 1) / 2^idx)^(n-1) * load
+ * load = (2^idx - 1) / 2^idx) * load + 1 / 2^idx * cur_load
+ *
+ * decay_load_missed() below does efficient calculation of
+ * load = ((2^idx - 1) / 2^idx)^(n-1) * load
+ * avoiding 0..n-1 loop doing load = ((2^idx - 1) / 2^idx) * load
+ *
+ * The calculation is approximated on a 128 point scale.
+ * degrade_zero_ticks is the number of ticks after which load at any
+ * particular idx is approximated to be zero.
+ * degrade_factor is a precomputed table, a row for each load idx.
+ * Each column corresponds to degradation factor for a power of two ticks,
+ * based on 128 point scale.
+ * Example:
+ * row 2, col 3 (=12) says that the degradation at load idx 2 after
+ * 8 ticks is 12/128 (which is an approximation of exact factor 3^8/4^8).
+ *
+ * With this power of 2 load factors, we can degrade the load n times
+ * by looking at 1 bits in n and doing as many mult/shift instead of
+ * n mult/shifts needed by the exact degradation.
+ */
+#define DEGRADE_SHIFT 7
+static const unsigned char
+ degrade_zero_ticks[CPU_LOAD_IDX_MAX] = {0, 8, 32, 64, 128};
+static const unsigned char
+ degrade_factor[CPU_LOAD_IDX_MAX][DEGRADE_SHIFT + 1] = {
+ {0, 0, 0, 0, 0, 0, 0, 0},
+ {64, 32, 8, 0, 0, 0, 0, 0},
+ {96, 72, 40, 12, 1, 0, 0},
+ {112, 98, 75, 43, 15, 1, 0},
+ {120, 112, 98, 76, 45, 16, 2} };
+
+/*
+ * Update cpu_load for any missed ticks, due to tickless idle. The backlog
+ * would be when CPU is idle and so we just decay the old load without
+ * adding any new load.
+ */
+static unsigned long
+decay_load_missed(unsigned long load, unsigned long missed_updates, int idx)
+{
+ int j = 0;
+
+ if (!missed_updates)
+ return load;
+
+ if (missed_updates >= degrade_zero_ticks[idx])
+ return 0;
+
+ if (idx == 1)
+ return load >> missed_updates;
+
+ while (missed_updates) {
+ if (missed_updates % 2)
+ load = (load * degrade_factor[idx][j]) >> DEGRADE_SHIFT;
+
+ missed_updates >>= 1;
+ j++;
+ }
+ return load;
+}
+
+/*
* Update rq->cpu_load[] statistics. This function is usually called every
- * scheduler tick (TICK_NSEC).
+ * scheduler tick (TICK_NSEC). With tickless idle this will not be called
+ * every tick. We fix it up based on jiffies.
*/
static void update_cpu_load(struct rq *this_rq)
{
unsigned long this_load = this_rq->load.weight;
+ unsigned long curr_jiffies = jiffies;
+ unsigned long pending_updates;
int i, scale;
this_rq->nr_load_updates++;
+ /* Avoid repeated calls on same jiffy, when moving in and out of idle */
+ if (curr_jiffies == this_rq->last_load_update_tick)
+ return;
+
+ pending_updates = curr_jiffies - this_rq->last_load_update_tick;
+ this_rq->last_load_update_tick = curr_jiffies;
+
/* Update our load: */
- for (i = 0, scale = 1; i < CPU_LOAD_IDX_MAX; i++, scale += scale) {
+ this_rq->cpu_load[0] = this_load; /* Fasttrack for idx 0 */
+ for (i = 1, scale = 2; i < CPU_LOAD_IDX_MAX; i++, scale += scale) {
unsigned long old_load, new_load;
/* scale is effectively 1 << i now, and >> i divides by scale */
old_load = this_rq->cpu_load[i];
+ old_load = decay_load_missed(old_load, pending_updates - 1, i);
new_load = this_load;
/*
* Round up the averaging division if load is increasing. This
@@ -3074,9 +3155,15 @@ static void update_cpu_load(struct rq *this_rq)
* example.
*/
if (new_load > old_load)
- new_load += scale-1;
- this_rq->cpu_load[i] = (old_load*(scale-1) + new_load) >> i;
+ new_load += scale - 1;
+
+ this_rq->cpu_load[i] = (old_load * (scale - 1) + new_load) >> i;
}
+}
+
+static void update_cpu_load_active(struct rq *this_rq)
+{
+ update_cpu_load(this_rq);
calc_load_account_active(this_rq);
}
@@ -3464,7 +3551,7 @@ void scheduler_tick(void)
raw_spin_lock(&rq->lock);
update_rq_clock(rq);
- update_cpu_load(rq);
+ update_cpu_load_active(rq);
curr->sched_class->task_tick(rq, curr, 0);
raw_spin_unlock(&rq->lock);
@@ -7688,6 +7775,9 @@ void __init sched_init(void)
for (j = 0; j < CPU_LOAD_IDX_MAX; j++)
rq->cpu_load[j] = 0;
+
+ rq->last_load_update_tick = jiffies;
+
#ifdef CONFIG_SMP
rq->sd = NULL;
rq->rd = NULL;
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index eed35ed..22b8b4f 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -3420,9 +3420,12 @@ static void run_rebalance_domains(struct softirq_action *h)
if (need_resched())
break;
+ rq = cpu_rq(balance_cpu);
+ raw_spin_lock_irq(&rq->lock);
+ update_cpu_load(rq);
+ raw_spin_unlock_irq(&rq->lock);
rebalance_domains(balance_cpu, CPU_IDLE);
- rq = cpu_rq(balance_cpu);
if (time_after(this_rq->next_balance, rq->next_balance))
this_rq->next_balance = rq->next_balance;
}

View File

@ -0,0 +1,651 @@
From: Venkatesh Pallipadi <venki@google.com>
Date: Sat, 22 May 2010 00:09:41 +0000 (-0700)
Subject: sched: Change nohz idle load balancing logic to push model
X-Git-Tag: v2.6.36-rc1~531^2~21
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=83cd4fe27ad8446619b2e030b171b858501de87d
sched: Change nohz idle load balancing logic to push model
In the new push model, all idle CPUs indeed go into nohz mode. There is
still the concept of idle load balancer (performing the load balancing
on behalf of all the idle cpu's in the system). Busy CPU kicks the nohz
balancer when any of the nohz CPUs need idle load balancing.
The kickee CPU does the idle load balancing on behalf of all idle CPUs
instead of the normal idle balance.
This addresses the below two problems with the current nohz ilb logic:
* the idle load balancer continued to have periodic ticks during idle and
wokeup frequently, even though it did not have any rebalancing to do on
behalf of any of the idle CPUs.
* On x86 and CPUs that have APIC timer stoppage on idle CPUs, this
periodic wakeup can result in a periodic additional interrupt on a CPU
doing the timer broadcast.
Also currently we are migrating the unpinned timers from an idle to the cpu
doing idle load balancing (when all the cpus in the system are idle,
there is no idle load balancing cpu and timers get added to the same idle cpu
where the request was made. So the existing optimization works only on semi idle
system).
And In semi idle system, we no longer have periodic ticks on the idle load
balancer CPU. Using that cpu will add more delays to the timers than intended
(as that cpu's timer base may not be uptodate wrt jiffies etc). This was
causing mysterious slowdowns during boot etc.
For now, in the semi idle case, use the nearest busy cpu for migrating timers
from an idle cpu. This is good for power-savings anyway.
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: Thomas Gleixner <tglx@linutronix.de>
LKML-Reference: <1274486981.2840.46.camel@sbs-t61.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
[ backported for 2.6.35 ]
diff --git a/include/linux/sched.h b/include/linux/sched.h
index c2d4316..a3e5b1c 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -271,13 +271,10 @@ extern int runqueue_is_locked(int cpu);
extern cpumask_var_t nohz_cpu_mask;
#if defined(CONFIG_SMP) && defined(CONFIG_NO_HZ)
-extern int select_nohz_load_balancer(int cpu);
-extern int get_nohz_load_balancer(void);
+extern void select_nohz_load_balancer(int stop_tick);
+extern int get_nohz_timer_target(void);
#else
-static inline int select_nohz_load_balancer(int cpu)
-{
- return 0;
-}
+static inline void select_nohz_load_balancer(int stop_tick) { }
#endif
/*
diff --git a/kernel/hrtimer.c b/kernel/hrtimer.c
index 5c69e99..e934339 100644
--- a/kernel/hrtimer.c
+++ b/kernel/hrtimer.c
@@ -144,12 +144,8 @@ struct hrtimer_clock_base *lock_hrtimer_base(const struct hrtimer *timer,
static int hrtimer_get_target(int this_cpu, int pinned)
{
#ifdef CONFIG_NO_HZ
- if (!pinned && get_sysctl_timer_migration() && idle_cpu(this_cpu)) {
- int preferred_cpu = get_nohz_load_balancer();
-
- if (preferred_cpu >= 0)
- return preferred_cpu;
- }
+ if (!pinned && get_sysctl_timer_migration() && idle_cpu(this_cpu))
+ return get_nohz_timer_target();
#endif
return this_cpu;
}
diff --git a/kernel/sched.c b/kernel/sched.c
index a757f6b..132950b 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -460,7 +460,7 @@ struct rq {
unsigned long last_load_update_tick;
#ifdef CONFIG_NO_HZ
u64 nohz_stamp;
- unsigned char in_nohz_recently;
+ unsigned char nohz_balance_kick;
#endif
unsigned int skip_clock_update;
@@ -1195,6 +1195,27 @@ static void resched_cpu(int cpu)
#ifdef CONFIG_NO_HZ
/*
+ * In the semi idle case, use the nearest busy cpu for migrating timers
+ * from an idle cpu. This is good for power-savings.
+ *
+ * We don't do similar optimization for completely idle system, as
+ * selecting an idle cpu will add more delays to the timers than intended
+ * (as that cpu's timer base may not be uptodate wrt jiffies etc).
+ */
+int get_nohz_timer_target(void)
+{
+ int cpu = smp_processor_id();
+ int i;
+ struct sched_domain *sd;
+
+ for_each_domain(cpu, sd) {
+ for_each_cpu(i, sched_domain_span(sd))
+ if (!idle_cpu(i))
+ return i;
+ }
+ return cpu;
+}
+/*
* When add_timer_on() enqueues a timer into the timer wheel of an
* idle CPU then this timer might expire before the next timer event
* which is scheduled to wake up that CPU. In case of a completely
@@ -7791,6 +7812,10 @@ void __init sched_init(void)
rq->idle_stamp = 0;
rq->avg_idle = 2*sysctl_sched_migration_cost;
rq_attach_root(rq, &def_root_domain);
+#ifdef CONFIG_NO_HZ
+ rq->nohz_balance_kick = 0;
+ init_sched_softirq_csd(&per_cpu(remote_sched_softirq_cb, i));
+#endif
#endif
init_rq_hrtick(rq);
atomic_set(&rq->nr_iowait, 0);
@@ -7835,8 +7860,11 @@ void __init sched_init(void)
zalloc_cpumask_var(&nohz_cpu_mask, GFP_NOWAIT);
#ifdef CONFIG_SMP
#ifdef CONFIG_NO_HZ
- zalloc_cpumask_var(&nohz.cpu_mask, GFP_NOWAIT);
- alloc_cpumask_var(&nohz.ilb_grp_nohz_mask, GFP_NOWAIT);
+ zalloc_cpumask_var(&nohz.idle_cpus_mask, GFP_NOWAIT);
+ alloc_cpumask_var(&nohz.grp_idle_mask, GFP_NOWAIT);
+ atomic_set(&nohz.load_balancer, nr_cpu_ids);
+ atomic_set(&nohz.first_pick_cpu, nr_cpu_ids);
+ atomic_set(&nohz.second_pick_cpu, nr_cpu_ids);
#endif
/* May be allocated at isolcpus cmdline parse time */
if (cpu_isolated_map == NULL)
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 22b8b4f..6ee2e0a 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -3091,13 +3091,40 @@ out_unlock:
}
#ifdef CONFIG_NO_HZ
+
+static DEFINE_PER_CPU(struct call_single_data, remote_sched_softirq_cb);
+
+static void trigger_sched_softirq(void *data)
+{
+ raise_softirq_irqoff(SCHED_SOFTIRQ);
+}
+
+static inline void init_sched_softirq_csd(struct call_single_data *csd)
+{
+ csd->func = trigger_sched_softirq;
+ csd->info = NULL;
+ csd->flags = 0;
+ csd->priv = 0;
+}
+
+/*
+ * idle load balancing details
+ * - One of the idle CPUs nominates itself as idle load_balancer, while
+ * entering idle.
+ * - This idle load balancer CPU will also go into tickless mode when
+ * it is idle, just like all other idle CPUs
+ * - When one of the busy CPUs notice that there may be an idle rebalancing
+ * needed, they will kick the idle load balancer, which then does idle
+ * load balancing for all the idle CPUs.
+ */
static struct {
atomic_t load_balancer;
- cpumask_var_t cpu_mask;
- cpumask_var_t ilb_grp_nohz_mask;
-} nohz ____cacheline_aligned = {
- .load_balancer = ATOMIC_INIT(-1),
-};
+ atomic_t first_pick_cpu;
+ atomic_t second_pick_cpu;
+ cpumask_var_t idle_cpus_mask;
+ cpumask_var_t grp_idle_mask;
+ unsigned long next_balance; /* in jiffy units */
+} nohz ____cacheline_aligned;
int get_nohz_load_balancer(void)
{
@@ -3151,17 +3178,17 @@ static inline struct sched_domain *lowest_flag_domain(int cpu, int flag)
*/
static inline int is_semi_idle_group(struct sched_group *ilb_group)
{
- cpumask_and(nohz.ilb_grp_nohz_mask, nohz.cpu_mask,
+ cpumask_and(nohz.grp_idle_mask, nohz.idle_cpus_mask,
sched_group_cpus(ilb_group));
/*
* A sched_group is semi-idle when it has atleast one busy cpu
* and atleast one idle cpu.
*/
- if (cpumask_empty(nohz.ilb_grp_nohz_mask))
+ if (cpumask_empty(nohz.grp_idle_mask))
return 0;
- if (cpumask_equal(nohz.ilb_grp_nohz_mask, sched_group_cpus(ilb_group)))
+ if (cpumask_equal(nohz.grp_idle_mask, sched_group_cpus(ilb_group)))
return 0;
return 1;
@@ -3194,7 +3221,7 @@ static int find_new_ilb(int cpu)
* Optimize for the case when we have no idle CPUs or only one
* idle CPU. Don't walk the sched_domain hierarchy in such cases
*/
- if (cpumask_weight(nohz.cpu_mask) < 2)
+ if (cpumask_weight(nohz.idle_cpus_mask) < 2)
goto out_done;
for_each_flag_domain(cpu, sd, SD_POWERSAVINGS_BALANCE) {
@@ -3202,7 +3229,7 @@ static int find_new_ilb(int cpu)
do {
if (is_semi_idle_group(ilb_group))
- return cpumask_first(nohz.ilb_grp_nohz_mask);
+ return cpumask_first(nohz.grp_idle_mask);
ilb_group = ilb_group->next;
@@ -3210,98 +3237,116 @@ static int find_new_ilb(int cpu)
}
out_done:
- return cpumask_first(nohz.cpu_mask);
+ return nr_cpu_ids;
}
#else /* (CONFIG_SCHED_MC || CONFIG_SCHED_SMT) */
static inline int find_new_ilb(int call_cpu)
{
- return cpumask_first(nohz.cpu_mask);
+ return nr_cpu_ids;
}
#endif
/*
+ * Kick a CPU to do the nohz balancing, if it is time for it. We pick the
+ * nohz_load_balancer CPU (if there is one) otherwise fallback to any idle
+ * CPU (if there is one).
+ */
+static void nohz_balancer_kick(int cpu)
+{
+ int ilb_cpu;
+
+ nohz.next_balance++;
+
+ ilb_cpu = get_nohz_load_balancer();
+
+ if (ilb_cpu >= nr_cpu_ids) {
+ ilb_cpu = cpumask_first(nohz.idle_cpus_mask);
+ if (ilb_cpu >= nr_cpu_ids)
+ return;
+ }
+
+ if (!cpu_rq(ilb_cpu)->nohz_balance_kick) {
+ struct call_single_data *cp;
+
+ cpu_rq(ilb_cpu)->nohz_balance_kick = 1;
+ cp = &per_cpu(remote_sched_softirq_cb, cpu);
+ __smp_call_function_single(ilb_cpu, cp, 0);
+ }
+ return;
+}
+
+/*
* This routine will try to nominate the ilb (idle load balancing)
* owner among the cpus whose ticks are stopped. ilb owner will do the idle
- * load balancing on behalf of all those cpus. If all the cpus in the system
- * go into this tickless mode, then there will be no ilb owner (as there is
- * no need for one) and all the cpus will sleep till the next wakeup event
- * arrives...
- *
- * For the ilb owner, tick is not stopped. And this tick will be used
- * for idle load balancing. ilb owner will still be part of
- * nohz.cpu_mask..
+ * load balancing on behalf of all those cpus.
*
- * While stopping the tick, this cpu will become the ilb owner if there
- * is no other owner. And will be the owner till that cpu becomes busy
- * or if all cpus in the system stop their ticks at which point
- * there is no need for ilb owner.
+ * When the ilb owner becomes busy, we will not have new ilb owner until some
+ * idle CPU wakes up and goes back to idle or some busy CPU tries to kick
+ * idle load balancing by kicking one of the idle CPUs.
*
- * When the ilb owner becomes busy, it nominates another owner, during the
- * next busy scheduler_tick()
+ * Ticks are stopped for the ilb owner as well, with busy CPU kicking this
+ * ilb owner CPU in future (when there is a need for idle load balancing on
+ * behalf of all idle CPUs).
*/
-int select_nohz_load_balancer(int stop_tick)
+void select_nohz_load_balancer(int stop_tick)
{
int cpu = smp_processor_id();
if (stop_tick) {
- cpu_rq(cpu)->in_nohz_recently = 1;
-
if (!cpu_active(cpu)) {
if (atomic_read(&nohz.load_balancer) != cpu)
- return 0;
+ return;
/*
* If we are going offline and still the leader,
* give up!
*/
- if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
+ if (atomic_cmpxchg(&nohz.load_balancer, cpu,
+ nr_cpu_ids) != cpu)
BUG();
- return 0;
+ return;
}
- cpumask_set_cpu(cpu, nohz.cpu_mask);
+ cpumask_set_cpu(cpu, nohz.idle_cpus_mask);
- /* time for ilb owner also to sleep */
- if (cpumask_weight(nohz.cpu_mask) == num_active_cpus()) {
- if (atomic_read(&nohz.load_balancer) == cpu)
- atomic_set(&nohz.load_balancer, -1);
- return 0;
- }
+ if (atomic_read(&nohz.first_pick_cpu) == cpu)
+ atomic_cmpxchg(&nohz.first_pick_cpu, cpu, nr_cpu_ids);
+ if (atomic_read(&nohz.second_pick_cpu) == cpu)
+ atomic_cmpxchg(&nohz.second_pick_cpu, cpu, nr_cpu_ids);
- if (atomic_read(&nohz.load_balancer) == -1) {
- /* make me the ilb owner */
- if (atomic_cmpxchg(&nohz.load_balancer, -1, cpu) == -1)
- return 1;
- } else if (atomic_read(&nohz.load_balancer) == cpu) {
+ if (atomic_read(&nohz.load_balancer) >= nr_cpu_ids) {
int new_ilb;
- if (!(sched_smt_power_savings ||
- sched_mc_power_savings))
- return 1;
+ /* make me the ilb owner */
+ if (atomic_cmpxchg(&nohz.load_balancer, nr_cpu_ids,
+ cpu) != nr_cpu_ids)
+ return;
+
/*
* Check to see if there is a more power-efficient
* ilb.
*/
new_ilb = find_new_ilb(cpu);
if (new_ilb < nr_cpu_ids && new_ilb != cpu) {
- atomic_set(&nohz.load_balancer, -1);
+ atomic_set(&nohz.load_balancer, nr_cpu_ids);
resched_cpu(new_ilb);
- return 0;
+ return;
}
- return 1;
+ return;
}
} else {
- if (!cpumask_test_cpu(cpu, nohz.cpu_mask))
- return 0;
+ if (!cpumask_test_cpu(cpu, nohz.idle_cpus_mask))
+ return;
- cpumask_clear_cpu(cpu, nohz.cpu_mask);
+ cpumask_clear_cpu(cpu, nohz.idle_cpus_mask);
if (atomic_read(&nohz.load_balancer) == cpu)
- if (atomic_cmpxchg(&nohz.load_balancer, cpu, -1) != cpu)
+ if (atomic_cmpxchg(&nohz.load_balancer, cpu,
+ nr_cpu_ids) != cpu)
BUG();
}
- return 0;
+ return;
}
#endif
@@ -3383,11 +3428,101 @@ out:
rq->next_balance = next_balance;
}
+#ifdef CONFIG_NO_HZ
/*
- * run_rebalance_domains is triggered when needed from the scheduler tick.
- * In CONFIG_NO_HZ case, the idle load balance owner will do the
+ * In CONFIG_NO_HZ case, the idle balance kickee will do the
* rebalancing for all the cpus for whom scheduler ticks are stopped.
*/
+static void nohz_idle_balance(int this_cpu, enum cpu_idle_type idle)
+{
+ struct rq *this_rq = cpu_rq(this_cpu);
+ struct rq *rq;
+ int balance_cpu;
+
+ if (idle != CPU_IDLE || !this_rq->nohz_balance_kick)
+ return;
+
+ for_each_cpu(balance_cpu, nohz.idle_cpus_mask) {
+ if (balance_cpu == this_cpu)
+ continue;
+
+ /*
+ * If this cpu gets work to do, stop the load balancing
+ * work being done for other cpus. Next load
+ * balancing owner will pick it up.
+ */
+ if (need_resched()) {
+ this_rq->nohz_balance_kick = 0;
+ break;
+ }
+
+ raw_spin_lock_irq(&this_rq->lock);
+ update_cpu_load(this_rq);
+ raw_spin_unlock_irq(&this_rq->lock);
+
+ rebalance_domains(balance_cpu, CPU_IDLE);
+
+ rq = cpu_rq(balance_cpu);
+ if (time_after(this_rq->next_balance, rq->next_balance))
+ this_rq->next_balance = rq->next_balance;
+ }
+ nohz.next_balance = this_rq->next_balance;
+ this_rq->nohz_balance_kick = 0;
+}
+
+/*
+ * Current heuristic for kicking the idle load balancer
+ * - first_pick_cpu is the one of the busy CPUs. It will kick
+ * idle load balancer when it has more than one process active. This
+ * eliminates the need for idle load balancing altogether when we have
+ * only one running process in the system (common case).
+ * - If there are more than one busy CPU, idle load balancer may have
+ * to run for active_load_balance to happen (i.e., two busy CPUs are
+ * SMT or core siblings and can run better if they move to different
+ * physical CPUs). So, second_pick_cpu is the second of the busy CPUs
+ * which will kick idle load balancer as soon as it has any load.
+ */
+static inline int nohz_kick_needed(struct rq *rq, int cpu)
+{
+ unsigned long now = jiffies;
+ int ret;
+ int first_pick_cpu, second_pick_cpu;
+
+ if (time_before(now, nohz.next_balance))
+ return 0;
+
+ if (!rq->nr_running)
+ return 0;
+
+ first_pick_cpu = atomic_read(&nohz.first_pick_cpu);
+ second_pick_cpu = atomic_read(&nohz.second_pick_cpu);
+
+ if (first_pick_cpu < nr_cpu_ids && first_pick_cpu != cpu &&
+ second_pick_cpu < nr_cpu_ids && second_pick_cpu != cpu)
+ return 0;
+
+ ret = atomic_cmpxchg(&nohz.first_pick_cpu, nr_cpu_ids, cpu);
+ if (ret == nr_cpu_ids || ret == cpu) {
+ atomic_cmpxchg(&nohz.second_pick_cpu, cpu, nr_cpu_ids);
+ if (rq->nr_running > 1)
+ return 1;
+ } else {
+ ret = atomic_cmpxchg(&nohz.second_pick_cpu, nr_cpu_ids, cpu);
+ if (ret == nr_cpu_ids || ret == cpu) {
+ if (rq->nr_running)
+ return 1;
+ }
+ }
+ return 0;
+}
+#else
+static void nohz_idle_balance(int this_cpu, enum cpu_idle_type idle) { }
+#endif
+
+/*
+ * run_rebalance_domains is triggered when needed from the scheduler tick.
+ * Also triggered for nohz idle balancing (with nohz_balancing_kick set).
+ */
static void run_rebalance_domains(struct softirq_action *h)
{
int this_cpu = smp_processor_id();
@@ -3397,40 +3532,12 @@ static void run_rebalance_domains(struct softirq_action *h)
rebalance_domains(this_cpu, idle);
-#ifdef CONFIG_NO_HZ
/*
- * If this cpu is the owner for idle load balancing, then do the
+ * If this cpu has a pending nohz_balance_kick, then do the
* balancing on behalf of the other idle cpus whose ticks are
* stopped.
*/
- if (this_rq->idle_at_tick &&
- atomic_read(&nohz.load_balancer) == this_cpu) {
- struct rq *rq;
- int balance_cpu;
-
- for_each_cpu(balance_cpu, nohz.cpu_mask) {
- if (balance_cpu == this_cpu)
- continue;
-
- /*
- * If this cpu gets work to do, stop the load balancing
- * work being done for other cpus. Next load
- * balancing owner will pick it up.
- */
- if (need_resched())
- break;
-
- rq = cpu_rq(balance_cpu);
- raw_spin_lock_irq(&rq->lock);
- update_cpu_load(rq);
- raw_spin_unlock_irq(&rq->lock);
- rebalance_domains(balance_cpu, CPU_IDLE);
-
- if (time_after(this_rq->next_balance, rq->next_balance))
- this_rq->next_balance = rq->next_balance;
- }
- }
-#endif
+ nohz_idle_balance(this_cpu, idle);
}
static inline int on_null_domain(int cpu)
@@ -3440,57 +3547,17 @@ static inline int on_null_domain(int cpu)
/*
* Trigger the SCHED_SOFTIRQ if it is time to do periodic load balancing.
- *
- * In case of CONFIG_NO_HZ, this is the place where we nominate a new
- * idle load balancing owner or decide to stop the periodic load balancing,
- * if the whole system is idle.
*/
static inline void trigger_load_balance(struct rq *rq, int cpu)
{
-#ifdef CONFIG_NO_HZ
- /*
- * If we were in the nohz mode recently and busy at the current
- * scheduler tick, then check if we need to nominate new idle
- * load balancer.
- */
- if (rq->in_nohz_recently && !rq->idle_at_tick) {
- rq->in_nohz_recently = 0;
-
- if (atomic_read(&nohz.load_balancer) == cpu) {
- cpumask_clear_cpu(cpu, nohz.cpu_mask);
- atomic_set(&nohz.load_balancer, -1);
- }
-
- if (atomic_read(&nohz.load_balancer) == -1) {
- int ilb = find_new_ilb(cpu);
-
- if (ilb < nr_cpu_ids)
- resched_cpu(ilb);
- }
- }
-
- /*
- * If this cpu is idle and doing idle load balancing for all the
- * cpus with ticks stopped, is it time for that to stop?
- */
- if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) == cpu &&
- cpumask_weight(nohz.cpu_mask) == num_online_cpus()) {
- resched_cpu(cpu);
- return;
- }
-
- /*
- * If this cpu is idle and the idle load balancing is done by
- * someone else, then no need raise the SCHED_SOFTIRQ
- */
- if (rq->idle_at_tick && atomic_read(&nohz.load_balancer) != cpu &&
- cpumask_test_cpu(cpu, nohz.cpu_mask))
- return;
-#endif
/* Don't need to rebalance while attached to NULL domain */
if (time_after_eq(jiffies, rq->next_balance) &&
likely(!on_null_domain(cpu)))
raise_softirq(SCHED_SOFTIRQ);
+#ifdef CONFIG_NO_HZ
+ else if (nohz_kick_needed(rq, cpu) && likely(!on_null_domain(cpu)))
+ nohz_balancer_kick(cpu);
+#endif
}
static void rq_online_fair(struct rq *rq)
diff --git a/kernel/time/tick-sched.c b/kernel/time/tick-sched.c
index 1d7b9bc..5f171f0 100644
--- a/kernel/time/tick-sched.c
+++ b/kernel/time/tick-sched.c
@@ -408,13 +408,7 @@ void tick_nohz_stop_sched_tick(int inidle)
* the scheduler tick in nohz_restart_sched_tick.
*/
if (!ts->tick_stopped) {
- if (select_nohz_load_balancer(1)) {
- /*
- * sched tick not stopped!
- */
- cpumask_clear_cpu(cpu, nohz_cpu_mask);
- goto out;
- }
+ select_nohz_load_balancer(1);
ts->idle_tick = hrtimer_get_expires(&ts->sched_timer);
ts->tick_stopped = 1;
diff --git a/kernel/timer.c b/kernel/timer.c
index ee305c8..48d6aec 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -679,12 +679,8 @@ __mod_timer(struct timer_list *timer, unsigned long expires,
cpu = smp_processor_id();
#if defined(CONFIG_NO_HZ) && defined(CONFIG_SMP)
- if (!pinned && get_sysctl_timer_migration() && idle_cpu(cpu)) {
- int preferred_cpu = get_nohz_load_balancer();
-
- if (preferred_cpu >= 0)
- cpu = preferred_cpu;
- }
+ if (!pinned && get_sysctl_timer_migration() && idle_cpu(cpu))
+ cpu = get_nohz_timer_target();
#endif
new_base = per_cpu(tvec_bases, cpu);

View File

@ -0,0 +1,28 @@
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Fri, 9 Jul 2010 13:19:54 +0000 (+0200)
Subject: sched: Update rq->clock for nohz balanced cpus
X-Git-Tag: v2.6.36-rc1~531^2~5
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=5343bdb8fd076f16edc9d113a9e35e2a1d1f4966
sched: Update rq->clock for nohz balanced cpus
Suresh spotted that we don't update the rq->clock in the nohz
load-balancer path.
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1278626014.2834.74.camel@sbs-t61.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index b4da534..e44a591 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -3596,6 +3596,7 @@ static void nohz_idle_balance(int this_cpu, enum cpu_idle_type idle)
}
raw_spin_lock_irq(&this_rq->lock);
+ update_rq_clock(this_rq);
update_cpu_load(this_rq);
raw_spin_unlock_irq(&this_rq->lock);

View File

@ -0,0 +1,38 @@
From: Peter Zijlstra <peterz@infradead.org>
Date: Thu, 19 Aug 2010 11:31:43 +0000 (+0200)
Subject: sched: Fix rq->clock synchronization when migrating tasks
X-Git-Tag: v2.6.36-rc3~25^2~1
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=861d034ee814917a83bd5de4b26e3b8336ddeeb8
sched: Fix rq->clock synchronization when migrating tasks
sched_fork() -- we do task placement in ->task_fork_fair() ensure we
update_rq_clock() so we work with current time. We leave the vruntime
in relative state, so the time delay until wake_up_new_task() doesn't
matter.
wake_up_new_task() -- Since task_fork_fair() left p->vruntime in
relative state we can safely migrate, the activate_task() on the
remote rq will call update_rq_clock() and causes the clock to be
synced (enough).
Tested-by: Jack Daniel <wanders.thirst@gmail.com>
Tested-by: Philby John <pjohn@mvista.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1281002322.1923.1708.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index 806d1b2..ab661eb 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -3752,6 +3752,8 @@ static void task_fork_fair(struct task_struct *p)
raw_spin_lock_irqsave(&rq->lock, flags);
+ update_rq_clock(rq);
+
if (unlikely(task_cpu(p) != this_cpu))
__set_task_cpu(p, this_cpu);

View File

@ -0,0 +1,58 @@
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Mon, 23 Aug 2010 20:42:51 +0000 (-0700)
Subject: sched: Move sched_avg_update() to update_cpu_load()
X-Git-Tag: v2.6.36-rc4~8^2~1
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=da2b71edd8a7db44fe1746261410a981f3e03632
sched: Move sched_avg_update() to update_cpu_load()
Currently sched_avg_update() (which updates rt_avg stats in the rq)
is getting called from scale_rt_power() (in the load balance context)
which doesn't take rq->lock.
Fix it by moving the sched_avg_update() to more appropriate
update_cpu_load() where the CFS load gets updated as well.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1282596171.2694.3.camel@sbsiddha-MOBL3>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
diff --git a/kernel/sched.c b/kernel/sched.c
index 09b574e..ed09d4f 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -1294,6 +1294,10 @@ static void resched_task(struct task_struct *p)
static void sched_rt_avg_update(struct rq *rq, u64 rt_delta)
{
}
+
+static void sched_avg_update(struct rq *rq)
+{
+}
#endif /* CONFIG_SMP */
#if BITS_PER_LONG == 32
@@ -3182,6 +3186,8 @@ static void update_cpu_load(struct rq *this_rq)
this_rq->cpu_load[i] = (old_load * (scale - 1) + new_load) >> i;
}
+
+ sched_avg_update(this_rq);
}
static void update_cpu_load_active(struct rq *this_rq)
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index ab661eb..f53ec75 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -2268,8 +2268,6 @@ unsigned long scale_rt_power(int cpu)
struct rq *rq = cpu_rq(cpu);
u64 total, available;
- sched_avg_update(rq);
-
total = sched_avg_period() + (rq->clock - rq->age_stamp);
available = total - rq->rt_avg;

View File

@ -0,0 +1,39 @@
From: Suresh Siddha <suresh.b.siddha@intel.com>
Date: Mon, 13 Sep 2010 18:02:21 +0000 (-0700)
Subject: sched: Fix nohz balance kick
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Ftorvalds%2Flinux-2.6.git;a=commitdiff_plain;h=f6c3f1686e7ec1dd8725a9a3dcb857dfd0c7a5bf
sched: Fix nohz balance kick
There's a situation where the nohz balancer will try to wake itself:
cpu-x is idle which is also ilb_cpu
got a scheduler tick during idle
and the nohz_kick_needed() in trigger_load_balance() checks for
rq_x->nr_running which might not be zero (because of someone waking a
task on this rq etc) and this leads to the situation of the cpu-x
sending a kick to itself.
And this can cause a lockup.
Avoid this by not marking ourself eligible for kicking.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1284400941.2684.19.camel@sbsiddha-MOBL3.sc.intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index a171138..db3f674 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -3630,7 +3630,7 @@ static inline int nohz_kick_needed(struct rq *rq, int cpu)
if (time_before(now, nohz.next_balance))
return 0;
- if (!rq->nr_running)
+ if (rq->idle_at_tick)
return 0;
first_pick_cpu = atomic_read(&nohz.first_pick_cpu);

View File

@ -0,0 +1,93 @@
From: Venkatesh Pallipadi <venki@google.com>
Date: Sat, 11 Sep 2010 01:19:17 +0000 (-0700)
Subject: sched: Increment cache_nice_tries only on periodic lb
X-Git-Url: http://git.kernel.org/?p=linux%2Fkernel%2Fgit%2Fmingo%2Flinux-2.6-x86.git;a=commitdiff_plain;h=58b26c4c025778c09c7a1438ff185080e11b7d0a
sched: Increment cache_nice_tries only on periodic lb
scheduler uses cache_nice_tries as an indicator to do cache_hot and
active load balance, when normal load balance fails. Currently,
this value is changed on any failed load balance attempt. That ends
up being not so nice to workloads that enter/exit idle often, as
they do more frequent new_idle balance and that pretty soon results
in cache hot tasks being pulled in.
Making the cache_nice_tries ignore failed new_idle balance seems to
make better sense. With that only the failed load balance in
periodic load balance gets accounted and the rate of accumulation
of cache_nice_tries will not depend on idle entry/exit (short
running sleep-wakeup kind of tasks). This reduces movement of
cache_hot tasks.
schedstat diff (after-before) excerpt from a workload that has
frequent and short wakeup-idle pattern (:2 in cpu col below refers
to NEWIDLE idx) This snapshot was across ~400 seconds.
Without this change:
domainstats: domain0
cpu cnt bln fld imb gain hgain nobusyq nobusyg
0:2 306487 219575 73167 110069413 44583 19070 1172 218403
1:2 292139 194853 81421 120893383 50745 21902 1259 193594
2:2 283166 174607 91359 129699642 54931 23688 1287 173320
3:2 273998 161788 93991 132757146 57122 24351 1366 160422
4:2 289851 215692 62190 83398383 36377 13680 851 214841
5:2 316312 222146 77605 117582154 49948 20281 988 221158
6:2 297172 195596 83623 122133390 52801 21301 929 194667
7:2 283391 178078 86378 126622761 55122 22239 928 177150
8:2 297655 210359 72995 110246694 45798 19777 1125 209234
9:2 297357 202011 79363 119753474 50953 22088 1089 200922
10:2 278797 178703 83180 122514385 52969 22726 1128 177575
11:2 272661 167669 86978 127342327 55857 24342 1195 166474
12:2 293039 204031 73211 110282059 47285 19651 948 203083
13:2 289502 196762 76803 114712942 49339 20547 1016 195746
14:2 264446 169609 78292 115715605 50459 21017 982 168627
15:2 260968 163660 80142 116811793 51483 21281 1064 162596
With this change:
domainstats: domain0
cpu cnt bln fld imb gain hgain nobusyq nobusyg
0:2 272347 187380 77455 105420270 24975 1 953 186427
1:2 267276 172360 86234 116242264 28087 6 1028 171332
2:2 259769 156777 93281 123243134 30555 1 1043 155734
3:2 250870 143129 97627 127370868 32026 6 1188 141941
4:2 248422 177116 64096 78261112 22202 2 757 176359
5:2 275595 180683 84950 116075022 29400 6 778 179905
6:2 262418 162609 88944 119256898 31056 4 817 161792
7:2 252204 147946 92646 122388300 32879 4 824 147122
8:2 262335 172239 81631 110477214 26599 4 864 171375
9:2 261563 164775 88016 117203621 28331 3 849 163926
10:2 243389 140949 93379 121353071 29585 2 909 140040
11:2 242795 134651 98310 124768957 30895 2 1016 133635
12:2 255234 166622 79843 104696912 26483 4 746 165876
13:2 244944 151595 83855 109808099 27787 3 801 150794
14:2 241301 140982 89935 116954383 30403 6 845 140137
15:2 232271 128564 92821 119185207 31207 4 1416 127148
Signed-off-by: Venkatesh Pallipadi <venki@google.com>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
LKML-Reference: <1284167957-3675-1-git-send-email-venki@google.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
[ 2.6.35.x backport ]
diff --git a/kernel/sched_fair.c b/kernel/sched_fair.c
index a171138..aa16cf1 100644
--- a/kernel/sched_fair.c
+++ b/kernel/sched_fair.c
@@ -3031,7 +3031,14 @@ redo:
if (!ld_moved) {
schedstat_inc(sd, lb_failed[idle]);
- sd->nr_balance_failed++;
+ /*
+ * Increment the failure counter only on periodic balance.
+ * We do not want newidle balance, which can be very
+ * frequent, pollute the failure counter causing
+ * excessive cache_hot migrations and active balances.
+ */
+ if (idle != CPU_NEWLY_IDLE)
+ sd->nr_balance_failed++;
if (need_active_balance(sd, sd_idle, idle)) {
raw_spin_lock_irqsave(&busiest->lock, flags);

View File

@ -0,0 +1,273 @@
From bounces.tip@hpa.at.zytor.com Wed Dec 8 15:40:48 2010
From: tip-bot for Peter Zijlstra <a.p.zijlstra@chello.nl>
In-Reply-To: <1291129145.32004.874.camel@laptop>
References: <1291129145.32004.874.camel@laptop>
Subject: [tip:sched/urgent] sched: Cure more NO_HZ load average woes
Message-ID: <tip-0f004f5a696a9434b7214d0d3cbd0525ee77d428@git.kernel.org>
Git-Commit-ID: 0f004f5a696a9434b7214d0d3cbd0525ee77d428
Commit-ID: 0f004f5a696a9434b7214d0d3cbd0525ee77d428
Gitweb: http://git.kernel.org/tip/0f004f5a696a9434b7214d0d3cbd0525ee77d428
Author: Peter Zijlstra <a.p.zijlstra@chello.nl>
AuthorDate: Tue, 30 Nov 2010 19:48:45 +0100
Committer: Ingo Molnar <mingo@elte.hu>
CommitDate: Wed, 8 Dec 2010 20:15:04 +0100
sched: Cure more NO_HZ load average woes
There's a long-running regression that proved difficult to fix and
which is hitting certain people and is rather annoying in its effects.
Damien reported that after 74f5187ac8 (sched: Cure load average vs
NO_HZ woes) his load average is unnaturally high, he also noted that
even with that patch reverted the load avgerage numbers are not
correct.
The problem is that the previous patch only solved half the NO_HZ
problem, it addressed the part of going into NO_HZ mode, not of
comming out of NO_HZ mode. This patch implements that missing half.
When comming out of NO_HZ mode there are two important things to take
care of:
- Folding the pending idle delta into the global active count.
- Correctly aging the averages for the idle-duration.
So with this patch the NO_HZ interaction should be complete and
behaviour between CONFIG_NO_HZ=[yn] should be equivalent.
Furthermore, this patch slightly changes the load average computation
by adding a rounding term to the fixed point multiplication.
Reported-by: Damien Wyart <damien.wyart@free.fr>
Reported-by: Tim McGrath <tmhikaru@gmail.com>
Tested-by: Damien Wyart <damien.wyart@free.fr>
Tested-by: Orion Poplawski <orion@cora.nwra.com>
Tested-by: Kyle McMartin <kyle@mcmartin.ca>
Signed-off-by: Peter Zijlstra <a.p.zijlstra@chello.nl>
Cc: stable@kernel.org
Cc: Chase Douglas <chase.douglas@canonical.com>
LKML-Reference: <1291129145.32004.874.camel@laptop>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
---
include/linux/sched.h | 2 +-
kernel/sched.c | 150 +++++++++++++++++++++++++++++++++++++++++++++----
kernel/timer.c | 2 +-
3 files changed, 141 insertions(+), 13 deletions(-)
diff --git a/include/linux/sched.h b/include/linux/sched.h
index 2c79e92..2238745 100644
--- a/include/linux/sched.h
+++ b/include/linux/sched.h
@@ -143,7 +143,7 @@ extern unsigned long nr_iowait_cpu(int cpu);
extern unsigned long this_cpu_load(void);
-extern void calc_global_load(void);
+extern void calc_global_load(unsigned long ticks);
extern unsigned long get_parent_ip(unsigned long addr);
diff --git a/kernel/sched.c b/kernel/sched.c
index dc91a4d..6b7c26a 100644
--- a/kernel/sched.c
+++ b/kernel/sched.c
@@ -3119,6 +3119,15 @@ static long calc_load_fold_active(struct rq *this_rq)
return delta;
}
+static unsigned long
+calc_load(unsigned long load, unsigned long exp, unsigned long active)
+{
+ load *= exp;
+ load += active * (FIXED_1 - exp);
+ load += 1UL << (FSHIFT - 1);
+ return load >> FSHIFT;
+}
+
#ifdef CONFIG_NO_HZ
/*
* For NO_HZ we delay the active fold to the next LOAD_FREQ update.
@@ -3148,6 +3157,128 @@ static long calc_load_fold_idle(void)
return delta;
}
+
+/**
+ * fixed_power_int - compute: x^n, in O(log n) time
+ *
+ * @x: base of the power
+ * @frac_bits: fractional bits of @x
+ * @n: power to raise @x to.
+ *
+ * By exploiting the relation between the definition of the natural power
+ * function: x^n := x*x*...*x (x multiplied by itself for n times), and
+ * the binary encoding of numbers used by computers: n := \Sum n_i * 2^i,
+ * (where: n_i \elem {0, 1}, the binary vector representing n),
+ * we find: x^n := x^(\Sum n_i * 2^i) := \Prod x^(n_i * 2^i), which is
+ * of course trivially computable in O(log_2 n), the length of our binary
+ * vector.
+ */
+static unsigned long
+fixed_power_int(unsigned long x, unsigned int frac_bits, unsigned int n)
+{
+ unsigned long result = 1UL << frac_bits;
+
+ if (n) for (;;) {
+ if (n & 1) {
+ result *= x;
+ result += 1UL << (frac_bits - 1);
+ result >>= frac_bits;
+ }
+ n >>= 1;
+ if (!n)
+ break;
+ x *= x;
+ x += 1UL << (frac_bits - 1);
+ x >>= frac_bits;
+ }
+
+ return result;
+}
+
+/*
+ * a1 = a0 * e + a * (1 - e)
+ *
+ * a2 = a1 * e + a * (1 - e)
+ * = (a0 * e + a * (1 - e)) * e + a * (1 - e)
+ * = a0 * e^2 + a * (1 - e) * (1 + e)
+ *
+ * a3 = a2 * e + a * (1 - e)
+ * = (a0 * e^2 + a * (1 - e) * (1 + e)) * e + a * (1 - e)
+ * = a0 * e^3 + a * (1 - e) * (1 + e + e^2)
+ *
+ * ...
+ *
+ * an = a0 * e^n + a * (1 - e) * (1 + e + ... + e^n-1) [1]
+ * = a0 * e^n + a * (1 - e) * (1 - e^n)/(1 - e)
+ * = a0 * e^n + a * (1 - e^n)
+ *
+ * [1] application of the geometric series:
+ *
+ * n 1 - x^(n+1)
+ * S_n := \Sum x^i = -------------
+ * i=0 1 - x
+ */
+static unsigned long
+calc_load_n(unsigned long load, unsigned long exp,
+ unsigned long active, unsigned int n)
+{
+
+ return calc_load(load, fixed_power_int(exp, FSHIFT, n), active);
+}
+
+/*
+ * NO_HZ can leave us missing all per-cpu ticks calling
+ * calc_load_account_active(), but since an idle CPU folds its delta into
+ * calc_load_tasks_idle per calc_load_account_idle(), all we need to do is fold
+ * in the pending idle delta if our idle period crossed a load cycle boundary.
+ *
+ * Once we've updated the global active value, we need to apply the exponential
+ * weights adjusted to the number of cycles missed.
+ */
+static void calc_global_nohz(unsigned long ticks)
+{
+ long delta, active, n;
+
+ if (time_before(jiffies, calc_load_update))
+ return;
+
+ /*
+ * If we crossed a calc_load_update boundary, make sure to fold
+ * any pending idle changes, the respective CPUs might have
+ * missed the tick driven calc_load_account_active() update
+ * due to NO_HZ.
+ */
+ delta = calc_load_fold_idle();
+ if (delta)
+ atomic_long_add(delta, &calc_load_tasks);
+
+ /*
+ * If we were idle for multiple load cycles, apply them.
+ */
+ if (ticks >= LOAD_FREQ) {
+ n = ticks / LOAD_FREQ;
+
+ active = atomic_long_read(&calc_load_tasks);
+ active = active > 0 ? active * FIXED_1 : 0;
+
+ avenrun[0] = calc_load_n(avenrun[0], EXP_1, active, n);
+ avenrun[1] = calc_load_n(avenrun[1], EXP_5, active, n);
+ avenrun[2] = calc_load_n(avenrun[2], EXP_15, active, n);
+
+ calc_load_update += n * LOAD_FREQ;
+ }
+
+ /*
+ * Its possible the remainder of the above division also crosses
+ * a LOAD_FREQ period, the regular check in calc_global_load()
+ * which comes after this will take care of that.
+ *
+ * Consider us being 11 ticks before a cycle completion, and us
+ * sleeping for 4*LOAD_FREQ + 22 ticks, then the above code will
+ * age us 4 cycles, and the test in calc_global_load() will
+ * pick up the final one.
+ */
+}
#else
static void calc_load_account_idle(struct rq *this_rq)
{
@@ -3157,6 +3288,10 @@ static inline long calc_load_fold_idle(void)
{
return 0;
}
+
+static void calc_global_nohz(unsigned long ticks)
+{
+}
#endif
/**
@@ -3174,24 +3309,17 @@ void get_avenrun(unsigned long *loads, unsigned long offset, int shift)
loads[2] = (avenrun[2] + offset) << shift;
}
-static unsigned long
-calc_load(unsigned long load, unsigned long exp, unsigned long active)
-{
- load *= exp;
- load += active * (FIXED_1 - exp);
- return load >> FSHIFT;
-}
-
/*
* calc_load - update the avenrun load estimates 10 ticks after the
* CPUs have updated calc_load_tasks.
*/
-void calc_global_load(void)
+void calc_global_load(unsigned long ticks)
{
- unsigned long upd = calc_load_update + 10;
long active;
- if (time_before(jiffies, upd))
+ calc_global_nohz(ticks);
+
+ if (time_before(jiffies, calc_load_update + 10))
return;
active = atomic_long_read(&calc_load_tasks);
diff --git a/kernel/timer.c b/kernel/timer.c
index 68a9ae7..7bd715f 100644
--- a/kernel/timer.c
+++ b/kernel/timer.c
@@ -1319,7 +1319,7 @@ void do_timer(unsigned long ticks)
{
jiffies_64 += ticks;
update_wall_time();
- calc_global_load();
+ calc_global_load(ticks);
}
#ifdef __ARCH_WANT_SYS_ALARM

View File

@ -0,0 +1,57 @@
From kernel-bounces@lists.fedoraproject.org Thu Sep 9 09:23:49 2010
From: Stanislaw Gruszka <sgruszka@redhat.com>
Subject: [PATCH 1/3] sdhci: 8-bit data transfer width support
Date: Thu, 9 Sep 2010 15:24:10 +0200
From: Kyungmin Park <kyungmin.park@samsung.com>
Upstream ae6d6c92212e94b12ab9365c23fb73acc2c3c2e7 commit.
Some host controllers such as s5pc110 support the WIDE8 feature.
Signed-off-by: Kyungmin Park <kyungmin.park@samsung.com>
Cc: Grant Likely <grant.likely@secretlab.ca>
Cc: <linux-mmc@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
drivers/mmc/host/sdhci.c | 5 +++++
drivers/mmc/host/sdhci.h | 1 +
2 files changed, 6 insertions(+), 0 deletions(-)
diff --git a/drivers/mmc/host/sdhci.c b/drivers/mmc/host/sdhci.c
index c6d1bd8..955cad9 100644
--- a/drivers/mmc/host/sdhci.c
+++ b/drivers/mmc/host/sdhci.c
@@ -1159,6 +1159,11 @@ static void sdhci_set_ios(struct mmc_host *mmc, struct mmc_ios *ios)
ctrl = sdhci_readb(host, SDHCI_HOST_CONTROL);
+ if (ios->bus_width == MMC_BUS_WIDTH_8)
+ ctrl |= SDHCI_CTRL_8BITBUS;
+ else
+ ctrl &= ~SDHCI_CTRL_8BITBUS;
+
if (ios->bus_width == MMC_BUS_WIDTH_4)
ctrl |= SDHCI_CTRL_4BITBUS;
else
diff --git a/drivers/mmc/host/sdhci.h b/drivers/mmc/host/sdhci.h
index c846813..eb5efe0 100644
--- a/drivers/mmc/host/sdhci.h
+++ b/drivers/mmc/host/sdhci.h
@@ -72,6 +72,7 @@
#define SDHCI_CTRL_ADMA1 0x08
#define SDHCI_CTRL_ADMA32 0x10
#define SDHCI_CTRL_ADMA64 0x18
+#define SDHCI_CTRL_8BITBUS 0x20
#define SDHCI_POWER_CONTROL 0x29
#define SDHCI_POWER_ON 0x01
--
1.7.1
_______________________________________________
kernel mailing list
kernel@lists.fedoraproject.org
https://admin.fedoraproject.org/mailman/listinfo/kernel

View File

@ -1 +1,2 @@
091abeb4684ce03d1d936851618687b6 linux-2.6.35.tar.bz2
0741b3219a009fdfc41c766416007fcf patch-2.6.35.10.bz2

View File

@ -1,155 +0,0 @@
From 4d9d1ff88f920e9fcdde155c0a1366b7e0462d14 Mon Sep 17 00:00:00 2001
From: John W. Linville <linville@tuxdriver.com>
Date: Fri, 19 Mar 2010 14:58:01 -0400
Subject: [PATCH v4] ssb: do not read SPROM if it does not exist
Attempting to read registers that don't exist on the SSB bus can cause
hangs on some boxes. At least some b43 devices are 'in the wild' that
don't have SPROMs at all. When the SSB bus support loads, it attempts
to read these (non-existant) SPROMs and causes hard hangs on the box --
no console output, etc.
This patch adds some intelligence to determine whether or not the SPROM
is present before attempting to read it. This avoids those hard hangs
on those devices with no SPROM attached to their SSB bus. The
SSB-attached devices (e.g. b43, et al.) won't work, but at least the box
will survive to test further patches. :-)
Signed-off-by: John W. Linville <linville@tuxdriver.com>
Cc: Larry Finger <Larry.Finger@lwfinger.net>
Cc: Michael Buesch <mb@bu3sch.de>
Cc: stable@kernel.org
---
drivers/ssb/driver_chipcommon.c | 3 +++
drivers/ssb/pci.c | 3 +++
drivers/ssb/sprom.c | 26 ++++++++++++++++++++++++++
include/linux/ssb/ssb.h | 3 +++
include/linux/ssb/ssb_driver_chipcommon.h | 15 +++++++++++++++
5 files changed, 50 insertions(+), 0 deletions(-)
diff --git a/drivers/ssb/driver_chipcommon.c b/drivers/ssb/driver_chipcommon.c
index 9681536..6cf288d 100644
--- a/drivers/ssb/driver_chipcommon.c
+++ b/drivers/ssb/driver_chipcommon.c
@@ -233,6 +233,9 @@ void ssb_chipcommon_init(struct ssb_chipcommon *cc)
{
if (!cc->dev)
return; /* We don't have a ChipCommon */
+ if (cc->dev->id.revision >= 11) {
+ cc->status = chipco_read32(cc, SSB_CHIPCO_CHIPSTAT);
+ }
ssb_pmu_init(cc);
chipco_powercontrol_init(cc);
ssb_chipco_set_clockmode(cc, SSB_CLKMODE_FAST);
diff --git a/drivers/ssb/pci.c b/drivers/ssb/pci.c
index a8dbb06..89d7ab1 100644
--- a/drivers/ssb/pci.c
+++ b/drivers/ssb/pci.c
@@ -621,6 +621,9 @@ static int ssb_pci_sprom_get(struct ssb_bus *bus,
int err = -ENOMEM;
u16 *buf;
+ if (!ssb_is_sprom_available(bus))
+ return -ENODEV;
+
buf = kcalloc(SSB_SPROMSIZE_WORDS_R123, sizeof(u16), GFP_KERNEL);
if (!buf)
goto out;
diff --git a/drivers/ssb/sprom.c b/drivers/ssb/sprom.c
index f2f920f..c690f58 100644
--- a/drivers/ssb/sprom.c
+++ b/drivers/ssb/sprom.c
@@ -176,3 +176,29 @@ const struct ssb_sprom *ssb_get_fallback_sprom(void)
{
return fallback_sprom;
}
+
+bool ssb_is_sprom_available(struct ssb_bus *bus)
+{
+ /* some older devices don't have chipcommon, but they have sprom */
+ if (!bus->chipco.dev)
+ return true;
+
+ /* status register only exists on chipcomon rev >= 11 */
+ if (bus->chipco.dev->id.revision < 11)
+ return true;
+
+ switch (bus->chip_id) {
+ case 0x4312:
+ return SSB_CHIPCO_CHST_4312_SPROM_PRESENT(bus->chipco.status);
+ case 0x4322:
+ return SSB_CHIPCO_CHST_4322_SPROM_PRESENT(bus->chipco.status);
+ case 0x4325:
+ return SSB_CHIPCO_CHST_4325_SPROM_PRESENT(bus->chipco.status);
+ default:
+ break;
+ }
+ if (bus->chipco.dev->id.revision >= 31)
+ return bus->chipco.capabilities & SSB_CHIPCO_CAP_SPROM;
+
+ return true;
+}
diff --git a/include/linux/ssb/ssb.h b/include/linux/ssb/ssb.h
index 24f9885..3b4da23 100644
--- a/include/linux/ssb/ssb.h
+++ b/include/linux/ssb/ssb.h
@@ -394,6 +394,9 @@ extern int ssb_bus_sdiobus_register(struct ssb_bus *bus,
extern void ssb_bus_unregister(struct ssb_bus *bus);
+/* Does the device have an SPROM? */
+extern bool ssb_is_sprom_available(struct ssb_bus *bus);
+
/* Set a fallback SPROM.
* See kdoc at the function definition for complete documentation. */
extern int ssb_arch_set_fallback_sprom(const struct ssb_sprom *sprom);
diff --git a/include/linux/ssb/ssb_driver_chipcommon.h b/include/linux/ssb/ssb_driver_chipcommon.h
index 4e27acf..2cdf249 100644
--- a/include/linux/ssb/ssb_driver_chipcommon.h
+++ b/include/linux/ssb/ssb_driver_chipcommon.h
@@ -53,6 +53,7 @@
#define SSB_CHIPCO_CAP_64BIT 0x08000000 /* 64-bit Backplane */
#define SSB_CHIPCO_CAP_PMU 0x10000000 /* PMU available (rev >= 20) */
#define SSB_CHIPCO_CAP_ECI 0x20000000 /* ECI available (rev >= 20) */
+#define SSB_CHIPCO_CAP_SPROM 0x40000000 /* SPROM present */
#define SSB_CHIPCO_CORECTL 0x0008
#define SSB_CHIPCO_CORECTL_UARTCLK0 0x00000001 /* Drive UART with internal clock */
#define SSB_CHIPCO_CORECTL_SE 0x00000002 /* sync clk out enable (corerev >= 3) */
@@ -385,6 +386,7 @@
/** Chip specific Chip-Status register contents. */
+#define SSB_CHIPCO_CHST_4322_SPROM_EXISTS 0x00000040 /* SPROM present */
#define SSB_CHIPCO_CHST_4325_SPROM_OTP_SEL 0x00000003
#define SSB_CHIPCO_CHST_4325_DEFCIS_SEL 0 /* OTP is powered up, use def. CIS, no SPROM */
#define SSB_CHIPCO_CHST_4325_SPROM_SEL 1 /* OTP is powered up, SPROM is present */
@@ -398,6 +400,18 @@
#define SSB_CHIPCO_CHST_4325_RCAL_VALUE_SHIFT 4
#define SSB_CHIPCO_CHST_4325_PMUTOP_2B 0x00000200 /* 1 for 2b, 0 for to 2a */
+/** Macros to determine SPROM presence based on Chip-Status register. */
+#define SSB_CHIPCO_CHST_4312_SPROM_PRESENT(status) \
+ ((status & SSB_CHIPCO_CHST_4325_SPROM_OTP_SEL) != \
+ SSB_CHIPCO_CHST_4325_OTP_SEL)
+#define SSB_CHIPCO_CHST_4322_SPROM_PRESENT(status) \
+ (status & SSB_CHIPCO_CHST_4322_SPROM_EXISTS)
+#define SSB_CHIPCO_CHST_4325_SPROM_PRESENT(status) \
+ (((status & SSB_CHIPCO_CHST_4325_SPROM_OTP_SEL) != \
+ SSB_CHIPCO_CHST_4325_DEFCIS_SEL) && \
+ ((status & SSB_CHIPCO_CHST_4325_SPROM_OTP_SEL) != \
+ SSB_CHIPCO_CHST_4325_OTP_SEL))
+
/** Clockcontrol masks and values **/
@@ -564,6 +578,7 @@ struct ssb_chipcommon_pmu {
struct ssb_chipcommon {
struct ssb_device *dev;
u32 capabilities;
+ u32 status;
/* Fast Powerup Delay constant */
u16 fast_pwrup_delay;
struct ssb_chipcommon_pmu pmu;
--
1.7.0.1

View File

@ -1,21 +1,46 @@
diff -up linux-2.6.34.noarch/drivers/platform/x86/thinkpad_acpi.c.orig linux-2.6.34.noarch/drivers/platform/x86/thinkpad_acpi.c
--- linux-2.6.34.noarch/drivers/platform/x86/thinkpad_acpi.c.orig 2010-05-17 16:28:13.254200070 -0400
+++ linux-2.6.34.noarch/drivers/platform/x86/thinkpad_acpi.c 2010-05-17 16:29:56.471200083 -0400
@@ -3397,7 +3397,7 @@ static int __init hotkey_init(struct ibm
/* update bright_acpimode... */
tpacpi_check_std_acpi_brightness_support();
From baa7ac7527ee3990007fb28b59ec1d73b922ddaa Mon Sep 17 00:00:00 2001
From: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Date: Mon, 9 Aug 2010 23:48:19 -0300
Subject: [PATCH] thinkpad-acpi: untangle ACPI/vendor backlight selection
acpi_video_backlight_support() already tells us if ACPI is handling
backlight control through the generic ACPI handle. It is better to just
trust it.
While at it, adjust down a printk priority, and test earlier for
brightness_enable=0.
Signed-off-by: Henrique de Moraes Holschuh <hmh@hmh.eng.br>
Signed-off-by: Matthew Garrett <mjg@redhat.com>
---
Documentation/laptops/thinkpad-acpi.txt | 4 +++
drivers/platform/x86/thinkpad_acpi.c | 42 ++++++++++++++----------------
2 files changed, 24 insertions(+), 22 deletions(-)
diff --git a/Documentation/laptops/thinkpad-acpi.txt b/Documentation/laptops/thinkpad-acpi.txt
index fc15538..720ba04 100644
--- a/Documentation/laptops/thinkpad-acpi.txt
+++ b/Documentation/laptops/thinkpad-acpi.txt
@@ -1073,6 +1073,10 @@ ThinkPad-specific interface. The driver will disable its native
backlight brightness control interface if it detects that the standard
ACPI interface is available in the ThinkPad.
- if (tp_features.bright_acpimode && acpi_video_backlight_support()) {
+ if (acpi_video_backlight_support()) {
printk(TPACPI_INFO
"This ThinkPad has standard ACPI backlight "
"brightness control, supported by the ACPI "
@@ -6189,26 +6189,24 @@ static int __init brightness_init(struct
* going to publish a backlight interface
*/
b = tpacpi_check_std_acpi_brightness_support();
- if (b > 0) {
+If you want to use the thinkpad-acpi backlight brightness control
+instead of the generic ACPI video backlight brightness control for some
+reason, you should use the acpi_backlight=vendor kernel parameter.
+
The brightness_enable module parameter can be used to control whether
the LCD brightness control feature will be enabled when available.
brightness_enable=0 forces it to be disabled. brightness_enable=1
diff --git a/drivers/platform/x86/thinkpad_acpi.c b/drivers/platform/x86/thinkpad_acpi.c
index 4bdb137..b383cc6 100644
--- a/drivers/platform/x86/thinkpad_acpi.c
+++ b/drivers/platform/x86/thinkpad_acpi.c
@@ -6313,28 +6313,6 @@ static int __init brightness_init(struct ibm_init_struct *iibm)
if (tp_features.bright_unkfw)
return 1;
- if (tp_features.bright_acpimode) {
- if (acpi_video_backlight_support()) {
- if (brightness_enable > 1) {
- printk(TPACPI_NOTICE
@ -34,23 +59,39 @@ diff -up linux-2.6.34.noarch/drivers/platform/x86/thinkpad_acpi.c.orig linux-2.6
- "available, thinkpad_acpi native "
- "brightness control enabled\n");
- }
- }
- }
-
if (!brightness_enable) {
dbg_printk(TPACPI_DBG_INIT | TPACPI_DBG_BRGHT,
"brightness support disabled by "
@@ -6342,6 +6320,26 @@ static int __init brightness_init(struct ibm_init_struct *iibm)
return 1;
}
+ if (acpi_video_backlight_support()) {
+ if (brightness_enable > 1) {
+ printk(TPACPI_NOTICE
+ printk(TPACPI_INFO
+ "Standard ACPI backlight interface "
+ "available, not loading native one.\n");
+ return 1;
+ } else if (brightness_enable == 1) {
+ printk(TPACPI_NOTICE
+ "Backlight control force enabled, even if standard "
+ "ACPI backlight interface is available\n");
+ printk(TPACPI_WARN
+ "Cannot enable backlight brightness support, "
+ "ACPI is already handling it. Refer to the "
+ "acpi_backlight kernel parameter\n");
+ return 1;
+ }
+ } else {
+ if (brightness_enable > 1) {
+ printk(TPACPI_NOTICE
+ "Standard ACPI backlight interface not "
+ "available, thinkpad_acpi native "
+ "brightness control enabled\n");
}
}
+ } else if (tp_features.bright_acpimode && brightness_enable > 1) {
+ printk(TPACPI_NOTICE
+ "Standard ACPI backlight interface not "
+ "available, thinkpad_acpi native "
+ "brightness control enabled\n");
+ }
+
/*
* Check for module parameter bogosity, note that we
* init brightness_mode to TPACPI_BRGHT_MODE_MAX in order to be
--
1.7.3.2

View File

@ -0,0 +1,65 @@
commit 8cf5102c84dba60b2ea29b7e89f1a65100e20bb9
Author: Matthew Garrett <mjg@redhat.com>
Date: Thu Oct 21 17:31:56 2010 -0400
tpm: Autodetect itpm devices
Some Lenovos have TPMs that require a quirk to function correctly. This can
be autodetected by checking whether the device has a _HID of INTC0102. This
is an invalid PNPid, and as such is discarded by the pnp layer - however
it's still present in the ACPI code, so we can pull it out that way. This
means that the quirk won't be automatically applied on non-ACPI systems,
but without ACPI we don't have any way to identify the chip anyway so I
don't think that's a great concern.
Signed-off-by: Matthew Garrett <mjg@redhat.com>
diff --git a/drivers/char/tpm/tpm_tis.c b/drivers/char/tpm/tpm_tis.c
index 1030f84..c17a305 100644
--- a/drivers/char/tpm/tpm_tis.c
+++ b/drivers/char/tpm/tpm_tis.c
@@ -25,6 +25,7 @@
#include <linux/slab.h>
#include <linux/interrupt.h>
#include <linux/wait.h>
+#include <linux/acpi.h>
#include "tpm.h"
#define TPM_HEADER_SIZE 10
@@ -78,6 +79,26 @@ enum tis_defaults {
static LIST_HEAD(tis_chips);
static DEFINE_SPINLOCK(tis_lock);
+#ifdef CONFIG_ACPI
+static int is_itpm(struct pnp_dev *dev)
+{
+ struct acpi_device *acpi = pnp_acpi_device(dev);
+ struct acpi_hardware_id *id;
+
+ list_for_each_entry(id, &acpi->pnp.ids, list) {
+ if (!strcmp("INTC0102", id->id))
+ return 1;
+ }
+
+ return 0;
+}
+#else
+static int is_itpm(struct pnp_dev *dev)
+{
+ return 0;
+}
+#endif
+
static int check_locality(struct tpm_chip *chip, int l)
{
if ((ioread8(chip->vendor.iobase + TPM_ACCESS(l)) &
@@ -472,6 +493,9 @@ static int tpm_tis_init(struct device *dev, resource_size_t start,
"1.2 TPM (device-id 0x%X, rev-id %d)\n",
vendor >> 16, ioread8(chip->vendor.iobase + TPM_RID(0)));
+ if (is_itpm(to_pnp_dev(dev)))
+ itpm = 1;
+
if (itpm)
dev_info(dev, "Intel iTPM workaround enabled\n");

View File

@ -0,0 +1,24 @@
Fix TPM timeouts on boot (#530393)
--- a/drivers/char/tpm/tpm.c
+++ a/drivers/char/tpm/tpm.c
@@ -354,12 +354,14 @@ unsigned long tpm_calc_ordinal_duration(struct tpm_chip *chip,
tpm_protected_ordinal_duration[ordinal &
TPM_PROTECTED_ORDINAL_MASK];
- if (duration_idx != TPM_UNDEFINED)
+ if (duration_idx != TPM_UNDEFINED) {
duration = chip->vendor.duration[duration_idx];
- if (duration <= 0)
+ /* if duration is 0, it's because chip->vendor.duration wasn't */
+ /* filled yet, so we set the lowest timeout just to give enough */
+ /* time to tpm_get_timeouts() succeed */
+ return (duration <= 0 ? HZ : duration);
+ } else
return 2 * 60 * HZ;
- else
- return duration;
}
EXPORT_SYMBOL_GPL(tpm_calc_ordinal_duration);

View File

@ -0,0 +1,982 @@
From f815dfecf23bfd19d4b5e9f1a660b1f5dbe70472 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Thu, 16 Sep 2010 18:21:52 +0100
Subject: [PATCH 2/2] tty: icount changeover for other main devices
Again basically cut and paste
Convert the main driver set to use the hooks for GICOUNT
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
arch/ia64/hp/sim/simserial.c | 12 +------
drivers/char/amiserial.c | 56 ++++++++++++++++-------------
drivers/char/cyclades.c | 49 +++++++++++++------------
drivers/char/ip2/ip2main.c | 72 ++++++++++++++++++++++---------------
drivers/char/mxser.c | 62 ++++++++++++++++++--------------
drivers/char/nozomi.c | 37 +++++++++----------
drivers/char/pcmcia/synclink_cs.c | 60 ++++++++++++++-----------------
drivers/char/synclink.c | 73 +++++++++++++++++--------------------
drivers/char/synclink_gt.c | 56 +++++++++++++++--------------
drivers/char/synclinkmp.c | 61 ++++++++++++++-----------------
drivers/serial/68360serial.c | 51 +++++++++++++-------------
net/bluetooth/rfcomm/tty.c | 4 --
12 files changed, 297 insertions(+), 296 deletions(-)
diff --git a/arch/ia64/hp/sim/simserial.c b/arch/ia64/hp/sim/simserial.c
index 2bef526..204f650 100644
--- a/arch/ia64/hp/sim/simserial.c
+++ b/arch/ia64/hp/sim/simserial.c
@@ -395,7 +395,7 @@ static int rs_ioctl(struct tty_struct *tty, struct file * file,
{
if ((cmd != TIOCGSERIAL) && (cmd != TIOCSSERIAL) &&
(cmd != TIOCSERCONFIG) && (cmd != TIOCSERGSTRUCT) &&
- (cmd != TIOCMIWAIT) && (cmd != TIOCGICOUNT)) {
+ (cmd != TIOCMIWAIT)) {
if (tty->flags & (1 << TTY_IO_ERROR))
return -EIO;
}
@@ -433,16 +433,6 @@ static int rs_ioctl(struct tty_struct *tty, struct file * file,
case TIOCMIWAIT:
printk(KERN_INFO "rs_ioctl: TIOCMIWAIT: called\n");
return 0;
- /*
- * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
- * Return: write counters to the user passed counter struct
- * NB: both 1->0 and 0->1 transitions are counted except for
- * RI where only 0->1 is counted.
- */
- case TIOCGICOUNT:
- printk(KERN_INFO "rs_ioctl: TIOCGICOUNT called\n");
- return 0;
-
case TIOCSERGWILD:
case TIOCSERSWILD:
/* "setserial -W" is called in Debian boot */
diff --git a/drivers/char/amiserial.c b/drivers/char/amiserial.c
index 4f8d60c..2918d5d 100644
--- a/drivers/char/amiserial.c
+++ b/drivers/char/amiserial.c
@@ -1263,6 +1263,36 @@ static int rs_break(struct tty_struct *tty, int break_state)
return 0;
}
+/*
+ * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
+ * Return: write counters to the user passed counter struct
+ * NB: both 1->0 and 0->1 transitions are counted except for
+ * RI where only 0->1 is counted.
+ */
+static int rs_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
+{
+ struct async_struct *info = tty->driver_data;
+ struct async_icount cnow;
+ unsigned long flags;
+
+ local_irq_save(flags);
+ cnow = info->state->icount;
+ local_irq_restore(flags);
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+ icount->rx = cnow.rx;
+ icount->tx = cnow.tx;
+ icount->frame = cnow.frame;
+ icount->overrun = cnow.overrun;
+ icount->parity = cnow.parity;
+ icount->brk = cnow.brk;
+ icount->buf_overrun = cnow.buf_overrun;
+
+ return 0;
+}
static int rs_ioctl(struct tty_struct *tty, struct file * file,
unsigned int cmd, unsigned long arg)
@@ -1332,31 +1362,6 @@ static int rs_ioctl(struct tty_struct *tty, struct file * file,
}
/* NOTREACHED */
- /*
- * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
- * Return: write counters to the user passed counter struct
- * NB: both 1->0 and 0->1 transitions are counted except for
- * RI where only 0->1 is counted.
- */
- case TIOCGICOUNT:
- local_irq_save(flags);
- cnow = info->state->icount;
- local_irq_restore(flags);
- icount.cts = cnow.cts;
- icount.dsr = cnow.dsr;
- icount.rng = cnow.rng;
- icount.dcd = cnow.dcd;
- icount.rx = cnow.rx;
- icount.tx = cnow.tx;
- icount.frame = cnow.frame;
- icount.overrun = cnow.overrun;
- icount.parity = cnow.parity;
- icount.brk = cnow.brk;
- icount.buf_overrun = cnow.buf_overrun;
-
- if (copy_to_user(argp, &icount, sizeof(icount)))
- return -EFAULT;
- return 0;
case TIOCSERGWILD:
case TIOCSERSWILD:
/* "setserial -W" is called in Debian boot */
@@ -1949,6 +1954,7 @@ static const struct tty_operations serial_ops = {
.wait_until_sent = rs_wait_until_sent,
.tiocmget = rs_tiocmget,
.tiocmset = rs_tiocmset,
+ .get_icount = rs_get_icount,
.proc_fops = &rs_proc_fops,
};
diff --git a/drivers/char/cyclades.c b/drivers/char/cyclades.c
index 9824b41..2364df8 100644
--- a/drivers/char/cyclades.c
+++ b/drivers/char/cyclades.c
@@ -2791,29 +2791,6 @@ cy_ioctl(struct tty_struct *tty, struct file *file,
* NB: both 1->0 and 0->1 transitions are counted except for
* RI where only 0->1 is counted.
*/
- case TIOCGICOUNT: {
- struct serial_icounter_struct sic = { };
-
- spin_lock_irqsave(&info->card->card_lock, flags);
- cnow = info->icount;
- spin_unlock_irqrestore(&info->card->card_lock, flags);
-
- sic.cts = cnow.cts;
- sic.dsr = cnow.dsr;
- sic.rng = cnow.rng;
- sic.dcd = cnow.dcd;
- sic.rx = cnow.rx;
- sic.tx = cnow.tx;
- sic.frame = cnow.frame;
- sic.overrun = cnow.overrun;
- sic.parity = cnow.parity;
- sic.brk = cnow.brk;
- sic.buf_overrun = cnow.buf_overrun;
-
- if (copy_to_user(argp, &sic, sizeof(sic)))
- ret_val = -EFAULT;
- break;
- }
default:
ret_val = -ENOIOCTLCMD;
}
@@ -2825,6 +2802,31 @@ cy_ioctl(struct tty_struct *tty, struct file *file,
return ret_val;
} /* cy_ioctl */
+static int cy_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *sic)
+{
+ struct cyclades_port *info = tty->driver_data;
+ struct cyclades_icount cnow; /* Used to snapshot */
+ unsigned long flags;
+
+ spin_lock_irqsave(&info->card->card_lock, flags);
+ cnow = info->icount;
+ spin_unlock_irqrestore(&info->card->card_lock, flags);
+
+ sic->cts = cnow.cts;
+ sic->dsr = cnow.dsr;
+ sic->rng = cnow.rng;
+ sic->dcd = cnow.dcd;
+ sic->rx = cnow.rx;
+ sic->tx = cnow.tx;
+ sic->frame = cnow.frame;
+ sic->overrun = cnow.overrun;
+ sic->parity = cnow.parity;
+ sic->brk = cnow.brk;
+ sic->buf_overrun = cnow.buf_overrun;
+ return 0;
+}
+
/*
* This routine allows the tty driver to be notified when
* device's termios settings have changed. Note that a
@@ -4086,6 +4088,7 @@ static const struct tty_operations cy_ops = {
.wait_until_sent = cy_wait_until_sent,
.tiocmget = cy_tiocmget,
.tiocmset = cy_tiocmset,
+ .get_icount = cy_get_icount,
.proc_fops = &cyclades_proc_fops,
};
diff --git a/drivers/char/ip2/ip2main.c b/drivers/char/ip2/ip2main.c
index 911e1da..c0f864c 100644
--- a/drivers/char/ip2/ip2main.c
+++ b/drivers/char/ip2/ip2main.c
@@ -183,6 +183,8 @@ static void ip2_hangup(PTTY);
static int ip2_tiocmget(struct tty_struct *tty, struct file *file);
static int ip2_tiocmset(struct tty_struct *tty, struct file *file,
unsigned int set, unsigned int clear);
+static int ip2_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount);
static void set_irq(int, int);
static void ip2_interrupt_bh(struct work_struct *work);
@@ -454,6 +456,7 @@ static const struct tty_operations ip2_ops = {
.hangup = ip2_hangup,
.tiocmget = ip2_tiocmget,
.tiocmset = ip2_tiocmset,
+ .get_icount = ip2_get_icount,
.proc_fops = &ip2_proc_fops,
};
@@ -2124,7 +2127,6 @@ ip2_ioctl ( PTTY tty, struct file *pFile, UINT cmd, ULONG arg )
i2ChanStrPtr pCh = DevTable[tty->index];
i2eBordStrPtr pB;
struct async_icount cprev, cnow; /* kernel counter temps */
- struct serial_icounter_struct __user *p_cuser;
int rc = 0;
unsigned long flags;
void __user *argp = (void __user *)arg;
@@ -2293,34 +2295,6 @@ ip2_ioctl ( PTTY tty, struct file *pFile, UINT cmd, ULONG arg )
break;
/*
- * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
- * Return: write counters to the user passed counter struct
- * NB: both 1->0 and 0->1 transitions are counted except for RI where
- * only 0->1 is counted. The controller is quite capable of counting
- * both, but this done to preserve compatibility with the standard
- * serial driver.
- */
- case TIOCGICOUNT:
- ip2trace (CHANN, ITRC_IOCTL, 11, 1, rc );
-
- write_lock_irqsave(&pB->read_fifo_spinlock, flags);
- cnow = pCh->icount;
- write_unlock_irqrestore(&pB->read_fifo_spinlock, flags);
- p_cuser = argp;
- rc = put_user(cnow.cts, &p_cuser->cts);
- rc = put_user(cnow.dsr, &p_cuser->dsr);
- rc = put_user(cnow.rng, &p_cuser->rng);
- rc = put_user(cnow.dcd, &p_cuser->dcd);
- rc = put_user(cnow.rx, &p_cuser->rx);
- rc = put_user(cnow.tx, &p_cuser->tx);
- rc = put_user(cnow.frame, &p_cuser->frame);
- rc = put_user(cnow.overrun, &p_cuser->overrun);
- rc = put_user(cnow.parity, &p_cuser->parity);
- rc = put_user(cnow.brk, &p_cuser->brk);
- rc = put_user(cnow.buf_overrun, &p_cuser->buf_overrun);
- break;
-
- /*
* The rest are not supported by this driver. By returning -ENOIOCTLCMD they
* will be passed to the line discipline for it to handle.
*/
@@ -2344,6 +2318,46 @@ ip2_ioctl ( PTTY tty, struct file *pFile, UINT cmd, ULONG arg )
return rc;
}
+static int ip2_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
+{
+ i2ChanStrPtr pCh = DevTable[tty->index];
+ i2eBordStrPtr pB;
+ struct async_icount cnow; /* kernel counter temp */
+ unsigned long flags;
+
+ if ( pCh == NULL )
+ return -ENODEV;
+
+ pB = pCh->pMyBord;
+
+ /*
+ * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
+ * Return: write counters to the user passed counter struct
+ * NB: both 1->0 and 0->1 transitions are counted except for RI where
+ * only 0->1 is counted. The controller is quite capable of counting
+ * both, but this done to preserve compatibility with the standard
+ * serial driver.
+ */
+
+ write_lock_irqsave(&pB->read_fifo_spinlock, flags);
+ cnow = pCh->icount;
+ write_unlock_irqrestore(&pB->read_fifo_spinlock, flags);
+
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+ icount->rx = cnow.rx;
+ icount->tx = cnow.tx;
+ icount->frame = cnow.frame;
+ icount->overrun = cnow.overrun;
+ icount->parity = cnow.parity;
+ icount->brk = cnow.brk;
+ icount->buf_overrun = cnow.buf_overrun;
+ return 0;
+}
+
/******************************************************************************/
/* Function: GetSerialInfo() */
/* Parameters: Pointer to channel structure */
diff --git a/drivers/char/mxser.c b/drivers/char/mxser.c
index d2692d4..65aeae8 100644
--- a/drivers/char/mxser.c
+++ b/drivers/char/mxser.c
@@ -1700,7 +1700,7 @@ static int mxser_ioctl(struct tty_struct *tty, struct file *file,
return 0;
}
- if (cmd != TIOCGSERIAL && cmd != TIOCMIWAIT && cmd != TIOCGICOUNT &&
+ if (cmd != TIOCGSERIAL && cmd != TIOCMIWAIT &&
test_bit(TTY_IO_ERROR, &tty->flags))
return -EIO;
@@ -1730,32 +1730,6 @@ static int mxser_ioctl(struct tty_struct *tty, struct file *file,
return wait_event_interruptible(info->port.delta_msr_wait,
mxser_cflags_changed(info, arg, &cnow));
- /*
- * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
- * Return: write counters to the user passed counter struct
- * NB: both 1->0 and 0->1 transitions are counted except for
- * RI where only 0->1 is counted.
- */
- case TIOCGICOUNT: {
- struct serial_icounter_struct icnt = { 0 };
- spin_lock_irqsave(&info->slock, flags);
- cnow = info->icount;
- spin_unlock_irqrestore(&info->slock, flags);
-
- icnt.frame = cnow.frame;
- icnt.brk = cnow.brk;
- icnt.overrun = cnow.overrun;
- icnt.buf_overrun = cnow.buf_overrun;
- icnt.parity = cnow.parity;
- icnt.rx = cnow.rx;
- icnt.tx = cnow.tx;
- icnt.cts = cnow.cts;
- icnt.dsr = cnow.dsr;
- icnt.rng = cnow.rng;
- icnt.dcd = cnow.dcd;
-
- return copy_to_user(argp, &icnt, sizeof(icnt)) ? -EFAULT : 0;
- }
case MOXA_HighSpeedOn:
return put_user(info->baud_base != 115200 ? 1 : 0, (int __user *)argp);
case MOXA_SDS_RSTICOUNTER:
@@ -1828,6 +1802,39 @@ static int mxser_ioctl(struct tty_struct *tty, struct file *file,
return 0;
}
+ /*
+ * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
+ * Return: write counters to the user passed counter struct
+ * NB: both 1->0 and 0->1 transitions are counted except for
+ * RI where only 0->1 is counted.
+ */
+
+static int mxser_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
+
+{
+ struct mxser_port *info = tty->driver_data;
+ struct async_icount cnow;
+ unsigned long flags;
+
+ spin_lock_irqsave(&info->slock, flags);
+ cnow = info->icount;
+ spin_unlock_irqrestore(&info->slock, flags);
+
+ icount->frame = cnow.frame;
+ icount->brk = cnow.brk;
+ icount->overrun = cnow.overrun;
+ icount->buf_overrun = cnow.buf_overrun;
+ icount->parity = cnow.parity;
+ icount->rx = cnow.rx;
+ icount->tx = cnow.tx;
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+ return 0;
+}
+
static void mxser_stoprx(struct tty_struct *tty)
{
struct mxser_port *info = tty->driver_data;
@@ -2326,6 +2333,7 @@ static const struct tty_operations mxser_ops = {
.wait_until_sent = mxser_wait_until_sent,
.tiocmget = mxser_tiocmget,
.tiocmset = mxser_tiocmset,
+ .get_icount = mxser_get_icount,
};
struct tty_port_operations mxser_port_ops = {
diff --git a/drivers/char/nozomi.c b/drivers/char/nozomi.c
index 18af923..0924435 100644
--- a/drivers/char/nozomi.c
+++ b/drivers/char/nozomi.c
@@ -1805,24 +1805,24 @@ static int ntty_cflags_changed(struct port *port, unsigned long flags,
return ret;
}
-static int ntty_ioctl_tiocgicount(struct port *port, void __user *argp)
+static int ntty_tiocgicount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
{
+ struct port *port = tty->driver_data;
const struct async_icount cnow = port->tty_icount;
- struct serial_icounter_struct icount;
-
- icount.cts = cnow.cts;
- icount.dsr = cnow.dsr;
- icount.rng = cnow.rng;
- icount.dcd = cnow.dcd;
- icount.rx = cnow.rx;
- icount.tx = cnow.tx;
- icount.frame = cnow.frame;
- icount.overrun = cnow.overrun;
- icount.parity = cnow.parity;
- icount.brk = cnow.brk;
- icount.buf_overrun = cnow.buf_overrun;
-
- return copy_to_user(argp, &icount, sizeof(icount)) ? -EFAULT : 0;
+
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+ icount->rx = cnow.rx;
+ icount->tx = cnow.tx;
+ icount->frame = cnow.frame;
+ icount->overrun = cnow.overrun;
+ icount->parity = cnow.parity;
+ icount->brk = cnow.brk;
+ icount->buf_overrun = cnow.buf_overrun;
+ return 0;
}
static int ntty_ioctl(struct tty_struct *tty, struct file *file,
@@ -1841,9 +1841,7 @@ static int ntty_ioctl(struct tty_struct *tty, struct file *file,
rval = wait_event_interruptible(port->tty_wait,
ntty_cflags_changed(port, arg, &cprev));
break;
- } case TIOCGICOUNT:
- rval = ntty_ioctl_tiocgicount(port, argp);
- break;
+ }
default:
DBG1("ERR: 0x%08X, %d", cmd, cmd);
break;
@@ -1923,6 +1921,7 @@ static const struct tty_operations tty_ops = {
.chars_in_buffer = ntty_chars_in_buffer,
.tiocmget = ntty_tiocmget,
.tiocmset = ntty_tiocmset,
+ .get_icount = ntty_tiocgicount,
.install = ntty_install,
.cleanup = ntty_cleanup,
};
diff --git a/drivers/char/pcmcia/synclink_cs.c b/drivers/char/pcmcia/synclink_cs.c
index 824d67c..1dc493f 100644
--- a/drivers/char/pcmcia/synclink_cs.c
+++ b/drivers/char/pcmcia/synclink_cs.c
@@ -2220,6 +2220,32 @@ static int mgslpc_break(struct tty_struct *tty, int break_state)
return 0;
}
+static int mgslpc_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
+{
+ MGSLPC_INFO * info = (MGSLPC_INFO *)tty->driver_data;
+ struct mgsl_icount cnow; /* kernel counter temps */
+ unsigned long flags;
+
+ spin_lock_irqsave(&info->lock,flags);
+ cnow = info->icount;
+ spin_unlock_irqrestore(&info->lock,flags);
+
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+ icount->rx = cnow.rx;
+ icount->tx = cnow.tx;
+ icount->frame = cnow.frame;
+ icount->overrun = cnow.overrun;
+ icount->parity = cnow.parity;
+ icount->brk = cnow.brk;
+ icount->buf_overrun = cnow.buf_overrun;
+
+ return 0;
+}
+
/* Service an IOCTL request
*
* Arguments:
@@ -2235,11 +2261,7 @@ static int mgslpc_ioctl(struct tty_struct *tty, struct file * file,
unsigned int cmd, unsigned long arg)
{
MGSLPC_INFO * info = (MGSLPC_INFO *)tty->driver_data;
- int error;
- struct mgsl_icount cnow; /* kernel counter temps */
- struct serial_icounter_struct __user *p_cuser; /* user space */
void __user *argp = (void __user *)arg;
- unsigned long flags;
if (debug_level >= DEBUG_LEVEL_INFO)
printk("%s(%d):mgslpc_ioctl %s cmd=%08X\n", __FILE__,__LINE__,
@@ -2249,7 +2271,7 @@ static int mgslpc_ioctl(struct tty_struct *tty, struct file * file,
return -ENODEV;
if ((cmd != TIOCGSERIAL) && (cmd != TIOCSSERIAL) &&
- (cmd != TIOCMIWAIT) && (cmd != TIOCGICOUNT)) {
+ (cmd != TIOCMIWAIT)) {
if (tty->flags & (1 << TTY_IO_ERROR))
return -EIO;
}
@@ -2279,34 +2301,6 @@ static int mgslpc_ioctl(struct tty_struct *tty, struct file * file,
return wait_events(info, argp);
case TIOCMIWAIT:
return modem_input_wait(info,(int)arg);
- case TIOCGICOUNT:
- spin_lock_irqsave(&info->lock,flags);
- cnow = info->icount;
- spin_unlock_irqrestore(&info->lock,flags);
- p_cuser = argp;
- PUT_USER(error,cnow.cts, &p_cuser->cts);
- if (error) return error;
- PUT_USER(error,cnow.dsr, &p_cuser->dsr);
- if (error) return error;
- PUT_USER(error,cnow.rng, &p_cuser->rng);
- if (error) return error;
- PUT_USER(error,cnow.dcd, &p_cuser->dcd);
- if (error) return error;
- PUT_USER(error,cnow.rx, &p_cuser->rx);
- if (error) return error;
- PUT_USER(error,cnow.tx, &p_cuser->tx);
- if (error) return error;
- PUT_USER(error,cnow.frame, &p_cuser->frame);
- if (error) return error;
- PUT_USER(error,cnow.overrun, &p_cuser->overrun);
- if (error) return error;
- PUT_USER(error,cnow.parity, &p_cuser->parity);
- if (error) return error;
- PUT_USER(error,cnow.brk, &p_cuser->brk);
- if (error) return error;
- PUT_USER(error,cnow.buf_overrun, &p_cuser->buf_overrun);
- if (error) return error;
- return 0;
default:
return -ENOIOCTLCMD;
}
diff --git a/drivers/char/synclink.c b/drivers/char/synclink.c
index 0658fc5..9970aca 100644
--- a/drivers/char/synclink.c
+++ b/drivers/char/synclink.c
@@ -2920,6 +2920,38 @@ static int mgsl_break(struct tty_struct *tty, int break_state)
} /* end of mgsl_break() */
+/*
+ * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
+ * Return: write counters to the user passed counter struct
+ * NB: both 1->0 and 0->1 transitions are counted except for
+ * RI where only 0->1 is counted.
+ */
+static int msgl_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
+
+{
+ struct mgsl_struct * info = tty->driver_data;
+ struct mgsl_icount cnow; /* kernel counter temps */
+ unsigned long flags;
+
+ spin_lock_irqsave(&info->irq_spinlock,flags);
+ cnow = info->icount;
+ spin_unlock_irqrestore(&info->irq_spinlock,flags);
+
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+ icount->rx = cnow.rx;
+ icount->tx = cnow.tx;
+ icount->frame = cnow.frame;
+ icount->overrun = cnow.overrun;
+ icount->parity = cnow.parity;
+ icount->brk = cnow.brk;
+ icount->buf_overrun = cnow.buf_overrun;
+ return 0;
+}
+
/* mgsl_ioctl() Service an IOCTL request
*
* Arguments:
@@ -2945,7 +2977,7 @@ static int mgsl_ioctl(struct tty_struct *tty, struct file * file,
return -ENODEV;
if ((cmd != TIOCGSERIAL) && (cmd != TIOCSSERIAL) &&
- (cmd != TIOCMIWAIT) && (cmd != TIOCGICOUNT)) {
+ (cmd != TIOCMIWAIT)) {
if (tty->flags & (1 << TTY_IO_ERROR))
return -EIO;
}
@@ -2958,11 +2990,7 @@ static int mgsl_ioctl(struct tty_struct *tty, struct file * file,
static int mgsl_ioctl_common(struct mgsl_struct *info, unsigned int cmd, unsigned long arg)
{
- int error;
- struct mgsl_icount cnow; /* kernel counter temps */
void __user *argp = (void __user *)arg;
- struct serial_icounter_struct __user *p_cuser; /* user space */
- unsigned long flags;
switch (cmd) {
case MGSL_IOCGPARAMS:
@@ -2991,40 +3019,6 @@ static int mgsl_ioctl_common(struct mgsl_struct *info, unsigned int cmd, unsigne
case TIOCMIWAIT:
return modem_input_wait(info,(int)arg);
- /*
- * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
- * Return: write counters to the user passed counter struct
- * NB: both 1->0 and 0->1 transitions are counted except for
- * RI where only 0->1 is counted.
- */
- case TIOCGICOUNT:
- spin_lock_irqsave(&info->irq_spinlock,flags);
- cnow = info->icount;
- spin_unlock_irqrestore(&info->irq_spinlock,flags);
- p_cuser = argp;
- PUT_USER(error,cnow.cts, &p_cuser->cts);
- if (error) return error;
- PUT_USER(error,cnow.dsr, &p_cuser->dsr);
- if (error) return error;
- PUT_USER(error,cnow.rng, &p_cuser->rng);
- if (error) return error;
- PUT_USER(error,cnow.dcd, &p_cuser->dcd);
- if (error) return error;
- PUT_USER(error,cnow.rx, &p_cuser->rx);
- if (error) return error;
- PUT_USER(error,cnow.tx, &p_cuser->tx);
- if (error) return error;
- PUT_USER(error,cnow.frame, &p_cuser->frame);
- if (error) return error;
- PUT_USER(error,cnow.overrun, &p_cuser->overrun);
- if (error) return error;
- PUT_USER(error,cnow.parity, &p_cuser->parity);
- if (error) return error;
- PUT_USER(error,cnow.brk, &p_cuser->brk);
- if (error) return error;
- PUT_USER(error,cnow.buf_overrun, &p_cuser->buf_overrun);
- if (error) return error;
- return 0;
default:
return -ENOIOCTLCMD;
}
@@ -4325,6 +4319,7 @@ static const struct tty_operations mgsl_ops = {
.hangup = mgsl_hangup,
.tiocmget = tiocmget,
.tiocmset = tiocmset,
+ .get_icount = msgl_get_icount,
.proc_fops = &mgsl_proc_fops,
};
diff --git a/drivers/char/synclink_gt.c b/drivers/char/synclink_gt.c
index 4561ce2..54fa0ee4 100644
--- a/drivers/char/synclink_gt.c
+++ b/drivers/char/synclink_gt.c
@@ -1025,9 +1025,6 @@ static int ioctl(struct tty_struct *tty, struct file *file,
unsigned int cmd, unsigned long arg)
{
struct slgt_info *info = tty->driver_data;
- struct mgsl_icount cnow; /* kernel counter temps */
- struct serial_icounter_struct __user *p_cuser; /* user space */
- unsigned long flags;
void __user *argp = (void __user *)arg;
int ret;
@@ -1036,7 +1033,7 @@ static int ioctl(struct tty_struct *tty, struct file *file,
DBGINFO(("%s ioctl() cmd=%08X\n", info->device_name, cmd));
if ((cmd != TIOCGSERIAL) && (cmd != TIOCSSERIAL) &&
- (cmd != TIOCMIWAIT) && (cmd != TIOCGICOUNT)) {
+ (cmd != TIOCMIWAIT)) {
if (tty->flags & (1 << TTY_IO_ERROR))
return -EIO;
}
@@ -1089,25 +1086,6 @@ static int ioctl(struct tty_struct *tty, struct file *file,
case MGSL_IOCWAITGPIO:
ret = wait_gpio(info, argp);
break;
- case TIOCGICOUNT:
- spin_lock_irqsave(&info->lock,flags);
- cnow = info->icount;
- spin_unlock_irqrestore(&info->lock,flags);
- p_cuser = argp;
- if (put_user(cnow.cts, &p_cuser->cts) ||
- put_user(cnow.dsr, &p_cuser->dsr) ||
- put_user(cnow.rng, &p_cuser->rng) ||
- put_user(cnow.dcd, &p_cuser->dcd) ||
- put_user(cnow.rx, &p_cuser->rx) ||
- put_user(cnow.tx, &p_cuser->tx) ||
- put_user(cnow.frame, &p_cuser->frame) ||
- put_user(cnow.overrun, &p_cuser->overrun) ||
- put_user(cnow.parity, &p_cuser->parity) ||
- put_user(cnow.brk, &p_cuser->brk) ||
- put_user(cnow.buf_overrun, &p_cuser->buf_overrun))
- ret = -EFAULT;
- ret = 0;
- break;
default:
ret = -ENOIOCTLCMD;
}
@@ -1115,6 +1093,33 @@ static int ioctl(struct tty_struct *tty, struct file *file,
return ret;
}
+static int get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
+
+{
+ struct slgt_info *info = tty->driver_data;
+ struct mgsl_icount cnow; /* kernel counter temps */
+ unsigned long flags;
+
+ spin_lock_irqsave(&info->lock,flags);
+ cnow = info->icount;
+ spin_unlock_irqrestore(&info->lock,flags);
+
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+ icount->rx = cnow.rx;
+ icount->tx = cnow.tx;
+ icount->frame = cnow.frame;
+ icount->overrun = cnow.overrun;
+ icount->parity = cnow.parity;
+ icount->brk = cnow.brk;
+ icount->buf_overrun = cnow.buf_overrun;
+
+ return 0;
+}
+
/*
* support for 32 bit ioctl calls on 64 bit systems
*/
@@ -1204,10 +1209,6 @@ static long slgt_compat_ioctl(struct tty_struct *tty, struct file *file,
case MGSL_IOCSGPIO:
case MGSL_IOCGGPIO:
case MGSL_IOCWAITGPIO:
- case TIOCGICOUNT:
- rc = ioctl(tty, file, cmd, (unsigned long)(compat_ptr(arg)));
- break;
-
case MGSL_IOCSTXIDLE:
case MGSL_IOCTXENABLE:
case MGSL_IOCRXENABLE:
@@ -3638,6 +3639,7 @@ static const struct tty_operations ops = {
.hangup = hangup,
.tiocmget = tiocmget,
.tiocmset = tiocmset,
+ .get_icount = get_icount,
.proc_fops = &synclink_gt_proc_fops,
};
diff --git a/drivers/char/synclinkmp.c b/drivers/char/synclinkmp.c
index 2b18adc..beffc24 100644
--- a/drivers/char/synclinkmp.c
+++ b/drivers/char/synclinkmp.c
@@ -1255,10 +1255,6 @@ static int do_ioctl(struct tty_struct *tty, struct file *file,
unsigned int cmd, unsigned long arg)
{
SLMP_INFO *info = tty->driver_data;
- int error;
- struct mgsl_icount cnow; /* kernel counter temps */
- struct serial_icounter_struct __user *p_cuser; /* user space */
- unsigned long flags;
void __user *argp = (void __user *)arg;
if (debug_level >= DEBUG_LEVEL_INFO)
@@ -1269,7 +1265,7 @@ static int do_ioctl(struct tty_struct *tty, struct file *file,
return -ENODEV;
if ((cmd != TIOCGSERIAL) && (cmd != TIOCSSERIAL) &&
- (cmd != TIOCMIWAIT) && (cmd != TIOCGICOUNT)) {
+ (cmd != TIOCMIWAIT)) {
if (tty->flags & (1 << TTY_IO_ERROR))
return -EIO;
}
@@ -1307,34 +1303,6 @@ static int do_ioctl(struct tty_struct *tty, struct file *file,
* NB: both 1->0 and 0->1 transitions are counted except for
* RI where only 0->1 is counted.
*/
- case TIOCGICOUNT:
- spin_lock_irqsave(&info->lock,flags);
- cnow = info->icount;
- spin_unlock_irqrestore(&info->lock,flags);
- p_cuser = argp;
- PUT_USER(error,cnow.cts, &p_cuser->cts);
- if (error) return error;
- PUT_USER(error,cnow.dsr, &p_cuser->dsr);
- if (error) return error;
- PUT_USER(error,cnow.rng, &p_cuser->rng);
- if (error) return error;
- PUT_USER(error,cnow.dcd, &p_cuser->dcd);
- if (error) return error;
- PUT_USER(error,cnow.rx, &p_cuser->rx);
- if (error) return error;
- PUT_USER(error,cnow.tx, &p_cuser->tx);
- if (error) return error;
- PUT_USER(error,cnow.frame, &p_cuser->frame);
- if (error) return error;
- PUT_USER(error,cnow.overrun, &p_cuser->overrun);
- if (error) return error;
- PUT_USER(error,cnow.parity, &p_cuser->parity);
- if (error) return error;
- PUT_USER(error,cnow.brk, &p_cuser->brk);
- if (error) return error;
- PUT_USER(error,cnow.buf_overrun, &p_cuser->buf_overrun);
- if (error) return error;
- return 0;
default:
return -ENOIOCTLCMD;
}
@@ -1351,6 +1319,32 @@ static int ioctl(struct tty_struct *tty, struct file *file,
return ret;
}
+static int get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
+{
+ SLMP_INFO *info = tty->driver_data;
+ struct mgsl_icount cnow; /* kernel counter temps */
+ unsigned long flags;
+
+ spin_lock_irqsave(&info->lock,flags);
+ cnow = info->icount;
+ spin_unlock_irqrestore(&info->lock,flags);
+
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+ icount->rx = cnow.rx;
+ icount->tx = cnow.tx;
+ icount->frame = cnow.frame;
+ icount->overrun = cnow.overrun;
+ icount->parity = cnow.parity;
+ icount->brk = cnow.brk;
+ icount->buf_overrun = cnow.buf_overrun;
+
+ return 0;
+}
+
/*
* /proc fs routines....
*/
@@ -3908,6 +3902,7 @@ static const struct tty_operations ops = {
.hangup = hangup,
.tiocmget = tiocmget,
.tiocmset = tiocmset,
+ .get_icount = get_icount,
.proc_fops = &synclinkmp_proc_fops,
};
diff --git a/drivers/serial/68360serial.c b/drivers/serial/68360serial.c
index 24661cd..1e4f831 100644
--- a/drivers/serial/68360serial.c
+++ b/drivers/serial/68360serial.c
@@ -1381,6 +1381,30 @@ static void send_break(ser_info_t *info, unsigned int duration)
}
+/*
+ * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
+ * Return: write counters to the user passed counter struct
+ * NB: both 1->0 and 0->1 transitions are counted except for
+ * RI where only 0->1 is counted.
+ */
+static int rs_360_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
+{
+ ser_info_t *info = (ser_info_t *)tty->driver_data;
+ struct async_icount cnow;
+
+ local_irq_disable();
+ cnow = info->state->icount;
+ local_irq_enable();
+
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+
+ return 0;
+}
+
static int rs_360_ioctl(struct tty_struct *tty, struct file * file,
unsigned int cmd, unsigned long arg)
{
@@ -1394,7 +1418,7 @@ static int rs_360_ioctl(struct tty_struct *tty, struct file * file,
if (serial_paranoia_check(info, tty->name, "rs_ioctl"))
return -ENODEV;
- if ((cmd != TIOCMIWAIT) && (cmd != TIOCGICOUNT)) {
+ if (cmd != TIOCMIWAIT) {
if (tty->flags & (1 << TTY_IO_ERROR))
return -EIO;
}
@@ -1477,31 +1501,6 @@ static int rs_360_ioctl(struct tty_struct *tty, struct file * file,
return 0;
#endif
- /*
- * Get counter of input serial line interrupts (DCD,RI,DSR,CTS)
- * Return: write counters to the user passed counter struct
- * NB: both 1->0 and 0->1 transitions are counted except for
- * RI where only 0->1 is counted.
- */
- case TIOCGICOUNT:
- local_irq_disable();
- cnow = info->state->icount;
- local_irq_enable();
- p_cuser = (struct serial_icounter_struct *) arg;
-/* error = put_user(cnow.cts, &p_cuser->cts); */
-/* if (error) return error; */
-/* error = put_user(cnow.dsr, &p_cuser->dsr); */
-/* if (error) return error; */
-/* error = put_user(cnow.rng, &p_cuser->rng); */
-/* if (error) return error; */
-/* error = put_user(cnow.dcd, &p_cuser->dcd); */
-/* if (error) return error; */
-
- put_user(cnow.cts, &p_cuser->cts);
- put_user(cnow.dsr, &p_cuser->dsr);
- put_user(cnow.rng, &p_cuser->rng);
- put_user(cnow.dcd, &p_cuser->dcd);
- return 0;
default:
return -ENOIOCTLCMD;
diff --git a/net/bluetooth/rfcomm/tty.c b/net/bluetooth/rfcomm/tty.c
index 309b6c2..8996006 100644
--- a/net/bluetooth/rfcomm/tty.c
+++ b/net/bluetooth/rfcomm/tty.c
@@ -844,10 +844,6 @@ static int rfcomm_tty_ioctl(struct tty_struct *tty, struct file *filp, unsigned
BT_DBG("TIOCMIWAIT");
break;
- case TIOCGICOUNT:
- BT_DBG("TIOCGICOUNT");
- break;
-
case TIOCGSERIAL:
BT_ERR("TIOCGSERIAL is not supported");
return -ENOIOCTLCMD;
--
1.7.3.2

View File

@ -0,0 +1,218 @@
From 52583e6fde34587a38c72a5c094c17e0a3503791 Mon Sep 17 00:00:00 2001
From: Alan Cox <alan@linux.intel.com>
Date: Thu, 16 Sep 2010 18:21:24 +0100
Subject: [PATCH 1/2] tty: Make tiocgicount a handler
Dan Rosenberg noted that various drivers return the struct with uncleared
fields. Instead of spending forever trying to stomp all the drivers that
get it wrong (and every new driver) do the job in one place.
This first patch adds the needed operations and hooks them up, including
the needed USB midlayer and serial core plumbing.
Signed-off-by: Alan Cox <alan@linux.intel.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
---
drivers/char/tty_io.c | 21 +++++++++++++++++++++
drivers/serial/serial_core.c | 37 +++++++++++++++++--------------------
drivers/usb/serial/usb-serial.c | 13 +++++++++++++
include/linux/tty_driver.h | 9 +++++++++
include/linux/usb/serial.h | 2 ++
5 files changed, 62 insertions(+), 20 deletions(-)
diff --git a/drivers/char/tty_io.c b/drivers/char/tty_io.c
index 507441a..3a69c39 100644
--- a/drivers/char/tty_io.c
+++ b/drivers/char/tty_io.c
@@ -96,6 +96,7 @@
#include <linux/bitops.h>
#include <linux/delay.h>
#include <linux/seq_file.h>
+#include <linux/serial.h>
#include <linux/uaccess.h>
#include <asm/system.h>
@@ -2456,6 +2457,20 @@ static int tty_tiocmset(struct tty_struct *tty, struct file *file, unsigned int
return tty->ops->tiocmset(tty, file, set, clear);
}
+static int tty_tiocgicount(struct tty_struct *tty, void __user *arg)
+{
+ int retval = -EINVAL;
+ struct serial_icounter_struct icount;
+ memset(&icount, 0, sizeof(icount));
+ if (tty->ops->get_icount)
+ retval = tty->ops->get_icount(tty, &icount);
+ if (retval != 0)
+ return retval;
+ if (copy_to_user(arg, &icount, sizeof(icount)))
+ return -EFAULT;
+ return 0;
+}
+
struct tty_struct *tty_pair_get_tty(struct tty_struct *tty)
{
if (tty->driver->type == TTY_DRIVER_TYPE_PTY &&
@@ -2576,6 +2591,12 @@ long tty_ioctl(struct file *file, unsigned int cmd, unsigned long arg)
case TIOCMBIC:
case TIOCMBIS:
return tty_tiocmset(tty, file, cmd, p);
+ case TIOCGICOUNT:
+ retval = tty_tiocgicount(tty, p);
+ /* For the moment allow fall through to the old method */
+ if (retval != -EINVAL)
+ return retval;
+ break;
case TCFLSH:
switch (arg) {
case TCIFLUSH:
diff --git a/drivers/serial/serial_core.c b/drivers/serial/serial_core.c
index 7f28307..232e2bb 100644
--- a/drivers/serial/serial_core.c
+++ b/drivers/serial/serial_core.c
@@ -1074,10 +1074,10 @@ uart_wait_modem_status(struct uart_state *state, unsigned long arg)
* NB: both 1->0 and 0->1 transitions are counted except for
* RI where only 0->1 is counted.
*/
-static int uart_get_count(struct uart_state *state,
- struct serial_icounter_struct __user *icnt)
+static int uart_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
{
- struct serial_icounter_struct icount;
+ struct uart_state *state = tty->driver_data;
struct uart_icount cnow;
struct uart_port *uport = state->uart_port;
@@ -1085,19 +1085,19 @@ static int uart_get_count(struct uart_state *state,
memcpy(&cnow, &uport->icount, sizeof(struct uart_icount));
spin_unlock_irq(&uport->lock);
- icount.cts = cnow.cts;
- icount.dsr = cnow.dsr;
- icount.rng = cnow.rng;
- icount.dcd = cnow.dcd;
- icount.rx = cnow.rx;
- icount.tx = cnow.tx;
- icount.frame = cnow.frame;
- icount.overrun = cnow.overrun;
- icount.parity = cnow.parity;
- icount.brk = cnow.brk;
- icount.buf_overrun = cnow.buf_overrun;
+ icount->cts = cnow.cts;
+ icount->dsr = cnow.dsr;
+ icount->rng = cnow.rng;
+ icount->dcd = cnow.dcd;
+ icount->rx = cnow.rx;
+ icount->tx = cnow.tx;
+ icount->frame = cnow.frame;
+ icount->overrun = cnow.overrun;
+ icount->parity = cnow.parity;
+ icount->brk = cnow.brk;
+ icount->buf_overrun = cnow.buf_overrun;
- return copy_to_user(icnt, &icount, sizeof(icount)) ? -EFAULT : 0;
+ return 0;
}
/*
@@ -1150,10 +1150,6 @@ uart_ioctl(struct tty_struct *tty, struct file *filp, unsigned int cmd,
case TIOCMIWAIT:
ret = uart_wait_modem_status(state, arg);
break;
-
- case TIOCGICOUNT:
- ret = uart_get_count(state, uarg);
- break;
}
if (ret != -ENOIOCTLCMD)
@@ -2305,6 +2301,7 @@ static const struct tty_operations uart_ops = {
#endif
.tiocmget = uart_tiocmget,
.tiocmset = uart_tiocmset,
+ .get_icount = uart_get_icount,
#ifdef CONFIG_CONSOLE_POLL
.poll_init = uart_poll_init,
.poll_get_char = uart_poll_get_char,
diff --git a/drivers/usb/serial/usb-serial.c b/drivers/usb/serial/usb-serial.c
index 941c2d4..8aea96b 100644
--- a/drivers/usb/serial/usb-serial.c
+++ b/drivers/usb/serial/usb-serial.c
@@ -519,6 +519,18 @@ static int serial_tiocmset(struct tty_struct *tty, struct file *file,
return -EINVAL;
}
+static int serial_get_icount(struct tty_struct *tty,
+ struct serial_icounter_struct *icount)
+{
+ struct usb_serial_port *port = tty->driver_data;
+
+ dbg("%s - port %d", __func__, port->number);
+
+ if (port->serial->type->get_icount)
+ return port->serial->type->get_icount(tty, icount);
+ return -EINVAL;
+}
+
/*
* We would be calling tty_wakeup here, but unfortunately some line
* disciplines have an annoying habit of calling tty->write from
@@ -1208,6 +1220,7 @@ static const struct tty_operations serial_ops = {
.chars_in_buffer = serial_chars_in_buffer,
.tiocmget = serial_tiocmget,
.tiocmset = serial_tiocmset,
+ .get_icount = serial_get_icount,
.cleanup = serial_cleanup,
.install = serial_install,
.proc_fops = &serial_proc_fops,
diff --git a/include/linux/tty_driver.h b/include/linux/tty_driver.h
index b086779..db2d227 100644
--- a/include/linux/tty_driver.h
+++ b/include/linux/tty_driver.h
@@ -224,6 +224,12 @@
* unless the tty also has a valid tty->termiox pointer.
*
* Optional: Called under the termios lock
+ *
+ * int (*get_icount)(struct tty_struct *tty, struct serial_icounter *icount);
+ *
+ * Called when the device receives a TIOCGICOUNT ioctl. Passed a kernel
+ * structure to complete. This method is optional and will only be called
+ * if provided (otherwise EINVAL will be returned).
*/
#include <linux/fs.h>
@@ -232,6 +238,7 @@
struct tty_struct;
struct tty_driver;
+struct serial_icounter_struct;
struct tty_operations {
struct tty_struct * (*lookup)(struct tty_driver *driver,
@@ -268,6 +275,8 @@ struct tty_operations {
unsigned int set, unsigned int clear);
int (*resize)(struct tty_struct *tty, struct winsize *ws);
int (*set_termiox)(struct tty_struct *tty, struct termiox *tnew);
+ int (*get_icount)(struct tty_struct *tty,
+ struct serial_icounter_struct *icount);
#ifdef CONFIG_CONSOLE_POLL
int (*poll_init)(struct tty_driver *driver, int line, char *options);
int (*poll_get_char)(struct tty_driver *driver, int line);
diff --git a/include/linux/usb/serial.h b/include/linux/usb/serial.h
index 84a4c44..8288b57 100644
--- a/include/linux/usb/serial.h
+++ b/include/linux/usb/serial.h
@@ -271,6 +271,8 @@ struct usb_serial_driver {
int (*tiocmget)(struct tty_struct *tty, struct file *file);
int (*tiocmset)(struct tty_struct *tty, struct file *file,
unsigned int set, unsigned int clear);
+ int (*get_icount)(struct tty_struct *tty,
+ struct serial_icounter_struct *icount);
/* Called by the tty layer for port level work. There may or may not
be an attached tty at this point */
void (*dtr_rts)(struct usb_serial_port *port, int on);
--
1.7.3.2

Some files were not shown because too many files have changed in this diff Show More