Commit Graph

247 Commits

Author SHA1 Message Date
Martin Hicks
a994018a5f [IA64] uncached allocator: use generic (not sn2 specific) functions
Change sn2-specific calls into generic functions.  Without this change
the uncached allocator will not work on non-sn2 platforms.

Signed-off-by: Greg Edwards <edwardsg@sgi.com>
Signed-off-by: Martin Hicks <mort@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-31 14:18:04 -07:00
Tony Luck
288ceb8f14 Auto-update from upstream 2005-08-30 09:30:09 -07:00
Tony Luck
3290580285 Pull rationalise-regions into release branch 2005-08-29 15:50:32 -07:00
Tony Luck
bcdd3a9114 Pull ngam-maule-steiner into release branch 2005-08-29 15:48:51 -07:00
Tony Luck
b946ecbb11 Pull pending-2.6.14 into release branch 2005-08-29 15:48:23 -07:00
Tony Luck
7115c13bd6 Pull acpi-p-state into release branch 2005-08-29 14:15:10 -07:00
Tony Luck
7d2e423add Pull asm-segment into release branch 2005-08-29 14:14:40 -07:00
Tony Luck
3ec8a773d2 Pull swiotlb-size into release branch 2005-08-29 14:14:18 -07:00
Steven Rostedt
69be8f1896 [PATCH] convert signal handling of NODEFER to act like other Unix boxes.
It has been reported that the way Linux handles NODEFER for signals is
not consistent with the way other Unix boxes handle it.  I've written a
program to test the behavior of how this flag affects signals and had
several reports from people who ran this on various Unix boxes,
confirming that Linux seems to be unique on the way this is handled.

The way NODEFER affects signals on other Unix boxes is as follows:

1) If NODEFER is set, other signals in sa_mask are still blocked.

2) If NODEFER is set and the signal is in sa_mask, then the signal is
still blocked. (Note: this is the behavior of all tested but Linux _and_
NetBSD 2.0 *).

The way NODEFER affects signals on Linux:

1) If NODEFER is set, other signals are _not_ blocked regardless of
sa_mask (Even NetBSD doesn't do this).

2) If NODEFER is set and the signal is in sa_mask, then the signal being
handled is not blocked.

The patch converts signal handling in all current Linux architectures to
the way most Unix boxes work.

Unix boxes that were tested:  DU4, AIX 5.2, Irix 6.5, NetBSD 2.0, SFU
3.5 on WinXP, AIX 5.3, Mac OSX, and of course Linux 2.6.13-rcX.

* NetBSD was the only other Unix to behave like Linux on point #2. The
main concern was brought up by point #1 which even NetBSD isn't like
Linux.  So with this patch, we leave NetBSD as the lonely one that
behaves differently here with #2.

Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-08-29 10:03:11 -07:00
Kenneth Chen
f8220c7f15 [IA64] Delete erroneous copy_page.o in global lib-y list
copy_page.o appeared twice in arch/ia64/lib/Makefile. The
one in global lib-y is wrong where it should be just in
lib-$(CONFIG_ITANIUM).

Both copy_page.o and copy_page_mck.o are build for Itanium2
processor and the link order will pick up the low performing
copy_page function (originally written for itanium processor).
In this case, we really want the copy_page_mck.o for optimized
version.

Signed-off-by: Kenneth Chen <kenneth.w.chen@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-26 16:01:25 -07:00
Venkatesh Pallipadi
4db8699bcf [IA64] Add ACPI based P-state support
Patch to support P-state transitions on ia64. This driver is based on ACPI,
and uses the ACPI processor driver interface to find out the P-state support
information for the processor. This driver plugs into generic cpufreq
infrastructure.

Once this driver is loaded successfully, ondemand/userspace governor can be
used to change the CPU frequency dynamically based on load or on request from
userspace process.

Refer :
ACPI specification -
      http://www.acpi.info
P-state related PAL calls -
      http://developer.intel.com/design/itanium/downloads/24869909.pdf

Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-26 15:09:24 -07:00
Russ Anderson
d1e079b3fc [IA64-SGI] fix bte_copy() calling smp_processor_id() while preemptible
bte_copy() calls calls smp_processor_id(), which will get flagged if
preemption if enabled.  raw_smp_processor_id() is used instead 
because we are just using it to pick a BTE interface and are not 
tied to a specific cpu.

Signed-off-by: Russ Anderson (rja@sgi.com)
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-26 15:01:37 -07:00
Mark Maule
8409668b56 [IA64] altix: Abstract irq_affinity at the sn pci provider
Altix patch to abstract irq_affinity down to the pci provider level since
different SGI hardware implements this in different ways.

Signed-off-by: Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-26 12:09:01 -07:00
Russ Anderson
5b9021bc58 [IA64] SGI SN remove redundant partition SAL call
Clean up of SGI SN partitioning related code.
The SN_SAL_GET_SN_INFO SAL call returns the partition ID, making
the SN_SAL_SYSCTL_PARTITION_GET SAL call redundant.  Remove sn_partid
and use sn_partition_id.

Signed-off-by: Russ Anderson (rja@sgi.com)
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-24 16:26:13 -07:00
Mark Goodwin
5390970d1c [IA64] - SGI SN hwperf enhancements -
Update the SN pci device info to use the nearest node function
to allocate driver memory on the nearest node (rather than
defaulting to node 0).

Signed-off-by: Mark Goodwin <markgw@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-24 16:25:21 -07:00
Mark Goodwin
60a3ba0bb4 [IA64] - SGI SN hwperf enhancements -
Add a new exported function for determining the nearest node
with CPUs for I/O nodes and fix a bug where the hwperf dynamic
misc device was being registered before misc_init(). 

Signed-off-by: Mark Goodwin <markgw@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-24 16:24:42 -07:00
Mark Goodwin
ecc3c30ae3 [IA64] - SGI SN hwperf enhancements - export_pci_topology
Bugfix to export PCI topology information in /proc/sgi_sn/sn_topology.

Signed-off-by: Mark Goodwin <markgw@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-24 16:23:39 -07:00
Peter Chubb
0a41e25011 [IA64] Rationalise Region Definitions
Currently, region numbers are defined in several files, with several 
names.  For example, we have REGION_KERNEL in asm/page.h and 
RGN_KERNEL in pgtable.h 
 
We also have address definitions that should depend on the 
RGN_XXX macros, but are currently just long constants. 
 
The following patch reorganises all the definitions so that they have 
the same form (RGN_XXX), are in one place, and that addresses that 
depend on RGN_XXX are derived from them. 

(This is a necessary but not sufficient patch to allow UML-like 
operation on IA64). 

Thanks to David Mosberger for catching the change I missed in mmu_context.h.
 
Signed-off-by: Peter Chubb <peterc@gelato.unsw.edu.au> 
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-24 15:35:41 -07:00
Adrian Bunk
38d26b9f57 [IA64] arch/ia64/hp/sim/boot/fw-emu.c: remove egcs workaround
Kernel 2.6 doesn't support egcs, and I didn't find any user of this 
function.

Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-24 14:50:37 -07:00
Kumar Gala
a877bd36f7 [IA64] remove use of asm/segment.h
Removed IA64 architecture specific users of asm/segment.h
The removal of asm-ia64/segment.h itself can wait until all
of the kernel source has been purged of references.

Signed-off-by: Kumar Gala <kumar.gala@freescale.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-24 13:23:30 -07:00
Keith Owens
41290c1464 [PATCH] Export pcibios_bus_to_resource
pcibios_bus_to_resource is exported on all architectures except ia64
and sparc.  Add exports for the two missing architectures.  Needed when
Yenta socket support is compiled as a module.

Signed-off-by: Keith Owens <kaos@sgi.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-08-24 10:22:44 -07:00
Peter Chubb
a4cce10492 [IA64] Fix simulator boot (for real this time).
Thanks to Stephane, we've now worked out the real cause of the
`Linux  will not boot on simulator' problem.  Turns out it's a stack
overflow because the stack pointer wasn't being initialised properly
in boot_head.S (it was being initialised to the lowest instead of the
highest address of the stack, so the first push started to overwrite
data in the BSS).

Signed-off-by: Peter Chubb <peterc@gelato.unsw.edu.au>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-23 07:41:56 -07:00
Tony Luck
62d75f3753 [IA64] backout incorrect fix for simulator boot issue
Earlier fix in 4aec0fb122 just
masked the real problem.

Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-23 07:39:15 -07:00
Tony Luck
034e5356a4 Pull prarit-bus-sysdata into release branch 2005-08-23 07:27:28 -07:00
Alex Williamson
e8579e72ca [IA64, X86_64] fix swiotlb sizing
Fix swiotlb sizing to match what the comments and the kernel
parameters documentation indicate.  Given a default 16k page size kernel
(ia64) and a 2k swiotlb page size, we're off by a multiple of 8 trying
to size the swiotlb.  When specified on the boot line, the swiotlb is
made 8x bigger than requested.  When left to the default value, it's 8x
smaller than the comments indicate.  For x86_64 the multiplier would be
2x.  The patch below fixes this.  Now, what's a good default swiotlb
size?  Apparently we don't really need 64MB.

Signed-off-by: Alex Williamson <alex.williamson@hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-18 16:33:28 -07:00
Ian Wienand
4aec0fb122 [IA64] Simulator bootloader fails with gcc 4
After building a fresh tree with gcc 4 I can't boot the simulator as
the bootloader loader dies with 

loading /home/ianw/kerntest/kerncomp//build/sim_defconfig/vmlinux...
failed to read phdr

After some investigation I believe this is do with differences between
the alignment of variables on the stack between gcc 3 and 4 and the
ski simulator.  If you trace through with the simulator you can see
that the disk_stat structure value returned from the SSC_WAIT_COMPLETION
call seems to be only half loaded.  I guess it doesn't like the alignment
of the input.

Signed-off-by: Ian Wienand <ianw@gelato.unsw.edu.au>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-18 14:10:41 -07:00
Jack Steiner
470ceb05d9 [IA64-SGI] - New SN hardware support - ptc_fixes
Shub2 provides a much improved mechanism for issuing internode
TLB purges. Add code to support the newer mechanism. There is also 
some debug code (disabled) that is useful for testing.

Collect statistics on the number, type & duration of TLB purges.
This data will be useful for making future improvements in the algorithms.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-17 15:32:24 -07:00
Jack Steiner
68b9753f47 [IA64-SGI] - New SN hardware support - cpu_relax
Add a few missing calls to "hint @pause".

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-17 15:31:53 -07:00
Jack Steiner
7e95b9d6e2 [IA64-SGI] - New SN hardware support - bte_fixes
Change the BTE driver so that it works for both shub1 and
shub2. Most of the changes are related to the number of cores
that use the BTE engine, to the MMR addresses of various
shub registers, and to using the correct processor or network
physical address.

Signed-off-by: Russ Anderson <rja@sgi.com>
Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-17 15:31:14 -07:00
Jack Steiner
1007d02160 [IA64-SGI] - New SN hardware support - no_wars
Disable some shub1-specific code when running on systems with shub2.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-17 15:30:44 -07:00
Jack Steiner
2fdbb590e4 [IA64-SGI] - New SN hardware support - boot_init_shub2
Update the addresses of the pio_write_status_addr so that
they are correct for newer processors. Shub2 did not number
the threads in the order that I had expected.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-17 15:30:14 -07:00
Jack Steiner
0aa2c72e59 [IA64-SGI] - New SN hardware support - use_alias_space
Use local SHUB alias space when referencing MMRs that are known
to be node local. There is a slight performance benefit & code 
simplification.

Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-17 15:29:43 -07:00
Tony Luck
c149ec05dc [IA64] Updated tiger defconfig
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-17 10:24:17 -07:00
Peter Chubb
7f09d6f935 [IA64] Updated zx1 defconfig
Just `make oldconfig' doesn't help for the zx1 defconfig ---
because we need the MPT Fusion drivers, which are picked up as not
selected.
Tested on HP ZX2000 and ZX2600.

Signed-off-by: Peter Chubb <peterc@gelato.unsw.edu.au>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-16 17:34:31 -07:00
Greg Edwards
7b1a843f46 [IA64] Refresh arch/ia64/configs/sn2_defconfig.
Signed-off-by: Greg Edwards <edwardsg@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-16 16:26:36 -07:00
Keith Owens
71841b8fe7 [IA64] Initialize some spinlocks
Some IA64 spinlocks are not being initialized, make it so.

Signed-off-by: Keith Owens <kaos@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-16 15:33:26 -07:00
Prarit Bhargava
c1ffb910f7 [IA64]: SN fix bus->sysdata pointer and memory cleanups
The main issue is that bus_fixup calls may potentially call
functions that require a valid bus->sysdata pointer.  Since
this is the case, we must set the bus->sysdata pointer before
calling the bus_fixup functions.  The remaining changes are
simple fixes to make sure memory is cleaned up in the function.

Signed-off-by: Prarit Bhargava <prarit@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-16 15:24:48 -07:00
Tony Luck
f7001e8f1f Auto-update from upstream 2005-08-16 11:29:57 -07:00
John Hawkes
367ae3cd74 [PATCH] fix for ia64 sched-domains code
Fix for ia64 sched domain building triggered by cpuset code.

Acked-by: Nick Piggin <npiggin@suse.de>
Acked-by: Dinakar Guniguntala <dino@in.ibm.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-08-16 08:54:00 -07:00
Christoph Hellwig
85f265d887 [IA64] update CONFIG_PCI description
The current one doesn't even make sense anymore on i386 where it
apparently came from.

Follow-up wordsmithing by Matthew Wilcox and Tony Luck.


Signed-off-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-15 15:03:12 -07:00
Mark Maule
c9221da9f2 [IA64-SGI] sn pci provider for TIOCE (pci
Altix patch to add an SN pci provider for TIOCE, which is SGI's 
PCI Express implementation.

Signed-off-by: Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-11 15:56:31 -07:00
Mark Maule
5b53ed1f2e [IA64-SGI] add support for TIO huge-window
Altix patch to add TIO "huge-window" address support to sn_dma_flush().

Update copyright in affected files.

Signed-off-by: Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-11 15:37:13 -07:00
Mark Maule
735e60f4c6 [IA64-SGI] abstract force_interrupt() mechanism
Altix patch to abstract the force_interrupt() mechanism away from the
pcibr provider.

Signed-off-by: Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-11 15:36:39 -07:00
Mark Maule
89963d16dc [IA64-SGI] altix: cosmetic rename of SGI_PCIBR_ERROR
Cosmetic altix patch to rename SGI_PCIBR_ERROR to something more generic and
remove a duplicate #define.

Signed-off-by: Mark Maule <maule@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-11 15:36:00 -07:00
Colin Ngam
674c6479b7 [IA64-SGI] Altix only: Add PCI Domain number support.
This patch enables PCI Domain numbering on Altix.

Signed-off-by: Colin Ngam <cngam@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-11 15:35:13 -07:00
stephane.eranian@hp.com
6bf11e8c70 [IA64] fix perfmon context load
The PFM_LOAD_CONTEXT may fail silently and cause a session
to remain reserved even though it should not. This can happen
when the commands succeeds in reserving the session but fails
when it actually tries to attach to the load_pid. In that case,
the command has failed but will return 0. More importantly,
the session will remain reserved. This patch fixes the problem.

Signed-off-by: <stephane.eranian@hp.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-10 16:21:58 -07:00
Ken Chen
fb573856b2 [IA64] fix nohalt boot option
this changeset broke the "nohalt" kernel boot option.
  8df5a500a3

default_idle() is looking at new variable can_do_pal_halt.  However,
that variable did not get cleared upon "nohalt" boot option.  Result
is that "nohalt" option is ignored until perfmon is exercised.

Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-08-08 15:39:47 -07:00
Ingo Molnar
6cb54819d7 [PATCH] remove sys_set_zone_reclaim()
This removes sys_set_zone_reclaim() for now.  While i'm sure Martin is
trying to solve a real problem, we must not hard-code an incomplete and
insufficient approach into a syscall, because syscalls are pretty much
for eternity.  I am quite strongly convinced that this syscall must not
hit v2.6.13 in its current form.

Firstly, the syscall lacks basic syscall design: e.g. it allows the
global setting of VM policy for unprivileged users. (!) [ Imagine an
Oracle installation and a SAP installation on the same NUMA box fighting
over the 'optimal' setting for this flag. What will they do? Will they
try to set the flag to their own preferred value every second or so? ]

Secondly, it was added based on a single datapoint from Martin:

 http://marc.theaimsgroup.com/?l=linux-mm&m=111763597218177&w=2

where Martin characterizes the numbers the following way:

 ' Run-to-run variability for "make -j" is huge, so these numbers aren't
   terribly useful except to see that with reclaim the benchmark still
   finishes in a reasonable amount of time. '

in other words: the fundamental problem has likely not been solved, only
a tendential move into the right direction has been observed, and a
handful of numbers were picked out of a set of hugely variable results,
without showing the variability data. How much variance is there
run-to-run?

I'd really suggest to first walk the walk and see what's needed to get
stable & predictable kernel compilation numbers on that NUMA box, before
adding random syscalls to tune a particular aspect of the VM ... which
approach might not even matter once the whole picture has been analyzed
and understood!

The third, most important point is that the syscall exposes VM tuning
internals in a completely unstructured way. What sense does it make to
have a _GLOBAL_ per-node setting for 'should we go to another node for
reclaim'? If then it might make sense to do this per-app, via numalib or
so.

The change is minimalistic in that it doesnt remove the syscall and the
underlying infrastructure changes, only the user-visible changes.  We
could perhaps add a CAP_SYS_ADMIN-only sysctl for this hack, a'ka
/proc/sys/vm/swappiness, but even that looks quite counterproductive
when the generic approach is that we are trying to reduce the number of
external factors in the VM balance picture.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
2005-08-01 10:03:56 -07:00
Keith Owens
b833961bd3 [IA64] unwind.c uses wrong unat from switch_stack
unwind.c can read the wrong unat bits from switch_stack.
sw->caller_unat is the value of ar.unat when the task was blocked.
sw->ar_unat is the value of ar.unat after doing st8.spill for r4-7.
IOW, ar_unat is caller_unat with 4 bits changed.

unw_access_gr() uses sw->ar_unat for r4-7 (correct), but it also uses
sw->ar_unat for other scratch registers (incorrect).  sw->ar_unat
should only be used for r4-7, everything else should use
sw->caller_unat, unless modified by unwind info.  Using sw->ar_unat
risks picking up the 4 bits that were overwritten when r4-7 were saved.

Also this line is wrong
	unw.sw_off[unw.preg_index[UNW_REG_PFS]] = SW(AR_UNAT);
and should be
	unw.sw_off[unw.preg_index[UNW_REG_PFS]] = SW(AR_PFS);

Signed-off-by: Keith Owens <kaos@sgi.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-07-27 14:18:08 -07:00
Robert Love
d108919b2b [IA64] inotify: ia64 syscalls.
Attached patch adds the inotify syscalls to ia64.

Signed-off-by: Robert Love <rml@novell.com>
Signed-off-by: Tony Luck <tony.luck@intel.com>
2005-07-27 10:46:12 -07:00