Provide the architecture specific implementation for SPARSEMEM for i386 SMP
and NUMA systems.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin Bligh <mbligh@aracnet.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Sparsemem abstracts the use of discontiguous mem_maps[]. This kind of
mem_map[] is needed by discontiguous memory machines (like in the old
CONFIG_DISCONTIGMEM case) as well as memory hotplug systems. Sparsemem
replaces DISCONTIGMEM when enabled, and it is hoped that it can eventually
become a complete replacement.
A significant advantage over DISCONTIGMEM is that it's completely separated
from CONFIG_NUMA. When producing this patch, it became apparent in that NUMA
and DISCONTIG are often confused.
Another advantage is that sparse doesn't require each NUMA node's ranges to be
contiguous. It can handle overlapping ranges between nodes with no problems,
where DISCONTIGMEM currently throws away that memory.
Sparsemem uses an array to provide different pfn_to_page() translations for
each SECTION_SIZE area of physical memory. This is what allows the mem_map[]
to be chopped up.
In order to do quick pfn_to_page() operations, the section number of the page
is encoded in page->flags. Part of the sparsemem infrastructure enables
sharing of these bits more dynamically (at compile-time) between the
page_zone() and sparsemem operations. However, on 32-bit architectures, the
number of bits is quite limited, and may require growing the size of the
page->flags type in certain conditions. Several things might force this to
occur: a decrease in the SECTION_SIZE (if you want to hotplug smaller areas of
memory), an increase in the physical address space, or an increase in the
number of used page->flags.
One thing to note is that, once sparsemem is present, the NUMA node
information no longer needs to be stored in the page->flags. It might provide
speed increases on certain platforms and will be stored there if there is
room. But, if out of room, an alternate (theoretically slower) mechanism is
used.
This patch introduces CONFIG_FLATMEM. It is used in almost all cases where
there used to be an #ifndef DISCONTIG, because SPARSEMEM and DISCONTIGMEM
often have to compile out the same areas of code.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin Bligh <mbligh@aracnet.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Yasunori Goto <y-goto@jp.fujitsu.com>
Signed-off-by: Bob Picco <bob.picco@hp.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Allow architectures to indicate that they will be providing hooks to indice
installed memory areas, memory_present(). Provide prototypes for the i386
implementation.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin Bligh <mbligh@aracnet.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Provide a default implementation for early_pfn_to_nid returning node 0. Allow
architectures to override this with their own implementation out of
asm/mmzone.h.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin Bligh <mbligh@aracnet.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
For all architectures, this just means that you'll see a "Memory Model"
choice in your architecture menu. For those that implement DISCONTIGMEM,
you may eventually want to make your ARCH_DISCONTIGMEM_ENABLE a "def_bool
y" and make your users select DISCONTIGMEM right out of the new choice
menu. The only disadvantage might be if you have some specific things that
you need in your help option to explain something about DISCONTIGMEM.
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
discontig.c has some assumptions that mem_map[]s inside of a node are
contiguous. Teach it to make sure that each region that it's bringing online
is actually made up of valid ranges of ram.
Written-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Introduce a simple allocator for the NUMA remap space. This space is very
scarce, used for structures which are best allocated node local.
This mechanism is also used on non-NUMA ia64 systems with a vmem_map to keep
the pgdat->node_mem_map initialized in a consistent place for all
architectures.
Issues:
o alloc_remap takes a node_id where we might expect a pgdat which was intended
to allow us to allocate the pgdat's using this mechanism; which we do not yet
do. Could have alloc_remap_node() and alloc_remap_nid() for this purpose.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The following four patches provide the last needed changes before the
introduction of sparsemem. For a more complete description of what this
will do, please see this patch:
http://www.sr71.net/patches/2.6.11/2.6.11-bk7-mhp1/broken-out/B-sparse-150-sparsemem.patch
or previous posts on the subject:
http://marc.theaimsgroup.com/?t=110868540700001&r=1&w=2http://marc.theaimsgroup.com/?l=linux-mm&m=109897373315016&w=2
Three of these are i386-only, but one of them reorganizes the macros
used to manage the space in page->flags, and will affect all platforms.
There are analogous patches to the i386 ones for ppc64, ia64, and
x86_64, but those will be submitted by the normal arch maintainers.
The combination of the four patches has been test-booted on a variety of
i386 hardware, and compiled for ppc64, i386, and x86-64 with about 17
different .configs. It's also been runtime-tested on ia64 configs (with
more patches on top).
This patch:
We _know_ which node pages in general belong to, at least at a very gross
level in node_{start,end}_pfn[]. Use those to target the allocations of
pages.
Signed-off-by: Andy Whitcroft <apw@shadowen.org>
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch effectively eliminates direct use of pgdat->node_mem_map outside
of the DISCONTIG code. On a flat memory system, these fields aren't
currently used, neither are they on a sparsemem system.
There was also a node_mem_map(nid) macro on many architectures. Its use
along with the use of ->node_mem_map itself was not consistent. It has
been removed in favor of two new, more explicit, arch-independent macros:
pgdat_page_nr(pgdat, pagenr)
nid_page_nr(nid, pagenr)
I called them "pgdat" and "nid" because we overload the term "node" to mean
"NUMA node", "DISCONTIG node" or "pg_data_t" in very confusing ways. I
believe the newer names are much clearer.
These macros can be overridden in the sparsemem case with a theoretically
slower operation using node_start_pfn and pfn_to_page(), instead. We could
make this the only behavior if people want, but I don't want to change too
much at once. One thing at a time.
This patch removes more code than it adds.
Compile tested on alpha, alpha discontig, arm, arm-discontig, i386, i386
generic, NUMAQ, Summit, ppc64, ppc64 discontig, and x86_64. Full list
here: http://sr71.net/patches/2.6.12/2.6.12-rc1-mhp2/configs/
Boot tested on NUMAQ, x86 SMP and ppc64 power4/5 LPARs.
Signed-off-by: Dave Hansen <haveblue@us.ibm.com>
Signed-off-by: Martin J. Bligh <mbligh@aracnet.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
I am always trying to make sure I've booted the right kernel after a new
install. Too paranoid maybe. But I guess there're other people like me.
So let's make kbuild display the compile version number at the end to give
us a hint. I know we may be booting vmlinux someday, but don't care about
it for now.
Signed-off-by: Coywolf Qi Hunt <coywolf@lovecn.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Remove PG_highmem, to save a page flag. Use is_highmem() instead. It'll
generate a little more code, but we don't use PageHigheMem() in many places.
Signed-off-by: Badari Pulavarty <pbadari@us.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Ingo recently introduced a great speedup for allocating new mmaps using the
free_area_cache pointer which boosts the specweb SSL benchmark by 4-5% and
causes huge performance increases in thread creation.
The downside of this patch is that it does lead to fragmentation in the
mmap-ed areas (visible via /proc/self/maps), such that some applications
that work fine under 2.4 kernels quickly run out of memory on any 2.6
kernel.
The problem is twofold:
1) the free_area_cache is used to continue a search for memory where
the last search ended. Before the change new areas were always
searched from the base address on.
So now new small areas are cluttering holes of all sizes
throughout the whole mmap-able region whereas before small holes
tended to close holes near the base leaving holes far from the base
large and available for larger requests.
2) the free_area_cache also is set to the location of the last
munmap-ed area so in scenarios where we allocate e.g. five regions of
1K each, then free regions 4 2 3 in this order the next request for 1K
will be placed in the position of the old region 3, whereas before we
appended it to the still active region 1, placing it at the location
of the old region 2. Before we had 1 free region of 2K, now we only
get two free regions of 1K -> fragmentation.
The patch addresses thes issues by introducing yet another cache descriptor
cached_hole_size that contains the largest known hole size below the
current free_area_cache. If a new request comes in the size is compared
against the cached_hole_size and if the request can be filled with a hole
below free_area_cache the search is started from the base instead.
The results look promising: Whereas 2.6.12-rc4 fragments quickly and my
(earlier posted) leakme.c test program terminates after 50000+ iterations
with 96 distinct and fragmented maps in /proc/self/maps it performs nicely
(as expected) with thread creation, Ingo's test_str02 with 20000 threads
requires 0.7s system time.
Taking out Ingo's patch (un-patch available per request) by basically
deleting all mentions of free_area_cache from the kernel and starting the
search for new memory always at the respective bases we observe: leakme
terminates successfully with 11 distinctive hardly fragmented areas in
/proc/self/maps but thread creating is gringdingly slow: 30+s(!) system
time for Ingo's test_str02 with 20000 threads.
Now - drumroll ;-) the appended patch works fine with leakme: it ends with
only 7 distinct areas in /proc/self/maps and also thread creation seems
sufficiently fast with 0.71s for 20000 threads.
Signed-off-by: Wolfgang Wander <wwc@rentec.com>
Credit-to: "Richard Purdie" <rpurdie@rpsys.net>
Signed-off-by: Ken Chen <kenneth.w.chen@intel.com>
Acked-by: Ingo Molnar <mingo@elte.hu> (partly)
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A lot of the code in arch/*/mm/hugetlbpage.c is quite similar. This patch
attempts to consolidate a lot of the code across the arch's, putting the
combined version in mm/hugetlb.c. There are a couple of uglyish hacks in
order to covert all the hugepage archs, but the result is a very large
reduction in the total amount of code. It also means things like hugepage
lazy allocation could be implemented in one place, instead of six.
Tested, at least a little, on ppc64, i386 and x86_64.
Notes:
- this patch changes the meaning of set_huge_pte() to be more
analagous to set_pte()
- does SH4 need s special huge_ptep_get_and_clear()??
Acked-by: William Lee Irwin <wli@holomorphy.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This is the core of the (much simplified) early reclaim. The goal of this
patch is to reclaim some easily-freed pages from a zone before falling back
onto another zone.
One of the major uses of this is NUMA machines. With the default allocator
behavior the allocator would look for memory in another zone, which might be
off-node, before trying to reclaim from the current zone.
This adds a zone tuneable to enable early zone reclaim. It is selected on a
per-zone basis and is turned on/off via syscall.
Adding some extra throttling on the reclaim was also required (patch
4/4). Without the machine would grind to a crawl when doing a "make -j"
kernel build. Even with this patch the System Time is higher on
average, but it seems tolerable. Here are some numbers for kernbench
runs on a 2-node, 4cpu, 8Gig RAM Altix in the "make -j" run:
wall user sys %cpu ctx sw. sleeps
---- ---- --- ---- ------ ------
No patch 1009 1384 847 258 298170 504402
w/patch, no reclaim 880 1376 667 288 254064 396745
w/patch & reclaim 1079 1385 926 252 291625 548873
These numbers are the average of 2 runs of 3 "make -j" runs done right
after system boot. Run-to-run variability for "make -j" is huge, so
these numbers aren't terribly useful except to seee that with reclaim
the benchmark still finishes in a reasonable amount of time.
I also looked at the NUMA hit/miss stats for the "make -j" runs and the
reclaim doesn't make any difference when the machine is thrashing away.
Doing a "make -j8" on a single node that is filled with page cache pages
takes 700 seconds with reclaim turned on and 735 seconds without reclaim
(due to remote memory accesses).
The simple zone_reclaim syscall program is at
http://www.bork.org/~mort/sgi/zone_reclaim.c
Signed-off-by: Martin Hicks <mort@sgi.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch implements a number of smp_processor_id() cleanup ideas that
Arjan van de Ven and I came up with.
The previous __smp_processor_id/_smp_processor_id/smp_processor_id API
spaghetti was hard to follow both on the implementational and on the
usage side.
Some of the complexity arose from picking wrong names, some of the
complexity comes from the fact that not all architectures defined
__smp_processor_id.
In the new code, there are two externally visible symbols:
- smp_processor_id(): debug variant.
- raw_smp_processor_id(): nondebug variant. Replaces all existing
uses of _smp_processor_id() and __smp_processor_id(). Defined
by every SMP architecture in include/asm-*/smp.h.
There is one new internal symbol, dependent on DEBUG_PREEMPT:
- debug_smp_processor_id(): internal debug variant, mapped to
smp_processor_id().
Also, i moved debug_smp_processor_id() from lib/kernel_lock.c into a new
lib/smp_processor_id.c file. All related comments got updated and/or
clarified.
I have build/boot tested the following 8 .config combinations on x86:
{SMP,UP} x {PREEMPT,!PREEMPT} x {DEBUG_PREEMPT,!DEBUG_PREEMPT}
I have also build/boot tested x64 on UP/PREEMPT/DEBUG_PREEMPT. (Other
architectures are untested, but should work just fine.)
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Arjan van de Ven <arjan@infradead.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch causes the ignore_normal_resume flag to be set slightly earlier,
before there is a chance that the apm driver will receive the normal resume
event from the BIOS. (Addresses Debian bug #310865)
Signed-off-by: Thomas Hood <jdthood@yahoo.co.uk>
Acked-by: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
arch/i386/kernel/vsyscall-note.o is not listed as a target so its .cmd file
is neither considered as a target nor is it read on the next build. This
causes vsyscall-note.o to be rebuilt every time that you run make, which
causes vmlinux to be rebuilt every time.
Signed-off-by: Keith Owens <kaos@ocs.com.au>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
As mandated by the spec, disable timer around transitions.
From code by : Ken Staton <ken_staton@agilent.com
Signed-off-by: Dave Jones <davej@redhat.com>
The spec states that we have to do this, which is *horrid*.
Based on code from: Ken Staton <ken_staton@agilent.com>
Signed-off-by: Dave Jones <davej@redhat.com>
With the release of the dual-core AMD Opterons last week,
it's high time that cpufreq supported them. The attached
patch applies cleanly to 2.6.12-rc3 and updates powernow-k8
to support the latest Athlon 64 and Opteron processors.
Update the driver to version 1.40.0 and provide support
for dual-core processors.
Signed-off-by: Mark Langsdorf <mark.langsdorf@amd.com>
Signed-off-by: Dave Jones <davej@redhat.com>
Some cpufreq drivers (at that time, only powernow-k7) need to recalibrate the
cpu_khz at runtime.
Signed-off-by: Bruno Ducrot <ducrot@poupinou.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Dave Jones <davej@redhat.com>
We have to recalibrate cpu_khz in order to use the current FID instead the max
FID since some BIOS do not put the processor at maximum frequency at POST.
Also, some BIOS will change the processor frequency at our back after cpu_khz
was calibrate. Finally, this will fix a long standing bug when we do
something like this:
# rmmod powernow-k7
# modprobe powernow-k7
Signed-off-by: Bruno Ducrot <ducrot@poupinou.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Dave Jones <davej@redhat.com>
The speedstep-smi driver actually works on >=1 notebook with a
Pentium 4-M CPU where all other cpufreq drivers fail. Therefore,
allow speedstep-smi on P4Ms again, but warn users of likely failure
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Dave Jones <davej@redhat.com>
The Pentium 4 - Ms (HT) with CPUID 0xF34 and 0xF41 seem to support
centrino-like enhanced speedstep; however, no "table" support is possible.
Therefore, put NULL entries into speedstep-centrino.c
Signed-off-by: Dominik Brodowski <linux@dominikbrodowski.net>
Signed-off-by: Dave Jones <davej@redhat.com>
This is a neverending story
linux/acpi.h contains empty declarations for acpi_boot_init() &
acpi_boot_table_init() but they are nested inside #ifdef CONFIG_ACPI.
So we'll have to #ifdef in arch/i386/kernel/setup.c: setup_arch()
Signed-off-by: Alexander Nyberg <alexn@telia.com>
Cc: "Brown, Len" <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This fixes 'smp_num_siblings' value on the systems with a buggy bios,
which sets number of siblings to '2' even when HT is disabled. (more
details are at http://bugzilla.kernel.org/show_bug.cgi?id=4359)
I am planning to do more cleanup in this area (like moving smp_num_siblings
to per cpuinfo) shortly.
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
num_cache_leaves is used in __devexit cache_remove_dev() and can therefore
not be __devinit.
Signed-off-by: Adrian Bunk <bunk@stusta.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Even after the previous fix you can still set CONFIG_ACPI_BOOT
indirectly even without CONFIG_ACPI by choosing CONFIG_PCI and
CONFIG_PCI_MMCONFIG.
That doesn't build very well either.
This makes PCI_MMCONFIG depend on ACPI, fixing that hole.
[ I guess in theory Kconfig could follow the whole chain of dependencies
for things that get selected, but that sounds insanely complicated, so
we'll just fix up these things by hand. --Linus ]
Signed-off-by: Alexander Nyberg <alexn@telia.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Delete quirk_via_bridge(), restore quirk_via_irqpic() -- but now
improved to be invoked upon device ENABLE, and now only for VIA devices
-- not all devices behind VIA bridges.
Signed-off-by: Bjorn Helgaas <bjorn.helgaas@hp.com>
Signed-off-by: Len Brown <len.brown@intel.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
This patch fixes a compile bug by moving a static inline function to the
right place. The body of a static inline function has to be declared
before the use of this function.
Signed-off-by: Dominik Hackl <dominik@hackl.dhs.org>
Cc: James Bottomley <James.Bottomley@steeleye.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
We need to hold the vmlist_lock while doing change_page_attr, otherwise we
could reset someone else's mapping.
Requires previous patch to add __remove_vm_area
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- Remove duplicated ifdef
- Make core_id match what Intel uses
- Initialize phys_proc_id correctly for non DC case
- Handle non power of two core numbers.
Fixes for both i386 and x86-64
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The GET_INDEX() macro should use just the low three bits of the devfn,
otherwise we have a memory scribble in pcie_rootport_aspm_quirk that
overwrites ptype_all
Fix it to be more careful about its arguments while at it.
Acked by Dely Sy <dely.l.sy@intel.com>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Last round hopefully of cpu_core_id changes hopefully fow now:
- Always initialize cpu_core_id for all CPUs, even when no dual core setup
is detected. This prevents funny /proc/cpuinfo output
- Do the same with phys_proc_id[] even when no HyperThreading - dito.
- Use the CPU APIC-ID from CPUID 1 instead of the linux virtual CPU number
to identify the core for AMD dual core setups.
Patch for i386/x86-64.
Signed-off-by: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
- Changed Name/defines from "Geode GX" to "Geode GX1" for clarification
- Dropped "-march=i586" in favor of "-march=i486"
- Dopped X86_OOSTORE support for Geode GX1
Signed-off-by: Kianusch Sayah Karadji <kianusch@sk-tech.net>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
When I do a "diff -Nur arch/i386 arch/x86_64" to see what is different between these two
architectures, I see some differences due to whitespace issues only. The attached patch removes
some of the noise by fixing up the following files:
- arch/i386/boot/bootsect.S
- arch/i386/boot/video.S
- arch/x86_64/boot/bootsect.S
Signed-off-by: Daniel Dickman <didickman@yahoo.com>
Signed-off-by: Domen Puncer <domen@coderock.org>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Kprobes could not handle the insertion of a probe on the ret/lret
instruction and used to oops after single stepping since kprobes was
modifying eip/rip incorrectly. Adjustment of eip/rip is not required after
single stepping in case of ret/lret instruction, because eip/rip points to
the correct location after execution of the ret/lret instruction. This
patch fixes the above problem.
Signed-off-by: Prasanna S Panchamukhi <prasanna@in.ibm.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
In include/asm-x86_64/string.h there are such comments:
/* Use C out of line version for memcmp */
#define memcmp __builtin_memcmp
int memcmp(const void * cs,const void * ct,size_t count);
This would mean that if the compiler does not decide to use __builtin_memcmp,
it emits a call to memcmp to be satisfied by the C out-of-line version in
lib/string.c. What happens is that after preprocessing, in lib/string.i you
may find the definition of "__builtin_strcmp".
Actually, by accident, in the object you will find the definition of strcmp
and such (maybe a trick intended to redirect calls to __builtin_memcmp to the
default memcmp when the definition is not expanded); however, this particular
case is not a documented feature as far as I can see.
Also, the EXPORT_SYMBOL does not work, so it's duplicated in the arch.
I simply added some #undef to lib/string.c and removed the (now duplicated)
exports in x86-64 and UML/x86_64 subarchs (the second ones are introduced by
another patch I just posted for -mm).
Signed-off-by: Paolo 'Blaisorblade' Giarrusso <blaisorblade@yahoo.it>
CC: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
The recent change fix-crash-in-entrys-restore_all.patch
childregs->esp = esp;
p->thread.esp = (unsigned long) childregs;
- p->thread.esp0 = (unsigned long) (childregs+1);
+ p->thread.esp0 = (unsigned long) (childregs+1) - 8;
p->thread.eip = (unsigned long) ret_from_fork;
introduces an inconsistency between esp and esp0 before the task is run the
first time. esp0 is no longer the actual start of the stack, but 8 bytes
off.
This shows itself clearly in a scenario when a ptracer that is set to also
ptrace eventual children traces program1 which then clones thread1. Now
the ptracer wants to modify the registers of thread1. The x86 ptrace
implementation bases it's knowledge about saved user-space registers upon
p->thread.esp0. But this will be a few bytes off causing certain writes to
the kernel stack to overwrite a saved kernel function address making the
kernel when actually running thread1 jump out into user-space. Very
spectacular.
The testcase I've used is:
/* start with strace -f ./a.out */
#include <pthread.h>
#include <stdio.h>
void *do_thread(void *p)
{
for (;;);
}
int main()
{
pthread_t one;
pthread_create(&one, NULL, &do_thread, NULL);
for (;;);
return 0;
}
So, my solution is to instead of just adjusting esp0 that creates an
inconsitent state I adjust where the user-space registers are saved with -8
bytes. This gives us the wanted extra bytes on the start of the stack and
esp0 is now correct. This solves the issues I saw from the original
testcase from Mateusz Berezecki and has survived testing here. I think
this should go into -mm a round or two first however as there might be some
cruft around depending on pt_regs lying on the start of the stack. That
however would have broken with the first change too!
It's actually a 2-line diff but I had to move the comment of why the -8 bytes
are there a few lines up. Thanks to Zwane for helping me with this.
Signed-off-by: Alexander Nyberg <alexn@telia.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
A bunch of drivers use ISA DMA helpers or their equivalents for
platforms that have ISA with different DMA controller (a lot of ARM
boxen). Currently there is no way to put such dependency in Kconfig -
CONFIG_ISA is not it (e.g. it is not set on platforms that have no ISA
slots, but have on-board devices that pretend to be ISA ones).
New symbol added - ISA_DMA_API. Set when we have functional
enable_dma()/set_dma_mode()/etc. set of helpers. Next patches in the
series will add missing dependencies for drivers that need them.
I'm very carefully staying the hell out of the recurring flamefest on
what exactly CONFIG_ISA would mean in ideal world - added symbol has a
well-defined meaning and for now I really want to treat it as completely
independent from the mess around CONFIG_ISA.
Signed-off-by: Al Viro <viro@parcelfarce.linux.theplanet.co.uk>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>
