Vivek Goyal [Wed, 2 May 2007 17:27:08 +0000 (19:27 +0200)]
[PATCH] i386: modpost apic related warning fixes
o Modpost generates warnings for i386 if compiled with CONFIG_RELOCATABLE=y
WARNING: vmlinux - Section mismatch: reference to .init.text:find_unisys_acpi_oem_table from .text between 'acpi_madt_oem_check' (at offset 0xc0101eda) and 'enable_apic_mode'
WARNING: vmlinux - Section mismatch: reference to .init.text:acpi_get_table_header_early from .text between 'acpi_madt_oem_check' (at offset 0xc0101ef0) and 'enable_apic_mode'
WARNING: vmlinux - Section mismatch: reference to .init.text:parse_unisys_oem from .text between 'acpi_madt_oem_check' (at offset 0xc0101f2e) and 'enable_apic_mode'
WARNING: vmlinux - Section mismatch: reference to .init.text:setup_unisys from .text between 'acpi_madt_oem_check' (at offset 0xc0101f37) and 'enable_apic_mode'WARNING: vmlinux - Section mismatch: reference to .init.text:parse_unisys_oem from .text between 'mps_oem_check' (at offset 0xc0101ec7) and 'acpi_madt_oem_check'
WARNING: vmlinux - Section mismatch: reference to .init.text:es7000_sw_apic from .text between 'enable_apic_mode' (at offset 0xc0101f48) and 'check_apicid_present'
o Some functions which are inline (acpi_madt_oem_check) are not inlined by
compiler as these functions are accessed using function pointer. These
functions are put in .text section and they in-turn access __init type
functions hence modpost generates warnings.
o Do not iniline acpi_madt_oem_check, instead make it __init.
Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de> Cc: Andi Kleen <ak@suse.de> Cc: Len Brown <lenb@kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
[PATCH] x86-64: Set HASHDIST_DEFAULT to 1 for x86_64 NUMA
Enable system hashtable memory to be distributed among nodes on x86_64 NUMA
Forcing the kernel to use node interleaved vmalloc instead of bootmem for
the system hashtable memory (alloc_large_system_hash) reduces the memory
imbalance on node 0 by around 40MB on a 8 node x86_64 NUMA box:
Before the following patch, on bootup of a 8 node box:
I guess it is a good idea to keep HASHDIST_DEFAULT "on" for x86_64 NUMA
since x86_64 has no dearth of vmalloc space? Or maybe enable hash
distribution for all 64bit NUMA arches? The following patch does it only
for x86_64.
I ran a HPC MPI benchmark -- 'Ansys wingsolid', which takes up quite a bit of
memory and uses up tlb entries. This was on a 4 way, 2 socket
Tyan AMD box (non vsmp), with 8G total memory (4G pernode).
The results with and without hash distribution are:
1. Vanilla - runtime of 1188.000s
2. With hashdist=1 runtime of 1154.000s
Oprofile output for the duration of run is:
1. Vanilla:
PU: AMD64 processors, speed 2411.16 MHz (estimated)
Counted L1_AND_L2_DTLB_MISSES events (L1 and L2 DTLB misses) with a unit
mask of 0x00 (No unit mask) count 500
samples % app name symbol name
163054 6.5513 libansys1.so MultiFront::decompose(int, int,
Elemset *, int *, int, int, int)
162061 6.5114 libansys3.so blockSaxpy6L_fd
162042 6.5107 libansys3.so blockInnerProduct6L_fd
156286 6.2794 libansys3.so maxb33_
87879 3.5309 libansys1.so elmatrixmultpcg_
84857 3.4095 libansys4.so saxpy_pcg
58637 2.3560 libansys4.so .st4560
46612 1.8728 libansys4.so .st4282
43043 1.7294 vmlinux-t copy_user_generic_string
41326 1.6604 libansys3.so blockSaxpyBackSolve6L_fd
41288 1.6589 libansys3.so blockInnerProductBackSolve6L_fd
2. With hashdist=1
CPU: AMD64 processors, speed 2411.13 MHz (estimated)
Counted L1_AND_L2_DTLB_MISSES events (L1 and L2 DTLB misses) with a unit
mask of 0x00 (No unit mask) count 500
samples % app name symbol name
162993 6.9814 libansys1.so MultiFront::decompose(int, int,
Elemset *, int *, int, int, int)
160799 6.8874 libansys3.so blockInnerProduct6L_fd
160459 6.8729 libansys3.so blockSaxpy6L_fd
156018 6.6826 libansys3.so maxb33_
84700 3.6279 libansys4.so saxpy_pcg
83434 3.5737 libansys1.so elmatrixmultpcg_
58074 2.4875 libansys4.so .st4560
46000 1.9703 libansys4.so .st4282
41166 1.7632 libansys3.so blockSaxpyBackSolve6L_fd
41033 1.7575 libansys3.so blockInnerProductBackSolve6L_fd
35762 1.5318 libansys1.so inner_product_sub
35591 1.5245 libansys1.so inner_product_sub2
28259 1.2104 libansys4.so addVectors
Signed-off-by: Pravin B. Shelar <pravin.shelar@calsoftinc.com> Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org> Signed-off-by: Shai Fultheim <shai@scalex86.org> Signed-off-by: Andi Kleen <ak@suse.de> Acked-by: Christoph Lameter <clameter@engr.sgi.com> Cc: Andi Kleen <ak@suse.de> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
When compiling with -Os (which is default) the compiler defaults to it
anyways. And with -O2 it probably generates somewhat better (although
also larger) code.
Vivek Goyal [Wed, 2 May 2007 17:27:08 +0000 (19:27 +0200)]
[PATCH] x86-64: Move cpu verification code to common file
o This patch moves the code to verify long mode and SSE to a common file.
This code is now shared by trampoline.S, wakeup.S, boot/setup.S and
boot/compressed/head.S
o So far we used to do very limited check in trampoline.S, wakeup.S and
in 32bit entry point. Now all the entry paths are forced to do the
exhaustive check, including SSE because verify_cpu is shared.
o I am keeping this patch as last in the x86 relocatable series because
previous patches have got quite some amount of testing done and don't want
to distrub that. So that if there is problem introduced by this patch, at
least it can be easily isolated.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:08 +0000 (19:27 +0200)]
[PATCH] x86-64: Extend bzImage protocol for relocatable bzImage
o Extend the bzImage protocol (same as i386) to allow bzImage loaders to
load the protected mode kernel at non-1MB address. Now protected mode
component is relocatable and can be loaded at non-1MB addresses.
o As of today kdump uses it to run a second kernel from a reserved memory
area.
Vivek Goyal [Wed, 2 May 2007 17:27:08 +0000 (19:27 +0200)]
[PATCH] x86-64: build-time checking
o X86_64 kernel should run from 2MB aligned address for two reasons.
- Performance.
- For relocatable kernels, page tables are updated based on difference
between compile time address and load time physical address.
This difference should be multiple of 2MB as kernel text and data
is mapped using 2MB pages and PMD should be pointing to a 2MB
aligned address. Life is simpler if both compile time and load time
kernel addresses are 2MB aligned.
o Flag the error at compile time if one is trying to build a kernel which
does not meet alignment restrictions.
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: Relocatable Kernel Support
This patch modifies the x86_64 kernel so that it can be loaded and run
at any 2M aligned address, below 512G. The technique used is to
compile the decompressor with -fPIC and modify it so the decompressor
is fully relocatable. For the main kernel the page tables are
modified so the kernel remains at the same virtual address. In
addition a variable phys_base is kept that holds the physical address
the kernel is loaded at. __pa_symbol is modified to add that when
we take the address of a kernel symbol.
When loaded with a normal bootloader the decompressor will decompress
the kernel to 2M and it will run there. This both ensures the
relocation code is always working, and makes it easier to use 2M
pages for the kernel and the cpu.
AK: changed to not make RELOCATABLE default in Kconfig
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86: __pa and __pa_symbol address space separation
Currently __pa_symbol is for use with symbols in the kernel address
map and __pa is for use with pointers into the physical memory map.
But the code is implemented so you can usually interchange the two.
__pa which is much more common can be implemented much more cheaply
if it is it doesn't have to worry about any other kernel address
spaces. This is especially true with a relocatable kernel as
__pa_symbol needs to peform an extra variable read to resolve
the address.
There is a third macro that is added for the vsyscall data
__pa_vsymbol for finding the physical addesses of vsyscall pages.
Most of this patch is simply sorting through the references to
__pa or __pa_symbol and using the proper one. A little of
it is continuing to use a physical address when we have it
instead of recalculating it several times.
swapper_pgd is now NULL. leave_mm now uses init_mm.pgd
and init_mm.pgd is initialized at boot (instead of compile time)
to the physmem virtual mapping of init_level4_pgd. The
physical address changed.
Except for the for EMPTY_ZERO page all of the remaining references
to __pa_symbol appear to be during kernel initialization. So this
should reduce the cost of __pa in the common case, even on a relocated
kernel.
As this is technically a semantic change we need to be on the lookout
for anything I missed. But it works for me (tm).
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: do not use virt_to_page on kernel data address
o virt_to_page() call should be used on kernel linear addresses and not
on kernel text and data addresses. Swsusp code uses it on kernel data
(statically allocated swsusp_header).
o Allocate swsusp_header dynamically so that virt_to_page() can be used
safely.
o I am changing this because in next few patches, __pa() on x86_64 will
no longer support kernel text and data addresses and hibernation breaks.
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86: Move swsusp __pa() dependent code to arch portion
o __pa() should be used only on kernel linearly mapped virtual addresses
and not on kernel text and data addresses.
o Hibernation code needs to determine the physical address associated
with kernel symbol to mark a section boundary which contains pages which
don't have to be saved and restored during hibernate/resume operation.
o Move this piece of code in arch dependent section. So that architectures
which don't have kernel text/data mapped into kernel linearly mapped
region can come up with their own ways of determining physical addresses
associated with a kernel text.
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: Remove the identity mapping as early as possible
With the rewrite of the SMP trampoline and the early page
allocator there is nothing that needs identity mapped pages,
once we start executing C code.
So add zap_identity_mappings into head64.c and remove
zap_low_mappings() from much later in the code. The functions
are subtly different thus the name change.
This also kills boot_level4_pgt which was from an earlier
attempt to move the identity mappings as early as possible,
and is now no longer needed. Essentially I have replaced
boot_level4_pgt with trampoline_level4_pgt in trampoline.S
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: 64bit ACPI wakeup trampoline
o Moved wakeup_level4_pgt into the wakeup routine so we can
run the kernel above 4G.
o Now we first go to 64bit mode and continue to run from trampoline and
then then start accessing kernel symbols and restore processor context.
This enables us to resume even in relocatable kernel context when
kernel might not be loaded at physical addr it has been compiled for.
o Removed the need for modifying any existing kernel page table.
o Increased the size of the wakeup routine to 8K. This is required as
wake page tables are on trampoline itself and they got to be at 4K
boundary, hence one page is not sufficient.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: wakeup.S misc cleanups
o Various cleanups. One of the main purpose of cleanups is that make
wakeup.S as close as possible to trampoline.S.
o Following are the changes
- Indentations for comments.
- Changed the gdt table to compact form and to resemble the
one in trampoline.S
- Take the jump to 32bit from real mode using ljmpl. Makes code
more readable.
- After enabling long mode, directly take a long jump for 64bit
mode. No need to take an extra jump to "reach_comaptibility_mode"
- Stack is not used after real mode. So don't load stack in
32 bit mode.
- No need to enable PGE here.
- No need to do extra EFER read, anyway we trash the read contents.
- No need to enable system call (EFER_SCE). Anyway it will be
enabled when original EFER is restored.
- No need to set MP, ET, NE, WP, AM bits in cr0. Very soon we will
reload the original cr0 while restroing the processor state.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: Get rid of dead code in suspend resume
o Get rid of dead code in wakeup.S
o We never restore from saved_gdt, saved_idt, saved_ltd, saved_tss, saved_cr3,
saved_cr4, saved_cr0, real_save_gdt, saved_efer, saved_efer2. Get rid
of of associated code.
o Get rid of bogus_magic, bogus_31_magic and bogus_magic2. No longer being
used.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: 64bit PIC SMP trampoline
This modifies the SMP trampoline and all of the associated code so
it can jump to a 64bit kernel loaded at an arbitrary address.
The dependencies on having an idenetity mapped page in the kernel
page tables for SMP bootup have all been removed.
In addition the trampoline has been modified to verify
that long mode is supported. Asking if long mode is implemented is
down right silly but we have traditionally had some of these checks,
and they can't hurt anything. So when the totally ludicrous happens
we just might handle it correctly.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: Add EFER to the register set saved by save_processor_state
EFER varies like %cr4 depending on the cpu capabilities, and which cpu
capabilities we want to make use of. So save/restore it make certain
we have the same EFER value when we are done.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: cleanup segments
Move __KERNEL32_CS up into the unused gdt entry. __KERNEL32_CS is
used when entering the kernel so putting it first is useful when
trying to keep boot gdt sizes to a minimum.
Set the accessed bit on all gdt entries. We don't care
so there is no need for the cpu to burn the extra cycles,
and it potentially allows the pages to be immutable. Plus
it is confusing when debugging and your gdt entries mysteriously
change.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:07 +0000 (19:27 +0200)]
[PATCH] x86-64: modify copy_bootdata to use virtual addresses
Use virtual addresses instead of physical addresses
in copy bootdata. In addition fix the implementation
of the old bootloader convention. Everything is
at real_mode_data always. It is just that sometimes
real_mode_data was relocated by setup.S to not sit at
0x90000.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:06 +0000 (19:27 +0200)]
[PATCH] x86-64: Clean up the early boot page table
- Merge physmem_pgt and ident_pgt, removing physmem_pgt. The merge
is broken as soon as mm/init.c:init_memory_mapping is run.
- As physmem_pgt is gone don't export it in pgtable.h.
- Use defines from pgtable.h for page permissions.
- Fix the physical memory identity mapping so it is at the correct
address.
- Remove the physical memory mapping from wakeup_level4_pgt it
is at the wrong address so we can't possibly be usinging it.
- Simply NEXT_PAGE the work to calculate the phys_ alias
of the labels was very cool. Unfortuantely it was a brittle
special purpose hack that makes maitenance more difficult.
Instead just use label - __START_KERNEL_map like we do
everywhere else in assembly.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:06 +0000 (19:27 +0200)]
[PATCH] x86-64: Kill temp boot pmds
Early in the boot process we need the ability to set
up temporary mappings, before our normal mechanisms are
initialized. Currently this is used to map pages that
are part of the page tables we are building and pages
during the dmi scan.
The core problem is that we are using the user portion of
the page tables to implement this. Which means that while
this mechanism is active we cannot catch NULL pointer dereferences
and we deviate from the normal ways of handling things.
In this patch I modify early_ioremap to map pages into
the kernel portion of address space, roughly where
we will later put modules, and I make the discovery of
which addresses we can use dynamic which removes all
kinds of static limits and remove the dependencies
on implementation details between different parts of the code.
Now alloc_low_page() and unmap_low_page() use
early_iomap() and early_iounmap() to allocate/map and
unmap a page.
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
Vivek Goyal [Wed, 2 May 2007 17:27:06 +0000 (19:27 +0200)]
[PATCH] x86-64: Assembly safe page.h and pgtable.h
This patch makes pgtable.h and page.h safe to include
in assembly files like head.S. Allowing us to use
symbolic constants instead of hard coded numbers when
refering to the page tables.
This patch copies asm-sparc64/const.h to asm-x86_64 to
get a definition of _AC() a very convinient macro that
allows us to force the type when we are compiling the
code in C and to drop all of the type information when
we are using the constant in assembly. Previously this
was done with multiple definition of the same constant.
const.h was modified slightly so that it works when given
CONFIG options as arguments.
This patch adds #ifndef __ASSEMBLY__ ... #endif
and _AC(1,UL) where appropriate so the assembler won't
choke on the header files. Otherwise nothing
should have changed.
AK: added const.h to exported headers to fix headers_check
Signed-off-by: Eric W. Biederman <ebiederm@xmission.com> Signed-off-by: Vivek Goyal <vgoyal@in.ibm.com> Signed-off-by: Andi Kleen <ak@suse.de>
[PATCH] i386: Simplify smp_call_function*() by using common implementation
smp_call_function and smp_call_function_single are almost complete
duplicates of the same logic. This patch combines them by
implementing them in terms of the more general
smp_call_function_mask().
[PATCH] x86-64: Remove duplicated code for reading control registers
On Tue, Mar 13, 2007 at 05:33:09AM -0700, Randy.Dunlap wrote:
> On Tue, 13 Mar 2007, Glauber de Oliveira Costa wrote:
>
> > Tiny cleanup:
> >
> > In x86_64, the same functions for reading cr3 and writing cr{3,4} are
> > defined in tlbflush.h and system.h, whith just a name change.
> > The only difference is the clobbering of memory, which seems a safe, and
> > even needed change for the write_cr4. This patch removes the duplicate.
> > write_cr3() is moved to system.h for consistency.
>
> missing patch.....
>
thanks. Attached now
--
Glauber de Oliveira Costa
Red Hat Inc.
"Free as in Freedom"
Lasse Collin [Wed, 2 May 2007 17:27:06 +0000 (19:27 +0200)]
[PATCH] i386: Fix usage of -mtune when X86_GENERIC=y or CONFIG_MCORE2=y
Hi!
I sent this simple patch to lkml about two weeks ago and also cc'ed
to Linus, but seems that the patch got ignored. I decided to write to
you, because you have modified the relevant file most recently.
Below is a copy of the mail that is also available at
<http://lkml.org/lkml/2007/2/28/230>.
The reboot_fixups stuff seems to be a bit of a mess, specifically the
header is in linux/ when its a purely i386-specific piece of code. I'm
not sure why it has its config option; its only currently needed for
"geode-gx1/cs5530a", so perhaps whatever config option controls that
hardware should enable this?
Signed-off-by: Jeremy Fitzhardinge <jeremy@xensource.com> Signed-off-by: Andi Kleen <ak@suse.de>
Prarit Bhargava [Wed, 2 May 2007 17:27:05 +0000 (19:27 +0200)]
[PATCH] i386: Change sysenter_setup to __cpuinit & improve __INIT, __INITDATA
Change sysenter_setup to __cpuinit.
Change __INIT & __INITDATA to be cpu hotplug aware.
Resolve MODPOST warnings similar to:
WARNING: vmlinux - Section mismatch: reference to .init.text:sysenter_setup from
.text between 'identify_cpu' (at offset 0xc040a380) and 'detect_ht'
and
WARNING: vmlinux - Section mismatch: reference to .init.data:vsyscall_int80_end
from .text between 'sysenter_setup' (at offset 0xc041a269) and 'enable_sep_cpu'
WARNING: vmlinux - Section mismatch: reference to
.init.data:vsyscall_int80_start from .text between 'sysenter_setup' (at offset
0xc041a26e) and 'enable_sep_cpu'
WARNING: vmlinux - Section mismatch: reference to
.init.data:vsyscall_sysenter_end from .text between 'sysenter_setup' (at offset
0xc041a275) and 'enable_sep_cpu'
WARNING: vmlinux - Section mismatch: reference to
.init.data:vsyscall_sysenter_start from .text between 'sysenter_setup' (at
offset 0xc041a27a) and 'enable_sep_cpu'
[PATCH] x86-64: x86_64 make NMI use PERFCTR1 for architectural perfmon (take 2)
Hello,
This patch against 2.6.20-git14 makes the NMI watchdog use PERFSEL1/PERFCTR1
instead of PERFSEL0/PERFCTR0 on processors supporting Intel architectural
perfmon, such as Intel Core 2. Although all PMU events can work on
both counters, the Precise Event-Based Sampling (PEBS) requires that the
event be in PERFCTR0 to work correctly (see section 18.14.4.1 in the
IA32 SDM Vol 3b). This versions has 3 chunks compared to previous where
we had missed on check.
Changelog:
- make the x86-64 NMI watchdog use PERFSEL1/PERFCTR1 instead of PERFSEL0/PERFCTR0
on processors supporting the Intel architectural perfmon (e.g. Core 2 Duo).
This allows PEBS to work when the NMI watchdog is active.
[PATCH] i386: i386 make NMI use PERFCTR1 for architectural perfmon (take 2)
Hello,
This patch against 2.6.20-git14 makes the NMI watchdog use PERFSEL1/PERFCTR1
instead of PERFSEL0/PERFCTR0 on processors supporting Intel architectural
perfmon, such as Intel Core 2. Although all PMU events can work on
both counters, the Precise Event-Based Sampling (PEBS) requires that the
event be in PERFCTR0 to work correctly (see section 18.14.4.1 in the
IA32 SDM Vol 3b).
A similar patch for x86-64 is to follow.
Changelog:
- make the i386 NMI watchdog use PERFSEL1/PERFCTR1 instead of PERFSEL0/PERFCTR0
on processors supporting the Intel architectural perfmon (e.g. Core 2 Duo).
This allows PEBS to work when the NMI watchdog is active.
Andi Kleen [Wed, 2 May 2007 17:27:05 +0000 (19:27 +0200)]
[PATCH] x86: Fix i386 and x86_64 fault information pollution
a userspace fault or a kernelspace fault which will result in the
immediate death of the process. They should not be filled in as a
result of a kernelspace fault which can be fixed up.
Otherwise, if the process is handling SIGSEGV and examining the fault
information, this can result in the kernel space fault trashing the
previously stored fault information if it arrives between the
userspace fault happening and the SIGSEGV being delivered to the process.
Signed-off-by: Jeff Dike <jdike@addtoit.com> Signed-off-by: Andi Kleen <ak@suse.de> Acked-by: Jan Beulich <jbeulich@novell.com>
--
arch/i386/kernel/traps.c | 24 ++++++++++++++++++------
arch/x86_64/kernel/traps.c | 30 +++++++++++++++++++++++-------
2 files changed, 41 insertions(+), 13 deletions(-)
Rene Herman [Wed, 2 May 2007 17:27:05 +0000 (19:27 +0200)]
[PATCH] i386: probe_roms() cleanup
Remove the assumption that if the first page of a legacy ROM is mapped,
it'll all be mapped. This'll also stop people reading this code from
wondering if they're looking at a bug...
Signed-off-by: Rene Herman <rene.herman@gmail.com> Signed-off-by: Martin Murray <murrayma@citi.umich.edu> Signed-off-by: Andi Kleen <ak@suse.de> Cc: Andi Kleen <ak@suse.de> Cc: Zachary Amsden <zach@vmware.com> Cc: Jeremy Fitzhardinge <jeremy@goop.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Simon Arlott [Wed, 2 May 2007 17:27:05 +0000 (19:27 +0200)]
[PATCH] i386: Add an option for the VIA C7 which sets appropriate L1 cache
The VIA C7 is a 686 (with TSC) that supports MMX, SSE and SSE2, it also has
a cache line length of 64 according to
http://www.digit-life.com/articles2/cpu/rmma-via-c7.html. This patch sets
gcc to -march=686 and select s the correct cache shift.
Signed-off-by: Simon Arlott <simon@fire.lp0.eu> Signed-off-by: Andi Kleen <ak@suse.de> Cc: Andi Kleen <ak@suse.de> Cc: Dave Jones <davej@codemonkey.org.uk> Cc: Alan Cox <alan@lxorguk.ukuu.org.uk> Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
takada [Wed, 2 May 2007 17:27:05 +0000 (19:27 +0200)]
[PATCH] i386: pit_latch_buggy has no effect
Eliminated the arch/i386/kernel/timers in 2.6.18, use clocksoures instead.
pit_latch_buggy was referred in timers/timer_tsc.c, and currently removed.
Therefore nobody refer it.
Until 2.6.17, MediaGX's TSC works correctly. after 2.6.18, warned "TSC
appears to be running slowly. Marking it as unstable". So marked unstable
TSC when CS55x0.
Jan Beulich [Wed, 2 May 2007 17:27:05 +0000 (19:27 +0200)]
[PATCH] x86: consolidate smp_send_stop()
Synchronize i386's smp_send_stop() with x86-64's in only try-locking
the call lock to prevent deadlocks when called from panic().
In both version, disable interrupts before clearing the CPU off the
online map to eliminate races with IRQ handlers inspecting this map.
Also in both versions, save/restore interrupts rather than disabling/
enabling them.
On x86-64, eliminate one function used here by folding it into its
single caller, convert to static, and rename for consistency with i386
(lkcd may like this).
Signed-off-by: Jan Beulich <jbeulich@novell.com> Signed-off-by: Andi Kleen <ak@suse.de>
Jan Beulich [Wed, 2 May 2007 17:27:04 +0000 (19:27 +0200)]
[PATCH] x86: adjust inclusion of asm/fixmap.h
Move inclusion of asm/fixmap.h to where it is really used rather than
where it may have been used long ago (requires a few other adjustments
to includes due to previous implicit dependencies).
Signed-off-by: Jan Beulich <jbeulich@novell.com> Signed-off-by: Andi Kleen <ak@suse.de>
Jan Beulich [Wed, 2 May 2007 17:27:04 +0000 (19:27 +0200)]
[PATCH] i386: adjustments to page table dump during oops (v4)
- make the page table contents printing PAE capable
- make sure the address stored in current->thread.cr2 is unmodified
from what was read from CR2
- don't call oops_may_print() multiple times, when one time suffices
- print pte even in highpte case, as long as the pte page isn't in
actually in high memory (which is specifically the case for all page
tables covering kernel space)
(Changes to v3: Use sizeof()*2 rather than the suggested sizeof()*4 for
printing width, use fixed 16-nibble width for PAE, and also apply the
max_low_pfn range check to the middle level lookup on PAE.)
Signed-off-by: Jan Beulich <jbeulich@novell.com> Signed-off-by: Andi Kleen <ak@suse.de>
Ingo Molnar [Wed, 2 May 2007 17:27:04 +0000 (19:27 +0200)]
[PATCH] x86-64: always use physical delivery mode on > 8 CPUs
Remove clustered APIC mode. There's little point in the use of clustered APIC
mode, broadcasting is limited to within the cluster only, and chipsets have
bugs in this area as well. So default to physical APIC mode when the CPU
count is large, and default to logical APIC mode when the CPU count is 8 or
smaller.
(this patch only removes the use of genapic_cluster and cleans up the
resulting genapic.c file - removal of all remaining traces of clustered
mode will be done by another patch.)
Ingo Molnar [Wed, 2 May 2007 17:27:04 +0000 (19:27 +0200)]
[PATCH] x86-64: optimize & fix APIC mode setup
Fix a couple of inconsistencies/problems I found while reviewing the x86_64
genapic code (when I was chasing mysterious eth0 timeouts that would only
trigger if CPU_HOTPLUG is enabled):
- AMD systems defaulted to the slower flat-physical mode instead
of the flat-logical mode. The only restriction on AMD systems
is that they should not use clustered APIC mode.
- removed the CPU hotplug hacks, switching the default for small
systems back from phys-flat to logical-flat. The switching to logical
flat mode on small systems fixed sporadic ethernet driver timeouts i
was getting on a dual-core Athlon64 system:
NETDEV WATCHDOG: eth0: transmit timed out
eth0: Transmit timeout, status 0c 0005 c07f media 80.
eth0: Tx queue start entry 32 dirty entry 28.
eth0: Tx descriptor 0 is 0008a04a. (queue head)
eth0: Tx descriptor 1 is 0008a04a.
eth0: Tx descriptor 2 is 0008a04a.
eth0: Tx descriptor 3 is 0008a04a.
eth0: link up, 100Mbps, full-duplex, lpa 0xC5E1
- The use of '<= 8' was a bug by itself (the valid APIC ids
for logical flat mode go from 0 to 7, not 0 to 8). The new logic
is to use logical flat mode on both AMD and Intel systems, and
to only switch to physical mode when logical mode cannot be used.
If CPU hotplug is racy wrt. APIC shutdown then CPU hotplug needs
fixing, not the whole IRQ system be made inconsistent and slowed
down.
- minor cleanups: simplified some code constructs
build & booted on a couple of AMD and Intel SMP systems.
Linus Torvalds [Tue, 1 May 2007 00:43:48 +0000 (17:43 -0700)]
libata: honour host controllers that want just one host
The Marvell IDE interface on my machine would hit a BUG_ON() in
lib/iomem.c because it was calling ata_pci_init_one() specifying just a
single port on the host, but that would actually end up trying to
initialize two ports, the second one with bogus information.
This fixes "ata_pci_init_one()" so that it actually passes down the
n_ports variable that it got from the low-level driver to the host
allocation routine ("ata_host_alloc_pinfo()"), which results in the ATA
layer actually having the correct port number information.
And in order to make it all work, I also needed to fix a few places that
had incorrectly hard-coded the fact that a host always had exactly two
ports (both ata_pci_init_bmdma() and ata_request_legacy_irqs() would
just always iterate over both ports).
Acked-by: Jeff Garzik <jeff@garzik.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Johannes Berg [Mon, 30 Apr 2007 22:09:55 +0000 (15:09 -0700)]
power management: force pm_ops.valid callback to be assigned
This patch changes the docs and behaviour from "all states valid" to "no
states valid" if no .valid callback is assigned. Users of pm_ops that only
need mem sleep can assign pm_valid_only_mem without any overhead, others
will require more elaborate callbacks.
Now that all users of pm_ops have a .valid callback this is a safe thing to
do and prevents things from getting messy again as they were before.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Pavel Machek <pavel@ucw.cz> Looks-okay-to: Rafael J. Wysocki <rjw@sisk.pl> Cc: <linux-pm@lists.linux-foundation.org> Cc: Greg KH <greg@kroah.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Johannes Berg [Mon, 30 Apr 2007 22:09:54 +0000 (15:09 -0700)]
power management: implement pm_ops.valid for everybody
Almost all users of pm_ops only support mem sleep, don't check in .valid and
don't reject any others in .prepare so users can be confused if they check
/sys/power/state, especially when new states are added (these would then
result in s-t-r although they're supposed to be something different).
This patch implements a generic pm_valid_only_mem function that is then
exported for users and puts it to use in almost all existing pm_ops.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: David Brownell <david-b@pacbell.net> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: linux-pm@lists.linux-foundation.org Cc: Len Brown <lenb@kernel.org> Acked-by: Russell King <rmk@arm.linux.org.uk> Cc: Greg KH <greg@kroah.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Johannes Berg [Mon, 30 Apr 2007 22:09:53 +0000 (15:09 -0700)]
power management: remove firmware disk mode
This patch removes the firmware disk suspend mode which is the wrong approach,
it is supposed to be used for implementing firmware-based disk suspend but
cannot actually be used for that.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: <linux-pm@lists.linux-foundation.org> Cc: David Brownell <david-b@pacbell.net> Cc: Len Brown <lenb@kernel.org> Acked-by: Russell King <rmk@arm.linux.org.uk> Cc: Greg KH <greg@kroah.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Cc: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Johannes Berg [Mon, 30 Apr 2007 22:09:51 +0000 (15:09 -0700)]
rework pm_ops pm_disk_mode, kill misuse
This patch series cleans up some misconceptions about pm_ops. Some users of
the pm_ops structure attempt to use it to stop the user from entering suspend
to disk, this, however, is not possible since the user can always use
"shutdown" in /sys/power/disk and then the pm_ops are never invoked. Also,
platforms that don't support suspend to disk simply should not allow
configuring SOFTWARE_SUSPEND (read the help text on it, it only selects
suspend to disk and nothing else, all the other stuff depends on PM).
The pm_ops structure is actually intended to provide a way to enter
platform-defined sleep states (currently supported states are "standby" and
"mem" (suspend to ram)) and additionally (if SOFTWARE_SUSPEND is configured)
allows a platform to support a platform specific way to enter low-power mode
once everything has been saved to disk. This is currently only used by ACPI
(S4).
This patch:
The pm_ops.pm_disk_mode is used in totally bogus ways since nobody really
seems to understand what it actually does.
This patch clarifies the pm_disk_mode description.
It also removes all the arm and sh users that think they can veto suspend to
disk via pm_ops; not so since the user can always do echo shutdown >
/sys/power/disk, they need to find a better way involving Kconfig or such.
ACPI is the only user left with a non-zero pm_disk_mode.
The patch also sets the default mode to shutdown again, but when a new pm_ops
is registered its pm_disk_mode is selected as default, that way the default
stays for ACPI where it is apparently required.
Signed-off-by: Johannes Berg <johannes@sipsolutions.net> Cc: David Brownell <david-b@pacbell.net> Acked-by: Pavel Machek <pavel@ucw.cz> Cc: <linux-pm@lists.linux-foundation.org> Cc: Len Brown <lenb@kernel.org> Acked-by: Russell King <rmk@arm.linux.org.uk> Cc: Greg KH <greg@kroah.com> Cc: "Rafael J. Wysocki" <rjw@sisk.pl> Acked-by: Paul Mundt <lethal@linux-sh.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Robert Peterson [Mon, 30 Apr 2007 22:09:48 +0000 (15:09 -0700)]
Extend print_symbol capability
Today's print_symbol function dumps a kernel symbol with printk. This
patch extends the functionality of kallsyms.c so that the symbol lookup
function may be used without the printk. This is useful for modules that
want to dump symbols elsewhere, for example, to debugfs. I intend to use
the new function call in the GFS2 file system (which will be a separate
patch).
[akpm@linux-foundation.org: build fix]
[clameter@sgi.com: sprint_symbol should return length of string like sprintf] Signed-off-by: Robert Peterson <rpeterso@redhat.com> Cc: Rusty Russell <rusty@rustcorp.com.au> Cc: Roman Zippel <zippel@linux-m68k.org> Cc: "Randy.Dunlap" <rdunlap@xenotime.net> Cc: Sam Ravnborg <sam@ravnborg.org> Acked-by: Paulo Marques <pmarques@grupopie.com> Signed-off-by: Christoph Lameter <clameter@sgi.com> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
David S. Miller [Mon, 30 Apr 2007 21:51:58 +0000 (14:51 -0700)]
[UDP]: Do not allow specific bind when wildcard bind exists.
When allocating local ports, do not allow a bind to a port
with a specific local address when a bind to that port with
a wildcard local address already exists.
Noticed by Linus.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Mon, 30 Apr 2007 20:35:29 +0000 (13:35 -0700)]
[IPV4] UDP: Fix endianness bugs in hashing changes.
I accidently applied an earlier version of Eric Dumazet's patch, from
March 21st. His version from March 30th didn't have these bugs, so
this just interdiffs to the correct patch.
Signed-off-by: David S. Miller <davem@davemloft.net>
Merge branch 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6
* 'for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/ieee1394/linux1394-2.6: (56 commits)
ieee1394: remove garbage from Kconfig
ieee1394: more help in Kconfig
ieee1394: ohci1394: Fix mistake in printk message.
ieee1394: ohci1394: remove unnecessary rcvPhyPkt bit flipping in LinkControl register
ieee1394: ohci1394: fix cosmetic problem in error logging
ieee1394: eth1394: send async streams at S100 on 1394b buses
ieee1394: eth1394: fix error path in module_init
ieee1394: eth1394: correct return codes in hard_start_xmit
ieee1394: eth1394: hard_start_xmit is called in atomic context
ieee1394: eth1394: some conditions are unlikely
ieee1394: eth1394: clean up fragment_overlap
ieee1394: eth1394: don't use alloc_etherdev
ieee1394: eth1394: omit useless set_mac_address callback
ieee1394: eth1394: CONFIG_INET is always defined
ieee1394: eth1394: allow MTU bigger than 1500
ieee1394: unexport highlevel_host_reset
ieee1394: eth1394: contain host reset
ieee1394: eth1394: shorter error messages
ieee1394: eth1394: correct a memset argument
ieee1394: eth1394: refactor .probe and .update
...
Merge branch 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jikos/hid
* 'for-linus' of master.kernel.org:/pub/scm/linux/kernel/git/jikos/hid: (21 commits)
USB HID: don't warn on idVendor == 0
USB HID: add 'quirks' module parameter
USB HID: add support for dynamically-created quirks
USB HID: clarify static quirk handling as squirks
USB HID: encapsulate quirk handling into hid-quirks.c
USB HID: EMS USBII device needs HID_QUIRK_MULTI_INPUT
HID: update copyright and authorship macro
HID: introduce proper zeroing of unused bits in output reports
USB HID: add support for WiseGroup MP-8800 Quad Joypad
USB HID: add FF support for Logitech Force 3D Pro Joystick
USB HID: numlock quirk for dell W7658 keyboard
USB HID: Logitech MX3000 keyboard needs report descriptor quirk
USB HID: extend quirk for Logitech S510 keyboard
USB HID: usbkbd/usbmouse - handle errors when registering devices
USB HID: add QUIRK_HIDDEV for Belkin Flip KVM
HID: enable dead keys on a belkin wireless keyboard
USB HID: Thustmaster firestorm dual power v1 support
USB HID: specify explicit size for hid_blacklist.quirks
USB HID: fix retry & reset logic
USB HID: consolidate vendor/product ids
...
* master.kernel.org:/pub/scm/linux/kernel/git/davem/net-2.6: (21 commits)
[IPV4] SNMP: Support OutMcastPkts and OutBcastPkts
[IPV4] SNMP: Support InMcastPkts and InBcastPkts
[IPV4] SNMP: Support InTruncatedPkts
[IPV4] SNMP: Support InNoRoutes
[SNMP]: Add definitions for {In,Out}BcastPkts
[TCP] FRTO: RFC4138 allows Nagle override when new data must be sent
[TCP] FRTO: Delay skb available check until it's mandatory
[XFRM]: Restrict upper layer information by bundle.
[TCP]: Catch skb with S+L bugs earlier
[PATCH] INET : IPV4 UDP lookups converted to a 2 pass algo
[L2TP]: Add the ability to autoload a pppox protocol module.
[SKB]: Introduce skb_queue_walk_safe()
[AF_IUCV/IUCV]: smp_call_function deadlock
[IPV6]: Fix slab corruption running ip6sic
[TCP]: Update references in two old comments
[XFRM]: Export SPD info
[IPV6]: Track device renames in snmp6.
[SCTP]: Fix sctp_getsockopt_local_addrs_old() to use local storage.
[NET]: Remove NETIF_F_INTERNAL_STATS, default to internal stats.
[NETPOLL]: Remove CONFIG_NETPOLL_RX
...
Merge branch 'for-linus' of git://git.kernel.dk/data/git/linux-2.6-block
* 'for-linus' of git://git.kernel.dk/data/git/linux-2.6-block:
[PATCH] elevator: elv_list_lock does not need irq disabling
[BLOCK] Don't pin lots of memory in mempools
cfq-iosched: speedup cic rb lookup
ll_rw_blk: add io_context private pointer
cfq-iosched: get rid of cfqq hash
cfq-iosched: tighten queue request overlap condition
cfq-iosched: improve sync vs async workloads
cfq-iosched: never allow an async queue idling
cfq-iosched: get rid of ->dispatch_slice
cfq-iosched: don't pass unused preemption variable around
cfq-iosched: get rid of ->cur_rr and ->cfq_list
cfq-iosched: slice offset should take ioprio into account
[PATCH] cfq-iosched: style cleanups and comments
cfq-iosched: sort IDLE queues into the rbtree
cfq-iosched: sort RT queues into the rbtree
[PATCH] cfq-iosched: speed up rbtree handling
cfq-iosched: rework the whole round-robin list concept
cfq-iosched: minor updates
cfq-iosched: development update
cfq-iosched: improve preemption for cooperating tasks
[IPV4] SNMP: Support OutMcastPkts and OutBcastPkts
A transmitted IP multicast datagram should be counted as OutMcastPkts.
By the same token, a transmitted IP broadcast datagram should be
counted as OutBcastPkts.
Signed-off-by: Mitsuru Chinen <mitch@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
A received IP multicast datagram should be counted as InMcastPkts.
By the same token, a received IP broadcast datagram should be
counted as InBcastPkts.
Signed-off-by: Mitsuru Chinen <mitch@linux.vnet.ibm.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Ilpo Järvinen [Mon, 30 Apr 2007 07:42:20 +0000 (00:42 -0700)]
[TCP] FRTO: RFC4138 allows Nagle override when new data must be sent
This is a corner case where less than MSS sized new data thingie
is awaiting in the send queue. For F-RTO to work correctly, a
new data segment must be sent at certain point or F-RTO cannot
be used at all. RFC4138 allows overriding of Nagle at that
point.
Implementation uses frto_counter states 2 and 3 to distinguish
when Nagle override is needed.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by: David S. Miller <davem@davemloft.net>
[XFRM]: Restrict upper layer information by bundle.
On MIPv6 usage, XFRM sub policy is enabled.
When main (IPsec) and sub (MIPv6) policy selectors have the same
address set but different upper layer information (i.e. protocol
number and its ports or type/code), multiple bundle should be created.
However, currently we have issue to use the same bundle created for
the first time with all flows covered by the case.
It is useful for the bundle to have the upper layer information
to be restructured correctly if it does not match with the flow.
1. Bundle was created by two policies
Selector from another policy is added to xfrm_dst.
If the flow does not match the selector, it goes to slow path to
restructure new bundle by single policy.
2. Bundle was created by one policy
Flow cache is added to xfrm_dst as originated one. If the flow does
not match the cache, it goes to slow path to try searching another
policy.
Signed-off-by: Masahide NAKAMURA <nakam@linux-ipv6.org> Signed-off-by: David S. Miller <davem@davemloft.net>
Ilpo Järvinen [Mon, 30 Apr 2007 07:57:33 +0000 (00:57 -0700)]
[TCP]: Catch skb with S+L bugs earlier
SACKED_ACKED and LOST are mutually exclusive with SACK, thus
having their sum larger than packets_out is bug with SACK.
Eventually these bugs trigger traps in the tcp_clean_rtx_queue
with SACK but it's much more informative to do this here.
Non-SACK TCP, however, could get more than packets_out duplicate
ACKs which each increment sacked_out, so it makes sense to do
this kind of limitting for non-SACK TCP but not for SACK enabled
one. Perhaps the author had the opposite in mind but did the
logic accidently wrong way around? Anyway, the sacked_out
incrementer code for non-SACK already deals this issue before
calling sync_left_out so this trapping can be done
unconditionally.
Signed-off-by: Ilpo Järvinen <ilpo.jarvinen@helsinki.fi> Signed-off-by: David S. Miller <davem@davemloft.net>
Eric Dumazet [Mon, 30 Apr 2007 07:26:00 +0000 (00:26 -0700)]
[PATCH] INET : IPV4 UDP lookups converted to a 2 pass algo
Some people want to have many UDP sockets, binded to a single port but
many different addresses. We currently hash all those sockets into a
single chain. Processing of incoming packets is very expensive,
because the whole chain must be examined to find the best match.
I chose in this patch to hash UDP sockets with a hash function that
take into account both their port number and address : This has a
drawback because we need two lookups : one with a given address, one
with a wildcard (null) address.
Signed-off-by: Eric Dumazet <dada1@cosmosbay.com> Signed-off-by: David S. Miller <davem@davemloft.net>
Currently we scale the mempool sizes depending on memory installed
in the machine, except for the bio pool itself which sits at a fixed
256 entry pre-allocation.
There's really no point in "optimizing" this OOM path, we just need
enough preallocated to make progress. A single unit is enough, lets
scale it down to 2 just to be on the safe side.
This patch saves ~150kb of pinned kernel memory on a 32-bit box.
cfq hash is no more necessary. We always can get cfqq from io context.
cfq_get_io_context_noalloc() function is introduced, because we don't
want to allocate cic on merging and checking may_queue. In order to
identify sync queue we've used hash key = CFQ_KEY_ASYNC. Since hash is
eliminated we need to use other criterion: sync flag for queue is added.
In all places where we dig in rb_tree we're in current context, so no
additional locking is required.
Advantages of this patch: no additional memory for hash, no seeking in
hash, code is cleaner. But it is necessary now to seek cic in per-ioc
rbtree, but it is faster:
- most processes work only with few devices
- most systems have only few block devices
- it is a rb-tree
Signed-off-by: Vasily Tarasov <vtaras@openvz.org>
Changes by me:
- Merge into CFQ devel branch
- Get rid of cfq_get_io_context_noalloc()
- Fix various bugs with dereferencing cic->cfqq[] with offset other
than 0 or 1.
- Fix bug in cfqq setup, is_sync condition was reversed.
- Fix bug where only bio_sync() is used, we need to check for a READ too
projects / linux-2.6 / log