Commit Graph

95 Commits

Author SHA1 Message Date
Wolfgang Wander 1363c3cd86 [PATCH] Avoiding mmap fragmentation
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>
2005-06-21 18:46:16 -07:00
Linus Torvalds 1d345dac1f Merge master.kernel.org:/pub/scm/linux/kernel/git/gregkh/driver-2.6 2005-06-20 16:00:33 -07:00
Yani Ioannou ff381d2223 [PATCH] Driver Core: arch: update device attribute callbacks
Signed-off-by: Yani Ioannou <yani.ioannou@gmail.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
2005-06-20 15:15:32 -07:00
Lennert Buytenhek f0ffeddc89 [PATCH] ARM: 2719/1: enable module support in ixp2000 defconfigs by default
Patch from Lennert Buytenhek

The ixp2000 defconfigs are among the few that do not enable module
support by default.  I keep enabling module support by hand for every
new kernel version, so let's just make this change upstream.

Signed-off-by: Lennert Buytenhek
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 18:51:08 +01:00
Richard Purdie d67947a1bd [PATCH] ARM: 2716/1: SharpSL Param: Fix typo
Patch from Richard Purdie

Fix typo in sharpsl_param.c so it works correctly on collie.

Signed-off-by: Richard Purdie
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 18:51:07 +01:00
Lennert Buytenhek e4fe19819e [PATCH] ARM: 2701/1: free up ixp2000 timer 4 for the watchdog
Patch from Lennert Buytenhek

The IXP2000 has four timers, but if we're on an A-step IXP2800, timer
2 and 3 don't work.  We need two timers for timekeeping (one for the
timer interrupt and one for tracking missed jiffies), so on early
IXP2800s we have no other choice but to use timer 1 and 4 for that,
but on all other IXP2000s we'd rather leave timer 4 free since that's
the only timer we can use for the watchdog.
So, on buggy IXP2000s (i.e. the A-step IXP2800) we use timer 4 for
tracking missed jiffies, and on all all non-buggy IXP2000s (i.e.
everything but the A-step IXP2800) we use timer 2.
On a pre-production IXP2800, this patch should print these messages
on boot:
	Enabling IXP2800 erratum #25 workaround
	Unable to use IXP2000 watchdog due to IXP2800 erratum #25
On any non-buggy IXP2800 (as well as on IXP2400s) you shouldn't see
anything at all, and the watchdog should be usable again.

Signed-off-by: Lennert Buytenhek
Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 18:51:07 +01:00
Catalin Marinas c0da085ad2 [PATCH] ARM: 2693/1: Add PCI support for Versatile/PB
Patch from Catalin Marinas

This patch adds PCI support for the Versatile PB926 platform.

Signed-off-by: Colin King
Signed-off-by: Catalin Marinas
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 18:51:06 +01:00
Bellido Nicolas 038c5b6025 [PATCH] ARM: 2686/2: AAEC-2000 Core support
Patch from Bellido Nicolas

Core support for AAEC-2000 based platforms.
This is an updated version of the previous patch, and takes
into account Russell's comments.
AAED-2000 default configuration will follow as soon
as some problems with the bootloader are sorted out...

Signed-off-by: Nicolas Bellido
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 18:51:05 +01:00
Russell King 09f0551d20 [PATCH] ARM: Add iomap support for ARM
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 18:44:37 +01:00
Russell King a507ef3ac6 [PATCH] ARM: Remove nmi_tick() from Integrator.
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 16:57:17 +01:00
Russell King 14eb75b6f8 [PATCH] ARM: Add missed include for dmabounce.c
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 16:56:08 +01:00
Russell King 3ade2fe0fd [PATCH] ARM: Lindent GCC helper functions
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 16:45:32 +01:00
Russell King f29481c0e7 [PATCH] ARM: Remove gcc type-isms from GCC helper functions
Convert ugly GCC types to Linux types:

	UQImode -> u8
	SImode -> s32
	USImode -> u32
	DImode -> s64
	UDImode -> u64
	word_type -> int

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 15:49:59 +01:00
Russell King 34c8eacab6 [PATCH] ARM: Remove obsolete arch/arm/kernel/arch.c
This is not used anymore - RiscPC now contains the necessary
supporting code.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 12:56:40 +01:00
Russell King 5abc100e88 [PATCH] ARM: Ensure DMA-bounced buffers are properly written to RAM
When DMA bounce buffers were unmapped and the data was memcpy'd to
the original buffer, we were not ensuring that the data was written
to RAM.  This means that there was the potential for page cache
pages to have different cache states depending whether they've been
bounced or not.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 12:31:14 +01:00
Russell King b8a9b66fbe [PATCH] ARM: Add common CACHE_COLOUR macro
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 11:31:09 +01:00
Russell King 8830f04a09 [PATCH] ARM: Fix delayed dcache flush for ARMv6 non-aliasing caches
flush_dcache_page() did nothing for these caches, but since they
suffer from I/D cache coherency issues, we need to ensure that data
is written back to RAM.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-20 09:51:03 +01:00
Russell King 0908db22b1 [PATCH] ARM SMP: Messages about CPUs should be prefixed by CPU%u
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-19 19:48:16 +01:00
Russell King ea4423c3b6 Merge with ../linux-2.6-smp 2005-06-19 19:26:54 +01:00
Russell King 36c5ed23b9 [PATCH] ARM SMP: Fix PXA/SA11x0 suspend resume crash
We need to re-initialise the stack pointers for undefined, IRQ
and abort mode handlers whenever we resume.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-19 18:39:33 +01:00
Russell King fe6ef2daa2 [PATCH] ARM SMP: Add missed files from Integrator/CP platform
Add missed new files from basic SMP support for the Integrator/CP platform.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-19 09:52:07 +01:00
Russell King 20cf33ea16 [PATCH] ARM SMP: Add basic support Integrator/CP platform
Add basic SMP support for the Integrator/CP platform.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-18 10:15:46 +01:00
Russell King e65f38ed0b [PATCH] ARM SMP: Add support for startup of secondary processors
Create a temporary page table to startup secondary processors.  This
page table must have a 1:1 virtual/physical mapping for the kernel
in addition to the standard mappings to ensure that the secondary
CPU can enable its MMU safely.

Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-18 09:33:31 +01:00
Russell King 5ab6091db0 Merge with ../linux-2.6-smp 2005-06-18 09:06:59 +01:00
Nicolas Pitre 22f11c4e66 [PATCH] ARM: 2715/1: restore CPLD interrupts upon resume for Lubbock and Mainstone
Patch from Nicolas Pitre

Without this some devices fail to work again after a suspend event.

Signed-off-by: Nicolas Pitre
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-16 21:23:56 +01:00
Catalin Marinas fea7722fd7 [PATCH] ARM: 2713/1: Fix the GPIO base for Integrator/CP
Patch from Catalin Marinas

The GPIO base for Integrator/CP is different from the
Integrator/AP. This patch sets the correct value for
INTEGRATOR_GPIO_BASE.

Signed-off-by: Catalin Marinas
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-16 18:01:11 +01:00
Catalin Marinas 90ef713b63 [PATCH] ARM: 2712/1: Fix the RGB order for the Versatile CLCD
Patch from Catalin Marinas

The current red and blue colours on the Versatile CLCD are
reversed when the 5:6:5 mode is used. The patch sets the proper
bit in the SYS_CLCD register value.

Signed-off-by: Catalin Marinas
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-16 18:01:11 +01:00
Nicolas Pitre a8fa3f0c59 [PATCH] ARM: 2711/1: fix compilation on PXA targets with CONFIG_PM=n
Patch from Nicolas Pitre

Signed-off-by: Nicolas Pitre
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-13 22:35:41 +01:00
David Brownell bb011b8e8e [PATCH] ARM: 2709/1: Systems with PCMCIA should also see IDE options (for CompactFlash memories)
Patch from David Brownell

The ARM generic Kconfig filters out IDE options ... except for
an error prone ARMload of special cases.
This adds one general case to the systems that will offer IDE options:
kernels with PCMCIA support, which probably want to use IDE to access
CompactFlash cards.  This might allow many (most?) of the other cases
to disappear, for systems that only see IDE hardware through CF cards.
Right now this one patch is used to gate access to CF cards, including
MicroDrives, for both omap_cf and at91_cf drivers.

Signed-off-by: David Brownell
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-12 23:26:05 +01:00
Vincent Sanders 07c6d48fd1 [PATCH] ARM: 2708/1: Fix hackkit CPU Frequency build faliure
Patch from Vincent Sanders

This fixes the "multiple definitions of cpufreq_get" build faliure on
the hackkit SA1100 platform.

Signed-off-by: Vincent Sanders
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-09 21:59:22 +01:00
Vincent Sanders 7aa0b0d5ab [PATCH] ARM: 2707/2: Fix badge4 CPU Frequency build faliure
Patch from Vincent Sanders

This fixes the "multiple definitions of cpufreq_get" build faliure on
the Badge4 SA1100 platform.

Signed-off-by: Vincent Sanders
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-09 21:59:21 +01:00
Dave Neuer 12035d6456 [PATCH] ARM: 2706/1: Fix compile on SA-based iPAQs and remove stale CREDITS info
Patch from Dave Neuer

This fixes the "multiple definitions of cpufreq_get" errors on
StrongARM-based iPAQs.

Signed-off-by: Dave Neuer
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-09 17:40:55 +01:00
Russell King 4e71e47da3 [PATCH] ARM: Remove zero-byte sized file
Remove the remaining zero byte file left over from the Xscale
fixes.

Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-06-09 16:53:28 +01:00
Nicolas Pitre dcef1f6346 [PATCH] ARM: 2664/2: add support for atomic ops on pre-ARMv6 SMP systems
Patch from Nicolas Pitre

Not that there might be many of them on the planet, but at least RMK
apparently has one.

Signed-off-by: Nicolas Pitre
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-08 19:00:47 +01:00
Nicolas Pitre aeabbbbe12 [PATCH] ARM: 2705/1: fix writesw for misaligned source pointer
Patch from Nicolas Pitre

Signed-off-by: Nicolas Pitre
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-08 19:00:16 +01:00
Russell King f8f98a9335 [PATCH] ARM: Fix Xscale copy_page implementation
The ARM copypage changes in 2.6.12-rc4-git1 removed the preempt locking
from the copypage functions which broke the XScale implementation.
This patch fixes the locking on XScale and removes the now unneeded
minicache code.

Signed-off-by: Russell King <rmk@arm.linux.org.uk>
Checked-by: Richard Purdie
2005-06-08 15:28:24 +01:00
Deepak Saxena 17d82fcc6a [PATCH] ARM: 2700/1: Disable IXP2000 IRQs at bootup
Patch from Deepak Saxena

The IXDP2800 bootloader does not disable IRQs before jumping into
the kernel and this is causing the Grand Unified KGDB to crash
the system when we do an early call to trap_init() and irq handlers
have not yet been registered. This patch disables IRQs before we
jump into the kernel.

Signed-off-by: Deepak Saxena
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-03 22:18:52 +01:00
Todd Poynor 8775420d2f [PATCH] ARM: 2691/1: PXA27x sleep fixes take 2
Patch from Todd Poynor

PXA27x sleep fixes:
* set additional sleep/wakeup registers for Mainstone boards.
* move CKEN=0 to pxa25x-specific code; that value is harmful on pxa27x.
* save/restore additional registers, including some found necessary for
C5 processors and/or newer blob versions.
* enable future support of additional sleep modes for PXA27x (eg,
standby, deep sleep).
* split off cpu-specific sleep processing between pxa27x and pxa25x into
separate files (partly in preparation for additional sleep modes).
Includes fixes from David Burrage.

Signed-off-by: Todd Poynor
Signed-off-by: Nicolas Pitre
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-03 20:52:27 +01:00
Albrecht Dreß 105bb2698b [PATCH] ARM: 2694/1: [s3c2410/dma] release irq properly to fix kernel oops
Patch from Albrecht Dreß

Problem:
When a module requests a DMA channel via the function s3c2410_dma_request(), this function requests the appropriate irq under the name of the client module. When the client module is unloaded, it calls s3c2410_dma_free() which does not free the irq. Consequently, when e.g. running "cat /proc/interrupts", the irq owner points to freed memory, leading to a kernel oops.
File:
linux/arch/arm/mach-s3c2410/dma.c
Fix:
trivial, below

Signed-off-by: Albrecht Dreß
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-06-03 20:52:26 +01:00
Russell King ccea7a19e5 [PATCH] ARM SMP: Fix vector entry
The current vector entry system does not allow for SMP.  In
order to work around this, we need to eliminate our reliance
on the fixed save areas, which breaks the way we enable
alignment traps.  This patch changes the way we handle the
save areas such that we can have one per CPU.

Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-05-31 22:22:32 +01:00
Russell King 49f680ea7b [PATCH] ARM SMP: convert alignment enable
The current vector entry system does not allow for SMP.  In
order to work around this, we need to eliminate our reliance
on the fixed save areas, which breaks the way we enable
alignment traps.  This patch makes the alignment trap enable
code independent of the way we handle the save areas.

Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-05-31 18:02:00 +01:00
Russell King 791be9b976 [PATCH] ARM SMP: add IPI support
Add support for inter-processor interrupts to the main IRQ
handling code.

Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-05-21 18:16:44 +01:00
Russell King 706fdd9faa [PATCH] ARM SMP: reallocate main IRQ handler code registers
By changing r9 -> r8 and r8 to 'tsk' (r9) we are able to remove
one instruction from the preempt path.

Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-05-21 18:15:45 +01:00
Russell King 187a51ad11 [PATCH] ARM SMP: consolidate main IRQ handler code
Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-05-21 18:14:44 +01:00
Russell King 8711a1b902 [PATCH] ARM: Fix build error
Mainline kernels don't have VECTORS_HIGH nor COPYPAGE_MINICACHE yet.

Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-05-16 23:36:22 +01:00
Ben Dooks bfd4e0709f [PATCH] ARM: 2678/1: S3C2440 - cpu fixes, hdiv divisors and nand dev name
Patch from Ben Dooks

Fix the setting of hdiv when set to divide-by-2. Thanks to
Jeonghoon Yoon for pointing this out.
Change name of the NAND device to "s3c2440-nand" as it
is not similar enough to the "s3c2410-nand" device.

Signed-off-by: Ben Dooks
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-05-12 19:27:14 +01:00
Ben Dooks 4ad3a443c9 [PATCH] ARM: 2677/1: S3C2440 - UPLL frequency doubled
Patch from Ben Dooks

S3C2440 UPLL is the same as the S3C2410 UPLL, it is only the
MPLL which has an extra multiplication factor of 2 in the
multiplier.

Signed-off-by: Ben Dooks
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-05-12 19:27:13 +01:00
Nicolas Pitre 70489c88d0 [PATCH] ARM: 2680/1: refine TLS reg availability some more again
Patch from Nicolas Pitre

Not all ARMv6 processors implement the TLS register.

Signed-off-by: Nicolas Pitre
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-05-12 19:27:12 +01:00
Nicolas Pitre fa4354359f [PATCH] ARM: 2663/2: I can't type
Patch from Nicolas Pitre

Signed-off-by: Nicolas Pitre
Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
2005-05-10 17:36:29 +01:00
Russell King 8d802d28c2 [PATCH] ARM: Add V6 aliasing cache flush
Add cache flushing support for aliased V6 caches to
flush_dcache_page.

Signed-off-by: Russell King <rmk@arm.linux.org.uk>
2005-05-10 17:31:43 +01:00