Skip to content

L
linux-davinci-2.6.23
Switch branch/tag
  1. 01 Nov, 2005 10 commits
    • Nicolas Pitre's avatar
      [ARM] 2947/1: copy template with new memcpy/memmove · 75494230
      Nicolas Pitre authored 
      Patch from Nicolas Pitre

This patch provides a new implementation for optimized memory copy
functions on ARM.  It is made of two levels: a template that consists of
the core copy code and separate files that define macros to be used with
the core code depending on the type of copy needed. This allows for best
performances while sharing the same core for implementing memcpy(),
copy_from_user() and copy_to_user() for instance.

Two reasons for this work:

1) the current copy_to_user/copy_from_user implementation assumes no
   task switch will ever occur in the middle of each copied page making
   it completely unsafe with CONFIG_PREEMPT=y.

2) current copy implementations are measurably suboptimal and optimizing
   different implementations separately is a pain and more opportunities
   for bugs.

The reason for (1) is the fact that copy inside user pages are performed
with the ldm instruction which has no mean for testing user protections
and could possibly race with process preemption bypassing the COW mechanism
for example.  This is a longstanding issue that we said ought to be fixed
for about two years now.  The solution is to substitute those ldm insns
with a series of ldrt or strt insns to enforce user memory protection.
At least on StrongARM and XScale cores the ldm is not faster than the
equivalent ldr/str insns with a warm i-cache so there is no measurable
performance degradation with that change. The fact that the copy code is
a template makes it pretty easy to reuse the same core code as for memcpy
and benefit from the same performance optimizations.

Now (2) is best demonstrated with actual throughput measurements.
First, here is a summary of memcopy tests performed on a StrongARM core:

	PTR alignment	buffer size	kernel version	this version
	------------------------------------------------------------
	  aligned	     32		 59.73		107.43
	unaligned	     32		 61.31		 74.72
	  aligned	    100		132.47		136.15
	unaligned	    100	    	103.84		123.76
	  aligned	   4096		130.67		130.80
	unaligned	   4096	    	130.68		130.64
	  aligned	1048576		 68.03		68.18
	unaligned	1048576		 68.03		68.18

The buffer size is in bytes and the measured speed in MB/s.  The copy
was performed repeatedly with given buffer and throughput averaged over
3 seconds.

Here we can see that the current kernel version has a higher entry cost
that shows up with small buffers.  As buffer size grows both implementation
converge to the same throughput.

Now here's the exact same test performed on an XScale core (PXA255):

	PTR alignment	buffer size	kernel version	this version
	------------------------------------------------------------
	  aligned	     32		 46.99		 77.58
	unaligned	     32		 53.61		 59.59
	  aligned	    100		107.19		136.59
	unaligned	    100		 83.61		 97.58
	  aligned	   4096		129.13		129.98
	unaligned	   4096		128.36		128.53
	  aligned	1048576		 53.76		 59.41
	unaligned	1048576		 33.67		 56.96

Again we can see the entry setup cost being higher for the current kernel
before getting to the main copy loop.  Then throughput results converge
as long as the buffer remains in the cache. Then the 1MB case shows more
differences probably due to better pld placement and/or less instruction
interlocks in this proposed implementation.

Disclaimer: The PXA system was running with slower clocks than the
StrongARM system so trying to infer any conclusion by comparing those
separate sets of results side by side would be completely inappropriate.

So...  What this patch does is to replace both memcpy and memmove with
an implementation based on the provided copy code template.  The memmove
code is kept separate since it is used only if the memory areas involved
do overlap in which case the code is a transposition of the template but
with the copy occurring in the opposite direction (trying to fit that
mode into the template turned it into a mess not worth it for memmove
alone).  And obviously both memcpy and memmove were tested with all kinds
of pointer alignments and buffer sizes to exercise all code paths for
correctness.

The next patch will provide the now trivial replacement implementation
copy_to_user and copy_from_user.
Signed-off-by: default avatarNicolas Pitre <nico@cam.org>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      75494230
    • Nicolas Pitre's avatar
      [ARM] 2946/2: split --arch_clear_user() out of lib/uaccess.S · a0c6fdb9
      Nicolas Pitre authored 
      Patch from Nicolas Pitre

Required for future enhancement patches.
Signed-off-by: default avatarNicolas Pitre <nico@cam.org>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      a0c6fdb9
    • David Brownell's avatar
      [ARM] 3078/1: lubbock platform updates, mostly mmc detection · 85eb226c
      David Brownell authored 
      Patch from David Brownell

Lubbock updates:

  * Provide an address for the SMC91x chip that doesn't generate
    a boot-time warning (matching the EEPROM).

  * Update MMC support to (a) detect card insert/remove, and
    (b) report the readonly switch setting for SD cards.

Previously, MMC/SD cards had to be present at boot time else they
couldn't be detected.
Signed-off-by: default avatarDavid Brownell <dbrownell@users.sourceforge.net>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      85eb226c
    • Ben Dooks's avatar
      [ARM] 3077/1: S3C2410 - regs-iis.h missing mask for IISMOD_FS · 4ebc3364
      Ben Dooks authored 
      Patch from Ben Dooks

Add definition for S3C2410_IISMOD_FS_MASK
Signed-off-by: default avatarBen Dooks <ben-linux@fluff.org>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      4ebc3364
    • Ben Dooks's avatar
      [ARM] 3076/1: S3C2410 - updated documentation for platfrom data init · 6ff8f59f
      Ben Dooks authored 
      Patch from Ben Dooks

Update the Documentation/arm/Samsung-S3C24XX to add
example platform data initialisation, and add the
linux-arm mailing list URL.
Signed-off-by: default avatarBen Dooks <ben-linux@fluff.org>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      6ff8f59f
    • Ben Dooks's avatar
      [ARM] 3071/1: RX3715 - add lcd/fb platform setup · e838ffc2
      Ben Dooks authored 
      Patch from Ben Dooks

Platform data for the LCD/framebuffer driver for
the RX3715 LCD panel.
Signed-off-by: default avatarBen Dooks <ben-linux@fluff.org>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      e838ffc2
    • Lennert Buytenhek's avatar
      [ARM] 3065/1: ixp2000 typo and whitespace fixes · fa87cedd
      Lennert Buytenhek authored 
      Patch from Lennert Buytenhek

Misc ixp2000 typo and whitespace fixes.
Signed-off-by: default avatarLennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: default avatarDeepak Saxena <dsaxena@plexity.net>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      fa87cedd
    • Lennert Buytenhek's avatar
      [ARM] 3064/1: start using ixp2000_reg_wrb · e9b72e43
      Lennert Buytenhek authored 
      Patch from Lennert Buytenhek

Switch the users of ixp2000_reg_write that depend on writes being
flushed out of the write buffer by the time that function returns
over to ixp2000_reg_wrb.

When using XCB=101, writes to the same functional unit are still
guaranteed to complete in order, so we only need to protect against:
- reordering of writes to different functional units
- masking an interrupt and then reenabling the IRQ bit in CPSR
Signed-off-by: default avatarLennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: default avatarDeepak Saxena <dsaxena@plexity.net>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      e9b72e43
    • Lennert Buytenhek's avatar
      [ARM] 3063/1: allow slave ixp2000 cpu reset · 69a85761
      Lennert Buytenhek authored 
      Patch from Lennert Buytenhek

On the ixdp2x00, the slave CPU is currently not allowed to reset itself
for fear that it will do something 'funky' on the PCI bus.  This fear is
ungrounded -- the slave CPU is wired up such that a CPU reset will not
cause a PCI bus reset to be done.  This patch changes arch_reset() so
that the slave CPU also executes the reset sequence, allowing it to
reboot itself using /sbin/reboot.
Signed-off-by: default avatarLennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: default avatarDeepak Saxena <dsaxena@plexity.net>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      69a85761
    • Lennert Buytenhek's avatar
      [ARM] 3062/1: map in various enp2611 peripherals for the ixp2000 netdev driver · a6f1063b
      Lennert Buytenhek authored 
      Patch from Lennert Buytenhek

The enp2611 version of the ixp2000 netdev driver needs to be able to
access a number of on-board peripherals.  ioremap() is not suitable
for this, as that will cause XCB=000 mappings to be done, which will
make the cpu susceptible to crashing on ixp2400 erratum #66.  Properly
aligned iotable mappings with MT_IXP2000_DEVICE will cause section
mappings with XCB=101 to be done, which is safe.
Signed-off-by: default avatarLennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: default avatarDeepak Saxena <dsaxena@plexity.net>
Signed-off-by: default avatarRussell King <rmk+kernel@arm.linux.org.uk>
      a6f1063b
  3. 31 Oct, 2005 30 commits