Skip to content

cloud-kernel
cloud-kernel

bug2833 / cloud-kernel
与 Fork 源项目一致

Fork自 openanolis / cloud-kernel

通知 1
Star 0
Fork 0
cloud-kernel
cloud-kernel

体验新版 GitCode，发现更多精彩内容 >>

切换分支/标签
  1. 27 6月, 2005 1 次提交
  3. 26 6月, 2005 3 次提交
    • R
      [PATCH] ARM: Generic Dynamic Tick Timer support for ARM, take 4 · 8749af68
      由 Russell King 提交于 
      This patch adds support for Dynamic Tick Timer for ARM. Dynamic Tick is
also known as VST (Variable Scheduling Timeouts).

Dynamic Tick has been in use in the OMAP tree since last October.  The
patch is not intrusive, and does not do anything unless CONFIG_NO_IDLE_HZ
is defined.  This patch has the following fixed based on comments from
RMK:
- Time is updated before calling interrupt handlers.
- Added new interrupt flag SA_TIMER to avoid duplicate timer interrupts
- Moved struct dyn_tick_timer to time.h until we at some point probably
  have an arch independent dyn-tick.h
- Cleaned up testing for DYN_TICK_ENABLED in irq.c

 I've cleaned up this patch to fix some remaining issues:
 - Call the timer tick handler with irqs disabled, as it would be from
   a normal interrupt
 - if we have a dyn_tick, we better implement all methods.
 - generic timer_dyn_reprogram() call, to be called before sleeping
 - added command line option - "dyntick=" to allow boot-time control
   of this feature
    -- rmk

Signed-off-by: Tony Lindgren
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
      8749af68
    • L
      [PATCH] ARM: 2752/1: disable ixp2000 PCI I/O software workaround on chips that don't need it · 321ab6a5
      由 Lennert Buytenhek 提交于 
      Patch from Lennert Buytenhek

The later ixp2000 models don't need the PCI I/O workaround that we
currently perform.  Add a config option to disable the workaround,
and panic on boot if a kernel without the workaround is booted on a
buggy chip.  As only pre-production ixp2000s need the workaround,
the default is for it not to be configured in.
Signed-off-by: NLennert Buytenhek <buytenh@wantstofly.org>
Signed-off-by: Deepak Saxena
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
      321ab6a5
    • R
      [PATCH] ARM: Fix discontigmem · 3cd9e19e
      由 Russell King 提交于 
      The merge of sparsemem broke ARM discontigmem.  Fix it.
Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
      3cd9e19e
  5. 25 6月, 2005 6 次提交
  7. 24 6月, 2005 5 次提交
  9. 23 6月, 2005 4 次提交
  11. 22 6月, 2005 4 次提交
    • R
      [PATCH] ARM: Allow clps7500 to build without parsing "acorn" tag · 05216219
      由 Russell King 提交于 
      Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
      05216219
    • R
      [PATCH] ARM: Allow riscpc to parse "acorn" boot info tag · ebe2a9ff
      由 Russell King 提交于 
      Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
      ebe2a9ff
    • R
      [PATCH] ARM: Fix sa1111.c build error caused by klist changes · 522c37b9
      由 Russell King 提交于 
      Signed-off-by: NRussell King <rmk+kernel@arm.linux.org.uk>
      522c37b9
    • W
      [PATCH] Avoiding mmap fragmentation · 1363c3cd
      由 Wolfgang Wander 提交于 
      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: NWolfgang Wander <wwc@rentec.com>
Credit-to: "Richard Purdie" <rpurdie@rpsys.net>
Signed-off-by: NKen Chen <kenneth.w.chen@intel.com>
Acked-by: Ingo Molnar <mingo@elte.hu> (partly)
Signed-off-by: NAndrew Morton <akpm@osdl.org>
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
      1363c3cd
  13. 21 6月, 2005 7 次提交
  15. 20 6月, 2005 10 次提交