Skip to content

K
Kernel

openeuler / Kernel
1 年多 前同步成功

通知 8
Star 0
Fork 0
K
Kernel
切换分支/标签
  1. 10 2月, 2008 1 次提交
  3. 30 1月, 2008 1 次提交
    • C
      x86: 64-bit, make sparsemem vmemmap the only memory model · b263295d
      由 Christoph Lameter 提交于 
      Use sparsemem as the only memory model for UP, SMP and NUMA.  Measurements
indicate that DISCONTIGMEM has a higher overhead than sparsemem.  And
FLATMEMs benefits are minimal.  So I think its best to simply standardize
on sparsemem.

Results of page allocator tests (test can be had via git from slab git
tree branch tests)

Measurements in cycle counts. 1000 allocations were performed and then the
average cycle count was calculated.

Order	FlatMem	Discontig	SparseMem
0	  639	  665		  641
1	  567	  647		  593
2	  679	  774		  692
3	  763	  967		  781
4	  961	 1501		  962
5	 1356	 2344		 1392
6	 2224	 3982		 2336
7	 4869	 7225		 5074
8	12500	14048		12732
9	27926	28223		28165
10	58578	58714		58682

(Note that FlatMem is an SMP config and the rest NUMA configurations)

Memory use:

SMP Sparsemem
-------------

Kernel size:

   text    data     bss     dec     hex filename
3849268  397739 1264856 5511863  541ab7 vmlinux

             total       used       free     shared    buffers     cached
Mem:       8242252      41164    8201088          0        352      11512
-/+ buffers/cache:      29300    8212952
Swap:      9775512          0    9775512

SMP Flatmem
-----------

Kernel size:

   text    data     bss     dec     hex filename
3844612  397739 1264536 5506887  540747 vmlinux

So 4.5k growth in text size vs. FLATMEM.

             total       used       free     shared    buffers     cached
Mem:       8244052      40544    8203508          0        352      11484
-/+ buffers/cache:      28708    8215344

2k growth in overall memory use after boot.

NUMA discontig:

   text    data     bss     dec     hex filename
3888124  470659 1276504 5635287  55fcd7 vmlinux

             total       used       free     shared    buffers     cached
Mem:       8256256      56908    8199348          0        352      11496
-/+ buffers/cache:      45060    8211196
Swap:      9775512          0    9775512

NUMA sparse:

   text    data     bss     dec     hex filename
3896428  470659 1276824 5643911  561e87 vmlinux

8k text growth. Given that we fully inline virt_to_page and friends now
that is rather good.

             total       used       free     shared    buffers     cached
Mem:       8264720      57240    8207480          0        352      11516
-/+ buffers/cache:      45372    8219348
Swap:      9775512          0    9775512

The total available memory is increased by 8k.

This patch makes sparsemem the default and removes discontig and
flatmem support from x86.

[ akpm@linux-foundation.org: allnoconfig build fix ]
Acked-by: NAndi Kleen <ak@suse.de>
Signed-off-by: NChristoph Lameter <clameter@sgi.com>
Signed-off-by: NAndrew Morton <akpm@linux-foundation.org>
Signed-off-by: NIngo Molnar <mingo@elte.hu>
Signed-off-by: NThomas Gleixner <tglx@linutronix.de>
      b263295d
  5. 30 10月, 2007 1 次提交
  7. 26 10月, 2007 1 次提交
    • S
      x86: move defconfig files for i386 and x86_64 to x86 · 2266cfd5
      由 Sam Ravnborg 提交于 
      With some small changes to kconfig makefile we can now
locate the defconfig files for i386 and x86_64 in
the configs/ subdirectory under x86.
make ARCH=i386 defconfig and make defconfig
works as expected also after this change.
But arch maintainers shall now update a defconfig file in
the configs/ directory.
Signed-off-by: NSam Ravnborg <sam@ravnborg.org>
      2266cfd5
  9. 31 8月, 2007 1 次提交
  11. 30 7月, 2007 1 次提交
  13. 22 7月, 2007 1 次提交
  15. 22 5月, 2007 1 次提交
  17. 03 5月, 2007 1 次提交
  19. 29 4月, 2007 1 次提交
    • J
      libata/IDE: remove combined mode quirk · 8cdfb29c
      由 Jeff Garzik 提交于 
      Both old-IDE and libata should be able handle all controllers and
devices found using normal resource reservation methods.

This eliminates the awful, low-performing split-driver configuration
where old-IDE drove the PATA portion of a PCI device, in PIO-only mode,
and libata drove the SATA portion of the /same/ PCI device, in DMA mode.
Typically vendors would ship SATA hard drive / PATA optical
configuration, which would lend itself to slow (PIO-only) CD-ROM
performance.

For Intel users running in combined mode, it is now wholly dependent on
your driver choice (potentially link order, if you compile both drivers
in) whether old-IDE or libata will drive your hardware.

In either case, you will get full performance from both SATA and PATA
ports now, without having to pass a kernel command line parameter.
Signed-off-by: NJeff Garzik <jeff@garzik.org>
      8cdfb29c
  21. 17 3月, 2007 1 次提交
  23. 13 2月, 2007 1 次提交
  25. 11 1月, 2007 1 次提交
  27. 31 12月, 2006 1 次提交
  29. 16 12月, 2006 1 次提交
    • L
      Remove stack unwinder for now · d1526e2c
      由 Linus Torvalds 提交于 
      It has caused more problems than it ever really solved, and is
apparently not getting cleaned up and fixed.  We can put it back when
it's stable and isn't likely to make warning or bug events worse.

In the meantime, enable frame pointers for more readable stack traces.
Signed-off-by: NLinus Torvalds <torvalds@osdl.org>
      d1526e2c
  31. 10 12月, 2006 1 次提交
  33. 07 12月, 2006 1 次提交
  35. 22 10月, 2006 1 次提交
  37. 06 10月, 2006 1 次提交
  39. 30 9月, 2006 1 次提交
  41. 26 9月, 2006 1 次提交
  43. 31 8月, 2006 1 次提交
  45. 30 7月, 2006 1 次提交
  47. 11 7月, 2006 1 次提交
  49. 27 6月, 2006 1 次提交
  51. 19 4月, 2006 1 次提交
  53. 10 4月, 2006 1 次提交
  55. 26 3月, 2006 1 次提交
  57. 21 2月, 2006 1 次提交
  59. 18 2月, 2006 1 次提交
  61. 05 2月, 2006 1 次提交
  63. 19 1月, 2006 1 次提交
  65. 17 1月, 2006 1 次提交
  67. 12 1月, 2006 1 次提交
  69. 15 11月, 2005 1 次提交
  71. 13 9月, 2005 1 次提交
  73. 26 8月, 2005 1 次提交
  75. 25 8月, 2005 3 次提交