lkml.org 
[lkml]   [1997]   [Feb]   [10]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: MMX performance....
On 6 Feb 1997, Robert Krawitz wrote:

> Is the Pentium memcpy patch in 2.1.x yet? If not, why not? Such a low
> level change should go into the development kernels as early as possible,
> surely...
>
> I've been sending it off to Linus periodically. He hasn't responded.
> I would like to see it in the kernel.
>
> Actually it's about time I started using development kernels on my
> personal box. I like to always run with the memcpy patch though. Is there
> at least a 2.1.x patch for it?

Well, after the recent discussions about the memcpy patch, I finally
decided to try it on my system with 2.0.28. After testing the newly
patched kernel, I ran a "make -j" for a while and it seemed to be
going okay so I left it running with the patch all night. Still no
problems...

Today I grabbed the 2.0.29 patch, reversed the memcpy, applied 2.0.29
and reapplied the memcpy patch. During the compile of 2.0.29 kernel
with a straight forward make zImage I received:

Feb 8 17:18:04 gomer kernel: general protection: 0000
Feb 8 17:18:04 gomer kernel: CPU: 0
Feb 8 17:18:04 gomer kernel: EIP: 0010:[kmalloc+115/656]
Feb 8 17:18:04 gomer kernel: EFLAGS: 00010282
Feb 8 17:18:04 gomer kernel: eax: ffff037f ebx: fffffff7 ecx:
00000084 edx: 0306e018
Feb 8 17:18:04 gomer kernel: esi: ffff037f edi: 08052b08 ebp:
bfffe1bc esp: 00a50fa0
Feb 8 17:18:04 gomer kernel: ds: 0018 es: 0018 fs: 002b gs: 002b
ss: 0018
Feb 8 17:18:04 gomer kernel: Process make (pid: 1914, process nr: 90,
stackpage=00a50000)
Feb 8 17:18:04 gomer kernel: Stack: fffffff7 00000100 08052b08 00120d5c
ffff037f 003ff810 0010a605 00000084
Feb 8 17:18:04 gomer kernel: 08017b8c 00000009 00000100 08052b08
bfffe1bc ffffffda 0000002b 0000002b
Feb 8 17:18:04 gomer kernel: 0000002b 0000002b 00000006 4007610a
00000023 00000286 bfffe1b8 0000002b
Feb 8 17:18:04 gomer kernel: Call Trace: [kmalloc+268/656]
[floppy_track_buffer+17925/24576]
Feb 8 17:18:04 gomer kernel: Code: 66 83 7e 0e 00 75 16 68 b2 56 1b 00 e8
0c 1d ff ff 31 c0 83

This is the first kernel fault I have seen in a while (since my CPU fan
died) and it occurred only after applying the memcpy patch. My system is
a Pentium 120mhz w/ 64mb RAM. It's certainly not a top of the line system
but it is generally quite reliable so I wouldn't be too quick to blame it
on the hardware.

My second compile of the kernel *seems* to have completed but I will be
removing the patch once I reboot/recompile. It doesn't seem to be stable
enough for me yet.

==============================================================
| Greg Patterson | EMAIL: root@gomer.mlink.net |
| | gomer@mlink.net |
| | IRC: Wizird (#Linux) |
| Linux is the Future! | WWW: http://gomer.mlink.net/~root |
==============================================================


\
 
 \ /
  Last update: 2005-03-22 13:38    [W:0.941 / U:0.208 seconds]
©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site