lkml.org 
[lkml]   [2000]   [Feb]   [11]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
/
Date
From
SubjectRe: 2.3.43 alpha broken with >2g of ram
On Fri, 11 Feb 2000, Richard Henderson wrote:

>Don't bother. I didn't finish rawhide.

Richard at this alpha breakpoint I think we can as well merge my SMP irq
rewrite before 2.4.0 (that allows PCI irq to scale all over the CPUs in
SMP and that make the irq code SMP safe).

andrea@alpha:~ > cat /proc/interrupts
CPU0 CPU1 TRY0 TRY1
1: 112800 0 112800 0 XT-PIC keyboard
2: 0 0 0 0 XT-PIC cascade
8: 29063124 0 29063124 0 RTC +timer
12: 362593 0 362593 0 XT-PIC PS/2 Mouse
27: 247873 187199 427009 427898 TSUNAMI eth0
46: 44693 23916 68481 68537 TSUNAMI sym53c8xx
47: 29 8 37 37 TSUNAMI sym53c8xx
LOC: 29063125 29062986
My rewrite is necessary if we want to scale with more than one CPU and
more than one NIC and it fixes a few bugs in the alpha irq code.

My irq breakage in arch/alpha/kernel is shown explicitly by the code that
doesn't compile and fixing the 7 not yet uptodate platforms is not a pain
and can be done based on the previous code. The design is the same of x86.
I only run my machine with my irq scaling code applyed and I _never_ had
one problem. I know you had problems with irq probing but unless I get
more details and bugreports I can't certainly fix that since I can't
reproduce anything of strange here.

The alpha SMP irq patch I run all the time against 2.3.42 is here:

ftp://ftp.*.kernel.org/pub/linux/kernel/people/andrea/patches/v2.3/2.3.42/alpha-irq-4.gz

I checked it generates two silly (trivially fixable) rejects on the top of
2.3.43. I'll produce anyway soon a 2.3.43 version (as soon as I'll run
2.3.43 on alpha ;).

Comments?

Andrea


-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

\
 
 \ /
  Last update: 2005-03-22 13:56    [from the cache]
©2003-2011 Jasper Spaans