[Date Prev][Date Next] [Thread Prev][Thread Next] [Date Index] [Thread Index]

Bug#455909: marked as done (linux-image-2,6-iop32x: many misaligned memory accesses)

Your message dated Tue, 10 Feb 2009 15:05:21 +0100
with message-id <20090210140521.GA15056@stro.at>
and subject line Re: linux-image-2,6-iop32x: many misaligned memory accesses
has caused the Debian Bug report #455909,
regarding linux-image-2,6-iop32x: many misaligned memory accesses
to be marked as done.

This means that you claim that the problem has been dealt with.
If this is not the case it is now your responsibility to reopen the
Bug report if necessary, and/or fix the problem forthwith.

(NB: If you are a system administrator and have no idea what this
message is talking about, this may indicate a serious mail system
misconfiguration somewhere. Please contact owner@bugs.debian.org
immediately.)


-- 
455909: http://bugs.debian.org/cgi-bin/bugreport.cgi?bug=455909
Debian Bug Tracking System
Contact owner@bugs.debian.org with problems
--- Begin Message --- 
Package: linux-image-2.6-iop32x
Version: 2.6.22+11
User: debian-arm@lists.debian.org
Usertags: eabi
Severity: wishlist

After running for a mere 20 hours, /proc/cpu/alignment reports
millions of misaligned word accesses from the kernel:
$ cat /proc/cpu/alignment
User:           0
System:         2765980
Skipped:        0
Half:           0
Word:           2765980
DWord:          0
Multi:          0
User faults:    0 (ignored)

I gather this has a performance penalty, as misaligned kernel memory
accesses are always trapped and fixed up.

None or the other EABI kernels I am running exhibits the same behaviour:
Angstrom's 2.6.20-rc1-h1940 reports 3 system-word alignment fixups in 30 days;
Angstrom's 2.6.23 running Angstrom GPE reports 9 user-word fixups in 12 days
a Debian-armel system with custom kernel reports no misalignments in 9 days.

--- End Message ---
--- Begin Message --- 
> I'd be inclined to close this item for lack of importance.


ok doing so as no further activity.

--- End Message ---
Reply to: