Skip to contentSkip to conversation

compile errors in 2.6.8.1 in drivers/md/*

DISCUSSION Sebastian Henschel 2 participants 3 messages Aug 27, 2004 MD RAID

lore
Share & tools
lore
#0 lore raw

hello folks...

hopefully i am right here... i just compiled a kernel 2.6.8.1 and two similar errors occurred:

drivers/md/dm-raid.c:1241 - the workqueue creation misses an argument. i have set the flag to 0.
drivers/md/kcopyd.c:612 - the singlethread workqueue creation misses an argument. i have set the flag to 0.

the kernel compiles now, but i do not if it really is a fix. since the machine this kernel was compiled on, does not make use of a software raid, i cannot test it.

cheers,
sebastian

#1 lore raw
Quoted 5 lines · Sebastian Henschel

hello folks...

hopefully i am right here... i just compiled a kernel 2.6.8.1 and two similar errors occurred:

Quoted 8 lines

drivers/md/dm-raid.c:1241 - the workqueue creation misses an argument. i have set the flag to 0.
drivers/md/kcopyd.c:612 - the singlethread workqueue creation misses an argument. i have set the flag to 0.

the kernel compiles now, but i do not if it really is a fix. since the machine this kernel was compiled on, does not make use of a software raid, i cannot test it.

Are you building a stock 2.6.8.1 kernel? Could it be -mm or one of the other common extra patchsets? I don't see any compile warnings or errors anywhere in the drivers/md/ directory when I build my kernel, which is mostly a stock 2.6.8.1 with a few extra DM patches.

Signature · 3 lines
Kevin Corry
[email protected]
http://evms.sourceforge.net
#2 lore raw

hello kevin...

Quoted 10 lines · Kevin Corry
Quoted 9 lines · Sebastian Henschel

hopefully i am right here... i just compiled a kernel 2.6.8.1 and two similar errors occurred:

drivers/md/dm-raid.c:1241 - the workqueue creation misses an argument. i have set the flag to 0.
drivers/md/kcopyd.c:612 - the singlethread workqueue creation misses an argument. i have set the flag to 0.

Quoted 8 lines
Quoted 3 lines

the kernel compiles now, but i do not if it really is a fix. since the machine this kernel was compiled on, does not make use of a software raid, i cannot test it.

Are you building a stock 2.6.8.1 kernel? Could it be -mm or one of the other common extra patchsets? I don't see any compile warnings or errors anywhere in the drivers/md/ directory when I build my kernel, which is mostly a stock 2.6.8.1 with a few extra DM patches.

oops... you hit me here. when i read your post, a memory was crawling back into my mind. i already had this workqueue issue some months ago at another place. it happens due to SWSUSP2 modifying include/linux/workqueue.h so that the creation functions get one more argument. so SWSUSP2 should patch the two aforementioned files and not the linux-raid team.

sorry for the fuzz,
sebastian

Signature · 4 lines
::: .O.
::: ..O
::: OOO
::: lynx -source http://www.kodeaffe.de/shensche.pub | gpg --import