Bug 21342 - Failure to compile or use dm-raid4-5 in Linux Kernel
Summary: Failure to compile or use dm-raid4-5 in Linux Kernel
Status: CLOSED OBSOLETE
Alias: None
Product: Other
Classification: Unclassified
Component: Modules (show other bugs)
Hardware: All Linux
: P1 high
Assignee: other_modules
URL:
Keywords:
Depends on:
Blocks:
 
Reported: 2010-10-28 09:20 UTC by Erick Couts
Modified: 2012-05-17 15:37 UTC (History)
2 users (show)

See Also:
Kernel Version: 2.6.35
Subsystem:
Regression: No
Bisected commit-id:

Attachments
Add an attachment (proposed patch, testcase, etc.)

Description Erick Couts 2010-10-28 09:20:29 UTC 
Hi, I've noticed that a great number of people have had trouble with getting their software RAID 4/5/6 arrays to work properly with dmraid.  While investigating this problem, I decided to recompile my kernel to see if it could be the problem.  Lo and behold:

ubuntu/built-in.o: In function `dm_rh_dec':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:527: multiple definition of `dm_rh_dec'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:535: first defined here
ubuntu/built-in.o: In function `dm_rh_delay':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:673: multiple definition of `dm_rh_delay'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:687: first defined here
ubuntu/built-in.o: In function `dm_rh_recovery_in_flight':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:661: multiple definition of `dm_rh_recovery_in_flight'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:675: first defined here
ubuntu/built-in.o: In function `dm_rh_update_states':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:431: multiple definition of `dm_rh_update_states'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:437: first defined here
ubuntu/built-in.o: In function `dm_rh_start_recovery':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:707: multiple definition of `dm_rh_start_recovery'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:708: first defined here
ubuntu/built-in.o: In function `dm_rh_mark_nosync':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:393: multiple definition of `dm_rh_mark_nosync'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:395: first defined here
ubuntu/built-in.o: In function `dm_region_hash_destroy':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:230: multiple definition of `dm_region_hash_destroy'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:235: first defined here
ubuntu/built-in.o: In function `dm_rh_region_to_sector':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:118: multiple definition of `dm_rh_region_to_sector'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:122: first defined here
ubuntu/built-in.o: In function `dm_rh_dirty_log':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:255: multiple definition of `dm_rh_dirty_log'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:260: first defined here
ubuntu/built-in.o: In function `dm_rh_stop_recovery':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:697: multiple definition of `dm_rh_stop_recovery'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:698: first defined here
ubuntu/built-in.o: In function `dm_rh_bio_to_region':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:124: multiple definition of `dm_rh_bio_to_region'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:128: first defined here
ubuntu/built-in.o: In function `dm_rh_recovery_prepare':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:604: multiple definition of `dm_rh_recovery_prepare'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:619: first defined here
ubuntu/built-in.o: In function `dm_rh_recovery_start':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:627: multiple definition of `dm_rh_recovery_start'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:642: first defined here
ubuntu/built-in.o: In function `dm_rh_flush':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:667: multiple definition of `dm_rh_flush'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:681: first defined here
ubuntu/built-in.o: In function `dm_rh_region_context':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:130: multiple definition of `dm_rh_region_context'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:134: first defined here
ubuntu/built-in.o: In function `dm_region_hash_create':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:163: multiple definition of `dm_region_hash_create'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:167: first defined here
ubuntu/built-in.o: In function `dm_rh_get_region_key':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:136: multiple definition of `dm_rh_get_region_key'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:140: first defined here
ubuntu/built-in.o: In function `dm_rh_inc_pending':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:518: multiple definition of `dm_rh_inc_pending'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:523: first defined here
ubuntu/built-in.o: In function `dm_rh_get_state':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:333: multiple definition of `dm_rh_get_state'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:338: first defined here
ubuntu/built-in.o: In function `dm_rh_get_region_size':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:142: multiple definition of `dm_rh_get_region_size'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:146: first defined here
ubuntu/built-in.o: In function `dm_rh_recovery_end':
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c:643: multiple definition of `dm_rh_recovery_end'
drivers/built-in.o:/home/erick/Desktop/linux-source-2.6.35/drivers/md/dm-region-hash.c:658: first defined here
make: *** [vmlinux.o] Error 1


Now, at first, I thought it was that I tried using -O3 optimizations instead of -O2 so I changed everything back to the way it "should" have been and ran make clean.  Upon my next attempt, the same thing happened.  I should mention that I'm using Ubuntu 10.10 and there are a great number of bugs they need to work through.  But, I feel that this is more of a source code problem than an Ubuntu problem.  Please review and let me know what you think.  Thanks!
Comment 1 Randy Dunlap 2010-10-28 17:20:22 UTC 
dm-region-hash.c builds fine in 2.6.35 for me, but then I don't have
the conflicting Ubuntu file(s), e.g.:
/home/erick/Desktop/linux-source-2.6.35/ubuntu/dm-raid4-5/dm-region-hash.c

so it looks like an Ubuntu-only problem to me.
Comment 2 Erick Couts 2010-10-29 02:14:26 UTC 
Yeah, I noticed that afterwards.  However, it seems that it wasn't just Ubuntu that had this problem for a while.  Initially, a few other distributions were having this as well.  I was thinking that it might be a better idea to include dm-raid4/5/6 support in the original kernel as it already has the required modules that it's based upon.  This would keep all distributions pretty well covered as far as being able to use dm-raid4/5/6 on devices otherwise unseen by Linux.  It would also allow for a universal fix and hopefully someone cleaning up the code (because OMG!  I so don't want to try going through that again [hasn't anyone heard of typedef?!]).

If you're wondering what other distro's had this problem, Fedora did for a while until they finally fixed it.
Note You need to log in before you can comment on or make changes to this bug.