Linux 4.8 compat: rw_semaphore atomic_long_t count #564
Conversation
|
@tuxoko @dweeezil thus far this appears to be the only compatibility issue with the latest kernel. If you guys could review this we can resolve the new kernel.org buildbot failures. You'll probably want to wait for the buildbot test results in openzfs/zfs#4887 even though I did test this locally. |
|
Looks like I spoke too soon, there's at least one other build issue to resolve. |
| #ifdef HAVE_RWSEM_ACTIVITY | ||
| #define RWSEM_COUNT(sem) sem->activity | ||
| #else | ||
| #ifdef HAVE_RWSEM_ATOMIC_LONG_COUNT |
There was a problem hiding this comment.
Yeah, I went back and forth about doing this as an #elif defined(). I'll refresh the patch with that change since after looking at the whole patch I think I like that better too.
|
submit_bio change |
While I'll all set up for this I'll work up a patch. It looks straight forward, just update |
For non-rwsem-spinlocks the "count" member was changed from a "long" to "atomic_long_t" type. A configure check has been added to detect this change along with new versions of the _rwsem_tryupgrade() function and RWSEM_COUNT() macro. See torvalds/linux/commit/8ee62b18 for complete details. Signed-off-by: Brian Behlendorf <[email protected]> Issue openzfs#563
|
@behlendorf [EDIT... oops, thought I was commenting on the ZFS PR] |
|
Merged as: b7c7008 Linux 4.8 compat: rw_semaphore atomic_long_t count |
For non-rwsem-spinlocks the "count" member was changed from a
"long" to "atomic_long_t" type. A configure check has been
added to detect this change along with new versions of the
_rwsem_tryupgrade() function and RWSEM_COUNT() macro. See
torvalds/linux/commit/8ee62b18 for complete details.
Signed-off-by: Brian Behlendorf [email protected]
Issue #563