Commit Graph

2761 Commits

Author SHA1 Message Date
Brian Behlendorf 9aedfd95f6 Merge branch 'feature-branch' into refs/top-bases/zfs-branch 2009-08-04 14:39:06 -07:00
Brian Behlendorf 7b71127c1c Merge commit 'refs/top-bases/feature-branch' into feature-branch 2009-08-04 14:39:04 -07:00
Brian Behlendorf 0741d789d2 Merge branch 'feature-pthreads' into refs/top-bases/feature-branch 2009-08-04 14:39:04 -07:00
Brian Behlendorf 0ebcbe2e74 Revert unneeded ztest_resume() prototype changes to match upstream. 2009-08-04 14:29:50 -07:00
Ricardo M. Correia c0ce3b40f2 Reimplement user kthreads based on pthread thread specific data.
This resolves previous scalabily concerns about the cost of calling
curthread which previously required a list walk.  The kthread address
is now tracked as thread specific data which can be quickly returned.

Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2009-08-04 13:55:15 -07:00
Brian Behlendorf 9be5ad2924 Flag ZIO_FLAG_SPECULATIVE now issues READ instead of READA.
There is concern that READA may do more than simply reorder the queue.
There may be an increased chance that a requested marked READA will
fail because the elevator considers it optional.  For this reason, all
read requests, even speculative ones, have been converted back to READ.
2009-08-04 10:37:40 -07:00
Brian Behlendorf 7c2850b819 Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-08-03 16:30:46 -07:00
Brian Behlendorf 48cbfae8bd Merge commit 'refs/top-bases/zfs-branch' into zfs-branch 2009-08-03 16:28:49 -07:00
Brian Behlendorf f2f52087f3 Merge branch 'gcc-branch' into refs/top-bases/zfs-branch 2009-08-03 16:28:48 -07:00
Brian Behlendorf 7a8bfe1597 Merge commit 'refs/top-bases/gcc-branch' into gcc-branch 2009-08-03 16:28:47 -07:00
Brian Behlendorf 6cb98ae6b9 Merge branch 'gcc-unused' into refs/top-bases/gcc-branch 2009-08-03 16:28:46 -07:00
Brian Behlendorf 68a80a8b87 refcount_add() mistakenly moved to ASSERT() breaks --disable-debug builds. 2009-08-03 16:20:27 -07:00
Brian Behlendorf 934ac68348 Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-07-31 14:04:57 -07:00
Brian Behlendorf 9f5e45ef4d Merge commit 'refs/top-bases/zfs-branch' into zfs-branch 2009-07-31 14:02:49 -07:00
Brian Behlendorf cf44f97b96 Merge branch 'gcc-branch' into refs/top-bases/zfs-branch 2009-07-31 14:02:48 -07:00
Brian Behlendorf af8e850408 Merge commit 'refs/top-bases/gcc-branch' into gcc-branch 2009-07-31 14:02:47 -07:00
Brian Behlendorf 252846a145 Merge branch 'gcc-shadow' into refs/top-bases/gcc-branch 2009-07-31 14:02:46 -07:00
Brian Behlendorf cfd9e9936f Fix an accidentally introduced formatting issue in zfs output.
Because the local 'index' variable shadows the index() function
it was replaced by 'i'.  Unfortunately when I made this change
I accidentally replaced one instance with 'j' resulting in the
short decimal values being printed incorrectly.
2009-07-31 13:50:21 -07:00
Brian Behlendorf 6c00d99d48 Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-07-29 21:29:41 -07:00
Brian Behlendorf 2d86fb2788 Merge commit 'refs/top-bases/zfs-branch' into zfs-branch 2009-07-29 21:27:13 -07:00
Brian Behlendorf 3185bb677e Merge branch 'fix-branch' into refs/top-bases/zfs-branch 2009-07-29 21:27:12 -07:00
Brian Behlendorf 83136344ea Merge commit 'refs/top-bases/fix-branch' into fix-branch 2009-07-29 21:27:11 -07:00
Brian Behlendorf 653cd5153a Merge branch 'fix-stack' into refs/top-bases/fix-branch 2009-07-29 21:27:10 -07:00
Brian Behlendorf e2bbeb5e61 Merge branch 'gcc-branch' into refs/top-bases/zfs-branch 2009-07-29 21:27:06 -07:00
Brian Behlendorf 5809ed62dd Merge commit 'refs/top-bases/gcc-branch' into gcc-branch 2009-07-29 21:27:05 -07:00
Brian Behlendorf f80c505f4d Merge branch 'gcc-uninit' into refs/top-bases/gcc-branch 2009-07-29 21:27:04 -07:00
Brian Behlendorf ee435f260c Excessively large stack frames detected.
The 2.6.30 kernel build systems sets -Wframe-larger-than=2048 which causes
a warning to be generated when an individual stack frame exceeds 2048.
This caught the spa_history_log() and dmu_objset_snapshot() functions
which declared a data structure on the stack which contained a char
array of MAXPATHLEN.  This in defined to be 4096 in the linux kernel
and I imagine it is quite large under Solaris as well.  Regardless, the
offending data structures were moved to the heap to correctly keep the
stack depth to a minimum.  We might consider setting this value even
lower to catch additional offenders because we are expecting deep stacks.
2009-07-29 21:18:48 -07:00
Brian Behlendorf 72d2035ed3 Updated gcc in fc11 flagged a new uninitialized variable. 2009-07-29 21:17:47 -07:00
Brian Behlendorf ed7ca273df Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-07-29 17:51:31 -07:00
Brian Behlendorf 4014bd401f Major vdev_disk feature and API (thru 2.6.30) update.
Tested under CHAOS4.2, RHEL5, SLES11, and FC11 (all x86_64)

Features:
Honor spa_mode() when opening the block device.  Previously this
was ignored and devices were always opened read/write.

Integrated DKIOCFLUSHWRITECACHE zio operation with linux WRITE_BARRIER
for kernels post 2.6.24 where empty bio requests are supported.  For
earlier kernels ENOTSUP is returned and no barriers are performed.  If
RHEL5 based kernels are intended to be supported long term we may need
make use of the old akward API.

With the addition of WRITE_BARRIER support all writes which were
WRITE_SYNC can now be safely made WRITE bios.  They will now take
advantage of aggregation in the elevator and improved write performance
is likely.

Notice the ZIO_FLAG_SPECULATIVE flag and pass along the hint to the
elevator by using READA instead of READ.  This provides the elevator
the ability to prioritize the real READs ahead of the speculative IO
if needed.

Implement an initial version of vdev_disk_io_done() which in the case
of an EIO error triggers a media change check.  If it determines a
media change has occured we fail the device and remove it from the
config.  This logic I'm sure can be improved further but for now it
is an improvement over the VERIFY() that no error will ever happen.

APIs:
2.6.22 API change
Unused destroy_dirty_buffers arg removed from prototype.

2.6.24 API change
Empty write barriers are now supported and we should use them.

2.6.24 API change
Size argument dropped from bio_endio and bi_end_io, because the
bi_end_io is only called once now when the request is complete.
There is no longer any need for a size argument.  This also means
that partial IO's are no longer possibe and the end_io callback
should not check bi->bi_size.  Finally, the return type was updated
to void.

2.6.28 API change
open/close_bdev_excl() renamed to open/close_bdev_exclusive().

2.6.29 API change
BIO_RW_SYNC renamed to BIO_RW_SYNCIO.
2009-07-29 17:24:08 -07:00
Brian Behlendorf 9cabcac115 Required autoconf support for vdev_disk integration with 2.6.30 kernels.
2.6.22 API change
Unused destroy_dirty_buffers arg removed from prototype.

2.6.24 API change
Empty write barriers are now supported and we should use them.

2.6.24 API change
Size argument dropped from bio_endio and bi_end_io, because the
bi_end_io is only called once now when the request is complete.
There is no longer any need for a size argument.  This also means
that partial IO's are no longer possibe and the end_io callback
should not check bi->bi_size.  Finally, the return type was updated
to void.

2.6.28 API change
open/close_bdev_excl() renamed to open/close_bdev_exclusive().

2.6.29 API change
BIO_RW_SYNC renamed to BIO_RW_SYNCIO.
2009-07-29 17:14:23 -07:00
Brian Behlendorf 5f0c206b77 Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-07-27 10:49:38 -07:00
Brian Behlendorf 3a431c68c2 Remove LINUXINCLUDE from autoconf wrapper, breaks 2.6.28+ kernels.
Modern kernel build systems at least post 2.6.16 will set this properly
so we should not.  In fact post 2.6.28 the include headers have moved
under arch so the guess we make here is completely wrong.  Letting
the kernel build system set this ensure it will be correct.  Also
drop the ulimit from the Makefile which, not surprisingly, turns out
to be very non-portable.  If your expecting failures set the ulimit
in your shell before kicking off the test suite.
2009-07-27 10:39:28 -07:00
Brian Behlendorf 1fba9cdbd8 Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-07-24 16:35:11 -07:00
Brian Behlendorf 18ffe73d3f Check arch/default/ path when detecting kernel objects on SLES
We still preferentially use arch/arch looking for a native version
but if that fails it is acceptable to use default.
2009-07-24 16:26:17 -07:00
Brian Behlendorf 4a50aa2518 Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-07-24 16:02:57 -07:00
Brian Behlendorf 5ecd180d91 ulimit is a function of the shell and should be invoked as such. 2009-07-24 15:50:57 -07:00
Brian Behlendorf 1fc5fb504c BIO_RW_FAILFAST replaced with BIO_RW_FAILFAST_{DEV|_TRANSPORT|_DRIVER}
Use the legacy BIO_RW_FAILFAST flag if it exists.  If it is missing it
means we are running against a kernel with the newer API.  We should
be able to enable some fairly smart behavior one we intergrate with the
new API, but until I get around to writing that code just remove the
flag entirely.  It's not critical for correctness.
2009-07-24 15:16:13 -07:00
Brian Behlendorf 67d31d626f The bi_end_io API changes make partial IO's impossible
Kernel commit 6712ecf8f648118c3363c142196418f89a510b90 which removes the
size argument from bio_endio and bi_end_io, also removes the need to
handle partial IOs in the handler.
2009-07-24 15:08:09 -07:00
Brian Behlendorf 9d80e40fbb Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-07-24 11:15:08 -07:00
Brian Behlendorf db3967ba86 Merge commit 'refs/top-bases/zfs-branch' into zfs-branch 2009-07-24 11:08:43 -07:00
Brian Behlendorf 9a2275c1c5 Merge branch 'feature-branch' into refs/top-bases/zfs-branch 2009-07-24 11:08:42 -07:00
Brian Behlendorf 77901ca759 Merge commit 'refs/top-bases/feature-branch' into feature-branch 2009-07-24 11:08:41 -07:00
Brian Behlendorf 0435a79314 Merge branch 'feature-pthreads' into refs/top-bases/feature-branch 2009-07-24 11:08:40 -07:00
Ricardo M. Correia 51760e3c7e Change VERIFY() into VERIFY3U() when creating threads in ztest, so that we can see the return code when it fails.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2009-07-24 11:00:23 -07:00
Ricardo M. Correia 08d148d418 Add ulimit -c unlimited before running ztest, to get a core dump.
Signed-off-by: Brian Behlendorf <behlendorf1@llnl.gov>
2009-07-24 10:58:55 -07:00
Brian Behlendorf e09a06470b Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-07-24 10:35:02 -07:00
Brian Behlendorf 710c21f2e9 Use LOSETUP variable for all calls to /sbin/losetup 2009-07-24 10:25:01 -07:00
Brian Behlendorf af92edb1a5 Merge commit 'refs/top-bases/linux-kernel-disk' into linux-kernel-disk 2009-07-24 09:18:04 -07:00
Brian Behlendorf 82ca18caae Merge commit 'refs/top-bases/zfs-branch' into zfs-branch 2009-07-24 09:15:56 -07:00