* [PATCH] src/seek_sanity_test: Fix for filesystems without delayed allocation @ 2017-05-10 11:46 Andreas Gruenbacher 2017-05-11 5:58 ` Eryu Guan 0 siblings, 1 reply; 3+ messages in thread From: Andreas Gruenbacher @ 2017-05-10 11:46 UTC (permalink / raw) To: fstests; +Cc: Andreas Gruenbacher src/seek_sanity_test (test generic/285) assumes that after fallocating space in a file, fseek SEEK_HOLE / SEEK_DATA will still report the allocated space as a hole. This isn't true on filesystems that don't support delayed allocation, so skip the affected tests in that case. Tested on xfs and gfs2 + patches for fseek SEEK_HOLE / SEEK_DATA. Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> --- src/seek_sanity_test.c | 27 +++++++++++++++++++++++++++ 1 file changed, 27 insertions(+) diff --git a/src/seek_sanity_test.c b/src/seek_sanity_test.c index a6dd48c..86ddf1c 100644 --- a/src/seek_sanity_test.c +++ b/src/seek_sanity_test.c @@ -37,6 +37,7 @@ static blksize_t alloc_size; int default_behavior = 0; +int delayed_allocation = 1; char *base_file_path; static void get_file_system(int fd) @@ -282,6 +283,12 @@ static int test09(int fd, int testnum) int bufsz = alloc_size; int filsz = 8 << 20; + if (!delayed_allocation) { + /* Report success if fs doesn't support delayed allocation */ + fprintf(stdout, "Test skipped as fs doesn't support delayed allocation\n"); + goto out; + } + /* * HOLE - unwritten DATA in dirty page - HOLE - * unwritten DATA in writeback page @@ -338,6 +345,12 @@ static int test08(int fd, int testnum) int bufsz = alloc_size; int filsz = 4 << 20; + if (!delayed_allocation) { + /* Report success if fs doesn't support delayed allocation */ + fprintf(stdout, "Test skipped as fs doesn't support delayed allocation\n"); + goto out; + } + /* HOLE - unwritten DATA in writeback page */ /* Each unit is bufsz */ buf = do_malloc(bufsz); @@ -387,6 +400,12 @@ static int test07(int fd, int testnum) int bufsz = alloc_size; int filsz = 4 << 20; + if (!delayed_allocation) { + /* Report success if fs doesn't support delayed allocation */ + fprintf(stdout, "Test skipped as fs doesn't support delayed allocation\n"); + goto out; + } + /* HOLE - unwritten DATA in dirty page */ /* Each unit is bufsz */ buf = do_malloc(bufsz); @@ -776,6 +795,14 @@ static int test_basic_support(void) fprintf(stderr, "File system supports the default behavior.\n"); } + ftruncate(fd, 0); + fallocate(fd, 0, 0, 1 << 20); + pos = lseek(fd, 0, SEEK_DATA); + if (pos == 0) { + fprintf(stderr, "File system does not support delayed allocation.\n"); + delayed_allocation = 0; + } + printf("\n"); out: -- 2.7.4 ^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH] src/seek_sanity_test: Fix for filesystems without delayed allocation 2017-05-10 11:46 [PATCH] src/seek_sanity_test: Fix for filesystems without delayed allocation Andreas Gruenbacher @ 2017-05-11 5:58 ` Eryu Guan 2017-05-11 11:28 ` Andreas Gruenbacher 0 siblings, 1 reply; 3+ messages in thread From: Eryu Guan @ 2017-05-11 5:58 UTC (permalink / raw) To: Andreas Gruenbacher; +Cc: fstests On Wed, May 10, 2017 at 01:46:04PM +0200, Andreas Gruenbacher wrote: > src/seek_sanity_test (test generic/285) assumes that after fallocating > space in a file, fseek SEEK_HOLE / SEEK_DATA will still report the > allocated space as a hole. This isn't true on filesystems that don't > support delayed allocation, so skip the affected tests in that case. > > Tested on xfs and gfs2 + patches for fseek SEEK_HOLE / SEEK_DATA. I'm not sure if this is really about delayed allocation, I tried nodelalloc mounted ext4, test still passed. And the following test reported fallocated space as HOLE, on both delalloc and nodelalloc mounted ext4: [root@bootp-73-5-205 xfstests]# grep vda6 /proc/mounts /dev/vda6 /mnt/ext4 ext4 rw,seclabel,relatime,nodelalloc,data=ordered 0 0 [root@bootp-73-5-205 xfstests]# rm /mnt/ext4/testfile [root@bootp-73-5-205 xfstests]# xfs_io -fc "falloc 0 1m" -c "seek -h 0" -c "seek -d 0" /mnt/ext4/testfile Whence Result HOLE 0 Whence Result DATA EOF [root@bootp-73-5-205 xfstests]# mount -o remount,delalloc /mnt/ext4 [root@bootp-73-5-205 xfstests]# grep vda6 /proc/mounts /dev/vda6 /mnt/ext4 ext4 rw,seclabel,relatime,data=ordered 0 0 [root@bootp-73-5-205 xfstests]# rm /mnt/ext4/testfile [root@bootp-73-5-205 xfstests]# xfs_io -fc "falloc 0 1m" -c "seek -h 0" -c "seek -d 0" /mnt/ext4/testfile Whence Result HOLE 0 Whence Result DATA EOF I noticed that test07/8/9 are the only three that test unwritten extents, perhaps not supporting unwritten extents are the reason? Thanks, Eryu > > Signed-off-by: Andreas Gruenbacher <agruenba@redhat.com> > --- > src/seek_sanity_test.c | 27 +++++++++++++++++++++++++++ > 1 file changed, 27 insertions(+) > > diff --git a/src/seek_sanity_test.c b/src/seek_sanity_test.c > index a6dd48c..86ddf1c 100644 > --- a/src/seek_sanity_test.c > +++ b/src/seek_sanity_test.c > @@ -37,6 +37,7 @@ > > static blksize_t alloc_size; > int default_behavior = 0; > +int delayed_allocation = 1; > char *base_file_path; > > static void get_file_system(int fd) > @@ -282,6 +283,12 @@ static int test09(int fd, int testnum) > int bufsz = alloc_size; > int filsz = 8 << 20; > > + if (!delayed_allocation) { > + /* Report success if fs doesn't support delayed allocation */ > + fprintf(stdout, "Test skipped as fs doesn't support delayed allocation\n"); > + goto out; > + } > + > /* > * HOLE - unwritten DATA in dirty page - HOLE - > * unwritten DATA in writeback page > @@ -338,6 +345,12 @@ static int test08(int fd, int testnum) > int bufsz = alloc_size; > int filsz = 4 << 20; > > + if (!delayed_allocation) { > + /* Report success if fs doesn't support delayed allocation */ > + fprintf(stdout, "Test skipped as fs doesn't support delayed allocation\n"); > + goto out; > + } > + > /* HOLE - unwritten DATA in writeback page */ > /* Each unit is bufsz */ > buf = do_malloc(bufsz); > @@ -387,6 +400,12 @@ static int test07(int fd, int testnum) > int bufsz = alloc_size; > int filsz = 4 << 20; > > + if (!delayed_allocation) { > + /* Report success if fs doesn't support delayed allocation */ > + fprintf(stdout, "Test skipped as fs doesn't support delayed allocation\n"); > + goto out; > + } > + > /* HOLE - unwritten DATA in dirty page */ > /* Each unit is bufsz */ > buf = do_malloc(bufsz); > @@ -776,6 +795,14 @@ static int test_basic_support(void) > fprintf(stderr, "File system supports the default behavior.\n"); > } > > + ftruncate(fd, 0); > + fallocate(fd, 0, 0, 1 << 20); > + pos = lseek(fd, 0, SEEK_DATA); > + if (pos == 0) { > + fprintf(stderr, "File system does not support delayed allocation.\n"); > + delayed_allocation = 0; > + } > + > printf("\n"); > > out: > -- > 2.7.4 > > -- > To unsubscribe from this list: send the line "unsubscribe fstests" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html ^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] src/seek_sanity_test: Fix for filesystems without delayed allocation 2017-05-11 5:58 ` Eryu Guan @ 2017-05-11 11:28 ` Andreas Gruenbacher 0 siblings, 0 replies; 3+ messages in thread From: Andreas Gruenbacher @ 2017-05-11 11:28 UTC (permalink / raw) To: Eryu Guan; +Cc: fstests Eryu, thanks for the review. On Thu, May 11, 2017 at 7:58 AM, Eryu Guan <eguan@redhat.com> wrote: > On Wed, May 10, 2017 at 01:46:04PM +0200, Andreas Gruenbacher wrote: >> src/seek_sanity_test (test generic/285) assumes that after fallocating >> space in a file, fseek SEEK_HOLE / SEEK_DATA will still report the >> allocated space as a hole. This isn't true on filesystems that don't >> support delayed allocation, so skip the affected tests in that case. >> >> Tested on xfs and gfs2 + patches for fseek SEEK_HOLE / SEEK_DATA. > > I'm not sure if this is really about delayed allocation, I tried > nodelalloc mounted ext4, test still passed. And the following test > reported fallocated space as HOLE, on both delalloc and nodelalloc > mounted ext4: > > [root@bootp-73-5-205 xfstests]# grep vda6 /proc/mounts > /dev/vda6 /mnt/ext4 ext4 rw,seclabel,relatime,nodelalloc,data=ordered 0 0 > [root@bootp-73-5-205 xfstests]# rm /mnt/ext4/testfile > [root@bootp-73-5-205 xfstests]# xfs_io -fc "falloc 0 1m" -c "seek -h 0" -c "seek -d 0" /mnt/ext4/testfile > Whence Result > HOLE 0 > Whence Result > DATA EOF > [root@bootp-73-5-205 xfstests]# mount -o remount,delalloc /mnt/ext4 > [root@bootp-73-5-205 xfstests]# grep vda6 /proc/mounts > /dev/vda6 /mnt/ext4 ext4 rw,seclabel,relatime,data=ordered 0 0 > [root@bootp-73-5-205 xfstests]# rm /mnt/ext4/testfile > [root@bootp-73-5-205 xfstests]# xfs_io -fc "falloc 0 1m" -c "seek -h 0" -c "seek -d 0" /mnt/ext4/testfile > Whence Result > HOLE 0 > Whence Result > DATA EOF > > I noticed that test07/8/9 are the only three that test unwritten > extents, perhaps not supporting unwritten extents are the reason? You are right, unwritten extents is the correct term for that particular technique. I noticed that some additional error checking makes sense as well, so I'll repost. Thanks, Andreas ^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2017-05-11 11:28 UTC | newest] Thread overview: 3+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2017-05-10 11:46 [PATCH] src/seek_sanity_test: Fix for filesystems without delayed allocation Andreas Gruenbacher 2017-05-11 5:58 ` Eryu Guan 2017-05-11 11:28 ` Andreas Gruenbacher
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox