search for: punch_hole

If we may not trim, we tried ZERO_RANGE, but this is not well supported yet, for example it is not available on NFS 4.2. ZERO_RANGE and PUNCH_HOLE are supported now on block devices, but not on RHRL 7, so we fallback to slow manual zeroing there. Change the logic to support block devices on RHEL 7, and file systems that do not support ZERO_RANGE. The new logic: - If we may trim, try PUNCH_HOLE - If we can zero range, Try ZERO_RANGE - If we...

...to manual zeroing. When > trimming, we used to try unsupported fallocate() on every call. > > Change file handle to remember if punching holes or zeroing range are > supported, and avoid unsupported calls. > > - zero changed to: > 1. If we can punch hole and may trim, try PUNCH_HOLE > 2. If we can zero range, try ZERO_RANGE > 3. Fall back to manual writing > > - trim changed to: > 1. If we can punch hole, try PUNCH_HOLE > 2. Succeed > --- > plugins/file/file.c | 80 ++++++++++++++++++++++++++++++--------------- > 1 file changed, 53 insertio...

If we may not trim, we tried ZERO_RANGE, but this is not well supported yet, for example it is not available on NFS 4.2. ZERO_RANGE and PUNCH_HOLE are supported now on block devices, but not on RHRL 7, so we fallback to slow manual zeroing there. Change the logic to support block devices on RHEL 7, and file systems that do not support ZERO_RANGE. The new logic: - If we may trim, try PUNCH_HOLE - If we can zero range, Try ZERO_RANGE - If we...

This is the second version to support efficient zero for block devices on older kernels (e.g. RHEL 7.5), and file systems that do not support yet FALLOC_FS_ZERO_RANGE (e.g. NFS 4.2). Changes since v1: - Split to smaller patches - Skip linux only includes on other systems - Skip code using BLKZEROOUT if the macro is not defined - Try BLKZEROOUT only if the offset and count are aligned to device

...fail with EOPNOTSUPP, and fallback to manual zeroing. When trimming, we used to try unsupported fallocate() on every call. Change file handle to remember if punching holes or zeroing range are supported, and avoid unsupported calls. - zero changed to: 1. If we can punch hole and may trim, try PUNCH_HOLE 2. If we can zero range, try ZERO_RANGE 3. Fall back to manual writing - trim changed to: 1. If we can punch hole, try PUNCH_HOLE 2. Succeed --- plugins/file/file.c | 80 ++++++++++++++++++++++++++++++--------------- 1 file changed, 53 insertions(+), 27 deletions(-) diff --git a/plugins/...

...fail with EOPNOTSUPP, and fallback to manual zeroing. When trimming, we used to try unsupported fallocate() on every call. Change file handle to remember if punching holes or zeroing range are supported, and avoid unsupported calls. - zero changed to: 1. If we can punch hole and may trim, try PUNCH_HOLE 2. If we can zero range, try ZERO_RANGE 3. Fall back to manual writing - trim changed to: 1. If we can punch hole, try PUNCH_HOLE 2. Succeed --- plugins/file/file.c | 80 ++++++++++++++++++++++++++++++--------------- 1 file changed, 53 insertions(+), 27 deletions(-) diff --git a/plugins/...

...fail with EOPNOTSUPP, and fallback to manual zeroing. When trimming, we used to try unsupported fallocate() on every call. Change file handle to remember if punching holes or zeroing range are supported, and avoid unsupported calls. - zero changed to: 1. If we can punch hole and may trim, try PUNCH_HOLE 2. If we can zero range, try ZERO_RANGE 3. Fall back to manual writing - trim changed to: 1. If we can punch hole, try PUNCH_HOLE 2. Succeed --- plugins/file/file.c | 84 +++++++++++++++++++++++++++++---------------- 1 file changed, 55 insertions(+), 29 deletions(-) diff --git a/plugins/...

On 07/29/2018 07:04 AM, Nir Soffer wrote: > If we may not trim, we tried ZERO_RANGE, but this is not well supported > yet, for example it is not available on NFS 4.2. ZERO_RANGE and > PUNCH_HOLE are supported now on block devices, but not on RHRL 7, so we > fallback to slow manual zeroing there. > > Change the logic to support block devices on RHEL 7, and file systems > that do not support ZERO_RANGE. > > The new logic: > - If we may trim, try PUNCH_HOLE > - If we...

.... Change file handle to remember if punching holes or zeroing range are supported, and avoid unsupported calls. - can_trim changed to report the actual capability once we tried to punch a hole. I don't think this is useful yet. - zero changed to: 1. If we can punch hole and may trim, try PUNCH_HOLE 2. If we can zero range, try ZERO_RANGE 3. Fall back to manual writing - trim changed to: 1. If we can punch hole, try PUNCH_HOLE 2. Succeed --- plugins/file/file.c | 96 ++++++++++++++++++++++++++++----------------- 1 file changed, 59 insertions(+), 37 deletions(-) diff --git a/plugins/...

...++++++++++++++++++++++++--------- 1 file changed, 24 insertions(+), 9 deletions(-) diff --git a/plugins/file/file.c b/plugins/file/file.c index a7c07fb..4210adb 100644 --- a/plugins/file/file.c +++ b/plugins/file/file.c @@ -50,6 +50,21 @@ static char *filename = NULL; +#if defined(FALLOC_FL_PUNCH_HOLE) || defined(FALLOC_FL_ZERO_RANGE) +static int +do_fallocate(int fd, int mode, off_t offset, off_t len) +{ + int r = -1; + r = fallocate (fd, mode, offset, len); + /* kernel 3.10 fails with ENODEV for block device. Kernel >= 4.9 fails + with EOPNOTSUPP in this case. Normlize errno to simpl...

...letions(-) diff --git a/plugins/file/file.c b/plugins/file/file.c index b6e33de..a7c07fb 100644 --- a/plugins/file/file.c +++ b/plugins/file/file.c @@ -243,7 +243,7 @@ file_zero (void *handle, uint32_t count, uint64_t offset, int may_trim) if (may_trim) { r = fallocate (h->fd, FALLOC_FL_PUNCH_HOLE | FALLOC_FL_KEEP_SIZE, offset, count); - if (r == -1 && errno != EOPNOTSUPP) { + if (r == -1 && errno != EOPNOTSUPP && errno != ENODEV) { nbdkit_error ("zero: %m"); } /* PUNCH_HOLE is older; if it is not supported, it is likely that @@...

...to manual zeroing. When > trimming, we used to try unsupported fallocate() on every call. > > Change file handle to remember if punching holes or zeroing range are > supported, and avoid unsupported calls. > > - zero changed to: > 1. If we can punch hole and may trim, try PUNCH_HOLE > 2. If we can zero range, try ZERO_RANGE > 3. Fall back to manual writing > > - trim changed to: > 1. If we can punch hole, try PUNCH_HOLE > 2. Succeed > --- > plugins/file/file.c | 80 ++++++++++++++++++++++++++++++--------------- > 1 file changed, 53 in...

...mbers it was off; turning the feature off is detrimental as the client already assumes it works). So the best you can do is make subsequent connections more efficient after the initial connection has learned state. > > - zero changed to: > 1. If we can punch hole and may trim, try PUNCH_HOLE > 2. If we can zero range, try ZERO_RANGE > 3. Fall back to manual writing > > - trim changed to: > 1. If we can punch hole, try PUNCH_HOLE > 2. Succeed Seems reasonable from the description. > @@ -118,6 +119,8 @@ file_config_complete (void) > /* The per-con...

This version addresses some of the comments on v2. Changes since v2: - file_zero: Add missing space in function call - is_aligned: Assert that align is indeed a power of 2 - Spelling in commit message Not changed: - Eric commented that spacing was off: https://www.redhat.com/archives/libguestfs/2018-August/msg00113.html but I could not find anything wrong. - Eric asked if ioctl.h will cause

This version addresses comments on v3. Changes since v3: - Finally got spacing right (Eric) - Reorder includes (Richard) - Return 0 or -1 instead of r (Richard) - Add common/include/isaligned.h to Makefile.am (Richard) v3 was here: https://www.redhat.com/archives/libguestfs/2018-August/msg00177.html Nir Soffer (4): file: Avoid unsupported fallocate() calls file: Support zero without

...learned state. > Sure disabling trim does not help the client. The only advantage is not calling trim when it does nothing. Do you think it is better to leave the file_can_trim as is, to avoid confusion? > > > - zero changed to: > > 1. If we can punch hole and may trim, try PUNCH_HOLE > > 2. If we can zero range, try ZERO_RANGE > > 3. Fall back to manual writing > > > > - trim changed to: > > 1. If we can punch hole, try PUNCH_HOLE > > 2. Succeed > > Seems reasonable from the description. > > > @@ -118,6 +119,8 @@ fi...

...only includes on other systems - Skip code using BLKZEROOUT if the macro is not defined - Try BLKZEROOUT only if the offset and count are aligned to device sector size. - initialize h->can_* properly. Before they were uninitialized if FALLOC_FL_* macros were not defined. - Use new h->can_punch_hole in file_can_trim, so now we report the actual capability once we detected it. - Use h->can_punch_hole in file_trim, so we try only once if the operation is not supported. v1 was here: https://www.redhat.com/archives/libguestfs/2018-July/msg00084.html Nir Soffer (4): file: Avoid unsupport...

...y. (If fd is a > >pipe or FIFO, a different error results.) > > The man page is out-of-date; newer kernels support > FALLOC_FL_ZERO_RANGE on block devices, in a manner that leaves the > pages marked allocated (that is, no discard happens). The kernel > also supports FALLOC_FL_PUNCH_HOLE to guarantee a read-zeroes result > (using discard, if that works), and > FALLOC_FL_PUNCH_HOLE|FALLOC_FL_NO_HIDE_STALE (which only discards, > and no longer guarantees a read-zeroes result). But you are also > right that in all cases, the errno returned when the operation > cannot b...

...++++++++++++++++++++++++--------- 1 file changed, 24 insertions(+), 9 deletions(-) diff --git a/plugins/file/file.c b/plugins/file/file.c index a7c07fb..a8a6253 100644 --- a/plugins/file/file.c +++ b/plugins/file/file.c @@ -50,6 +50,21 @@ static char *filename = NULL; +#if defined(FALLOC_FL_PUNCH_HOLE) || defined(FALLOC_FL_ZERO_RANGE) +static int +do_fallocate(int fd, int mode, off_t offset, off_t len) +{ + int r = -1; + r = fallocate (fd, mode, offset, len); + /* kernel 3.10 fails with ENODEV for block device. Kernel >= 4.9 fails + with EOPNOTSUPP in this case. Normalize errno to simp...

...4 insertions(+), 9 deletions(-) > > diff --git a/plugins/file/file.c b/plugins/file/file.c > index a7c07fb..4210adb 100644 > --- a/plugins/file/file.c > +++ b/plugins/file/file.c > @@ -50,6 +50,21 @@ > > static char *filename = NULL; > > +#if defined(FALLOC_FL_PUNCH_HOLE) || defined(FALLOC_FL_ZERO_RANGE) > +static int > +do_fallocate(int fd, int mode, off_t offset, off_t len) > +{ Similar name to the normalizing wrapper that qemu uses, but enough differences that I think you're safe here. (Be careful that you aren't directly copying code from th...