search for: blkzeroout

On 08/02/2018 02:05 PM, Nir Soffer wrote: > fallocate(FALLOC_FL_ZERO_RANGE) is supportd for block devices with > modern kernel, but when it is not, we fall back to manual zeroing. > > Check if the underlying file is a block device when opening the file, > and fall back to ioctl(BLKZEROOUT) for aligned zero requests for a > block device. > > +++ b/plugins/file/file.c > @@ -45,6 +45,7 @@ > > #if defined(__linux__) && !defined(FALLOC_FL_PUNCH_HOLE) > #include <linux/falloc.h> /* For FALLOC_FL_*, glibc < 2.18 */ > +#include <linux/fs...

...PM, Nir Soffer wrote: > > fallocate(FALLOC_FL_ZERO_RANGE) is supportd for block devices with > > modern kernel, but when it is not, we fall back to manual zeroing. > > > > Check if the underlying file is a block device when opening the file, > > and fall back to ioctl(BLKZEROOUT) for aligned zero requests for a > > block device. > > > > +++ b/plugins/file/file.c > > @@ -45,6 +45,7 @@ > > > > #if defined(__linux__) && !defined(FALLOC_FL_PUNCH_HOLE) > > #include <linux/falloc.h> /* For FALLOC_FL_*, glibc < 2.18...

...k 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 can punch hole and fallocate, try fallocate(PUNCH_HOLE) followed by fallocate(0). - If underlying file is a block device, try ioctl(BLKZEROOUT) - Otherwise fallback to manual zeroing The handle keeps now the underlying file capabilities, so once we discover that an operation is not supported, we never try it again. Here are examples runs on a server based on Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz, using XtremIO storage via 4G FC HBA...

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 sector size. v1 was here: https://www.redhat.com/archives/libguestfs/2018-July/msg00084.html Nir Soffer (3): file: Avoid unsupported fallocate() calls file: Support zero without ZERO_RANGE file...

...); \ +}) #endif /* NBDKIT_ISALIGNED_H */ diff --git a/plugins/file/file.c b/plugins/file/file.c index 9d03f18..1391f62 100644 --- a/plugins/file/file.c +++ b/plugins/file/file.c @@ -397,13 +397,13 @@ file_zero (void *handle, uint32_t count, uint64_t offset, int may_trim) #ifdef BLKZEROOUT /* For aligned range and block device, we can use BLKZEROOUT. */ - if (h->can_zeroout && is_aligned (offset | count, h->sector_size)) { + if (h->can_zeroout && IS_ALIGNED (offset | count, h->sector_size)) { uint64_t range[2] = {offset, count}; r = ioctl...

...; 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 can punch hole and fallocate, try fallocate(PUNCH_HOLE) followed > by fallocate(0). > - If underlying file is a block device, try ioctl(BLKZEROOUT) > - Otherwise fallback to manual zeroing > > The handle keeps now the underlying file capabilities, so once we > discover that an operation is not supported, we never try it again. > > > Issues: > - ioctl(BLKZEROOUT) will fail if offset or count are not aligned to >...

fallocate(FALLOC_FL_ZERO_RANGE) is supportd for block devices with modern kernel, but when it is not, we fall back to manual zeroing. Check if the underlying file is a block device when opening the file, and fall back to ioctl(BLKZEROOUT) for aligned zero requests for a block device. Here is an example run without this change on RHEL 7.5: $ export SOCK=/tmp/nbd.sock $ export BLOCK=/dev/e30bfac2-8e13-479d-8cd6-c6da5e306c4e/c9864222-bc52-4359-80d7-76e47d619b15 $ src/nbdkit plugins/file/.libs/nbdkit-file-plugin.so file=$BLOCK -U $SO...

fallocate(FALLOC_FL_ZERO_RANGE) is supported for block devices with modern kernel, but when it is not, we fall back to manual zeroing. For block device, try also to use ioctl(BLKZEROOUT) if offset and count are aligned to block device sector size. Here is an example run without this change on RHEL 7.5: $ export SOCK=/tmp/nbd.sock $ export BLOCK=/dev/e30bfac2-8e13-479d-8cd6-c6da5e306c4e/c9864222-bc52-4359-80d7-76e47d619b15 $ src/nbdkit plugins/file/.libs/nbdkit-file-plugin.so fil...

fallocate(FALLOC_FL_ZERO_RANGE) is supportd for block devices with modern kernel, but when it is not, we fall back to manual zeroing. For block device, try also to use ioctl(BLKZEROOUT) if offset and count are aligned to block device sector size. Here is an example run without this change on RHEL 7.5: $ export SOCK=/tmp/nbd.sock $ export BLOCK=/dev/e30bfac2-8e13-479d-8cd6-c6da5e306c4e/c9864222-bc52-4359-80d7-76e47d619b15 $ src/nbdkit plugins/file/.libs/nbdkit-file-plugin.so fil...

...); \ +}) #endif /* NBDKIT_ISALIGNED_H */ diff --git a/plugins/file/file.c b/plugins/file/file.c index 9d03f18..1391f62 100644 --- a/plugins/file/file.c +++ b/plugins/file/file.c @@ -397,13 +397,13 @@ file_zero (void *handle, uint32_t count, uint64_t offset, int may_trim) #ifdef BLKZEROOUT /* For aligned range and block device, we can use BLKZEROOUT. */ - if (h->can_zeroout && is_aligned (offset | count, h->sector_size)) { + if (h->can_zeroout && IS_ALIGNED (offset | count, h->sector_size)) { uint64_t range[2] = {offset, count}; r = ioctl...

...BDKIT_ISALIGNED_H */ > diff --git a/plugins/file/file.c b/plugins/file/file.c > index 9d03f18..1391f62 100644 > --- a/plugins/file/file.c > +++ b/plugins/file/file.c > @@ -397,13 +397,13 @@ file_zero (void *handle, uint32_t count, uint64_t > offset, int may_trim) > > #ifdef BLKZEROOUT > /* For aligned range and block device, we can use BLKZEROOUT. */ > - if (h->can_zeroout && is_aligned (offset | count, h->sector_size)) { > + if (h->can_zeroout && IS_ALIGNED (offset | count, h->sector_size)) { > uint64_t range[2] = {offset, count...

...LLOC_FL_* when including <linux/fs.h>. - Add common/includes/isaligned.h for is_aligned helper, implemented in a more efficient way with bitwise math. - If getting sector size fail, fall back to safe guess instead of hard failure. - More efficient alignment check using bitwise math. - For BLKZEROOUT, treat ENOTTY as EOPNOTSUPP. Theoretically possible with a mix of new headers and old or strangely configured kernel. - Use default multi line comment style. - Fix few typos in comments and commit message Issues to explore later: - Eric suggested to use the new FALLOC_FL_NO_HIDE_STALE. Requires...

...k 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 can punch hole and fallocate, try fallocate(PUNCH_HOLE) followed by fallocate(0). - If underlying file is a block device, try ioctl(BLKZEROOUT) - Otherwise fallback to manual zeroing The handle keeps now the underlying file capabilities, so once we discover that an operation is not supported, we never try it again. Here are examples runs on a server based on Intel(R) Xeon(R) CPU E5-2630 v4 @ 2.20GHz, using XtremIO storage via 4G FC HBA...

...<sys/stat.h> +#include <sys/ioctl.h> + +#include <pthread.h> + +#if defined (__linux__) && !defined (FALLOC_FL_PUNCH_HOLE) +#include <linux/falloc.h> /* For FALLOC_FL_*, glibc < 2.18 */ +#endif + +#if defined (__linux__) +#include <linux/fs.h> /* For BLKZEROOUT */ +#endif + +#define NBDKIT_API_VERSION 2 +#include <nbdkit-plugin.h> + +#include "cleanup.h" +#include "fileops.h" +#include "isaligned.h" + +#ifndef HAVE_FDATASYNC +#define fdatasync fsync +#endif + +/* Any callbacks using lseek must be protected by this lock....

There's a lot of code in nbdkit-file-plugin which it would be nice to reuse elsewhere. One possible approach (as outlined here) is simply to move the file callbacks (like file.pread, file.pwrite, file.zero etc) to a new mini-library. They can then be consumed by other plugins fairly easily by doing: static void * foo_open (int readonly) { struct fileops *fops; int fd, flags; /*

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

...ps://gerrit.ovirt.org/#/c/85413/ We have a demo here: https://gerrit.ovirt.org/#/c/85468/ This will not help you use case when you want to mix read/write/zero requests, but we can use the same infrastructure. We plan to use fallocate for file based storage: https://gerrit.ovirt.org/#/c/85512/ and BLKZEROOUT for block storage: https://gerrit.ovirt.org/#/c/85537/ and some dumb zero loop if these options are not available. So we need to map the zero operation to http - how about: POST /images/ticket-id ... ... { "op": "zero", "offset": X, "size": Y }...

...| 62 ++++++++++++++++++++++++++++++--------------- 1 file changed, 42 insertions(+), 20 deletions(-) diff --git a/plugins/file/file.c b/plugins/file/file.c index 2a8c9b5..dcff0ee 100644 --- a/plugins/file/file.c +++ b/plugins/file/file.c @@ -52,6 +52,8 @@ #include <linux/fs.h> /* For BLKZEROOUT */ #endif +#define NBDKIT_API_VERSION 2 + #include <nbdkit-plugin.h> #include "isaligned.h" @@ -263,9 +265,30 @@ file_can_trim (void *handle) #endif } +static int +file_can_fua (void *handle) +{ + return NBDKIT_FUA_NATIVE; +} + +/* Flush the file to disk. */ +static int...

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

Needs some work still, see in particular the commit message for patch 3. Rich.