IronPort-SDR: 667a2450_2wWuU7aDEIaQNNHZBJ1fn3OP/++O90G1bonscVc/ptkgyFe SEO9gQS+BHzokeNg1gp79VSLLPZ7r8+Jd0D1+Nw== WDCIronportException: Internal From: Shin'ichiro Kawasaki To: linux-block@vger.kernel.org Cc: Bryan Gurney , Christoph Hellwig , Shin'ichiro Kawasaki Subject: [PATCH blktests] dm/002: do not assume 512 byte block size Date: Tue, 25 Jun 2024 11:51:43 +0900 Message-ID: <20240625025143.405254-1-shinichiro.kawasaki@wdc.com> X-Mailing-List: linux-block@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Xref: photonic.trudheim.com org.kernel.vger.linux-block:93431 Newsgroups: org.kernel.vger.linux-block Path: photonic.trudheim.com!nntp.lore.kernel.org!not-for-mail The test case assumes that TEST_DEV would have 512 byte block size always. However, TEST_DEV may have non 512 byte, 4k block size. In that case, the test case fails with I/O errors. To avoid the errors, refer to the block size of TEST_DEV. Also record dd command output to the FULL file to help debug work in the future. Reported-by: Christoph Hellwig Fixes: 7308e11c595a ("tests/dm: add dm-dust general functionality test") Link: https://lore.kernel.org/linux-block/ZmqrzUyLcUORPdOe@infradead.org/ Signed-off-by: Shin'ichiro Kawasaki --- tests/dm/002 | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-) diff --git a/tests/dm/002 b/tests/dm/002 index 6635c43..fae3986 100755 --- a/tests/dm/002 +++ b/tests/dm/002 @@ -14,10 +14,12 @@ requires() { test_device() { + local sz bsz echo "Running ${TEST_NAME}" - TEST_DEV_SZ=$(blockdev --getsz "$TEST_DEV") - dmsetup create dust1 --table "0 $TEST_DEV_SZ dust $TEST_DEV 0 512" + sz=$(blockdev --getsz "$TEST_DEV") + bsz=$(blockdev --getbsz "$TEST_DEV") + dmsetup create dust1 --table "0 $sz dust $TEST_DEV 0 $bsz" dmsetup message dust1 0 addbadblock 60 dmsetup message dust1 0 addbadblock 67 dmsetup message dust1 0 addbadblock 72 @@ -30,7 +32,8 @@ test_device() { dmsetup message dust1 0 addbadblock 72 dmsetup message dust1 0 countbadblocks dmsetup message dust1 0 enable - dd if=/dev/zero of=/dev/mapper/dust1 bs=512 count=128 oflag=direct >/dev/null 2>&1 || return $? + dd if=/dev/zero of=/dev/mapper/dust1 bs="$bsz" count=128 oflag=direct \ + >"$FULL" 2>&1 || return $? sync dmsetup message dust1 0 countbadblocks sync -- 2.45.0 . From: Christoph Hellwig To: roger.pau@citrix.com Cc: jgross@suse.com, marmarek@invisiblethingslab.com, xen-devel@lists.xenproject.org, linux-block@vger.kernel.org, Rusty Bird Subject: [PATCH] xen-blkfront: fix sector_size propagation to the block layer Date: Tue, 25 Jun 2024 07:52:38 +0200 Message-ID: <20240625055238.7934-1-hch@lst.de> X-Mailing-List: linux-block@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Xref: photonic.trudheim.com org.kernel.vger.linux-block:93433 org.xenproject.lists.xen-devel:51248 Newsgroups: org.kernel.vger.linux-block,org.xenproject.lists.xen-devel Path: photonic.trudheim.com!nntp.lore.kernel.org!not-for-mail Ensure that info->sector_size and info->physical_sector_size are set before the call to blkif_set_queue_limits by doing away with the local variables and arguments that propagate them. Thanks to Marek Marczykowski-Górecki and Jürgen Groß for root causing the issue. Fixes: ba3f67c11638 ("xen-blkfront: atomically update queue limits") Reported-by: Rusty Bird Signed-off-by: Christoph Hellwig --- drivers/block/xen-blkfront.c | 16 +++++----------- 1 file changed, 5 insertions(+), 11 deletions(-) diff --git a/drivers/block/xen-blkfront.c b/drivers/block/xen-blkfront.c index fa3a2ba525458b..59ce113b882a0e 100644 --- a/drivers/block/xen-blkfront.c +++ b/drivers/block/xen-blkfront.c @@ -1070,8 +1070,7 @@ static char *encode_disk_name(char *ptr, unsigned int n) } static int xlvbd_alloc_gendisk(blkif_sector_t capacity, - struct blkfront_info *info, u16 sector_size, - unsigned int physical_sector_size) + struct blkfront_info *info) { struct queue_limits lim = {}; struct gendisk *gd; @@ -1165,8 +1164,6 @@ static int xlvbd_alloc_gendisk(blkif_sector_t capacity, info->rq = gd->queue; info->gd = gd; - info->sector_size = sector_size; - info->physical_sector_size = physical_sector_size; xlvbd_flush(info); @@ -2320,8 +2317,6 @@ static void blkfront_gather_backend_features(struct blkfront_info *info) static void blkfront_connect(struct blkfront_info *info) { unsigned long long sectors; - unsigned long sector_size; - unsigned int physical_sector_size; int err, i; struct blkfront_ring_info *rinfo; @@ -2360,7 +2355,7 @@ static void blkfront_connect(struct blkfront_info *info) err = xenbus_gather(XBT_NIL, info->xbdev->otherend, "sectors", "%llu", §ors, "info", "%u", &info->vdisk_info, - "sector-size", "%lu", §or_size, + "sector-size", "%lu", &info->sector_size, NULL); if (err) { xenbus_dev_fatal(info->xbdev, err, @@ -2374,9 +2369,9 @@ static void blkfront_connect(struct blkfront_info *info) * provide this. Assume physical sector size to be the same as * sector_size in that case. */ - physical_sector_size = xenbus_read_unsigned(info->xbdev->otherend, + info->physical_sector_size = xenbus_read_unsigned(info->xbdev->otherend, "physical-sector-size", - sector_size); + info->sector_size); blkfront_gather_backend_features(info); for_each_rinfo(info, rinfo, i) { err = blkfront_setup_indirect(rinfo); @@ -2388,8 +2383,7 @@ static void blkfront_connect(struct blkfront_info *info) } } - err = xlvbd_alloc_gendisk(sectors, info, sector_size, - physical_sector_size); + err = xlvbd_alloc_gendisk(sectors, info); if (err) { xenbus_dev_fatal(info->xbdev, err, "xlvbd_add at %s", info->xbdev->otherend); -- 2.43.0 . From: Christoph Hellwig To: Jens Axboe Cc: Damien Le Moal , Niklas Cassel , Song Liu , Yu Kuai , "Martin K. Petersen" , Alim Akhtar , Avri Altman , Bart Van Assche , linux-block@vger.kernel.org, linux-ide@vger.kernel.org, linux-raid@vger.kernel.org, linux-scsi@vger.kernel.org Subject: queue_limits fixups and tidyups Date: Tue, 25 Jun 2024 13:05:40 +0200 Message-ID: <20240625110603.50885-1-hch@lst.de> X-Mailing-List: linux-block@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Xref: photonic.trudheim.com org.kernel.vger.linux-block:93441 Newsgroups: org.kernel.vger.linux-block,org.kernel.vger.linux-ide,org.kernel.vger.linux-raid,org.kernel.vger.linux-scsi Path: photonic.trudheim.com!nntp.lore.kernel.org!not-for-mail Hi Jens, this series has a few fixes for the queue_limits conversion in the first few patches and then has a bunch more cleanups and improvements in that area. Diffstat: block/bio-integrity.c | 2 +- block/blk-map.c | 2 +- block/blk-settings.c | 46 +++++++++++----------------------------------- block/blk-sysfs.c | 3 ++- block/blk.h | 2 ++ block/genhd.c | 2 +- drivers/ata/libata-scsi.c | 3 +-- drivers/ata/pata_macio.c | 4 ++-- drivers/md/md.c | 13 ++++++++----- drivers/md/md.h | 1 + drivers/md/raid0.c | 2 +- drivers/md/raid1.c | 2 +- drivers/md/raid10.c | 2 +- drivers/md/raid5.c | 2 +- drivers/scsi/scsi_lib.c | 4 ++-- drivers/ufs/core/ufshcd.c | 9 +++++---- include/linux/blkdev.h | 17 ++++++++++------- 17 files changed, 51 insertions(+), 65 deletions(-) . IronPort-SDR: 667a9a27_uh0ymX6MYUb+mQmwCUICXFBrqWDrmx1LSEEr+8bQATyySDz Wi5/rQ9o7PKY1lz57UsKcMcKlLdD/gIfOVxepVQ== WDCIronportException: Internal From: Shin'ichiro Kawasaki To: linux-block@vger.kernel.org Cc: Gulam Mohamed , Chaitanya Kulkarni , Shin'ichiro Kawasaki Subject: [PATCH blktests v2] loop/010: do not assume /dev/loop0 Date: Tue, 25 Jun 2024 20:20:11 +0900 Message-ID: <20240625112011.409282-1-shinichiro.kawasaki@wdc.com> X-Mailing-List: linux-block@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Xref: photonic.trudheim.com org.kernel.vger.linux-block:93449 Newsgroups: org.kernel.vger.linux-block Path: photonic.trudheim.com!nntp.lore.kernel.org!not-for-mail The current implementation of the test case loop/010 assumes that the prepared loop device is /dev/loop0, which is not always true. When other loop devices are set up before the test case run, the assumption is wrong and the test case fails. To avoid the failure, use the prepared loop device name stored in $loop_device instead of /dev/loop0. Adjust the grep string to meet the device name. Also use "losetup --detach" instead of "losetup --detach-all" to not detach the loop devices which existed before the test case runs. Fixes: 1c4ae4fed9b4 ("loop: Detect a race condition between loop detach and open") Signed-off-by: Shin'ichiro Kawasaki --- Changes from v1: * Replaced the losetup --find option with the loop device name * Added the missing "p1" postfix to the blkid argument tests/loop/010 | 23 +++++++++++++++-------- 1 file changed, 15 insertions(+), 8 deletions(-) diff --git a/tests/loop/010 b/tests/loop/010 index ea396ec..ade8044 100755 --- a/tests/loop/010 +++ b/tests/loop/010 @@ -16,18 +16,23 @@ requires() { } create_loop() { + local dev=$1 + while true do - loop_device="$(losetup --partscan --find --show "${image_file}")" - blkid /dev/loop0p1 >& /dev/null + if losetup --partscan "$dev" "${image_file}" &> /dev/null; then + blkid "$dev"p1 >& /dev/null + fi done } detach_loop() { + local dev=$1 + while true do - if [ -e /dev/loop0 ]; then - losetup --detach /dev/loop0 >& /dev/null + if [[ -e "$dev" ]]; then + losetup --detach "$dev" >& /dev/null fi done } @@ -38,6 +43,7 @@ test() { local create_pid local detach_pid local image_file="$TMPDIR/loopImg" + local grep_str truncate --size 1G "${image_file}" parted --align none --script "${image_file}" mklabel gpt @@ -53,9 +59,9 @@ test() { mkfs.xfs --force "${loop_device}p1" >& /dev/null losetup --detach "${loop_device}" >& /dev/null - create_loop & + create_loop "${loop_device}" & create_pid=$! - detach_loop & + detach_loop "${loop_device}" & detach_pid=$! sleep "${TIMEOUT:-90}" @@ -66,8 +72,9 @@ test() { sleep 1 } 2>/dev/null - losetup --detach-all >& /dev/null - if _dmesg_since_test_start | grep --quiet "partition scan of loop0 failed (rc=-16)"; then + losetup --detach "${loop_device}" >& /dev/null + grep_str="partition scan of ${loop_device##*/} failed (rc=-16)" + if _dmesg_since_test_start | grep --quiet "$grep_str"; then echo "Fail" fi echo "Test complete" -- 2.45.0 . From: linan666@huaweicloud.com To: axboe@kernel.dk Cc: linux-block@vger.kernel.org, linux-kernel@vger.kernel.org, hch@lst.de, linan666@huaweicloud.com, yukuai3@huawei.com, yi.zhang@huawei.com, houtao1@huawei.com, yangerkun@huawei.com Subject: [PATCH] block: clean up the check in blkdev_iomap_begin() Date: Tue, 25 Jun 2024 19:55:17 +0800 Message-Id: <20240625115517.1472120-1-linan666@huaweicloud.com> X-Mailing-List: linux-block@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Xref: photonic.trudheim.com org.kernel.vger.linux-block:93453 org.kernel.vger.linux-kernel:1258050 Newsgroups: org.kernel.vger.linux-block,org.kernel.vger.linux-kernel Path: photonic.trudheim.com!nntp.lore.kernel.org!not-for-mail From: Li Nan It is odd to check the offset amidst a series of assignments. Moving this check to the beginning of the function makes the code look better. Signed-off-by: Li Nan --- block/fops.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-) diff --git a/block/fops.c b/block/fops.c index 376265935714..bc5ad3e6197f 100644 --- a/block/fops.c +++ b/block/fops.c @@ -383,10 +383,11 @@ static int blkdev_iomap_begin(struct inode *inode, loff_t offset, loff_t length, struct block_device *bdev = I_BDEV(inode); loff_t isize = i_size_read(inode); - iomap->bdev = bdev; - iomap->offset = ALIGN_DOWN(offset, bdev_logical_block_size(bdev)); if (offset >= isize) return -EIO; + + iomap->bdev = bdev; + iomap->offset = ALIGN_DOWN(offset, bdev_logical_block_size(bdev)); iomap->type = IOMAP_MAPPED; iomap->addr = iomap->offset; iomap->length = isize - iomap->offset; -- 2.39.2 . Date: Tue, 25 Jun 2024 22:49:29 +0800 From: kernel test robot To: Yu Kuai CC: , , , Jens Axboe , , , , , Subject: [linus:master] [block] 060406c61c: sysbench-fileio.write_operations/s -1.7% regression Message-ID: <202406252157.348c160f-oliver.sang@intel.com> Content-Type: text/plain; charset="iso-8859-1" Content-Disposition: inline Content-Transfer-Encoding: 8bit X-Mailing-List: linux-block@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Xref: photonic.trudheim.com org.kernel.vger.linux-block:93458 org.kernel.vger.linux-kernel:1258315 Newsgroups: org.kernel.vger.linux-block,dev.linux.lists.oe-lkp,org.kernel.vger.linux-kernel Path: photonic.trudheim.com!nntp.lore.kernel.org!not-for-mail Hello, kernel test robot noticed a -1.7% regression of sysbench-fileio.write_operations/s on: commit: 060406c61c7cb4bbd82a02d179decca9c9bb3443 ("block: add plug while submitting IO") https://git.kernel.org/cgit/linux/kernel/git/torvalds/linux.git master testcase: sysbench-fileio test machine: 128 threads 2 sockets Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz (Ice Lake) with 128G memory parameters: period: 600s nr_threads: 100% disk: 1HDD fs: xfs size: 64G filenum: 1024f rwmode: rndwr iomode: sync cpufreq_governor: performance If you fix the issue in a separate patch/commit (i.e. not just a new version of the same patch/commit), kindly add following tags | Reported-by: kernel test robot | Closes: https://lore.kernel.org/oe-lkp/202406252157.348c160f-oliver.sang@intel.com Details are as below: --------------------------------------------------------------------------------------------------> The kernel config and materials to reproduce are available at: https://download.01.org/0day-ci/archive/20240625/202406252157.348c160f-oliver.sang@intel.com ========================================================================================= compiler/cpufreq_governor/disk/filenum/fs/iomode/kconfig/nr_threads/period/rootfs/rwmode/size/tbox_group/testcase: gcc-13/performance/1HDD/1024f/xfs/sync/x86_64-rhel-8.3/100%/600s/debian-12-x86_64-20240206.cgz/rndwr/64G/lkp-icl-2sp6/sysbench-fileio commit: 3a861560cc ("bcache: fix variable length array abuse in btree_iter") 060406c61c ("block: add plug while submitting IO") 3a861560ccb35f2a 060406c61c7cb4bbd82a02d179d ---------------- --------------------------- %stddev %change %stddev \ | \ 24.53 ± 5% -3.0 21.49 ± 3% mpstat.cpu.all.idle% 52604 ± 70% +99.5% 104921 ± 10% numa-numastat.node1.other_node 24267 ± 4% -10.3% 21772 ± 3% uptime.idle 8218 -1.6% 8084 vmstat.system.cs 24.75 ± 5% -12.3% 21.72 ± 3% iostat.cpu.idle 75.08 +4.0% 78.12 iostat.cpu.iowait 132522 ± 11% +23.4% 163559 ± 6% numa-meminfo.node0.Inactive(file) 237987 ± 6% -15.5% 201059 ± 5% numa-meminfo.node1.Inactive(file) 22.53 ± 11% -6.3 16.22 ± 25% perf-profile.children.cycles-pp.native_flush_tlb_one_user 0.58 ± 29% +0.5 1.05 ± 28% perf-profile.children.cycles-pp.set_pte_vaddr_p4d 22.53 ± 11% -6.3 16.22 ± 25% perf-profile.self.cycles-pp.native_flush_tlb_one_user 0.58 ± 29% +0.5 1.05 ± 28% perf-profile.self.cycles-pp.set_pte_vaddr_p4d 0.03 ± 6% +13.9% 0.03 ± 4% sched_debug.cfs_rq:/.h_nr_running.avg 0.03 ± 6% +13.9% 0.03 ± 4% sched_debug.cfs_rq:/.nr_running.avg -226.99 +41.3% -320.83 sched_debug.cpu.nr_uninterruptible.min 25.90 ± 20% +40.7% 36.44 ± 10% sched_debug.cpu.nr_uninterruptible.stddev 194997 -1.7% 191663 proc-vmstat.nr_dirtied 92615 -1.6% 91119 proc-vmstat.nr_inactive_file 194997 -1.7% 191663 proc-vmstat.nr_written 92615 -1.6% 91119 proc-vmstat.nr_zone_inactive_file 952833 -1.7% 936954 proc-vmstat.pgpgout 1035 -1.4% 1020 sysbench-fileio.fsync_operations/s 140.12 +1.7% 142.56 sysbench-fileio.latency_avg_ms 1559978 -1.7% 1533306 sysbench-fileio.time.file_system_outputs 1497 ± 3% -51.2% 730.50 ± 10% sysbench-fileio.time.involuntary_context_switches 1871907 -1.5% 1843314 sysbench-fileio.time.voluntary_context_switches 1.31 -1.7% 1.29 sysbench-fileio.write_bytes_MB/s 1.25 -1.7% 1.23 sysbench-fileio.write_bytes_MiB/s 80.13 -1.7% 78.74 sysbench-fileio.write_operations/s 71858 ± 7% +18.7% 85272 ± 3% numa-vmstat.node0.nr_dirtied 33121 ± 11% +23.4% 40882 ± 6% numa-vmstat.node0.nr_inactive_file 71858 ± 7% +18.7% 85272 ± 3% numa-vmstat.node0.nr_written 33121 ± 11% +23.4% 40882 ± 6% numa-vmstat.node0.nr_zone_inactive_file 123139 ± 4% -13.6% 106391 ± 2% numa-vmstat.node1.nr_dirtied 59483 ± 6% -15.5% 50254 ± 5% numa-vmstat.node1.nr_inactive_file 123139 ± 4% -13.6% 106391 ± 2% numa-vmstat.node1.nr_written 59483 ± 6% -15.5% 50254 ± 5% numa-vmstat.node1.nr_zone_inactive_file 52604 ± 70% +99.5% 104921 ± 10% numa-vmstat.node1.numa_other 2.55 +4.8% 2.68 perf-stat.i.MPKI 92991408 ± 2% -3.7% 89545806 ± 2% perf-stat.i.branch-instructions 6.14 +0.3 6.41 perf-stat.i.branch-miss-rate% 18893060 -3.6% 18215592 perf-stat.i.cache-references 1.83 +4.9% 1.92 perf-stat.i.cpi 6.848e+08 -2.5% 6.675e+08 ± 2% perf-stat.i.cpu-cycles 214.11 ± 2% -7.2% 198.60 perf-stat.i.cpu-migrations 4.678e+08 ± 2% -3.9% 4.495e+08 ± 2% perf-stat.i.instructions 0.58 -4.3% 0.56 perf-stat.i.ipc 92993821 ± 2% -3.7% 89564915 ± 2% perf-stat.ps.branch-instructions 18876617 -3.6% 18201664 perf-stat.ps.cache-references 6.863e+08 -2.5% 6.692e+08 ± 2% perf-stat.ps.cpu-cycles 213.74 ± 2% -7.2% 198.25 perf-stat.ps.cpu-migrations 4.678e+08 ± 2% -3.9% 4.497e+08 ± 2% perf-stat.ps.instructions 3.003e+11 ± 2% -3.9% 2.887e+11 ± 2% perf-stat.total.instructions Disclaimer: Results have been estimated based on internal Intel analysis and are provided for informational purposes only. Any difference in system hardware or software design or configuration may affect actual performance. -- 0-DAY CI Kernel Test Service https://github.com/intel/lkp-tests/wiki . From: Christoph Hellwig To: Jens Axboe Cc: Damien Le Moal , Niklas Cassel , Song Liu , Yu Kuai , "Martin K. Petersen" , Alim Akhtar , Avri Altman , Bart Van Assche , linux-block@vger.kernel.org, linux-ide@vger.kernel.org, linux-raid@vger.kernel.org, linux-scsi@vger.kernel.org Subject: queue_limits fixups and tidyups v2 Date: Tue, 25 Jun 2024 16:59:45 +0200 Message-ID: <20240625145955.115252-1-hch@lst.de> X-Mailing-List: linux-block@vger.kernel.org List-Id: List-Subscribe: List-Unsubscribe: MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Xref: photonic.trudheim.com org.kernel.vger.linux-block:93460 Newsgroups: org.kernel.vger.linux-block,org.kernel.vger.linux-ide,org.kernel.vger.linux-raid,org.kernel.vger.linux-scsi Path: photonic.trudheim.com!nntp.lore.kernel.org!not-for-mail Hi Jens, this series has a few fixes for the queue_limits conversion in the first few patches and then has a bunch more cleanups and improvements in that area. Changes since v1: - remove an incorrect use of a flag on the features field - add a patch to switch to __bitwise annotations for features and flags - update a commit log Diffstat: block/bio-integrity.c | 2 block/blk-map.c | 2 block/blk-settings.c | 46 +++++---------------- block/blk-sysfs.c | 9 ++-- block/blk.h | 2 block/genhd.c | 2 drivers/ata/libata-scsi.c | 3 - drivers/ata/pata_macio.c | 4 - drivers/md/md.c | 13 +++-- drivers/md/md.h | 1 drivers/md/raid0.c | 2 drivers/md/raid1.c | 2 drivers/md/raid10.c | 2 drivers/md/raid5.c | 2 drivers/scsi/scsi_lib.c | 4 - drivers/ufs/core/ufshcd.c | 9 ++-- include/linux/blkdev.h | 100 +++++++++++++++++++++++----------------------- 17 files changed, 96 insertions(+), 109 deletions(-) .