kernel/linux.git/block/blk-iocost.c, branch v7.1

blk-iocost: fix busy_level reset when no IOs complete

2026-03-31T19:56:38+00:00

When a disk is saturated, it is common for no IOs to complete within a timer period. Currently, in this case, rq_wait_pct and missed_ppm are calculated as 0, the iocost incorrectly interprets this as meeting QoS targets and resets busy_level to 0. This reset prevents busy_level from reaching the threshold (4) needed to reduce vrate. On certain cloud storage, such as Azure Premium SSD, we observed that iocost may fail to reduce vrate for tens of seconds during saturation, failing to mitigate noisy neighbor issues. Fix this by tracking the number of IO completions (nr_done) in a period. If nr_done is 0 and there are lagging IOs, the saturation status is unknown, so we keep busy_level unchanged. The issue is consistently reproducible on Azure Standard_D8as_v5 (Dasv5) VMs with 512GB Premium SSD (P20) using the script below. It was not observed on GCP n2d VMs (with 100G pd-ssd and 1.5T local-ssd), and no regressions were found with this patch. In this script, cgA performs large IOs with iodepth=128, while cgB performs small IOs with iodepth=1 rate_iops=100 rw=randrw. With iocost enabled, we expect it to throttle cgA, the submission latency (slat) of cgA should be significantly higher, cgB can reach 200 IOPS and the completion latency (clat) should below. BLK_DEVID="8:0" MODEL="rbps=173471131 rseqiops=3566 rrandiops=3566 wbps=173333269 wseqiops=3566 wrandiops=3566" QOS="rpct=90 rlat=3500 wpct=90 wlat=3500 min=80 max=10000" echo "$BLK_DEVID ctrl=user model=linear $MODEL" > /sys/fs/cgroup/io.cost.model echo "$BLK_DEVID enable=1 ctrl=user $QOS" > /sys/fs/cgroup/io.cost.qos CG_A="/sys/fs/cgroup/cgA" CG_B="/sys/fs/cgroup/cgB" FILE_A="/path/to/sda/A.fio.testfile" FILE_B="/path/to/sda/B.fio.testfile" RESULT_DIR="./iocost_results_$(date +%Y%m%d_%H%M%S)" mkdir -p "$CG_A" "$CG_B" "$RESULT_DIR" get_result() { local file=$1 local label=$2 local results=$(jq -r ' .jobs[0].mixed | ( .iops | tonumber | round ) as $iops | ( .bw_bytes / 1024 / 1024 ) as $bps | ( .slat_ns.mean / 1000000 ) as $slat | ( .clat_ns.mean / 1000000 ) as $avg | ( .clat_ns.max / 1000000 ) as $max | ( .clat_ns.percentile["90.000000"] / 1000000 ) as $p90 | ( .clat_ns.percentile["99.000000"] / 1000000 ) as $p99 | ( .clat_ns.percentile["99.900000"] / 1000000 ) as $p999 | ( .clat_ns.percentile["99.990000"] / 1000000 ) as $p9999 | "\($iops)|\($bps)|\($slat)|\($avg)|\($max)|\($p90)|\($p99)|\($p999)|\($p9999)" ' "$file") IFS='|' read -r iops bps slat avg max p90 p99 p999 p9999 <<<"$results" printf "%-8s %-6s %-7.2f %-8.2f %-8.2f %-8.2f %-8.2f %-8.2f %-8.2f %-8.2f\n" \ "$label" "$iops" "$bps" "$slat" "$avg" "$max" "$p90" "$p99" "$p999" "$p9999" } run_fio() { local cg_path=$1 local filename=$2 local name=$3 local bs=$4 local qd=$5 local out=$6 shift 6 local extra=$@ ( pid=$(sh -c 'echo $PPID') echo $pid >"${cg_path}/cgroup.procs" fio --name="$name" --filename="$filename" --direct=1 --rw=randrw --rwmixread=50 \ --ioengine=libaio --bs="$bs" --iodepth="$qd" --size=4G --runtime=10 \ --time_based --group_reporting --unified_rw_reporting=mixed \ --output-format=json --output="$out" $extra >/dev/null 2>&1 ) & } echo "Starting Test ..." for bs_b in "4k" "32k" "256k"; do echo "Running iteration: BS=$bs_b" out_a="${RESULT_DIR}/cgA_1m.json" out_b="${RESULT_DIR}/cgB_${bs_b}.json" # cgA: Heavy background (BS 1MB, QD 128) run_fio "$CG_A" "$FILE_A" "cgA" "1m" 128 "$out_a" # cgB: Latency sensitive (Variable BS, QD 1, Read/Write IOPS limit 100) run_fio "$CG_B" "$FILE_B" "cgB" "$bs_b" 1 "$out_b" "--rate_iops=100" wait SUMMARY_DATA+="$(get_result "$out_a" "cgA-1m")"$'\n' SUMMARY_DATA+="$(get_result "$out_b" "cgB-$bs_b")"$'\n\n' done echo -e "\nFinal Results Summary:\n" printf "%-8s %-6s %-7s %-8s %-8s %-8s %-8s %-8s %-8s %-8s\n" \ "" "" "" "slat" "clat" "clat" "clat" "clat" "clat" "clat" printf "%-8s %-6s %-7s %-8s %-8s %-8s %-8s %-8s %-8s %-8s\n\n" \ "CGROUP" "IOPS" "MB/s" "avg(ms)" "avg(ms)" "max(ms)" "P90(ms)" "P99" "P99.9" "P99.99" echo "$SUMMARY_DATA" echo "Results saved in $RESULT_DIR" Before: slat clat clat clat clat clat clat CGROUP IOPS MB/s avg(ms) avg(ms) max(ms) P90(ms) P99 P99.9 P99.99 cgA-1m 166 166.37 3.44 748.95 1298.29 977.27 1233.13 1300.23 1300.23 cgB-4k 5 0.02 0.02 181.74 761.32 742.39 759.17 759.17 759.17 cgA-1m 167 166.51 1.98 748.68 1549.41 809.50 1451.23 1551.89 1551.89 cgB-32k 6 0.18 0.02 169.98 761.76 742.39 759.17 759.17 759.17 cgA-1m 166 165.55 2.89 750.89 1540.37 851.44 1451.23 1535.12 1535.12 cgB-256k 5 1.30 0.02 191.35 759.51 750.78 759.17 759.17 759.17 After: slat clat clat clat clat clat clat CGROUP IOPS MB/s avg(ms) avg(ms) max(ms) P90(ms) P99 P99.9 P99.99 cgA-1m 162 162.48 6.14 749.69 850.02 826.28 834.67 843.06 851.44 cgB-4k 199 0.78 0.01 1.95 42.12 2.57 7.50 34.87 42.21 cgA-1m 146 146.20 6.83 833.04 908.68 893.39 901.78 910.16 910.16 cgB-32k 200 6.25 0.01 2.32 31.40 3.06 7.50 16.58 31.33 cgA-1m 110 110.46 9.04 1082.67 1197.91 1182.79 1199.57 1199.57 1199.57 cgB-256k 200 49.98 0.02 3.69 22.20 4.88 9.11 20.05 22.15 Signed-off-by: Jialin Wang Acked-by: Tejun Heo Link: https://patch.msgid.link/20260331100509.182882-1-wjl.linux@gmail.com Signed-off-by: Jens Axboe

Convert 'alloc_obj' family to use the new default GFP_KERNEL argument

2026-02-22T01:09:51+00:00

This was done entirely with mindless brute force, using git grep -l '\

treewide: Replace kmalloc with kmalloc_obj for non-scalar types

2026-02-21T09:02:28+00:00

This is the result of running the Coccinelle script from scripts/coccinelle/api/kmalloc_objs.cocci. The script is designed to avoid scalar types (which need careful case-by-case checking), and instead replace kmalloc-family calls that allocate struct or union object instances: Single allocations: kmalloc(sizeof(TYPE), ...) are replaced with: kmalloc_obj(TYPE, ...) Array allocations: kmalloc_array(COUNT, sizeof(TYPE), ...) are replaced with: kmalloc_objs(TYPE, COUNT, ...) Flex array allocations: kmalloc(struct_size(PTR, FAM, COUNT), ...) are replaced with: kmalloc_flex(*PTR, FAM, COUNT, ...) (where TYPE may also be *VAR) The resulting allocations no longer return "void *", instead returning "TYPE *". Signed-off-by: Kees Cook

block: introduce blk_queue_rot()

2026-01-29T20:15:50+00:00

To check if a request queue is for a rotational device, a double negation is needed with the pattern "!blk_queue_nonrot(q)". Simplify this with the introduction of the helper blk_queue_rot() which tests if a requests queue limit has the BLK_FEAT_ROTATIONAL feature set. All call sites of blk_queue_nonrot() are modified to use blk_queue_rot() and blk_queue_nonrot() definition removed. No functional changes. Signed-off-by: Damien Le Moal Reviewed-by: Nitesh Shetty Reviewed-by: Martin K. Petersen Signed-off-by: Jens Axboe

block: use min() instead of min_t()

2025-11-20T14:44:29+00:00

min_t(unsigned int, a, b) casts an 'unsigned long' to 'unsigned int'. Use min(a, b) instead as it promotes any 'unsigned int' to 'unsigned long' and so cannot discard significant bits. In this case the 'unsigned long' value is small enough that the result is ok. (Similarly for max_t() and clamp_t().) Detected by an extra check added to min_t(). Signed-off-by: David Laight Signed-off-by: Jens Axboe

Merge tag 'for-6.15/block-20250322' of git://git.kernel.dk/linux

2025-03-27T01:08:55+00:00

Pull block updates from Jens Axboe: - Fixes for integrity handling - NVMe pull request via Keith: - Secure concatenation for TCP transport (Hannes) - Multipath sysfs visibility (Nilay) - Various cleanups (Qasim, Baruch, Wang, Chen, Mike, Damien, Li) - Correct use of 64-bit BARs for pci-epf target (Niklas) - Socket fix for selinux when used in containers (Peijie) - MD pull request via Yu: - fix recovery can preempt resync (Li Nan) - fix md-bitmap IO limit (Su Yue) - fix raid10 discard with REQ_NOWAIT (Xiao Ni) - fix raid1 memory leak (Zheng Qixing) - fix mddev uaf (Yu Kuai) - fix raid1,raid10 IO flags (Yu Kuai) - some refactor and cleanup (Yu Kuai) - Series cleaning up and fixing bugs in the bad block handling code - Improve support for write failure simulation in null_blk - Various lock ordering fixes - Fixes for locking for debugfs attributes - Various ublk related fixes and improvements - Cleanups for blk-rq-qos wait handling - blk-throttle fixes - Fixes for loop dio and sync handling - Fixes and cleanups for the auto-PI code - Block side support for hardware encryption keys in blk-crypto - Various cleanups and fixes * tag 'for-6.15/block-20250322' of git://git.kernel.dk/linux: (105 commits) nvmet: replace max(a, min(b, c)) by clamp(val, lo, hi) nvme-tcp: fix selinux denied when calling sock_sendmsg nvmet: pci-epf: Always configure BAR0 as 64-bit nvmet: Remove duplicate uuid_copy nvme: zns: Simplify nvme_zone_parse_entry() nvmet: pci-epf: Remove redundant 'flush_workqueue()' calls nvmet-fc: Remove unused functions nvme-pci: remove stale comment nvme-fc: Utilise min3() to simplify queue count calculation nvme-multipath: Add visibility for queue-depth io-policy nvme-multipath: Add visibility for numa io-policy nvme-multipath: Add visibility for round-robin io-policy nvmet: add tls_concat and tls_key debugfs entries nvmet-tcp: support secure channel concatenation nvmet: Add 'sq' argument to alloc_ctrl_args nvme-fabrics: reset admin connection for secure concatenation nvme-tcp: request secure channel concatenation nvme-keyring: add nvme_tls_psk_refresh() nvme: add nvme_auth_derive_tls_psk() nvme: add nvme_auth_generate_digest() ...

block/blk-iocost: ensure 'ret' is set on error

2025-03-19T20:51:36+00:00

In case blkg_conf_open_bdev_frozen() fails, ioc_qos_write() jumps to the error path without assigning a value to 'ret'. Ensure that it inherits the error from the passed back error value. Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-kbuild-all/202503200454.QWpwKeJu-lkp@intel.com/ Fixes: 9730763f4756 ("block: correct locking order for protecting blk-wbt parameters") Signed-off-by: Jens Axboe

block: correct locking order for protecting blk-wbt parameters

2025-03-19T17:35:45+00:00

The commit '245618f8e45f ("block: protect wbt_lat_usec using q-> elevator_lock")' introduced q->elevator_lock to protect updates to blk-wbt parameters when writing to the sysfs attribute wbt_ lat_usec and the cgroup attribute io.cost.qos. However, both these attributes also acquire q->rq_qos_mutex, leading to the following lockdep warning: ====================================================== WARNING: possible circular locking dependency detected 6.14.0-rc5+ #138 Not tainted ------------------------------------------------------ bash/5902 is trying to acquire lock: c000000085d495a0 (&q->rq_qos_mutex){+.+.}-{4:4}, at: wbt_init+0x164/0x238 but task is already holding lock: c000000085d498c8 (&q->elevator_lock){+.+.}-{4:4}, at: queue_wb_lat_store+0xb0/0x20c which lock already depends on the new lock. the existing dependency chain (in reverse order) is: -> #1 (&q->elevator_lock){+.+.}-{4:4}: __mutex_lock+0xf0/0xa58 ioc_qos_write+0x16c/0x85c cgroup_file_write+0xc4/0x32c kernfs_fop_write_iter+0x1b8/0x29c vfs_write+0x410/0x584 ksys_write+0x84/0x140 system_call_exception+0x134/0x360 system_call_vectored_common+0x15c/0x2ec -> #0 (&q->rq_qos_mutex){+.+.}-{4:4}: __lock_acquire+0x1b6c/0x2ae0 lock_acquire+0x140/0x430 __mutex_lock+0xf0/0xa58 wbt_init+0x164/0x238 queue_wb_lat_store+0x1dc/0x20c queue_attr_store+0x12c/0x164 sysfs_kf_write+0x6c/0xb0 kernfs_fop_write_iter+0x1b8/0x29c vfs_write+0x410/0x584 ksys_write+0x84/0x140 system_call_exception+0x134/0x360 system_call_vectored_common+0x15c/0x2ec other info that might help us debug this: Possible unsafe locking scenario: CPU0 CPU1 ---- ---- lock(&q->elevator_lock); lock(&q->rq_qos_mutex); lock(&q->elevator_lock); lock(&q->rq_qos_mutex); *** DEADLOCK *** 6 locks held by bash/5902: #0: c000000051122400 (sb_writers#3){.+.+}-{0:0}, at: ksys_write+0x84/0x140 #1: c00000007383f088 (&of->mutex#2){+.+.}-{4:4}, at: kernfs_fop_write_iter+0x174/0x29c #2: c000000008550428 (kn->active#182){.+.+}-{0:0}, at: kernfs_fop_write_iter+0x180/0x29c #3: c000000085d493a8 (&q->q_usage_counter(io)#5){++++}-{0:0}, at: blk_mq_freeze_queue_nomemsave+0x28/0x40 #4: c000000085d493e0 (&q->q_usage_counter(queue)#5){++++}-{0:0}, at: blk_mq_freeze_queue_nomemsave+0x28/0x40 #5: c000000085d498c8 (&q->elevator_lock){+.+.}-{4:4}, at: queue_wb_lat_store+0xb0/0x20c stack backtrace: CPU: 17 UID: 0 PID: 5902 Comm: bash Kdump: loaded Not tainted 6.14.0-rc5+ #138 Hardware name: IBM,9043-MRX POWER10 (architected) 0x800200 0xf000006 of:IBM,FW1060.00 (NM1060_028) hv:phyp pSeries Call Trace: [c0000000721ef590] [c00000000118f8a8] dump_stack_lvl+0x108/0x18c (unreliable) [c0000000721ef5c0] [c00000000022563c] print_circular_bug+0x448/0x604 [c0000000721ef670] [c000000000225a44] check_noncircular+0x24c/0x26c [c0000000721ef740] [c00000000022bf28] __lock_acquire+0x1b6c/0x2ae0 [c0000000721ef870] [c000000000229240] lock_acquire+0x140/0x430 [c0000000721ef970] [c0000000011cfbec] __mutex_lock+0xf0/0xa58 [c0000000721efaa0] [c00000000096c46c] wbt_init+0x164/0x238 [c0000000721efaf0] [c0000000008f8cd8] queue_wb_lat_store+0x1dc/0x20c [c0000000721efb50] [c0000000008f8fa0] queue_attr_store+0x12c/0x164 [c0000000721efc60] [c0000000007c11cc] sysfs_kf_write+0x6c/0xb0 [c0000000721efca0] [c0000000007bfa4c] kernfs_fop_write_iter+0x1b8/0x29c [c0000000721efcf0] [c0000000006a281c] vfs_write+0x410/0x584 [c0000000721efdc0] [c0000000006a2cc8] ksys_write+0x84/0x140 [c0000000721efe10] [c000000000031b64] system_call_exception+0x134/0x360 [c0000000721efe50] [c00000000000cedc] system_call_vectored_common+0x15c/0x2ec >From the above log it's apparent that method which writes to sysfs attr wbt_lat_usec acquires q->elevator_lock first, and then acquires q->rq_ qos_mutex. However the another method which writes to io.cost.qos, acquires q->rq_qos_mutex first, and then acquires q->rq_qos_mutex. So this could potentially cause the deadlock. A closer look at ioc_qos_write shows that correcting the lock order is non-trivial because q->rq_qos_mutex is acquired in blkg_conf_open_bdev and released in blkg_conf_exit. The function blkg_conf_open_bdev is responsible for parsing user input and finding the corresponding block device (bdev) from the user provided major:minor number. Since we do not know the bdev until blkg_conf_open_bdev completes, we cannot simply move q->elevator_lock acquisition before blkg_conf_open_ bdev. So to address this, we intoduce new helpers blkg_conf_open_bdev_ frozen and blkg_conf_exit_frozen which are just wrappers around blkg_ conf_open_bdev and blkg_conf_exit respectively. The helper blkg_conf_ open_bdev_frozen is similar to blkg_conf_open_bdev, but additionally freezes the queue, acquires q->elevator_lock and ensures the correct locking order is followed between q->elevator_lock and q->rq_qos_mutex. Similarly another helper blkg_conf_exit_frozen in addition to unfreezing the queue ensures that we release the locks in correct order. By using these helpers, now we maintain the same locking order in all code paths where we update blk-wbt parameters. Fixes: 245618f8e45f ("block: protect wbt_lat_usec using q->elevator_lock") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-lkp/202503171650.cc082b66-lkp@intel.com Signed-off-by: Nilay Shroff Link: https://lore.kernel.org/r/20250319105518.468941-3-nilay@linux.ibm.com Signed-off-by: Jens Axboe

block: release q->elevator_lock in ioc_qos_write

2025-03-19T17:35:45+00:00

The ioc_qos_write method acquires q->elevator_lock to protect updates to blk-wbt parameters. Once these updates are complete, the lock should be released before returning from ioc_qos_write. However, in one code path, the release of q->elevator_lock was mistakenly omitted, potentially leading to a lock leak. This commit fixes the issue by ensuring that q->elevator_lock is properly released in all return paths of ioc_qos_write. Fixes: 245618f8e45f ("block: protect wbt_lat_usec using q->elevator_lock") Reported-by: kernel test robot Closes: https://lore.kernel.org/oe-lkp/202503171650.cc082b66-lkp@intel.com Signed-off-by: Nilay Shroff Reviewed-by: Ming Lei Link: https://lore.kernel.org/r/20250319105518.468941-2-nilay@linux.ibm.com Signed-off-by: Jens Axboe

block: protect wbt_lat_usec using q->elevator_lock

2025-03-10T13:30:18+00:00

The wbt latency and state could be updated while initializing the elevator or exiting the elevator. It could be also updated while configuring IO latency QoS parameters using cgroup. The elevator code path is now protected with q->elevator_lock. So we should protect the access to sysfs attribute wbt_lat_usec using q->elevator _lock instead of q->sysfs_lock. White we're at it, also protect ioc_qos_write(), which configures wbt parameters via cgroup, using q->elevator_lock. Reviewed-by: Christoph Hellwig Reviewed-by: Hannes Reinecke Reviewed-by: Ming Lei Signed-off-by: Nilay Shroff Link: https://lore.kernel.org/r/20250304102551.2533767-7-nilay@linux.ibm.com Signed-off-by: Jens Axboe