On Mon, Oct 29, 2018 at 09:04:33AM -0600, Jens Axboe wrote: > On 10/29/18 8:50 AM, Jens Axboe wrote: > > On 10/29/18 6:00 AM, Ming Lei wrote: > >> On Thu, Oct 25, 2018 at 03:10:11PM -0600, Jens Axboe wrote: > >>> The first round of this went into 4.20-rc, but we've still some of > >>> them pending. This patch series converts the remaining drivers to > >>> blk-mq. The ones that support dual paths (like SCSI and DM) have > >>> the non-mq path removed. At the end, legacy IO code and schedulers > >>> are killed off. > >>> > >>> This patch series is on top of my for-linus branch. It can also > >>> be bound in my mq-conversions branch. > >>> > >>> Documentation/block/biodoc.txt | 88 - > >>> Documentation/block/cfq-iosched.txt | 291 -- > >>> Documentation/scsi/scsi-parameters.txt | 5 - > >>> block/Kconfig | 6 - > >>> block/Kconfig.iosched | 61 - > >>> block/Makefile | 5 +- > >>> block/bfq-iosched.c | 1 - > >>> block/blk-cgroup.c | 55 - > >>> block/blk-core.c | 1860 +----------- > >>> block/blk-exec.c | 20 +- > >>> block/blk-flush.c | 154 +- > >>> block/blk-ioc.c | 46 +- > >>> block/blk-merge.c | 35 +- > >>> block/blk-mq-debugfs.c | 2 - > >>> block/blk-mq-tag.c | 6 +- > >>> block/blk-mq.c | 13 +- > >>> block/blk-settings.c | 49 - > >>> block/blk-softirq.c | 20 - > >>> block/blk-sysfs.c | 39 +- > >>> block/blk-tag.c | 378 --- > >>> block/blk-timeout.c | 99 +- > >>> block/blk-wbt.c | 3 +- > >>> block/blk.h | 60 +- > >>> block/bsg-lib.c | 131 +- > >>> block/cfq-iosched.c | 4916 -------------------------------- > >>> block/deadline-iosched.c | 560 ---- > >>> block/elevator.c | 447 +-- > >>> block/kyber-iosched.c | 1 - > >>> block/mq-deadline.c | 1 - > >>> block/noop-iosched.c | 124 - > >>> drivers/block/sunvdc.c | 149 +- > >>> drivers/ide/ide-atapi.c | 25 +- > >>> drivers/ide/ide-cd.c | 175 +- > >>> drivers/ide/ide-disk.c | 5 +- > >>> drivers/ide/ide-io.c | 101 +- > >>> drivers/ide/ide-park.c | 4 +- > >>> drivers/ide/ide-pm.c | 28 +- > >>> drivers/ide/ide-probe.c | 68 +- > >>> drivers/infiniband/ulp/srp/ib_srp.c | 7 - > >>> drivers/md/Kconfig | 11 - > >>> drivers/md/dm-core.h | 10 - > >>> drivers/md/dm-mpath.c | 18 +- > >>> drivers/md/dm-rq.c | 293 +- > >>> drivers/md/dm-rq.h | 4 - > >>> drivers/md/dm-sysfs.c | 3 +- > >>> drivers/md/dm-table.c | 36 +- > >>> drivers/md/dm.c | 21 +- > >>> drivers/md/dm.h | 1 - > >>> drivers/memstick/core/ms_block.c | 110 +- > >>> drivers/memstick/core/ms_block.h | 1 + > >>> drivers/memstick/core/mspro_block.c | 121 +- > >>> drivers/s390/block/dasd_ioctl.c | 22 +- > >>> drivers/scsi/Kconfig | 12 - > >>> drivers/scsi/cxlflash/main.c | 6 - > >>> drivers/scsi/hosts.c | 29 +- > >>> drivers/scsi/lpfc/lpfc_scsi.c | 2 +- > >>> drivers/scsi/osd/osd_initiator.c | 4 +- > >>> drivers/scsi/osst.c | 2 +- > >>> drivers/scsi/qedi/qedi_main.c | 3 +- > >>> drivers/scsi/qla2xxx/qla_os.c | 30 +- > >>> drivers/scsi/scsi.c | 5 +- > >>> drivers/scsi/scsi_debug.c | 3 +- > >>> drivers/scsi/scsi_error.c | 4 +- > >>> drivers/scsi/scsi_lib.c | 624 +--- > >>> drivers/scsi/scsi_priv.h | 1 - > >>> drivers/scsi/scsi_scan.c | 10 +- > >>> drivers/scsi/scsi_sysfs.c | 8 +- > >>> drivers/scsi/scsi_transport_fc.c | 72 +- > >>> drivers/scsi/scsi_transport_iscsi.c | 9 +- > >>> drivers/scsi/scsi_transport_sas.c | 10 +- > >>> drivers/scsi/sg.c | 2 +- > >>> drivers/scsi/st.c | 2 +- > >>> drivers/scsi/ufs/ufshcd.c | 6 - > >>> drivers/target/target_core_pscsi.c | 2 +- > >>> include/linux/blk-cgroup.h | 108 - > >>> include/linux/blkdev.h | 174 +- > >>> include/linux/bsg-lib.h | 3 +- > >>> include/linux/elevator.h | 90 +- > >>> include/linux/ide.h | 13 +- > >>> include/linux/init.h | 1 - > >>> include/scsi/scsi_host.h | 18 +- > >>> include/scsi/scsi_tcq.h | 14 +- > >>> init/do_mounts_initrd.c | 3 - > >>> init/initramfs.c | 6 - > >>> init/main.c | 12 - > >>> 85 files changed, 833 insertions(+), 11144 deletions(-) > >> > >> Hi Jens, > >> > >> Kernel oops[2] is triggered when running elevator switch stress test[1]. > >> > >> [1] https://people.redhat.com/~minlei/tests/tools/elv-switch > >> > >> [2] kernel oops > >> > >> [ 399.702365] scsi 8:0:0:0: Direct-Access Linux scsi_debug 0188 PQ: 0 ANSI: 7 > >> [ 399.703603] sd 8:0:0:0: Power-on or device reset occurred > >> [ 399.709833] sd 8:0:0:0: [sdd] 16384 512-byte logical blocks: (8.39 MB/8.00 MiB) > >> [ 399.711745] sd 8:0:0:0: [sdd] Write Protect is off > >> [ 399.712395] sd 8:0:0:0: [sdd] Mode Sense: 73 00 10 08 > >> [ 399.715119] sd 8:0:0:0: [sdd] Write cache: enabled, read cache: enabled, supports DPO and FUA > >> [ 399.803561] sd 8:0:0:0: [sdd] Attached SCSI disk > >> [ 399.844825] sd 8:0:0:0: [sdd] Synchronizing SCSI cache > >> [ 399.968945] scsi 8:0:0:0: Direct-Access Linux scsi_debug 0188 PQ: 0 ANSI: 7 > >> [ 399.970077] sd 8:0:0:0: Power-on or device reset occurred > >> [ 399.972904] sd 8:0:0:0: [sdd] 16384 512-byte logical blocks: (8.39 MB/8.00 MiB) > >> [ 399.974823] sd 8:0:0:0: [sdd] Write Protect is off > >> [ 399.975411] sd 8:0:0:0: [sdd] Mode Sense: 73 00 10 08 > >> [ 399.978052] sd 8:0:0:0: [sdd] Write cache: enabled, read cache: enabled, supports DPO and FUA > >> [ 400.030588] BUG: unable to handle kernel NULL pointer dereference at 0000000000000500 > >> [ 400.032195] PGD 0 P4D 0 > >> [ 400.032482] Oops: 0000 [#1] PREEMPT SMP PTI > >> [ 400.032945] CPU: 2 PID: 318 Comm: kworker/u8:5 Not tainted 4.19.0_0ffdea6679a9_mq-conversions+ #1 > >> [ 400.033927] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.10.2-2.fc27 04/01/2014 > >> [ 400.034915] Workqueue: events_unbound async_run_entry_fn > >> [ 400.035536] RIP: 0010:bfq_reset_rate_computation+0x52/0x85 > >> [ 400.036177] Code: 00 48 89 83 b0 00 00 00 8b 45 20 c1 e8 09 89 83 d8 00 00 00 48 89 83 d0 00 00 00 eb 0a c7 87 c8 00 00 00 00 00 00 00 48 8b 03 <48> 8b b8 00 05 00 00 48 85 ff 74 24 8b 8b c8 00 00 00 4c 8b 8b d0 > >> [ 400.038300] RSP: 0018:ffffc90000a47c50 EFLAGS: 00010046 > >> [ 400.038915] RAX: 0000000000000000 RBX: ffff88017ac2bc00 RCX: 0000000017d6067a > >> [ 400.039746] RDX: 0000000000000000 RSI: 0000000000000000 RDI: ffff88017ac2bc00 > >> [ 400.040573] RBP: ffff88017ac2bc00 R08: 00000000f461e6ac R09: 0000000000000006 > >> [ 400.041361] R10: 000000000000020f R11: 0000000000000000 R12: ffff88016c130000 > >> [ 400.042127] R13: 0000005d1c095013 R14: ffff88017ac2bf78 R15: 0000000000000020 > >> [ 400.042957] FS: 0000000000000000(0000) GS:ffff88017bb00000(0000) knlGS:0000000000000000 > >> [ 400.043895] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 > >> [ 400.044518] CR2: 0000000000000500 CR3: 0000000179c32004 CR4: 0000000000760ee0 > >> [ 400.045294] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 > >> [ 400.046070] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 > >> [ 400.046843] PKRU: 55555554 > >> [ 400.047144] Call Trace: > >> [ 400.047438] bfq_finish_requeue_request+0x145/0x2fd > >> [ 400.047977] blk_mq_free_request+0x49/0xea > >> [ 400.048429] __scsi_execute+0x16c/0x17b > >> [ 400.048885] scsi_mode_sense+0x113/0x265 > >> [ 400.049356] sd_revalidate_disk+0xb5b/0x109a [sd_mod] > >> [ 400.049947] ? __device_add_disk+0x3d2/0x3f6 > >> [ 400.050415] sd_probe_async+0x10d/0x18b [sd_mod] > >> [ 400.050924] async_run_entry_fn+0x6d/0x12b > >> [ 400.051374] process_one_work+0x1da/0x313 > >> [ 400.051841] ? rescuer_thread+0x282/0x282 > >> [ 400.052313] worker_thread+0x1ca/0x295 > >> [ 400.052769] kthread+0x115/0x11d > >> [ 400.053127] ? kthread_park+0x76/0x76 > >> [ 400.053531] ret_from_fork+0x35/0x40 > >> [ 400.053929] Modules linked in: scsi_debug null_blk isofs iTCO_wdt iTCO_vendor_support i2c_i801 i2c_core lpc_ich mfd_core ip_tables sr_mod cdrom usb_storage sd_mod ahci libahci crc32c_intel libata virtio_scsi qemu_fw_cfg dm_mirror dm_region_hash dm_log dm_mod [last unloaded: null_blk] > >> [ 400.056640] Dumping ftrace buffer: > >> [ 400.057022] (ftrace buffer empty) > >> [ 400.057418] CR2: 0000000000000500 > >> [ 400.057820] ---[ end trace 812cbc0fe8802fdd ]--- > > > > Doesn't look related to the series, more like a bug in BFQ. Are you > > sure it passes on current -git? Or maybe the bug is exposed by the > > series. > > > > I'll try and reproduce it here and see what happens. > > I think I see the error, it is introduced by my series. This incremental > should help, I'm going to test it now. > > > diff --git a/block/blk-ioc.c b/block/blk-ioc.c > index b8afdd6ec4b8..391128456aec 100644 > --- a/block/blk-ioc.c > +++ b/block/blk-ioc.c > @@ -212,6 +212,21 @@ void exit_io_context(struct task_struct *task) > put_io_context_active(ioc); > } > > +static void __ioc_clear_queue(struct list_head *icq_list) > +{ > + unsigned long flags; > + > + while (!list_empty(icq_list)) { > + struct io_cq *icq = list_entry(icq_list->next, > + struct io_cq, q_node); > + struct io_context *ioc = icq->ioc; > + > + spin_lock_irqsave(&ioc->lock, flags); > + ioc_destroy_icq(icq); > + spin_unlock_irqrestore(&ioc->lock, flags); > + } > +} > + > /** > * ioc_clear_queue - break any ioc association with the specified queue > * @q: request_queue being cleared > @@ -225,6 +240,8 @@ void ioc_clear_queue(struct request_queue *q) > spin_lock_irq(q->queue_lock); > list_splice_init(&q->icq_list, &icq_list); > spin_unlock_irq(q->queue_lock); > + > + __ioc_clear_queue(&icq_list); > } > > int create_task_io_context(struct task_struct *task, gfp_t gfp_flags, int node) Not see such issue anymore on today's mq-conversions. -- Ming
