On 2018/05/31 20:42, Jan Kara wrote: > On Thu 31-05-18 01:00:08, Tetsuo Handa wrote: >> So, we have no idea what is happening... >> Then, what about starting from temporary debug printk() patch shown below? >> >> >From 4f70f72ad3c9ae6ce1678024ef740aca4958e5b0 Mon Sep 17 00:00:00 2001 >> From: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx> >> Date: Wed, 30 May 2018 09:57:10 +0900 >> Subject: [PATCH] bdi: Add temporary config for debugging wb_workfn() versus >> bdi_unregister() race bug. >> >> syzbot is hitting NULL pointer dereference at wb_workfn() [1]. But due to >> limitations that syzbot cannot find reproducer for this bug (frequency is >> once or twice per a day) nor we can't capture vmcore in the environment >> which syzbot is using, for now we need to rely on printk() debugging. >> >> [1] https://syzkaller.appspot.com/bug?id=e0818ccb7e46190b3f1038b0c794299208ed4206 >> >> Signed-off-by: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx> > > Hum a bit ugly solution but if others are fine with this, I can live with > it for a while as well. Or would it be possible for syzkaller to just test > some git tree where this patch is included? Then we would not even have to > have the extra config option... If syzbot can reproduce this bug that way. While it is possible to add/remove git trees syzbot tests, frequently adding/removing trees is bothering. syzbot can enable extra config option. Maybe the config name should be something like CONFIG_DEBUG_FOR_SYZBOT rather than individual topic. I think that syzbot is using many VM instances. I don't know how many instances will be needed for reproducing this bug within reasonable period. More git trees syzbot tests, (I assume that) longer period will be needed for reproducing this bug. The most reliable way is to use the shared part of all trees (i.e. linux.git). > >> diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c >> index 471d863..b4dd078 100644 >> --- a/fs/fs-writeback.c >> +++ b/fs/fs-writeback.c >> @@ -1934,6 +1934,37 @@ void wb_workfn(struct work_struct *work) >> struct bdi_writeback, dwork); >> long pages_written; >> >> +#ifdef CONFIG_BLK_DEBUG_WB_WORKFN_RACE >> + if (!wb->bdi->dev) { >> + pr_warn("WARNING: %s: device is NULL\n", __func__); >> + pr_warn("wb->state=%lx\n", wb->state); >> + pr_warn("list_empty(&wb->work_list)=%u\n", >> + list_empty(&wb->work_list)); >> + if (!wb->bdi) > > This is not possible when we dereferences wb->bdi above... Oops. I missed it. > >> + pr_warn("wb->bdi == NULL\n"); >> + else { >> + pr_warn("list_empty(&wb->bdi->bdi_list)=%u\n", >> + list_empty(&wb->bdi->bdi_list)); >> + pr_warn("wb->bdi->wb.state=%lx\n", wb->bdi->wb.state); >> + } > > It would be also good to print whether wb == wb->bdi->wb (i.e. it is the > default writeback structure or one for some cgroup) and also > wb->bdi->wb.state. > wb->bdi->wb.state is already printed. Updated patch is shown below. Anything else to print? >From 3f3346d42b804e59d12caaa525365a8482505f08 Mon Sep 17 00:00:00 2001 From: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx> Date: Thu, 31 May 2018 22:07:20 +0900 Subject: [PATCH v2] bdi: Add temporary config for debugging wb_workfn() versus bdi_unregister() race bug. syzbot is hitting NULL pointer dereference at wb_workfn() [1]. But due to limitations that syzbot cannot find reproducer for this bug (frequency is once or twice per a day) nor we can't capture vmcore in the environment which syzbot is using, for now we need to rely on printk() debugging. [1] https://syzkaller.appspot.com/bug?id=e0818ccb7e46190b3f1038b0c794299208ed4206 Signed-off-by: Tetsuo Handa <penguin-kernel@xxxxxxxxxxxxxxxxxxx> --- block/Kconfig | 7 +++++++ fs/fs-writeback.c | 28 ++++++++++++++++++++++++++++ 2 files changed, 35 insertions(+) diff --git a/block/Kconfig b/block/Kconfig index 28ec557..fbce13e 100644 --- a/block/Kconfig +++ b/block/Kconfig @@ -139,6 +139,13 @@ config BLK_CMDLINE_PARSER See Documentation/block/cmdline-partition.txt for more information. +config BLK_DEBUG_WB_WORKFN_RACE + bool "Dump upon hitting wb_workfn() versus bdi_unregister() race bug." + default n + ---help--- + This is a temporary option used for obtaining information for + specific bug. This option will be removed after the bug is fixed. + config BLK_WBT bool "Enable support for block device writeback throttling" default n diff --git a/fs/fs-writeback.c b/fs/fs-writeback.c index 471d863..14ab873 100644 --- a/fs/fs-writeback.c +++ b/fs/fs-writeback.c @@ -1934,6 +1934,34 @@ void wb_workfn(struct work_struct *work) struct bdi_writeback, dwork); long pages_written; +#ifdef CONFIG_BLK_DEBUG_WB_WORKFN_RACE + if (!wb->bdi->dev) { + pr_warn("WARNING: %s: device is NULL\n", __func__); + pr_warn("wb->state=%lx\n", wb->state); + pr_warn("(wb == &wb->bdi->wb)=%u\n", wb == &wb->bdi->wb); + pr_warn("list_empty(&wb->work_list)=%u\n", + list_empty(&wb->work_list)); + pr_warn("list_empty(&wb->bdi->bdi_list)=%u\n", + list_empty(&wb->bdi->bdi_list)); + pr_warn("wb->bdi->wb.state=%lx\n", wb->bdi->wb.state); + if (!wb->congested) + pr_warn("wb->congested == NULL\n"); +#ifdef CONFIG_CGROUP_WRITEBACK + else if (!wb->congested->__bdi) + pr_warn("wb->congested->__bdi == NULL\n"); + else { + pr_warn("(wb->congested->__bdi == wb->bdi)=%u\n", + wb->congested->__bdi == wb->bdi); + pr_warn("list_empty(&wb->congested->__bdi->bdi_list)=%u\n", + list_empty(&wb->congested->__bdi->bdi_list)); + pr_warn("wb->congested->__bdi->wb.state=%lx\n", + wb->congested->__bdi->wb.state); + } +#endif + /* Will halt shortly due to NULL pointer dereference... */ + } +#endif + set_worker_desc("flush-%s", dev_name(wb->bdi->dev)); current->flags |= PF_SWAPWRITE; -- 1.8.3.1
