's opinion about tasklets
(https://lore.kernel.org/all/yvovfxmjqaubs...@linutronix.de/).
Thanks,
Bart.
igned int bio_max_size(struct bio *bio);
>>
>> You may want to define bio_max_size() as an inline function in bio.h
>> such that no additional function calls are introduced in the hot path.
>
> I tried, but it is not easy. because request_queue structure of blkdev.h
> should be referred in bio.h. I think it's not good to apply as a inline
> function.
Please don't worry about this. Inlining bio_max_size() is not a big
concern to me.
Thanks,
Bart.
of the 2 patches.
>
> If Théo is "Tejun Heo" ( (maintainer:WORKQUEUE) ), he is already in
> the To: line.
Linus wants to see a "Cc: ${maintainer}" tag in patches that he receives
from a maintainer and that modify another subsystem than the subsystem
maintained by that maintainer.
Thanks,
Bart.
On 4/18/21 2:26 PM, Christophe JAILLET wrote:
> Improve 'create_workqueue', 'create_freezable_workqueue' and
> 'create_singlethread_workqueue' so that they accept a format
> specifier and a variable number of arguments.
>
> This will put these macros more in line with 'alloc_ordered_workqueue' and
o out_err;
>
> - strcpy(name, "mlx5_health");
> - strcat(name, dev_name(dev->device));
> health->wq = create_singlethread_workqueue(name);
> kfree(name);
> if (!health->wq)
Instead of modifying the mlx5 driver, please change the definition of
the create_singlethread_workqueue() such that it accept a format
specifier and a variable number of arguments.
Thanks,
Bart.
tacked drivers for which the bio size has to be
limited than for which the bio size has not to be limited. Hence the
proposal to set t->bio_max_bytes to UINT_MAX in blk_stack_limits() and
to let the stacked driver (e.g. dm-crypt) decide whether or not to lower
that value.
Thanks,
Bart.
- * to avoid deadlock between ufshcd_suspend
> - * and exception event handler.
> - */
> - pm_runtime_get_noresume(hba->dev);
> - }
> + ufshcd_is_exception_event(lrbp->ucd_rsp_ptr))
> + /* Flushed in suspend */
> + schedule_work(&hba->eeh_work);
What makes it safe to leave out the above pm_runtime_get_noresume() call?
Thanks,
Bart.
d63da3af 100644
> --- a/include/linux/bio.h
> +++ b/include/linux/bio.h
> @@ -106,6 +106,8 @@ static inline void *bio_data(struct bio *bio)
> return NULL;
> }
>
> +extern unsigned int bio_max_size(struct bio *bio);
You may want to define bio_max_size() as an inline function in bio.h
such that no additional function calls are introduced in the hot path.
Thanks,
Bart.
ne_mask;
> -}
A similar patch is already in Jens' for-5.13/block branch. See also
commit 540ad3f3da25 ("blk-zoned: Remove the definition of
blk_zone_start()").
Bart.
o
initialize the maximum bio size to UINT_MAX in blk_set_default_limits().
Thanks,
Bart.
As you may know blk_queue_max_hw_sectors() may be
called by any block driver (stacking and non-stacking).
* In blk_stack_limits(), set the maximum BIO size to UINT_MAX. Stacking
block drivers must call blk_stack_limits().
Thanks,
Bart.
_name(&sdev->device->dev), port_num);
> mutex_unlock(&sport->mutex);
> + ret = -EINVAL;
> goto reject;
> }
Reviewed-by: Bart Van Assche
On 4/12/21 11:09 AM, Jason Gunthorpe wrote:
> On Thu, Apr 08, 2021 at 09:50:30AM -0700, Bart Van Assche wrote:
>> On 4/8/21 4:31 AM, Wang Wensheng wrote:
>>> Fix to return a negative error code from the error handling
>>> case instead of 0, as done elsewhere in t
^~
How about adding a Fixes: tag?
Thanks,
Bart.
ret = qla2xxx_get_ini_stats(fc_bsg_to_shost(bsg_job),
> req_data->stat_type,
Since the above looks good to me:
Reviewed-by: Bart Van Assche
hanks,
Bart.
_name(&sdev->device->dev), port_num);
> mutex_unlock(&sport->mutex);
> + ret = -EINVAL;
> goto reject;
> }
Please fix the Hulk Robot. The following code occurs three lines above
the modified code:
ret = -EINVAL;
Thanks,
Bart.
On 4/5/21 11:39 PM, luojiaxing wrote:
On 2021/4/3 0:01, Bart Van Assche wrote:
On 4/2/21 2:08 AM, Luo Jiaxing wrote:
#define AAP1_MEMMAP(r, c) \
- (*(u32 *)((u8*)pm8001_ha->memoryMap.region[AAP1].virt_ptr + (r)
* 32 \
+ (*(u32 *)((u8 *)pm8001_ha->memoryMap.region[AAP1].virt_pt
r,
> &queue_iostats_entry.attr,
> &queue_stable_writes_entry.attr,
> + &queue_limit_bio_size_entry.attr,
> &queue_random_entry.attr,
> &queue_poll_entry.attr,
> &queue_wc_entry.attr,
Has it been considered to introduce a function to set the BIO size limit
instead of introducing a new sysfs attribute? See also
blk_queue_max_hw_sectors().
Thanks,
Bart.
riable]
Has it been verified that the compiler is wrong, or in other words that
this patch does not suppress a useful warning?
Thanks,
Bart.
On 3/12/21 2:55 AM, Lee Jones wrote:
> Fixes the following W=1 kernel build warning(s):
> [ ... ]
Reviewed-by: Bart Van Assche
Card Event Handler ---*/
> -static const char * const rsxx_card_state_to_str(unsigned int state)
> +static const char *rsxx_card_state_to_str(unsigned int state)
> {
> static const char * const state_strings[] = {
> "Unknown", "Shutdown", "Starting", "Formatting",
Reviewed-by: Bart Van Assche
ax_num_meta_sg,
> + u32 access);
> void ib_mr_pool_destroy(struct ib_qp *qp, struct list_head *list);
>
> #endif /* _RDMA_MR_POOL_H */
Does the new 'access' argument only control whether or not PCIe relaxed
ordering is enabled? It seems wrong to me to make enabling of PCIe
relaxed ordering configurable. I think this mechanism should be enabled
unconditionally if the HCA supports it.
Thanks,
Bart.
w image file",FAIL_OPEN_BIOS_FILE},
> {"image header mismatch", FLASH_UPDATE_HDR_ERR},
> {"image offset mismatch", FLASH_UPDATE_OFFSET_ERR},
Can 'flash_error_table' be declared const?
Thanks,
Bart.
please convert it into an inline
function such that the type of the arguments can be verified by the
compiler.
Thanks,
Bart.
, TMR always ends up with completion timeout. Fix it by
> calling blk_mq_start_request() in __ufshcd_issue_tm_cmd().
Reviewed-by: Bart Van Assche
ULL. Changing IS_ERR_OR_NULL() into IS_ERR() and
removing the WARN_ON_ONCE(ch->sess == NULL) may be a better solution.
Thanks,
Bart.
just set req->state to MQ_RQ_IN_FLIGHT
> Without all other irrelevant initializations such as add timeout etc.
Hmm ... I'm not sure that any of the actions performed by
blk_mq_start_request() are irrelevant in this context. Additionally, no
other block or SCSI driver sets MQ_RQ_IN_FLIGHT directly.
Thanks,
Bart.
GE unit attention
after resuming from runtime suspend".
Thanks,
Bart.
an
approach that does not require to introduce a new state variable, e.g.
to send a REQUEST SENSE command after a resume?
Thanks,
Bart.
in VPD page 0x83?
Thanks,
Bart.
bb ("scsi: devinfo:
BLIST_RETRY_ASC_C1 for Fujitsu ETERNUS").
Thanks,
Bart.
;
Please change this WARN_ON() into WARN_ON_ONCE() to prevent a storm of
call traces in the kernel log if this warning statement would be hit
repeatedly.
Thanks,
Bart.
On 3/25/21 6:39 PM, Satya Tangirala wrote:
> On Thu, Mar 25, 2021 at 02:51:31PM -0700, Bart Van Assche wrote:
>> Are you sure that the block layer core splits bios at logical block
>> boundaries? Commit 9cc5169cd478 ("block: Improve physical block
>> alignment of split
nd the above reply.
Thanks,
Bart.
tch series, can the same
effect be achieved by reporting the crypto data unit size as the
physical block size?
Thanks,
Bart.
r with git send-email if
you are not yet familiar with it.
Thanks,
Bart.
only freed if its reference count drops to
zero. The function that frees the partition information,
bdev_free_inode(), is invoked via call_rcu(). bdgrab() fails if the
refcount of a partition is zero. Does that mean that it is not necessary
to call synchronize_rcu() between xa_erase() and put_device()?
Thanks,
Bart.
On 3/21/21 7:35 PM, Ming Lei wrote:
> On Mon, Mar 22, 2021 at 7:03 AM Bart Van Assche wrote:
>>
>> On 3/14/21 4:08 AM, syzbot wrote:
>>> syzbot found the following issue on:
>>>
>>> HEAD commit:280d542f Merge tag 'drm-fixes-2021-03-05&#
On 3/14/21 4:08 AM, syzbot wrote:
> syzbot found the following issue on:
>
> HEAD commit:280d542f Merge tag 'drm-fixes-2021-03-05' of git://anongit..
> git tree: upstream
> console output: https://syzkaller.appspot.com/x/log.txt?x=15ade5aed0
> kernel config: https://syzkaller.appspo
/scm/linux/kernel/git/mkp/scsi.git.
Thanks,
Bart.
inux-block/20210319010009.10041-1-bvanass...@acm.org/)?
Thanks,
Bart.
* resume fails, irq and clocks can be OFF, and powers
> * can be OFF or in LPM.
>*/
Please make better use of the horizontal space in the above comment by
making comment lines longer.
Thanks,
Bart.
Hex data needs to be parsed before it can be used by any software. Has
it been considered to make the "raw" attributes binary attributes
instead of hex-encoded binary? See also sysfs_create_bin_file().
Thanks,
Bart.
On 3/16/21 10:43 AM, John Garry wrote:
> On 16/03/2021 17:00, Bart Van Assche wrote:
>> I agree that Jens asked at the end of 2018 not to touch the fast path
>> to fix this use-after-free (maybe that request has been repeated more
>> recently). If Jens or anyone else fee
odern CPU pipelines store instructions so the performance
impact of adding an additional store should be small.
- Since the block layer has a tendency to reuse tags that have been
freed recently, it is likely that hctx->tags->rqs[rq->tag] will be used
for a next request and hence that it will have to be loaded into the CPU
cache anyway.
Bart.
are added to the block layer code.
- No early return is inserted in blk_mq_tagset_busy_iter().
Thanks,
Bart.
>From a0e534012a766bd6e53cdd466eec0a811164c12a Mon Sep 17 00:00:00 2001
From: Bart Van Assche
Date: Wed, 10 Mar 2021 19:11:47 -0800
Subject: [PATCH] blk-mq: Fix races between iteratin
On 3/10/21 12:52 AM, John Garry wrote:
On 09/03/2021 19:21, Bart Van Assche wrote:
Regarding this patch series, I have shared the feedback I wanted to
share so I would appreciate it if someone else could also take a look.
So I can incorporate any changes and suggestions so far and send a
non
{
+ if (tmp == q)
+ continue;
+ blk_mq_unquiesce_queue(tmp);
+ blk_mq_unfreeze_queue(tmp);
+ }
+ mutex_unlock(&set->tag_list_lock);
}
Reviewed-by: Bart Van Assche
() and
blk_mq_alloc_request_hctx(). So I think all allocated tags are
associated with a request queue.
Regarding this patch series, I have shared the feedback I wanted to
share so I would appreciate it if someone else could also take a look.
Thanks,
Bart.
On 3/8/21 3:17 AM, John Garry wrote:
On 06/03/2021 04:43, Bart Van Assche wrote:
On 3/5/21 7:14 AM, John Garry wrote:
diff --git a/block/blk-mq-tag.c b/block/blk-mq-tag.c
index 7ff1b20d58e7..5950fee490e8 100644
--- a/block/blk-mq-tag.c
+++ b/block/blk-mq-tag.c
@@ -358,11 +358,16 @@ void
On 3/8/21 2:50 AM, John Garry wrote:
Please let me know further thoughts.
Hi John,
My guess is that it is safe to nest these two locks. I was asking
because I had not found any information about the nesting in the patch
description.
Bart.
tags[...]
before the cmpxcg() call and dereferencing it after blk_mq_free_rqs()
has called __free_pages()?
Thanks,
Bart.
chg(&set->iter_usage_counter, 1, 0) != 1);
Isn't it recommended to call cpu_relax() inside busy-waiting loops?
> blk_mq_sched_free_requests(q);
> __elevator_exit(q, e);
>
> + atomic_set(&set->iter_usage_counter, 1);
Can it happen that the above atomic_set() call happens while a
blk_mq_tagset_busy_iter() call is in progress? Should that atomic_set()
call perhaps be changed into an atomic_inc() call?
Thanks,
Bart.
amp;set->tag_list_lock);
> }
This patch introduces nesting of tag_list_lock inside sysfs_lock. The
latter is per request queue while the former can be shared across
multiple request queues. Has it been analyzed whether this is safe?
Thanks,
Bart.
gt; -#ifndef __QLA_TARGET_H
> -#define __QLA_TARGET_H
> -
> +#pragma once
> #include "qla_def.h"
> #include "qla_dsd.h"
Please insert a blank line between #pragma once and the #include directives.
Thanks,
Bart.
dev_err(&sdev->sdev_gendev, "%s failed: %d\n", __func__, ret);
> +
> + trace_ufshcd_wl_suspend(dev_name(dev), ret,
> + ktime_to_us(ktime_sub(ktime_get(), start)),
> + hba->curr_dev_pwr_mode, hba->uic_link_state);
> +
> + return ret;
> +
> +}
Please remove the blank line after the return statement.
Otherwise this patch looks good to me. Hence:
Reviewed-by: Bart Van Assche
\
> kfree(desc_buf);\
Please do not introduce a mode variable but instead introduce a new
attribute such that there is one attribute for the unicode output and
one attribute for the ASCII output. Mode variables are troublesome when
e.g. two scripts try to set the mode attribute concurrently.
Thanks,
Bart.
anagement Input Parameters
> Field Description
> Input Parameter 2 LSB: Task Tag of the task/command operated by the task
> management function.
Thanks for the clarification. Feel free to add the following to this patch:
Reviewed-by: Bart Van Assche
On 1/26/21 8:00 PM, Asutosh Das wrote:
> Resumes the scsi device before accessing it.
>
> Change-Id: I2929af60f2a92c89704a582fcdb285d35b429fde
> Signed-off-by: Asutosh Das
> Signed-off-by: Can Guo
> Signed-off-by: Bao D. Nguyen
No Change-Id tags in upstream patches please.
Thanks,
Bart.
On 1/28/21 9:57 PM, Can Guo wrote:
> On 2021-01-29 11:15, Bart Van Assche wrote:
>> On 1/27/21 8:16 PM, Can Guo wrote:
>>> In __ufshcd_issue_tm_cmd(), it is not right to use hba->nutrs +
>>> req->tag as
>>> the Task Tag in one TMR UPIU. Directly use
of host spin lock, ufshcd_compl_tm() shall not run
> into the scenario where req->end_io_data is set but
> REG_UTP_TASK_REQ_DOOR_BELL
> has not been set. What do you think?
That sounds reasonable to me.
Thanks,
Bart.
flight but also
starts a timer. However, no timeout handler has been defined in
ufshcd_tmf_ops. Should a timeout handler be defined in that data structure?
Thanks,
Bart.
equest completion and timeout handling.
Thanks,
Bart.
hba);
An explanation for this change is missing from the patch description.
Thanks,
Bart.
size_read(bdev->bd_inode);
> +
> + while (bsize < PAGE_SIZE) {
> + if (size & bsize)
> + break;
> + bsize <<= 1;
> + }
> + bdev->bd_inode->i_blkbits = blksize_bits(bsize);
> }
>
> int set_blocksize(struct block_device *bdev, int size)
How can this patch affect write speed? I haven't found any calls of
set_init_blocksize() in the I/O path. Did I perhaps overlook something?
Bart.
blem: In __ufshcd_issue_tm_cmd(), it is not right to use hba->nutrs +
> req->tag as the Task Tag in one TMR UPIU.
> Fix: Directly use req->tag as Task Tag.
Please split this patch into three separate patches - one patch per
problem that has been described above.
Thanks,
Bart.
> ha->fw_info.fw_patch, ha->fw_info.fw_build);
>From the sysfs_emit_at() source code:
WARN(... || at >= PAGE_SIZE, "invalid sysfs_emit_at: buf:%p at:%d\n",
buf, at)
In other words, this patch is wrong. sysfs_emit() should have been used
instead of sysfs_emit_at().
Bart.
On 1/13/21 3:47 AM, Pavel Machek wrote:
>> From: Bart Van Assche
>>
>> [ Upstream commit cfefd9f8240a7b9fdd96fcd54cb029870b6d8d88 ]
>>
>> Disable runtime power management during domain validation. Since a later
>> patch removes RQF_PREEMPT, set RQF_PM for dom
On 1/9/21 7:53 AM, Pavel Begunkov wrote:
> iov_iter_bvec() initialises iterators well, no need to pre-zero it
> beforehand as done in fd_execute_rw_aio(). Compilers can't optimise it
> out and generate extra code for that (confirmed with assembly).
Reviewed-by: Bart Van Assche
On 1/4/21 2:50 PM, Finn Thain wrote:
> On Mon, 4 Jan 2021, Bart Van Assche wrote:
>> Additionally, there is a good alternative available for the sbp driver.
>> Every system I know of that is equipped with a Firewire port also has an
>> Ethernet port. So users who want t
On 6/16/20 7:07 PM, Finn Thain wrote:
> On Tue, 16 Jun 2020, Bart Van Assche wrote:
>> As far as I know the sbp driver only has had one user ever and that user
>> is no longer user the sbp driver.
>
> So, you estimate the userbase at zero. Can you give a confidence level?
On 1/4/21 7:33 AM, John Garry wrote:
> On 23/12/2020 15:47, Bart Van Assche wrote:
>> I propose to change the order in which blk_mq_sched_free_requests(q) and
>> blk_mq_debugfs_unregister(q) are called. Today blk_mq_sched_free_requests(q)
>> is called by blk_cleanup_queue()
p;hba->eh_sem);
> + if (!hba)
> return -EINVAL;
> - }
>
> if (!hba->is_powered || pm_runtime_suspended(hba->dev))
> /*
Hi Can,
How can ufshcd_system_suspend() or ufshcd_system_resume() be called with a
NULL argument? In ufshcd_pci_probe() I see that pci_set_drvdata() is called
before pm_runtime_allow(). ufshcd_pci_remove() calls pm_runtime_forbid().
Thanks,
Bart.
ssages and data that is shared between
kernel and user space (documented in Documentation/target/tcmu-design.rst).
Is one system call per read and per write operation fast enough for all
block-device-in-user-space implementations?
Thanks,
Bart.
debugfs attributes earlier is less risky.
Although this will make it harder to debug lockups that happen while
removing a request queue, kernel developers who are analyzing such an issue
can undo this change in their development kernel tree.
Thanks,
Bart.
it is guaranteed that the tag set
won't disappear, e.g. the call from inside sdk_mq_queue_rq().
How about using a mutex inside blk_mq_queue_tag_busy_iter() instead? As
far as I can see all blk_mq_queue_tag_busy_iter() happen from a context
where it is allowed to sleep.
Thanks,
Bart.
uggestion to serialize blk_mq_tagset_busy_iter() and
blk_mq_free_rqs() looks interesting to me.
Thanks,
Bart.
On 12/21/20 4:06 AM, John Garry wrote:
> On 18/12/2020 22:43, Bart Van Assche wrote:
>> Does this mean that we do not yet have
>> a full explanation about why the above call stack can be triggered?
>
> We understand it, and I'll describe my experiment in detail:
>
rt with freezing the request queue.
blk_mq_freeze_queue() starts with killing q->q_usage_counter and waits
until that counter has dropped to zero. In other words,
blk_mq_queue_tag_busy_iter() should not iterate over a tag set while a tag
set is being freed or reallocated. Does this mean that we do not yet have
a full explanation about why the above call stack can be triggered?
Thanks,
Bart.
| 156.26 |
> | 32 GB | 277.4 | 153.25 |
> +--++-+
Hi Daejun,
What are the units of the numbers in columns 2 and 3?
Thanks,
Bart.
more expensive than the corresponding RCU
primitives. In other words, I would be much happier if this patch
series would use RCU instead of atomics.
Thanks,
Bart.
y is
> chomping at the bit. I'm sending it out in this early state to try and
> figure
> out if it's interesting enough to keep going.
Cc-ing Josef and Mike since their nbd contributions make me wonder
whether this new driver could be useful to their use cases?
Thanks,
Bart.
On 12/8/20 9:55 AM, Alan Stern wrote:
Yes, that certainly is the proper fix. It's all to easy to miss these
issues that depend on your kernel configuration.
Bart, can you fold it into a new version of the patch?
Sure, I will do that.
Thanks,
Bart.
_upiu(dev_name(hba->dev), str, &descp->rsp_header,
> + &descp->output_param1);
> }
Same comment here: please change the type of the 'str' argument in an
enum such that the strcmp() call can be changed into an integer comparison.
Thanks,
Bart.
a->dev), str, &rq->header, &rq->qr);
> + trace_ufshcd_upiu(dev_name(hba->dev), str, &rq_rsp->header,
> + &rq_rsp->qr);
> }
Please change the 'str' argument into an enum and let
ufshcd_add_query_upiu_trace() do the enum-to-string conversion. That
will allow to convert the strcmp() call into an integer comparison.
Thanks,
Bart.
On 11/29/20 11:04 PM, Hannes Reinecke wrote:
> On 11/26/20 5:49 PM, John Garry wrote:
>> On 26/11/2020 16:27, Bart Van Assche wrote:
>>> On 11/26/20 7:02 AM, Pradeep P V K wrote:
>>>> Observes below crash while accessing (use-after-free) request queue
atic inline void __blk_mq_put_driver_tag(struct blk_mq_hw_ctx *hctx,
> struct request *rq)
> {
> + hctx->tags->rqs[rq->tag] = NULL;
> blk_mq_put_tag(hctx->tags, rq->mq_ctx, rq->tag);
> rq->tag = BLK_MQ_NO_TAG;
Is this perhaps a block driver bug instead of a block layer core bug? If
this would be a block layer core bug, it would have been reported before.
Bart.
e into the power management core (drivers/base/power)?
Thanks,
Bart.
cessary. Maybe these calls are a leftover from a time
when runtime suspended devices were not resumed automatically by the
block layer? Anyway, I'm fine with removing these calls.
Thanks,
Bart.
Please fix the test bot. The DID_ERROR messages are reported during test
block/001 and in the attached dmesg output I found the following:
block/001 (stress device hotplugging)[passed]
Bart.
ve won't compile if CONFIG_PM=n. How about adding a function in
include/linux/blk-pm.h to check whether or not runtime PM has been enabled?
Otherwise this patch looks good to me.
Thanks,
Bart.
On 11/12/20 9:20 AM, Jason Gunthorpe wrote:
I think it should be like this, Bart?
diff --git a/drivers/infiniband/ulp/srpt/ib_srpt.c
b/drivers/infiniband/ulp/srpt/ib_srpt.c
index 6017d525084a0c..80f9673956ced2 100644
--- a/drivers/infiniband/ulp/srpt/ib_srpt.c
+++ b/drivers/infiniband/ulp/srpt
On 10/27/20 2:51 AM, Mauro Carvalho Chehab wrote:
> Add a description for it.
It seems like my Reviewed-by tag is missing? See also
https://lore.kernel.org/linux-rdma/cb5944e2-bdea-e320-d4d1-2f9bc9539...@acm.org/
Bart.
dditionally, since this functionality can be implemented easily in user
space, I think it should be implemented in user space instead of in the
kernel. Even when not using systemd or udev, it is easy to implement a loop
that waits until a certain LUN or WWID appears.
Bart.
N]) directly from kernel.
Please use udev or systemd instead of adding code in the UFS driver that is
not necessary when udev or systemd is used.
Thanks,
Bart.
Number of CQEs in @cq.
> * @zw_cqe: Zero-length write CQE.
> * @rcu: RCU head.
> * @kref: kref for this channel.
Reviewed-by: Bart Van Assche
y wrong to me. No code should depend on the order in
which LUNs are scanned. Please use the soft links created by udev instead
of serializing LUN scanning.
Thanks,
Bart.
Size of the @cq pool.
> * @zw_cqe: Zero-length write CQE.
> * @rcu: RCU head.
> * @kref: kref for this channel.
That doesn't seem correct to me. My understanding is that cq_size is the
number of CQEs in @cq. @cq is a completion queue and not a CQ pool.
Bart.
/repro.syz?x=10ce0a8b90
>
> The issue was bisected to:
>
> commit 2004bfdef945fe55196db6b9cdf321fbc75bb0de
> Author: Bart Van Assche
> Date: Tue Mar 10 04:26:21 2020 +
>
> null_blk: Fix the null_add_dev() error path
#syz wrong-bisect
1 - 100 of 1026 matches
Mail list logo
