Re: Problem with SPCC 256GB NVMe 1.3 drive - refcount_t: underflow; use-after-free.

On 1/18/21 10:33 AM, Bradley Chapman wrote:

Good afternoon!

On 1/17/21 11:36 PM, Chaitanya Kulkarni wrote:

quoted

On 1/17/21 11:05 AM, Bradley Chapman wrote:

quoted

[ 2836.554298] nvme nvme1: I/O 415 QID 3 timeout, disable controller
[ 2836.672064] blk_update_request: I/O error, dev nvme1n1, sector 16350
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672072] blk_update_request: I/O error, dev nvme1n1, sector 16093
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672074] blk_update_request: I/O error, dev nvme1n1, sector 15836
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672076] blk_update_request: I/O error, dev nvme1n1, sector 15579
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672078] blk_update_request: I/O error, dev nvme1n1, sector 15322
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672080] blk_update_request: I/O error, dev nvme1n1, sector 15065
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672082] blk_update_request: I/O error, dev nvme1n1, sector 14808
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672083] blk_update_request: I/O error, dev nvme1n1, sector 14551
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672085] blk_update_request: I/O error, dev nvme1n1, sector 14294
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672087] blk_update_request: I/O error, dev nvme1n1, sector 14037
op 0x9:(WRITE_ZEROES) flags 0x0 phys_seg 0 prio class 0
[ 2836.672121] nvme nvme1: failed to mark controller live state
[ 2836.672123] nvme nvme1: Removing after probe failure status: -19
[ 2836.689016] Aborting journal on device dm-0-8.
[ 2836.689024] Buffer I/O error on dev dm-0, logical block 25198592,
lost sync page write
[ 2836.689027] JBD2: Error -5 detected when updating journal superblock
for dm-0-8.

Without the knowledge of fs mount/format command I can only suspect that
super
block zeroing issued with write-zeroes request is translated into
REQ_OP_WRITE_ZEROES which controller is not able to process resulting in
the error. This analysis maybe wrong.

Can you please share following details :-

nvme id-ns /dev/nvme0n1 -H (we are interested in oncs part here)

I ran the requested command against /dev/nvme1n1 (since /dev/nvme0n1 
works perfectly so far) and here is the result:

Sorry my bad it suppose to be nvme id-ctrl /dev/nvme0n1 -H

quoted

Also for above device what is the value for the queue block write-zeroes

parameter that is present in the
/sys/block/<nvmeXnY>/queue/write_zeroes_max_bytes ?

$ cat /sys/block/nvme1n1/queue/write_zeroes_max_bytes
131584

So write-zeroes is configured from the setup.

quoted

You can also try blkdiscard -z 0 -l 1024 /dev/<nvmeXnY> to see if the
problem is with
write zeroes.

# blkdiscard -z -l 1024 /dev/nvme1n1
blkdiscard: /dev/nvme1n1: BLKZEROOUT ioctl failed: Device or resource busy

This is exactly what I thought, we need to add a quirk for this model
and make sure
we don't set the write-zeroes support and make blk-lib emulate the
write-zeroes.

quoted

Also can you please also try the latest nvme tree branch nvme-5.11 ?

Where do I get that code from? Is it already in the 5.11-rc tree or do I 
need to look somewhere else? I checked https://github.com/linux-nvme but 
I did not see it there.

Here is the link :-git://git.infradead.org/nvme.git
Branch 5.12.

Brad

_______________________________________________
Linux-nvme mailing list
Linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme