* Page fault on branch nvme-5.6
@ 2020-01-10 19:14 Wunderlich, Mark
2020-01-10 19:24 ` Keith Busch
0 siblings, 1 reply; 3+ messages in thread
From: Wunderlich, Mark @ 2020-01-10 19:14 UTC (permalink / raw)
To: linux-nvme
Ping: Anyone tracking this already? Do not see note on this list, maybe I missed it? Will dig into it more unless already a known issue.
Thanks --- Mark
Just updated to this branch. Performing baseline testing before attempting to make any patch changes.
Fio libaio seemed to work fine. Fio io_uring hitting PF. Attempting I/O to single nvmf tcp device.
fio --filename=/dev/nvme0n1 --time_based --runtime=60 --thread --rw=randrw --rwmixread=100 --refill_buffers --direct=1 --ioengine=io_uring --hipri --fixedbufs --bs=4k --iodepth=1 --iodepth_batch_complete_min=1 --iodepth_batch_complete_max=1 --iodepth_batch=1 --numjobs=1 --group_reporting --gtod_reduce=0 --disable_lat=0 --name=cpu3 --cpus_allowed=3
cpu3: (g=0): rw=randrw, bs=(R) 4096B-4096B, (W) 4096B-4096B, (T) 4096B-4096B, ioengine=io_uring, iodepth=1
Tip of this pulled branch currently:
commit 2fcb73048f5a98e3cc18124f07067825e9f18b9d
Author: Edmund Nadolski <edmund.nadolski@intel.com>
Date: Wed Nov 27 10:17:43 2019 -0700
nvme: remove unused return code from nvme_alloc_ns
The return code of nvme_alloc_ns is never used, so change it
to void.
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Edmund Nadolski <edmund.nadolski@intel.com>
Signed-off-by: Keith Busch mailto:kbusch@kernel.org
PF Info:
[ 415.518095] BUG: kernel NULL pointer dereference, address: 0000000000000018
[ 415.518508] #PF: supervisor read access in kernel mode
[ 415.518809] #PF: error_code(0x0000) - not-present page
[ 415.519110] PGD 800000100bdd7067 P4D 800000100bdd7067 PUD 103b154067 PMD 0
[ 415.519528] Oops: 0000 [#1] SMP PTI
[ 415.519732] CPU: 3 PID: 3665 Comm: fio Tainted: G O 5.5.0-rc3stable+ #2
[ 415.520194] Hardware name: Dell Inc. PowerEdge R740/00WGD1, BIOS 1.4.9 06/29/2018
[ 415.520638] RIP: 0010:io_free_req_many.part.67+0x16e/0x1d0
[ 415.520959] Code: 48 0b a9 01 48 89 da e8 50 ab f6 ff 48 63 53 40 49 8b 46 08 a8 03 75 3e 48 f7 da 65 48 01 10 48 63 4b 40 49 8b 96 f8 00 00 00 <48> 8b 42 18 a8 03 75 3a 48 f7 d9 65 48 01 08 c7 43 44 00 00 00 00
[ 415.522053] RSP: 0018:ffff9f61c901bd90 EFLAGS: 00010213
[ 415.522357] RAX: 00002d3e00009e78 RBX: ffff9f61c901be10 RCX: 0000000000000001
[ 415.522773] RDX: 0000000000000000 RSI: 0000000000000001 RDI: ffff9213bf020e00
[ 415.523189] RBP: ffff9f61c901be18 R08: ffff9223ac3e3160 R09: ffff9f61c901be00
[ 415.523605] R10: 0000000000000352 R11: 071c71c71c71c71c R12: 0000000000000001
[ 415.524020] R13: 0000000000000000 R14: ffff9223add33800 R15: ffff9223add339d0
[ 415.524436] FS: 00007f9f2818b700(0000) GS:ffff9223bf040000(0000) knlGS:0000000000000000
[ 415.524908] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[ 415.525244] CR2: 0000000000000018 CR3: 00000010154e6005 CR4: 00000000007606e0
[ 415.525659] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
[ 415.526075] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
[ 415.526490] PKRU: 55555554
[ 415.526647] Call Trace:
[ 415.526793] io_iopoll_getevents+0x27b/0x2f0
[ 415.527044] ? blkdev_ioctl+0x669/0xa50
[ 415.527272] ? __check_object_size+0x182/0x1d0
[ 415.527531] __io_iopoll_check+0x4b/0xa0
[ 415.527760] __x64_sys_io_uring_enter+0x19c/0x360
[ 415.528038] do_syscall_64+0x5b/0x1b0
[ 415.528255] entry_SYSCALL_64_after_hwframe+0x44/0xa9
_______________________________________________
linux-nvme mailing list
linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: Page fault on branch nvme-5.6
2020-01-10 19:14 Page fault on branch nvme-5.6 Wunderlich, Mark
@ 2020-01-10 19:24 ` Keith Busch
2020-01-10 21:13 ` Wunderlich, Mark
0 siblings, 1 reply; 3+ messages in thread
From: Keith Busch @ 2020-01-10 19:24 UTC (permalink / raw)
To: Wunderlich, Mark; +Cc: linux-nvme
On Fri, Jan 10, 2020 at 07:14:46PM +0000, Wunderlich, Mark wrote:
> Ping: Anyone tracking this already? Do not see note on this list, maybe I missed it? Will dig into it more unless already a known issue.
> Thanks --- Mark
>
> Just updated to this branch. Performing baseline testing before attempting to make any patch changes.
> Fio libaio seemed to work fine. Fio io_uring hitting PF. Attempting I/O to single nvmf tcp device.
Thanks for the notice. Looks like I've used a bad baseline to start the
5.6 nvme series. I'll find a more suitable starting point and push this
branch back out.
_______________________________________________
linux-nvme mailing list
linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
^ permalink raw reply [flat|nested] 3+ messages in thread
* RE: Page fault on branch nvme-5.6
2020-01-10 19:24 ` Keith Busch
@ 2020-01-10 21:13 ` Wunderlich, Mark
0 siblings, 0 replies; 3+ messages in thread
From: Wunderlich, Mark @ 2020-01-10 21:13 UTC (permalink / raw)
To: Keith Busch; +Cc: linux-nvme
> Thanks for the notice. Looks like I've used a bad baseline to start the
> 5.6 nvme series. I'll find a more suitable starting point and push this branch back out.a
Thanks Keith,
I stepped back to 5.5-rc and do not see the PF there.
_______________________________________________
linux-nvme mailing list
linux-nvme@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-nvme
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2020-01-10 21:13 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2020-01-10 19:14 Page fault on branch nvme-5.6 Wunderlich, Mark
2020-01-10 19:24 ` Keith Busch
2020-01-10 21:13 ` Wunderlich, Mark
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).