From: Leon Romanovsky <leon@kernel.org>
To: Li Zhijian <lizhijian@fujitsu.com>
Cc: Yanjun Zhu <yanjun.zhu@linux.dev>, Jason Gunthorpe <jgg@ziepe.ca>,
linux-rdma@vger.kernel.org, Bob Pearson <rpearsonhpe@gmail.com>,
Haris Iqbal <haris.iqbal@ionos.com>,
Guoqing Jiang <guoqing.jiang@linux.dev>
Subject: Re: [PATCH for-next v2] Revert "RDMA/rxe: Create duplicate mapping tables for FMRs"
Date: Wed, 27 Jul 2022 12:26:12 +0300 [thread overview]
Message-ID: <YuEEtGW0BPUoPnr3@unreal> (raw)
In-Reply-To: <1658805386-2-1-git-send-email-lizhijian@fujitsu.com>
On Tue, Jul 26, 2022 at 03:16:26AM +0000, Li Zhijian wrote:
> Below 2 commits will be reverted:
> commit 8ff5f5d9d8cf ("RDMA/rxe: Prevent double freeing rxe_map_set()")
> commit 647bf13ce944 ("RDMA/rxe: Create duplicate mapping tables for FMRs")
>
> The community has a few bug reports which pointed this commit at last.
> Some proposals are raised up in the meantime but all of them have no
> follow-up operation.
>
> The previous commit led the map_set of FMR to be not available any more if
> the MR is registered again after invalidating. Although the mentioned
> patch try to fix a potential race in building/accessing the same table
> for fast memory regions, it broke rtrs etc ULPs. Since the latter could
> be worse, revert this patch.
>
> With previous commit, it's observed that a same MR in rnbd server will
> trigger below code path:
> -> rxe_mr_init_fast()
> |-> alloc map_set() # map_set is uninitialized
> |...-> rxe_map_mr_sg() # build the map_set
> |-> rxe_mr_set_page()
> |...-> rxe_reg_fast_mr() # mr->state change to VALID from FREE that means
> # we can access host memory(such rxe_mr_copy)
> |...-> rxe_invalidate_mr() # mr->state change to FREE from VALID
> |...-> rxe_reg_fast_mr() # mr->state change to VALID from FREE,
> # but map_set was not built again
> |...-> rxe_mr_copy() # kernel crash due to access wild addresses
> # that lookup from the map_set
>
> The backtraces are not always identical.
> [1st]----------
> [ 80.163579] RIP: 0010:lookup_iova+0x66/0xa0 [rdma_rxe]
> [ 80.164825] Code: 00 00 00 48 d3 ee 89 32 c3 4c 8b 18 49 8b 3b 48 8b 47 08 48 39 c6 72 38 48 29 c6 45 31 d2 b8 01 00 00 00 48 63 c8 48 c1 e1 04 <48> 8b 4c 0f 08 48 39 f1 77 21 83 c0 01 48 29 ce 3d 00 01 00 00 75
> [ 80.168935] RSP: 0018:ffffb7ff80063bf0 EFLAGS: 00010246
> [ 80.170333] RAX: 0000000000000000 RBX: ffff9b9949d86800 RCX: 0000000000000000
> [ 80.171976] RDX: ffffb7ff80063c00 RSI: 0000000049f6b378 RDI: 002818da00000004
> [ 80.173606] RBP: 0000000000000120 R08: ffffb7ff80063c08 R09: ffffb7ff80063c04
> [ 80.176933] R10: 0000000000000002 R11: ffff9b9916f7eef8 R12: ffff9b99488a0038
> [ 80.178526] R13: ffff9b99488a0038 R14: ffff9b9914fb346a R15: ffff9b990ab27000
> [ 80.180378] FS: 0000000000000000(0000) GS:ffff9b997dc00000(0000) knlGS:0000000000000000
> [ 80.182257] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 80.183577] CR2: 00007efc33a98ed0 CR3: 0000000014f32004 CR4: 00000000001706f0
> [ 80.185210] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> [ 80.186890] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> [ 80.188517] Call Trace:
> [ 80.189269] <TASK>
> [ 80.189949] rxe_mr_copy.part.0+0x6f/0x140 [rdma_rxe]
> [ 80.191173] rxe_responder+0x12ee/0x1b60 [rdma_rxe]
> [ 80.192409] ? rxe_icrc_check+0x7e/0x100 [rdma_rxe]
> [ 80.193576] ? rxe_rcv+0x1d0/0x780 [rdma_rxe]
> [ 80.194668] ? rxe_icrc_hdr.isra.0+0xf6/0x160 [rdma_rxe]
> [ 80.195952] rxe_do_task+0x67/0xb0 [rdma_rxe]
> [ 80.197081] rxe_xmit_packet+0xc7/0x210 [rdma_rxe]
> [ 80.198253] rxe_requester+0x680/0xee0 [rdma_rxe]
> [ 80.199439] ? update_load_avg+0x5f/0x690
> [ 80.200530] ? update_load_avg+0x5f/0x690
> [ 80.213968] ? rtrs_clt_recv_done+0x1b/0x30 [rtrs_client]
>
> [2nd]----------
> [ 5213.049494] RIP: 0010:rxe_mr_copy.part.0+0xa8/0x140 [rdma_rxe]
> [ 5213.050978] Code: 00 00 49 c1 e7 04 48 8b 00 4c 8d 2c d0 48 8b 44 24 10 4d 03 7d 00 85 ed 7f 10 eb 6c 89 54 24 0c 49 83 c7 10 31 c0 85 ed 7e 5e <49> 8b 3f 8b 14 24 4c 89 f6 48 01 c7 85 d2 74 06 48 89 fe 4c 89 f7
> [ 5213.056463] RSP: 0018:ffffae3580063bf8 EFLAGS: 00010202
> [ 5213.057986] RAX: 0000000000018978 RBX: ffff9d7ef7a03600 RCX: 0000000000000008
> [ 5213.059797] RDX: 000000000000007c RSI: 000000000000007c RDI: ffff9d7ef7a03600
> [ 5213.061720] RBP: 0000000000000120 R08: ffffae3580063c08 R09: ffffae3580063c04
> [ 5213.063532] R10: ffff9d7efece0038 R11: ffff9d7ec4b1db00 R12: ffff9d7efece0038
> [ 5213.065445] R13: ffff9d7ef4098260 R14: ffff9d7f11e23c6a R15: 4c79500065708144
> [ 5213.067264] FS: 0000000000000000(0000) GS:ffff9d7f3dc00000(0000) knlGS:0000000000000000
> [ 5213.069442] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033
> [ 5213.071004] CR2: 00007fce47276c60 CR3: 0000000003f66004 CR4: 00000000001706f0
> [ 5213.072827] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
> [ 5213.074484] DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
> [ 5213.076292] Call Trace:
> [ 5213.077027] <TASK>
> [ 5213.077718] rxe_responder+0x12ee/0x1b60 [rdma_rxe]
> [ 5213.079019] ? rxe_icrc_check+0x7e/0x100 [rdma_rxe]
> [ 5213.080380] ? rxe_rcv+0x1d0/0x780 [rdma_rxe]
> [ 5213.081708] ? rxe_icrc_hdr.isra.0+0xf6/0x160 [rdma_rxe]
> [ 5213.082990] rxe_do_task+0x67/0xb0 [rdma_rxe]
> [ 5213.084030] rxe_xmit_packet+0xc7/0x210 [rdma_rxe]
> [ 5213.085156] rxe_requester+0x680/0xee0 [rdma_rxe]
> [ 5213.088258] ? update_load_avg+0x5f/0x690
> [ 5213.089381] ? update_load_avg+0x5f/0x690
> [ 5213.090446] ? rtrs_clt_recv_done+0x1b/0x30 [rtrs_client]
> [ 5213.092087] rxe_do_task+0x67/0xb0 [rdma_rxe]
> [ 5213.093125] tasklet_action_common.constprop.0+0x92/0xc0
> [ 5213.094366] __do_softirq+0xe1/0x2d8
> [ 5213.095287] run_ksoftirqd+0x21/0x30
> [ 5213.096456] smpboot_thread_fn+0x183/0x220
> [ 5213.097519] ? sort_range+0x20/0x20
> [ 5213.098761] kthread+0xe2/0x110
> [ 5213.099638] ? kthread_complete_and_exit+0x20/0x20
> [ 5213.100948] ret_from_fork+0x22/0x30
>
> Link: https://lore.kernel.org/all/20220210073655.42281-1-guoqing.jiang@linux.dev/T/
> Link: https://www.spinics.net/lists/linux-rdma/msg110836.html
> Link: https://lore.kernel.org/lkml/94a5ea93-b8bb-3a01-9497-e2021f29598a@linux.dev/t/
>
> CC: Bob Pearson <rpearsonhpe@gmail.com>
> CC: Haris Iqbal <haris.iqbal@ionos.com>
> CC: Guoqing Jiang <guoqing.jiang@linux.dev>
> Tested-by: Md Haris Iqbal <haris.iqbal@ionos.com>
> Reviewed-by: Bob Pearson <rpearsonhpe@gmail.com>
> Signed-off-by: Li Zhijian <lizhijian@fujitsu.com>
> ---
> V2: fix commit log
> ---
> drivers/infiniband/sw/rxe/rxe_loc.h | 1 -
> drivers/infiniband/sw/rxe/rxe_mr.c | 199 +++++++++++-----------------------
> drivers/infiniband/sw/rxe/rxe_mw.c | 6 +-
> drivers/infiniband/sw/rxe/rxe_verbs.c | 39 +++++--
> drivers/infiniband/sw/rxe/rxe_verbs.h | 21 ++--
> 5 files changed, 104 insertions(+), 162 deletions(-)
>
Thanks, applied.
prev parent reply other threads:[~2022-07-27 9:26 UTC|newest]
Thread overview: 2+ messages / expand[flat|nested] mbox.gz Atom feed top
2022-07-26 3:16 [PATCH for-next v2] Revert "RDMA/rxe: Create duplicate mapping tables for FMRs" Li Zhijian
2022-07-27 9:26 ` Leon Romanovsky [this message]
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=YuEEtGW0BPUoPnr3@unreal \
--to=leon@kernel.org \
--cc=guoqing.jiang@linux.dev \
--cc=haris.iqbal@ionos.com \
--cc=jgg@ziepe.ca \
--cc=linux-rdma@vger.kernel.org \
--cc=lizhijian@fujitsu.com \
--cc=rpearsonhpe@gmail.com \
--cc=yanjun.zhu@linux.dev \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).