netdev.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Mohith Kumar Thummaluru <mohith.k.kumar.thummaluru@oracle.com>
To: Mark Bloch <mbloch@nvidia.com>,
	"saeedm@nvidia.com" <saeedm@nvidia.com>,
	"leon@kernel.org" <leon@kernel.org>,
	"tariqt@nvidia.com" <tariqt@nvidia.com>,
	"netdev@vger.kernel.org" <netdev@vger.kernel.org>
Cc: "andrew+netdev@lunn.ch" <andrew+netdev@lunn.ch>,
	"davem@davemloft.net" <davem@davemloft.net>,
	"edumazet@google.com" <edumazet@google.com>,
	"kuba@kernel.org" <kuba@kernel.org>,
	"pabeni@redhat.com" <pabeni@redhat.com>,
	"jacob.e.keller@intel.com" <jacob.e.keller@intel.com>,
	"shayd@nvidia.com" <shayd@nvidia.com>,
	"elic@nvidia.com" <elic@nvidia.com>,
	"linux-rdma@vger.kernel.org" <linux-rdma@vger.kernel.org>,
	"linux-kernel@vger.kernel.org" <linux-kernel@vger.kernel.org>,
	Anand Khoje <anand.a.khoje@oracle.com>,
	Manjunath Patil <manjunath.b.patil@oracle.com>,
	Rama Nichanamatlu <rama.nichanamatlu@oracle.com>,
	Rajesh Sivaramasubramaniom
	<rajesh.sivaramasubramaniom@oracle.com>,
	Rohit Sajan Kumar <rohit.sajan.kumar@oracle.com>,
	Qing Huang <qing.huang@oracle.com>
Subject: Re: [RESEND PATCH net-next 1/1] net/mlx5: Clean up only new IRQ glue on request_irq() failure
Date: Fri, 27 Jun 2025 12:13:28 +0530	[thread overview]
Message-ID: <d6851f08-e7ce-42d8-a154-412c8403e97c@oracle.com> (raw)
In-Reply-To: <2fc66048-36be-408d-a79f-0393ce2b4040@nvidia.com>


On 26/06/25 5:28 pm, Mark Bloch wrote:
>
> On 26/06/2025 9:04, Mohith Kumar Thummaluru wrote:
>> The mlx5_irq_alloc() function can inadvertently free the entire rmap
>> and end up in a crash[1] when the other threads tries to access this,
>> when request_irq() fails due to exhausted IRQ vectors. This commit
>> modifies the cleanup to remove only the specific IRQ mapping that was
>> just added.
>>
>> This prevents removal of other valid mappings and ensures precise
>> cleanup of the failed IRQ allocation's associated glue object.
>>
>> Note: This error is observed when both fwctl and rds configs are enabled.
>>
> Please target net and not net-next.
>
> Mark

Thanks, Mark. Let me do that.

- Mohith.

>> [1]
>> mlx5_core 0000:05:00.0: Successfully registered panic handler for port 1
>> mlx5_core 0000:05:00.0: mlx5_irq_alloc:293:(pid 66740): Failed to request irq. err = -28
>> infiniband mlx5_0: mlx5_ib_test_wc:290:(pid 66740): Error -28 while trying to test write-combining support
>> mlx5_core 0000:05:00.0: Successfully unregistered panic handler for port 1
>> mlx5_core 0000:06:00.0: Successfully registered panic handler for port 1
>> mlx5_core 0000:06:00.0: mlx5_irq_alloc:293:(pid 66740): Failed to request irq. err = -28
>> infiniband mlx5_0: mlx5_ib_test_wc:290:(pid 66740): Error -28 while trying to test write-combining support
>> mlx5_core 0000:06:00.0: Successfully unregistered panic handler for port 1
>> mlx5_core 0000:03:00.0: mlx5_irq_alloc:293:(pid 28895): Failed to request irq. err = -28
>> mlx5_core 0000:05:00.0: mlx5_irq_alloc:293:(pid 28895): Failed to request irq. err = -28
>> general protection fault, probably for non-canonical address 0xe277a58fde16f291: 0000 [#1] SMP NOPTI
>>
>> RIP: 0010:free_irq_cpu_rmap+0x23/0x7d
>> Call Trace:
>>    <TASK>
>>    ? show_trace_log_lvl+0x1d6/0x2f9
>>    ? show_trace_log_lvl+0x1d6/0x2f9
>>    ? mlx5_irq_alloc.cold+0x5d/0xf3 [mlx5_core]
>>    ? __die_body.cold+0x8/0xa
>>    ? die_addr+0x39/0x53
>>    ? exc_general_protection+0x1c4/0x3e9
>>    ? dev_vprintk_emit+0x5f/0x90
>>    ? asm_exc_general_protection+0x22/0x27
>>    ? free_irq_cpu_rmap+0x23/0x7d
>>    mlx5_irq_alloc.cold+0x5d/0xf3 [mlx5_core]
>>    irq_pool_request_vector+0x7d/0x90 [mlx5_core]
>>    mlx5_irq_request+0x2e/0xe0 [mlx5_core]
>>    mlx5_irq_request_vector+0xad/0xf7 [mlx5_core]
>>    comp_irq_request_pci+0x64/0xf0 [mlx5_core]
>>    create_comp_eq+0x71/0x385 [mlx5_core]
>>    ? mlx5e_open_xdpsq+0x11c/0x230 [mlx5_core]
>>    mlx5_comp_eqn_get+0x72/0x90 [mlx5_core]
>>    ? xas_load+0x8/0x91
>>    mlx5_comp_irqn_get+0x40/0x90 [mlx5_core]
>>    mlx5e_open_channel+0x7d/0x3c7 [mlx5_core]
>>    mlx5e_open_channels+0xad/0x250 [mlx5_core]
>>    mlx5e_open_locked+0x3e/0x110 [mlx5_core]
>>    mlx5e_open+0x23/0x70 [mlx5_core]
>>    __dev_open+0xf1/0x1a5
>>    __dev_change_flags+0x1e1/0x249
>>    dev_change_flags+0x21/0x5c
>>    do_setlink+0x28b/0xcc4
>>    ? __nla_parse+0x22/0x3d
>>    ? inet6_validate_link_af+0x6b/0x108
>>    ? cpumask_next+0x1f/0x35
>>    ? __snmp6_fill_stats64.constprop.0+0x66/0x107
>>    ? __nla_validate_parse+0x48/0x1e6
>>    __rtnl_newlink+0x5ff/0xa57
>>    ? kmem_cache_alloc_trace+0x164/0x2ce
>>    rtnl_newlink+0x44/0x6e
>>    rtnetlink_rcv_msg+0x2bb/0x362
>>    ? __netlink_sendskb+0x4c/0x6c
>>    ? netlink_unicast+0x28f/0x2ce
>>    ? rtnl_calcit.isra.0+0x150/0x146
>>    netlink_rcv_skb+0x5f/0x112
>>    netlink_unicast+0x213/0x2ce
>>    netlink_sendmsg+0x24f/0x4d9
>>    __sock_sendmsg+0x65/0x6a
>>    ____sys_sendmsg+0x28f/0x2c9
>>    ? import_iovec+0x17/0x2b
>>    ___sys_sendmsg+0x97/0xe0
>>    __sys_sendmsg+0x81/0xd8
>>    do_syscall_64+0x35/0x87
>>    entry_SYSCALL_64_after_hwframe+0x6e/0x0
>> RIP: 0033:0x7fc328603727
>> Code: c3 66 90 41 54 41 89 d4 55 48 89 f5 53 89 fb 48 83 ec 10 e8 0b ed ff ff 44 89 e2 48 89 ee 89 df 41 89 c0 b8 2e 00 00 00 0f 05 <48> 3d 00 f0 ff ff 77 35 44 89 c7 48 89 44 24 08 e8 44 ed ff ff 48
>> RSP: 002b:00007ffe8eb3f1a0 EFLAGS: 00000293 ORIG_RAX: 000000000000002e
>> RAX: ffffffffffffffda RBX: 000000000000000d RCX: 00007fc328603727
>> RDX: 0000000000000000 RSI: 00007ffe8eb3f1f0 RDI: 000000000000000d
>> RBP: 00007ffe8eb3f1f0 R08: 0000000000000000 R09: 0000000000000000
>> R10: 0000000000000000 R11: 0000000000000293 R12: 0000000000000000
>> R13: 0000000000000000 R14: 00007ffe8eb3f3c8 R15: 00007ffe8eb3f3bc
>>    </TASK>
>> ---[ end trace f43ce73c3c2b13a2 ]---
>> RIP: 0010:free_irq_cpu_rmap+0x23/0x7d
>> Code: 0f 1f 80 00 00 00 00 48 85 ff 74 6b 55 48 89 fd 53 66 83 7f 06 00 74 24 31 db 48 8b 55 08 0f b7 c3 48 8b 04 c2 48 85 c0 74 09 <8b> 38 31 f6 e8 c4 0a b8 ff 83 c3 01 66 3b 5d 06 72 de b8 ff ff ff
>> RSP: 0018:ff384881640eaca0 EFLAGS: 00010282
>> RAX: e277a58fde16f291 RBX: 0000000000000000 RCX: 0000000000000000
>> RDX: ff2335e2e20b3600 RSI: 0000000000000000 RDI: ff2335e2e20b3400
>> RBP: ff2335e2e20b3400 R08: 0000000000000000 R09: 0000000000000000
>> R10: 0000000000000000 R11: 00000000ffffffe4 R12: ff384881640ead88
>> R13: ff2335c3760751e0 R14: ff2335e2e1672200 R15: ff2335c3760751f8
>> FS:  00007fc32ac22480(0000) GS:ff2335e2d6e00000(0000) knlGS:0000000000000000
>> CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
>> CR2: 00007f651ab54000 CR3: 00000029f1206003 CR4: 0000000000771ef0
>> DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
>> DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400
>> PKRU: 55555554
>> Kernel panic - not syncing: Fatal exception
>> Kernel Offset: 0x1dc00000 from 0xffffffff81000000 (relocation range: 0xffffffff80000000-0xffffffffbfffffff)
>> kvm-guest: disable async PF for cpu 0
>>
>>
>> Fixes: 3354822cde5a ("net/mlx5: Use dynamic msix vectors allocation")
>> Signed-off-by: Mohith Kumar Thummaluru<mohith.k.kumar.thummaluru@oracle.com>
>> Tested-by: Mohith Kumar Thummaluru<mohith.k.kumar.thummaluru@oracle.com>
>> Reviewed-by: Moshe Shemesh<moshe@nvidia.com>
>> ---
>>    drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c | 3 +--
>>    1 file changed, 1 insertion(+), 2 deletions(-)
>>
>> diff --git a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
>> index 40024cfa3099..822e92ed2d45 100644
>> --- a/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
>> +++ b/drivers/net/ethernet/mellanox/mlx5/core/pci_irq.c
>> @@ -325,8 +325,7 @@ struct mlx5_irq *mlx5_irq_alloc(struct mlx5_irq_pool *pool, int i,
>>    err_req_irq:
>>    #ifdef CONFIG_RFS_ACCEL
>>        if (i && rmap && *rmap) {
>> -        free_irq_cpu_rmap(*rmap);
>> -        *rmap = NULL;
>> +        irq_cpu_rmap_remove(*rmap, irq->map.virq);
>>        }
>>    err_irq_rmap:
>>    #endif

  reply	other threads:[~2025-06-27  6:44 UTC|newest]

Thread overview: 6+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-06-26  6:04 [RESEND PATCH net-next 1/1] net/mlx5: Clean up only new IRQ glue on request_irq() failure Mohith Kumar Thummaluru
2025-06-26 11:58 ` Mark Bloch
2025-06-27  6:43   ` Mohith Kumar Thummaluru [this message]
     [not found] <7cb171c4-3c36-42ea-bd6f-52dfe6bc5dab@oracle.com>
2025-07-01 20:51 ` Jacob Keller
2025-07-02  5:07   ` Mohith Kumar Thummaluru
2025-07-02 17:56     ` Jacob Keller

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=d6851f08-e7ce-42d8-a154-412c8403e97c@oracle.com \
    --to=mohith.k.kumar.thummaluru@oracle.com \
    --cc=anand.a.khoje@oracle.com \
    --cc=andrew+netdev@lunn.ch \
    --cc=davem@davemloft.net \
    --cc=edumazet@google.com \
    --cc=elic@nvidia.com \
    --cc=jacob.e.keller@intel.com \
    --cc=kuba@kernel.org \
    --cc=leon@kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-rdma@vger.kernel.org \
    --cc=manjunath.b.patil@oracle.com \
    --cc=mbloch@nvidia.com \
    --cc=netdev@vger.kernel.org \
    --cc=pabeni@redhat.com \
    --cc=qing.huang@oracle.com \
    --cc=rajesh.sivaramasubramaniom@oracle.com \
    --cc=rama.nichanamatlu@oracle.com \
    --cc=rohit.sajan.kumar@oracle.com \
    --cc=saeedm@nvidia.com \
    --cc=shayd@nvidia.com \
    --cc=tariqt@nvidia.com \
    /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).