ghsa-5322-f7jr-cv4m
Vulnerability from github
In the Linux kernel, the following vulnerability has been resolved:
net/smc: Reset connection when trying to use SMCRv2 fails.
We found a crash when using SMCRv2 with 2 Mellanox ConnectX-4. It can be reproduced by:
- smc_run nginx
- smc_run wrk -t 32 -c 500 -d 30 http://:
BUG: kernel NULL pointer dereference, address: 0000000000000014 #PF: supervisor read access in kernel mode #PF: error_code(0x0000) - not-present page PGD 8000000108713067 P4D 8000000108713067 PUD 151127067 PMD 0 Oops: 0000 [#1] PREEMPT SMP PTI CPU: 4 PID: 2441 Comm: kworker/4:249 Kdump: loaded Tainted: G W E 6.4.0-rc1+ #42 Workqueue: smc_hs_wq smc_listen_work [smc] RIP: 0010:smc_clc_send_confirm_accept+0x284/0x580 [smc] RSP: 0018:ffffb8294b2d7c78 EFLAGS: 00010a06 RAX: ffff8f1873238880 RBX: ffffb8294b2d7dc8 RCX: 0000000000000000 RDX: 00000000000000b4 RSI: 0000000000000001 RDI: 0000000000b40c00 RBP: ffffb8294b2d7db8 R08: ffff8f1815c5860c R09: 0000000000000000 R10: 0000000000000400 R11: 0000000000000000 R12: ffff8f1846f56180 R13: ffff8f1815c5860c R14: 0000000000000001 R15: 0000000000000001 FS: 0000000000000000(0000) GS:ffff8f1aefd00000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000014 CR3: 00000001027a0001 CR4: 00000000003706e0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: ? mlx5_ib_map_mr_sg+0xa1/0xd0 [mlx5_ib] ? smcr_buf_map_link+0x24b/0x290 [smc] ? __smc_buf_create+0x4ee/0x9b0 [smc] smc_clc_send_accept+0x4c/0xb0 [smc] smc_listen_work+0x346/0x650 [smc] ? __schedule+0x279/0x820 process_one_work+0x1e5/0x3f0 worker_thread+0x4d/0x2f0 ? __pfx_worker_thread+0x10/0x10 kthread+0xe5/0x120 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x2c/0x50
During the CLC handshake, server sequentially tries available SMCRv2 and SMCRv1 devices in smc_listen_work().
If an SMCRv2 device is found. SMCv2 based link group and link will be assigned to the connection. Then assumed that some buffer assignment errors happen later in the CLC handshake, such as RMB registration failure, server will give up SMCRv2 and try SMCRv1 device instead. But the resources assigned to the connection won't be reset.
When server tries SMCRv1 device, the connection creation process will be executed again. Since conn->lnk has been assigned when trying SMCRv2, it will not be set to the correct SMCRv1 link in smcr_lgr_conn_assign_link(). So in such situation, conn->lgr points to correct SMCRv1 link group but conn->lnk points to the SMCRv2 link mistakenly.
Then in smc_clc_send_confirm_accept(), conn->rmb_desc->mr[link->link_idx] will be accessed. Since the link->link_idx is not correct, the related MR may not have been initialized, so crash happens.
| Try SMCRv2 device first | |-> conn->lgr: assign existed SMCRv2 link group; | |-> conn->link: assign existed SMCRv2 link (link_idx may be 1 in SMC_LGR_SYMMETRIC); | |-> sndbuf & RMB creation fails, quit; | | Try SMCRv1 device then | |-> conn->lgr: create SMCRv1 link group and assign; | |-> conn->link: keep SMCRv2 link mistakenly; | |-> sndbuf & RMB creation succeed, only RMB->mr[link_idx = 0] | initialized. | | Then smc_clc_send_confirm_accept() accesses | conn->rmb_desc->mr[conn->link->link_idx, which is 1], then crash. v
This patch tries to fix this by cleaning conn->lnk before assigning link. In addition, it is better to reset the connection and clean the resources assigned if trying SMCRv2 failed in buffer creation or registration.
{
"affected": [],
"aliases": [
"CVE-2023-53382"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2025-09-18T14:15:41Z",
"severity": null
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\nnet/smc: Reset connection when trying to use SMCRv2 fails.\n\nWe found a crash when using SMCRv2 with 2 Mellanox ConnectX-4. It\ncan be reproduced by:\n\n- smc_run nginx\n- smc_run wrk -t 32 -c 500 -d 30 http://\u003cip\u003e:\u003cport\u003e\n\n BUG: kernel NULL pointer dereference, address: 0000000000000014\n #PF: supervisor read access in kernel mode\n #PF: error_code(0x0000) - not-present page\n PGD 8000000108713067 P4D 8000000108713067 PUD 151127067 PMD 0\n Oops: 0000 [#1] PREEMPT SMP PTI\n CPU: 4 PID: 2441 Comm: kworker/4:249 Kdump: loaded Tainted: G W E 6.4.0-rc1+ #42\n Workqueue: smc_hs_wq smc_listen_work [smc]\n RIP: 0010:smc_clc_send_confirm_accept+0x284/0x580 [smc]\n RSP: 0018:ffffb8294b2d7c78 EFLAGS: 00010a06\n RAX: ffff8f1873238880 RBX: ffffb8294b2d7dc8 RCX: 0000000000000000\n RDX: 00000000000000b4 RSI: 0000000000000001 RDI: 0000000000b40c00\n RBP: ffffb8294b2d7db8 R08: ffff8f1815c5860c R09: 0000000000000000\n R10: 0000000000000400 R11: 0000000000000000 R12: ffff8f1846f56180\n R13: ffff8f1815c5860c R14: 0000000000000001 R15: 0000000000000001\n FS: 0000000000000000(0000) GS:ffff8f1aefd00000(0000) knlGS:0000000000000000\n CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033\n CR2: 0000000000000014 CR3: 00000001027a0001 CR4: 00000000003706e0\n DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000\n DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400\n Call Trace:\n \u003cTASK\u003e\n ? mlx5_ib_map_mr_sg+0xa1/0xd0 [mlx5_ib]\n ? smcr_buf_map_link+0x24b/0x290 [smc]\n ? __smc_buf_create+0x4ee/0x9b0 [smc]\n smc_clc_send_accept+0x4c/0xb0 [smc]\n smc_listen_work+0x346/0x650 [smc]\n ? __schedule+0x279/0x820\n process_one_work+0x1e5/0x3f0\n worker_thread+0x4d/0x2f0\n ? __pfx_worker_thread+0x10/0x10\n kthread+0xe5/0x120\n ? __pfx_kthread+0x10/0x10\n ret_from_fork+0x2c/0x50\n \u003c/TASK\u003e\n\nDuring the CLC handshake, server sequentially tries available SMCRv2\nand SMCRv1 devices in smc_listen_work().\n\nIf an SMCRv2 device is found. SMCv2 based link group and link will be\nassigned to the connection. Then assumed that some buffer assignment\nerrors happen later in the CLC handshake, such as RMB registration\nfailure, server will give up SMCRv2 and try SMCRv1 device instead. But\nthe resources assigned to the connection won\u0027t be reset.\n\nWhen server tries SMCRv1 device, the connection creation process will\nbe executed again. Since conn-\u003elnk has been assigned when trying SMCRv2,\nit will not be set to the correct SMCRv1 link in\nsmcr_lgr_conn_assign_link(). So in such situation, conn-\u003elgr points to\ncorrect SMCRv1 link group but conn-\u003elnk points to the SMCRv2 link\nmistakenly.\n\nThen in smc_clc_send_confirm_accept(), conn-\u003ermb_desc-\u003emr[link-\u003elink_idx]\nwill be accessed. Since the link-\u003elink_idx is not correct, the related\nMR may not have been initialized, so crash happens.\n\n | Try SMCRv2 device first\n | |-\u003e conn-\u003elgr:\tassign existed SMCRv2 link group;\n | |-\u003e conn-\u003elink:\tassign existed SMCRv2 link (link_idx may be 1 in SMC_LGR_SYMMETRIC);\n | |-\u003e sndbuf \u0026 RMB creation fails, quit;\n |\n | Try SMCRv1 device then\n | |-\u003e conn-\u003elgr:\tcreate SMCRv1 link group and assign;\n | |-\u003e conn-\u003elink:\tkeep SMCRv2 link mistakenly;\n | |-\u003e sndbuf \u0026 RMB creation succeed, only RMB-\u003emr[link_idx = 0]\n | initialized.\n |\n | Then smc_clc_send_confirm_accept() accesses\n | conn-\u003ermb_desc-\u003emr[conn-\u003elink-\u003elink_idx, which is 1], then crash.\n v\n\nThis patch tries to fix this by cleaning conn-\u003elnk before assigning\nlink. In addition, it is better to reset the connection and clean the\nresources assigned if trying SMCRv2 failed in buffer creation or\nregistration.",
"id": "GHSA-5322-f7jr-cv4m",
"modified": "2025-09-18T15:30:34Z",
"published": "2025-09-18T15:30:34Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2023-53382"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/35112271672ae98f45df7875244a4e33aa215e31"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/9540765d1882d15497d880096de99fafabcfa08c"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/d33be18917ffe69865dfed18b0a67b0dee0b47d7"
}
],
"schema_version": "1.4.0",
"severity": []
}
Sightings
| Author | Source | Type | Date |
|---|
Nomenclature
- Seen: The vulnerability was mentioned, discussed, or seen somewhere by the user.
- Confirmed: The vulnerability is confirmed from an analyst perspective.
- Published Proof of Concept: A public proof of concept is available for this vulnerability.
- Exploited: This vulnerability was exploited and seen by the user reporting the sighting.
- Patched: This vulnerability was successfully patched by the user reporting the sighting.
- Not exploited: This vulnerability was not exploited or seen by the user reporting the sighting.
- Not confirmed: The user expresses doubt about the veracity of the vulnerability.
- Not patched: This vulnerability was not successfully patched by the user reporting the sighting.