ghsa-rcm4-5vxv-3g8q
Vulnerability from github
In the Linux kernel, the following vulnerability has been resolved:
powerpc/smp: do not decrement idle task preempt count in CPU offline
With PREEMPT_COUNT=y, when a CPU is offlined and then onlined again, we get:
BUG: scheduling while atomic: swapper/1/0/0x00000000 no locks held by swapper/1/0. CPU: 1 PID: 0 Comm: swapper/1 Not tainted 5.15.0-rc2+ #100 Call Trace: dump_stack_lvl+0xac/0x108 __schedule_bug+0xac/0xe0 __schedule+0xcf8/0x10d0 schedule_idle+0x3c/0x70 do_idle+0x2d8/0x4a0 cpu_startup_entry+0x38/0x40 start_secondary+0x2ec/0x3a0 start_secondary_prolog+0x10/0x14
This is because powerpc's arch_cpu_idle_dead() decrements the idle task's preempt count, for reasons explained in commit a7c2bb8279d2 ("powerpc: Re-enable preemption before cpu_die()"), specifically "start_secondary() expects a preempt_count() of 0."
However, since commit 2c669ef6979c ("powerpc/preempt: Don't touch the idle task's preempt_count during hotplug") and commit f1a0a376ca0c ("sched/core: Initialize the idle task with preemption disabled"), that justification no longer holds.
The idle task isn't supposed to re-enable preemption, so remove the vestigial preempt_enable() from the CPU offline path.
Tested with pseries and powernv in qemu, and pseries on PowerVM.
{
"affected": [],
"aliases": [
"CVE-2021-47454"
],
"database_specific": {
"cwe_ids": [],
"github_reviewed": false,
"github_reviewed_at": null,
"nvd_published_at": "2024-05-22T07:15:10Z",
"severity": null
},
"details": "In the Linux kernel, the following vulnerability has been resolved:\n\npowerpc/smp: do not decrement idle task preempt count in CPU offline\n\nWith PREEMPT_COUNT=y, when a CPU is offlined and then onlined again, we\nget:\n\nBUG: scheduling while atomic: swapper/1/0/0x00000000\nno locks held by swapper/1/0.\nCPU: 1 PID: 0 Comm: swapper/1 Not tainted 5.15.0-rc2+ #100\nCall Trace:\n dump_stack_lvl+0xac/0x108\n __schedule_bug+0xac/0xe0\n __schedule+0xcf8/0x10d0\n schedule_idle+0x3c/0x70\n do_idle+0x2d8/0x4a0\n cpu_startup_entry+0x38/0x40\n start_secondary+0x2ec/0x3a0\n start_secondary_prolog+0x10/0x14\n\nThis is because powerpc\u0027s arch_cpu_idle_dead() decrements the idle task\u0027s\npreempt count, for reasons explained in commit a7c2bb8279d2 (\"powerpc:\nRe-enable preemption before cpu_die()\"), specifically \"start_secondary()\nexpects a preempt_count() of 0.\"\n\nHowever, since commit 2c669ef6979c (\"powerpc/preempt: Don\u0027t touch the idle\ntask\u0027s preempt_count during hotplug\") and commit f1a0a376ca0c (\"sched/core:\nInitialize the idle task with preemption disabled\"), that justification no\nlonger holds.\n\nThe idle task isn\u0027t supposed to re-enable preemption, so remove the\nvestigial preempt_enable() from the CPU offline path.\n\nTested with pseries and powernv in qemu, and pseries on PowerVM.",
"id": "GHSA-rcm4-5vxv-3g8q",
"modified": "2024-05-22T09:31:45Z",
"published": "2024-05-22T09:31:45Z",
"references": [
{
"type": "ADVISORY",
"url": "https://nvd.nist.gov/vuln/detail/CVE-2021-47454"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/3ea0b497a7a2fff6a4b7090310c9f52c91975934"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/53770a411559cf7bc0906d1df319cc533d2f4f58"
},
{
"type": "WEB",
"url": "https://git.kernel.org/stable/c/787252a10d9422f3058df9a4821f389e5326c440"
}
],
"schema_version": "1.4.0",
"severity": []
}