In the Linux kernel, the following vulnerability has been resolved:
perf/x86/intel: Fix segfault with PEBS-via-PT with sample_freq
Currently, using PEBS-via-PT with a sample frequency instead of a sample period, causes a segfault. For example:
BUG: kernel NULL pointer dereference, address: 0000000000000195
<NMI>
? __die_body.cold+0x19/0x27
? page_fault_oops+0xca/0x290
? exc_page_fault+0x7e/0x1b0
? asm_exc_page_fault+0x26/0x30
? intel_pmu_pebs_event_update_no_drain+0x40/0x60
? intel_pmu_pebs_event_update_no_drain+0x32/0x60
intel_pmu_drain_pebs_icl+0x333/0x350
handle_pmi_common+0x272/0x3c0
intel_pmu_handle_irq+0x10a/0x2e0
perf_event_nmi_handler+0x2a/0x50
That happens because intel_pmu_pebs_event_update_no_drain() assumes all the pebs_enabled bits represent counter indexes, which is not always the case. In this particular case, bits 60 and 61 are set for PEBS-via-PT purposes.
The behaviour of PEBS-via-PT with sample frequency is questionable because although a PMI is generated (PEBS_PMI_AFTER_EACH_RECORD), the period is not adjusted anyway.
Putting that aside, fix intel_pmu_pebs_event_update_no_drain() by passing the mask of counter bits instead of 'size'. Note, prior to the Fixes commit, 'size' would be limited to the maximum counter index, so the issue was not hit.
| Software | From | Fixed in |
|---|---|---|
| linux / linux_kernel | 6.10.5 | 6.12.31 |
| linux / linux_kernel | 6.13 | 6.14.9 |
| linux / linux_kernel | 6.15-rc1 | 6.15-rc1.x |
| linux / linux_kernel | 6.15-rc2 | 6.15-rc2.x |
| linux / linux_kernel | 6.15-rc3 | 6.15-rc3.x |
| linux / linux_kernel | 6.15-rc4 | 6.15-rc4.x |
| linux / linux_kernel | 6.15-rc5 | 6.15-rc5.x |
| linux / linux_kernel | 6.15-rc6 | 6.15-rc6.x |