{"id":"CVE-2026-64070","title":"In the Linux kernel, the following vulnerability has been resolved:\n\npowerpc/hv-gpci: fix preempt count leak in sysfs show paths\n\nFour sysfs show() callbacks in hv-gpci take get_cpu_var(hv_gpci_reqb)\n(which calls preempt_disable()) but o…","summary":"In the Linux kernel, the following vulnerability has been resolved:\n\npowerpc/hv-gpci: fix preempt count leak in sysfs show paths\n\nFour sysfs show() callbacks in hv-gpci take get_cpu_var(hv_gpci_reqb)\n(which calls preempt_disable()) but o…","severity":"medium","cvss":5.5,"cvssVector":"CVSS:3.1/AV:L/AC:L/PR:L/UI:N/S:U/C:N/I:N/A:H","vendor":"linux","product":"linux_kernel","affected":["linux_kernel >= 6.6, < 6.6.156","linux_kernel >= 6.7, < 6.12.108","linux_kernel >= 6.13, < 6.18.49","linux_kernel >= 6.19, < 7.0.11","linux_kernel = 7.1"],"patched":["linux_kernel 7.0.11"],"published":"2026-07-19","updated":"2026-09-03","source":"NVD","sourceUrl":"https://nvd.nist.gov/vuln/detail/CVE-2026-64070","references":[{"url":"https://git.kernel.org/stable/c/45afabe7f99c5d8746434ee41c86584c01d70147","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/903409000a07ac8e31ffedeb8516f4f8d67150c8","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/b300312562fd6e4729697adb9f777f00ca0429f3","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/b61ebb2826ca1852c2bbd92a676fb622c358cab4","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"},{"url":"https://git.kernel.org/stable/c/dbc30a57bd8e026995e9fa8e8c31cffd18542c01","label":"416baaa9-dc9f-4396-8d5f-8c081fb06d67"}],"tags":["nvd"],"epss":0.00122,"epssPercentile":0.02291,"ingestedAt":"2026-09-04T04:14:25.745Z","slug":"CVE-2026-64070","body":"## Overview\n\nIn the Linux kernel, the following vulnerability has been resolved:\n\npowerpc/hv-gpci: fix preempt count leak in sysfs show paths\n\nFour sysfs show() callbacks in hv-gpci take get_cpu_var(hv_gpci_reqb)\n(which calls preempt_disable()) but only call the matching put_cpu_var()\non the error path under the 'out:' label. Every successful read leaks\none preempt_disable():\n\n  processor_bus_topology_show()\n  processor_config_show()\n  affinity_domain_via_virtual_processor_show()\n  affinity_domain_via_domain_show()\n\n(affinity_domain_via_partition_show() was already correct.)\n\nOn a CONFIG_PREEMPT=y kernel, repeated reads raise preempt_count and\neventually return to userspace with preemption still disabled. The\nnext user-mode page fault then hits faulthandler_disabled() == 1,\ngets forced to SIGSEGV, and the resulting coredump trips\n'BUG: scheduling while atomic' in call_usermodehelper_exec ->\nwait_for_completion_state -> schedule:\n\n  BUG: scheduling while atomic: <task>/<pid>/0x00000004\n  ...\n  __schedule_bug+0x6c/0x90\n  __schedule+0x58c/0x13a0\n  schedule+0x48/0x1a0\n  schedule_timeout+0x104/0x170\n  wait_for_completion_state+0x16c/0x330\n  call_usermodehelper_exec+0x254/0x2d0\n  vfs_coredump+0x1050/0x2590\n  get_signal+0xb9c/0xc80\n  do_notify_resume+0xf8/0x470\n\nAdd an out_success label that calls put_cpu_var() before returning\nthe byte count, mirroring affinity_domain_via_partition_show().\n\n## Affected\n\n- `linux_kernel >= 6.6, < 6.6.156`\n- `linux_kernel >= 6.7, < 6.12.108`\n- `linux_kernel >= 6.13, < 6.18.49`\n- `linux_kernel >= 6.19, < 7.0.11`\n- `linux_kernel = 7.1`\n\n## Remediation\n\nUpgrade past the affected range:\n\n- `linux_kernel 7.0.11`","depth":"sunlit","depthScore":30,"depthScoreParts":{"impact":30.3,"likelihood":0,"exploitation":0,"ransomware":0},"changes":[]}