{"id":"CVE-2026-98919","title":"drm/xe: harden adjust_idledly() against divide-by-zero and overflow","summary":"In the Linux kernel, the following vulnerability has been resolved:\n\ndrm/xe: harden adjust_idledly() against divide-by-zero and overflow\n\nadjust_idledly() has several corner-case issues flagged during review:\n\n  1. If xe_gt_clock_init() …","severity":"none","vendor":"Linux","product":"Linux","affected":["Linux >= d2de4410a88ffc6053300d03041cc73b8c85dbaf < 2bcbfc5e7bfa9cc76818c54db698da5bdc13d9ff","Linux >= d2de4410a88ffc6053300d03041cc73b8c85dbaf < 531d4cd03118224022dc21ae25af052ce2a9e267","Linux >= d2de4410a88ffc6053300d03041cc73b8c85dbaf < 90f467577ebc28c5bc4ba2f0f1b83a4419fb0740","Linux 6.15"],"published":"2026-10-23","updated":"2026-10-23","sourceUpdated":"2026-10-23T12:05:08.723Z","source":"CVEORG","sourceUrl":"https://www.cve.org/CVERecord?id=CVE-2026-98919","references":[{"url":"https://git.kernel.org/stable/c/2bcbfc5e7bfa9cc76818c54db698da5bdc13d9ff"},{"url":"https://git.kernel.org/stable/c/531d4cd03118224022dc21ae25af052ce2a9e267"},{"url":"https://git.kernel.org/stable/c/90f467577ebc28c5bc4ba2f0f1b83a4419fb0740"}],"tags":["cve.org"],"ingestedAt":"2026-10-23T13:24:55.568Z","slug":"CVE-2026-98919","body":"## Overview\n\nIn the Linux kernel, the following vulnerability has been resolved:\n\ndrm/xe: harden adjust_idledly() against divide-by-zero and overflow\n\nadjust_idledly() has several corner-case issues flagged during review:\n\n  1. If xe_gt_clock_init() failed to recognise the crystal clock,\n     gt->info.timestamp_base is 0, which makes idledly_units_ps also 0.\n     The subsequent DIV_ROUND_CLOSEST(..., idledly_units_ps) is then a\n     divide-by-zero and panics the kernel.\n\n  2. The tick-to-ns conversions are done in u32:\n        idledly * idledly_units_ps, (maxcnt - 1) * 1000\n     Both overflow u32 before DIV_ROUND_CLOSEST() sees them.\n\n  3. If IDLE_WAIT_TIME reads back as 0, maxcnt evaluates to 0 and\n     the maxcnt - 1 clamp wraps to 0xFFFFFFFF in u32.\n\n  4. The register only stores whole ticks, so the clamped ns value has\n     to be converted to ticks and back. DIV_ROUND_CLOSEST() can round\n     that conversion up past maxcnt:\n\n       maxcnt = 640 ns, one tick = 666664 ps\n\n       clamp:        maxcnt - 1        = 639 ns\n       ns -> ticks:  639000 / 666664   = 0.958 -> rounds to 1 tick\n       tick -> ns:    1 * 666664 / 1000 = 667 ns\n\n     667 ns is programmed into RING_IDLEDLY, but 667 >= maxcnt (640),\n     so xe_gt_WARN_ON() fires again on every subsequent init.\n\nReturn early if timestamp_base is 0 (the unknown-crystal path).\nDo the conversions in u64 via the *_ULL() helpers so they cannot wrap.\nClamp with a floor (DIV_ROUND_DOWN_ULL) so the programmed delay stays\nstrictly below maxcnt, and guard the maxcnt == 0 case with a zero delay\nwhile still writing RING_IDLEDLY so INHIBIT_SWITCH_UNTIL_PREEMPTED is\ncleared.\n\nv2: Drop the redundant warn on the timestamp_base == 0 path;\n    xe_gt_clock_init() already warns on an unrecognised crystal clock.\n    Keep the early return to avoid the divide-by-zero. - Vinay\n\nv3: Field-mask the RING_IDLEDLY write with REG_FIELD_PREP(IDLE_DELAY, ...)\n    instead of writing the raw tick count, which could clobber\n    INHIBIT_SWITCH_UNTIL_PREEMPTED and reserved bits. Split the\n    inhibit-switch clear from the maxcnt clamp so a set inhibit bit no\n    longer forces a needless delay overwrite when the delay itself is\n    already valid. Use gt_to_xe(gt) instead of gt_to_xe(hwe->gt).\n\n(cherry picked from commit d864065ea25e9d12897c175de9176ce46677e176)\n\n## Affected\n\n- `Linux >= d2de4410a88ffc6053300d03041cc73b8c85dbaf < 2bcbfc5e7bfa9cc76818c54db698da5bdc13d9ff`\n- `Linux >= d2de4410a88ffc6053300d03041cc73b8c85dbaf < 531d4cd03118224022dc21ae25af052ce2a9e267`\n- `Linux >= d2de4410a88ffc6053300d03041cc73b8c85dbaf < 90f467577ebc28c5bc4ba2f0f1b83a4419fb0740`\n- `Linux 6.15`\n\n## Remediation\n\nRefer to the linked advisories for vendor-supplied fixes and affected version ranges.","depth":"sunlit","depthScore":3,"depthScoreParts":{"impact":2.8,"likelihood":0,"exploitation":0,"ransomware":0},"changes":[]}