{"id":"CVE-2026-90100","published":"2026-09-17T17:17:01.680","lastModified":"2026-09-17T17:17:01.680","description":"In the Linux kernel, the following vulnerability has been resolved:\n\nptp: netc: fix period truncation and potential divide-by-zero in PEROUT\n\nThe max_period bound in net_timer_enable_perout() was computed as:\n\n  max_period = (u64)NETC_TMR_DEFAULT_FIPER + integral_period;\n\nwhich exceeds U32_MAX when integral_period > 0 (e.g. 0x100000002 for\nthe default 333333333 Hz clock). A period_ns that passes this check but\nexceeds U32_MAX is then silently truncated when stored into the u32\nstruct netc_pp::period field.\n\nA truncated value of zero can reach netc_timer_set_perout_alarm(), where\nthe local u32 period variable would also be 0, causing a divide-by-zero\nin roundup_u64(delta, period) whenever the stime < min_time branch is\ntaken (which always happens for a start time of {0, 0}).\n\nAdditionally, netc_timer_enable_periodic_pulse() and\nnetc_timer_enable_fiper() both compute:\n\n  fiper = pp->period - integral_period;\n\nA zero pp->period results in an unsigned wraparound to 0xFFFFFFFD,\nmis-programming the FIPER hardware register.\n\nFix all three issues by capping max_period at NETC_TMR_DEFAULT_FIPER\n(0xFFFFFFFF). This ensures that any period_ns passing the range check\nfits in a u32 without truncation, so the stored value is always valid\nand non-zero. The accepted range is reduced by integral_period ns\n(typically only a few nanoseconds), which is negligible in practice.","cvssScore":null,"cvssSeverity":null,"cvssVector":null,"cwes":[],"vendors":[],"products":[],"references":[{"url":"https://git.kernel.org/stable/c/08988d1941e180f0ad30d91233cb64f3418ba23c","tags":[]},{"url":"https://git.kernel.org/stable/c/51fe3fe0ffec033bd831c71d9bee3dee827e491c","tags":[]},{"url":"https://git.kernel.org/stable/c/777dbc9914b2f003f1d44af80c7a4a395c5961b2","tags":[]}],"exploitRefs":[],"hasPoc":false,"ai":null}