CVE-2026-23052

*** Pendiente de traducción *** In the Linux kernel, the following vulnerability has been resolved:<br /> <br /> ftrace: Do not over-allocate ftrace memory<br /> <br /> The pg_remaining calculation in ftrace_process_locs() assumes that<br /> ENTRIES_PER_PAGE multiplied by 2^order equals the actual capacity of the<br /> allocated page group. However, ENTRIES_PER_PAGE is PAGE_SIZE / ENTRY_SIZE<br /> (integer division). When PAGE_SIZE is not a multiple of ENTRY_SIZE (e.g.<br /> 4096 / 24 = 170 with remainder 16), high-order allocations (like 256 pages)<br /> have significantly more capacity than 256 * 170. This leads to pg_remaining<br /> being underestimated, which in turn makes skip (derived from skipped -<br /> pg_remaining) larger than expected, causing the WARN(skip != remaining)<br /> to trigger.<br /> <br /> Extra allocated pages for ftrace: 2 with 654 skipped<br /> WARNING: CPU: 0 PID: 0 at kernel/trace/ftrace.c:7295 ftrace_process_locs+0x5bf/0x5e0<br /> <br /> A similar problem in ftrace_allocate_records() can result in allocating<br /> too many pages. This can trigger the second warning in<br /> ftrace_process_locs().<br /> <br /> Extra allocated pages for ftrace<br /> WARNING: CPU: 0 PID: 0 at kernel/trace/ftrace.c:7276 ftrace_process_locs+0x548/0x580<br /> <br /> Use the actual capacity of a page group to determine the number of pages<br /> to allocate. Have ftrace_allocate_pages() return the number of allocated<br /> pages to avoid having to calculate it. Use the actual page group capacity<br /> when validating the number of unused pages due to skipped entries.<br /> Drop the definition of ENTRIES_PER_PAGE since it is no longer used.