From: Joonsoo Kim <iamjoonsoo.kim@...> commit 40b44137971c2e5865a78f9f7de274449983ccb5 upstream. Currently, open code for checking DEBUG_PAGEALLOC cache is spread to some sites. It makes code unreadab

From: Laura Abbott <labbott@...> commit 8823b1dbc05fab1a8bec275eeae4709257c2661d upstream. Page poisoning is currently set up as a feature if architectures don't have architecture debug page_alloc to

From: Laura Abbott <labbott@...> commit 1414c7f4f7d72d138fff35f00151d15749b5beda upstream. By default, page poisoning uses a poison value (0xaa) on free. If this is changed to 0, the page is not only

This is a backport of changes in 4.5 to add support for GCC's Undefined Behaviour Sanitizer (UBSAN), and fixes for undefined behaviour that it has found (that are not already in 4.4.30). UBSAN detects

From: Andrey Ryabinin <aryabinin@...> commit c6d308534aef6c99904bf5862066360ae067abc4 upstream. UBSAN uses compile-time instrumentation to catch undefined behavior (UB). Compiler inserts code that per

From: Yang Shi <yang.shi@...> commit 7707535ab95e2231b6d7f2bfb4f27558e83c4dc2 upstream. When enabling UBSAN_SANITIZE_ALL, the kernel image size gets increased significantly (~3x). So, it sounds better

From: Eric Biggers <ebiggers3@...> commit f97238373b8662a6d580e204df2e7bcbfa43e27a upstream. Accessing more than one byte from a symbol declared simply 'char' is undefined behavior, as reported by UBS

From: Chris Bainbridge <chris.bainbridge@...> commit bc864af13f34d19c911f5691d87bdacc9ce109f5 upstream. Microcode checksum verification should be done using unsigned 32-bit values otherwise the calcul

From: Andrey Ryabinin <aryabinin@...> commit d59b1087a98e402ed9a7cc577f4da435f9a555f5 upstream. Calculation of dirty_ratelimit sometimes is not correct. E.g. initial values of dirty_ratelimit == INIT_

From: Peter Zijlstra <peterz@...> commit 8184059e93c200757f5c0805dae0f14e880eab5d upstream. Sasha reported: [ 3494.030114] UBSAN: Undefined behaviour in kernel/events/ring_buffer.c:685:22 [ 3494.03064

From: Andrey Ryabinin <aryabinin@...> commit dde5cf39d4d2cce71f2997c37210dd624d0e4bf6 upstream. -fsanitize=* options makes GCC less smart than usual and increase number of 'maybe-uninitialized' false-

From: Nicolai Stange <nicstange@...> commit e7080a439a6f507abbc860847c33efc39b5c1c6d upstream. If - generic_file_read_iter() gets called with a zero read length, - the read offset is at a page boundar

From: Adam Borowski <kilobyte@...> commit 0a25556f84d5f79e68e9502bb1f32a43377ab2bf upstream. The entry for PERF_COUNT_HW_REF_CPU_CYCLES is not used on AMD, but is referenced by filter_events() which e

From: Nicolai Stange <nicstange@...> commit b2c0cbd657173f024138d6421774007690ceeffd upstream. The values of all but the RADEON_HPD_NONE members of the radeon_hpd_id enum transform 1:1 into bit positi

From: Adam Borowski <kilobyte@...> commit 8eb0dfdbda3f56bf7d248ed87fcc383df114ecbb upstream. UBSAN: Undefined behaviour in fs/btrfs/extent-tree.c:4623:21 signed integer overflow: 10808 * 262144 cannot

From: Bartlomiej Zolnierkiewicz <b.zolnierkie@...> commit b3a834b1596ac668df206aa2bb1f191c31f5f5e4 upstream. When this_order variable in blk_mq_init_rq_map() becomes zero the code incorrectly decremen

From: Sven Eckelmann <sven.eckelmann@...> commit d285f52cc0f23564fd61976d43fd5b991b4828f6 upstream. The undefined behavior sanatizer detected an signed integer overflow in a setup with near perfect li

From: Oleg Nesterov <oleg@...> commit 5c8ccefdf46c5f87d87b694c7fbc04941c2c99a5 upstream. All the users of siginmask() must ensure that sig < SIGRTMIN. sig_fatal() doesn't and this is wrong: UBSAN: Und

From: Seung-Woo Kim <sw0312.kim@...> commit 65257a0deed5aee66b4e3708944f0be62a64cabc upstream. This patch removes following UBSAN warnings in dw_mci_setup_bus(). UBSAN: Undefined behaviour in drivers/

From: Nicolas Iooss <nicolas.iooss_linux@...> commit 901d805c33fc4c029fc6b2d94ee5fb7d30278045 upstream. handle_object_size_mismatch() used %pk to format a kernel pointer with pr_err(). This seemed to