casaroli opened a new pull request, #4079: URL: https://github.com/apache/nuttx-apps/pull/4079
depends-on: apache/nuttx/pull/21961 ## Summary `malloc()`, `calloc()` and `realloc()` must return memory that is aligned for every type (C11 7.22.3), that is to `alignof(max_align_t)`. No test checks this today. This PR adds case 9 to the mm cmocka suite (`cmocka_mm_test`, `CONFIG_CM_MM_TEST`): it allocates 1024 blocks of many sizes, frees one in three, reallocates the others and fills the holes with `calloc()`, and counts the pointers that are not aligned to `alignof(max_align_t)`. The case gets the alignment from `offsetof()` of a `max_align_t` member, so it needs no C11 keyword. It reads each address back through a `volatile`: GCC assumes that `malloc()` memory is aligned to 16 on 64-bit targets, and at -O2 it removes a check such as `(uintptr_t)p % 16`. Depends on apache/nuttx#21961, which makes the default heap alignment 16 on 64-bit architectures. With the old default 8, this case fails there. ## Impact Test only. The new file is in the existing `kernel/mm/cases/` directory, which the Make and CMake builds take with a wildcard. ## Testing `cmocka_mm_test` with `CONFIG_CM_MM_TEST` (and cmocka: `ALLOW_MIT_COMPONENTS`, `LIBC_REGEX`, `ARCH_SETJMP_H`): | configuration | nuttx master | with apache/nuttx#21961 | |---|---|---| | qemu-armv8a:nsh (flat, arm64, QEMU HVF) | `test_nuttx_mm09` FAIL: 1023 pointers not aligned; cases 1-8 PASS | 9 of 9 PASS | | rv-virt:nsh64 (flat, RV64, QEMU TCG) | `test_nuttx_mm09` FAIL: 997 pointers not aligned; cases 1-8 PASS | 9 of 9 PASS | ostest passes (exit status 0) in the same qemu-armv8a:nsh images. tools/checkpatch.sh passes. -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
