Hi "André, [FYI, it's a private test report for your RFC patch.] [auto build test WARNING on tip/locking/core] [also build test WARNING on tip/x86/asm arm64/for-next/core tip/perf/core linus/master v5.11] [cannot apply to next-20210215] [If your patch is applied to the wrong git tree, kindly drop us a note. And when submitting patch, we suggest to use '--base' as documented in https://git-scm.com/docs/git-format-patch] url: https://github.com/0day-ci/linux/commits/Andr-Almeida/Add-futex2-syscalls... base: https://git.kernel.org/pub/scm/linux/kernel/git/tip/tip.git 3765d01bab73bdb920ef711203978f02cd26e4da config: x86_64-randconfig-s022-20210215 (attached as .config) compiler: gcc-9 (Debian 9.3.0-15) 9.3.0 reproduce: # apt-get install sparse # sparse version: v0.6.3-215-g0fb77bb6-dirty # https://github.com/0day-ci/linux/commit/d1b45b031c4017655fa7507e130260a54... git remote add linux-review https://github.com/0day-ci/linux git fetch --no-tags linux-review Andr-Almeida/Add-futex2-syscalls/20210215-233004 git checkout d1b45b031c4017655fa7507e130260a5419e791b # save the attached .config to linux build tree make W=1 C=1 CF='-fdiagnostic-prefix -D__CHECK_ENDIAN__' ARCH=x86_64 If you fix the issue, kindly add following tag as appropriate Reported-by: kernel test robot <lkp(a)intel.com&gt; "sparse warnings: (new ones prefixed by >>)" kernel/futex2.c:107:21: sparse: sparse: symbol 'futex_table' was not declared. Should it be static? kernel/futex2.c:108:14: sparse: sparse: symbol 'futex2_hashsize' was not declared. Should it be static? kernel/futex2.c:354:13: sparse: sparse: incorrect type in argument 1 (different base types) @@ expected void const volatile [noderef] __user *ptr @@ got unsigned long [usertype] address @@ kernel/futex2.c:354:13: sparse: expected void const volatile [noderef] __user *ptr kernel/futex2.c:354:13: sparse: got unsigned long [usertype] address kernel/futex2.c:513:51: sparse: expected void [noderef] __user *uaddr kernel/futex2.c:513:51: sparse: got void * kernel/futex2.c:528:45: sparse: sparse: incorrect type in argument 2 (different address spaces) @@ expected unsigned int [noderef] [usertype] __user *uaddr @@ got unsigned int [usertype] *[assigned] uaddr @@ kernel/futex2.c:528:45: sparse: expected unsigned int [noderef] [usertype] __user *uaddr kernel/futex2.c:528:45: sparse: got unsigned int [usertype] *[assigned] uaddr kernel/futex2.c:537:29: sparse: sparse: incorrect type in argument 1 (different address spaces) @@ expected void const volatile [noderef] __user *ptr @@ got unsigned int [usertype] *[assigned] uaddr @@ kernel/futex2.c:537:29: sparse: expected void const volatile [noderef] __user *ptr kernel/futex2.c:537:29: sparse: got unsigned int [usertype] *[assigned] uaddr vim +513 kernel/futex2.c 456 457 /** 458 * futex_enqueue - Check the value and enqueue a futex on a wait list 459 * 460 * @futexv: List of futexes 461 * @nr_futexes: Number of futexes in the list 462 * @awakened: If a futex was awakened during enqueueing, store the index here 463 * 464 * Get the value from the userspace address and compares with the expected one. 465 * 466 * Getting the value from user futex address: 467 * 468 * Since we are in a hurry, we use a spin lock and we can't sleep. 469 * Try to get the value with page fault disabled (when enable, we might 470 * sleep). 471 * 472 * If we fail, we aren't sure if the address is invalid or is just a 473 * page fault. Then, release the lock (so we can sleep) and try to get 474 * the value with page fault enabled. In order to trigger a page fault 475 * handling, we just call __get_user() again. If we sleep with enqueued 476 * futexes, we might miss a wake, so dequeue everything before sleeping. 477 * 478 * If get_user succeeds, this mean that the address is valid and we do 479 * the work again. Since we just handled the page fault, the page is 480 * likely pinned in memory and we should be luckier this time and be 481 * able to get the value. If we fail anyway, we will try again. 482 * 483 * If even with page faults enabled we get and error, this means that 484 * the address is not valid and we return from the syscall. 485 * 486 * If we got an unexpected value or need to treat a page fault and realized that 487 * a futex was awakened, we can priority this and return success. 488 * 489 * In success, enqueue the futex in the correct bucket 490 * 491 * Return: 492 * * 1 - We were awake in the process and nothing is enqueued 493 * * 0 - Everything is enqueued and we are ready to sleep 494 * * 0< - Something went wrong, nothing is enqueued, return error code 495 */ 496 static int futex_enqueue(struct futexv_head *futexv, unsigned int nr_futexes, 497 int *awakened) 498 { 499 int i, ret; 500 bool retry = false; 501 u32 uval, *uaddr, val; 502 struct futex_bucket *bucket; 503 504 retry: 505 set_current_state(TASK_INTERRUPTIBLE); 506 507 for (i = 0; i < nr_futexes; i++) { 508 uaddr = (u32 * __user)futexv->objects[i].uaddr; 509 val = (u32)futexv->objects[i].val; 510 511 if (is_object_shared && retry) { 512 struct futex_bucket *tmp = 514 &futexv->objects[i].key, true); 515 if (IS_ERR(tmp)) { 516 __set_current_state(TASK_RUNNING); 517 futex_dequeue_multiple(futexv, i); 518 return PTR_ERR(tmp); 519 } 520 futexv->objects[i].bucket = tmp; 521 } 522 523 bucket = futexv->objects[i].bucket; 524 525 bucket_inc_waiters(bucket); 526 spin_lock(&bucket->lock); 527 528 ret = futex_get_user(&uval, uaddr); 529 530 if (unlikely(ret)) { 531 spin_unlock(&bucket->lock); 532 533 bucket_dec_waiters(bucket); 534 __set_current_state(TASK_RUNNING); 535 *awakened = futex_dequeue_multiple(futexv, i); 536 537 if (__get_user(uval, uaddr)) 538 return -EFAULT; 539 540 if (*awakened >= 0) 541 return 1; 542 543 retry = true; 544 goto retry; 545 } 546 547 if (uval != val) { 548 spin_unlock(&bucket->lock); 549 550 bucket_dec_waiters(bucket); 551 __set_current_state(TASK_RUNNING); 552 *awakened = futex_dequeue_multiple(futexv, i); 553 554 if (*awakened >= 0) 555 return 1; 556 557 return -EAGAIN; 558 } 559 560 list_add_tail(&futexv->objects[i].list, &bucket->list); 561 spin_unlock(&bucket->lock); 562 } 563 564 return 0; 565 } 566 --- 0-DAY CI Kernel Test Service, Intel Corporation https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org