)]}'
{
  "commit": "137f4607b36c0dc2df5910d60cd322fd2cc8c5dc",
  "tree": "69997402d9f68c3630566211626f0335802c5001",
  "parents": [
    "d0e2b5bf0557a9ff0d5ac02d9afaa8cd3b8087e4"
  ],
  "author": {
    "name": "Yu Kuai",
    "email": "yukuai3@huawei.com",
    "time": "Fri Nov 29 17:15:09 2024 +0800"
  },
  "committer": {
    "name": "Arnav Kansal",
    "email": "rnv@google.com",
    "time": "Tue Mar 18 16:07:21 2025 -0700"
  },
  "message": "block, bfq: fix bfqq uaf in bfq_limit_depth()\n\ncommit e8b8344de3980709080d86c157d24e7de07d70ad upstream.\n\nSet new allocated bfqq to bic or remove freed bfqq from bic are both\nprotected by bfqd-\u003elock, however bfq_limit_depth() is deferencing bfqq\nfrom bic without the lock, this can lead to UAF if the io_context is\nshared by multiple tasks.\n\nFor example, test bfq with io_uring can trigger following UAF in v6.6:\n\n\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\u003d\nBUG: KASAN: slab-use-after-free in bfqq_group+0x15/0x50\n\nCall Trace:\n \u003cTASK\u003e\n dump_stack_lvl+0x47/0x80\n print_address_description.constprop.0+0x66/0x300\n print_report+0x3e/0x70\n kasan_report+0xb4/0xf0\n bfqq_group+0x15/0x50\n bfqq_request_over_limit+0x130/0x9a0\n bfq_limit_depth+0x1b5/0x480\n __blk_mq_alloc_requests+0x2b5/0xa00\n blk_mq_get_new_requests+0x11d/0x1d0\n blk_mq_submit_bio+0x286/0xb00\n submit_bio_noacct_nocheck+0x331/0x400\n __block_write_full_folio+0x3d0/0x640\n writepage_cb+0x3b/0xc0\n write_cache_pages+0x254/0x6c0\n write_cache_pages+0x254/0x6c0\n do_writepages+0x192/0x310\n filemap_fdatawrite_wbc+0x95/0xc0\n __filemap_fdatawrite_range+0x99/0xd0\n filemap_write_and_wait_range.part.0+0x4d/0xa0\n blkdev_read_iter+0xef/0x1e0\n io_read+0x1b6/0x8a0\n io_issue_sqe+0x87/0x300\n io_wq_submit_work+0xeb/0x390\n io_worker_handle_work+0x24d/0x550\n io_wq_worker+0x27f/0x6c0\n ret_from_fork_asm+0x1b/0x30\n \u003c/TASK\u003e\n\nAllocated by task 808602:\n kasan_save_stack+0x1e/0x40\n kasan_set_track+0x21/0x30\n __kasan_slab_alloc+0x83/0x90\n kmem_cache_alloc_node+0x1b1/0x6d0\n bfq_get_queue+0x138/0xfa0\n bfq_get_bfqq_handle_split+0xe3/0x2c0\n bfq_init_rq+0x196/0xbb0\n bfq_insert_request.isra.0+0xb5/0x480\n bfq_insert_requests+0x156/0x180\n blk_mq_insert_request+0x15d/0x440\n blk_mq_submit_bio+0x8a4/0xb00\n submit_bio_noacct_nocheck+0x331/0x400\n __blkdev_direct_IO_async+0x2dd/0x330\n blkdev_write_iter+0x39a/0x450\n io_write+0x22a/0x840\n io_issue_sqe+0x87/0x300\n io_wq_submit_work+0xeb/0x390\n io_worker_handle_work+0x24d/0x550\n io_wq_worker+0x27f/0x6c0\n ret_from_fork+0x2d/0x50\n ret_from_fork_asm+0x1b/0x30\n\nFreed by task 808589:\n kasan_save_stack+0x1e/0x40\n kasan_set_track+0x21/0x30\n kasan_save_free_info+0x27/0x40\n __kasan_slab_free+0x126/0x1b0\n kmem_cache_free+0x10c/0x750\n bfq_put_queue+0x2dd/0x770\n __bfq_insert_request.isra.0+0x155/0x7a0\n bfq_insert_request.isra.0+0x122/0x480\n bfq_insert_requests+0x156/0x180\n blk_mq_dispatch_plug_list+0x528/0x7e0\n blk_mq_flush_plug_list.part.0+0xe5/0x590\n __blk_flush_plug+0x3b/0x90\n blk_finish_plug+0x40/0x60\n do_writepages+0x19d/0x310\n filemap_fdatawrite_wbc+0x95/0xc0\n __filemap_fdatawrite_range+0x99/0xd0\n filemap_write_and_wait_range.part.0+0x4d/0xa0\n blkdev_read_iter+0xef/0x1e0\n io_read+0x1b6/0x8a0\n io_issue_sqe+0x87/0x300\n io_wq_submit_work+0xeb/0x390\n io_worker_handle_work+0x24d/0x550\n io_wq_worker+0x27f/0x6c0\n ret_from_fork+0x2d/0x50\n ret_from_fork_asm+0x1b/0x30\n\nFix the problem by protecting bic_to_bfqq() with bfqd-\u003elock.\n\nBUG\u003db/390599687\nTEST\u003dpresubmit\nRELEASE_NOTE\u003dFixed CVE-2024-53166 in the Linux kernel.\n\ncos-patch: security-moderate\nChange-Id: I688d384d32a7e30f75c7023287663686932c787b\nSigned-off-by: Arnav Kansal \u003crnv@google.com\u003e\nCC: Jan Kara \u003cjack@suse.cz\u003e\nFixes: 76f1df88bbc2 (\"bfq: Limit number of requests consumed by each cgroup\")\nSigned-off-by: Yu Kuai \u003cyukuai3@huawei.com\u003e\nLink: https://lore.kernel.org/r/20241129091509.2227136-1-yukuai1@huaweicloud.com\nSigned-off-by: Jens Axboe \u003caxboe@kernel.dk\u003e\nSigned-off-by: Hagar Hemdan \u003chagarhem@amazon.com\u003e\nSigned-off-by: Greg Kroah-Hartman \u003cgregkh@linuxfoundation.org\u003e\nReviewed-on: https://cos-review.googlesource.com/c/third_party/kernel/+/96460\nTested-by: Cusky Presubmit Bot \u003cpresubmit@cos-infra-prod.iam.gserviceaccount.com\u003e\nReviewed-by: Shuo Yang \u003cgshuoy@google.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "76b2e66d09070d602bdd37e0b1ea1ce442780f62",
      "old_mode": 33188,
      "old_path": "block/bfq-iosched.c",
      "new_id": "d1c62a4e8c357bd39c4783674bc5d2b5b7cd2e86",
      "new_mode": 33188,
      "new_path": "block/bfq-iosched.c"
    }
  ]
}