LinuxLists.cc - [ext4:dev 6/8] fs/ext4/extents.c:4456 ext4_alloc_file_blocks() error: uninitialized symbol 'ret'.

2021-02-19 08:30:30

Subject: [ext4:dev 6/8] fs/ext4/extents.c:4456 ext4_alloc_file_blocks() error: uninitialized symbol 'ret'.

tree: https://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4.git dev
b578b311efdfedf92e6c
99d2d0b7395e14c8d097 [6/8] ext4: reset retry counter when ext4_alloc_file_blocks() makes progress
(attached as .config)
9.3.0-15) 9.3.0
kindly add following tag as appropriate
test robot <[email protected]>
<[email protected]>
ext4_alloc_file_blocks() error: uninitialized symbol 'ret'.
ext4_rereserve_cluster() warn: should '(1) << sbi->s_cluster_bits' be a 64 bit type?
i_size_write() warn: statement has no effect 31
ext4_clu_mapped() warn: should 'lclu << sbi->s_cluster_bits' be a 64 bit type?
ext4_ext_replay_set_iblocks() warn: should 'numblks << (inode->i_sb->s_blocksize_bits - 9)' be a 64 bit type?
2014-03-18 4379 static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset,
2014-08-27 4380 ext4_lblk_t len, loff_t new_size,
^^^^^^^^^^^^^^^
If so then we have a problem, if not then this is
2017-08-05 4381 int flags)
2014-03-18 4382 {
2014-03-18 4383 struct inode *inode = file_inode(file);
2014-03-18 4384 handle_t *handle;
2021-01-13 4385 int ret, ret2 = 0, ret3 = 0;
2014-03-18 4386 int retries = 0;
2015-06-15 4387 int depth = 0;
2014-03-18 4388 struct ext4_map_blocks map;
2014-03-18 4389 unsigned int credits;
2014-08-27 4390 loff_t epos;
2014-03-18 4391
2016-09-15 4392 BUG_ON(!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS));
2014-03-18 4393 map.m_lblk = offset;
2014-08-27 4394 map.m_len = len;
2014-03-18 4395 /*
2014-03-18 4396 * Don't normalize the request if it can fit in one extent so
2014-03-18 4397 * that it doesn't get unnecessarily split into multiple
2014-03-18 4398 * extents.
2014-03-18 4399 */
2014-04-20 4400 if (len <= EXT_UNWRITTEN_MAX_LEN)
2014-03-18 4401 flags |= EXT4_GET_BLOCKS_NO_NORMALIZE;
2014-03-18 4402
2014-03-18 4403 /*
2014-03-18 4404 * credits to insert 1 extent into extent tree
2014-03-18 4405 */
2014-03-18 4406 credits = ext4_chunk_trans_blocks(inode, len);
2015-06-15 4407 depth = ext_depth(inode);
2014-03-18 4408
2014-03-18 4409 retry:
2021-01-13 4410 while (len) {
^^^^^^^^^^^^^
2015-06-15 4411 /*
2015-06-15 4412 * Recalculate credits when extent tree depth changes.
2015-06-15 4413 */
2016-12-03 4414 if (depth != ext_depth(inode)) {
2015-06-15 4415 credits = ext4_chunk_trans_blocks(inode, len);
2015-06-15 4416 depth = ext_depth(inode);
2015-06-15 4417 }
2015-06-15 4418
2014-03-18 4419 handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS,
2014-03-18 4420 credits);
2014-03-18 4421 if (IS_ERR(handle)) {
2014-03-18 4422 ret = PTR_ERR(handle);
2014-03-18 4423 break;
2014-03-18 4424 }
2014-03-18 4425 ret = ext4_map_blocks(handle, inode, &map, flags);
2014-03-18 4426 if (ret <= 0) {
2014-03-18 4427 ext4_debug("inode #%lu: block %u: len %u: "
2014-03-18 4428 "ext4_ext_map_blocks returned %d",
2014-03-18 4429 inode->i_ino, map.m_lblk,
2014-03-18 4430 map.m_len, ret);
2014-03-18 4431 ext4_mark_inode_dirty(handle, inode);
2021-01-13 4432 ext4_journal_stop(handle);
2014-03-18 4433 break;
2014-03-18 4434 }
2021-01-13 4435 /*
2021-01-13 4436 * allow a full retry cycle for any remaining allocations
2021-01-13 4437 */
2021-01-13 4438 retries = 0;
2014-08-27 4439 map.m_lblk += ret;
2014-08-27 4440 map.m_len = len = len - ret;
2014-08-27 4441 epos = (loff_t)map.m_lblk << inode->i_blkbits;
2016-11-14 4442 inode->i_ctime = current_time(inode);
2014-08-27 4443 if (new_size) {
2014-08-27 4444 if (epos > new_size)
2014-08-27 4445 epos = new_size;
2014-08-27 4446 if (ext4_update_inode_size(inode, epos) & 0x1)
2014-08-27 4447 inode->i_mtime = inode->i_ctime;
2014-08-27 4448 }
Shirwadkar 2020-04-26 4449 ret2 = ext4_mark_inode_dirty(handle, inode);
2017-12-03 4450 ext4_update_inode_fsync_trans(handle, inode, 1);
Shirwadkar 2020-04-26 4451 ret3 = ext4_journal_stop(handle);
Shirwadkar 2020-04-26 4452 ret2 = ret3 ? ret3 : ret2;
Shirwadkar 2020-04-26 4453 if (unlikely(ret2))
2014-03-18 4454 break;
2014-03-18 4455 }
2021-01-13 @4456 if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
^^^^^^^^^^^^^^
2014-03-18 4457 goto retry;
2014-03-18 4458
2014-03-18 4459 return ret > 0 ? ret2 : ret;
2014-03-18 4460 }
Service, Intel Corporation
rg/hyperkitty/list/kbuild-all@lists.01.org">https://lists.01.org/hyperkitty/list/[email protected]

Attachments:

(No filename) (7.85 kB)
.config.gz (37.16 kB)
Download all attachments

2021-02-19 16:10:01

by Eric Whitney

[permalink] [raw]

Subject: Re: [ext4:dev 6/8] fs/ext4/extents.c:4456 ext4_alloc_file_blocks() error: uninitialized symbol 'ret'.

* Dan Carpenter <[email protected]>:
> tree: https://git.kernel.org/pub/scm/linux/kernel/git/tytso/ext4.git dev
> head: 0a76945fd1ba2ab44da7b578b311efdfedf92e6c
> commit: 3258386aba670e3406a499d2d0b7395e14c8d097 [6/8] ext4: reset retry counter when ext4_alloc_file_blocks() makes progress
> config: i386-randconfig-m021-20210215 (attached as .config)
> compiler: gcc-9 (Debian 9.3.0-15) 9.3.0
>
> If you fix the issue, kindly add following tag as appropriate
> Reported-by: kernel test robot <[email protected]>
> Reported-by: Dan Carpenter <[email protected]>
>
> New smatch warnings:
> fs/ext4/extents.c:4456 ext4_alloc_file_blocks() error: uninitialized symbol 'ret'.
>
> Old smatch warnings:
> fs/ext4/extents.c:2396 ext4_rereserve_cluster() warn: should '(1) << sbi->s_cluster_bits' be a 64 bit type?
> include/linux/fs.h:861 i_size_write() warn: statement has no effect 31
> fs/ext4/extents.c:5760 ext4_clu_mapped() warn: should 'lclu << sbi->s_cluster_bits' be a 64 bit type?
> fs/ext4/extents.c:6009 ext4_ext_replay_set_iblocks() warn: should 'numblks << (inode->i_sb->s_blocksize_bits - 9)' be a 64 bit type?
>
> vim +/ret +4456 fs/ext4/extents.c
>
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4379 static int ext4_alloc_file_blocks(struct file *file, ext4_lblk_t offset,
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4380 ext4_lblk_t len, loff_t new_size,
> ^^^^^^^^^^^^^^^
> Can "len" be zero? If so then we have a problem, if not then this is
> a false positive.

This should be a false positive, but there's no reason not to silence this
warning. I'll post a v2 correcting the original patch.

Eric

>
> 77a2e84d51729d Tahsin Erdogan 2017-08-05 4381 int flags)
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4382 {
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4383 struct inode *inode = file_inode(file);
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4384 handle_t *handle;
> 3258386aba670e Eric Whitney 2021-01-13 4385 int ret, ret2 = 0, ret3 = 0;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4386 int retries = 0;
> 4134f5c88dcd5b Lukas Czerner 2015-06-15 4387 int depth = 0;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4388 struct ext4_map_blocks map;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4389 unsigned int credits;
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4390 loff_t epos;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4391
> c3fe493ccdb1f4 Fabian Frederick 2016-09-15 4392 BUG_ON(!ext4_test_inode_flag(inode, EXT4_INODE_EXTENTS));
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4393 map.m_lblk = offset;
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4394 map.m_len = len;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4395 /*
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4396 * Don't normalize the request if it can fit in one extent so
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4397 * that it doesn't get unnecessarily split into multiple
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4398 * extents.
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4399 */
> 556615dcbf38b0 Lukas Czerner 2014-04-20 4400 if (len <= EXT_UNWRITTEN_MAX_LEN)
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4401 flags |= EXT4_GET_BLOCKS_NO_NORMALIZE;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4402
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4403 /*
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4404 * credits to insert 1 extent into extent tree
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4405 */
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4406 credits = ext4_chunk_trans_blocks(inode, len);
> 4134f5c88dcd5b Lukas Czerner 2015-06-15 4407 depth = ext_depth(inode);
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4408
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4409 retry:
> 3258386aba670e Eric Whitney 2021-01-13 4410 while (len) {
> ^^^^^^^^^^^^^
>
> 4134f5c88dcd5b Lukas Czerner 2015-06-15 4411 /*
> 4134f5c88dcd5b Lukas Czerner 2015-06-15 4412 * Recalculate credits when extent tree depth changes.
> 4134f5c88dcd5b Lukas Czerner 2015-06-15 4413 */
> 011c88e36c26a0 Dan Carpenter 2016-12-03 4414 if (depth != ext_depth(inode)) {
> 4134f5c88dcd5b Lukas Czerner 2015-06-15 4415 credits = ext4_chunk_trans_blocks(inode, len);
> 4134f5c88dcd5b Lukas Czerner 2015-06-15 4416 depth = ext_depth(inode);
> 4134f5c88dcd5b Lukas Czerner 2015-06-15 4417 }
> 4134f5c88dcd5b Lukas Czerner 2015-06-15 4418
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4419 handle = ext4_journal_start(inode, EXT4_HT_MAP_BLOCKS,
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4420 credits);
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4421 if (IS_ERR(handle)) {
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4422 ret = PTR_ERR(handle);
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4423 break;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4424 }
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4425 ret = ext4_map_blocks(handle, inode, &map, flags);
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4426 if (ret <= 0) {
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4427 ext4_debug("inode #%lu: block %u: len %u: "
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4428 "ext4_ext_map_blocks returned %d",
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4429 inode->i_ino, map.m_lblk,
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4430 map.m_len, ret);
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4431 ext4_mark_inode_dirty(handle, inode);
> 3258386aba670e Eric Whitney 2021-01-13 4432 ext4_journal_stop(handle);
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4433 break;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4434 }
> 3258386aba670e Eric Whitney 2021-01-13 4435 /*
> 3258386aba670e Eric Whitney 2021-01-13 4436 * allow a full retry cycle for any remaining allocations
> 3258386aba670e Eric Whitney 2021-01-13 4437 */
> 3258386aba670e Eric Whitney 2021-01-13 4438 retries = 0;
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4439 map.m_lblk += ret;
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4440 map.m_len = len = len - ret;
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4441 epos = (loff_t)map.m_lblk << inode->i_blkbits;
> eeca7ea1baa939 Deepa Dinamani 2016-11-14 4442 inode->i_ctime = current_time(inode);
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4443 if (new_size) {
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4444 if (epos > new_size)
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4445 epos = new_size;
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4446 if (ext4_update_inode_size(inode, epos) & 0x1)
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4447 inode->i_mtime = inode->i_ctime;
> c174e6d6979a04 Dmitry Monakhov 2014-08-27 4448 }
> 4209ae12b12265 Harshad Shirwadkar 2020-04-26 4449 ret2 = ext4_mark_inode_dirty(handle, inode);
> c894aa97577e47 Eryu Guan 2017-12-03 4450 ext4_update_inode_fsync_trans(handle, inode, 1);
> 4209ae12b12265 Harshad Shirwadkar 2020-04-26 4451 ret3 = ext4_journal_stop(handle);
> 4209ae12b12265 Harshad Shirwadkar 2020-04-26 4452 ret2 = ret3 ? ret3 : ret2;
> 4209ae12b12265 Harshad Shirwadkar 2020-04-26 4453 if (unlikely(ret2))
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4454 break;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4455 }
> 3258386aba670e Eric Whitney 2021-01-13 @4456 if (ret == -ENOSPC && ext4_should_retry_alloc(inode->i_sb, &retries))
> ^^^^^^^^^^^^^^
>
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4457 goto retry;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4458
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4459 return ret > 0 ? ret2 : ret;
> 0e8b6879f3c234 Lukas Czerner 2014-03-18 4460 }
>
> ---
> 0-DAY CI Kernel Test Service, Intel Corporation
> https://lists.01.org/hyperkitty/list/[email protected]