* [brauner-vfs:vfs.all 117/195] fs/iomap/direct-io.c:1074 iomap_dio_simple() warn: bitwise AND condition is false here
@ 2026-07-02 10:10 kernel test robot
0 siblings, 0 replies; only message in thread
From: kernel test robot @ 2026-07-02 10:10 UTC (permalink / raw)
To: oe-kbuild; +Cc: lkp, Dan Carpenter
BCC: lkp@intel.com
CC: oe-kbuild-all@lists.linux.dev
CC: Christian Brauner <christianvanbrauner@gmail.com>
TO: Fengnan Chang <changfengnan@bytedance.com>
CC: Christian Brauner <brauner@kernel.org>
CC: Christoph Hellwig <hch@lst.de>
tree: https://git.kernel.org/pub/scm/linux/kernel/git/vfs/vfs.git vfs.all
head: cf6f88615485a68df77092de1f90f88708a32fa6
commit: c45650e4f8ba25d02e860094a77bf85ec3a35f13 [117/195] iomap: add simple dio path for small direct I/O
:::::: branch date: 2 hours ago
:::::: commit date: 21 hours ago
config: i386-randconfig-141-20260702 (https://download.01.org/0day-ci/archive/20260702/202607021845.yVHnfllM-lkp@intel.com/config)
compiler: gcc-14 (Debian 14.2.0-19) 14.2.0
smatch: v0.5.0-9185-gbcc58b9c
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Reported-by: Dan Carpenter <error27@gmail.com>
| Closes: https://lore.kernel.org/r/202607021845.yVHnfllM-lkp@intel.com/
New smatch warnings:
fs/iomap/direct-io.c:1074 iomap_dio_simple() warn: bitwise AND condition is false here
Old smatch warnings:
fs/iomap/direct-io.c:376 iomap_dio_bio_iter_one() warn: bitwise AND condition is false here
fs/iomap/internal.h:15 iomap_max_bio_size() warn: bitwise AND condition is false here
vim +1074 fs/iomap/direct-io.c
c45650e4f8ba25 Fengnan Chang 2026-07-01 1002
c45650e4f8ba25 Fengnan Chang 2026-07-01 1003 /*
c45650e4f8ba25 Fengnan Chang 2026-07-01 1004 * Fast path for small, block-aligned direct I/Os that map to a single
c45650e4f8ba25 Fengnan Chang 2026-07-01 1005 * contiguous on-disk extent.
c45650e4f8ba25 Fengnan Chang 2026-07-01 1006 *
c45650e4f8ba25 Fengnan Chang 2026-07-01 1007 * iomap_dio_simple_supported() enforces the cheap up-front constraints before
c45650e4f8ba25 Fengnan Chang 2026-07-01 1008 * entering this path.
c45650e4f8ba25 Fengnan Chang 2026-07-01 1009 *
c45650e4f8ba25 Fengnan Chang 2026-07-01 1010 * @dops must be NULL: a non-NULL @dops means the caller wants its
c45650e4f8ba25 Fengnan Chang 2026-07-01 1011 * ->end_io / ->submit_io hooks invoked, and in particular wants its bios to be
c45650e4f8ba25 Fengnan Chang 2026-07-01 1012 * allocated from the filesystem-private @dops->bio_set (whose front_pad sizes a
c45650e4f8ba25 Fengnan Chang 2026-07-01 1013 * filesystem-private wrapper around the bio). The fast path instead allocates
c45650e4f8ba25 Fengnan Chang 2026-07-01 1014 * from the shared iomap_dio_simple_pool, whose front_pad matches struct
c45650e4f8ba25 Fengnan Chang 2026-07-01 1015 * iomap_dio_simple; the two wrappers are not interchangeable, so we must fall
c45650e4f8ba25 Fengnan Chang 2026-07-01 1016 * back to __iomap_dio_rw() in that case.
c45650e4f8ba25 Fengnan Chang 2026-07-01 1017 *
c45650e4f8ba25 Fengnan Chang 2026-07-01 1018 * @done_before must be zero: a non-zero caller-accumulated residual cannot be
c45650e4f8ba25 Fengnan Chang 2026-07-01 1019 * carried through a single-bio inline completion.
c45650e4f8ba25 Fengnan Chang 2026-07-01 1020 *
c45650e4f8ba25 Fengnan Chang 2026-07-01 1021 * @iter must describe a non-empty READ no larger than the inode block size:
c45650e4f8ba25 Fengnan Chang 2026-07-01 1022 * writes, zero-length I/O, and larger requests need the generic iomap direct
c45650e4f8ba25 Fengnan Chang 2026-07-01 1023 * I/O path.
c45650e4f8ba25 Fengnan Chang 2026-07-01 1024 *
c45650e4f8ba25 Fengnan Chang 2026-07-01 1025 * @dio_flags must not request IOMAP_DIO_FORCE_WAIT, IOMAP_DIO_PARTIAL, or
c45650e4f8ba25 Fengnan Chang 2026-07-01 1026 * IOMAP_DIO_BOUNCE: this path does not support forced waiting, partial direct
c45650e4f8ba25 Fengnan Chang 2026-07-01 1027 * I/O, or bouncing. The range must also stay within i_size and encrypted
c45650e4f8ba25 Fengnan Chang 2026-07-01 1028 * inodes must use the generic iomap direct I/O path.
c45650e4f8ba25 Fengnan Chang 2026-07-01 1029 *
c45650e4f8ba25 Fengnan Chang 2026-07-01 1030 * -ENOTBLK is the private sentinel returned by iomap_dio_simple() when it
c45650e4f8ba25 Fengnan Chang 2026-07-01 1031 * decides the request does not fit the fast path. In that case we proceed to
c45650e4f8ba25 Fengnan Chang 2026-07-01 1032 * the generic __iomap_dio_rw() slow path. Any other errno is a real result and
c45650e4f8ba25 Fengnan Chang 2026-07-01 1033 * is propagated as-is, in particular -EAGAIN for IOCB_NOWAIT must reach the
c45650e4f8ba25 Fengnan Chang 2026-07-01 1034 * caller.
c45650e4f8ba25 Fengnan Chang 2026-07-01 1035 */
c45650e4f8ba25 Fengnan Chang 2026-07-01 1036 static ssize_t
c45650e4f8ba25 Fengnan Chang 2026-07-01 1037 iomap_dio_simple(struct kiocb *iocb, struct iov_iter *iter,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1038 const struct iomap_ops *ops, void *private,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1039 unsigned int dio_flags)
c45650e4f8ba25 Fengnan Chang 2026-07-01 1040 {
c45650e4f8ba25 Fengnan Chang 2026-07-01 1041 struct inode *inode = file_inode(iocb->ki_filp);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1042 size_t count = iov_iter_count(iter);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1043 bool wait_for_completion = is_sync_kiocb(iocb);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1044 struct iomap_iter iomi = {
c45650e4f8ba25 Fengnan Chang 2026-07-01 1045 .inode = inode,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1046 .pos = iocb->ki_pos,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1047 .len = count,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1048 .flags = IOMAP_DIRECT,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1049 .private = private,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1050 };
c45650e4f8ba25 Fengnan Chang 2026-07-01 1051 struct iomap_dio_simple *sr;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1052 unsigned int alignment;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1053 struct bio *bio;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1054 ssize_t ret;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1055
c45650e4f8ba25 Fengnan Chang 2026-07-01 1056 if (iocb->ki_flags & IOCB_NOWAIT)
c45650e4f8ba25 Fengnan Chang 2026-07-01 1057 iomi.flags |= IOMAP_NOWAIT;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1058
c45650e4f8ba25 Fengnan Chang 2026-07-01 1059 ret = kiocb_write_and_wait(iocb, count);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1060 if (ret)
c45650e4f8ba25 Fengnan Chang 2026-07-01 1061 return ret;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1062
c45650e4f8ba25 Fengnan Chang 2026-07-01 1063 inode_dio_begin(inode);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1064
c45650e4f8ba25 Fengnan Chang 2026-07-01 1065 ret = ops->iomap_begin(inode, iomi.pos, count, iomi.flags,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1066 &iomi.iomap, &iomi.srcmap);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1067 if (ret) {
c45650e4f8ba25 Fengnan Chang 2026-07-01 1068 inode_dio_end(inode);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1069 return ret;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1070 }
c45650e4f8ba25 Fengnan Chang 2026-07-01 1071
c45650e4f8ba25 Fengnan Chang 2026-07-01 1072 if (iomi.iomap.type != IOMAP_MAPPED ||
c45650e4f8ba25 Fengnan Chang 2026-07-01 1073 iomi.iomap.offset + iomi.iomap.length < iomi.pos + count ||
c45650e4f8ba25 Fengnan Chang 2026-07-01 @1074 (iomi.iomap.flags & IOMAP_F_INTEGRITY)) {
c45650e4f8ba25 Fengnan Chang 2026-07-01 1075 ret = -ENOTBLK;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1076 goto out_iomap_end;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1077 }
c45650e4f8ba25 Fengnan Chang 2026-07-01 1078
c45650e4f8ba25 Fengnan Chang 2026-07-01 1079 alignment = iomap_dio_alignment(inode, iomi.iomap.bdev, dio_flags);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1080 if ((iomi.pos | count) & (alignment - 1)) {
c45650e4f8ba25 Fengnan Chang 2026-07-01 1081 ret = -EINVAL;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1082 goto out_iomap_end;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1083 }
c45650e4f8ba25 Fengnan Chang 2026-07-01 1084
c45650e4f8ba25 Fengnan Chang 2026-07-01 1085 if (!wait_for_completion && unlikely(!inode->i_sb->s_dio_done_wq)) {
c45650e4f8ba25 Fengnan Chang 2026-07-01 1086 ret = sb_init_dio_done_wq(inode->i_sb);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1087 if (ret < 0)
c45650e4f8ba25 Fengnan Chang 2026-07-01 1088 goto out_iomap_end;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1089 }
c45650e4f8ba25 Fengnan Chang 2026-07-01 1090
c45650e4f8ba25 Fengnan Chang 2026-07-01 1091 trace_iomap_dio_rw_begin(iocb, iter, dio_flags, 0);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1092
c45650e4f8ba25 Fengnan Chang 2026-07-01 1093 if (user_backed_iter(iter))
c45650e4f8ba25 Fengnan Chang 2026-07-01 1094 dio_flags |= IOMAP_DIO_USER_BACKED;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1095
c45650e4f8ba25 Fengnan Chang 2026-07-01 1096 bio = bio_alloc_bioset(iomi.iomap.bdev,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1097 bio_iov_vecs_to_alloc(iter, BIO_MAX_VECS),
c45650e4f8ba25 Fengnan Chang 2026-07-01 1098 REQ_OP_READ, GFP_KERNEL, &iomap_dio_simple_pool);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1099 sr = container_of(bio, struct iomap_dio_simple, bio);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1100 sr->iocb = iocb;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1101 sr->dio_flags = dio_flags;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1102
c45650e4f8ba25 Fengnan Chang 2026-07-01 1103 bio->bi_iter.bi_sector = iomap_sector(&iomi.iomap, iomi.pos);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1104 bio->bi_ioprio = iocb->ki_ioprio;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1105
c45650e4f8ba25 Fengnan Chang 2026-07-01 1106 ret = bio_iov_iter_get_pages(bio, iter, alignment - 1);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1107 if (unlikely(ret))
c45650e4f8ba25 Fengnan Chang 2026-07-01 1108 goto out_bio_put;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1109
c45650e4f8ba25 Fengnan Chang 2026-07-01 1110 if (bio->bi_iter.bi_size != count) {
c45650e4f8ba25 Fengnan Chang 2026-07-01 1111 iov_iter_revert(iter, bio->bi_iter.bi_size);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1112 ret = -ENOTBLK;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1113 goto out_bio_release_pages;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1114 }
c45650e4f8ba25 Fengnan Chang 2026-07-01 1115
c45650e4f8ba25 Fengnan Chang 2026-07-01 1116 sr->size = bio->bi_iter.bi_size;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1117
c45650e4f8ba25 Fengnan Chang 2026-07-01 1118 if (dio_flags & IOMAP_DIO_USER_BACKED)
c45650e4f8ba25 Fengnan Chang 2026-07-01 1119 bio_set_pages_dirty(bio);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1120
c45650e4f8ba25 Fengnan Chang 2026-07-01 1121 if (iocb->ki_flags & IOCB_NOWAIT)
c45650e4f8ba25 Fengnan Chang 2026-07-01 1122 bio->bi_opf |= REQ_NOWAIT;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1123 if ((iocb->ki_flags & IOCB_HIPRI) && !wait_for_completion) {
c45650e4f8ba25 Fengnan Chang 2026-07-01 1124 bio->bi_opf |= REQ_POLLED;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1125 WRITE_ONCE(iocb->private, bio);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1126 }
c45650e4f8ba25 Fengnan Chang 2026-07-01 1127
c45650e4f8ba25 Fengnan Chang 2026-07-01 1128 if (ops->iomap_end)
c45650e4f8ba25 Fengnan Chang 2026-07-01 1129 ops->iomap_end(inode, iomi.pos, count, count, iomi.flags,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1130 &iomi.iomap);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1131
c45650e4f8ba25 Fengnan Chang 2026-07-01 1132 if (!wait_for_completion) {
c45650e4f8ba25 Fengnan Chang 2026-07-01 1133 bio->bi_end_io = iomap_dio_simple_end_io;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1134 submit_bio(bio);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1135 trace_iomap_dio_rw_queued(inode, iomi.pos, count);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1136 return -EIOCBQUEUED;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1137 }
c45650e4f8ba25 Fengnan Chang 2026-07-01 1138
c45650e4f8ba25 Fengnan Chang 2026-07-01 1139 submit_bio_wait(bio);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1140 return iomap_dio_simple_complete(sr);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1141
c45650e4f8ba25 Fengnan Chang 2026-07-01 1142 out_bio_release_pages:
c45650e4f8ba25 Fengnan Chang 2026-07-01 1143 bio_release_pages(bio, false);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1144 out_bio_put:
c45650e4f8ba25 Fengnan Chang 2026-07-01 1145 bio_put(bio);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1146 out_iomap_end:
c45650e4f8ba25 Fengnan Chang 2026-07-01 1147 if (ops->iomap_end)
c45650e4f8ba25 Fengnan Chang 2026-07-01 1148 ops->iomap_end(inode, iomi.pos, count, 0, iomi.flags,
c45650e4f8ba25 Fengnan Chang 2026-07-01 1149 &iomi.iomap);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1150 inode_dio_end(inode);
c45650e4f8ba25 Fengnan Chang 2026-07-01 1151 return ret;
c45650e4f8ba25 Fengnan Chang 2026-07-01 1152 }
c45650e4f8ba25 Fengnan Chang 2026-07-01 1153
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2026-07-02 10:10 UTC | newest]
Thread overview: (only message) (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2026-07-02 10:10 [brauner-vfs:vfs.all 117/195] fs/iomap/direct-io.c:1074 iomap_dio_simple() warn: bitwise AND condition is false here kernel test robot
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox