Skip to content

Commit 0d24f1b

Browse files
nathanchancemasahir0y
authored andcommitted
padata: Mark padata_work_init() as __ref
When building arm64 allmodconfig + ThinLTO with clang and a proposed modpost update to account for -ffuncton-sections, the following warning appears: WARNING: modpost: vmlinux.o: section mismatch in reference: padata_work_init (section: .text.padata_work_init) -> padata_mt_helper (section: .init.text) WARNING: modpost: vmlinux.o: section mismatch in reference: padata_work_init (section: .text.padata_work_init) -> padata_mt_helper (section: .init.text) LLVM has optimized padata_work_init() to include the address of padata_mt_helper() directly because it inlined the other call to padata_work_init() with padata_parallel_worker(), meaning the remaining uses of padata_work_init() use padata_mt_helper() as the work_fn argument. This optimization causes modpost to complain since padata_work_init() is not __init, whereas padata_mt_helper() is. Since padata_work_init() is only called from __init code when padata_mt_helper() is passed as the work_fn argument, mark padata_work_init() as __ref, which makes it clear to modpost that this scenario is okay. Suggested-by: Daniel Jordan <[email protected]> Signed-off-by: Nathan Chancellor <[email protected]> Acked-by: Daniel Jordan <[email protected]> Signed-off-by: Masahiro Yamada <[email protected]>
1 parent 87d599f commit 0d24f1b

File tree

1 file changed

+10
-2
lines changed

1 file changed

+10
-2
lines changed

kernel/padata.c

Lines changed: 10 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -83,8 +83,16 @@ static struct padata_work *padata_work_alloc(void)
8383
return pw;
8484
}
8585

86-
static void padata_work_init(struct padata_work *pw, work_func_t work_fn,
87-
void *data, int flags)
86+
/*
87+
* This function is marked __ref because this function may be optimized in such
88+
* a way that it directly refers to work_fn's address, which causes modpost to
89+
* complain when work_fn is marked __init. This scenario was observed with clang
90+
* LTO, where padata_work_init() was optimized to refer directly to
91+
* padata_mt_helper() because the calls to padata_work_init() with other work_fn
92+
* values were eliminated or inlined.
93+
*/
94+
static void __ref padata_work_init(struct padata_work *pw, work_func_t work_fn,
95+
void *data, int flags)
8896
{
8997
if (flags & PADATA_WORK_ONSTACK)
9098
INIT_WORK_ONSTACK(&pw->pw_work, work_fn);

0 commit comments

Comments
 (0)