The patch titled
Subject: drivers/acpi/nfit/core.c: fix the abuse of COMPLETION_INITIALIZER_ONSTACK()
has been removed from the -mm tree. Its filename was
nfit-use-init_completion-in-acpi_nfit_flush_probe.patch
This patch was dropped because it was merged into mainline or a subsystem tree
------------------------------------------------------
From: Boqun Feng <boqun.feng@xxxxxxxxx>
Subject: drivers/acpi/nfit/core.c: fix the abuse of COMPLETION_INITIALIZER_ONSTACK()
Patch series "completion: Reduce stack usage caused by
COMPLETION_INITIALIZER_ONSTACK()".
With LOCKDEP_CROSSRELEASE and LOCKDEP_COMPLETIONS introduced, the growth
in kernel stack usage of several functions were reported:
https://marc.info/?l=linux-kernel&m=150270063231284&w=2
The root cause of this is in COMPLETION_INITIALIZER_ONSTACK(), we use
({init_completion(&work); work})
which will create a temporary object when returned. However this
temporary object is unnecessary. And this patch fixes it by making the
statement expression in COMPLETION_INITIALIZER_ONSTACK() return a pointer
rather than a whole structure. This will reduce the stack usage even if
!LOCKDEP.
However, such a change does make one COMPLETION_INITIALIZER_ONSTACK()
callsite invalid, so we fix this first via converting to
init_completion().
This patch (of 2):
COMPLETION_INITIALIZER_ONSTACK() is supposed to used as an initializer, in
other words, it should only be used in assignment expressions or compound
literals. So the usage in drivers/acpi/nfit/core.c:
COMPLETION_INITIALIZER_ONSTACK(flush.cmp);
is inappropriate.
Besides, this usage could also break compilations for another fix to
reduce stack sizes caused by COMPLETION_INITIALIZER_ONSTACK(), because
that fix changes COMPLETION_INITIALIZER_ONSTACK() from rvalue to lvalue,
and usage as above will report error:
drivers/acpi/nfit/core.c: In function 'acpi_nfit_flush_probe':
include/linux/completion.h:77:3: error: value computed is not used [-Werror=unused-value]
(*({ init_completion(&work); &work; }))
This patch fixes this by replacing COMPLETION_INITIALIZER_ONSTACK() with
init_completion() in acpi_nfit_flush_probe(), which does the same
initialization without any other problem.
Link: http://lkml.kernel.org/r/20170823152542.5150-2-boqun.feng@xxxxxxxxx
Signed-off-by: Boqun Feng <boqun.feng@xxxxxxxxx>
Acked-by: Arnd Bergmann <arnd@xxxxxxxx>
Acked-by: Dan Williams <dan.j.williams@xxxxxxxxx>
Cc: Peter Zijlstra <peterz@xxxxxxxxxxxxx>
Cc: Ingo Molnar <mingo@xxxxxxxxxx>
Cc: Thomas Gleixner <tglx@xxxxxxxxxxxxx>
Cc: Michel Lespinasse <walken@xxxxxxxxxx>
Cc: Byungchul Park <byungchul.park@xxxxxxx>
Cc: Nicholas Piggin <npiggin@xxxxxxxxx>
Cc: "Rafael J. Wysocki" <rjw@xxxxxxxxxxxxx>
Signed-off-by: Andrew Morton <akpm@xxxxxxxxxxxxxxxxxxxx>
---
drivers/acpi/nfit/core.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff -puN drivers/acpi/nfit/core.c~nfit-use-init_completion-in-acpi_nfit_flush_probe drivers/acpi/nfit/core.c
--- a/drivers/acpi/nfit/core.c~nfit-use-init_completion-in-acpi_nfit_flush_probe
+++ a/drivers/acpi/nfit/core.c
@@ -2884,7 +2884,7 @@ static int acpi_nfit_flush_probe(struct
* need to be interruptible while waiting.
*/
INIT_WORK_ONSTACK(&flush.work, flush_probe);
- COMPLETION_INITIALIZER_ONSTACK(flush.cmp);
+ init_completion(&flush.cmp);
queue_work(nfit_wq, &flush.work);
mutex_unlock(&acpi_desc->init_mutex);
_
Patches currently in -mm which might be from boqun.feng@xxxxxxxxx are
--
To unsubscribe from this list: send the line "unsubscribe mm-commits" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
