From: mwilck@suse.com
To: Christophe Varoqui <christophe.varoqui@opensvc.com>,
Benjamin Marzinski <bmarzins@redhat.com>
Cc: dm-devel@redhat.com, Martin Wilck <mwilck@suse.com>
Subject: [dm-devel] [PATCH v3 38/38] libmultipath: avoid -Warray-bounds error in uatomic operations
Date: Thu, 14 Sep 2023 16:51:31 +0200 [thread overview]
Message-ID: <20230914145131.15165-4-mwilck@suse.com> (raw)
In-Reply-To: <20230914145131.15165-1-mwilck@suse.com>
From: Martin Wilck <mwilck@suse.com>
The use of uatomic_xchg() in alias.c causes a -Warray-bounds error
on distributions using gcc 12, such as Fedora 37. This is a similar
error to 2534c4f ("libmultipath: avoid -Warray-bounds error with gcc
12 and musl libc"). This happens only with liburcu 0.13 and earlier,
and only with certain gcc versions. See liburcu commit 835b9ab
("Fix: x86 and s390 uatomic: __hp() macro warning with gcc 11").
Enhance the fix for 2534c4f by a adding a workaround for uatomic_xchg(),
and introduce the macro URCU_VERSION (originally only used for multipathd)
globally.
Signed-off-by: Martin Wilck <mwilck@suse.com>
---
Makefile.inc | 2 +-
create-config.mk | 5 +++++
libmultipath/alias.c | 5 +++--
libmultipath/lock.h | 23 ++++++++++++++---------
multipathd/Makefile | 2 --
5 files changed, 23 insertions(+), 14 deletions(-)
diff --git a/Makefile.inc b/Makefile.inc
index 6e384e6..04bfa56 100644
--- a/Makefile.inc
+++ b/Makefile.inc
@@ -95,7 +95,7 @@ OPTFLAGS := -O2 -g $(STACKPROT) --param=ssp-buffer-size=4
WARNFLAGS := -Werror -Wall -Wextra -Wformat=2 $(WFORMATOVERFLOW) -Werror=implicit-int \
-Werror=implicit-function-declaration -Werror=format-security \
$(WNOCLOBBERED) -Werror=cast-qual $(ERROR_DISCARDED_QUALIFIERS) $(W_URCU_TYPE_LIMITS)
-CPPFLAGS := $(FORTIFY_OPT) $(CPPFLAGS) \
+CPPFLAGS := $(FORTIFY_OPT) $(CPPFLAGS) $(D_URCU_VERSION) \
-DBIN_DIR=\"$(bindir)\" -DMULTIPATH_DIR=\"$(plugindir)\" \
-DRUNTIME_DIR=\"$(runtimedir)\" -DCONFIG_DIR=\"$(configdir)\" \
-DDEFAULT_CONFIGFILE=\"$(configfile)\" -DSTATE_DIR=\"$(statedir)\" \
diff --git a/create-config.mk b/create-config.mk
index d125597..4d318b9 100644
--- a/create-config.mk
+++ b/create-config.mk
@@ -73,6 +73,10 @@ TEST_URCU_TYPE_LIMITS = $(shell \
$(CC) -c -Werror=type-limits -o /dev/null -xc - 2>/dev/null \
|| echo -Wno-type-limits )
+URCU_VERSION = $(shell \
+ $(PKG_CONFIG) --modversion liburcu 2>/dev/null | \
+ awk -F. '{ printf("-DURCU_VERSION=0x%06x", 256 * ( 256 * $$1 + $$2) + $$3); }')
+
DEFINES :=
ifneq ($(call check_func,dm_task_no_flush,$(devmapper_incdir)/libdevmapper.h),0)
@@ -168,6 +172,7 @@ $(TOPDIR)/config.mk: $(multipathdir)/autoconfig.h
@echo creating $@
@echo "FPIN_SUPPORT := $(FPIN_SUPPORT)" >$@
@echo "FORTIFY_OPT := $(FORTIFY_OPT)" >>$@
+ @echo "D_URCU_VERSION := $(call URCU_VERSION)" >>$@
@echo "SYSTEMD := $(SYSTEMD)" >>$@
@echo "ANA_SUPPORT := $(ANA_SUPPORT)" >>$@
@echo "STACKPROT := $(call TEST_CC_OPTION,-fstack-protector-strong,-fstack-protector)" >>$@
diff --git a/libmultipath/alias.c b/libmultipath/alias.c
index e5d3f15..74431f3 100644
--- a/libmultipath/alias.c
+++ b/libmultipath/alias.c
@@ -24,6 +24,7 @@
#include "devmapper.h"
#include "strbuf.h"
#include "time-util.h"
+#include "lock.h"
/*
* significant parts of this file were taken from iscsi-bindings.c of the
@@ -300,7 +301,7 @@ void handle_bindings_file_inotify(const struct inotify_event *event)
pthread_mutex_unlock(×tamp_mutex);
if (changed) {
- uatomic_xchg(&bindings_file_changed, 1);
+ uatomic_xchg_int(&bindings_file_changed, 1);
condlog(3, "%s: bindings file must be re-read, new timestamp: %ld.%06ld",
__func__, (long)ts.tv_sec, (long)ts.tv_nsec / 1000);
} else
@@ -775,7 +776,7 @@ static int _read_bindings_file(const struct config *conf, Bindings *bindings,
int rc = 0, ret, fd;
FILE *file;
struct stat st;
- int has_changed = uatomic_xchg(&bindings_file_changed, 0);
+ int has_changed = uatomic_xchg_int(&bindings_file_changed, 0);
if (!force) {
if (!has_changed) {
diff --git a/libmultipath/lock.h b/libmultipath/lock.h
index 9814be7..ac80d1d 100644
--- a/libmultipath/lock.h
+++ b/libmultipath/lock.h
@@ -13,17 +13,22 @@ struct mutex_lock {
int waiters; /* uatomic access only */
};
-#if !defined(__GLIBC__) && defined(__GNUC__) && __GNUC__ == 12
-#pragma GCC diagnostic push
-#pragma GCC diagnostic ignored "-Warray-bounds"
-#endif
-
static inline void init_lock(struct mutex_lock *a)
{
pthread_mutex_init(&a->mutex, NULL);
uatomic_set(&a->waiters, 0);
}
+#if defined(__GNUC__) && __GNUC__ == 12 && URCU_VERSION < 0xe00
+#pragma GCC diagnostic push
+#pragma GCC diagnostic ignored "-Warray-bounds"
+#endif
+
+static inline int uatomic_xchg_int(int *ptr, int val)
+{
+ return uatomic_xchg(ptr, val);
+}
+
static inline void lock(struct mutex_lock *a)
{
uatomic_inc(&a->waiters);
@@ -31,6 +36,10 @@ static inline void lock(struct mutex_lock *a)
uatomic_dec(&a->waiters);
}
+#if defined(__GNUC__) && __GNUC__ == 12 && URCU_VERSION < 0xe00
+#pragma GCC diagnostic pop
+#endif
+
static inline int trylock(struct mutex_lock *a)
{
return pthread_mutex_trylock(&a->mutex);
@@ -51,10 +60,6 @@ static inline bool lock_has_waiters(struct mutex_lock *a)
return (uatomic_read(&a->waiters) > 0);
}
-#if !defined(__GLIBC__) && defined(__GNUC__) && __GNUC__ == 12
-#pragma GCC diagnostic pop
-#endif
-
#define lock_cleanup_pop(a) pthread_cleanup_pop(1)
void cleanup_lock (void * data);
diff --git a/multipathd/Makefile b/multipathd/Makefile
index cdba3db..0ba6ecb 100644
--- a/multipathd/Makefile
+++ b/multipathd/Makefile
@@ -5,8 +5,6 @@ CLI := multipathc
MANPAGES := multipathd.8
CPPFLAGS += -I$(multipathdir) -I$(mpathutildir) -I$(mpathpersistdir) -I$(mpathcmddir) -I$(thirdpartydir) \
- $(shell $(PKG_CONFIG) --modversion liburcu 2>/dev/null | \
- awk -F. '{ printf("-DURCU_VERSION=0x%06x", 256 * ( 256 * $$1 + $$2) + $$3); }') \
-DBINDIR='"$(bindir)"' $(SYSTEMD_CPPFLAGS)
#
--
2.42.0
--
dm-devel mailing list
dm-devel@redhat.com
https://listman.redhat.com/mailman/listinfo/dm-devel
next prev parent reply other threads:[~2023-09-14 14:53 UTC|newest]
Thread overview: 7+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-09-14 14:51 [dm-devel] [PATCH v3 00/38] multipath-tools: user-friendly names rework mwilck
2023-09-14 14:51 ` [dm-devel] [PATCH v3 26/38] multipath-tools tests: add test for ordering of bindings mwilck
2023-09-14 21:25 ` Benjamin Marzinski
2023-09-14 14:51 ` [dm-devel] [PATCH v3 27/38] multipathd: watch bindings file with inotify + timestamp mwilck
2023-09-14 22:28 ` Benjamin Marzinski
2023-09-14 14:51 ` mwilck [this message]
2023-09-14 22:31 ` [dm-devel] [PATCH v3 38/38] libmultipath: avoid -Warray-bounds error in uatomic operations Benjamin Marzinski
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20230914145131.15165-4-mwilck@suse.com \
--to=mwilck@suse.com \
--cc=bmarzins@redhat.com \
--cc=christophe.varoqui@opensvc.com \
--cc=dm-devel@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox