From: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
To: linux-kernel@vger.kernel.org,
Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>,
Peter Zijlstra <peterz@infradead.org>,
Darren Hart <dvhart@infradead.org>
Cc: Maxim Samoylov <max7255@yandex-team.ru>,
Linus Torvalds <torvalds@linux-foundation.org>,
linux-api@vger.kernel.org
Subject: [PATCH] futex: send SIGBUS if argument is not aligned on a four-byte boundary
Date: Fri, 15 May 2020 18:36:47 +0300 [thread overview]
Message-ID: <158955700764.647498.18025770126733698386.stgit@buzz> (raw)
Userspace implementations of mutexes (including glibc) in some cases
retries operation without checking error code from syscall futex.
This is good for performance because most errors are impossible when
locking code trusts itself.
Some errors which could came from outer code are handled automatically,
for example invalid address triggers SIGSEGV on atomic fast path.
But one case turns into nasty busy-loop: when address is unaligned.
futex(FUTEX_WAIT) returns EINVAL immediately and loop goes to retry.
Example which loops inside second call rather than hung peacefully:
#include <stdlib.h>
#include <pthread.h>
int main(int argc, char **argv)
{
char buf[sizeof(pthread_mutex_t) + 1];
pthread_mutex_t *mutex = (pthread_mutex_t *)(buf + 1);
pthread_mutex_init(mutex, NULL);
pthread_mutex_lock(mutex);
pthread_mutex_lock(mutex);
}
It seems there is no practical usage for calling syscall futex for
unaligned address. This may be only bug in user space. Let's help
and handle this gracefully without adding extra code on fast path.
This patch sends SIGBUS signal to slay task and break busy-loop.
Signed-off-by: Konstantin Khlebnikov <khlebnikov@yandex-team.ru>
Reported-by: Maxim Samoylov <max7255@yandex-team.ru>
---
kernel/futex.c | 13 ++++++++++++-
1 file changed, 12 insertions(+), 1 deletion(-)
diff --git a/kernel/futex.c b/kernel/futex.c
index b59532862bc0..8a6d35fa56bc 100644
--- a/kernel/futex.c
+++ b/kernel/futex.c
@@ -508,10 +508,21 @@ get_futex_key(u32 __user *uaddr, int fshared, union futex_key *key, enum futex_a
/*
* The futex address must be "naturally" aligned.
+ * Also send signal to break busy-loop if user-space ignore error.
+ * EFAULT case should trigger SIGSEGV at access from user-space.
*/
key->both.offset = address % PAGE_SIZE;
- if (unlikely((address % sizeof(u32)) != 0))
+ if (unlikely((address % sizeof(u32)) != 0)) {
+ struct kernel_siginfo info;
+
+ clear_siginfo(&info);
+ info.si_signo = SIGBUS;
+ info.si_code = BUS_ADRALN;
+ info.si_addr = uaddr;
+ force_sig_info(&info);
+
return -EINVAL;
+ }
address -= key->both.offset;
if (unlikely(!access_ok(uaddr, sizeof(u32))))
next reply other threads:[~2020-05-15 15:37 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-05-15 15:36 Konstantin Khlebnikov [this message]
2020-05-15 16:27 ` [PATCH] futex: send SIGBUS if argument is not aligned on a four-byte boundary Peter Zijlstra
2020-05-15 17:50 ` Carlos O'Donell
2020-05-19 17:22 ` Carlos O'Donell
2020-05-19 22:53 ` Thomas Gleixner
2020-05-20 8:22 ` Konstantin Khlebnikov
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=158955700764.647498.18025770126733698386.stgit@buzz \
--to=khlebnikov@yandex-team.ru \
--cc=dvhart@infradead.org \
--cc=linux-api@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=max7255@yandex-team.ru \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=torvalds@linux-foundation.org \
/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