Messages in this thread Patch in this message | | | Date | Thu, 23 May 2013 15:22:55 +0200 (CEST) | From | Jiri Kosina <> | Subject | [PATCH umpteenth RESEND] random: fix accounting race condition with lockless irq entropy_count update |
| |
Hi everybody,
this is an umpteenth resend of patch that has been queued in -mm tree for quite some time, but apart from that, hasn't been acted upon for some reason. A few more CCs added.
As the bug is causing observable failures of userspace services relying on /dev/urandom never EOFing (openssh and apache reported to fail) on certain architectures (s390 known to be hit by this), I think it shouldn't be unnecessarily delayed.
Thanks in advance.
From: Jiri Kosina <jkosina@suse.cz> Subject: random: fix accounting race condition with lockless irq entropy_count update
Commit 902c098a3663 ("random: use lockless techniques in the interrupt path") turned IRQ path from being spinlock protected into lockless cmpxchg-retry update.
That commit removed r->lock serialization between crediting entropy bits from IRQ context and accounting when extracting entropy on userspace read path, but didn't turn the r->entropy_count reads/updates in account() to use cmpxchg as well.
It has been observed, that under certain circumstances this leads to read() on /dev/urandom to return 0 (EOF), as r->entropy_count gets corrupted and becomes negative, which in turn results in propagating 0 all the way from account() to the actual read() call.
Convert the accounting code to be the proper lockless counterpart of what has been partially done by 902c098a3663.
Signed-off-by: Jiri Kosina <jkosina@suse.cz> Cc: Theodore Ts'o <tytso@mit.edu> Cc: Greg KH <greg@kroah.com> Cc: <stable@vger.kernel.org> Signed-off-by: Andrew Morton <akpm@linux-foundation.org> ---
drivers/char/random.c | 22 +++++++++++++++------- 1 file changed, 15 insertions(+), 7 deletions(-)
diff -puN drivers/char/random.c~random-fix-accounting-race-condition-with-lockless-irq-entropy_count-update drivers/char/random.c --- a/drivers/char/random.c~random-fix-accounting-race-condition-with-lockless-irq-entropy_count-update +++ a/drivers/char/random.c @@ -865,16 +865,24 @@ static size_t account(struct entropy_sto if (r->entropy_count / 8 < min + reserved) { nbytes = 0; } else { + int entropy_count, orig; +retry: + entropy_count = orig = ACCESS_ONCE(r->entropy_count); /* If limited, never pull more than available */ - if (r->limit && nbytes + reserved >= r->entropy_count / 8) - nbytes = r->entropy_count/8 - reserved; + if (r->limit && nbytes + reserved >= entropy_count / 8) + nbytes = entropy_count/8 - reserved; - if (r->entropy_count / 8 >= nbytes + reserved) - r->entropy_count -= nbytes*8; - else - r->entropy_count = reserved; + if (entropy_count / 8 >= nbytes + reserved) { + entropy_count -= nbytes*8; + if (cmpxchg(&r->entropy_count, orig, entropy_count) != orig) + goto retry; + } else { + entropy_count = reserved; + if (cmpxchg(&r->entropy_count, orig, entropy_count) != orig) + goto retry; + } - if (r->entropy_count < random_write_wakeup_thresh) + if (entropy_count < random_write_wakeup_thresh) wakeup_write = 1; } _
| |