Re: frequent lockups in 3.18rc4

From: Jens Axboe
Date: Mon Nov 17 2014 - 21:23:46 EST


On 11/17/2014 04:17 PM, Thomas Gleixner wrote:
On Mon, 17 Nov 2014, Jens Axboe wrote:
On 11/17/2014 02:22 PM, Linus Torvalds wrote:
On Fri, Nov 14, 2014 at 5:59 PM, Linus Torvalds
<torvalds@xxxxxxxxxxxxxxxxxxxx> wrote:

Judging by the code disassembly, it's the "csd_lock_wait(csd)" at the
end.

Btw, looking at this, I grew really suspicious of this code in csd_unlock():

WARN_ON((csd->flags & CSD_FLAG_WAIT) && !(csd->flags & CSD_FLAG_LOCK));

because that makes no sense at all. It basically removes a sanity
check, yet that sanity check makes a hell of a lot of sense. Unlocking
a CSD that is not locked is *wrong*.

The crazy code code comes from commit c84a83e2aaab ("smp: don't warn
about csd->flags having CSD_FLAG_LOCK cleared for !wait") by Jens, but
the explanation and the code is pure crap.

There is no way in hell that it is ever correct to unlock an entry
that isn't locked, so that whole CSD_FLAG_WAIT thing is buggy as hell.

The explanation in commit c84a83e2aaab says that "blk-mq reuses the
request potentially immediately" and claims that that is somehow ok,
but that's utter BS. Even if you don't ever wait for it, the CSD lock
bit fundamentally also protects the "csd->llist" pointer. So what that
commit actually does is to just remove a safety check, and do so in a
very unsafe manner. And apparently block-mq re-uses something THAT IS
STILL ACTIVELY IN USE. That's just horrible.

I agree that this description is probably utter crap. And now I do
actually remember the issue at hand. The resource here is the tag, that
decides what request we'll use, and subsequently what call_single_data
storage is used. When this was originally done, blk-mq cleared the
request from the function callback, instead of doing it at allocation
time. The assumption here was cache hotness. That in turn also cleared
->csd, which meant that the flags got zeroed and csd_unlock() was
naturally unhappy.

So that's exactly what I described in my other reply.

csd_lock(csd);
queue_csd();
ipi();
csd->func(csd->info);
wait_for_completion(csd);
complete(csd);
reuse_csd(csd);
csd_unlock(csd);

When you call complete() nothing can rely on csd anymore, except for
the smp core code ....

Right, and I didn't. It was the core use of csd->flags afterwards that complained. blk-mq merely cleared ->flags in csd->func(), which (granted) was a bit weird. So it was just storing to csd (before unlock), but in an inappropriate way. It would obviously have broken a sync invocation, but the block layer never does that.

THAT was the reuse case, not that the request would get reused
before we had finished the IPI fn callback since that would
obviously create other badness. Now I'm not sure what made me create
that patch, which in retrospect is a bad hammer for this problem.

Performance blindness?

Possibly...

blk-mq doesn't do the init-at-finish time anymore, so it should not be
hit by the issue. But if we do bring that back, then it would still work
fine with Thomas' patch, since we unlock prior to running the callback.

So if blk-mq is not relying on that, then we really should back out
that stuff for 3.18 and tag it for stable.

Yeah, I'd be fine with doing that. I don't recall at the top of my head when when we stopped doing the clear at free time, but it was relatively early. OK, so checked, and 3.15 does init at free time, and 3.16 and later does it at allocation time. So the revert can only safely be applied to 3.16 and later...

Treating sync and async function calls differently makes sense,
because any async caller which cannot deal with the unlock before call
scheme is broken by definition already today. But that's material for
next.

Agree.

--
Jens Axboe

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/