Re: [RFC PATCH] rtmutex: Do not prio boost when timeout is used

[Steven Rostedt]

Glad I put on my shark repellent.

On Sun, 27 Apr 2014 20:20:26 +0200 (CEST)
Thomas Gleixner <tglx@xxxxxxxxxxxxx> wrote:
> 
> We really have better things to do, than changing the rules just to
> please some random (probably commercial) application which has been
> programmed along the coding rules from hell.
> 

Peter had already pointed out that the current behavior is actually
specified by the standards. This patch is thus NULL and void. I only
posted this patch because you haven't bitched at me in a while and I
figured I was due.

I just noticed this when we were trying to debug why an application is
reporting EDEADLOCK on a lock. This was a side discussion as there were
no timeout locks in the application.

It looks more like a bug in the glibc implementation, as they have
both RECURSIVE and PRIO_INHERIT set on the mutex, and it seems that it
is deadlocking on a lock it already owns when it tries to retake it.

-- Steve
--
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/