Re: PROBLEM: pthread-safety bug in write(2) on Linux 2.6.x

[Linus Torvalds]

On Thu, 13 Apr 2006, Linus Torvalds wrote:

> 
> 
> On Fri, 14 Apr 2006, Alan Cox wrote:
> > 
> > Quality for whom ? There is a measurable cost to all that extra locking
> > which will hurt everyone. Given existing kernels don't make the
> > guarantee and SuS v3 does not make the guarantee the apps that need it
> > will continue to do the extra work themselves anyway.
> 
> True. 

Side note: if you want the strange POSIX guarantees that Dan quoted 
(atomicity between write and fstat), you'd almost have to do it with 
user-space crapola magic locks. Doing it in the kernel would just be 
insane, you'd have to have some per-process "IO semaphore".

(Doing it in user space _also_ sounds insane, but then you could have a 
switch like "POSIX me harder, and do all the really stupid things" at 
compile time or something)

			Linus
-
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/