Re: [PATCH 2/4] set_restore_sigmask TIF_SIGPENDING

[David Woodhouse <dwmw2-wEGCiKHe2LqWVfeAwA7xHQ@public.gmane.org>]

On Wed, 2008-04-09 at 15:39 +0400, Oleg Nesterov wrote:
> As I see it, the main disadvantage of ERESTART_ approach is that we need 2
> new ERESTART_ codes, one for ERESTARTNOHAND, another for ERESTART_RESTARTBLOCK.
> And yes, while I personally think this is "more clean", it is very subjective.

Subjective, yeah.... personally, I don't like using ERESTART_xxx much,
because you're _not_ necessarily restarting the system call. The
separate flag for TIF_RESTORE_SIGMASK (or TLF_RESTORE_SIGMASK) seems
cleaner to me -- especially once you observe that you need new codes for
ERESTART_xxx_AND_RESTORE_SIGMASK for each ERESTART_xxx that you might
want to use in conjunction with the flags.

But I don't really care much either, if you want to change it and get
the details right.

One of the supposed advantages of TIF_RESTORE_SIGMASK in the first
place, iirc, was that it allowed us to return a result code other than
-EINTR as _well_ as restoring the signal mask. But we don't actually
make use of that possibility now anyway.

-- 
dwmw2