[Bug malloc/22611] malloc/tst-realloc wrongly assumes that errno must not be modified in case of success

["aurelien at aurel32 dot net" <sourceware-bugzilla at sourceware dot org>]

https://sourceware.org/bugzilla/show_bug.cgi?id=22611

--- Comment #9 from Aurelien Jarno <aurelien at aurel32 dot net> ---
(In reply to Carlos O'Donell from comment #8)
> (In reply to Florian Weimer from comment #7)
> > (In reply to Aurelien Jarno from comment #6)
> > > (In reply to Florian Weimer from comment #3)
> > > > (In reply to Aurelien Jarno from comment #0)
> > > > > tst-realloc tests, among others things that malloc and friends to not set
> > > > > errno to ENOMEM in case of successful memory allocation. This test always
> > > > > fail on sparc64 with "Error: errno is set but should not be".
> > > > 
> > > > Isn't this test invalid?
> > 
> > > Do you mean this is correct to set ENOMEM even when malloc is successful?
> > 
> > Yes, I think as long as we do not set errno to zero on success, we can
> > clobber errno.  POSIX explicitly says that applications should check errno
> > only after failure (or if the function specification provides additional
> > scenarios where it has a defined value).
> 
> You are right. I forgot about this wile reviewing.
> 
> Only for functions whose return value cannot be disambiguated from an error
> should we set errno to 0, and then check it after the call e.g. atoi.
> 
> For reference:
> https://wiki.sei.cmu.edu/confluence/pages/viewpage.action?pageId=87152351

Thanks for the pointer. This clearly show that tst-realloc must be fixed. I
have therefore changed the title of the bug, and I'll send a patch soon.

-- 
You are receiving this mail because:
You are on the CC list for the bug.