This is the mail archive of the glibc-bugs@sourceware.org mailing list for the glibc project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug libc/20111] struct sockaddr_storage cannot be aggregate-copied


https://sourceware.org/bugzilla/show_bug.cgi?id=20111

--- Comment #14 from rguenther at suse dot de ---
On Tue, 24 May 2016, fweimer at redhat dot com wrote:

> https://sourceware.org/bugzilla/show_bug.cgi?id=20111
> 
> --- Comment #12 from Florian Weimer <fweimer at redhat dot com> ---
> (In reply to rguenther from comment #11)
> > I might want to call it a security issue (the miscompile ends up
> > using uninitialized data), so maybe backport it to active release
> > branches.
> 
> Changing struct definitions is always a bit iffy.  I plan to backport this
> change to 2.23 though.
> 
> For our downstream backports, we will perhaps follow a more conservative
> approach which is easy for us because we do not have to support m68k.
> 
> Considering that it's established practice to compile with -fstrict-aliasing
> (the default), for most users, this âfixâ is very brittle anyway.

Agreed.  But as you said the "real" fix isn't possible in glibc
(using may_alias attribute) but would need GCC assistance (automagically
adding the attribute to structs with that magic name).  Let me open
an enhancement bug for that.

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=71255

Maybe you can try listing the struct identifiers that would need such
handling there.

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

Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]