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 hurd/12187] [PATCH] sysdeps/mach/hurd: critical error values get overwritten


http://sourceware.org/bugzilla/show_bug.cgi?id=12187

Thomas Schwinge <tschwinge at sourceware dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |WAITING
                 CC|                            |tschwinge at sourceware dot
                   |                            |org
         AssignedTo|roland at gnu dot org       |tschwinge at sourceware dot
                   |                            |org

--- Comment #1 from Thomas Schwinge <tschwinge at sourceware dot org> 2012-02-18 09:21:08 UTC ---
Hi Nicolas!

Thanks for your report, and sorry for the delay in handling it.  My I ask
how/why you found this issue?

(The same pattern is also found in sendmsg.c.)

With the change you suggest, these calls would then only continue to work
for AF_LOCAL; don't you agree?  Here, that assignment to err is only to
initialize it to a non-zero value.  Then, in the HURD_DPORT_USE
construct's second argument, this is used to differentiate between the
AF_LOCAL case (where aport has already been bound), and all other AF_*
(where aport will now be bound by the socket_create_address call).  I
suspect that the value of EIEIO for initializing err has been chosen to
make it more obvious that it's a programming error, should this value
ever ``escape'' from the call.  Do you agree with this analyis?

I would happily accept a patch to make this construct easier to grasp.
>From a quick glance, we could probably do the following: initialize aport
to MACH_PORT_NULL and err to 0 at their definition sites, remote the err
assignment in else branch (not AF_LOCAL case), and instead use aport !=
MACH_PORT_NULL to conditionalize the socket_create_address call.  Would
that make it more obvious what is going on there?

-- 
Configure bugmail: http://sourceware.org/bugzilla/userprefs.cgi?tab=email
------- 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]