Re: [PATCH,eglibc] Re: Questions about patches for hurdselect.c

[Svante Signell]

On Tue, 2012-12-18 at 14:51 +0100, Svante Signell wrote:
> On Mon, 2012-12-17 at 23:02 +0100, Svante Signell wrote:
> > On Mon, 2012-12-17 at 20:58 +0100, Richard Braun wrote:
> > > On Mon, Dec 17, 2012 at 03:12:10AM +0100, Svante Signell wrote:
...
> As a follow-up the attached patch combines Richards zero timeout for
> select()-based calls to hurdselect to also work with poll(). One example
> is ntpdate. The timeout also seems to be OK, maybe further testing
> needed?. This is a workaround until the split into three cases are
> ready, and until the poll code is updated.

Now the 3-way split of hurdselect is updated, see the attached patch.
Hopefully the trivial faults are solved by now and we can proceed with
the poll-specific changes towards POSIX compliance. Attached is also
some test code to check the functionality of the patched version,.That
can be checked with the poll/select patches as proposed previously (and
with the buggy current version).

Of course the #if 1 case has to be resolved, My findings is that they
are needed (with the new code, not the old) until a safe timeout is
communicated for the file descriptors to be ready in the poll case.

Thanks and Merry Christmas,
Svante (note my change of ISP mail supplier) 

Comments and feedbacks are of course encouraged :-)

hurdselect.c.patch
Description: Text Data

test_poll+select.c
Description: Text Data