[PATCH] Prepare pflocal/io.c for a revised hurdselect.c: Pipe case (for review)

Hello, Attached is a patch for review adding return values of pflocal/io.c for the case of pipes to detect broken ones. With an upgraded hurdselect.c:poll() (in the works) the test from e.g. gnulib passes. Remaning work is to solve the corresponding problem for sockets in pfinet (and eventually al

io_select EPIPE error case? [Was: [PATCH] Prepare pflocal/io.c for a revised hurdselect.c: Pipe case (for review)]

Svante Signell, le Thu 22 Nov 2012 20:29:32 +0100, a écrit : > @@ -231,15 +232,21 @@ S_io_select (struct sock_user *user, >if (valid & SELECT_READ) > { > pipe_acquire_reader (read_pipe); > - if (pipe_wait_readable (read_pipe, 1, 1) != EWOULDBLOCK) > + err = pipe_wa

Re: io_select EPIPE error case? [Was: [PATCH] Prepare pflocal/io.c for a revised hurdselect.c: Pipe case (for review)]

On Fri, 2012-11-23 at 01:55 +0100, Samuel Thibault wrote: > Svante Signell, le Thu 22 Nov 2012 20:29:32 +0100, a écrit : > > @@ -231,15 +232,21 @@ S_io_select (struct sock_user *user, > >if (valid & SELECT_READ) > > { > > pipe_acquire_reader (read_pipe); > > - if (pipe_wait_re