From: Jeff King <peff@peff.net>
Date: Wed, 23 Sep 2015 01:51:27 GMT
Subject: Re: [PATCHv3 02/13] xread: poll on non blocking fds
Message-ID: <20150923015127.GC21294@sigill.intra.peff.net>
In-Reply-To: <CAGZ79kYKxNAXhbmXEtMeBYeGqu8i4bcUW_hpjLaPSF4p5O=uBg@mail.gmail.com>

On Tue, Sep 22, 2015 at 05:14:42PM -0700, Stefan Beller wrote:

> We should not care if the call to poll failed, as we're in an infinite loop and
> can only get out with the correct read(..). So maybe an implementation like this
> would already suffice:
> 
> ssize_t xread(int fd, void *buf, size_t len)
> {
>     ssize_t nr;
>     if (len > MAX_IO_SIZE)
>         len = MAX_IO_SIZE;
>     while (1) {
>         nr = read(fd, buf, len);
>         if (nr < 0) {
>             if (errno == EINTR)
>                 continue;
>             if (errno == EAGAIN || errno == EWOULDBLOCK) {
>                 struct pollfd pfd;
>                 pfd.events = POLLIN;
>                 pfd.fd = fd;
>                 /* We deliberately ignore the return value of poll. */
>                 poll(&pfd, 1, -1);
>                 continue;
>             }
>         }
>         return nr;
>     }
> }

FWIW, that is what I had imagined.

-Peff