Re: O_NONBLOCK under Windows (was: git.git as of tonight)

From: Stefan Beller <sbeller@google.com>
Date: Nov 4, 2015, 20:07 UTC
Message-ID: <CAGZ79kZk-oAcdVNHsFe_cRhR+TuGi8RqLYigHajC9KJu7ATJJw@mail.gmail.com>
In-Reply-To: <563A63BA.9020407@web.de>

On Wed, Nov 4, 2015 at 11:59 AM, Torsten Bögershausen <tboegi@web.de> wrote:

Show 21 quoted lines

> On 11/04/2015 12:00 AM, Stefan Beller wrote:
>> On Tue, Nov 3, 2015 at 1:03 PM, Johannes Sixt <j6t@kdbg.org> wrote:
>>
> The #ifdef assumes that Windows never will have O_NONBLOCK/F_GETFL)
>
> Does the following make more sense ?
> #if defined (O_NONBLOCK) && defined (F_GETFL)
>
> Or may be:
> #ifndef NO_O_NONBLOCK
>>> #ifndef GIT_WINDOWS_NATIVE
>>>         int flags = fcntl(fd, F_GETFL);
>>>         if (flags < 0)
>>>                 warning("Could not get file status flags, "
>>>                         "output will be degraded");
>>>         else if (fcntl(fd, F_SETFL, flags | O_NONBLOCK))
>>> #endif
>>>                 warning("Could not set file status flags, "
>>>                         "output will be degraded");
>>> }
>>>

Reading Junios answer to the resent patch[1], I am currently debating if this is the right way to go anyway. As Junio points out, this is not a warning but rather a critical issue such that we'd maybe rather die(...) than just warning(...), which would make the discussion about the correct condition in the #ifdef moot.

[1] [PATCHv3 02/11] run-command: report failure for degraded output just once