RE: [BUG] 2.44.0 t7704.9 Fails on NonStop ia64

From: rsbecker@nexbridge.com <rsbecker@nexbridge.com>
Date: Feb 25, 2024, 20:36 UTC
Message-ID: <01ca01da682a$5f6a7b60$1e3f7220$@nexbridge.com>
In-Reply-To: <20240225191954.GA28646@tb-raspi4>

On Sunday, February 25, 2024 2:20 PM, Torsten Bögershausen wrote:

Show 12 quoted lines

>On Sun, Feb 25, 2024 at 02:08:35PM -0500, rsbecker@nexbridge.com wrote:
>> On Sunday, February 25, 2024 1:45 PM, I wrote:
>> >To: git@vger.kernel.org
>> >Subject: [BUG] 2.44.0 t7704.9 Fails on NonStop ia64
>> >
>> >This appears to be a new issue introduced at 2.44.0. It only occurs
>> >on
>> NonStop ia64
>> >1..9
>[snip]
>>
>> I did find the following calls to write(), one of which might be

involved.

>> write() should not be used directly unless the count is clearly very

small.

Show 7 quoted lines

>> Xwrite() should be used instead. There are other calls but those are
>> either small or not on platform.
>
>(Probably a typ0: Xwrite() -> xwrite()
>
>But I think that this should be used:
>write_in_full()

My mailer autocorrected, yes, xwrite. write_in_full() would be safe, although a bit redundant since xwrite() does similar things and is used by write_in_full(). The question is which call is bad? The cruft stuff is relatively new and I don't know the code.

Show 5 quoted lines

>> reftable/writer.c:              int n = w->write(w->write_arg, zeroed,
>> w->pending_padding);
>> reftable/writer.c:      n = w->write(w->write_arg, data, len);
>> run-command.c:                  len = write(io->fd, io->u.out.buf,
>> t/helper/test-path-utils.c:                     if (write(1, buffer,

count)

>> < 0)
>> t/helper/test-windows-named-pipe.c:             write(1, buf, nbr);
>> t/helper/test-windows-named-pipe.c:             write(1, buf, nbr);
>> trace2/tr2_dst.c:       bytes = write(fd, buf_line->buf, buf_line->len);