RE: [PATCH v1 2/4] builtin/receive-pack.c: change xwrite to write_in_full to allow large sizes.

From: rsbecker@nexbridge.com <rsbecker@nexbridge.com>
Date: Feb 27, 2024, 00:15 UTC
Message-ID: <027101da6912$23cf31c0$6b6d9540$@nexbridge.com>
In-Reply-To: <xmqq34tekcoo.fsf@gitster.g>

On Monday, February 26, 2024 6:51 PM, Junio C Hamano wrote:

Show 7 quoted lines

>"Randall S. Becker" <the.n.e.key@gmail.com> writes:
>
>> From: "Randall S. Becker" <rsbecker@nexbridge.com>
>>
>> This change is required because some platforms do not support file
>> writes of arbitrary sizes (e.g, NonStop). xwrite ends up truncating
>> the output to the maximum single I/O size possible for the destination

device.

>
>As msg[] here is 4k on-stack buffer, if the I/O size is small enough, the

above may happen, and I think write-in-full is warranted here.  If

>your I/O must be done in 1k chunks, it would be very slow to run things

like writing a pack stream to clone any non-toy projects,

>though X-<.

On the x86 platform, we get a size large enough not to trigger the failure in t7704. However, on ia64, the limit is 56Kb, which apparently does. I'm hoping no one else has a 1Kb limit - although some TCP stacks might experience it. Either way, truncating a package is bad. Fortunately the I/O subsystem on NonStop is very fast (basically DMA) between process memory space.

Show 23 quoted lines

>
>> Signed-off-by: Randall S. Becker <rsbecker@nexbridge.com>
>> ---
>>  builtin/receive-pack.c | 5 +++--
>>  1 file changed, 3 insertions(+), 2 deletions(-)
>>
>> diff --git a/builtin/receive-pack.c b/builtin/receive-pack.c index
>> db65607485..5064f3d300 100644
>> --- a/builtin/receive-pack.c
>> +++ b/builtin/receive-pack.c
>> @@ -455,8 +455,9 @@ static void report_message(const char *prefix,
>> const char *err, va_list params)
>>
>>  	if (use_sideband)
>>  		send_sideband(1, 2, msg, sz, use_sideband);
>> -	else
>> -		xwrite(2, msg, sz);
>> +	else {
>> +		write_in_full(2, msg, sz);
>> +	}
>>  }
>>
>>  __attribute__((format (printf, 1, 2)))