Re: [PATCH] Fix gdb crash due to SIGPIPE when the compile command fails

[Bernd Edlinger <bernd.edlinger@hotmail.de>]

On 6/5/21 2:04 PM, Andrew Burgess wrote:
> * Bernd Edlinger <bernd.edlinger@hotmail.de> [2021-06-05 13:44:03 +0200]:
> 
>> On 6/4/21 3:39 PM, Tom Tromey wrote:
>>>>>>>> "Bernd" == Bernd Edlinger <bernd.edlinger@hotmail.de> writes:
>>> Bernd> I believe it is right to avoid the SIGPIPE before calling the plugin,
>>> Bernd> instead of doing that in gcc-trunk, since we don't know which version
>>> Bernd> we will be calling, and all versions I tried have failed like this.
>>>
>>> That seems fine, but I think it would be better to install the handler
>>> just when working with the plugin, and then uninstall it afterward, sort
>>> of like what class scoped_ignore_sigttou does.
>>>
>> Okay, done, that works for me.
>>
>> Is this OK?
>>
>>
>> Thanks
>> Bernd.
>>
>> From 3bb3966601a89c18ea2700c3eae4aa4f4e195269 Mon Sep 17 00:00:00 2001
>> From: Bernd Edlinger <bernd.edlinger@hotmail.de>
>> Date: Wed, 2 Jun 2021 19:21:15 +0200
>> Subject: [PATCH] Fix gdb crash due to SIGPIPE when the compile command fails
>>
>> Due to the SIGPIPE the gdb process is killed here, which is
>> not helpful.
>>
>> 2021-06-02  Bernd Edlinger  <bernd.edlinger@hotmail.de>
>>
>> 	* compile/compile.c (compile_to_object): Ignore SIGPIPE before calling
>> 	the plugin.
>> ---
>>  gdb/compile/compile.c | 31 +++++++++++++++++++++++++++++++
>>  1 file changed, 31 insertions(+)
>>
>> diff --git a/gdb/compile/compile.c b/gdb/compile/compile.c
>> index 8481d14..3431d4c 100644
>> --- a/gdb/compile/compile.c
>> +++ b/gdb/compile/compile.c
>> @@ -633,6 +633,33 @@ struct compile_options
>>    fputs_filtered (message, gdb_stderr);
>>  }
>>  
>> +/* RAII class used to ignore SIGPIPE in a scope.  */
>> +
>> +class scoped_ignore_sigpipe
>> +{
>> +public:
>> +  scoped_ignore_sigpipe ()
>> +  {
>> +#ifdef SIGPIPE
>> +    m_osigpipe = signal (SIGPIPE, SIG_IGN);
>> +#endif
>> +  }
>> +
>> +  ~scoped_ignore_sigpipe ()
>> +  {
>> +#ifdef SIGTTOU
> SIGPIPE?
> 

Ah, yes of course!

Consider this fixed.


Thanks,
Bernd.