Re: [PATCH] gdb: fix IA64 build failure of linux-nat

[Tom de Vries <tdevries@suse.de>]

On 8/16/20 11:51 PM, Simon Marchi wrote:
> On 2020-08-16 4:45 a.m., Sergei Trofimovich via Gdb-patches wrote:
>> On Tue, 19 May 2020 15:00:41 -0700
>> Kevin Buettner <kevinb@redhat.com> wrote:
>>
>>> On Tue, 19 May 2020 22:27:10 +0100
>>> Sergei Trofimovich via Gdb-patches <gdb-patches@sourceware.org> wrote:
>>>
>>>> From: Sergei Trofimovich <siarheit@google.com>
>>>>
>>>> On IA64 built failed as:
>>>>
>>>> ```
>>>> ia64-linux-nat.c:352:29: error: 'gdbarch_num_regs' was not declared in this scope
>>>>   352 |   if (regno < 0 || regno >= gdbarch_num_regs (gdbarch))
>>>>       |                             ^~~~~~~~~~~~~~~~
>>>> ```
>>>>
>>>> The fix includes "gdbarch.h" header where symbol is declared.
>>>>
>>>> 	* gdb/ia64-linux-nat.c: include "gdbarch.h" to declare used
>>>> 	'gdbarch_num_regs'.  
>>>
>>> Okay, but please capitalize "include" in the ChangeLog entry prior
>>> to pushing this change.
>>
>> Attached v2-* patch with capitalization changes.
>>
>> I don't have a 'gdb' write access yet (I think), but I do have GCC one.
>>
>> Should I request 'gdb' access as well as specified in
>>     https://sourceware.org/cgi-bin/pdw/ps_form.cgi ?
> 
> Yes, that would be useful if you plan on contributing regularly.  If it's just
> a occasional patch, we can also push for you.  As you wish.
> 
>>
>> Thank you for the review! 
> 
> A few more styling nits:
> 
>> diff --git a/gdb/ChangeLog b/gdb/ChangeLog
>> index 9cc7e44cba7..8865e6949d6 100644
>> --- a/gdb/ChangeLog
>> +++ b/gdb/ChangeLog
>> @@ -4564,6 +4564,11 @@
>>
>>  	* sparc64-tdep.c (adi_tag_fd): Update call to target_fileio_open.
>>
>> +2020-05-19  Sergei Trofimovich  <siarheit@google.com>
>> +
>> +	* gdb/ia64-linux-nat.c: Include "gdbarch.h" to declare used
>> +	'gdbarch_num_regs'.
> 
> The file path should be relative to the ChangeLog location, so here just "ia64-linux-nat.c".
> 
> Also, make sure that your new entry is at the top of the file (here, it's at line 4564), and
> that you update the date to $TODAY when you push.  For these reasons, most people don't include
> the ChangeLog bits in the patch directly, but just include it in the commit log (as you did).
> Of course, when pushing the patch, then you need to insert it in the ChangeLog file.
> 

Hi,

IA64 support was just obsoleted in bfd (commit 73d0dc162e "Obsolete ia64").

So, AFAIU, this should now be built with --enable-obsolete.

What are the consequences for gdb IA64 patches ?

Thanks,
- Tom