Re: [PATCH 1/5] [func_call] Add possible spelling of linker error message.

[Keith Seitz <keiths@redhat.com>]

On 3/17/22 10:58, Metzger, Markus T wrote:
> Hello Ari,
> 
>> Seen with RHEL 7.5 and Fedora 26.
>>

Is there a little more that can be said? I gather this patch is
desired because the error message is simply different on some systems?

>> gdb/testsuite/ChangeLog:
>> 2019-02-18  Tim Wiederhake  <tim.wiederhake@intel.com>
>>
>> 	* lib/gdb.exp (gdb_compile_pthreads): Add new error regexp.
>>

As mentioned, you can drop these ChangeLog entries.

>> Signed-off-by: Tim Wiederhake <tim.wiederhake@intel.com>
>> Signed-off-by: Ari Hannula <ari.hannula@intel.com>
>> ---
>> gdb/testsuite/lib/gdb.exp | 3 +++
>> 1 file changed, 3 insertions(+)
>>
>> diff --git a/gdb/testsuite/lib/gdb.exp b/gdb/testsuite/lib/gdb.exp
>> index a3717a4022..363289caa1 100644
>> --- a/gdb/testsuite/lib/gdb.exp
>> +++ b/gdb/testsuite/lib/gdb.exp
>> @@ -4618,6 +4618,9 @@ proc gdb_compile_pthreads {source dest type options}
>> {
>>                  set why_msg "missing threads include file"
>>                  break
>>              }
>> +            ".*cannot find -lpthread.*" {
>> +                set why_msg "missing runtime threads library"
>> +            }
>>              ".*cannot open -lpthread.*" {
>>                  set why_msg "missing runtime threads library"
>>              }

If these are truly the same error condition, then since this switch statement
uses regexp-matching, we should be able to merge the two "missing runtime
threads library" cases with "(find|open)".

If they are actually different errors, I'd like to see `why_msg' reflect the
distinction.

> LGTM but you need Keith or a global maintainer to approve this.

This otherwise looks okay to me, too, but I am unable to officially approve
patches.

Keith