public inbox for gdb-patches@sourceware.org
 help / color / mirror / Atom feed
From: Tsukasa OI <research_trasio@irq.a4lg.com>
To: Enze Li <enze.li@hotmail.com>
Cc: gdb-patches@sourceware.org
Subject: Re: [PATCH v2 3/4] gdb/unittests: PR28413, suppress warnings generated by Gnulib
Date: Wed, 19 Oct 2022 01:14:55 +0900	[thread overview]
Message-ID: <3415b2ae-bcf4-b995-15a3-e8f173461c5a@irq.a4lg.com> (raw)
In-Reply-To: <OS3P286MB21524D2170537CBC3F7BF86FF0289@OS3P286MB2152.JPNP286.PROD.OUTLOOK.COM>

On 2022/10/18 22:44, Enze Li wrote:
> Hi Tsukasa,
> 
> Thanks for doing this.  I tested this patch on macOS and found one nit.
> Please see below.
> 
> On Thu, Sep 22 2022 at 08:25:46 AM +0000, Tsukasa OI wrote:
> 
>> Gnulib generates a warning if the system version of certain functions
>> are used (to redirect the developer to use Gnulib version).  It caused a
>> compiler error when...
>>
>> -   Compiled with Clang
>> -   -Werror is specified (by default)
>> -   C++ standard used by Clang is before C++17 (by default as of 15.0.0)
>>     when this unit test is activated.
>>
>> This issue is raised as PR28413.
>>
>> However, previous proposal to fix this issue (a "fix" to Gnulib):
>> <https://lists.gnu.org/archive/html/bug-gnulib/2021-10/msg00003.html>
>> was rejected because it ruins the intent of Gnulib warnings.
>>
>> So, we need a Binutils/GDB-side solution.
>>
>> This commit tries to deal with this issue on the GDB side.  We have
>> "include/diagnostics.h" to disable certain warnings only when necessary.
>>
>> This commit suppresses the Gnulib warnings by adding
>> DIAGNOSTIC_IGNORE_USER_DEFINED_WARNINGS before including "defs.h".
>>
>> gdb/ChangeLog:
>>
>> 	pr 28413
>> 	* unittests/string_view-selftests.c: Suppress Gnulib-generated
>> 	warnings when "defs.h" is included.
>> ---
>>  gdb/unittests/string_view-selftests.c | 5 +++++
>>  1 file changed, 5 insertions(+)
>>
>> diff --git a/gdb/unittests/string_view-selftests.c b/gdb/unittests/string_view-selftests.c
>> index 2d7261d18d3..c1f7799d94c 100644
>> --- a/gdb/unittests/string_view-selftests.c
>> +++ b/gdb/unittests/string_view-selftests.c
>> @@ -23,7 +23,12 @@
>>  
>>  #define GNULIB_NAMESPACE gnulib
>>  
>> +#include "diagnostics.h"
>> +
>> +DIAGNOSTIC_PUSH
>> +DIAGNOSTIC_IGNORE_USER_DEFINED_WARNINGS
>>  #include "defs.h"
>> +DIAGNOSTIC_POP
>>  #include "gdbsupport/selftest.h"
>>  #include "gdbsupport/gdb_string_view.h"
> 
> With your patch applied on macOS 15.10.7, it still says,
> 
> ==========================================================================================
>   CXX    unittests/string_view-selftests.o
> In file included from unittests/string_view-selftests.c:39:
> /Library/Developer/CommandLineTools/usr/bin/../include/c++/v1/fstream:579:17: error: The symbol
>       ::fdopen refers to the system function. Use gnulib::fdopen instead.
>       [-Werror,-Wuser-defined-warnings]
>       __file_ = fdopen(__fd, __mdstr);
>                 ^
> ./../gnulib/import/stdio.h:826:1: note: from 'diagnose_if' attribute on 'fdopen':
> _GL_CXXALIASWARN (fdopen);
> ^~~~~~~~~~~~~~~~~~~~~~~~~
> ./../gnulib/import/stdio.h:461:4: note: expanded from macro '_GL_CXXALIASWARN'
>    _GL_CXXALIASWARN_1 (func, GNULIB_NAMESPACE)
>    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ./../gnulib/import/stdio.h:463:4: note: expanded from macro '_GL_CXXALIASWARN_1'
>    _GL_CXXALIASWARN_2 (func, namespace)
>    ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> ./../gnulib/import/stdio.h:468:5: note: expanded from macro '_GL_CXXALIASWARN_2'
>     _GL_WARN_ON_USE (func, \
>     ^~~~~~~~~~~~~~~~~~~~~~~~
> ./../gnulib/import/stdio.h:632:19: note: expanded from macro '_GL_WARN_ON_USE'
>   __attribute__ ((__diagnose_if__ (1, message, "warning")))
>                   ^                ~
> 1 error generated.
> make[2]: *** [unittests/string_view-selftests.o] Error 1
> ==========================================================================================
> 
> I think this can be solved by just adding
> DIAGNOSTIC_IGNORE_USER_DEFINED_WARNINGS before including <fstream>
> according to your method.
> 
> WDYT?

Thanks for testing this on real Mac (since only Mac machine I own is
very old so that support is dropped).  Thinking of possible differences
in the standard C++ library (as you tested), it'd be better to surround
entire #include block with DIAGNOSTIC_IGNORE_USER_DEFINED_WARNINGS
(rather than just "defs.h" and <fstream>).

I will submit PATCH v3 consisting only with this patch (because 2 of 4
patches [1-2/4] are merged already and Simon Marchi made a change better
than my PATCH v2 4/4).

Regards,
Tsukasa

> ==========================================================================================
> diff --git a/gdb/unittests/string_view-selftests.c b/gdb/unittests/string_view-selftests.c
> index 2d7261d18d3..b779b9b765a 100644
> --- a/gdb/unittests/string_view-selftests.c
> +++ b/gdb/unittests/string_view-selftests.c
> @@ -23,7 +23,12 @@
> 
>  #define GNULIB_NAMESPACE gnulib
> 
> +#include "diagnostics.h"
> +
> +DIAGNOSTIC_PUSH
> +DIAGNOSTIC_IGNORE_USER_DEFINED_WARNINGS
>  #include "defs.h"
> +DIAGNOSTIC_POP
>  #include "gdbsupport/selftest.h"
>  #include "gdbsupport/gdb_string_view.h"
> 
> @@ -31,7 +36,10 @@
>     included test files are wrapped in a namespace.  */
>  #include <string>
>  #include <sstream>
> +DIAGNOSTIC_PUSH
> +DIAGNOSTIC_IGNORE_USER_DEFINED_WARNINGS
>  #include <fstream>
> +DIAGNOSTIC_POP
>  #include <iostream>
> 
>  /* libstdc++'s testsuite uses VERIFY.  */
> ==========================================================================================
> 
> Other than that, looks good to me.
> 
> Reviewed-by: Enze Li <enze.li@hotmail.com>
> 
> Thanks,
> Enze
> 

  reply	other threads:[~2022-10-18 16:15 UTC|newest]

Thread overview: 21+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-09-15  3:10 [PATCH 0/4] gdb: (includes PR28413), Suppress some general warnings if built with Clang Tsukasa OI
2022-09-15  3:10 ` [PATCH 1/4] include: Add macro to ignore -Wuser-defined-warnings Tsukasa OI
2022-09-20 16:36   ` Nick Clifton
2022-09-22 13:02     ` Enze Li
2022-09-15  3:10 ` [PATCH 2/4] include: Add macro to ignore -Wunused-but-set-variable Tsukasa OI
2022-09-15  3:10 ` [PATCH 3/4] gdb/unittests: PR28413, suppress warnings generated by Gnulib Tsukasa OI
2022-09-15  3:10 ` [PATCH 4/4] gdb: Suppress "unused" variable warning on Clang Tsukasa OI
2022-09-22  8:25 ` [PATCH v2 0/4] gdb: (includes PR28413), Suppress some general warnings if built with Clang Tsukasa OI
2022-09-22  8:25   ` [PATCH v2 1/4] include: Add macro to ignore -Wuser-defined-warnings Tsukasa OI
2022-09-22 11:26     ` Nick Clifton
2022-09-22  8:25   ` [PATCH v2 2/4] include: Add macro to ignore -Wunused-but-set-variable Tsukasa OI
2022-09-22 11:27     ` Nick Clifton
2022-09-22  8:25   ` [PATCH v2 3/4] gdb/unittests: PR28413, suppress warnings generated by Gnulib Tsukasa OI
2022-10-18 13:44     ` Enze Li
2022-10-18 16:14       ` Tsukasa OI [this message]
2022-09-22  8:25   ` [PATCH v2 4/4] gdb: Suppress "unused" variable warning on Clang Tsukasa OI
2022-10-12 17:36     ` Simon Marchi
2022-10-16 13:37       ` Tsukasa OI
2022-10-17 12:35         ` Simon Marchi
2022-10-18 17:11   ` [PATCH v3] gdb/unittests: PR28413, suppress warnings generated by Gnulib Tsukasa OI
2022-11-14 14:02     ` Simon Marchi

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=3415b2ae-bcf4-b995-15a3-e8f173461c5a@irq.a4lg.com \
    --to=research_trasio@irq.a4lg.com \
    --cc=enze.li@hotmail.com \
    --cc=gdb-patches@sourceware.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).