[PATCH] Fix procfs.c compilation on Solaris

[PATCH] Fix procfs.c compilation on Solaris

[Rainer Orth]

[-- Attachment #1: Type: text/plain, Size: 2707 bytes --]

procfs.c doesn't compile on trunk and the gdb-18 branch:

procfs.c: In function ‘int procfs_notice_thread(procinfo*, procinfo*, void*)’:
procfs.c:2857:33: error: invalid use of non-static member function ‘thread_state thread_info::state() const’
 2857 |   if (thr == NULL || thr->state == THREAD_EXITED)
      |                      ~~~~~~~~~~~^~~~~~~~~~~~~~~~

Fixed by calling the member function.

procfs.c: In function ‘bool find_memory_regions_callback(prmap*, find_memory_region_ftype)’:
procfs.c:3152:15: error: no match for call to ‘(find_memory_region_ftype {aka gdb::function_view<bool(long unsigned int, long unsigned int, bool, bool, bool, bool, bool, bool)>}) (CORE_ADDR, std::size_t&, bool, bool, bool, bool, bool)’
 3152 |   return func ((CORE_ADDR) map->pr_vaddr,
      |          ~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
 3153 |                map->pr_size,
      |                ~~~~~~~~~~~~~
 3154 |                (map->pr_mflags & MA_READ) != 0,
      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 3155 |                (map->pr_mflags & MA_WRITE) != 0,
      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 3156 |                (map->pr_mflags & MA_EXEC) != 0,
      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 3157 |                true, /* MODIFIED is unknown, pass it as true.  */
      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
 3158 |                false);
      |                ~~~~~~

Fixed by passing the new HOLE arg.

procfs.c: In function ‘void info_proc_mappings(procinfo*, int)’:
procfs.c:3257:25: error: too many arguments to function ‘bool iterate_over_mappings(procinfo*, find_memory_region_ftype, bool (*)(prmap*, find_memory_region_ftype))’
 3257 |   iterate_over_mappings (pi, NULL, NULL, info_mappings_callback);
      |   ~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~

Fixed by removing the unused arg.

Tested on sparcv9-sun-solaris2.11 and amd64-pc-solaris2.11 with results
similar to the gdb-17 ones.

Committed to trunk.


This patch should also go to the gdb-18 branch: I'm not sure if I should
just commit it or it needs approval.

Besides, this has happened repeatedly over the last couple of years, so
I'm now working to set up Solaris/sparcv9 and amd64 builders parallel to
the recent binutils-solaris11-* ones so failures are noticed early.

There are a few issues with the subset of patches run by the gdb
builders: I'll report once I'm closer to clean results.

	Rainer

-- 
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University



[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: gdb-18.patch --]
[-- Type: text/x-patch, Size: 1295 bytes --]

diff --git a/gdb/procfs.c b/gdb/procfs.c
--- a/gdb/procfs.c
+++ b/gdb/procfs.c
@@ -2854,7 +2854,7 @@ procfs_notice_thread (procinfo *pi, proc
   ptid_t gdb_threadid = ptid_t (pi->pid, thread->tid, 0);
 
   thread_info *thr = the_procfs_target.find_thread (gdb_threadid);
-  if (thr == NULL || thr->state == THREAD_EXITED)
+  if (thr == NULL || thr->state () == THREAD_EXITED)
     add_thread (&the_procfs_target, gdb_threadid);
 
   return 0;
@@ -3155,7 +3155,7 @@ find_memory_regions_callback (struct prm
 	       (map->pr_mflags & MA_WRITE) != 0,
 	       (map->pr_mflags & MA_EXEC) != 0,
 	       true, /* MODIFIED is unknown, pass it as true.  */
-	       false);
+	       false, false);
 }
 
 /* External interface.  Calls a callback function once for each
@@ -3205,8 +3205,7 @@ mappingflags (long flags)
    mappings'.  */
 
 static bool
-info_mappings_callback (struct prmap *map, find_memory_region_ftype ignore,
-			void *unused)
+info_mappings_callback (struct prmap *map, find_memory_region_ftype ignore)
 {
   unsigned int pr_off;
 
@@ -3254,7 +3253,7 @@ info_proc_mappings (procinfo *pi, int su
 		"    Offset",
 		"Flags");
 
-  iterate_over_mappings (pi, NULL, NULL, info_mappings_callback);
+  iterate_over_mappings (pi, NULL, info_mappings_callback);
   gdb_printf ("\n");
 }
 

Re: [PATCH] Fix procfs.c compilation on Solaris

[Eli Zaretskii]

> From: Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
> Cc: Andrew Burgess <aburgess@redhat.com>
> Date: Fri, 14 Aug 2026 11:34:14 +0200
> 
> procfs.c doesn't compile on trunk and the gdb-18 branch:
> 
> procfs.c: In function ‘int procfs_notice_thread(procinfo*, procinfo*, void*)’:
> procfs.c:2857:33: error: invalid use of non-static member function ‘thread_state thread_info::state() const’
>  2857 |   if (thr == NULL || thr->state == THREAD_EXITED)
>       |                      ~~~~~~~~~~~^~~~~~~~~~~~~~~~
> 
> Fixed by calling the member function.
> 
> procfs.c: In function ‘bool find_memory_regions_callback(prmap*, find_memory_region_ftype)’:
> procfs.c:3152:15: error: no match for call to ‘(find_memory_region_ftype {aka gdb::function_view<bool(long unsigned int, long unsigned int, bool, bool, bool, bool, bool, bool)>}) (CORE_ADDR, std::size_t&, bool, bool, bool, bool, bool)’
>  3152 |   return func ((CORE_ADDR) map->pr_vaddr,
>       |          ~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
>  3153 |                map->pr_size,
>       |                ~~~~~~~~~~~~~
>  3154 |                (map->pr_mflags & MA_READ) != 0,
>       |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>  3155 |                (map->pr_mflags & MA_WRITE) != 0,
>       |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>  3156 |                (map->pr_mflags & MA_EXEC) != 0,
>       |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>  3157 |                true, /* MODIFIED is unknown, pass it as true.  */
>       |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>  3158 |                false);
>       |                ~~~~~~
> 
> Fixed by passing the new HOLE arg.
> 
> procfs.c: In function ‘void info_proc_mappings(procinfo*, int)’:
> procfs.c:3257:25: error: too many arguments to function ‘bool iterate_over_mappings(procinfo*, find_memory_region_ftype, bool (*)(prmap*, find_memory_region_ftype))’
>  3257 |   iterate_over_mappings (pi, NULL, NULL, info_mappings_callback);
>       |   ~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> 
> Fixed by removing the unused arg.
> 
> Tested on sparcv9-sun-solaris2.11 and amd64-pc-solaris2.11 with results
> similar to the gdb-17 ones.
> 
> Committed to trunk.
> 
> 
> This patch should also go to the gdb-18 branch: I'm not sure if I should
> just commit it or it needs approval.

The rule until now was that each bugfix on the release branch needs a
Bugzilla report.

Thanks.

Re: [PATCH] Fix procfs.c compilation on Solaris

[Hannes Domani]

Am Freitag, 14. August 2026 um 13:34:30 MESZ hat Eli Zaretskii <eliz@gnu.org> Folgendes geschrieben:

> > From: Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
> > Cc: Andrew Burgess <aburgess@redhat.com>
> > Date: Fri, 14 Aug 2026 11:34:14 +0200
> >
> > procfs.c doesn't compile on trunk and the gdb-18 branch:
> >
> > procfs.c: In function ‘int procfs_notice_thread(procinfo*, procinfo*, void*)’:
> > procfs.c:2857:33: error: invalid use of non-static member function ‘thread_state thread_info::state() const’
> >  2857 |  if (thr == NULL || thr->state == THREAD_EXITED)
> >      |                      ~~~~~~~~~~~^~~~~~~~~~~~~~~~
> >
> > Fixed by calling the member function.
> >
> > procfs.c: In function ‘bool find_memory_regions_callback(prmap*, find_memory_region_ftype)’:
> > procfs.c:3152:15: error: no match for call to ‘(find_memory_region_ftype {aka gdb::function_view<bool(long unsigned int, long unsigned int, bool, bool, bool, bool, bool, bool)>}) (CORE_ADDR, std::size_t&, bool, bool, bool, bool, bool)’
> >  3152 |  return func ((CORE_ADDR) map->pr_vaddr,
> >      |          ~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
> >  3153 |                map->pr_size,
> >      |                ~~~~~~~~~~~~~
> >  3154 |                (map->pr_mflags & MA_READ) != 0,
> >      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >  3155 |                (map->pr_mflags & MA_WRITE) != 0,
> >      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >  3156 |                (map->pr_mflags & MA_EXEC) != 0,
> >      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >  3157 |                true, /* MODIFIED is unknown, pass it as true.  */
> >      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >  3158 |                false);
> >      |                ~~~~~~
> >
> > Fixed by passing the new HOLE arg.
> >
> > procfs.c: In function ‘void info_proc_mappings(procinfo*, int)’:
> > procfs.c:3257:25: error: too many arguments to function ‘bool iterate_over_mappings(procinfo*, find_memory_region_ftype, bool (*)(prmap*, find_memory_region_ftype))’
> >  3257 |  iterate_over_mappings (pi, NULL, NULL, info_mappings_callback);
> >      |  ~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >
> > Fixed by removing the unused arg.
> >
> > Tested on sparcv9-sun-solaris2.11 and amd64-pc-solaris2.11 with results
> > similar to the gdb-17 ones.
> >
> > Committed to trunk.
> >
> >
> > This patch should also go to the gdb-18 branch: I'm not sure if I should
> > just commit it or it needs approval.
> 
> The rule until now was that each bugfix on the release branch needs a
> Bugzilla report.

AFAIK a bugzilla report is only necessary after the first release on that
branch, but a global maintainer has to approve in any case, see:
https://sourceware.org/gdb/wiki/PushingToReleaseBranch


Hannes

Re: [PATCH] Fix procfs.c compilation on Solaris

[Andrew Burgess]

Hannes Domani <ssbssa@yahoo.de> writes:

> Am Freitag, 14. August 2026 um 13:34:30 MESZ hat Eli Zaretskii <eliz@gnu.org> Folgendes geschrieben:
>
>> > From: Rainer Orth <ro@CeBiTec.Uni-Bielefeld.DE>
>> > Cc: Andrew Burgess <aburgess@redhat.com>
>> > Date: Fri, 14 Aug 2026 11:34:14 +0200
>> >
>> > procfs.c doesn't compile on trunk and the gdb-18 branch:
>> >
>> > procfs.c: In function ‘int procfs_notice_thread(procinfo*, procinfo*, void*)’:
>> > procfs.c:2857:33: error: invalid use of non-static member function ‘thread_state thread_info::state() const’
>> >  2857 |  if (thr == NULL || thr->state == THREAD_EXITED)
>> >      |                      ~~~~~~~~~~~^~~~~~~~~~~~~~~~
>> >
>> > Fixed by calling the member function.
>> >
>> > procfs.c: In function ‘bool find_memory_regions_callback(prmap*, find_memory_region_ftype)’:
>> > procfs.c:3152:15: error: no match for call to ‘(find_memory_region_ftype {aka gdb::function_view<bool(long unsigned int, long unsigned int, bool, bool, bool, bool, bool, bool)>}) (CORE_ADDR, std::size_t&, bool, bool, bool, bool, bool)’
>> >  3152 |  return func ((CORE_ADDR) map->pr_vaddr,
>> >      |          ~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~
>> >  3153 |                map->pr_size,
>> >      |                ~~~~~~~~~~~~~
>> >  3154 |                (map->pr_mflags & MA_READ) != 0,
>> >      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> >  3155 |                (map->pr_mflags & MA_WRITE) != 0,
>> >      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> >  3156 |                (map->pr_mflags & MA_EXEC) != 0,
>> >      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> >  3157 |                true, /* MODIFIED is unknown, pass it as true.  */
>> >      |                ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> >  3158 |                false);
>> >      |                ~~~~~~
>> >
>> > Fixed by passing the new HOLE arg.
>> >
>> > procfs.c: In function ‘void info_proc_mappings(procinfo*, int)’:
>> > procfs.c:3257:25: error: too many arguments to function ‘bool iterate_over_mappings(procinfo*, find_memory_region_ftype, bool (*)(prmap*, find_memory_region_ftype))’
>> >  3257 |  iterate_over_mappings (pi, NULL, NULL, info_mappings_callback);
>> >      |  ~~~~~~~~~~~~~~~~~~~~~~^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
>> >
>> > Fixed by removing the unused arg.
>> >
>> > Tested on sparcv9-sun-solaris2.11 and amd64-pc-solaris2.11 with results
>> > similar to the gdb-17 ones.
>> >
>> > Committed to trunk.
>> >
>> >
>> > This patch should also go to the gdb-18 branch: I'm not sure if I should
>> > just commit it or it needs approval.
>> 
>> The rule until now was that each bugfix on the release branch needs a
>> Bugzilla report.
>
> AFAIK a bugzilla report is only necessary after the first release on that
> branch, but a global maintainer has to approve in any case, see:
> https://sourceware.org/gdb/wiki/PushingToReleaseBranch

This is how I've always understood the rule too.  I'm happy to see this
pushed to gdb-18-branch if it has not already been done so.

Approved-By: Andrew Burgess <aburgess@redhat.com>

thanks,
Andrew

Re: [PATCH] Fix procfs.c compilation on Solaris

[Rainer Orth]

Hi Andrew,

>>> > This patch should also go to the gdb-18 branch: I'm not sure if I should
>>> > just commit it or it needs approval.
>>> 
>>> The rule until now was that each bugfix on the release branch needs a
>>> Bugzilla report.
>>
>> AFAIK a bugzilla report is only necessary after the first release on that
>> branch, but a global maintainer has to approve in any case, see:
>> https://sourceware.org/gdb/wiki/PushingToReleaseBranch
>
> This is how I've always understood the rule too.  I'm happy to see this
> pushed to gdb-18-branch if it has not already been done so.
>
> Approved-By: Andrew Burgess <aburgess@redhat.com>

done now.

Thanks.
	Rainer

-- 
-----------------------------------------------------------------------------
Rainer Orth, Center for Biotechnology, Bielefeld University