[ARI regression] About score-tdep commit

[ARI regression] About score-tdep commit

[Pierre Muller]

  Hi Don,

  looking at 
http://sourceware.org/gdb/current/ari/
I found a revival of TARGET_HAS_HARDWARE_WATCHPOINTS 
macros, which should not be there.

  your commit of score-tdep.h was apparently approved in march
but you committed it only in august.
  In the mean time, the macro
TARGET_HAS_HARDWARE_WATCHPOINTS has been removed
and several other macros have been renamed
TARGET_CAN_USE_HARDWARE_WATCHPOINT => target_can_use_hardware_watchpoint
STOPPED_BY_WATCHPOINT => target_stopped_by_watchpoint
HAVE_CONTINUABLE_HARDWARE_WATCHPOINT => target_have_continuable_watchpoint
TARGET_REGION_OK_FOR_HW_WATCHPOINT => target_region_ok_for_hw_watchpoint

  Thus the current code is not functional.

  All the hardware watchpoint related macros
that you set in score-tdep.c should be replaced by
appropriate entry in a target_ops struct...

  But I saw no target_ops for your system,
thus I don't really know how to fix your code.
Could you try to look into this?

Pierre Muller
GDB ARI (Awk Regression Index) maintainer


> -----Message d'origine-----
> De : gdb-cvs-owner@sourceware.org [mailto:gdb-cvs-owner@sourceware.org]
> De la part de donlee@sourceware.org
> Envoyé : Thursday, August 06, 2009 12:29 PM
> À : gdb-cvs@sourceware.org
> Objet : src/gdb ChangeLog MAINTAINERS NEWS configure.t ...
> 
> CVSROOT:	/cvs/src
> Module name:	src
> Changes by:	donlee@sourceware.org	2009-08-06 10:28:38
> 
> Modified files:
> 	gdb            : ChangeLog MAINTAINERS NEWS configure.tgt
> 	                 score-tdep.c score-tdep.h
> 
> Log message:
> 	* score-tdep.c: Support a new Sunplus CT S+core variant, S+core
> 3.
> 	* score-tdep.h: Ditto.
> 	* NEWS: Ditto.
> 	* configure.tgt: Test do we have a simulator.
> 	* MAINTAINERS: Add myself under target score
> 
> Patches:
> http://sources.redhat.com/cgi-
> bin/cvsweb.cgi/src/gdb/ChangeLog.diff?cvsroot=src&r1=1.10781&r2=1.10782
> http://sources.redhat.com/cgi-
> bin/cvsweb.cgi/src/gdb/MAINTAINERS.diff?cvsroot=src&r1=1.422&r2=1.423
> http://sources.redhat.com/cgi-
> bin/cvsweb.cgi/src/gdb/NEWS.diff?cvsroot=src&r1=1.324&r2=1.325
> http://sources.redhat.com/cgi-
> bin/cvsweb.cgi/src/gdb/configure.tgt.diff?cvsroot=src&r1=1.223&r2=1.224
> http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/gdb/score-
> tdep.c.diff?cvsroot=src&r1=1.20&r2=1.21
> http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/gdb/score-
> tdep.h.diff?cvsroot=src&r1=1.5&r2=1.6

[PING] About score-tdep commit

[Pierre Muller]

  Branch 7.0 is open now,
but no one reacted to this email I sent
where I said that the watchpoint
related code for score-tdep
is non functional.
  Can we leave this as is in 7.0?


Pierre Muller
ARI maintainer for GDB







First message:

  Hi Don,

  looking at 
http://sourceware.org/gdb/current/ari/
I found a revival of TARGET_HAS_HARDWARE_WATCHPOINTS 
macros, which should not be there.

  your commit of score-tdep.h was apparently approved in march
but you committed it only in august.
  In the mean time, the macro
TARGET_HAS_HARDWARE_WATCHPOINTS has been removed
and several other macros have been renamed
TARGET_CAN_USE_HARDWARE_WATCHPOINT => target_can_use_hardware_watchpoint
STOPPED_BY_WATCHPOINT => target_stopped_by_watchpoint
HAVE_CONTINUABLE_HARDWARE_WATCHPOINT => target_have_continuable_watchpoint
TARGET_REGION_OK_FOR_HW_WATCHPOINT => target_region_ok_for_hw_watchpoint

  Thus the current code is not functional.

  All the hardware watchpoint related macros
that you set in score-tdep.c should be replaced by
appropriate entry in a target_ops struct...

  But I saw no target_ops for your system,
thus I don't really know how to fix your code.
Could you try to look into this?

Pierre Muller
GDB ARI (Awk Regression Index) maintainer


> -----Message d'origine-----
> De : gdb-cvs-owner@sourceware.org [mailto:gdb-cvs-owner@sourceware.org]
> De la part de donlee@sourceware.org
> Envoyé : Thursday, August 06, 2009 12:29 PM
> À : gdb-cvs@sourceware.org
> Objet : src/gdb ChangeLog MAINTAINERS NEWS configure.t ...
> 
> CVSROOT:	/cvs/src
> Module name:	src
> Changes by:	donlee@sourceware.org	2009-08-06 10:28:38
> 
> Modified files:
> 	gdb            : ChangeLog MAINTAINERS NEWS configure.tgt
> 	                 score-tdep.c score-tdep.h
> 
> Log message:
> 	* score-tdep.c: Support a new Sunplus CT S+core variant, S+core
> 3.
> 	* score-tdep.h: Ditto.
> 	* NEWS: Ditto.
> 	* configure.tgt: Test do we have a simulator.
> 	* MAINTAINERS: Add myself under target score
> 
> Patches:
> http://sources.redhat.com/cgi-
> bin/cvsweb.cgi/src/gdb/ChangeLog.diff?cvsroot=src&r1=1.10781&r2=1.10782
> http://sources.redhat.com/cgi-
> bin/cvsweb.cgi/src/gdb/MAINTAINERS.diff?cvsroot=src&r1=1.422&r2=1.423
> http://sources.redhat.com/cgi-
> bin/cvsweb.cgi/src/gdb/NEWS.diff?cvsroot=src&r1=1.324&r2=1.325
> http://sources.redhat.com/cgi-
> bin/cvsweb.cgi/src/gdb/configure.tgt.diff?cvsroot=src&r1=1.223&r2=1.224
> http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/gdb/score-
> tdep.c.diff?cvsroot=src&r1=1.20&r2=1.21
> http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/gdb/score-
> tdep.h.diff?cvsroot=src&r1=1.5&r2=1.6

Re: [PING] About score-tdep commit

[Joel Brobecker]

>   Branch 7.0 is open now, but no one reacted to this email I sent
>   where I said that the watchpoint related code for score-tdep is non
>   functional. Can we leave this as is in 7.0?

Thanks for the ping.

I was assuming that the score maintainer would be reacting to your
message.  Perhaps he does not receive emails sent to his sourceware
address. Trying on his regular email address...

> 
> 
> Pierre Muller
> ARI maintainer for GDB
> 
> 
> 
> 
> 
> 
> 
> First message:
> 
>   Hi Don,
> 
>   looking at 
> http://sourceware.org/gdb/current/ari/
> I found a revival of TARGET_HAS_HARDWARE_WATCHPOINTS 
> macros, which should not be there.
> 
>   your commit of score-tdep.h was apparently approved in march
> but you committed it only in august.
>   In the mean time, the macro
> TARGET_HAS_HARDWARE_WATCHPOINTS has been removed
> and several other macros have been renamed
> TARGET_CAN_USE_HARDWARE_WATCHPOINT => target_can_use_hardware_watchpoint
> STOPPED_BY_WATCHPOINT => target_stopped_by_watchpoint
> HAVE_CONTINUABLE_HARDWARE_WATCHPOINT => target_have_continuable_watchpoint
> TARGET_REGION_OK_FOR_HW_WATCHPOINT => target_region_ok_for_hw_watchpoint
> 
>   Thus the current code is not functional.
> 
>   All the hardware watchpoint related macros
> that you set in score-tdep.c should be replaced by
> appropriate entry in a target_ops struct...
> 
>   But I saw no target_ops for your system,
> thus I don't really know how to fix your code.
> Could you try to look into this?
> 
> Pierre Muller
> GDB ARI (Awk Regression Index) maintainer
> 
> 
> > -----Message d'origine-----
> > De : gdb-cvs-owner@sourceware.org [mailto:gdb-cvs-owner@sourceware.org]
> > De la part de donlee@sourceware.org
> > Envoyé : Thursday, August 06, 2009 12:29 PM
> > À : gdb-cvs@sourceware.org
> > Objet : src/gdb ChangeLog MAINTAINERS NEWS configure.t ...
> > 
> > CVSROOT:	/cvs/src
> > Module name:	src
> > Changes by:	donlee@sourceware.org	2009-08-06 10:28:38
> > 
> > Modified files:
> > 	gdb            : ChangeLog MAINTAINERS NEWS configure.tgt
> > 	                 score-tdep.c score-tdep.h
> > 
> > Log message:
> > 	* score-tdep.c: Support a new Sunplus CT S+core variant, S+core
> > 3.
> > 	* score-tdep.h: Ditto.
> > 	* NEWS: Ditto.
> > 	* configure.tgt: Test do we have a simulator.
> > 	* MAINTAINERS: Add myself under target score
> > 
> > Patches:
> > http://sources.redhat.com/cgi-
> > bin/cvsweb.cgi/src/gdb/ChangeLog.diff?cvsroot=src&r1=1.10781&r2=1.10782
> > http://sources.redhat.com/cgi-
> > bin/cvsweb.cgi/src/gdb/MAINTAINERS.diff?cvsroot=src&r1=1.422&r2=1.423
> > http://sources.redhat.com/cgi-
> > bin/cvsweb.cgi/src/gdb/NEWS.diff?cvsroot=src&r1=1.324&r2=1.325
> > http://sources.redhat.com/cgi-
> > bin/cvsweb.cgi/src/gdb/configure.tgt.diff?cvsroot=src&r1=1.223&r2=1.224
> > http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/gdb/score-
> > tdep.c.diff?cvsroot=src&r1=1.20&r2=1.21
> > http://sources.redhat.com/cgi-bin/cvsweb.cgi/src/gdb/score-
> > tdep.h.diff?cvsroot=src&r1=1.5&r2=1.6
> 

-- 
Joel

Re: [PING] About score-tdep commit

[don.lee]

Joel Brobecker <brobecker@adacore.com> 写于 2009-09-18 00:05:37:
> 
> Thanks for the ping.
> 
> I was assuming that the score maintainer would be reacting to your
> message.  Perhaps he does not receive emails sent to his sourceware
> address. Trying on his regular email address...
> 

Sorry for the delay, busy for other project, didn't check these emails for 
a while.

> > First message:
> > 
> >   Hi Don,
> > 
> >   looking at 
> > http://sourceware.org/gdb/current/ari/
> > I found a revival of TARGET_HAS_HARDWARE_WATCHPOINTS 
> > macros, which should not be there.
> > 

I'll fix these problems as soon as possible.

don

RE: [PING] About score-tdep commit

[Pierre Muller]

Hi Don,
it looks like you have committed
some changes to score-tdep.c and score-tdep.h
on September 27. 
  I only realized this today...

  This commit is only in main branch,
shouldn't this also be included in gdb 7.0 branch
as it removes code that is surely broken in 
the branch?

Pierre

> -----Message d'origine-----
> De : gdb-patches-owner@sourceware.org [mailto:gdb-patches-
> owner@sourceware.org] De la part de don.lee@sunplusct.com
> Envoyé : Sunday, September 27, 2009 6:14 AM
> À : Joel Brobecker
> Cc : don.lee@sunplusct.com; donlee@sourceware.org; gdb-
> patches@sourceware.org; Pierre Muller
> Objet : Re: [PING] About score-tdep commit
> 
> Joel Brobecker <brobecker@adacore.com> 写于 2009-09-18 00:05:37:
> >
> > Thanks for the ping.
> >
> > I was assuming that the score maintainer would be reacting to your
> > message.  Perhaps he does not receive emails sent to his sourceware
> > address. Trying on his regular email address...
> >
> 
> Sorry for the delay, busy for other project, didn't check these emails
> for
> a while.
> 
> > > First message:
> > >
> > >   Hi Don,
> > >
> > >   looking at
> > > http://sourceware.org/gdb/current/ari/
> > > I found a revival of TARGET_HAS_HARDWARE_WATCHPOINTS
> > > macros, which should not be there.
> > >
> 
> I'll fix these problems as soon as possible.
> 
> don

RE: [PING] About score-tdep commit

[don.lee]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="GB2312", Size: 641 bytes --]

"Pierre Muller" <muller@ics.u-strasbg.fr> дÓÚ 2009-10-03 21:06:15:

> Hi Don,
> it looks like you have committed
> some changes to score-tdep.c and score-tdep.h
> on September 27. 
>   I only realized this today...

My fault, I didn't post my patch to gdb-patches, next time
I'll do this.

> 
>   This commit is only in main branch,
> shouldn't this also be included in gdb 7.0 branch
> as it removes code that is surely broken in 
> the branch?
> 

Yes, it should be. Should I write this change to 7.0 branch (looks
like I can't write to this branch) or other one will do this?

don

\x16º&Öéj×!zÊÞ¶êçë¾|æX¬µªÜ†\a[¥«\…ë

Re: [PING] About score-tdep commit

[Joel Brobecker]

> My fault, I didn't post my patch to gdb-patches, next time
> I'll do this.

I would prefer if you deleted the code completely, instead of commenting
it out.  We're still cleaning up dead code that was commented-out years
ago.  If someone needs the code back, it's very easy to get it back from
a previous version of that file.  Can you do that cleanup, please?

Note that we've occasionally commented out code like this, but there
should be a comment explaining why the code was commented out, and we
should have a plan to address the issue in the short to medium term.
Even then, I still think that commenting code out is usually wrong.

> >   This commit is only in main branch, shouldn't this also be
> >   included in gdb 7.0 branch as it removes code that is surely
> >   broken in the branch?
> 
> Yes, it should be. Should I write this change to 7.0 branch (looks
> like I can't write to this branch) or other one will do this?

If you don't mind, can you do the commit? However, please do not
commit the change as you made it on Sept 27th.  Instead, do the
cleanup I suggested above, and commit both patches on the branch,
or a merged version of both patches.

Thank you,
-- 
Joel

Re: [PING] About score-tdep commit

[don.lee]

[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #1: Type: text/plain; charset="GB2312", Size: 1256 bytes --]

Joel Brobecker <brobecker@adacore.com> дÓÚ 2009-10-14 10:04:43:

> > My fault, I didn't post my patch to gdb-patches, next time
> > I'll do this.
> 
> I would prefer if you deleted the code completely, instead of commenting
> it out.  We're still cleaning up dead code that was commented-out years
> ago.  If someone needs the code back, it's very easy to get it back from
> a previous version of that file.  Can you do that cleanup, please?
> 
> Note that we've occasionally commented out code like this, but there
> should be a comment explaining why the code was commented out, and we
> should have a plan to address the issue in the short to medium term.
> Even then, I still think that commenting code out is usually wrong.
> 
> > >   This commit is only in main branch, shouldn't this also be
> > >   included in gdb 7.0 branch as it removes code that is surely
> > >   broken in the branch?
> > 
> > Yes, it should be. Should I write this change to 7.0 branch (looks
> > like I can't write to this branch) or other one will do this?
> 
> If you don't mind, can you do the commit? 

These dead codes in gdb/score-tdep.[ch] have been deleted, both in main
branch and gdb 7.0 branch.

don

\x16º&Öéj×!zÊÞ¶êçë¿4æX¬µªÜ†\a[¥«\…ë