-stack-list-frames HIGH_FRAME changes

-stack-list-frames HIGH_FRAME changes

[Vladimir Prus]

[-- Attachment #1: Type: text/plain, Size: 149 bytes --]


Hi,
attached patch stops -stack-list-frames from emitting error when the
HIGH_FRAME parameter is larger than the actual number of frames.

- Volodya

[-- Attachment #2: @-stack-list-frames.diff --]
[-- Type: text/plain, Size: 5814 bytes --]

? .gdbinit
? @-stack-list-frames.diff
? Makefile
? ada-exp.c
? ada-lex.c
? c-exp.c
? config.cache
? config.h
? config.log
? config.status
? cp-name-parser.c
? f-exp.c
? gdb
? gdbtui
? init.c
? jv-exp.c
? m2-exp.c
? objc-exp.c
? observer.h
? observer.inc
? p-exp.c
? stamp-h
? version.c
? doc/Makefile
? doc/config.log
? doc/config.status
? gdbserver/Makefile
? gdbserver/config.h
? gdbserver/config.log
? gdbserver/config.status
? gdbserver/gdbreplay
? gdbserver/gdbserver
? gdbserver/reg-i386-linux.c
? gdbserver/stamp-h
? gdbserver/version.c
? mi/ChangeLog
? testsuite/Makefile
? testsuite/config.log
? testsuite/config.status
? testsuite/gdb.log
? testsuite/gdb.sum
? testsuite/site.exp
? testsuite/gdb.ada/Makefile
? testsuite/gdb.arch/Makefile
? testsuite/gdb.asm/Makefile
? testsuite/gdb.base/Makefile
? testsuite/gdb.cp/Makefile
? testsuite/gdb.disasm/Makefile
? testsuite/gdb.dwarf2/Makefile
? testsuite/gdb.fortran/Makefile
? testsuite/gdb.java/Makefile
? testsuite/gdb.mi/Makefile
? testsuite/gdb.mi/basics
? testsuite/gdb.mi/gdb669-pthreads
? testsuite/gdb.mi/gdb701
? testsuite/gdb.mi/gdb792
? testsuite/gdb.mi/mi-console
? testsuite/gdb.mi/mi-pthreads
? testsuite/gdb.mi/mi-read-memory
? testsuite/gdb.mi/mi-stack
? testsuite/gdb.mi/mi-syn-frame
? testsuite/gdb.mi/mi-var-child
? testsuite/gdb.mi/mi2-pthreads
? testsuite/gdb.mi/until
? testsuite/gdb.mi/var-cmd
? testsuite/gdb.objc/Makefile
? testsuite/gdb.server/Makefile
? testsuite/gdb.stabs/Makefile
? testsuite/gdb.stabs/config.log
? testsuite/gdb.stabs/config.status
? testsuite/gdb.threads/Makefile
? testsuite/gdb.trace/Makefile
Index: ChangeLog
===================================================================
RCS file: /cvs/src/src/gdb/ChangeLog,v
retrieving revision 1.7714
diff -u -r1.7714 ChangeLog
--- ChangeLog	3 May 2006 22:59:38 -0000	1.7714
+++ ChangeLog	5 May 2006 07:50:28 -0000
@@ -1,3 +1,10 @@
+2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
+
+	* mi/mi-cmd-stack.c (mi_cmd_stack_list_frames): Don't emit error
+	if high requested frame number is larger then number of available 
+	frames.
+	
+
 2006-05-03  Vladimir Prus  <ghost@cs.msu.su>
 
         * varobj.c (c_value_of_variable): Ignore top-level references.
Index: doc/ChangeLog
===================================================================
RCS file: /cvs/src/src/gdb/doc/ChangeLog,v
retrieving revision 1.572
diff -u -r1.572 ChangeLog
--- doc/ChangeLog	27 Apr 2006 23:03:42 -0000	1.572
+++ doc/ChangeLog	5 May 2006 07:50:29 -0000
@@ -1,3 +1,9 @@
+2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
+
+	* gdb.texinfo (GDB/MI Stack Manipulation): Mention that
+	-stack-list-locals HIGH_FRAME argument can be larger then the
+	actual number of frames.
+
 2006-04-27  Michael Snyder  <msnyder@redhat.com>
 
 	* gdb.texinfo (delete-fork): Command renamed to "delete fork".
Index: doc/gdb.texinfo
===================================================================
RCS file: /cvs/src/src/gdb/doc/gdb.texinfo,v
retrieving revision 1.325
diff -u -r1.325 gdb.texinfo
--- doc/gdb.texinfo	27 Apr 2006 23:03:42 -0000	1.325
+++ doc/gdb.texinfo	5 May 2006 07:50:32 -0000
@@ -19922,7 +19922,10 @@
 If invoked without arguments, this command prints a backtrace for the
 whole stack.  If given two integer arguments, it shows the frames whose
 levels are between the two arguments (inclusive).  If the two arguments
-are equal, it shows the single frame at the corresponding level.
+are equal, it shows the single frame at the corresponding level. It is
+an error if @var{low-frame} is larger than the actual number of frames. On
+the other hand, @var{high-frame} may be larger then the actual number of
+frames, in which case only existing frames will be returned.
 
 @subsubheading @value{GDBN} Command
 
Index: mi/mi-cmd-stack.c
===================================================================
RCS file: /cvs/src/src/gdb/mi/mi-cmd-stack.c,v
retrieving revision 1.29
diff -u -r1.29 mi-cmd-stack.c
--- mi/mi-cmd-stack.c	23 Dec 2005 18:57:46 -0000	1.29
+++ mi/mi-cmd-stack.c	5 May 2006 07:50:32 -0000
@@ -88,8 +88,6 @@
     }
 
   do_cleanups (cleanup_stack);
-  if (i < frame_high)
-    error (_("mi_cmd_stack_list_frames: Not enough frames in stack."));
 
   return MI_CMD_DONE;
 }
Index: testsuite/ChangeLog
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/ChangeLog,v
retrieving revision 1.1208
diff -u -r1.1208 ChangeLog
--- testsuite/ChangeLog	1 May 2006 22:21:35 -0000	1.1208
+++ testsuite/ChangeLog	5 May 2006 07:50:34 -0000
@@ -1,3 +1,9 @@
+2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
+
+	* gdb.mi/mi2-stack.exp (test_stack_frame_listing): Test that
+	HIGH_FRAME argument to -stack-list-locals can be larger than
+	the number of frames.
+
 2006-05-02  Mark Kettenis  <kettenis@gnu.org>
 
 	* gdb.asm/asm-source.exp: Add powerpc-*-netbsd* to the list of
Index: testsuite/gdb.mi/mi2-stack.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.mi/mi2-stack.exp,v
retrieving revision 1.6
diff -u -r1.6 mi2-stack.exp
--- testsuite/gdb.mi/mi2-stack.exp	27 Sep 2005 22:39:03 -0000	1.6
+++ testsuite/gdb.mi/mi2-stack.exp	5 May 2006 07:50:34 -0000
@@ -74,6 +74,11 @@
     mi_gdb_test "235-stack-info-frame" \
             "235\\^done,frame=\{level=\"0\",addr=\"$hex\",func=\"callee4\",file=\".*basics.c\",fullname=\"${fullname_syntax}${srcfile}\",line=\"$line_callee4_body\"\}" \
 	       "selected frame listing"
+
+    mi_gdb_test "236-stack-list-frames 1 300" \
+	    "236\\^done,stack=\\\[frame=\{level=\"1\",addr=\"$hex\",func=\"callee3\",.*\},frame=\{level=\"2\",addr=\"$hex\",func=\"callee2\",.*\},frame=\{level=\"3\",addr=\"$hex\",func=\"callee1\",.*\}\\\]" \
+                "stack frame listing 1 300"
+
 }
 
 proc test_stack_args_listing {} {

Re: -stack-list-frames HIGH_FRAME changes

[Eli Zaretskii]

> From:  Vladimir Prus <ghost@cs.msu.su>
> Date:  Fri, 05 May 2006 11:54:08 +0400
> 
> +2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> +
> +	* gdb.texinfo (GDB/MI Stack Manipulation): Mention that
> +	-stack-list-locals HIGH_FRAME argument can be larger then the
> +	actual number of frames.

This part of the patch is approved.

Thanks.

Re: -stack-list-frames HIGH_FRAME changes

[Daniel Jacobowitz]

On Fri, May 05, 2006 at 11:54:08AM +0400, Vladimir Prus wrote:
> Hi,
> attached patch stops -stack-list-frames from emitting error when the
> HIGH_FRAME parameter is larger than the actual number of frames.

As far as I'm concerned, this change is reasonable; but let's wait
until next week to see if anyone else has a reason for the current
behavior. (I don't think anyone will; just being cautious.)

> ? .gdbinit

Please trim all this next time :-)

> Index: ChangeLog
> ===================================================================
> RCS file: /cvs/src/src/gdb/ChangeLog,v
> retrieving revision 1.7714
> diff -u -r1.7714 ChangeLog
> --- ChangeLog	3 May 2006 22:59:38 -0000	1.7714
> +++ ChangeLog	5 May 2006 07:50:28 -0000
> @@ -1,3 +1,10 @@

It's usually better to include ChangeLog entries as text, not as diffs.
When they're diffs, they invariably generate patch rejects later on.

> +2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> +
> +	* mi/mi-cmd-stack.c (mi_cmd_stack_list_frames): Don't emit error
> +	if high requested frame number is larger then number of available 
> +	frames.
> +	
> +

Just one blank line between entries, please.

> -are equal, it shows the single frame at the corresponding level.
> +are equal, it shows the single frame at the corresponding level. It is
> +an error if @var{low-frame} is larger than the actual number of frames. On
> +the other hand, @var{high-frame} may be larger then the actual number of
> +frames, in which case only existing frames will be returned.

Two spaces after periods.  (Eli, does TeX actually care about that, or
is it just good for consistency?)

> Index: testsuite/ChangeLog
> ===================================================================
> RCS file: /cvs/src/src/gdb/testsuite/ChangeLog,v
> retrieving revision 1.1208
> diff -u -r1.1208 ChangeLog
> --- testsuite/ChangeLog	1 May 2006 22:21:35 -0000	1.1208
> +++ testsuite/ChangeLog	5 May 2006 07:50:34 -0000
> @@ -1,3 +1,9 @@
> +2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> +
> +	* gdb.mi/mi2-stack.exp (test_stack_frame_listing): Test that
> +	HIGH_FRAME argument to -stack-list-locals can be larger than
> +	the number of frames.


Please don't add new tests to just mi2-stack.exp.  I don't much care
whether mi2-stack.exp is updated or not, but when we finalize mi3 and
create mi3-* tests, they're going to be copied from mi-*.  So it's
important that new tests go there also.

-- 
Daniel Jacobowitz
CodeSourcery

Re: -stack-list-frames HIGH_FRAME changes

[Vladimir Prus]

[-- Attachment #1: Type: text/plain, Size: 3548 bytes --]

On Friday 05 May 2006 17:51, Daniel Jacobowitz wrote:
> On Fri, May 05, 2006 at 11:54:08AM +0400, Vladimir Prus wrote:
> > Hi,
> > attached patch stops -stack-list-frames from emitting error when the
> > HIGH_FRAME parameter is larger than the actual number of frames.
>
> As far as I'm concerned, this change is reasonable; but let's wait
> until next week to see if anyone else has a reason for the current
> behavior. (I don't think anyone will; just being cautious.)
>
> > ? .gdbinit
>
> Please trim all this next time :-)

Sure, sorry.

> > ===================================================================
> > RCS file: /cvs/src/src/gdb/ChangeLog,v
> > retrieving revision 1.7714
> > diff -u -r1.7714 ChangeLog
> > --- ChangeLog	3 May 2006 22:59:38 -0000	1.7714
> > +++ ChangeLog	5 May 2006 07:50:28 -0000
> > @@ -1,3 +1,10 @@
>
> It's usually better to include ChangeLog entries as text, not as diffs.
> When they're diffs, they invariably generate patch rejects later on.

The reason I've included them in patch is that I've learned about Emacs's 
add-change-log-entry command, which conveniently adds entry to existing 
file ;-) So, in case if there are several ChangeLog files in patch, I should 
put those entries separately, and indicate which file each entry goes to, 
right?

> > +2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> > +
> > +	* mi/mi-cmd-stack.c (mi_cmd_stack_list_frames): Don't emit error
> > +	if high requested frame number is larger then number of available
> > +	frames.
> > +
> > +
>
> Just one blank line between entries, please.

Ok.

> > -are equal, it shows the single frame at the corresponding level.
> > +are equal, it shows the single frame at the corresponding level. It is
> > +an error if @var{low-frame} is larger than the actual number of frames.
> > On +the other hand, @var{high-frame} may be larger then the actual number
> > of +frames, in which case only existing frames will be returned.
>
> Two spaces after periods.  

Ok.

>
> > Index: testsuite/ChangeLog
> > ===================================================================
> > RCS file: /cvs/src/src/gdb/testsuite/ChangeLog,v
> > retrieving revision 1.1208
> > diff -u -r1.1208 ChangeLog
> > --- testsuite/ChangeLog	1 May 2006 22:21:35 -0000	1.1208
> > +++ testsuite/ChangeLog	5 May 2006 07:50:34 -0000
> > @@ -1,3 +1,9 @@
> > +2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> > +
> > +	* gdb.mi/mi2-stack.exp (test_stack_frame_listing): Test that
> > +	HIGH_FRAME argument to -stack-list-locals can be larger than
> > +	the number of frames.
>
> Please don't add new tests to just mi2-stack.exp.  I don't much care
> whether mi2-stack.exp is updated or not, but when we finalize mi3 and
> create mi3-* tests, they're going to be copied from mi-*.  So it's
> important that new tests go there also.

So, what's the point of mi2 tests at all? 
In any case, there's revised patch:

./Changelog:
2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>

	* mi/mi-cmd-stack.c (mi_cmd_stack_list_frames): Don't emit error
	if high requested frame number is larger then number of available 
	frames.


doc/ChangeLog
2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>

	* gdb.texinfo (GDB/MI Stack Manipulation): Mention that
	-stack-list-locals HIGH_FRAME argument can be larger then the
	actual number of frames.


testsuite/ChangeLog

2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>

	* gdb.mi/mi-stack.exp (test_stack_frame_listing): Test that
	HIGH_FRAME argument to -stack-list-locals can be larger than
	the number of frames.


- Volodya


[-- Attachment #2: %40-stack-list-frames.diff --]
[-- Type: text/x-diff, Size: 2354 bytes --]

Index: doc/gdb.texinfo
===================================================================
RCS file: /cvs/src/src/gdb/doc/gdb.texinfo,v
retrieving revision 1.325
diff -u -r1.325 gdb.texinfo
--- doc/gdb.texinfo	27 Apr 2006 23:03:42 -0000	1.325
+++ doc/gdb.texinfo	5 May 2006 14:37:09 -0000
@@ -19922,7 +19922,10 @@
 If invoked without arguments, this command prints a backtrace for the
 whole stack.  If given two integer arguments, it shows the frames whose
 levels are between the two arguments (inclusive).  If the two arguments
-are equal, it shows the single frame at the corresponding level.
+are equal, it shows the single frame at the corresponding level.  It is
+an error if @var{low-frame} is larger than the actual number of
+frames.  On the other hand, @var{high-frame} may be larger then the
+actual number of frames, in which case only existing frames will be returned.
 
 @subsubheading @value{GDBN} Command
 
Index: mi/mi-cmd-stack.c
===================================================================
RCS file: /cvs/src/src/gdb/mi/mi-cmd-stack.c,v
retrieving revision 1.29
diff -u -r1.29 mi-cmd-stack.c
--- mi/mi-cmd-stack.c	23 Dec 2005 18:57:46 -0000	1.29
+++ mi/mi-cmd-stack.c	5 May 2006 14:37:09 -0000
@@ -88,8 +88,6 @@
     }
 
   do_cleanups (cleanup_stack);
-  if (i < frame_high)
-    error (_("mi_cmd_stack_list_frames: Not enough frames in stack."));
 
   return MI_CMD_DONE;
 }
Index: testsuite/gdb.mi/mi-stack.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.mi/mi-stack.exp,v
retrieving revision 1.19
diff -u -r1.19 mi-stack.exp
--- testsuite/gdb.mi/mi-stack.exp	27 Sep 2005 22:39:02 -0000	1.19
+++ testsuite/gdb.mi/mi-stack.exp	5 May 2006 14:37:10 -0000
@@ -74,6 +74,10 @@
     mi_gdb_test "235-stack-info-frame" \
             "235\\^done,frame=\{level=\"0\",addr=\"$hex\",func=\"callee4\",file=\".*${srcfile}\",fullname=\"${fullname_syntax}${srcfile}\",line=\"$line_callee4_body\"\}" \
 	       "selected frame listing"
+
+    mi_gdb_test "236-stack-list-frames 1 300" \
+	    "236\\^done,stack=\\\[frame=\{level=\"1\",addr=\"$hex\",func=\"callee3\",.*\},frame=\{level=\"2\",addr=\"$hex\",func=\"callee2\",.*\},frame=\{level=\"3\",addr=\"$hex\",func=\"callee1\",.*\}\\\]" \
+                "stack frame listing 1 300"
 }
 
 proc test_stack_args_listing {} {

Re: -stack-list-frames HIGH_FRAME changes

[Daniel Jacobowitz]

On Fri, May 05, 2006 at 06:38:19PM +0400, Vladimir Prus wrote:
> The reason I've included them in patch is that I've learned about Emacs's 
> add-change-log-entry command, which conveniently adds entry to existing 
> file ;-) So, in case if there are several ChangeLog files in patch, I should 
> put those entries separately, and indicate which file each entry goes to, 
> right?

Right.  I usually don't bother when it's obvious (e.g. there's only one
file named gdb.texinfo, and it's obviously in gdb/doc/).

Alexandre Oliva wrote a script to pull the entries out of diffs
automatically, if you prefer.


-- 
Daniel Jacobowitz
CodeSourcery

Re: -stack-list-frames HIGH_FRAME changes

[Jim Blandy]

Daniel Jacobowitz <drow@false.org> writes:
> On Fri, May 05, 2006 at 06:38:19PM +0400, Vladimir Prus wrote:
>> The reason I've included them in patch is that I've learned about Emacs's 
>> add-change-log-entry command, which conveniently adds entry to existing 
>> file ;-) So, in case if there are several ChangeLog files in patch, I should 
>> put those entries separately, and indicate which file each entry goes to, 
>> right?
>
> Right.  I usually don't bother when it's obvious (e.g. there's only one
> file named gdb.texinfo, and it's obviously in gdb/doc/).
>
> Alexandre Oliva wrote a script to pull the entries out of diffs
> automatically, if you prefer.

If you use Emacs, I have a patch-cleanup command that:

- pulls out ChangeLog patches

- removes patches to generated files *if* there is also a patch to
  the presumed source file (so Makefile.in patches get removed if
  there is an accompanying Makefile.am patch, but not otherwise).

Let me know if you're interested.

Re: -stack-list-frames HIGH_FRAME changes

[Bob Rossi]

> > Please don't add new tests to just mi2-stack.exp.  I don't much care
> > whether mi2-stack.exp is updated or not, but when we finalize mi3 and
> > create mi3-* tests, they're going to be copied from mi-*.  So it's
> > important that new tests go there also.
> 
> So, what's the point of mi2 tests at all? 
> In any case, there's revised patch:

The mi- tests are the "current" MI tests. The mi2- tests make sure that
gdb -i=mi2 work. 

So, at some point in the future we will make mi- tests become mi3-, we
will still have mi2- and mi- will become the new "current" MI tests.

Bob Rossi

Re: -stack-list-frames HIGH_FRAME changes

[Daniel Jacobowitz]

On Fri, May 05, 2006 at 10:48:43AM -0400, Bob Rossi wrote:
> > > Please don't add new tests to just mi2-stack.exp.  I don't much care
> > > whether mi2-stack.exp is updated or not, but when we finalize mi3 and
> > > create mi3-* tests, they're going to be copied from mi-*.  So it's
> > > important that new tests go there also.
> > 
> > So, what's the point of mi2 tests at all? 
> > In any case, there's revised patch:
> 
> The mi- tests are the "current" MI tests. The mi2- tests make sure that
> gdb -i=mi2 work. 
> 
> So, at some point in the future we will make mi- tests become mi3-, we
> will still have mi2- and mi- will become the new "current" MI tests.

Well, actually, shortly after that point we may drop the mi2-* tests.

We discuss this every couple of months and get ourselves more confused 
each time :-)  Since -i=mi currently sets MI == 2, it's all quite
unclear.  I'm just trying not to rock the boat.  If someone wants to
propose a reorganization, more power to them.

-- 
Daniel Jacobowitz
CodeSourcery

Re: -stack-list-frames HIGH_FRAME changes

[Daniel Jacobowitz]

On Fri, May 05, 2006 at 06:38:19PM +0400, Vladimir Prus wrote:
> In any case, there's revised patch:
> 
> ./Changelog:
> 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> 
> 	* mi/mi-cmd-stack.c (mi_cmd_stack_list_frames): Don't emit error
> 	if high requested frame number is larger then number of available 
> 	frames.
> 
> 
> doc/ChangeLog
> 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> 
> 	* gdb.texinfo (GDB/MI Stack Manipulation): Mention that
> 	-stack-list-locals HIGH_FRAME argument can be larger then the
> 	actual number of frames.
> 
> 
> testsuite/ChangeLog
> 
> 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> 
> 	* gdb.mi/mi-stack.exp (test_stack_frame_listing): Test that
> 	HIGH_FRAME argument to -stack-list-locals can be larger than
> 	the number of frames.

This patch is OK.  Sorry for forgetting about it.


-- 
Daniel Jacobowitz
CodeSourcery

Re: -stack-list-frames HIGH_FRAME changes

[Vladimir Prus]

On Thursday 07 September 2006 20:10, Daniel Jacobowitz wrote:
> On Fri, May 05, 2006 at 06:38:19PM +0400, Vladimir Prus wrote:
> > In any case, there's revised patch:
> >
> > ./Changelog:
> > 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> >
> > 	* mi/mi-cmd-stack.c (mi_cmd_stack_list_frames): Don't emit error
> > 	if high requested frame number is larger then number of available
> > 	frames.
> >
> >
> > doc/ChangeLog
> > 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> >
> > 	* gdb.texinfo (GDB/MI Stack Manipulation): Mention that
> > 	-stack-list-locals HIGH_FRAME argument can be larger then the
> > 	actual number of frames.
> >
> >
> > testsuite/ChangeLog
> >
> > 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> >
> > 	* gdb.mi/mi-stack.exp (test_stack_frame_listing): Test that
> > 	HIGH_FRAME argument to -stack-list-locals can be larger than
> > 	the number of frames.
>
> This patch is OK.  Sorry for forgetting about it.

Checked in.

- Volodya

Re: -stack-list-frames HIGH_FRAME changes

[Eli Zaretskii]

> Date: Fri, 5 May 2006 09:51:24 -0400
> From: Daniel Jacobowitz <drow@false.org>
> Cc: gdb-patches@sources.redhat.com
> 
> Two spaces after periods.  (Eli, does TeX actually care about that, or
> is it just good for consistency?)

TeX usually guesses correctly where the sentence ends even if you
leave only one space (and we have @:, @!, @., etc. commands for when
it could err), but leaving 2 spaces also facilitates Emacs commands
that move by sentences.

Re: -stack-list-frames HIGH_FRAME changes

[Vladimir Prus]

On Friday 05 May 2006 17:51, Daniel Jacobowitz wrote:
> On Fri, May 05, 2006 at 11:54:08AM +0400, Vladimir Prus wrote:
> > Hi,
> > attached patch stops -stack-list-frames from emitting error when the
> > HIGH_FRAME parameter is larger than the actual number of frames.
>
> As far as I'm concerned, this change is reasonable; but let's wait
> until next week to see if anyone else has a reason for the current
> behavior. (I don't think anyone will; just being cautious.)

Quite some time has passed. Ok to commit now?

- Volodya

Re: -stack-list-frames HIGH_FRAME changes

[Nick Roberts]

> Checked in.

I can't remember that far back but was there any reason for not making the same
change to -stack-list-arguments?

-- 
Nick                                           http://www.inet.net.nz/~nickrob

Re: -stack-list-frames HIGH_FRAME changes

[Daniel Jacobowitz]

On Fri, Sep 08, 2006 at 10:35:29AM +1200, Nick Roberts wrote:
> 
> 
> > Checked in.
> 
> I can't remember that far back but was there any reason for not making the same
> change to -stack-list-arguments?

There wasn't; if -stack-list-arguments has the same problem, could one
of you post a patch to remove the error?  TIA!

-- 
Daniel Jacobowitz
CodeSourcery

Re: -stack-list-frames HIGH_FRAME changes

[Vladimir Prus]

[-- Attachment #1: Type: text/plain, Size: 1096 bytes --]

On Friday 08 September 2006 02:35, Nick Roberts wrote:
> > Checked in.
>
> I can't remember that far back but was there any reason for not making the
> same change to -stack-list-arguments?

Other than the fact that KDevelop does not need -stack-list-arguments at the 
moment, so I even did not look at -stack-list-arguments? I think there are no 
other reasons. 

Here we go, OK?

- Volodya

./Changelog:
2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>

        * mi/mi-cmd-stack.c (mi_cmd_stack_list_args): Don't emit error
        if high requested frame number is larger then number of available 
        frames.


doc/ChangeLog
2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>

        * gdb.texinfo (GDB/MI Stack Manipulation): Mention that
        -stack-list-arguments HIGH_FRAME argument can be larger then the
        actual number of frames.


testsuite/ChangeLog

2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>

        * gdb.mi/mi-stack.exp (test_stack_args_listing): Test that
        HIGH_FRAME argument to -stack-list-arguments can be larger than
        the number of frames.

[-- Attachment #2: stack_list_arguments_high_frame.diff --]
[-- Type: text/x-diff, Size: 2669 bytes --]

Index: doc/gdb.texinfo
===================================================================
RCS file: /cvs/src/src/gdb/doc/gdb.texinfo,v
retrieving revision 1.351
diff -u -r1.351 gdb.texinfo
--- doc/gdb.texinfo	7 Sep 2006 16:40:18 -0000	1.351
+++ doc/gdb.texinfo	8 Sep 2006 06:24:41 -0000
@@ -19230,8 +19230,12 @@
 
 Display a list of the arguments for the frames between @var{low-frame}
 and @var{high-frame} (inclusive).  If @var{low-frame} and
-@var{high-frame} are not provided, list the arguments for the whole call
-stack.
+@var{high-frame} are not provided, list the arguments for the whole
+call stack. If the two arguments are equal, it shows the single frame
+at the corresponding level.  It is an error if @var{low-frame} is
+larger than the actual number of frames.  On the other hand,
+@var{high-frame} may be larger then the actual number of frames, in
+which case only existing frames will be returned.
 
 The @var{show-values} argument must have a value of 0 or 1.  A value of
 0 means that only the names of the arguments are listed, a value of 1
Index: mi/mi-cmd-stack.c
===================================================================
RCS file: /cvs/src/src/gdb/mi/mi-cmd-stack.c,v
retrieving revision 1.30
diff -u -r1.30 mi-cmd-stack.c
--- mi/mi-cmd-stack.c	7 Sep 2006 16:40:18 -0000	1.30
+++ mi/mi-cmd-stack.c	8 Sep 2006 06:24:41 -0000
@@ -205,8 +205,6 @@
     }
 
   do_cleanups (cleanup_stack_args);
-  if (i < frame_high)
-    error (_("mi_cmd_stack_list_args: Not enough frames in stack."));
 
   return MI_CMD_DONE;
 }
Index: testsuite/gdb.mi/mi-stack.exp
===================================================================
RCS file: /cvs/src/src/gdb/testsuite/gdb.mi/mi-stack.exp,v
retrieving revision 1.21
diff -u -r1.21 mi-stack.exp
--- testsuite/gdb.mi/mi-stack.exp	7 Sep 2006 16:40:19 -0000	1.21
+++ testsuite/gdb.mi/mi-stack.exp	8 Sep 2006 06:24:41 -0000
@@ -122,6 +122,10 @@
     mi_gdb_test "234-stack-list-arguments" \
 	    "&.*234\\^error,msg=\"mi_cmd_stack_list_args: Usage.*PRINT_VALUES.*FRAME_LOW FRAME_HIGH.*\"" \
 	    "stack args listing wrong"
+
+    mi_gdb_test "235-stack-list-arguments 1 1 3" \
+	    "235\\^done,stack-args=\\\[frame=\{level=\"1\",args=\\\[\{name=\"strarg\",value=\"$hex \\\\\"A string argument.\\\\\"\"\}\\\]\},frame=\{level=\"2\",args=\\\[\{name=\"intarg\",value=\"2\"\},\{name=\"strarg\",value=\"$hex \\\\\"A string argument.\\\\\"\"\}\\\]\},frame=\{level=\"3\",args=\\\[\{name=\"intarg\",value=\"2\"\},\{name=\"strarg\",value=\"$hex \\\\\"A string argument.\\\\\"\"\},\{name=\"fltarg\",value=\"3.5\"\}\\\]\}\\\]" \
+                "stack args listing 1 1 300"
 }
 
 proc test_stack_info_depth {} {

Re: -stack-list-frames HIGH_FRAME changes

[Nick Roberts]

 > Here we go, OK?

Looks fine to me apart from:

 > +
 > +    mi_gdb_test "235-stack-list-arguments 1 1 3" \
                                                 ^^^ (should be 300?)
 > +	    "235\\^done,stack-args=\\\[frame=\{level=\"1\",args=\\\[\{name=\"strarg\",value=\"$hex \\\\\"A string argument.\\\\\"\"\}\\\]\},frame=\{level=\"2\",args=\\\[\{name=\"intarg\",value=\"2\"\},\{name=\"strarg\",value=\"$hex \\\\\"A string argument.\\\\\"\"\}\\\]\},frame=\{level=\"3\",args=\\\[\{name=\"intarg\",value=\"2\"\},\{name=\"strarg\",value=\"$hex \\\\\"A string argument.\\\\\"\"\},\{name=\"fltarg\",value=\"3.5\"\}\\\]\}\\\]" \
 > +                "stack args listing 1 1 300"
 >  }
 >  
 >  proc test_stack_info_depth {} {


-- 
Nick                                           http://www.inet.net.nz/~nickrob

Re: -stack-list-frames HIGH_FRAME changes

[Vladimir Prus]

On Friday 08 September 2006 12:27, Nick Roberts wrote:
>  > Here we go, OK?
>
> Looks fine to me apart from:
>  > +
>  > +    mi_gdb_test "235-stack-list-arguments 1 1 3" \
>
>                                                  ^^^ (should be 300?)

Yes, you're right.

- Volodya

Re: -stack-list-frames HIGH_FRAME changes

[Daniel Jacobowitz]

On Fri, Sep 08, 2006 at 10:27:50AM +0400, Vladimir Prus wrote:
> Here we go, OK?
> 
> - Volodya
> 
> ./Changelog:
> 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> 
>         * mi/mi-cmd-stack.c (mi_cmd_stack_list_args): Don't emit error
>         if high requested frame number is larger then number of available 
>         frames.
> 
> 
> doc/ChangeLog
> 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> 
>         * gdb.texinfo (GDB/MI Stack Manipulation): Mention that
>         -stack-list-arguments HIGH_FRAME argument can be larger then the
>         actual number of frames.
> 
> 
> testsuite/ChangeLog
> 
> 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> 
>         * gdb.mi/mi-stack.exp (test_stack_args_listing): Test that
>         HIGH_FRAME argument to -stack-list-arguments can be larger than
>         the number of frames.

With Nick's correction, and assuming the documentation change is OK
with Eli, this is approved.  Thanks.


-- 
Daniel Jacobowitz
CodeSourcery

Re: -stack-list-frames HIGH_FRAME changes

[Eli Zaretskii]

> From: Vladimir Prus <ghost@cs.msu.su>
> Date: Fri, 8 Sep 2006 10:27:50 +0400
> Cc: gdb-patches@sources.redhat.com
> 
> ./Changelog:
> 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> 
>         * mi/mi-cmd-stack.c (mi_cmd_stack_list_args): Don't emit error
>         if high requested frame number is larger then number of available 
>         frames.                                  ^^^^
> 
> 
> doc/ChangeLog
> 2006-05-05  Vladimir Prus  <ghost@lvk.cs.msu.su>
> 
>         * gdb.texinfo (GDB/MI Stack Manipulation): Mention that
>         -stack-list-arguments HIGH_FRAME argument can be larger then the
>         actual number of frames.                                ^^^^

You mean "than", not "then".

>  Display a list of the arguments for the frames between @var{low-frame}
>  and @var{high-frame} (inclusive).  If @var{low-frame} and
> -@var{high-frame} are not provided, list the arguments for the whole call
> -stack.
> +@var{high-frame} are not provided, list the arguments for the whole
> +call stack. If the two arguments are equal, it shows the single frame
             ^^
Two spaces, please.

Also, ``it shows'' is not in line with the style of the rest of the
description; ``show'' is better:

  If the two arguments are equal, show the single frame at the
  given level.

> +larger than the actual number of frames.  On the other hand,
> +@var{high-frame} may be larger then the actual number of frames, in
                                  ^^^^
"than".

With these fixes, you have green light to commit the patch.  Thanks!