[6.0] PROBLEMS and NEWS

[6.0] PROBLEMS and NEWS

[Andrew Cagney]

[-- Attachment #1: Type: text/plain, Size: 261 bytes --]

Hello,

This patch significantly expands the 6.0 NEWS file.  It also drops a 
strong hint about PPC, MIPS, SPARC and ARM in the problems file.

Comments, objections, complaints?

Taking advantange of the unfreeze, I'll leave this for at least two days.

Andrew

[-- Attachment #2: diffs --]
[-- Type: text/plain, Size: 3766 bytes --]

2003-09-20  Andrew Cagney  <cagney@redhat.com>

	* NEWS: Mention the new backtrace mechanism, DWARF 2 CFI, hosted
	file I/O, multi-arch, TLS and NPTL, and Location Expressions.
	* PROBLEMS: Mention that mips*-*-*, powerpc*-*-*, sparc*-*-* and
	arm*-*-* do not use the new frame code.

Index: NEWS
===================================================================
RCS file: /cvs/src/src/gdb/NEWS,v
retrieving revision 1.120
diff -u -r1.120 NEWS
--- NEWS	11 Sep 2003 14:00:54 -0000	1.120
+++ NEWS	20 Sep 2003 20:55:43 -0000
@@ -16,11 +16,60 @@
 
 *** Changes in GDB 6.0:
 
+* New back-trace mechanism (includes DWARF 2 Call Frame Identification).
+
+DWARF 2's Call Frame Identification makes available compiler generated
+information that more exactly describes the program's run-time stack.
+By using this information, GDB is able to provide more robust stack
+back-traces.
+
+The i386, amd64 (nee, x86-64), Alpha, m68hc11, ia64, and m32r targets
+have been updated to use a new back-trace mechanism which includes
+DWARF 2 CFI support.
+
+* Hosted file I/O.
+
+GDB's remote protocol has been extended to include support for hosted
+file I/O (where the remote target uses GDB's file system).  See GDB's
+remote protocol documentation for details.
+
+* All targets using the new architecture framework.
+
+All of GDB's targets have been updated to use the new internal
+architecture framework.  The way is now open for future GDB releases
+to include cross-architecture native debugging support (i386 on amd64,
+ppc32 on ppc64).
+
+* GNU/Linux's Thread Local Storage (TLS)
+
+GDB now includes support for for the GNU/Linux implementation of
+per-thread variables.
+
+* GNU/Linux's Native POSIX Thread Library (NPTL)
+
+GDB's thread code has been updated to work with either the new
+GNU/Linux NPTL thread library or the older "LinuxThreads" library.
+
+* Separate debug info.
+
+GDB, in conjunction with BINUTILS, now supports a mechanism for
+automatically loading debug information from a separate file.  Instead
+of shipping full debug and non-debug versions of system libraries,
+system integrators can now instead ship just the stripped libraries
+and optional debug files.
+
 * GNU/Linux support for fork, vfork, and exec.
 
 The "catch fork", "catch exec", "catch vfork", and "set follow-fork-mode"
 commands are now implemented for GNU/Linux.  They require a 2.5.x or later
 kernel.
+
+* DWARF 2 Location Expressions
+
+DWARF 2 Location Expressions allow the compiler to more exactly
+describe the location of variables to the debugger.  By using this
+information, GDB is able to more robustly display the value of
+variables (even in optomized code).
 
 * GDB supports logging output to a file
 
Index: PROBLEMS
===================================================================
RCS file: /cvs/src/src/gdb/PROBLEMS,v
retrieving revision 1.18
diff -u -r1.18 PROBLEMS
--- PROBLEMS	26 Aug 2003 03:08:32 -0000	1.18
+++ PROBLEMS	20 Sep 2003 20:55:43 -0000
@@ -3,6 +3,26 @@
 
 		See also: http://www.gnu.org/software/gdb/bugs/
 
+mips*-*-*
+powerpc*-*-*
+sparc*-*-*
+
+GDB's SPARC, MIPS and PowerPC targets, in 6.0, have not been updated
+to use the new frame mechanism.
+
+People encountering problems with these targets should consult GDB's
+web pages and mailing lists (http://www.gnu.org/software/gdb/) to see
+if there is an update.
+
+arm-*-*
+
+GDB's ARM target, in 6.0, has not been updated to use the new frame
+mechanism.
+
+Fortunatly the ARM target, in the GDB's mainline sources, has been
+updated so people encountering problems should consider downloading a
+more current GDB (http://www.gnu.org/software/gdb/current).
+
 gdb/1091: Constructor breakpoints ignored
 gdb/1193: g++ 3.3 creates multiple constructors: gdb 5.3 can't set breakpoints
 

Re: [6.0] PROBLEMS and NEWS

[Daniel Jacobowitz]

On Sat, Sep 20, 2003 at 04:59:17PM -0400, Andrew Cagney wrote:
> Hello,
> 
> This patch significantly expands the 6.0 NEWS file.  It also drops a 
> strong hint about PPC, MIPS, SPARC and ARM in the problems file.
> 
> Comments, objections, complaints?

Looks good to me.  Are PPC, MIPS, and SPARC the only ones left on
mainline?  I know Mark's got a SPARC update in the works, too.

> +variables (even in optomized code).

"optimized".

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer

Re: [6.0] PROBLEMS and NEWS

[Andrew Cagney]

> On Sat, Sep 20, 2003 at 04:59:17PM -0400, Andrew Cagney wrote:
> 
>> Hello,
>> 
>> This patch significantly expands the 6.0 NEWS file.  It also drops a 
>> strong hint about PPC, MIPS, SPARC and ARM in the problems file.
>> 
>> Comments, objections, complaints?
> 
> 
> Looks good to me.  Are PPC, MIPS, and SPARC the only ones left on
> mainline?  I know Mark's got a SPARC update in the works, too.

I didn't what to say anything that could put Mark on the spot (which is 
why I mentioned the Arm as it's up-to-date).

I picked those as I know they are going to get used.  That can't, for 
instance, be said of the vax :-)

>> +variables (even in optomized code).
> 
> 
> "optimized".

And I even spell checked it, just not enough.  Doh!

Andrew

Re: [6.0] PROBLEMS and NEWS

[Andrew Cagney]

> someone>   A long-standing bug involving Java and dynamic libraries has been
> someone>   fixed.  Support for Java debugging remains minimal, however.
> 
> ac> Do you have something more specific?  That could mean anything.
> 
> The "long-standing bug" was gdb/1322.
> 
>   http://sources.redhat.com/gdb/bugs/1322
>   "internal-error: sect_index_text not initialized" after printing a java type

Thanks, I'll add:

* Java

A long standing bug involving Java and dynamic libraries has been fixed 
(Ref: gdb/1322: "internal-error: sect_index_text not initialized" after 
printing a java type).

Hmm, did the equally long standing print java string bug also get fixed?

> I don't know what "minimal" means, specifically.

I'm told that It works really well at backtraces, but starts to struggle 
when printing variables, values, and objects, and really struggles with 
dynamic compiled/loaded shared libraries.  As for byte codes :-)

Andrew

Re: [6.0] PROBLEMS and NEWS

[Andrew Cagney]

[-- Attachment #1: Type: text/plain, Size: 145 bytes --]

Here's an update.  Given a recent thread, I watered down location 
expressions - there is a very large chunk of work still out there :-(

Andrew

[-- Attachment #2: diffs --]
[-- Type: text/plain, Size: 3629 bytes --]

2003-09-20  Andrew Cagney  <cagney@redhat.com>

	* NEWS: Mention the new backtrace mechanism, DWARF 2 CFI, hosted
	file I/O, multi-arch, TLS and NPTL, and Location Expressions.
	* PROBLEMS: Mention that mips*-*-*, powerpc*-*-*, sparc*-*-* and
	arm*-*-* do not use the new frame code.

Index: NEWS
===================================================================
RCS file: /cvs/src/src/gdb/NEWS,v
retrieving revision 1.120
diff -u -r1.120 NEWS
--- NEWS	11 Sep 2003 14:00:54 -0000	1.120
+++ NEWS	22 Sep 2003 19:09:26 -0000
@@ -16,6 +16,56 @@
 
 *** Changes in GDB 6.0:
 
+* New back-trace mechanism (includes DWARF 2 Call Frame Identification).
+
+DWARF 2's Call Frame Identification makes available compiler generated
+information that more exactly describes the program's run-time stack.
+By using this information, GDB is able to provide more robust stack
+back-traces.
+
+The i386, amd64 (nee, x86-64), Alpha, m68hc11, ia64, and m32r targets
+have been updated to use a new back-trace mechanism which includes
+DWARF 2 CFI support.
+
+* Hosted file I/O.
+
+GDB's remote protocol has been extended to include support for hosted
+file I/O (where the remote target uses GDB's file system).  See GDB's
+remote protocol documentation for details.
+
+* All targets using the new architecture framework.
+
+All of GDB's targets have been updated to use the new internal
+architecture framework.  The way is now open for future GDB releases
+to include cross-architecture native debugging support (i386 on amd64,
+ppc32 on ppc64).
+
+* GNU/Linux's Thread Local Storage (TLS)
+
+GDB now includes support for for the GNU/Linux implementation of
+per-thread variables.
+
+* GNU/Linux's Native POSIX Thread Library (NPTL)
+
+GDB's thread code has been updated to work with either the new
+GNU/Linux NPTL thread library or the older "LinuxThreads" library.
+
+* Separate debug info.
+
+GDB, in conjunction with BINUTILS, now supports a mechanism for
+automatically loading debug information from a separate file.  Instead
+of shipping full debug and non-debug versions of system libraries,
+system integrators can now instead ship just the stripped libraries
+and optional debug files.
+
+* DWARF 2 Location Expressions
+
+DWARF 2 Location Expressions allow the compiler to more exactly
+describe the location of variables to the debugger.
+
+GDB now includes preliminary support for location expressions (support
+for DW_OP_piece is still missing).
+
 * GNU/Linux support for fork, vfork, and exec.
 
 The "catch fork", "catch exec", "catch vfork", and "set follow-fork-mode"
Index: PROBLEMS
===================================================================
RCS file: /cvs/src/src/gdb/PROBLEMS,v
retrieving revision 1.18
diff -u -r1.18 PROBLEMS
--- PROBLEMS	26 Aug 2003 03:08:32 -0000	1.18
+++ PROBLEMS	22 Sep 2003 19:09:26 -0000
@@ -3,6 +3,26 @@
 
 		See also: http://www.gnu.org/software/gdb/bugs/
 
+mips*-*-*
+powerpc*-*-*
+sparc*-*-*
+
+GDB's SPARC, MIPS and PowerPC targets, in 6.0, have not been updated
+to use the new frame mechanism.
+
+People encountering problems with these targets should consult GDB's
+web pages and mailing lists (http://www.gnu.org/software/gdb/) to see
+if there is an update.
+
+arm-*-*
+
+GDB's ARM target, in 6.0, has not been updated to use the new frame
+mechanism.
+
+Fortunatly the ARM target, in the GDB's mainline sources, has been
+updated so people encountering problems should consider downloading a
+more current GDB (http://www.gnu.org/software/gdb/current).
+
 gdb/1091: Constructor breakpoints ignored
 gdb/1193: g++ 3.3 creates multiple constructors: gdb 5.3 can't set breakpoints
 

Re: [6.0] PROBLEMS and NEWS

[Jim Blandy]

Looks good, just some picky comments:

Andrew Cagney <ac131313@redhat.com> writes:
> +* New back-trace mechanism (includes DWARF 2 Call Frame Identification).
> +
> +DWARF 2's Call Frame Identification makes available compiler generated

s/Identification/Information/; that's what the spec calls it.

> +information that more exactly describes the program's run-time stack.
> +By using this information, GDB is able to provide more robust stack
> +back-traces.

The GDB manual uses "backtrace" throughout, never "back-trace".  There
are other uses of "back-trace" in the proposed text.

> +* DWARF 2 Location Expressions
> +
> +DWARF 2 Location Expressions allow the compiler to more exactly
> +describe the location of variables to the debugger.

It'd be nice to give some indication of why the user would care about
this, like: "Taken together with Dwarf 2 Call Frame Information,
location expressions give GDB the information it needs to debug
optimized code much more effectively."

Re: [6.0] PROBLEMS and NEWS

[Andrew Cagney]

> Looks good, just some picky comments:
> 
> Andrew Cagney <ac131313@redhat.com> writes:
> 
>> +* New back-trace mechanism (includes DWARF 2 Call Frame Identification).
>> +
>> +DWARF 2's Call Frame Identification makes available compiler generated
> 
> 
> s/Identification/Information/; that's what the spec calls it.

Oops.

>> +information that more exactly describes the program's run-time stack.
>> +By using this information, GDB is able to provide more robust stack
>> +back-traces.
> 
> 
> The GDB manual uses "backtrace" throughout, never "back-trace".  There
> are other uses of "back-trace" in the proposed text.

That will learn me for spellchecking it ...

>> +* DWARF 2 Location Expressions
>> +
>> +DWARF 2 Location Expressions allow the compiler to more exactly
>> +describe the location of variables to the debugger.
> 
> 
> It'd be nice to give some indication of why the user would care about
> this, like: "Taken together with Dwarf 2 Call Frame Information,
> location expressions give GDB the information it needs to debug
> optimized code much more effectively."

At one stage I had the word optimized, but took it out.  I've changed it to:

DWARF 2 Location Expressions allow the compiler to more completly
describe the location of variables (even in optimized code) to the
debugger.

IA64 ABI, for instance,  specifies something other than CFI for doing 
unwinds so the LE <-> CFI link isn't that strong.

Andrew

Re: [6.0] PROBLEMS and NEWS

[Jim Blandy]

Andrew Cagney <ac131313@redhat.com> writes:
> >> +DWARF 2 Location Expressions allow the compiler to more exactly
> >> +describe the location of variables to the debugger.
> > It'd be nice to give some indication of why the user would care about
> > this, like: "Taken together with Dwarf 2 Call Frame Information,
> > location expressions give GDB the information it needs to debug
> > optimized code much more effectively."
> 
> At one stage I had the word optimized, but took it out.  I've changed it to:
> 
> DWARF 2 Location Expressions allow the compiler to more completly
> describe the location of variables (even in optimized code) to the
> debugger.

You're right --- it's always better to describe things that way.  How
hard the compiler worked to generate the code isn't the point: how
accurately the compiler describes whatever it produces is.

I'd say "more accurately" instead of "more completely", since the most
common manifestation of the old problem was that GDB was wrong about
the location of the variable, not that it couldn't find the variable.

Re: [6.0] PROBLEMS and NEWS

[Andrew Cagney]

[-- Attachment #1: Type: text/plain, Size: 38 bytes --]

This is what I've checked in.

Andrew

[-- Attachment #2: diffs --]
[-- Type: text/plain, Size: 3833 bytes --]

2003-09-25  Andrew Cagney  <cagney@redhat.com>

	* NEWS: Mention the new backtrace mechanism, DWARF 2 CFI, hosted
	file I/O, multi-arch, TLS and NPTL, DWARF 2 Location Expressions,
	and Java.
	* PROBLEMS: Mention that mips*-*-*, powerpc*-*-*, sparc*-*-* and
	arm*-*-* do not use the new frame code.

Index: NEWS
===================================================================
RCS file: /cvs/src/src/gdb/NEWS,v
retrieving revision 1.120
diff -u -r1.120 NEWS
--- NEWS	11 Sep 2003 14:00:54 -0000	1.120
+++ NEWS	25 Sep 2003 18:22:53 -0000
@@ -16,6 +16,63 @@
 
 *** Changes in GDB 6.0:
 
+* New backtrace mechanism (includes DWARF 2 Call Frame Information).
+
+DWARF 2's Call Frame Information makes available compiler generated
+information that more exactly describes the program's run-time stack.
+By using this information, GDB is able to provide more robust stack
+backtraces.
+
+The i386, amd64 (nee, x86-64), Alpha, m68hc11, ia64, and m32r targets
+have been updated to use a new backtrace mechanism which includes
+DWARF 2 CFI support.
+
+* Hosted file I/O.
+
+GDB's remote protocol has been extended to include support for hosted
+file I/O (where the remote target uses GDB's file system).  See GDB's
+remote protocol documentation for details.
+
+* All targets using the new architecture framework.
+
+All of GDB's targets have been updated to use the new internal
+architecture framework.  The way is now open for future GDB releases
+to include cross-architecture native debugging support (i386 on amd64,
+ppc32 on ppc64).
+
+* GNU/Linux's Thread Local Storage (TLS)
+
+GDB now includes support for for the GNU/Linux implementation of
+per-thread variables.
+
+* GNU/Linux's Native POSIX Thread Library (NPTL)
+
+GDB's thread code has been updated to work with either the new
+GNU/Linux NPTL thread library or the older "LinuxThreads" library.
+
+* Separate debug info.
+
+GDB, in conjunction with BINUTILS, now supports a mechanism for
+automatically loading debug information from a separate file.  Instead
+of shipping full debug and non-debug versions of system libraries,
+system integrators can now instead ship just the stripped libraries
+and optional debug files.
+
+* DWARF 2 Location Expressions
+
+DWARF 2 Location Expressions allow the compiler to more completely
+describe the location of variables (even in optimized code) to the
+debugger.
+
+GDB now includes preliminary support for location expressions (support
+for DW_OP_piece is still missing).
+
+* Java
+
+A number of long standing bugs that caused GDB to die while starting a
+Java application have been fixed.  GDB's Java support is now
+considered "useable".
+
 * GNU/Linux support for fork, vfork, and exec.
 
 The "catch fork", "catch exec", "catch vfork", and "set follow-fork-mode"
Index: PROBLEMS
===================================================================
RCS file: /cvs/src/src/gdb/PROBLEMS,v
retrieving revision 1.18
diff -u -r1.18 PROBLEMS
--- PROBLEMS	26 Aug 2003 03:08:32 -0000	1.18
+++ PROBLEMS	25 Sep 2003 18:22:53 -0000
@@ -3,6 +3,26 @@
 
 		See also: http://www.gnu.org/software/gdb/bugs/
 
+mips*-*-*
+powerpc*-*-*
+sparc*-*-*
+
+GDB's SPARC, MIPS and PowerPC targets, in 6.0, have not been updated
+to use the new frame mechanism.
+
+People encountering problems with these targets should consult GDB's
+web pages and mailing lists (http://www.gnu.org/software/gdb/) to see
+if there is an update.
+
+arm-*-*
+
+GDB's ARM target, in 6.0, has not been updated to use the new frame
+mechanism.
+
+Fortunatly the ARM target, in the GDB's mainline sources, has been
+updated so people encountering problems should consider downloading a
+more current GDB (http://www.gnu.org/software/gdb/current).
+
 gdb/1091: Constructor breakpoints ignored
 gdb/1193: g++ 3.3 creates multiple constructors: gdb 5.3 can't set breakpoints
 

Re: [6.0] PROBLEMS and NEWS

[Michael Elizabeth Chastain]

ac> Hmm, did the equally long standing print java string bug also get fixed?

I don't know.

Michael C