[parisc-linux] [PATCH] space register 0 changes for hppa

[parisc-linux] [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

Hi.  I believe I should still have a copyright assignment on file.

Index: opcodes/ChangeLog
===================================================================
RCS file: /cvs/src/src/opcodes/ChangeLog,v
retrieving revision 1.547
diff -u -p -r1.547 ChangeLog
--- opcodes/ChangeLog	12 Feb 2003 22:33:59 -0000	1.547
+++ opcodes/ChangeLog	23 Feb 2003 18:54:00 -0000
@@ -1,3 +1,9 @@
+2003-02-23  Matthew Wilcox  <willy@debian.org>
+
+	* hppa-dis.c: Do not emit a space register if the space register
+	field is 2 bits and the value is 0 as this is implicit addressing,
+	not use of space register 0.
+
 2003-02-12  Dave Brolley  <brolley@redhat.com>
 
 	* fr30-desc.c: Regenerate.
Index: opcodes/hppa-dis.c
===================================================================
RCS file: /cvs/src/src/opcodes/hppa-dis.c,v
retrieving revision 1.41
diff -u -p -r1.41 hppa-dis.c
--- opcodes/hppa-dis.c	31 Dec 2001 23:44:08 -0000	1.41
+++ opcodes/hppa-dis.c	23 Feb 2003 18:54:00 -0000
@@ -541,8 +541,11 @@ print_insn_hppa (memaddr, info)
 		  fput_const (extract_5_load (insn), info);
 		  break;
 		case 's':
-		  (*info->fprintf_func) (info->stream,
-					 "sr%d", GET_FIELD (insn, 16, 17));
+		  {
+		    int space = GET_FIELD (insn, 16, 17);
+		    if (space != 0)
+		      (*info->fprintf_func) (info->stream, "sr%d", space);
+		  }
 		  break;
 
 		case 'S':

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Alan Modra]

On Sun, Feb 23, 2003 at 06:58:47PM +0000, Matthew Wilcox wrote:
> +	* hppa-dis.c: Do not emit a space register if the space register
> +	field is 2 bits and the value is 0 as this is implicit addressing,
> +	not use of space register 0.

Hi Matthew,
  ChangeLog entries shouldn't be sent as diffs.  Write this one as

	* hppa-dis.c (print_insn_hppa <2 bit space register>): Do not print
	the space register when the value is zero.

and put the reason for the change as a comment in the code if you feel
you need some justification for the change.

Did you run the binutils testsuite?

-- 
Alan Modra
IBM OzLabs - Linux Technology Centre

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 04:54:21PM +1030, Alan Modra wrote:
>   ChangeLog entries shouldn't be sent as diffs.  Write this one as
> 
> 	* hppa-dis.c (print_insn_hppa <2 bit space register>): Do not print
> 	the space register when the value is zero.
> 
> and put the reason for the change as a comment in the code if you feel
> you need some justification for the change.

OK.  I'm just used to changelogs which're a little more verbose.  I
don't think it really needs justifying.

> Did you run the binutils testsuite?

I hadn't... but I have now, and it doesn't introduce any failures.

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 04:54:21PM +1030, Alan Modra wrote:
>   ChangeLog entries shouldn't be sent as diffs.  Write this one as
> 
> 	* hppa-dis.c (print_insn_hppa <2 bit space register>): Do not print
> 	the space register when the value is zero.
> 
> and put the reason for the change as a comment in the code if you feel
> you need some justification for the change.

OK.  I'm just used to changelogs which're a little more verbose.  I
don't think it really needs justifying.

> Did you run the binutils testsuite?

I hadn't... but I have now, and it doesn't introduce any failures.

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[law]

In message <20030224062421.GA31111@bubble.sa.bigpond.net.au>, Alan Modra writes
:
 >Hi Matthew,
 >  ChangeLog entries shouldn't be sent as diffs.  Write this one as
 >
 >	* hppa-dis.c (print_insn_hppa <2 bit space register>): Do not print
 >	the space register when the value is zero.
 >
 >and put the reason for the change as a comment in the code if you feel
 >you need some justification for the change.
 >
 >Did you run the binutils testsuite?
And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
it's going to fail after this change.

jeff

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 08:01:31AM -0700, law@redhat.com wrote:
> And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
> it's going to fail after this change.

Is it OK if I fix the testsuite then?

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[law]

In message <20030224145957.K2938@parcelfarce.linux.theplanet.co.uk>, Matthew Wi
lcox writes:
 >On Mon, Feb 24, 2003 at 08:01:31AM -0700, law@redhat.com wrote:
 >> And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
 >> it's going to fail after this change.
 >
 >Is it OK if I fix the testsuite then?
No strong opinions either way.

jeff

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[law]

In message <20030224145957.K2938@parcelfarce.linux.theplanet.co.uk>, Matthew Wi
lcox writes:
 >On Mon, Feb 24, 2003 at 08:01:31AM -0700, law@redhat.com wrote:
 >> And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
 >> it's going to fail after this change.
 >
 >Is it OK if I fix the testsuite then?
No strong opinions either way.

jeff

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 08:01:31AM -0700, law@redhat.com wrote:
> And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
> it's going to fail after this change.

Is it OK if I fix the testsuite then?

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 08:01:31AM -0700, law@redhat.com wrote:
> And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
> it's going to fail after this change.

i would, if i could.  every attempt i made to get gdb's testsuite to
run met with disaster.  this was supposed to be a quick bugfix not a
major expedition.  i give up.  someone who can cope with the ridiculous
requirements for building gdb targetting hppa can fix this.

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[law]

In message <20030224182328.M2938@parcelfarce.linux.theplanet.co.uk>, Matthew Wi
lcox writes:
 >On Mon, Feb 24, 2003 at 08:01:31AM -0700, law@redhat.com wrote:
 >> And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
 >> it's going to fail after this change.
 >
 >i would, if i could.  every attempt i made to get gdb's testsuite to
 >run met with disaster.  this was supposed to be a quick bugfix not a
 >major expedition.  i give up.  someone who can cope with the ridiculous
 >requirements for building gdb targetting hppa can fix this.
Err, I built it without difficulty a few months ago (I don't do nearly as
much PA work as I used to).  Presumably you have an HPUX box, right?  That's
the easiest way to get it build and running.

Jeff

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 11:39:52AM -0700, law@redhat.com wrote:
> Err, I built it without difficulty a few months ago (I don't do nearly as
> much PA work as I used to).  Presumably you have an HPUX box, right?  That's
> the easiest way to get it build and running.

nope, no hpux, only i386-linux and hppa-linux around here.  trying to
get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
directory), trying with just `make all-binutils all-gdb' dies being
unable to find dl.h or machine/save_state.h.

so switch to trying native on an hppa-linux host with gdb-5.3.  that dies
claiming hppa-linux is unsupported.  apply the debian patches.  that build
dies trying to find texc.pro.

i'm willing to put in a certain amount of time & effort, but this is
too much.

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 11:39:52AM -0700, law@redhat.com wrote:
> Err, I built it without difficulty a few months ago (I don't do nearly as
> much PA work as I used to).  Presumably you have an HPUX box, right?  That's
> the easiest way to get it build and running.

nope, no hpux, only i386-linux and hppa-linux around here.  trying to
get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
directory), trying with just `make all-binutils all-gdb' dies being
unable to find dl.h or machine/save_state.h.

so switch to trying native on an hppa-linux host with gdb-5.3.  that dies
claiming hppa-linux is unsupported.  apply the debian patches.  that build
dies trying to find texc.pro.

i'm willing to put in a certain amount of time & effort, but this is
too much.

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Daniel Jacobowitz]

On Mon, Feb 24, 2003 at 06:42:25PM +0000, Matthew Wilcox wrote:
> On Mon, Feb 24, 2003 at 11:39:52AM -0700, law@redhat.com wrote:
> > Err, I built it without difficulty a few months ago (I don't do nearly as
> > much PA work as I used to).  Presumably you have an HPUX box, right?  That's
> > the easiest way to get it build and running.
> 
> nope, no hpux, only i386-linux and hppa-linux around here.  trying to
> get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
> directory), trying with just `make all-binutils all-gdb' dies being
> unable to find dl.h or machine/save_state.h.
> 
> so switch to trying native on an hppa-linux host with gdb-5.3.  that dies
> claiming hppa-linux is unsupported.  apply the debian patches.  that build
> dies trying to find texc.pro.
> 
> i'm willing to put in a certain amount of time & effort, but this is
> too much.

That means the version of texinfo install on that machine is broken. 
It was fixed in Debian/unstable some months ago.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Daniel Jacobowitz]

On Mon, Feb 24, 2003 at 06:42:25PM +0000, Matthew Wilcox wrote:
> On Mon, Feb 24, 2003 at 11:39:52AM -0700, law@redhat.com wrote:
> > Err, I built it without difficulty a few months ago (I don't do nearly as
> > much PA work as I used to).  Presumably you have an HPUX box, right?  That's
> > the easiest way to get it build and running.
> 
> nope, no hpux, only i386-linux and hppa-linux around here.  trying to
> get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
> directory), trying with just `make all-binutils all-gdb' dies being
> unable to find dl.h or machine/save_state.h.
> 
> so switch to trying native on an hppa-linux host with gdb-5.3.  that dies
> claiming hppa-linux is unsupported.  apply the debian patches.  that build
> dies trying to find texc.pro.
> 
> i'm willing to put in a certain amount of time & effort, but this is
> too much.

That means the version of texinfo install on that machine is broken. 
It was fixed in Debian/unstable some months ago.

-- 
Daniel Jacobowitz
MontaVista Software                         Debian GNU/Linux Developer

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Andrew Cagney]

> On Mon, Feb 24, 2003 at 11:39:52AM -0700, law@redhat.com wrote:
> 
>> Err, I built it without difficulty a few months ago (I don't do nearly as
>> much PA work as I used to).  Presumably you have an HPUX box, right?  That's
>> the easiest way to get it build and running.
> 
> 
> nope, no hpux, only i386-linux and hppa-linux around here.  trying to
> get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
> directory), trying with just `make all-binutils all-gdb' dies being
> unable to find dl.h or machine/save_state.h.

Unfortunatly, no one got the i386->hppa GDB to build so asking you to do 
that isn't reasonable.  If you've a hppa-linux-gnu box, can you just try 
binutils, `objdump --disassemble`?

Andrew

> so switch to trying native on an hppa-linux host with gdb-5.3.  that dies
> claiming hppa-linux is unsupported.  apply the debian patches.  that build
> dies trying to find texc.pro.
> 
> i'm willing to put in a certain amount of time & effort, but this is
> too much.
> 
> -- "It's not Hollywood. War is real, war is primarily not about defeat or victory, it is about death. I've seen thousands and thousands of dead bodies. Do you think I want to have an academic debate on this subject?" -- Robert Fisk 

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Andrew Cagney]

> On Mon, Feb 24, 2003 at 11:39:52AM -0700, law@redhat.com wrote:
> 
>> Err, I built it without difficulty a few months ago (I don't do nearly as
>> much PA work as I used to).  Presumably you have an HPUX box, right?  That's
>> the easiest way to get it build and running.
> 
> 
> nope, no hpux, only i386-linux and hppa-linux around here.  trying to
> get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
> directory), trying with just `make all-binutils all-gdb' dies being
> unable to find dl.h or machine/save_state.h.

Unfortunatly, no one got the i386->hppa GDB to build so asking you to do 
that isn't reasonable.  If you've a hppa-linux-gnu box, can you just try 
binutils, `objdump --disassemble`?

Andrew

> so switch to trying native on an hppa-linux host with gdb-5.3.  that dies
> claiming hppa-linux is unsupported.  apply the debian patches.  that build
> dies trying to find texc.pro.
> 
> i'm willing to put in a certain amount of time & effort, but this is
> too much.
> 
> -- "It's not Hollywood. War is real, war is primarily not about defeat or victory, it is about death. I've seen thousands and thousands of dead bodies. Do you think I want to have an academic debate on this subject?" -- Robert Fisk 

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 04:49:35PM -0500, Andrew Cagney wrote:
> > get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
> > directory), trying with just `make all-binutils all-gdb' dies being
> > unable to find dl.h or machine/save_state.h.
> 
> Unfortunatly, no one got the i386->hppa GDB to build so asking you to do 
> that isn't reasonable.  If you've a hppa-linux-gnu box, can you just try 
> binutils, `objdump --disassemble`?

Yep, I can build binutils head-of-cvs just fine as a crosscompiler,
and objdump --disassemble does the right thing.  Jeff's concern is that
this intended output change might break the gdb testsuite.

fwiw, it doesn't break the binutils `make check' ;-)

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Andrew Cagney]

> On Mon, Feb 24, 2003 at 04:49:35PM -0500, Andrew Cagney wrote:
> 
>> > get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
>> > directory), trying with just `make all-binutils all-gdb' dies being
>> > unable to find dl.h or machine/save_state.h.
> 
>> 
>> Unfortunatly, no one got the i386->hppa GDB to build so asking you to do 
>> that isn't reasonable.  If you've a hppa-linux-gnu box, can you just try 
>> binutils, `objdump --disassemble`?
> 
> 
> Yep, I can build binutils head-of-cvs just fine as a crosscompiler,
> and objdump --disassemble does the right thing.  Jeff's concern is that
> this intended output change might break the gdb testsuite.
> 
> fwiw, it doesn't break the binutils `make check' ;-)

If that happens `oops'.  GDB's HP/PA support isn't maintained.

Andrew

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Andrew Cagney]

> On Mon, Feb 24, 2003 at 04:49:35PM -0500, Andrew Cagney wrote:
> 
>> > get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
>> > directory), trying with just `make all-binutils all-gdb' dies being
>> > unable to find dl.h or machine/save_state.h.
> 
>> 
>> Unfortunatly, no one got the i386->hppa GDB to build so asking you to do 
>> that isn't reasonable.  If you've a hppa-linux-gnu box, can you just try 
>> binutils, `objdump --disassemble`?
> 
> 
> Yep, I can build binutils head-of-cvs just fine as a crosscompiler,
> and objdump --disassemble does the right thing.  Jeff's concern is that
> this intended output change might break the gdb testsuite.
> 
> fwiw, it doesn't break the binutils `make check' ;-)

If that happens `oops'.  GDB's HP/PA support isn't maintained.

Andrew

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 05:11:19PM -0500, Andrew Cagney wrote:
> > Yep, I can build binutils head-of-cvs just fine as a crosscompiler,
> > and objdump --disassemble does the right thing.  Jeff's concern is that
> > this intended output change might break the gdb testsuite.
> 
> If that happens `oops'.  GDB's HP/PA support isn't maintained.

OK.  If at some point it does start to be maintained, I volunteer to
fix up the testsuite.  It looks fairly straightforward -- deleting
the '%sr0,' from these lines:

.*ldw 0\\(sr0,r4\\),r26.*
.*ldh 0\\(sr0,r4\\),r26.*
.*ldb 0\\(sr0,r4\\),r26.*

would be a good start.  not sure which other ones will fail.  all the ld*
and wt*, i'm pretty sure.  maybe a few others.  i could work it out from
the manual, but let's wait for the computer to be able to do it for us?

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Alan Modra]

I've committed the patch.

-- 
Alan Modra
IBM OzLabs - Linux Technology Centre

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Alan Modra]

I've committed the patch.

-- 
Alan Modra
IBM OzLabs - Linux Technology Centre

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 05:11:19PM -0500, Andrew Cagney wrote:
> > Yep, I can build binutils head-of-cvs just fine as a crosscompiler,
> > and objdump --disassemble does the right thing.  Jeff's concern is that
> > this intended output change might break the gdb testsuite.
> 
> If that happens `oops'.  GDB's HP/PA support isn't maintained.

OK.  If at some point it does start to be maintained, I volunteer to
fix up the testsuite.  It looks fairly straightforward -- deleting
the '%sr0,' from these lines:

.*ldw 0\\(sr0,r4\\),r26.*
.*ldh 0\\(sr0,r4\\),r26.*
.*ldb 0\\(sr0,r4\\),r26.*

would be a good start.  not sure which other ones will fail.  all the ld*
and wt*, i'm pretty sure.  maybe a few others.  i could work it out from
the manual, but let's wait for the computer to be able to do it for us?

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 04:49:35PM -0500, Andrew Cagney wrote:
> > get binutils cvs to build a cross i386->hppa gdb dies in tix (no tk8.4
> > directory), trying with just `make all-binutils all-gdb' dies being
> > unable to find dl.h or machine/save_state.h.
> 
> Unfortunatly, no one got the i386->hppa GDB to build so asking you to do 
> that isn't reasonable.  If you've a hppa-linux-gnu box, can you just try 
> binutils, `objdump --disassemble`?

Yep, I can build binutils head-of-cvs just fine as a crosscompiler,
and objdump --disassemble does the right thing.  Jeff's concern is that
this intended output change might break the gdb testsuite.

fwiw, it doesn't break the binutils `make check' ;-)

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[law]

In message <20030224182328.M2938@parcelfarce.linux.theplanet.co.uk>, Matthew Wi
lcox writes:
 >On Mon, Feb 24, 2003 at 08:01:31AM -0700, law@redhat.com wrote:
 >> And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
 >> it's going to fail after this change.
 >
 >i would, if i could.  every attempt i made to get gdb's testsuite to
 >run met with disaster.  this was supposed to be a quick bugfix not a
 >major expedition.  i give up.  someone who can cope with the ridiculous
 >requirements for building gdb targetting hppa can fix this.
Err, I built it without difficulty a few months ago (I don't do nearly as
much PA work as I used to).  Presumably you have an HPUX box, right?  That's
the easiest way to get it build and running.

Jeff

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Matthew Wilcox]

On Mon, Feb 24, 2003 at 08:01:31AM -0700, law@redhat.com wrote:
> And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
> it's going to fail after this change.

i would, if i could.  every attempt i made to get gdb's testsuite to
run met with disaster.  this was supposed to be a quick bugfix not a
major expedition.  i give up.  someone who can cope with the ridiculous
requirements for building gdb targetting hppa can fix this.

-- 
"It's not Hollywood.  War is real, war is primarily not about defeat or
victory, it is about death.  I've seen thousands and thousands of dead bodies.
Do you think I want to have an academic debate on this subject?" -- Robert Fisk

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[law]

In message <20030224062421.GA31111@bubble.sa.bigpond.net.au>, Alan Modra writes
:
 >Hi Matthew,
 >  ChangeLog entries shouldn't be sent as diffs.  Write this one as
 >
 >	* hppa-dis.c (print_insn_hppa <2 bit space register>): Do not print
 >	the space register when the value is zero.
 >
 >and put the reason for the change as a comment in the code if you feel
 >you need some justification for the change.
 >
 >Did you run the binutils testsuite?
And the GDB testsuite, which tests the PA disassembler.  I'm 99% sure
it's going to fail after this change.

jeff

[parisc-linux] Re: [PATCH] space register 0 changes for hppa

[Alan Modra]

On Sun, Feb 23, 2003 at 06:58:47PM +0000, Matthew Wilcox wrote:
> +	* hppa-dis.c: Do not emit a space register if the space register
> +	field is 2 bits and the value is 0 as this is implicit addressing,
> +	not use of space register 0.

Hi Matthew,
  ChangeLog entries shouldn't be sent as diffs.  Write this one as

	* hppa-dis.c (print_insn_hppa <2 bit space register>): Do not print
	the space register when the value is zero.

and put the reason for the change as a comment in the code if you feel
you need some justification for the change.

Did you run the binutils testsuite?

-- 
Alan Modra
IBM OzLabs - Linux Technology Centre