* Re: [PATCH] Provide `long double' support for most i386 targets
[not found] ` <200004030845.EAA11820@indy.delorie.com>
@ 2000-04-03 13:05 ` Mark Kettenis
0 siblings, 0 replies; only message in thread
From: Mark Kettenis @ 2000-04-03 13:05 UTC (permalink / raw)
To: eliz; +Cc: gdb-patches
Date: Mon, 3 Apr 2000 04:45:36 -0400 (EDT)
From: Eli Zaretskii <eliz@delorie.com>
I really like it if code is being removed :-).
> The LD_I387 definition can probably also be removed. Its
> only purpose would be the HEX_LONG_DOUBLE_INPUT macro, but that macro
> is never used in the GDB sources.
Hmm, this HEX_LONG_DOUBLE_INPUT seems like a useful feature, but it
doesn't work, including in GDB 4.18. Did it ever work?
I removed the macro.
I think such a feature is in HJ's version Linux-only version of gdb
4.17. The implementation there is a bit kludgey, and AFAIK never
submitted as an isolated feature. Moreover I don't see why this
should be i386 and/or Linux only. Probably a wish-list item.
Mark
From jtc@redback.com Mon Apr 03 14:35:00 2000
From: jtc@redback.com (J.T. Conklin)
To: gdb-patches@sourceware.cygnus.com
Subject: FYI: patched i386/xm-nbsd.h
Date: Mon, 03 Apr 2000 14:35:00 -0000
Message-id: <5mem8msw9v.fsf@jtc.redback.com>
X-SW-Source: 2000-04/msg00043.html
Content-length: 693
As suggested by Mark Kettenis, I've checked in the enclosed patch
--jtc
2000-04-03 J.T. Conklin <jtc@redback.com>
* config/i386/xm-nbsd.h (HOST_LONG_DOUBLE_FORMAT): Define.
Index: config/i386/xm-nbsd.h
===================================================================
RCS file: /cvs/src/src/gdb/config/i386/xm-nbsd.h,v
retrieving revision 1.1.1.2
diff -c -3 -p -r1.1.1.2 xm-nbsd.h
*** xm-nbsd.h 1999/07/07 20:13:42 1.1.1.2
--- xm-nbsd.h 2000/04/03 21:30:47
***************
*** 20,22 ****
--- 20,24 ----
/* Get generic NetBSD host definitions. */
#include "xm-nbsd.h"
+
+ #define HOST_LONG_DOUBLE_FORMAT &floatformat_i387_ext
--
J.T. Conklin
RedBack Networks
From jtc@redback.com Mon Apr 03 14:50:00 2000
From: jtc@redback.com (J.T. Conklin)
To: Andrew Cagney <ac131313@cygnus.com>
Cc: Jimmy Guo <guo@cup.hp.com>, gdb-patches@sourceware.cygnus.com
Subject: Re: regression analysis: gdb.sum postprocessing script
Date: Mon, 03 Apr 2000 14:50:00 -0000
Message-id: <5maejasvl6.fsf@jtc.redback.com>
References: <Pine.LNX.4.10.10003241103440.15236-100000@hpcll168.cup.hp.com> <38DEFEB1.EB573B72@cygnus.com>
X-SW-Source: 2000-04/msg00044.html
Content-length: 1319
>>>>> "Andrew" == Andrew Cagney <ac131313@cygnus.com> writes:
Jimmy> Here is a Perl script I wrote to postprocess one or more
Jimmy> gdb.sum files and provide a complete dejagnu test outcome
Jimmy> summary (or diff if more than one gdb.sum files are given).
Jimmy> It's most useful if you want to have a complete view of
Jimmy> regression status between two source trees and/or if you set up
Jimmy> your testing to run multiple passes (to see how a test point
Jimmy> behaves in different passes), however you can use it just on
Jimmy> one test run's gdb.sum file and play around with the options to
Jimmy> get test outcome presented in a much organized and useful way
Jimmy> than going through the flat gdb.sum file.
Andrew> I have a personal preference for keeping perl scripts out of
Andrew> the main GDB sources :-( Now if someone rewrote it to use
Andrew> either TCL, DEJAGNU or SH then I'd have to think about it
Andrew> again - they are required to run the testsuite.
Jimmy,
Is your script specific to the GDB testsuite or would it work with any
dejagnu suite? I skimmed through the sources, and didn't see anything
which obviously ties it to the GDB suite.
If the script is generic, one could argue that it belongs in the deja-
gnu distribution itself.
--jtc
--
J.T. Conklin
RedBack Networks
From kevinb@cygnus.com Mon Apr 03 15:13:00 2000
From: Kevin Buettner <kevinb@cygnus.com>
To: gdb-patches@sourceware.cygnus.com
Subject: [PATCH] Mention GNU/Linux PowerPC port in NEWS
Date: Mon, 03 Apr 2000 15:13:00 -0000
Message-id: <1000403221318.ZM28000@ocotillo.lan>
X-SW-Source: 2000-04/msg00045.html
Content-length: 437
I just committed the following change:
* NEWS (powerpc-*-linux*): Mention.
Index: NEWS
===================================================================
RCS file: /cvs/src/src/gdb/NEWS,v
retrieving revision 1.2
diff -u -r1.2 NEWS
--- NEWS 2000/04/03 13:24:08 1.2
+++ NEWS 2000/04/03 22:07:29
@@ -15,6 +15,7 @@
* New native configurations
ARM GNU/Linux arm*-*-linux*
+PowerPC GNU/Linux powerpc-*-linux*
* New targets
From ac131313@cygnus.com Mon Apr 03 17:30:00 2000
From: Andrew Cagney <ac131313@cygnus.com>
To: egcs@cygnus.com
Cc: gdb-patches@sourceware.cygnus.com
Subject: Re: FYI: patched i386/xm-nbsd.h
Date: Mon, 03 Apr 2000 17:30:00 -0000
Message-id: <38E93731.91F14979@cygnus.com>
References: <5mem8msw9v.fsf@jtc.redback.com>
X-SW-Source: 2000-04/msg00046.html
Content-length: 918
"J.T. Conklin" wrote:
>
> As suggested by Mark Kettenis, I've checked in the enclosed patch
>
> --jtc
>
> 2000-04-03 J.T. Conklin <jtc@redback.com>
>
> * config/i386/xm-nbsd.h (HOST_LONG_DOUBLE_FORMAT): Define.
>
> Index: config/i386/xm-nbsd.h
> ===================================================================
> RCS file: /cvs/src/src/gdb/config/i386/xm-nbsd.h,v
> retrieving revision 1.1.1.2
> diff -c -3 -p -r1.1.1.2 xm-nbsd.h
> *** xm-nbsd.h 1999/07/07 20:13:42 1.1.1.2
> --- xm-nbsd.h 2000/04/03 21:30:47
> ***************
> *** 20,22 ****
> --- 20,24 ----
>
> /* Get generic NetBSD host definitions. */
> #include "xm-nbsd.h"
> +
> + #define HOST_LONG_DOUBLE_FORMAT &floatformat_i387_ext
Just an asside on this one. If memory serves correct, I think the long
long term plan was to eliminating xm-*.h and instead use autoconf for
these host side things.
enjoy,
Andrew
From blizzard@mozilla.org Mon Apr 03 18:28:00 2000
From: Christopher Blizzard <blizzard@mozilla.org>
To: gdb-patches@sourceware.cygnus.com
Subject: source listing change
Date: Mon, 03 Apr 2000 18:28:00 -0000
Message-id: <38E94593.668AAF44@mozilla.org>
X-SW-Source: 2000-04/msg00047.html
Content-length: 6459
This was forwarded to me from James Nance. He asked me to forward it on
to the gdb-patches list.
--Chris
-------- Original Message --------
Subject: gdb patches
Date: Tue, 28 Mar 2000 06:26:02 -0500
From: James Lewis Nance <jlnance@worldnet.att.net>
To: blizzard@mozilla.org
Chris,
As long as you are putting together patches for gdb, here is one I
have
been trying to get included for a while. I got it approved by Cygnus
about
2 years ago, but then the person I was working up there quit. I have
not
been able to find anyone really interested in it since then. This patch
adds a command like the "w" command in the Digital Unix ladebug debugger
(though the Cygnus people had me call it "list -l"). What it does is
when you type "list -l" it displays your souce code but it puts a '> '
character in front of the current line so you can find out where you
are. Its one of those things you dont appreciate until you have used
it a few times. Then you cant live with out it.
I am attaching my patch. Please give it a try.
Thanks,
Jim
PS: If you alias list -l to w in your .gdbinit, its much more
convienient to
use.
diff -ru gdb-4.18.orig/gdb/ChangeLog gdb-4.18.new/gdb/ChangeLog
--- gdb-4.18.orig/gdb/ChangeLog Wed Apr 7 17:34:27 1999
+++ gdb-4.18.new/gdb/ChangeLog Wed Jan 12 17:47:10 2000
@@ -1,3 +1,9 @@
+2000-01-12 Jim Nance <jlnance@worldnet.att.net>
+
+ * source.c symtab.h: Added a -w option to the list command. Allows
+ the currently executing line of code in the current stack frame to
+ be annotated.
+
1999-04-07 Jim Blandy <jimb@zwingli.cygnus.com>
* GDB 4.18 released.
Only in gdb-4.18.new/gdb: ChangeLog.orig
Only in gdb-4.18.new/gdb: linuxthreads.c.threads
diff -ru gdb-4.18.orig/gdb/source.c gdb-4.18.new/gdb/source.c
--- gdb-4.18.orig/gdb/source.c Thu Dec 31 16:58:08 1998
+++ gdb-4.18.new/gdb/source.c Wed Jan 12 17:45:48 2000
@@ -976,11 +976,11 @@
/* Print source lines from the file of symtab S,
starting with line number LINE and stopping before line number STOPLINE. */
-static void
-print_source_lines_base (s, line, stopline, noerror)
+void
+print_source_lines_base (s, line, stopline, noerror, markline)
struct symtab *s;
int line, stopline;
- int noerror;
+ int noerror, markline;
{
register int c;
register int desc;
@@ -1046,8 +1046,11 @@
{
c = fgetc (stream);
if (c == EOF) break;
- last_line_listed = current_source_line;
- printf_filtered ("%d\t", current_source_line++);
+ last_line_listed = current_source_line++;
+ if (last_line_listed != markline)
+ printf_filtered ("%d\t", last_line_listed);
+ else
+ printf_filtered ("%d >\t", last_line_listed);
do
{
if (c < 040 && c != '\t' && c != '\n' && c != '\r')
@@ -1081,7 +1084,7 @@
#if defined(TUI)
if (!tui_version ||
m_winPtrIsNull(srcWin) || !srcWin->generic.isVisible )
- print_source_lines_base(s, line, stopline, noerror);
+ print_source_lines_base(s, line, stopline, noerror, 0);
else
{
TuiGenWinInfoPtr locator = locatorWinInfoPtr();
@@ -1101,7 +1104,7 @@
tuiDo((TuiOpaqueFuncPtr)tui_vUpdateLocatorFilename, s->filename);
}
#else
- print_source_lines_base(s, line, stopline, noerror);
+ print_source_lines_base(s, line, stopline, noerror, 0);
#endif
}
\f
@@ -1123,6 +1126,18 @@
sals->sals[i].symtab->filename, sals->sals[i].line);
}
+static void printw (s, line, stopline, noerror)
+ struct symtab *s;
+ int line;
+ int stopline;
+ int noerror;
+{
+ int nlines = lines_to_list / 2;
+ int bline = line > nlines ? line - nlines : 1;
+ int eline = line + nlines;
+ print_source_lines_base (s, bline, eline, noerror, line);
+}
+
static void
list_command (arg, from_tty)
char *arg;
@@ -1157,8 +1172,25 @@
return;
}
- /* "l -" lists previous ten lines, the ones before the ten just listed. */
- if (STREQ (arg, "-"))
+ /* We check for 2 cases here:
+ * "l -" lists previous ten lines, the ones before the ten just listed.
+ * "l -w" gives a context listing showing the current position.
+ */
+ if (STREQN (arg, "-", 1))
+ if (STREQ (arg, "-w"))
+ {
+ extern void (*print_frame_info_listing_hook)
+ PARAMS ((struct symtab *s, int line, int stopline, int noerror));
+ void (*savefn)
+ PARAMS ((struct symtab *s, int line, int stopline, int noerror));
+
+ savefn = print_frame_info_listing_hook;
+ print_frame_info_listing_hook = printw;
+ frame_command (NULL, 0);
+ print_frame_info_listing_hook = savefn;
+ return;
+ }
+ else
{
if (current_source_symtab == 0)
error ("No default source file yet. Do \"help list\".");
@@ -1500,7 +1532,7 @@
/* Match! */
fclose (stream);
if (tui_version)
- print_source_lines_base (current_source_symtab, line, line+1, 0);
+ print_source_lines_base (current_source_symtab, line, line+1, 0, 0);
print_source_lines (current_source_symtab, line, line+1, 0);
set_internalvar (lookup_internalvar ("_"),
value_from_longest (builtin_type_int,
@@ -1598,7 +1630,7 @@
/* Match! */
fclose (stream);
if (tui_version)
- print_source_lines_base (current_source_symtab, line, line+1, 0);
+ print_source_lines_base (current_source_symtab, line, line+1, 0, 0);
print_source_lines (current_source_symtab, line, line+1, 0);
set_internalvar (lookup_internalvar ("_"),
value_from_longest (builtin_type_int,
@@ -1695,6 +1727,7 @@
add_com ("list", class_files, list_command,
concat ("List specified function or line.\n\
With no argument, lists ten more lines after or around previous listing.\n\
+\"list -w\" lists the currently executing line of souce code with context.\n\
\"list -\" lists the ten lines before a previous ten-line listing.\n\
One argument specifies a line, and ten lines are listed around that line.\n\
Two arguments with comma between specify starting and ending lines to list.\n\
diff -ru gdb-4.18.orig/gdb/symtab.h gdb-4.18.new/gdb/symtab.h
--- gdb-4.18.orig/gdb/symtab.h Wed Feb 10 19:24:40 1999
+++ gdb-4.18.new/gdb/symtab.h Wed Jan 12 17:42:58 2000
@@ -1403,6 +1403,9 @@
identify_source_line PARAMS ((struct symtab *, int, int, CORE_ADDR));
extern void
+print_source_lines_base PARAMS ((struct symtab *, int, int, int, int));
+
+extern void
print_source_lines PARAMS ((struct symtab *, int, int, int));
extern void
^ permalink raw reply [flat|nested] only message in thread