* [PATCH 2/8] make symtab::dirname const
2013-11-26 21:50 [PATCH 0/8] trivia Tom Tromey
@ 2013-11-26 21:50 ` Tom Tromey
2013-11-26 21:57 ` [PATCH 1/8] make symtab::filename const Tom Tromey
` (7 subsequent siblings)
8 siblings, 0 replies; 14+ messages in thread
From: Tom Tromey @ 2013-11-26 21:50 UTC (permalink / raw)
To: gdb-patches; +Cc: Tom Tromey
This makes symtab::dirname const and updates one spot to avoid an
intermediate constless result.
2013-11-26 Tom Tromey <tromey@redhat.com>
* buildsym.c (end_symtab_from_static_block): Use obstack_copy0.
* symtab.h (struct symtab) <dirname>: Now const.
---
gdb/ChangeLog | 5 +++++
gdb/buildsym.c | 8 ++++----
gdb/symtab.h | 2 +-
3 files changed, 10 insertions(+), 5 deletions(-)
diff --git a/gdb/buildsym.c b/gdb/buildsym.c
index 0326e26..8d9bdb1 100644
--- a/gdb/buildsym.c
+++ b/gdb/buildsym.c
@@ -1204,10 +1204,10 @@ end_symtab_from_static_block (struct block *static_block,
if (subfile->dirname)
{
/* Reallocate the dirname on the symbol obstack. */
- symtab->dirname = (char *)
- obstack_alloc (&objfile->objfile_obstack,
- strlen (subfile->dirname) + 1);
- strcpy (symtab->dirname, subfile->dirname);
+ symtab->dirname =
+ obstack_copy0 (&objfile->objfile_obstack,
+ subfile->dirname,
+ strlen (subfile->dirname));
}
else
{
diff --git a/gdb/symtab.h b/gdb/symtab.h
index 821479a..3be85ca 100644
--- a/gdb/symtab.h
+++ b/gdb/symtab.h
@@ -879,7 +879,7 @@ struct symtab
/* Directory in which it was compiled, or NULL if we don't know. */
- char *dirname;
+ const char *dirname;
/* Total number of lines found in source file. */
--
1.8.1.4
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 1/8] make symtab::filename const
2013-11-26 21:50 [PATCH 0/8] trivia Tom Tromey
2013-11-26 21:50 ` [PATCH 2/8] make symtab::dirname const Tom Tromey
@ 2013-11-26 21:57 ` Tom Tromey
2013-11-26 22:12 ` [PATCH 3/8] put the psymtab filename in the filename bcache Tom Tromey
` (6 subsequent siblings)
8 siblings, 0 replies; 14+ messages in thread
From: Tom Tromey @ 2013-11-26 21:57 UTC (permalink / raw)
To: gdb-patches; +Cc: Tom Tromey
This makes symtab::filename const and removes a newly unnecessary
cast.
2013-11-26 Tom Tromey <tromey@redhat.com>
* symfile.c (allocate_symtab): Remove cast.
* symtab.h (struct symtab) <filename>: Now const.
---
gdb/ChangeLog | 5 +++++
gdb/symfile.c | 4 ++--
gdb/symtab.h | 2 +-
3 files changed, 8 insertions(+), 3 deletions(-)
diff --git a/gdb/symfile.c b/gdb/symfile.c
index 59d0583..71f6b3e 100644
--- a/gdb/symfile.c
+++ b/gdb/symfile.c
@@ -2883,8 +2883,8 @@ allocate_symtab (const char *filename, struct objfile *objfile)
symtab = (struct symtab *)
obstack_alloc (&objfile->objfile_obstack, sizeof (struct symtab));
memset (symtab, 0, sizeof (*symtab));
- symtab->filename = (char *) bcache (filename, strlen (filename) + 1,
- objfile->per_bfd->filename_cache);
+ symtab->filename = bcache (filename, strlen (filename) + 1,
+ objfile->per_bfd->filename_cache);
symtab->fullname = NULL;
symtab->language = deduce_language_from_filename (filename);
symtab->debugformat = "unknown";
diff --git a/gdb/symtab.h b/gdb/symtab.h
index 7cc6667..821479a 100644
--- a/gdb/symtab.h
+++ b/gdb/symtab.h
@@ -875,7 +875,7 @@ struct symtab
/* Name of this source file. This pointer is never NULL. */
- char *filename;
+ const char *filename;
/* Directory in which it was compiled, or NULL if we don't know. */
--
1.8.1.4
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 3/8] put the psymtab filename in the filename bcache
2013-11-26 21:50 [PATCH 0/8] trivia Tom Tromey
2013-11-26 21:50 ` [PATCH 2/8] make symtab::dirname const Tom Tromey
2013-11-26 21:57 ` [PATCH 1/8] make symtab::filename const Tom Tromey
@ 2013-11-26 22:12 ` Tom Tromey
2013-11-26 22:27 ` [PATCH 8/8] update free_objfile comment Tom Tromey
` (5 subsequent siblings)
8 siblings, 0 replies; 14+ messages in thread
From: Tom Tromey @ 2013-11-26 22:12 UTC (permalink / raw)
To: gdb-patches; +Cc: Tom Tromey
This puts the psymtab filename in the filename bcache.
This saves a small amount of memory.
2013-11-26 Tom Tromey <tromey@redhat.com>
* psymtab.c (allocate_psymtab): Put the filename in the filename
bcache.
---
gdb/ChangeLog | 5 +++++
gdb/psymtab.c | 4 ++--
2 files changed, 7 insertions(+), 2 deletions(-)
diff --git a/gdb/psymtab.c b/gdb/psymtab.c
index 8eac5e1..1d33662 100644
--- a/gdb/psymtab.c
+++ b/gdb/psymtab.c
@@ -1742,8 +1742,8 @@ allocate_psymtab (const char *filename, struct objfile *objfile)
sizeof (struct partial_symtab));
memset (psymtab, 0, sizeof (struct partial_symtab));
- psymtab->filename = obstack_copy0 (&objfile->objfile_obstack,
- filename, strlen (filename));
+ psymtab->filename = bcache (filename, strlen (filename) + 1,
+ objfile->per_bfd->filename_cache);
psymtab->symtab = NULL;
/* Prepend it to the psymtab list for the objfile it belongs to.
--
1.8.1.4
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 8/8] update free_objfile comment
2013-11-26 21:50 [PATCH 0/8] trivia Tom Tromey
` (2 preceding siblings ...)
2013-11-26 22:12 ` [PATCH 3/8] put the psymtab filename in the filename bcache Tom Tromey
@ 2013-11-26 22:27 ` Tom Tromey
2013-11-26 22:37 ` [PATCH 7/8] remove objfile_to_front Tom Tromey
` (4 subsequent siblings)
8 siblings, 0 replies; 14+ messages in thread
From: Tom Tromey @ 2013-11-26 22:27 UTC (permalink / raw)
To: gdb-patches; +Cc: Tom Tromey
The introductory comment to free_objfile is obsolete.
This patch fixes it by removing all the obsolete bits.
2013-11-26 Tom Tromey <tromey@redhat.com>
* objfiles.c (free_objfile): Update comment.
---
gdb/ChangeLog | 4 ++++
gdb/objfiles.c | 16 +---------------
2 files changed, 5 insertions(+), 15 deletions(-)
diff --git a/gdb/objfiles.c b/gdb/objfiles.c
index e8d641f..4f84ff6 100644
--- a/gdb/objfiles.c
+++ b/gdb/objfiles.c
@@ -509,21 +509,7 @@ free_objfile_separate_debug (struct objfile *objfile)
}
}
-/* Destroy an objfile and all the symtabs and psymtabs under it. Note
- that as much as possible is allocated on the objfile_obstack
- so that the memory can be efficiently freed.
-
- Things which we do NOT free because they are not in malloc'd memory
- or not in memory specific to the objfile include:
-
- objfile -> sf
-
- FIXME: If the objfile is using reusable symbol information (via mmalloc),
- then we need to take into account the fact that more than one process
- may be using the symbol information at the same time (when mmalloc is
- extended to support cooperative locking). When more than one process
- is using the mapped symbol info, we need to be more careful about when
- we free objects in the reusable area. */
+/* Destroy an objfile and all the symtabs and psymtabs under it. */
void
free_objfile (struct objfile *objfile)
--
1.8.1.4
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 7/8] remove objfile_to_front
2013-11-26 21:50 [PATCH 0/8] trivia Tom Tromey
` (3 preceding siblings ...)
2013-11-26 22:27 ` [PATCH 8/8] update free_objfile comment Tom Tromey
@ 2013-11-26 22:37 ` Tom Tromey
2013-11-26 22:44 ` [PATCH 5/8] pack partial_symtab for space Tom Tromey
` (3 subsequent siblings)
8 siblings, 0 replies; 14+ messages in thread
From: Tom Tromey @ 2013-11-26 22:37 UTC (permalink / raw)
To: gdb-patches; +Cc: Tom Tromey
I happened to notice that nothing uses objfile_to_front.
This patch removes it.
2013-11-26 Tom Tromey <tromey@redhat.com>
* objfiles.h (objfile_to_front): Remove.
* objfiles.c (objfile_to_front): Remove.
---
gdb/ChangeLog | 5 +++++
gdb/objfiles.c | 20 --------------------
gdb/objfiles.h | 2 --
3 files changed, 5 insertions(+), 22 deletions(-)
diff --git a/gdb/objfiles.c b/gdb/objfiles.c
index ba930fa..e8d641f 100644
--- a/gdb/objfiles.c
+++ b/gdb/objfiles.c
@@ -438,26 +438,6 @@ put_objfile_before (struct objfile *objfile, struct objfile *before_this)
_("put_objfile_before: before objfile not in list"));
}
-/* Put OBJFILE at the front of the list. */
-
-void
-objfile_to_front (struct objfile *objfile)
-{
- struct objfile **objp;
- for (objp = &object_files; *objp != NULL; objp = &((*objp)->next))
- {
- if (*objp == objfile)
- {
- /* Unhook it from where it is. */
- *objp = objfile->next;
- /* Put it in the front. */
- objfile->next = object_files;
- object_files = objfile;
- break;
- }
- }
-}
-
/* Unlink OBJFILE from the list of known objfiles, if it is found in the
list.
diff --git a/gdb/objfiles.h b/gdb/objfiles.h
index 9bca812..3f08d4b 100644
--- a/gdb/objfiles.h
+++ b/gdb/objfiles.h
@@ -453,8 +453,6 @@ extern struct objfile *objfile_separate_debug_iterate (const struct objfile *,
extern void put_objfile_before (struct objfile *, struct objfile *);
-extern void objfile_to_front (struct objfile *);
-
extern void add_separate_debug_objfile (struct objfile *, struct objfile *);
extern void unlink_objfile (struct objfile *);
--
1.8.1.4
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 5/8] pack partial_symtab for space
2013-11-26 21:50 [PATCH 0/8] trivia Tom Tromey
` (4 preceding siblings ...)
2013-11-26 22:37 ` [PATCH 7/8] remove objfile_to_front Tom Tromey
@ 2013-11-26 22:44 ` Tom Tromey
2013-11-27 8:34 ` [PATCH 6/8] remove unnecessary declaration Tom Tromey
` (2 subsequent siblings)
8 siblings, 0 replies; 14+ messages in thread
From: Tom Tromey @ 2013-11-26 22:44 UTC (permalink / raw)
To: gdb-patches; +Cc: Tom Tromey
This improves the packing of struct partial_symtab. I noticed with
pahole that were were a couple of holes. This consolidates the holes
without, I think, affecting readability -- it just moves the "user"
field a bit earlier in the struct. This change saves a small amount
of memory.
2013-11-26 Tom Tromey <tromey@redhat.com>
* psympriv.h (struct partial_symtab) <user>: Move earlier.
---
gdb/ChangeLog | 4 ++++
gdb/psympriv.h | 26 +++++++++++++-------------
2 files changed, 17 insertions(+), 13 deletions(-)
diff --git a/gdb/psympriv.h b/gdb/psympriv.h
index 7a62bcc..9df687d 100644
--- a/gdb/psympriv.h
+++ b/gdb/psympriv.h
@@ -111,19 +111,6 @@ struct partial_symtab
CORE_ADDR textlow;
CORE_ADDR texthigh;
- /* Array of pointers to all of the partial_symtab's which this one
- depends on. Since this array can only be set to previous or
- the current (?) psymtab, this dependency tree is guaranteed not
- to have any loops. "depends on" means that symbols must be read
- for the dependencies before being read for this psymtab; this is
- for type references in stabs, where if foo.c includes foo.h, declarations
- in foo.h may use type numbers defined in foo.c. For other debugging
- formats there may be no need to use dependencies. */
-
- struct partial_symtab **dependencies;
-
- int number_of_dependencies;
-
/* If NULL, this is an ordinary partial symbol table.
If non-NULL, this holds a single includer of this partial symbol
@@ -153,6 +140,19 @@ struct partial_symtab
struct partial_symtab *user;
+ /* Array of pointers to all of the partial_symtab's which this one
+ depends on. Since this array can only be set to previous or
+ the current (?) psymtab, this dependency tree is guaranteed not
+ to have any loops. "depends on" means that symbols must be read
+ for the dependencies before being read for this psymtab; this is
+ for type references in stabs, where if foo.c includes foo.h, declarations
+ in foo.h may use type numbers defined in foo.c. For other debugging
+ formats there may be no need to use dependencies. */
+
+ struct partial_symtab **dependencies;
+
+ int number_of_dependencies;
+
/* Global symbol list. This list will be sorted after readin to
improve access. Binary search will be the usual method of
finding a symbol within it. globals_offset is an integer offset
--
1.8.1.4
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 6/8] remove unnecessary declaration
2013-11-26 21:50 [PATCH 0/8] trivia Tom Tromey
` (5 preceding siblings ...)
2013-11-26 22:44 ` [PATCH 5/8] pack partial_symtab for space Tom Tromey
@ 2013-11-27 8:34 ` Tom Tromey
2013-11-27 9:49 ` [PATCH 4/8] remove some stale FIXMEs Tom Tromey
2013-12-06 19:12 ` [PATCH 0/8] trivia Tom Tromey
8 siblings, 0 replies; 14+ messages in thread
From: Tom Tromey @ 2013-11-27 8:34 UTC (permalink / raw)
To: gdb-patches; +Cc: Tom Tromey
This removes an unnecessary declaration from minsyms.c.
2013-11-26 Tom Tromey <tromey@redhat.com>
* minsyms.c (get_symbol_leading_char): Remove unnecessary
declaration.
---
gdb/ChangeLog | 5 +++++
gdb/minsyms.c | 2 --
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/gdb/minsyms.c b/gdb/minsyms.c
index 9044803..db830c7 100644
--- a/gdb/minsyms.c
+++ b/gdb/minsyms.c
@@ -843,8 +843,6 @@ lookup_minimal_symbol_and_objfile (const char *name)
/* Return leading symbol character for a BFD. If BFD is NULL,
return the leading symbol character from the main objfile. */
-static int get_symbol_leading_char (bfd *);
-
static int
get_symbol_leading_char (bfd *abfd)
{
--
1.8.1.4
^ permalink raw reply [flat|nested] 14+ messages in thread* [PATCH 4/8] remove some stale FIXMEs
2013-11-26 21:50 [PATCH 0/8] trivia Tom Tromey
` (6 preceding siblings ...)
2013-11-27 8:34 ` [PATCH 6/8] remove unnecessary declaration Tom Tromey
@ 2013-11-27 9:49 ` Tom Tromey
2013-11-27 12:25 ` Pedro Alves
2013-12-06 19:12 ` [PATCH 0/8] trivia Tom Tromey
8 siblings, 1 reply; 14+ messages in thread
From: Tom Tromey @ 2013-11-27 9:49 UTC (permalink / raw)
To: gdb-patches; +Cc: Tom Tromey
This removes a couple of old "32x64" FIXME comments that are no longer
true -- the code they annotate is correct now.
2013-11-26 Tom Tromey <tromey@redhat.com>
* cli/cli-cmds.c (edit_command): Remove stale FIXME.
(list_command): Likewise.
---
gdb/ChangeLog | 5 +++++
gdb/cli/cli-cmds.c | 2 --
2 files changed, 5 insertions(+), 2 deletions(-)
diff --git a/gdb/cli/cli-cmds.c b/gdb/cli/cli-cmds.c
index 85f1713..6815d9f 100644
--- a/gdb/cli/cli-cmds.c
+++ b/gdb/cli/cli-cmds.c
@@ -817,7 +817,6 @@ edit_command (char *arg, int from_tty)
struct gdbarch *gdbarch;
if (sal.symtab == 0)
- /* FIXME-32x64--assumes sal.pc fits in long. */
error (_("No source file for address %s."),
hex_string ((unsigned long) sal.pc));
@@ -982,7 +981,6 @@ list_command (char *arg, int from_tty)
struct gdbarch *gdbarch;
if (sal.symtab == 0)
- /* FIXME-32x64--assumes sal.pc fits in long. */
error (_("No source file for address %s."),
hex_string ((unsigned long) sal.pc));
--
1.8.1.4
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH 4/8] remove some stale FIXMEs
2013-11-27 9:49 ` [PATCH 4/8] remove some stale FIXMEs Tom Tromey
@ 2013-11-27 12:25 ` Pedro Alves
2013-11-27 15:56 ` Tom Tromey
0 siblings, 1 reply; 14+ messages in thread
From: Pedro Alves @ 2013-11-27 12:25 UTC (permalink / raw)
To: Tom Tromey; +Cc: gdb-patches
On 11/26/2013 09:49 PM, Tom Tromey wrote:
> This removes a couple of old "32x64" FIXME comments that are no longer
> true -- the code they annotate is correct now.
Not really true. unsigned long is usually 32-bit on 32-bit hosts.
So on 32x64 (32-bit host, 64-bit target), the unsigned long casts
trim the target address to 32-bit.
The casts should be removed, or even better, the hex_string
calls replaced by:
paddress (get_current_arch (), sal.pc))
>
> if (sal.symtab == 0)
> - /* FIXME-32x64--assumes sal.pc fits in long. */
> error (_("No source file for address %s."),
> hex_string ((unsigned long) sal.pc));
>
> @@ -982,7 +981,6 @@ list_command (char *arg, int from_tty)
> struct gdbarch *gdbarch;
>
> if (sal.symtab == 0)
> - /* FIXME-32x64--assumes sal.pc fits in long. */
> error (_("No source file for address %s."),
> hex_string ((unsigned long) sal.pc));
>
--
Pedro Alves
^ permalink raw reply [flat|nested] 14+ messages in thread* Re: [PATCH 4/8] remove some stale FIXMEs
2013-11-27 12:25 ` Pedro Alves
@ 2013-11-27 15:56 ` Tom Tromey
2013-11-27 17:21 ` Tom Tromey
0 siblings, 1 reply; 14+ messages in thread
From: Tom Tromey @ 2013-11-27 15:56 UTC (permalink / raw)
To: Pedro Alves; +Cc: gdb-patches
>>>>> "Pedro" == Pedro Alves <palves@redhat.com> writes:
Pedro> Not really true. unsigned long is usually 32-bit on 32-bit hosts.
Pedro> So on 32x64 (32-bit host, 64-bit target), the unsigned long casts
Pedro> trim the target address to 32-bit.
I'll fix it.
Tom
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 4/8] remove some stale FIXMEs
2013-11-27 15:56 ` Tom Tromey
@ 2013-11-27 17:21 ` Tom Tromey
2013-11-27 17:52 ` Pedro Alves
0 siblings, 1 reply; 14+ messages in thread
From: Tom Tromey @ 2013-11-27 17:21 UTC (permalink / raw)
To: Pedro Alves; +Cc: gdb-patches
Tom> I'll fix it.
Here's a new version.
Tom
commit 370f78a0c0bdb770d21abc3dccb61da4ed3574eb
Author: Tom Tromey <tromey@redhat.com>
Date: Mon Nov 25 08:47:51 2013 -0700
fix a couple of FIXMEs
This fixes a couple of old "32x64" FIXME comments by using paddress
with current_gdbarch rather than hex_string and a cast to long.
2013-11-26 Tom Tromey <tromey@redhat.com>
* cli/cli-cmds.c (edit_command): Use paddress, not hex_string.
(list_command): Likewise.
diff --git a/gdb/ChangeLog b/gdb/ChangeLog
index 0bda3bb..d059474 100644
--- a/gdb/ChangeLog
+++ b/gdb/ChangeLog
@@ -1,5 +1,10 @@
2013-11-26 Tom Tromey <tromey@redhat.com>
+ * cli/cli-cmds.c (edit_command): Use paddress, not hex_string.
+ (list_command): Likewise.
+
+2013-11-26 Tom Tromey <tromey@redhat.com>
+
* psymtab.c (allocate_psymtab): Put the filename in the filename
bcache.
diff --git a/gdb/cli/cli-cmds.c b/gdb/cli/cli-cmds.c
index 85f1713..52a6bc9 100644
--- a/gdb/cli/cli-cmds.c
+++ b/gdb/cli/cli-cmds.c
@@ -817,9 +817,8 @@ edit_command (char *arg, int from_tty)
struct gdbarch *gdbarch;
if (sal.symtab == 0)
- /* FIXME-32x64--assumes sal.pc fits in long. */
error (_("No source file for address %s."),
- hex_string ((unsigned long) sal.pc));
+ paddress (get_current_arch (), sal.pc));
gdbarch = get_objfile_arch (sal.symtab->objfile);
sym = find_pc_function (sal.pc);
@@ -982,9 +981,8 @@ list_command (char *arg, int from_tty)
struct gdbarch *gdbarch;
if (sal.symtab == 0)
- /* FIXME-32x64--assumes sal.pc fits in long. */
error (_("No source file for address %s."),
- hex_string ((unsigned long) sal.pc));
+ paddress (get_current_arch (), sal.pc));
gdbarch = get_objfile_arch (sal.symtab->objfile);
sym = find_pc_function (sal.pc);
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 0/8] trivia
2013-11-26 21:50 [PATCH 0/8] trivia Tom Tromey
` (7 preceding siblings ...)
2013-11-27 9:49 ` [PATCH 4/8] remove some stale FIXMEs Tom Tromey
@ 2013-12-06 19:12 ` Tom Tromey
8 siblings, 0 replies; 14+ messages in thread
From: Tom Tromey @ 2013-12-06 19:12 UTC (permalink / raw)
To: gdb-patches
>>>>> "Tom" == Tom Tromey <tromey@redhat.com> writes:
Tom> I have a few trivial changes I've accumulated over the years that I
Tom> never got around to submitting. I've collected them all in this
Tom> series.
I'm pushing this in now.
Tom
^ permalink raw reply [flat|nested] 14+ messages in thread