* [PATCH][gdb/symtab] Factor out have_complaint
@ 2022-10-04 15:10 Tom de Vries
2022-10-04 15:51 ` Tom de Vries
0 siblings, 1 reply; 4+ messages in thread
From: Tom de Vries @ 2022-10-04 15:10 UTC (permalink / raw)
To: gdb-patches
Hi,
After committing 8ba677d3560 ("[gdb/symtab] Don't complain about function
decls") I noticed that quite a bit of code in read_func_scope is used to decide
whether to issue the "cannot get low and high bounds for subprogram DIE at
$hex" complaint, which executes unnecessarily if we have the default
"set complaints 0".
Fix this by (NFC):
- factoring out new static function have_complaint from macro complaint, and
- using it to wrap the relevant code in read_func_scope.
Tested on x86_64-linux.
Any comments?
Thanks,
- Tom
[gdb/symtab] Factor out have_complaint
---
gdb/complaints.h | 11 ++++++++++-
gdb/dwarf2/read.c | 31 +++++++++++++++++--------------
2 files changed, 27 insertions(+), 15 deletions(-)
diff --git a/gdb/complaints.h b/gdb/complaints.h
index 68c79bd6d0a..3ba2db76b0b 100644
--- a/gdb/complaints.h
+++ b/gdb/complaints.h
@@ -31,6 +31,15 @@ extern void complaint_internal (const char *fmt, ...)
extern int stop_whining;
+/* Return true if complaints are enabled. This can be used to guard code
+ that is used only to decide whether to issue a complaint. */
+
+static bool
+have_complaint ()
+{
+ return stop_whining > 0;
+}
+
/* Register a complaint. This is a macro around complaint_internal to
avoid computing complaint's arguments when complaints are disabled.
Running FMT via gettext [i.e., _(FMT)] can be quite expensive, for
@@ -38,7 +47,7 @@ extern int stop_whining;
#define complaint(FMT, ...) \
do \
{ \
- if (stop_whining > 0) \
+ if (have_complaints ()) \
complaint_internal (FMT, ##__VA_ARGS__); \
} \
while (0)
diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c
index 8e42c0f4d8d..78f4cc1f60d 100644
--- a/gdb/dwarf2/read.c
+++ b/gdb/dwarf2/read.c
@@ -12043,20 +12043,23 @@ read_func_scope (struct die_info *die, struct dwarf2_cu *cu)
if (dwarf2_get_pc_bounds (die, &lowpc, &highpc, cu, nullptr, nullptr)
<= PC_BOUNDS_INVALID)
{
- attr = dwarf2_attr (die, DW_AT_external, cu);
- bool external_p = attr != nullptr && attr->as_boolean ();
- attr = dwarf2_attr (die, DW_AT_inline, cu);
- bool inlined_p
- = (attr != nullptr
- && attr->is_nonnegative ()
- && (attr->as_nonnegative () == DW_INL_inlined
- || attr->as_nonnegative () == DW_INL_declared_inlined));
- attr = dwarf2_attr (die, DW_AT_declaration, cu);
- bool decl_p = attr != nullptr && attr->as_boolean ();
- if (!external_p && !inlined_p && !decl_p)
- complaint (_("cannot get low and high bounds "
- "for subprogram DIE at %s"),
- sect_offset_str (die->sect_off));
+ if (have_complaint ())
+ {
+ attr = dwarf2_attr (die, DW_AT_external, cu);
+ bool external_p = attr != nullptr && attr->as_boolean ();
+ attr = dwarf2_attr (die, DW_AT_inline, cu);
+ bool inlined_p
+ = (attr != nullptr
+ && attr->is_nonnegative ()
+ && (attr->as_nonnegative () == DW_INL_inlined
+ || attr->as_nonnegative () == DW_INL_declared_inlined));
+ attr = dwarf2_attr (die, DW_AT_declaration, cu);
+ bool decl_p = attr != nullptr && attr->as_boolean ();
+ if (!external_p && !inlined_p && !decl_p)
+ complaint (_("cannot get low and high bounds "
+ "for subprogram DIE at %s"),
+ sect_offset_str (die->sect_off));
+ }
return;
}
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH][gdb/symtab] Factor out have_complaint
2022-10-04 15:10 [PATCH][gdb/symtab] Factor out have_complaint Tom de Vries
@ 2022-10-04 15:51 ` Tom de Vries
2022-10-04 16:43 ` Tom de Vries
0 siblings, 1 reply; 4+ messages in thread
From: Tom de Vries @ 2022-10-04 15:51 UTC (permalink / raw)
To: gdb-patches
On 10/4/22 17:10, Tom de Vries via Gdb-patches wrote:
> Hi,
>
> After committing 8ba677d3560 ("[gdb/symtab] Don't complain about function
> decls") I noticed that quite a bit of code in read_func_scope is used to decide
> whether to issue the "cannot get low and high bounds for subprogram DIE at
> $hex" complaint, which executes unnecessarily if we have the default
> "set complaints 0".
>
> Fix this by (NFC):
> - factoring out new static function have_complaint from macro complaint, and
> - using it to wrap the relevant code in read_func_scope.
>
> Tested on x86_64-linux.
>
> Any comments?
>
Well, I didn't check properly, this doesn't even build, I'll repost once
I got a properly build & tested version.
Sorry for the noise.
Thanks,
- Tom
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH][gdb/symtab] Factor out have_complaint
2022-10-04 15:51 ` Tom de Vries
@ 2022-10-04 16:43 ` Tom de Vries
2022-10-05 9:43 ` Bruno Larsen
0 siblings, 1 reply; 4+ messages in thread
From: Tom de Vries @ 2022-10-04 16:43 UTC (permalink / raw)
To: gdb-patches
[-- Attachment #1: Type: text/plain, Size: 820 bytes --]
On 10/4/22 17:51, Tom de Vries wrote:
> On 10/4/22 17:10, Tom de Vries via Gdb-patches wrote:
>> Hi,
>>
>> After committing 8ba677d3560 ("[gdb/symtab] Don't complain about function
>> decls") I noticed that quite a bit of code in read_func_scope is used
>> to decide
>> whether to issue the "cannot get low and high bounds for subprogram
>> DIE at
>> $hex" complaint, which executes unnecessarily if we have the default
>> "set complaints 0".
>>
>> Fix this by (NFC):
>> - factoring out new static function have_complaint from macro
>> complaint, and
>> - using it to wrap the relevant code in read_func_scope.
>>
>> Tested on x86_64-linux.
>>
>> Any comments?
>>
>
> Well, I didn't check properly, this doesn't even build, I'll repost once
> I got a properly build & tested version.
And here it is.
Thanks,
- Tom
[-- Attachment #2: 0001-gdb-symtab-Factor-out-have_complaint.patch --]
[-- Type: text/x-patch, Size: 3305 bytes --]
[gdb/symtab] Factor out have_complaint
After committing 8ba677d3560 ("[gdb/symtab] Don't complain about function
decls") I noticed that quite a bit of code in read_func_scope is used to decide
whether to issue the "cannot get low and high bounds for subprogram DIE at
$hex" complaint, which executes unnecessarily if we have the default
"set complaints 0".
Fix this by (NFC):
- factoring out new static function have_complaint from macro complaint, and
- using it to wrap the relevant code in read_func_scope.
Tested on x86_64-linux.
---
gdb/complaints.h | 11 ++++++++++-
gdb/dwarf2/read.c | 31 +++++++++++++++++--------------
2 files changed, 27 insertions(+), 15 deletions(-)
diff --git a/gdb/complaints.h b/gdb/complaints.h
index 68c79bd6d0a..02c8c2c558b 100644
--- a/gdb/complaints.h
+++ b/gdb/complaints.h
@@ -31,6 +31,15 @@ extern void complaint_internal (const char *fmt, ...)
extern int stop_whining;
+/* Return true if complaints are enabled. This can be used to guard code
+ that is used only to decide whether to issue a complaint. */
+
+static inline bool
+have_complaint ()
+{
+ return stop_whining > 0;
+}
+
/* Register a complaint. This is a macro around complaint_internal to
avoid computing complaint's arguments when complaints are disabled.
Running FMT via gettext [i.e., _(FMT)] can be quite expensive, for
@@ -38,7 +47,7 @@ extern int stop_whining;
#define complaint(FMT, ...) \
do \
{ \
- if (stop_whining > 0) \
+ if (have_complaint ()) \
complaint_internal (FMT, ##__VA_ARGS__); \
} \
while (0)
diff --git a/gdb/dwarf2/read.c b/gdb/dwarf2/read.c
index 8e42c0f4d8d..78f4cc1f60d 100644
--- a/gdb/dwarf2/read.c
+++ b/gdb/dwarf2/read.c
@@ -12043,20 +12043,23 @@ read_func_scope (struct die_info *die, struct dwarf2_cu *cu)
if (dwarf2_get_pc_bounds (die, &lowpc, &highpc, cu, nullptr, nullptr)
<= PC_BOUNDS_INVALID)
{
- attr = dwarf2_attr (die, DW_AT_external, cu);
- bool external_p = attr != nullptr && attr->as_boolean ();
- attr = dwarf2_attr (die, DW_AT_inline, cu);
- bool inlined_p
- = (attr != nullptr
- && attr->is_nonnegative ()
- && (attr->as_nonnegative () == DW_INL_inlined
- || attr->as_nonnegative () == DW_INL_declared_inlined));
- attr = dwarf2_attr (die, DW_AT_declaration, cu);
- bool decl_p = attr != nullptr && attr->as_boolean ();
- if (!external_p && !inlined_p && !decl_p)
- complaint (_("cannot get low and high bounds "
- "for subprogram DIE at %s"),
- sect_offset_str (die->sect_off));
+ if (have_complaint ())
+ {
+ attr = dwarf2_attr (die, DW_AT_external, cu);
+ bool external_p = attr != nullptr && attr->as_boolean ();
+ attr = dwarf2_attr (die, DW_AT_inline, cu);
+ bool inlined_p
+ = (attr != nullptr
+ && attr->is_nonnegative ()
+ && (attr->as_nonnegative () == DW_INL_inlined
+ || attr->as_nonnegative () == DW_INL_declared_inlined));
+ attr = dwarf2_attr (die, DW_AT_declaration, cu);
+ bool decl_p = attr != nullptr && attr->as_boolean ();
+ if (!external_p && !inlined_p && !decl_p)
+ complaint (_("cannot get low and high bounds "
+ "for subprogram DIE at %s"),
+ sect_offset_str (die->sect_off));
+ }
return;
}
^ permalink raw reply [flat|nested] 4+ messages in thread
* Re: [PATCH][gdb/symtab] Factor out have_complaint
2022-10-04 16:43 ` Tom de Vries
@ 2022-10-05 9:43 ` Bruno Larsen
0 siblings, 0 replies; 4+ messages in thread
From: Bruno Larsen @ 2022-10-05 9:43 UTC (permalink / raw)
To: Tom de Vries, gdb-patches
On 04/10/2022 18:43, Tom de Vries via Gdb-patches wrote:
> On 10/4/22 17:51, Tom de Vries wrote:
>> On 10/4/22 17:10, Tom de Vries via Gdb-patches wrote:
>>> Hi,
>>>
>>> After committing 8ba677d3560 ("[gdb/symtab] Don't complain about
>>> function
>>> decls") I noticed that quite a bit of code in read_func_scope is
>>> used to decide
>>> whether to issue the "cannot get low and high bounds for subprogram
>>> DIE at
>>> $hex" complaint, which executes unnecessarily if we have the default
>>> "set complaints 0".
>>>
>>> Fix this by (NFC):
>>> - factoring out new static function have_complaint from macro
>>> complaint, and
>>> - using it to wrap the relevant code in read_func_scope.
>>>
>>> Tested on x86_64-linux.
>>>
>>> Any comments?
>>>
>>
>> Well, I didn't check properly, this doesn't even build, I'll repost
>> once I got a properly build & tested version.
>
> And here it is.
I tested this locally and it introduces no regressions. And the changes
themselves look contained enough, so I recommend you approve your patch!
Cheers,
Bruno
>
> Thanks,
> - Tom
^ permalink raw reply [flat|nested] 4+ messages in thread
end of thread, other threads:[~2022-10-05 9:43 UTC | newest]
Thread overview: 4+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-10-04 15:10 [PATCH][gdb/symtab] Factor out have_complaint Tom de Vries
2022-10-04 15:51 ` Tom de Vries
2022-10-04 16:43 ` Tom de Vries
2022-10-05 9:43 ` Bruno Larsen
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for read-only IMAP folder(s) and NNTP newsgroup(s).