* [PATCH 0/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
@ 2026-03-02 11:48 Tom de Vries
2026-03-02 11:48 ` [PATCH 1/3] [gdb/symtab] Rename lnp_state_machine::m_last_line to m_last_recorded_line Tom de Vries
` (2 more replies)
0 siblings, 3 replies; 14+ messages in thread
From: Tom de Vries @ 2026-03-02 11:48 UTC (permalink / raw)
To: gdb-patches
This patch series fixes PR33930, a FAIL in test-case
gdb.cp/step-and-next-inline.exp on ppc64le-linux.
The first patch is a simple refactoring patch, renaming
lnp_state_machine::m_last_line (tracking the last recorded line) to the more
accurate m_last_recorded_line.
The second patch reintroduces lnp_state_machine::m_last_line, with semantics
matching the name (tracking all lines, not just recorded ones).
The third patch uses lnp_state_machine::m_last_line in the actual fix.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=33930
Tom de Vries (3):
[gdb/symtab] Rename lnp_state_machine::m_last_line to
m_last_recorded_line
[gdb/symtab] Re-add lnp_state_machine::m_last_line
[gdb/symtab] Tweak fix-up of truncated inline function block ranges
gdb/dwarf2/line-program.c | 26 +++++++++++++++++++++-----
1 file changed, 21 insertions(+), 5 deletions(-)
base-commit: bd0babf07de17da6e8e536ceb4dd940d0ed6a0d0
--
2.51.0
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH 1/3] [gdb/symtab] Rename lnp_state_machine::m_last_line to m_last_recorded_line
2026-03-02 11:48 [PATCH 0/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges Tom de Vries
@ 2026-03-02 11:48 ` Tom de Vries
2026-03-02 11:48 ` [PATCH 2/3] [gdb/symtab] Re-add lnp_state_machine::m_last_line Tom de Vries
2026-03-02 11:48 ` [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges Tom de Vries
2 siblings, 0 replies; 14+ messages in thread
From: Tom de Vries @ 2026-03-02 11:48 UTC (permalink / raw)
To: gdb-patches
In class lnp_state_machine, we have member m_last_line:
...
/* The last line number that was recorded, used to coalesce
consecutive entries for the same line. This can happen, for
example, when discriminators are present. PR 17276. */
unsigned int m_last_line = 0;
...
representing the last recorded line.
Let's make this situation more clear, and rename it to m_last_recorded_line.
---
gdb/dwarf2/line-program.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/gdb/dwarf2/line-program.c b/gdb/dwarf2/line-program.c
index e6e9235f7bc..a28d589e251 100644
--- a/gdb/dwarf2/line-program.c
+++ b/gdb/dwarf2/line-program.c
@@ -196,7 +196,7 @@ class lnp_state_machine
/* The last line number that was recorded, used to coalesce
consecutive entries for the same line. This can happen, for
example, when discriminators are present. PR 17276. */
- unsigned int m_last_line = 0;
+ unsigned int m_last_recorded_line = 0;
bool m_line_has_non_zero_discriminator = false;
};
@@ -301,7 +301,7 @@ lnp_state_machine::record_line_p ()
{
if (m_builder->get_current_subfile () != m_last_subfile)
return true;
- if (m_line != m_last_line)
+ if (m_line != m_last_recorded_line)
return true;
/* Same line for the same file that we've seen already.
As a last check, for pr 17276, only record the line if the line
@@ -495,7 +495,7 @@ lnp_state_machine::record_line (bool end_sequence)
{
m_last_subfile = m_builder->get_current_subfile ();
record_line_1 (m_line, lte_flags);
- m_last_line = m_line;
+ m_last_recorded_line = m_line;
}
}
}
--
2.51.0
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH 2/3] [gdb/symtab] Re-add lnp_state_machine::m_last_line
2026-03-02 11:48 [PATCH 0/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges Tom de Vries
2026-03-02 11:48 ` [PATCH 1/3] [gdb/symtab] Rename lnp_state_machine::m_last_line to m_last_recorded_line Tom de Vries
@ 2026-03-02 11:48 ` Tom de Vries
2026-03-02 11:48 ` [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges Tom de Vries
2 siblings, 0 replies; 14+ messages in thread
From: Tom de Vries @ 2026-03-02 11:48 UTC (permalink / raw)
To: gdb-patches
In class lnp_state_machine, we have member:
...
/* The address of the last line entry. */
unrelocated_addr m_last_address;
...
representing the address of the last line entry.
Add a similar member for the line of the last line entry.
---
gdb/dwarf2/line-program.c | 5 ++++-
1 file changed, 4 insertions(+), 1 deletion(-)
diff --git a/gdb/dwarf2/line-program.c b/gdb/dwarf2/line-program.c
index a28d589e251..b9d93bb7b0e 100644
--- a/gdb/dwarf2/line-program.c
+++ b/gdb/dwarf2/line-program.c
@@ -181,8 +181,9 @@ class lnp_state_machine
/* The last file a line number was recorded for. */
struct subfile *m_last_subfile = NULL;
- /* The address of the last line entry. */
+ /* The address and line of the last line entry. */
unrelocated_addr m_last_address;
+ unsigned int m_last_line = 0;
/* Set to true when a previous line at the same address (using
m_last_address) had LEF_IS_STMT set in m_flags. This is reset to false
@@ -508,6 +509,8 @@ lnp_state_machine::record_line (bool end_sequence)
m_last_address = m_address;
}
m_stmt_at_address |= (m_flags & LEF_IS_STMT) != 0;
+
+ m_last_line = m_line;
}
lnp_state_machine::lnp_state_machine (struct dwarf2_cu *cu, gdbarch *arch)
--
2.51.0
^ permalink raw reply [flat|nested] 14+ messages in thread
* [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-02 11:48 [PATCH 0/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges Tom de Vries
2026-03-02 11:48 ` [PATCH 1/3] [gdb/symtab] Rename lnp_state_machine::m_last_line to m_last_recorded_line Tom de Vries
2026-03-02 11:48 ` [PATCH 2/3] [gdb/symtab] Re-add lnp_state_machine::m_last_line Tom de Vries
@ 2026-03-02 11:48 ` Tom de Vries
2026-03-10 16:05 ` Andrew Burgess
2026-03-19 15:39 ` Andrew Burgess
2 siblings, 2 replies; 14+ messages in thread
From: Tom de Vries @ 2026-03-02 11:48 UTC (permalink / raw)
To: gdb-patches
Consider test-case gdb.cp/step-and-next-inline.exp on ppc64le-linux.
The corresponding source file step-and-next-inline.cc contains functions
tree_check and get_alias_set:
...
35 #define TREE_TYPE(NODE) (*tree_check (NODE, 0))
36
37 inline tree *
38 tree_check (tree *t, int i)
39 {
40 if (t->x != i)
41 abort();
42 tree *x = t;
43 return x;
44 }
...
48 int __attribute__((noinline, noclone))
49 get_alias_set (tree *t)
50 {
51 if (t != NULL
52 && TREE_TYPE (t).z != 1
53 && TREE_TYPE (t).z != 2
54 && TREE_TYPE (t).z != 3)
55 return 0;
56 return 1;
57 }
...
as well as a trivial function main calling get_alias_set.
Say we step into the first call to tree_check, and then step to the return
at line 43:
...
(gdb) s
tree_check (i=0, t=0x10020030 <xx>) at step-and-next-inline.cc:40
40 if (t->x != i)
(gdb) s
43 return x;
(gdb)
...
At that point, we have pc 0x1000071c:
...
(gdb) p $pc
$1 = (void (*)(void)) 0x1000071c <get_alias_set(tree*)+28>
(gdb)
...
and the backtrace looks like this:
...
(gdb) bt
#0 tree_check (i=<optimized out>, t=<optimized out>) at
step-and-next-inline.cc:43
#1 get_alias_set (t=t@entry=0x10020030 <xx>) at step-and-next-inline.cc:52
#2 0x0000000010000560 in main () at step-and-next-inline.cc:64
(gdb)
...
which shows all 3 functions.
This seems trivial, but it's not.
All three calls to tree_check are inlined, and the first call is represented
by:
...
<2><877>: Abbrev Number: 40 (DW_TAG_inlined_subroutine)
<878> DW_AT_abstract_origin: <0x967>
<87c> DW_AT_entry_pc : 0x10000710
<884> DW_AT_GNU_entry_view: 0
<885> DW_AT_ranges : 0xc
<88a> DW_AT_call_line : 52
...
with DW_AT_ranges referring to:
...
Contents of the .debug_rnglists section:
Offset Begin End
0000000c 0000000010000710 (base address)
00000015 0000000010000710 000000001000071c
00000018 000000001000077c 000000001000077c (start == end)
0000001b 0000000010000788 0000000010000790
0000001f <End of list>
...
The range at offset 0x15 is [0x10000710, 0x1000071c), so address 0x1000071c
does not fall in the range, and consequently the debug info does not consider
0x1000071c part of the inlined tree_check.
However, since commit 8efed40efd6 ("gdb: fix-up truncated inline function
block ranges"), gdb contains a fix in lnp_state_machine::record_line:
...
if (m_address != m_last_address
&& m_stmt_at_address
&& m_cu->producer_is_gcc ()
&& (m_flags & LEF_IS_STMT) == 0)
dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
m_address, m_line);
...
that looks at the corresponding line number information:
...
File name Line number Starting address View Stmt
step-and-next-inline.cc 42 0x1000071c x
step-and-next-inline.cc 43 0x1000071c 1 x
step-and-next-inline.cc 43 0x1000071c 2
step-and-next-inline.cc 52 0x1000071c 3
step-and-next-inline.cc 52 0x10000720
...
and extends the range of the inlined tree_check to include
[0x1000071c, 0x10000720).
[ Please read the commit message of aforementioned commit to understand why
the fix is correct. ]
Let's look in more detail at how the call to
dwarf_find_and_extend_inline_block_range is activated for the fix. It's
activated for the last entry (52/0x10000720), with:
- m_last_address == 0x1000071c,
- m_address == 0x10000720, and
- m_line == 52 (matching the DW_AT_call_line).
It's easy to see that for the last entry, (m_flags & LEF_IS_STMT) == 0 holds,
because it doesn't have an x in the "Stmt" column.
I found it less obvious that m_stmt_at_address also holds.
[ The documentation clarifies that this is related to m_last_address:
....
/* Set to true when a previous line at the same address (using
m_last_address) had LEF_IS_STMT set in m_flags. This is reset to false
when a line entry at a new address (m_address different to
m_last_address) is processed. */
bool m_stmt_at_address = false;
...
To get maximum clarity, I checked the value for each entry:
...
address m_stmt_at_address
---------------------------------
before false
42/0x1000071c false->true
43/0x1000071c/1 true
43/0x1000071c/2 true
52/0x1000071c/3 true
52/0x10000720 true->false
...
Again it's easy to relate the transitions for particular entries to the "Stmt"
column.
The tricky bit is that the transition takes place at the end of
lnp_state_machine::record_line, so while processing entry 52/0x10000720, we
sample m_stmt_at_address while it's still true. ]
Likewise, the fix works for the second inlined call.
But not for the third. The debug info has the same problem, but the fix is
not applied.
The corresponding line info looks slightly different:
...
File name Line number Starting address View Stmt
step-and-next-inline.cc 42 0x1000074c x
step-and-next-inline.cc 43 0x1000074c 1 x
step-and-next-inline.cc 43 0x1000074c 2
step-and-next-inline.cc 54 0x1000074c 3
step-and-next-inline.cc 55 0x10000750
...
In this case, dwarf_find_and_extend_inline_block_range gets called with:
- m_last_address == 0x1000074c,
- m_address == 0x10000750, and
- m_line == 55,
but since line 55 doesn't match DW_AT_call_line 54:
...
<2><918>: Abbrev Number: 44 (DW_TAG_inlined_subroutine)
<919> DW_AT_abstract_origin: <0x967>
<91d> DW_AT_entry_pc : 0x10000740
<925> DW_AT_GNU_entry_view: 0
<926> DW_AT_low_pc : 0x10000740
<92e> DW_AT_high_pc : 0xc
<937> DW_AT_call_line : 54
...
the fix is not applied.
I'm proposing the following simple tweak, to handle the third inlined call as
well: instead of using m_line, use m_last_line.
Tested on x86_64-linux and ppc64le-linux.
Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=33930
---
gdb/dwarf2/line-program.c | 15 ++++++++++++++-
1 file changed, 14 insertions(+), 1 deletion(-)
diff --git a/gdb/dwarf2/line-program.c b/gdb/dwarf2/line-program.c
index b9d93bb7b0e..5aa97ea3714 100644
--- a/gdb/dwarf2/line-program.c
+++ b/gdb/dwarf2/line-program.c
@@ -445,12 +445,25 @@ lnp_state_machine::record_line (bool end_sequence)
(end_sequence ? "\t(end sequence)" : ""));
}
+ /* Activate dwarf_find_and_extend_inline_block_range for line number info:
+
+ Line number Starting address View Stmt
+ 42 0x1000074c x
+ 43 0x1000074c 1 x
+ 43 0x1000074c 2
+ 54 0x1000074c 3
+ 55 0x10000750
+
+ at entry 55/0x10000750 with:
+ - m_last_address == 0x1000074c
+ - m_address == 0x10000750
+ - m_last_line == 54. */
if (m_address != m_last_address
&& m_stmt_at_address
&& m_cu->producer_is_gcc ()
&& (m_flags & LEF_IS_STMT) == 0)
dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
- m_address, m_line);
+ m_address, m_last_line);
file_entry *fe = current_file ();
--
2.51.0
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-02 11:48 ` [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges Tom de Vries
@ 2026-03-10 16:05 ` Andrew Burgess
2026-03-10 20:34 ` Tom de Vries
2026-03-11 13:24 ` Tom de Vries
2026-03-19 15:39 ` Andrew Burgess
1 sibling, 2 replies; 14+ messages in thread
From: Andrew Burgess @ 2026-03-10 16:05 UTC (permalink / raw)
To: Tom de Vries, gdb-patches
Tom de Vries <tdevries@suse.de> writes:
> Consider test-case gdb.cp/step-and-next-inline.exp on ppc64le-linux.
>
> The corresponding source file step-and-next-inline.cc contains functions
> tree_check and get_alias_set:
> ...
> 35 #define TREE_TYPE(NODE) (*tree_check (NODE, 0))
> 36
> 37 inline tree *
> 38 tree_check (tree *t, int i)
> 39 {
> 40 if (t->x != i)
> 41 abort();
> 42 tree *x = t;
> 43 return x;
> 44 }
> ...
> 48 int __attribute__((noinline, noclone))
> 49 get_alias_set (tree *t)
> 50 {
> 51 if (t != NULL
> 52 && TREE_TYPE (t).z != 1
> 53 && TREE_TYPE (t).z != 2
> 54 && TREE_TYPE (t).z != 3)
> 55 return 0;
> 56 return 1;
> 57 }
> ...
> as well as a trivial function main calling get_alias_set.
>
> Say we step into the first call to tree_check, and then step to the return
> at line 43:
> ...
> (gdb) s
> tree_check (i=0, t=0x10020030 <xx>) at step-and-next-inline.cc:40
> 40 if (t->x != i)
> (gdb) s
> 43 return x;
> (gdb)
> ...
>
> At that point, we have pc 0x1000071c:
> ...
> (gdb) p $pc
> $1 = (void (*)(void)) 0x1000071c <get_alias_set(tree*)+28>
> (gdb)
> ...
> and the backtrace looks like this:
> ...
> (gdb) bt
> #0 tree_check (i=<optimized out>, t=<optimized out>) at
> step-and-next-inline.cc:43
> #1 get_alias_set (t=t@entry=0x10020030 <xx>) at step-and-next-inline.cc:52
> #2 0x0000000010000560 in main () at step-and-next-inline.cc:64
> (gdb)
> ...
> which shows all 3 functions.
>
> This seems trivial, but it's not.
>
> All three calls to tree_check are inlined, and the first call is represented
> by:
> ...
> <2><877>: Abbrev Number: 40 (DW_TAG_inlined_subroutine)
> <878> DW_AT_abstract_origin: <0x967>
> <87c> DW_AT_entry_pc : 0x10000710
> <884> DW_AT_GNU_entry_view: 0
> <885> DW_AT_ranges : 0xc
> <88a> DW_AT_call_line : 52
> ...
> with DW_AT_ranges referring to:
> ...
> Contents of the .debug_rnglists section:
>
> Offset Begin End
> 0000000c 0000000010000710 (base address)
> 00000015 0000000010000710 000000001000071c
> 00000018 000000001000077c 000000001000077c (start == end)
> 0000001b 0000000010000788 0000000010000790
> 0000001f <End of list>
> ...
>
> The range at offset 0x15 is [0x10000710, 0x1000071c), so address 0x1000071c
> does not fall in the range, and consequently the debug info does not consider
> 0x1000071c part of the inlined tree_check.
>
> However, since commit 8efed40efd6 ("gdb: fix-up truncated inline function
> block ranges"), gdb contains a fix in lnp_state_machine::record_line:
> ...
> if (m_address != m_last_address
> && m_stmt_at_address
> && m_cu->producer_is_gcc ()
> && (m_flags & LEF_IS_STMT) == 0)
> dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
> m_address, m_line);
> ...
> that looks at the corresponding line number information:
> ...
> File name Line number Starting address View Stmt
> step-and-next-inline.cc 42 0x1000071c x
> step-and-next-inline.cc 43 0x1000071c 1 x
> step-and-next-inline.cc 43 0x1000071c 2
> step-and-next-inline.cc 52 0x1000071c 3
> step-and-next-inline.cc 52 0x10000720
> ...
> and extends the range of the inlined tree_check to include
> [0x1000071c, 0x10000720).
>
> [ Please read the commit message of aforementioned commit to understand why
> the fix is correct. ]
>
> Let's look in more detail at how the call to
> dwarf_find_and_extend_inline_block_range is activated for the fix. It's
> activated for the last entry (52/0x10000720), with:
> - m_last_address == 0x1000071c,
> - m_address == 0x10000720, and
> - m_line == 52 (matching the DW_AT_call_line).
>
> It's easy to see that for the last entry, (m_flags & LEF_IS_STMT) == 0 holds,
> because it doesn't have an x in the "Stmt" column.
>
> I found it less obvious that m_stmt_at_address also holds.
>
> [ The documentation clarifies that this is related to m_last_address:
> ....
> /* Set to true when a previous line at the same address (using
> m_last_address) had LEF_IS_STMT set in m_flags. This is reset to false
> when a line entry at a new address (m_address different to
> m_last_address) is processed. */
> bool m_stmt_at_address = false;
> ...
>
> To get maximum clarity, I checked the value for each entry:
> ...
> address m_stmt_at_address
> ---------------------------------
> before false
> 42/0x1000071c false->true
> 43/0x1000071c/1 true
> 43/0x1000071c/2 true
> 52/0x1000071c/3 true
> 52/0x10000720 true->false
> ...
>
> Again it's easy to relate the transitions for particular entries to the "Stmt"
> column.
>
> The tricky bit is that the transition takes place at the end of
> lnp_state_machine::record_line, so while processing entry 52/0x10000720, we
> sample m_stmt_at_address while it's still true. ]
>
> Likewise, the fix works for the second inlined call.
>
> But not for the third. The debug info has the same problem, but the fix is
> not applied.
>
> The corresponding line info looks slightly different:
> ...
> File name Line number Starting address View Stmt
> step-and-next-inline.cc 42 0x1000074c x
> step-and-next-inline.cc 43 0x1000074c 1 x
> step-and-next-inline.cc 43 0x1000074c 2
> step-and-next-inline.cc 54 0x1000074c 3
> step-and-next-inline.cc 55 0x10000750
> ...
>
> In this case, dwarf_find_and_extend_inline_block_range gets called with:
> - m_last_address == 0x1000074c,
> - m_address == 0x10000750, and
> - m_line == 55,
> but since line 55 doesn't match DW_AT_call_line 54:
> ...
> <2><918>: Abbrev Number: 44 (DW_TAG_inlined_subroutine)
> <919> DW_AT_abstract_origin: <0x967>
> <91d> DW_AT_entry_pc : 0x10000740
> <925> DW_AT_GNU_entry_view: 0
> <926> DW_AT_low_pc : 0x10000740
> <92e> DW_AT_high_pc : 0xc
> <937> DW_AT_call_line : 54
> ...
> the fix is not applied.
>
> I'm proposing the following simple tweak, to handle the third inlined call as
> well: instead of using m_line, use m_last_line.
>
> Tested on x86_64-linux and ppc64le-linux.
>
> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=33930
Hi Tom,
Thanks for looking at this failure. Can you confirm which gcc version
you're using please. I don't see the same failure on cfarm29 (from gcc
compile farm), using gcc 'gcc (Debian 14.2.0-19) 14.2.0'. But that
step-and-next-inline test can change behaviour based on compiler
version, so I'm not surprised you're seeing failures on some specific
combinations.
Which is why I tried to write DWARF assembler tests to cover edge cases
as I found them, and I think it would be great if we could get such a
test to cover this fix too. I'm happy to help with, or even write, the
test, once I can reproduce the failure.
I'll try some other ppc64le machines I have access too, maybe one of
those will have the right compiler version already installed, otherwise
I'll have to rebuild gcc once you let me know which version is causing
problems.
Thanks,
Andrew
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-10 16:05 ` Andrew Burgess
@ 2026-03-10 20:34 ` Tom de Vries
2026-03-14 14:22 ` Andrew Burgess
2026-03-11 13:24 ` Tom de Vries
1 sibling, 1 reply; 14+ messages in thread
From: Tom de Vries @ 2026-03-10 20:34 UTC (permalink / raw)
To: Andrew Burgess, gdb-patches
On 3/10/26 5:05 PM, Andrew Burgess wrote:
> Tom de Vries <tdevries@suse.de> writes:
>
>> Consider test-case gdb.cp/step-and-next-inline.exp on ppc64le-linux.
>>
>> The corresponding source file step-and-next-inline.cc contains functions
>> tree_check and get_alias_set:
>> ...
>> 35 #define TREE_TYPE(NODE) (*tree_check (NODE, 0))
>> 36
>> 37 inline tree *
>> 38 tree_check (tree *t, int i)
>> 39 {
>> 40 if (t->x != i)
>> 41 abort();
>> 42 tree *x = t;
>> 43 return x;
>> 44 }
>> ...
>> 48 int __attribute__((noinline, noclone))
>> 49 get_alias_set (tree *t)
>> 50 {
>> 51 if (t != NULL
>> 52 && TREE_TYPE (t).z != 1
>> 53 && TREE_TYPE (t).z != 2
>> 54 && TREE_TYPE (t).z != 3)
>> 55 return 0;
>> 56 return 1;
>> 57 }
>> ...
>> as well as a trivial function main calling get_alias_set.
>>
>> Say we step into the first call to tree_check, and then step to the return
>> at line 43:
>> ...
>> (gdb) s
>> tree_check (i=0, t=0x10020030 <xx>) at step-and-next-inline.cc:40
>> 40 if (t->x != i)
>> (gdb) s
>> 43 return x;
>> (gdb)
>> ...
>>
>> At that point, we have pc 0x1000071c:
>> ...
>> (gdb) p $pc
>> $1 = (void (*)(void)) 0x1000071c <get_alias_set(tree*)+28>
>> (gdb)
>> ...
>> and the backtrace looks like this:
>> ...
>> (gdb) bt
>> #0 tree_check (i=<optimized out>, t=<optimized out>) at
>> step-and-next-inline.cc:43
>> #1 get_alias_set (t=t@entry=0x10020030 <xx>) at step-and-next-inline.cc:52
>> #2 0x0000000010000560 in main () at step-and-next-inline.cc:64
>> (gdb)
>> ...
>> which shows all 3 functions.
>>
>> This seems trivial, but it's not.
>>
>> All three calls to tree_check are inlined, and the first call is represented
>> by:
>> ...
>> <2><877>: Abbrev Number: 40 (DW_TAG_inlined_subroutine)
>> <878> DW_AT_abstract_origin: <0x967>
>> <87c> DW_AT_entry_pc : 0x10000710
>> <884> DW_AT_GNU_entry_view: 0
>> <885> DW_AT_ranges : 0xc
>> <88a> DW_AT_call_line : 52
>> ...
>> with DW_AT_ranges referring to:
>> ...
>> Contents of the .debug_rnglists section:
>>
>> Offset Begin End
>> 0000000c 0000000010000710 (base address)
>> 00000015 0000000010000710 000000001000071c
>> 00000018 000000001000077c 000000001000077c (start == end)
>> 0000001b 0000000010000788 0000000010000790
>> 0000001f <End of list>
>> ...
>>
>> The range at offset 0x15 is [0x10000710, 0x1000071c), so address 0x1000071c
>> does not fall in the range, and consequently the debug info does not consider
>> 0x1000071c part of the inlined tree_check.
>>
>> However, since commit 8efed40efd6 ("gdb: fix-up truncated inline function
>> block ranges"), gdb contains a fix in lnp_state_machine::record_line:
>> ...
>> if (m_address != m_last_address
>> && m_stmt_at_address
>> && m_cu->producer_is_gcc ()
>> && (m_flags & LEF_IS_STMT) == 0)
>> dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
>> m_address, m_line);
>> ...
>> that looks at the corresponding line number information:
>> ...
>> File name Line number Starting address View Stmt
>> step-and-next-inline.cc 42 0x1000071c x
>> step-and-next-inline.cc 43 0x1000071c 1 x
>> step-and-next-inline.cc 43 0x1000071c 2
>> step-and-next-inline.cc 52 0x1000071c 3
>> step-and-next-inline.cc 52 0x10000720
>> ...
>> and extends the range of the inlined tree_check to include
>> [0x1000071c, 0x10000720).
>>
>> [ Please read the commit message of aforementioned commit to understand why
>> the fix is correct. ]
>>
>> Let's look in more detail at how the call to
>> dwarf_find_and_extend_inline_block_range is activated for the fix. It's
>> activated for the last entry (52/0x10000720), with:
>> - m_last_address == 0x1000071c,
>> - m_address == 0x10000720, and
>> - m_line == 52 (matching the DW_AT_call_line).
>>
>> It's easy to see that for the last entry, (m_flags & LEF_IS_STMT) == 0 holds,
>> because it doesn't have an x in the "Stmt" column.
>>
>> I found it less obvious that m_stmt_at_address also holds.
>>
>> [ The documentation clarifies that this is related to m_last_address:
>> ....
>> /* Set to true when a previous line at the same address (using
>> m_last_address) had LEF_IS_STMT set in m_flags. This is reset to false
>> when a line entry at a new address (m_address different to
>> m_last_address) is processed. */
>> bool m_stmt_at_address = false;
>> ...
>>
>> To get maximum clarity, I checked the value for each entry:
>> ...
>> address m_stmt_at_address
>> ---------------------------------
>> before false
>> 42/0x1000071c false->true
>> 43/0x1000071c/1 true
>> 43/0x1000071c/2 true
>> 52/0x1000071c/3 true
>> 52/0x10000720 true->false
>> ...
>>
>> Again it's easy to relate the transitions for particular entries to the "Stmt"
>> column.
>>
>> The tricky bit is that the transition takes place at the end of
>> lnp_state_machine::record_line, so while processing entry 52/0x10000720, we
>> sample m_stmt_at_address while it's still true. ]
>>
>> Likewise, the fix works for the second inlined call.
>>
>> But not for the third. The debug info has the same problem, but the fix is
>> not applied.
>>
>> The corresponding line info looks slightly different:
>> ...
>> File name Line number Starting address View Stmt
>> step-and-next-inline.cc 42 0x1000074c x
>> step-and-next-inline.cc 43 0x1000074c 1 x
>> step-and-next-inline.cc 43 0x1000074c 2
>> step-and-next-inline.cc 54 0x1000074c 3
>> step-and-next-inline.cc 55 0x10000750
>> ...
>>
>> In this case, dwarf_find_and_extend_inline_block_range gets called with:
>> - m_last_address == 0x1000074c,
>> - m_address == 0x10000750, and
>> - m_line == 55,
>> but since line 55 doesn't match DW_AT_call_line 54:
>> ...
>> <2><918>: Abbrev Number: 44 (DW_TAG_inlined_subroutine)
>> <919> DW_AT_abstract_origin: <0x967>
>> <91d> DW_AT_entry_pc : 0x10000740
>> <925> DW_AT_GNU_entry_view: 0
>> <926> DW_AT_low_pc : 0x10000740
>> <92e> DW_AT_high_pc : 0xc
>> <937> DW_AT_call_line : 54
>> ...
>> the fix is not applied.
>>
>> I'm proposing the following simple tweak, to handle the third inlined call as
>> well: instead of using m_line, use m_last_line.
>>
>> Tested on x86_64-linux and ppc64le-linux.
>>
>> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=33930
>
> Hi Tom,
>
> Thanks for looking at this failure. Can you confirm which gcc version
> you're using please. I don't see the same failure on cfarm29 (from gcc
> compile farm), using gcc 'gcc (Debian 14.2.0-19) 14.2.0'. But that
> step-and-next-inline test can change behaviour based on compiler
> version, so I'm not surprised you're seeing failures on some specific
> combinations.
Hi Andrew,
As stated here ( https://sourceware.org/bugzilla/show_bug.cgi?id=33930#c2 )
...
Used compiler version:
...
[vries@cfarm120 gdb]$ gcc --version
gcc (GCC) 11.5.0 20240719 (Red Hat 11.5.0-11)
...
So I'd try cfarm120.
Thanks,
- Tom
> Which is why I tried to write DWARF assembler tests to cover edge cases
> as I found them, and I think it would be great if we could get such a
> test to cover this fix too. I'm happy to help with, or even write, the
> test, once I can reproduce the failure.
>
> I'll try some other ppc64le machines I have access too, maybe one of
> those will have the right compiler version already installed, otherwise
> I'll have to rebuild gcc once you let me know which version is causing
> problems.
>
> Thanks,
> Andrew
>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-10 16:05 ` Andrew Burgess
2026-03-10 20:34 ` Tom de Vries
@ 2026-03-11 13:24 ` Tom de Vries
2026-03-11 15:13 ` Tom de Vries
1 sibling, 1 reply; 14+ messages in thread
From: Tom de Vries @ 2026-03-11 13:24 UTC (permalink / raw)
To: Andrew Burgess, gdb-patches
On 3/10/26 5:05 PM, Andrew Burgess wrote:
> Which is why I tried to write DWARF assembler tests to cover edge cases
> as I found them, and I think it would be great if we could get such a
> test to cover this fix too.
Agreed.
> I'm happy to help with, or even write, the
> test, once I can reproduce the failure.
My first try:
...
diff --git a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
b/gdb/testsuite/gdb.dwarf2/dw2-ext
end-inline-block.exp
index 9e4798b53f3..82d3f7e0310 100644
--- a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
+++ b/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
@@ -115,7 +115,7 @@ proc build_dwarf_for_contiguous_block { asm_file } {
DW_LNS_copy
DW_LNE_set_address main_3
- line $::foo_call_line
+ line [expr $::foo_call_line + 1]
DW_LNS_negate_stmt
DW_LNS_copy
...
doesn't seem to work (in the sense of causing failure with the fix
applied), I'll investigate a bit further to understand why.
Thanks,
- Tom
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-11 13:24 ` Tom de Vries
@ 2026-03-11 15:13 ` Tom de Vries
2026-03-12 6:45 ` Tom de Vries
0 siblings, 1 reply; 14+ messages in thread
From: Tom de Vries @ 2026-03-11 15:13 UTC (permalink / raw)
To: Andrew Burgess, gdb-patches
On 3/11/26 2:24 PM, Tom de Vries wrote:
> On 3/10/26 5:05 PM, Andrew Burgess wrote:
>> Which is why I tried to write DWARF assembler tests to cover edge cases
>> as I found them, and I think it would be great if we could get such a
>> test to cover this fix too.
>
> Agreed.
>
>> I'm happy to help with, or even write, the
>> test, once I can reproduce the failure.
>
> My first try:
> ...
> diff --git a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp b/gdb/
> testsuite/gdb.dwarf2/dw2-ext
> end-inline-block.exp
> index 9e4798b53f3..82d3f7e0310 100644
> --- a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
> +++ b/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
> @@ -115,7 +115,7 @@ proc build_dwarf_for_contiguous_block { asm_file } {
> DW_LNS_copy
>
> DW_LNE_set_address main_3
> - line $::foo_call_line
> + line [expr $::foo_call_line + 1]
> DW_LNS_negate_stmt
> DW_LNS_copy
> ...
> doesn't seem to work (in the sense of causing failure with the fix
> applied), I'll investigate a bit further to understand why.
OK, I think this:
...
diff --git a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
b/gdb/testsuite/gdb.dwarf2/dw2-ext
end-inline-block.exp
index 9e4798b53f3..4a88250d927 100644
--- a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
+++ b/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
@@ -120,10 +120,10 @@ proc build_dwarf_for_contiguous_block { asm_file } {
DW_LNS_copy
DW_LNE_set_address main_4
+ DW_LNS_advance_line 1
DW_LNS_copy
DW_LNE_set_address main_5
- DW_LNS_advance_line 1
DW_LNS_negate_stmt
DW_LNS_copy
...
is the patch that represents the behavior I was seeing.
The testcase doesn't completely pass though, I get:
...
(gdb) PASS: $exp: contiguous block: step to second line of foo
step^M
28 /* foo:3 */^M
(gdb) PASS: $exp: contiguous block: step to third line of foo
step^M
37 /* main:4 */^M
(gdb) FAIL: $exp: contiguous block: set back to main
...
so the message saying we're back in main is missing.
Thanks,
- Tom
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-11 15:13 ` Tom de Vries
@ 2026-03-12 6:45 ` Tom de Vries
2026-03-12 12:07 ` Tom de Vries
0 siblings, 1 reply; 14+ messages in thread
From: Tom de Vries @ 2026-03-12 6:45 UTC (permalink / raw)
To: Andrew Burgess, gdb-patches
On 3/11/26 4:13 PM, Tom de Vries wrote:
> On 3/11/26 2:24 PM, Tom de Vries wrote:
>> On 3/10/26 5:05 PM, Andrew Burgess wrote:
>>> Which is why I tried to write DWARF assembler tests to cover edge cases
>>> as I found them, and I think it would be great if we could get such a
>>> test to cover this fix too.
>>
>> Agreed.
>>
>>> I'm happy to help with, or even write, the
>>> test, once I can reproduce the failure.
>>
>> My first try:
>> ...
>> diff --git a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp b/
>> gdb/ testsuite/gdb.dwarf2/dw2-ext
>> end-inline-block.exp
>> index 9e4798b53f3..82d3f7e0310 100644
>> --- a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
>> +++ b/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
>> @@ -115,7 +115,7 @@ proc build_dwarf_for_contiguous_block { asm_file } {
>> DW_LNS_copy
>>
>> DW_LNE_set_address main_3
>> - line $::foo_call_line
>> + line [expr $::foo_call_line + 1]
>> DW_LNS_negate_stmt
>> DW_LNS_copy
>> ...
>> doesn't seem to work (in the sense of causing failure with the fix
>> applied), I'll investigate a bit further to understand why.
>
> OK, I think this:
> ...
> diff --git a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp b/gdb/
> testsuite/gdb.dwarf2/dw2-ext
> end-inline-block.exp
> index 9e4798b53f3..4a88250d927 100644
> --- a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
> +++ b/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
> @@ -120,10 +120,10 @@ proc build_dwarf_for_contiguous_block { asm_file } {
> DW_LNS_copy
>
> DW_LNE_set_address main_4
> + DW_LNS_advance_line 1
> DW_LNS_copy
>
> DW_LNE_set_address main_5
> - DW_LNS_advance_line 1
> DW_LNS_negate_stmt
> DW_LNS_copy
> ...
> is the patch that represents the behavior I was seeing.
>
> The testcase doesn't completely pass though, I get:
> ...
> (gdb) PASS: $exp: contiguous block: step to second line of foo
> step^M
> 28 /* foo:3 */^M
> (gdb) PASS: $exp: contiguous block: step to third line of foo
> step^M
> 37 /* main:4 */^M
> (gdb) FAIL: $exp: contiguous block: set back to main
> ...
> so the message saying we're back in main is missing.
>
I've filed a PR ( https://sourceware.org/bugzilla/show_bug.cgi?id=33981
) with some analysis.
Thanks,
- Tom
> Thanks,
> - Tom
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-12 6:45 ` Tom de Vries
@ 2026-03-12 12:07 ` Tom de Vries
0 siblings, 0 replies; 14+ messages in thread
From: Tom de Vries @ 2026-03-12 12:07 UTC (permalink / raw)
To: Andrew Burgess, gdb-patches
On 3/12/26 7:45 AM, Tom de Vries wrote:
> On 3/11/26 4:13 PM, Tom de Vries wrote:
>> On 3/11/26 2:24 PM, Tom de Vries wrote:
>>> On 3/10/26 5:05 PM, Andrew Burgess wrote:
>>>> Which is why I tried to write DWARF assembler tests to cover edge cases
>>>> as I found them, and I think it would be great if we could get such a
>>>> test to cover this fix too.
>>>
>>> Agreed.
>>>
>>>> I'm happy to help with, or even write, the
>>>> test, once I can reproduce the failure.
>>>
>>> My first try:
>>> ...
>>> diff --git a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp b/
>>> gdb/ testsuite/gdb.dwarf2/dw2-ext
>>> end-inline-block.exp
>>> index 9e4798b53f3..82d3f7e0310 100644
>>> --- a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
>>> +++ b/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
>>> @@ -115,7 +115,7 @@ proc build_dwarf_for_contiguous_block { asm_file } {
>>> DW_LNS_copy
>>>
>>> DW_LNE_set_address main_3
>>> - line $::foo_call_line
>>> + line [expr $::foo_call_line + 1]
>>> DW_LNS_negate_stmt
>>> DW_LNS_copy
>>> ...
>>> doesn't seem to work (in the sense of causing failure with the fix
>>> applied), I'll investigate a bit further to understand why.
>>
>> OK, I think this:
>> ...
>> diff --git a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp b/
>> gdb/ testsuite/gdb.dwarf2/dw2-ext
>> end-inline-block.exp
>> index 9e4798b53f3..4a88250d927 100644
>> --- a/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
>> +++ b/gdb/testsuite/gdb.dwarf2/dw2-extend-inline-block.exp
>> @@ -120,10 +120,10 @@ proc build_dwarf_for_contiguous_block
>> { asm_file } {
>> DW_LNS_copy
>>
>> DW_LNE_set_address main_4
>> + DW_LNS_advance_line 1
>> DW_LNS_copy
>>
>> DW_LNE_set_address main_5
>> - DW_LNS_advance_line 1
>> DW_LNS_negate_stmt
>> DW_LNS_copy
>> ...
>> is the patch that represents the behavior I was seeing.
>>
>> The testcase doesn't completely pass though, I get:
>> ...
>> (gdb) PASS: $exp: contiguous block: step to second line of foo
>> step^M
>> 28 /* foo:3 */^M
>> (gdb) PASS: $exp: contiguous block: step to third line of foo
>> step^M
>> 37 /* main:4 */^M
>> (gdb) FAIL: $exp: contiguous block: set back to main
>> ...
>> so the message saying we're back in main is missing.
>>
>
> I've filed a PR ( https://sourceware.org/bugzilla/show_bug.cgi?
> id=33981 ) with some analysis.
I've submitted a patch (
https://sourceware.org/pipermail/gdb-patches/2026-March/225916.html )
for this, which also contains a disabled test for the PR in this thread
(PR33930).
Thanks,
- Tom
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-10 20:34 ` Tom de Vries
@ 2026-03-14 14:22 ` Andrew Burgess
2026-03-14 14:53 ` Tom de Vries
0 siblings, 1 reply; 14+ messages in thread
From: Andrew Burgess @ 2026-03-14 14:22 UTC (permalink / raw)
To: Tom de Vries, gdb-patches
Tom de Vries <tdevries@suse.de> writes:
> On 3/10/26 5:05 PM, Andrew Burgess wrote:
>> Tom de Vries <tdevries@suse.de> writes:
>>
>>> Consider test-case gdb.cp/step-and-next-inline.exp on ppc64le-linux.
>>>
>>> The corresponding source file step-and-next-inline.cc contains functions
>>> tree_check and get_alias_set:
>>> ...
>>> 35 #define TREE_TYPE(NODE) (*tree_check (NODE, 0))
>>> 36
>>> 37 inline tree *
>>> 38 tree_check (tree *t, int i)
>>> 39 {
>>> 40 if (t->x != i)
>>> 41 abort();
>>> 42 tree *x = t;
>>> 43 return x;
>>> 44 }
>>> ...
>>> 48 int __attribute__((noinline, noclone))
>>> 49 get_alias_set (tree *t)
>>> 50 {
>>> 51 if (t != NULL
>>> 52 && TREE_TYPE (t).z != 1
>>> 53 && TREE_TYPE (t).z != 2
>>> 54 && TREE_TYPE (t).z != 3)
>>> 55 return 0;
>>> 56 return 1;
>>> 57 }
>>> ...
>>> as well as a trivial function main calling get_alias_set.
>>>
>>> Say we step into the first call to tree_check, and then step to the return
>>> at line 43:
>>> ...
>>> (gdb) s
>>> tree_check (i=0, t=0x10020030 <xx>) at step-and-next-inline.cc:40
>>> 40 if (t->x != i)
>>> (gdb) s
>>> 43 return x;
>>> (gdb)
>>> ...
>>>
>>> At that point, we have pc 0x1000071c:
>>> ...
>>> (gdb) p $pc
>>> $1 = (void (*)(void)) 0x1000071c <get_alias_set(tree*)+28>
>>> (gdb)
>>> ...
>>> and the backtrace looks like this:
>>> ...
>>> (gdb) bt
>>> #0 tree_check (i=<optimized out>, t=<optimized out>) at
>>> step-and-next-inline.cc:43
>>> #1 get_alias_set (t=t@entry=0x10020030 <xx>) at step-and-next-inline.cc:52
>>> #2 0x0000000010000560 in main () at step-and-next-inline.cc:64
>>> (gdb)
>>> ...
>>> which shows all 3 functions.
>>>
>>> This seems trivial, but it's not.
>>>
>>> All three calls to tree_check are inlined, and the first call is represented
>>> by:
>>> ...
>>> <2><877>: Abbrev Number: 40 (DW_TAG_inlined_subroutine)
>>> <878> DW_AT_abstract_origin: <0x967>
>>> <87c> DW_AT_entry_pc : 0x10000710
>>> <884> DW_AT_GNU_entry_view: 0
>>> <885> DW_AT_ranges : 0xc
>>> <88a> DW_AT_call_line : 52
>>> ...
>>> with DW_AT_ranges referring to:
>>> ...
>>> Contents of the .debug_rnglists section:
>>>
>>> Offset Begin End
>>> 0000000c 0000000010000710 (base address)
>>> 00000015 0000000010000710 000000001000071c
>>> 00000018 000000001000077c 000000001000077c (start == end)
>>> 0000001b 0000000010000788 0000000010000790
>>> 0000001f <End of list>
>>> ...
>>>
>>> The range at offset 0x15 is [0x10000710, 0x1000071c), so address 0x1000071c
>>> does not fall in the range, and consequently the debug info does not consider
>>> 0x1000071c part of the inlined tree_check.
>>>
>>> However, since commit 8efed40efd6 ("gdb: fix-up truncated inline function
>>> block ranges"), gdb contains a fix in lnp_state_machine::record_line:
>>> ...
>>> if (m_address != m_last_address
>>> && m_stmt_at_address
>>> && m_cu->producer_is_gcc ()
>>> && (m_flags & LEF_IS_STMT) == 0)
>>> dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
>>> m_address, m_line);
>>> ...
>>> that looks at the corresponding line number information:
>>> ...
>>> File name Line number Starting address View Stmt
>>> step-and-next-inline.cc 42 0x1000071c x
>>> step-and-next-inline.cc 43 0x1000071c 1 x
>>> step-and-next-inline.cc 43 0x1000071c 2
>>> step-and-next-inline.cc 52 0x1000071c 3
>>> step-and-next-inline.cc 52 0x10000720
>>> ...
>>> and extends the range of the inlined tree_check to include
>>> [0x1000071c, 0x10000720).
>>>
>>> [ Please read the commit message of aforementioned commit to understand why
>>> the fix is correct. ]
>>>
>>> Let's look in more detail at how the call to
>>> dwarf_find_and_extend_inline_block_range is activated for the fix. It's
>>> activated for the last entry (52/0x10000720), with:
>>> - m_last_address == 0x1000071c,
>>> - m_address == 0x10000720, and
>>> - m_line == 52 (matching the DW_AT_call_line).
>>>
>>> It's easy to see that for the last entry, (m_flags & LEF_IS_STMT) == 0 holds,
>>> because it doesn't have an x in the "Stmt" column.
>>>
>>> I found it less obvious that m_stmt_at_address also holds.
>>>
>>> [ The documentation clarifies that this is related to m_last_address:
>>> ....
>>> /* Set to true when a previous line at the same address (using
>>> m_last_address) had LEF_IS_STMT set in m_flags. This is reset to false
>>> when a line entry at a new address (m_address different to
>>> m_last_address) is processed. */
>>> bool m_stmt_at_address = false;
>>> ...
>>>
>>> To get maximum clarity, I checked the value for each entry:
>>> ...
>>> address m_stmt_at_address
>>> ---------------------------------
>>> before false
>>> 42/0x1000071c false->true
>>> 43/0x1000071c/1 true
>>> 43/0x1000071c/2 true
>>> 52/0x1000071c/3 true
>>> 52/0x10000720 true->false
>>> ...
>>>
>>> Again it's easy to relate the transitions for particular entries to the "Stmt"
>>> column.
>>>
>>> The tricky bit is that the transition takes place at the end of
>>> lnp_state_machine::record_line, so while processing entry 52/0x10000720, we
>>> sample m_stmt_at_address while it's still true. ]
>>>
>>> Likewise, the fix works for the second inlined call.
>>>
>>> But not for the third. The debug info has the same problem, but the fix is
>>> not applied.
>>>
>>> The corresponding line info looks slightly different:
>>> ...
>>> File name Line number Starting address View Stmt
>>> step-and-next-inline.cc 42 0x1000074c x
>>> step-and-next-inline.cc 43 0x1000074c 1 x
>>> step-and-next-inline.cc 43 0x1000074c 2
>>> step-and-next-inline.cc 54 0x1000074c 3
>>> step-and-next-inline.cc 55 0x10000750
>>> ...
>>>
>>> In this case, dwarf_find_and_extend_inline_block_range gets called with:
>>> - m_last_address == 0x1000074c,
>>> - m_address == 0x10000750, and
>>> - m_line == 55,
>>> but since line 55 doesn't match DW_AT_call_line 54:
>>> ...
>>> <2><918>: Abbrev Number: 44 (DW_TAG_inlined_subroutine)
>>> <919> DW_AT_abstract_origin: <0x967>
>>> <91d> DW_AT_entry_pc : 0x10000740
>>> <925> DW_AT_GNU_entry_view: 0
>>> <926> DW_AT_low_pc : 0x10000740
>>> <92e> DW_AT_high_pc : 0xc
>>> <937> DW_AT_call_line : 54
>>> ...
>>> the fix is not applied.
>>>
>>> I'm proposing the following simple tweak, to handle the third inlined call as
>>> well: instead of using m_line, use m_last_line.
>>>
>>> Tested on x86_64-linux and ppc64le-linux.
>>>
>>> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=33930
>>
>> Hi Tom,
>>
>> Thanks for looking at this failure. Can you confirm which gcc version
>> you're using please. I don't see the same failure on cfarm29 (from gcc
>> compile farm), using gcc 'gcc (Debian 14.2.0-19) 14.2.0'. But that
>> step-and-next-inline test can change behaviour based on compiler
>> version, so I'm not surprised you're seeing failures on some specific
>> combinations.
> Hi Andrew,
>
> As stated here ( https://sourceware.org/bugzilla/show_bug.cgi?id=33930#c2 )
> ...
> Used compiler version:
> ...
> [vries@cfarm120 gdb]$ gcc --version
> gcc (GCC) 11.5.0 20240719 (Red Hat 11.5.0-11)
> ...
>
> So I'd try cfarm120.
Thanks. Just wanted to say I've not forgotten this patch. I managed to
reproduce the failure, and I understand what's going on now. I'm going
to think about this some more next week.
Thanks,
Andrew
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-14 14:22 ` Andrew Burgess
@ 2026-03-14 14:53 ` Tom de Vries
0 siblings, 0 replies; 14+ messages in thread
From: Tom de Vries @ 2026-03-14 14:53 UTC (permalink / raw)
To: Andrew Burgess, gdb-patches
On 3/14/26 3:22 PM, Andrew Burgess wrote:
> Tom de Vries <tdevries@suse.de> writes:
>
>> On 3/10/26 5:05 PM, Andrew Burgess wrote:
>>> Tom de Vries <tdevries@suse.de> writes:
>>>
>>>> Consider test-case gdb.cp/step-and-next-inline.exp on ppc64le-linux.
>>>>
>>>> The corresponding source file step-and-next-inline.cc contains functions
>>>> tree_check and get_alias_set:
>>>> ...
>>>> 35 #define TREE_TYPE(NODE) (*tree_check (NODE, 0))
>>>> 36
>>>> 37 inline tree *
>>>> 38 tree_check (tree *t, int i)
>>>> 39 {
>>>> 40 if (t->x != i)
>>>> 41 abort();
>>>> 42 tree *x = t;
>>>> 43 return x;
>>>> 44 }
>>>> ...
>>>> 48 int __attribute__((noinline, noclone))
>>>> 49 get_alias_set (tree *t)
>>>> 50 {
>>>> 51 if (t != NULL
>>>> 52 && TREE_TYPE (t).z != 1
>>>> 53 && TREE_TYPE (t).z != 2
>>>> 54 && TREE_TYPE (t).z != 3)
>>>> 55 return 0;
>>>> 56 return 1;
>>>> 57 }
>>>> ...
>>>> as well as a trivial function main calling get_alias_set.
>>>>
>>>> Say we step into the first call to tree_check, and then step to the return
>>>> at line 43:
>>>> ...
>>>> (gdb) s
>>>> tree_check (i=0, t=0x10020030 <xx>) at step-and-next-inline.cc:40
>>>> 40 if (t->x != i)
>>>> (gdb) s
>>>> 43 return x;
>>>> (gdb)
>>>> ...
>>>>
>>>> At that point, we have pc 0x1000071c:
>>>> ...
>>>> (gdb) p $pc
>>>> $1 = (void (*)(void)) 0x1000071c <get_alias_set(tree*)+28>
>>>> (gdb)
>>>> ...
>>>> and the backtrace looks like this:
>>>> ...
>>>> (gdb) bt
>>>> #0 tree_check (i=<optimized out>, t=<optimized out>) at
>>>> step-and-next-inline.cc:43
>>>> #1 get_alias_set (t=t@entry=0x10020030 <xx>) at step-and-next-inline.cc:52
>>>> #2 0x0000000010000560 in main () at step-and-next-inline.cc:64
>>>> (gdb)
>>>> ...
>>>> which shows all 3 functions.
>>>>
>>>> This seems trivial, but it's not.
>>>>
>>>> All three calls to tree_check are inlined, and the first call is represented
>>>> by:
>>>> ...
>>>> <2><877>: Abbrev Number: 40 (DW_TAG_inlined_subroutine)
>>>> <878> DW_AT_abstract_origin: <0x967>
>>>> <87c> DW_AT_entry_pc : 0x10000710
>>>> <884> DW_AT_GNU_entry_view: 0
>>>> <885> DW_AT_ranges : 0xc
>>>> <88a> DW_AT_call_line : 52
>>>> ...
>>>> with DW_AT_ranges referring to:
>>>> ...
>>>> Contents of the .debug_rnglists section:
>>>>
>>>> Offset Begin End
>>>> 0000000c 0000000010000710 (base address)
>>>> 00000015 0000000010000710 000000001000071c
>>>> 00000018 000000001000077c 000000001000077c (start == end)
>>>> 0000001b 0000000010000788 0000000010000790
>>>> 0000001f <End of list>
>>>> ...
>>>>
>>>> The range at offset 0x15 is [0x10000710, 0x1000071c), so address 0x1000071c
>>>> does not fall in the range, and consequently the debug info does not consider
>>>> 0x1000071c part of the inlined tree_check.
>>>>
>>>> However, since commit 8efed40efd6 ("gdb: fix-up truncated inline function
>>>> block ranges"), gdb contains a fix in lnp_state_machine::record_line:
>>>> ...
>>>> if (m_address != m_last_address
>>>> && m_stmt_at_address
>>>> && m_cu->producer_is_gcc ()
>>>> && (m_flags & LEF_IS_STMT) == 0)
>>>> dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
>>>> m_address, m_line);
>>>> ...
>>>> that looks at the corresponding line number information:
>>>> ...
>>>> File name Line number Starting address View Stmt
>>>> step-and-next-inline.cc 42 0x1000071c x
>>>> step-and-next-inline.cc 43 0x1000071c 1 x
>>>> step-and-next-inline.cc 43 0x1000071c 2
>>>> step-and-next-inline.cc 52 0x1000071c 3
>>>> step-and-next-inline.cc 52 0x10000720
>>>> ...
>>>> and extends the range of the inlined tree_check to include
>>>> [0x1000071c, 0x10000720).
>>>>
>>>> [ Please read the commit message of aforementioned commit to understand why
>>>> the fix is correct. ]
>>>>
>>>> Let's look in more detail at how the call to
>>>> dwarf_find_and_extend_inline_block_range is activated for the fix. It's
>>>> activated for the last entry (52/0x10000720), with:
>>>> - m_last_address == 0x1000071c,
>>>> - m_address == 0x10000720, and
>>>> - m_line == 52 (matching the DW_AT_call_line).
>>>>
>>>> It's easy to see that for the last entry, (m_flags & LEF_IS_STMT) == 0 holds,
>>>> because it doesn't have an x in the "Stmt" column.
>>>>
>>>> I found it less obvious that m_stmt_at_address also holds.
>>>>
>>>> [ The documentation clarifies that this is related to m_last_address:
>>>> ....
>>>> /* Set to true when a previous line at the same address (using
>>>> m_last_address) had LEF_IS_STMT set in m_flags. This is reset to false
>>>> when a line entry at a new address (m_address different to
>>>> m_last_address) is processed. */
>>>> bool m_stmt_at_address = false;
>>>> ...
>>>>
>>>> To get maximum clarity, I checked the value for each entry:
>>>> ...
>>>> address m_stmt_at_address
>>>> ---------------------------------
>>>> before false
>>>> 42/0x1000071c false->true
>>>> 43/0x1000071c/1 true
>>>> 43/0x1000071c/2 true
>>>> 52/0x1000071c/3 true
>>>> 52/0x10000720 true->false
>>>> ...
>>>>
>>>> Again it's easy to relate the transitions for particular entries to the "Stmt"
>>>> column.
>>>>
>>>> The tricky bit is that the transition takes place at the end of
>>>> lnp_state_machine::record_line, so while processing entry 52/0x10000720, we
>>>> sample m_stmt_at_address while it's still true. ]
>>>>
>>>> Likewise, the fix works for the second inlined call.
>>>>
>>>> But not for the third. The debug info has the same problem, but the fix is
>>>> not applied.
>>>>
>>>> The corresponding line info looks slightly different:
>>>> ...
>>>> File name Line number Starting address View Stmt
>>>> step-and-next-inline.cc 42 0x1000074c x
>>>> step-and-next-inline.cc 43 0x1000074c 1 x
>>>> step-and-next-inline.cc 43 0x1000074c 2
>>>> step-and-next-inline.cc 54 0x1000074c 3
>>>> step-and-next-inline.cc 55 0x10000750
>>>> ...
>>>>
>>>> In this case, dwarf_find_and_extend_inline_block_range gets called with:
>>>> - m_last_address == 0x1000074c,
>>>> - m_address == 0x10000750, and
>>>> - m_line == 55,
>>>> but since line 55 doesn't match DW_AT_call_line 54:
>>>> ...
>>>> <2><918>: Abbrev Number: 44 (DW_TAG_inlined_subroutine)
>>>> <919> DW_AT_abstract_origin: <0x967>
>>>> <91d> DW_AT_entry_pc : 0x10000740
>>>> <925> DW_AT_GNU_entry_view: 0
>>>> <926> DW_AT_low_pc : 0x10000740
>>>> <92e> DW_AT_high_pc : 0xc
>>>> <937> DW_AT_call_line : 54
>>>> ...
>>>> the fix is not applied.
>>>>
>>>> I'm proposing the following simple tweak, to handle the third inlined call as
>>>> well: instead of using m_line, use m_last_line.
>>>>
>>>> Tested on x86_64-linux and ppc64le-linux.
>>>>
>>>> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=33930
>>>
>>> Hi Tom,
>>>
>>> Thanks for looking at this failure. Can you confirm which gcc version
>>> you're using please. I don't see the same failure on cfarm29 (from gcc
>>> compile farm), using gcc 'gcc (Debian 14.2.0-19) 14.2.0'. But that
>>> step-and-next-inline test can change behaviour based on compiler
>>> version, so I'm not surprised you're seeing failures on some specific
>>> combinations.
>> Hi Andrew,
>>
>> As stated here ( https://sourceware.org/bugzilla/show_bug.cgi?id=33930#c2 )
>> ...
>> Used compiler version:
>> ...
>> [vries@cfarm120 gdb]$ gcc --version
>> gcc (GCC) 11.5.0 20240719 (Red Hat 11.5.0-11)
>> ...
>>
>> So I'd try cfarm120.
>
> Thanks. Just wanted to say I've not forgotten this patch. I managed to
> reproduce the failure, and I understand what's going on now. I'm going
> to think about this some more next week.
>
Great, thanks for the update.
- Tom
> Thanks,
> Andrew
>
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-02 11:48 ` [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges Tom de Vries
2026-03-10 16:05 ` Andrew Burgess
@ 2026-03-19 15:39 ` Andrew Burgess
2026-04-20 14:38 ` Tom de Vries
1 sibling, 1 reply; 14+ messages in thread
From: Andrew Burgess @ 2026-03-19 15:39 UTC (permalink / raw)
To: Tom de Vries, gdb-patches
Hi Tom,
Thanks for looking at this.
Tom de Vries <tdevries@suse.de> writes:
> Consider test-case gdb.cp/step-and-next-inline.exp on ppc64le-linux.
>
> The corresponding source file step-and-next-inline.cc contains functions
> tree_check and get_alias_set:
> ...
> 35 #define TREE_TYPE(NODE) (*tree_check (NODE, 0))
> 36
> 37 inline tree *
> 38 tree_check (tree *t, int i)
> 39 {
> 40 if (t->x != i)
> 41 abort();
> 42 tree *x = t;
> 43 return x;
> 44 }
> ...
> 48 int __attribute__((noinline, noclone))
> 49 get_alias_set (tree *t)
> 50 {
> 51 if (t != NULL
> 52 && TREE_TYPE (t).z != 1
> 53 && TREE_TYPE (t).z != 2
> 54 && TREE_TYPE (t).z != 3)
> 55 return 0;
> 56 return 1;
> 57 }
> ...
> as well as a trivial function main calling get_alias_set.
>
> Say we step into the first call to tree_check, and then step to the return
> at line 43:
> ...
> (gdb) s
> tree_check (i=0, t=0x10020030 <xx>) at step-and-next-inline.cc:40
> 40 if (t->x != i)
> (gdb) s
> 43 return x;
> (gdb)
> ...
>
> At that point, we have pc 0x1000071c:
> ...
> (gdb) p $pc
> $1 = (void (*)(void)) 0x1000071c <get_alias_set(tree*)+28>
> (gdb)
> ...
> and the backtrace looks like this:
> ...
> (gdb) bt
> #0 tree_check (i=<optimized out>, t=<optimized out>) at
> step-and-next-inline.cc:43
> #1 get_alias_set (t=t@entry=0x10020030 <xx>) at step-and-next-inline.cc:52
> #2 0x0000000010000560 in main () at step-and-next-inline.cc:64
> (gdb)
> ...
> which shows all 3 functions.
>
> This seems trivial, but it's not.
>
> All three calls to tree_check are inlined, and the first call is represented
> by:
> ...
> <2><877>: Abbrev Number: 40 (DW_TAG_inlined_subroutine)
> <878> DW_AT_abstract_origin: <0x967>
> <87c> DW_AT_entry_pc : 0x10000710
> <884> DW_AT_GNU_entry_view: 0
> <885> DW_AT_ranges : 0xc
> <88a> DW_AT_call_line : 52
> ...
> with DW_AT_ranges referring to:
> ...
> Contents of the .debug_rnglists section:
>
> Offset Begin End
> 0000000c 0000000010000710 (base address)
> 00000015 0000000010000710 000000001000071c
> 00000018 000000001000077c 000000001000077c (start == end)
> 0000001b 0000000010000788 0000000010000790
> 0000001f <End of list>
> ...
>
> The range at offset 0x15 is [0x10000710, 0x1000071c), so address 0x1000071c
> does not fall in the range, and consequently the debug info does not consider
> 0x1000071c part of the inlined tree_check.
>
> However, since commit 8efed40efd6 ("gdb: fix-up truncated inline function
> block ranges"), gdb contains a fix in lnp_state_machine::record_line:
> ...
> if (m_address != m_last_address
> && m_stmt_at_address
> && m_cu->producer_is_gcc ()
> && (m_flags & LEF_IS_STMT) == 0)
> dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
> m_address, m_line);
> ...
> that looks at the corresponding line number information:
> ...
> File name Line number Starting address View Stmt
> step-and-next-inline.cc 42 0x1000071c x
> step-and-next-inline.cc 43 0x1000071c 1 x
> step-and-next-inline.cc 43 0x1000071c 2
> step-and-next-inline.cc 52 0x1000071c 3
> step-and-next-inline.cc 52 0x10000720
> ...
> and extends the range of the inlined tree_check to include
> [0x1000071c, 0x10000720).
>
> [ Please read the commit message of aforementioned commit to understand why
> the fix is correct. ]
I think it's worth discussing the idea behind the aforementioned commit
at this point because it is the reason behind my later thoughts.
What that commit does is modify the debug information. It has GDB
trying to decide that it, GDB, knows better than the compiler, what the
debug information should look like. This is always going to come with a
degree of risk. To restrictive and the "fix" will not trigger often
enough to provide real benefits to the user. To loose and the "fix"
ends up triggering in cases where it shouldn't, making the debug
experience worse.
The pattern I observed, and sort-of tried to match (more on this below),
is as follows:
1. A series of line entries at the end address for (one of) an inline
function's ranges.
2. The first of these line entries is a STMT, while the subsequent
entries are non-STMT.
3. The last of these line entries is a non-STMT and is associated with
the calling line of the inline function.
4. The next line entry is a non-STMT, and is also associated with the
calling line of the function.
Now I said above that I 'sort-of' tried to match this pattern.
Honestly, I don't think I did a great job, only some of these
characteristics are actually matched. Lets go through and see which
characteristics are, or are not actually checked for:
1. This is checked inside dwarf_find_and_extend_inline_block_range
where we check if original_address matches the end address of an
inline function range.
2. For this we check m_stmt_at_address, but this isn't exactly
correct. If _any_ of the line entries at the previous address are
marked as STMT then this will be true (and count at a match), this
is probably OK in most cases except for the next point....
3. We don't really check this at all; as you point out we check the
current line, not the previous one, and m_stmt_at_address will be
true if the previous entry is a STMT, which I think would be bad.
4. We do check the non-STMT part, and we check the line number part.
So I was a little sloppy with the STMT checking, and should have taken
care to add both line number checks.
I mention this because at this point your commit message seem (IMHO) to
focus on the is-stmt logic even though that's not really the part you
end up fixing.
>
> Let's look in more detail at how the call to
> dwarf_find_and_extend_inline_block_range is activated for the fix. It's
> activated for the last entry (52/0x10000720), with:
> - m_last_address == 0x1000071c,
> - m_address == 0x10000720, and
> - m_line == 52 (matching the DW_AT_call_line).
>
> It's easy to see that for the last entry, (m_flags & LEF_IS_STMT) == 0 holds,
> because it doesn't have an x in the "Stmt" column.
>
> I found it less obvious that m_stmt_at_address also holds.
>
> [ The documentation clarifies that this is related to m_last_address:
> ....
> /* Set to true when a previous line at the same address (using
> m_last_address) had LEF_IS_STMT set in m_flags. This is reset to false
> when a line entry at a new address (m_address different to
> m_last_address) is processed. */
> bool m_stmt_at_address = false;
> ...
>
> To get maximum clarity, I checked the value for each entry:
> ...
> address m_stmt_at_address
> ---------------------------------
> before false
> 42/0x1000071c false->true
> 43/0x1000071c/1 true
> 43/0x1000071c/2 true
> 52/0x1000071c/3 true
> 52/0x10000720 true->false
> ...
>
> Again it's easy to relate the transitions for particular entries to the "Stmt"
> column.
>
> The tricky bit is that the transition takes place at the end of
> lnp_state_machine::record_line, so while processing entry 52/0x10000720, we
> sample m_stmt_at_address while it's still true. ]
>
> Likewise, the fix works for the second inlined call.
>
> But not for the third. The debug info has the same problem, but the fix is
> not applied.
>
> The corresponding line info looks slightly different:
> ...
> File name Line number Starting address View Stmt
> step-and-next-inline.cc 42 0x1000074c x
> step-and-next-inline.cc 43 0x1000074c 1 x
> step-and-next-inline.cc 43 0x1000074c 2
> step-and-next-inline.cc 54 0x1000074c 3
> step-and-next-inline.cc 55 0x10000750
> ...
>
> In this case, dwarf_find_and_extend_inline_block_range gets called with:
> - m_last_address == 0x1000074c,
> - m_address == 0x10000750, and
> - m_line == 55,
> but since line 55 doesn't match DW_AT_call_line 54:
Your discussion of the is-stmt handling / tracking is all correct, but
the cause of the problem is captured in these last 5 lines, the next
line is not associated with the function's calling line, but with the
next source line.
> ...
> <2><918>: Abbrev Number: 44 (DW_TAG_inlined_subroutine)
> <919> DW_AT_abstract_origin: <0x967>
> <91d> DW_AT_entry_pc : 0x10000740
> <925> DW_AT_GNU_entry_view: 0
> <926> DW_AT_low_pc : 0x10000740
> <92e> DW_AT_high_pc : 0xc
> <937> DW_AT_call_line : 54
> ...
> the fix is not applied.
>
> I'm proposing the following simple tweak, to handle the third inlined call as
> well: instead of using m_line, use m_last_line.
If I'd done my job better when I wrote the original commit then we would
have been checking BOTH lines already, and requiring that they both
match the calling line; that after all was the original pattern I
spotted.
So in my mind the question is, if I had done a better job of matching
the pattern, would it be OK to loosen that matching? I'm a little
conflicted on this, but I think it's probably worth the risk.
>
> Tested on x86_64-linux and ppc64le-linux.
>
> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=33930
> ---
> gdb/dwarf2/line-program.c | 15 ++++++++++++++-
> 1 file changed, 14 insertions(+), 1 deletion(-)
>
> diff --git a/gdb/dwarf2/line-program.c b/gdb/dwarf2/line-program.c
> index b9d93bb7b0e..5aa97ea3714 100644
> --- a/gdb/dwarf2/line-program.c
> +++ b/gdb/dwarf2/line-program.c
> @@ -445,12 +445,25 @@ lnp_state_machine::record_line (bool end_sequence)
> (end_sequence ? "\t(end sequence)" : ""));
> }
>
> + /* Activate dwarf_find_and_extend_inline_block_range for line number info:
> +
> + Line number Starting address View Stmt
> + 42 0x1000074c x
> + 43 0x1000074c 1 x
> + 43 0x1000074c 2
> + 54 0x1000074c 3
> + 55 0x10000750
> +
> + at entry 55/0x10000750 with:
> + - m_last_address == 0x1000074c
> + - m_address == 0x10000750
> + - m_last_line == 54. */
> if (m_address != m_last_address
> && m_stmt_at_address
> && m_cu->producer_is_gcc ()
> && (m_flags & LEF_IS_STMT) == 0)
I wonder if should extend the condition with `&& m_line >= m_last_line`?
Under my original scheme I _should_ have been checking `& m_line ==
m_last_line` and your change is to relax that constraint to `>=`.
> dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
> - m_address, m_line);
> + m_address, m_last_line);
>
This patch will need rebasing onto your frame printing patch once that's
merged, and I guess you'll be enabling the additional tests that are in
that patch and currently disabled so that this commit gets some tests.
But I think this change is the right way to go, we just need to be
mindful that as we relax the match criteria we run the risk of having
GDB change the debug info in non-helpful ways.
Thanks,
Andrew
^ permalink raw reply [flat|nested] 14+ messages in thread
* Re: [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges
2026-03-19 15:39 ` Andrew Burgess
@ 2026-04-20 14:38 ` Tom de Vries
0 siblings, 0 replies; 14+ messages in thread
From: Tom de Vries @ 2026-04-20 14:38 UTC (permalink / raw)
To: Andrew Burgess, gdb-patches
On 3/19/26 4:39 PM, Andrew Burgess wrote:
>
> Hi Tom,
>
> Thanks for looking at this.
>
Hi Andrew,
thanks for the review.
> Tom de Vries <tdevries@suse.de> writes:
>
>> Consider test-case gdb.cp/step-and-next-inline.exp on ppc64le-linux.
>>
>> The corresponding source file step-and-next-inline.cc contains functions
>> tree_check and get_alias_set:
>> ...
>> 35 #define TREE_TYPE(NODE) (*tree_check (NODE, 0))
>> 36
>> 37 inline tree *
>> 38 tree_check (tree *t, int i)
>> 39 {
>> 40 if (t->x != i)
>> 41 abort();
>> 42 tree *x = t;
>> 43 return x;
>> 44 }
>> ...
>> 48 int __attribute__((noinline, noclone))
>> 49 get_alias_set (tree *t)
>> 50 {
>> 51 if (t != NULL
>> 52 && TREE_TYPE (t).z != 1
>> 53 && TREE_TYPE (t).z != 2
>> 54 && TREE_TYPE (t).z != 3)
>> 55 return 0;
>> 56 return 1;
>> 57 }
>> ...
>> as well as a trivial function main calling get_alias_set.
>>
>> Say we step into the first call to tree_check, and then step to the return
>> at line 43:
>> ...
>> (gdb) s
>> tree_check (i=0, t=0x10020030 <xx>) at step-and-next-inline.cc:40
>> 40 if (t->x != i)
>> (gdb) s
>> 43 return x;
>> (gdb)
>> ...
>>
>> At that point, we have pc 0x1000071c:
>> ...
>> (gdb) p $pc
>> $1 = (void (*)(void)) 0x1000071c <get_alias_set(tree*)+28>
>> (gdb)
>> ...
>> and the backtrace looks like this:
>> ...
>> (gdb) bt
>> #0 tree_check (i=<optimized out>, t=<optimized out>) at
>> step-and-next-inline.cc:43
>> #1 get_alias_set (t=t@entry=0x10020030 <xx>) at step-and-next-inline.cc:52
>> #2 0x0000000010000560 in main () at step-and-next-inline.cc:64
>> (gdb)
>> ...
>> which shows all 3 functions.
>>
>> This seems trivial, but it's not.
>>
>> All three calls to tree_check are inlined, and the first call is represented
>> by:
>> ...
>> <2><877>: Abbrev Number: 40 (DW_TAG_inlined_subroutine)
>> <878> DW_AT_abstract_origin: <0x967>
>> <87c> DW_AT_entry_pc : 0x10000710
>> <884> DW_AT_GNU_entry_view: 0
>> <885> DW_AT_ranges : 0xc
>> <88a> DW_AT_call_line : 52
>> ...
>> with DW_AT_ranges referring to:
>> ...
>> Contents of the .debug_rnglists section:
>>
>> Offset Begin End
>> 0000000c 0000000010000710 (base address)
>> 00000015 0000000010000710 000000001000071c
>> 00000018 000000001000077c 000000001000077c (start == end)
>> 0000001b 0000000010000788 0000000010000790
>> 0000001f <End of list>
>> ...
>>
>> The range at offset 0x15 is [0x10000710, 0x1000071c), so address 0x1000071c
>> does not fall in the range, and consequently the debug info does not consider
>> 0x1000071c part of the inlined tree_check.
>>
>> However, since commit 8efed40efd6 ("gdb: fix-up truncated inline function
>> block ranges"), gdb contains a fix in lnp_state_machine::record_line:
>> ...
>> if (m_address != m_last_address
>> && m_stmt_at_address
>> && m_cu->producer_is_gcc ()
>> && (m_flags & LEF_IS_STMT) == 0)
>> dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
>> m_address, m_line);
>> ...
>> that looks at the corresponding line number information:
>> ...
>> File name Line number Starting address View Stmt
>> step-and-next-inline.cc 42 0x1000071c x
>> step-and-next-inline.cc 43 0x1000071c 1 x
>> step-and-next-inline.cc 43 0x1000071c 2
>> step-and-next-inline.cc 52 0x1000071c 3
>> step-and-next-inline.cc 52 0x10000720
>> ...
>> and extends the range of the inlined tree_check to include
>> [0x1000071c, 0x10000720).
>>
>> [ Please read the commit message of aforementioned commit to understand why
>> the fix is correct. ]
>
> I think it's worth discussing the idea behind the aforementioned commit
> at this point because it is the reason behind my later thoughts.
>
> What that commit does is modify the debug information. It has GDB
> trying to decide that it, GDB, knows better than the compiler, what the
> debug information should look like. This is always going to come with a
> degree of risk. To restrictive and the "fix" will not trigger often
> enough to provide real benefits to the user. To loose and the "fix"
> ends up triggering in cases where it shouldn't, making the debug
> experience worse.
>
Ack.
> The pattern I observed, and sort-of tried to match (more on this below),
> is as follows:
>
> 1. A series of line entries at the end address for (one of) an inline
> function's ranges.
>
> 2. The first of these line entries is a STMT, while the subsequent
> entries are non-STMT.
>
> 3. The last of these line entries is a non-STMT and is associated with
> the calling line of the inline function.
>
> 4. The next line entry is a non-STMT, and is also associated with the
> calling line of the function.
>
Thanks for spelling this out, this is helpful for me.
I understand that you're explaining here what you tried to match, but I
wonder if the 4th item is indeed necessary. That is, we're trying to
match something like this:
...
Line number Starting address View Stmt
42 0x1000071c x
43 0x1000071c 1 x
43 0x1000071c 2
52 0x1000071c 3
52 0x10000720
...
but if you annotate each entry using the following address like this:
...
Line number Address range View Stmt
42 [0x1000071c,0x1000071c) x
43 [0x1000071c,0x1000071c) 1 x
43 [0x1000071c,0x1000071c) 2
52 [0x1000071c,0x10000720) 3
52 [0x10000720,...)
...
then would we still look at the last item at all? ISTM we just need the
address.
> Now I said above that I 'sort-of' tried to match this pattern.
> Honestly, I don't think I did a great job, only some of these
> characteristics are actually matched. Lets go through and see which
> characteristics are, or are not actually checked for:
>
> 1. This is checked inside dwarf_find_and_extend_inline_block_range
> where we check if original_address matches the end address of an
> inline function range.
>
> 2. For this we check m_stmt_at_address, but this isn't exactly
> correct. If _any_ of the line entries at the previous address are
> marked as STMT then this will be true (and count at a match), this
> is probably OK in most cases except for the next point....
>
> 3. We don't really check this at all; as you point out we check the
> current line, not the previous one, and m_stmt_at_address will be
> true if the previous entry is a STMT, which I think would be bad.
We could add an m_last_is_stmt field, and use that.
>
> 4. We do check the non-STMT part, and we check the line number part.
>
> So I was a little sloppy with the STMT checking, and should have taken
> care to add both line number checks.
>
> I mention this because at this point your commit message seem (IMHO) to
> focus on the is-stmt logic even though that's not really the part you
> end up fixing.
>
Yes, the part about the is-stmt was merely a reflection of my difficulty
in understanding it. I've submitted a v2 that has two refactoring
patches added that focus on that part, leaving the commit message of
this patch much cleaner, I hope.
>>
>> Let's look in more detail at how the call to
>> dwarf_find_and_extend_inline_block_range is activated for the fix. It's
>> activated for the last entry (52/0x10000720), with:
>> - m_last_address == 0x1000071c,
>> - m_address == 0x10000720, and
>> - m_line == 52 (matching the DW_AT_call_line).
>>
>> It's easy to see that for the last entry, (m_flags & LEF_IS_STMT) == 0 holds,
>> because it doesn't have an x in the "Stmt" column.
>>
>> I found it less obvious that m_stmt_at_address also holds.
>>
>> [ The documentation clarifies that this is related to m_last_address:
>> ....
>> /* Set to true when a previous line at the same address (using
>> m_last_address) had LEF_IS_STMT set in m_flags. This is reset to false
>> when a line entry at a new address (m_address different to
>> m_last_address) is processed. */
>> bool m_stmt_at_address = false;
>> ...
>>
>> To get maximum clarity, I checked the value for each entry:
>> ...
>> address m_stmt_at_address
>> ---------------------------------
>> before false
>> 42/0x1000071c false->true
>> 43/0x1000071c/1 true
>> 43/0x1000071c/2 true
>> 52/0x1000071c/3 true
>> 52/0x10000720 true->false
>> ...
>>
>> Again it's easy to relate the transitions for particular entries to the "Stmt"
>> column.
>>
>> The tricky bit is that the transition takes place at the end of
>> lnp_state_machine::record_line, so while processing entry 52/0x10000720, we
>> sample m_stmt_at_address while it's still true. ]
>>
>> Likewise, the fix works for the second inlined call.
>>
>> But not for the third. The debug info has the same problem, but the fix is
>> not applied.
>>
>> The corresponding line info looks slightly different:
>> ...
>> File name Line number Starting address View Stmt
>> step-and-next-inline.cc 42 0x1000074c x
>> step-and-next-inline.cc 43 0x1000074c 1 x
>> step-and-next-inline.cc 43 0x1000074c 2
>> step-and-next-inline.cc 54 0x1000074c 3
>> step-and-next-inline.cc 55 0x10000750
>> ...
>>
>> In this case, dwarf_find_and_extend_inline_block_range gets called with:
>> - m_last_address == 0x1000074c,
>> - m_address == 0x10000750, and
>> - m_line == 55,
>> but since line 55 doesn't match DW_AT_call_line 54:
>
> Your discussion of the is-stmt handling / tracking is all correct, but
> the cause of the problem is captured in these last 5 lines, the next
> line is not associated with the function's calling line, but with the
> next source line.
>
>> ...
>> <2><918>: Abbrev Number: 44 (DW_TAG_inlined_subroutine)
>> <919> DW_AT_abstract_origin: <0x967>
>> <91d> DW_AT_entry_pc : 0x10000740
>> <925> DW_AT_GNU_entry_view: 0
>> <926> DW_AT_low_pc : 0x10000740
>> <92e> DW_AT_high_pc : 0xc
>> <937> DW_AT_call_line : 54
>> ...
>> the fix is not applied.
>>
>> I'm proposing the following simple tweak, to handle the third inlined call as
>> well: instead of using m_line, use m_last_line.
>
> If I'd done my job better when I wrote the original commit then we would
> have been checking BOTH lines already, and requiring that they both
> match the calling line; that after all was the original pattern I
> spotted.
>
> So in my mind the question is, if I had done a better job of matching
> the pattern, would it be OK to loosen that matching? I'm a little
> conflicted on this, but I think it's probably worth the risk.
>
>>
>> Tested on x86_64-linux and ppc64le-linux.
>>
>> Bug: https://sourceware.org/bugzilla/show_bug.cgi?id=33930
>> ---
>> gdb/dwarf2/line-program.c | 15 ++++++++++++++-
>> 1 file changed, 14 insertions(+), 1 deletion(-)
>>
>> diff --git a/gdb/dwarf2/line-program.c b/gdb/dwarf2/line-program.c
>> index b9d93bb7b0e..5aa97ea3714 100644
>> --- a/gdb/dwarf2/line-program.c
>> +++ b/gdb/dwarf2/line-program.c
>> @@ -445,12 +445,25 @@ lnp_state_machine::record_line (bool end_sequence)
>> (end_sequence ? "\t(end sequence)" : ""));
>> }
>>
>> + /* Activate dwarf_find_and_extend_inline_block_range for line number info:
>> +
>> + Line number Starting address View Stmt
>> + 42 0x1000074c x
>> + 43 0x1000074c 1 x
>> + 43 0x1000074c 2
>> + 54 0x1000074c 3
>> + 55 0x10000750
>> +
>> + at entry 55/0x10000750 with:
>> + - m_last_address == 0x1000074c
>> + - m_address == 0x10000750
>> + - m_last_line == 54. */
>> if (m_address != m_last_address
>> && m_stmt_at_address
>> && m_cu->producer_is_gcc ()
>> && (m_flags & LEF_IS_STMT) == 0)
>
> I wonder if should extend the condition with `&& m_line >= m_last_line`?
>
As I tried to explain above, I don't think so, and having said that, I
think we could probably also drop the '(m_flags & LEF_IS_STMT) == 0'
part. Haven't done so in a v2 though.
> Under my original scheme I _should_ have been checking `& m_line ==
> m_last_line` and your change is to relax that constraint to `>=`.
>
>> dwarf_find_and_extend_inline_block_range (m_cu, m_last_address,
>> - m_address, m_line);
>> + m_address, m_last_line);
>>
>
> This patch will need rebasing onto your frame printing patch once that's
> merged, and I guess you'll be enabling the additional tests that are in
> that patch and currently disabled so that this commit gets some tests.
>
> But I think this change is the right way to go, we just need to be
> mindful that as we relax the match criteria we run the risk of having
> GDB change the debug info in non-helpful ways.
>
Agreed.
V2 submitted here (
https://sourceware.org/pipermail/gdb-patches/2026-April/226646.html ).
Thanks,
- Tom
> Thanks,
> Andrew
>
^ permalink raw reply [flat|nested] 14+ messages in thread
end of thread, other threads:[~2026-04-20 14:39 UTC | newest]
Thread overview: 14+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-03-02 11:48 [PATCH 0/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges Tom de Vries
2026-03-02 11:48 ` [PATCH 1/3] [gdb/symtab] Rename lnp_state_machine::m_last_line to m_last_recorded_line Tom de Vries
2026-03-02 11:48 ` [PATCH 2/3] [gdb/symtab] Re-add lnp_state_machine::m_last_line Tom de Vries
2026-03-02 11:48 ` [PATCH 3/3] [gdb/symtab] Tweak fix-up of truncated inline function block ranges Tom de Vries
2026-03-10 16:05 ` Andrew Burgess
2026-03-10 20:34 ` Tom de Vries
2026-03-14 14:22 ` Andrew Burgess
2026-03-14 14:53 ` Tom de Vries
2026-03-11 13:24 ` Tom de Vries
2026-03-11 15:13 ` Tom de Vries
2026-03-12 6:45 ` Tom de Vries
2026-03-12 12:07 ` Tom de Vries
2026-03-19 15:39 ` Andrew Burgess
2026-04-20 14:38 ` Tom de Vries
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox