public inbox for gdb-prs@sourceware.org
help / color / mirror / Atom feed
* [Bug gdb/30319] New: GDB shows "Incompatible DWARF Stack" when it's actually compatible
@ 2023-04-06 6:13 hluaw at connect dot ust.hk
2023-04-13 14:39 ` [Bug gdb/30319] " tromey at sourceware dot org
0 siblings, 1 reply; 2+ messages in thread
From: hluaw at connect dot ust.hk @ 2023-04-06 6:13 UTC (permalink / raw)
To: gdb-prs
https://sourceware.org/bugzilla/show_bug.cgi?id=30319
Bug ID: 30319
Summary: GDB shows "Incompatible DWARF Stack" when it's
actually compatible
Product: gdb
Version: 13.1
Status: UNCONFIRMED
Severity: normal
Priority: P2
Component: gdb
Assignee: unassigned at sourceware dot org
Reporter: hluaw at connect dot ust.hk
Target Milestone: ---
Given the following code:
#include "stdint.h"
int a, b = 9067;
volatile uint16_t *c = (volatile uint16_t*)&b;
int *d = &a;
int func_1() {
uint16_t e[7][2][8];
uint32_t f = 37455;
int32_t l_146;
a = 0;
l_146 = *c * f;
*d = e[3][1][5];
return l_146;
}
int main() { func_1(); return 0; }
Compiled with clang 15.0.7 and -O1 -g.
For l_146, the following DWARF is generated.
(gdb) info addr l_146
Symbol "l_146" is multi-location:
Range 0x555555554674-0x55555555467b: a complex DWARF expression:
0: DW_OP_breg0 0 [$rax]
2: DW_OP_constu 65535
6: DW_OP_and
7: DW_OP_convert<DW_ATE_unsigned_16 [0x27]>
12: DW_OP_convert<DW_ATE_unsigned_32 [0x2b]>
17: DW_OP_constu 37455
21: DW_OP_mul
22: DW_OP_stack_value
(gdb) p l_146
Incompatible types on DWARF stack
GDB complaining that this DWARF expression is invalid since on DW_OP_mul, but
this DWARF is in fact valid since it performs a explicit conversion, and
DW_OP_constu 37455 is automatically unsigned.
--
You are receiving this mail because:
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 2+ messages in thread
* [Bug gdb/30319] GDB shows "Incompatible DWARF Stack" when it's actually compatible
2023-04-06 6:13 [Bug gdb/30319] New: GDB shows "Incompatible DWARF Stack" when it's actually compatible hluaw at connect dot ust.hk
@ 2023-04-13 14:39 ` tromey at sourceware dot org
0 siblings, 0 replies; 2+ messages in thread
From: tromey at sourceware dot org @ 2023-04-13 14:39 UTC (permalink / raw)
To: gdb-prs
https://sourceware.org/bugzilla/show_bug.cgi?id=30319
Tom Tromey <tromey at sourceware dot org> changed:
What |Removed |Added
----------------------------------------------------------------------------
CC| |tromey at sourceware dot org
Resolution|--- |INVALID
Status|UNCONFIRMED |RESOLVED
--- Comment #1 from Tom Tromey <tromey at sourceware dot org> ---
I think this is invalid DWARF.
The default type of a stack element is the so-called "generic type".
See DWARF 5 section 2.5.1.
Arithmetic operations are only defined when the types match;
either both operations must have the same base type, or be
of the generic type. See DWARF 5 section 2.5.1.4:
Operands of an
operation with two operands must have the same type, either the same base
type
or the generic type.
DW_OP_constu is not actually automatically unsigned.
Instead it pushes a value of the generic type. This
is documented in DWARF 5 section 2.5.1.1.
So, I think this is a compiler bug.
I'm going to close this PR; if you think this is in error,
feel free to comment with an explanation and I'll be
happy to reopen it.
--
You are receiving this mail because:
You are on the CC list for the bug.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2023-04-13 14:39 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2023-04-06 6:13 [Bug gdb/30319] New: GDB shows "Incompatible DWARF Stack" when it's actually compatible hluaw at connect dot ust.hk
2023-04-13 14:39 ` [Bug gdb/30319] " tromey at sourceware dot org
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).