* [PATCH] gdb/arm: Don't rely on loop detection to stop unwind
@ 2022-10-14 13:46 Torbjörn SVENSSON
2022-10-14 17:13 ` Luis Machado
0 siblings, 1 reply; 3+ messages in thread
From: Torbjörn SVENSSON @ 2022-10-14 13:46 UTC (permalink / raw)
To: gdb-patches; +Cc: luis.machado, pedro, Torbjörn SVENSSON
Setting SP of the next frame to the same address as the current frame
is an ugly way to stop the unwind. A cleaner way is to rely on the
frame_unwind_stop_reason function to return UNWIND_OUTERMOST.
Signed-off-by: Torbjörn SVENSSON <torbjorn.svensson@foss.st.com>
---
gdb/arm-tdep.c | 13 ++++++-------
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/gdb/arm-tdep.c b/gdb/arm-tdep.c
index fa6b08e4a54..b5facae8a5e 100644
--- a/gdb/arm-tdep.c
+++ b/gdb/arm-tdep.c
@@ -3356,7 +3356,6 @@ arm_m_exception_cache (frame_info_ptr this_frame)
to the exception and if FPU is used (causing extended stack frame). */
CORE_ADDR lr = get_frame_register_unsigned (this_frame, ARM_LR_REGNUM);
- CORE_ADDR sp = get_frame_register_unsigned (this_frame, ARM_SP_REGNUM);
/* ARMv7-M Architecture Reference "A2.3.1 Arm core registers"
states that LR is set to 0xffffffff on reset. ARMv8-M Architecture
@@ -3364,8 +3363,8 @@ arm_m_exception_cache (frame_info_ptr this_frame)
reset if Main Extension is implemented, otherwise the value is unknown. */
if (lr == 0xffffffff)
{
- /* Terminate any further stack unwinding by referring to self. */
- arm_cache_set_active_sp_value (cache, tdep, sp);
+ /* Terminate any further stack unwinding. */
+ arm_cache_set_active_sp_value (cache, tdep, 0);
return cache;
}
@@ -3387,8 +3386,8 @@ arm_m_exception_cache (frame_info_ptr this_frame)
{
warning (_("Non-secure to secure stack unwinding disabled."));
- /* Terminate any further stack unwinding by referring to self. */
- arm_cache_set_active_sp_value (cache, tdep, sp);
+ /* Terminate any further stack unwinding. */
+ arm_cache_set_active_sp_value (cache, tdep, 0);
return cache;
}
@@ -3452,8 +3451,8 @@ arm_m_exception_cache (frame_info_ptr this_frame)
{
warning (_("Non-secure to secure stack unwinding disabled."));
- /* Terminate any further stack unwinding by referring to self. */
- arm_cache_set_active_sp_value (cache, tdep, sp);
+ /* Terminate any further stack unwinding. */
+ arm_cache_set_active_sp_value (cache, tdep, 0);
return cache;
}
--
2.25.1
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] gdb/arm: Don't rely on loop detection to stop unwind
2022-10-14 13:46 [PATCH] gdb/arm: Don't rely on loop detection to stop unwind Torbjörn SVENSSON
@ 2022-10-14 17:13 ` Luis Machado
2022-10-15 9:40 ` Torbjorn SVENSSON
0 siblings, 1 reply; 3+ messages in thread
From: Luis Machado @ 2022-10-14 17:13 UTC (permalink / raw)
To: Torbjörn SVENSSON, gdb-patches; +Cc: pedro
On 10/14/22 14:46, Torbjörn SVENSSON wrote:
> Setting SP of the next frame to the same address as the current frame
> is an ugly way to stop the unwind. A cleaner way is to rely on the
stop the unwind -> stop the unwinding.
> frame_unwind_stop_reason function to return UNWIND_OUTERMOST.
>
> Signed-off-by: Torbjörn SVENSSON <torbjorn.svensson@foss.st.com>
> ---
> gdb/arm-tdep.c | 13 ++++++-------
> 1 file changed, 6 insertions(+), 7 deletions(-)
>
> diff --git a/gdb/arm-tdep.c b/gdb/arm-tdep.c
> index fa6b08e4a54..b5facae8a5e 100644
> --- a/gdb/arm-tdep.c
> +++ b/gdb/arm-tdep.c
> @@ -3356,7 +3356,6 @@ arm_m_exception_cache (frame_info_ptr this_frame)
> to the exception and if FPU is used (causing extended stack frame). */
>
> CORE_ADDR lr = get_frame_register_unsigned (this_frame, ARM_LR_REGNUM);
> - CORE_ADDR sp = get_frame_register_unsigned (this_frame, ARM_SP_REGNUM);
>
> /* ARMv7-M Architecture Reference "A2.3.1 Arm core registers"
> states that LR is set to 0xffffffff on reset. ARMv8-M Architecture
> @@ -3364,8 +3363,8 @@ arm_m_exception_cache (frame_info_ptr this_frame)
> reset if Main Extension is implemented, otherwise the value is unknown. */
> if (lr == 0xffffffff)
> {
> - /* Terminate any further stack unwinding by referring to self. */
> - arm_cache_set_active_sp_value (cache, tdep, sp);
> + /* Terminate any further stack unwinding. */
> + arm_cache_set_active_sp_value (cache, tdep, 0);
> return cache;
> }
>
> @@ -3387,8 +3386,8 @@ arm_m_exception_cache (frame_info_ptr this_frame)
> {
> warning (_("Non-secure to secure stack unwinding disabled."));
>
> - /* Terminate any further stack unwinding by referring to self. */
> - arm_cache_set_active_sp_value (cache, tdep, sp);
> + /* Terminate any further stack unwinding. */
> + arm_cache_set_active_sp_value (cache, tdep, 0);
> return cache;
> }
>
> @@ -3452,8 +3451,8 @@ arm_m_exception_cache (frame_info_ptr this_frame)
> {
> warning (_("Non-secure to secure stack unwinding disabled."));
>
> - /* Terminate any further stack unwinding by referring to self. */
> - arm_cache_set_active_sp_value (cache, tdep, sp);
> + /* Terminate any further stack unwinding. */
> + arm_cache_set_active_sp_value (cache, tdep, 0);
> return cache;
> }
>
LGTM. Thanks.
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH] gdb/arm: Don't rely on loop detection to stop unwind
2022-10-14 17:13 ` Luis Machado
@ 2022-10-15 9:40 ` Torbjorn SVENSSON
0 siblings, 0 replies; 3+ messages in thread
From: Torbjorn SVENSSON @ 2022-10-15 9:40 UTC (permalink / raw)
To: Luis Machado, gdb-patches; +Cc: pedro
On 2022-10-14 19:13, Luis Machado wrote:
> On 10/14/22 14:46, Torbjörn SVENSSON wrote:
>> Setting SP of the next frame to the same address as the current frame
>> is an ugly way to stop the unwind. A cleaner way is to rely on the
>
> stop the unwind -> stop the unwinding.
>
>> frame_unwind_stop_reason function to return UNWIND_OUTERMOST.
>>
>> Signed-off-by: Torbjörn SVENSSON <torbjorn.svensson@foss.st.com>
>> ---
>> gdb/arm-tdep.c | 13 ++++++-------
>> 1 file changed, 6 insertions(+), 7 deletions(-)
>>
>> diff --git a/gdb/arm-tdep.c b/gdb/arm-tdep.c
>> index fa6b08e4a54..b5facae8a5e 100644
>> --- a/gdb/arm-tdep.c
>> +++ b/gdb/arm-tdep.c
>> @@ -3356,7 +3356,6 @@ arm_m_exception_cache (frame_info_ptr this_frame)
>> to the exception and if FPU is used (causing extended stack
>> frame). */
>> CORE_ADDR lr = get_frame_register_unsigned (this_frame,
>> ARM_LR_REGNUM);
>> - CORE_ADDR sp = get_frame_register_unsigned (this_frame,
>> ARM_SP_REGNUM);
>> /* ARMv7-M Architecture Reference "A2.3.1 Arm core registers"
>> states that LR is set to 0xffffffff on reset. ARMv8-M
>> Architecture
>> @@ -3364,8 +3363,8 @@ arm_m_exception_cache (frame_info_ptr this_frame)
>> reset if Main Extension is implemented, otherwise the value is
>> unknown. */
>> if (lr == 0xffffffff)
>> {
>> - /* Terminate any further stack unwinding by referring to self. */
>> - arm_cache_set_active_sp_value (cache, tdep, sp);
>> + /* Terminate any further stack unwinding. */
>> + arm_cache_set_active_sp_value (cache, tdep, 0);
>> return cache;
>> }
>> @@ -3387,8 +3386,8 @@ arm_m_exception_cache (frame_info_ptr this_frame)
>> {
>> warning (_("Non-secure to secure stack unwinding disabled."));
>> - /* Terminate any further stack unwinding by referring to self. */
>> - arm_cache_set_active_sp_value (cache, tdep, sp);
>> + /* Terminate any further stack unwinding. */
>> + arm_cache_set_active_sp_value (cache, tdep, 0);
>> return cache;
>> }
>> @@ -3452,8 +3451,8 @@ arm_m_exception_cache (frame_info_ptr this_frame)
>> {
>> warning (_("Non-secure to secure stack unwinding
>> disabled."));
>> - /* Terminate any further stack unwinding by referring to
>> self. */
>> - arm_cache_set_active_sp_value (cache, tdep, sp);
>> + /* Terminate any further stack unwinding. */
>> + arm_cache_set_active_sp_value (cache, tdep, 0);
>> return cache;
>> }
> LGTM. Thanks.
Pushed.
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2022-10-15 9:40 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2022-10-14 13:46 [PATCH] gdb/arm: Don't rely on loop detection to stop unwind Torbjörn SVENSSON
2022-10-14 17:13 ` Luis Machado
2022-10-15 9:40 ` Torbjorn SVENSSON
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).