On 03.07.2025 01:50, Steven Rostedt wrote:
On Tue, 1 Jul 2025 17:36:55 +0200
Jens Remus [off-list ref] wrote:
quoted
On s390 the prev_frame_sp may be equal to curr_frame_sp for the topmost
frame, as long as the topmost function did not allocate any stack. For
instance when early in the prologue or when in a leaf function that does
not require any stack space. My s390 sframe support patches would
therefore currently change above check to:
/* stack going in wrong direction? */
if (sp <= state->sp - topmost)
goto done;
How do you calculate "topmost" then?
Is it another field you add to "state"?
Correct. It is a boolean set to true in unwind_user_start() and set to
false in unwind_user_next() when updating the state.
I assume most architectures need above change, as their SP at function
entry should be equal to the SP at call site (unlike x86-64 due to CALL).
s390 also needs this information to allow restoring of FP/RA saved in
other registers (instead of on the stack) only for the topmost frame.
For any other frame arbitrary register contents would not be available,
as user unwind only unwinds SP, FP, and RA.
I would post my s390 sframe support patches as RFC once you have
provided a merged sframe branch as discussed in:
https://lore.kernel.org/all/20250702124737.565934b5@batman.local.home/ (local)
Regards,
Jens
--
Jens Remus
Linux on Z Development (D3303)
+49-7031-16-1128 Office
jremus@de.ibm.com
IBM
IBM Deutschland Research & Development GmbH; Vorsitzender des Aufsichtsrats: Wolfgang Wendt; Geschäftsführung: David Faller; Sitz der Gesellschaft: Böblingen; Registergericht: Amtsgericht Stuttgart, HRB 243294
IBM Data Privacy Statement: https://www.ibm.com/privacy/