On 11/10/26 7:17 AM, Matt Turner wrote:
The alternatives table treats 0x418, 0x50f and 0x60f as one event, but
the manual (MPC7450UM Rev. 5, Tables 11-12 to 11-14) describes three
different events that share the name "snoop retries":
0x418 load/store snoops retried by the load/store unit
0x50f internal requests that are internally retried
0x60f snoop responses of "retry" to any access
On a 7448, over the same workload, counting r418, r50f and r60f
together, and then r418 in a group with r41c:
r418 1793
r50f 1546267
r60f 2266850
r418 in {r41c,r418} 670562
r418 and r41c both need PMC4, so in the group the scheduler moves r418
to PMC5 or PMC6, which count other events, and reports 374 times the
real count. Remove the row.
Fixes: 7325927e5a20 ("perf_counter: powerpc: Add processor back-end for MPC7450
family")
Signed-off-by: Matt Turner <[email protected]>
Hi Matt,
I have checked the three "Snoop retries" entries in the manual:
0x418 Table 11-12, PMC4 select 24: load-store snoops retried by the
load-store unit, not counting MSS collisions
0x50f Table 11-13, PMC5 select 15: internal requests that are
internally retried, including some MSS collision cases
0x60f Table 11-14, PMC6 select 15: times a snoop response to any
access is "retry"
The descriptions differ in what they count, so they do not look like
the same event.
Fixing it by removing the row from the alternatives table makes sense.
Reviewed-by: Tanushree Shah <[email protected]>
Thanks
Tanushree Shah
---
arch/powerpc/perf/mpc7450-pmu.c | 1 -
1 file changed, 1 deletion(-)
diff --git a/arch/powerpc/perf/mpc7450-pmu.c b/arch/powerpc/perf/mpc7450-pmu.c
index 5a882d934645..9890c87c58ef 100644
--- a/arch/powerpc/perf/mpc7450-pmu.c
+++ b/arch/powerpc/perf/mpc7450-pmu.c
@@ -184,7 +184,6 @@ static int mpc7450_get_constraint(u64 event, unsigned long
*maskp,
static const unsigned int event_alternatives[][MAX_ALT] = {
{ 0x217, 0x317 }, /* PM_L1_DCACHE_MISS */
- { 0x418, 0x50f, 0x60f }, /* PM_SNOOP_RETRY */
{ 0x502, 0x602 }, /* PM_L2_HIT */
{ 0x503, 0x603 }, /* PM_L3_HIT */
{ 0x504, 0x604 }, /* PM_L2_ICACHE_MISS */