perf: Use sample_flags for txn

Use the new sample_flags to indicate whether the txn field is filled by
the PMU driver.

Remove the txn field from the perf_sample_data_init() to minimize the
number of cache lines touched.

Signed-off-by: Kan Liang <kan.liang@linux.intel.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lore.kernel.org/r/20220901130959.1285717-7-kan.liang@linux.intel.com

diff --git a/arch/x86/events/intel/ds.c b/arch/x86/events/intel/ds.c
index bde73d492889d513229c42c1e7e5db62efa3bdc5..a5275c235c2aff755f7afc1c66859933b1f29e74 100644 (file)
--- a/arch/x86/events/intel/ds.c
+++ b/arch/x86/events/intel/ds.c
@@ -1628,9 +1628,11 @@ static void setup_pebs_fixed_sample_data(struct perf_event *event,
                        data->weight.full = intel_get_tsx_weight(pebs->tsx_tuning);
                        data->sample_flags |= PERF_SAMPLE_WEIGHT_TYPE;
                }
-               if (sample_type & PERF_SAMPLE_TRANSACTION)
+               if (sample_type & PERF_SAMPLE_TRANSACTION) {
                        data->txn = intel_get_tsx_transaction(pebs->tsx_tuning,
                                                              pebs->ax);
+                       data->sample_flags |= PERF_SAMPLE_TRANSACTION;
+               }
        }
 
        /*
@@ -1780,9 +1782,11 @@ static void setup_pebs_adaptive_sample_data(struct perf_event *event,
                if (sample_type & PERF_SAMPLE_ADDR_TYPE)
                        data->addr = meminfo->address;
 
-               if (sample_type & PERF_SAMPLE_TRANSACTION)
+               if (sample_type & PERF_SAMPLE_TRANSACTION) {
                        data->txn = intel_get_tsx_transaction(meminfo->tsx_tuning,
                                                          gprs ? gprs->ax : 0);
+                       data->sample_flags |= PERF_SAMPLE_TRANSACTION;
+               }
        }
 
        if (format_size & PEBS_DATACFG_XMMS) {

diff --git a/include/linux/perf_event.h b/include/linux/perf_event.h
index 6849f10dfc7ece3bdd2affd6093f16d508c4cfb5..581880ddb9efb0e032746461e564adc741a3b8ca 100644 (file)
--- a/include/linux/perf_event.h
+++ b/include/linux/perf_event.h
@@ -1012,7 +1012,6 @@ struct perf_sample_data {
        u64                             addr;
        struct perf_raw_record          *raw;
        u64                             period;
-       u64                             txn;
 
        /*
         * The other fields, optionally {set,used} by
@@ -1021,6 +1020,7 @@ struct perf_sample_data {
        struct perf_branch_stack        *br_stack;
        union perf_sample_weight        weight;
        union  perf_mem_data_src        data_src;
+       u64                             txn;
 
        u64                             type;
        u64                             ip;
@@ -1063,7 +1063,6 @@ static inline void perf_sample_data_init(struct perf_sample_data *data,
        data->addr = addr;
        data->raw  = NULL;
        data->period = period;
-       data->txn = 0;
 }
 
 /*

diff --git a/kernel/events/core.c b/kernel/events/core.c
index 163e2f478e616610f26277c3bb01ed27c285a0e2..15d27b14c827d2e964db8d3c6ce1fa3cb6aa431e 100644 (file)
--- a/kernel/events/core.c
+++ b/kernel/events/core.c
@@ -7414,6 +7414,9 @@ void perf_prepare_sample(struct perf_event_header *header,
        if (filtered_sample_type & PERF_SAMPLE_DATA_SRC)
                data->data_src.val = PERF_MEM_NA;
 
+       if (filtered_sample_type & PERF_SAMPLE_TRANSACTION)
+               data->txn = 0;
+
        if (sample_type & PERF_SAMPLE_REGS_INTR) {
                /* regs dump ABI info */
                int size = sizeof(u64);