overlay: Remove busy estimation

Too inaccurate to be useful.

Signed-off-by: Chris Wilson <chris@chris-wilson.co.uk>

diff --git a/overlay/gpu-perf.c b/overlay/gpu-perf.c
index bbebc5a..d5e461c 100644 (file)
--- a/overlay/gpu-perf.c
+++ b/overlay/gpu-perf.c
@@ -241,51 +241,6 @@ static int request_add(struct gpu_perf *gp, const void *event)
        return 1;
 }
 
-static int busy_start(struct gpu_perf *gp, const void *event)
-{
-       const struct sample_event *sample = event;
-       struct gpu_perf_comm *comm;
-       struct gpu_perf_time *busy;
-
-       comm = lookup_comm(gp, sample->pid);
-       if (comm == NULL)
-               return 0;
-
-       busy = malloc(sizeof(*busy));
-       if (busy == NULL)
-               return 0;
-
-       busy->ring = sample->raw[1];
-       busy->seqno = sample->raw[2];
-       busy->time = sample->time;
-       busy->comm = comm;
-       busy->next = gp->busy;
-       gp->busy = busy;
-
-       return 0;
-}
-
-static int busy_end(struct gpu_perf *gp, const void *event)
-{
-       const struct sample_event *sample = event;
-       struct gpu_perf_time *busy, **prev;
-
-       for (prev = &gp->busy; (busy = *prev) != NULL; prev = &busy->next) {
-               if (busy->ring != sample->raw[1])
-                       continue;
-
-               if (busy->seqno != sample->raw[2])
-                       continue;
-
-               busy->comm->busy_time += sample->time - busy->time;
-               *prev = busy->next;
-               free(busy);
-               return 1;
-       }
-
-       return 0;
-}
-
 static int flip_complete(struct gpu_perf *gp, const void *event)
 {
        const struct sample_event *sample = event;
@@ -309,11 +264,10 @@ static int wait_begin(struct gpu_perf *gp, const void *event)
                return 0;
 
        wait->comm = comm;
-       wait->ring = sample->raw[1];
        wait->seqno = sample->raw[2];
        wait->time = sample->time;
-       wait->next = gp->wait;
-       gp->wait = wait;
+       wait->next = gp->wait[sample->raw[1]];
+       gp->wait[sample->raw[1]] = wait;
 
        return 0;
 }
@@ -323,10 +277,7 @@ static int wait_end(struct gpu_perf *gp, const void *event)
        const struct sample_event *sample = event;
        struct gpu_perf_time *wait, **prev;
 
-       for (prev = &gp->wait; (wait = *prev) != NULL; prev = &wait->next) {
-               if (wait->ring != sample->raw[1])
-                       continue;
-
+       for (prev = &gp->wait[sample->raw[1]]; (wait = *prev) != NULL; prev = &wait->next) {
                if (wait->seqno != sample->raw[2])
                        continue;
 
@@ -346,8 +297,6 @@ void gpu_perf_init(struct gpu_perf *gp, unsigned flags)
        gp->page_size = getpagesize();
 
        perf_tracepoint_open(gp, "i915", "i915_gem_request_add", request_add);
-       if (perf_tracepoint_open(gp, "i915", "i915_gem_ring_complete", busy_end) == 0)
-               perf_tracepoint_open(gp, "i915", "i915_gem_ring_dispatch", busy_start);
        if (perf_tracepoint_open(gp, "i915", "i915_gem_request_wait_begin", wait_begin) == 0)
                perf_tracepoint_open(gp, "i915", "i915_gem_request_wait_end", wait_end);
        perf_tracepoint_open(gp, "i915", "i915_flip_complete", flip_complete);

diff --git a/overlay/gpu-perf.h b/overlay/gpu-perf.h
index 56c783d..3b91bd5 100644 (file)
--- a/overlay/gpu-perf.h
+++ b/overlay/gpu-perf.h
@@ -27,6 +27,8 @@
 
 #include <stdint.h>
 
+#define MAX_RINGS 4
+
 struct gpu_perf {
        int page_size;
        int nr_cpus;
@@ -52,10 +54,9 @@ struct gpu_perf {
        struct gpu_perf_time {
                struct gpu_perf_time *next;
                struct gpu_perf_comm *comm;
-               int ring;
                uint32_t seqno;
                uint64_t time;
-       } *wait, *busy;
+       } *wait[MAX_RINGS];
 };
 
 void gpu_perf_init(struct gpu_perf *gp, unsigned flags);