Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]
Groups > linux.kernel > #1235173
| From | Robert Bragg <robert@sixbynine.org> |
|---|---|
| Newsgroups | linux.kernel |
| Subject | [RFC 6/6] drm/i915: add oa_event_min_timer_exponent sysctl |
| Date | 2015-09-29 16:50 +0200 |
| Message-ID | <qe4oa-ZN-25@gated-at.bofh.it> (permalink) |
| References | <qe4ev-Oh-49@gated-at.bofh.it> |
| Organization | linux.* mail to news gateway |
The minimal sampling period is now configurable via a
dev.i915.oa_event_min_timer_exponent sysctl parameter.
Following the precedent set by perf, the default is the minimum that
won't (on its own) exceed the default kernel.perf_event_max_sample_rate
default of 100000 samples/s.
Signed-off-by: Robert Bragg <robert@sixbynine.org>
---
drivers/gpu/drm/i915/i915_perf.c | 37 ++++++++++++++++++++++++++++---------
1 file changed, 28 insertions(+), 9 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index ab82857..5ef7d92 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -42,6 +42,23 @@ static u32 i915_perf_event_paranoid = true;
#define OA_EXPONENT_MAX 0x3f
+/* for sysctl proc_dointvec_minmax of i915_oa_event_min_timer_exponent */
+static int zero;
+static int oa_exponent_max = OA_EXPONENT_MAX;
+
+/* Theoretically we can program the OA unit to sample every 160ns but don't
+ * allow that by default unless root...
+ *
+ * The period is derived from the exponent as:
+ *
+ * period = 80ns * 2^(exponent + 1)
+ *
+ * Referring to perf's kernel.perf_event_max_sample_rate for a precedent
+ * (100000 by default); with an OA exponent of 6 we get a period of 10.240
+ * microseconds - just under 100000Hz
+ */
+static u32 i915_oa_event_min_timer_exponent = 6;
+
static struct i915_oa_format hsw_oa_formats[I915_OA_FORMAT_MAX] = {
[I915_OA_FORMAT_A13] = { 0, 64 },
[I915_OA_FORMAT_A29] = { 1, 128 },
@@ -674,15 +691,8 @@ static int i915_oa_event_init(struct i915_perf_event *event,
if (period_exponent > OA_EXPONENT_MAX)
return -EINVAL;
- /* Theoretically we can program the OA unit to sample every
- * 160ns but don't allow that by default unless root...
- *
- * Referring to perf's kernel.perf_event_max_sample_rate for
- * a precedent (100000 by default); with an OA exponent of
- * 6 we get a period of 10.240 microseconds -just under
- * 100000Hz
- */
- if (period_exponent < 6 && !capable(CAP_SYS_ADMIN)) {
+ if (period_exponent < i915_oa_event_min_timer_exponent &&
+ !capable(CAP_SYS_ADMIN)) {
DRM_ERROR("Sampling period too high without root privileges\n");
return -EACCES;
}
@@ -1113,6 +1123,15 @@ static struct ctl_table oa_table[] = {
.mode = 0644,
.proc_handler = proc_dointvec,
},
+ {
+ .procname = "oa_event_min_timer_exponent",
+ .data = &i915_oa_event_min_timer_exponent,
+ .maxlen = sizeof(i915_oa_event_min_timer_exponent),
+ .mode = 0644,
+ .proc_handler = proc_dointvec_minmax,
+ .extra1 = &zero,
+ .extra2 = &oa_exponent_max,
+ },
{}
};
--
2.5.2
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread
[RFC 0/6] Non perf based Gen Graphics OA unit driver Robert Bragg <robert@sixbynine.org> - 2015-09-29 16:40 +0200 [RFC 6/6] drm/i915: add oa_event_min_timer_exponent sysctl Robert Bragg <robert@sixbynine.org> - 2015-09-29 16:50 +0200 [RFC 5/6] drm/i915: Add dev.i915.perf_event_paranoid sysctl option Robert Bragg <robert@sixbynine.org> - 2015-09-29 16:50 +0200 [RFC 1/6] drm/i915: Add i915 perf infrastructure Robert Bragg <robert@sixbynine.org> - 2015-09-29 16:50 +0200 [RFC 2/6] drm/i915: rename OACONTROL GEN7_OACONTROL Robert Bragg <robert@sixbynine.org> - 2015-09-29 16:50 +0200 [RFC 3/6] drm/i915: Add static '3D' Haswell OA unit config Robert Bragg <robert@sixbynine.org> - 2015-09-29 16:50 +0200 Re: [RFC 0/6] Non perf based Gen Graphics OA unit driver Zhenyu Wang <zhenyuw@linux.intel.com> - 2015-09-30 05:30 +0200 Re: [RFC 0/6] Non perf based Gen Graphics OA unit driver Chris Wilson <chris@chris-wilson.co.uk> - 2015-09-30 10:40 +0200
csiph-web