Add new OA formats for DG2. Some of the newer OA formats are not
multples of 64 bytes and are not powers of 2. For those formats, adjust
hw_tail accordingly when checking for new reports.
Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.rama...@intel.com>
---
drivers/gpu/drm/i915/i915_perf.c | 63 ++++++++++++++++++++------------
include/uapi/drm/i915_drm.h | 6 +++
2 files changed, 46 insertions(+), 23 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
index 735244a3aedd..c8331b549d31 100644
--- a/drivers/gpu/drm/i915/i915_perf.c
+++ b/drivers/gpu/drm/i915/i915_perf.c
@@ -306,7 +306,8 @@ static u32 i915_oa_max_sample_rate = 100000;
/* XXX: beware if future OA HW adds new report formats that the current
* code assumes all reports have a power-of-two size and ~(size - 1) can
- * be used as a mask to align the OA tail pointer.
+ * be used as a mask to align the OA tail pointer. In some of the
+ * formats, R is used to denote reserved field.
*/
static const struct i915_oa_format oa_formats[I915_OA_FORMAT_MAX] = {
[I915_OA_FORMAT_A13] = { 0, 64 },
@@ -320,6 +321,10 @@ static const struct i915_oa_format
oa_formats[I915_OA_FORMAT_MAX] = {
[I915_OA_FORMAT_A12] = { 0, 64 },
[I915_OA_FORMAT_A12_B8_C8] = { 2, 128 },
[I915_OA_FORMAT_A32u40_A4u32_B8_C8] = { 5, 256 },
+ [I915_OAR_FORMAT_A32u40_A4u32_B8_C8] = { 5, 256 },
+ [I915_OA_FORMAT_A24u40_A14u32_B8_C8] = { 5, 256 },
+ [I915_OAR_FORMAT_A36u64_B8_C8] = { 1, 384 },
+ [I915_OA_FORMAT_A38u64_R2u64_B8_C8] = { 1, 448 },
};
#define SAMPLE_OA_REPORT (1<<0)
@@ -467,6 +472,7 @@ static bool oa_buffer_check_unlocked(struct
i915_perf_stream *stream)
bool pollin;
u32 hw_tail;
u64 now;
+ u32 partial_report_size;
/* We have to consider the (unlikely) possibility that read() errors
* could result in an OA buffer reset which might reset the head and
@@ -476,10 +482,16 @@ static bool oa_buffer_check_unlocked(struct
i915_perf_stream *stream)
hw_tail = stream->perf->ops.oa_hw_tail_read(stream);
- /* The tail pointer increases in 64 byte increments,
- * not in report_size steps...
+ /* The tail pointer increases in 64 byte increments, whereas report
+ * sizes need not be integral multiples or 64 or powers of 2.
+ * Compute potentially partially landed report in the OA buffer
*/
- hw_tail &= ~(report_size - 1);
+ partial_report_size = OA_TAKEN(hw_tail, stream->oa_buffer.tail);
+ partial_report_size %= report_size;
+
+ /* Subtract partial amount off the tail */
+ hw_tail = gtt_offset + ((hw_tail - partial_report_size) &
+ (stream->oa_buffer.vma->size - 1));
now = ktime_get_mono_fast_ns();
@@ -601,6 +613,8 @@ static int append_oa_sample(struct i915_perf_stream *stream,
{
int report_size = stream->oa_buffer.format_size;
struct drm_i915_perf_record_header header;
+ int report_size_partial;
+ u8 *oa_buf_end;
header.type = DRM_I915_PERF_RECORD_SAMPLE;
header.pad = 0;
@@ -614,7 +628,19 @@ static int append_oa_sample(struct i915_perf_stream
*stream,
return -EFAULT;
buf += sizeof(header);
- if (copy_to_user(buf, report, report_size))
+ oa_buf_end = stream->oa_buffer.vaddr +
+ stream->oa_buffer.vma->size;
+ report_size_partial = oa_buf_end - report;
+
+ if (report_size_partial < report_size) {
+ if(copy_to_user(buf, report, report_size_partial))
+ return -EFAULT;
+ buf += report_size_partial;
+
+ if(copy_to_user(buf, stream->oa_buffer.vaddr,
+ report_size - report_size_partial))
+ return -EFAULT;