From: akash.goel@intel.com
To: intel-gfx@lists.freedesktop.org
Cc: Akash Goel <akash.goel@intel.com>
Subject: [PATCH 11/18] drm/i915: Optimization to reduce the sampling time of GuC log buffer
Date: Thu, 8 Sep 2016 16:09:11 +0530 [thread overview]
Message-ID: <1473331158-21082-12-git-send-email-akash.goel@intel.com> (raw)
In-Reply-To: <1473331158-21082-1-git-send-email-akash.goel@intel.com>
From: Akash Goel <akash.goel@intel.com>
GuC firmware sends an interrupt to flush the log buffer when it becomes
half full, so Driver doesn't really need to sample the complete buffer
and can just copy only the newly written data by GuC into the local
buffer, i.e. as per the read & write pointer values.
Moreover the flush interrupt would generally come for one type of log
buffer, when it becomes half full, so at that time the other 2 types of
log buffer would comparatively have much lesser unread data in them.
In case of overflow reported by GuC, Driver do need to copy the entire
buffer as the whole buffer would contain the unread data.
v2: Rebase.
v3: Fix the blooper of doing the copy twice. (Tvrtko)
v4: Add curlies for 'else' case also, matching the 'if'. (Tvrtko)
Signed-off-by: Akash Goel <akash.goel@intel.com>
Reviewed-by: Tvrtko Ursulin <tvrtko.ursulin@intel.com>
---
drivers/gpu/drm/i915/i915_guc_submission.c | 28 +++++++++++++++++++++++++---
1 file changed, 25 insertions(+), 3 deletions(-)
diff --git a/drivers/gpu/drm/i915/i915_guc_submission.c b/drivers/gpu/drm/i915/i915_guc_submission.c
index dbf7c84..ae52d10 100644
--- a/drivers/gpu/drm/i915/i915_guc_submission.c
+++ b/drivers/gpu/drm/i915/i915_guc_submission.c
@@ -1036,11 +1036,12 @@ static unsigned int guc_get_log_buffer_size(enum guc_log_buffer_type type)
static void guc_read_update_log_buffer(struct intel_guc *guc)
{
+ unsigned int buffer_size, read_offset, write_offset, bytes_to_copy, full_cnt;
struct guc_log_buffer_state *log_buf_state, *log_buf_snapshot_state;
struct guc_log_buffer_state log_buf_state_local;
- unsigned int buffer_size, write_offset, full_cnt;
enum guc_log_buffer_type type;
void *src_data, *dst_data;
+ bool new_overflow;
if (WARN_ON(!guc->log.buf_addr))
return;
@@ -1063,12 +1064,13 @@ static void guc_read_update_log_buffer(struct intel_guc *guc)
memcpy(&log_buf_state_local, log_buf_state,
sizeof(struct guc_log_buffer_state));
buffer_size = guc_get_log_buffer_size(type);
+ read_offset = log_buf_state_local.read_ptr;
write_offset = log_buf_state_local.sampled_write_ptr;
full_cnt = log_buf_state_local.buffer_full_cnt;
/* Bookkeeping stuff */
guc->log.flush_count[type] += log_buf_state_local.flush_to_file;
- guc_check_log_buf_overflow(guc, type, full_cnt);
+ new_overflow = guc_check_log_buf_overflow(guc, type, full_cnt);
/* Update the state of shared log buffer */
log_buf_state->read_ptr = write_offset;
@@ -1091,7 +1093,27 @@ static void guc_read_update_log_buffer(struct intel_guc *guc)
log_buf_snapshot_state++;
/* Now copy the actual logs. */
- memcpy(dst_data, src_data, buffer_size);
+ if (unlikely(new_overflow)) {
+ /* copy the whole buffer in case of overflow */
+ read_offset = 0;
+ write_offset = buffer_size;
+ } else if (unlikely((read_offset > buffer_size) ||
+ (write_offset > buffer_size))) {
+ DRM_ERROR("invalid log buffer state\n");
+ /* copy whole buffer as offsets are unreliable */
+ read_offset = 0;
+ write_offset = buffer_size;
+ }
+
+ /* Just copy the newly written data */
+ if (read_offset > write_offset) {
+ memcpy(dst_data, src_data, write_offset);
+ bytes_to_copy = buffer_size - read_offset;
+ } else {
+ bytes_to_copy = write_offset - read_offset;
+ }
+ memcpy(dst_data + read_offset,
+ src_data + read_offset, bytes_to_copy);
src_data += buffer_size;
dst_data += buffer_size;
--
1.9.2
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
next prev parent reply other threads:[~2016-09-08 10:25 UTC|newest]
Thread overview: 25+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-09-08 10:39 [PATCH v9 00/18] Support for sustained capturing of GuC firmware logs akash.goel
2016-09-08 10:39 ` [PATCH 01/18] drm/i915: Decouple GuC log setup from verbosity parameter akash.goel
2016-09-08 10:39 ` [PATCH 02/18] drm/i915: Add GuC ukernel logging related fields to fw interface file akash.goel
2016-09-08 10:39 ` [PATCH 03/18] drm/i915: New structure to contain GuC logging related fields akash.goel
2016-09-08 10:39 ` [PATCH 04/18] drm/i915: Add low level set of routines for programming PM IER/IIR/IMR register set akash.goel
2016-09-08 10:39 ` [PATCH 05/18] drm/i915: Support for GuC interrupts akash.goel
2016-09-08 10:39 ` [PATCH 06/18] drm/i915: Handle log buffer flush interrupt event from GuC akash.goel
2016-09-08 10:39 ` [PATCH 07/18] relay: Use per CPU constructs for the relay channel buffer pointers akash.goel
2016-09-08 10:39 ` [PATCH 08/18] drm/i915: Add a relay backed debugfs interface for capturing GuC logs akash.goel
2016-09-08 10:39 ` [PATCH 09/18] drm/i915: New lock to serialize the Host2GuC actions akash.goel
2016-09-08 10:39 ` [PATCH 10/18] drm/i915: Add stats for GuC log buffer flush interrupts akash.goel
2016-09-08 10:39 ` akash.goel [this message]
2016-09-08 10:39 ` [PATCH 12/18] drm/i915: Increase GuC log buffer size to reduce " akash.goel
2016-09-08 10:39 ` [PATCH 13/18] drm/i915: Augment i915 error state to include the dump of GuC log buffer akash.goel
2016-09-08 10:39 ` [PATCH 14/18] drm/i915: Support for forceful flush " akash.goel
2016-09-08 10:39 ` [PATCH 15/18] drm/i915: Debugfs support for GuC logging control akash.goel
2016-09-08 10:39 ` [PATCH 16/18] drm/i915: Use SSE4.1 movntdqa based memcpy for sampling GuC log buffer akash.goel
2016-09-08 10:39 ` [PATCH 17/18] drm/i915: Early creation of relay channel for capturing boot time logs akash.goel
2016-09-08 10:39 ` [PATCH 18/18] drm/i915: Mark the GuC log buffer flush interrupts handling WQ as freezable akash.goel
2016-09-12 8:23 ` ✗ Fi.CI.BAT: failure for Support for sustained capturing of GuC firmware logs (rev10) Patchwork
2016-09-12 13:51 ` Goel, Akash
-- strict thread matches above, loose matches on Subject: below --
2016-10-12 16:24 [PATCH v10 00/18] Support for sustained capturing of GuC firmware logs akash.goel
2016-10-12 16:24 ` [PATCH 11/18] drm/i915: Optimization to reduce the sampling time of GuC log buffer akash.goel
2016-08-15 14:49 [PATCH v6 00/18] Support for sustained capturing of GuC firmware logs akash.goel
2016-08-15 14:49 ` [PATCH 11/18] drm/i915: Optimization to reduce the sampling time of GuC log buffer akash.goel
2016-08-15 15:36 ` Tvrtko Ursulin
2016-08-15 16:13 ` Goel, Akash
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1473331158-21082-12-git-send-email-akash.goel@intel.com \
--to=akash.goel@intel.com \
--cc=intel-gfx@lists.freedesktop.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox