intel-gfx.lists.freedesktop.org archive mirror
 help / color / mirror / Atom feed
From: Arun Siluvery <arun.siluvery@linux.intel.com>
To: intel-gfx@lists.freedesktop.org
Subject: [PATCH 5/5] drm/i915/error: Capture WA ctx batch in error state
Date: Thu, 28 Jan 2016 19:01:24 +0000	[thread overview]
Message-ID: <1454007684-16777-6-git-send-email-arun.siluvery@linux.intel.com> (raw)
In-Reply-To: <1454007684-16777-1-git-send-email-arun.siluvery@linux.intel.com>

From Gen8 onwards we apply ctx workarounds using special batch buffers that
execute during save/restore, good to have them in error state.

Signed-off-by: Arun Siluvery <arun.siluvery@linux.intel.com>
---
 drivers/gpu/drm/i915/i915_drv.h       |  2 +-
 drivers/gpu/drm/i915/i915_gpu_error.c | 25 +++++++++++++++++++++++++
 2 files changed, 26 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
index 4b199a4..8440c35 100644
--- a/drivers/gpu/drm/i915/i915_drv.h
+++ b/drivers/gpu/drm/i915/i915_drv.h
@@ -567,7 +567,7 @@ struct drm_i915_error_state {
 			int page_count;
 			u64 gtt_offset;
 			u32 *pages[0];
-		} *req_ringbuffer, *hw_ringbuffer, *batchbuffer, *wa_batchbuffer, *ctx, *hws_page;
+		} *req_ringbuffer, *hw_ringbuffer, *batchbuffer, *wa_batchbuffer, *ctx, *hws_page, *wa_ctx;
 
 		struct drm_i915_error_request {
 			u64 ctx_desc;
diff --git a/drivers/gpu/drm/i915/i915_gpu_error.c b/drivers/gpu/drm/i915/i915_gpu_error.c
index 8b1a1c0..e2c32d4 100644
--- a/drivers/gpu/drm/i915/i915_gpu_error.c
+++ b/drivers/gpu/drm/i915/i915_gpu_error.c
@@ -561,6 +561,24 @@ int i915_error_state_to_str(struct drm_i915_error_state_buf *m,
 			}
 		}
 
+		if ((obj = error->ring[i].wa_ctx)) {
+			u64 wa_ctx_offset = obj->gtt_offset;
+			u32 *wa_ctx_page = &obj->pages[0][0];
+
+			err_printf(m, "\n%s --- WA Ctx batch buffer = 0x%08llx\n",
+				   dev_priv->ring[i].name, wa_ctx_offset);
+			offset = 0;
+			for (elt = 0; elt < PAGE_SIZE/32; elt += 4) {
+				err_printf(m, "[%04x] %08x %08x %08x %08x\n",
+					   offset,
+					   wa_ctx_page[elt],
+					   wa_ctx_page[elt+1],
+					   wa_ctx_page[elt+2],
+					   wa_ctx_page[elt+3]);
+				offset += 16;
+			}
+		}
+
 		if ((obj = error->ring[i].ctx)) {
 			err_printf(m, "%s --- HW Context = 0x%08x\n",
 				   dev_priv->ring[i].name,
@@ -654,6 +672,8 @@ static void i915_error_state_free(struct kref *error_ref)
 		i915_error_object_free(error->ring[i].hws_page);
 		i915_error_object_free(error->ring[i].ctx);
 		kfree(error->ring[i].requests);
+		if (i == RCS)
+			i915_error_object_free(error->ring[i].wa_ctx);
 	}
 
 	i915_error_object_free(error->semaphore_obj);
@@ -1165,6 +1185,11 @@ static void i915_gem_record_rings(struct drm_device *dev,
 		error->ring[i].hws_page =
 			i915_error_ggtt_object_create(dev_priv, ring->status_page.obj);
 
+		if (INTEL_INFO(dev)->gen >= 8 && ring->id == RCS) {
+			error->ring[i].wa_ctx =
+				i915_error_ggtt_object_create(dev_priv, ring->wa_ctx.obj);
+		}
+
 		i915_gem_record_active_context(ring, error, &error->ring[i]);
 
 		count = 0;
-- 
1.9.1

_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/intel-gfx

  parent reply	other threads:[~2016-01-28 19:01 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-01-28 19:01 [PATCH 0/5] Capture more useful details in error state Arun Siluvery
2016-01-28 19:01 ` [PATCH 1/5] drm/i915/error: capture execlist state on error Arun Siluvery
2016-01-29  7:49   ` Mika Kuoppala
2016-01-29 11:45     ` Chris Wilson
2016-01-29 12:25       ` Arun Siluvery
2016-01-29 12:38         ` Chris Wilson
2016-01-28 19:01 ` [PATCH 2/5] drm/i915/error: capture ringbuffer pointed to by START Arun Siluvery
2016-01-29 11:47   ` Chris Wilson
2016-02-01 21:30     ` Arun Siluvery
2016-01-28 19:01 ` [PATCH 3/5] drm/i915/error: report ctx id & desc for each request in the queue Arun Siluvery
2016-01-29  8:17   ` Mika Kuoppala
2016-01-29  9:48     ` Arun Siluvery
2016-01-28 19:01 ` [PATCH 4/5] drm/i915/error: improve CSB reporting Arun Siluvery
2016-01-28 19:01 ` Arun Siluvery [this message]
2016-01-29  7:52   ` [PATCH 5/5] drm/i915/error: Capture WA ctx batch in error state Mika Kuoppala
2016-01-29 10:09     ` Arun Siluvery
2016-01-29 10:59 ` ✗ Fi.CI.BAT: failure for Capture more useful details " Patchwork

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=1454007684-16777-6-git-send-email-arun.siluvery@linux.intel.com \
    --to=arun.siluvery@linux.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;
as well as URLs for NNTP newsgroup(s).