public inbox for intel-gfx@lists.freedesktop.org
 help / color / mirror / Atom feed
From: Xiaolin Zhang <xiaolin.zhang@intel.com>
To: intel-gvt-dev@lists.freedesktop.org, intel-gfx@lists.freedesktop.org
Cc: zhenyu.z.wang@intel.com, hang.yuan@intel.com, zhiyuan.lv@intel.com
Subject: [PATCH v6 8/8] drm/i915/gvt: GVTg support context submission pv optimization
Date: Mon,  3 Jun 2019 14:02:49 +0800	[thread overview]
Message-ID: <1559541769-25279-9-git-send-email-xiaolin.zhang@intel.com> (raw)
In-Reply-To: <1559541769-25279-1-git-send-email-xiaolin.zhang@intel.com>

implemented context submission pv optimizaiton within GVTg.

GVTg to read context submission data (elsp_data) from the shared_page
directly without trap cost and eliminate execlist HW behavior emulation
without injecting context switch interrupt to guest under PV
submisison mechanism.

v0: RFC.
v1: rebase.
v2: rebase.
v3: report pv context submission cap and handle VGT_G2V_ELSP_SUBMIT
g2v pv notification.
v4: eliminate execlist HW emulation and don't inject context switch
interrupt to guest under PV submisison mechanism.
v5: rebase.
v6: rebase.

Signed-off-by: Xiaolin Zhang <xiaolin.zhang@intel.com>
---
 drivers/gpu/drm/i915/gvt/execlist.c |  6 ++++++
 drivers/gpu/drm/i915/gvt/handlers.c | 29 ++++++++++++++++++++++++++++-
 drivers/gpu/drm/i915/gvt/vgpu.c     |  1 +
 3 files changed, 35 insertions(+), 1 deletion(-)

diff --git a/drivers/gpu/drm/i915/gvt/execlist.c b/drivers/gpu/drm/i915/gvt/execlist.c
index f21b8fb..e52bfd6 100644
--- a/drivers/gpu/drm/i915/gvt/execlist.c
+++ b/drivers/gpu/drm/i915/gvt/execlist.c
@@ -382,6 +382,9 @@ static int prepare_execlist_workload(struct intel_vgpu_workload *workload)
 	int ring_id = workload->ring_id;
 	int ret;
 
+	if (VGPU_PVCAP(vgpu, PV_SUBMISSION))
+		return 0;
+
 	if (!workload->emulate_schedule_in)
 		return 0;
 
@@ -429,6 +432,9 @@ static int complete_execlist_workload(struct intel_vgpu_workload *workload)
 		goto out;
 	}
 
+	if (VGPU_PVCAP(vgpu, PV_SUBMISSION))
+		goto out;
+
 	ret = emulate_execlist_ctx_schedule_out(execlist, &workload->ctx_desc);
 out:
 	intel_vgpu_unpin_mm(workload->shadow_mm);
diff --git a/drivers/gpu/drm/i915/gvt/handlers.c b/drivers/gpu/drm/i915/gvt/handlers.c
index 1e09c23..9cff9396 100644
--- a/drivers/gpu/drm/i915/gvt/handlers.c
+++ b/drivers/gpu/drm/i915/gvt/handlers.c
@@ -1692,6 +1692,31 @@ static int mmio_read_from_hw(struct intel_vgpu *vgpu,
 	return intel_vgpu_default_mmio_read(vgpu, offset, p_data, bytes);
 }
 
+static int handle_pv_submission(struct intel_vgpu *vgpu, int ring_id)
+{
+	struct intel_vgpu_execlist *execlist;
+	u32 hw_id = vgpu->gvt->dev_priv->engine[ring_id]->hw_id;
+	u32 pv_elsp_off = offsetof(struct gvt_shared_page, buf.pv_elsp);
+	u32 submitted_off = offsetof(struct gvt_shared_page, buf.submitted);
+	bool submitted = true;
+	int ret;
+
+	execlist = &vgpu->submission.execlist[ring_id];
+
+	pv_elsp_off += hw_id * sizeof(struct pv_submission);
+	if (intel_gvt_read_shared_page(vgpu, pv_elsp_off,
+		&execlist->elsp_dwords.data, sizeof(struct pv_submission)))
+		return -EINVAL;
+
+	ret = intel_vgpu_submit_execlist(vgpu, ring_id);
+	if (ret)
+		submitted = false;
+
+	submitted_off += hw_id;
+	ret = intel_gvt_write_shared_page(vgpu, submitted_off, &submitted, 1);
+	return ret;
+}
+
 static int elsp_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
 		void *p_data, unsigned int bytes)
 {
@@ -1703,8 +1728,10 @@ static int elsp_mmio_write(struct intel_vgpu *vgpu, unsigned int offset,
 	if (WARN_ON(ring_id < 0 || ring_id >= I915_NUM_ENGINES))
 		return -EINVAL;
 
-	execlist = &vgpu->submission.execlist[ring_id];
+	if (VGPU_PVCAP(vgpu, PV_SUBMISSION) && VGT_G2V_PV_SUBMISSION == data)
+		return handle_pv_submission(vgpu, ring_id);
 
+	execlist = &vgpu->submission.execlist[ring_id];
 	execlist->elsp_dwords.data[3 - execlist->elsp_dwords.index] = data;
 	if (execlist->elsp_dwords.index == 3) {
 		ret = intel_vgpu_submit_execlist(vgpu, ring_id);
diff --git a/drivers/gpu/drm/i915/gvt/vgpu.c b/drivers/gpu/drm/i915/gvt/vgpu.c
index 57eaf56..debdb88 100644
--- a/drivers/gpu/drm/i915/gvt/vgpu.c
+++ b/drivers/gpu/drm/i915/gvt/vgpu.c
@@ -51,6 +51,7 @@ void populate_pvinfo_page(struct intel_vgpu *vgpu)
 
 	if (!intel_vtd_active())
 		vgpu_vreg_t(vgpu, vgtif_reg(pv_caps)) = PV_PPGTT_UPDATE;
+	vgpu_vreg_t(vgpu, vgtif_reg(pv_caps)) |= PV_SUBMISSION;
 
 	vgpu_vreg_t(vgpu, vgtif_reg(avail_rs.mappable_gmadr.base)) =
 		vgpu_aperture_gmadr_base(vgpu);
-- 
2.7.4

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

  parent reply	other threads:[~2019-06-03  6:02 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-03  6:02 [PATCH v6 0/8] i915 vgpu PV to improve vgpu performance Xiaolin Zhang
2019-06-03  6:02 ` [PATCH v6 1/8] drm/i915: introduced vgpu pv capability Xiaolin Zhang
2019-06-03  6:02 ` [PATCH v6 2/8] drm/i915: vgpu shared memory setup for pv optimization Xiaolin Zhang
2019-06-03  6:02 ` [PATCH v6 3/8] drm/i915: vgpu ppgtt update " Xiaolin Zhang
2019-06-04  9:00   ` Chris Wilson
2019-06-10  1:32     ` Zhang, Xiaolin
2019-06-10  7:44       ` Chris Wilson
2019-06-10  7:47         ` Zhang, Xiaolin
2019-06-03  6:02 ` [PATCH v6 4/8] drm/i915: vgpu context submission " Xiaolin Zhang
2019-06-03  6:02 ` [PATCH v6 5/8] drm/i915/gvt: GVTg handle pv_caps PVINFO register Xiaolin Zhang
2019-06-03  6:02 ` [PATCH v6 6/8] drm/i915/gvt: GVTg handle shared_page setup Xiaolin Zhang
2019-06-03  6:02 ` [PATCH v6 7/8] drm/i915/gvt: GVTg support ppgtt pv optimization Xiaolin Zhang
2019-06-03  6:02 ` Xiaolin Zhang [this message]
2019-06-03  7:23 ` ✗ Fi.CI.CHECKPATCH: warning for i915 vgpu PV to improve vgpu performance Patchwork
2019-06-03  7:27 ` ✗ Fi.CI.SPARSE: " Patchwork
2019-06-03  8:16 ` ✓ Fi.CI.BAT: success " Patchwork
2019-06-03  9:51 ` ✓ Fi.CI.IGT: " 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=1559541769-25279-9-git-send-email-xiaolin.zhang@intel.com \
    --to=xiaolin.zhang@intel.com \
    --cc=hang.yuan@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=intel-gvt-dev@lists.freedesktop.org \
    --cc=zhenyu.z.wang@intel.com \
    --cc=zhiyuan.lv@intel.com \
    /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