AMD-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: "Christian König" <ckoenig.leichtzumerken-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
To: amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.org
Subject: [PATCH 4/8] drm/amdgpu: allow direct submission in the VM backends
Date: Fri, 29 Mar 2019 11:45:03 +0100	[thread overview]
Message-ID: <20190329104507.2602-4-christian.koenig@amd.com> (raw)
In-Reply-To: <20190329104507.2602-1-christian.koenig-5C7GfCeVMHo@public.gmane.org>

This allows us to update page tables directly while in a page fault.

Signed-off-by: Christian König <christian.koenig@amd.com>
---
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h      |  5 ++++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c  |  4 +++
 drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c | 29 +++++++++++++--------
 3 files changed, 27 insertions(+), 11 deletions(-)

diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
index 91baf95212a6..66e27b941a55 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm.h
@@ -191,6 +191,11 @@ struct amdgpu_vm_update_params {
 	 */
 	struct amdgpu_vm *vm;
 
+	/**
+	 * @direct: if changes should be made directly
+	 */
+	bool direct;
+
 	/**
 	 * @pages_addr:
 	 *
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
index 5222d165abfc..f94e4896079c 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_cpu.c
@@ -49,6 +49,10 @@ static int amdgpu_vm_cpu_prepare(struct amdgpu_vm_update_params *p, void *owner,
 {
 	int r;
 
+	/* Don't wait for anything during page fault */
+	if (p->direct)
+		return 0;
+
 	/* Wait for PT BOs to be idle. PTs share the same resv. object
 	 * as the root PD BO
 	 */
diff --git a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
index ddd181f5ed37..891d597063cb 100644
--- a/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
+++ b/drivers/gpu/drm/amd/amdgpu/amdgpu_vm_sdma.c
@@ -68,17 +68,17 @@ static int amdgpu_vm_sdma_prepare(struct amdgpu_vm_update_params *p,
 	if (r)
 		return r;
 
-	r = amdgpu_sync_fence(p->adev, &p->job->sync, exclusive, false);
-	if (r)
-		return r;
+	p->num_dw_left = ndw;
+
+	if (p->direct)
+		return 0;
 
-	r = amdgpu_sync_resv(p->adev, &p->job->sync, root->tbo.resv,
-			     owner, false);
+	r = amdgpu_sync_fence(p->adev, &p->job->sync, exclusive, false);
 	if (r)
 		return r;
 
-	p->num_dw_left = ndw;
-	return 0;
+	return amdgpu_sync_resv(p->adev, &p->job->sync, root->tbo.resv,
+				owner, false);
 }
 
 /**
@@ -99,13 +99,21 @@ static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p,
 	struct dma_fence *f;
 	int r;
 
-	ring = container_of(p->vm->entity.rq->sched, struct amdgpu_ring, sched);
+	if (p->direct)
+		ring = p->adev->vm_manager.page_fault;
+	else
+		ring = container_of(p->vm->entity.rq->sched,
+				    struct amdgpu_ring, sched);
 
 	WARN_ON(ib->length_dw == 0);
 	amdgpu_ring_pad_ib(ring, ib);
 	WARN_ON(ib->length_dw > p->num_dw_left);
-	r = amdgpu_job_submit(p->job, &p->vm->entity,
-			      AMDGPU_FENCE_OWNER_VM, &f);
+
+	if (p->direct)
+		r = amdgpu_job_submit_direct(p->job, ring, &f);
+	else
+		r = amdgpu_job_submit(p->job, &p->vm->entity,
+				      AMDGPU_FENCE_OWNER_VM, &f);
 	if (r)
 		goto error;
 
@@ -120,7 +128,6 @@ static int amdgpu_vm_sdma_commit(struct amdgpu_vm_update_params *p,
 	return r;
 }
 
-
 /**
  * amdgpu_vm_sdma_copy_ptes - copy the PTEs from mapping
  *
-- 
2.17.1

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

  parent reply	other threads:[~2019-03-29 10:45 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-03-29 10:45 [PATCH 1/8] drm/amdgpu: fix ATC handling for Ryzen Christian König
     [not found] ` <20190329104507.2602-1-christian.koenig-5C7GfCeVMHo@public.gmane.org>
2019-03-29 10:45   ` [PATCH 2/8] drm/amdgpu: handle leaf PDEs as PTEs on Vega Christian König
2019-03-29 10:45   ` [PATCH 3/8] drm/amdgpu: provide the page fault queue to the VM code Christian König
     [not found]     ` <20190329104507.2602-3-christian.koenig-5C7GfCeVMHo@public.gmane.org>
2019-03-29 14:23       ` Alex Deucher
     [not found]         ` <CADnq5_Mv5dVc2o5yvnFRJXLAwxuv5ysFLOLGc9OTWeiVBZAeZA-JsoAwUIsXosN+BqQ9rBEUg@public.gmane.org>
2019-03-29 14:24           ` Christian König
2019-03-29 10:45   ` Christian König [this message]
2019-03-29 10:45   ` [PATCH 5/8] drm/amdgpu: allow direct submission of PDE updates Christian König
2019-03-29 10:45   ` [PATCH 6/8] drm/amdgpu: allow direct submission of PTE updates Christian König
2019-03-29 10:45   ` [PATCH 7/8] drm/amdgpu: allow direct submission of clears Christian König
2019-03-29 10:45   ` [PATCH 8/8] drm/amdgpu: add graceful VM fault handling Christian König
2019-03-30  0:41   ` [PATCH 1/8] drm/amdgpu: fix ATC handling for Ryzen Kuehling, Felix
     [not found]     ` <9cd61a18-fe16-f5aa-5b61-fef6d91d143b-5C7GfCeVMHo@public.gmane.org>
2019-04-01 11:23       ` Christian König
     [not found]         ` <7317b472-f24a-346a-7e6c-97c98b04e600-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2019-04-01 17:59           ` Kuehling, Felix
     [not found]             ` <ba45371a-7ea9-5144-c3c9-c59f93b434d8-5C7GfCeVMHo@public.gmane.org>
2019-04-01 18:03               ` Christian König
     [not found]                 ` <a40597e6-1f76-583b-55ad-089dcd9eeed4-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org>
2019-04-01 18:58                   ` Kuehling, Felix
     [not found]                     ` <32035eda-563d-b26b-83a6-71db67c846e6-5C7GfCeVMHo@public.gmane.org>
2019-04-03 17:24                       ` Koenig, Christian
     [not found]                         ` <c2cfeddb-7217-2110-d335-824b0a22371c-5C7GfCeVMHo@public.gmane.org>
2019-04-03 21:41                           ` Kuehling, Felix

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=20190329104507.2602-4-christian.koenig@amd.com \
    --to=ckoenig.leichtzumerken-re5jqeeqqe8avxtiumwx3w@public.gmane.org \
    --cc=amd-gfx-PD4FTy7X32lNgt0PjOBp9y5qC8QIuHrW@public.gmane.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