Igt-dev Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Sunil Khatri <sunil.khatri@amd.com>
To: igt-dev@lists.freedesktop.org
Cc: "Alex Deucher" <alexander.deucher@amd.com>,
	"Christian König" <christian.koenig@amd.com>,
	"Vitaly Prosyak" <vitaly.prosyak@amd.com>,
	"Sunil Khatri" <sunil.khatri@amd.com>
Subject: [PATCH v3 10/19] lib/amdgpu: add macro for adding cmds in user queue
Date: Fri, 28 Mar 2025 13:54:07 +0530	[thread overview]
Message-ID: <20250328082416.1469810-10-sunil.khatri@amd.com> (raw)
In-Reply-To: <20250328082416.1469810-1-sunil.khatri@amd.com>

Add macros which enable adding cmds in user queue
for size more than ring size. Ring is treate as
a circular buffer and ovewrite from beginning
when write ptr reaches the end of queue.

Signed-off-by: Sunil Khatri <sunil.khatri@amd.com>
---
 lib/amdgpu/amd_user_queue.h | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)

diff --git a/lib/amdgpu/amd_user_queue.h b/lib/amdgpu/amd_user_queue.h
index 355f16f19..539a78819 100644
--- a/lib/amdgpu/amd_user_queue.h
+++ b/lib/amdgpu/amd_user_queue.h
@@ -15,9 +15,24 @@
 #define PAGE_SIZE 4096
 #endif
 
-#define USERMODE_QUEUE_SIZE		(PAGE_SIZE * 256)   //In bytes
+#define USERMODE_QUEUE_SIZE		(PAGE_SIZE * 256)   //In bytes with total size as 1 Mbyte
 #define ALIGNMENT			4096
 #define DOORBELL_INDEX			4
+#define USERMODE_QUEUE_SIZE_DW		(USERMODE_QUEUE_SIZE >> 2)
+#define USERMODE_QUEUE_SIZE_DW_MASK	(USERMODE_QUEUE_SIZE_DW - 1)
+
+#define amdgpu_pkt_begin() uint32_t __num_dw_written = 0; \
+	uint32_t __ring_start = *ring_context->wptr_cpu & USERMODE_QUEUE_SIZE_DW_MASK;
+
+#define amdgpu_pkt_add_dw(value) do { \
+	*(ring_context->queue_cpu + \
+	((__ring_start + __num_dw_written) & USERMODE_QUEUE_SIZE_DW_MASK)) \
+	= value; \
+	__num_dw_written++;\
+} while (0)
+
+#define amdgpu_pkt_end() \
+	*ring_context->wptr_cpu += __num_dw_written
 
 void amdgpu_alloc_doorbell(amdgpu_device_handle device_handle, struct amdgpu_userq_bo *doorbell_bo,
 			   unsigned int size, unsigned int domain);
-- 
2.43.0


  parent reply	other threads:[~2025-03-28  8:24 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-03-28  8:23 [PATCH v3 01/19] drm-uapi/amdgpu: sync with drm-next Sunil Khatri
2025-03-28  8:23 ` [PATCH v3 02/19] " Sunil Khatri
2025-03-31 19:11   ` vitaly prosyak
2025-04-01  4:39     ` Khatri, Sunil
2025-04-01  4:50       ` vitaly prosyak
2025-04-01  5:46         ` Khatri, Sunil
2025-04-01 16:09   ` Kamil Konieczny
2025-03-28  8:24 ` [PATCH v3 03/19] lib/amdgpu: Add user mode queue support in ring context Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 04/19] lib/amdgpu: Add support of amd user queues Sunil Khatri
2025-04-01  4:21   ` vitaly prosyak
2025-04-01  4:41     ` Khatri, Sunil
2025-03-28  8:24 ` [PATCH v3 05/19] lib/amdgpu: add func amdgpu_bo_alloc_and_map_sync Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 06/19] tests/amdgpu: Add user queue support for gfx and compute Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 07/19] tests/amdgpu: Add UMQ submission tests " Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 08/19] tests/amdgpu: Add amdgpu_sync_dependency_test with UMQ Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 09/19] tests/amdgpu: use memory API's from amd_memory.h Sunil Khatri
2025-03-28  8:24 ` Sunil Khatri [this message]
2025-03-28  8:24 ` [PATCH v3 11/19] lib/amdgpu: use macro to add cmds in the user ring Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 12/19] tests/amdgpu: Add amdgpu_cp_nops tests for UMQ Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 13/19] drm-uapi/amdgpu: sync with drm-next Sunil Khatri
2025-04-01 16:06   ` Kamil Konieczny
2025-04-01 23:52     ` vitaly prosyak
2025-04-02 10:51       ` Kamil Konieczny
2025-04-01 23:57     ` vitaly prosyak
2025-03-28  8:24 ` [PATCH v3 14/19] lib/amdgpu: use right API to get the correct size Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 15/19] lib/amdgpu: use a memory fence to serialize write Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 16/19] tests/amdgpu: disable check for IP presense with no kernel queue Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 17/19] lib/amdgpu: make the local functions as static Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 18/19] lib/amdgpu: enable UMQ function under macro Sunil Khatri
2025-03-28  8:24 ` [PATCH v3 19/19] tests/amdgpu: Disable the UMQ tests under a macro Sunil Khatri
2025-03-28 13:01 ` ✓ Xe.CI.BAT: success for series starting with [v3,01/19] drm-uapi/amdgpu: sync with drm-next Patchwork
2025-03-28 13:12 ` ✗ i915.CI.BAT: failure " Patchwork
2025-03-29  0:43 ` ✗ Xe.CI.Full: " Patchwork
2025-04-01 23:46 ` [PATCH v3 01/19] " vitaly prosyak
2025-04-06 18:47 ` ✗ Xe.CI.Full: failure for series starting with [v3,01/19] " 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=20250328082416.1469810-10-sunil.khatri@amd.com \
    --to=sunil.khatri@amd.com \
    --cc=alexander.deucher@amd.com \
    --cc=christian.koenig@amd.com \
    --cc=igt-dev@lists.freedesktop.org \
    --cc=vitaly.prosyak@amd.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