public inbox for linux-tegra@vger.kernel.org
 help / color / mirror / Atom feed
From: Thierry Reding <thierry.reding@gmail.com>
To: Thierry Reding <thierry.reding@gmail.com>
Cc: linux-tegra@vger.kernel.org, Dmitry Osipenko <digetx@gmail.com>,
	dri-devel@lists.freedesktop.org,
	Mikko Perttunen <mperttunen@nvidia.com>
Subject: [PATCH v3 09/16] gpu: host1x: Optimize CDMA push buffer memory usage
Date: Fri,  1 Feb 2019 14:28:30 +0100	[thread overview]
Message-ID: <20190201132837.12327-10-thierry.reding@gmail.com> (raw)
In-Reply-To: <20190201132837.12327-1-thierry.reding@gmail.com>

From: Thierry Reding <treding@nvidia.com>

The host1x CDMA push buffer is terminated by a special opcode (RESTART)
that tells the CDMA to wrap around to the beginning of the push buffer.
To accomodate the RESTART opcode, an extra 4 bytes are allocated on top
of the 512 * 8 = 4096 bytes needed for the 512 slots (1 slot = 2 words)
that are used for other commands passed to CDMA. This requires that two
memory pages are allocated, but most of the second page (4092 bytes) is
never used.

Decrease the number of slots to 511 so that the RESTART opcode fits
within the page. Adjust the push buffer wraparound code to take into
account push buffer sizes that are not a power of two.

Signed-off-by: Thierry Reding <treding@nvidia.com>
---
 drivers/gpu/host1x/cdma.c | 29 +++++++++++++++++++++++++----
 1 file changed, 25 insertions(+), 4 deletions(-)

diff --git a/drivers/gpu/host1x/cdma.c b/drivers/gpu/host1x/cdma.c
index a96c4dd1e449..50c1370b56c7 100644
--- a/drivers/gpu/host1x/cdma.c
+++ b/drivers/gpu/host1x/cdma.c
@@ -42,7 +42,17 @@
  * means that the push buffer is full, not empty.
  */
 
-#define HOST1X_PUSHBUFFER_SLOTS	512
+/*
+ * Typically the commands written into the push buffer are a pair of words. We
+ * use slots to represent each of these pairs and to simplify things. Note the
+ * strange number of slots allocated here. 512 slots will fit exactly within a
+ * single memory page. We also need one additional word at the end of the push
+ * buffer for the RESTART opcode that will instruct the CDMA to jump back to
+ * the beginning of the push buffer. With 512 slots, this means that we'll use
+ * 2 memory pages and waste 4092 bytes of the second page that will never be
+ * used.
+ */
+#define HOST1X_PUSHBUFFER_SLOTS	511
 
 /*
  * Clean up push buffer resources
@@ -148,7 +158,10 @@ static void host1x_pushbuffer_push(struct push_buffer *pb, u32 op1, u32 op2)
 	WARN_ON(pb->pos == pb->fence);
 	*(p++) = op1;
 	*(p++) = op2;
-	pb->pos = (pb->pos + 8) & (pb->size - 1);
+	pb->pos += 8;
+
+	if (pb->pos >= pb->size)
+		pb->pos -= pb->size;
 }
 
 /*
@@ -158,7 +171,10 @@ static void host1x_pushbuffer_push(struct push_buffer *pb, u32 op1, u32 op2)
 static void host1x_pushbuffer_pop(struct push_buffer *pb, unsigned int slots)
 {
 	/* Advance the next write position */
-	pb->fence = (pb->fence + slots * 8) & (pb->size - 1);
+	pb->fence += slots * 8;
+
+	if (pb->fence >= pb->size)
+		pb->fence -= pb->size;
 }
 
 /*
@@ -166,7 +182,12 @@ static void host1x_pushbuffer_pop(struct push_buffer *pb, unsigned int slots)
  */
 static u32 host1x_pushbuffer_space(struct push_buffer *pb)
 {
-	return ((pb->fence - pb->pos) & (pb->size - 1)) / 8;
+	unsigned int fence = pb->fence;
+
+	if (pb->fence < pb->pos)
+		fence += pb->size;
+
+	return (fence - pb->pos) / 8;
 }
 
 /*
-- 
2.19.1

_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/dri-devel

  parent reply	other threads:[~2019-02-01 13:28 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-02-01 13:28 [PATCH v3 00/16] drm/tegra: Fix IOVA space on Tegra186 and later Thierry Reding
2019-02-01 13:28 ` [PATCH v3 01/16] gpu: host1x: Set up stream ID table Thierry Reding
2019-02-01 13:28 ` [PATCH v3 02/16] gpu: host1x: Program the channel stream ID Thierry Reding
2019-02-01 13:28 ` [PATCH v3 03/16] gpu: host1x: Introduce support for wide opcodes Thierry Reding
2019-02-01 13:28 ` [PATCH v3 04/16] gpu: host1x: Support 40-bit addressing Thierry Reding
2019-02-01 13:28 ` [PATCH v3 05/16] gpu: host1x: Use direct DMA with IOMMU API usage Thierry Reding
2019-02-01 14:46   ` Dmitry Osipenko
2019-02-01 13:28 ` [PATCH v3 06/16] gpu: host1x: Restrict IOVA space to DMA mask Thierry Reding
2019-02-01 14:47   ` Dmitry Osipenko
2019-02-01 13:28 ` [PATCH v3 07/16] gpu: host1x: Support 40-bit addressing on Tegra186 Thierry Reding
2019-02-01 13:28 ` [PATCH v3 08/16] gpu: host1x: Use correct semantics for HOST1X_CHANNEL_DMAEND Thierry Reding
2019-02-01 13:37   ` Dmitry Osipenko
2019-02-01 13:41     ` Thierry Reding
2019-02-01 13:48       ` Dmitry Osipenko
2019-02-01 14:10         ` Thierry Reding
2019-02-01 14:40           ` Dmitry Osipenko
2019-02-01 14:47   ` Dmitry Osipenko
2019-02-01 13:28 ` Thierry Reding [this message]
2019-02-01 14:48   ` [PATCH v3 09/16] gpu: host1x: Optimize CDMA push buffer memory usage Dmitry Osipenko
2019-02-01 13:28 ` [PATCH v3 10/16] drm/tegra: Store parent pointer in Tegra DRM clients Thierry Reding
2019-02-01 14:48   ` Dmitry Osipenko
2019-02-01 13:28 ` [PATCH v3 11/16] drm/tegra: vic: Load firmware on demand Thierry Reding
2019-02-01 13:28 ` [PATCH v3 12/16] drm/tegra: Setup shared IOMMU domain after initialization Thierry Reding
2019-02-01 14:48   ` Dmitry Osipenko
2019-02-01 13:28 ` [PATCH v3 13/16] drm/tegra: Restrict IOVA space to DMA mask Thierry Reding
2019-02-01 14:49   ` Dmitry Osipenko
2019-02-01 13:28 ` [PATCH v3 14/16] drm/tegra: vic: Do not clear driver data Thierry Reding
2019-02-01 13:28 ` [PATCH v3 15/16] drm/tegra: vic: Support stream ID register programming Thierry Reding
2019-02-01 13:28 ` [PATCH v3 16/16] arm64: tegra: Enable SMMU for VIC on Tegra186 Thierry Reding
2019-02-01 14:46 ` [PATCH v3 00/16] drm/tegra: Fix IOVA space on Tegra186 and later Dmitry Osipenko

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=20190201132837.12327-10-thierry.reding@gmail.com \
    --to=thierry.reding@gmail.com \
    --cc=digetx@gmail.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=linux-tegra@vger.kernel.org \
    --cc=mperttunen@nvidia.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