All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Sasha Levin <sashal@kernel.org>,
	tomeu.vizoso@collabora.com, airlied@linux.ie,
	dri-devel@lists.freedesktop.org,
	Steven Price <steven.price@arm.com>,
	Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Subject: [PATCH AUTOSEL 5.10 06/46] drm/panfrost: Don't set L2_MMU_CONFIG quirks
Date: Thu, 11 Aug 2022 12:03:30 -0400	[thread overview]
Message-ID: <20220811160421.1539956-6-sashal@kernel.org> (raw)
In-Reply-To: <20220811160421.1539956-1-sashal@kernel.org>

From: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>

[ Upstream commit d8e53d8a4e0ae842ef5e83e0dfb0796980f710cf ]

L2_MMU_CONFIG is an implementation-defined register. Different Mali GPUs
define slightly different MAX_READS and MAX_WRITES fields, which
throttle outstanding reads and writes when set to non-zero values. When
left as zero, reads and writes are not throttled.

Both kbase and panfrost always zero these registers. Per discussion with
Steven Price, there are two reasons these quirks may be used:

1. Simulating slower memory subsystems. This use case is only of
   interest to system-on-chip designers; it is not relevant to mainline.

2. Working around broken memory subsystems. Hopefully we never see this
   case in mainline. If we do, we'll need to set this register based on
   an SoC-compatible, rather than generally matching on the GPU model.

To the best of our knowledge, these fields are zero at reset, so the
write is not necessary. Let's remove the write to aid porting to new
Mali GPUs, which have different layouts for the L2_MMU_CONFIG register.

Suggested-by: Steven Price <steven.price@arm.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220525145754.25866-8-alyssa.rosenzweig@collabora.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/panfrost/panfrost_gpu.c | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.c b/drivers/gpu/drm/panfrost/panfrost_gpu.c
index d22bb2589cb1..ce310b4496d8 100644
--- a/drivers/gpu/drm/panfrost/panfrost_gpu.c
+++ b/drivers/gpu/drm/panfrost/panfrost_gpu.c
@@ -127,18 +127,6 @@ static void panfrost_gpu_init_quirks(struct panfrost_device *pfdev)
 	gpu_write(pfdev, GPU_TILER_CONFIG, quirks);
 
 
-	quirks = gpu_read(pfdev, GPU_L2_MMU_CONFIG);
-
-	/* Limit read & write ID width for AXI */
-	if (panfrost_has_hw_feature(pfdev, HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG))
-		quirks &= ~(L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS |
-			    L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES);
-	else
-		quirks &= ~(L2_MMU_CONFIG_LIMIT_EXTERNAL_READS |
-			    L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES);
-
-	gpu_write(pfdev, GPU_L2_MMU_CONFIG, quirks);
-
 	quirks = 0;
 	if ((panfrost_model_eq(pfdev, 0x860) || panfrost_model_eq(pfdev, 0x880)) &&
 	    pfdev->features.revision >= 0x2000)
-- 
2.35.1


WARNING: multiple messages have this Message-ID (diff)
From: Sasha Levin <sashal@kernel.org>
To: linux-kernel@vger.kernel.org, stable@vger.kernel.org
Cc: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>,
	Steven Price <steven.price@arm.com>,
	Sasha Levin <sashal@kernel.org>,
	robh@kernel.org, tomeu.vizoso@collabora.com, airlied@linux.ie,
	daniel@ffwll.ch, dri-devel@lists.freedesktop.org
Subject: [PATCH AUTOSEL 5.10 06/46] drm/panfrost: Don't set L2_MMU_CONFIG quirks
Date: Thu, 11 Aug 2022 12:03:30 -0400	[thread overview]
Message-ID: <20220811160421.1539956-6-sashal@kernel.org> (raw)
In-Reply-To: <20220811160421.1539956-1-sashal@kernel.org>

From: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>

[ Upstream commit d8e53d8a4e0ae842ef5e83e0dfb0796980f710cf ]

L2_MMU_CONFIG is an implementation-defined register. Different Mali GPUs
define slightly different MAX_READS and MAX_WRITES fields, which
throttle outstanding reads and writes when set to non-zero values. When
left as zero, reads and writes are not throttled.

Both kbase and panfrost always zero these registers. Per discussion with
Steven Price, there are two reasons these quirks may be used:

1. Simulating slower memory subsystems. This use case is only of
   interest to system-on-chip designers; it is not relevant to mainline.

2. Working around broken memory subsystems. Hopefully we never see this
   case in mainline. If we do, we'll need to set this register based on
   an SoC-compatible, rather than generally matching on the GPU model.

To the best of our knowledge, these fields are zero at reset, so the
write is not necessary. Let's remove the write to aid porting to new
Mali GPUs, which have different layouts for the L2_MMU_CONFIG register.

Suggested-by: Steven Price <steven.price@arm.com>
Reviewed-by: Steven Price <steven.price@arm.com>
Signed-off-by: Alyssa Rosenzweig <alyssa.rosenzweig@collabora.com>
Link: https://patchwork.freedesktop.org/patch/msgid/20220525145754.25866-8-alyssa.rosenzweig@collabora.com
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/gpu/drm/panfrost/panfrost_gpu.c | 12 ------------
 1 file changed, 12 deletions(-)

diff --git a/drivers/gpu/drm/panfrost/panfrost_gpu.c b/drivers/gpu/drm/panfrost/panfrost_gpu.c
index d22bb2589cb1..ce310b4496d8 100644
--- a/drivers/gpu/drm/panfrost/panfrost_gpu.c
+++ b/drivers/gpu/drm/panfrost/panfrost_gpu.c
@@ -127,18 +127,6 @@ static void panfrost_gpu_init_quirks(struct panfrost_device *pfdev)
 	gpu_write(pfdev, GPU_TILER_CONFIG, quirks);
 
 
-	quirks = gpu_read(pfdev, GPU_L2_MMU_CONFIG);
-
-	/* Limit read & write ID width for AXI */
-	if (panfrost_has_hw_feature(pfdev, HW_FEATURE_3BIT_EXT_RW_L2_MMU_CONFIG))
-		quirks &= ~(L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_READS |
-			    L2_MMU_CONFIG_3BIT_LIMIT_EXTERNAL_WRITES);
-	else
-		quirks &= ~(L2_MMU_CONFIG_LIMIT_EXTERNAL_READS |
-			    L2_MMU_CONFIG_LIMIT_EXTERNAL_WRITES);
-
-	gpu_write(pfdev, GPU_L2_MMU_CONFIG, quirks);
-
 	quirks = 0;
 	if ((panfrost_model_eq(pfdev, 0x860) || panfrost_model_eq(pfdev, 0x880)) &&
 	    pfdev->features.revision >= 0x2000)
-- 
2.35.1


  parent reply	other threads:[~2022-08-11 16:05 UTC|newest]

Thread overview: 74+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-08-11 16:03 [PATCH AUTOSEL 5.10 01/46] drm/r128: Fix undefined behavior due to shift overflowing the constant Sasha Levin
2022-08-11 16:03 ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 02/46] ath10k: htt_tx: do not interpret Eth frames as WiFi Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 03/46] ath10k: fix misreported tx bandwidth for 160Mhz Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [Nouveau] [PATCH AUTOSEL 5.10 04/46] drm/nouveau: clear output poll workers before nouveau_fbcon_destroy() Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 05/46] drm/panfrost: Handle HW_ISSUE_TTRX_2968_TTRX_3162 Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-13 13:44   ` Pavel Machek
2022-08-13 13:44     ` Pavel Machek
2022-08-11 16:03 ` Sasha Levin [this message]
2022-08-11 16:03   ` [PATCH AUTOSEL 5.10 06/46] drm/panfrost: Don't set L2_MMU_CONFIG quirks Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 07/46] ath10k: fix regdomain info of iw reg set/get Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 08/46] drm/amd/display: Detect dpcd_rev when hotplug mst monitor Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 09/46] drm/amd/display: Fix dpp dto for disabled pipes Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 10/46] drm/radeon: integer overflow in radeon_mode_dumb_create() Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 11/46] drm/radeon: Initialize fences array entries in radeon_sa_bo_next_hole Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 12/46] udmabuf: Set the DMA mask for the udmabuf device (v2) Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 13/46] net/mlx5: Add HW definitions of vport debug counters Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 14/46] drm/amd/display: Fix monitor flash issue Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 15/46] selftests: mlxsw: resource_scale: Allow skipping a test Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 16/46] net: dsa: ar9331: fix potential dead lock on mdio access Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 17/46] media: davinci: vpif: add missing of_node_put() in vpif_probe() Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 18/46] media: mediatek: vcodec: prevent kernel crash when scp ipi timeout Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 19/46] media: airspy: respect the DMA coherency rules Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 20/46] media: pvrusb2: fix memory leak in pvr_probe Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 21/46] rcu: Apply noinstr to rcu_idle_enter() and rcu_idle_exit() Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 22/46] drm/bridge/tc358775: Fix DSI clock division for vsync delay calculation Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 23/46] uprobe: gate bpf call behind BPF_EVENTS Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 24/46] rcuscale: Fix smp_processor_id()-in-preemptible warnings Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 25/46] mlxsw: cmd: Increase 'config_profile.flood_mode' length Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 26/46] media: atmel: atmel-isc-base: allow wb ctrls to be changed when isc is not configured Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 27/46] kselftests: Enable the echo command to print newlines in Makefile Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 28/46] ipv6/addrconf: fix timing bug in tempaddr regen Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 29/46] drm/amdgpu/display/dc: Fix null pointer exception Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 30/46] crypto: vmx - Fix warning on p8_ghash_alg Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [Nouveau] [PATCH AUTOSEL 5.10 31/46] drm/nouveau/nvkm: use list_add_tail() when building object tree Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03   ` Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 32/46] crypto: ccree - Add missing clk_disable_unprepare() in cc_pm_resume() Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 33/46] bpf: Fix check against plain integer v 'NULL' Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 34/46] media: rkisp1: Disable runtime PM in probe error path Sasha Levin
2022-08-11 16:03 ` [PATCH AUTOSEL 5.10 35/46] bpf: Don't redirect packets with invalid pkt_len Sasha Levin
2022-08-11 16:04 ` [PATCH AUTOSEL 5.10 36/46] can: sja1000: Add Quirk for RZ/N1 SJA1000 CAN controller Sasha Levin
2022-08-11 16:04 ` [PATCH AUTOSEL 5.10 37/46] net/cdc_ncm: Increase NTB max RX/TX values to 64kb Sasha Levin
2022-08-11 16:04 ` [PATCH AUTOSEL 5.10 38/46] Bluetooth: hci_bcm: Add BCM4349B1 variant Sasha Levin
2022-08-11 16:04 ` [PATCH AUTOSEL 5.10 39/46] Bluetooth: Collect kcov coverage from hci_rx_work Sasha Levin
2022-08-11 16:04 ` [PATCH AUTOSEL 5.10 40/46] bpf/selftests: Fix couldn't retrieve pinned program in xdp veth test Sasha Levin
2022-08-11 16:04 ` [PATCH AUTOSEL 5.10 41/46] net: ethernet: stmicro: stmmac: first disable all queues and disconnect in release Sasha Levin
2022-08-11 16:04   ` Sasha Levin
2022-08-11 16:04 ` [PATCH AUTOSEL 5.10 42/46] wifi: rtl8xxxu: Fix the error handling of the probe function Sasha Levin
2022-08-12 19:06 ` [PATCH AUTOSEL 5.10 01/46] drm/r128: Fix undefined behavior due to shift overflowing the constant Pavel Machek
2022-08-12 19:06   ` Pavel Machek

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=20220811160421.1539956-6-sashal@kernel.org \
    --to=sashal@kernel.org \
    --cc=airlied@linux.ie \
    --cc=alyssa.rosenzweig@collabora.com \
    --cc=dri-devel@lists.freedesktop.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=steven.price@arm.com \
    --cc=tomeu.vizoso@collabora.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.