From: Daniel Vetter <daniel@ffwll.ch>
To: Xinliang Liu <xinliang.liu@linaro.org>
Cc: linux-doc@vger.kernel.org, catalin.marinas@arm.com,
will.deacon@arm.com, linuxarm@huawei.com,
dri-devel@lists.freedesktop.org, xuwei5@hisilicon.com,
benjamin.gaignard@linaro.org, zourongrong@huawei.com,
xuyiping@hisilicon.com, corbet@lwn.net, w.f@huawei.com,
lijianhua@huawei.com, devicetree@vger.kernel.org,
bintian.wang@huawei.com, puck.chen@hisilicon.com,
linux-arm-kernel@lists.infradead.org, andy.green@linaro.org,
liguozhu@hisilicon.com, haojian.zhuang@linaro.org
Subject: Re: [PATCH v2 06/10] drm/hisilicon: Add vblank feature
Date: Mon, 30 Nov 2015 08:54:32 +0100 [thread overview]
Message-ID: <20151130075432.GF17050@phenom.ffwll.local> (raw)
In-Reply-To: <1448707145-69348-7-git-send-email-xinliang.liu@linaro.org>
On Sat, Nov 28, 2015 at 06:39:01PM +0800, Xinliang Liu wrote:
> Add vblank handle for ADE.
>
> Signed-off-by: Xinliang Liu <xinliang.liu@linaro.org>
> Signed-off-by: Xinwei Kong <kong.kongxinwei@hisilicon.com>
> Signed-off-by: Andy Green <andy.green@linaro.org>
> ---
> drivers/gpu/drm/hisilicon/hisi_drm_ade.c | 78 ++++++++++++++++++++++++++++++++
> drivers/gpu/drm/hisilicon/hisi_drm_ade.h | 16 +++++++
> drivers/gpu/drm/hisilicon/hisi_drm_drv.c | 19 +++++++-
> 3 files changed, 112 insertions(+), 1 deletion(-)
> create mode 100644 drivers/gpu/drm/hisilicon/hisi_drm_ade.h
>
> diff --git a/drivers/gpu/drm/hisilicon/hisi_drm_ade.c b/drivers/gpu/drm/hisilicon/hisi_drm_ade.c
> index b0976c3..acb11e7 100644
> --- a/drivers/gpu/drm/hisilicon/hisi_drm_ade.c
> +++ b/drivers/gpu/drm/hisilicon/hisi_drm_ade.c
> @@ -267,7 +267,79 @@ static void ade_power_down(struct ade_hw_ctx *ctx)
> ctx->power_on = false;
> }
>
> +static struct drm_crtc *hisi_get_crtc_from_index(struct drm_device *dev,
> + unsigned int index)
Ugly that you had to add this, but unfortunately necessary :( Fixing up
the drm vblank hooks so that they deal with struct drm_crtc directly is
somewhere on my todo. But drm_irq.c is still a bit a mess, so this is some
ways off still.
What might be possible as a follow-up cleanup though is to add
vblank_enable and vblank_disable functions to struct
drm_crtc_helper_funcs. And then provide this code here to map from int
index to struct drm_crtc * as helpers in a new drm_vblank_helper.c file.
That might be a good intermediate step.
But nothing that needs to be done before merging hisilicon, that's for
sure.
One more comment below.
> +{
> + unsigned int index_tmp = 0;
> + struct drm_crtc *crtc;
> +
> + list_for_each_entry(crtc, &dev->mode_config.crtc_list, head) {
> + if (index_tmp == index)
> + return crtc;
> +
> + index_tmp++;
> + }
> +
> + WARN_ON(true);
> +}
> +
> +int ade_enable_vblank(struct drm_device *dev, int crtc_index)
> +{
> + struct drm_crtc *crtc = hisi_get_crtc_from_index(dev, crtc_index);
> + struct ade_crtc *acrtc = to_ade_crtc(crtc);
> + struct ade_hw_ctx *ctx = acrtc->ctx;
> + void __iomem *base = ctx->base;
> + u32 intr_en;
> +
> + DRM_INFO("enable_vblank enter.\n");
> + if (!ctx->power_on)
> + (void)ade_power_up(ctx);
> +
> + intr_en = readl(base + LDI_INT_EN);
> + intr_en |= LDI_ISR_FRAME_END_INT;
> + writel(intr_en, base + LDI_INT_EN);
>
> + return 0;
> +}
> +
> +void ade_disable_vblank(struct drm_device *dev, int crtc_index)
> +{
> + struct drm_crtc *crtc = hisi_get_crtc_from_index(dev, crtc_index);
> + struct ade_crtc *acrtc = to_ade_crtc(crtc);
> + struct ade_hw_ctx *ctx = acrtc->ctx;
> + void __iomem *base = ctx->base;
> + u32 intr_en;
> +
> + DRM_INFO("disable_vblank enter.\n");
> + if (!ctx->power_on) {
> + DRM_ERROR("power is down! vblank disable fail\n");
> + return;
> + }
> + intr_en = readl(base + LDI_INT_EN);
> + intr_en &= ~LDI_ISR_FRAME_END_INT;
> + writel(intr_en, base + LDI_INT_EN);
> +}
> +
> +static irqreturn_t ade_irq_handler(int irq, void *data)
> +{
> + struct ade_crtc *acrtc = data;
> + struct ade_hw_ctx *ctx = acrtc->ctx;
> + struct drm_crtc *crtc = &acrtc->base;
> + struct drm_device *dev = crtc->dev;
> + void __iomem *base = ctx->base;
> + u32 status;
> +
> + status = readl(base + LDI_MSK_INT);
> + /* DRM_INFO("LDI IRQ: status=0x%X\n",status); */
> +
> + /* vblank irq */
> + if (status & LDI_ISR_FRAME_END_INT) {
> + writel(LDI_ISR_FRAME_END_INT, base + LDI_INT_CLR);
> + drm_handle_vblank(dev, drm_crtc_index(crtc));
drm_crtc_handle_vblank please. At least when calling into the vblank code
drivers don't have to do the struct drm_crtc -> int index conversion any
more. Please make sure you do that everywhere, in case I've missed one.
Cheers, Daniel
> + }
> +
> + return IRQ_HANDLED;
> +}
>
> /*
> * set modules' reset mode: by software or hardware
> @@ -858,6 +930,12 @@ static int ade_bind(struct device *dev, struct device *master, void *data)
> if (ret)
> return ret;
>
> + /* vblank irq init */
> + ret = request_irq(ctx->irq, ade_irq_handler, DRIVER_IRQ_SHARED,
> + drm_dev->driver->name, acrtc);
> + if (ret)
> + return ret;
> +
> return 0;
> }
>
> diff --git a/drivers/gpu/drm/hisilicon/hisi_drm_ade.h b/drivers/gpu/drm/hisilicon/hisi_drm_ade.h
> new file mode 100644
> index 0000000..d1d7b5d
> --- /dev/null
> +++ b/drivers/gpu/drm/hisilicon/hisi_drm_ade.h
> @@ -0,0 +1,16 @@
> +/*
> + * Copyright (c) 2014-2015 Hisilicon Limited.
> + *
> + * This program is free software; you can redistribute it and/or modify
> + * it under the terms of the GNU General Public License version 2 as
> + * published by the Free Software Foundation.
> + *
> + */
> +
> +#ifndef __HISI_DRM_ADE_H__
> +#define __HISI_DRM_ADE_H__
> +
> +int ade_enable_vblank(struct drm_device *dev, int crtc_index);
> +void ade_disable_vblank(struct drm_device *dev, int crtc_index);
> +
> +#endif
> diff --git a/drivers/gpu/drm/hisilicon/hisi_drm_drv.c b/drivers/gpu/drm/hisilicon/hisi_drm_drv.c
> index d0eca80..13f59aa 100644
> --- a/drivers/gpu/drm/hisilicon/hisi_drm_drv.c
> +++ b/drivers/gpu/drm/hisilicon/hisi_drm_drv.c
> @@ -21,6 +21,7 @@
> #include <drm/drm_fb_cma_helper.h>
> #include <drm/drm_atomic_helper.h>
>
> +#include "hisi_drm_ade.h"
> #include "hisi_drm_drv.h"
>
> #define DRIVER_NAME "hisi-drm"
> @@ -29,6 +30,7 @@ static int hisi_drm_unload(struct drm_device *dev)
> {
> struct hisi_drm_private *priv = dev->dev_private;
>
> + drm_vblank_cleanup(dev);
> drm_mode_config_cleanup(dev);
> devm_kfree(dev->dev, priv);
> dev->dev_private = NULL;
> @@ -76,11 +78,22 @@ static int hisi_drm_load(struct drm_device *dev, unsigned long flags)
> goto err_mode_config_cleanup;
> }
>
> + /* vblank init */
> + ret = drm_vblank_init(dev, dev->mode_config.num_crtc);
> + if (ret) {
> + DRM_ERROR("failed to initialize vblank.\n");
> + goto err_unbind_all;
> + }
> + /* with irq_enabled = true, we can use the vblank feature. */
> + dev->irq_enabled = true;
> +
> /* reset all the states of crtc/plane/encoder/connector */
> drm_mode_config_reset(dev);
>
> return 0;
>
> +err_unbind_all:
> + component_unbind_all(dev->dev, dev);
> err_mode_config_cleanup:
> drm_mode_config_cleanup(dev);
> devm_kfree(dev->dev, priv);
> @@ -126,7 +139,7 @@ static int hisi_gem_cma_dumb_create(struct drm_file *file,
>
> static struct drm_driver hisi_drm_driver = {
> .driver_features = DRIVER_GEM | DRIVER_MODESET | DRIVER_PRIME |
> - DRIVER_ATOMIC,
> + DRIVER_ATOMIC | DRIVER_HAVE_IRQ,
> .load = hisi_drm_load,
> .unload = hisi_drm_unload,
> .fops = &hisi_drm_fops,
> @@ -148,6 +161,10 @@ static struct drm_driver hisi_drm_driver = {
> .gem_prime_vunmap = drm_gem_cma_prime_vunmap,
> .gem_prime_mmap = drm_gem_cma_prime_mmap,
>
> + .get_vblank_counter = drm_vblank_count,
> + .enable_vblank = ade_enable_vblank,
> + .disable_vblank = ade_disable_vblank,
> +
> .name = "hisi",
> .desc = "Hisilicon SoCs' DRM Driver",
> .date = "20150718",
> --
> 1.9.1
>
--
Daniel Vetter
Software Engineer, Intel Corporation
http://blog.ffwll.ch
_______________________________________________
dri-devel mailing list
dri-devel@lists.freedesktop.org
http://lists.freedesktop.org/mailman/listinfo/dri-devel
next prev parent reply other threads:[~2015-11-30 7:54 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2015-11-28 10:38 [PATCH v2 00/10] Add New DRM Driver for HiSilicon hi6220 SoC Xinliang Liu
2015-11-28 10:38 ` [PATCH v2 01/10] arm64: dts: hisilicon: Add display subsystem DT nodes for hi6220 Xinliang Liu
2015-11-28 10:38 ` [PATCH v2 02/10] drm/hisilicon: Add DT binding docs for hi6220 display subsystem Xinliang Liu
2015-11-30 19:31 ` Rob Herring
2015-12-01 7:17 ` Xinliang Liu
2015-12-01 13:58 ` Rob Herring
2015-11-28 10:38 ` [PATCH v2 03/10] drm/hisilicon: Add hisilicon DRM master driver Xinliang Liu
2015-12-03 16:21 ` Rob Herring
2015-12-05 1:25 ` Xinliang Liu
2015-11-28 10:38 ` [PATCH v2 04/10] drm/hisilicon: Add crtc funcs for ADE Xinliang Liu
[not found] ` <1448707145-69348-5-git-send-email-xinliang.liu-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
2015-11-28 15:56 ` Emil Velikov
2015-12-01 2:52 ` Xinliang Liu
2015-11-28 10:39 ` [PATCH v2 05/10] drm/hisilicon: Add plane " Xinliang Liu
2015-11-28 10:39 ` [PATCH v2 06/10] drm/hisilicon: Add vblank feature Xinliang Liu
2015-11-30 7:54 ` Daniel Vetter [this message]
2015-12-01 3:16 ` Xinliang Liu
2015-12-01 7:13 ` Daniel Vetter
2015-12-01 10:34 ` Xinliang Liu
2015-12-01 12:54 ` Daniel Vetter
2015-11-28 10:39 ` [PATCH v2 07/10] drm/hisilicon: Add cma fbdev and hotplug Xinliang Liu
2015-11-28 10:39 ` [PATCH v2 08/10] drm/hisilicon: Add dsi encoder driver Xinliang Liu
2015-12-01 8:58 ` Archit Taneja
[not found] ` <565D612A.1010307-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2015-12-01 11:16 ` Xinliang Liu
2015-11-28 10:39 ` [PATCH v2 09/10] drm/hisilicon: Add dsi host driver Xinliang Liu
[not found] ` <1448707145-69348-1-git-send-email-xinliang.liu-QSEj5FYQhm4dnm+yROfE0A@public.gmane.org>
2015-11-28 10:39 ` [PATCH v2 10/10] drm/hisilicon: Add support for external bridge Xinliang Liu
2015-12-01 9:04 ` Archit Taneja
2015-12-01 14:50 ` Xinliang Liu
2015-12-02 8:20 ` Archit Taneja
[not found] ` <565EA9D4.4090905-sgV2jX0FEOL9JmXXK+q4OQ@public.gmane.org>
2015-12-02 11:24 ` Xinliang Liu
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=20151130075432.GF17050@phenom.ffwll.local \
--to=daniel@ffwll.ch \
--cc=andy.green@linaro.org \
--cc=benjamin.gaignard@linaro.org \
--cc=bintian.wang@huawei.com \
--cc=catalin.marinas@arm.com \
--cc=corbet@lwn.net \
--cc=devicetree@vger.kernel.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=haojian.zhuang@linaro.org \
--cc=liguozhu@hisilicon.com \
--cc=lijianhua@huawei.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-doc@vger.kernel.org \
--cc=linuxarm@huawei.com \
--cc=puck.chen@hisilicon.com \
--cc=w.f@huawei.com \
--cc=will.deacon@arm.com \
--cc=xinliang.liu@linaro.org \
--cc=xuwei5@hisilicon.com \
--cc=xuyiping@hisilicon.com \
--cc=zourongrong@huawei.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