From: "Heiko Stübner" <heiko@sntech.de>
To: Andy Yan <andyshrk@163.com>
Cc: hjc@rock-chips.com, dri-devel@lists.freedesktop.org,
linux-rockchip@lists.infradead.org, linux-kernel@vger.kernel.org,
krzysztof.kozlowski+dt@linaro.org, robh+dt@kernel.org,
devicetree@vger.kernel.org, sebastian.reichel@collabora.com,
kever.yang@rock-chips.com, chris.obbard@collabora.com,
s.hauer@pengutronix.de, Andy Yan <andy.yan@rock-chips.com>
Subject: Re: [PATCH v2 10/12] drm/rockchip: vop2: Add support for rk3588
Date: Mon, 27 Nov 2023 16:29:45 +0100 [thread overview]
Message-ID: <4788319.uZKlY2gecq@diego> (raw)
In-Reply-To: <20231122125544.3454918-1-andyshrk@163.com>
Hi Andy,
Am Mittwoch, 22. November 2023, 13:55:44 CET schrieb Andy Yan:
> From: Andy Yan <andy.yan@rock-chips.com>
>
> VOP2 on rk3588:
>
> Four video ports:
> VP0 Max 4096x2160
> VP1 Max 4096x2160
> VP2 Max 4096x2160
> VP3 Max 2048x1080
>
> 4 4K Cluster windows with AFBC/line RGB and AFBC-only YUV support
> 4 4K Esmart windows with line RGB/YUV support
>
> Signed-off-by: Andy Yan <andy.yan@rock-chips.com>
>
> ---
>
> Changes in v2:
> - add rk3588_ prefix for functions which are rk3588 only
> - make some calculation as fixed value and keep calculation formula as
> comment
> - check return value for some cru calculation functions.
> - check return value for syscon_regmap_lookup_by_phandle
> - add NV20/NV30 for esmart plane
>
> drivers/gpu/drm/rockchip/rockchip_drm_vop2.c | 381 ++++++++++++++++++-
> drivers/gpu/drm/rockchip/rockchip_drm_vop2.h | 66 ++++
> drivers/gpu/drm/rockchip/rockchip_vop2_reg.c | 221 +++++++++++
> 3 files changed, 660 insertions(+), 8 deletions(-)
>
> diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c
> index 4bcc405bcf11..9eecbe1f71f9 100644
> --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c
> +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c
> @@ -271,9 +282,12 @@ static bool vop2_cluster_window(const struct vop2_win *win)
> static void vop2_cfg_done(struct vop2_video_port *vp)
> {
> struct vop2 *vop2 = vp->vop2;
> + u32 val;
> +
> + val = BIT(vp->id) | (BIT(vp->id) << 16) |
> + RK3568_REG_CFG_DONE__GLB_CFG_DONE_EN;
>
> - regmap_set_bits(vop2->map, RK3568_REG_CFG_DONE,
> - BIT(vp->id) | RK3568_REG_CFG_DONE__GLB_CFG_DONE_EN);
> + regmap_set_bits(vop2->map, RK3568_REG_CFG_DONE, val);
I don't fully understand that code:
(1) the write mask is also present on the rk3568, so should this change
be a separate patch with a fixes tag?
(2) RK3568_REG_CFG_DONE__GLB_CFG_DONE_EN does not contain the part for
the write-mask
#define RK3568_REG_CFG_DONE__GLB_CFG_DONE_EN BIT(15)
why is this working then?
> }
>
> static void vop2_win_disable(struct vop2_win *win)
[...]
> @@ -1298,7 +1346,11 @@ static void vop2_plane_atomic_update(struct drm_plane *plane,
> vop2_win_write(win, VOP2_WIN_AFBC_ENABLE, 1);
> vop2_win_write(win, VOP2_WIN_AFBC_FORMAT, afbc_format);
> vop2_win_write(win, VOP2_WIN_AFBC_UV_SWAP, uv_swap);
> - vop2_win_write(win, VOP2_WIN_AFBC_AUTO_GATING_EN, 0);
> + if (vop2->data->soc_id == 3566 || vop2->data->soc_id == 3568)
> + vop2_win_write(win, VOP2_WIN_AFBC_AUTO_GATING_EN, 0);
> + else
> + vop2_win_write(win, VOP2_WIN_AFBC_AUTO_GATING_EN, 1);
> +
I think this at least warrants a comment, what is happening here. Also,
can you already see how future vop2-users are behaving - aka are all new
socs in the "else" part of the conditional, or would a switch-case better
represent future socs?
> vop2_win_write(win, VOP2_WIN_AFBC_BLOCK_SPLIT_EN, 0);
> transform_offset = vop2_afbc_transform_offset(pstate, half_block_en);
> vop2_win_write(win, VOP2_WIN_AFBC_HDR_PTR, yrgb_mst);
> @@ -1627,9 +1937,17 @@ static void vop2_crtc_atomic_enable(struct drm_crtc *crtc,
> drm_for_each_encoder_mask(encoder, crtc->dev, crtc_state->encoder_mask) {
> struct rockchip_encoder *rkencoder = to_rockchip_encoder(encoder);
>
> - rk3568_set_intf_mux(vp, rkencoder->crtc_endpoint_id, polflags);
> + /*
> + * for drive a high resolution(4KP120, 8K), vop on rk3588/rk3576 need
> + * process multi(1/2/4/8) pixels per cycle, so the dclk feed by the
> + * system cru may be the 1/2 or 1/4 of mode->clock.
> + */
> + clock = vop2_set_intf_mux(vp, rkencoder->crtc_endpoint_id, polflags);
> }
>
> + if (!clock)
> + return;
> +
hmm, shouldn't the check for the validity of a mode happen before
atomic_enable is run? So this shouldn't error out in the middle of the
function?
> if (vcstate->output_mode == ROCKCHIP_OUT_MODE_AAAA &&
> !(vp_data->feature & VOP_FEATURE_OUTPUT_10BIT))
> out_mode = ROCKCHIP_OUT_MODE_P888;
Thanks
Heiko
next prev parent reply other threads:[~2023-11-27 15:29 UTC|newest]
Thread overview: 39+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-11-22 12:53 [PATCH v2 00/12] Add VOP2 support on rk3588 Andy Yan
2023-11-22 12:53 ` [PATCH v2 01/12] drm/rockchip: move output interface related definition to rockchip_drm_drv.h Andy Yan
2023-11-27 14:03 ` Sascha Hauer
2023-11-22 12:54 ` [PATCH v2 02/12] Revert "drm/rockchip: vop2: Use regcache_sync() to fix suspend/resume" Andy Yan
2023-11-27 14:04 ` Sascha Hauer
2023-11-22 12:54 ` [PATCH v2 03/12] drm/rockchip: vop2: set half_block_en bit in all mode Andy Yan
2023-11-27 14:04 ` Sascha Hauer
2023-11-27 15:01 ` Heiko Stübner
2023-11-22 12:54 ` [PATCH v2 04/12] drm/rockchip: vop2: clear afbc en and transform bit for cluster window at linear mode Andy Yan
2023-11-27 14:04 ` Sascha Hauer
2023-11-27 15:02 ` Heiko Stübner
2023-11-28 8:03 ` Andy Yan
2023-11-28 8:30 ` Heiko Stübner
2023-11-28 9:44 ` Andy Yan
2023-11-22 12:54 ` [PATCH v2 05/12] drm/rockchip: vop2: Set YUV/RGB overlay mode Andy Yan
2023-11-27 14:16 ` Sascha Hauer
2023-11-29 6:52 ` Andy Yan
2023-11-22 12:54 ` [PATCH v2 06/12] drm/rockchip: vop2: rename grf to sys_grf Andy Yan
2023-11-27 14:17 ` Sascha Hauer
2023-11-22 12:55 ` [PATCH v2 07/12] dt-bindings: soc: rockchip: add rk3588 vop/vo syscon Andy Yan
2023-11-22 12:55 ` [PATCH v2 08/12] dt-bindings: display: vop2: Add rk3588 support Andy Yan
2023-11-22 19:07 ` Krzysztof Kozlowski
2023-11-30 12:32 ` Andy Yan
2023-11-22 12:55 ` [PATCH v2 09/12] dt-bindings: soc: vop2: Add more endpoint definition Andy Yan
2023-11-22 19:08 ` Krzysztof Kozlowski
2023-11-22 12:55 ` [PATCH v2 10/12] drm/rockchip: vop2: Add support for rk3588 Andy Yan
2023-11-27 11:19 ` Sascha Hauer
2023-11-29 8:28 ` Andy Yan
2023-11-27 15:29 ` Heiko Stübner [this message]
2023-11-28 9:32 ` Andy Yan
2023-11-28 9:44 ` Heiko Stübner
2023-11-28 10:11 ` Andy Yan
2023-11-22 12:56 ` [PATCH v2 11/12] drm/rockchip: vop2: Add debugfs support Andy Yan
2023-11-27 10:13 ` Sascha Hauer
[not found] ` <ea24a638-d10f-4f58-9992-1c80bafdd6d4@rock-chips.com>
2023-11-29 8:52 ` Sascha Hauer
2023-11-29 11:01 ` Andy Yan
2023-11-29 12:59 ` Sascha Hauer
2023-11-30 1:15 ` Andy Yan
2023-11-22 12:56 ` [PATCH v2 12/12] arm64: dts: rockchip: Add vop on rk3588 Andy Yan
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=4788319.uZKlY2gecq@diego \
--to=heiko@sntech.de \
--cc=andy.yan@rock-chips.com \
--cc=andyshrk@163.com \
--cc=chris.obbard@collabora.com \
--cc=devicetree@vger.kernel.org \
--cc=dri-devel@lists.freedesktop.org \
--cc=hjc@rock-chips.com \
--cc=kever.yang@rock-chips.com \
--cc=krzysztof.kozlowski+dt@linaro.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-rockchip@lists.infradead.org \
--cc=robh+dt@kernel.org \
--cc=s.hauer@pengutronix.de \
--cc=sebastian.reichel@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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).