linux-kernel - Re: [PATCH v5 05/18] drm/rockchip: vop2: Set AXI id for rk3588

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <5843712.DvuYhMxLoT@bootstrap>
Date: Wed, 11 Dec 2024 10:45:01 -0500
From: Detlev Casanova <detlev.casanova@...labora.com>
To: Andy Yan <andyshrk@....com>
Cc: heiko@...ech.de, hjc@...k-chips.com, krzk+dt@...nel.org,
 s.hauer@...gutronix.de, devicetree@...r.kernel.org,
 dri-devel@...ts.freedesktop.org, linux-arm-kernel@...ts.infradead.org,
 linux-kernel@...r.kernel.org, linux-rockchip@...ts.infradead.org,
 derek.foreman@...labora.com, Andy Yan <andy.yan@...k-chips.com>,
 kernel@...labora.com
Subject: Re: [PATCH v5 05/18] drm/rockchip: vop2: Set AXI id for rk3588

On Wednesday, 11 December 2024 01:34:34 EST Andy Yan wrote:
> Hi Detlev,
> 
> At 2024-12-11 02:40:14, "Detlev Casanova" <detlev.casanova@...labora.com> 
wrote:
> >Hi Andy,
> >
> >On Monday, 9 December 2024 07:29:18 EST Andy Yan wrote:
> >> From: Andy Yan <andy.yan@...k-chips.com>
> >> 
> >> There are two AXI bus in vop2, windows attached on the same bus must
> >> have a unique channel YUV and RGB channel ID.
> >> 
> >> The default IDs will conflict with each other on the rk3588, so they
> >> need to be reassigned.
> >> 
> >> Fixes: 5a028e8f062f ("drm/rockchip: vop2: Add support for rk3588")
> >> Signed-off-by: Andy Yan <andy.yan@...k-chips.com>
> >> Tested-by: Derek Foreman <derek.foreman@...labora.com>
> >> 
> >> ---
> >> 
> >> Changes in v5:
> >> - Added in V5
> >> 
> >>  drivers/gpu/drm/rockchip/rockchip_drm_vop2.c | 14 +++++++++++
> >>  drivers/gpu/drm/rockchip/rockchip_drm_vop2.h |  9 +++++++
> >>  drivers/gpu/drm/rockchip/rockchip_vop2_reg.c | 26 +++++++++++++++++++-
> >>  3 files changed, 48 insertions(+), 1 deletion(-)
> >> 
> >> diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c
> >> b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c index
> >> dc4edd65bc9e..8b9ca046eeeb 100644
> >> --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c
> >> +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop2.c
> >> @@ -1426,6 +1426,12 @@ static void vop2_plane_atomic_update(struct
> >> drm_plane *plane, &fb->format->format,
> >> 
> >>  		afbc_en ? "AFBC" : "", &yrgb_mst);
> >> 
> >> +	if (vop2->data->soc_id > 3568) {
> >
> >Shouldn't this be done only for rk3588, as specified in the comments below
> >? The test we did before showed that it is failing on rk3576 and 3576 is >
> >3588.
> I think this is because you tested before with the patch I gave Derek 
> without axi id assigned for rk3576.
> I assigned axi id for rk3576 in this version 18/18（The new ID assignment can
> adapt to more application scenarios.）, can you test it with the whole V5
> version again?

Yes, I will do that. But patch 18 mentions:
+       /* Read only bit on rk3576*/
+       [VOP2_WIN_AXI_BUS_ID] = REG_FIELD(RK3568_CLUSTER_CTRL, 13, 13),

And the bit is being written here. If it is indeed writable, then I would drop 
that comment.

> >I suggest
> >
> >	if (vop2->data->soc_id == 3588) {
> >
> >Regards,
> >Detlev
> >
> >> +		vop2_win_write(win, VOP2_WIN_AXI_BUS_ID, win->data-
> >>
> >>axi_bus_id);
> >>
> >> +		vop2_win_write(win, VOP2_WIN_AXI_YRGB_R_ID, win->data-
> >>
> >>axi_yrgb_r_id);
> >>
> >> +		vop2_win_write(win, VOP2_WIN_AXI_UV_R_ID, win->data-
> >>
> >>axi_uv_r_id);
> >>
> >> +	}
> >> +
> >> 
> >>  	if (vop2_cluster_window(win))
> >>  	
> >>  		vop2_win_write(win, VOP2_WIN_AFBC_HALF_BLOCK_EN,
> >
> >half_block_en);
> >

Detlev.