linux-kernel - Re: [PATCH v2 5/5] drm/rockchip: default enable win2/3 area0 bit

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: <CAAFQd5CDYuZkO_1QGdwK80PkTuyiK95VY+KGX-0UBh1hex-5RA@mail.gmail.com>
Date:	Tue, 21 Jul 2015 16:33:39 +0900
From:	Tomasz Figa <tfiga@...omium.org>
To:	Mark yao <mark.yao@...k-chips.com>
Cc:	dri-devel <dri-devel@...ts.freedesktop.org>,
	David Airlie <airlied@...ux.ie>,
	Heiko Stuebner <heiko@...ech.de>,
	Daniel Kurtz <djkurtz@...omium.org>,
	Philipp Zabel <p.zabel@...gutronix.de>,
	Daniel Vetter <daniel@...ll.ch>,
	Rob Clark <robdclark@...il.com>,
	"linux-arm-kernel@...ts.infradead.org" 
	<linux-arm-kernel@...ts.infradead.org>,
	"open list:ARM/Rockchip SoC..." <linux-rockchip@...ts.infradead.org>,
	"linux-kernel@...r.kernel.org" <linux-kernel@...r.kernel.org>,
	sandy.huang@...k-chips.com, dkm@...k-chips.com, zwl@...k-chips.com,
	xw@...k-chips.com
Subject: Re: [PATCH v2 5/5] drm/rockchip: default enable win2/3 area0 bit

On Fri, Jul 3, 2015 at 7:08 PM, Mark yao <mark.yao@...k-chips.com> wrote:
> On 2015年07月03日 17:24, Tomasz Figa wrote:
>>
>> On Fri, Jul 3, 2015 at 5:19 PM, Mark yao <mark.yao@...k-chips.com> wrote:
>>>
>>> On 2015年07月03日 16:02, Tomasz Figa wrote:
>>>>
>>>> Hi Mark,
>>>>
>>>> Please see my comments inline.
>>>>
>>>> On Fri, Jun 26, 2015 at 7:10 PM, Mark Yao <mark.yao@...k-chips.com>
>>>> wrote:
>>>>>
>>>>> Win2/3 support 4 area display, but now havn't found a suitable
>>>>> way to use it, and it enable by win gate and area gate,
>>>>> so default enable area0 gate, so that its behaviour just like a
>>>>> win.
>>>>
>>>> So I assume this means that currently, without those bits set, win2
>>>> and win3 do not work? This would make this patch a fix maybe even with
>>>> a potential backportability.
>>>
>>>
>>> Yes, without this patch, all win2/3 area gate default disabled.
>>> vop_update_plane_event call win enable only enable the win gate.
>>>
>>>
>>>>> Signed-off-by: Mark Yao <mark.yao@...k-chips.com>
>>>>>
>>>>> ---
>>>>> Changes in v2: None
>>>>>
>>>>>    drivers/gpu/drm/rockchip/rockchip_drm_vop.c |    6 ++++++
>>>>>    1 file changed, 6 insertions(+)
>>>>>
>>>>> diff --git a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
>>>>> b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
>>>>> index 40107bb..e001d26 100644
>>>>> --- a/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
>>>>> +++ b/drivers/gpu/drm/rockchip/rockchip_drm_vop.c
>>>>> @@ -337,6 +337,12 @@ static const struct vop_reg_data
>>>>> vop_init_reg_table[] = {
>>>>>           {DSP_CTRL0, 0x00000000},
>>>>>           {WIN0_CTRL0, 0x00000080},
>>>>>           {WIN1_CTRL0, 0x00000080},
>>>>> +       /*
>>>>> +        * Todo: win2/3 support area func, but now havn't found a
>>>>> suitable
>>>>> +        * way to use it, so default enable area0 as a win display.
>>>>
>>>> TODO: Win2/3 support multiple area function, but we haven't found
>>>> a suitable way to use it yet, so let's just use them as other windows
>>>> with only area 0 enabled.
>>>>
>>>>> +        */
>>>>> +       {WIN2_CTRL0, 0x00000010},
>>>>> +       {WIN3_CTRL0, 0x00000010},
>>>>
>>>> Anyway, is it enough to program those registers one time in
>>>> vop_initial()? Won't they get cleared when VOP is power cycled, e.g.
>>>> in case of DPMS off and on? Maybe instead this could be done in
>>>> vop_update_plane_event() for windows that need it?
>>>
>>> There are two gate for Win2/3,
>>> at VOP_WIN3_CTRL0:
>>>          bit[0], "win3_en"
>>>              this gating all the area.
>>>
>>>          bit[4], win3_mst0_en
>>>          bit[5], win3_mst1_en
>>>          bit[6], win3_mst2_en
>>>          bit[7], win3_mst3_en
>>>              those gate each area.
>>>
>>> This patch default enable win3_mst0_en, so control bit[0]"win3_en" that
>>> cat
>>> power on/off this window.
>>>
>>> vop_update_plane_event()/ vop_disable_plane() only can control
>>> bit[0]"win3_en".
>>>
>>>
>>> So this patch is enough to enable window2/3 area 0.
>>
>> That's right. However, the vop_init_reg_table[] is only used at probe
>> time by vop_initial() and register settings listed there are not
>> applied any time later. If we call DPMS off, which will turn the VOP
>> off and in turn also the whole power domain off, won't the registers
>> be reset to default values (e.g. zeroed)?
>
> Right, the vop registers would be reset to default values when power domain
> off.
>
> But the cursor can works after resume. because the initial value save to the
> regbak cache,
> and cursor area gate, win gate are at the same regs, so it can be restore
> when do cursor enable.
>
> But if we add other regs, this may cause bug, maybe no one restore them.
> So I think we need do like under to force restore all the regs when resume:
>     memcpy(vop->regs, vop->regsbak, vop->len);
>

Actually, you're right. The value will be restored on next write to
WINx_CTRL0 using the bit field accessors. So:

Reviewed-by: Tomasz Figa <tfiga@...omium.org>

Best regards,
Tomasz
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/