| 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
| ||
|
Date: Mon, 20 May 2019 13:30:27 +0800
From: CK Hu <ck.hu@...iatek.com>
To: Jitao Shi <jitao.shi@...iatek.com>
CC: Rob Herring <robh+dt@...nel.org>, Pawel Moll <pawel.moll@....com>,
"Mark Rutland" <mark.rutland@....com>,
Ian Campbell <ijc+devicetree@...lion.org.uk>,
Kumar Gala <galak@...eaurora.org>, <linux-pwm@...r.kernel.org>,
David Airlie <airlied@...ux.ie>,
"Matthias Brugger" <matthias.bgg@...il.com>,
Thierry Reding <treding@...dia.com>,
"Ajay Kumar" <ajaykumar.rs@...sung.com>,
Inki Dae <inki.dae@...sung.com>,
"Rahul Sharma" <rahul.sharma@...sung.com>,
Sean Paul <seanpaul@...omium.org>,
Vincent Palatin <vpalatin@...omium.org>,
Andy Yan <andy.yan@...k-chips.com>,
Philipp Zabel <p.zabel@...gutronix.de>,
"Russell King" <rmk+kernel@....linux.org.uk>,
<devicetree@...r.kernel.org>, <linux-kernel@...r.kernel.org>,
<dri-devel@...ts.freedesktop.org>,
<linux-arm-kernel@...ts.infradead.org>,
<linux-mediatek@...ts.infradead.org>,
<srv_heupstream@...iatek.com>,
Sascha Hauer <kernel@...gutronix.de>,
<yingjoe.chen@...iatek.com>, <eddie.huang@...iatek.com>,
<cawa.cheng@...iatek.com>, <bibby.hsieh@...iatek.com>,
<stonea168@....com>
Subject: Re: [v2 3/3] drm/mediatek: add mipi_tx driver for mt8183
On Sat, 2019-05-18 at 15:51 +0800, Jitao Shi wrote:
> On Mon, 2019-05-06 at 17:17 +0800, CK Hu wrote:
> > Hi, Jitao:
> >
> > On Tue, 2019-04-16 at 13:42 +0800, Jitao Shi wrote:
> > > This patch add mt8183 mipi_tx driver.
> > > And also support other chips that use the same binding and driver.
> > >
> > > Signed-off-by: Jitao Shi <jitao.shi@...iatek.com>
> > > ---
> > > drivers/gpu/drm/mediatek/Makefile | 1 +
> > > drivers/gpu/drm/mediatek/mtk_mipi_tx.c | 2 +
> > > drivers/gpu/drm/mediatek/mtk_mipi_tx.h | 1 +
> > > drivers/gpu/drm/mediatek/mtk_mt8183_mipi_tx.c | 154 ++++++++++++++++++
> > > 4 files changed, 158 insertions(+)
> > > create mode 100644 drivers/gpu/drm/mediatek/mtk_mt8183_mipi_tx.c
> > >
> >
> > [snip]
> >
> > > +
> > > +static int mtk_mipi_tx_pll_prepare(struct clk_hw *hw)
> > > +{
> > > + struct mtk_mipi_tx *mipi_tx = mtk_mipi_tx_from_clk_hw(hw);
> > > + unsigned int txdiv, txdiv0;
> > > + u64 pcw;
> > > + int ret;
> > > +
> > > + dev_dbg(mipi_tx->dev, "prepare: %u bps\n", mipi_tx->data_rate);
> > > +
> > > + if (mipi_tx->data_rate >= 2000000000) {
> > > + txdiv = 1;
> > > + txdiv0 = 0;
> > > + } else if (mipi_tx->data_rate >= 1000000000) {
> > > + txdiv = 2;
> > > + txdiv0 = 1;
> > > + } else if (mipi_tx->data_rate >= 500000000) {
> > > + txdiv = 4;
> > > + txdiv0 = 2;
> > > + } else if (mipi_tx->data_rate > 250000000) {
> > > + txdiv = 8;
> > > + txdiv0 = 3;
> > > + } else if (mipi_tx->data_rate >= 125000000) {
> > > + txdiv = 16;
> > > + txdiv0 = 4;
> > > + } else {
> > > + return -EINVAL;
> > > + }
> > > +
> > > + ret = clk_prepare_enable(mipi_tx->ref_clk);
> > > + if (ret < 0) {
> > > + dev_err(mipi_tx->dev,
> > > + "can't prepare and enable mipi_tx ref_clk %d\n", ret);
> > > + return ret;
> > > + }
> >
> > You enable the parent clock when prepare this clock here, this behavior
> > looks strange. I think the flow should be:
> >
> > 1. Parent clock prepare
> > 2. This clock prepare
> > 3. Parent clock enable
> > 4. This clock enable
> >
> > Maybe you should implement 'enable callback' so that parent clock would
> > be already enabled.
> >
> > One question is, mipi_tx_pll is used by dsi driver, but I does not see
> > dsi prepare_enable() mipi_tx_pll, how does this work?
> >
> > Regards,
> > CK
> >
>
> The mipi_tx can be accessed after clk_prepare_enable(mipi_tx->ref_clk);
>
> So place the clk_prepare_enable(mipi_tx->ref_clk) before accessing
> mipitx.
>
> mipi_tx_pll is enable by mtk_mipi_tx_power_on() in mtk_mip_tx.c.
> clk_prepare_enable(mipi_tx->pll) will enable mipi_tx_pll.
OK, so it start from dsi driver. The callstack is:
phy_power_on(dsi->phy);
-> mtk_mipi_tx_power_on()
--> clk_prepare_enable(mipi_tx->pll);
---> mtk_mipi_tx_pll_prepare();
In clk_prepare_enable(), it separately call clk_prepare() and
clk_enable(). When clk_prepare(), it prepare the parent clock then
prepare this clock. When clk_enable(), it enable the parent clock then
enable this clock. So this would result in the sequence:
1. Prepare mipi_tx->ref_clk
2. Prepare mipi_tx->pll
3. Enable mipi_tx->ref_clk
4. Enable mipi_tx->pll
You say 'So place the clk_prepare_enable(mipi_tx->ref_clk) before
accessing mipitx.', so the step 1 and step 3 is equal to
clk_prepare_enable(mipi_tx->ref_clk), so I require you to access mipitx
in step 4, not in step 2.
Regards,
CK
>
> Beset Regards
> Jitao
>
> > > +
> > > + mtk_mipi_tx_clear_bits(mipi_tx, MIPITX_PLL_CON4, RG_DSI_PLL_IBIAS);
> > > +
> > > + mtk_mipi_tx_set_bits(mipi_tx, MIPITX_PLL_PWR, AD_DSI_PLL_SDM_PWR_ON);
> > > + usleep_range(30, 100);
> > > + mtk_mipi_tx_clear_bits(mipi_tx, MIPITX_PLL_PWR, AD_DSI_PLL_SDM_ISO_EN);
> > > + pcw = div_u64(((u64)mipi_tx->data_rate * txdiv) << 24, 26000000);
> > > + writel(pcw, mipi_tx->regs + MIPITX_PLL_CON0);
> > > + mtk_mipi_tx_update_bits(mipi_tx, MIPITX_PLL_CON1, RG_DSI_PLL_POSDIV,
> > > + txdiv0 << 8);
> > > + usleep_range(1000, 2000);
> > > + mtk_mipi_tx_set_bits(mipi_tx, MIPITX_PLL_CON1, RG_DSI_PLL_EN);
> > > +
> > > + return 0;
> > > +}
> > > +
> > > +static void mtk_mipi_tx_pll_unprepare(struct clk_hw *hw)
> > > +{
> > > + struct mtk_mipi_tx *mipi_tx = mtk_mipi_tx_from_clk_hw(hw);
> > > +
> > > + dev_dbg(mipi_tx->dev, "unprepare\n");
> > > +
> > > + mtk_mipi_tx_clear_bits(mipi_tx, MIPITX_PLL_CON1, RG_DSI_PLL_EN);
> > > +
> > > + mtk_mipi_tx_set_bits(mipi_tx, MIPITX_PLL_PWR, AD_DSI_PLL_SDM_ISO_EN);
> > > + mtk_mipi_tx_clear_bits(mipi_tx, MIPITX_PLL_PWR, AD_DSI_PLL_SDM_PWR_ON);
> > > + clk_disable_unprepare(mipi_tx->ref_clk);
> > > +}
> > > +
> >
> >
>
>
Powered by blists - more mailing lists