Re: [PATCH v2] clk: imx: imx8mp: Add delay after power up

From: Shengjiu Wang <shengjiu.wang@gmail.com>
To: Frank Li <Frank.li@nxp.com>
Cc: Shengjiu Wang <shengjiu.wang@nxp.com>,
	abelvesa@kernel.org, peng.fan@nxp.com,  mturquette@baylibre.com,
	sboyd@kernel.org, shawnguo@kernel.org,  s.hauer@pengutronix.de,
	kernel@pengutronix.de, festevam@gmail.com,  imx@lists.linux.dev,
	linux-clk@vger.kernel.org,  linux-arm-kernel@lists.infradead.org,
	linux-kernel@vger.kernel.org
Subject: Re: [PATCH v2] clk: imx: imx8mp: Add delay after power up
Date: Sat, 11 May 2024 11:09:34 +0800	[thread overview]
Message-ID: <CAA+D8ANgev0o0Z9HkmOnNuw5hACU7FgCDHFF6sbvXLd6Vd2T3w@mail.gmail.com> (raw)
In-Reply-To: <Zj5psV6ZIFZ/OPth@lizhi-Precision-Tower-5810>

On Sat, May 11, 2024 at 2:38 AM Frank Li <Frank.li@nxp.com> wrote:
>
> On Tue, May 07, 2024 at 04:04:14PM +0800, Shengjiu Wang wrote:
> > On Tue, May 7, 2024 at 12:02 PM Frank Li <Frank.li@nxp.com> wrote:
> > >
> > > On Tue, May 07, 2024 at 11:44:32AM +0800, Shengjiu Wang wrote:
> > > > On Tue, May 7, 2024 at 11:41 AM Frank Li <Frank.li@nxp.com> wrote:
> > > > >
> > > > > On Tue, May 07, 2024 at 09:44:19AM +0800, Shengjiu Wang wrote:
> > > > > > On Tue, May 7, 2024 at 2:22 AM Frank Li <Frank.li@nxp.com> wrote:
> > > > > > >
> > > > > > > On Mon, May 06, 2024 at 11:35:02AM +0800, Shengjiu Wang wrote:
> > > > > > > > According to comments in drivers/pmdomain/imx/gpcv2.c:
> > > > > > > >
> > > > > > > >       /* request the ADB400 to power up */
> > > > > > > >       if (domain->bits.hskreq) {
> > > > > > > >               regmap_update_bits(domain->regmap, domain->regs->hsk,
> > > > > > > >                                  domain->bits.hskreq, domain->bits.hskreq);
> > > > > > > >
> > > > > > > >               /*
> > > > > > > >                * ret = regmap_read_poll_timeout(domain->regmap, domain->regs->hsk, reg_val,
> > > > > > > >                *                                (reg_val & domain->bits.hskack), 0,
> > > > > > > >                *                                USEC_PER_MSEC);
> > > > > > > >                * Technically we need the commented code to wait handshake. But that needs
> > > > > > > >                * the BLK-CTL module BUS clk-en bit being set.
> > > > > > > >                *
> > > > > > > >                * There is a separate BLK-CTL module and we will have such a driver for it,
> > > > > > > >                * that driver will set the BUS clk-en bit and handshake will be triggered
> > > > > > > >                * automatically there. Just add a delay and suppose the handshake finish
> > > > > > > >                * after that.
> > > > > > > >                */
> > > > > > > >       }
> > > > > > > >
> > > > > > > > The BLK-CTL module needs to add delay to wait for a handshake request finished
> > > > > > > > before accessing registers, which is just after the enabling of the power domain.
> > > > > > > >
> > > > > > > > Otherwise there is error:
> > > > > > > >
> > > > > > > > [    2.181035] Kernel panic - not syncing: Asynchronous SError Interrupt
> > > > > > > > [    2.181038] CPU: 1 PID: 48 Comm: kworker/u16:2 Not tainted 6.9.0-rc5-next-20240424-00003-g21cec88845c6 #171
> > > > > > > > [    2.181047] Hardware name: NXP i.MX8MPlus EVK board (DT)
> > > > > > > > [    2.181050] Workqueue: events_unbound deferred_probe_work_func
> > > > > > > > [    2.181064] Call trace:
> > > > > > > > [...]
> > > > > > > > [    2.181142]  arm64_serror_panic+0x6c/0x78
> > > > > > > > [    2.181149]  do_serror+0x3c/0x70
> > > > > > > > [    2.181157]  el1h_64_error_handler+0x30/0x48
> > > > > > > > [    2.181164]  el1h_64_error+0x64/0x68
> > > > > > > > [    2.181171]  clk_imx8mp_audiomix_runtime_resume+0x34/0x44
> > > > > > > > [    2.181183]  __genpd_runtime_resume+0x30/0x80
> > > > > > > > [    2.181195]  genpd_runtime_resume+0x110/0x244
> > > > > > > > [    2.181205]  __rpm_callback+0x48/0x1d8
> > > > > > > > [    2.181213]  rpm_callback+0x68/0x74
> > > > > > > > [    2.181224]  rpm_resume+0x468/0x6c0
> > > > > > > > [    2.181234]  __pm_runtime_resume+0x50/0x94
> > > > > > > > [    2.181243]  pm_runtime_get_suppliers+0x60/0x8c
> > > > > > > > [    2.181258]  __driver_probe_device+0x48/0x12c
> > > > > > > > [    2.181268]  driver_probe_device+0xd8/0x15c
> > > > > > > > [    2.181278]  __device_attach_driver+0xb8/0x134
> > > > > > > > [    2.181290]  bus_for_each_drv+0x84/0xe0
> > > > > > > > [    2.181302]  __device_attach+0x9c/0x188
> > > > > > > > [    2.181312]  device_initial_probe+0x14/0x20
> > > > > > > > [    2.181323]  bus_probe_device+0xac/0xb0
> > > > > > > > [    2.181334]  deferred_probe_work_func+0x88/0xc0
> > > > > > > > [    2.181344]  process_one_work+0x150/0x290
> > > > > > > > [    2.181357]  worker_thread+0x2f8/0x408
> > > > > > > > [    2.181370]  kthread+0x110/0x114
> > > > > > > > [    2.181381]  ret_from_fork+0x10/0x20
> > > > > > > > [    2.181391] SMP: stopping secondary CPUs
> > > > > > > >
> > > > > > > > Fixes: 1496dd413b2e ("clk: imx: imx8mp: Add pm_runtime support for power saving")
> > > > > > > > Reported-by: Francesco Dolcini <francesco@dolcini.it>
> > > > > > > > Signed-off-by: Shengjiu Wang <shengjiu.wang@nxp.com>
> > > > > > > > Revewied-by: Peng Fan <peng.fan@nxp.com>
> > > > > > > > ---
> > > > > > > > changes in v2:
> > > > > > > > - reduce size of panic log in commit message
> > > > > > > >
> > > > > > > >  drivers/clk/imx/clk-imx8mp-audiomix.c | 7 +++++++
> > > > > > > >  1 file changed, 7 insertions(+)
> > > > > > > >
> > > > > > > > diff --git a/drivers/clk/imx/clk-imx8mp-audiomix.c b/drivers/clk/imx/clk-imx8mp-audiomix.c
> > > > > > > > index b381d6f784c8..ae2c0f254225 100644
> > > > > > > > --- a/drivers/clk/imx/clk-imx8mp-audiomix.c
> > > > > > > > +++ b/drivers/clk/imx/clk-imx8mp-audiomix.c
> > > > > > > > @@ -6,6 +6,7 @@
> > > > > > > >   */
> > > > > > > >
> > > > > > > >  #include <linux/clk-provider.h>
> > > > > > > > +#include <linux/delay.h>
> > > > > > > >  #include <linux/device.h>
> > > > > > > >  #include <linux/io.h>
> > > > > > > >  #include <linux/mod_devicetable.h>
> > > > > > > > @@ -360,6 +361,12 @@ static int clk_imx8mp_audiomix_runtime_suspend(struct device *dev)
> > > > > > > >
> > > > > > > >  static int clk_imx8mp_audiomix_runtime_resume(struct device *dev)
> > > > > > > >  {
> > > > > > > > +     /*
> > > > > > > > +      * According to the drivers/pmdomain/imx/gpcv2.c
> > > > > > > > +      * need to wait for handshake request to propagate
> > > > > > > > +      */
> > > > > > > > +     udelay(5);
> > > > > > > > +
> > > > > > >
> > > > > > > Did you address the issue I comments at v1?
> > > > > > > It should not fix at here, I think it should be gpcv2.c to delay 5us.
> > > > > >
> > > > > > Other BLK CTRL drivers already delay 5us in its own drivers, if
> > > > > > add delay in gpcv2.c, for these drivers, it will delay 10us totally.
> > > > >
> > > > > We should go forward as correct direction. If udelay should be gpcv2.c,
> > > > > it should be there and remove other udelay in BLK CTRL drivers gradually.
> > > > >
> > > > With Peng's reply:
> > > >
> > > > "No. Because BLK CTRL enable BUS_EN, before enable BUS_EN, udelay does
> > > > not help. For the audiomix, move to gpcv2 would work, but gpcv2 is
> > > > not only for i.MX8MP audiomix. For mixes, default not enable BUS_EN
> > > > after power on, the udelay must be in blk ctrl driver."
> > > >
> > > > So gpcv2.c is not correct place for all BLK CTRL drivers.
> > >
> > > where BLK-CTRL driver source code?
> >
> > drivers/pmdomain/imx/imx8m-blk-ctrl.c
> > drivers/pmdomain/imx/imx8mp-blk-ctrl.c
> > drivers/pmdomain/imx/imx93-blk-ctrl.c
>
> I still think it should put in gpcv2.c. Call power_on/off happen at very
> low frequency. Even there are additional 5us delay for other BLK-CTRL
> drivers, it will tiny impact to system performance. It is not worth to add
> additonal software check to disingiush these two cases.
>
> But correct power on is more important.
>
> So readl() follow a udelay(5) is more important then additional 5us delay
> for other BLK-CTRL driver since there are many 5us delay already in gpcv2.
>

I will send another v3 to gpcv2.c.  Let more people to review.

best regards
Shengjiu Wang
> Frank
>
> >
> > Best regards
> > Shengjiu Wang
> > >
> > > even if put clk_imx8mp_audiomix_runtime_resume(), it need read any
> > > register before udelay. all regiser read and write is strong ordered.
> > > when get value from a register, all previous write must be done.
> > >
> > > all udelay (5) in gpcv2 may not delay 5us at all.
> > >
> > > Frank
> > > >
> > > > Best regards
> > > > Shengjiu Wang
> > > >
> > > > > If sometime found 5us is not enough, need change to 6us, we just need
> > > > > change at one place.
> > > > >
> > > > > Frank
> > > > >
> > > > > >
> > > > > > Best regards
> > > > > > Shengjiu Wang
> > > > > >
> > > > > >
> > > > > >
> > > > > > >
> > > > > > > Frank
> > > > > > >
> > > > > > > >       clk_imx8mp_audiomix_save_restore(dev, false);
> > > > > > > >
> > > > > > > >       return 0;
> > > > > > > > --
> > > > > > > > 2.34.1
> > > > > > > >

_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel