Re: [External] Re: [PATCH 1/2] misc: Add clock control logic into Aspeed LPC SNOOP driver

From: John Wang
Date: Tue Dec 08 2020 - 04:07:50 EST


On Tue, Dec 8, 2020 at 10:19 AM Joel Stanley <joel@xxxxxxxxx> wrote:
>
> On Wed, 2 Dec 2020 at 05:16, John Wang <wangzhiqiang.bj@xxxxxxxxxxxxx> wrote:
> >
> > From: Jae Hyun Yoo <jae.hyun.yoo@xxxxxxxxx>
> >
> > If LPC SNOOP driver is registered ahead of lpc-ctrl module, LPC
> > SNOOP block will be enabled without heart beating of LCLK until
> > lpc-ctrl enables the LCLK. This issue causes improper handling on
> > host interrupts when the host sends interrupt in that time frame.
> > Then kernel eventually forcibly disables the interrupt with
> > dumping stack and printing a 'nobody cared this irq' message out.
> >
> > To prevent this issue, all LPC sub-nodes should enable LCLK
> > individually so this patch adds clock control logic into the LPC
> > SNOOP driver.
> >
> > Signed-off-by: Jae Hyun Yoo <jae.hyun.yoo@xxxxxxxxx>
> > Signed-off-by: Vernon Mauery <vernon.mauery@xxxxxxxxxxxxxxx>
> > Signed-off-by: John Wang <wangzhiqiang.bj@xxxxxxxxxxxxx>
>
> Reviewed-by: Joel Stanley <joel@xxxxxxxxx>
>
> Thanks for sending these John. It is an excellent idea to upstream
> fixes that have been developed.
>
> I assume we will have the same issue for all devices that use the LPC
> bus? eg. vuart, bt, kcs, lpc2ahb? It looks like only the lpc-ctrl
> (lpc2ahb) does this so far:

ok that's on my todo list. :)

>
> git grep -l clk drivers/soc/aspeed/aspeed-p2a-ctrl.c
> drivers/soc/aspeed/aspeed-lpc-ctrl.c
> drivers/char/ipmi/kcs_bmc_aspeed.c drivers/char/ipmi/bt-bmc.c
> drivers/soc/aspeed/aspeed-lpc-ctrl.c
>
>
>
>
> > ---
> > drivers/soc/aspeed/aspeed-lpc-snoop.c | 30 ++++++++++++++++++++++++---
> > 1 file changed, 27 insertions(+), 3 deletions(-)
> >
> > diff --git a/drivers/soc/aspeed/aspeed-lpc-snoop.c b/drivers/soc/aspeed/aspeed-lpc-snoop.c
> > index 682ba0eb4eba..20acac6342ef 100644
> > --- a/drivers/soc/aspeed/aspeed-lpc-snoop.c
> > +++ b/drivers/soc/aspeed/aspeed-lpc-snoop.c
> > @@ -11,6 +11,7 @@
> > */
> >
> > #include <linux/bitops.h>
> > +#include <linux/clk.h>
> > #include <linux/interrupt.h>
> > #include <linux/fs.h>
> > #include <linux/kfifo.h>
> > @@ -67,6 +68,7 @@ struct aspeed_lpc_snoop_channel {
> > struct aspeed_lpc_snoop {
> > struct regmap *regmap;
> > int irq;
> > + struct clk *clk;
> > struct aspeed_lpc_snoop_channel chan[NUM_SNOOP_CHANNELS];
> > };
> >
> > @@ -282,22 +284,42 @@ static int aspeed_lpc_snoop_probe(struct platform_device *pdev)
> > return -ENODEV;
> > }
> >
> > + lpc_snoop->clk = devm_clk_get(dev, NULL);
> > + if (IS_ERR(lpc_snoop->clk)) {
> > + rc = PTR_ERR(lpc_snoop->clk);
> > + if (rc != -EPROBE_DEFER)
> > + dev_err(dev, "couldn't get clock\n");
> > + return rc;
> > + }
> > + rc = clk_prepare_enable(lpc_snoop->clk);
> > + if (rc) {
> > + dev_err(dev, "couldn't enable clock\n");
> > + return rc;
> > + }
> > +
> > rc = aspeed_lpc_snoop_config_irq(lpc_snoop, pdev);
> > if (rc)
> > - return rc;
> > + goto err;
> >
> > rc = aspeed_lpc_enable_snoop(lpc_snoop, dev, 0, port);
> > if (rc)
> > - return rc;
> > + goto err;
> >
> > /* Configuration of 2nd snoop channel port is optional */
> > if (of_property_read_u32_index(dev->of_node, "snoop-ports",
> > 1, &port) == 0) {
> > rc = aspeed_lpc_enable_snoop(lpc_snoop, dev, 1, port);
> > - if (rc)
> > + if (rc) {
> > aspeed_lpc_disable_snoop(lpc_snoop, 0);
> > + goto err;
> > + }
> > }
> >
> > + return 0;
> > +
> > +err:
> > + clk_disable_unprepare(lpc_snoop->clk);
> > +
> > return rc;
> > }
> >
> > @@ -309,6 +331,8 @@ static int aspeed_lpc_snoop_remove(struct platform_device *pdev)
> > aspeed_lpc_disable_snoop(lpc_snoop, 0);
> > aspeed_lpc_disable_snoop(lpc_snoop, 1);
> >
> > + clk_disable_unprepare(lpc_snoop->clk);
> > +
> > return 0;
> > }
> >
> > --
> > 2.25.1
> >