linux-kernel - Re: [PATCH v1 v1 3/3] i2c-ocores: sifive: add polling mode workaround for FU540-C000 SoC.

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] [day] [month] [year] [list]

Message-ID: <CAARK3H=9frKMTB6aWBwEmCxXxQuZgjAij_Uam+U8of48hjq=bA@mail.gmail.com>
Date:   Mon, 6 May 2019 23:24:28 +0530
From:   Sagar Kadam <sagar.kadam@...ive.com>
To:     Andrew Lunn <andrew@...n.ch>
Cc:     robh+dt@...nel.org, mark.rutland@....com,
        Palmer Dabbelt <palmer@...ive.com>,
        Paul Walmsley <paul.walmsley@...ive.com>, peter@...sgaard.com,
        linux-i2c@...r.kernel.org, devicetree@...r.kernel.org,
        linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject: Re: [PATCH v1 v1 3/3] i2c-ocores: sifive: add polling mode workaround
 for FU540-C000 SoC.

On Mon, May 6, 2019 at 6:59 PM Andrew Lunn <andrew@...n.ch> wrote:
>
> >  /*
> >   * 'process_lock' exists because ocores_process() and ocores_process_timeout()
> > @@ -239,8 +240,13 @@ static irqreturn_t ocores_isr(int irq, void *dev_id)
> >       struct ocores_i2c *i2c = dev_id;
> >       u8 stat = oc_getreg(i2c, OCI2C_STATUS);
> >
> > -     if (!(stat & OCI2C_STAT_IF))
> > +     if (i2c->flags && SIFIVE_FLAG_POLL) {
>
> Do you really want && here?
>
> > +             if (stat & OCI2C_STAT_IF)
> > +                     if (!(stat & OCI2C_STAT_BUSY))
> > +                             return IRQ_NONE;
> > +     } else if (!(stat & OCI2C_STAT_IF)) {
> >               return IRQ_NONE;
> > +     }
> >
> >       ocores_process(i2c, stat);
> >
> > @@ -356,6 +362,11 @@ static void ocores_process_polling(struct ocores_i2c *i2c)
> >               ret = ocores_isr(-1, i2c);
> >               if (ret == IRQ_NONE)
> >                       break; /* all messages have been transferred */
> > +             else {
> > +                     if (i2c->flags && SIFIVE_FLAG_POLL)
>
> And here?
>
> > +                             if (i2c->state == STATE_DONE)
> > +                                     break;
> > +             }
> >       }
> >  }
> >
> > @@ -406,7 +417,7 @@ static int ocores_xfer(struct i2c_adapter *adap,
> >  {
> >       struct ocores_i2c *i2c = i2c_get_adapdata(adap);
> >
> > -     if (i2c->flags & OCORES_FLAG_POLL)
> > +     if ((i2c->flags & OCORES_FLAG_POLL) || (i2c->flags & SIFIVE_FLAG_POLL))
>
> You can combine this

Thanks for your suggestion's Andrew.
Yes, I will optimize this.
>
> if ((i2c->flags & (OCORES_FLAG_POLL | SIFIVE_FLAG_POLL))
>
> >               return ocores_xfer_polling(adap, msgs, num);
> >       return ocores_xfer_core(i2c, msgs, num, false);
> >  }
> > @@ -597,6 +608,7 @@ static int ocores_i2c_probe(struct platform_device *pdev)
> >  {
> >       struct ocores_i2c *i2c;
> >       struct ocores_i2c_platform_data *pdata;
> > +     const struct of_device_id *match;
> >       struct resource *res;
> >       int irq;
> >       int ret;
> > @@ -678,13 +690,21 @@ static int ocores_i2c_probe(struct platform_device *pdev)
> >
> >       irq = platform_get_irq(pdev, 0);
> >       if (irq == -ENXIO) {
> > -             i2c->flags |= OCORES_FLAG_POLL;
> > +             /*
> > +              * Set a SIFIVE_FLAG_POLL to enable workaround for FU540
> > +              * in polling mode interface of i2c-ocore driver.
> > +              */
> > +             match = of_match_node(ocores_i2c_match, pdev->dev.of_node);
> > +             if (match && (long)match->data == TYPE_SIFIVE_REV0)
> > +                     i2c->flags |= SIFIVE_FLAG_POLL;
> > +             else
> > +                     i2c->flags |= OCORES_FLAG_POLL;
>
> Please take a look at the whole code, and consider if it is better to
> set both SIFIVE_FLAG_POLL and OCORES_FLAG_POLL. Maybe rename
> SIFIVE_FLAG_POLL to OCORES_FLAG_BROKEN_IRQ_BIT?
>
The intent of this patch is to add a workaround for hardware errratum
of FU540 a SiFive Device,
hence I had named the flag accordingly. Yes,
OCORES_FLAG_BROKEN_IRQ_BIT is a better and generic term,
I will rename and resubmit this patch

-Thanks
Sagar

> Thanks
>         Andrew