Hi Geert,
Thank you for the review.
On Mon, Jul 26, 2021 at 10:53 AM Geert Uytterhoeven
<geert@xxxxxxxxxxxxxx> wrote:
>
> Hi Prabhakar,
>
> On Wed, Jul 21, 2021 at 9:50 PM Lad Prabhakar
> <prabhakar.mahadev-lad.rj@xxxxxxxxxxxxxx> wrote:
> > CANFD block on RZ/G2L SoC is almost identical to one found on
> > R-Car Gen3 SoC's. On RZ/G2L SoC interrupt sources for each channel
> > are split into different sources and the IP doesn't divide (1/2)
> > CANFD clock within the IP.
> >
> > This patch adds compatible string for RZ/G2L family and registers
> > the irq handlers required for CANFD operation. IRQ numbers are now
> > fetched based on names instead of indices. For backward compatibility
> > on non RZ/G2L SoC's we fallback reading based on indices.
> >
> > Signed-off-by: Lad Prabhakar <prabhakar.mahadev-lad.rj@xxxxxxxxxxxxxx>
>
> Thanks for the update!
>
> I think you misunderstood my comment on v1 about the interrupt
> handlers, cfr. below.
>
Argh my bad I took it the other way round!
> > --- a/drivers/net/can/rcar/rcar_canfd.c
> > +++ b/drivers/net/can/rcar/rcar_canfd.c
>
> > @@ -1577,6 +1586,53 @@ static int rcar_canfd_channel_probe(struct rcar_canfd_global *gpriv, u32 ch,
> > priv->can.clock.freq = fcan_freq;
> > dev_info(&pdev->dev, "can_clk rate is %u\n", priv->can.clock.freq);
> >
> > + if (gpriv->chip_id == RENESAS_RZG2L) {
> > + char *irq_name;
> > + int err_irq;
> > + int tx_irq;
> > +
> > + err_irq = platform_get_irq_byname(pdev, ch == 0 ? "ch0_err" : "ch1_err");
> > + if (err_irq < 0) {
> > + err = err_irq;
> > + goto fail;
> > + }
> > +
> > + tx_irq = platform_get_irq_byname(pdev, ch == 0 ? "ch0_trx" : "ch1_trx");
> > + if (tx_irq < 0) {
> > + err = tx_irq;
> > + goto fail;
> > + }
> > +
> > + irq_name = devm_kasprintf(&pdev->dev, GFP_KERNEL,
> > + "canfd.ch%d_err", ch);
> > + if (!irq_name) {
> > + err = -ENOMEM;
> > + goto fail;
> > + }
> > + err = devm_request_irq(&pdev->dev, err_irq,
> > + rcar_canfd_channel_interrupt, 0,
>
> This is the same interrupt handler...
>
> > + irq_name, gpriv);
> > + if (err) {
> > + dev_err(&pdev->dev, "devm_request_irq CH Err(%d) failed, error %d\n",
> > + err_irq, err);
> > + goto fail;
> > + }
> > + irq_name = devm_kasprintf(&pdev->dev, GFP_KERNEL,
> > + "canfd.ch%d_trx", ch);
> > + if (!irq_name) {
> > + err = -ENOMEM;
> > + goto fail;
> > + }
> > + err = devm_request_irq(&pdev->dev, tx_irq,
> > + rcar_canfd_channel_interrupt, 0,
>
> ... as this one.
>
> > + irq_name, gpriv);
> > + if (err) {
> > + dev_err(&pdev->dev, "devm_request_irq Tx (%d) failed, error %d\n",
> > + tx_irq, err);
> > + goto fail;
> > + }
> > + }
> > +
> > if (gpriv->fdmode) {
> > priv->can.bittiming_const = &rcar_canfd_nom_bittiming_const;
> > priv->can.data_bittiming_const =
>
> > @@ -1711,20 +1798,51 @@ static int rcar_canfd_probe(struct platform_device *pdev)
> > gpriv->base = addr;
> >
> > /* Request IRQ that's common for both channels */
> > - err = devm_request_irq(&pdev->dev, ch_irq,
> > - rcar_canfd_channel_interrupt, 0,
> > - "canfd.chn", gpriv);
> > - if (err) {
> > - dev_err(&pdev->dev, "devm_request_irq(%d) failed, error %d\n",
> > - ch_irq, err);
> > - goto fail_dev;
> > + if (gpriv->chip_id == RENESAS_RCAR_GEN3) {
> > + err = devm_request_irq(&pdev->dev, ch_irq,
> > + rcar_canfd_channel_interrupt, 0,
> > + "canfd.ch_int", gpriv);
> > + if (err) {
> > + dev_err(&pdev->dev, "devm_request_irq(%d) failed, error %d\n",
> > + ch_irq, err);
> > + goto fail_dev;
> > + }
> > +
> > + err = devm_request_irq(&pdev->dev, g_irq,
> > + rcar_canfd_global_interrupt, 0,
> > + "canfd.g_int", gpriv);
> > + if (err) {
> > + dev_err(&pdev->dev, "devm_request_irq(%d) failed, error %d\n",
> > + g_irq, err);
> > + goto fail_dev;
> > + }
> > + } else {
> > + err = devm_request_irq(&pdev->dev, g_recc_irq,
> > + rcar_canfd_global_interrupt, 0,
>
> This is the same interrupt handler...
>
> > + "canfd.g_recc", gpriv);
> > +
> > + if (err) {
> > + dev_err(&pdev->dev, "devm_request_irq(%d) failed, error %d\n",
> > + g_recc_irq, err);
> > + goto fail_dev;
> > + }
> > +
> > + err = devm_request_irq(&pdev->dev, g_err_irq,
> > + rcar_canfd_global_interrupt, 0,
>
> ... as this one.
>
> > + "canfd.g_err", gpriv);
> > + if (err) {
> > + dev_err(&pdev->dev, "devm_request_irq(%d) failed, error %d\n",
> > + g_err_irq, err);
> > + goto fail_dev;
> > + }
> > }
> > - err = devm_request_irq(&pdev->dev, g_irq,
> > - rcar_canfd_global_interrupt, 0,
> > - "canfd.gbl", gpriv);
> > +
> > + err = reset_control_reset(gpriv->rstc1);
> > + if (err)
> > + goto fail_dev;
> > + err = reset_control_reset(gpriv->rstc2);
> > if (err) {
> > - dev_err(&pdev->dev, "devm_request_irq(%d) failed, error %d\n",
> > - g_irq, err);
> > + reset_control_assert(gpriv->rstc1);
> > goto fail_dev;
> > }
>
> I did not object to having fine-grained interrupt handlers on RZ/G2L.
> I did object to duplicating code in global and fine-grained interrupt
> handlers.
>
> The trick to have both is to let the global interrupt handlers call
> (conditionally) into the fine-grained handlers. In pseudo-code:
>
> global_interrupt_handler()
> {
> if (...)
> fine_grained_handler1();
>
> if (...)
> fine_grained_handler2();
> ...
> }
>
> On R-Car Gen3, you register the global interrupt handlers, as before.
> On RZ/G2L, you register the fine-grained interrupt handlers instead.
>
Agreed will re-spin with the fine-grained version tomorrow.
Cheers,
Prabhakar
> Gr{oetje,eeting}s,
>
> Geert
>
> --
> Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@xxxxxxxxxxxxxx
>
> In personal conversations with technical people, I call myself a hacker. But
> when I'm talking to journalists I just say "programmer" or something like that.
> -- Linus Torvalds
