[PATCH v5 3/3] dmaengine: sf-pdma: Get number of channel by device tree

[Zong Li]

On Fri, Mar 4, 2022 at 4:43 AM Palmer Dabbelt <palmer at dabbelt.com> wrote:
>
> On Tue, 15 Feb 2022 22:52:14 PST (-0800), zong.li at sifive.com wrote:
> > On Tue, Feb 15, 2022 at 8:06 PM Vinod Koul <vkoul at kernel.org> wrote:
> >>
> >> On 07-02-22, 14:30, Zong Li wrote:
> >> > It currently assumes that there are always four channels, it would
> >> > cause the error if there is actually less than four channels. Change
> >> > that by getting number of channel from device tree.
> >> >
> >> > For backwards-compatibility, it uses the default value (i.e. 4) when
> >> > there is no 'dma-channels' information in dts.
> >> >
> >> > Signed-off-by: Zong Li <zong.li at sifive.com>
> >> > ---
> >> >  drivers/dma/sf-pdma/sf-pdma.c | 21 ++++++++++++++-------
> >> >  drivers/dma/sf-pdma/sf-pdma.h |  8 ++------
> >> >  2 files changed, 16 insertions(+), 13 deletions(-)
> >> >
> >> > diff --git a/drivers/dma/sf-pdma/sf-pdma.c b/drivers/dma/sf-pdma/sf-pdma.c
> >> > index f12606aeff87..2ae10b61dfa1 100644
> >> > --- a/drivers/dma/sf-pdma/sf-pdma.c
> >> > +++ b/drivers/dma/sf-pdma/sf-pdma.c
> >> > @@ -482,9 +482,7 @@ static void sf_pdma_setup_chans(struct sf_pdma *pdma)
> >> >  static int sf_pdma_probe(struct platform_device *pdev)
> >> >  {
> >> >       struct sf_pdma *pdma;
> >> > -     struct sf_pdma_chan *chan;
> >> >       struct resource *res;
> >> > -     int len, chans;
> >> >       int ret;
> >> >       const enum dma_slave_buswidth widths =
> >> >               DMA_SLAVE_BUSWIDTH_1_BYTE | DMA_SLAVE_BUSWIDTH_2_BYTES |
> >> > @@ -492,13 +490,21 @@ static int sf_pdma_probe(struct platform_device *pdev)
> >> >               DMA_SLAVE_BUSWIDTH_16_BYTES | DMA_SLAVE_BUSWIDTH_32_BYTES |
> >> >               DMA_SLAVE_BUSWIDTH_64_BYTES;
> >> >
> >> > -     chans = PDMA_NR_CH;
> >> > -     len = sizeof(*pdma) + sizeof(*chan) * chans;
> >> > -     pdma = devm_kzalloc(&pdev->dev, len, GFP_KERNEL);
> >> > +     pdma = devm_kzalloc(&pdev->dev, sizeof(*pdma), GFP_KERNEL);
> >> >       if (!pdma)
> >> >               return -ENOMEM;
> >> >
> >> > -     pdma->n_chans = chans;
> >> > +     ret = of_property_read_u32(pdev->dev.of_node, "dma-channels",
> >> > +                                &pdma->n_chans);
> >> > +     if (ret) {
> >> > +             dev_notice(&pdev->dev, "set number of channels to default value: 4\n");
> >>
> >> This is useful for only debug i think, so dev_dbg perhaps
> >>
> >
> > Thanks for your suggestion, let me change it in the next version.
>
> Not sure if I'm missing something, but I don't see a v6.  I'm going to
> assume that one will be sent, but the suggested changes look minor
> enough so
>

I have been sending the v6 patchset, thank you for the review.

> Acked-by: Palmer Dabbelt <palmer at rivosinc.com>
>
> LMK if you guys were expecting this to go in via the RISC-V tree,
> otherwise I'll assume this aimed at the dmaengine tree.  Probably best to keep
> all three together, so feel free to take the DTS updates as well -- having some
> shared tag never hurts, but the DTs don't move that much so any conflicts
> should be straight-forward to just fix at merge time.
>

Yes, if the v6 version is good enough and you could pick them into
RISC-V tree, I'd appreciate that. Thanks!

> Thanks!
>
> >> > +             pdma->n_chans = PDMA_MAX_NR_CH;
> >> > +     }
> >> > +
> >> > +     if (pdma->n_chans > PDMA_MAX_NR_CH) {
> >> > +             dev_err(&pdev->dev, "the number of channels exceeds the maximum\n");
> >> > +             return -EINVAL;
> >> > +     }
> >> >
> >> >       res = platform_get_resource(pdev, IORESOURCE_MEM, 0);
> >> >       pdma->membase = devm_ioremap_resource(&pdev->dev, res);
> >> > @@ -556,7 +562,7 @@ static int sf_pdma_remove(struct platform_device *pdev)
> >> >       struct sf_pdma_chan *ch;
> >> >       int i;
> >> >
> >> > -     for (i = 0; i < PDMA_NR_CH; i++) {
> >> > +     for (i = 0; i < pdma->n_chans; i++) {
> >> >               ch = &pdma->chans[i];
> >> >
> >> >               devm_free_irq(&pdev->dev, ch->txirq, ch);
> >> > @@ -574,6 +580,7 @@ static int sf_pdma_remove(struct platform_device *pdev)
> >> >
> >> >  static const struct of_device_id sf_pdma_dt_ids[] = {
> >> >       { .compatible = "sifive,fu540-c000-pdma" },
> >> > +     { .compatible = "sifive,pdma0" },
> >> >       {},
> >> >  };
> >> >  MODULE_DEVICE_TABLE(of, sf_pdma_dt_ids);
> >> > diff --git a/drivers/dma/sf-pdma/sf-pdma.h b/drivers/dma/sf-pdma/sf-pdma.h
> >> > index 0c20167b097d..8127d792f639 100644
> >> > --- a/drivers/dma/sf-pdma/sf-pdma.h
> >> > +++ b/drivers/dma/sf-pdma/sf-pdma.h
> >> > @@ -22,11 +22,7 @@
> >> >  #include "../dmaengine.h"
> >> >  #include "../virt-dma.h"
> >> >
> >> > -#define PDMA_NR_CH                                   4
> >> > -
> >> > -#if (PDMA_NR_CH != 4)
> >> > -#error "Please define PDMA_NR_CH to 4"
> >> > -#endif
> >> > +#define PDMA_MAX_NR_CH                                       4
> >> >
> >> >  #define PDMA_BASE_ADDR                                       0x3000000
> >> >  #define PDMA_CHAN_OFFSET                             0x1000
> >> > @@ -118,7 +114,7 @@ struct sf_pdma {
> >> >       void __iomem            *membase;
> >> >       void __iomem            *mappedbase;
> >> >       u32                     n_chans;
> >> > -     struct sf_pdma_chan     chans[PDMA_NR_CH];
> >> > +     struct sf_pdma_chan     chans[PDMA_MAX_NR_CH];
> >>
> >> why waste memory allocating max, we know number of channels in probe,
> >> why not allocate runtime?
> >>
> >
> > I kept it there because I'd like to do minimum change in this patch
> > set. You're right, let me change it in the next version.
> >
> >> --
> >> ~Vinod