questions about drivers/dma/amba-pl08x.c

[Linus Walleij]

On Tue, Apr 10, 2012 at 9:30 AM, Kassey Lee <kassey1216 at gmail.com> wrote:

>             1) how to add the devices resource for this driver ? i
> did not find any info in the kernel but this driver.

There are pending patches fore RealView for this driver.
http://marc.info/?l=linux-arm-kernel&m=128568461506103&w=2

(Yes I know I should finish this patch set...)

Hm, Viresh have you posted your platform code for PL080 on
SPEAr?

>             2) in the below function, if the slave is set true,
> kernel will panic, because slave_channels is NULL, and nobody
> initialize this member, should this member be initialized by vendor ?
>
>
> static int pl08x_dma_init_virtual_channels(struct pl08x_driver_data *pl08x,
>                struct dma_device *dmadev, unsigned int channels, bool slave)
> {
>        struct pl08x_dma_chan *chan;
>        int i;
>
>        INIT_LIST_HEAD(&dmadev->channels);
>
>        /*
>         * Register as many many memcpy as we have physical channels,
>         * we won't always be able to use all but the code will have
>         * to cope with that situation.
>         */
>        for (i = 0; i < channels; i++) {
>                chan = kzalloc(sizeof(*chan), GFP_KERNEL);
>                if (!chan) {
>                        dev_err(&pl08x->adev->dev,
>                                "%s no memory for channel\n", __func__);
>                        return -ENOMEM;
>                }
>
>                chan->host = pl08x;
>                chan->state = PL08X_CHAN_IDLE;
>
>                if (slave) {
>                        chan->cd = &pl08x->pd->slave_channels[i];
>                        pl08x_dma_slave_init(chan);
>                } else {
>

Sorry, I'm not following, so slave is true and there is still no slave
channels defined? It looks like your platform data is wrong.

But better error checking is always needed so if you want to,
please send a patch to improve it...

Yours,
Linus Walleij