[PATCH 2/3 v7] mailbox: Enable BCM2835 mailbox support

[Jassi Brar]

On Tue, May 5, 2015 at 6:27 AM, Eric Anholt <eric at anholt.net> wrote:
> Jassi Brar <jassisinghbrar at gmail.com> writes:
>
>> On Wed, Apr 29, 2015 at 10:39 PM, Eric Anholt <eric at anholt.net> wrote:
>>
>>> +
>>> +struct bcm2835_mbox {
>>> +       struct device *dev;
>>> +       void __iomem *regs;
>>> +       spinlock_t lock;
>>> +       struct mbox_controller controller;
>>> +};
>>> +
>>> +static struct bcm2835_mbox *mbox;
>>> +
>>> +static irqreturn_t bcm2835_mbox_irq(int irq, void *dev_id)
>>> +{
>>> +       struct device *dev = mbox->dev;
>>> +       struct mbox_chan *link = &mbox->controller.chans[0];
>>> +
>> I learn from Stephen's other post that the controller could have
>> multiple channels. In which case this driver is poorly setup. Actually
>> if the driver was designed properly there isn't anything special to be
>> done.
>>  Here you choose to waste 'dev_id' and hard-code dereferencing to channel-0
>
> There's only the one channel according to the docs.  I wish we wouldn't
> get derailed by speculation on the list when the documentation is
> available. :(
>
Can I have the pointer to the doc please, if its publicly available.

>>> +       while (!(readl(mbox->regs + MAIL0_STA) & ARM_MS_EMPTY)) {
>>> +               u32 msg = readl(mbox->regs + MAIL0_RD);
>>> +               dev_dbg(dev, "Reply 0x%08X\n", msg);
>>> +               mbox_chan_received_data(link, &msg);
>>> +       }
>>> +       return IRQ_HANDLED;
>>> +}
>>> +
>>> +static int bcm2835_send_data(struct mbox_chan *link, void *data)
>>> +{
>>> +       int ret = 0;
>>> +       u32 msg = *(u32 *)data;
>>> +
>>> +       spin_lock(&mbox->lock);
>>> +       writel(msg, mbox->regs + MAIL1_WRT);
>>> +       dev_dbg(mbox->dev, "Request 0x%08X\n", msg);
>>> +end:
>>>
>> Did you compile check the driver for errors and warnings?
>
> At this point I'm so burned out on repainting this bikeshed that I
> missed a spot of the current color.
>
I have just rechecked the thread and I didn't find any revision that
had only nits picked. Subsystems even dictate how the files are named
and ordered in Makefile, and you have been insisting on hardcoding
values.