On Mon, Jul 06, 2020 at 12:35:30AM +0200, Ahmad Fatoum wrote:
>
>
> On 7/5/20 8:42 PM, Sascha Hauer wrote:
> > On Wed, Jul 01, 2020 at 11:11:13AM +0200, Ahmad Fatoum wrote:
> >> +static void at91_fat_start_image(struct pbl_bio *bio,
> >> + void *buf, unsigned int len,
> >> + u32 r4)
> >> +{
> >> + void __noreturn (*bb)(void);
> >> + int ret;
> >> +
> >> + ret = pbl_fat_load(bio, "barebox.bin", buf, len);
> >> + if (ret < 0) {
> >> + pr_err("pbl_fat_load: error %d\n", ret);
> >> + return;
> >> + }
> >> +
> >> + bb = buf;
> >> +
> >> + sync_caches_for_execution();
> >> +
> >> + xload_bb(bb, r4);
> >> +}
> >> +
> >> +static const struct sdhci_instance {
> >> + void __iomem *base;
> >> + unsigned id;
> >> + u8 periph;
> >> + s8 pins[15];
> >> +} sdhci_instances[] = {
> >> + [0] = { SAMA5D2_BASE_SDHC0, SAMA5D2_ID_SDMMC0, AT91_MUX_PERIPH_A,
> >> + { 2, 3, 4, 5, 6, 7, 8, 9, 0, 1, 13, 10, 11, 12, -1 } },
> >> + [1] = { SAMA5D2_BASE_SDHC1, SAMA5D2_ID_SDMMC1, AT91_MUX_PERIPH_E,
> >> + { 18, 19, 20, 21, 22, 28, 30, -1 } },
> >> +};
> >
> > Can we use C99 initializer here please?
>
> Can do
>
> >
> >> +
> >> +/**
> >> + * sama5d2_sdhci_start_image - Load and start an image from FAT-formatted SDHCI
> >> + * @r4: value of r4 passed by BootROM
> >> + *
> >> + * Return: If successul, this function does not return. A negative error
> >
> > s/successul/successful/
> >
> >> + * code is returned when this function fails.
> >
> > This is not true, the function panics on error.
>
> It did when I first wrote the comment :D.
> Will reply with a fixup.
>
> >
> >> + */
> >> +void __noreturn sama5d2_sdhci_start_image(u32 r4)
> >> +{
> >> + void *buf = (void *)SAMA5_DDRCS;
> >> + const struct sdhci_instance *instance;
> >> + struct pbl_bio bio;
> >> + const s8 *pin;
> >> + int ret;
> >> +
> >> + instance = &sdhci_instances[!!sama5_bootsource_instance(r4)];
> >
> > Why "!!"? sama5_bootsource_instance() should return exactly 0 or 1 here.
>
> SAMA5_BOOTSOURCE_INSTANCE is GENMASK(7, 4). While it should only be 0 or 1
> on the sama5d2, better be on the safe side and force it to either 0 or 1.
"safe" would be to only accept 0 and 1 and refuse other values.
Interpreting everything != 0 as 1 is IMO not safe.
Sascha
--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
_______________________________________________
barebox mailing list
barebox@xxxxxxxxxxxxxxxxxxx
http://lists.infradead.org/mailman/listinfo/barebox
