On Sat, Jul 25, 2009 at 07:17:14PM +0300, Adrian Hunter wrote: > Matt Fleming wrote: >> >> And by saying what capabilities a host supports, when we add new >> capabilities we don't have to modify existing code to say that it >> doesn't support the new capability. > > If the capabilities are the other way around, then all existing drivers > must be changed. On the other hand, the if statement can easily be > improved: > > #define mmc_cap_mmc(host) (!((host)->caps & MMC_CAP_NOT_MMC)) > #define mmc_cap_sd(host) (!((host)->caps & MMC_CAP_NOT_SD)) > #define mmc_cap_sdio(host) (!((host)->caps & MMC_CAP_NOT_SDIO)) > > - mmc_send_if_cond(host, host->ocr_avail); > + if (mmc_cap_sdio(host) || mmc_cap_sd(host)) > + mmc_send_if_cond(host, host->ocr_avail); > + OK, that's a fair point, all drivers would need to be changed. However, I'm not adverse to making a one-time change to all drivers in the name of sanity. Enumerating what cards a controller doesn't support just doesn't sound correct to me. It wouldn't be a complicated change either, I'd even be happy to write the patch ;-) -- To unsubscribe from this list: send the line "unsubscribe linux-omap" in the body of a message to majordomo@xxxxxxxxxxxxxxx More majordomo info at http://vger.kernel.org/majordomo-info.html
