The core sends this command anyhow shortly after calling .init(). So doing it in bcm2835_mci_reset() is only overhead if it succeeds and if there is a problem it is noticed a tad earlier which has little benefit compared to the timeout in bcm2835_mci_wait_command_done().
So simplify and drop the transfer. Signed-off-by: Uwe Kleine-König <[email protected]> --- drivers/mci/mci-bcm2835.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-) diff --git a/drivers/mci/mci-bcm2835.c b/drivers/mci/mci-bcm2835.c index 4913dc27f8a3..10f588930cda 100644 --- a/drivers/mci/mci-bcm2835.c +++ b/drivers/mci/mci-bcm2835.c @@ -352,10 +352,7 @@ static int bcm2835_mci_reset(struct mci_host *mci, struct device_d *mci_dev) sdhci_write32(&host->sdhci, SDHCI_INT_STATUS, 0xFFFFFFFF); - /*Now write command 0 and see if we get response*/ - sdhci_write32(&host->sdhci, SDHCI_ARGUMENT, 0x0); - sdhci_write32(&host->sdhci, SDHCI_TRANSFER_MODE__COMMAND, 0x0); - return bcm2835_mci_wait_command_done(host); + return 0; } static int bcm2835_mci_probe(struct device_d *hw_dev) base-commit: f2dd8896846757829fb3033f41b89be3fc037d78 -- 2.30.2 _______________________________________________ barebox mailing list [email protected] http://lists.infradead.org/mailman/listinfo/barebox
