From: Pierre Ossman Date: Fri, 30 Jun 2006 09:22:28 +0000 (-0700) Subject: [MMC] sdhci: check only relevant inhibit bits X-Git-Tag: v2.6.18-rc1~146^2~5 X-Git-Url: https://err.no/cgi-bin/gitweb.cgi?a=commitdiff_plain;h=fd2208d7c72ef5995b730f1e23b082261499e334;p=linux-2.6 [MMC] sdhci: check only relevant inhibit bits Conform to the sdhci specification as to which inhibit bits should be checked at different times. Signed-off-by: Pierre Ossman Signed-off-by: Andrew Morton Signed-off-by: Russell King --- diff --git a/drivers/mmc/sdhci.c b/drivers/mmc/sdhci.c index 302dd5bde7..5324eae6d7 100644 --- a/drivers/mmc/sdhci.c +++ b/drivers/mmc/sdhci.c @@ -465,6 +465,7 @@ static void sdhci_finish_data(struct sdhci_host *host) static void sdhci_send_command(struct sdhci_host *host, struct mmc_command *cmd) { int flags; + u32 mask; unsigned long timeout; WARN_ON(host->cmd); @@ -473,11 +474,20 @@ static void sdhci_send_command(struct sdhci_host *host, struct mmc_command *cmd) /* Wait max 10 ms */ timeout = 10; - while (readl(host->ioaddr + SDHCI_PRESENT_STATE) & - (SDHCI_CMD_INHIBIT | SDHCI_DATA_INHIBIT)) { + + mask = SDHCI_CMD_INHIBIT; + if ((cmd->data != NULL) || (cmd->flags & MMC_RSP_BUSY)) + mask |= SDHCI_DATA_INHIBIT; + + /* We shouldn't wait for data inihibit for stop commands, even + though they might use busy signaling */ + if (host->mrq->data && (cmd == host->mrq->data->stop)) + mask &= ~SDHCI_DATA_INHIBIT; + + while (readl(host->ioaddr + SDHCI_PRESENT_STATE) & mask) { if (timeout == 0) { printk(KERN_ERR "%s: Controller never released " - "inhibit bits. Please report this to " + "inhibit bit(s). Please report this to " BUGMAIL ".\n", mmc_hostname(host->mmc)); sdhci_dumpregs(host); cmd->error = MMC_ERR_FAILED;