4.9-stable review patch. If anyone has any objections, please let me know.
------------------
From: Adrian Hunter <adrian.hunter@xxxxxxxxx>
commit 027fb89e61054b4aedd962adb3e2003dec78a716 upstream.
Disabling interrupts for even a millisecond can cause problems for some
devices. That can happen when Intel host controllers wait for the present
state to propagate.
The spin lock is not necessary here. Anything that is racing with changes
to the I/O state is already broken. The mmc core already provides
synchronization via "claiming" the host.
Although the spin lock probably should be removed from the code paths that
lead to this point, such a patch would touch too much code to be suitable
for stable trees. Consequently, for this patch, just drop the spin lock
while waiting.
Signed-off-by: Adrian Hunter <adrian.hunter@xxxxxxxxx>
Signed-off-by: Ulf Hansson <ulf.hansson@xxxxxxxxxx>
Tested-by: Ludovic Desroches <ludovic.desroches@xxxxxxxxxxxxx>
Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx>
---
drivers/mmc/host/sdhci-pci-core.c | 4 ++++
1 file changed, 4 insertions(+)
--- a/drivers/mmc/host/sdhci-pci-core.c
+++ b/drivers/mmc/host/sdhci-pci-core.c
@@ -412,6 +412,8 @@ static void sdhci_intel_set_power(struct
if (mode == MMC_POWER_OFF)
return;
+ spin_unlock_irq(&host->lock);
+
/*
* Bus power might not enable after D3 -> D0 transition due to the
* present state not yet having propagated. Retry for up to 2ms.
@@ -424,6 +426,8 @@ static void sdhci_intel_set_power(struct
reg |= SDHCI_POWER_ON;
sdhci_writeb(host, reg, SDHCI_POWER_CONTROL);
}
+
+ spin_lock_irq(&host->lock);
}
static const struct sdhci_ops sdhci_intel_byt_ops = {