diff options
| author | Nicolas Pitre <nico@cam.org> | 2007-10-03 15:32:57 -0400 |
|---|---|---|
| committer | Pierre Ossman <drzeus@drzeus.cx> | 2007-10-06 17:51:45 +0200 |
| commit | 599473cf15a3fae78cbc3192cfb38ca04d5abc72 (patch) | |
| tree | cba3a1c236711d4fc267ef6272c93d9dcf6a08e8 /drivers/mmc/core | |
| parent | 3e01e4bcdd56209e70c39293e0c4c355d09364b8 (diff) | |
sdio: make the IRQ thread more resilient in the presence of bad states
Currently we print a message about some bad states wrt function IRQ
handlers but return 0 from process_sdio_pending_irqs() nevertheless.
This can lead to an infinite loop as nothing might have cleared the
condition for the pending card interrupt from the host controller by
the time host->ops->enable_sdio_irq(host, 1) is called.
Signed-off-by: Nicolas Pitre <nico@marvell.com>
Signed-off-by: Pierre Ossman <drzeus@drzeus.cx>
Diffstat (limited to 'drivers/mmc/core')
| -rw-r--r-- | drivers/mmc/core/sdio_irq.c | 10 |
1 files changed, 8 insertions, 2 deletions
diff --git a/drivers/mmc/core/sdio_irq.c b/drivers/mmc/core/sdio_irq.c index e7865059e12..3bd3021f5e8 100644 --- a/drivers/mmc/core/sdio_irq.c +++ b/drivers/mmc/core/sdio_irq.c @@ -45,16 +45,22 @@ static int process_sdio_pending_irqs(struct mmc_card *card) printk(KERN_WARNING "%s: pending IRQ for " "non-existant function\n", mmc_card_id(card)); + ret = -EINVAL; } else if (func->irq_handler) { func->irq_handler(func); count++; - } else + } else { printk(KERN_WARNING "%s: pending IRQ with no handler\n", sdio_func_id(func)); + ret = -EINVAL; + } } } - return count; + if (count) + return count; + + return ret; } static int sdio_irq_thread(void *_host) |