| Message ID | 20161025205718.1637880-1-arnd@arndb.de (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
diff --git a/drivers/spi/spi-fsl-espi.c b/drivers/spi/spi-fsl-espi.c index 7451585..2c175b9 100644 --- a/drivers/spi/spi-fsl-espi.c +++ b/drivers/spi/spi-fsl-espi.c @@ -458,7 +458,7 @@ static void fsl_espi_cpu_irq(struct mpc8xxx_spi *mspi, u32 events) mspi->len -= rx_nr_bytes; - if (mspi->rx) + if (rx_nr_bytes && mspi->rx) mspi->get_rx(rx_data, mspi); }
When we get a spurious interrupt in fsl_espi_irq, we end up processing four uninitalized bytes of data, as shown in this warning message: drivers/spi/spi-fsl-espi.c: In function 'fsl_espi_irq': drivers/spi/spi-fsl-espi.c:462:4: warning: 'rx_data' may be used uninitialized in this function [-Wmaybe-uninitialized] This adds another check so we skip the data in this case. Fixes: 6319a68011b8 ("spi/fsl-espi: avoid infinite loops on fsl_espi_cpu_irq()") Cc: stable@vger.kernel.org Signed-off-by: Arnd Bergmann <arnd@arndb.de> --- drivers/spi/spi-fsl-espi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) This is a bugfix that we should get into v4.9. It conflicts with a rework of that function that is currently scheduled in spi-next for v4.10, but is also required there in some form. I suppose we also want this backported to -stable, as we also get the warning in older kernels, and the patch that introduced the condition apparently address to fix the hardware problem that would now lead to incorrect data instead of an infinite loop.