Message ID | 1402502222-19520-1-git-send-email-chiau.ee.chew@intel.com (mailing list archive) |
---|---|
State | New, archived |
Headers | show |
On Wed, Jun 11, 2014 at 11:57:02PM +0800, Chew Chiau Ee wrote: > It was observed that after module removal followed by insertion, > the SW mode chipselect is not properly set. Thus causing transfer > failure due to incorrect CS toggling. > - value &= ~SPI_CS_CONTROL_SW_MODE; > + orig = value &= ~SPI_CS_CONTROL_SW_MODE; > writel(value, drv_data->ioaddr + offset + SPI_CS_CONTROL); > value = readl(drv_data->ioaddr + offset + SPI_CS_CONTROL); > if (value != orig) { This is a *really* non-obvious fix, I'd have expected the original value to be re-read from the hardware here rather than just tweaked, or at least some comment explaining what's going on. Possibly also saving to a differently named temporary variable would do it but whatever I had to go look at the context to figure out what was happening and even there I was unclear.
diff --git a/drivers/spi/spi-pxa2xx.c b/drivers/spi/spi-pxa2xx.c index a98df7e..cfaf3e6 100644 --- a/drivers/spi/spi-pxa2xx.c +++ b/drivers/spi/spi-pxa2xx.c @@ -126,7 +126,7 @@ static void lpss_ssp_setup(struct driver_data *drv_data) goto detection_done; } - value &= ~SPI_CS_CONTROL_SW_MODE; + orig = value &= ~SPI_CS_CONTROL_SW_MODE; writel(value, drv_data->ioaddr + offset + SPI_CS_CONTROL); value = readl(drv_data->ioaddr + offset + SPI_CS_CONTROL); if (value != orig) {