On Fri, Apr 11, 2014 at 12:27 AM, Andrew Ruder andy@aeruder.net wrote:
On Fri, Apr 11, 2014 at 12:24:20AM +0530, Jagan Teki wrote:
At-least from zynq_spi.c case - I wouldn't find that obscure case as you pointed and even with dout NULL which is status poll from (sf_ops.c).
zynq_spi is correct, soft_spi is not.
zynq_spi.c: int spi_xfer(struct spi_slave *slave, unsigned int bitlen, const void *dout, void *din, unsigned long flags) { const u8 *tx_buf = dout; u8 *rx_buf = din, buf; [...] if (tx_buf) buf = *tx_buf++; else buf = 0; [...] if (rx_buf) *rx_buf++ = buf; [...]
It would be great if you mentioned issue scenario for status poll case drivers/mtd/spi/sf_ops.c: ret = spi_xfer(spi, 8, NULL, &status, 0);
That breaks on soft_spi, hence the patch.
OK - means issue only with soft_spi.c is it? Can you share the issue log or typical use case scenario w.r.t soft_spi.c, I need to understand how this got resolved with your change.
I understand you assigned '0' when dout is NULL and you took the buf only when din is !NULL.
thanks!