| Message ID | 20180419165333.32962-1-andriy.shevchenko@linux.intel.com (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 7956fadd40af82cec610a0ccf1ca937e0dfecf94 |
| Headers | show |
On 04/19/18 19:53, Andy Shevchenko wrote: > Currently the 32-bit device address only is supported for DMA. However, > starting from Intel Sunrisepoint PCH the DMA address of the device FIFO > can be 64-bit. > > Change the respective variable to be compatible with DMA engine > expectations, i.e. to phys_addr_t. > > Fixes: 34cadd9c1bcb ("spi: pxa2xx: Add support for Intel Sunrisepoint") > Cc: Jarkko Nikula <jarkko.nikula@linux.intel.com> > Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> > --- > drivers/spi/spi-pxa2xx.h | 2 +- > 1 file changed, 1 insertion(+), 1 deletion(-) > > diff --git a/drivers/spi/spi-pxa2xx.h b/drivers/spi/spi-pxa2xx.h > index 513ec6c6e25b..0ae7defd3492 100644 > --- a/drivers/spi/spi-pxa2xx.h > +++ b/drivers/spi/spi-pxa2xx.h > @@ -38,7 +38,7 @@ struct driver_data { > > /* SSP register addresses */ > void __iomem *ioaddr; > - u32 ssdr_physical; > + phys_addr_t ssdr_physical; > Nice catch! Sunrisepoint SPI PCI BAR indeed has BAR high register and BAR low type bits indicate 64 address range even BAR high seems to 0 in my test machines. Have you seen a real issue with this? And more over, can this cause memory corruption if DMA is writing into (ssp->phys_base + SSDR) & 0xFFFFFFFF? I don't know so well the DMA handshaking can it do so but both points are valid reasons to queue the fix into linux-stable. Reviewed-by: Jarkko Nikula <jarkko.nikula@linux.intel.com> -- To unsubscribe from this list: send the line "unsubscribe linux-spi" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On Fri, 2018-04-20 at 10:54 +0300, Jarkko Nikula wrote: > On 04/19/18 19:53, Andy Shevchenko wrote: > > Currently the 32-bit device address only is supported for DMA. > > However, > > starting from Intel Sunrisepoint PCH the DMA address of the device > > FIFO > > can be 64-bit. > > > > Change the respective variable to be compatible with DMA engine > > expectations, i.e. to phys_addr_t. > > > > Fixes: 34cadd9c1bcb ("spi: pxa2xx: Add support for Intel > > Sunrisepoint") > > Cc: Jarkko Nikula <jarkko.nikula@linux.intel.com> > > Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> > > --- > > drivers/spi/spi-pxa2xx.h | 2 +- > > 1 file changed, 1 insertion(+), 1 deletion(-) > > > > diff --git a/drivers/spi/spi-pxa2xx.h b/drivers/spi/spi-pxa2xx.h > > index 513ec6c6e25b..0ae7defd3492 100644 > > --- a/drivers/spi/spi-pxa2xx.h > > +++ b/drivers/spi/spi-pxa2xx.h > > @@ -38,7 +38,7 @@ struct driver_data { > > > > /* SSP register addresses */ > > void __iomem *ioaddr; > > - u32 ssdr_physical; > > + phys_addr_t ssdr_physical; > > > > Nice catch! Sunrisepoint SPI PCI BAR indeed has BAR high register and > BAR low type bits indicate 64 address range even BAR high seems to 0 > in > my test machines. > > Have you seen a real issue with this? Yes. It fixes an actual issue on our Kabylake and Coffeelake systems. > And more over, can this cause > memory corruption if DMA is writing into (ssp->phys_base + SSDR) & > 0xFFFFFFFF? I don't know so well the DMA handshaking can it do so but > both points are valid reasons to queue the fix into linux-stable. It requires DMA request / acknowledge lines to be asserted in certain time. It might be a corruption up to burst size though.
diff --git a/drivers/spi/spi-pxa2xx.h b/drivers/spi/spi-pxa2xx.h index 513ec6c6e25b..0ae7defd3492 100644 --- a/drivers/spi/spi-pxa2xx.h +++ b/drivers/spi/spi-pxa2xx.h @@ -38,7 +38,7 @@ struct driver_data { /* SSP register addresses */ void __iomem *ioaddr; - u32 ssdr_physical; + phys_addr_t ssdr_physical; /* SSP masks*/ u32 dma_cr1;
Currently the 32-bit device address only is supported for DMA. However, starting from Intel Sunrisepoint PCH the DMA address of the device FIFO can be 64-bit. Change the respective variable to be compatible with DMA engine expectations, i.e. to phys_addr_t. Fixes: 34cadd9c1bcb ("spi: pxa2xx: Add support for Intel Sunrisepoint") Cc: Jarkko Nikula <jarkko.nikula@linux.intel.com> Signed-off-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com> --- drivers/spi/spi-pxa2xx.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)