On Fri, Apr 19, 2013 at 8:23 PM, Peter Maydell <peter.mayd...@linaro.org> wrote:
> On 19 April 2013 03:06, <peter.crosthwa...@xilinx.com> wrote:
>> From: Peter Crosthwaite <peter.crosthwa...@xilinx.com>
>>
>> Invalidate the LQSPI cached page when transitioning into LQSPI mode.
>> Otherwise there is a possibility that the controller will return stale
>> data to the guest when transitioning back to LQ_MODE after a page
>> program.
>>
>> Signed-off-by: Peter Crosthwaite <peter.crosthwa...@xilinx.com>
>> ---
>> changed from v1:
>> Re-implemented using separate SPI/QSPI write handlers.
>>
>> hw/ssi/xilinx_spips.c | 26 +++++++++++++++++++++++++-
>> 1 files changed, 25 insertions(+), 1 deletions(-)
>>
>> diff --git a/hw/ssi/xilinx_spips.c b/hw/ssi/xilinx_spips.c
>> index e351cb2..6d38111 100644
>> --- a/hw/ssi/xilinx_spips.c
>> +++ b/hw/ssi/xilinx_spips.c
>> @@ -165,6 +165,8 @@ typedef struct {
>> typedef struct XilinxSPIPSClass {
>> SysBusDeviceClass parent_class;
>>
>> + const MemoryRegionOps *reg_ops;
>> +
>> uint32_t rx_fifo_size;
>> uint32_t tx_fifo_size;
>> } XilinxSPIPSClass;
>> @@ -446,6 +448,7 @@ static void xilinx_spips_write(void *opaque, hwaddr addr,
>> case R_TXD3:
>> tx_data_bytes(s, (uint32_t)value, 3);
>> goto no_reg_update;
>> + break;
>
> Why insert code after a goto?
>
Removed.
Thanks,
Peter
>> }
>> s->regs[addr] = (s->regs[addr] & ~mask) | (value & mask);
>> no_reg_update:
>
> -- PMM
>
