Re: [PATCH v5 3/9] spi: Add a driver for the Freescale/NXP QuadSPI controller

[Schrempf Frieder <frieder.schrempf@kontron.De>]

Hi Yogesh,

On 14.11.18 09:50, Frieder Schrempf wrote:
> Hi Yogesh,
> 
> On 14.11.18 09:39, Yogesh Narayan Gaur wrote:
>> Hi Frieder,
>>
>> I have tried v5 version of the patch and have observed that Read is 
>> failing for CS1.
> 
> Thanks a lot for doing the test. I really appreciate it.
> 
>> In my target 2 flash devices are connected on same bus i.e. A1 -> CS0 
>> and A2 -> CS1.
>>
>> On initial debugging, I figured that Read is failing for the AHB mode 
>> i.e. if I attempt to read data size less than rxfifo read is working 
>> fine without any issue.
>>
>> For data size more than rxfifo Read out data is content of same 
>> requested address of CS0.
>>     mtd_debug read /dev/mtd1 0xf00000 0x70 read --> Data is correct
>>     mtd_debug read /dev/mtd1 0xf00000 0x100 read --> Data is 
>> in-correct and data content are of the address 0xf00000 of CS0 
>> connected flash device.
> 
> Ok, I will have a look at what could make the chip selection fail in 
> case of AHB read.

Could you try with this change applied:

@@ -503,7 +503,7 @@ static void fsl_qspi_select_mem(struct fsl_qspi *q, 
struct spi_device *spi)
                         map_addr = q->memmap_phy;
                 else
                         map_addr = q->memmap_phy +
-                                  2 * q->devtype_data->ahb_buf_size;
+                                  q->devtype_data->ahb_buf_size;

                 qspi_writel(q, map_addr, q->iobase + QUADSPI_SFA1AD + 
(i * 4));
         }

> 
>> On the setup where you have done testing, did AHB mode read is being 
>> verified for CS1?
> 
> No, I currently have only hardware with CS0 connected. So I didn't test 
> with CS1.
> 
>>
>> I am doing further debugging of this issue.
> 
> Thanks,
> Frieder