[PATCH] soc: qcom: geni-se: ignore invalid PROG_RAM_DEPTH when checking for firmware size

[PATCH] soc: qcom: geni-se: ignore invalid PROG_RAM_DEPTH when checking for firmware size

[Neil Armstrong]

The I2C Hub serial elements lacks DMA, so the SE_HW_PARAM_2 register which
is in the DMA register space doesn't exist for those instances.

Simply ignore the ram depth if it return 0.

Fixes: 522bfb4f33c0 ("soc: qcom: geni-se: Use HW PROG_RAM_DEPTH to validate firmware size")
Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
---
 drivers/soc/qcom/qcom-geni-se.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/soc/qcom/qcom-geni-se.c b/drivers/soc/qcom/qcom-geni-se.c
index 873bfbd6b2b7..e75349f2c15d 100644
--- a/drivers/soc/qcom/qcom-geni-se.c
+++ b/drivers/soc/qcom/qcom-geni-se.c
@@ -1308,7 +1308,7 @@ static struct se_fw_hdr *geni_find_protocol_fw(struct geni_se *se, const struct
 
 		prog_ram_depth = FIELD_GET(PROG_RAM_DEPTH_MSK,
 					   readl_relaxed(se->base + SE_HW_PARAM_2));
-		if (fw_size >= prog_ram_depth) {
+		if (prog_ram_depth && fw_size >= prog_ram_depth) {
 			dev_err(dev, "Firmware size (%u) exceeds RAM size (%u)\n",
 				fw_size, prog_ram_depth);
 			continue;

---
base-commit: fd73f4a6659897191fa0d40695fe370925dd3780
change-id: 20260917-topic-sm8x50-i2c-hub-fw-ram-54c503be6038

Best regards,
--  
Neil Armstrong <neil.armstrong@linaro.org>

Re: [PATCH] soc: qcom: geni-se: ignore invalid PROG_RAM_DEPTH when checking for firmware size

[Konrad Dybcio]

On 9/17/26 11:07 AM, Neil Armstrong wrote:
> The I2C Hub serial elements lacks DMA, so the SE_HW_PARAM_2 register which
> is in the DMA register space doesn't exist for those instances.
> 
> Simply ignore the ram depth if it return 0.
> 
> Fixes: 522bfb4f33c0 ("soc: qcom: geni-se: Use HW PROG_RAM_DEPTH to validate firmware size")
> Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
> ---
>  drivers/soc/qcom/qcom-geni-se.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
> 
> diff --git a/drivers/soc/qcom/qcom-geni-se.c b/drivers/soc/qcom/qcom-geni-se.c
> index 873bfbd6b2b7..e75349f2c15d 100644
> --- a/drivers/soc/qcom/qcom-geni-se.c
> +++ b/drivers/soc/qcom/qcom-geni-se.c
> @@ -1308,7 +1308,7 @@ static struct se_fw_hdr *geni_find_protocol_fw(struct geni_se *se, const struct
>  
>  		prog_ram_depth = FIELD_GET(PROG_RAM_DEPTH_MSK,
>  					   readl_relaxed(se->base + SE_HW_PARAM_2));
> -		if (fw_size >= prog_ram_depth) {
> +		if (prog_ram_depth && fw_size >= prog_ram_depth) {
>  			dev_err(dev, "Firmware size (%u) exceeds RAM size (%u)\n",
>  				fw_size, prog_ram_depth);

The RAM size on 8650's i2c master hub SEs is 1020 registers (correct,
not 1024) - Viken, would that be true for all implementations?

Konrad

Re: [PATCH] soc: qcom: geni-se: ignore invalid PROG_RAM_DEPTH when checking for firmware size

[Neil Armstrong]

On 9/17/26 11:21, Konrad Dybcio wrote:
> On 9/17/26 11:07 AM, Neil Armstrong wrote:
>> The I2C Hub serial elements lacks DMA, so the SE_HW_PARAM_2 register which
>> is in the DMA register space doesn't exist for those instances.
>>
>> Simply ignore the ram depth if it return 0.
>>
>> Fixes: 522bfb4f33c0 ("soc: qcom: geni-se: Use HW PROG_RAM_DEPTH to validate firmware size")
>> Signed-off-by: Neil Armstrong <neil.armstrong@linaro.org>
>> ---
>>   drivers/soc/qcom/qcom-geni-se.c | 2 +-
>>   1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/drivers/soc/qcom/qcom-geni-se.c b/drivers/soc/qcom/qcom-geni-se.c
>> index 873bfbd6b2b7..e75349f2c15d 100644
>> --- a/drivers/soc/qcom/qcom-geni-se.c
>> +++ b/drivers/soc/qcom/qcom-geni-se.c
>> @@ -1308,7 +1308,7 @@ static struct se_fw_hdr *geni_find_protocol_fw(struct geni_se *se, const struct
>>   
>>   		prog_ram_depth = FIELD_GET(PROG_RAM_DEPTH_MSK,
>>   					   readl_relaxed(se->base + SE_HW_PARAM_2));
>> -		if (fw_size >= prog_ram_depth) {
>> +		if (prog_ram_depth && fw_size >= prog_ram_depth) {
>>   			dev_err(dev, "Firmware size (%u) exceeds RAM size (%u)\n",
>>   				fw_size, prog_ram_depth);
> 
> The RAM size on 8650's i2c master hub SEs is 1020 registers (correct,
> not 1024) - Viken, would that be true for all implementations?

Seems to be right for 8650, if all HUB Ses have 1020 I'll prepare
a v2 with the prog_ram_depth in the se_desc

Thanks,
Neil

> 
> Konrad