On 10.10.22 07:52, Sascha Hauer wrote:
> On Fri, Sep 30, 2022 at 03:37:02PM +0200, Johannes Zink wrote:
>> From: Ahmad Fatoum <a.fatoum@xxxxxxxxxxxxxx>
>>
>> BootROM on the i.MX7 doesn't set the SMP bit when booted
>> over serial download. This leads to vastly worse performance
>> when doing memory-heavy operations in a USB-booted system,
>> as the caches are not utilized. Example running md5sum over
>> a 25M image in ramfs:
>>
>> without patch: 10796ms
>> with patch: 457ms
>>
>> This issue isn't unique to the i.MX7, but exists for the i.MX6UL as
>> well, which also has the Cortex-A7 as CPU. Like with
>> imx6ul_cpu_lowlevel_init(), adapt imx7_cpu_lowlevel_init() to avoid this
>> slow down.
>>
>> Signed-off-by: Ahmad Fatoum <a.fatoum@xxxxxxxxxxxxxx>
>> Signed-off-by: Johannes Zink <j.zink@xxxxxxxxxxxxxx>
>> ---
>> arch/arm/mach-imx/cpu_init.c | 2 +-
>> 1 file changed, 1 insertion(+), 1 deletion(-)
>>
>> diff --git a/arch/arm/mach-imx/cpu_init.c b/arch/arm/mach-imx/cpu_init.c
>> index ea36215419..ede2076102 100644
>> --- a/arch/arm/mach-imx/cpu_init.c
>> +++ b/arch/arm/mach-imx/cpu_init.c
>> @@ -49,7 +49,7 @@ void imx6ul_cpu_lowlevel_init(void)
>>
>> void imx7_cpu_lowlevel_init(void)
>> {
>> - arm_cpu_lowlevel_init();
>> + cortex_a7_lowlevel_init();
>
> Why do you remove the call to arm_cpu_lowlevel_init() here?
Oversight, I thought cortex_a7_lowlevel_init() calls it
internally. Just sent out v2. Thanks.
>
> Sascha
>
--
Pengutronix e.K. | |
Steuerwalder Str. 21 | http://www.pengutronix.de/ |
31137 Hildesheim, Germany | Phone: +49-5121-206917-0 |
Amtsgericht Hildesheim, HRA 2686 | Fax: +49-5121-206917-5555 |
