[dpdk-dev] [EXT] Re: [PATCH 4/4] event/octeontx2: timer always use virtual counter

Pavan Nikhilesh Bhagavatula pbhagavatula at marvell.com
Sun Mar 21 08:11:55 CET 2021



>-----Original Message-----
>From: Jerin Jacob <jerinjacobk at gmail.com>
>Sent: Saturday, March 20, 2021 7:04 PM
>To: Pavan Nikhilesh Bhagavatula <pbhagavatula at marvell.com>
>Cc: Jerin Jacob Kollanukkaran <jerinj at marvell.com>; dpdk-dev
><dev at dpdk.org>
>Subject: [EXT] Re: [dpdk-dev] [PATCH 4/4] event/octeontx2: timer
>always use virtual counter
>
>External Email
>
>----------------------------------------------------------------------
>On Thu, Feb 25, 2021 at 5:53 PM <pbhagavatula at marvell.com> wrote:
>>
>> From: Pavan Nikhilesh <pbhagavatula at marvell.com>
>>
>> Use virtual counter for estimating current bucket as PMU cannot be
>> reliably used to estimate time.
>>
>> Signed-off-by: Pavan Nikhilesh <pbhagavatula at marvell.com>
>
>>
>> +#ifdef RTE_ARCH_ARM64
>> +static inline uint64_t
>> +tim_cntvct(void)
>> +{
>> +       uint64_t tsc;
>> +
>> +       asm volatile("mrs %0, cntvct_el0" : "=r"(tsc));
>> +       return tsc;
>
>
>Reuse __rte_arm64_cntvct()
>
>> +}
>> +
>> +static inline uint64_t
>> +tim_cntfrq(void)
>> +{
>> +       uint64_t freq;
>> +
>> +       asm volatile("mrs %0, cntfrq_el0" : "=r"(freq));
>> +       return freq;
>
>
>Reuse __rte_arm64_cntfrq()

Ack.

>
>
>Please fix the following checkpatch and check format errors too.
>
>
>[1]
>Wrong headline case:
>                        "event/octeontx2: optimize timer arm routine":
>arm --> Arm

I think checkpatch is interpreting this as Arm Inc. which I think we 
can ignore it in this context.

>
>Invalid patch(es) found - checked 4 patches
>
>
>[2]
>total: 0 errors, 47 warnings, 272 lines checked
>Warning in drivers/event/octeontx2/otx2_tim_worker.h:
>Using rte_smp_[r/w]mb
>Warning in drivers/event/octeontx2/otx2_tim_worker.c:
>Using __atomic_thread_fence
>

Ack.

>
>>


More information about the dev mailing list