On Sun, Feb 12, 2023 at 10:44:28PM +0200, Oded Gabbay wrote:
> From: Ofir Bitton
>
> Currently we support scenarios where a timestamp registration request
> of a certain offset is received during the interrupt handling of the
> same offset. In this case we give a grace period of up to 100us for
>
On Thu, Feb 16, 2023 at 16:24 PM, Oded Gabbay wrote:
>
> On Thu, Feb 16, 2023 at 12:53 PM Stanislaw Gruszka
> wrote:
> >
> > On Sun, Feb 12, 2023 at 10:44:28PM +0200, Oded Gabbay wrote:
> > > @@ -3178,11 +3181,12 @@ static int
> > > ts_buff_get_kernel_ts_record(struct hl_mmap_mem_buf *buf,
> > >
On Thu, Feb 16, 2023 at 12:53 PM Stanislaw Gruszka
wrote:
>
> On Sun, Feb 12, 2023 at 10:44:28PM +0200, Oded Gabbay wrote:
> > @@ -3178,11 +3181,12 @@ static int ts_buff_get_kernel_ts_record(struct
> > hl_mmap_mem_buf *buf,
> >
> > /* irq handling in the middle give it time
On Sun, Feb 12, 2023 at 10:44:28PM +0200, Oded Gabbay wrote:
> @@ -3178,11 +3181,12 @@ static int ts_buff_get_kernel_ts_record(struct
> hl_mmap_mem_buf *buf,
>
> /* irq handling in the middle give it time to finish */
> spin_unlock_irqrestore(wait_list
From: Ofir Bitton
Currently we support scenarios where a timestamp registration request
of a certain offset is received during the interrupt handling of the
same offset. In this case we give a grace period of up to 100us for
the interrupt handler to finish.
It seems that sometimes the interrupt h