[Mesa-dev] [PATCH 5/5] gallium/radeon: don't monitor SDMA busyness on EG/Cayman

[Marek Olšák]

It looks like I'll also have to skip the SDMA counter update on SI,
because it's always busy too. Consider it done.

Marek

On Sun, Mar 5, 2017 at 11:22 PM, Marek Olšák <maraeo at gmail.com> wrote:
> From: Marek Olšák <marek.olsak at amd.com>
>
> It's always busy.
>
> Bugzilla: https://bugs.freedesktop.org/show_bug.cgi?id=99955
> ---
>  src/gallium/drivers/radeon/r600_gpu_load.c | 2 +-
>  1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/src/gallium/drivers/radeon/r600_gpu_load.c b/src/gallium/drivers/radeon/r600_gpu_load.c
> index 24f54c0..2d3f000 100644
> --- a/src/gallium/drivers/radeon/r600_gpu_load.c
> +++ b/src/gallium/drivers/radeon/r600_gpu_load.c
> @@ -98,21 +98,21 @@ static void r600_update_mmio_counters(struct r600_common_screen *rscreen,
>         UPDATE_COUNTER(spi, SPI_BUSY);
>         UPDATE_COUNTER(bci, BCI_BUSY);
>         UPDATE_COUNTER(sc, SC_BUSY);
>         UPDATE_COUNTER(pa, PA_BUSY);
>         UPDATE_COUNTER(db, DB_BUSY);
>         UPDATE_COUNTER(cp, CP_BUSY);
>         UPDATE_COUNTER(cb, CB_BUSY);
>         UPDATE_COUNTER(gui, GUI_ACTIVE);
>         gui_busy = GUI_ACTIVE(value);
>
> -       if (rscreen->chip_class >= EVERGREEN) {
> +       if (rscreen->chip_class >= SI) {
>                 /* SRBM_STATUS2 */
>                 rscreen->ws->read_registers(rscreen->ws, SRBM_STATUS2, 1, &value);
>
>                 UPDATE_COUNTER(sdma, SDMA_BUSY);
>                 sdma_busy = SDMA_BUSY(value);
>         }
>
>         if (rscreen->chip_class >= VI) {
>                 /* CP_STAT */
>                 rscreen->ws->read_registers(rscreen->ws, CP_STAT, 1, &value);
> --
> 2.7.4
>