Quoting Mika Kuoppala (2017-10-31 15:27:34)
> From: Mika Kuoppala <mika.kuoppala@xxxxxxxxx>
>
> As all our access to execlist ports are through head and tail
> helpers, we can now move the head instead of memmoving the array.
>
> v2: use memset (Chris)
>
> Cc: Michał Winiarski <michal.winiarski@xxxxxxxxx>
> Cc: Joonas Lahtinen <joonas.lahtinen@xxxxxxxxxxxxxxx>
> Cc: Chris Wilson <chris@xxxxxxxxxxxxxxxxxx>
> Signed-off-by: Mika Kuoppala <mika.kuoppala@xxxxxxxxxxxxxxx>
> ---
> drivers/gpu/drm/i915/intel_ringbuffer.h | 10 +++++-----
> 1 file changed, 5 insertions(+), 5 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_ringbuffer.h b/drivers/gpu/drm/i915/intel_ringbuffer.h
> index 387667fe50d3..011c4b8f1339 100644
> --- a/drivers/gpu/drm/i915/intel_ringbuffer.h
> +++ b/drivers/gpu/drm/i915/intel_ringbuffer.h
> @@ -611,13 +611,13 @@ static inline struct execlist_port *
> execlists_head_complete(struct intel_engine_execlists * const execlists,
> struct execlist_port * const port)
> {
> - const unsigned int m = execlists->port_mask;
> -
> - GEM_BUG_ON(port_index(port, execlists) != 0);
> + GEM_BUG_ON(port_index(port, execlists) != execlists->port_head);
> + GEM_BUG_ON(!port_isset(port));
> GEM_BUG_ON(!execlists_is_active(execlists, EXECLISTS_ACTIVE_USER));
>
> - memmove(port, port + 1, m * sizeof(struct execlist_port));
> - memset(port + m, 0, sizeof(struct execlist_port));
> + memset(port, 0, sizeof(*port));
> +
> + execlists->port_head = port_head_add(execlists, 1);
Ok, I would have gone for
port = port_next(port);
execlists->port_head = port - execlists->port;
return port;
That to me looks more natural advance of port as we complete the
requests, and matches the loop in the irq handler.
Care to crunch the numbers and see which gcc favours?
-Chris
_______________________________________________
Intel-gfx mailing list
Intel-gfx@xxxxxxxxxxxxxxxxxxxxx
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
