On to, 2017-01-19 at 11:41 +0000, Chris Wilson wrote:
> Just create several batches of requests and expect it to not fall over!
>
> Signed-off-by: Chris Wilson <ch...@chris-wilson.co.uk>
<SNIP>
> +static int live_nop_request(void *arg)
> +{
<SNIP>
> + for_each_prime_number_from(prime, 1, 8192) {
#define MAGIC_PRIME_2 8192
> + times[1] = ktime_get_raw();
> +
> + for (n = 0; n < prime; n++) {
> + request = i915_gem_request_alloc(engine,
> +
> i915->kernel_context);
> + if (IS_ERR(request)) {
> + err = PTR_ERR(request);
> + goto out_unlock;
> + }
Emitting even a single MI_NOOP or making a comment that the request
will contain instructions even if we do nothing here, might help a
newcomer.
> +
> + i915_add_request(request);
> + }
> + i915_wait_request(request,
> + I915_WAIT_LOCKED,
> + MAX_SCHEDULE_TIMEOUT);
> +
> + times[1] = ktime_sub(ktime_get_raw(), times[1]);
> + if (prime == 1)
> + times[0] = times[1];
Having this as an array is just hiding names and gaining nothing, how
about calling them just time_first, time_last?
Reviewed-by: Joonas Lahtinen <joonas.lahti...@linux.intel.com>
Regards, Joonas
--
Joonas Lahtinen
Open Source Technology Center
Intel Corporation
_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
