On Sat, 30 Sep 2017 19:16:42 +0200, Sagar Arun Kamble
<[email protected]> wrote:
On 9/30/2017 10:36 PM, Sagar Arun Kamble wrote:
On 9/29/2017 11:11 PM, Michal Wajdeczko wrote:
We want to keep GuC functions together. While here move
send registers initialization to early init as this is
one time operation.
Signed-off-by: Michal Wajdeczko <[email protected]>
Cc: Joonas Lahtinen <[email protected]>
Cc: Chris Wilson <[email protected]>
Cc: Sagar Arun Kamble <[email protected]>
<snip>
+void intel_guc_init_early(struct intel_guc *guc)
+{
+ intel_guc_ct_init_early(&guc->ct);
+
+ mutex_init(&guc->send_mutex);
+ guc->send = intel_guc_send_nop;
+ guc->notify = gen8_guc_raise_irq;
+ guc_init_send_regs(guc);
Better to address guc_init_send_regs move in new patch as discussed.
Otherwise overall patch looks good to me.
Please s/dev_priv/i915 in entire patch. Missed earlier.
I can't as rename dev_priv as it is required indirect parameter for
#define I915_READ(reg) dev_priv->uncore.funcs.mmio_readl(dev_priv, (reg),
true)
^^^^^^^^ ^^^^^^^^
#define I915_WRITE(reg, val) dev_priv->uncore.funcs.mmio_writel(dev_priv,
(reg), (val), true)
^^^^^^^^ ^^^^^^^^
Btw, this limitation will make whole dev_priv/i915 cleanup questionable.
Michal
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
