On Mon, Apr 10, 2017 at 04:02:07PM +0100, Chris Wilson wrote:
> We acquire the forcewake and use I915_READ_FW instead for the atomic
> wait within intel_uncore_wait_for_register. However, this still leaves
> us vulnerable to concurrent mmio access to the register, which can cause
> system hangs on gen7. The protection is to acquire uncore.lock around
> each register, so lets add it back.
>
> Signed-off-by: Chris Wilson <[email protected]>
> Cc: Michal Wajdeczko <[email protected]>
> Cc: Joonas Lahtinen <[email protected]>
> ---
> drivers/gpu/drm/i915/intel_uncore.c | 10 +++++++---
> 1 file changed, 7 insertions(+), 3 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/intel_uncore.c
> b/drivers/gpu/drm/i915/intel_uncore.c
> index 53c8457869f6..d2ba3011c2f0 100644
> --- a/drivers/gpu/drm/i915/intel_uncore.c
> +++ b/drivers/gpu/drm/i915/intel_uncore.c
> @@ -1661,14 +1661,18 @@ int intel_wait_for_register(struct drm_i915_private
> *dev_priv,
> u32 value,
> unsigned int timeout_ms)
> {
> -
> unsigned fw =
> intel_uncore_forcewake_for_reg(dev_priv, reg, FW_REG_READ);
> int ret;
>
> - intel_uncore_forcewake_get(dev_priv, fw);
> + spin_lock_irq(&dev_priv->uncore.lock);
> + intel_uncore_forcewake_get__locked(dev_priv, fw);
> +
> ret = wait_for_us((I915_READ_FW(reg) & mask) == value, 2);
To further decrease driver size we may try to call here our helper function:
ret = __intel_wait_for_register_fw(dev_priv, reg, mask, value, 2, 0,
NULL);
-Michal
> - intel_uncore_forcewake_put(dev_priv, fw);
> +
> + intel_uncore_forcewake_put__locked(dev_priv, fw);
> + spin_unlock_irq(&dev_priv->uncore.lock);
> +
> if (ret)
> ret = wait_for((I915_READ_NOTRACE(reg) & mask) == value,
> timeout_ms);
> --
> 2.11.0
>
_______________________________________________
Intel-gfx mailing list
[email protected]
https://lists.freedesktop.org/mailman/listinfo/intel-gfx
