Re: [Intel-gfx] [PATCH 3/9] drm/i915/pcode: Extend pcode functions for multiple gt's

[Rodrigo Vivi <rodrigo.vivi@intel.com>]

On Thu, Apr 28, 2022 at 05:39:37PM -0700, Ashutosh Dixit wrote:
> Each gt contains an independent instance of pcode. Extend pcode functions
> to interface with pcode on different gt's. To avoid creating dependency of
> display functionality on intel_gt, new pcode function interfaces are
> exposed in terms of uncore rather than intel_gt. Previous struct
> drm_i915_private based pcode interfaces are preserved.
> 
> v2: Expose pcode functions in terms of uncore rather than gt (Jani/Rodrigo)

thank you for that! it looks better with the uncore.
sorry for not thinking about this earlier.
but some comments below...

> 
> Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
> Cc: Jani Nikula <jani.nikula@intel.com>
> Cc: Andi Shyti <andi.shyti@linux.intel.com>
> Signed-off-by: Ashutosh Dixit <ashutosh.dixit@intel.com>
> ---
>  drivers/gpu/drm/i915/gt/intel_gt.c | 17 +++++++
>  drivers/gpu/drm/i915/gt/intel_gt.h |  2 +
>  drivers/gpu/drm/i915/i915_driver.c |  4 +-
>  drivers/gpu/drm/i915/intel_pcode.c | 76 +++++++++++++++---------------
>  drivers/gpu/drm/i915/intel_pcode.h | 29 +++++++++---
>  5 files changed, 80 insertions(+), 48 deletions(-)
> 
> diff --git a/drivers/gpu/drm/i915/gt/intel_gt.c b/drivers/gpu/drm/i915/gt/intel_gt.c
> index 92394f13b42f..07cfe66dd0e8 100644
> --- a/drivers/gpu/drm/i915/gt/intel_gt.c
> +++ b/drivers/gpu/drm/i915/gt/intel_gt.c
> @@ -28,6 +28,7 @@
>  #include "intel_rps.h"
>  #include "intel_gt_sysfs.h"
>  #include "intel_uncore.h"
> +#include "intel_pcode.h"
>  #include "shmem_utils.h"
>  
>  static void __intel_gt_init_early(struct intel_gt *gt)
> @@ -1240,3 +1241,19 @@ void intel_gt_invalidate_tlbs(struct intel_gt *gt)
>  	intel_uncore_forcewake_put_delayed(uncore, FORCEWAKE_ALL);
>  	mutex_unlock(&gt->tlb_invalidate_lock);
>  }
> +
> +int intel_gt_pcode_init(struct drm_i915_private *i915)
> +{
> +	struct intel_gt *gt;
> +	int id, ret;
> +
> +	for_each_gt(gt, i915, id) {
> +		ret = intel_pcode_init(gt->uncore);
> +		if (ret) {
> +			drm_err(&gt->i915->drm, "gt %d: intel_pcode_init failed %d\n", id, ret);
> +			return ret;
> +		}
> +	}
> +
> +	return 0;
> +}
> diff --git a/drivers/gpu/drm/i915/gt/intel_gt.h b/drivers/gpu/drm/i915/gt/intel_gt.h
> index 44c6cb63ccbc..241d833fdb1e 100644
> --- a/drivers/gpu/drm/i915/gt/intel_gt.h
> +++ b/drivers/gpu/drm/i915/gt/intel_gt.h
> @@ -125,6 +125,8 @@ void intel_gt_watchdog_work(struct work_struct *work);
>  
>  void intel_gt_invalidate_tlbs(struct intel_gt *gt);
>  
> +int intel_gt_pcode_init(struct drm_i915_private *i915);
> +
>  struct resource intel_pci_resource(struct pci_dev *pdev, int bar);
>  
>  #endif /* __INTEL_GT_H__ */
> diff --git a/drivers/gpu/drm/i915/i915_driver.c b/drivers/gpu/drm/i915/i915_driver.c
> index 90b0ce5051af..518d6e357017 100644
> --- a/drivers/gpu/drm/i915/i915_driver.c
> +++ b/drivers/gpu/drm/i915/i915_driver.c
> @@ -629,7 +629,7 @@ static int i915_driver_hw_probe(struct drm_i915_private *dev_priv)
>  
>  	intel_opregion_setup(dev_priv);
>  
> -	ret = intel_pcode_init(dev_priv);
> +	ret = intel_gt_pcode_init(dev_priv);
>  	if (ret)
>  		goto err_msi;
>  
> @@ -1251,7 +1251,7 @@ static int i915_drm_resume(struct drm_device *dev)
>  
>  	disable_rpm_wakeref_asserts(&dev_priv->runtime_pm);
>  
> -	ret = intel_pcode_init(dev_priv);
> +	ret = intel_gt_pcode_init(dev_priv);

I didn't like we have this indirection i915 -> gt -> i915...
At the same time I understand you don't want to duplicate the for_each with
the error msg and all in here.

So, what about having in this file a
static int __init_pcode(dev_priv)
?!

>  	if (ret)
>  		return ret;
>  
> diff --git a/drivers/gpu/drm/i915/intel_pcode.c b/drivers/gpu/drm/i915/intel_pcode.c
> index ac727546868e..66020b2e461f 100644
> --- a/drivers/gpu/drm/i915/intel_pcode.c
> +++ b/drivers/gpu/drm/i915/intel_pcode.c
> @@ -52,14 +52,12 @@ static int gen7_check_mailbox_status(u32 mbox)
>  	}
>  }
>  
> -static int __snb_pcode_rw(struct drm_i915_private *i915, u32 mbox,
> +static int intel_pcode_rw(struct intel_uncore *uncore, u32 mbox,

I'm not sure if I like the idea of the renaming here...
I mean, it looks nicer indeed, but at the same time the "intel_"
make it looks it is exported one.

>  			  u32 *val, u32 *val1,
>  			  int fast_timeout_us, int slow_timeout_ms,
>  			  bool is_read)
>  {
> -	struct intel_uncore *uncore = &i915->uncore;
> -
> -	lockdep_assert_held(&i915->sb_lock);
> +	lockdep_assert_held(&uncore->i915->sb_lock);
>  
>  	/*
>  	 * GEN6_PCODE_* are outside of the forcewake domain, we can use
> @@ -88,22 +86,22 @@ static int __snb_pcode_rw(struct drm_i915_private *i915, u32 mbox,
>  	if (is_read && val1)
>  		*val1 = intel_uncore_read_fw(uncore, GEN6_PCODE_DATA1);
>  
> -	if (GRAPHICS_VER(i915) > 6)
> +	if (GRAPHICS_VER(uncore->i915) > 6)
>  		return gen7_check_mailbox_status(mbox);
>  	else
>  		return gen6_check_mailbox_status(mbox);
>  }
>  
> -int snb_pcode_read(struct drm_i915_private *i915, u32 mbox, u32 *val, u32 *val1)
> +int intel_pcode_read(struct intel_uncore *uncore, u32 mbox, u32 *val, u32 *val1)
>  {
>  	int err;
>  
> -	mutex_lock(&i915->sb_lock);
> -	err = __snb_pcode_rw(i915, mbox, val, val1, 500, 20, true);
> -	mutex_unlock(&i915->sb_lock);
> +	mutex_lock(&uncore->i915->sb_lock);
> +	err = intel_pcode_rw(uncore, mbox, val, val1, 500, 20, true);
> +	mutex_unlock(&uncore->i915->sb_lock);
>  
>  	if (err) {
> -		drm_dbg(&i915->drm,
> +		drm_dbg(&uncore->i915->drm,
>  			"warning: pcode (read from mbox %x) mailbox access failed for %ps: %d\n",
>  			mbox, __builtin_return_address(0), err);
>  	}
> @@ -111,18 +109,18 @@ int snb_pcode_read(struct drm_i915_private *i915, u32 mbox, u32 *val, u32 *val1)
>  	return err;
>  }
>  
> -int snb_pcode_write_timeout(struct drm_i915_private *i915, u32 mbox, u32 val,
> -			    int fast_timeout_us, int slow_timeout_ms)
> +int intel_pcode_write_timeout(struct intel_uncore *uncore, u32 mbox, u32 val,
> +			      int fast_timeout_us, int slow_timeout_ms)
>  {
>  	int err;
>  
> -	mutex_lock(&i915->sb_lock);
> -	err = __snb_pcode_rw(i915, mbox, &val, NULL,
> +	mutex_lock(&uncore->i915->sb_lock);
> +	err = intel_pcode_rw(uncore, mbox, &val, NULL,
>  			     fast_timeout_us, slow_timeout_ms, false);
> -	mutex_unlock(&i915->sb_lock);
> +	mutex_unlock(&uncore->i915->sb_lock);
>  
>  	if (err) {
> -		drm_dbg(&i915->drm,
> +		drm_dbg(&uncore->i915->drm,
>  			"warning: pcode (write of 0x%08x to mbox %x) mailbox access failed for %ps: %d\n",
>  			val, mbox, __builtin_return_address(0), err);
>  	}
> @@ -130,18 +128,18 @@ int snb_pcode_write_timeout(struct drm_i915_private *i915, u32 mbox, u32 val,
>  	return err;
>  }
>  
> -static bool skl_pcode_try_request(struct drm_i915_private *i915, u32 mbox,
> -				  u32 request, u32 reply_mask, u32 reply,
> -				  u32 *status)
> +static bool intel_pcode_try_request(struct intel_uncore *uncore, u32 mbox,
> +				    u32 request, u32 reply_mask, u32 reply,
> +				    u32 *status)
>  {
> -	*status = __snb_pcode_rw(i915, mbox, &request, NULL, 500, 0, true);
> +	*status = intel_pcode_rw(uncore, mbox, &request, NULL, 500, 0, true);
>  
>  	return (*status == 0) && ((request & reply_mask) == reply);
>  }
>  
>  /**
> - * skl_pcode_request - send PCODE request until acknowledgment
> - * @i915: device private
> + * intel_pcode_request - send PCODE request until acknowledgment
> + * @uncore: uncore
>   * @mbox: PCODE mailbox ID the request is targeted for
>   * @request: request ID
>   * @reply_mask: mask used to check for request acknowledgment
> @@ -158,16 +156,16 @@ static bool skl_pcode_try_request(struct drm_i915_private *i915, u32 mbox,
>   * Returns 0 on success, %-ETIMEDOUT in case of a timeout, <0 in case of some
>   * other error as reported by PCODE.
>   */
> -int skl_pcode_request(struct drm_i915_private *i915, u32 mbox, u32 request,
> -		      u32 reply_mask, u32 reply, int timeout_base_ms)
> +int intel_pcode_request(struct intel_uncore *uncore, u32 mbox, u32 request,
> +			u32 reply_mask, u32 reply, int timeout_base_ms)
>  {
>  	u32 status;
>  	int ret;
>  
> -	mutex_lock(&i915->sb_lock);
> +	mutex_lock(&uncore->i915->sb_lock);
>  
>  #define COND \
> -	skl_pcode_try_request(i915, mbox, request, reply_mask, reply, &status)
> +	intel_pcode_try_request(uncore, mbox, request, reply_mask, reply, &status)
>  
>  	/*
>  	 * Prime the PCODE by doing a request first. Normally it guarantees
> @@ -193,35 +191,35 @@ int skl_pcode_request(struct drm_i915_private *i915, u32 mbox, u32 request,
>  	 * requests, and for any quirks of the PCODE firmware that delays
>  	 * the request completion.
>  	 */
> -	drm_dbg_kms(&i915->drm,
> +	drm_dbg_kms(&uncore->i915->drm,
>  		    "PCODE timeout, retrying with preemption disabled\n");
> -	drm_WARN_ON_ONCE(&i915->drm, timeout_base_ms > 3);
> +	drm_WARN_ON_ONCE(&uncore->i915->drm, timeout_base_ms > 3);
>  	preempt_disable();
>  	ret = wait_for_atomic(COND, 50);
>  	preempt_enable();
>  
>  out:
> -	mutex_unlock(&i915->sb_lock);
> +	mutex_unlock(&uncore->i915->sb_lock);
>  	return status ? status : ret;
>  #undef COND
>  }
>  
> -int intel_pcode_init(struct drm_i915_private *i915)
> +int intel_pcode_init(struct intel_uncore *uncore)
>  {
> -	int ret = 0;
> +	int ret;
>  
> -	if (!IS_DGFX(i915))
> -		return ret;
> +	if (!IS_DGFX(uncore->i915))
> +		return 0;
>  
> -	ret = skl_pcode_request(i915, DG1_PCODE_STATUS,
> -				DG1_UNCORE_GET_INIT_STATUS,
> -				DG1_UNCORE_INIT_STATUS_COMPLETE,
> -				DG1_UNCORE_INIT_STATUS_COMPLETE, 180000);
> +	ret = intel_pcode_request(uncore, DG1_PCODE_STATUS,
> +				  DG1_UNCORE_GET_INIT_STATUS,
> +				  DG1_UNCORE_INIT_STATUS_COMPLETE,
> +				  DG1_UNCORE_INIT_STATUS_COMPLETE, 180000);
>  
> -	drm_dbg(&i915->drm, "PCODE init status %d\n", ret);
> +	drm_dbg(&uncore->i915->drm, "PCODE init status %d\n", ret);
>  
>  	if (ret)
> -		drm_err(&i915->drm, "Pcode did not report uncore initialization completion!\n");
> +		drm_err(&uncore->i915->drm, "Pcode did not report uncore initialization completion!\n");
>  
>  	return ret;
>  }
> diff --git a/drivers/gpu/drm/i915/intel_pcode.h b/drivers/gpu/drm/i915/intel_pcode.h
> index 0962a17fac48..a03d4ef688aa 100644
> --- a/drivers/gpu/drm/i915/intel_pcode.h
> +++ b/drivers/gpu/drm/i915/intel_pcode.h
> @@ -8,17 +8,32 @@
>  
>  #include <linux/types.h>
>  
> +struct intel_uncore;
>  struct drm_i915_private;
>  
> -int snb_pcode_read(struct drm_i915_private *i915, u32 mbox, u32 *val, u32 *val1);
> -int snb_pcode_write_timeout(struct drm_i915_private *i915, u32 mbox, u32 val,
> -			    int fast_timeout_us, int slow_timeout_ms);
> -#define snb_pcode_write(i915, mbox, val)			\
> +int intel_pcode_read(struct intel_uncore *uncore, u32 mbox, u32 *val, u32 *val1);
> +
> +int intel_pcode_write_timeout(struct intel_uncore *uncore, u32 mbox, u32 val,
> +			      int fast_timeout_us, int slow_timeout_ms);
> +
> +#define intel_pcode_write(uncore, mbox, val) \
> +	intel_pcode_write_timeout(uncore, mbox, val, 500, 0)
> +
> +int intel_pcode_request(struct intel_uncore *uncore, u32 mbox, u32 request,
> +			u32 reply_mask, u32 reply, int timeout_base_ms);
> +
> +#define snb_pcode_read(i915, mbox, val, val1) \
> +	intel_pcode_read(&(i915)->uncore, mbox, val, val1)
> +
> +#define snb_pcode_write_timeout(i915, mbox, val, fast_timeout_us, slow_timeout_ms) \
> +	intel_pcode_write_timeout(&(i915)->uncore, mbox, val, fast_timeout_us, slow_timeout_ms)
> +
> +#define snb_pcode_write(i915, mbox, val) \
>  	snb_pcode_write_timeout(i915, mbox, val, 500, 0)
>  
> -int skl_pcode_request(struct drm_i915_private *i915, u32 mbox, u32 request,
> -		      u32 reply_mask, u32 reply, int timeout_base_ms);
> +#define skl_pcode_request(i915, mbox, request, reply_mask, reply, timeout_base_ms) \
> +	intel_pcode_request(&(i915)->uncore, mbox, request, reply_mask, reply, timeout_base_ms)

and for the exported one, since we are renaming it, shouldn't we rename
all the users instead of creating these defines?

>  
> -int intel_pcode_init(struct drm_i915_private *i915);
> +int intel_pcode_init(struct intel_uncore *uncore);
>  
>  #endif /* _INTEL_PCODE_H */
> -- 
> 2.34.1
>