Intel-GFX Archive on lore.kernel.org
 help / color / mirror / Atom feed
From: Lionel Landwerlin <lionel.g.landwerlin@linux.intel.com>
To: Robert Bragg <robert@sixbynine.org>, intel-gfx@lists.freedesktop.org
Subject: Re: [PATCH v3 6/7] drm/i915/perf: per-gen timebase for checking sample freq
Date: Wed, 5 Apr 2017 17:49:03 +0100	[thread overview]
Message-ID: <66148a4d-4a1e-3836-7546-b3ee91de2dd5@linux.intel.com> (raw)
In-Reply-To: <20170405162320.30094-7-robert@sixbynine.org>

Hey Rob,

Thanks for sending this, it looks good to me.
I think we also need to update the oa_sample_rate_hard_limit & 
i915_oa_max_sample_rate variables.

This patch is :

Reviewed-by: Lionel Landwerlin <lionel.g.landwerlin@intel.com>


On 05/04/17 17:23, Robert Bragg wrote:
> An oa_exponent_to_ns() utility and per-gen timebase constants where
> recently removed when updating the tail pointer race condition WA, and
> this restores those so we can update the _PROP_OA_EXPONENT validation
> done in read_properties_unlocked() to not assume we have a 12.5KHz
> timebase as we did for Haswell.
>
> Signed-off-by: Robert Bragg <robert@sixbynine.org>
> Cc: Lionel Landwerlin <lionel.g.landwerlin@linux.intel.com>
> ---
>   drivers/gpu/drm/i915/i915_drv.h  |  1 +
>   drivers/gpu/drm/i915/i915_perf.c | 21 +++++++++++++++------
>   2 files changed, 16 insertions(+), 6 deletions(-)
>
> diff --git a/drivers/gpu/drm/i915/i915_drv.h b/drivers/gpu/drm/i915/i915_drv.h
> index 3a22b6fd0ee6..48b07d706f06 100644
> --- a/drivers/gpu/drm/i915/i915_drv.h
> +++ b/drivers/gpu/drm/i915/i915_drv.h
> @@ -2463,6 +2463,7 @@ struct drm_i915_private {
>   
>   			bool periodic;
>   			int period_exponent;
> +			int timestamp_frequency;
>   
>   			int metrics_set;
>   
> diff --git a/drivers/gpu/drm/i915/i915_perf.c b/drivers/gpu/drm/i915/i915_perf.c
> index 98eb6415b63a..87c0d1ce1b9f 100644
> --- a/drivers/gpu/drm/i915/i915_perf.c
> +++ b/drivers/gpu/drm/i915/i915_perf.c
> @@ -2549,6 +2549,12 @@ i915_perf_open_ioctl_locked(struct drm_i915_private *dev_priv,
>   	return ret;
>   }
>   
> +static u64 oa_exponent_to_ns(struct drm_i915_private *dev_priv, int exponent)
> +{
> +       return div_u64(1000000000ULL * (2ULL << exponent),
> +                      dev_priv->perf.oa.timestamp_frequency);
> +}
> +
>   /**
>    * read_properties_unlocked - validate + copy userspace stream open properties
>    * @dev_priv: i915 device instance
> @@ -2647,14 +2653,9 @@ static int read_properties_unlocked(struct drm_i915_private *dev_priv,
>   			/* Theoretically we can program the OA unit to sample
>   			 * every 160ns but don't allow that by default unless
>   			 * root.
> -			 *
> -			 * On Haswell the period is derived from the exponent
> -			 * as:
> -			 *
> -			 *   period = 80ns * 2^(exponent + 1)
>   			 */
>   			BUILD_BUG_ON(sizeof(oa_period) != 8);
> -			oa_period = 80ull * (2ull << value);
> +			oa_period = oa_exponent_to_ns(dev_priv, value);
>   
>   			/* This check is primarily to ensure that oa_period <=
>   			 * UINT32_MAX (before passing to do_div which only
> @@ -2910,6 +2911,8 @@ void i915_perf_init(struct drm_i915_private *dev_priv)
>   		dev_priv->perf.oa.ops.oa_hw_tail_read =
>   			gen7_oa_hw_tail_read;
>   
> +		dev_priv->perf.oa.timestamp_frequency = 12500000;
> +
>   		dev_priv->perf.oa.oa_formats = hsw_oa_formats;
>   
>   		dev_priv->perf.oa.n_builtin_sets =
> @@ -2923,6 +2926,8 @@ void i915_perf_init(struct drm_i915_private *dev_priv)
>   		 */
>   
>   		if (IS_GEN8(dev_priv)) {
> +			dev_priv->perf.oa.timestamp_frequency = 12500000;
> +
>   			dev_priv->perf.oa.ctx_oactxctrl_offset = 0x120;
>   			dev_priv->perf.oa.ctx_flexeu0_offset = 0x2ce;
>   			dev_priv->perf.oa.gen8_valid_ctx_bit = (1<<25);
> @@ -2939,6 +2944,8 @@ void i915_perf_init(struct drm_i915_private *dev_priv)
>   					i915_oa_select_metric_set_chv;
>   			}
>   		} else if (IS_GEN9(dev_priv)) {
> +			dev_priv->perf.oa.timestamp_frequency = 12000000;
> +
>   			dev_priv->perf.oa.ctx_oactxctrl_offset = 0x128;
>   			dev_priv->perf.oa.ctx_flexeu0_offset = 0x3de;
>   			dev_priv->perf.oa.gen8_valid_ctx_bit = (1<<16);
> @@ -2959,6 +2966,8 @@ void i915_perf_init(struct drm_i915_private *dev_priv)
>   				dev_priv->perf.oa.ops.select_metric_set =
>   					i915_oa_select_metric_set_sklgt4;
>   			} else if (IS_BROXTON(dev_priv)) {
> +				dev_priv->perf.oa.timestamp_frequency = 19200123;
> +
>   				dev_priv->perf.oa.n_builtin_sets =
>   					i915_oa_n_builtin_metric_sets_bxt;
>   				dev_priv->perf.oa.ops.select_metric_set =


_______________________________________________
Intel-gfx mailing list
Intel-gfx@lists.freedesktop.org
https://lists.freedesktop.org/mailman/listinfo/intel-gfx

  reply	other threads:[~2017-04-05 16:49 UTC|newest]

Thread overview: 22+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-04-05 16:23 [PATCH v3 0/7] Enable OA unit for Gen 8 and 9 in i915 perf Robert Bragg
2017-04-05 16:23 ` [PATCH v3 1/7] drm/i915: expose _SLICE_MASK GETPARM Robert Bragg
2017-04-05 16:23 ` [PATCH v3 2/7] drm/i915: expose _SUBSLICE_MASK GETPARM Robert Bragg
2017-04-05 16:23 ` [PATCH v3 3/7] drm/i915/perf: Add 'render basic' Gen8+ OA unit configs Robert Bragg
2017-04-05 16:23 ` [PATCH v3 4/7] drm/i915/perf: Add OA unit support for Gen 8+ Robert Bragg
2017-04-06 16:56   ` Lionel Landwerlin
2017-04-12 11:33   ` Matthew Auld
2017-04-12 14:57     ` Robert Bragg
2017-04-05 16:23 ` [PATCH v3 5/7] drm/i915/perf: Add more OA configs for BDW, CHV, SKL + BXT Robert Bragg
2017-04-05 16:23 ` [PATCH v3 6/7] drm/i915/perf: per-gen timebase for checking sample freq Robert Bragg
2017-04-05 16:49   ` Lionel Landwerlin [this message]
2017-04-05 17:06   ` Ville Syrjälä
2017-04-05 17:17     ` Lionel Landwerlin
2017-04-05 17:26       ` Ville Syrjälä
2017-04-05 17:59         ` Robert Bragg
2017-04-05 19:05   ` [PATCH v2] " Robert Bragg
2017-04-12 12:34     ` Matthew Auld
2017-04-12 15:07       ` Robert Bragg
2017-04-05 16:23 ` [PATCH v3 7/7] drm/i915/perf: remove perf.hook_lock Robert Bragg
2017-04-12 11:36   ` Matthew Auld
2017-04-05 16:42 ` ✓ Fi.CI.BAT: success for Enable OA unit for Gen 8 and 9 in i915 perf (rev4) Patchwork
2017-04-05 19:21 ` ✓ Fi.CI.BAT: success for Enable OA unit for Gen 8 and 9 in i915 perf (rev5) Patchwork

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=66148a4d-4a1e-3836-7546-b3ee91de2dd5@linux.intel.com \
    --to=lionel.g.landwerlin@linux.intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=robert@sixbynine.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox