All of lore.kernel.org
 help / color / mirror / Atom feed
From: Rodrigo Vivi <rodrigo.vivi@intel.com>
To: Tvrtko Ursulin <tursulin@ursulin.net>
Cc: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>,
	<intel-gfx@lists.freedesktop.org>,
	Ashutosh Dixit <ashutosh.dixit@intel.com>,
	 <andi.shyti@kernel.org>
Subject: Re: [PATCH] drm/i915: Fix conversion between clock ticks and nanoseconds
Date: Wed, 5 Nov 2025 14:38:18 -0500	[thread overview]
Message-ID: <aQunqj1t9tG-pK4t@intel.com> (raw)
In-Reply-To: <e60834fa-2d03-4531-a56e-fa8d1799ed11@ursulin.net>

On Fri, Oct 31, 2025 at 08:40:53AM +0000, Tvrtko Ursulin wrote:
> 
> On 29/10/2025 22:01, Umesh Nerlige Ramappa wrote:
> > On Wed, Oct 29, 2025 at 11:33:20AM +0000, Tvrtko Ursulin wrote:
> > > 
> > > On 16/10/2025 01:03, Umesh Nerlige Ramappa wrote:
> > > > When tick values are large, the multiplication by NSEC_PER_SEC is larger
> > > > than 64 bits and results in bad conversions.
> > > > 
> > > > The issue is seen in PMU busyness counters that look like they have
> > > > wrapped around due to bad conversion. i915 PMU implementation returns
> > > > monotonically increasing counters. If a count is lesser than previous
> > > > one, it will only return the larger value until the smaller value
> > > > catches up. The user will see this as zero delta between two
> > > > measurements even though the engines are busy.
> > > > 
> > > > Fix it by using mul_u64_u32_div()
> > > > 
> > > > Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14955
> > > > Signed-off-by: Umesh Nerlige Ramappa <umesh.nerlige.ramappa@intel.com>
> > > 
> > > There was no need for Fixes: or cc: stable?
> > 
> > Oh, sorry, I missed that. Should have been both. Is it too late to do
> > that now?
> > 
> > Fixes: 77cdd054dd2c ("drm/i915/pmu: Connect engine busyness stats from
> > GuC to pmu")
> 
> Possibly not. Lets ask Rodrigo if he could manually cherry pick it for 6.18.

Done, and added the fixes tag while at it.

Thanks for the heads up.

> 
> Regards,
> 
> Tvrtko
> 
> > > > ---
> > > > v2:
> > > > - Fix divide by zero for Gen11 (Andi)
> > > > - Update commit message
> > > > 
> > > > v3:
> > > > - Drop GCD and use mul_u64_u32_div() instead (Ashutosh)
> > > > ---
> > > >  drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c | 4 ++--
> > > >  1 file changed, 2 insertions(+), 2 deletions(-)
> > > > 
> > > > diff --git a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c b/
> > > > drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c
> > > > index 88b147fa5cb1..c90b35881a26 100644
> > > > --- a/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c
> > > > +++ b/drivers/gpu/drm/i915/gt/intel_gt_clock_utils.c
> > > > @@ -205,7 +205,7 @@ static u64 div_u64_roundup(u64 nom, u32 den)
> > > >  u64 intel_gt_clock_interval_to_ns(const struct intel_gt *gt, u64 count)
> > > >  {
> > > > -    return div_u64_roundup(count * NSEC_PER_SEC, gt->clock_frequency);
> > > > +    return mul_u64_u32_div(count, NSEC_PER_SEC, gt->clock_frequency);
> > > >  }
> > > >  u64 intel_gt_pm_interval_to_ns(const struct intel_gt *gt, u64 count)
> > > > @@ -215,7 +215,7 @@ u64 intel_gt_pm_interval_to_ns(const struct
> > > > intel_gt *gt, u64 count)
> > > >  u64 intel_gt_ns_to_clock_interval(const struct intel_gt *gt, u64 ns)
> > > >  {
> > > > -    return div_u64_roundup(gt->clock_frequency * ns, NSEC_PER_SEC);
> > > > +    return mul_u64_u32_div(ns, gt->clock_frequency, NSEC_PER_SEC);
> > > >  }
> > > >  u64 intel_gt_ns_to_pm_interval(const struct intel_gt *gt, u64 ns)
> > > 
> 

  reply	other threads:[~2025-11-05 19:38 UTC|newest]

Thread overview: 18+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2025-10-16  0:03 [PATCH] drm/i915: Fix conversion between clock ticks and nanoseconds Umesh Nerlige Ramappa
2025-10-16  1:17 ` ✓ i915.CI.BAT: success for drm/i915: Fix conversion between clock ticks and nanoseconds (rev3) Patchwork
2025-10-16 10:18 ` ✓ i915.CI.Full: " Patchwork
2025-10-16 20:07 ` [PATCH] drm/i915: Fix conversion between clock ticks and nanoseconds Dixit, Ashutosh
2025-10-16 23:36   ` Umesh Nerlige Ramappa
2025-10-17  2:01     ` Rodrigo Vivi
2025-10-29 11:33 ` Tvrtko Ursulin
2025-10-29 22:01   ` Umesh Nerlige Ramappa
2025-10-31  8:40     ` Tvrtko Ursulin
2025-11-05 19:38       ` Rodrigo Vivi [this message]
  -- strict thread matches above, loose matches on Subject: below --
2025-10-09 19:16 Umesh Nerlige Ramappa
2025-10-07 23:35 Umesh Nerlige Ramappa
2025-10-09 12:24 ` Andi Shyti
2025-10-10 17:45   ` Umesh Nerlige Ramappa
2025-10-14  8:58     ` Andi Shyti
2025-10-14  1:19 ` Dixit, Ashutosh
2025-10-14 20:31   ` Umesh Nerlige Ramappa
2025-10-14 21:43     ` Dixit, Ashutosh

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=aQunqj1t9tG-pK4t@intel.com \
    --to=rodrigo.vivi@intel.com \
    --cc=andi.shyti@kernel.org \
    --cc=ashutosh.dixit@intel.com \
    --cc=intel-gfx@lists.freedesktop.org \
    --cc=tursulin@ursulin.net \
    --cc=umesh.nerlige.ramappa@intel.com \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.