On Thu, Oct 16, 2025 at 01:07:17PM -0700, Dixit, Ashutosh wrote:
On Wed, 15 Oct 2025 17:03:51 -0700, 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()
Reviewed-by: Ashutosh Dixit <[email protected]>
Thanks, Pushed it.
@Lucas, @Rodrigo
fyi,
I was working on drm-intel-gt-next and did a dim ub at some point and it
automatically switched to drm-xe-next. Instead of pushing
drm-intel-gt-next, I accidentally pushed drm-xe-next without any
changes. Hope that does not cause any issues.
I later pushed the right branch.
Regards,
Umesh
Closes: https://gitlab.freedesktop.org/drm/i915/kernel/-/issues/14955
Signed-off-by: Umesh Nerlige Ramappa <[email protected]>
---
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)
--
2.43.0
