Skip to contentSkip to conversation

mm: fix arithmetic for bdi min_ratio and

PATCH Superseded Jingbo Xu 2 participants 7 messages Dec 19, 2023

Subsystems (4)

v2 of 3 Superseded Latest is v3 →

lore
Share & tools
lore

Jingbo Xu

author
#0 lore raw

v2 cover mm: fix arithmetic for bdi min_ratio and

changes since v1:

- split the previous v1 patch into two separate patches with corresponding "Fixes" tag (Andrew Morton)
- patch 2: remove "UL" suffix for the "100" constant

v1: https://lore.kernel.org/all/[email protected]/

Jingbo Xu (2):
mm: fix arithmetic for bdi min_ratio
mm: fix arithmetic for max_prop_frac when setting max_ratio

 mm/page-writeback.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
Signature · 1 line
2.19.1.6.gb485710b
author
#1 lore raw

PATCH v2 1/2 mm: fix arithmetic for bdi min_ratio

Since now bdi->min_ratio is part per million, fix the wrong arithmetic. Otherwise it will fail with -EINVAL when setting a reasonable min_ratio, as it tries to set min_ratio to (min_ratio * BDI_RATIO_SCALE) in percentage unit, which exceeds 100% anyway.

    # cat /sys/class/bdi/253\:0/min_ratio
    0
    # cat /sys/class/bdi/253\:0/max_ratio
    100
    # echo 1 > /sys/class/bdi/253\:0/min_ratio
    -bash: echo: write error: Invalid argument
Fixes 8021fb3232f2 ("mm: split off __bdi_set_min_ratio() function")
---
 mm/page-writeback.c | 1 -
 1 file changed, 1 deletion(-)
mm/page-writeback.c +0 -1
index ee2fd6a6af40..2140382dd768 100644
@@ -692,7 +692,6 @@ static int __bdi_set_min_ratio(struct backing_dev_info *bdi, unsigned int min_ra
 
 	if (min_ratio > 100 * BDI_RATIO_SCALE)
 		return -EINVAL;
-	min_ratio *= BDI_RATIO_SCALE;
 
 	spin_lock_bh(&bdi_lock);
 	if (min_ratio > bdi->max_ratio) {
Signature · 1 line
2.19.1.6.gb485710b
author
#2 lore raw

PATCH v2 2/2 mm: fix arithmetic for max_prop_frac when setting max_ratio

In reply to Jingbo Xu

Since now bdi->max_ratio is part per million, fix the wrong arithmetic for max_prop_frac when setting max_ratio. Otherwise the miscalculated max_prop_frac will affect the incrementing of writeout completion count when max_ratio is not 100%.

Fixes efc3e6ad53ea ("mm: split off __bdi_set_max_ratio() function")
---
 mm/page-writeback.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
mm/page-writeback.c +2 -1
index 2140382dd768..dda59b368c01 100644
@@ -728,7 +728,8 @@ static int __bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned int max_ra
 		ret = -EINVAL;
 	} else {
 		bdi->max_ratio = max_ratio;
-		bdi->max_prop_frac = (FPROP_FRAC_BASE * max_ratio) / 100;
+		bdi->max_prop_frac = div64_u64(FPROP_FRAC_BASE * max_ratio,
+					       100 * BDI_RATIO_SCALE);
 	}
 	spin_unlock_bh(&bdi_lock);
 
Signature · 1 line
2.19.1.6.gb485710b
maintainer
#3 lore raw

Re: [PATCH v2 2/2] mm: fix arithmetic for max_prop_frac when setting max_ratio

mm/page-writeback.c static int __bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned int max_ra · line 733

Quoted 6 lines · Jingbo Xu
mm/page-writeback.c +2 -1
 	} else {
 		bdi->max_ratio = max_ratio;
-		bdi->max_prop_frac = (FPROP_FRAC_BASE * max_ratio) / 100;
+		bdi->max_prop_frac = div64_u64(FPROP_FRAC_BASE * max_ratio,
+					       100 * BDI_RATIO_SCALE);
 	}

Why use div64_u64 here?

FPROP_FRAC_BASE is an unsigned long. max_ratio is an unsigned int, so the numerator is an unsigned long. BDI_RATIO_SCALE is 10,000, so the numerator is an unsigned int. There's no 64-bit arithmetic needed here.

author
#4 lore raw

Re: [PATCH v2 2/2] mm: fix arithmetic for max_prop_frac when setting max_ratio

Quoted 10 lines · Matthew Wilcox

mm/page-writeback.c static int __bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned int max_ra · line 733

Quoted 6 lines · Jingbo Xu
mm/page-writeback.c +2 -1
 	} else {
 		bdi->max_ratio = max_ratio;
-		bdi->max_prop_frac = (FPROP_FRAC_BASE * max_ratio) / 100;
+		bdi->max_prop_frac = div64_u64(FPROP_FRAC_BASE * max_ratio,
+					       100 * BDI_RATIO_SCALE);
 	}

Why use div64_u64 here?

Quoted 3 lines

FPROP_FRAC_BASE is an unsigned long. max_ratio is an unsigned int, so the numerator is an unsigned long. BDI_RATIO_SCALE is 10,000, so the numerator is an unsigned int. There's no 64-bit arithmetic needed here.

Yes, div64_u64() is actually not needed here. So it seems

bdi->max_prop_frac = FPROP_FRAC_BASE * max_ratio / 100 / BDI_RATIO_SCALE;

is adequate?

Signature · 2 lines
Thanks,
Jingbo
maintainer
#5 lore raw

Re: [PATCH v2 2/2] mm: fix arithmetic for max_prop_frac when setting max_ratio

Quoted 15 lines · Jingbo Xu
Quoted 10 lines · Matthew Wilcox

mm/page-writeback.c static int __bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned int max_ra · line 733

Quoted 6 lines · Jingbo Xu
mm/page-writeback.c +2 -1
 	} else {
 		bdi->max_ratio = max_ratio;
-		bdi->max_prop_frac = (FPROP_FRAC_BASE * max_ratio) / 100;
+		bdi->max_prop_frac = div64_u64(FPROP_FRAC_BASE * max_ratio,
+					       100 * BDI_RATIO_SCALE);
 	}

Why use div64_u64 here?

Quoted 3 lines

FPROP_FRAC_BASE is an unsigned long. max_ratio is an unsigned int, so the numerator is an unsigned long. BDI_RATIO_SCALE is 10,000, so the numerator is an unsigned int. There's no 64-bit arithmetic needed here.

Quoted 5 lines

Yes, div64_u64() is actually not needed here. So it seems

bdi->max_prop_frac = FPROP_FRAC_BASE * max_ratio / 100 / BDI_RATIO_SCALE;

is adequate?

I'd rather spell that as:

		bdi->max_prop_frac = (FPROP_FRAC_BASE * max_ratio) /
					(100 * BDI_RATIO_SCALE);

It's closer to how you'd write it out mathematically and so it reads more easily. At least for me.

author
#6 lore raw

Re: [PATCH v2 2/2] mm: fix arithmetic for max_prop_frac when setting max_ratio

Quoted 22 lines · Matthew Wilcox
Quoted 15 lines · Jingbo Xu
Quoted 10 lines · Matthew Wilcox

mm/page-writeback.c static int __bdi_set_max_ratio(struct backing_dev_info *bdi, unsigned int max_ra · line 733

Quoted 6 lines · Jingbo Xu
mm/page-writeback.c +2 -1
 	} else {
 		bdi->max_ratio = max_ratio;
-		bdi->max_prop_frac = (FPROP_FRAC_BASE * max_ratio) / 100;
+		bdi->max_prop_frac = div64_u64(FPROP_FRAC_BASE * max_ratio,
+					       100 * BDI_RATIO_SCALE);
 	}

Why use div64_u64 here?

Quoted 3 lines

FPROP_FRAC_BASE is an unsigned long. max_ratio is an unsigned int, so the numerator is an unsigned long. BDI_RATIO_SCALE is 10,000, so the numerator is an unsigned int. There's no 64-bit arithmetic needed here.

Quoted 5 lines

Yes, div64_u64() is actually not needed here. So it seems

bdi->max_prop_frac = FPROP_FRAC_BASE * max_ratio / 100 / BDI_RATIO_SCALE;

is adequate?

Quoted 7 lines

I'd rather spell that as:

		bdi->max_prop_frac = (FPROP_FRAC_BASE * max_ratio) /
					(100 * BDI_RATIO_SCALE);

It's closer to how you'd write it out mathematically and so it reads more easily. At least for me.

Thanks, I would send v3 soon.

Signature · 2 lines
Thanks,
Jingbo