I have been meaning to make this patch use UINT_MAX and submit to the
kernel for a long time. Anyone want to give it a spin?
---------- Forwarded message ---------
From: Dave Taht <dave.t...@gmail.com>
Date: Sat, Feb 11, 2023 at 5:59 PM
Subject: been a while since I did a kernel patch
To: Toke Høiland-Jørgensen <t...@toke.dk>
workflow still goes into net-next? You need a signed-off-by, and a tested-by?
(stupid patch attached)
--
This song goes out to all the folk that thought Stadia would work:
https://www.linkedin.com/posts/dtaht_the-mushroom-song-activity-6981366665607352320-FXtz
Dave Täht CEO, TekLibre, LLC
--
Podcast: https://www.youtube.com/watch?v=bxmoBr4cBKg
Dave Täht CSO, LibreQos
From cad12c8cd0e6031d33ea64209c4308ae10a7dd71 Mon Sep 17 00:00:00 2001
From: Dave Taht <dtaht@taht.net>
Date: Sun, 12 Feb 2023 01:53:40 +0000
Subject: [PATCH] sch_cake: use const invsqrt lookup table
It was silly to actually calculate these at run time, particularly
in an age where 10s of thousands of cake instances exist.
This lookup table is also mildly more accurate.
---
net/sched/sch_cake.c | 50 ++++++++++++++------------------------------
1 file changed, 16 insertions(+), 34 deletions(-)
diff --git a/net/sched/sch_cake.c b/net/sched/sch_cake.c
index 3ed0c3342189..ead5ea0605a3 100644
--- a/net/sched/sch_cake.c
+++ b/net/sched/sch_cake.c
@@ -360,8 +360,23 @@ static const u8 besteffort[] = {
static const u8 normal_order[] = {0, 1, 2, 3, 4, 5, 6, 7};
static const u8 bulk_order[] = {1, 0, 2, 3};
+/* There is a big difference in timing between the accurate values placed in
+ * the cache and the approximations given by a single Newton step for small
+ * count values, particularly when stepping from count 1 to 2 or vice versa.
+ * Above 16, a single Newton step gives sufficient accuracy in either
+ * direction, given the precision stored.
+ *
+ * The magnitude of the error when stepping up to count 2 is such as to give
+ * the value that *should* have been produced at count 4.
+ */
+
#define REC_INV_SQRT_CACHE (16)
-static u32 cobalt_rec_inv_sqrt_cache[REC_INV_SQRT_CACHE] = {0};
+static const u32 inv_sqrt_cache[REC_INV_SQRT_CACHE] = {
+ ~0, ~0, 3037000500, 2479700525,
+ 2147483647, 1920767767, 1753413056, 1623345051,
+ 1518500250, 1431655765, 1358187913, 1294981364,
+ 1239850263, 1191209601, 1147878294, 1108955788
+};
/* http://en.wikipedia.org/wiki/Methods_of_computing_square_roots
* new_invsqrt = (invsqrt / 2) * (3 - count * invsqrt^2)
@@ -392,42 +407,9 @@ static void cobalt_invsqrt(struct cobalt_vars *vars)
cobalt_newton_step(vars);
}
-/* There is a big difference in timing between the accurate values placed in
- * the cache and the approximations given by a single Newton step for small
- * count values, particularly when stepping from count 1 to 2 or vice versa.
- * Above 16, a single Newton step gives sufficient accuracy in either
- * direction, given the precision stored.
- *
- * The magnitude of the error when stepping up to count 2 is such as to give
- * the value that *should* have been produced at count 4.
- */
-
-static void cobalt_cache_init(void)
-{
- struct cobalt_vars v;
-
- memset(&v, 0, sizeof(v));
- v.rec_inv_sqrt = ~0U;
- cobalt_rec_inv_sqrt_cache[0] = v.rec_inv_sqrt;
-
- for (v.count = 1; v.count < REC_INV_SQRT_CACHE; v.count++) {
- cobalt_newton_step(&v);
- cobalt_newton_step(&v);
- cobalt_newton_step(&v);
- cobalt_newton_step(&v);
-
- cobalt_rec_inv_sqrt_cache[v.count] = v.rec_inv_sqrt;
- }
-}
-
static void cobalt_vars_init(struct cobalt_vars *vars)
{
memset(vars, 0, sizeof(*vars));
-
- if (!cobalt_rec_inv_sqrt_cache[0]) {
- cobalt_cache_init();
- cobalt_rec_inv_sqrt_cache[0] = ~0;
- }
}
/* CoDel control_law is t + interval/sqrt(count)
--
2.34.1
_______________________________________________
Cake mailing list
Cake@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/cake
