* [Cake] RHODIUM - nuking blue (for testing)
@ 2017-12-05 19:32 Dave Taht
2017-12-05 22:25 ` Georgios Amanakis
2017-12-06 8:14 ` [Cake] [OT] " Sebastian Moeller
0 siblings, 2 replies; 3+ messages in thread
From: Dave Taht @ 2017-12-05 19:32 UTC (permalink / raw)
To: Cake List
[-- Attachment #1: Type: text/plain, Size: 656 bytes --]
Given some of the results thus far at various RTTs, and at higher
loads, like the ginormous one georgios just ran, I thought it might be
useful to run a battery of tests with blue disabled.
I haven't gone to the trouble of creating a formal branch - tho if I
did, I'd call it rhodium, as rhodium detectors are used in nuclear
reactors to measure the neutron flux level.
untested-but-compiling-patch attached. I'm busy on a few other things
this week, and it is looking like we'll have to replace the xstats
structure with something smaller to go mainline.
--
Dave Täht
CEO, TekLibre, LLC
http://www.teklibre.com
Tel: 1-669-226-2619
[-- Attachment #2: 0001-RHODIUM-disable-blue-algorithm-for-testing.patch --]
[-- Type: text/x-patch, Size: 2614 bytes --]
From 56a55081199ed9ee36de8a2806a7c51d5a83a078 Mon Sep 17 00:00:00 2001
From: Dave Taht <dave.taht@gmail.com>
Date: Tue, 5 Dec 2017 11:22:08 -0800
Subject: [PATCH] RHODIUM: disable blue algorithm for testing
This patch disables the blue algorithm so the aqm becomes more
pure codel.
---
cobalt.h | 2 ++
sch_cake.c | 13 ++++++++++++-
2 files changed, 14 insertions(+), 1 deletion(-)
diff --git a/cobalt.h b/cobalt.h
index f10f9db..c1d9dff 100644
--- a/cobalt.h
+++ b/cobalt.h
@@ -126,8 +126,10 @@ struct cobalt_vars {
u32 count;
u32 rec_inv_sqrt;
cobalt_time_t drop_next;
+#ifndef RHODIUM
cobalt_time_t blue_timer;
u32 p_drop;
+#endif
bool dropping;
bool ecn_marked;
};
diff --git a/sch_cake.c b/sch_cake.c
index 4010388..134c6ca 100644
--- a/sch_cake.c
+++ b/sch_cake.c
@@ -62,6 +62,7 @@
#else
#include <net/flow_dissector.h>
#endif
+#define RHODIUM 1
#include "cobalt.h"
#if IS_ENABLED(CONFIG_NF_CONNTRACK)
@@ -443,6 +444,7 @@ static bool cobalt_queue_full(struct cobalt_vars *vars,
{
bool up = false;
+#ifndef RHODIUM
if ((now - vars->blue_timer) > p->target) {
up = !vars->p_drop;
vars->p_drop += p->p_inc;
@@ -450,6 +452,7 @@ static bool cobalt_queue_full(struct cobalt_vars *vars,
vars->p_drop = ~0;
vars->blue_timer = now;
}
+#endif
vars->dropping = true;
vars->drop_next = now;
if (!vars->count)
@@ -467,6 +470,7 @@ static bool cobalt_queue_empty(struct cobalt_vars *vars,
{
bool down = false;
+#ifndef RHODIUM
if (vars->p_drop && (now - vars->blue_timer) > p->target) {
if (vars->p_drop < p->p_dec)
vars->p_drop = 0;
@@ -475,6 +479,7 @@ static bool cobalt_queue_empty(struct cobalt_vars *vars,
vars->blue_timer = now;
down = !vars->p_drop;
}
+#endif
vars->dropping = false;
if (vars->count && (now - vars->drop_next) >= 0) {
@@ -559,10 +564,11 @@ static bool cobalt_should_drop(struct cobalt_vars *vars,
}
}
+#ifndef RHODIUM
/* Simple BLUE implementation. Lack of ECN is deliberate. */
if (vars->p_drop)
drop |= (prandom_u32() < vars->p_drop);
-
+#endif
/* Overload the drop_next field as an activity timeout */
if (!vars->count)
vars->drop_next = now + p->interval;
@@ -1818,8 +1824,13 @@ retry:
if (cobalt_queue_empty(&flow->cvars, &b->cparams, now))
b->unresponsive_flow_count--;
+#ifndef RHODIUM
if (flow->cvars.p_drop || flow->cvars.count ||
(now - flow->cvars.drop_next) < 0) {
+#else
+ if (flow->cvars.count ||
+ (now - flow->cvars.drop_next) < 0) {
+#endif
/* keep in the flowchain until the state has
* decayed to rest
*/
--
2.7.4
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [Cake] RHODIUM - nuking blue (for testing)
2017-12-05 19:32 [Cake] RHODIUM - nuking blue (for testing) Dave Taht
@ 2017-12-05 22:25 ` Georgios Amanakis
2017-12-06 8:14 ` [Cake] [OT] " Sebastian Moeller
1 sibling, 0 replies; 3+ messages in thread
From: Georgios Amanakis @ 2017-12-05 22:25 UTC (permalink / raw)
To: Dave Taht; +Cc: Cake List
[-- Attachment #1: Type: text/plain, Size: 1004 bytes --]
Great! I am repeating the 200mbit/10mbit 32/8 flows, RTT 20ms scenario with
the patch.
On Dec 5, 2017 2:32 PM, "Dave Taht" <dave.taht@gmail.com> wrote:
> Given some of the results thus far at various RTTs, and at higher
> loads, like the ginormous one georgios just ran, I thought it might be
> useful to run a battery of tests with blue disabled.
>
> I haven't gone to the trouble of creating a formal branch - tho if I
> did, I'd call it rhodium, as rhodium detectors are used in nuclear
> reactors to measure the neutron flux level.
>
> untested-but-compiling-patch attached. I'm busy on a few other things
> this week, and it is looking like we'll have to replace the xstats
> structure with something smaller to go mainline.
>
> --
>
> Dave Täht
> CEO, TekLibre, LLC
> http://www.teklibre.com
> Tel: 1-669-226-2619
>
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake
>
>
[-- Attachment #2: Type: text/html, Size: 1617 bytes --]
^ permalink raw reply [flat|nested] 3+ messages in thread
* [Cake] [OT] Re: RHODIUM - nuking blue (for testing)
2017-12-05 19:32 [Cake] RHODIUM - nuking blue (for testing) Dave Taht
2017-12-05 22:25 ` Georgios Amanakis
@ 2017-12-06 8:14 ` Sebastian Moeller
1 sibling, 0 replies; 3+ messages in thread
From: Sebastian Moeller @ 2017-12-06 8:14 UTC (permalink / raw)
To: Dave Täht; +Cc: Cake List
[-- Attachment #1: Type: text/plain, Size: 911 bytes --]
Hi Dave
> On Dec 5, 2017, at 20:32, Dave Taht <dave.taht@gmail.com> wrote:
>
> Given some of the results thus far at various RTTs, and at higher
> loads, like the ginormous one georgios just ran, I thought it might be
> useful to run a battery of tests with blue disabled.
>
> I haven't gone to the trouble of creating a formal branch - tho if I
> did, I'd call it rhodium, as rhodium detectors are used in nuclear
> reactors to measure the neutron flux level.
And the name Rhodium comes from the greek rhodeos meaning red like a rose; in other words "not blue" ;)
Best Regards
Sebastian
>
> untested-but-compiling-patch attached. I'm busy on a few other things
> this week, and it is looking like we'll have to replace the xstats
> structure with something smaller to go mainline.
>
> --
>
> Dave Täht
> CEO, TekLibre, LLC
> http://www.teklibre.com
> Tel: 1-669-226-2619
[-- Warning: decoded text below may be mangled, UTF-8 assumed --]
[-- Attachment #2: 0001-RHODIUM-disable-blue-algorithm-for-testing.patch --]
[-- Type: text/x-patch; x-unix-mode=0644; name="0001-RHODIUM-disable-blue-algorithm-for-testing.patch", Size: 2715 bytes --]
From 56a55081199ed9ee36de8a2806a7c51d5a83a078 Mon Sep 17 00:00:00 2001
From: Dave Taht <dave.taht@gmail.com>
Date: Tue, 5 Dec 2017 11:22:08 -0800
Subject: [PATCH] RHODIUM: disable blue algorithm for testing
This patch disables the blue algorithm so the aqm becomes more
pure codel.
---
cobalt.h | 2 ++
sch_cake.c | 13 ++++++++++++-
2 files changed, 14 insertions(+), 1 deletion(-)
diff --git a/cobalt.h b/cobalt.h
index f10f9db..c1d9dff 100644
--- a/cobalt.h
+++ b/cobalt.h
@@ -126,8 +126,10 @@ struct cobalt_vars {
u32 count;
u32 rec_inv_sqrt;
cobalt_time_t drop_next;
+#ifndef RHODIUM
cobalt_time_t blue_timer;
u32 p_drop;
+#endif
bool dropping;
bool ecn_marked;
};
diff --git a/sch_cake.c b/sch_cake.c
index 4010388..134c6ca 100644
--- a/sch_cake.c
+++ b/sch_cake.c
@@ -62,6 +62,7 @@
#else
#include <net/flow_dissector.h>
#endif
+#define RHODIUM 1
#include "cobalt.h"
#if IS_ENABLED(CONFIG_NF_CONNTRACK)
@@ -443,6 +444,7 @@ static bool cobalt_queue_full(struct cobalt_vars *vars,
{
bool up = false;
+#ifndef RHODIUM
if ((now - vars->blue_timer) > p->target) {
up = !vars->p_drop;
vars->p_drop += p->p_inc;
@@ -450,6 +452,7 @@ static bool cobalt_queue_full(struct cobalt_vars *vars,
vars->p_drop = ~0;
vars->blue_timer = now;
}
+#endif
vars->dropping = true;
vars->drop_next = now;
if (!vars->count)
@@ -467,6 +470,7 @@ static bool cobalt_queue_empty(struct cobalt_vars *vars,
{
bool down = false;
+#ifndef RHODIUM
if (vars->p_drop && (now - vars->blue_timer) > p->target) {
if (vars->p_drop < p->p_dec)
vars->p_drop = 0;
@@ -475,6 +479,7 @@ static bool cobalt_queue_empty(struct cobalt_vars *vars,
vars->blue_timer = now;
down = !vars->p_drop;
}
+#endif
vars->dropping = false;
if (vars->count && (now - vars->drop_next) >= 0) {
@@ -559,10 +564,11 @@ static bool cobalt_should_drop(struct cobalt_vars *vars,
}
}
+#ifndef RHODIUM
/* Simple BLUE implementation. Lack of ECN is deliberate. */
if (vars->p_drop)
drop |= (prandom_u32() < vars->p_drop);
-
+#endif
/* Overload the drop_next field as an activity timeout */
if (!vars->count)
vars->drop_next = now + p->interval;
@@ -1818,8 +1824,13 @@ retry:
if (cobalt_queue_empty(&flow->cvars, &b->cparams, now))
b->unresponsive_flow_count--;
+#ifndef RHODIUM
if (flow->cvars.p_drop || flow->cvars.count ||
(now - flow->cvars.drop_next) < 0) {
+#else
+ if (flow->cvars.count ||
+ (now - flow->cvars.drop_next) < 0) {
+#endif
/* keep in the flowchain until the state has
* decayed to rest
*/
--
2.7.4
[-- Attachment #3: Type: text/plain, Size: 146 bytes --]
> _______________________________________________
> Cake mailing list
> Cake@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/cake
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2017-12-06 8:14 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2017-12-05 19:32 [Cake] RHODIUM - nuking blue (for testing) Dave Taht
2017-12-05 22:25 ` Georgios Amanakis
2017-12-06 8:14 ` [Cake] [OT] " Sebastian Moeller
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox