mirror of
https://github.com/openwrt/openwrt.git
synced 2024-12-21 06:33:41 +00:00
f138833d0c
x86_64 platforms typically don't lack memory, so don't needlessly economize memory if fq_codel on capable platforms. Signed-off-by: Philip Prindeville <philipp@redfish-solutions.com> [Add a comment to the patch] Signed-off-by: Hauke Mehrtens <hauke@hauke-m.de>
28 lines
879 B
Diff
28 lines
879 B
Diff
From a6ccb238939b25851474a279b20367fd24a0e816 Mon Sep 17 00:00:00 2001
|
|
From: Felix Fietkau <nbd@nbd.name>
|
|
Date: Fri, 7 Jul 2017 17:21:53 +0200
|
|
Subject: hack: net: fq_codel: tune defaults for small devices
|
|
|
|
Assume that x86_64 devices always have a big memory and do not need this
|
|
optimization compared to devices with only 32 MB or 64 MB RAM.
|
|
|
|
Signed-off-by: Felix Fietkau <nbd@nbd.name>
|
|
---
|
|
net/sched/sch_fq_codel.c | 2 +-
|
|
1 file changed, 1 insertion(+), 1 deletion(-)
|
|
|
|
--- a/net/sched/sch_fq_codel.c
|
|
+++ b/net/sched/sch_fq_codel.c
|
|
@@ -479,7 +479,11 @@ static int fq_codel_init(struct Qdisc *s
|
|
|
|
sch->limit = 10*1024;
|
|
q->flows_cnt = 1024;
|
|
+#ifdef CONFIG_X86_64
|
|
q->memory_limit = 32 << 20; /* 32 MBytes */
|
|
+#else
|
|
+ q->memory_limit = 4 << 20; /* 4 MBytes */
|
|
+#endif
|
|
q->drop_batch_size = 64;
|
|
q->quantum = psched_mtu(qdisc_dev(sch));
|
|
q->perturbation = prandom_u32();
|