[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <YTQB5jCbvhmCWzNd@codewreck.org>
Date: Sun, 5 Sep 2021 08:31:50 +0900
From: Dominique Martinet <asmadeus@...ewreck.org>
To: Christian Schoenebeck <linux_oss@...debyte.com>
Cc: v9fs-developer@...ts.sourceforge.net, netdev@...r.kernel.org,
Eric Van Hensbergen <ericvh@...il.com>,
Latchesar Ionkov <lucho@...kov.net>, Greg Kurz <groug@...d.org>
Subject: Re: [PATCH 2/2] net/9p: increase default msize to 128k
Hi Christian,
thanks for the follow up, this has been on my todolist forever and
despite how simple it is I never took the time for it...
I've applied both patches to my -next branch... We're a bit tight for
this merge window (v5.15) but I'll send it to linux mid next week anyway.
I've also added this patch (sorry for laziness) for TCP, other transports
are OK iirc:
>From 657e35583c70bed86526cb8eb207abe3d55ea4ea Mon Sep 17 00:00:00 2001
From: Dominique Martinet <asmadeus@...ewreck.org>
Date: Sun, 5 Sep 2021 08:29:22 +0900
Subject: [PATCH] net/9p: increase tcp max msize to 1MB
Historically TCP has been limited to 64K buffers, but increasing msize provides
huge performance benefits especially as latency increase so allow for bigger buffers.
Ideally further improvements could change the allocation from the current contiguous chunk
in slab (kmem_cache) to some scatter-gather compatible API...
Signed-off-by: Dominique Martinet <asmadeus@...ewreck.org>
diff --git a/net/9p/trans_fd.c b/net/9p/trans_fd.c
index f4dd0456beaf..007bbcc68010 100644
--- a/net/9p/trans_fd.c
+++ b/net/9p/trans_fd.c
@@ -34,7 +34,7 @@
#include <linux/syscalls.h> /* killme */
#define P9_PORT 564
-#define MAX_SOCK_BUF (64*1024)
+#define MAX_SOCK_BUF (1024*1024)
#define MAXPOLLWADDR 2
static struct p9_trans_module p9_tcp_trans;
--
Dominique
Powered by blists - more mailing lists