This patch fixes a minor bug that mcheck picked up. I don't know if
it ever causes problems in practice.
-Ken
diff -burN qemu-snapshot-2006-03-27_23.orig/console.c
qemu-snapshot-2006-03-27_23/console.c
--- qemu-snapshot-2006-03-27_23.orig/console.c 2006-03-11 07:35:30.0
-0800
+++ qe
In qemu-0.8.0.20060327, there are three problems with sending large
packets from guest to host:
(1) the code in slirp's ip_reass() reads a next pointer out an mbuf
after freeing it via m_cat().
(2) the code in slirp's m_inc() calls realloc() on a large mbuf, but
fails to adjust m_data to point
Well that is a good question. When I made this patch, I wanted to
respect the comment in the code that the author felt that using a
signed data type was safer in some cases. However, I will admit that
I do not understand this reasoning, and I agree that switching to an
unsigned data type would b
Leo, thank you for exercising this stuff.
> 1. before your patches, the upstream transfers (guest->host) consumed
> almost no CPU at all, but of course were much slower. Now, about half
> the CPU gets used under heavy upstream load.
I am surprised that only half the CPU gets consumed --- that su
e up to date. I'd be glad to help with any testing
> if/when you have patches.
>
> Thanks,
>
> Leo Reiter
>
> Kenneth Duda wrote:
> > Thanks, Leo. It appears your patch or something similar has made it
> > into 0.8.0. I have already merged the select loo
The "qemu-slirp-performance" patch contains three improvements to qemu
slirp networking performance. Booting my virtual machine (which
NFS-mounts its root filesystem from the host) has been accelerated by
8x, from over 5 minutes to 40 seconds. TCP throughput has been
accelerated from about 2 mega
Thanks, Leo. It appears your patch or something similar has made it
into 0.8.0. I have already merged the select loops, but it didn't
help as much as I hoped, maybe 10%. A much bigger improvement was
made by fixing the badly hacked slirp DELACK behavior. Believe it or
not, slirp delays all TCP
m vl.c:
#if defined(CONFIG_SLIRP)
/* XXX: merge with the previous select() */
if (slirp_inited) {
I will take a swing at this first. Please let me know if there's
anything I should be aware of.
Thanks,
-Ken
On 4/11/06, Paul Brook <[EMAIL PROTECTED]> wrote:
> On Tuesday 11
I am also having severe performance problems using NFS-over-TCP on
qemu-0.8 with a Linux host and guest. I will be looking at this
today. My current theory is that the whole machine is going idle
before qemu decides to poll kernel ring buffers holding packets the
guest is transmitting, but if any
Hi everyone, here is another patch for a much less significant bug. If
your "vc" console width is 0, qemu corrupts the heap (because it
writes one character into a screen buffer that's been malloc'ed as
size 0). I don't know if this bug ever causes problems in practice
--- I picked it up using mch
Hi everyone,
I have patches for a few bugs in qemu, and am new to the list --- if
anyone could clue me in on the best way to get patches applied to the
qemu mainline, that would be great.
This patch fixes three problems (actually all in slirp) with sending
large packets from guest to host in qemu
11 matches
Mail list logo
