Heho,
I gave v3 a shot and it performs as expected; For a requested MSS of 1320, TSO
consistently uses a 1308 MSS. So for me, this patch works. Thanks for fixing
this. :-)
Sadly, I do not have boxes to test with .1q around; If none of you has either,
and that should be tested as well, I can giv
Heho,
Thanks, will test the three patches later.
With best regards,
Tobias
-Original Message-
From: Stefan Hajnoczi
Sent: Thursday, 17 November 2022 17:57
To: Tobias Fiebig
Cc: Jason Wang ; Stefan Hajnoczi ;
qemu-devel@nongnu.org; qemu-sta...@nongnu.org; Russell King - ARM Linux
, to test how the real hardware
behaved?
With best regards,
Tobias
-Original Message-
From: Stefan Hajnoczi
Sent: Thursday, 17 November 2022 12:16
To: Tobias Fiebig
Cc: Jason Wang ; Stefan Hajnoczi ;
qemu-devel@nongnu.org; qemu-sta...@nongnu.org; Russell King - ARM Linux
Subject: Re
Heho,
Ok, I just learned more C than I ever wanted to. There is a bit more amiss here
(ll from 7d7238c72b983cff5064734349d2d45be9c6282c):
In line 1916 of rtl8139.c we set txdw0; If we calculate the MSS at this point,
it is consistently 12 below requested, but generally accurate. The bits that
f
Heho,
Quick follow-up; Applied the change you suggested, but there are still some
things to test.
While this now works (mostly), MSS values are still off; Especially the
behavior below <=1036 is difficult, as for v4 the minimum MTU is 576 and
minimum MSS is 536:
RequestedDPRINT
1320
Heho,
> Ok, I think I found at least one issue:
>
> /* large send MSS mask, bits 16...25 */
> #define CP_TC_LGSEN_MSS_MASK ((1 << 12) - 1)
>
> First, MSS occupies 11 bits from 16 to 26 Second, the mask is wrong it should
> be ((1 << 11) - 1)
Awesome, thanks, will give this a shot later on and le
Heho,
I just tested around with the patch;
Good news: Certainly my builds are being executed. Also, if I patch the old
code to have a MAX_MTU <= the max MTU on my path, throughput is ok.
Bad news: Something is wrong with getting the MSS in the patch you shared. When
enabling DPRINT, values are o
doesn't show an effect, I'll hold tcpdump to the wire again.
With best regards,
Tobias
-Original Message-
From: Stefan Hajnoczi
Sent: Tuesday, 15 November 2022 17:37
To: qemu-devel@nongnu.org
Cc: jasow...@redhat.com; qemu-sta...@nongnu.org; Stefan Hajnoczi
; Russell King - A
