Re: [Qemu-devel] Is anyone able to load a web page from a guest operating system?

[Programmingkid]

On Apr 27, 2016, at 2:34 AM, Thomas Huth wrote:

> On 26.04.2016 22:19, Programmingkid wrote:
>> 
>> On Apr 26, 2016, at 4:12 PM, Thomas Huth wrote:
>> 
>>> On 26.04.2016 21:25, Programmingkid wrote:
>>>> 
>>>> On Apr 26, 2016, at 3:00 PM, Dr. David Alan Gilbert wrote:
>>>> 
>>>>> * Programmingkid (address@hidden) wrote:
>>>>>> My three guest operating systems can't load a web page. I think this is 
>>>>>> a bug with QEMU. Is there anyone who has the latest revision of QEMU 
>>>>>> that can access the web from a guest? Or are you experiencing the same 
>>>>>> problem?
> ...
>>>>> What's the qemu command line you use?
>>>> qemu-system-ppc -hda <hd1> -hdb <hd2> -m 512 -boot c -M mac99 -netdev 
>>>> user,id=mynet0 -device usb-net,netdev=mynet0 -cpu 750 -prom-env 
>>>> boot-args=-v -device ich9-usb-uhci1,id=newusb -device 
>>>> usb-audio,bus=newusb.0 
>>>> 
>>>> and
>>>> 
>>>> qemu-system-ppc -hda <hd1> -hdb <hd2> -m 512 -boot c -M mac99 -netdev 
>>>> user,id=mynet0 -device rtl8139,netdev=mynet0 -cpu 750 -prom-env 
>>>> boot-args=-v -device ich9-usb-uhci1,id=newusb -device 
>>>> usb-audio,bus=newusb.0 
>>> 
>>> Ok, that means you're using user-mode / slirp networking.
>>> I just tried it with a pseries guest, and it seems to be working fine
>>> for me with the current git version of QEMU (f419a626c76bcb266).
>> 
>> So you are saying you can view web pages on your guest?
> 
> Yes. Linux guest on a Linux host, and I was able to access a web page
> from the guest.
> 
>>> 
>>> Now, what kind of host do you use? Mac OS X? 
>> Yes. Mac OS 10.6.
> 
> Ok, at least not Windows ... because there have been some problems with
> Windows recently, which could be the culprit otherwise:
> 
> http://git.qemu.org/?p=qemu.git;a=commitdiff;h=3424c8a9c89a3bc0d29ad
> 
>>> Also can you determine a revision when it was still working fine for
>>> you? (and then maybe even bisect the problem?)
>> 
>> I will see what I can find out.
> 
> Thanks! Bisecting the problem is likely the best way to deal with this...

Found out which patch was causing problems. This one: 

commit 5379229a2708df3a1506113315214c3ce5325859
Author: Guillaume Subiron <address@hidden>
Date:   Sat Dec 19 22:24:59 2015 +0100

    slirp: Factorizing address translation
    
    This patch factorizes some duplicate code into a new function,
    sotranslate_out(). This function perform the address translation when a
    packet is transmitted to the host network. If the packet is destinated
    to the host, the loopback address is used, and if the packet is
    destinated to the virtual DNS, the real DNS address is used. This code
    is just a copy of the existent, but factorized and ready to manage the
    IPv6 case.