Re: [Qemu-devel] [PATCH RFC] Do not set SO_REUSEADDR on Windows

[Michael Tokarev <mjt@tls.msk.ru>]

04.09.2013 18:34, Jan Kiszka wrote:
> On 2013-09-04 16:27, Paolo Bonzini wrote:
>> Il 04/09/2013 16:22, Sebastian Ottlik ha scritto:
>>> This patchset disabels all use of SO_REUSEADDR on Windows. On Windows systems
>>> the default behavior is equivalent to SO_REUSEADDR on other operating
>>> systems. SO_REUSEADDR can still be set but results in undesired bahvior
>>> instead. It may even lead to situations were system behavior is
>>> unspecified. More information on this can be found at:
>>> http://msdn.microsoft.com/en-us/library/windows/desktop/ms740621.aspx
>>>
>>> I originally encountered this issue when accidentally launching two QEMU
>>> instances with identical GDB ports at the same time. In which case QEMU won't
>>> fail as one might expect. I am sending this as RFC as I A) only checked that
>>> this fixes issues for the GDB server and B) am not sure if this is the correct
>>> format for this patchset.
>>>
>>> gdbstub: do not set SO_REUSEADDR on Windows
>>> net: do not set SO_REUSEADDR on Windows
>>> slirp: do not set SO_REUSEADDR on Windows
>>> util: do not set SO_REUSEADDR on Windows
>>
>> Makes sense.
>>
>> Can you make a different patch that introduces a new function
>> qemu_set_reuseaddr is include/qemu/sockets.h & util/oslib-*, and makes
>> it a stub for Windows?

Maybe a better approach will be to introduce something like
qemu_[stream_]listen_socket(int family, sockaddr_t *sa, int salen) function
which opens and prepares a socket with all necessary platform-dependent
stuff?

> Yeah, this is definitely better then. IIRC, Michael has some version of
> patch 3 in his queue right now - should be dropped in this light.

It's been pulled by Anthony yesterday so it is a bit too late now.
But the change is trivial, and can be reworked easily.

Thanks,

/mjt