Re: [Qemu-devel] [PATCH v3] slirp: Handle whole 127.0.0.0/8 network as local addresses.

[Jan Kiszka <jan.kiszka@web.de>]

[-- Attachment #1: Type: text/plain, Size: 3225 bytes --]

On 2012-07-14 10:00, Jan Kiszka wrote:
> On 2012-07-13 22:54, Anders Waldenborg wrote:
>> Changes so translation of remote address to the host's ip address in
>> the virtual network happens for all addresses in the 127.0.0.0/8
>> network, not just 127.0.0.1.
>>
>> This fixes so that hostfwd bound to addresses such as 127.0.0.2 works.
>>
>> Signed-off-by: Anders Waldenborg <anders@0x63.nu>
>> ---
>>
>> On Fri, Jul 13, 2012 at 03:01:14PM +0000, Blue Swirl wrote:
>>>> Notice that the surrounding code in tcp_subr.c uses tabs for
>>>> indentation. Should I still use space as the coding style mandates
>>>> (and which makes checkpatch happy)?
>>>
>>> Yes, the goal is to get rid of tabs.
>>
>> OK. Thanks for clarifacion and review!
>>
>> Patch updated.
>>
>>  slirp/main.h     |    1 +
>>  slirp/slirp.c    |    3 +++
>>  slirp/tcp_subr.c |    7 +++++--
>>  3 files changed, 9 insertions(+), 2 deletions(-)
>>
>> diff --git a/slirp/main.h b/slirp/main.h
>> index 028df4b..bf601e2 100644
>> --- a/slirp/main.h
>> +++ b/slirp/main.h
>> @@ -31,6 +31,7 @@ extern char *exec_shell;
>>  extern u_int curtime;
>>  extern fd_set *global_readfds, *global_writefds, *global_xfds;
>>  extern struct in_addr loopback_addr;
>> +extern in_addr_t loopback_mask;
>>  extern char *username;
>>  extern char *socket_path;
>>  extern int towrite_max;
>> diff --git a/slirp/slirp.c b/slirp/slirp.c
>> index 90473eb..9787104 100644
>> --- a/slirp/slirp.c
>> +++ b/slirp/slirp.c
>> @@ -29,6 +29,8 @@
>>  
>>  /* host loopback address */
>>  struct in_addr loopback_addr;
>> +/* host loopback network mask */
>> +in_addr_t loopback_mask;
>>  
>>  /* emulated hosts use the MAC addr 52:55:IP:IP:IP:IP */
>>  static const uint8_t special_ethaddr[ETH_ALEN] = {
>> @@ -191,6 +193,7 @@ static void slirp_init_once(void)
>>  #endif
>>  
>>      loopback_addr.s_addr = htonl(INADDR_LOOPBACK);
>> +    loopback_mask = htonl(IN_CLASSA_NET);
>>  }
>>  
>>  static void slirp_state_save(QEMUFile *f, void *opaque);
>> diff --git a/slirp/tcp_subr.c b/slirp/tcp_subr.c
>> index 0a545c4..6b01f74 100644
>> --- a/slirp/tcp_subr.c
>> +++ b/slirp/tcp_subr.c
>> @@ -435,8 +435,11 @@ tcp_connect(struct socket *inso)
>>  	so->so_fport = addr.sin_port;
>>  	so->so_faddr = addr.sin_addr;
>>  	/* Translate connections from localhost to the real hostname */
>> -	if (so->so_faddr.s_addr == 0 || so->so_faddr.s_addr == loopback_addr.s_addr)
>> -	   so->so_faddr = slirp->vhost_addr;
>> +    if (so->so_faddr.s_addr == 0 ||
>> +        (so->so_faddr.s_addr & loopback_mask) ==
>> +        (loopback_addr.s_addr & loopback_mask)) {
>> +        so->so_faddr = slirp->vhost_addr;
>> +    }
> 
> Dropping tabs is desired, but as long as slirp is such a mess coding
> style wise, I prefer keeping the indention depth consistent. I fixed
> this up while merging it.
> 
> Out of curiosity: Which host platform requires this? Linux always
> reports 127.0.0.1 as source, even when you bind the hostfwd rule to a
> different localhost address.

Ah, found the test case: binding the host-side initiator of the
connection to some other localhost address provides a before-after effect.

Jan


[-- Attachment #2: OpenPGP digital signature --]
[-- Type: application/pgp-signature, Size: 262 bytes --]