Date: Wed, 28 Jun 2000 12:21:03 -0700 From: Wilbert de Graaf <wilbertdg@hetnet.nl> To: freebsd-hackers@freebsd.org Subject: Getting the plain void* through the generic ioctl processing Message-ID: <395A501F.50CA476E@hetnet.nl>
next in thread | raw e-mail | index | archive | help
--------------651756F39847F164F7E0CB33
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
Could anybody advise me on this:
For an implementation of IGMPv3, I had to extend the socket api
following the draft(s). I had to add two ioctl commands:
SIO_GET_MULTICAST_FILTER and SIO_SET_MULTICAST_FILTER. The argument is a
structure that has a variable size. FreeBSD uses some macros, and the
one I added looks like
#define SIO_SET_MULTICAST_FILTER _IOWR('s', 63, struct ip_msfilter*)
I couldn't use 'struct ip_msfilter' since this structure has a variable
size. Now this macro makes a unsigned long from all this, where the
length of the structure is encoded using sizeof(struct ip_msfilter*).
This makes the ioctl pre-processing (sys_generic.c) copy 4 bytes from
this pointer into the kernel. But this data is worthless to me, since it
is the first 4 bytes of the structure itself. What I want is the pointer
itself. I understand that I can do this by making people pass the
address of a pointer to ioctl(), but this is kind of ugly I think. And
besides, the other OS'es don't need this.
But the good news is that I found a way around, but the question is if
this is okay. If I follow the code, I can make the generic ioctl()
preprocessing giving me the datapointer itself when the length equals
zero. One way to make the length zero is to change the macro into
#define SIO_SET_MULTICAST_FILTER _IOWR('s', 63, struct {})
since sizeof(struct {}) returns zero. My question is, is the ioctl()
code in sys_generic.c meant to be able to handle cases where size = 0,
or is this a dirty 'trick' that probably won't make it in a next
version. Any reason not do solve it like this?
- Wilbert
--------------651756F39847F164F7E0CB33
Content-Type: text/html; charset=us-ascii
Content-Transfer-Encoding: 7bit
<!doctype html public "-//w3c//dtd html 4.0 transitional//en">
<html>
<font face="Courier New,Courier">Could anybody advise me on this:</font><font face="Courier New,Courier"></font>
<p><font face="Courier New,Courier">For an implementation of IGMPv3, I
had to extend the socket api</font>
<br><font face="Courier New,Courier">following the draft(s). I had to add
two ioctl commands:</font>
<br><font face="Courier New,Courier">SIO_GET_MULTICAST_FILTER and SIO_SET_MULTICAST_FILTER.
The argument is a</font>
<br><font face="Courier New,Courier">structure that has a variable size.
FreeBSD uses some macros, and the</font>
<br><font face="Courier New,Courier">one I added looks like</font><font face="Courier New,Courier"></font>
<p><font face="Courier New,Courier">#define SIO_SET_MULTICAST_FILTER
_IOWR('s', 63, struct ip_msfilter*)</font><font face="Courier New,Courier"></font>
<p><font face="Courier New,Courier">I couldn't use 'struct ip_msfilter'
since this structure has a variable</font>
<br><font face="Courier New,Courier">size. Now this macro makes a unsigned
long from all this, where the</font>
<br><font face="Courier New,Courier">length of the structure is encoded
using sizeof(struct ip_msfilter*).</font>
<br><font face="Courier New,Courier">This makes the ioctl pre-processing
(sys_generic.c) copy 4 bytes from</font>
<br><font face="Courier New,Courier">this pointer into the kernel. But
this data is worthless to me, since it</font>
<br><font face="Courier New,Courier">is the first 4 bytes of the structure
itself. What I want is the pointer</font>
<br><font face="Courier New,Courier">itself. I understand that I can do
this by making people pass the</font>
<br><font face="Courier New,Courier">address of a pointer to ioctl(), but
this is kind of ugly I think. And</font>
<br><font face="Courier New,Courier">besides, the other OS'es don't need
this.</font><font face="Courier New,Courier"></font>
<p><font face="Courier New,Courier">But the good news is that I found a
way around, but the question is if</font>
<br><font face="Courier New,Courier">this is okay. If I follow the code,
I can make the generic ioctl()</font>
<br><font face="Courier New,Courier">preprocessing giving me the datapointer
itself when the length equals</font>
<br><font face="Courier New,Courier">zero. One way to make the length zero
is to change the macro into</font><font face="Courier New,Courier"></font>
<p><font face="Courier New,Courier">#define SIO_SET_MULTICAST_FILTER
_IOWR('s', 63, struct {})</font><font face="Courier New,Courier"></font>
<p><font face="Courier New,Courier">since sizeof(struct {}) returns zero.
My question is, is the ioctl()</font>
<br><font face="Courier New,Courier">code in sys_generic.c meant to be
able to handle cases where size = 0,</font>
<br><font face="Courier New,Courier">or is this a dirty 'trick' that probably
won't make it in a next</font>
<br><font face="Courier New,Courier">version. Any reason not do solve it
like this?</font><font face="Courier New,Courier"></font>
<p><font face="Courier New,Courier">- Wilbert</font>
<br><font face="Courier New,Courier"></font>
<br> </html>
--------------651756F39847F164F7E0CB33--
To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?395A501F.50CA476E>