> Ok, I managed to run the tests today and got surprised a bit.
> ...
> Then, if my tests aren't wrong and if I'm not missing anything important
> I say Duncan is right and we have no benefits of having these functions
> as functions. :)
It's not that surprising! Have a look at usb_endpoint_dir_in for example
(choosing this one is a bit of a cheat, since it's one of the smallest):
000000e0 <usb_endpoint_dir_in>:
e0: 0f be 40 02 movsbl 0x2(%eax),%eax
e4: c1 e8 1f shr $0x1f,%eax
That's 7 bytes, not including the return. Here's a call to it:
1705: 89 d0 mov %edx,%eax
1707: e8 fc ff ff ff call 1708 <speedtch_bind+0x4a8>
1708: R_386_PC32 usb_endpoint_dir_in
That's also 7 bytes! So no bytes are being saved. You might as well
just do the usb_endpoint_dir_in logic directly in place, i.e. inline
it.
Ciao,
Duncan.
-------------------------------------------------------------------------
Using Tomcat but need to do more? Need to support web services, security?
Get stuff done quickly with pre-integrated technology to make your job easier
Download IBM WebSphere Application Server v.1.0.1 based on Apache Geronimo
http://sel.as-us.falkag.net/sel?cmd=lnk&kid=120709&bid=263057&dat=121642
_______________________________________________
[email protected]
To unsubscribe, use the last form field at:
https://lists.sourceforge.net/lists/listinfo/linux-usb-devel
