Hi On 3/25/08, Carlo Bertoldi <mcbain@xxxxxxxxxx> wrote: > Well, I may be wrong, sot I'll try to be more specific: what I' trying > to do is to monitor the sequence of hypercall (more or less Xen > equivalent of syscall) that the kernel issues. So I don't know when the > first hypercall is issued, and I don't think it's really efficient to > check if the needed symbol is available every time, since the kernel can > issues thousands of hcall per seconds. Hope I explained myself. Ehm, then maybe it's not supposed to be in the form of module? maybe it's better made as additional function patched into the kernel? This way you have advantages: 1. no overhead of symbol resolve and function indirect call 2. you can detect the hypercall event the kernel in the stage where it's not ready to accept modules (well, AFAIK, hypercalls might do something like preparing initial guess IDT/GDT, right?) regards, Mulyadi. -- To unsubscribe from this list: send an email with "unsubscribe kernelnewbies" to ecartis@xxxxxxxxxxxx Please read the FAQ at http://kernelnewbies.org/FAQ
