On Tue, 20 Nov 2007, Ingo Molnar wrote: > * Steven Rostedt <rostedt@xxxxxxxxxxx> wrote: > > > With PVOPS on it gives compiler warnings without that explict cast. > > Without looking at the code, IIRC with non-PVOPS it is a macro > > directly into asm, so it didn't matter what the cast was. But with > > PVOPS as a function, it gave compiler warnings. > > > > Take it out and try compiling it for both i386 and x86_64. One of them > > gave warnings. But maybe it's not a problem now. > > i dont think there's ever any true need (and good cause) to force > integer type casts like that at the callee site. I guess the problem is that we converted a macro to a function, where the macro did no type checking. Now we need to pick between integers and pointers. Some places uses intergers in wrmsrl and some use pointers. So changing this to a typechecking protocol is not going to be nice. Looking at the current code now, we have this: ==== checking_wrmsrl(MSR_IA32_SYSENTER_CS, (u64)__KERNEL_CS); checking_wrmsrl(MSR_IA32_SYSENTER_ESP, 0ULL); checking_wrmsrl(MSR_IA32_SYSENTER_EIP, (u64)ia32_sysenter_target); wrmsrl(MSR_CSTAR, ia32_cstar_target); ==== A typecast is already used in that same area. -- Steve _______________________________________________ Virtualization mailing list Virtualization@xxxxxxxxxxxxxxxxxxxxxxxxxx https://lists.linux-foundation.org/mailman/listinfo/virtualization
