On 17.01.2022 12:41, Andrew Cooper wrote: > On 17/01/2022 11:22, Jan Beulich wrote: >> On 13.01.2022 17:38, Andrew Cooper wrote: >>> --- a/xen/arch/x86/hvm/svm/entry.S >>> +++ b/xen/arch/x86/hvm/svm/entry.S >>> @@ -59,7 +59,7 @@ __UNLIKELY_END(nsvm_hap) >>> mov VCPUMSR_spec_ctrl_raw(%rax), %eax >>> >>> /* WARNING! `ret`, `call *`, `jmp *` not safe beyond this point. */ >>> - SPEC_CTRL_EXIT_TO_HVM /* Req: a=spec_ctrl %rsp=regs/cpuinfo, >>> Clob: cd */ >>> + /* SPEC_CTRL_EXIT_TO_SVM (nothing currently) */ >>> >>> pop %r15 >>> pop %r14 >>> @@ -86,7 +86,8 @@ __UNLIKELY_END(nsvm_hap) >>> >>> GET_CURRENT(bx) >>> >>> - SPEC_CTRL_ENTRY_FROM_HVM /* Req: b=curr %rsp=regs/cpuinfo, >>> Clob: acd */ >>> + /* SPEC_CTRL_ENTRY_FROM_SVM Req: b=curr %rsp=regs/cpuinfo, >>> Clob: ac */ >>> + ALTERNATIVE "", DO_OVERWRITE_RSB, X86_FEATURE_SC_RSB_HVM >> Just for my own understanding: The comments you add aren't commented >> out macro invocations (as I did read it first), but comments naming >> would-be-macros which are then expanded "manually"? That is partly >> because initially I read the description saying "Expand and drop the >> macros" as meaning that the macros grow in what they contain, which >> looked contradictory to them getting dropped at the same time. >> Perhaps me not sufficiently understanding the difference between all >> possible meanings of "expand" vs "extend" ... > > They're grep fodder to be able to easily locate where we're doing > entry/exit speculation logic.
And I don't suppose I could talk you into replacing the use of "expand" in the description, even if it was just me who was mislead? Perhaps by "Open-code and drop the macros"? > These paths will continue to diverge over > time, and cannot have a common implementation. That's understood. Jan
