Time::HiRes::stat() cleanup and optimize fake OP struct construction

-XPUSHs() requires saving the SV* retval of sv_2mortal(newSVsv()) around
 a possible Perl_stack_grow(), split the EXTEND from the PUSH, so SV*
 is held only in volatile registers (liveness).
-over EXTEND to 13 elements instead of 1 element.
 Why not? pp_stat()/pp_lstat() have to do the Perl_stack_grow() call if we
 don't do it.
-remove Zero() macro and use a function call free struct initializer.
 Just b/c GCC and its offshoot Clang will inline a fixed length memset()
 doesn't make it part of ISO C. MSVC compiler never inlines memset() calls
 on WinPerl (b/c P5P never added the magic sauce to ask for that feature).
 More portably, P5P has never verified the machine code output of all
 known commercial Unix CCs on all CPU archs regarding inlining memset().
-when filling out the fake OP, do some instruction level parallelism like
 filling in fakeop with 0s, while digging through my_perl->Iop->op_flags,
 my_perl->Icurstackinfo->si_cxsubix, my_perl->Icurstackinfo->si_cxstack,
 and etc, as part of GIMME_V macro, which used to be a libperl.so exported
 function call a very long time ago IIRC.

 Another example, translate ix?OP_LSTAT:OP_STAT while translating
 gm==G_LIST?OPf_WANT_LIST:gm==G_SCALAR?OPf_WANT_SCALAR:OPf_WANT_VOID.

 Dig through PLT/GOT/PE sym table as part of PL_ppaddr[op_type]
 while writing to C stk mem as part of fakeop.op_type = op_type

-change fakeop.op_ppaddr(aTHX); to ppaddr(aTHX); b/c some CCs have a low
 IQ and can't prove statement "PL_op = &fakeop;" won't modify field
 fakeop.op_ppaddr in our C auto storage OP struct var.
-don't execute the Perl_sv_2uv_flags() getter method pointlessly inside
 UV atime = SvUV(ST( 8)); if static function hrstatns() is a NOOP and
 is inlined and totally optimized away since in some build configs,
 hrstatns() only does atime_nsec = 0; mtime_nsec = 0; ctime_nsec = 0;
 Windows is an example.
-change SvUV(ST( 8)); to SvUV(SPBASE[ 8]); don't deref my_perl->Istack_base
 over and over

Author: bulk88

dist/Time-HiRes/HiRes.xs

@@ -1515,39 +1515,56 @@ void
 stat(...)
 PROTOTYPE: ;$
     PREINIT:
-        OP fakeop;
-        int nret;
+        SSize_t nret;
+        SV* sv_arg;
+        SV** SPBASE;
     ALIAS:
         Time::HiRes::lstat = 1
     PPCODE:
-        XPUSHs(sv_2mortal(newSVsv(items == 1 ? ST(0) : DEFSV)));
+        sv_arg = items == 1 ? ST(0) : DEFSV;
+        EXTEND(SP, 13);
+        /* XXX will pp_stat()/pp_lstat() really modify $_[0] ? */
+        PUSHs(sv_2mortal(newSVsv(sv_arg)));
         PUTBACK;
         ENTER;
         PL_laststatval = -1;
         SAVEOP();
-        Zero(&fakeop, 1, OP);
-        fakeop.op_type = ix ? OP_LSTAT : OP_STAT;
-        fakeop.op_ppaddr = PL_ppaddr[fakeop.op_type];
-        fakeop.op_flags = GIMME_V == G_LIST ? OPf_WANT_LIST :
-            GIMME_V == G_SCALAR ? OPf_WANT_SCALAR : OPf_WANT_VOID;
-        PL_op = &fakeop;
-        (void)fakeop.op_ppaddr(aTHX);
-        SPAGAIN;
+        {
+            OP* (*ppaddr)(pTHX);
+            U8 gimme = GIMME_V; /* ILP */
+/* extern "C" memset() doesn't know struct OP's alignment. ISO C doesn't
+   promise Zero(); and memset(); will inline.  But this does. Now the CC can
+   detangle for us, what OP fields will get a 0/NULL, or our values. */
+            OP fakeop = {0};
+            U16 op_type = ix ? OP_LSTAT : OP_STAT;
+            fakeop.op_flags = gimme == G_LIST ? OPf_WANT_LIST :
+                gimme == G_SCALAR ? OPf_WANT_SCALAR : OPf_WANT_VOID; /* ILP */
+            ppaddr = PL_ppaddr[op_type];
+            fakeop.op_type = op_type;
+            fakeop.op_ppaddr = ppaddr; /* ILP */
+            PL_op = &fakeop;
+            (void)ppaddr(aTHX);
+        }
         LEAVE;
-        nret = SP+1 - &ST(0);
+        SPAGAIN;
+        SPBASE = &ST(0);
+        nret = SP+1 - SPBASE;
         if (nret == 13) {
-            UV atime = SvUV(ST( 8));
-            UV mtime = SvUV(ST( 9));
-            UV ctime = SvUV(ST(10));
             UV atime_nsec;
             UV mtime_nsec;
             UV ctime_nsec;
             hrstatns(&atime_nsec, &mtime_nsec, &ctime_nsec);
-            if (atime_nsec)
-                ST( 8) = sv_2mortal(newSVnv(atime + (NV) atime_nsec / NV_1E9));
-            if (mtime_nsec)
-                ST( 9) = sv_2mortal(newSVnv(mtime + (NV) mtime_nsec / NV_1E9));
-            if (ctime_nsec)
-                ST(10) = sv_2mortal(newSVnv(ctime + (NV) ctime_nsec / NV_1E9));
+            if (atime_nsec) { /* on certain configs hrstatns() is a NOOP */
+                UV atime = SvUV(SPBASE[ 8]);
+                SPBASE[ 8] = sv_2mortal(newSVnv(atime + (NV) atime_nsec / NV_1E9));
+            }
+            if (mtime_nsec) {
+                UV mtime = SvUV(SPBASE[ 9]);
+                SPBASE[ 9] = sv_2mortal(newSVnv(mtime + (NV) mtime_nsec / NV_1E9));
+            }
+            if (ctime_nsec) {
+                UV ctime = SvUV(SPBASE[10]);
+                SPBASE[10] = sv_2mortal(newSVnv(ctime + (NV) ctime_nsec / NV_1E9));
+            }
         }
         XSRETURN(nret);