REFACTOR: move sources to src/

1 files changed, 308 insertions, 0 deletions

diff --git a/src/ir_simpl.c b/src/ir_simpl.c
new file mode 100644
index 0000000..a01879a
--- /dev/null
+++ b/src/ir_simpl.c
@@ -0,0 +1,308 @@
+#include "ir.h"
+
+static int
+mulk(struct instr *ins, struct block *blk, int *curi)
+{
+   vlong iv = intconval(ins->r);
+   enum irclass cls = ins->cls;
+   assert((uvlong)iv > 1 && "trivial mul not handled by irbinop() ?");
+   bool neg = iv < 0;
+   if (neg) iv = -iv;
+   /* This can be generalized to any sequence of shifts and
+    * adds/subtracts, but whether that's worth it depends on the number of them
+    * and the microarchitecture..  clang seems to stop after two shifts. Should
+    * we compute approximate cost of instrs to determine? For now just handle
+    * the po2 (+/- 1) case */
+   if (ispo2(iv)) {
+      /* x * 2^y ==> x << y */
+      ins->op = Oshl;
+      ins->r = mkref(RICON, ilog2(iv));
+   } else if (ispo2(iv-1)) {
+      /* x * 5 ==> (x << 2) + x */
+      ins->op = Oadd;
+      ins->r = ins->l;
+      ins->l = insertinstr(blk, (*curi)++, mkinstr(Oshl, cls, ins->l, mkref(RICON, ilog2(iv-1))));
+   } else if (ispo2(iv+1)) {
+      /* x * 7 ==> (x << 3) - x */
+      ins->op = Osub;
+      ins->r = ins->l;
+      ins->l = insertinstr(blk, (*curi)++, mkinstr(Oshl, cls, ins->l, mkref(RICON, ilog2(iv+1))));
+   } else return 0;
+   if (neg) {
+      ins->l = insertinstr(blk, (*curi)++, *ins);
+      ins->op = Oneg;
+      ins->r = NOREF;
+   }
+   return 1;
+}
+
+static int
+divmodk(struct instr *ins, struct block *blk, int *curi)
+{
+   enum op op = ins->op;
+   enum irclass cls = ins->cls;
+   vlong iv = intconval(ins->r);
+   uint nbit = 8 * cls2siz[cls];
+   bool neg = (op == Odiv || op == Orem) && iv < 0;
+   if (ispo2(iv) || (neg && ispo2(-iv))) { /* simple po2 cases */
+      union ref temp;
+      uint s = ilog2(neg ? -iv : iv);
+      switch (op) {
+      default: assert(0);
+      case Oudiv: /* x / 2^s ==> x >> s */
+         ins->op = Oslr, ins->r = mkref(RICON, s);
+         return 1;
+      case Ourem: /* x % 2^s ==> x & 2^s-1 */
+         ins->op = Oand, ins->r = mkintcon(cls, iv - 1);
+         return 1;
+      case Odiv: case Orem:
+         /* have to adjust to round negatives toward zero */
+         /* x' = (((x < 0 ? -1 : 0) >>> (Nbit - s)) + x) */
+         temp = insertinstr(blk, (*curi)++, mkinstr(Osar, cls, ins->l, mkref(RICON, nbit - 1)));
+         temp = insertinstr(blk, (*curi)++, mkinstr(Oslr, cls, temp, mkref(RICON, nbit - s)));
+         temp = insertinstr(blk, (*curi)++, mkinstr(Oadd, cls, ins->l, temp));
+         if (op == Odiv) {
+            /* (-) (x' >> s) */
+            struct instr sar = mkinstr(Osar, cls, temp, mkref(RICON, s));
+            if (!neg) *ins = sar;
+            else {
+               temp = insertinstr(blk, (*curi)++, sar);
+               ins->op = Oneg, ins->l = temp, ins->r = NOREF;
+            }
+         } else {
+            /* x - (x' & -(2^s)) */
+            temp = insertinstr(blk, (*curi)++, mkinstr(Oand, cls, temp, mkintcon(cls, neg ? iv : -iv)));
+            ins->op = Osub, ins->r = temp;
+         }
+         break;
+         return 0;
+      }
+   }
+   return 0;
+}
+
+static int
+doins(struct instr *ins, struct block *blk, int *curi)
+{
+   int narg = opnarg[ins->op];
+   if (oisarith(ins->op)) {
+      union ref r = narg == 1 ? irunop(NULL, ins->op, ins->cls, ins->l)
+                              : irbinop(NULL, ins->op, ins->cls, ins->l, ins->r);
+      if (r.bits) {
+         *ins = mkinstr(Onop,0,);
+         replcuses(mkref(RTMP, ins - instrtab), r);
+         return 1;
+      }
+   }
+   enum irclass k = ins->cls;
+   switch (ins->op) {
+   case Ocopy:
+      if ((ins->l.t == RTMP && k == instrtab[ins->l.i].cls)
+       || (isnumcon(ins->l) && k == concls(ins->l))
+       || (kisint(k) && ins->l.t == RICON)) {
+         union ref it = ins->l;
+         *ins = mkinstr(Onop,0,);
+         replcuses(mkref(RTMP, ins - instrtab), it);
+         return 1;
+      }
+      break;
+   case Omul:
+      if (kisflt(k)) break;
+      if (isnumcon(ins->l)) rswap(ins->l, ins->r); /* put const in rhs */
+      if (isintcon(ins->r)) return mulk(ins, blk, curi);
+      break;
+   case Odiv: case Oudiv: case Orem: case Ourem:
+      if (kisflt(k)) break;
+      if (isintcon(ins->r)) return divmodk(ins, blk, curi);
+      break;
+   case Oequ: case Oneq:
+      if (ins->l.t == RTMP && isintcon(ins->r)) {
+         /* optimize `x <op> C <cmp> Q` */
+         /* could apply with add/sub to lth/lte/gth/gte iff no signed wraparound, which isn't assumed */
+         struct instr *lhs = &instrtab[ins->l.i];
+         enum op o = lhs->op;
+         if (ins->cls == lhs->cls && (o == Oadd || o == Osub || o == Oxor) && isintcon(lhs->r)) {
+            uvlong c = intconval(ins->r), q = intconval(lhs->r);
+            switch (o) { default: assert(0);
+            case Oadd: c -= q; break; /* x + 3 == C ==> x == C - 3 */
+            case Osub: c += q; break; /* x - 3 == C ==> x == C + 3 */
+            case Oxor: c ^= q; break; /* x ^ 3 == C ==> x == C ^ 3 */
+            }
+            ins->l = lhs->l, ins->r = mkintcon(ins->cls, c);
+            deluse(blk, ins - instrtab, mkref(RTMP, lhs - instrtab));
+            if (!instruse[lhs - instrtab]) *lhs = mkinstr(Onop,0,);
+            return 1;
+         }
+      }
+   }
+   return 0;
+}
+
+static void
+jmpfind(struct block **final, struct block **pblk)
+{
+   struct block **p2 = &final[(*pblk)->id];
+   if (*p2 && !(*p2)->phi.n) {
+      jmpfind(final, p2);
+      *pblk = *p2;
+   }
+}
+
+static void
+fillpredsrec(struct block *blk)
+{
+   while (blk && !wasvisited(blk)) {
+      markvisited(blk);
+      if (!blk->s1) return;
+      if (!blk->s1->phi.n) addpred(blk->s1, blk);
+      if (!blk->s2) {
+         blk = blk->s1;
+         continue;
+      }
+      if (!blk->s2->phi.n) addpred(blk->s2, blk);
+      fillpredsrec(blk->s1);
+      blk = blk->s2;
+   }
+}
+
+static void
+fillpreds(struct function *fn)
+{
+   struct block *blk = fn->entry, *next;
+   do {
+      if (blk->phi.n) continue;
+      if (blk->npred > 1)
+         xbfree(blk->_pred);
+      blk->npred = 0;
+      blk->_pred = NULL;
+   } while ((blk = blk->lnext) != fn->entry);
+   startbbvisit();
+   fillpredsrec(fn->entry);
+   blk = fn->entry->lnext;
+   do { /* remove dead blocks */
+      next = blk->lnext;
+      if (!blk->npred) freeblk(fn, blk);
+   } while ((blk = next) != fn->entry);
+}
+
+static void
+mergeblks(struct function *fn, struct block *p, struct block *s)
+{
+   assert(s->npred == 1 && !s->phi.n);
+   vpushn(&p->ins, s->ins.p, s->ins.n);
+   p->jmp = p->s1->jmp;
+   for (int i = 0; i < 2; ++i) {
+      struct block *ss = (&s->s1)[i];
+      if (ss) for (int j = 0; j < ss->npred; ++j) {
+         if (blkpred(ss, j) == s) {
+            blkpred(ss, j) = p;
+            break;
+         }
+      }
+   }
+   /* breaks use for temps used in s */
+   if (s->ins.n || s->jmp.arg[0].t == RTMP || s->jmp.arg[1].t == RTMP)
+      fn->prop &= ~FNUSE;
+   p->s1 = s->s1;
+   p->s2 = s->s2;
+   freeblk(fn, s);
+}
+
+void
+simpl(struct function *fn)
+{
+   FREQUIRE(FNUSE);
+   int inschange = 0,
+       blkchange = 0;
+   struct block **jmpfinal = allocz(fn->passarena, fn->nblk * sizeof *jmpfinal, 0);
+   struct block *blk = fn->entry;
+
+   do {
+      /* merge blocks:
+       * @blk:
+       *   ... (1)
+       *   b @s
+       * @s:
+       *   ... (2)
+       *   b {...}
+       *=>
+       * @blk:
+       *   ... (1)
+       *   ... (2)
+       *   b {...}
+       * */
+      if (blk != fn->entry) while (blk->s1 && !blk->s2 && blk->s1->npred == 1 && !blk->phi.n) {
+         mergeblks(fn, blk, blk->s1);
+      }
+   } while ((blk = blk->lnext) != fn->entry);
+
+   if (!(fn->prop & FNUSE)) filluses(fn);
+
+   do {
+      for (int i = 0; i < blk->phi.n; ++i) {
+         int phi = blk->phi.p[i];
+         /* delete trivial phis */
+         union ref *args = phitab.p[instrtab[phi].l.i],
+               same = *args;
+         if (same.t == RTMP && instrtab[same.i].op == Ophi) goto Next;
+         if (blk->npred > 1) for (int j = 1; j < blk->npred; ++j) {
+            if (args[j].bits != same.bits) goto Next;
+         }
+         if (!(fn->prop & FNUSE)) filluses(fn);
+         replcuses(mkref(RTMP, phi), same);
+         delphi(blk, i);
+         Next:;
+      }
+
+      int curi = 0;
+   DoIns:
+      for (; curi < blk->ins.n; ++curi) {
+         struct instr *ins = &instrtab[blk->ins.p[curi]];
+         if (ins->op != Onop) {
+            if (!(fn->prop & FNUSE)) filluses(fn);
+            inschange += doins(ins, blk, &curi);
+         }
+      }
+
+      if (blk->s2 && isintcon(blk->jmp.arg[0])) {
+         /* simplify known conditional branch */
+         struct block *s = intconval(blk->jmp.arg[0]) ? blk->s1 : blk->s2;
+         delpred(s == blk->s1 ? blk->s2 : blk->s1, blk);
+         blk->s1 = s, blk->s2 = NULL;
+         blk->jmp.arg[0] = NOREF;
+         if (blk->s1 && !blk->s2 && blk->s1->npred == 1 && blk->s1->phi.n == 0) {
+            mergeblks(fn, blk, blk->s1);
+            goto DoIns;
+         }
+      }
+
+      if (blk != fn->entry && blk->npred == 0) {
+         freeblk(fn, blk);
+      } else if (!blk->phi.n && !blk->ins.n) { /* thread jumps.. */ 
+         if (blk->jmp.t == Jb && !blk->s2) {
+            jmpfind(jmpfinal, &blk->s1);
+            if (blk->s1 != blk) {
+               ++blkchange;
+               jmpfinal[blk->id] = blk->s1;
+            }
+         }
+      }
+   } while ((blk = blk->lnext) != fn->entry);
+
+   if (blkchange) {
+      do {
+         if (blk->s1) jmpfind(jmpfinal, &blk->s1);
+         if (blk->s2) jmpfind(jmpfinal, &blk->s2);
+         if (blk->s1 && blk->s1 == blk->s2) {
+            memset(blk->jmp.arg, 0, sizeof blk->jmp.arg);
+            blk->s2 = NULL;
+         }
+      } while ((blk = blk->lnext) != fn->entry);
+      fillpreds(fn);
+      sortrpo(fn);
+   }
+   if (!(fn->prop & FNUSE)) filluses(fn);
+   fn->prop &= ~FNDOM;
+}
+
+/* vim:set ts=3 sw=3 expandtab: */