qbe/main.c
Roland Paterson-Jones c2ff93e75e Global Value Numbering / Global Code Motion
More or less as proposed in its ninth iteration with the
addition of a gcmmove() functionality to restore coherent
local schedules.

Changes since RFC 8:

Features:

 - generalization of phi 1/0 detection
 - collapse linear jmp chains before GVN; simplifies if-graph
     detection used in 0/non-0 value inference and if-elim...
 - infer 0/non-0 values from dominating blk jnz; eliminates
     redundant cmp eq/ne 0 and associated jnz/blocks, for example
     redundant null pointer checks (hare codebase likes this)
 - remove (emergent) empty if-then-else graphlets between GVN and
     GCM; improves GCM instruction placement, particularly cmps.
 - merge %addr =l add %addr1, N sequences - reduces tmp count,
     register pressure.
 - squash consecutive associative ops with constant args, e.g.
     t1 = add t, N ... t2 = add t2, M -> t2 = add t, N+M

Bug Fixes:

 - remove "cmp eq/ne of non-identical RCon's " in copyref().
   RCon's are not guaranteed to be dedup'ed, and symbols can
   alias.

Codebase:

  - moved some stuff into cfg.c including blkmerge()
  - some refactoring in gvn.c
  - simplification of reassoc.c - always reassoc all cmp ops
      and Kl add %t, N. Better on coremark, smaller codebase.
  - minor simplification of movins() - use vins

Testing - standard QBE, cproc, hare, harec, coremark
          [still have Rust build issues with latest roland]

Benchmark
- coremark is ~15%+ faster than master
- hare "HARETEST_INCLUDE='slow' make check" ~8% faster
    (crypto::sha1::sha1_1gb is biggest obvious win - ~25% faster)

Changes since RFC 7:

Bug fixes:

- remove isbad4gcm() in GVN/GCM - it is unsound due to different state
    at GVN vs GCM time; replace with "reassociation" pass after GCM
- fix intra-blk use-before-def after GCM
- prevent GVN from deduping trapping instructions cos GCM will not
    move them
- remove cmp eq/ne identical arg copy detection for floating point, it
    is not valid for NaN
- fix cges/cged flagged as commutative in ops.h instead of cnes/cned
    respectively; just a typo

Minor features:

- copy detection handles cmp le/lt/ge/gt with identical args
- treat (integer) div/rem by non-zero constant as non-trapping
- eliminate add N/sub N pairs in copy detection
- maintain accurate tmp use in GVN; not strictly necessary but enables
    interim global state sanity checking
- "reassociation" of trivial constant offset load/store addresses, and
    cmp ops with point-of-use in pass after GCM
- normalise commutative op arg order - e.g. op con, tmp -> op tmp, con
    to simplify copy detection and GVN instruction dedup

Codebase:

- split out core copy detection and constant folding (back) out into
    copy.c, fold.c respectively; gvn.c was getting monolithic
- generic support for instruction moving in ins.c - used by GCM and
    reassoc
- new reassociation pass in reassoc.c
- other minor clean-up/refactor

Changes since RFC 6:

- More ext elimination in GVN by examination of def and use bit width
- elimination of redundant and mask by bit width examination
- Incorporation of Song's patch

Changes since RFC 5:

- avoidance of "bad" candidates for GVN/GCM - trivial address offset
    calculations, and comparisons
- more copy detection mostly around boolean values
- allow elimination of unused load, alloc, trapping instructions
- detection of trivial boolean v ? 1 : 0 phi patterns
- bug fix for (removal of) "chg" optimisation in ins recreation - it
    was missing removal of unused instructions in some cases

ifelim() between GVN and GCM; deeper nopunused()
2025-03-14 09:58:37 +01:00

205 lines
3.6 KiB
C

#include "all.h"
#include "config.h"
#include <ctype.h>
#include <getopt.h>
Target T;
char debug['Z'+1] = {
['P'] = 0, /* parsing */
['M'] = 0, /* memory optimization */
['N'] = 0, /* ssa construction */
['C'] = 0, /* copy elimination */
['F'] = 0, /* constant folding */
['A'] = 0, /* abi lowering */
['I'] = 0, /* instruction selection */
['L'] = 0, /* liveness */
['S'] = 0, /* spilling */
['R'] = 0, /* reg. allocation */
};
extern Target T_amd64_sysv;
extern Target T_amd64_apple;
extern Target T_arm64;
extern Target T_arm64_apple;
extern Target T_rv64;
static Target *tlist[] = {
&T_amd64_sysv,
&T_amd64_apple,
&T_arm64,
&T_arm64_apple,
&T_rv64,
0
};
static FILE *outf;
static int dbg;
static void
data(Dat *d)
{
if (dbg)
return;
emitdat(d, outf);
if (d->type == DEnd) {
fputs("/* end data */\n\n", outf);
freeall();
}
}
static void
func(Fn *fn)
{
uint n;
if (dbg)
fprintf(stderr, "**** Function %s ****", fn->name);
if (debug['P']) {
fprintf(stderr, "\n> After parsing:\n");
printfn(fn, stderr);
}
T.abi0(fn);
fillcfg(fn);
filluse(fn);
promote(fn);
filluse(fn);
ssa(fn);
filluse(fn);
ssacheck(fn);
fillalias(fn);
loadopt(fn);
filluse(fn);
fillalias(fn);
coalesce(fn);
filluse(fn);
filldom(fn);
ssacheck(fn);
gvn(fn);
fillcfg(fn);
filluse(fn);
ifelim(fn);
fillcfg(fn);
filluse(fn);
filldom(fn);
gcm(fn);
filluse(fn);
ssacheck(fn);
T.abi1(fn);
simpl(fn);
fillcfg(fn);
filluse(fn);
T.isel(fn);
fillcfg(fn);
filllive(fn);
fillloop(fn);
fillcost(fn);
spill(fn);
rega(fn);
fillcfg(fn);
simpljmp(fn);
fillcfg(fn);
assert(fn->rpo[0] == fn->start);
for (n=0;; n++)
if (n == fn->nblk-1) {
fn->rpo[n]->link = 0;
break;
} else
fn->rpo[n]->link = fn->rpo[n+1];
if (!dbg) {
T.emitfn(fn, outf);
fprintf(outf, "/* end function %s */\n\n", fn->name);
} else
fprintf(stderr, "\n");
freeall();
}
static void
dbgfile(char *fn)
{
emitdbgfile(fn, outf);
}
int
main(int ac, char *av[])
{
Target **t;
FILE *inf, *hf;
char *f, *sep;
int c;
T = Deftgt;
outf = stdout;
while ((c = getopt(ac, av, "hd:o:t:")) != -1)
switch (c) {
case 'd':
for (; *optarg; optarg++)
if (isalpha(*optarg)) {
debug[toupper(*optarg)] = 1;
dbg = 1;
}
break;
case 'o':
if (strcmp(optarg, "-") != 0) {
outf = fopen(optarg, "w");
if (!outf) {
fprintf(stderr, "cannot open '%s'\n", optarg);
exit(1);
}
}
break;
case 't':
if (strcmp(optarg, "?") == 0) {
puts(T.name);
exit(0);
}
for (t=tlist;; t++) {
if (!*t) {
fprintf(stderr, "unknown target '%s'\n", optarg);
exit(1);
}
if (strcmp(optarg, (*t)->name) == 0) {
T = **t;
break;
}
}
break;
case 'h':
default:
hf = c != 'h' ? stderr : stdout;
fprintf(hf, "%s [OPTIONS] {file.ssa, -}\n", av[0]);
fprintf(hf, "\t%-11s prints this help\n", "-h");
fprintf(hf, "\t%-11s output to file\n", "-o file");
fprintf(hf, "\t%-11s generate for a target among:\n", "-t <target>");
fprintf(hf, "\t%-11s ", "");
for (t=tlist, sep=""; *t; t++, sep=", ") {
fprintf(hf, "%s%s", sep, (*t)->name);
if (*t == &Deftgt)
fputs(" (default)", hf);
}
fprintf(hf, "\n");
fprintf(hf, "\t%-11s dump debug information\n", "-d <flags>");
exit(c != 'h');
}
do {
f = av[optind];
if (!f || strcmp(f, "-") == 0) {
inf = stdin;
f = "-";
} else {
inf = fopen(f, "r");
if (!inf) {
fprintf(stderr, "cannot open '%s'\n", f);
exit(1);
}
}
parse(inf, f, dbgfile, data, func);
fclose(inf);
} while (++optind < ac);
if (!dbg)
T.emitfin(outf);
exit(0);
}