pedroediaz/lexvm
1
0
Fork 0
Code Issues Pull Requests Actions Packages Projects Releases Wiki Activity

lower nsubs memory complexity

Browse Source 
Now it shouldn't have the whole regex worst case,
optimizes basic search cases with 1 alt, now giving
a sane nsubs size default that should be very hard
to ever surpass.
This commit is contained in:
Kyryl Melekhin
2021-09-08 14:15:12 +00:00
parent 6b37292d0f
commit 529ec9c722
1 changed files with 7 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

16
pike.c
View File

@@ -537,18 +537,12 @@ int re_pikevm(rcode *prog, const char *s, const char **subp, int nsubp)
{ {
int i, j, c, gen, subidx = 1, *npc; int i, j, c, gen, subidx = 1, *npc;
int rsubsize = sizeof(rsub)+(sizeof(char*)*nsubp); int rsubsize = sizeof(rsub)+(sizeof(char*)*nsubp);
int nsubssize = rsubsize * (prog->len+3 - prog->splits); int clistidx = 0, nlistidx = 0, pclistidx;
int clistidx = 0, nlistidx = 0;
const char *sp = s, *_sp = s; const char *sp = s, *_sp = s;
int *insts = prog->insts; int *insts = prog->insts;
int *pcs[prog->splits]; int *pcs[prog->splits];
rsub *subs[prog->splits]; rsub *subs[prog->splits];
/* Although worst case scenario nsubs size is prog->len, char nsubs[rsubsize * 512];
with moderate sized regexes it is easy to stack overflow
here. Most of the time only very small portion of memory
is actually used, but it is necessary to cover all cases
and posible paths, as it is nondeterministic. */
char nsubs[nsubssize > 500000 ? 500000 : nsubssize];
rsub *nsub, *s1, *matched = NULL, *freesub = NULL; rsub *nsub, *s1, *matched = NULL, *freesub = NULL;
rthread _clist[prog->len], _nlist[prog->len]; rthread _clist[prog->len], _nlist[prog->len];
rthread *clist = _clist, *nlist = _nlist, *tmp; rthread *clist = _clist, *nlist = _nlist, *tmp;
@@ -590,8 +584,11 @@ int re_pikevm(rcode *prog, const char *s, const char **subp, int nsubp)
nlist = tmp; nlist = tmp;
clistidx = nlistidx; clistidx = nlistidx;
nlistidx = 0; nlistidx = 0;
if (!matched) { if (clistidx != 1 && !matched) {
if (!clistidx && pclistidx)
_sp = sp;
jmp_start: jmp_start:
pclistidx = nlistidx;
newsub(for (i = 1; i < nsubp; i++) s1->sub[i] = NULL;, /*nop*/) newsub(for (i = 1; i < nsubp; i++) s1->sub[i] = NULL;, /*nop*/)
s1->ref = 1; s1->ref = 1;
s1->sub[0] = _sp; s1->sub[0] = _sp;
@@ -599,6 +596,7 @@ int re_pikevm(rcode *prog, const char *s, const char **subp, int nsubp)
addthread(1, clist, clistidx) addthread(1, clist, clistidx)
} else if (!clistidx) } else if (!clistidx)
break; break;
pclistidx = clistidx;
} }
if (matched) { if (matched) {
for (i = 0, j = i; i < nsubp; i+=2, j++) { for (i = 0, j = i; i < nsubp; i+=2, j++) {