Help Wanted Optimizing Win32 Compressor

Hi! I have an issue. I am working on compression, and the code is way too slow. I need help optimizing my code. A code snippet follows:
-----------------------
static unsigned char po_getnumlits2 (int pos)
{
register unsigned char i, j, l=0;
unsigned char* tmpcptr=&InBuffer[pos];
for (i=1; i<=100; ++i) {
--tmpcptr;
if (pos-i<0) break;
//l=1;
for (j=i-1; j>0; --j) {
if (tmpcptr[0]==tmpcptr[j])
{--l; break;}
} ++l;
} return l>100?100:l;
}

static unsigned char po_getnum (int in, unsigned char len)
{
unsigned i, j, k, l, l2;
unsigned x;
unsigned char comp=0;
unsigned char* y;
if (j<0) {po_compl=EstimDist (k, po_getnumlits(in))+1; po_offs[0]=k; return 1;}
po_offs[0]=-1;
return 0;
}
for (i=j=k=0; i<len; ++i) {
j=0; y=&InBuffer[in+i];
for (k=0; k<100 && j<100; /*++k*/) {
if (len<6 && j>=32) {j=100; break;}
m=0;
--y;
if ((unsigned)y<(unsigned)&InBuffer) {j= 100; break;}
if (cin==*y) break;
if (j>=0) {
for (l=1;l<=j; ++l) {
if (*y==y[l]) {
m=1; break;
}
}
}
if (m) {++j; continue;}

++j;
++k;
}
if (k<100 && j<100) {
po_offs=k; //comp+=/*5-(k<4)*/(k<6?1:getnumbits(po_getnumlits(vz.InPos-buflen+i+1)));
if (0 && 1 || len>=3 && len<4) comp+=(k<8?4:EstimDist(k-8, po_getnumlits2(l2)-8))+(0 && buflen>=4)+1; //n=1;
else if (len<6) comp+=EstimDist(k, po_getnumlits(in+i));
else comp+=EstimDist (k, po_getnumlits2(in+i)); continue;
}
else {
po_offs=-1;
if ((len<5)) x=po_writelit (in+i, InBuffer[in+i], &l);
else x=po_writelit1a (in+i, InBuffer[in+i], &l);
comp+=GetHuffLen (&cin, 1);
po_nl1=x; po_nl2=l;
}
}
po_compl=comp;
return 1;
}

static unsigned po_writelit1a (int pos, unsigned char c, unsigned *outlen)
{
register unsigned i, k, l, m;
register unsigned j;
register unsigned char* s=&InBuffer[pos];
memset (&litsused, 0, sizeof(litsused));
for (i=1; i<=100; ++i) {
if (pos-i<0) break;
--s;
litsused[*s]=1;
}
if (pos) {
k=InBuffer[pos-1];
} else k=0;
for (j=i=l=0; j<256; ++j) {
if (!litsused[j&255] && HuffEnv.entrynum[j]>=0) {
if ((j&255)<c) i++;
l++;
}
}
*outlen=(unsigned) l;
return i;
}
-------------------------
I'm compiling on a Win10/64 laptop with an Intel Celeron N2830 microprocessor using DMC.