Commit graph

18 commits

Author SHA1 Message Date
Bill Currie
10ad8bebbd Make files.dat actually useful. I really dislike the hardcoding, but it /is/
useful. Also, fix a bug in PrecacheModel where it was checking MAX_SOUNDS and
numsounds instead of models.
2001-07-18 17:23:42 +00:00
Bill Currie
0186732130 make initialized local variables work /properly/
local float urk = 0;
STORE_F    46(IMMEDIATE)  0.0   55(?)
2001-06-08 20:13:10 +00:00
Bill Currie
7ef4c2776e Immidiate sharing now /works/ (shaved off two pr_globals from CustomTF, too:).
Detect assignments to initialized globals and give an error, unless the --cow
(copy on write) option is given, and then allocate a new global for the var,
clear its initialized flag.
Relocate all globals.
2001-06-08 06:32:15 +00:00
Bill Currie
bbb37d0080 customTF is now down to 4989 pr_globals. all parameters, local veriables and
termporary variables sit in one pool of memory (at the end of the globals)
thus drasticly reducing globals requirements. This works because the whole
lot is declared to be in the function's local variable space which is copied
to the locals stack in the progs engine.
2001-06-05 08:09:12 +00:00
Bill Currie
a26f799de4 allocate temps based on size rather than type (get better re-usage this way)
put temps onto the local scope.
2001-06-05 05:22:11 +00:00
Bill Currie
17e19f3892 regain about 5000 globals for customTF, but this fixes possible breakage with
premature temp var re-usage (ie, it puts the temps back into the locals space).
2001-06-04 22:35:54 +00:00
Bill Currie
e090f3c9e3 do proper temp def handling so we don'e use any where near as many defs for
temporary variables (customTF went from 35941 to 12587 pr_globals).
2001-06-04 17:52:50 +00:00
Bill Currie
e339e82c01 hash.h is now const correct as is a lot of qfcc 2001-06-04 04:52:14 +00:00
Bill Currie
4aa47d728a use the correct STORE_ op for function arges, rather than STORE_V. Should
speed up progs in general.
2001-06-04 03:57:15 +00:00
Bill Currie
ddcd172a03 re-write the opcode selection to be a hash table lookup and statment generation
to get the opcode from the table record rather than the location within the
table (ewww). gives a nice speed boost /and/ makes the opcode table easier to
maintain.
2001-06-04 02:41:45 +00:00
Bill Currie
09118bc01e pr_comp.h:
add OP_ADD_S. WARNING!!! this /will/ move.
progs.h:
	add prototype for PR_PrintStatement
pr_edict.c:
	add OP_ADD_S support in the progs checker
pr_exec.c:
	implement OP_ADD_S
tools/qfcc/include/.gitignore:
	add config.h.in
qfcc.h:
	nuke PR_NameImmediate and change PR_ParseImmediate's prototype (see
	pr_imm.c)
pr_comp.c:
	add ADD_S, adjust for PR_ParseImmediate's prototype, make
	PR_ParseExpression work with non-sequential opcodes (slow, will work on
	that next). Fix up initialised global parsing.
pr_imm.c:
	nuke PR_NameImmediate. didn't work well and wasn't such a good idea anyway.
	PR_ParseImmediate now accepts a def_t * arg. if null, will allocate a
	new global def, otherwise it will initialize the def passed in.
qwaq/main.c:
	sports some debugging code (dumps info about the progs it's running)
qwaq/main.qc:
	better ADD_S testing
2001-06-03 17:36:49 +00:00
Bill Currie
b7aba0b643 allow named constants to be checked for immediate constants. qfcc now produces
the same number of globals etc as it used to
2001-04-02 18:43:10 +00:00
Bill Currie
02b09f4e5c no more redundant strings. costs ~7ms, though, but HALVING the strofs size
in frikbot seems worth it.
2001-04-01 06:40:51 +00:00
Bill Currie
59e55834ed Another massive speadup caused by using hash tables to lookup already
generated immediate values. frikbot now compiles in just over 1s on my
machine.
2001-04-01 06:01:02 +00:00
Bill Currie
ec16ffaa65 move to using hash tables for variable lookups.
WARNING: this fixes a scope bug in qcc and thus you WILL get different (but
now correct) code for the following:

float foo;

void () bar =
{
	local float foo;
	foo = 0;
}

ie, the local foo will, as is correct, now be a separate var to the global
foo.
2001-04-01 02:12:57 +00:00
Bill Currie
cb5c262ffc qtypes.h:
remove includes of qdefs.h and compat.h
pr_comp.h:
	merge pr_comp.h from quake and qfcc, removing the copy in qfcc
cmdlib.[ch]:
	nuke the endian code.
qendian.c:
	initialise the LittleLong etc pointers at compile time rather than run
	time
com.c (both nq and qw):
	nuke the LittleLong etc init code
everything else:
	fix up after the qtypes.h cleanup
2001-03-28 17:17:56 +00:00
Bill Currie
6247233e0e when checking a token, they type of the token must be taken into account. 2001-03-14 04:14:09 +00:00
Jeff Teunissen
2dd03876af QFCC -- the QuakeForge Code Compiler -- an autoconfiscated qcc.
It sucks, but it works, and will form the basis for something that Does
Not Suck.
2001-02-24 06:58:54 +00:00