qf/quakeforge
0
0
Fork 0
mirror of https://git.code.sf.net/p/quake/quakeforge synced 2024-11-14 08:50:58 +00:00
Code Issues Projects Releases Packages Wiki Activity
quakeforge/tools/qfcc/test/Makefile.am

159 lines
4.4 KiB
Text
Raw Normal View History

Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
AUTOMAKE_OPTIONS= foreign
INCLUDES= -I$(top_srcdir)/include $(QFCC_INCS)
QFCC_DEP=$(builddir)/../source/qfcc$(EXEEXT)
QFCC=$(QFCC_DEP)
Add vecinit.r to the test-suite. Ugh, I really need to take time out to do compile tests properly.
2012-12-10 12:09:53 +00:00
QCFLAGS=-qq -O -g --no-default-paths -Werror
Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
QCPPFLAGS=
Get automatic dependencies working for the progs files. Including dependencies on QFCC itself :)
2012-11-22 08:11:21 +00:00
QCOMPILE=$(QFCC) $(QCFLAGS) $(QCPPFLAGS)
Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
SUFFIXES=.qfo .r
.r.qfo:
Get automatic dependencies working for the progs files. Including dependencies on QFCC itself :)
2012-11-22 08:11:21 +00:00
$(QCOMPILE) -MT $@ -MD -MP -MF $(DEPDIR)/$*.Tqo -c -o $@ $<
sed -i -e '1s@:@: $(QFCC_DEP)@' $(DEPDIR)/$*.Tqo
$(am__mv) $(DEPDIR)/$*.Tqo $(DEPDIR)/$*.Qo
Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
QFCC_TEST_LIBS=@QFCC_TEST_LIBS@
QFCC_TEST_DEPS=@QFCC_TEST_DEPS@
QFCC_TEST_INCS=@QFCC_TEST_INCS@
Add a test case for the alias chain mangling. I think I need to rework the tests to better handle compile-only tests.
2012-11-26 12:42:37 +00:00
test_progs_dat=\
chewed-alias.dat \
Don't use the expr type for assignments. Instead, the type is taken from the source operand (this means the computed type of .return for that var). ctf now builds :)
2012-11-30 11:47:36 +00:00
chewed-return.dat \
Add a test for function local static variables. It turns out they're not getting allocated properly (they're put in the function's locals defspace rather than near data), but fixing it proved a little more problematic than expected, so the test is marked as XFAIL for now to remind me to fix it.
2012-12-03 07:27:31 +00:00
func-static.dat \
Add a test case for the alias chain mangling. I think I need to rework the tests to better handle compile-only tests.
2012-11-26 12:42:37 +00:00
deadbool.dat \
infloop.dat \
modulo.dat \
Add a test for returning @param into an array.
2012-12-26 02:49:39 +00:00
paramret.dat \
Force RETURN's operand to be live. This seems to be the final fix to get return-ivar.r working.
2012-12-20 12:01:46 +00:00
return-ivar.dat \
Add a test for compiling calls to obj_msg_sendv. It seems dag_set_live_vars still served a purpose after all, but I don't feel like bringing it as I'd rather implement its param handing in dagnode_set_edges. I've now got a test case for it, though the test currently causes the VM to segfault (even with pr_boundscheck 2!).
2012-12-22 05:05:12 +00:00
sendv.dat \
Start implementing vec = [x,y,z]. This is a nice feature found in fteqcc (also a bit of a challenge from Spike). Getting bison to accept the new expression required rewriting the state expression grammar, so this is mostly for the state expression. A test to ensure the state expression doesn't break is included.
2013-06-24 02:36:52 +00:00
state.dat \
Add a test for struct writes. This tests local kills via aliases.
2012-12-13 04:44:25 +00:00
structlive.dat \
Add a test case for the alias chain mangling. I think I need to rework the tests to better handle compile-only tests.
2012-11-26 12:42:37 +00:00
structptr.dat \
Add vecinit.r to the test-suite. Ugh, I really need to take time out to do compile tests properly.
2012-12-10 12:09:53 +00:00
vecinit.dat \
Append a new sblock for return when necessary. If the final block ends in a conditional statement, appending return to the block will hide the conditional statement from the flow analyzer. This may cause the conditional statement's destination node be become unreachable according to the analyzer and thus eliminated. The label for the branch then loses its target sblock and thus the code generator will produce a zero-distance jump resulting in an infinite loop. Thus, if the final block ends in a conditional statement (or, for completeness, a call statement), append a new empty block before adding the return statement.
2012-12-21 11:03:28 +00:00
while.dat \
voidfor.dat
Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
Add a test for function local static variables. It turns out they're not getting allocated properly (they're put in the function's locals defspace rather than near data), but fixing it proved a little more problematic than expected, so the test is marked as XFAIL for now to remind me to fix it.
2012-12-03 07:27:31 +00:00
fail_progs_dat=\
Remove func-static from XFAIL_TESTS. I realized that's not the way it's meant to be used, and it /is/ I bug I want fixed before release, so... :)
2012-12-04 04:27:39 +00:00
$E
Add a test for function local static variables. It turns out they're not getting allocated properly (they're put in the function's locals defspace rather than near data), but fixing it proved a little more problematic than expected, so the test is marked as XFAIL for now to remind me to fix it.
2012-12-03 07:27:31 +00:00
Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
TESTS=$(test_progs_dat:.dat=.run)
Add a test for function local static variables. It turns out they're not getting allocated properly (they're put in the function's locals defspace rather than near data), but fixing it proved a little more problematic than expected, so the test is marked as XFAIL for now to remind me to fix it.
2012-12-03 07:27:31 +00:00
XFAIL_TESTS=$(fail_progs_dat:.dat=.run)
Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
check_PROGRAMS=test-harness $(test_progs_dat)
test_harness_SOURCES= test-bi.c test-harness.c
test_harness_LDADD= $(QFCC_TEST_LIBS)
test_harness_DEPENDENCIES= $(QFCC_TEST_DEPS)
Add a test case for the alias chain mangling. I think I need to rework the tests to better handle compile-only tests.
2012-11-26 12:42:37 +00:00
chewed_alias_dat_SOURCES=chewed-alias.r
chewed_alias_obj=$(chewed_alias_dat_SOURCES:.r=.qfo)
chewed-alias.dat: $(chewed_alias_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(chewed_alias_obj)
chewed-alias.run: Makefile build-run
$(srcdir)/build-run $@
Don't use the expr type for assignments. Instead, the type is taken from the source operand (this means the computed type of .return for that var). ctf now builds :)
2012-11-30 11:47:36 +00:00
chewed_return_dat_SOURCES=chewed-return.r
chewed_return_obj=$(chewed_return_dat_SOURCES:.r=.qfo)
chewed-return.dat: $(chewed_return_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(chewed_return_obj)
chewed-return.run: Makefile build-run
TEST_HARNESS_OPTS=--float $(srcdir)/build-run $@
Add a test for function local static variables. It turns out they're not getting allocated properly (they're put in the function's locals defspace rather than near data), but fixing it proved a little more problematic than expected, so the test is marked as XFAIL for now to remind me to fix it.
2012-12-03 07:27:31 +00:00
func_static_dat_SOURCES=func-static.r
func_static_obj=$(func_static_dat_SOURCES:.r=.qfo)
func-static.dat: $(func_static_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(func_static_obj)
func-static.run: Makefile build-run
$(srcdir)/build-run $@
Add a test for the dead boolean expression. This tests the zombie label problem found via ctf.
2012-11-26 12:05:24 +00:00
deadbool_dat_SOURCES=deadbool.r
deadbool_obj=$(deadbool_dat_SOURCES:.r=.qfo)
deadbool.dat: $(deadbool_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(deadbool_obj)
deadbool.run: Makefile build-run
$(srcdir)/build-run $@
Add a test for writes to global vars in a loop.
2012-11-24 05:53:45 +00:00
infloop_dat_SOURCES=infloop.r
infloop_obj=$(infloop_dat_SOURCES:.r=.qfo)
infloop.dat: $(infloop_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(infloop_obj)
infloop.run: Makefile build-run
$(srcdir)/build-run $@
Add a new test case. It turns out that the optimizer completely breaks (a + b) % (a - b) with v6 code.
2012-11-22 11:49:12 +00:00
modulo_dat_SOURCES=modulo.r
modulo_obj=$(modulo_dat_SOURCES:.r=.qfo)
modulo.dat: $(modulo_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(modulo_obj)
modulo.run: Makefile build-run
TEST_HARNESS_OPTS=--float $(srcdir)/build-run $@
Add a test for returning @param into an array.
2012-12-26 02:49:39 +00:00
paramret_dat_SOURCES=paramret.r
paramret_obj=$(paramret_dat_SOURCES:.r=.qfo)
paramret.dat: $(paramret_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(paramret_obj)
paramret.run: Makefile build-run
TEST_HARNESS_OPTS=--float $(srcdir)/build-run $@
Force RETURN's operand to be live. This seems to be the final fix to get return-ivar.r working.
2012-12-20 12:01:46 +00:00
return_ivar_dat_SOURCES=return-ivar.r
return_ivar_obj=$(return_ivar_dat_SOURCES:.r=.qfo)
return-ivar.dat: $(return_ivar_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(return_ivar_obj)
return-ivar.run: Makefile build-run
$(srcdir)/build-run $@
Add a test for compiling calls to obj_msg_sendv. It seems dag_set_live_vars still served a purpose after all, but I don't feel like bringing it as I'd rather implement its param handing in dagnode_set_edges. I've now got a test case for it, though the test currently causes the VM to segfault (even with pr_boundscheck 2!).
2012-12-22 05:05:12 +00:00
sendv_dat_SOURCES=sendv.r
sendv_obj=$(sendv_dat_SOURCES:.r=.qfo)
sendv.dat: $(sendv_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(sendv_obj)
sendv.run: Makefile build-run
$(srcdir)/build-run $@
Start implementing vec = [x,y,z]. This is a nice feature found in fteqcc (also a bit of a challenge from Spike). Getting bison to accept the new expression required rewriting the state expression grammar, so this is mostly for the state expression. A test to ensure the state expression doesn't break is included.
2013-06-24 02:36:52 +00:00
state_dat_SOURCES=state.r
state_obj=$(state_dat_SOURCES:.r=.qfo)
state.dat: $(state_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(state_obj)
state.run: Makefile build-run
$(srcdir)/build-run $@
Add a test for struct writes. This tests local kills via aliases.
2012-12-13 04:44:25 +00:00
structlive_dat_SOURCES=structlive.r
structlive_obj=$(structlive_dat_SOURCES:.r=.qfo)
structlive.dat: $(structlive_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(structlive_obj)
structlive.run: Makefile build-run
$(srcdir)/build-run $@
Fix make dist for qfcc's test.
2012-11-22 06:47:28 +00:00
structptr_dat_SOURCES=structptr.r
structptr_obj=$(structptr_dat_SOURCES:.r=.qfo)
Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
structptr.dat: $(structptr_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(structptr_obj)
structptr.run: Makefile build-run
$(srcdir)/build-run $@
Add vecinit.r to the test-suite. Ugh, I really need to take time out to do compile tests properly.
2012-12-10 12:09:53 +00:00
vecinit_dat_SOURCES=vecinit.r
vecinit_obj=$(vecinit_dat_SOURCES:.r=.qfo)
vecinit.dat: $(vecinit_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(vecinit_obj)
vecinit.run: Makefile build-run
$(srcdir)/build-run $@
Fix make dist for qfcc's test.
2012-11-22 06:47:28 +00:00
while_dat_SOURCES=while.r
while_obj=$(while_dat_SOURCES:.r=.qfo)
Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
while.dat: $(while_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(while_obj)
while.run: Makefile build-run
$(srcdir)/build-run $@
Append a new sblock for return when necessary. If the final block ends in a conditional statement, appending return to the block will hide the conditional statement from the flow analyzer. This may cause the conditional statement's destination node be become unreachable according to the analyzer and thus eliminated. The label for the branch then loses its target sblock and thus the code generator will produce a zero-distance jump resulting in an infinite loop. Thus, if the final block ends in a conditional statement (or, for completeness, a call statement), append a new empty block before adding the return statement.
2012-12-21 11:03:28 +00:00
voidfor_dat_SOURCES=voidfor.r
voidfor_obj=$(voidfor_dat_SOURCES:.r=.qfo)
voidfor.dat: $(voidfor_obj) $(QFCC_DEP)
$(QFCC) $(QCFLAGS) -o $@ $(voidfor_obj)
voidfor.run: Makefile build-run
$(srcdir)/build-run $@
Get automatic dependencies working for the progs files. Including dependencies on QFCC itself :)
2012-11-22 08:11:21 +00:00
include ./$(DEPDIR)/*.Qo
Fix make dist for qfcc's test.
2012-11-22 06:47:28 +00:00
EXTRA_DIST= test-bi.h build-run
Add a test harness for qfcc. It's only the beginning, but finally make check is useful for qfcc :) It even has a failing test :D
2012-11-22 06:40:16 +00:00
CLEANFILES= *.dat *.sym *.qfo *.run
Reference in a new issue Copy permalink