Core 300 Mac build issues

[carli2]

Carliu, I think we are seen some very different you and I.
you keep saying the project is a mess, it is full of Bug, uninilized variables, while bunch of pejorative adjectives, but each time I ask waht it si that you want, you reply with generalities tha do not really help.

I'm sorry for that.

My current whish:

Code: Select all

gcc -c -Wall -Wno-strict-aliasing -D_POSIX_VER -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant -I../../source/core -I../../source/physics -I../../source/meshUtil -I../../source/openCL -I  -o ../../source/core/dgAABBPolygonSoup.o ../../source/core/dgAABBPolygonSoup.cpp
gcc -c -Wall -Wno-strict-aliasing -D_POSIX_VER -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant -I../../source/core -I../../source/physics -I../../source/meshUtil -I../../source/openCL -I  -o ../../source/core/dgAsyncThread.o ../../source/core/dgAsyncThread.cpp
gcc -c -Wall -Wno-strict-aliasing -D_POSIX_VER -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant -I../../source/core -I../../source/physics -I../../source/meshUtil -I../../source/openCL -I  -o ../../source/core/dgConvexHull3d.o ../../source/core/dgConvexHull3d.cpp
gcc: error: ../../source/core/dgAABBPolygonSoup.o: No such file or directory
gcc -c -Wall -Wno-strict-aliasing -D_POSIX_VER -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant -I../../source/core -I../../source/physics -I../../source/meshUtil -I../../source/openCL -I  -o ../../source/core/dgConvexHull4d.o ../../source/core/dgConvexHull4d.cpp
make: *** [../../source/core/dgAABBPolygonSoup.o] Error 1
make: *** Waiting for unfinished jobs....
gcc: error: ../../source/core/dgAsyncThread.o: No such file or directory
gcc: error: ../../source/core/dgConvexHull4d.o: No such file or directory
gcc: error: ../../source/core/dgConvexHull3d.o: No such file or directory
make: *** [../../source/core/dgConvexHull4d.o] Error 1
make: *** [../../source/core/dgAsyncThread.o] Error 1
make: *** [../../source/core/dgConvexHull3d.o] Error 1


I was not able compile Newton 3 the last 2 months and Newton 2 is unstable with lots of weird effects.

[Julio Jerez]

Ok teh mke file are up todate now

[carli2]

Did you ever try the make files?

Code: Select all

gcc -c -Wall -Wno-strict-aliasing -D_POSIX_VER -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant -I../../source/core -I../../source/physics -I../../source/meshUtil -I../../source/openCL -I  -o ../../source/core/dgAABBPolygonSoup.o ../../source/core/dgAABBPolygonSoup.cpp
gcc -c -Wall -Wno-strict-aliasing -D_POSIX_VER -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant -I../../source/core -I../../source/physics -I../../source/meshUtil -I../../source/openCL -I  -o ../../source/core/dgAsyncThread.o ../../source/core/dgAsyncThread.cpp
gcc -c -Wall -Wno-strict-aliasing -D_POSIX_VER -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant -I../../source/core -I../../source/physics -I../../source/meshUtil -I../../source/openCL -I  -o ../../source/core/dgConvexHull3d.o ../../source/core/dgConvexHull3d.cpp
gcc: error: ../../source/core/dgAABBPolygonSoup.o: No such file or directory
gcc: error: ../../source/core/dgConvexHull3d.o: No such file or directory
gcc: error: ../../source/core/dgAsyncThread.o: No such file or directory
make: *** [../../source/core/dgConvexHull3d.o] Error 1
make: *** Waiting for unfinished jobs....
make: *** [../../source/core/dgAABBPolygonSoup.o] Error 1
make: *** [../../source/core/dgAsyncThread.o] Error 1


[JoeJ]

Did you ever try the make files?

Why don't you see WHAT's wrong and post a hint or fix?
My opinion is: It's a waste of talent, if Julio needs to care for everything.
He gone open source and should get something back, don't you think?

There are lots of platforms, there are lots of newton users, but newton is only one man project.

[Julio Jerez]

yes I did, and it compeile all fine in ubunto 10.4
using GCC 4.3

those file has being in teh emake file for a long time

Code: Select all

DG_INCLUDED_PATH = ../../source/core
DG_PATH = $(DG_INCLUDED_PATH)/
DG_SRCS = \
   $(DG_PATH)dgAABBPolygonSoup.cpp \
   $(DG_PATH)dgAsyncThread.cpp \
   $(DG_PATH)dgConvexHull3d.cpp \
   $(DG_PATH)dgConvexHull4d.cpp \

I do not know why it fail to build for you there.

[carli2]

Code: Select all

gcc -c -Wall -Wno-strict-aliasing -D_POSIX_VER -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant -I../../source/core -I../../source/physics -I../../source/meshUtil -I../../source/openCL -I  -o ../../source/core/dgAABBPolygonSoup.o ../../source/core/dgAABBPolygonSoup.cpp


There is a -I without a parameter so "-o" is misinterpreted. That is because DG_AI_PATH is not defined. I removed the AI path and it started to build

Code: Select all

Index: makefile
===================================================================
--- makefile   (revision 2562)
+++ makefile   (working copy)
@@ -182,7 +182,7 @@
 #CPU_FLAGS = -O2 -fpic -g -msse -msse4.1 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant
 CPU_FLAGS = -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant
 
-FLAGS = -c -Wall -Wno-strict-aliasing -D_POSIX_VER $(CPU_FLAGS) -I$(DG_INCLUDED_PATH) -I$(DG_INCLUDED_PHYSICS_PATH) -I$(DG_INCLUDED_MESH_PATH) -I$(DG_INCLUDED_OPENCL_PATH) -I$(DG_AI_PATH)
+FLAGS = -c -Wall -Wno-strict-aliasing -D_POSIX_VER $(CPU_FLAGS) -I$(DG_INCLUDED_PATH) -I$(DG_INCLUDED_PHYSICS_PATH) -I$(DG_INCLUDED_MESH_PATH) -I$(DG_INCLUDED_OPENCL_PATH)
 
 .SUFFIXES : .o .cpp
 .cpp.o :


Next Error:

Code: Select all

gcc -c -Wall -Wno-strict-aliasing -D_POSIX_VER -O2 -fpic -g -msse -msse2 -mfpmath=sse -ffloat-store -ffast-math -freciprocal-math -funsafe-math-optimizations -fsingle-precision-constant -I../../source/core -I../../source/physics -I../../source/meshUtil -I../../source/openCL -o ../../source/physics/dgCollisionBVH.o ../../source/physics/dgCollisionBVH.cpp
../../source/core/dgTypes.cpp: Assembler messages:
../../source/core/dgTypes.cpp:169: Error: invalid instruction suffix for `push'
../../source/core/dgTypes.cpp:172: Error: invalid instruction suffix for `pop'
make: *** [../../source/core/dgTypes.o] Error 1


[carli2]

An other thing I had to correct to make it work with gcc is:

Code: Select all

--- coreLibrary_300/source/core/dgTypes.cpp   (revision 2562)
+++ coreLibrary_300/source/core/dgTypes.cpp   (working copy)
@@ -150,7 +150,7 @@
       void cpuid(dgUnsigned32 op, dgUnsigned32 reg[4])
       {
          asm volatile(
-            #ifdef _MINGW_64_VER
+            #if __x86_64__
                "pushq %%rbx      \n\t" /* save %rbx */
             #else
                "pushl %%ebx      \n\t" /* save %ebx */
@@ -159,7 +159,7 @@
             "cpuid            \n\t"
             "movl %%ebx, %1   \n\t" /* save what cpuid just put in %ebx */
 
-            #ifdef _MINGW_64_VER
+            #if __x86_64__
                "popq %%rbx       \n\t" /* restore the old %rbx */
             #else      
                "popl %%ebx       \n\t" /* restore the old %ebx */


And I had to remove - $(DG_PHYSICS_PATH)dgOpenCL.cpp \ from the compile list.
Now I have a libNewton.so and can test it

[Julio Jerez]

Oh those error are in the 64 bit make file, I did not test that, I made the changes now.
if you can test again it will be greta because I do not have a 64 linux system.

[carli2]

Yes, that works fine for me. Thanks.