From darksharkster at gmail.com Tue Nov 1 06:52:29 2005 From: darksharkster at gmail.com (darkshark atnight) Date: Tue, 1 Nov 2005 11:52:29 +0000 Subject: [quake3] quake3 on Solaris x86 In-Reply-To: References: Message-ID: I've attached the diffs to get things compiled on x86. the lcc/etc/gcc-solaris.c is really the lcc/etc/linux.c renamed and s/linux/sun/g And, I can start quake3 now :) Need to find me cd key, been a few years.. Thanks guys. Sean. . -------------- next part -------------- A non-text attachment was scrubbed... Name: diffs Type: application/octet-stream Size: 5388 bytes Desc: not available URL: From vincent at cojot.name Tue Nov 1 06:58:39 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Tue, 1 Nov 2005 12:58:39 +0100 (CET) Subject: [quake3] quake3 on Solaris x86 In-Reply-To: References: Message-ID: On Tue, 1 Nov 2005, darkshark atnight wrote: Hello Sean, > I've attached the diffs to get things compiled on x86. > > the lcc/etc/gcc-solaris.c is really the lcc/etc/linux.c renamed and > s/linux/sun/g Make sure you checked out the latest SVN, that part was already there.. > And, I can start quake3 now :) Need to find me cd key, been a few years.. Do you mean it is 'working'? Could it be that single "ID_INLINE" that made a difference..? I'm trying it on Sparc now.. Vincent -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From vincent at cojot.name Tue Nov 1 08:20:17 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Tue, 1 Nov 2005 14:20:17 +0100 (CET) Subject: [quake3] Re: Running q3ded on PPC In-Reply-To: References: <20051027221346.02796f81.tim@ngus.net> <1130505187.11098.3.camel@klamath.syndrom23.de> <20051028171532.25dc3a5a.tim@ngus.net> <20051028175842.1b1f8a09.tim@ngus.net> Message-ID: You know, This PPC problem made me realize something as Solaris/Sparc suffers from the same symptoms as PPC (no VM_compiled, using interpreter and then a sig 10 or sig 11). When running q3_endian_test on Solaris/Sparc it resulted in: raistlin at ergoth:[/net/thorbardin/export/home/raistlin ]cat q3_endian.c #include #include #include #include #include #include #include #include #include #include int main( void ) { printf( "Q3 thinks this machine is " ); #if __FLOAT_WORD_ORDER == __LITTLE_ENDIAN printf( "little endian\n" ); #else printf( "big endian\n" ); #endif return 0; } raistlin at ergoth:[/net/thorbardin/export/home/raistlin ]./q3_endian Q3 thinks this machine is little endian Obviously this is wrong for Solaris/Sparc and __FLOAT_WORD_ORDER is nowhere to be defined anyway.. I'll be checking the #ifdef in the VM stuff.. Vincent From vincent at cojot.name Tue Nov 1 09:16:23 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Tue, 1 Nov 2005 15:16:23 +0100 (CET) Subject: [quake3] quake3 on Solaris x86 In-Reply-To: References: Message-ID: Sean, Tim, I have produced a diff against current SVN and based upon Sean's diff. It also includes improved endianness detection for Solaris.. Please tell us if the patch's acceptable.. Vincent On Tue, 1 Nov 2005, darkshark atnight wrote: > I've attached the diffs to get things compiled on x86. > > the lcc/etc/gcc-solaris.c is really the lcc/etc/linux.c renamed and > s/linux/sun/g > > And, I can start quake3 now :) Need to find me cd key, been a few years.. > > Thanks guys. > > Sean. -------------- next part -------------- Index: code/unix/snapvectora.s =================================================================== --- code/unix/snapvectora.s (revision 204) +++ code/unix/snapvectora.s (working copy) @@ -76,7 +76,7 @@ // void Sys_SnapVectorCW( float *v, unsigned short int cw ) .global C(Sys_SnapVectorCW) -C(Sys_SnapVector_cw): +C(Sys_SnapVectorCW): pushl %eax pushl %ebp movl %esp,%ebp Index: code/unix/Makefile =================================================================== --- code/unix/Makefile (revision 204) +++ code/unix/Makefile (working copy) @@ -418,13 +418,13 @@ LDFLAGS=-lsocket -lnsl -ldl -lm ifeq ($(strip $(USE_SDL)),true) - GLLDFLAGS=$(shell sdl-config --libs) -lGLU -lX11 -lXext + GLLDFLAGS=$(shell sdl-config --libs) -L/usr/X11/lib -lGLU -lX11 -lXext else - GLLDFLAGS=-L/usr/openwin/$(LIB) -lGLU -lX11 -lXext + GLLDFLAGS=-L/usr/openwin/$(LIB) -L/usr/X11/lib -lGLU -lX11 -lXext endif ifeq ($(ARCH),i386) - # linux32 make ... + # Solarix x86 make ... BASE_CFLAGS += -m32 LDFLAGS+=-m32 endif Index: code/tools/lcc/etc/gcc-solaris.c =================================================================== --- code/tools/lcc/etc/gcc-solaris.c (revision 204) +++ code/tools/lcc/etc/gcc-solaris.c (working copy) @@ -1,4 +1,4 @@ -/* x86s running Linux */ +/* Solaris with GNU Compilers */ #include @@ -25,7 +25,7 @@ // TTimo experimental: do not compile with the __linux__ define, we are doing bytecode! char *cpp[] = { LCCDIR "q3cpp" BINEXT, "-U__GNUC__", "-D_POSIX_SOURCE", "-D__STDC__=1", "-D__STRICT_ANSI__", - "-Dunix", "-Di386", "-Dlinux", + "-Dunix", "-Di386", "-Dsun", "-D__unix__", "-D__i386__", "-D__signed__=signed", "$1", "$2", "$3", 0 }; char *include[] = {"-I" LCCDIR "include", "-I" LCCDIR "gcc/include", "-I/usr/include", Index: code/renderer/qgl.h =================================================================== --- code/renderer/qgl.h (revision 204) +++ code/renderer/qgl.h (working copy) @@ -94,7 +94,7 @@ #endif /* defined(__sun) */ // NOTE: some Linux platforms would need those prototypes -#if defined(MACOS_X) || defined(__sun) +#if defined(MACOS_X) || ( defined(__sun) && defined(__sparc) ) typedef void (APIENTRY * PFNGLMULTITEXCOORD1DARBPROC) (GLenum target, GLdouble s); typedef void (APIENTRY * PFNGLMULTITEXCOORD1DVARBPROC) (GLenum target, const GLdouble *v); typedef void (APIENTRY * PFNGLMULTITEXCOORD1FARBPROC) (GLenum target, GLfloat s); Index: code/qcommon/q_shared.h =================================================================== --- code/qcommon/q_shared.h (revision 204) +++ code/qcommon/q_shared.h (working copy) @@ -346,6 +346,7 @@ #ifdef __sun #include +#include // bk001205 - from Makefile #define stricmp strcasecmp @@ -360,20 +361,24 @@ #define PATH_SEP '/' -#if defined(_LITTLE_ENDIAN) +#if defined(_BIG_ENDIAN) && !defined(_LITTLE_ENDIAN) +#define BigShort +inline static short LittleShort(short l) { return ShortSwap(l); } +#define BigLong +inline static int LittleLong (int l) { return LongSwap(l); } +#define BigFloat +inline static float LittleFloat (const float l) { return FloatSwap(&l); } + +#elif defined(_LITTLE_ENDIAN) && !defined(_BIG_ENDIAN) inline static short BigShort( short l) { return ShortSwap(l); } #define LittleShort inline static int BigLong(int l) { return LongSwap(l); } #define LittleLong inline static float BigFloat(const float l) { return FloatSwap(&l); } #define LittleFloat -#else /* Must be _BIG_ENDIAN */ -#define BigShort -inline static short LittleShort(short l) { return ShortSwap(l); } -#define BigLong -inline static int LittleLong (int l) { return LongSwap(l); } -#define BigFloat -inline static float LittleFloat (const float l) { return FloatSwap(&l); } + +#else +#error "Either _BIG_ENDIAN or _LITTLE_ENDIAN must be #defined, but not both." #endif #endif @@ -688,7 +693,7 @@ float v[3]; } vec3struct_t; #define VectorCopy(a,b) (*(vec3struct_t *)b=*(vec3struct_t *)a) -#define ID_INLINE static +// #define ID_INLINE static #endif #endif From cgatesman at gmail.com Tue Nov 1 10:25:00 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Tue, 1 Nov 2005 10:25:00 -0500 Subject: [quake3] Re: Running q3ded on PPC In-Reply-To: References: <20051027221346.02796f81.tim@ngus.net> <1130505187.11098.3.camel@klamath.syndrom23.de> <20051028171532.25dc3a5a.tim@ngus.net> <20051028175842.1b1f8a09.tim@ngus.net> Message-ID: <7bd2d8660511010725v6e4aceb5tbeda1b9501e921b6@mail.gmail.com> I get the same thing on my Sparc, but if I modify your code: int main( void ) { #ifndef __FLOAT_WORD_ORDER printf( "Q3 has no clue!\n" ); return 0; #endif printf( "Q3 thinks this machine is " ); #if __FLOAT_WORD_ORDER == __LITTLE_ENDIAN printf( "little endian\n" ); #else printf( "big endian\n" ); #endif return 0; } I get: bacardi:/tmp> ./a.out Q3 has no clue! I just checked a couple other architectures (ai32, Alpha, Power4), and I can't find __FLOAT_WORD_ORDER or __LITTLE_ENDIAN defined on any machine. This is why the test chooses little--because they are both undefined therefore equal one another. Perhaps a more generic #define should be used in the code, and the appropriate value be set for each arch in the Makefile. I guess I was lucky when building on my Alpha as it is little endian. On 11/1/05, vincent at cojot.name wrote: > > > You know, This PPC problem made me realize something as Solaris/Sparc > suffers from the same symptoms as PPC (no VM_compiled, using interpreter > and then a sig 10 or sig 11). > > When running q3_endian_test on Solaris/Sparc it resulted in: > > raistlin at ergoth:[/net/thorbardin/export/home/raistlin ]cat q3_endian.c > #include > #include > #include > #include > #include > #include > #include > #include > #include > #include > > int main( void ) > { > printf( "Q3 thinks this machine is " ); > > #if __FLOAT_WORD_ORDER == __LITTLE_ENDIAN > printf( "little endian\n" ); > #else > printf( "big endian\n" ); > #endif > > return 0; > } > > raistlin at ergoth:[/net/thorbardin/export/home/raistlin ]./q3_endian > Q3 thinks this machine is little endian > > Obviously this is wrong for Solaris/Sparc and __FLOAT_WORD_ORDER is > nowhere to be defined anyway.. I'll be checking the #ifdef in the VM > stuff.. > > Vincent > -- Chad Gatesman http://gatesman.dyndns.org/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From cgatesman at gmail.com Tue Nov 1 10:33:50 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Tue, 1 Nov 2005 10:33:50 -0500 Subject: Alpha Build Message-ID: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> I'm building on Alpha, and I found that vm_interpreted.c includes: #if __WORDSIZE == 64 This is not defined anywhere. Is it a problem for other platforms as well? Also, I was wondering what the status was for the Alpha port as I noticed the following comment in the Makefile: # Linux Alpha (dedicated server only) (TTimo: dropped) I got the code to build and run, but I'm experiencing some hang conditions when multiple people connect. I'm not sure if others had experienced this before, but when I get the time, I'll run it in the debugger, and see what's going on. -- Chad Gatesman http://gatesman.dyndns.org/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim at ngus.net Tue Nov 1 10:57:12 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 1 Nov 2005 15:57:12 +0000 Subject: Running q3ded on PPC In-Reply-To: <7bd2d8660511010725v6e4aceb5tbeda1b9501e921b6@mail.gmail.com> References: <20051027221346.02796f81.tim@ngus.net> <1130505187.11098.3.camel@klamath.syndrom23.de> <20051028171532.25dc3a5a.tim@ngus.net> <20051028175842.1b1f8a09.tim@ngus.net> <7bd2d8660511010725v6e4aceb5tbeda1b9501e921b6@mail.gmail.com> Message-ID: <20051101155712.2daa600a.tim@ngus.net> On Tue, 1 Nov 2005 10:25:00 -0500 Chad wrote: > #ifndef __FLOAT_WORD_ORDER > printf( "Q3 has no clue!\n" ); > return 0; > #endif __FLOAT_WORD_ORDER and __BYTE_ORDER are derived from the header endian.h, which is available on Linux. It is not tested for with the Solaris path however. From tim at ngus.net Tue Nov 1 10:59:40 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 1 Nov 2005 15:59:40 +0000 Subject: Alpha Build In-Reply-To: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> Message-ID: <20051101155940.02e84f22.tim@ngus.net> On Tue, 1 Nov 2005 10:33:50 -0500 Chad wrote: > I'm building on Alpha, and I found that vm_interpreted.c includes: > > #if __WORDSIZE == 64 > > This is not defined anywhere. Is it a problem for other platforms as > well? Not AFAIK. > Also, I was wondering what the status was for the Alpha port as I > noticed the following comment in the Makefile: > > # Linux Alpha (dedicated server only) (TTimo: dropped) That comment is at least 3 years old, probably more, so should be ignored. > I got the code to build and run, but I'm experiencing some hang > conditions when multiple people connect. I'm not sure if others had > experienced this before, but when I get the time, I'll run it in the > debugger, and see what's going on. This is the status of the Alpha port :). Presumably this is Linux? From ludwig.nussel at suse.de Tue Nov 1 11:09:07 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Tue, 1 Nov 2005 17:09:07 +0100 Subject: [quake3] Alpha Build In-Reply-To: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> Message-ID: <200511011709.07425.ludwig.nussel@suse.de> Chad Gatesman wrote: > I'm building on Alpha, and I found that vm_interpreted.c includes: > > #if __WORDSIZE == 64 > > This is not defined anywhere. Is it a problem for other platforms as well? The code that follows that statement needs to be activated for any architecture where sizeof(int) != sizeof(long). __WORDSIZE is defined by some glibc header. If your OS defines this elsewhere feel free to send a patch. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From cgatesman at gmail.com Tue Nov 1 11:43:16 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Tue, 1 Nov 2005 11:43:16 -0500 Subject: [quake3] Re: Alpha Build In-Reply-To: <20051101155940.02e84f22.tim@ngus.net> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> Message-ID: <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> I gotch ya! ;-) I assumed as much. I'm assuming my Alpha 333MHz should be able to handle the dedicated server. It had no problems with the official Q2 port. However, I'm uncertain with this "interpreter" (I don't know how the Q2 was implemented). On 11/1/05, Tim Angus wrote: > > > I got the code to build and run, but I'm experiencing some hang > > conditions when multiple people connect. I'm not sure if others had > > experienced this before, but when I get the time, I'll run it in the > > debugger, and see what's going on. > > This is the status of the Alpha port :). Presumably this is Linux? > -- Chad Gatesman Software Engineer -------------- next part -------------- An HTML attachment was scrubbed... URL: From cgatesman at gmail.com Tue Nov 1 11:59:11 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Tue, 1 Nov 2005 11:59:11 -0500 Subject: [quake3] Alpha Build In-Reply-To: <200511011709.07425.ludwig.nussel@suse.de> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <200511011709.07425.ludwig.nussel@suse.de> Message-ID: <7bd2d8660511010859h3fb3df79g37fdb7cae2ae4d9a@mail.gmail.com> Ah! I see now. It's in which is ultimately picked up by < stdint.h>, but that must not be included in the code. I'll look into it further, and submit a diff. Sorry, I'm new to this list. Is bugzilla or zakk at icculus.org still the prefered way of submitting them as the website states? Or just here? On 11/1/05, Ludwig Nussel wrote: > > Chad Gatesman wrote: > > I'm building on Alpha, and I found that vm_interpreted.c includes: > > > > #if __WORDSIZE == 64 > > > > This is not defined anywhere. Is it a problem for other platforms as > well? > > The code that follows that statement needs to be activated for any > architecture where sizeof(int) != sizeof(long). __WORDSIZE is > defined by some glibc header. If your OS defines this elsewhere feel > free to send a patch. > > cu > Ludwig > > -- > (o_ Ludwig Nussel > //\ SUSE LINUX Products GmbH, Development > V_/_ http://www.suse.de/ > -- Chad Gatesman Software Engineer -------------- next part -------------- An HTML attachment was scrubbed... URL: From ludwig.nussel at suse.de Tue Nov 1 12:08:09 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Tue, 1 Nov 2005 18:08:09 +0100 Subject: [quake3] Alpha Build In-Reply-To: <7bd2d8660511010859h3fb3df79g37fdb7cae2ae4d9a@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <200511011709.07425.ludwig.nussel@suse.de> <7bd2d8660511010859h3fb3df79g37fdb7cae2ae4d9a@mail.gmail.com> Message-ID: <200511011808.09409.ludwig.nussel@suse.de> Chad Gatesman wrote: > Ah! I see now. It's in which is ultimately picked up by < > stdint.h>, but that must not be included in the code. I'll look into it > further, and submit a diff. > > Sorry, I'm new to this list. Is bugzilla or zakk at icculus.org still the > prefered way of submitting them as the website states? Or just here? Personally I prefer to have patches be sent to this list. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From vincent at cojot.name Tue Nov 1 12:12:11 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Tue, 1 Nov 2005 18:12:11 +0100 (CET) Subject: [quake3] quake3 on Solaris x86 In-Reply-To: References: Message-ID: Here's a diff (backported from ioq2 code base) that fixes LoadTGA on Solaris/Sparc. I bet it applies to other arches too.. I'm going to audit the tr_image* code to check for other fixes.. Best regards, PS: Now the client dies in as follows: ----- finished R_Init ----- ------- sound initialization ------- SDL Audio driver initializing... Calling SDL_Init(SDL_INIT_AUDIO)... SDL_Init(SDL_INIT_AUDIO) passed. SDL audio driver is "audio" Format we requested from SDL audio device: Format: AUDIO_S16MSB Freq: 22050 Samples: 512 Channels: 2 [New LWP 1] [New LWP 2] Format we actually got: Format: AUDIO_S16MSB Freq: 22050 Samples: 512 Channels: 2 WARNING: sdlmixsamps wasn't a power of two (10240), so we made it one (16384). Starting SDL audio callback... SDL audio initialized. ------------------------------------ ----- Sound Info ----- 1 stereo 16384 samples 16 samplebits 1 submission_chunk 22050 speed 0x54576a0 dma buffer No background file. ---------------------- Sound memory manager started Loading vm file vm/ui.qvm... Architecture doesn't have a bytecode compiler, using interpreter Program received signal SIGSEGV, Segmentation fault. 0x000be1a0 in VM_PrepareInterpreter (vm=0x9c2efc, header=0x51a4ab8) at ../qcommon/vm_interpreted.c:208 208 codeBase[pc+0] = loadWord(&code[pc]); (gdb) -------------- next part -------------- Index: tr_image.c =================================================================== --- tr_image.c (revision 204) +++ tr_image.c (working copy) @@ -1135,6 +1135,7 @@ byte *buffer; TargaHeader targa_header; byte *targa_rgba; + byte tmp[2]; *pic = NULL; @@ -1152,18 +1153,22 @@ targa_header.colormap_type = *buf_p++; targa_header.image_type = *buf_p++; - targa_header.colormap_index = LittleShort ( *(short *)buf_p ); + tmp[0] = buf_p[0]; + tmp[1] = buf_p[1]; + targa_header.colormap_index = LittleShort ( *((short *)tmp) ); buf_p += 2; - targa_header.colormap_length = LittleShort ( *(short *)buf_p ); + tmp[0] = buf_p[0]; + tmp[1] = buf_p[1]; + targa_header.colormap_length = LittleShort ( *((short *)tmp) ); buf_p += 2; targa_header.colormap_size = *buf_p++; - targa_header.x_origin = LittleShort ( *(short *)buf_p ); + targa_header.x_origin = LittleShort ( *((short *)buf_p) ); buf_p += 2; - targa_header.y_origin = LittleShort ( *(short *)buf_p ); + targa_header.y_origin = LittleShort ( *((short *)buf_p) ); buf_p += 2; - targa_header.width = LittleShort ( *(short *)buf_p ); + targa_header.width = LittleShort ( *((short *)buf_p) ); buf_p += 2; - targa_header.height = LittleShort ( *(short *)buf_p ); + targa_header.height = LittleShort ( *((short *)buf_p) ); buf_p += 2; targa_header.pixel_size = *buf_p++; targa_header.attributes = *buf_p++; Index: qgl.h =================================================================== --- qgl.h (revision 204) +++ qgl.h (working copy) @@ -94,7 +94,7 @@ #endif /* defined(__sun) */ // NOTE: some Linux platforms would need those prototypes -#if defined(MACOS_X) || defined(__sun) +#if defined(MACOS_X) || ( defined(__sun) && defined(__sparc) ) typedef void (APIENTRY * PFNGLMULTITEXCOORD1DARBPROC) (GLenum target, GLdouble s); typedef void (APIENTRY * PFNGLMULTITEXCOORD1DVARBPROC) (GLenum target, const GLdouble *v); typedef void (APIENTRY * PFNGLMULTITEXCOORD1FARBPROC) (GLenum target, GLfloat s); From zakk at timedoctor.org Tue Nov 1 13:27:47 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Tue, 1 Nov 2005 10:27:47 -0800 Subject: [quake3] Alpha Build In-Reply-To: <7bd2d8660511010859h3fb3df79g37fdb7cae2ae4d9a@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <200511011709.07425.ludwig.nussel@suse.de> <7bd2d8660511010859h3fb3df79g37fdb7cae2ae4d9a@mail.gmail.com> Message-ID: <20051101182747.GA12100@timedoctor.org> On Tue, Nov 01, 2005 at 11:59:11AM -0500, Chad Gatesman wrote: > Sorry, I'm new to this list. Is bugzilla or zakk at icculus.org still the > prefered way of submitting them as the website states? Or just here? > From ludwig.nussel at suse.de Tue Nov 1 15:02:37 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Tue, 1 Nov 2005 21:02:37 +0100 Subject: [quake3] quake3 on Solaris x86 In-Reply-To: References: Message-ID: <200511012102.38184.ludwig.nussel@suse.de> vincent at cojot.name wrote: > Here's a diff (backported from ioq2 code base) that fixes LoadTGA on > Solaris/Sparc. I bet it applies to other arches too.. > I'm going to audit the tr_image* code to check for other fixes.. > [...] > Program received signal SIGSEGV, Segmentation fault. > 0x000be1a0 in VM_PrepareInterpreter (vm=0x9c2efc, header=0x51a4ab8) at > ../qcommon/vm_interpreted.c:208 208 > codeBase[pc+0] = loadWord(&code[pc]); (gdb) $SUBJECT says x86 but you are talking about SPARC right? I switched both places to using memcpy rather than ugly casts. I suppose memcpy can deal with unaligned data. Try rev 208. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From tim at ngus.net Tue Nov 1 17:18:44 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 1 Nov 2005 22:18:44 +0000 Subject: Alpha Build In-Reply-To: <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> Message-ID: <20051101221844.32e0086e.tim@ngus.net> On Tue, 1 Nov 2005 11:43:16 -0500 Chad wrote: > I gotch ya! ;-) I assumed as much. > > I'm assuming my Alpha 333MHz should be able to handle the dedicated > server. It had no problems with the official Q2 port. However, I'm > uncertain with this "interpreter" (I don't know how the Q2 was > implemented). Possibly that is touch and go. The interpreter runs the game code quite slowly. You could see if this is the problem by starting q3 as follows: $ ./linuxq3ded +set vm_game 0 +set vm_cgame 0 +set vm_ui 0 +set sv_pure 0 +map q3dm17 This will load the platform libs (ala q2) instead of the bytecode, and should be much faster. From cgatesman at gmail.com Tue Nov 1 17:29:49 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Tue, 1 Nov 2005 17:29:49 -0500 Subject: [quake3] Re: Alpha Build In-Reply-To: <20051101221844.32e0086e.tim@ngus.net> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> Message-ID: <7bd2d8660511011429i32c45eaai7339c06af8450850@mail.gmail.com> They aren't part of the Alpha build. Were they not ported? Do I have some more work ahead of me? ifeq ($(ARCH),axp) TARGETS=\ $(B)/$(PLATFORM)q3ded else TARGETS=\ $(B)/$(PLATFORM)quake3 \ $(B)/$(PLATFORM)q3ded \ $(B)/baseq3/cgame$(ARCH).$(SHLIBEXT) \ $(B)/baseq3/qagame$(ARCH).$(SHLIBEXT) \ $(B)/baseq3/ui$(ARCH).$(SHLIBEXT) \ $(B)/missionpack/cgame$(ARCH).$(SHLIBEXT) \ $(B)/missionpack/qagame$(ARCH).$(SHLIBEXT) \ $(B)/missionpack/ui$(ARCH).$(SHLIBEXT) \ $(B)/baseq3/vm/cgame.qvm \ $(B)/baseq3/vm/qagame.qvm \ $(B)/baseq3/vm/ui.qvm \ $(B)/missionpack/vm/qagame.qvm \ $(B)/missionpack/vm/cgame.qvm \ $(B)/missionpack/vm/ui.qvm # $(B)/$(PLATFORM)quake3-smp \ On 11/1/05, Tim Angus wrote: > > On Tue, 1 Nov 2005 11:43:16 -0500 Chad wrote: > > I gotch ya! ;-) I assumed as much. > > > > I'm assuming my Alpha 333MHz should be able to handle the dedicated > > server. It had no problems with the official Q2 port. However, I'm > > uncertain with this "interpreter" (I don't know how the Q2 was > > implemented). > > Possibly that is touch and go. The interpreter runs the game code quite > slowly. You could see if this is the problem by starting q3 as follows: > > $ ./linuxq3ded +set vm_game 0 +set vm_cgame 0 +set vm_ui 0 +set sv_pure > 0 +map q3dm17 > > This will load the platform libs (ala q2) instead of the bytecode, and > should be much faster. > -- Chad Gatesman Software Engineer -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim at ngus.net Tue Nov 1 17:32:13 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 1 Nov 2005 22:32:13 +0000 Subject: Alpha Build In-Reply-To: <7bd2d8660511011429i32c45eaai7339c06af8450850@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511011429i32c45eaai7339c06af8450850@mail.gmail.com> Message-ID: <20051101223213.397cadeb.tim@ngus.net> On Tue, 1 Nov 2005 17:29:49 -0500 Chad wrote: > They aren't part of the Alpha build. Were they not ported? Do I have > some more work ahead of me? Try: > ifeq ($(ARCH),axp) > TARGETS=\ > $(B)/$(PLATFORM)q3ded \ > $(B)/baseq3/cgame$(ARCH).$(SHLIBEXT) \ > $(B)/baseq3/qagame$(ARCH).$(SHLIBEXT) \ > $(B)/baseq3/ui$(ARCH).$(SHLIBEXT) \ > $(B)/missionpack/cgame$(ARCH).$(SHLIBEXT) \ > $(B)/missionpack/qagame$(ARCH).$(SHLIBEXT) \ > $(B)/missionpack/ui$(ARCH).$(SHLIBEXT) > else > TARGETS=\ > $(B)/$(PLATFORM)quake3 \ > $(B)/$(PLATFORM)q3ded \ > $(B)/baseq3/cgame$(ARCH).$(SHLIBEXT) \ > $(B)/baseq3/qagame$(ARCH).$(SHLIBEXT) \ > $(B)/baseq3/ui$(ARCH).$(SHLIBEXT) \ > $(B)/missionpack/cgame$(ARCH).$(SHLIBEXT) \ > $(B)/missionpack/qagame$(ARCH).$(SHLIBEXT) \ > $(B)/missionpack/ui$(ARCH).$(SHLIBEXT) \ > $(B)/baseq3/vm/cgame.qvm \ > $(B)/baseq3/vm/qagame.qvm \ > $(B)/baseq3/vm/ui.qvm \ > $(B)/missionpack/vm/qagame.qvm \ > $(B)/missionpack/vm/cgame.qvm \ > $(B)/missionpack/vm/ui.qvm > # $(B)/$(PLATFORM)quake3-smp \ From zakk at timedoctor.org Tue Nov 1 22:50:44 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Tue, 1 Nov 2005 19:50:44 -0800 Subject: Aaron Gyes Message-ID: <20051102035043.GA9588@timedoctor.org> FYI: Aaron "Floam" Gyes now has svn access. He's written a web site for us, and it too will go into SVN. At some point a script will be set up similarly to the way the lgfaq updates. That is, every 20 minutes the lgfaq is pulled from svn and applied to the website. This allows for some leeway if you accidentally ci a change that makes the website display a naked picture of Ryan Gordon instead of what you wanted. Aaron probably won't be committing any patches, though I certainly won't mind if he does commit small ones, or larger ones at the behest of myself or others. Please feel free to comment on this, or if you're Ryan "fragmaster" Gordon, set up that script from the next svn checkout that has the web directory in it. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From floam at sh.nu Tue Nov 1 22:59:29 2005 From: floam at sh.nu (Aaron Gyes) Date: Tue, 01 Nov 2005 19:59:29 -0800 Subject: [quake3] Aaron Gyes In-Reply-To: <20051102035043.GA9588@timedoctor.org> References: <20051102035043.GA9588@timedoctor.org> Message-ID: <1130903969.7247.1.camel@localhost> I'd just like to ask for feedback regarding this new XHTML/CSS'ized version of the site. The colors are by no means final, I threw the whole thing together in a few minutes. The only person to okay it so far indeed /is/ colorblind. Don't be afraid to tell me how much [and why] it Sucks. Aaron Gyes From icculus at clutteredmind.org Tue Nov 1 23:42:18 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Tue, 01 Nov 2005 23:42:18 -0500 Subject: [quake3] Aaron Gyes In-Reply-To: <1130903969.7247.1.camel@localhost> References: <20051102035043.GA9588@timedoctor.org> <1130903969.7247.1.camel@localhost> Message-ID: <436843AA.8010602@clutteredmind.org> Aaron Gyes wrote: > I'd just like to ask for feedback regarding this new XHTML/CSS'ized > version of the site. The colors are by no means final, I threw the whole > thing together in a few minutes. The only person to okay it so far > indeed /is/ colorblind. Don't be afraid to tell me how much [and why] it > Sucks. I like it. However, the project is actually making really good progress on the usual icculus.org-version-of-open-sourced-game goals now, so it might be time to get rid of some of the jokes on the website (title, "world of warcraft client", "clap your hands", etc)...but that's just me. zakk, find me in IRC and we'll work out the svn-to-website details when you're ready. floam, nice work. --ryan. From jaq at spacepants.org Tue Nov 1 23:48:22 2005 From: jaq at spacepants.org (Jamie Wilkinson) Date: Wed, 2 Nov 2005 15:48:22 +1100 Subject: [quake3] Aaron Gyes In-Reply-To: <436843AA.8010602@clutteredmind.org> References: <20051102035043.GA9588@timedoctor.org> <1130903969.7247.1.camel@localhost> <436843AA.8010602@clutteredmind.org> Message-ID: <20051102044822.GE23420@spacepants.org> This one time, at band camp, Ryan C. Gordon wrote: >However, the project is actually making really good progress on the >usual icculus.org-version-of-open-sourced-game goals now, so it might be >time to get rid of some of the jokes on the website (title, "world of >warcraft client", "clap your hands", etc)...but that's just me. I like the title :) From zakk at timedoctor.org Wed Nov 2 00:14:01 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Tue, 1 Nov 2005 21:14:01 -0800 Subject: [quake3] Aaron Gyes In-Reply-To: <436843AA.8010602@clutteredmind.org> References: <20051102035043.GA9588@timedoctor.org> <1130903969.7247.1.camel@localhost> <436843AA.8010602@clutteredmind.org> Message-ID: <20051102051401.GB9588@timedoctor.org> On Tue, Nov 01, 2005 at 11:42:18PM -0500, Ryan C. Gordon wrote: > I like it. > > However, the project is actually making really good progress on the > usual icculus.org-version-of-open-sourced-game goals now, so it might be > time to get rid of some of the jokes on the website (title, "world of > warcraft client", "clap your hands", etc)...but that's just me. Anyone else feel this way? The title would change to a normal one too. The project name will never change from icculus.org/quake3 If I can get at least one more person to agree I'll go with it. Otherwise I feel it adds character to the site. > zakk, find me in IRC and we'll work out the svn-to-website details when > you're ready. floam, nice work. > > --ryan. > Ryan set it up after we chatted. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From icculus at clutteredmind.org Wed Nov 2 00:16:13 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Wed, 02 Nov 2005 00:16:13 -0500 Subject: webspace wired to svn repo... Message-ID: <43684B9D.4060209@clutteredmind.org> Ok, zakk and I got it all sorted out...anything that is committed to the web/ directory in the quake3 svn repository shows up immediately on the website...this means anyone with write access to svn can modify the webspace, and we get full revision history to boot. Unlike the lgfaq, which also does this, we do NOT have this on a 20 minute delay...so do not commit anything to web/ that you don't want the world to see immediately. Also, please don't put big files into subversion meant for the webspace...if you, say, package up a multi-megabyte installer for the webspace, then everyone pulling the source from Subversion will have to sit through the download, which is undesirable. If/when there are files like this, please speak to zakk and he will arrange for posting them, outside of the usual webspace and make sure their final URLs are referenced in svn. Otherwise, please commit responsibly, and you know where to find me if there are any issues. --ryan. From floam at sh.nu Wed Nov 2 00:26:42 2005 From: floam at sh.nu (Aaron Gyes) Date: Tue, 01 Nov 2005 21:26:42 -0800 Subject: [quake3] Aaron Gyes In-Reply-To: <20051102051401.GB9588@timedoctor.org> References: <20051102035043.GA9588@timedoctor.org> <1130903969.7247.1.camel@localhost> <436843AA.8010602@clutteredmind.org> <20051102051401.GB9588@timedoctor.org> Message-ID: <1130909202.7691.3.camel@localhost> On Tue, 2005-11-01 at 21:14 -0800, Zachary J. Slater wrote: > On Tue, Nov 01, 2005 at 11:42:18PM -0500, Ryan C. Gordon wrote: > > I like it. > > > > However, the project is actually making really good progress on the > > usual icculus.org-version-of-open-sourced-game goals now, so it might be > > time to get rid of some of the jokes on the website (title, "world of > > warcraft client", "clap your hands", etc)...but that's just me. > > Anyone else feel this way? I think I agree. I think you can probably have character and voice in it without scaring potential users/contributors away. My first website had a lot of character! "WELCOME TO AARONX NETWORK HERE U WILLF IND MY AOL FADERS I HAVE CODED AND MY ELETE MAGIC 8 BALL AND WORD OF THE DAY APPLETS GEOCITES PROVIDES PRESS ALT-F4 TO USE LOOL JUST JK" But I don't know if it was good character. Aaron Gyes From vincent at cojot.name Wed Nov 2 05:24:56 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 2 Nov 2005 11:24:56 +0100 (CET) Subject: Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: <20051101221844.32e0086e.tim@ngus.net> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> Message-ID: On Tue, 1 Nov 2005, Tim Angus wrote: This is just to let you know I got much further with the latest SVN fixes (Thanks Ludwig!) on Solaris/Sparc (unless that's the startup arguments..) raistlin at ergoth:[/usr/local/share/games/quake3 ]gdb ./q3ded GNU gdb 6.2.1 Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "sparc-sun-solaris2.8"...(no debugging symbols found)... (gdb) r +set vm_game 0 +set vm_cgame 0 +set vm_ui 0 +set sv_pure 0 +map q3dm17 +set ttycon 0 Starting program: /usr/local/share/games/quake3/q3ded +set vm_game 0 +set vm_cgame 0 +set vm_ui 0 +set sv_pure 0 +map q3dm17 +set ttycon 0 (no debugging symbols found)...(no debugging symbols found)...warning: Lowest section in /lib/libdl.so.1 is .dynamic at 00000094 (no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...ioQ3 1.33 Solaris-sparc Nov 2 2005 ----- FS_Startup ----- Current search path: /export/home/raistlin/.q3a/baseq3 /usr/local/share/games/quake3/baseq3/pak8.pk3 (9 files) /usr/local/share/games/quake3/baseq3/pak7.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak6.pk3 (64 files) /usr/local/share/games/quake3/baseq3/pak5.pk3 (7 files) /usr/local/share/games/quake3/baseq3/pak4.pk3 (272 files) /usr/local/share/games/quake3/baseq3/pak3.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak2.pk3 (148 files) /usr/local/share/games/quake3/baseq3/pak1.pk3 (26 files) /usr/local/share/games/quake3/baseq3/pak0.pk3 (3539 files) /usr/local/share/games/quake3/baseq3 ---------------------- 4073 files in pk3 files execing default.cfg execing q3config.cfg couldn't exec autoexec.cfg Hunk_Clear: reset the hunk ok --- Common Initialization Complete --- Opening IP socket: localhost:27960 Hostname: ergoth Alias: loghost IP: 192.168.128.2 ------ Server Initialization ------ Server: q3dm17 Hunk_Clear: reset the hunk ok ----- FS_Startup ----- Current search path: /export/home/raistlin/.q3a/baseq3 /usr/local/share/games/quake3/baseq3/pak8.pk3 (9 files) /usr/local/share/games/quake3/baseq3/pak7.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak6.pk3 (64 files) /usr/local/share/games/quake3/baseq3/pak5.pk3 (7 files) /usr/local/share/games/quake3/baseq3/pak4.pk3 (272 files) /usr/local/share/games/quake3/baseq3/pak3.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak2.pk3 (148 files) /usr/local/share/games/quake3/baseq3/pak1.pk3 (26 files) /usr/local/share/games/quake3/baseq3/pak0.pk3 (3539 files) /usr/local/share/games/quake3/baseq3 ---------------------- 8146 files in pk3 files Loading dll file qagame. Sys_LoadDll(/usr/local/share/games/quake3/baseq3/qagamesparc.so)... (no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...(no debugging symbols found)...Sys_LoadDll(/usr/local/share/games/quake3/baseq3/qagamesparc.so): succeeded ... Sys_LoadDll(qagame) found **vmMain** at fee960a4 Sys_LoadDll(qagame) succeeded! ------- Game Initialization ------- gamename: baseq3 gamedate: Nov 2 2005 ------------------------------------------------------------ InitGame: \sv_hostname\noname\sv_maxRate\0\sv_minPing\0\sv_maxPing\0\sv_floodProtect\1\version\ioQ3 1.33 Solaris-sparc Nov 2 2005\dmflags\0\fraglimit\20\timelimit\0\g_gametype\0\protocol\68\mapname\q3dm17\sv_privateClients\0\sv_maxclients\8\sv_allowDownload\0\gamename\baseq3\g_maxGameClients\0\capturelimit\8\g_needpass\0 0 teams with 0 entities 16 items registered ----------------------------------- ------- BotLib Initialization ------- loaded weapons.c loaded items.c Program received signal SIGSEGV, Segmentation fault. 0x000b3f98 in BotLoadSynonyms () (gdb) bt #0 0x000b3f98 in BotLoadSynonyms () #1 0x000ba868 in BotSetupChatAI () #2 0x000d15fc in Export_BotLibSetup () #3 0x00023304 in SV_BotLibSetup () #4 0x0002d7e4 in SV_GameSystemCalls () #5 0x0007bd6c in VM_DllSyscall () #6 0xfef0fce0 in trap_BotLibSetup () from /usr/local/share/games/quake3/baseq3/qagamesparc.so #7 0xfeecb448 in BotInitLibrary () from /usr/local/share/games/quake3/baseq3/qagamesparc.so #8 0xfeecb784 in BotAISetup () from /usr/local/share/games/quake3/baseq3/qagamesparc.so #9 0xfee96de4 in G_InitGame () from /usr/local/share/games/quake3/baseq3/qagamesparc.so #10 0xfee96124 in vmMain () from /usr/local/share/games/quake3/baseq3/qagamesparc.so #11 0x0007cd10 in VM_Call () #12 0x00030fe0 in SV_ShutdownGameProgs () #13 0x000311cc in SV_InitGameProgs () #14 0x0003253c in SV_SpawnServer () #15 0x00029cbc in SV_ExecuteClientMessage () #16 0x000522d0 in Cmd_ExecuteString () #17 0x00051668 in Cbuf_Execute () #18 0x00058754 in Com_Frame () #19 0x000e1e98 in main () (gdb) I'm investigating... Vincent From tim at ngus.net Wed Nov 2 09:25:11 2005 From: tim at ngus.net (Tim Angus) Date: Wed, 2 Nov 2005 14:25:11 +0000 Subject: Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: References: Message-ID: <20051102142511.146b62d2.tim@ngus.net> On Wed, 2 Nov 2005 11:24:56 +0100 (CET) vincent at cojot.name wrote: > This is just to let you know I got much further with the latest SVN > fixes (Thanks Ludwig!) on Solaris/Sparc (unless that's the startup > arguments..) Well, the code path that it crashed in before will be avoided by using +set vm_* 0 on the command line. These should be set to 1 to use the interpreter and 2 to use the JIT compiler (if it exists on your platform, which it doesn't on sparc). > (gdb) r +set vm_game 0 +set vm_cgame 0 +set vm_ui 0 +set sv_pure 0 > +map q3dm17 +set ttycon 0 From cgatesman at gmail.com Wed Nov 2 11:16:26 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Wed, 2 Nov 2005 11:16:26 -0500 Subject: [quake3] Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> Message-ID: <7bd2d8660511020816i2453580dt743d40f991f2a3a8@mail.gmail.com> You should use the executable from a 'make debug' if you want to debug it in gdb. It would be a lot more useful. ;-) -- Chad Gatesman Software Engineer -------------- next part -------------- An HTML attachment was scrubbed... URL: From vincent at cojot.name Wed Nov 2 11:19:49 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 2 Nov 2005 17:19:49 +0100 (CET) Subject: [quake3] Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: <7bd2d8660511020816i2453580dt743d40f991f2a3a8@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511020816i2453580dt743d40f991f2a3a8@mail.gmail.com> Message-ID: On Wed, 2 Nov 2005, Chad Gatesman wrote: BTW, I got the client to come up on Solaris/Sparc: Screenshot here: http://step.polymtl.ca/~coyote/quake3_Sol10sparc.png It still dies when starting a game but I was able to navigate the menus using the keyboard.. The cinematics work and also the introduction screen.. More on this later once I get home (on a more powerful machine).. -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From cgatesman at gmail.com Wed Nov 2 11:20:43 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Wed, 2 Nov 2005 11:20:43 -0500 Subject: [quake3] Re: Alpha Build In-Reply-To: <20051101223213.397cadeb.tim@ngus.net> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511011429i32c45eaai7339c06af8450850@mail.gmail.com> <20051101223213.397cadeb.tim@ngus.net> Message-ID: <7bd2d8660511020820k2602c631wa90fe14ebac0dff0@mail.gmail.com> UPDATE: Thanks for the help, guys. I had to make a few fixes building the game libraries. I got it all built, and the server appears to be running beautifully! Not a single hitch warning like I was getting with the vm_interpreter. I'll try to put together a full patch and submit a Bugzilla PR. Quake 3 on Alpha! On 11/1/05, Tim Angus wrote: > > On Tue, 1 Nov 2005 17:29:49 -0500 Chad wrote: > > They aren't part of the Alpha build. Were they not ported? Do I have > > some more work ahead of me? > > Try: > > > ifeq ($(ARCH),axp) > > TARGETS=\ > > $(B)/$(PLATFORM)q3ded \ > > $(B)/baseq3/cgame$(ARCH).$(SHLIBEXT) \ > > $(B)/baseq3/qagame$(ARCH).$(SHLIBEXT) \ > > $(B)/baseq3/ui$(ARCH).$(SHLIBEXT) \ > > $(B)/missionpack/cgame$(ARCH).$(SHLIBEXT) \ > > $(B)/missionpack/qagame$(ARCH).$(SHLIBEXT) \ > > $(B)/missionpack/ui$(ARCH).$(SHLIBEXT) > > else > > TARGETS=\ > > $(B)/$(PLATFORM)quake3 \ > > $(B)/$(PLATFORM)q3ded \ > > $(B)/baseq3/cgame$(ARCH).$(SHLIBEXT) \ > > $(B)/baseq3/qagame$(ARCH).$(SHLIBEXT) \ > > $(B)/baseq3/ui$(ARCH).$(SHLIBEXT) \ > > $(B)/missionpack/cgame$(ARCH).$(SHLIBEXT) \ > > $(B)/missionpack/qagame$(ARCH).$(SHLIBEXT) \ > > $(B)/missionpack/ui$(ARCH).$(SHLIBEXT) \ > > $(B)/baseq3/vm/cgame.qvm \ > > $(B)/baseq3/vm/qagame.qvm \ > > $(B)/baseq3/vm/ui.qvm \ > > $(B)/missionpack/vm/qagame.qvm \ > > $(B)/missionpack/vm/cgame.qvm \ > > $(B)/missionpack/vm/ui.qvm > > # $(B)/$(PLATFORM)quake3-smp \ > -- Chad Gatesman Software Engineer -------------- next part -------------- An HTML attachment was scrubbed... URL: From vincent at cojot.name Wed Nov 2 11:21:11 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 2 Nov 2005 17:21:11 +0100 (CET) Subject: [quake3] Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: <7bd2d8660511020816i2453580dt743d40f991f2a3a8@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511020816i2453580dt743d40f991f2a3a8@mail.gmail.com> Message-ID: On Wed, 2 Nov 2005, Chad Gatesman wrote: > You should use the executable from a 'make debug' if you want to debug it in > gdb. It would be a lot more useful. ;-) But that's what I did.. :) Otherwise I wouldn't even have gotten the backtrace.. Vincent -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From cgatesman at gmail.com Wed Nov 2 11:26:38 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Wed, 2 Nov 2005 11:26:38 -0500 Subject: [quake3] Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511020816i2453580dt743d40f991f2a3a8@mail.gmail.com> Message-ID: <7bd2d8660511020826p1c6e2d76l54a75717b255c075@mail.gmail.com> You're missing line numbers and arguments. Are you using gcc? If the Makefile contains -g in the CFLAGS, change it to -ggdb (if you're using gcc). On 11/2/05, vincent at cojot.name wrote: > > On Wed, 2 Nov 2005, Chad Gatesman wrote: > > > You should use the executable from a 'make debug' if you want to debug > it in > > gdb. It would be a lot more useful. ;-) > > But that's what I did.. :) Otherwise I wouldn't even have gotten the > backtrace.. > > Vincent > > -- > > ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, > Vincent S. Cojot, Computer Engineering. STEP project. > _.,-*~'`^`'~*-,._.,-*~ > Ecole Polytechnique de Montreal, Comite Micro-Informatique. > _.,-*~'`^`'~*-,. > Linux Xview/OpenLook resources page > _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' > http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name > > They cannot scare me with their empty spaces > Between stars - on stars where no human race is > I have it in me so much nearer home > To scare myself with my own desert places. - Robert Frost > > > -- Chad Gatesman Software Engineer -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim at ngus.net Wed Nov 2 11:29:57 2005 From: tim at ngus.net (Tim Angus) Date: Wed, 2 Nov 2005 16:29:57 +0000 Subject: Alpha Build In-Reply-To: <7bd2d8660511020820k2602c631wa90fe14ebac0dff0@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511011429i32c45eaai7339c06af8450850@mail.gmail.com> <20051101223213.397cadeb.tim@ngus.net> <7bd2d8660511020820k2602c631wa90fe14ebac0dff0@mail.gmail.com> Message-ID: <20051102162957.52f9f740.tim@ngus.net> On Wed, 2 Nov 2005 11:20:43 -0500 Chad wrote: > UPDATE: > > Thanks for the help, guys. I had to make a few fixes building the game > libraries. I got it all built, and the server appears to be running > beautifully! Not a single hitch warning like I was getting with the > vm_interpreter. > > I'll try to put together a full patch and submit a Bugzilla PR. > > Quake 3 on Alpha! Neat. You know on second thought it probably doesn't make a great deal of sense to build the cgame and ui libs when your target only supports the dedicated server, so you could remove these lines for a faster build, if you wanted. > > > $(B)/baseq3/cgame$(ARCH).$(SHLIBEXT) \ > > > $(B)/baseq3/ui$(ARCH).$(SHLIBEXT) \ > > > $(B)/missionpack/cgame$(ARCH).$(SHLIBEXT) \ > > > $(B)/missionpack/ui$(ARCH).$(SHLIBEXT) From cgatesman at gmail.com Wed Nov 2 11:34:57 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Wed, 2 Nov 2005 11:34:57 -0500 Subject: [quake3] Re: Alpha Build In-Reply-To: <20051102162957.52f9f740.tim@ngus.net> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511011429i32c45eaai7339c06af8450850@mail.gmail.com> <20051101223213.397cadeb.tim@ngus.net> <7bd2d8660511020820k2602c631wa90fe14ebac0dff0@mail.gmail.com> <20051102162957.52f9f740.tim@ngus.net> Message-ID: <7bd2d8660511020834p67a1162bh70a5329f15d5de4f@mail.gmail.com> I may move on to the client anyways, though. -Chad On 11/2/05, Tim Angus wrote: > > On Wed, 2 Nov 2005 11:20:43 -0500 Chad wrote: > > UPDATE: > > > > Thanks for the help, guys. I had to make a few fixes building the game > > libraries. I got it all built, and the server appears to be running > > beautifully! Not a single hitch warning like I was getting with the > > vm_interpreter. > > > > I'll try to put together a full patch and submit a Bugzilla PR. > > > > Quake 3 on Alpha! > > Neat. > > You know on second thought it probably doesn't make a great deal of > sense to build the cgame and ui libs when your target only supports the > dedicated server, so you could remove these lines for a faster build, if > you wanted. > > > > > $(B)/baseq3/cgame$(ARCH).$(SHLIBEXT) \ > > > > $(B)/baseq3/ui$(ARCH).$(SHLIBEXT) \ > > > > $(B)/missionpack/cgame$(ARCH).$(SHLIBEXT) \ > > > > $(B)/missionpack/ui$(ARCH).$(SHLIBEXT) > -- Chad Gatesman http://gatesman.dyndns.org/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From tim at ngus.net Wed Nov 2 11:49:21 2005 From: tim at ngus.net (Tim Angus) Date: Wed, 2 Nov 2005 16:49:21 +0000 Subject: Alpha Build In-Reply-To: <7bd2d8660511020834p67a1162bh70a5329f15d5de4f@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511011429i32c45eaai7339c06af8450850@mail.gmail.com> <20051101223213.397cadeb.tim@ngus.net> <7bd2d8660511020820k2602c631wa90fe14ebac0dff0@mail.gmail.com> <20051102162957.52f9f740.tim@ngus.net> <7bd2d8660511020834p67a1162bh70a5329f15d5de4f@mail.gmail.com> Message-ID: <20051102164921.1ca15cb0.tim@ngus.net> On Wed, 2 Nov 2005 11:34:57 -0500 Chad wrote: > I may move on to the client anyways, though. Bear in mind that in order to play on a pure server (which validates which files the client references), the client must use the vm interpreter or compiler, so you may need a faster alpha :). In practice the client native libs (cgame and ui) are only really useful for debugging. From vincent at cojot.name Wed Nov 2 16:02:16 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 2 Nov 2005 22:02:16 +0100 (CET) Subject: [quake3] Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> Message-ID: Ludwig, Your recent commit made it go somewhat further on rev. 232: ---------------------- 8146 files in pk3 files Loading vm file vm/qagame.qvm... qagame loaded in 4431200 bytes on the hunk ------- Game Initialization ------- gamename: baseq3 gamedate: Sep 30 2002 ------------------------------------------------------------ InitGame: \sv_hostname\noname\sv_maxRate\0\sv_minPing\0\sv_maxPing\0\sv_floodProtect\1\version\ioQ3 1.33 Solaris-sparc Nov 2 2005\dmflags\0\fraglimit\20\timelimit\0\g_gametype\0\protocol\68\mapname\q3dm17\sv_privateClients\0\sv_maxclients\8\sv_allowDownload\0\gamename\baseq3\g_maxGameClients\0\capturelimit\8\g_needpass\0 0 teams with 0 entities 16 items registered ----------------------------------- ------- BotLib Initialization ------- loaded weapons.c loaded items.c loaded syn.c Program received signal SIGSEGV, Segmentation fault. 0x000b4ed4 in BotLoadRandomStrings (filename=0x42e934 "rnd.c") at ../botlib/be_ai_chat.c:1022 1022 randomstring->string = ptr; (gdb) bt #0 0x000b4ed4 in BotLoadRandomStrings (filename=0x42e934 "rnd.c") at ../botlib/be_ai_chat.c:1022 #1 0x000ba89c in BotSetupChatAI () at ../botlib/be_ai_chat.c:2964 #2 0x000d15f4 in Export_BotLibSetup () at ../botlib/be_interface.c:159 #3 0x00023304 in SV_BotLibSetup () at ../server/sv_bot.c:454 #4 0x0002d7e4 in SV_GameSystemCalls (args=0x1862ac0) at ../server/sv_game.c:451 #5 0x0007df00 in VM_CallInterpreted (vm=0x2905f8, args=0xffbfa8e8) at ../qcommon/vm_interpreted.c:538 #6 0x0007cd90 in VM_Call (vm=0x2905f8, callnum=0) at ../qcommon/vm.c:767 #7 0x00030fe0 in SV_InitGameVM (restart=qfalse) at ../server/sv_game.c:901 #8 0x000311cc in SV_InitGameProgs () at ../server/sv_game.c:955 #9 0x0003253c in SV_SpawnServer (server=0xffbfeb30 "q3dm17", killBots=qfalse) at ../server/sv_init.c:442 #10 0x00029cbc in SV_Map_f () at ../server/sv_ccmds.c:187 #11 0x000522d0 in Cmd_ExecuteString (text=0xffbfecb0 "map q3dm17 ") at ../qcommon/cmd.c:645 #12 0x00051668 in Cbuf_Execute () at ../qcommon/cmd.c:220 #13 0x00058758 in Com_Frame () at ../qcommon/common.c:2695 #14 0x000e1e90 in main (argc=18, argv=0xffbff65c) at ../unix/unix_main.c:1411 (gdb) Thanks for this!!! :) -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From vincent at cojot.name Wed Nov 2 16:21:48 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 2 Nov 2005 22:21:48 +0100 (CET) Subject: [quake3] Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511020816i2453580dt743d40f991f2a3a8@mail.gmail.com> Message-ID: On Wed, 2 Nov 2005, vincent at cojot.name wrote: > BTW, I got the client to come up on Solaris/Sparc: > > Screenshot here: > http://step.polymtl.ca/~coyote/quake3_Sol10sparc.png Ok, here's the current status that I have on Solaris/Sparc: 1) I cannot run the dedicated server without a crash (investigating). 2) The client "works" with SDL sound but crashes if I try to run a single-player game (probably related to 1). ). 3) I can run a demo quite nicely (If I turn OpenGL extensions off in q3). Heavy screenshot here: http://step.polymtl.ca/~coyote/quake3_Sol10sparc.png 4) I'm doing this with "+set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set sv_pure 1 +set ttycon 0 ".. Still going.... -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From vincent at cojot.name Wed Nov 2 16:46:17 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 2 Nov 2005 22:46:17 +0100 (CET) Subject: [quake3] Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511020816i2453580dt743d40f991f2a3a8@mail.gmail.com> Message-ID: Sorry, Wrong link: http://step.polymtl.ca/~coyote/quake3_Sol10sparc_v4.png BTW, the mouse cannot be used to navigate the menus... -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From ludwig.nussel at suse.de Wed Nov 2 16:48:39 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Wed, 2 Nov 2005 22:48:39 +0100 Subject: [quake3] Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> Message-ID: <200511022248.39749.ludwig.nussel@suse.de> vincent at cojot.name wrote: > On Wed, 2 Nov 2005, vincent at cojot.name wrote: > > BTW, I got the client to come up on Solaris/Sparc: > > > > Screenshot here: > > http://step.polymtl.ca/~coyote/quake3_Sol10sparc.png > > Ok, here's the current status that I have on Solaris/Sparc: > > 1) I cannot run the dedicated server without a crash (investigating). > > 2) The client "works" with SDL sound but crashes if I try to run a > single-player game (probably related to 1). ). I've comitted more alignment fixes of the same kind as the first one. Your tool is helpful to find them. Spotting them just by looking at the code is hard. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From darksharkster at gmail.com Wed Nov 2 17:18:31 2005 From: darksharkster at gmail.com (darkshark atnight) Date: Wed, 2 Nov 2005 22:18:31 +0000 Subject: possible stack corruption with solaris/x86 ? Message-ID: I'm falling behind Vincent now in the solaris stakes :( When starting a game, it crashes. Seems the pointer, cvar_t *showpackets, gets modified somewhere. In VM_Call() the showpackets pointer is ok, it points to good data. Then after VM_Call() does its vm->entryPoint showpackets is changed. Debug info below: (using the v230 release) (Oh mouse doesn't work, keeps itself to the bottom right corner, more on that later) Sean. . Breakpoint 2, VM_Call (vm=0x896f780, callnum=0) at ../qcommon/vm.c:737 737 r = vm->entryPoint( callnum, args[0], args[1], args[2], args[3], (gdb) print showpackets $29 = (cvar_t *) 0x893c528 (gdb) n ------- Game Initialization ------- gamename: baseq3 gamedate: Nov 2 2005 Not logging to disk. Gametype changed, clearing session data. info_spectator_start doesn't have a spawn function info_spectator_start doesn't have a spawn function info_spectator_start doesn't have a spawn function info_spectator_start doesn't have a spawn function 1 teams with 3 entities 14 items registered ----------------------------------- Opened log botlib.log ------- BotLib Initialization ------- loaded weapons.c loaded items.c loaded syn.c loaded rnd.c loaded match.c ^3Warning: file rchat.c, line 635: variables from the match template(s) could be invalid when outputting one of the chat messages ^3Warning: file rchat.c, line 732: the key say with prefix ! is inside the key sayonara ^3Warning: file rchat.c, line 1243: variables from the match template(s) could be invalid when outputting one of the chat messages loaded rchat.c ------------ Map Loading ------------ trying to load maps/q3dm3.aas loaded maps/q3dm3.aas found 24 level items ------------------------------------- 32 bots parsed 30 arenas parsed 771 if ( oldVM != NULL ) // bk001220 - assert(currentVM!=NULL) for oldVM==NULL (gdb) print showpackets $30 = (cvar_t *) 0x30 (gdb) where #0 VM_Call (vm=0x896f780, callnum=0) at ../qcommon/vm.c:771 #1 0x080b41b8 in SV_InitGameVM (restart=qfalse) at ../server/sv_game.c:901 #2 0x080b42cc in SV_InitGameProgs () at ../server/sv_game.c:955 #3 0x080b4ed7 in SV_SpawnServer (server=0x8047360 "q3dm3", killBots=qtrue) at ../server/sv_init.c:442 #4 0x080acd65 in SV_Map_f () at ../server/sv_ccmds.c:187 #5 0x08096a6b in Cmd_ExecuteString (text=0x8047430 "spmap q3dm3") at ../qcommon/cmd.c:645 #6 0x080963f7 in Cbuf_Execute () at ../qcommon/cmd.c:220 #7 0x0809a510 in Com_Frame () at ../qcommon/common.c:2695 #8 0x08143fe0 in main (argc=7, argv=0x8047cdc) at ../unix/unix_main.c:1411 (gdb) delete Delete all breakpoints? (y or n) y (gdb) c Continuing. ----------------------------------- loaded skill 1 from bots/default_c.c loaded skill 1 from bots/orbb_c.c loaded bots/orbb_i.c loaded bots/orbb_w.c loaded orbb from bots/orbb_t.c loaded cached skill 1.000000 from bots/default_c.c loaded skill 1 from bots/mynx_c.c loaded bots/mynx_i.c loaded bots/mynx_w.c loaded mynx from bots/mynx_t.c Program received signal SIGSEGV, Segmentation fault. 0x080a537e in NET_SendPacket (sock=NS_SERVER, length=19, data=0x8022770, to= {type = NA_LOOPBACK, ip = "\000\000\000", ipx = "\000\000\000\000\000\000\000\000\000", port = 0}) at ../qcommon/net_chan.c:622 622 if ( showpackets->integer && *(int *)data == -1 ) { (gdb) c Continuing. Received signal 11, exiting... ----- CL_Shutdown ----- From darksharkster at gmail.com Wed Nov 2 18:01:17 2005 From: darksharkster at gmail.com (darkshark atnight) Date: Wed, 2 Nov 2005 23:01:17 +0000 Subject: [quake3] Some (further) results with latest SVN on Solaris/Sparc In-Reply-To: <200511022248.39749.ludwig.nussel@suse.de> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <200511022248.39749.ludwig.nussel@suse.de> Message-ID: On 11/2/05, Ludwig Nussel wrote: > vincent at cojot.name wrote: > > On Wed, 2 Nov 2005, vincent at cojot.name wrote: > > > BTW, I got the client to come up on Solaris/Sparc: > > > > > > Screenshot here: > > > http://step.polymtl.ca/~coyote/quake3_Sol10sparc.png > > > > Ok, here's the current status that I have on Solaris/Sparc: > > > > 1) I cannot run the dedicated server without a crash (investigating). > > > > 2) The client "works" with SDL sound but crashes if I try to run a > > single-player game (probably related to 1). ). I'm seeing something similar on solaris x86. Though I'm crashing cause the showpackets pointer gets mangled somehow. I'm using v236 here. However, like Vincent I can play the demo ! Though hearing it run with software opengl is rather strange. Sean. . Breakpoint 2, VM_Call (vm=0x896f780, callnum=0) at ../qcommon/vm.c:737 737 r = vm->entryPoint( callnum, args[0], args[1], args[2], args[3], (gdb) print showpackets $29 = (cvar_t *) 0x893c528 (gdb) n ------- Game Initialization ------- gamename: baseq3 gamedate: Nov 2 2005 Not logging to disk. Gametype changed, clearing session data. info_spectator_start doesn't have a spawn function info_spectator_start doesn't have a spawn function info_spectator_start doesn't have a spawn function info_spectator_start doesn't have a spawn function 1 teams with 3 entities 14 items registered ----------------------------------- Opened log botlib.log ------- BotLib Initialization ------- loaded weapons.c loaded items.c loaded syn.c loaded rnd.c loaded match.c ^3Warning: file rchat.c, line 635: variables from the match template(s) could be invalid when outputting one of the chat messages ^3Warning: file rchat.c, line 732: the key say with prefix ! is inside the key sayonara ^3Warning: file rchat.c, line 1243: variables from the match template(s) could be invalid when outputting one of the chat messages loaded rchat.c ------------ Map Loading ------------ trying to load maps/q3dm3.aas loaded maps/q3dm3.aas found 24 level items ------------------------------------- 32 bots parsed 30 arenas parsed 771 if ( oldVM != NULL ) // bk001220 - assert(currentVM!=NULL) for oldVM==NULL (gdb) print showpackets $30 = (cvar_t *) 0x30 (gdb) where #0 VM_Call (vm=0x896f780, callnum=0) at ../qcommon/vm.c:771 #1 0x080b41b8 in SV_InitGameVM (restart=qfalse) at ../server/sv_game.c:901 #2 0x080b42cc in SV_InitGameProgs () at ../server/sv_game.c:955 #3 0x080b4ed7 in SV_SpawnServer (server=0x8047360 "q3dm3", killBots=qtrue) at ../server/sv_init.c:442 #4 0x080acd65 in SV_Map_f () at ../server/sv_ccmds.c:187 #5 0x08096a6b in Cmd_ExecuteString (text=0x8047430 "spmap q3dm3") at ../qcommon/cmd.c:645 #6 0x080963f7 in Cbuf_Execute () at ../qcommon/cmd.c:220 #7 0x0809a510 in Com_Frame () at ../qcommon/common.c:2695 #8 0x08143fe0 in main (argc=7, argv=0x8047cdc) at ../unix/unix_main.c:1411 (gdb) delete Delete all breakpoints? (y or n) y (gdb) c Continuing. ----------------------------------- loaded skill 1 from bots/default_c.c loaded skill 1 from bots/orbb_c.c loaded bots/orbb_i.c loaded bots/orbb_w.c loaded orbb from bots/orbb_t.c loaded cached skill 1.000000 from bots/default_c.c loaded skill 1 from bots/mynx_c.c loaded bots/mynx_i.c loaded bots/mynx_w.c loaded mynx from bots/mynx_t.c Program received signal SIGSEGV, Segmentation fault. 0x080a537e in NET_SendPacket (sock=NS_SERVER, length=19, data=0x8022770, to= {type = NA_LOOPBACK, ip = "\000\000\000", ipx = "\000\000\000\000\000\000\000\000\000", port = 0}) at ../qcommon/net_chan.c:622 622 if ( showpackets->integer && *(int *)data == -1 ) { (gdb) c Continuing. Received signal 11, exiting... > > I've comitted more alignment fixes of the same kind as the first > one. Your tool is helpful to find them. Spotting them just by > looking at the code is hard. > > cu > Ludwig > > -- > (o_ Ludwig Nussel > //\ SUSE LINUX Products GmbH, Development > V_/_ http://www.suse.de/ > > From floam at sh.nu Wed Nov 2 23:11:45 2005 From: floam at sh.nu (Aaron Gyes) Date: Wed, 02 Nov 2005 20:11:45 -0800 Subject: [quake3] Aaron Gyes In-Reply-To: <436843AA.8010602@clutteredmind.org> References: <20051102035043.GA9588@timedoctor.org> <1130903969.7247.1.camel@localhost> <436843AA.8010602@clutteredmind.org> Message-ID: <1130991105.24058.2.camel@localhost> Thought I'd make a note of this here: While the new design is mostly stable, we are working a completely new version of the the website that uses some PHP and has multiple sections, along with completely rewritten everything. It will be here until it's completed and renamed: http://icculus.org/quake3/newindex.php Remember, this is in-progress and it can't be gotten to by outside world. Aaron Gyes From xipher at xipher.us Thu Nov 3 00:13:35 2005 From: xipher at xipher.us (Joshua Hoppes) Date: Wed, 02 Nov 2005 23:13:35 -0600 Subject: [quake3] Aaron Gyes In-Reply-To: <1130991105.24058.2.camel@localhost> References: <20051102035043.GA9588@timedoctor.org> <1130903969.7247.1.camel@localhost> <436843AA.8010602@clutteredmind.org> <1130991105.24058.2.camel@localhost> Message-ID: <43699C7F.6020003@xipher.us> Just took a quick look, and it seems great barring one minor thing. The footer the "Valid ..." part seems to be a bit close to the left edge, if that could be nudged over just a wee bit it might help. Aaron Gyes wrote: > Thought I'd make a note of this here: > > While the new design is mostly stable, we are working a completely new > version of the the website that uses some PHP and has multiple sections, > along with completely rewritten everything. > > It will be here until it's completed and renamed: > http://icculus.org/quake3/newindex.php > > Remember, this is in-progress and it can't be gotten to by outside > world. > > Aaron Gyes > From floam at sh.nu Thu Nov 3 00:33:18 2005 From: floam at sh.nu (Aaron Gyes) Date: Wed, 02 Nov 2005 21:33:18 -0800 Subject: [quake3] Aaron Gyes In-Reply-To: <43699C7F.6020003@xipher.us> References: <20051102035043.GA9588@timedoctor.org> <1130903969.7247.1.camel@localhost> <436843AA.8010602@clutteredmind.org> <1130991105.24058.2.camel@localhost> <43699C7F.6020003@xipher.us> Message-ID: <1130995998.24309.0.camel@localhost> On Wed, 2005-11-02 at 23:13 -0600, Joshua Hoppes wrote: > Just took a quick look, and it seems great barring one minor thing. The > footer the "Valid ..." part seems to be a bit close to the left edge, if > that could be nudged over just a wee bit it might help. Thanks -- I've added some padding now. Aaron Gyes From vincent at cojot.name Thu Nov 3 01:53:55 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Thu, 3 Nov 2005 07:53:55 +0100 (CET) Subject: [quake3] possible stack corruption with solaris/x86 ? In-Reply-To: References: Message-ID: On Wed, 2 Nov 2005, darkshark atnight wrote: > (Oh mouse doesn't work, keeps itself to the bottom right corner, more > on that later) > Sean. Btw, the mouse problem is identical on sparc. I tried compiling with USE_SDL_VIDEO but it was even worse (black screen). I suppose I'll check that one by comparing the mouse handling code in ioq2 (it came from SUN). Vincent From paszczus at gmail.com Thu Nov 3 05:27:41 2005 From: paszczus at gmail.com (PaSzCzUs) Date: Thu, 3 Nov 2005 11:27:41 +0100 Subject: [quake3] Re: Running q3ded on PPC In-Reply-To: <20051101155712.2daa600a.tim@ngus.net> References: <1130505187.11098.3.camel@klamath.syndrom23.de> <20051028171532.25dc3a5a.tim@ngus.net> <20051028175842.1b1f8a09.tim@ngus.net> <7bd2d8660511010725v6e4aceb5tbeda1b9501e921b6@mail.gmail.com> <20051101155712.2daa600a.tim@ngus.net> Message-ID: Hi! Thanks a lot for fixing quake3 at ppc, now server and client works fine :) I bought a amd64 processor (sempron) and i tried to run quake3.. it compiled by q3ded and quake3-game not working :( here is some info: [quake3 at amd64 ~]$ q3ded +map q3dm17 ioQ3 1.33 linux-x86_64 Oct 30 2005 ----- FS_Startup ----- Current search path: /var/games/quake3/.q3a/baseq3 /usr/share/games/quake3/baseq3/pak8.pk3 (9 files) /usr/share/games/quake3/baseq3/pak7.pk3 (4 files) /usr/share/games/quake3/baseq3/pak6.pk3 (64 files) /usr/share/games/quake3/baseq3/pak5.pk3 (7 files) /usr/share/games/quake3/baseq3/pak4.pk3 (272 files) /usr/share/games/quake3/baseq3/pak3.pk3 (4 files) /usr/share/games/quake3/baseq3/pak2.pk3 (148 files) /usr/share/games/quake3/baseq3/pak1.pk3 (26 files) /usr/share/games/quake3/baseq3/pak0.pk3 (3539 files) /usr/share/games/quake3/baseq3 ./baseq3 ---------------------- 4073 files in pk3 files execing default.cfg couldn't exec q3config.cfg couldn't exec autoexec.cfg Hunk_Clear: reset the hunk ok --- Common Initialization Complete --- Opening IP socket: localhost:27960 Hostname: amd64 IP: 87.207.157.109 Started tty console (use +set ttycon 0 to disable) ------ Server Initialization ------ Server: q3dm17 Hunk_Clear: reset the hunk ok ----- FS_Startup ----- Current search path: /var/games/quake3/.q3a/baseq3 /usr/share/games/quake3/baseq3/pak8.pk3 (9 files) /usr/share/games/quake3/baseq3/pak7.pk3 (4 files) /usr/share/games/quake3/baseq3/pak6.pk3 (64 files) /usr/share/games/quake3/baseq3/pak5.pk3 (7 files) /usr/share/games/quake3/baseq3/pak4.pk3 (272 files) /usr/share/games/quake3/baseq3/pak3.pk3 (4 files) /usr/share/games/quake3/baseq3/pak2.pk3 (148 files) /usr/share/games/quake3/baseq3/pak1.pk3 (26 files) /usr/share/games/quake3/baseq3/pak0.pk3 (3539 files) /usr/share/games/quake3/baseq3 ./baseq3 ---------------------- 8146 files in pk3 files Loading vm file vm/qagame.qvm... compiling qagame running assembler < /var/games/quake3/.q3a/baseq3/qagame.s > /var/games/quake3/.q3a/baseq3/qagame.o done computing jump table Received signal 11, exiting... Shutdown tty console and in quake3-game: SDL_SetVideoMode failed: Couldn't find matching GLX visual Xlib: extension "GLX" missing on display ":0.0". Xlib: extension "GLX" missing on display ":0.0". SDL_SetVideoMode failed: Couldn't find matching GLX visual Couldn't get a visual ...WARNING: could not set the given mode (3) ----- CL_Shutdown ----- RE_Shutdown( 1 ) Received signal 11, exiting... ----- CL_Shutdown ----- recursive shutdown ^ - but maybe here it`s my fault? From ludwig.nussel at suse.de Thu Nov 3 05:49:49 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Thu, 3 Nov 2005 11:49:49 +0100 Subject: [quake3] Q3 x86_64 (was: Re: Running q3ded on PPC) In-Reply-To: References: <1130505187.11098.3.camel@klamath.syndrom23.de> <20051028171532.25dc3a5a.tim@ngus.net> <20051028175842.1b1f8a09.tim@ngus.net> <7bd2d8660511010725v6e4aceb5tbeda1b9501e921b6@mail.gmail.com> <20051101155712.2daa600a.tim@ngus.net> Message-ID: <20051103104949.GA7477@suse.de> PaSzCzUs wrote: > Hi! Thanks a lot for fixing quake3 at ppc, now server and client works fine :) > > I bought a amd64 processor (sempron) and i tried to run quake3.. it > compiled by q3ded and quake3-game not working :( here is some info: Please start a new thread if you are talking about a different issue. > [...] > Loading vm file vm/qagame.qvm... > compiling qagame > running assembler < /var/games/quake3/.q3a/baseq3/qagame.s > > /var/games/quake3/.q3a/baseq3/qagame.o > done > computing jump table > Received signal 11, exiting... > Shutdown tty console The vm compiler is disabled by default on purpose. It works for me though (except for bots). Which distro/compiler/assembler/CFLAGS are you using? cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From vincent at cojot.name Thu Nov 3 06:10:22 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Thu, 3 Nov 2005 12:10:22 +0100 (CET) Subject: More debut traces from quake3 on Solaris/Sparc through purify Message-ID: Hi everyone, Hi Ludwig, Tim and Sean, I just ran rev. 248 of q3ded through Purify on Sparc. It went much further this time but it still crashed with bots on q3dm17. I have attached three logs: quake3_v248_log.txt is the complete log from the console. quake3_v248_purify.txt is a copy from the relevant parts of the purify files. quake3_v248_gdb.txt is a copy from the matching gdb run. I hope this helps.. I'm looking into this also but since I'm no pro (code-wise) I will take longer to diagnose something from these logs so I'd rather share them early.. :) -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost -------------- next part -------------- gdb /export/home/vcojot/SVN/icculus/quake3-Solaris/code/unix/debugs parc/SunOSq3ded GNU gdb 6.2.1 Copyright 2004 Free Software Foundation, Inc. GDB is free software, covered by the GNU General Public License, and you are welcome to change it and/or distribute copies of it under certain conditions. Type "show copying" to see the conditions. There is absolutely no warranty for GDB. Type "show warranty" for details. This GDB was configured as "sparc-sun-solaris2.10"... (gdb) r +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set sv_pure 1 +set ttycon 0 +map q3dm17 Starting program: /export/home/vcojot/SVN/icculus/quake3-Solaris/code/unix/debugsparc/SunOSq3ded +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set sv_pure 1 +set ttycon 0 +map q3dm17 warning: Lowest section in /lib/libdl.so.1 is .dynamic at 00000094 ioQ3 1.33 Solaris-sparc Nov 3 2005 ----- FS_Startup ----- Current search path: /export/home/vcojot/.q3a/baseq3 /usr/local/share/games/quake3/baseq3/pak8.pk3 (9 files) /usr/local/share/games/quake3/baseq3/pak7.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak6.pk3 (64 files) /usr/local/share/games/quake3/baseq3/pak5.pk3 (7 files) /usr/local/share/games/quake3/baseq3/pak4.pk3 (272 files) /usr/local/share/games/quake3/baseq3/pak3.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak2.pk3 (148 files) /usr/local/share/games/quake3/baseq3/pak1.pk3 (26 files) /usr/local/share/games/quake3/baseq3/pak0.pk3 (3539 files) /usr/local/share/games/quake3/baseq3 /export/home/vcojot/SVN/icculus/quake3-Solaris/code/unix/debugsparc/baseq3 ---------------------- 4073 files in pk3 files execing default.cfg execing q3config.cfg couldn't exec autoexec.cfg Hunk_Clear: reset the hunk ok --- Common Initialization Complete --- Opening IP socket: localhost:27960 Hostname: flotsam Alias: flotsam. Alias: loghost IP: 10.179.104.54 ------ Server Initialization ------ Server: q3dm17 Hunk_Clear: reset the hunk ok ----- FS_Startup ----- Current search path: /export/home/vcojot/.q3a/baseq3 /usr/local/share/games/quake3/baseq3/pak8.pk3 (9 files) /usr/local/share/games/quake3/baseq3/pak7.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak6.pk3 (64 files) /usr/local/share/games/quake3/baseq3/pak5.pk3 (7 files) /usr/local/share/games/quake3/baseq3/pak4.pk3 (272 files) /usr/local/share/games/quake3/baseq3/pak3.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak2.pk3 (148 files) /usr/local/share/games/quake3/baseq3/pak1.pk3 (26 files) /usr/local/share/games/quake3/baseq3/pak0.pk3 (3539 files) /usr/local/share/games/quake3/baseq3 /export/home/vcojot/SVN/icculus/quake3-Solaris/code/unix/debugsparc/baseq3 ---------------------- 8146 files in pk3 files Loading vm file vm/qagame.qvm... qagame loaded in 4431200 bytes on the hunk ------- Game Initialization ------- gamename: baseq3 gamedate: Sep 30 2002 ------------------------------------------------------------ InitGame: \sv_floodProtect\1\sv_maxPing\0\sv_minPing\0\sv_maxRate\0\sv_hostname\noname\version\ioQ3 1.33 Solaris-sparc Nov 3 2005\dmflags\0\fraglimit\20\timelimit\0\g_gametype\0\protocol\68\mapname\q3dm17\sv_privateClients\0\sv_maxclients\8\sv_allowDownload\0\gamename\baseq3\g_maxGameClients\0\capturelimit\8\g_needpass\0 0 teams with 0 entities 16 items registered ----------------------------------- ------- BotLib Initialization ------- loaded weapons.c loaded items.c loaded syn.c loaded rnd.c loaded match.c loaded rchat.c ------------ Map Loading ------------ trying to load maps/q3dm17.aas loaded maps/q3dm17.aas found 29 level items ------------------------------------- 32 bots parsed 35 arenas parsed Program received signal SIGSEGV, Segmentation fault. 0x000a4a24 in AAS_CalculateAreaTravelTimes () at ../botlib/be_aas_route.c:543 543 aasworld.areatraveltimes[i][l] = (unsigned short *) ptr; (gdb) bt #0 0x000a4a24 in AAS_CalculateAreaTravelTimes () at ../botlib/be_aas_route.c:543 #1 0x000a6660 in AAS_InitRouting () at ../botlib/be_aas_route.c:1229 #2 0x0008abc0 in AAS_ContinueInit (time=0.100000001) at ../botlib/be_aas_main.c:227 #3 0x0008ac08 in AAS_StartFrame (time=0.100000001) at ../botlib/be_aas_main.c:246 #4 0x000d17c0 in Export_BotLibStartFrame (time=0.100000001) at ../botlib/be_interface.c:247 #5 0x0002dac0 in SV_GameSystemCalls (args=0x1862f04) at ../server/sv_game.c:471 #6 0x0007ded0 in VM_CallInterpreted (vm=0x2905b0, args=0xffbfa528) at ../qcommon/vm_interpreted.c:538 #7 0x0007cd60 in VM_Call (vm=0x2905b0, callnum=10) at ../qcommon/vm.c:766 #8 0x00023280 in SV_BotFrame (time=100) at ../server/sv_bot.c:436 #9 0x000325a8 in SV_SpawnServer (server=0xffbfe6d0 "q3dm17", killBots=qfalse) at ../server/sv_init.c:451 #10 0x00029cbc in SV_Map_f () at ../server/sv_ccmds.c:187 #11 0x000522d0 in Cmd_ExecuteString (text=0xffbfe850 "map q3dm17") at ../qcommon/cmd.c:645 #12 0x00051668 in Cbuf_Execute () at ../qcommon/cmd.c:220 #13 0x00058758 in Com_Frame () at ../qcommon/common.c:2695 #14 0x000e1e48 in main (argc=18, argv=0xffbff1fc) at ../unix/unix_main.c:1411 (gdb) -------------- next part -------------- Finished SunOSq3ded ( 6 errors, 1072 leaked bytes) Purify instrumented /export/home/vcojot/SVN/icculus/quake3-Solaris/code/unix/debugsparc/SunOSq3ded (pid 19531 at Thu Nov 3 11:18:05 2005) Purify 2003a.06.15 FixPack 0188 050822 Solaris 2 (32-bit) (c) Copyright IBM Corp. 1992, 2005 All rights reserved. For contact information type: "purify -help" For TTY output, use the option "-windows=no" Options settings: -purify \ -purify-home=/opt/Rational/releases/purify.sol.2003a.06.15.FixPack.0188 \ -gcc3_path=/usr/sfw/bin/gcc \ -cache-dir=/opt/Rational/releases/purify.sol.2003a.06.15.FixPack.0188/cache \ -threads=yes -use-internal-locks=yes -thread_stack_change=0x4000 \ -mt_safe_malloc=yes License successfully checked out. Command-line: \ /export/home/vcojot/SVN/icculus/quake3-Solaris/code/unix/debugsparc/SunOSq3ded \ +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set sv_pure 1 +set ttycon 0 SBR: Stack array bounds read (2 times) This is occurring while in: VM_Call [vm.c:758] a.callnum = callnum; va_start(ap, callnum); for (i = 0; i < sizeof (a.args) / sizeof (a.args[0]); i++) { => a.args[i] = va_arg(ap, long); } va_end(ap); #ifndef NO_VM_COMPILED SV_BotFrame [sv_bot.c:436] if (!bot_enable) return; //NOTE: maybe the game is already shutdown if (!gvm) return; => VM_Call( gvm, BOTAI_START_FRAME, time ); } /* SV_SpawnServer [sv_init.c:451] for (i = 0;i < 3; i++) { VM_Call (gvm, GAME_RUN_FRAME, sv.time); => SV_BotFrame (sv.time); sv.time += 100; svs.time += 100; } SV_Map_f [sv_ccmds.c:187] Q_strncpyz(mapname, map, sizeof(mapname)); // start up the map => SV_SpawnServer( mapname, killBots ); // set the cheat value // if the level was started with "map ", then Cmd_ExecuteString [cmd.c:645] // let the cgame or game handle it break; } else { => cmd->function (); } return; } Cbuf_Execute [cmd.c:220] // execute the command line => Cmd_ExecuteString (line); } } Reading 4 bytes from 0xffbfa5b0. Frame pointer 0xffbfa5b0 Address 0xffbfa5b0 is 0 bytes above stack pointer in function SV_SpawnServer. MAW: Misaligned memory write This is occurring while in: AAS_CalculateAreaTravelTimes [be_aas_route.c:543] // for (l = 0; l < settings->numreachableareas; l++) { => aasworld.areatraveltimes[i][l] = (unsigned short *) ptr; ptr += revreach->numlinks * sizeof(unsigned short); //reachability link reach = &aasworld.reachability[settings->firstreachablearea + l]; AAS_InitRouting [be_aas_route.c:1229] //initialize portal cache AAS_InitPortalCache(); //initialize the area travel times => AAS_CalculateAreaTravelTimes(); //calculate the maximum travel times through portals AAS_InitPortalMaxTravelTimes(); //get the areas reachabilities go through AAS_ContinueInit [be_aas_main.c:227] } //end else } //end if //initialize the routing => AAS_InitRouting(); //at this point AAS is initialized AAS_SetInitialized(); } //end of the function AAS_ContinueInit AAS_StartFrame [be_aas_main.c:246] //invalidate the entities AAS_InvalidateEntities(); //initialize AAS => AAS_ContinueInit(time); // aasworld.frameroutingupdates = 0; // Export_BotLibStartFrame [be_interface.c:247] int Export_BotLibStartFrame(float time) { if (!BotLibSetup("BotStartFrame")) return BLERR_LIBRARYNOTSETUP; => return AAS_StartFrame(time); } //end of the function Export_BotLibStartFrame //=========================================================================== // SV_GameSystemCalls [sv_game.c:471] return botlib_export->PC_SourceFileAndLine( args[1], VMA(2), VMA(3) ); case BOTLIB_START_FRAME: => return botlib_export->BotLibStartFrame( VMF(1) ); case BOTLIB_LOAD_MAP: return botlib_export->BotLibLoadMap( VMA(1) ); case BOTLIB_UPDATENTITY: Misaligned write of 4 bytes to 0x57b44e in the heap. Address 0x57b44e is 1338358 bytes into a malloc'd block at 0x434858 of 16777216 bytes. This block was allocated from: malloc [rtlib.o] calloc [rtlib.o] Com_InitZoneMemory [common.c:1396] } // bk001205 - was malloc => mainzone = calloc( s_zoneTotal, 1 ); if ( !mainzone ) { Com_Error( ERR_FATAL, "Zone data failed to allocate %i megs", s_zoneTotal / (1024*1024) ); } Com_Init [common.c:2370] // Swap_Init (); Cbuf_Init (); => Com_InitZoneMemory(); Cmd_Init (); // override anything from the config files with command line args main [unix_main.c:1394] memset( &eventQue[0], 0, MAX_QUED_EVENTS*sizeof(sysEvent_t) ); memset( &sys_packetReceived[0], 0, MAX_MSGLEN*sizeof(byte) ); => Com_Init(cmdline); NET_Init(); Sys_ConsoleInputInit(); _start [crt1.o] SIG: Signal handled This is occurring while in: AAS_CalculateAreaTravelTimes [be_aas_route.c:543] AAS_InitRouting [be_aas_route.c:1229] AAS_ContinueInit [be_aas_main.c:227] AAS_StartFrame [be_aas_main.c:246] Export_BotLibStartFrame [be_interface.c:247] SV_GameSystemCalls [sv_game.c:471] Received signal 10 (SIGBUS - Bus Error) Faulting address = 0x57b44e Signal mask: (SIGHUP | SIGINT | SIGQUIT | SIGILL | SIGTRAP | \ SIGABRT | SIGEMT | SIGFPE | SIGSEGV | SIGSYS | SIGPIPE | SIGALRM | \ SIGTERM | SIGUSR1 | SIGUSR2 | SIGCHLD | SIGPWR | SIGWINCH | SIGURG | \ SIGPOLL | SIGTSTP | SIGCONT | SIGTTIN | SIGTTOU | SIGVTALRM | \ SIGPROF | SIGXCPU | SIGXFSZ | SIGWAITING | SIGLWP) Pending signals: UMR: Uninitialized memory read This is occurring while in: __systemcall6 [libc.so.1] call_user_handler [libc.so.1] pure_signal_handler_wrapper [rtlib.o] pure_sigtramp [rtlib.o] AAS_InitRouting [be_aas_route.c:1229] //initialize portal cache AAS_InitPortalCache(); //initialize the area travel times => AAS_CalculateAreaTravelTimes(); //calculate the maximum travel times through portals AAS_InitPortalMaxTravelTimes(); //get the areas reachabilities go through AAS_ContinueInit [be_aas_main.c:227] } //end else } //end if //initialize the routing => AAS_InitRouting(); //at this point AAS is initialized AAS_SetInitialized(); } //end of the function AAS_ContinueInit Reading 4 bytes from 0xffbf979c on the stack. Address 0xffbf979c is 12 bytes below frame pointer in function __systemcall6. UMR: Uninitialized memory read This is occurring while in: __systemcall6 [libc.so.1] call_user_handler [libc.so.1] pure_signal_handler_wrapper [rtlib.o] pure_sigtramp [rtlib.o] AAS_InitRouting [be_aas_route.c:1229] AAS_ContinueInit [be_aas_main.c:227] Reading 4 bytes from 0xffbf97a0 on the stack. Address 0xffbf97a0 is 8 bytes below frame pointer in function __systemcall6. Current file descriptors in use: 18 FIU: file descriptor 0: FIU: file descriptor 1: FIU: file descriptor 2: FIU: file descriptor 3: "/usr/local/share/games/quake3/baseq3/pak0.pk3", O_RDONLY File info: -rw-r--r-- 1 root root 479493658 Sep 19 2002 File position: 479360651 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] unzOpen [unzip.c:1330] FS_LoadZipFile [files.c:1694] FIU: file descriptor 4: "/usr/local/share/games/quake3/baseq3/pak1.pk3", O_RDONLY File info: -rw-r--r-- 1 root root 374405 Oct 7 2003 File position: 374405 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] unzOpen [unzip.c:1330] FS_LoadZipFile [files.c:1694] FIU: file descriptor 5: FIU: file descriptor 6: "/usr/local/share/games/quake3/baseq3/pak2.pk3", O_RDONLY File info: -rw-r--r-- 1 root root 7511182 Oct 7 2003 File position: 7511182 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] unzOpen [unzip.c:1330] FS_LoadZipFile [files.c:1694] FIU: file descriptor 7: "/usr/local/share/games/quake3/baseq3/pak3.pk3", O_RDONLY File info: -rw-r--r-- 1 root root 276305 Oct 7 2003 File position: 276305 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] unzOpen [unzip.c:1330] FS_LoadZipFile [files.c:1694] FIU: file descriptor 8: "/usr/local/share/games/quake3/baseq3/pak4.pk3", O_RDONLY File info: -rw-r--r-- 1 root root 9600350 Oct 7 2003 File position: 9600350 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] unzOpen [unzip.c:1330] FS_LoadZipFile [files.c:1694] FIU: file descriptor 9: "/usr/local/share/games/quake3/baseq3/pak5.pk3", O_RDONLY File info: -rw-r--r-- 1 root root 191872 Oct 7 2003 File position: 191872 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] unzOpen [unzip.c:1330] FS_LoadZipFile [files.c:1694] FIU: file descriptor 10: "/usr/local/share/games/quake3/baseq3/pak6.pk3", O_RDONLY File info: -rw-r--r-- 1 root root 7346884 Oct 7 2003 File position: 7346884 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] unzOpen [unzip.c:1330] FS_LoadZipFile [files.c:1694] FIU: file descriptor 11: "/usr/local/share/games/quake3/baseq3/pak7.pk3", O_RDONLY File info: -rw-r--r-- 1 root root 320873 Oct 7 2003 File position: 320873 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] unzOpen [unzip.c:1330] FS_LoadZipFile [files.c:1694] FIU: file descriptor 12: "/usr/local/share/games/quake3/baseq3/pak8.pk3", O_RDONLY File info: -rw-r--r-- 1 root root 454478 Oct 7 2003 File position: 454478 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] unzOpen [unzip.c:1330] FS_LoadZipFile [files.c:1694] FIU: file descriptor 13: "/var/run/name_service_door", O_RDONLY File info: ?r--r--r-- 1 root root 0 Oct 25 11:30 This file descriptor was allocated from: __open64 [libc.so.1] _open64 [libc.so.1] _nsc_trydoorcall [libc.so.1] getpwuid_r [libc.so.1] Sys_GetCurrentUser [unix_shared.c:422] Sys_Init [unix_main.c:416] FIU: file descriptor 14: , O_RDWR This file descriptor was allocated from: _so_socket [libc.so.1] _socket_create [libsocket.so.1] NET_IPSocket [unix_net.c:546] NET_OpenIP [unix_net.c:502] NET_Init [unix_net.c:523] main [unix_main.c:1395] FIU: file descriptor 15: "/export/home/vcojot/.q3a/baseq3/games.log", O_WRONLY | O_APPEND | O_CREAT File info: -rw-r--r-- 1 vcojot other 2031 Nov 3 11:07 File position: 2031 This file descriptor was allocated from: __open [libc.so.1] _open [libc.so.1] _endopen [libc.so.1] fopen [libc.so.1] FS_FOpenFileAppend [files.c:916] FS_FOpenFileByMode [files.c:3374] FIU: file descriptor 26: FIU: file descriptor 27: Memory leaked: 1072 bytes (0.00141%); potentially leaked: 58720287 bytes (77.1%) MLK: 1072 bytes leaked at 0x4c48ce8 This memory was allocated from: malloc [rtlib.o] _nss_XbyY_buf_alloc [libc.so.1] get_pwbuf [libc.so.1] getpwuid [libc.so.1] Sys_GetCurrentUser [unix_shared.c:422] { struct passwd *p; => if ( (p = getpwuid( getuid() )) == NULL ) { return "player"; } return p->pw_name; Sys_Init [unix_main.c:416] Cvar_Set( "arch", "unknown" ); #endif => Cvar_Set( "username", Sys_GetCurrentUser() ); //IN_Init(); // rcg08312005 moved into glimp. PLK: 58720287 bytes potentially leaked at 0x1448c70 This memory was allocated from: malloc [rtlib.o] calloc [rtlib.o] Com_InitHunkMemory [common.c:1522] Com_Init [common.c:2407] main [unix_main.c:1394] _start [crt1.o] Purify Heap Analysis (combining suppressed and unsuppressed blocks) Blocks Bytes Leaked 1 1072 Potentially Leaked 11 58802287 In-Use 53 17310917 ---------------------------------------- Total Allocated 65 76114276 Thread Summary : 1 threads in existence Thread 0 [main thread] Stack Limit : (0xff400000 0xffc00000), size = 0x800000 Program exited with status code 0. * Basic memory usage (including Purify overhead): 1972520 code 1781244 data/bss 76365828 heap (peak use) 27032 stack * Shared library memory usage (including Purify overhead): 1412 libpure_solaris2_init.so (shared code) 252 libpure_solaris2_init.so (private data) 61648 libsocket.so.1_pure_p3_c0_108222112_510_32 (shared code) 5152 libsocket.so.1_pure_p3_c0_108222112_510_32 (private data) 936280 libnsl.so.1_pure_p3_c0_108222112_510_32 (shared code) 64036 libnsl.so.1_pure_p3_c0_108222112_510_32 (private data) 1928 libdl.so.1_pure_p3_c0_108222112_510_32 (shared code) 76 libdl.so.1_pure_p3_c0_108222112_510_32 (private data) 1924448 libm.so.2_pure_p3_c0_108222112_510_32 (shared code) 23328 libm.so.2_pure_p3_c0_108222112_510_32 (private data) 1390024 libc.so.1_pure_p3_c0_108222112_510_32 (shared code) 35196 libc.so.1_pure_p3_c0_108222112_510_32 (private data) 15416 libinternal_stubs.so.1 (shared code) 1044 libinternal_stubs.so.1 (private data) * Memory mapped usage: 24576 mmap'd at 0xff3a0000 -------------- next part -------------- /usr/local/share/games/quake3/start_q3.sh ioQ3 1.33 Solaris-sparc Nov 3 2005 ----- FS_Startup ----- Current search path: /export/home/vcojot/.q3a/baseq3 /usr/local/share/games/quake3/baseq3/pak8.pk3 (9 files) /usr/local/share/games/quake3/baseq3/pak7.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak6.pk3 (64 files) /usr/local/share/games/quake3/baseq3/pak5.pk3 (7 files) /usr/local/share/games/quake3/baseq3/pak4.pk3 (272 files) /usr/local/share/games/quake3/baseq3/pak3.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak2.pk3 (148 files) /usr/local/share/games/quake3/baseq3/pak1.pk3 (26 files) /usr/local/share/games/quake3/baseq3/pak0.pk3 (3539 files) /usr/local/share/games/quake3/baseq3 /export/home/vcojot/SVN/icculus/quake3-Solaris/code/unix/debugsparc/baseq3 ---------------------- 4073 files in pk3 files execing default.cfg execing q3config.cfg couldn't exec autoexec.cfg Hunk_Clear: reset the hunk ok --- Common Initialization Complete --- Opening IP socket: localhost:27960 Hostname: flotsam Alias: flotsam. Alias: loghost IP: 10.179.104.54 Hitch warning: 1261 msec frame time map q3dm17 ------ Server Initialization ------ Server: q3dm17 Hunk_Clear: reset the hunk ok ----- FS_Startup ----- Current search path: /export/home/vcojot/.q3a/baseq3 /usr/local/share/games/quake3/baseq3/pak8.pk3 (9 files) /usr/local/share/games/quake3/baseq3/pak7.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak6.pk3 (64 files) /usr/local/share/games/quake3/baseq3/pak5.pk3 (7 files) /usr/local/share/games/quake3/baseq3/pak4.pk3 (272 files) /usr/local/share/games/quake3/baseq3/pak3.pk3 (4 files) /usr/local/share/games/quake3/baseq3/pak2.pk3 (148 files) /usr/local/share/games/quake3/baseq3/pak1.pk3 (26 files) /usr/local/share/games/quake3/baseq3/pak0.pk3 (3539 files) /usr/local/share/games/quake3/baseq3 /export/home/vcojot/SVN/icculus/quake3-Solaris/code/unix/debugsparc/baseq3 ---------------------- 8146 files in pk3 files Loading vm file vm/qagame.qvm... qagame loaded in 4431200 bytes on the hunk ------- Game Initialization ------- gamename: baseq3 gamedate: Sep 30 2002 ------------------------------------------------------------ InitGame: \sv_hostname\noname\sv_maxRate\0\sv_minPing\0\sv_maxPing\0\sv_floodProtect\1\version\ioQ3 1.33 Solaris-sparc Nov 3 2005\dmflags\0\fraglimit\20\timelimit\0\g_gametype\0\protocol\68\mapname\q3dm17\sv_privateClients\0\sv_maxclients\8\sv_allowDownload\0\gamename\baseq3\g_maxGameClients\0\capturelimit\8\g_needpass\0 0 teams with 0 entities 16 items registered ----------------------------------- ------- BotLib Initialization ------- loaded weapons.c loaded items.c loaded syn.c loaded rnd.c loaded match.c loaded rchat.c ------------ Map Loading ------------ trying to load maps/q3dm17.aas loaded maps/q3dm17.aas found 29 level items ------------------------------------- 32 bots parsed 35 arenas parsed Received signal 10, exiting... vcojot at flotsam:[~][505]$ exit From paszczus at gmail.com Thu Nov 3 12:51:26 2005 From: paszczus at gmail.com (PaSzCzUs) Date: Thu, 3 Nov 2005 18:51:26 +0100 Subject: [quake3] Q3 x86_64 Message-ID: > > [...] > > Loading vm file vm/qagame.qvm... > > compiling qagame > > running assembler < /var/games/quake3/.q3a/baseq3/qagame.s > > > /var/games/quake3/.q3a/baseq3/qagame.o > > done > > computing jump table > > Received signal 11, exiting... > > Shutdown tty console > > The vm compiler is disabled by default on purpose. It works for me > though (except for bots). Which distro/compiler/assembler/CFLAGS are > you using? I`m using PLD Linux Distribution with gcc 3.3.6 and compiled using: CFLAGS="%{rpmcflags}" CFLAGS="$CFLAGS -DDEFAULT_BASEDIR=\\\"%{_datadir}/games/%{name}\\\"" CFLAGS="$CFLAGS -Wall -Wimplicit -Wstrict-prototypes" CFLAGS="$CFLAGS -DUSE_SDL_VIDEO=1 -DUSE_SDL_SOUND=1 $(sdl-config --cflags)" CFLAGS="$CFLAGS -DNDEBUG -MMD" %ifnarch %{ix86} #%{x8664} - experimental CFLAGS="$CFLAGS -DNO_VM_COMPILED" %endif %{__make} -C code/unix makedirs targets \ B="release-%{_target}" \ CC="%{__cc}" \rpm --showrc|grep CFLAGS CFLAGS="${CFLAGS:-%optflags}" ; export CFLAGS; CFLAGS="${CFLAGS:-%rpmcflags}" \ CFLAGS="${CFLAGS:-%rpmcflags}" ; export CFLAGS ; \ %optflags -O2 CFLAGS="$CFLAGS" Complete spec file is on http://cvs.pld-linux.org/SPECS/quake3.spec From ramercer at gmail.com Thu Nov 3 14:45:48 2005 From: ramercer at gmail.com (Adam Mercer) Date: Thu, 3 Nov 2005 19:45:48 +0000 Subject: Game exits with "Sys_Error: recursive error after: Z_CheckHeap: next block doesn't have proper back link" Message-ID: <799406d60511031145v25ea85b1i@mail.gmail.com> Hi I'm getting the following error from a fresh build of r249 on x86 Debian Sid, this happens after the level loads - instead of entering into the game it quits will the following displayed on the console: ram at skymoo:/opt/quake3$ ./quake3 ioQ3 1.33 linux-i386 Nov 3 2005 ----- FS_Startup ----- Current search path: /home/ram/.q3a/baseq3 /opt/quake3/baseq3/pak8.pk3 (9 files) /opt/quake3/baseq3/pak7.pk3 (4 files) /opt/quake3/baseq3/pak6.pk3 (64 files) /opt/quake3/baseq3/pak5.pk3 (7 files) /opt/quake3/baseq3/pak4.pk3 (272 files) /opt/quake3/baseq3/pak3.pk3 (4 files) /opt/quake3/baseq3/pak2.pk3 (148 files) /opt/quake3/baseq3/pak1.pk3 (26 files) /opt/quake3/baseq3/pak0.pk3 (3539 files) /opt/quake3/baseq3 ./baseq3/pak8.pk3 (9 files) ./baseq3/pak7.pk3 (4 files) ./baseq3/pak6.pk3 (64 files) ./baseq3/pak5.pk3 (7 files) ./baseq3/pak4.pk3 (272 files) ./baseq3/pak3.pk3 (4 files) ./baseq3/pak2.pk3 (148 files) ./baseq3/pak1.pk3 (26 files) ./baseq3/pak0.pk3 (3539 files) ./baseq3 ---------------------- 8146 files in pk3 files execing default.cfg execing q3config.cfg couldn't exec autoexec.cfg Hunk_Clear: reset the hunk ok ----- Client Initialization ----- ----- Initializing Renderer ---- ------------------------------- ----- Client Initialization Complete ----- ----- R_Init ----- ------- Input Initialization ------- Joystick is not active. ------------------------------------ ...loading libGL.so.1: Calling SDL_Init(SDL_INIT_VIDEO)... SDL_Init(SDL_INIT_VIDEO) passed. Initializing OpenGL display ...setting mode 4: 800 600 Using 8/8/8 Color bits, 24 depth, 0 stencil display. GL_RENDERER: GeForce4 Ti 4800 SE/AGP/SSE/3DNOW! Initializing OpenGL extensions ...ignoring GL_S3_s3tc ...ignoring GL_EXT_texture_env_add ...using GL_ARB_multitexture ...using GL_EXT_compiled_vertex_array GL_VENDOR: NVIDIA Corporation GL_RENDERER: GeForce4 Ti 4800 SE/AGP/SSE/3DNOW! GL_VERSION: 1.5.3 NVIDIA 71.74 GL_EXTENSIONS: GL_ARB_depth_texture GL_ARB_imaging GL_ARB_multisample GL_ARB_multitexture GL_ARB_occlusion_query GL_ARB_point_parameters GL_ARB_point_sprite GL_ARB_shadow GL_ARB_shader_objects GL_ARB_shading_language_100 GL_ARB_texture_border_clamp GL_ARB_texture_compression GL_ARB_texture_cube_map GL_ARB_texture_env_add GL_ARB_texture_env_combine GL_ARB_texture_env_dot3 GL_ARB_texture_mirrored_repeat GL_ARB_texture_rectangle GL_ARB_transpose_matrix GL_ARB_vertex_buffer_object GL_ARB_vertex_program GL_ARB_vertex_shader GL_ARB_window_pos GL_S3_s3tc GL_EXT_texture_env_add GL_EXT_abgr GL_EXT_bgra GL_EXT_blend_color GL_EXT_blend_minmax GL_EXT_blend_subtract GL_EXT_compiled_vertex_array GL_EXT_Cg_shader GL_EXT_draw_range_elements GL_EXT_fog_coord GL_EXT_multi_draw_arrays GL_EXT_packed_pixels GL_EXT_paletted_texture GL_EXT_pixel_buffer_object GL_EXT_point_parameters GL_EXT_rescale_normal GL_EXT_secondary_color GL_EXT_separate_specular_color GL_EXT_shadow_funcs GL_EXT_shared_texture_palette GL_EXT_stencil_wrap GL_EXT_texture3D GL_EXT_texture_compression_s3tc GL_EXT_texture_cube_map GL_EXT_texture_edge_clamp GL_EXT_texture_env_combine GL_EXT_texture_env_dot3 GL_EXT_texture_filter_anisotropic GL_EXT_texture_lod GL_EXT_texture_lod_bias GL_EXT_texture_object GL_EXT_vertex_array GL_HP_occlusion_test GL_IBM_rasterpos_clip GL_IBM_texture_mirrored_repeat GL_KTX_buffer_region GL_NV_blend_square GL_NV_copy_depth_to_color GL_NV_depth_clamp GL_NV_fence GL_NV_fog_distance GL_NV_light_max_exponent GL_NV_multisample_filter_hint GL_NV_occlusion_query GL_NV_packed_depth_stencil GL_NV_pixel_data_range GL_NV_point_sprite GL_NV_register_combiners GL_NV_register_combiners2 GL_NV_texgen_reflection GL_NV_texture_compression_vtc GL_NV_texture_env_combine4 GL_NV_texture_rectangle GL_NV_texture_shader GL_NV_texture_shader2 GL_NV_texture_shader3 GL_NV_vertex_array_range GL_NV_vertex_array_range2 GL_NV_vertex_program GL_NV_vertex_program1_1 GL_SGIS_generate_mipmap GL_SGIS_multitexture GL_SGIS_texture_lod GL_SGIX_depth_texture GL_SGIX_shadow GL_SUN_slice_accum GL_MAX_TEXTURE_SIZE: 4096 GL_MAX_ACTIVE_TEXTURES_ARB: 4 PIXELFORMAT: color(24-bits) Z(24-bit) stencil(0-bits) MODE: 4, 800 x 600 fullscreen hz:N/A GAMMA: hardware w/ 0 overbright bits CPU: rendering primitives: single glDrawElements texturemode: GL_LINEAR_MIPMAP_NEAREST picmip: 1 texture bits: 0 multitexture: enabled compiled vertex arrays: enabled texenv add: disabled compressed textures: disabled HACK: using vertex lightmap approximation Initializing Shaders ...loading 'scripts/lightningnew.shader' ...loading 'scripts/explode1.shader' ...loading 'scripts/gfx.shader' ...loading 'scripts/tim.shader' ...loading 'scripts/base.shader' ...loading 'scripts/base_button.shader' ...loading 'scripts/base_floor.shader' ...loading 'scripts/base_light.shader' ...loading 'scripts/base_object.shader' ...loading 'scripts/base_support.shader' ...loading 'scripts/base_trim.shader' ...loading 'scripts/base_wall.shader' ...loading 'scripts/common.shader' ...loading 'scripts/ctf.shader' ...loading 'scripts/eerie.shader' ...loading 'scripts/gothic_block.shader' ...loading 'scripts/gothic_floor.shader' ...loading 'scripts/gothic_light.shader' ...loading 'scripts/gothic_trim.shader' ...loading 'scripts/gothic_wall.shader' ...loading 'scripts/hell.shader' ...loading 'scripts/liquid.shader' ...loading 'scripts/menu.shader' ...loading 'scripts/models.shader' ...loading 'scripts/organics.shader' ...loading 'scripts/sfx.shader' ...loading 'scripts/shrine.shader' ...loading 'scripts/skin.shader' ...loading 'scripts/sky.shader' ...loading 'scripts/test.shader' ----- finished R_Init ----- ------- sound initialization ------- SDL Audio driver initializing... Calling SDL_Init(SDL_INIT_AUDIO)... SDL_Init(SDL_INIT_AUDIO) passed. SDL audio driver is "dsp" Format we requested from SDL audio device: Format: AUDIO_S16LSB Freq: 22050 Samples: 512 Channels: 2 Format we actually got: Format: AUDIO_S16LSB Freq: 22050 Samples: 512 Channels: 2 WARNING: sdlmixsamps wasn't a power of two (10240), so we made it one (16384). Starting SDL audio callback... SDL audio initialized. ------------------------------------ ----- Sound Info ----- 1 stereo 16384 samples 16 samplebits 1 submission_chunk 22050 speed 0x8ac6910 dma buffer No background file. ---------------------- Sound memory manager started Loading vm file vm/ui.qvm... VM file ui compiled to 594408 bytes of code ui loaded in 1963008 bytes on the hunk 35 arenas parsed 32 bots parsed --- Common Initialization Complete --- Opening IP socket: localhost:27960 Hostname: skymoo.dyndns.org Alias: localhost Alias: skymoo IP: 127.0.0.1 Started tty console (use +set ttycon 0 to disable) ------ Server Initialization ------ Server: q3dm0 RE_Shutdown( 0 ) Hunk_Clear: reset the hunk ok ----- FS_Startup ----- Current search path: /home/ram/.q3a/baseq3 /opt/quake3/baseq3/pak8.pk3 (9 files) /opt/quake3/baseq3/pak7.pk3 (4 files) /opt/quake3/baseq3/pak6.pk3 (64 files) /opt/quake3/baseq3/pak5.pk3 (7 files) /opt/quake3/baseq3/pak4.pk3 (272 files) /opt/quake3/baseq3/pak3.pk3 (4 files) /opt/quake3/baseq3/pak2.pk3 (148 files) /opt/quake3/baseq3/pak1.pk3 (26 files) /opt/quake3/baseq3/pak0.pk3 (3539 files) /opt/quake3/baseq3 ./baseq3/pak8.pk3 (9 files) ./baseq3/pak7.pk3 (4 files) ./baseq3/pak6.pk3 (64 files) ./baseq3/pak5.pk3 (7 files) ./baseq3/pak4.pk3 (272 files) ./baseq3/pak3.pk3 (4 files) ./baseq3/pak2.pk3 (148 files) ./baseq3/pak1.pk3 (26 files) ./baseq3/pak0.pk3 (3539 files) ./baseq3 ---------------------- 16292 files in pk3 files Loading vm file vm/qagame.qvm... VM file qagame compiled to 1137840 bytes of code qagame loaded in 3821696 bytes on the hunk ------- Game Initialization ------- gamename: baseq3 gamedate: Sep 30 2002 Not logging to disk. Gametype changed, clearing session data. 1 teams with 2 entities 11 items registered ----------------------------------- Opened log botlib.log ------- BotLib Initialization ------- loaded weapons.c loaded items.c loaded syn.c loaded rnd.c loaded match.c loaded rchat.c ------------ Map Loading ------------ trying to load maps/q3dm0.aas loaded maps/q3dm0.aas found 19 level items ------------------------------------- 32 bots parsed 35 arenas parsed AAS initialized. ----------------------------------- loaded skill 1 from bots/default_c.c loaded skill 1 from bots/crash_c.c loaded bots/crash_i.c loaded bots/crash_w.c loaded crash from bots/crash_t.c RE_Shutdown( 0 ) ----- R_Init ----- GL_VENDOR: NVIDIA Corporation GL_RENDERER: GeForce4 Ti 4800 SE/AGP/SSE/3DNOW! GL_VERSION: 1.5.3 NVIDIA 71.74 GL_EXTENSIONS: GL_ARB_depth_texture GL_ARB_imaging GL_ARB_multisample GL_ARB_multitexture GL_ARB_occlusion_query GL_ARB_point_parameters GL_ARB_point_sprite GL_ARB_shadow GL_ARB_shader_objects GL_ARB_shading_language_100 GL_ARB_texture_border_clamp GL_ARB_texture_compression GL_ARB_texture_cube_map GL_ARB_texture_env_add GL_ARB_texture_env_combine GL_ARB_texture_env_dot3 GL_ARB_texture_mirrored_repeat GL_ARB_texture_rectangle GL_ARB_transpose_matrix GL_ARB_vertex_buffer_object GL_ARB_vertex_program GL_ARB_vertex_shader GL_ARB_window_pos GL_S3_s3tc GL_EXT_texture_env_add GL_EXT_abgr GL_EXT_bgra GL_EXT_blend_color GL_EXT_blend_minmax GL_EXT_blend_subtract GL_EXT_compiled_vertex_array GL_EXT_Cg_shader GL_EXT_draw_range_elements GL_EXT_fog_coord GL_EXT_multi_draw_arrays GL_EXT_packed_pixels GL_EXT_paletted_texture GL_EXT_pixel_buffer_object GL_EXT_point_parameters GL_EXT_rescale_normal GL_EXT_secondary_color GL_EXT_separate_specular_color GL_EXT_shadow_funcs GL_EXT_shared_texture_palette GL_EXT_stencil_wrap GL_EXT_texture3D GL_EXT_texture_compression_s3tc GL_EXT_texture_cube_map GL_EXT_texture_edge_clamp GL_EXT_texture_env_combine GL_EXT_texture_env_dot3 GL_EXT_texture_filter_anisotropic GL_EXT_texture_lod GL_EXT_texture_lod_bias GL_EXT_texture_object GL_EXT_vertex_array GL_HP_occlusion_test GL_IBM_rasterpos_clip GL_IBM_texture_mirrored_repeat GL_KTX_buffer_region GL_NV_blend_square GL_NV_copy_depth_to_color GL_NV_depth_clamp GL_NV_fence GL_NV_fog_distance GL_NV_light_max_exponent GL_NV_multisample_filter_hint GL_NV_occlusion_query GL_NV_packed_depth_stencil GL_NV_pixel_data_range GL_NV_point_sprite GL_NV_register_combiners GL_NV_register_combiners2 GL_NV_texgen_reflection GL_NV_texture_compression_vtc GL_NV_texture_env_combine4 GL_NV_texture_rectangle GL_NV_texture_shader GL_NV_texture_shader2 GL_NV_texture_shader3 GL_NV_vertex_array_range GL_NV_vertex_array_range2 GL_NV_vertex_program GL_NV_vertex_program1_1 GL_SGIS_generate_mipmap GL_SGIS_multitexture GL_SGIS_texture_lod GL_SGIX_depth_texture GL_SGIX_shadow GL_SUN_slice_accum GL_MAX_TEXTURE_SIZE: 4096 GL_MAX_ACTIVE_TEXTURES_ARB: 4 PIXELFORMAT: color(24-bits) Z(24-bit) stencil(0-bits) MODE: 4, 800 x 600 fullscreen hz:N/A GAMMA: hardware w/ 0 overbright bits CPU: rendering primitives: single glDrawElements texturemode: GL_LINEAR_MIPMAP_NEAREST picmip: 1 texture bits: 0 multitexture: enabled compiled vertex arrays: enabled texenv add: disabled compressed textures: disabled HACK: using vertex lightmap approximation Initializing Shaders ...loading 'scripts/lightningnew.shader' ...loading 'scripts/explode1.shader' ...loading 'scripts/gfx.shader' ...loading 'scripts/tim.shader' ...loading 'scripts/base.shader' ...loading 'scripts/base_button.shader' ...loading 'scripts/base_floor.shader' ...loading 'scripts/base_light.shader' ...loading 'scripts/base_object.shader' ...loading 'scripts/base_support.shader' ...loading 'scripts/base_trim.shader' ...loading 'scripts/base_wall.shader' ...loading 'scripts/common.shader' ...loading 'scripts/ctf.shader' ...loading 'scripts/eerie.shader' ...loading 'scripts/gothic_block.shader' ...loading 'scripts/gothic_floor.shader' ...loading 'scripts/gothic_light.shader' ...loading 'scripts/gothic_trim.shader' ...loading 'scripts/gothic_wall.shader' ...loading 'scripts/hell.shader' ...loading 'scripts/liquid.shader' ...loading 'scripts/menu.shader' ...loading 'scripts/models.shader' ...loading 'scripts/organics.shader' ...loading 'scripts/sfx.shader' ...loading 'scripts/shrine.shader' ...loading 'scripts/skin.shader' ...loading 'scripts/sky.shader' ...loading 'scripts/test.shader' ----- finished R_Init ----- Loading vm file vm/ui.qvm... VM file ui compiled to 594408 bytes of code ui loaded in 1963008 bytes on the hunk 35 arenas parsed 32 bots parsed Loading vm file vm/cgame.qvm... VM file cgame compiled to 786818 bytes of code cgame loaded in 5380768 bytes on the hunk stitched 0 LoD cracks ...loaded 4806 faces, 46 meshes, 43 trisurfs, 128 flares ^3WARNING: could not find sound/player/announce/Adam.wav - using default ^3WARNING: could not find sound/player/crash/taunt.wav - using default CL_InitCGame: 1.62 seconds 6 msec to draw all images ----- CL_Shutdown ----- Closing SDL audio device... SDL audio device shut down. RE_Shutdown( 1 ) ----------------------- ----- Server Shutdown ----- ==== ShutdownGame ==== ----- CL_Shutdown ----- ----------------------- Sys_Error: recursive error after: Z_CheckHeap: next block doesn't have proper back link Shutdown tty console Cheers Adam From tim at ngus.net Thu Nov 3 15:08:16 2005 From: tim at ngus.net (Tim Angus) Date: Thu, 3 Nov 2005 20:08:16 +0000 Subject: Game exits with "Sys_Error: recursive error after: Z_CheckHeap: next block doesn't have proper back link" In-Reply-To: <799406d60511031145v25ea85b1i@mail.gmail.com> References: <799406d60511031145v25ea85b1i@mail.gmail.com> Message-ID: <20051103200816.11e09dc6.tim@ngus.net> On Thu, 3 Nov 2005 19:45:48 +0000 Adam wrote: > I'm getting the following error from a fresh build of r249 on x86 > Debian Sid, this happens after the level loads - instead of entering > into the game it quits will the following displayed on the console: https://bugzilla.icculus.org/show_bug.cgi?id=2482 From ludwig.nussel at suse.de Thu Nov 3 15:32:57 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Thu, 3 Nov 2005 21:32:57 +0100 Subject: [quake3] Q3 x86_64 In-Reply-To: References: Message-ID: <200511032132.57805.ludwig.nussel@suse.de> PaSzCzUs wrote: > > > [...] > > > Loading vm file vm/qagame.qvm... > > > compiling qagame > > > running assembler < /var/games/quake3/.q3a/baseq3/qagame.s > > > > /var/games/quake3/.q3a/baseq3/qagame.o > > > done > > > computing jump table > > > Received signal 11, exiting... > > > Shutdown tty console > > > > The vm compiler is disabled by default on purpose. It works for me > > though (except for bots). Which distro/compiler/assembler/CFLAGS are > > you using? > > I`m using PLD Linux Distribution with gcc 3.3.6 and compiled using: > > CFLAGS="%{rpmcflags}" > CFLAGS="$CFLAGS -DDEFAULT_BASEDIR=\\\"%{_datadir}/games/%{name}\\\"" > CFLAGS="$CFLAGS -Wall -Wimplicit -Wstrict-prototypes" > CFLAGS="$CFLAGS -DUSE_SDL_VIDEO=1 -DUSE_SDL_SOUND=1 $(sdl-config --cflags)" > CFLAGS="$CFLAGS -DNDEBUG -MMD" > %ifnarch %{ix86} #%{x8664} - experimental > CFLAGS="$CFLAGS -DNO_VM_COMPILED" > %endif > > %{__make} -C code/unix makedirs targets \ > B="release-%{_target}" \ > CC="%{__cc}" \rpm --showrc|grep CFLAGS > CFLAGS="${CFLAGS:-%optflags}" ; export CFLAGS; > CFLAGS="${CFLAGS:-%rpmcflags}" \ > CFLAGS="${CFLAGS:-%rpmcflags}" ; export CFLAGS ; \ > > %optflags -O2 > > CFLAGS="$CFLAGS" > > Complete spec file is on http://cvs.pld-linux.org/SPECS/quake3.spec make %{?jobs:-j%jobs} \ OPTIMIZE="%{optflags} -O3 -ffast-math -fno-strict-aliasing" \ LCC_CFLAGS="%{optflags} -fno-strict-aliasing" \ Q3ASM_CFLAGS="%{optflags} -fno-strict-aliasing" \ release That's what I use. You better override OPTIMIZE rather than CFLAGS. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From vincent at cojot.name Fri Nov 4 11:43:28 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Fri, 4 Nov 2005 17:43:28 +0100 (CET) Subject: Problems in botlib/be_aas_route.c ? Message-ID: I'm trying to find the source of the botlib segfaults on Solaris/Sparc. I'm running the dedicated server with: gdb> ....3ded gdb> b AAS_AreaTravelTime gdb> r +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set sv_pure 1 +set ttycon 0 +map q3dm17 On Linux/i386/RHEL3, I get this run: ------------ Map Loading ------------ trying to load maps/q3dm17.aas loaded maps/q3dm17.aas found 29 level items ------------------------------------- 32 bots parsed 35 arenas parsed Breakpoint 1, AAS_AreaTravelTime (areanum=4, start=0xbfff8bc0, end=0xb622b69c) at ../botlib/be_aas_route.c:480 480 VectorSubtract(start, end, dir); (gdb) n 481 dist = VectorLength(dir); (gdb) n 483 if (AAS_AreaCrouch(areanum)) dist *= DISTANCEFACTOR_CROUCH; (gdb) n 485 else if (AAS_AreaSwim(areanum)) dist *= DISTANCEFACTOR_SWIM; (gdb) n 487 else dist *= DISTANCEFACTOR_WALK; (gdb) n 489 intdist = (int) dist; (gdb) print dir $6 = {0, 5.1000061, 0.125} (gdb) print dist $7 = 1.68350756 (gdb) print intdist $8 = 0 On Solaris/Sparc, I get this run: ------------ Map Loading ------------ trying to load maps/q3dm17.aas loaded maps/q3dm17.aas found 29 level items ------------------------------------- 32 bots parsed 35 arenas parsed Breakpoint 1, AAS_AreaTravelTime (areanum=4, start=0xffbf9d28, end=0x496b1bc) at ../botlib/be_aas_route.c:480 480 VectorSubtract(start, end, dir); (gdb) n 481 dist = VectorLength(dir); (gdb) n 483 if (AAS_AreaCrouch(areanum)) dist *= DISTANCEFACTOR_CROUCH; (gdb) n 485 else if (AAS_AreaSwim(areanum)) dist *= DISTANCEFACTOR_SWIM; (gdb) n 487 else dist *= DISTANCEFACTOR_WALK; (gdb) n 489 intdist = (int) dist; (gdb) print dir $5 = {0, 5.1000061, 0.125} (gdb) print dist $6 = 1.68350756 (gdb) print intdist $7 = 4543008 I cannot understand why intdist would be so different in both cases.. I cannot understand either why it isn't being cast as 'unsigned short' since that's what AAS_AreaTravelTime should return.. Eventually, this results in segfaults on Solaris/Sparc.. Program received signal SIGSEGV, Segmentation fault. 0x000a4a24 in AAS_CalculateAreaTravelTimes () at ../botlib/be_aas_route.c:543 543 aasworld.areatraveltimes[i][l] = (unsigned short *) ptr; I can override that segfault by using playing with LittleShort but then it dies on line 522 on be_aas_route.c... *Sigh* Vincent From cgatesman at gmail.com Fri Nov 4 12:15:04 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Fri, 4 Nov 2005 12:15:04 -0500 Subject: [quake3] Problems in botlib/be_aas_route.c ? In-Reply-To: References: Message-ID: <7bd2d8660511040915m603d401dgd7e029b5212dea2d@mail.gmail.com> You didn't step over "intdist = (int) dist;" yet. intdist is uninitialized, and that is why the values are different on the two systems. -Chad On 11/4/05, vincent at cojot.name wrote: > > > I'm trying to find the source of the botlib segfaults on Solaris/Sparc. > > I'm running the dedicated server with: > > gdb> ....3ded > gdb> b AAS_AreaTravelTime > gdb> r +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set sv_pure 1 +set > ttycon 0 +map q3dm17 > > On Linux/i386/RHEL3, I get this run: > > ------------ Map Loading ------------ > trying to load maps/q3dm17.aas > loaded maps/q3dm17.aas > found 29 level items > ------------------------------------- > 32 bots parsed > 35 arenas parsed > > Breakpoint 1, AAS_AreaTravelTime (areanum=4, start=0xbfff8bc0, > end=0xb622b69c) > at ../botlib/be_aas_route.c:480 > 480 VectorSubtract(start, end, dir); > (gdb) n > 481 dist = VectorLength(dir); > (gdb) n > 483 if (AAS_AreaCrouch(areanum)) dist *= > DISTANCEFACTOR_CROUCH; > (gdb) n > 485 else if (AAS_AreaSwim(areanum)) dist *= > DISTANCEFACTOR_SWIM; > (gdb) n > 487 else dist *= DISTANCEFACTOR_WALK; > (gdb) n > 489 intdist = (int) dist; > (gdb) print dir > $6 = {0, 5.1000061, 0.125} > (gdb) print dist > $7 = 1.68350756 > (gdb) print intdist > $8 = 0 > > On Solaris/Sparc, I get this run: > ------------ Map Loading ------------ > trying to load maps/q3dm17.aas > loaded maps/q3dm17.aas > found 29 level items > ------------------------------------- > 32 bots parsed > 35 arenas parsed > > Breakpoint 1, AAS_AreaTravelTime (areanum=4, start=0xffbf9d28, > end=0x496b1bc) at ../botlib/be_aas_route.c:480 > 480 VectorSubtract(start, end, dir); > (gdb) n > 481 dist = VectorLength(dir); > (gdb) n > 483 if (AAS_AreaCrouch(areanum)) dist *= > DISTANCEFACTOR_CROUCH; > (gdb) n > 485 else if (AAS_AreaSwim(areanum)) dist *= > DISTANCEFACTOR_SWIM; > (gdb) n > 487 else dist *= DISTANCEFACTOR_WALK; > (gdb) n > 489 intdist = (int) dist; > (gdb) print dir > $5 = {0, 5.1000061, 0.125} > (gdb) print dist > $6 = 1.68350756 > (gdb) print intdist > $7 = 4543008 > > I cannot understand why intdist would be so different in both cases.. I > cannot understand either why it isn't being cast as 'unsigned short' since > that's what AAS_AreaTravelTime should return.. > > Eventually, this results in segfaults on Solaris/Sparc.. > > Program received signal SIGSEGV, Segmentation fault. > 0x000a4a24 in AAS_CalculateAreaTravelTimes () at > ../botlib/be_aas_route.c:543 > 543 aasworld.areatraveltimes[i][l] = (unsigned short *) ptr; > > I can override that segfault by using playing with LittleShort but then it > dies on line 522 on be_aas_route.c... > > *Sigh* > > Vincent > -- Chad Gatesman Software Engineer -------------- next part -------------- An HTML attachment was scrubbed... URL: From vincent at cojot.name Fri Nov 4 12:58:42 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Fri, 4 Nov 2005 18:58:42 +0100 (CET) Subject: [quake3] Problems in botlib/be_aas_route.c ? In-Reply-To: <7bd2d8660511040915m603d401dgd7e029b5212dea2d@mail.gmail.com> References: <7bd2d8660511040915m603d401dgd7e029b5212dea2d@mail.gmail.com> Message-ID: On Fri, 4 Nov 2005, Chad Gatesman wrote: > You didn't step over "intdist = (int) dist;" yet. intdist is uninitialized, > and that is why the values are different on the two systems. Hi Chad, Thanks for the reply.. Yes, I had noticed that a few moments later.. Dohhhh.. Vincent From darksharkster at gmail.com Fri Nov 4 15:51:37 2005 From: darksharkster at gmail.com (darkshark atnight) Date: Fri, 4 Nov 2005 20:51:37 +0000 Subject: quake3 on Solaris x86 In-Reply-To: References: Message-ID: Some good news. Playing quake3 on Solaris x86 now 8-) Seems ludwig's changes to the bot code (realignments) got things working ! The in_dgamouse cvar needs to be 0. details; snv v268 just changed the Makefile ala patch Vincent already sent out, no other changes. Both single and multi player works. Will be trying out the bytecode compiler next. Thanks guys ! off to pub. Sean. . On 10/31/05, darkshark atnight wrote: > Hi folks, > I'm trying to get quake3 on solaris x86, specifically running on > SolarisExpress. Everything > compiles, but then running it fails to start OpenGL. It seems that > the the qgl* function > pointers aren't pointing to the correct addresses of the gl* functions. > > Some details: > OpenGL version is Mesa (io quake2 compiles and runs on same system). > Using the quake2 source base as a reference, since theres a lot of > common code there. > I've got solaris_qgl.c and solaris_glimp.c done, which are really > linux_*.c with a bunch > of #def's. > > More details at: > http://blogs.sun.com/roller/page/smg?entry=getting_there_with_quake3_on > > I'm trying to get this port done on x86 platform. Will be using the > nVidia opengl drivers later on another box, once I get these problems sorted. > > Any help would really > be appreciated. > Regards, > Sean. > . > From paszczus at gmail.com Fri Nov 4 16:27:03 2005 From: paszczus at gmail.com (PaSzCzUs) Date: Fri, 4 Nov 2005 22:27:03 +0100 Subject: [quake3] Q3 x86_64 In-Reply-To: <200511032132.57805.ludwig.nussel@suse.de> References: <200511032132.57805.ludwig.nussel@suse.de> Message-ID: Ok, fixed in spec file, thanks! 2005/11/3, Ludwig Nussel : > PaSzCzUs wrote: > > > > [...] > > > > Loading vm file vm/qagame.qvm... > > > > compiling qagame > > > > running assembler < /var/games/quake3/.q3a/baseq3/qagame.s > > > > > /var/games/quake3/.q3a/baseq3/qagame.o > > > > done > > > > computing jump table > > > > Received signal 11, exiting... > > > > Shutdown tty console > > > > > > The vm compiler is disabled by default on purpose. It works for me > > > though (except for bots). Which distro/compiler/assembler/CFLAGS are > > > you using? > > > > I`m using PLD Linux Distribution with gcc 3.3.6 and compiled using: > > > > CFLAGS="%{rpmcflags}" > > CFLAGS="$CFLAGS -DDEFAULT_BASEDIR=\\\"%{_datadir}/games/%{name}\\\"" > > CFLAGS="$CFLAGS -Wall -Wimplicit -Wstrict-prototypes" > > CFLAGS="$CFLAGS -DUSE_SDL_VIDEO=1 -DUSE_SDL_SOUND=1 $(sdl-config --cflags)" > > CFLAGS="$CFLAGS -DNDEBUG -MMD" > > %ifnarch %{ix86} #%{x8664} - experimental > > CFLAGS="$CFLAGS -DNO_VM_COMPILED" > > %endif > > > > %{__make} -C code/unix makedirs targets \ > > B="release-%{_target}" \ > > CC="%{__cc}" \rpm --showrc|grep CFLAGS > > CFLAGS="${CFLAGS:-%optflags}" ; export CFLAGS; > > CFLAGS="${CFLAGS:-%rpmcflags}" \ > > CFLAGS="${CFLAGS:-%rpmcflags}" ; export CFLAGS ; \ > > > > %optflags -O2 > > > > CFLAGS="$CFLAGS" > > > > Complete spec file is on http://cvs.pld-linux.org/SPECS/quake3.spec > > make %{?jobs:-j%jobs} \ > OPTIMIZE="%{optflags} -O3 -ffast-math -fno-strict-aliasing" \ > LCC_CFLAGS="%{optflags} -fno-strict-aliasing" \ > Q3ASM_CFLAGS="%{optflags} -fno-strict-aliasing" \ > release > > That's what I use. You better override OPTIMIZE rather than CFLAGS. > > cu > Ludwig > > -- > (o_ Ludwig Nussel > //\ SUSE LINUX Products GmbH, Development > V_/_ http://www.suse.de/ > > From darksharkster at gmail.com Sat Nov 5 07:54:59 2005 From: darksharkster at gmail.com (darkshark atnight) Date: Sat, 5 Nov 2005 12:54:59 +0000 Subject: svn 269 breaks solaris x86 compile Message-ID: The change in qcommon/q_platform.h broke solaris x86 compiling. The __sun doesn't really have to be there (does it?). patch: diff -u q_platform.h.orig q_platform.h --- q_platform.h.orig Sat Nov 5 12:45:04 2005 +++ q_platform.h Sat Nov 5 12:07:28 2005 @@ -32,8 +32,7 @@ #else -#if (defined _M_IX86 || defined __i386__) && \ - !defined __sun && !defined(C_ONLY) +#if (defined _M_IX86 || defined __i386__) && !defined(C_ONLY) #define id386 1 #else #define id386 0 From tim at ngus.net Sat Nov 5 10:14:17 2005 From: tim at ngus.net (Tim Angus) Date: Sat, 5 Nov 2005 15:14:17 +0000 Subject: svn 269 breaks solaris x86 compile In-Reply-To: References: Message-ID: <20051105151417.01aa24fc.tim@ngus.net> On Sat, 5 Nov 2005 12:54:59 +0000 darkshark wrote: > The change in qcommon/q_platform.h broke solaris x86 compiling. The > __sun doesn't really have to be there (does it?). I suspect the actual reason it broke was a typo in snapvectora.s (fixed in 279). However, if you mean to say it works fine with your patch, then yes, the __sun should go. That has been there since day one, I assumed for a reason, but not having any sun hardware to play with I just left it alone. Assuming this does work, could you also try the attached patch to enable the JIT compiler on solaris x86? Make sure that your vm_* cvars are set to 2 (they should be by default). Thanks. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: enable-x86-jit-compiler-on-solaris.diff URL: From tim at ngus.net Sat Nov 5 11:04:30 2005 From: tim at ngus.net (Tim Angus) Date: Sat, 5 Nov 2005 16:04:30 +0000 Subject: Alpha Build In-Reply-To: <7bd2d8660511020820k2602c631wa90fe14ebac0dff0@mail.gmail.com> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511011429i32c45eaai7339c06af8450850@mail.gmail.com> <20051101223213.397cadeb.tim@ngus.net> <7bd2d8660511020820k2602c631wa90fe14ebac0dff0@mail.gmail.com> Message-ID: <20051105160430.396b3393.tim@ngus.net> On Wed, 2 Nov 2005 11:20:43 -0500 Chad wrote: > I'll try to put together a full patch and submit a Bugzilla PR. > > Quake 3 on Alpha! Any progress on this? Even if it's for the dedicated server only... From darksharkster at gmail.com Sat Nov 5 11:31:35 2005 From: darksharkster at gmail.com (darkshark atnight) Date: Sat, 5 Nov 2005 16:31:35 +0000 Subject: [quake3] Re: svn 269 breaks solaris x86 compile In-Reply-To: <20051105151417.01aa24fc.tim@ngus.net> References: <20051105151417.01aa24fc.tim@ngus.net> Message-ID: On 11/5/05, Tim Angus wrote: > On Sat, 5 Nov 2005 12:54:59 +0000 darkshark wrote: > > The change in qcommon/q_platform.h broke solaris x86 compiling. The > > __sun doesn't really have to be there (does it?). > > I suspect the actual reason it broke was a typo in snapvectora.s (fixed > in 279). However, if you mean to say it works fine with your patch, then > yes, the __sun should go. That has been there since day one, I assumed > for a reason, but not having any sun hardware to play with I just left > it alone. The typo fix does indeed allow compile and running, but on x86 with the __ sun there we don't get the asm SnapVector, just the C version. I'd guess the __sun bit was there because it was assumed that __sun == sparc which isn't the case today :) > > Assuming this does work, could you also try the attached patch to enable > the JIT compiler on solaris x86? Make sure that your vm_* cvars are set > to 2 (they should be by default). Yup, applied, compiled and running ok, both game and dedicated server. > > Thanks. Thanks you :) Sean. . > > > From tim at ngus.net Sat Nov 5 13:39:16 2005 From: tim at ngus.net (Tim Angus) Date: Sat, 5 Nov 2005 18:39:16 +0000 Subject: svn 269 breaks solaris x86 compile In-Reply-To: References: <20051105151417.01aa24fc.tim@ngus.net> Message-ID: <20051105183916.698ba9f0.tim@ngus.net> On Sat, 5 Nov 2005 16:31:35 +0000 darkshark wrote: > I'd guess the __sun bit was there because it was assumed that __sun > == sparc which > isn't the case today :) Strange that it should specifically turn id386 off for x86 machines that define __sun though. It's pointless speculating now really... > Yup, applied, compiled and running ok, both game and dedicated > server. OK, good. I guess I'll apply these patches then. From ludwig.nussel at suse.de Sat Nov 5 14:06:44 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Sat, 5 Nov 2005 20:06:44 +0100 Subject: [quake3] Problems in botlib/be_aas_route.c ? In-Reply-To: References: Message-ID: <200511052006.44808.ludwig.nussel@suse.de> vincent at cojot.name wrote: > I'm trying to find the source of the botlib segfaults on Solaris/Sparc. > > [...] > Program received signal SIGSEGV, Segmentation fault. > 0x000a4a24 in AAS_CalculateAreaTravelTimes () at > ../botlib/be_aas_route.c:543 > 543 aasworld.areatraveltimes[i][l] = (unsigned > short *) ptr; purify it harder. Alignment again. r282. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From cgatesman at gmail.com Sat Nov 5 14:13:19 2005 From: cgatesman at gmail.com (Chad Gatesman) Date: Sat, 5 Nov 2005 14:13:19 -0500 Subject: [quake3] Re: Alpha Build In-Reply-To: <20051105160430.396b3393.tim@ngus.net> References: <7bd2d8660511010733j57b912fdm5ed2b6b0aacbd48b@mail.gmail.com> <20051101155940.02e84f22.tim@ngus.net> <7bd2d8660511010843w4391abcbo36331a951bb73ba@mail.gmail.com> <20051101221844.32e0086e.tim@ngus.net> <7bd2d8660511011429i32c45eaai7339c06af8450850@mail.gmail.com> <20051101223213.397cadeb.tim@ngus.net> <7bd2d8660511020820k2602c631wa90fe14ebac0dff0@mail.gmail.com> <20051105160430.396b3393.tim@ngus.net> Message-ID: <7bd2d8660511051113j7268e2f8tae5730a7da0c04b6@mail.gmail.com> No. Haven't had time to put anything together yet. On 11/5/05, Tim Angus wrote: > > On Wed, 2 Nov 2005 11:20:43 -0500 Chad wrote: > > I'll try to put together a full patch and submit a Bugzilla PR. > > > > Quake 3 on Alpha! > > Any progress on this? Even if it's for the dedicated server only... > -- Chad Gatesman Home Page - http://www.chadgatesman.com/ Weblog - http://blog.chadgatesman.com/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From vincent at cojot.name Sat Nov 5 14:16:41 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sat, 5 Nov 2005 20:16:41 +0100 (CET) Subject: [quake3] Re: svn 269 breaks solaris x86 compile In-Reply-To: References: <20051105151417.01aa24fc.tim@ngus.net> Message-ID: On Sat, 5 Nov 2005, darkshark atnight wrote: > I'd guess the __sun bit was there because it was assumed that __sun > == sparc which > isn't the case today :) My typo, most likely.. :) I'm trying, guys.. :) I'll test the alignment fixes tomorrow.. :) Thanks, -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From vincent at cojot.name Sun Nov 6 11:26:16 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sun, 6 Nov 2005 17:26:16 +0100 (CET) Subject: Quake3 client + server on Solaris/Sparc In-Reply-To: <200511052006.44808.ludwig.nussel@suse.de> References: <200511052006.44808.ludwig.nussel@suse.de> Message-ID: Hi everyone, Kudos to Ludwig for the alignment fixes and also to tim for the other stuff. I just used both the client and server (vm_* set to 1) on Solaris10/Sparc (rev 292 to 295). The client runs a little bit slow but it works. Also, I verified that q3 1.32b Linux client were able to connect and play on the serveur running on Solaris/Sparc. For the record, I used these flags on the serveur: debugsparc/SunOSq3ded +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set \ sv_pure 1 +set ttycon 0 +map q3dm17 And these on the client: debugsparc/SunOSquake3 +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set \ sv_pure 1 +set ttycon 0 +set in_dgamouse 0 +map q3dm3 It works! :) I bow before your coding/debugging skills, Ludwig.. :) Many thanks! :) For the record, I tried it on an Sb2k with XVR-1000 graphics (which isn't exactly the top graphics card for OpenGL). I'll try it again tomorrow on an XVR-600... Best regards and many thanks again. :) -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From tim at ngus.net Sun Nov 6 12:00:38 2005 From: tim at ngus.net (Tim Angus) Date: Sun, 6 Nov 2005 17:00:38 +0000 Subject: Quake3 client + server on Solaris/Sparc In-Reply-To: References: Message-ID: <20051106170038.2d87c79a.tim@ngus.net> On Sun, 6 Nov 2005 17:26:16 +0100 (CET) vincent at cojot.name wrote: > And these on the client: > debugsparc/SunOSquake3 +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 > +set \ sv_pure 1 +set ttycon 0 +set in_dgamouse 0 +map q3dm3 Please could you try the attached patch, starting with +set in_dgamouse 1. This setting should really be ignored completely if dga isn't available. If it works I'll apply it. Should I change the status (http://icculus.org/quake3/?page=status) for Solaris sparc to working then? -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ignore-in_dgamouse-when-do-not-HAVE_XF86DGA.diff URL: From vincent at cojot.name Sun Nov 6 12:25:00 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Sun, 6 Nov 2005 18:25:00 +0100 (CET) Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: <20051106170038.2d87c79a.tim@ngus.net> References: <20051106170038.2d87c79a.tim@ngus.net> Message-ID: On Sun, 6 Nov 2005, Tim Angus wrote: > On Sun, 6 Nov 2005 17:26:16 +0100 (CET) vincent at cojot.name wrote: >> And these on the client: >> debugsparc/SunOSquake3 +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 >> +set \ sv_pure 1 +set ttycon 0 +set in_dgamouse 0 +map q3dm3 > > Please could you try the attached patch, starting with +set in_dgamouse > 1. This setting should really be ignored completely if dga isn't > available. If it works I'll apply it. > > Should I change the status (http://icculus.org/quake3/?page=status) for > Solaris sparc to working then? Hello Tim, Thank you very much. The patch works fine on rev v299 on Solaris/Sparc. :) Hurrrah... :) -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From vincent at cojot.name Mon Nov 7 09:53:11 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Mon, 7 Nov 2005 15:53:11 +0100 (CET) Subject: Quake3 client + server on Solaris/Sparc In-Reply-To: References: <20051106170038.2d87c79a.tim@ngus.net> Message-ID: I think there might still be some alignment problems lurking: When I run the debug client, it just works. When I try the 'release' version on Solaris/Sparc, the client crashes immediately after opening an X11 window and displaying the first image of the intro cinematics (the wall which is fired at). Here's the gdb backtrace..: --- Common Initialization Complete --- Opening IP socket: localhost:27960 Hostname: flotsam Alias: flotsam. Alias: loghost IP: 10.179.104.54 Program received signal SIGSEGV, Segmentation fault. 0x00031f30 in blitVQQuad32fs () (gdb) bt #0 0x00031f30 in blitVQQuad32fs () #1 0x000328ac in RoQInterrupt () at ../client/cl_cin.c:1198 #2 0x00036990 in CIN_RunCinematic (handle=7251) at ../client/cl_cin.c:1449 #3 0x00045b88 in CL_Frame (msec=0) at ../client/cl_main.c:2064 #4 0x00068e20 in Com_Frame () at ../qcommon/common.c:2763 #5 0x00162888 in main (argc=16, argv=0xffbff20c) at ../unix/unix_main.c:1323 (gdb) q The program is running. Exit anyway? (y or n) y I don't know if this could be related to the HUNK_DEBUG stuff.. -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From tim at ngus.net Mon Nov 7 10:30:14 2005 From: tim at ngus.net (Tim Angus) Date: Mon, 7 Nov 2005 15:30:14 +0000 Subject: Quake3 client + server on Solaris/Sparc In-Reply-To: References: <20051106170038.2d87c79a.tim@ngus.net> Message-ID: <20051107153014.6f190e11.tim@ngus.net> On Mon, 7 Nov 2005 15:53:11 +0100 (CET) vincent at cojot.name wrote: > When I run the debug client, it just works. When I try the 'release' > version on Solaris/Sparc, the client crashes immediately after opening > an X11 window and displaying the first image of the intro cinematics > (the wall which is fired at). Here's the gdb backtrace..: > #0 0x00031f30 in blitVQQuad32fs () > #1 0x000328ac in RoQInterrupt () at ../client/cl_cin.c:1198 > #2 0x00036990 in CIN_RunCinematic (handle=7251) at If you put a command on the commandline it will skip playing the splash movie. e.g. quake3 +echo "hi mum". I guess if it works, it's a problem specific to the cinematic playback code. From icculus at clutteredmind.org Mon Nov 7 10:54:50 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Mon, 07 Nov 2005 10:54:50 -0500 Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: References: <20051106170038.2d87c79a.tim@ngus.net> Message-ID: <436F78CA.7000900@clutteredmind.org> vincent at cojot.name wrote: > I think there might still be some alignment problems lurking: > > When I run the debug client, it just works. When I try the 'release' > version on Solaris/Sparc, the client crashes immediately after opening an > X11 window and displaying the first image of the intro cinematics (the > wall which is fired at). Here's the gdb backtrace..: This happens on PowerPC linux, too. Well, not on the first cinematic (the id logo), but the intro movie with Sarge. There's video corruption and an eventual crash in that same function. Is the Sparc bigendian? Perhaps there's a byteorder issue? --ryan. From tim at ngus.net Mon Nov 7 11:07:50 2005 From: tim at ngus.net (Tim Angus) Date: Mon, 7 Nov 2005 16:07:50 +0000 Subject: Quake3 client + server on Solaris/Sparc In-Reply-To: <436F78CA.7000900@clutteredmind.org> References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> Message-ID: <20051107160750.7084af16.tim@ngus.net> On Mon, 07 Nov 2005 10:54:50 -0500 Ryan wrote: > This happens on PowerPC linux, too. > Is the Sparc bigendian? Perhaps there's a byteorder issue? You may be onto something. Patch attached. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: cinematic-endianess.diff URL: From tim at ngus.net Mon Nov 7 11:10:37 2005 From: tim at ngus.net (Tim Angus) Date: Mon, 7 Nov 2005 16:10:37 +0000 Subject: Quake3 client + server on Solaris/Sparc In-Reply-To: <20051107160750.7084af16.tim@ngus.net> References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> Message-ID: <20051107161037.730f8779.tim@ngus.net> On Mon, 7 Nov 2005 16:07:50 +0000 Tim wrote: > On Mon, 07 Nov 2005 10:54:50 -0500 Ryan wrote: > > This happens on PowerPC linux, too. > > Is the Sparc bigendian? Perhaps there's a byteorder issue? > > You may be onto something. Patch attached. Hmm, and perhaps this too... -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: cinematic-endianess2.diff URL: From vincent at cojot.name Mon Nov 7 11:16:26 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Mon, 7 Nov 2005 17:16:26 +0100 (CET) Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: <436F78CA.7000900@clutteredmind.org> References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> Message-ID: On Mon, 7 Nov 2005, Ryan C. Gordon wrote: > This happens on PowerPC linux, too. > > Well, not on the first cinematic (the id logo), but the intro movie with > Sarge. > > There's video corruption and an eventual crash in that same function. > > Is the Sparc bigendian? Perhaps there's a byteorder issue? Yup, the sparc is bigendian. What I don't understand is why it worked fine using the debug version and crashed using the non-debug/release.. Tim: I'm testing your fix.. -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From vincent at cojot.name Mon Nov 7 11:38:25 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Mon, 7 Nov 2005 17:38:25 +0100 (CET) Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: <20051107161037.730f8779.tim@ngus.net> References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> Message-ID: On Mon, 7 Nov 2005, Tim Angus wrote: > On Mon, 7 Nov 2005 16:07:50 +0000 Tim wrote: >> On Mon, 07 Nov 2005 10:54:50 -0500 Ryan wrote: >>> This happens on PowerPC linux, too. >>> Is the Sparc bigendian? Perhaps there's a byteorder issue? >> >> You may be onto something. Patch attached. > > Hmm, and perhaps this too... Hi Tim, I think your diffs are fine and should be commited. However, for the Solaris/Sparc problem, they made no difference.. I still got the crash in blitVQQuad32fs(). I have also noticed that the dedicated server crashes in its release version (again in botlib) whereas the debug version works fine: $ gdb \ /export/home/vcojot/SVN/icculus/quake3-Solaris/code/unix/releasesparc/SunOSq3ded (gdb) r +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set sv_pure 1 +set ttycon 0 +map q3dm17 [....] 0 teams with 0 entities 16 items registered ----------------------------------- ------- BotLib Initialization ------- Program received signal SIGSEGV, Segmentation fault. PC_EvaluateTokens (source=0x414338, tokens=0x43df2c, intvalue=0xffbf7364, floatvalue=0x0, integer=1) at ../botlib/l_precomp.c:1800 1800 v->floatvalue = t->floatvalue; (gdb) bt #0 PC_EvaluateTokens (source=0x414338, tokens=0x43df2c, intvalue=0xffbf7364, floatvalue=0x0, integer=1) at ../botlib/l_precomp.c:1800 #1 0x000beae4 in PC_DollarEvaluate (source=0x414338, intvalue=0xffbf7364, floatvalue=0x0, integer=1) at ../botlib/l_precomp.c:2325 #2 0x000bf490 in PC_DollarDirective_evalint (source=0x414338) at ../botlib/l_precomp.c:2558 #3 0x000bf9b4 in PC_ReadDollarDirective (source=0x414338) at ../botlib/l_precomp.c:2641 #4 0x000bfb40 in PC_ReadToken (source=0x414338, token=0xffbf8638) at ../botlib/l_precomp.c:2735 #5 0x000c0dd4 in PC_ExpectAnyToken (source=0x414338, token=0xffbf8638) at ../botlib/l_precomp.c:2870 #6 0x000c4dd4 in ReadNumber (source=0x414338, fd=0xe8ed4, p=0x4a9c5b8) at ../botlib/l_struct.c:83 #7 0x000c5348 in ReadStructure (source=0x414338, def=0xe8e40, structure=0x4a9c504 "grenade") at ../botlib/l_struct.c:288 #8 0x000b586c in LoadWeaponConfig (filename=0x4a97d84 "") at ../botlib/be_ai_weap.c:271 #9 0x000b62f8 in BotSetupWeaponAI () at ../botlib/be_ai_weap.c:520 #10 0x000b983c in Export_BotLibSetup () at ../botlib/be_interface.c:155 #11 0x00022e84 in SV_BotLibSetup () at ../server/sv_bot.c:454 #12 0x0002b81c in SV_GameSystemCalls (args=0x18486a0) at ../server/sv_game.c:451 #13 0x0006eb34 in VM_CallInterpreted (vm=0x2761d8, args=0xffbfa130) at ../qcommon/vm_interpreted.c:548 #14 0x0006e098 in VM_Call (vm=0x2761d8, callnum=0) at ../qcommon/vm.c:769 #15 0x0002dadc in SV_SpawnServer (server=0xffbfe730 "q3dm17", killBots=qfalse) at ../server/sv_init.c:442 #16 0x00027fe8 in SV_Map_f () at ../server/sv_ccmds.c:187 #17 0x00046c68 in Cmd_ExecuteString (text=0xffbfe890 "map q3dm17") at ../qcommon/cmd.c:645 #18 0x00046f50 in Cbuf_Execute () at ../qcommon/cmd.c:220 #19 0x0004c894 in Com_Frame () at ../qcommon/common.c:2707 #20 0x000c7dcc in main (argc=18, argv=0xffbff1fc) at ../unix/unix_main.c:1323 I'll try to gather more info when I get home tonight on a faster machine.. -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From tim at ngus.net Mon Nov 7 11:47:54 2005 From: tim at ngus.net (Tim Angus) Date: Mon, 7 Nov 2005 16:47:54 +0000 Subject: Quake3 client + server on Solaris/Sparc In-Reply-To: References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> Message-ID: <20051107164754.47525d88.tim@ngus.net> On Mon, 7 Nov 2005 17:38:25 +0100 (CET) vincent at cojot.name wrote: > I think your diffs are fine and should be commited. Heh, I'm not so sure. I don't really understand what the second MACOS_X define in cl_cin.c is about, and the first shouldn't be necessary anyway since LittleLong is used in the non-OSX version. I'm not really sure how much validity should be placed on those defines anyway since cinematics are disabled by default on OS X (cl_main.c): #ifdef MACOS_X // In game video is REALLY slow in Mac OS X right now due to driver slowness cl_inGameVideo = Cvar_Get ("r_inGameVideo", "0", CVAR_ARCHIVE); #else cl_inGameVideo = Cvar_Get ("r_inGameVideo", "1", CVAR_ARCHIVE); #endif From vincent at cojot.name Tue Nov 8 04:38:42 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Tue, 8 Nov 2005 10:38:42 +0100 (CET) Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: <20051107164754.47525d88.tim@ngus.net> References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> <20051107164754.47525d88.tim@ngus.net> Message-ID: Some more information on this problem: It's the compiler that kills the 'release' versions.. or rather the optimizations performed by the compiler uncover some problems (alignments maybe?). On sparc, I'm using this version: $ gcc --version gcc (GCC) 3.4.3 (csl-sol210-3_4-branch+sol_rpath) Copyright (C) 2004 Free Software Foundation, Inc. (Note the 'csl-sol210-3_4-branch+sol_rpath', it's a SUN-customized version). I just compiled and ran both the release client and server by setting OPTIMIZE to "-O0" for sparc. I'll be investigating the alignment optimizations performed by that compiler... Stay tuned, -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From vincent at cojot.name Wed Nov 9 09:33:53 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 9 Nov 2005 15:33:53 +0100 (CET) Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> <20051107164754.47525d88.tim@ngus.net> Message-ID: Tim, Ludwig, Could you please commit the following workaround? It makes the 'release' version work on Solaris/Sparc (s9 and s10) by working around some of the optimizations that make the executables segfault. It's ugly but at least it will make both the debug and release versions work out of the box for people who compile from source. Tested on: # SunOS 5.9 Generic_118558-14 sun4u SUNW,Sun-Blade-100 / gcc3 (GCC) 3.3.5 # SunOS 5.10 Generic_118822-11 sun4u SUNW,Ultra-5_10 / gcc (GCC) 3.4.3 (csl-sol210-3_4-branch+sol_rpath) # SunOS 5.10 Generic_118822-11 sun4u SUNW,Sun-Blade-1000 / gcc (GCC) 3.4.3 (csl-sol210-3_4-branch+sol_rpath) Thanks, -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost -------------- next part -------------- Index: code/unix/Makefile =================================================================== --- code/unix/Makefile (revision 323) +++ code/unix/Makefile (working copy) @@ -391,12 +391,20 @@ GL_CFLAGS = -I/usr/openwin/include endif - OPTIMIZE = -O3 -ffast-math -funroll-loops -fomit-frame-pointer - + # Here's the scoop: Q3 'release' dies when using -Os, -O1, -02 but + # it survives some less global optimization flags. So we're getting + # this ugly 'OPTIMIZE' until we get the other stuff fixed.. + # Verified on: + # SunOS 5.9 Generic_118558-14 sun4u SUNW,Sun-Blade-100 / gcc3 (GCC) 3.3.5 + # SunOS 5.10 Generic_118822-11 sun4u SUNW,Ultra-5_10 / gcc (GCC) 3.4.3 (csl-sol210-3_4-branch+sol_rpath) + # SunOS 5.10 Generic_118822-11 sun4u SUNW,Sun-Blade-1000 / gcc (GCC) 3.4.3 (csl-sol210-3_4-branch+sol_rpath) ifeq ($(ARCH),sparc) - OPTIMIZE = -O3 -fomit-frame-pointer -ffast-math -falign-loops=2 \ - -falign-jumps=2 -falign-functions=2 -fstrength-reduce \ - -funroll-loops + OPTIMIZE = -O0 -m32 -mptr32 -fomit-frame-pointer -munaligned-doubles \ + -floop-optimize -fif-conversion -fmerge-constants -fthread-jumps \ + -fexpensive-optimizations -frename-registers -fdelayed-branch \ + -fschedule-insns -finline-functions -fstrength-reduce \ + -funroll-loops -falign-loops=2 -falign-jumps=2 -fomit-frame-pointer \ + -mcpu=ultrasparc -mtune=ultrasparc -mv8plus -ffast-math -mhard-float BASE_CFLAGS += -DNO_VM_COMPILED else ifeq ($(ARCH),i386) @@ -406,7 +414,7 @@ endif endif - DEBUG_CFLAGS = $(BASE_CFLAGS) -g -O0 + DEBUG_CFLAGS = $(BASE_CFLAGS) -ggdb -O0 RELEASE_CFLAGS=$(BASE_CFLAGS) -DNDEBUG $(OPTIMIZE) From ludwig.nussel at suse.de Wed Nov 9 09:54:18 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Wed, 9 Nov 2005 15:54:18 +0100 Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> <20051107164754.47525d88.tim@ngus.net> Message-ID: <20051109145417.GA7223@suse.de> vincent at cojot.name wrote: > Could you please commit the following workaround? > > It makes the 'release' version work on Solaris/Sparc (s9 and s10) by > working around some of the optimizations that make the executables > segfault. > > It's ugly but at least it will make both the debug and release versions > work out of the box for people who compile from source. A note about possible miscompilation/undiscovered bugs triggered by optimizations belong to the README. I vote against this ugly thing in the Makefile. Did you run the optimized version through purify? Maybe it helps us to get a clue about what happens. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From vincent at cojot.name Wed Nov 9 11:02:08 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Wed, 9 Nov 2005 17:02:08 +0100 (CET) Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: <20051109145417.GA7223@suse.de> References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> <20051107164754.47525d88.tim@ngus.net> <20051109145417.GA7223@suse.de> Message-ID: On Wed, 9 Nov 2005, Ludwig Nussel wrote: > A note about possible miscompilation/undiscovered bugs triggered by > optimizations belong to the README. I vote against this ugly thing > in the Makefile. The comments, the compile flags or both? :) > Did you run the optimized version through purify? Maybe it helps us > to get a clue about what happens. I ran it and didn't get anything isolately significant.. I will try harder over the week-end.. Thanks, -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From vincent at cojot.name Thu Nov 10 14:56:44 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Thu, 10 Nov 2005 20:56:44 +0100 (CET) Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: <20051109145417.GA7223@suse.de> References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> <20051107164754.47525d88.tim@ngus.net> <20051109145417.GA7223@suse.de> Message-ID: Hi Ludwig and Tim, On Wed, 9 Nov 2005, Ludwig Nussel wrote: > A note about possible miscompilation/undiscovered bugs triggered by > optimizations belong to the README. I vote against this ugly thing > in the Makefile. I agree, how about this patch? Can we at least get this into the Makefile so it makes the 'release' builds work on Sparc until we can fix the problems and use heavier optimizations? > Did you run the optimized version through purify? Maybe it helps us > to get a clue about what happens. Yes, I ran it and with the optimized version and purify, I got a lot more UMR's (Uninitialized Memory Read) than with the debug version. I ran it with: releasesparc/SunOSquake3 +set vm_game 1 +set vm_cgame 1 +set vm_ui 1 +set sv_pure 1 +set ttycon 0 +map q3dm17 It died inside the botlib initialization. Purify shows 'MAW' (Misaligned Memory Write) just before the crash. I got the following trace but I cannot make anything of it: MAW: Misaligned memory write This is occurring while in: PC_ClearTokenWhiteSpace [l_precomp.c:1103] void PC_ClearTokenWhiteSpace(token_t *token) { token->whitespace_p = NULL; => token->endwhitespace_p = NULL; token->linescrossed = 0; } //end of the function PC_ClearTokenWhiteSpace //============================================================================ PC_ReadDirective [l_precomp.c:2540] { if (!strcmp(directives[i].name, token.string)) { => return directives[i].func(source); } //end if } //end for } //end if PC_ReadToken [l_precomp.c:2724] #endif //QUAKC { //read the precompiler directive => if (!PC_ReadDirective(source)) return qfalse; continue; } //end if } //end if LoadWeaponConfig [be_ai_weap.c:240] wc->numweapons = max_weaponinfo; wc->numprojectiles = 0; //parse the source file => while(PC_ReadToken(source, &token)) { if (!strcmp(token.string, "weaponinfo")) { BotSetupWeaponAI [be_ai_weap.c:509] char *file; file = LibVarString("weaponconfig", "weapons.c"); => weaponconfig = LoadWeaponConfig(file); if (!weaponconfig) { botimport.Print(PRT_FATAL, "couldn't load the weapon config\n"); Export_BotLibSetup [be_interface.c:155] if (errnum != BLERR_NOERROR) return errnum; errnum = EA_Setup(); //be_ea.c if (errnum != BLERR_NOERROR) return errnum; => errnum = BotSetupWeaponAI(); //be_ai_weap.c if (errnum != BLERR_NOERROR)return errnum; errnum = BotSetupGoalAI(); //be_ai_goal.c if (errnum != BLERR_NOERROR) return errnum; Misaligned write of 8 bytes to 0xc18a04 in the heap. Address 0xc18a04 is 1177756 bytes into a malloc'd block at 0xaf9168 of 16777216 bytes. This block was allocated from: malloc [rtlib.o] calloc [rtlib.o] Com_InitZoneMemory [common.c:1397] } // bk001205 - was malloc => mainzone = calloc( s_zoneTotal, 1 ); if ( !mainzone ) { Com_Error( ERR_FATAL, "Zone data failed to allocate %i megs", s_zoneTotal / (1024*1024) ); } Com_Init [common.c:2382] main [unix_main.c:1306] _start [crt1.o] Then there's the crash: SIG: Signal handled This is occurring while in: PC_ClearTokenWhiteSpace [l_precomp.c:1103] PC_ReadDirective [l_precomp.c:2540] PC_ReadToken [l_precomp.c:2724] LoadWeaponConfig [be_ai_weap.c:240] BotSetupWeaponAI [be_ai_weap.c:509] Export_BotLibSetup [be_interface.c:155] Received signal 10 (SIGBUS - Bus Error) Faulting address = 0xc18a04 Signal mask: (SIGHUP | SIGINT | SIGQUIT | SIGILL | SIGTRAP | \ SIGABRT | SIGEMT | SIGFPE | SIGSEGV | SIGSYS | SIGPIPE | SIGALRM | \ SIGTERM | SIGUSR1 | SIGUSR2 | SIGCHLD | SIGPWR | SIGWINCH | SIGURG | \ SIGPOLL | SIGTSTP | SIGCONT | SIGTTIN | SIGTTOU | SIGVTALRM | \ SIGPROF | SIGXCPU | SIGXFSZ | SIGWAITING | SIGLWP) Pending signals: Best regards, -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost -------------- next part -------------- Index: i_o-q3-readme =================================================================== --- i_o-q3-readme (revision 323) +++ i_o-q3-readme (working copy) @@ -116,6 +116,28 @@ 8. ???? 9. Profit! +Compiling and Running under Solaris/x86 or Solaris/Sparc +~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ +Follow the Linux instructions on installing, that should get you started. + +ioQ3 was tested on the the following systems: +- Solaris Express on x86 using Software OpenGL. +- Solaris 10/Sparc GA on a SunBlade 2000 with XVR-1000 graphics. +- Solaris 9/U8 on a SunBlade 150 with XVR-600 graphics. + +The dedicated server and client work on x86. +The dedicated server and client work on sparc + +Don't use anyting other than -O0 on sparc, otherwise it will crash due to alignment errors. +We're investigating this issue (Nov 10th 2005). + +The executables survive some less global optimization flags. So we're getting +this ugly 'OPTIMIZE' for the release build until we get the other stuff fixed.. +Verified on: +SunOS 5.9 Generic_118558-14 sun4u SUNW,Sun-Blade-100 / gcc3 (GCC) 3.3.5 +SunOS 5.10 Generic_118822-11 sun4u SUNW,Ultra-5_10 / gcc (GCC) 3.4.3 (csl-sol210-3_4-branch+sol_rpath) +SunOS 5.10 Generic_118822-11 sun4u SUNW,Sun-Blade-1000 / gcc (GCC) 3.4.3 (csl-sol210-3_4-branch+sol_rpath) + Creating mods compatible with Q3 1.32b ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ If you're using this package to create mods for the last official release of Index: code/unix/Makefile =================================================================== --- code/unix/Makefile (revision 323) +++ code/unix/Makefile (working copy) @@ -391,12 +391,13 @@ GL_CFLAGS = -I/usr/openwin/include endif - OPTIMIZE = -O3 -ffast-math -funroll-loops -fomit-frame-pointer - ifeq ($(ARCH),sparc) - OPTIMIZE = -O3 -fomit-frame-pointer -ffast-math -falign-loops=2 \ - -falign-jumps=2 -falign-functions=2 -fstrength-reduce \ - -funroll-loops + OPTIMIZE = -O0 -m32 -mptr32 -fomit-frame-pointer -munaligned-doubles \ + -floop-optimize -fif-conversion -fmerge-constants -fthread-jumps \ + -fexpensive-optimizations -frename-registers -fdelayed-branch \ + -fschedule-insns -finline-functions -fstrength-reduce \ + -funroll-loops -falign-loops=2 -falign-jumps=2 -fomit-frame-pointer \ + -mcpu=ultrasparc -mtune=ultrasparc -mv8plus -ffast-math -mhard-float BASE_CFLAGS += -DNO_VM_COMPILED else ifeq ($(ARCH),i386) @@ -406,7 +407,7 @@ endif endif - DEBUG_CFLAGS = $(BASE_CFLAGS) -g -O0 + DEBUG_CFLAGS = $(BASE_CFLAGS) -ggdb -O0 RELEASE_CFLAGS=$(BASE_CFLAGS) -DNDEBUG $(OPTIMIZE) From proton at mrtech.ru Thu Nov 10 16:34:06 2005 From: proton at mrtech.ru (Empty World) Date: Fri, 11 Nov 2005 00:34:06 +0300 Subject: global ways to develop? Message-ID: <4373BCCE.7000501@mrtech.ru> hello guyz.. i'm new to this list, so i want to ask some questions and give some ideas.. the world is a little happier now - q3 is opensource. what can it give to us? a) bring q3 to new platforms, such as solaris, mac os 9, etc.. b) make existing q3 faster c) bring new features in q3 i wanted to talk about new stuff that can be brought.. a year ago i dreamed about software-render in q3 (don't laugh ;)) but now i understand that it's 99,5% impossible. everybody agrees? there can be done lost of new features, that usual mod-making could not do. there can be a new, next-generation q3, maybe q3 for pro_s. but i think that is possible only if people make a community and start developing their hacks on this source code away from official (icculus.org), usual q3 sources.. anybody knows such already existing communities? i'd love to help if anybody decides make next-gen, pro-q3. p.s. hey guyz, maybe software-render is not hard to make, huh? :) anyway, sorry for my english etc bla bla bla.. -- ______ /_____/\ Alex KORCHAGIN /____ \\ \ //blurry// /_____\ \\ / Java & Developer Platforms Group /_____/ \/ / / /_____/ / \//\ s.r.o. \_____\//\ / / Empty World \____/ / /\ / q3.defrag->cpm.runrl.map, world record. \____/ \\ \ \____\ \\ Tel: +7 905 7699629 ICQ: 963975 \____\/ Email: proton at mrtech.ru -- From icculus at clutteredmind.org Thu Nov 10 16:49:47 2005 From: icculus at clutteredmind.org (Ryan C. Gordon) Date: Thu, 10 Nov 2005 16:49:47 -0500 Subject: [quake3] global ways to develop? In-Reply-To: <4373BCCE.7000501@mrtech.ru> References: <4373BCCE.7000501@mrtech.ru> Message-ID: <4373C07B.1020808@clutteredmind.org> > i wanted to talk about new stuff that can be brought.. a year ago i > dreamed about software-render in q3 (don't laugh ;)) but now i > understand that it's 99,5% impossible. everybody agrees? I don't agree; CPUs are much faster than they were in 1999, and there have been advances (Pixomatic, Swiftshader) that suggest this is possible...but not something I'm going to work on. > ______ > /_____/\ Alex KORCHAGIN > /____ \\ \ //blurry// > /_____\ \\ / Java & Developer Platforms Group > /_____/ \/ / / > /_____/ / \//\ s.r.o. > \_____\//\ / / Empty World > \____/ / /\ / q3.defrag->cpm.runrl.map, world record. > \____/ \\ \ > \____\ \\ Tel: +7 905 7699629 ICQ: 963975 > \____\/ Email: proton at mrtech.ru I think that's supposed to be a Sun Microsystems logo, but it really looks like a swastica from here. --ryan. From zakk at timedoctor.org Thu Nov 10 16:52:36 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 10 Nov 2005 16:52:36 -0500 Subject: [quake3] global ways to develop? In-Reply-To: <4373BCCE.7000501@mrtech.ru> References: <4373BCCE.7000501@mrtech.ru> Message-ID: <1131659556.4373c124a4e22@mail.icculus.org> Quoting Empty World : > a) bring q3 to new platforms, such as solaris, mac os 9, etc.. We're already working on that, solaris is being worked on I believe. Check out the status part of the web page for updates on that or just keep reading the ML. > b) make existing q3 faster time, right?> This isn't particularly interesting, though I'm not adverse to improvments as long as they don't break compatability. Quake 3 runs ridiculously fast on any nearly modern system, even using laptop chipsets like the intel 855GM in my thinkpad. > c) bring new features in q3 Well, new features are mostly mods, you can do more comprhensive mods these days with regard to additional features, post-source release. Though it would be prudent for them to be stand-alone games since you can't have 1.32 compatability once you've gone beyond the standard mod SDK and into post-source modify-the-whole-tree-for-your-mod land. Which isn't to say you can't have 1.32 compatability if you set your sights within the confines of the old SDK. Timbo could probably explain that part better :) > i wanted to talk about new stuff that can be brought.. a year ago i > dreamed about software-render in q3 (don't laugh ;)) but now i > understand that it's 99,5% impossible. everybody agrees? > there can be done lost of new features, that usual mod-making could not > do. there can be a new, next-generation q3, maybe q3 for pro_s. but i > think that is possible only if people make a community and start > developing their hacks on this source code away from official > (icculus.org), usual q3 sources.. I'd certainly welcome a software renderer patch, though I don't know how difficult it would be. When you say next generation, I assume you mean "enhanced graphics". That is certainly doable though it might require art talent in addition to coding skills. Nobody has ponied up to doing that kind of enhanced graphics feature yet... As far as "pro_s" I'm not sure what you're talking about, professional gamers, the CPL, etc? I'm not particularly interested in them, and their needs have always been fufilled by standard mods. With regard to people breaking away from icculus.org/quake3, I'm not sure why you would discuss that here. The modifications made are certainly GPL and people are welcome to take them and develop their own. I fully expect to see Larry C0d3h4x0r taking these changes to make his cel-shaded quake 3 or whatever. However, this isn't the place to discuss that, larry and his friends would have made their own project somewhere else. Though I would suggest they track the changes made here, since there will almost inevitably be more bugfixes and platforms supported down the road. I fully intend to see SDL support for quake3 on windows before I die, and that might be interesting for other people as well. It would probably make mingw building easier, too. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Thu Nov 10 16:53:23 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 10 Nov 2005 16:53:23 -0500 Subject: [quake3] global ways to develop? In-Reply-To: <4373C07B.1020808@clutteredmind.org> References: <4373BCCE.7000501@mrtech.ru> <4373C07B.1020808@clutteredmind.org> Message-ID: <1131659603.4373c15360367@mail.icculus.org> Quoting "Ryan C. Gordon" : > > > > ______ > > /_____/\ Alex KORCHAGIN > > /____ \\ \ //blurry// > > /_____\ \\ / Java & Developer Platforms Group > > /_____/ \/ / / > > /_____/ / \//\ s.r.o. > > \_____\//\ / / Empty World > > \____/ / /\ / q3.defrag->cpm.runrl.map, world record. > > \____/ \\ \ > > \____\ \\ Tel: +7 905 7699629 ICQ: 963975 > > \____\/ Email: proton at mrtech.ru > > I think that's supposed to be a Sun Microsystems logo, but it really > looks like a swastica from here. > > --ryan. I think I just piss pantsed myself. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From proton at mrtech.ru Thu Nov 10 17:10:19 2005 From: proton at mrtech.ru (Empty World) Date: Fri, 11 Nov 2005 01:10:19 +0300 Subject: [quake3] global ways to develop? In-Reply-To: <1131659603.4373c15360367@mail.icculus.org> References: <4373BCCE.7000501@mrtech.ru> <4373C07B.1020808@clutteredmind.org> <1131659603.4373c15360367@mail.icculus.org> Message-ID: <4373C54B.2000806@mrtech.ru> Zachary J. Slater wrote: > Quoting "Ryan C. Gordon" : > >> >>> ______ >>> /_____/\ Alex KORCHAGIN >>> /____ \\ \ //blurry// >>> /_____\ \\ / Java & Developer Platforms Group >>> /_____/ \/ / / >>> /_____/ / \//\ s.r.o. >>> \_____\//\ / / Empty World >>> \____/ / /\ / q3.defrag->cpm.runrl.map, world record. >>> \____/ \\ \ >>> \____\ \\ Tel: +7 905 7699629 ICQ: 963975 >>> \____\/ Email: proton at mrtech.ru >>> >> I think that's supposed to be a Sun Microsystems logo, but it really >> looks like a swastica from here. >> >> --ryan. >> > > I think I just piss pantsed myself. > > i agree, logo is a mess.. it was good til someone touched my thunderbird (: and look of this logo depends on font you use by default ;) anyway, that's not the point. the point is that i want someone to write a software render (or take if from any opensouce project) and make q3 work with it. if i could, i'd have done it by myself (: -- -------------- next part -------------- An HTML attachment was scrubbed... URL: From laurent.desnogues at wanadoo.fr Thu Nov 10 17:15:28 2005 From: laurent.desnogues at wanadoo.fr (Laurent Desnogues) Date: Thu, 10 Nov 2005 23:15:28 +0100 Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> <20051107164754.47525d88.tim@ngus.net> <20051109145417.GA7223@suse.de> Message-ID: <4373C680.3010706@wanadoo.fr> vincent at cojot.name wrote: > Yes, I ran it and with the optimized version and purify, I got a lot > more UMR's (Uninitialized Memory Read) than with the debug version. I often have this problem with Purify. It's mostly due to structures being copied as a whole even when some of their bytes are not used. One can call it noise, but there can be some bug lurking there :/ > It died inside the botlib initialization. Purify shows 'MAW' (Misaligned > Memory Write) just before the crash. I got the following trace but I > cannot make anything of it: > > MAW: Misaligned memory write > This is occurring while in: > PC_ClearTokenWhiteSpace [l_precomp.c:1103] > void PC_ClearTokenWhiteSpace(token_t *token) > { > token->whitespace_p = NULL; > => token->endwhitespace_p = NULL; > token->linescrossed = 0; > } //end of the function PC_ClearTokenWhiteSpace I think you can launch a debugger when Purify finds such a problem (never tried, but saw it documented). Anyway the problem here is that the offset into the aligned block is a multiple of 4 and not of 8 and you are compiling in 64 bit mode (could you try in 32 bit mode?). My *very wild* guess is that given the size of the original allocation (16 MB) Q3 is doing its own memory management into the area and aligns on multiple of 4 bytes instead of 8. By looking very quickly at the svn repository another wild guess is that the GetMemory in l_memory.c might be wrong. There are two instances of it depending on whether MEMORYMANEGER is defined or not. Anyway I don't see any aligment enforcement there, whereas the alignment is enforced in qcommon/common.c/Z_TagMalloc. Hope it may provide some hint... Regards, Laurent From laurent.desnogues at wanadoo.fr Thu Nov 10 17:19:27 2005 From: laurent.desnogues at wanadoo.fr (Laurent Desnogues) Date: Thu, 10 Nov 2005 23:19:27 +0100 Subject: [quake3] global ways to develop? In-Reply-To: <1131659556.4373c124a4e22@mail.icculus.org> References: <4373BCCE.7000501@mrtech.ru> <1131659556.4373c124a4e22@mail.icculus.org> Message-ID: <4373C76F.3000305@wanadoo.fr> Zachary J. Slater wrote: > > I'd certainly welcome a software renderer patch, though I don't know how > difficult it would be. The very best way to start from is static linking Mesa and optimize it for the paths being used in Q3. I did that for Q2 and it was fun. Anyway in the end you'd be left with a big amount of C code to optimize at assembly level ;-) Laurent From zakk at timedoctor.org Thu Nov 10 17:25:12 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Thu, 10 Nov 2005 17:25:12 -0500 Subject: [quake3] global ways to develop? In-Reply-To: <4373C54B.2000806@mrtech.ru> References: <4373BCCE.7000501@mrtech.ru> <4373C07B.1020808@clutteredmind.org> <1131659603.4373c15360367@mail.icculus.org> <4373C54B.2000806@mrtech.ru> Message-ID: <1131661512.4373c8c85d04a@mail.icculus.org> Quoting Empty World : > anyway, that's not the point. the point is that i want someone to write > a software render (or take if from any opensouce project) and make q3 > work with it. if i could, I'd have done it by myself (: > I'd like a 7800 to magically pop into existence on top of a pony that rides up to my house and tells me which patches to accept and which to not. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From tim at ngus.net Thu Nov 10 18:22:59 2005 From: tim at ngus.net (Tim Angus) Date: Thu, 10 Nov 2005 23:22:59 +0000 Subject: global ways to develop? In-Reply-To: <4373C07B.1020808@clutteredmind.org> References: <4373BCCE.7000501@mrtech.ru> <4373C07B.1020808@clutteredmind.org> Message-ID: <20051110232259.05094caa.tim@ngus.net> On Thu, 10 Nov 2005 16:49:47 -0500 Ryan wrote: > I think that's supposed to be a Sun Microsystems logo, but it really > looks like a swastica from here. One represents an oppressive regime bent on tyranny and world domination, whereas the other represents a political party... easy to get the two mixed up I guess. From vincent at cojot.name Fri Nov 11 02:45:53 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Fri, 11 Nov 2005 08:45:53 +0100 (CET) Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: <4373C680.3010706@wanadoo.fr> References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> <20051107164754.47525d88.tim@ngus.net> <20051109145417.GA7223@suse.de> <4373C680.3010706@wanadoo.fr> Message-ID: On Thu, 10 Nov 2005, Laurent Desnogues wrote: Hi Laurent, >> It died inside the botlib initialization. Purify shows 'MAW' (Misaligned >> Memory Write) just before the crash. I got the following trace but I cannot >> make anything of it: >> >> MAW: Misaligned memory write >> This is occurring while in: >> PC_ClearTokenWhiteSpace [l_precomp.c:1103] >> void PC_ClearTokenWhiteSpace(token_t *token) >> { >> token->whitespace_p = NULL; >> => token->endwhitespace_p = NULL; >> token->linescrossed = 0; >> } //end of the function PC_ClearTokenWhiteSpace > > I think you can launch a debugger when Purify finds > such a problem (never tried, but saw it documented). > > Anyway the problem here is that the offset into the > aligned block is a multiple of 4 and not of 8 and > you are compiling in 64 bit mode (could you try in > 32 bit mode?). That's the problem.. I'm -not- trying to compile with -m64, I compiling with -m32, of course.. > My *very wild* guess is that given the size of the > original allocation (16 MB) Q3 is doing its own > memory management into the area and aligns on > multiple of 4 bytes instead of 8. By looking very > quickly at the svn repository another wild guess is > that the GetMemory in l_memory.c might be wrong. > There are two instances of it depending on whether > MEMORYMANEGER is defined or not. Anyway I don't see > any aligment enforcement there, whereas the > alignment is enforced in qcommon/common.c/Z_TagMalloc. Ok, thanks for the ideas.. You've read more q3 source than I did.. :) Cheers, -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From laurent.desnogues at wanadoo.fr Fri Nov 11 05:14:11 2005 From: laurent.desnogues at wanadoo.fr (Laurent Desnogues) Date: Fri, 11 Nov 2005 11:14:11 +0100 Subject: [quake3] Re: Quake3 client + server on Solaris/Sparc In-Reply-To: References: <20051106170038.2d87c79a.tim@ngus.net> <436F78CA.7000900@clutteredmind.org> <20051107160750.7084af16.tim@ngus.net> <20051107161037.730f8779.tim@ngus.net> <20051107164754.47525d88.tim@ngus.net> <20051109145417.GA7223@suse.de> <4373C680.3010706@wanadoo.fr> Message-ID: <43746EF3.6060106@wanadoo.fr> vincent at cojot.name wrote: > On Thu, 10 Nov 2005, Laurent Desnogues wrote: > > Hi Laurent, > >>> It died inside the botlib initialization. Purify shows 'MAW' >>> (Misaligned Memory Write) just before the crash. I got the following >>> trace but I cannot make anything of it: >>> >>> MAW: Misaligned memory write >>> This is occurring while in: >>> PC_ClearTokenWhiteSpace [l_precomp.c:1103] >>> void PC_ClearTokenWhiteSpace(token_t *token) >>> { >>> token->whitespace_p = NULL; >>> => token->endwhitespace_p = NULL; >>> token->linescrossed = 0; >>> } //end of the function PC_ClearTokenWhiteSpace >> >> >> I think you can launch a debugger when Purify finds >> such a problem (never tried, but saw it documented). >> >> Anyway the problem here is that the offset into the >> aligned block is a multiple of 4 and not of 8 and >> you are compiling in 64 bit mode (could you try in >> 32 bit mode?). > > > That's the problem.. I'm -not- trying to compile with -m64, I compiling > with -m32, of course.. Ha... Anyway my hypothesis about the memory not being aligned may still hold: gcc is trying to clear memory 64 bits at a time ("Misaligned write of 8 bytes" as Purify says). Could you provide the assembly listing of PC_ClearTokenWhiteSpace? Either gdb output or the result of gcc -S. Laurent From zakk at timedoctor.org Sun Nov 13 20:04:20 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Sun, 13 Nov 2005 17:04:20 -0800 Subject: OpenAL support Message-ID: <20051114010420.GB11103@timedoctor.org> Just because nobody else has mentioned it, Timbo worked on BlackAura's openal patch a bit, then ci'd it. Please test it out on your systems and comment on it here and in the irc channel. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From F.Everdij at citg.tudelft.nl Mon Nov 14 15:22:32 2005 From: F.Everdij at citg.tudelft.nl (Frank Everdij) Date: Mon, 14 Nov 2005 21:22:32 +0100 Subject: IRIX port running Message-ID: <1131999752.4378f20880459@mech001.citg.tudelft.nl> Hi all, I'm Frank/dexter1 and together with Mare and some of my mates we have ported icculus.org Quake3 to SGI IRIX systems. Currently we have it running pretty clean on Octane2 systems and O2 systems. I have it also working on older I2 Impact systems, but am experiencing some texture glitches. Therefore, please do not drop the IRIX section from the Makefile, as we obviously need it. It could use some touchups, but we're still testing builds and trying out Makefile changes.... Sound is clean and working fine. We've ported SDL as a community free software project residing at www.nekochan.net, where one can download the necessary supporting libraries and programs for quake3 That's a first holler from me. Hope to keep you updated with specifics, like broken Cinematics and bots... See ya dexter1 -- drs Frank Everdij Email:F.Everdij at citg.tudelft.nl Tel:01527-88202 Room:6.34 System Administrator for Structural Mechanics Dept. of Civil Engineering TU Delft From zakk at timedoctor.org Mon Nov 14 16:32:44 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 14 Nov 2005 16:32:44 -0500 Subject: [quake3] IRIX port running In-Reply-To: <1131999752.4378f20880459@mech001.citg.tudelft.nl> References: <1131999752.4378f20880459@mech001.citg.tudelft.nl> Message-ID: <1132003964.4379027c9f4c5@mail.icculus.org> Sounds good, let me know when you can give us a patch. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From vincent at cojot.name Mon Nov 14 16:45:58 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Mon, 14 Nov 2005 22:45:58 +0100 (CET) Subject: [quake3] IRIX port running In-Reply-To: <1131999752.4378f20880459@mech001.citg.tudelft.nl> References: <1131999752.4378f20880459@mech001.citg.tudelft.nl> Message-ID: Hey Frank, Congratulations!!!! Can't wait to test it on my dual-V12 Octane2.. :) Many thanks, Vincent On Mon, 14 Nov 2005, Frank Everdij wrote: > Hi all, > > I'm Frank/dexter1 and together with Mare and some of my mates we have ported > icculus.org Quake3 to SGI IRIX systems. Currently we have it running pretty > clean on Octane2 systems and O2 systems. I have it also working on older I2 > Impact systems, but am experiencing some texture glitches. > > Therefore, please do not drop the IRIX section from the Makefile, as we > obviously need it. It could use some touchups, but we're still testing builds > and trying out Makefile changes.... > > Sound is clean and working fine. We've ported SDL as a community free software > project residing at www.nekochan.net, where one can download the necessary > supporting libraries and programs for quake3 > > That's a first holler from me. Hope to keep you updated with specifics, like > broken Cinematics and bots... > > See ya > > dexter1 From lists at michaelstather.com Mon Nov 14 18:21:29 2005 From: lists at michaelstather.com (Michael Stather) Date: Tue, 15 Nov 2005 00:21:29 +0100 Subject: =?ISO-8859-1?Q?Can=B4t_compile_q3_svn_snaphot?= Message-ID: <43791BF9.4020507@michaelstather.com> Hi, I wanted to download and try the q3 version on my SuSE 10.0 system , but I can?t compile it: The error is: gcc -Wall -fno-strict-aliasing -Wimplicit -Wstrict-prototypes -DUSE_OPENAL=1 -DUSE_SDL_VIDEO=1 -DUSE_SDL_SOUND=1 -I/usr/include/SDL -D_REENTRANT -m32 -DNDEBUG -O3 -march=i686 -fomit-frame-pointer -ffast-math -funroll-loops -falign-loops=2 -falign-jumps=2 -falign-functions=2 -fstrength-reduce -MMD -o releasei386-glibc/client/qal.o -c ../client/qal.c In file included from ../client/qal.c:28: ../client/qal.h:35:19: error: AL/al.h: Datei oder Verzeichnis nicht gefunden ../client/qal.h:36:20: error: AL/alc.h: Datei oder Verzeichnis nicht gefunden make[2]: *** [releasei386-glibc/client/qal.o] Fehler 1 make[2]: Leaving directory `/home/michael/Desktop/quake3/code/unix' make[1]: *** [build_release] Fehler 2 make[1]: Leaving directory `/home/michael/Desktop/quake3/code/unix' make: *** [release] Fehler 2 What can I do? thanks in advance for your help! regards Michael From zakk at timedoctor.org Mon Nov 14 18:24:23 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 14 Nov 2005 18:24:23 -0500 Subject: [quake3] =?iso-8859-1?b?Q2FutHQ=?= compile q3 svn snaphot In-Reply-To: <43791BF9.4020507@michaelstather.com> References: <43791BF9.4020507@michaelstather.com> Message-ID: <1132010663.43791ca791317@mail.icculus.org> Quoting Michael Stather : > ../client/qal.h:35:19: error: AL/al.h: Datei oder Verzeichnis nicht gefunden > ../client/qal.h:36:20: error: AL/alc.h: Datei oder Verzeichnis nicht Install openal. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From lists at michaelstather.com Mon Nov 14 18:39:13 2005 From: lists at michaelstather.com (Michael Stather) Date: Tue, 15 Nov 2005 00:39:13 +0100 Subject: [quake3] =?ISO-8859-1?Q?Can=B4t_compile_q3_svn_snaph?= =?ISO-8859-1?Q?ot?= In-Reply-To: <1132010663.43791ca791317@mail.icculus.org> References: <43791BF9.4020507@michaelstather.com> <1132010663.43791ca791317@mail.icculus.org> Message-ID: <43792021.2020409@michaelstather.com> An HTML attachment was scrubbed... URL: From zakk at timedoctor.org Mon Nov 14 18:41:56 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 14 Nov 2005 18:41:56 -0500 Subject: [quake3] =?iso-8859-1?b?Q2FutHQ=?= compile q3 svn snaphot In-Reply-To: <43792021.2020409@michaelstather.com> References: <43791BF9.4020507@michaelstather.com> <1132010663.43791ca791317@mail.icculus.org> <43792021.2020409@michaelstather.com> Message-ID: <1132011716.437920c4946ec@mail.icculus.org> Please don't send HTML e-mail to this list. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From davebytes at comcast.net Tue Nov 15 00:21:28 2005 From: davebytes at comcast.net (David Chait) Date: Tue, 15 Nov 2005 00:21:28 -0500 Subject: did anyone get the demo content to work? References: <43791BF9.4020507@michaelstather.com> Message-ID: <00c001c5e9a4$6e541640$6e01a8c0@sixfour> I've been away from things for a while, but when I last tried to run against the demo content it failed (I recall some kind of version check or something). Did that ever get solved/workable (even if via a local makefile or #def change..)? I'm about to jump back into my porting work (ARM and MIPS embedded), excited to hear some OpenAL stuff just got in (since I'll be using a no-SDL base...). Oh, and I'm asking about the demo content 'path' as the platforms I'm looking toward have low ram, low storage, and the demo content would just be easier to work with (i.e., people could download the demo and grab the content, rather than needing a full Q3A install...) -- and this is just platform testing/demo purposes, not to have the full game playable... Thanks all, -d From vincent at cojot.name Tue Nov 15 03:05:45 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Tue, 15 Nov 2005 09:05:45 +0100 (CET) Subject: [quake3] IRIX port running In-Reply-To: <1131999752.4378f20880459@mech001.citg.tudelft.nl> References: <1131999752.4378f20880459@mech001.citg.tudelft.nl> Message-ID: On Mon, 14 Nov 2005, Frank Everdij wrote: > Hi all, > > I'm Frank/dexter1 and together with Mare and some of my mates we have ported > icculus.org Quake3 to SGI IRIX systems. Currently we have it running pretty > clean on Octane2 systems and O2 systems. I have it also working on older I2 > Impact systems, but am experiencing some texture glitches. Just curious, what texture glitches are you experiencing? My Solaris/Sparc experience has been less than satisfactory, even at the latest OpenGL and Graphics patch level: - on XVR-1000 graphics (MAJC-5200 based, 360Mb RAM), I had to disable q3's OpenGL extensions otherwise the walls would look mostly pitch black. - on XVR-600 graphics (3DLabs based, 128Mb RAM), I could run with increased speed using q3's OpenGL extensions -but- the sky (q3dm3, etc.) would never be right: The walls and buildings textures would 'splash' the sky, leading to a confusing display.. Anything similar on your port? Cheers, -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From F.Everdij at citg.tudelft.nl Tue Nov 15 03:40:09 2005 From: F.Everdij at citg.tudelft.nl (Frank Everdij) Date: Tue, 15 Nov 2005 09:40:09 +0100 Subject: [quake3] did anyone get the demo content to work? In-Reply-To: <00c001c5e9a4$6e541640$6e01a8c0@sixfour> References: <43791BF9.4020507@michaelstather.com> <00c001c5e9a4$6e541640$6e01a8c0@sixfour> Message-ID: <1132044009.43799ee93b325@mech001.citg.tudelft.nl> Quoting David Chait : > I've been away from things for a while, but when I last tried to run against > the demo content it failed (I recall some kind of version check or > something). Did that ever get solved/workable (even if via a local makefile > or #def change..)? I'm about to jump back into my porting work (ARM and > MIPS embedded), excited to hear some OpenAL stuff just got in (since I'll be > using a no-SDL base...). For the IRIX port i've been using the demo content and as long as you put the pak0.pk3 in baseq3, it's fine. You should use different startup options, as explained in the i_o-q3-readme I use: quake3 +set sv_pure 0 +set vm_cgame 0 +set vm_game 0 +set vm_ui 0 +set ttycon 0 +set r_fullscreen 0 +set bot_enable 0 +set in_nograb 1 +disconnect HTH Frank -- drs Frank Everdij Email:F.Everdij at citg.tudelft.nl Tel:01527-88202 Room:6.34 System Administrator for Structural Mechanics Dept. of Civil Engineering TU Delft From F.Everdij at citg.tudelft.nl Tue Nov 15 06:23:07 2005 From: F.Everdij at citg.tudelft.nl (Frank Everdij) Date: Tue, 15 Nov 2005 12:23:07 +0100 Subject: [quake3] IRIX port running In-Reply-To: References: <1131999752.4378f20880459@mech001.citg.tudelft.nl> Message-ID: <1132053787.4379c51b1d390@mech001.citg.tudelft.nl> Quoting vincent at cojot.name: > On Mon, 14 Nov 2005, Frank Everdij wrote: > > > Hi all, > > > > I'm Frank/dexter1 and together with Mare and some of my mates we have > > ported icculus.org Quake3 to SGI IRIX systems. Currently we have it > > running pretty clean on Octane2 systems and O2 systems. I have it also > > working on older I2 Impact systems, but am experiencing some > > texture glitches. > > Just curious, what texture glitches are you experiencing? > > My Solaris/Sparc experience has been less than satisfactory, even at the > latest OpenGL and Graphics patch level: > > - on XVR-1000 graphics (MAJC-5200 based, 360Mb RAM), I had to disable q3's > OpenGL extensions otherwise the walls would look mostly pitch black. > > - on XVR-600 graphics (3DLabs based, 128Mb RAM), I could run with > increased speed using q3's OpenGL extensions -but- the sky (q3dm3, etc.) > would never be right: The walls and buildings textures would 'splash' the > sky, leading to a confusing display.. > > Anything similar on your port? Hm, nor really. On Impact systems (I2 and Octane) we experience texture corruption like: - Rainbow effects on objects, i.e. rotating armors looks like a patchwork of color-rotating banded textures. - Sky is sometimes black, or has greenish stripe texture, which change when you move (!) - Digits, like the Hitpoint counter, is readable, but the textures are shifted downwards halfway, so i only see the top of the digits. Where the top of the digits should be, i only see garbage. - Similar to the digits, the crosshair, which is default a circle with a point in the center, looks like a crescent moon. The menu and initial screens are fine, only when the arena is loaded i see these artifacts. I infer that this is a deficiency on IMPACT systems, possibly the limited texture size of 4 MB. O2's which have arbitrary sized texture support, and Octane2 VPRO with texture sizes of atleast 8-13 MB do not seem to be affected. I'll test again someday soon with extensions disabled, to see if this is the culprit. HTH, Frank -- drs Frank Everdij Email:F.Everdij at citg.tudelft.nl Tel:01527-88202 Room:6.34 System Administrator for Structural Mechanics Dept. of Civil Engineering TU Delft From dino at e4a.it Tue Nov 15 14:27:42 2005 From: dino at e4a.it (Dino Puller) Date: Tue, 15 Nov 2005 20:27:42 +0100 Subject: OpenAL + linu Message-ID: <437A36AE.8080500@e4a.it> Hi all, how can i use q3 with openal? I've recompiled with USE_OPENAL=1 and seta s_useOpenAL "1" in q3config but the game insist use the sdl backend. If i execute ls -n | grep linuxquake | grep openal i see the correct libopenal loaded. tnx, Dino From tim at ngus.net Tue Nov 15 14:39:34 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 15 Nov 2005 19:39:34 +0000 Subject: OpenAL + linu In-Reply-To: <437A36AE.8080500@e4a.it> References: <437A36AE.8080500@e4a.it> Message-ID: <20051115193934.2ce2997e.tim@ngus.net> On Tue, 15 Nov 2005 20:27:42 +0100 Dino wrote: > how can i use q3 with openal? I've recompiled with USE_OPENAL=1 and > > seta s_useOpenAL "1" in q3config but the game insist use the sdl > backend. If i execute ls -n | grep linuxquake | grep openal i see the > correct libopenal loaded. Please provide a complete condump. From dino at e4a.it Tue Nov 15 14:49:14 2005 From: dino at e4a.it (Dino Puller) Date: Tue, 15 Nov 2005 20:49:14 +0100 Subject: [quake3] Re: OpenAL + linu In-Reply-To: <20051115193934.2ce2997e.tim@ngus.net> References: <437A36AE.8080500@e4a.it> <20051115193934.2ce2997e.tim@ngus.net> Message-ID: <437A3BBA.4050406@e4a.it> Tim Angus wrote: >On Tue, 15 Nov 2005 20:27:42 +0100 Dino wrote: > > >> how can i use q3 with openal? I've recompiled with USE_OPENAL=1 and >> >>seta s_useOpenAL "1" in q3config but the game insist use the sdl >>backend. If i execute ls -n | grep linuxquake | grep openal i see the >>correct libopenal loaded. >> >> > >Please provide a complete condump. > > -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: out URL: From tim at ngus.net Tue Nov 15 15:07:21 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 15 Nov 2005 20:07:21 +0000 Subject: OpenAL + linu In-Reply-To: <437A3BBA.4050406@e4a.it> References: <437A36AE.8080500@e4a.it> <20051115193934.2ce2997e.tim@ngus.net> <437A3BBA.4050406@e4a.it> Message-ID: <20051115200721.35ec5b7f.tim@ngus.net> On Tue, 15 Nov 2005 20:49:14 +0100 Dino wrote: > >Please provide a complete condump. ------- sound initialization ------- SDL Audio driver initializing... SDL audio driver is "alsa" Starting SDL audio callback... SDL audio initialized. ------------------------------------ ----- Sound Info ----- sound system is muted 1 stereo 32768 samples 16 samplebits 1 submission_chunk 22050 speed 0x8a66808 dma buffer No background file. ---------------------- Are you certain OpenAL support is compiled in an s_useOpenAL is 1? Furthermore, this output is rather odd. You should expect to see something like this when SDL is used: ------ Initializing Sound ------ Initializing SDL audio driver... SDL audio driver is "dsp". SDL_AudioSpec: Format: AUDIO_S16LSB Freq: 22050 Samples: 512 Channels: 2 Starting SDL audio callback... SDL audio initialized. ----- Sound Info ----- 1 stereo 16384 samples 16 samplebits 1 submission_chunk 22050 speed 0x8b644f8 dma buffer No background file. ---------------------- Sound intialization successful. -------------------------------- Are you sure you're using the latest version from svn? From anthaus at tutopia.com Tue Nov 15 16:18:04 2005 From: anthaus at tutopia.com (Carlos Alberto Lopez Romero) Date: Tue, 15 Nov 2005 15:18:04 -0600 Subject: [quake3] OpenAL + linu In-Reply-To: <437A36AE.8080500@e4a.it> References: <437A36AE.8080500@e4a.it> Message-ID: <437A508C.9050009@tutopia.com> I had to link symbollically my libopenal.so from /usr/local/lib to /usr/lib/ My sound initialization output is the following: ------ Initializing Sound ------ Allocated 64 sources. OpenAL info: Vendor: OpenAL Community Version: 1.1 Renderer: Software Extensions: AL_LOKI_quadriphonic AL_LOKI_play_position AL_LOKI_WAVE_format AL_LOKI_IMA_ADPCM_format AL_LOKI_buffer_data_callback ALC_LOKI_audio_channel Sound intialization successful. -------------------------------- I hope this helps. Dino Puller wrote: > Hi all, > how can i use q3 with openal? I've recompiled with USE_OPENAL=1 and > seta s_useOpenAL "1" in q3config but the game insist use the sdl > backend. If i execute ls -n | grep linuxquake | grep openal i see the > correct libopenal loaded. > > tnx, > Dino > > > > ______________________________ Visita http://www.tutopia.com y comienza a navegar m?s r?pido en Internet. Tutopia es Internet para todos. From lists at michaelstather.com Tue Nov 15 19:11:04 2005 From: lists at michaelstather.com (Michael Stather) Date: Wed, 16 Nov 2005 01:11:04 +0100 Subject: Performance win/lin Message-ID: <437A7918.6090607@michaelstather.com> Hi, I?ve a notebook with a "Mobile Radeon M6" chip with 16MB VRAM in it. Under Windows the performance of quake3 SVN is ok, but when I try it under linux it?s almost unplayable. I?m using the open-source "radeon" driver and SuSE 10.0. What can I do here? regards Michael From zakk at timedoctor.org Tue Nov 15 19:16:56 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Tue, 15 Nov 2005 16:16:56 -0800 Subject: [quake3] Performance win/lin In-Reply-To: <437A7918.6090607@michaelstather.com> References: <437A7918.6090607@michaelstather.com> Message-ID: <1132100216.437a7a78573dd@mail.icculus.org> Quoting Michael Stather : > Hi, > > I?ve a notebook with a "Mobile Radeon M6" chip with 16MB VRAM in it. > Under Windows the performance of quake3 SVN is ok, but when I try it > under linux it?s almost unplayable. I?m using the open-source "radeon" > driver and SuSE 10.0. What can I do here? > > regards > > Michael > Either the drivers aren't working under Linux, or they just suck. Neither is surprising, both are possible. Most likely is that they aren't working and you either need the official ati driver or to figure out DRI better. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From anthaus at tutopia.com Tue Nov 15 20:17:35 2005 From: anthaus at tutopia.com (Carlos Alberto Lopez Romero) Date: Tue, 15 Nov 2005 19:17:35 -0600 Subject: [quake3] Performance win/lin In-Reply-To: <437A7918.6090607@michaelstather.com> References: <437A7918.6090607@michaelstather.com> Message-ID: <437A88AF.90403@tutopia.com> Get the proprietary ATi Drivers for Linux at ATi's homepage. Quake 3 should be playable, but don't expect those drivers to perform quite as well as the ones for windows. If you need help installing them, please, PLEASE go to www.rage3d.com and ask on the linux forum, not here. Michael Stather wrote: > Hi, > > I?ve a notebook with a "Mobile Radeon M6" chip with 16MB VRAM in it. > Under Windows the performance of quake3 SVN is ok, but when I try it > under linux it?s almost unplayable. I?m using the open-source "radeon" > driver and SuSE 10.0. What can I do here? > > regards > > Michael > ______________________________ Visita http://www.tutopia.com y comienza a navegar m?s r?pido en Internet. Tutopia es Internet para todos. From arny at ats.s.bawue.de Tue Nov 15 19:20:11 2005 From: arny at ats.s.bawue.de (Thilo Schulz) Date: Wed, 16 Nov 2005 01:20:11 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <437A7918.6090607@michaelstather.com> References: <437A7918.6090607@michaelstather.com> Message-ID: <200511160120.13547.arny@ats.s.bawue.de> On Wednesday 16 November 2005 01:11, Michael Stather wrote: > I?ve a notebook with a "Mobile Radeon M6" chip with 16MB VRAM in it. > Under Windows the performance of quake3 SVN is ok, but when I try it > under linux it?s almost unplayable. I?m using the open-source "radeon" > driver and SuSE 10.0. What can I do here? thilo at Thilo /mnt $ glxinfo name of display: :0.0 display: :0 screen: 0 direct rendering: Yes glxinfo must return: direct rendering: Yes For your series, I think you need the DRI drivers working or else software rendering is used which is very slow. -- Thilo Schulz -------------- next part -------------- A non-text attachment was scrubbed... Name: not available Type: application/pgp-signature Size: 189 bytes Desc: not available URL: From badcdev at gmail.com Tue Nov 15 23:00:46 2005 From: badcdev at gmail.com (BlackAura) Date: Wed, 16 Nov 2005 14:00:46 +1000 Subject: [quake3] OpenAL support In-Reply-To: <20051114010420.GB11103@timedoctor.org> References: <20051114010420.GB11103@timedoctor.org> Message-ID: <200511161400.46391.badcdev@gmail.com> On Mon, 14 Nov 2005 11:04 am, Zachary J. Slater wrote: > Just because nobody else has mentioned it, Timbo > worked on BlackAura's openal patch a bit, then ci'd it. Ah, thanks. I'd forgotten about that. I think I might have a couple of updates to that laying around somewhere too. From lists at michaelstather.com Wed Nov 16 03:32:13 2005 From: lists at michaelstather.com (Michael Stather) Date: Wed, 16 Nov 2005 09:32:13 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <437A88AF.90403@tutopia.com> References: <437A7918.6090607@michaelstather.com> <437A88AF.90403@tutopia.com> Message-ID: <437AEE8D.2000002@michaelstather.com> Carlos Alberto Lopez Romero wrote: > Get the proprietary ATi Drivers for Linux at ATi's homepage. Quake 3 > should be playable, but don't expect those drivers to perform quite as > well as the ones for windows. If you need help installing them, > please, PLEASE go to www.rage3d.com and ask on the linux forum, not here. > > Michael Stather wrote: > >> Hi, >> >> I?ve a notebook with a "Mobile Radeon M6" chip with 16MB VRAM in it. >> Under Windows the performance of quake3 SVN is ok, but when I try it >> under linux it?s almost unplayable. I?m using the open-source >> "radeon" driver and SuSE 10.0. What can I do here? >> >> regards >> >> Michael >> > > ______________________________ > Visita http://www.tutopia.com y comienza a navegar m?s r?pido en > Internet. Tutopia es Internet para todos. > > As said I?ve a Mobile Radeon and the proprietary drivers only support newer cards. Direct rendering is enabled too. So If there?s no other thing I can do thanks for your help. But I wonder anyway why the open-source drivers are so slow compared to the windows ones... From michel at daenzer.net Wed Nov 16 04:58:36 2005 From: michel at daenzer.net (Michel =?ISO-8859-1?Q?D=E4nzer?=) Date: Wed, 16 Nov 2005 10:58:36 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <437A7918.6090607@michaelstather.com> References: <437A7918.6090607@michaelstather.com> Message-ID: <1132135116.20075.73.camel@localhost> On Wed, 2005-11-16 at 01:11 +0100, Michael Stather wrote: > > I?ve a notebook with a "Mobile Radeon M6" chip with 16MB VRAM in it. > Under Windows the performance of quake3 SVN is ok, but when I try it > under linux it?s almost unplayable. I?m using the open-source "radeon" > driver and SuSE 10.0. What can I do here? Hard to say without seeing your X server log file at least, but with that little video RAM, my first guess would be that texture memory is tight. You can try increasing the amount of AGP memory used with Option "GARTSize", or if you're running in depth 24, try 16 instead. -- Earthling Michel D?nzer | Debian (powerpc), X and DRI developer Libre software enthusiast | http://svcs.affero.net/rm.php?r=daenzer From dino at e4a.it Wed Nov 16 05:26:17 2005 From: dino at e4a.it (Dino Puller) Date: Wed, 16 Nov 2005 11:26:17 +0100 Subject: [quake3] OpenAL + linu In-Reply-To: <437A508C.9050009@tutopia.com> References: <437A36AE.8080500@e4a.it> <437A508C.9050009@tutopia.com> Message-ID: <437B0949.4090407@e4a.it> Solved! I'm a bit silly, i've used the old name of the executable : linuxquake3 instead of the new quake3 :\ thanks anyway for your help Dino Carlos Alberto Lopez Romero wrote: > I had to link symbollically my libopenal.so from /usr/local/lib to > /usr/lib/ > > My sound initialization output is the following: > > ------ Initializing Sound ------ > Allocated 64 sources. > OpenAL info: > Vendor: OpenAL Community > Version: 1.1 > Renderer: Software > Extensions: AL_LOKI_quadriphonic AL_LOKI_play_position > AL_LOKI_WAVE_format AL_LOKI_IMA_ADPCM_format > AL_LOKI_buffer_data_callback ALC_LOKI_audio_channel > Sound intialization successful. > -------------------------------- > > > I hope this helps. > > Dino Puller wrote: > >> Hi all, >> how can i use q3 with openal? I've recompiled with USE_OPENAL=1 and >> seta s_useOpenAL "1" in q3config but the game insist use the sdl >> backend. If i execute ls -n | grep linuxquake | grep openal i see >> the correct libopenal loaded. >> >> tnx, >> Dino >> >> >> >> > > ______________________________ > Visita http://www.tutopia.com y comienza a navegar m?s r?pido en > Internet. Tutopia es Internet para todos. From eugenydzh at jdevelop.com Wed Nov 16 06:30:39 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Wed, 16 Nov 2005 13:30:39 +0200 Subject: Sys_Error: ZIP FILE FSEEK NOT YET IMPLEMENTED Message-ID: <20051116113039.GN4637@115-2.trifle.net> Hi there! I recently upgraded sources from SVN, and now i'm gettign this error: Sys_Error: ZIP FILE FSEEK NOT YET IMPLEMENTED any ideas why is that and how to fix? Thanks in advance! -- Eugene N Dzhurinsky From tim at ngus.net Wed Nov 16 15:23:55 2005 From: tim at ngus.net (Tim Angus) Date: Wed, 16 Nov 2005 20:23:55 +0000 Subject: Sys_Error: ZIP FILE FSEEK NOT YET IMPLEMENTED In-Reply-To: <20051116113039.GN4637@115-2.trifle.net> References: <20051116113039.GN4637@115-2.trifle.net> Message-ID: <20051116202355.0ac75806.tim@ngus.net> On Wed, 16 Nov 2005 13:30:39 +0200 Eugeny wrote: > Hi there! > I recently upgraded sources from SVN, and now i'm gettign this error: > Sys_Error: ZIP FILE FSEEK NOT YET IMPLEMENTED > > any ideas why is that and how to fix? Fixed in revision 363. From demodevil at gmail.com Wed Nov 16 18:51:24 2005 From: demodevil at gmail.com (Slash) Date: Wed, 16 Nov 2005 18:51:24 -0500 Subject: [quake3] Performance win/lin In-Reply-To: <1132135116.20075.73.camel@localhost> References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> Message-ID: The open source drivers don't have 3D acceleration support at all, afaik. Same with the "nv" open source nvidia drivers. If you want to do 3D and such, you need the proprietary drivers in both cases. And of course they don't support your card, so you are out of luck. Maybe you can find an old version of their Linux drivers or something? Good luck :P On 11/16/05, Michel D?nzer wrote: > On Wed, 2005-11-16 at 01:11 +0100, Michael Stather wrote: > > > > I?ve a notebook with a "Mobile Radeon M6" chip with 16MB VRAM in it. > > Under Windows the performance of quake3 SVN is ok, but when I try it > > under linux it?s almost unplayable. I?m using the open-source "radeon" > > driver and SuSE 10.0. What can I do here? > > Hard to say without seeing your X server log file at least, but with > that little video RAM, my first guess would be that texture memory is > tight. You can try increasing the amount of AGP memory used with Option > "GARTSize", or if you're running in depth 24, try 16 instead. > > > -- > Earthling Michel D?nzer | Debian (powerpc), X and DRI developer > Libre software enthusiast | http://svcs.affero.net/rm.php?r=daenzer > From zakk at timedoctor.org Wed Nov 16 18:56:34 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 16 Nov 2005 15:56:34 -0800 Subject: [quake3] Performance win/lin In-Reply-To: References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> Message-ID: <1132185394.437bc73221c63@mail.icculus.org> Quoting Slash : > The open source drivers don't have 3D acceleration support at all, > afaik. Same with the "nv" open source nvidia drivers. If you want to > do 3D and such, you need the proprietary drivers in both cases. And of > course they don't support your card, so you are out of luck. Maybe you > can find an old version of their Linux drivers or something? > > Good luck :P This is untrue, glxinfo showing that there is direct rendering indicates support. What might be happening is that the game is simply not working well at the bit depth he is using, or something else wrong. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From lists at michaelstather.com Fri Nov 18 06:52:27 2005 From: lists at michaelstather.com (Michael Stather) Date: Fri, 18 Nov 2005 12:52:27 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <1132135116.20075.73.camel@localhost> References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> Message-ID: <437DC07B.90208@michaelstather.com> An HTML attachment was scrubbed... URL: From michel at daenzer.net Fri Nov 18 07:08:42 2005 From: michel at daenzer.net (Michel =?ISO-8859-1?Q?D=E4nzer?=) Date: Fri, 18 Nov 2005 13:08:42 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <437DC07B.90208@michaelstather.com> References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> <437DC07B.90208@michaelstather.com> Message-ID: <1132315723.7235.6.camel@localhost> On Fri, 2005-11-18 at 12:52 +0100, Michael Stather wrote: > Michel D?nzer wrote: > > On Wed, 2005-11-16 at 01:11 +0100, Michael Stather wrote: > > > > > I?ve a notebook with a "Mobile Radeon M6" chip with 16MB VRAM in it. > > > Under Windows the performance of quake3 SVN is ok, but when I try it > > > under linux it?s almost unplayable. I?m using the open-source "radeon" > > > driver and SuSE 10.0. What can I do here? > > > > > > > Hard to say without seeing your X server log file at least, but with > > that little video RAM, my first guess would be that texture memory is > > tight. You can try increasing the amount of AGP memory used with Option > > "GARTSize", or if you're running in depth 24, try 16 instead. > > > I?m just curious about the performance on Windows and the one on Linux > with the same settings. > Is this normal that the drivers are slower wit a factor about 2? Make sure that everything is really the same. The windows drivers will most definitely use more AGP memory than the free Linux drivers do by default, and Windows apps can switch depth at runtime whereas X apps can't. All that said though, the Windows drivers obviously receive a lot more engineering effort from people with better access to information about the hardware, so some difference is always to be expected. -- Earthling Michel D?nzer | Debian (powerpc), X and DRI developer Libre software enthusiast | http://svcs.affero.net/rm.php?r=daenzer From lists at michaelstather.com Fri Nov 18 07:20:29 2005 From: lists at michaelstather.com (Michael Stather) Date: Fri, 18 Nov 2005 13:20:29 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <1132315723.7235.6.camel@localhost> References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> <437DC07B.90208@michaelstather.com> <1132315723.7235.6.camel@localhost> Message-ID: <437DC70D.4080102@michaelstather.com> An HTML attachment was scrubbed... URL: From michel at daenzer.net Fri Nov 18 07:32:22 2005 From: michel at daenzer.net (Michel =?ISO-8859-1?Q?D=E4nzer?=) Date: Fri, 18 Nov 2005 13:32:22 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <437DC70D.4080102@michaelstather.com> References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> <437DC07B.90208@michaelstather.com> <1132315723.7235.6.camel@localhost> <437DC70D.4080102@michaelstather.com> Message-ID: <1132317142.7235.7.camel@localhost> On Fri, 2005-11-18 at 13:20 +0100, Michael Stather wrote: > > Is the GARTSize an option in the x config file? Yes, see the radeon(4x) manpage. -- Earthling Michel D?nzer | Debian (powerpc), X and DRI developer Libre software enthusiast | http://svcs.affero.net/rm.php?r=daenzer From lists at michaelstather.com Fri Nov 18 08:45:27 2005 From: lists at michaelstather.com (Michael Stather) Date: Fri, 18 Nov 2005 14:45:27 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <1132317142.7235.7.camel@localhost> References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> <437DC07B.90208@michaelstather.com> <1132315723.7235.6.camel@localhost> <437DC70D.4080102@michaelstather.com> <1132317142.7235.7.camel@localhost> Message-ID: <437DDAF7.1010702@michaelstather.com> An HTML attachment was scrubbed... URL: From davebytes at comcast.net Fri Nov 18 11:47:52 2005 From: davebytes at comcast.net (David Chait) Date: Fri, 18 Nov 2005 11:47:52 -0500 Subject: [quake3] Performance win/lin References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> <437DC07B.90208@michaelstather.com> Message-ID: <001b01c5ec5f$d173c280$6e01a8c0@sixfour> Hugely. If you're only seeing a factor of 2, that's pretty good. ATI has an enormous driver development team, many of whom are dedicated to raw performance optimization. I can promise you that the generic radeon linux drivers don't have that level of tuning, though ATI's linux drivers should incorporate the major enhancements. -d ----- Original Message ----- From: Michael Stather To: quake3 at icculus.org Sent: Friday, November 18, 2005 6:52 AM Subject: Re: [quake3] Performance win/lin Michel D?nzer wrote: On Wed, 2005-11-16 at 01:11 +0100, Michael Stather wrote: I?ve a notebook with a "Mobile Radeon M6" chip with 16MB VRAM in it. Under Windows the performance of quake3 SVN is ok, but when I try it under linux it?s almost unplayable. I?m using the open-source "radeon" driver and SuSE 10.0. What can I do here? Hard to say without seeing your X server log file at least, but with that little video RAM, my first guess would be that texture memory is tight. You can try increasing the amount of AGP memory used with Option "GARTSize", or if you're running in depth 24, try 16 instead. I?m just curious about the performance on Windows and the one on Linux with the same settings. Is this normal that the drivers are slower wit a factor about 2? -------------- next part -------------- An HTML attachment was scrubbed... URL: From zakk at timedoctor.org Fri Nov 18 13:32:36 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Fri, 18 Nov 2005 10:32:36 -0800 Subject: [quake3] Performance win/lin In-Reply-To: <437DDAF7.1010702@michaelstather.com> References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> <437DC07B.90208@michaelstather.com> <1132315723.7235.6.camel@localhost> <437DC70D.4080102@michaelstather.com> <1132317142.7235.7.camel@localhost> <437DDAF7.1010702@michaelstather.com> Message-ID: <20051118183236.GA9787@timedoctor.org> STOP SENDING HTML E-MAIL TO THIS LIST. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From wakko at generation.no Fri Nov 18 13:37:52 2005 From: wakko at generation.no (Anders Storsveen) Date: Fri, 18 Nov 2005 19:37:52 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <20051118183236.GA9787@timedoctor.org> References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> <437DC07B.90208@michaelstather.com> <1132315723.7235.6.camel@localhost> <437DC70D.4080102@michaelstather.com> <1132317142.7235.7.camel@localhost> <437DDAF7.1010702@michaelstather.com> <20051118183236.GA9787@timedoctor.org> Message-ID: <437E1F80.8080306@generation.no> Zachary J. Slater wrote: >STOP SENDING HTML E-MAIL TO THIS LIST. > > I second that From debeuk at gmail.com Fri Nov 18 13:56:04 2005 From: debeuk at gmail.com (Sven Beukenex) Date: Fri, 18 Nov 2005 19:56:04 +0100 Subject: [quake3] Performance win/lin In-Reply-To: <437E1F80.8080306@generation.no> References: <437A7918.6090607@michaelstather.com> <1132135116.20075.73.camel@localhost> <437DC07B.90208@michaelstather.com> <1132315723.7235.6.camel@localhost> <437DC70D.4080102@michaelstather.com> <1132317142.7235.7.camel@localhost> <437DDAF7.1010702@michaelstather.com> <20051118183236.GA9787@timedoctor.org> <437E1F80.8080306@generation.no> Message-ID: <9a224f8b0511181056r3abde321s60b83023b7529485@mail.gmail.com> On 11/18/05, Anders Storsveen wrote: > > Zachary J. Slater wrote: > > >STOP SENDING HTML E-MAIL TO THIS LIST. > > > > > I second that > hear hear -- Why are the pretty ones always insane? -- J.G. Thirlwell -------------- next part -------------- An HTML attachment was scrubbed... URL: From oliver.boettcher at onlinehome.de Sat Nov 19 05:08:53 2005 From: oliver.boettcher at onlinehome.de (=?UTF-8?B?T2xpdmVyIELDtnR0Y2hlcg==?=) Date: Sat, 19 Nov 2005 11:08:53 +0100 Subject: Distorted Sound on OSX Message-ID: <437EF9B5.6090607@onlinehome.de> Hi. I've build basic Quake3 w/ Xcode 1.5 on my G4 with non-altivec for OSX 10.3, because of these graphic glitches. I use the pak0.pk3 packed in the demo for testing. I get terrible sound with this build, while the roq-videos are sounding good. The latest ID-pointrelese is also working. I've no glue if something is messed up on my Panther, or if this is a problem with Xcode 1.5 with gcc3.3. Unfortunatly I'm not able to test on other systems. If other informations should be provided, I'll do. Btw. porting Q3 from the OSX soundcore to OpenAL would also be as good as porting the input to SDL. I'm working on this but it's a huge question of my spare time. Sys-Specs: G4 (400Mhz) "appled" GF5200 (stock PC card) 256MB RAM (runs better than expected on this system) --oli From tim at ngus.net Sat Nov 19 07:35:20 2005 From: tim at ngus.net (Tim Angus) Date: Sat, 19 Nov 2005 12:35:20 +0000 Subject: Distorted Sound on OSX In-Reply-To: <437EF9B5.6090607@onlinehome.de> References: <437EF9B5.6090607@onlinehome.de> Message-ID: <20051119123520.150cd8c5.tim@ngus.net> On Sat, 19 Nov 2005 11:08:53 +0100 Oliver wrote: > I've build basic Quake3 w/ Xcode 1.5 on my G4 with non-altivec for OSX > 10.3, because of these graphic glitches. I use the pak0.pk3 packed in > the demo for testing. I get terrible sound with this build, while the > roq-videos are sounding good. The latest ID-pointrelese is also > working. I've no glue if something is messed up on my Panther, or if > this is a problem with Xcode 1.5 with gcc3.3. Unfortunatly I'm not > able to test on other systems. > If other informations should be provided, I'll do. It's probably worth trying a build pre revision 343. This is where OpenAL and more importantly, the new codec system were introduced. It's possible that the wav decoder doesn't deal with a particular big endian case. > Btw. porting Q3 from the OSX soundcore to OpenAL would also be as good > as porting the input to SDL. I'm working on this but it's a huge > question of my spare time. Cool. > Sys-Specs: > G4 (400Mhz) > "appled" GF5200 (stock PC card) > 256MB RAM > (runs better than expected on this system) Neat. From ludwig.nussel at suse.de Sat Nov 19 10:08:23 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Sat, 19 Nov 2005 16:08:23 +0100 Subject: versioning scheme Message-ID: <200511191608.23169.ludwig.nussel@suse.de> Hi, - XQF uses the first bytes of the version cvar to detect the actual type of a server that speaks the q3 query protocol. Since ioQ3 != Q3 that doesn't work with this project. If other (Q3 compatible) projects use the same designation scheme detecting Q3 becomes impossible. - claiming we are the official successor of Q3 by continuing the version scheme of Q3 is probably a bit arrogant (1.32 -> 1.33). - the pk3 files stay the same as on 1.32, so the game in fact is still Q3 1.32 I therefore like to propose to use a new versioning scheme in a new namespace, e.g. "icculus.org/quake3 0.1" or "ioquake3 0.1". The latter would be file system friendlier. cu Ludwig Index: code/unix/Makefile =================================================================== --- code/unix/Makefile (Revision 367) +++ code/unix/Makefile (Arbeitskopie) @@ -83,10 +83,6 @@ JPDIR=$(MOUNT_DIR)/jpeg-6 SPLNDIR=$(MOUNT_DIR)/splines -# extract version info -VERSION=$(shell grep Q3_VERSION ../qcommon/q_shared.h | \ - sed -e 's/.*Q3\ \(.*\)"/\1/') - ############################################################################# # SETUP AND BUILD -- LINUX ############################################################################# Index: code/qcommon/q_shared.h =================================================================== --- code/qcommon/q_shared.h (Revision 367) +++ code/qcommon/q_shared.h (Arbeitskopie) @@ -26,7 +26,8 @@ // q_shared.h -- included first by ALL program modules. // A user mod should never modify this file -#define Q3_VERSION "ioQ3 1.33" +#define Q3_VERSION "Q3 1.32b" +#define Q3DERIVATE_VERSION "icculus.org/quake3 0.1" // 1.32 released 7-10-2002 #define MAX_TEAMNAME 32 Index: code/qcommon/qcommon.h =================================================================== --- code/qcommon/qcommon.h (Revision 367) +++ code/qcommon/qcommon.h (Arbeitskopie) @@ -828,6 +828,8 @@ void Com_TouchMemory( void ); +const char* Com_Q3Version( void ); + // commandLine should not include the executable name (argv[0]) void Com_Init( char *commandLine ); void Com_Frame( void ); Index: code/qcommon/common.c =================================================================== --- code/qcommon/common.c (Revision 367) +++ code/qcommon/common.c (Arbeitskopie) @@ -2351,17 +2351,20 @@ } #endif +const char* Com_Q3Version(void) +{ + return Q3_VERSION " " PLATFORM_STRING " " __DATE__ " (" Q3DERIVATE_VERSION ")"; +} + /* ================= Com_Init ================= */ void Com_Init( char *commandLine ) { - char *s; + Com_Printf( "%s\n", Com_Q3Version()); - Com_Printf( "%s %s %s\n", Q3_VERSION, PLATFORM_STRING, __DATE__ ); - if ( setjmp (abortframe) ) { Sys_Error ("Error during initialization"); } @@ -2467,8 +2470,7 @@ Cmd_AddCommand ("changeVectors", MSG_ReportChangeVectors_f ); Cmd_AddCommand ("writeconfig", Com_WriteConfig_f ); - s = va("%s %s %s", Q3_VERSION, PLATFORM_STRING, __DATE__ ); - com_version = Cvar_Get ("version", s, CVAR_ROM | CVAR_SERVERINFO ); + com_version = Cvar_Get ("version", Com_Q3Version(), CVAR_ROM | CVAR_SERVERINFO ); Sys_Init(); Netchan_Init( Com_Milliseconds() & 0xffff ); // pick a port value that should be nice and random Index: code/client/cl_console.c =================================================================== --- code/client/cl_console.c (Revision 367) +++ code/client/cl_console.c (Arbeitskopie) @@ -586,6 +586,7 @@ // qhandle_t conShader; int currentColor; vec4_t color; + const char ver[] = Q3_VERSION " (" Q3DERIVATE_VERSION ") "; lines = cls.glconfig.vidHeight * frac; if (lines <= 0) @@ -618,13 +619,13 @@ re.SetColor( g_color_table[ColorIndex(COLOR_RED)] ); - i = strlen( Q3_VERSION ); + i = strlen( ver ); for (x=0 ; x References: <200511191608.23169.ludwig.nussel@suse.de> Message-ID: <20051119195255.GA9980@timedoctor.org> Ludwig, Please don't commit this just yet, I want to think about it for a bit. -- -Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From oliver.boettcher at onlinehome.de Sun Nov 20 07:09:20 2005 From: oliver.boettcher at onlinehome.de (=?ISO-8859-1?Q?Oliver_B=F6ttcher?=) Date: Sun, 20 Nov 2005 13:09:20 +0100 Subject: [quake3] Re: Distorted Sound on OSX In-Reply-To: <20051119123520.150cd8c5.tim@ngus.net> References: <437EF9B5.6090607@onlinehome.de> <20051119123520.150cd8c5.tim@ngus.net> Message-ID: <43806770.3030800@onlinehome.de> > It's probably worth trying a build pre revision 343. This is where > OpenAL and more importantly, the new codec system were introduced. It's > possible that the wav decoder doesn't deal with a particular big endian > case. Does the trick. rev342 works ok, and the current revision w/ OpenAL Framework enabled too, with some better performance, I think. Wasn't that big deal, exept that I'm very unfamiliar with that XCode thing. From icculus at icculus.org Sun Nov 20 09:47:11 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Sun, 20 Nov 2005 09:47:11 -0500 Subject: [quake3] Re: Distorted Sound on OSX In-Reply-To: <43806770.3030800@onlinehome.de> References: <437EF9B5.6090607@onlinehome.de> <20051119123520.150cd8c5.tim@ngus.net> <43806770.3030800@onlinehome.de> Message-ID: <43808C6F.8090408@icculus.org> > Does the trick. rev342 works ok, and the current revision w/ OpenAL > Framework enabled too, with some better performance, I think. Wasn't > that big deal, exept that I'm very unfamiliar with that XCode thing. Yeah, it's definitely not the codecs...OpenAL fixes it here. I'll take a look at the mixer, since I'm in the middle of ripping up the Mac code. --ryan. From icculus at icculus.org Sun Nov 20 10:36:13 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Sun, 20 Nov 2005 10:36:13 -0500 Subject: [quake3] Re: Distorted Sound on OSX In-Reply-To: <43808C6F.8090408@icculus.org> References: <437EF9B5.6090607@onlinehome.de> <20051119123520.150cd8c5.tim@ngus.net> <43806770.3030800@onlinehome.de> <43808C6F.8090408@icculus.org> Message-ID: <438097ED.2020700@icculus.org> > Yeah, it's definitely not the codecs...OpenAL fixes it here. I'll take a > look at the mixer, since I'm in the middle of ripping up the Mac code. Fixed in svn revision #372. --ryan. From icculus at icculus.org Sun Nov 20 10:42:13 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Sun, 20 Nov 2005 10:42:13 -0500 Subject: Mac work... Message-ID: <43809955.4030709@icculus.org> I spent some time moving the Mac version to the Makefiles and SDL, in hopes that we can eventually have one codepath for all platforms, more or less. This still needs some work, and there's some other things addressed in this patch that aren't necessarily Mac-specific (dlopen is replaced with SDL_LoadObject and SDL_GL_LoadLibrary, etc)...if anyone wants to take a look at the current state, here's the patch. There are still broken things...debug builds are hosed if you use Altivec (presumably for alignment reasons, etc), plus some other glitches that need to be weeded out, etc. This is not ready for primetime, especially if we're talking about making a release soon. --ryan. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: ioq3-macosx-ryan-1.diff URL: From yannick.francken at gmail.com Mon Nov 21 09:40:06 2005 From: yannick.francken at gmail.com (Yannick Francken) Date: Mon, 21 Nov 2005 15:40:06 +0100 Subject: Q3 Source documentation Message-ID: <9af75400511210640o16f4bce4r9926bdfa7579c95@mail.gmail.com> Hi, I am Yannick, a PhD student at the University of Hasselt in Belgium. I have to make some modifications to the Q3 engine for a side-project. I have to do things like projecting the rendered result into a CAVE, adding accurate shadows, projecting movies into the world, etc. Of course first I have to learn about those tons of c-code. Is there somewhere documentation about the code available? Or are there already possibilities in the engine to project a movie (with shadows where occluded). Or are there people who wants to cooperate with me for implementing some common techniques? Hope to hear from you, Yannick From eugenydzh at jdevelop.com Mon Nov 21 10:51:40 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Mon, 21 Nov 2005 17:51:40 +0200 Subject: weird sound Message-ID: <20051121155140.GA1081@115-2.trifle.net> After upgradem I noticed there are a lot of issues with sound, like this: WARNING: music file music/fla22k_06.wav is not 22k stereo Any ideas why is that and how to fix? -- Eugene N Dzhurinsky From tim at ngus.net Tue Nov 22 12:19:57 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 22 Nov 2005 17:19:57 +0000 Subject: weird sound In-Reply-To: <20051121155140.GA1081@115-2.trifle.net> References: <20051121155140.GA1081@115-2.trifle.net> Message-ID: <20051122171957.2de0990e.tim@ngus.net> On Mon, 21 Nov 2005 17:51:40 +0200 Eugeny wrote: > After upgradem I noticed there are a lot of issues with sound, like > this: > > WARNING: music file music/fla22k_06.wav is not 22k stereo You must have the "developer" cvar set to 1, otherwise you wouldn't get this warning. It is a benign warning by and large. > Any ideas why is that and how to fix? I assume you're using OpenAL and baseq3 (no mods). Perhaps your OpenAL installation is not properly functional. Try running with "+set s_useOpenAL 0" on the command line. From tim at ngus.net Tue Nov 22 12:26:43 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 22 Nov 2005 17:26:43 +0000 Subject: Q3 Source documentation In-Reply-To: <9af75400511210640o16f4bce4r9926bdfa7579c95@mail.gmail.com> References: <9af75400511210640o16f4bce4r9926bdfa7579c95@mail.gmail.com> Message-ID: <20051122172643.0470349b.tim@ngus.net> On Mon, 21 Nov 2005 15:40:06 +0100 Yannick wrote: > Of course first I have to learn about those tons of c-code. Is there > somewhere documentation about the code available? http://www.linux.ucla.edu/~phaethon/q3mc/q3mc.html http://www.soclose.de/q3doc/ These only really deal with the game code side of things (they predate the GPL release), so perhaps they aren't that much use to you. > Or are there already > possibilities in the engine to project a movie (with shadows where > occluded). Or are there people who wants to cooperate with me for > implementing some common techniques? I'm not aware of anybody having implemented shadow volumes in Q3 yet. There are plenty of sites explaining how on the web though, albeit not Q3 specific. The book (remember those? :)) "Game Programming Gems 1" has a relatively good article on shadow map based shadows; it may be worth a look. From tim at ngus.net Tue Nov 22 12:45:18 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 22 Nov 2005 17:45:18 +0000 Subject: Mac work... In-Reply-To: <43809955.4030709@icculus.org> References: <43809955.4030709@icculus.org> Message-ID: <20051122174518.02db29dc.tim@ngus.net> On Sun, 20 Nov 2005 10:42:13 -0500 Ryan wrote: > I spent some time moving the Mac version to the Makefiles and SDL, in > hopes that we can eventually have one codepath for all platforms, more > or less. Yay. > This still needs some work, and there's some other things addressed in > this patch that aren't necessarily Mac-specific (dlopen is replaced > with SDL_LoadObject and SDL_GL_LoadLibrary, etc)...if anyone wants to > take a look at the current state, here's the patch. > + BASE_CFLAGS += -DMACOS_X=1 -fno-common -faltivec I noticed while poking about with the Xcode build (via pearpc) that MACOS_X is defined by the build system. This seems a bit poo. Does any body know what the "definitive" OS X define is? A cursory glance over http://developer.apple.com/technotes/tn2002/tn2071.html suggests there isn't one! > -#include "qasm.h" > > - > #if id386 > +#include "qasm.h" I'm not sure I understand this. id386 is (indirectly) defined by qasm.h, so moving the #include to after the #if means that the contents of the #if will never been seen -- on any platform. A powerpc build should not #be including these files anyway since they are x86. > There are still broken things...debug builds are hosed if you use > Altivec (presumably for alignment reasons, etc) It looks as if the Altivec build is fairly broken in general (and has been since day 1). It would be nice if somebody skilled in the ways of __VEC__ could take a look at it, but it seems Mac developers are thin on the ground. From eugenydzh at jdevelop.com Tue Nov 22 13:21:11 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Tue, 22 Nov 2005 20:21:11 +0200 Subject: [quake3] Re: weird sound In-Reply-To: <20051122171957.2de0990e.tim@ngus.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> Message-ID: <20051122182111.GA1150@jdevelop.zssm.zp.ua> On Tue, Nov 22, 2005 at 05:19:57PM +0000, Tim Angus wrote: > > Any ideas why is that and how to fix? > I assume you're using OpenAL and baseq3 (no mods). Perhaps your OpenAL > installation is not properly functional. Try running with "+set > s_useOpenAL 0" on the command line. well, i tried this - still no luck, sound crepitates and it is not possible at all to play game with sound turned on... -- Eugene N Dzhurinsky From tim at ngus.net Tue Nov 22 15:22:14 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 22 Nov 2005 20:22:14 +0000 Subject: weird sound In-Reply-To: <20051122182111.GA1150@jdevelop.zssm.zp.ua> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> Message-ID: <20051122202214.62d46cc9.tim@ngus.net> On Tue, 22 Nov 2005 20:21:11 +0200 Eugeny wrote: > On Tue, Nov 22, 2005 at 05:19:57PM +0000, Tim Angus wrote: > > > Any ideas why is that and how to fix? > > I assume you're using OpenAL and baseq3 (no mods). Perhaps your > > OpenAL installation is not properly functional. Try running with > > "+set s_useOpenAL 0" on the command line. > > well, i tried this - still no luck, sound crepitates and it is not > possible at all to play game with sound turned on... I'm not sure what else to suggest based on the scant information you have provided. What OS/arch is this? Condumps of initialisation using Open AL and not might be useful. From icculus at icculus.org Tue Nov 22 16:20:44 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Tue, 22 Nov 2005 16:20:44 -0500 Subject: [quake3] Re: Mac work... In-Reply-To: <20051122174518.02db29dc.tim@ngus.net> References: <43809955.4030709@icculus.org> <20051122174518.02db29dc.tim@ngus.net> Message-ID: <43838BAC.40706@icculus.org> > I noticed while poking about with the Xcode build (via pearpc) that > MACOS_X is defined by the build system. This seems a bit poo. Does any > body know what the "definitive" OS X define is? A cursory glance over > http://developer.apple.com/technotes/tn2002/tn2071.html suggests there > isn't one! (PearPC?! Dear lord, that must be painful...) There's isn't one...you can look for something like ((__DARWIN__) && (__APPLE__)), but that's always been awkward imho. Forcibly defining MACOS_X is fine, as far as I'm concerned, as long as we don't make the usual mistake of using it when we mean "powerpc" or "bigendian architecture" ... I'm sure we have a few of those in there, which I'll weed out next week when I build this on an x86 Mac. > I'm not sure I understand this. id386 is (indirectly) defined by qasm.h, > so moving the #include to after the #if means that the contents of the > #if will never been seen -- on any platform. A powerpc build should not > #be including these files anyway since they are x86. I was just trying to get it to compile at all here...I didn't realize id386 was defined in there. I'll change that back, and have the makefile only include that file on Intel Mac builds. > It looks as if the Altivec build is fairly broken in general (and has > been since day 1). It would be nice if somebody skilled in the ways of > __VEC__ could take a look at it, but it seems Mac developers are thin > on the ground. Actually, LerpMeshVertexes() is definitely broken in debug builds with Altivec, but everything else seems fine (including this function in release builds)...there are some other rendering issues, but they don't appear to be Altivec-related. --ryan. From eugenydzh at jdevelop.com Wed Nov 23 02:32:03 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Wed, 23 Nov 2005 09:32:03 +0200 Subject: [quake3] Re: weird sound In-Reply-To: <20051122202214.62d46cc9.tim@ngus.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> Message-ID: <20051123073203.GB770@115-2.trifle.net> On Tue, Nov 22, 2005 at 08:22:14PM +0000, Tim Angus wrote: > On Tue, 22 Nov 2005 20:21:11 +0200 Eugeny wrote: > > well, i tried this - still no luck, sound crepitates and it is not > > possible at all to play game with sound turned on... > > I'm not sure what else to suggest based on the scant information you > have provided. If you could describe what information you may need - it would be great! > What OS/arch is this? FreeBSD 5.4 stable on P-IV (i386). > Condumps of initialisation using Open AL and not might be useful. here you go 1.log is produced executing the command: quake3 +set s_useOpenAL 0 2.log is produced by just invoking quake3 Hope that's what you need. -- Eugene N Dzhurinsky -------------- next part -------------- A non-text attachment was scrubbed... Name: logs.tar.gz Type: application/octet-stream Size: 3657 bytes Desc: not available URL: From icculus at icculus.org Wed Nov 23 03:07:59 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Wed, 23 Nov 2005 03:07:59 -0500 Subject: [quake3] Re: weird sound In-Reply-To: <20051123073203.GB770@115-2.trifle.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> Message-ID: <4384235F.1090501@icculus.org> >>What OS/arch is this? > > > FreeBSD 5.4 stable on P-IV (i386). Possibly an SDL bug? Perhaps this just needs the s_sdl* cvars tweaked? --ryan. From eugenydzh at jdevelop.com Wed Nov 23 03:11:57 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Wed, 23 Nov 2005 10:11:57 +0200 Subject: [quake3] Re: weird sound In-Reply-To: <4384235F.1090501@icculus.org> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <4384235F.1090501@icculus.org> Message-ID: <20051123081157.GE770@115-2.trifle.net> On Wed, Nov 23, 2005 at 03:07:59AM -0500, Ryan C. Gordon wrote: > >FreeBSD 5.4 stable on P-IV (i386). > Possibly an SDL bug? Perhaps this just needs the s_sdl* cvars tweaked? So what, where and how needs to be fixed? =) -- Eugene N Dzhurinsky From icculus at icculus.org Wed Nov 23 04:00:30 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Wed, 23 Nov 2005 04:00:30 -0500 Subject: [quake3] Re: weird sound In-Reply-To: <20051123081157.GE770@115-2.trifle.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <4384235F.1090501@icculus.org> <20051123081157.GE770@115-2.trifle.net> Message-ID: <43842FAE.7090907@icculus.org> > So what, where and how needs to be fixed? =) These are the cvars in question: s_sdlBits = Cvar_Get("s_sdlBits", "16", CVAR_ARCHIVE); s_sdlSpeed = Cvar_Get("s_sdlSpeed", "0", CVAR_ARCHIVE); s_sdlChannels = Cvar_Get("s_sdlChannels", "2", CVAR_ARCHIVE); s_sdlDevSamps = Cvar_Get("s_sdlDevSamps", "0", CVAR_ARCHIVE); s_sdlMixSamps = Cvar_Get("s_sdlMixSamps", "0", CVAR_ARCHIVE); Obviously, this has nothing to do with the OpenAL codepath. You can tweak those from the command line and look at code/unix/sdl_snd.c for more information. You might want to try some other program that uses SDL audio and see if it's hosed, too. --ryan. From tim at ngus.net Wed Nov 23 07:21:47 2005 From: tim at ngus.net (Tim Angus) Date: Wed, 23 Nov 2005 12:21:47 +0000 Subject: weird sound In-Reply-To: <20051123073203.GB770@115-2.trifle.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> Message-ID: <20051123122147.22e5edb3.tim@ngus.net> On Wed, 23 Nov 2005 09:32:03 +0200 Eugeny wrote: > FreeBSD 5.4 stable on P-IV (i386). Neither the OpenAL or SDL code paths are used on FreeBSD. It still uses the DMA based sound system. That still doesn't really explain why it doesn't work. Does revision 342 work? It's possible there is some bad interaction between the new codec system and FreeBSD. Attached are two patches to enable SDL and OpenAL respectively. Could you test them (SDL first) and report whether either or both work. If so, they should really be enabled by default. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: enable-sdl-on-freebsd.diff URL: -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: enable-openal-on-freebsd.diff URL: From eugenydzh at jdevelop.com Wed Nov 23 07:36:38 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Wed, 23 Nov 2005 14:36:38 +0200 Subject: [quake3] Re: weird sound In-Reply-To: <20051123122147.22e5edb3.tim@ngus.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <20051123122147.22e5edb3.tim@ngus.net> Message-ID: <20051123123638.GC20429@115-2.trifle.net> On Wed, Nov 23, 2005 at 12:21:47PM +0000, Tim Angus wrote: > On Wed, 23 Nov 2005 09:32:03 +0200 Eugeny wrote: > > FreeBSD 5.4 stable on P-IV (i386). > > Neither the OpenAL or SDL code paths are used on FreeBSD. It still uses > the DMA based sound system. That still doesn't really explain why it > doesn't work. Does revision 342 work? It's possible there is some bad > interaction between the new codec system and FreeBSD. Attached are two > patches to enable SDL and OpenAL respectively. Could you test them (SDL > first) and report whether either or both work. If so, they should really > be enabled by default. first patch (openAL) won't build with error: In file included from ../client/qal.c:28: ../client/qal.h:35:19: AL/al.h: No such file or directory ../client/qal.h:36:20: AL/alc.h: No such file or directory and second (SDL) won't build too 386/client/jerror.o -c ../jpeg-6/jerror.c In file included from ../jpeg-6/../renderer/tr_local.h:31, from ../jpeg-6/jerror.c:22: ../jpeg-6/../renderer/qgl.h:53:19: GL/gl.h: No such file or directory ../jpeg-6/../renderer/qgl.h:54:20: GL/glx.h: No such file or directory In file included from ../jpeg-6/../renderer/tr_local.h:31, from ../jpeg-6/jerror.c:22: ../jpeg-6/../renderer/qgl.h:153: error: syntax error before "texture" ../jpeg-6/../renderer/qgl.h:153: warning: function declaration isn't a prototype ../jpeg-6/../renderer/qgl.h:154: error: syntax error before "texture" ../jpeg-6/../renderer/qgl.h:154: warning: function declaration isn't a prototype ../jpeg-6/../renderer/qgl.h:155: error: syntax error before "texture" ../jpeg-6/../renderer/qgl.h:155: warning: function declaration isn't a prototype ../jpeg-6/../renderer/qgl.h:157: warning: parameter names (without types) in function declaration -- Eugene N Dzhurinsky From tim at ngus.net Wed Nov 23 07:46:14 2005 From: tim at ngus.net (Tim Angus) Date: Wed, 23 Nov 2005 12:46:14 +0000 Subject: weird sound In-Reply-To: <20051123123638.GC20429@115-2.trifle.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <20051123122147.22e5edb3.tim@ngus.net> <20051123123638.GC20429@115-2.trifle.net> Message-ID: <20051123124614.38680068.tim@ngus.net> On Wed, 23 Nov 2005 14:36:38 +0200 Eugeny wrote: > first patch (openAL) won't build with error: > In file included from ../client/qal.c:28: > ../client/qal.h:35:19: AL/al.h: No such file or directory > ../client/qal.h:36:20: AL/alc.h: No such file or directory Obviously you need to install OpenAL headers to build with OpenAL. > and second (SDL) won't build too > 386/client/jerror.o -c ../jpeg-6/jerror.c > In file included from ../jpeg-6/../renderer/tr_local.h:31, > from ../jpeg-6/jerror.c:22: > ../jpeg-6/../renderer/qgl.h:53:19: GL/gl.h: No such file or directory > ../jpeg-6/../renderer/qgl.h:54:20: GL/glx.h: No such file or directory This makes no sense. The non-SDL build requires the GL headers too obviously, so I don't understand why the non-SDL build compiles (assuming it does) if this doesn't. Bottom line: check you have SDL, OpenAL and OpenGL correctly installed with headers for each. From eugenydzh at jdevelop.com Wed Nov 23 07:50:57 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Wed, 23 Nov 2005 14:50:57 +0200 Subject: [quake3] Re: weird sound In-Reply-To: <20051123124614.38680068.tim@ngus.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <20051123122147.22e5edb3.tim@ngus.net> <20051123123638.GC20429@115-2.trifle.net> <20051123124614.38680068.tim@ngus.net> Message-ID: <20051123125057.GD20429@115-2.trifle.net> On Wed, Nov 23, 2005 at 12:46:14PM +0000, Tim Angus wrote: > On Wed, 23 Nov 2005 14:36:38 +0200 Eugeny wrote: > > first patch (openAL) won't build with error: > > In file included from ../client/qal.c:28: > > ../client/qal.h:35:19: AL/al.h: No such file or directory > > ../client/qal.h:36:20: AL/alc.h: No such file or directory > Obviously you need to install OpenAL headers to build with OpenAL. ok > > and second (SDL) won't build too > > 386/client/jerror.o -c ../jpeg-6/jerror.c > > In file included from ../jpeg-6/../renderer/tr_local.h:31, > > from ../jpeg-6/jerror.c:22: > > ../jpeg-6/../renderer/qgl.h:53:19: GL/gl.h: No such file or directory > > ../jpeg-6/../renderer/qgl.h:54:20: GL/glx.h: No such file or directory > > This makes no sense. The non-SDL build requires the GL headers too > obviously, so I don't understand why the non-SDL build compiles > (assuming it does) if this doesn't. > > Bottom line: check you have SDL, OpenAL and OpenGL correctly installed > with headers for each. SDL is installed, openGL is installed together with X.org and nvidia driver, w/o these patches you sent quake builds w/o any error :( -- Eugene N Dzhurinsky From tim at ngus.net Wed Nov 23 08:08:09 2005 From: tim at ngus.net (Tim Angus) Date: Wed, 23 Nov 2005 13:08:09 +0000 Subject: weird sound In-Reply-To: <20051123125057.GD20429@115-2.trifle.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <20051123122147.22e5edb3.tim@ngus.net> <20051123123638.GC20429@115-2.trifle.net> <20051123124614.38680068.tim@ngus.net> <20051123125057.GD20429@115-2.trifle.net> Message-ID: <20051123130809.6dd97abd.tim@ngus.net> On Wed, 23 Nov 2005 14:50:57 +0200 Eugeny wrote: > SDL is installed, openGL is installed together with X.org and nvidia > driver, w/o these patches you sent quake builds w/o any error :( Try that attached one instead. Apparently "sdl-config --cflags" doesn't include an include path for the GL headers. -------------- next part -------------- An embedded and charset-unspecified text was scrubbed... Name: enable-sdl-on-freebsd_2.diff URL: From eugenydzh at jdevelop.com Wed Nov 23 08:27:29 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Wed, 23 Nov 2005 15:27:29 +0200 Subject: [quake3] Re: weird sound In-Reply-To: <20051123130809.6dd97abd.tim@ngus.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <20051123122147.22e5edb3.tim@ngus.net> <20051123123638.GC20429@115-2.trifle.net> <20051123124614.38680068.tim@ngus.net> <20051123125057.GD20429@115-2.trifle.net> <20051123130809.6dd97abd.tim@ngus.net> Message-ID: <20051123132729.GE20429@115-2.trifle.net> On Wed, Nov 23, 2005 at 01:08:09PM +0000, Tim Angus wrote: > On Wed, 23 Nov 2005 14:50:57 +0200 Eugeny wrote: > > SDL is installed, openGL is installed together with X.org and nvidia > > driver, w/o these patches you sent quake builds w/o any error :( > > Try that attached one instead. Apparently "sdl-config --cflags" doesn't > include an include path for the GL headers. Well, I tried this, still no luck, it won't build gcc -Wall -fno-strict-aliasing -Wimplicit -Wstrict-prototypes -DUSE_SDL_VIDEO=1 -DUSE_SDL_SOUND=1 -DNDEBUG -O3 -mtune=pentiumpro -march=pentium -fomit-frame-pointer -pipe -ffast-math -falign-loops=2 -falign-jumps=2 -falign-functions=2 -funroll-loops -fstrength-reduce -MMD -o releasei386/client/sdl_glimp.o -c ../unix/sdl_glimp.c -I/usr/X11R6/include ../unix/sdl_glimp.c:61:17: SDL.h: No such file or directory ../unix/sdl_glimp.c:91: error: syntax error before '*' token ../unix/sdl_glimp.c:91: warning: type defaults to `int' in declaration of `screen' SDL.h is located in /usr/X11R6/include/SDL11/SDL.h sdl11-config --cflags -I/usr/local/include/SDL11 -I/usr/local/include -D_REENTRANT I modified patch you sent me to use sdl11-config instead of sdl-config should i invoke gmake USE_SDL=1 or just gmake ? -- Eugene N Dzhurinsky -------------- next part -------------- Index: code/unix/Makefile =================================================================== --- code/unix/Makefile (revision 372) +++ code/unix/Makefile (working copy) @@ -298,6 +298,10 @@ BASE_CFLAGS = -Wall -fno-strict-aliasing -Wimplicit -Wstrict-prototypes + ifeq ($(USE_SDL),1) + BASE_CFLAGS += -DUSE_SDL_VIDEO=1 -DUSE_SDL_SOUND=1 $(shell sdl-config --cflags) + endif + GL_CFLAGS = -I/usr/X11R6/include DEBUG_CFLAGS=$(BASE_CFLAGS) -g @@ -327,7 +331,11 @@ # don't need -ldl (FreeBSD) LDFLAGS=-lm - CLIENT_LDFLAGS=-L/usr/X11R6/$(LIB) -lGL -lX11 -lXext -lXxf86dga -lXxf86vm + ifeq ($(USE_SDL),1) + CLIENT_LDFLAGS=$(shell sdl11-config --libs) + else + CLIENT_LDFLAGS=-L/usr/X11R6/$(LIB) -lGL -lX11 -lXext -lXxf86dga -lXxf86vm + endif ifeq ($(ARCH),axp) TARGETS=\ From eugenydzh at jdevelop.com Wed Nov 23 08:42:04 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Wed, 23 Nov 2005 15:42:04 +0200 Subject: [quake3] Re: weird sound In-Reply-To: <20051123130809.6dd97abd.tim@ngus.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <20051123122147.22e5edb3.tim@ngus.net> <20051123123638.GC20429@115-2.trifle.net> <20051123124614.38680068.tim@ngus.net> <20051123125057.GD20429@115-2.trifle.net> <20051123130809.6dd97abd.tim@ngus.net> Message-ID: <20051123134204.GG20429@115-2.trifle.net> On Wed, Nov 23, 2005 at 01:08:09PM +0000, Tim Angus wrote: > On Wed, 23 Nov 2005 14:50:57 +0200 Eugeny wrote: > > SDL is installed, openGL is installed together with X.org and nvidia > > driver, w/o these patches you sent quake builds w/o any error :( > > Try that attached one instead. Apparently "sdl-config --cflags" doesn't > include an include path for the GL headers. Okay i built it with gmake USE_SDL=1 installed, but still same problem - it complains about WARNING: music file music\sonic5.wav is not 22k stereo and sound still weird :( in previous versions everything was working fine -- Eugene N Dzhurinsky From tim at ngus.net Wed Nov 23 09:15:48 2005 From: tim at ngus.net (Tim Angus) Date: Wed, 23 Nov 2005 14:15:48 +0000 Subject: weird sound In-Reply-To: <20051123132729.GE20429@115-2.trifle.net> References: <20051121155140.GA1081@115-2.trifle.net> <20051122171957.2de0990e.tim@ngus.net> <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <20051123122147.22e5edb3.tim@ngus.net> <20051123123638.GC20429@115-2.trifle.net> <20051123124614.38680068.tim@ngus.net> <20051123125057.GD20429@115-2.trifle.net> <20051123130809.6dd97abd.tim@ngus.net> <20051123132729.GE20429@115-2.trifle.net> Message-ID: <20051123141548.1b9398b9.tim@ngus.net> On Wed, 23 Nov 2005 15:27:29 +0200 Eugeny wrote: > should i invoke > gmake USE_SDL=1 or just gmake ? Either. > in previous versions everything was working fine In which revision do you start to have problems? Try 342 first, then 353. From eugenydzh at jdevelop.com Wed Nov 23 09:37:39 2005 From: eugenydzh at jdevelop.com (Eugeny N Dzhurinsky) Date: Wed, 23 Nov 2005 16:37:39 +0200 Subject: [quake3] Re: weird sound In-Reply-To: <20051123141548.1b9398b9.tim@ngus.net> References: <20051122182111.GA1150@jdevelop.zssm.zp.ua> <20051122202214.62d46cc9.tim@ngus.net> <20051123073203.GB770@115-2.trifle.net> <20051123122147.22e5edb3.tim@ngus.net> <20051123123638.GC20429@115-2.trifle.net> <20051123124614.38680068.tim@ngus.net> <20051123125057.GD20429@115-2.trifle.net> <20051123130809.6dd97abd.tim@ngus.net> <20051123132729.GE20429@115-2.trifle.net> <20051123141548.1b9398b9.tim@ngus.net> Message-ID: <20051123143739.GA32612@115-2.trifle.net> On Wed, Nov 23, 2005 at 02:15:48PM +0000, Tim Angus wrote: > On Wed, 23 Nov 2005 15:27:29 +0200 Eugeny wrote: > > should i invoke > > gmake USE_SDL=1 or just gmake ? > Either. > > in previous versions everything was working fine > In which revision do you start to have problems? Try 342 first, then > 353. i did svn up -r 342 then built quake, everythign works fine, no problems with sound then i tried to go with svn up -r 353 and sound stopped to work correctly, it is keeping crepitating -- Eugene N Dzhurinsky From benh at kernel.crashing.org Wed Nov 23 22:38:53 2005 From: benh at kernel.crashing.org (Benjamin Herrenschmidt) Date: Thu, 24 Nov 2005 14:38:53 +1100 Subject: [quake3] Re: Mac work... In-Reply-To: <43838BAC.40706@icculus.org> References: <43809955.4030709@icculus.org> <20051122174518.02db29dc.tim@ngus.net> <43838BAC.40706@icculus.org> Message-ID: <1132803533.26560.425.camel@gaston> > Forcibly defining MACOS_X is fine, as far as I'm concerned, as long as > we don't make the usual mistake of using it when we mean "powerpc" or > "bigendian architecture" ... I'm sure we have a few of those in there, > which I'll weed out next week when I build this on an x86 Mac. Yes, beware, there are people around using it on linux/ppc :) That reminds me that I still need to fix the jit, I've fixed a few bugs in there a while ago but not all, there are still issues with argument passing. From yannick.francken at gmail.com Thu Nov 24 04:43:23 2005 From: yannick.francken at gmail.com (Yannick Francken) Date: Thu, 24 Nov 2005 10:43:23 +0100 Subject: Compiler error Message-ID: <9af75400511240143g43dbc80q57bc5d51c8936714@mail.gmail.com> Hi, I have added a new .c file to the renderer part in the visual studio (.NET 2003) project file. In that file there is only one dummy function "void test(void) {}". When I try to call this function from tr_backend.c, this results in a linking error. If I put this dummy function in a random other .c file, there are no problems. The error I get is: quake3 error LNK2019: unresolved external symbol _test referenced in function _RB_DrawSurfs Any ideas wath the reason can be? From yannick.francken at gmail.com Thu Nov 24 17:29:59 2005 From: yannick.francken at gmail.com (Yannick Francken) Date: Thu, 24 Nov 2005 23:29:59 +0100 Subject: [quake3] Re: Q3 Source documentation In-Reply-To: <20051122172643.0470349b.tim@ngus.net> References: <9af75400511210640o16f4bce4r9926bdfa7579c95@mail.gmail.com> <20051122172643.0470349b.tim@ngus.net> Message-ID: <9af75400511241429s21bb22d7ve4f284e0a93206db@mail.gmail.com> > http://www.linux.ucla.edu/~phaethon/q3mc/q3mc.html > http://www.soclose.de/q3doc/ > > These only really deal with the game code side of things (they predate > the GPL release), so perhaps they aren't that much use to you. This is indeed not exactly where I am looking for. > I'm not aware of anybody having implemented shadow volumes in Q3 yet. Isn't there a kind of shadow volume implemention (using the stencil buffer) available in Q3? I think only entities can be shadowed that way. > There are plenty of sites explaining how on the web though, albeit not > Q3 specific. The book (remember those? :)) "Game Programming Gems 1" has > a relatively good article on shadow map based shadows; it may be worth a > look. I think the main difficulty lies in integrating the shadowing code into the engine, not into programming the shadow volumes. This is also explained in the "Real-time rendering book". From rbromley256 at gmail.com Thu Nov 24 19:22:40 2005 From: rbromley256 at gmail.com (Richard Bromley) Date: Thu, 24 Nov 2005 16:22:40 -0800 Subject: OpenAL doesn't work in the latest build of Quake 3 Message-ID: <1f4734040511241622vd180973y4650427c90ac8120@mail.gmail.com> OpenAL doesn't work for me in the latest build of Quake 3 (1.33 alpha 364). SDL sound still works as normal. Here's the relevent console output: Code: ------ Initializing Sound ------ Failed to open OpenAL device. Initializing SDL audio driver... SDL audio driver is "dsp". SDL_AudioSpec: Format: AUDIO_S16LSB Freq: 22050 Samples: 512 Channels: 2 Starting SDL audio callback... SDL audio initialized. ----- Sound Info ----- 1 stereo 16384 samples 16 samplebits 1 submission_chunk 22050 speed 0x8b065c0 dma buffer No background file. ---------------------- Sound intialization successful. -------------------------------- I have a number of other games that work with OpenAL, both games that have their own openal.so and ones that don't. Just in case it matters, here's my .openalrc: Code: # Contains user settings for OpenAL # Goes in ~/.openalrc # Use ALSA (also valid: sdl, native) (define devices '(alsa)) # Four speaker surround with ALSA (define speaker-num 4) (define alsa-out-device "surround40:0,0") Is this solely my problem or a bug that can be fixed? Thanks for your reply. From yannick.francken at gmail.com Fri Nov 25 04:42:55 2005 From: yannick.francken at gmail.com (Yannick Francken) Date: Fri, 25 Nov 2005 10:42:55 +0100 Subject: Compiler error In-Reply-To: <9af75400511240143g43dbc80q57bc5d51c8936714@mail.gmail.com> References: <9af75400511240143g43dbc80q57bc5d51c8936714@mail.gmail.com> Message-ID: <9af75400511250142q5782222fi184aba3e1d492f72@mail.gmail.com> > I have added a new .c file to the renderer part in the visual studio > (.NET 2003) project file. In that file there is only one dummy > function "void test(void) {}". When I try to call this function from > tr_backend.c, this results in a linking error. If I put this dummy > function in a random other .c file, there are no problems. It seems that the added .c file was recognized as a C++ file for some reason, which causes the linking error. From fatty at gsklan.de Fri Nov 25 21:11:57 2005 From: fatty at gsklan.de (fatty) Date: Sat, 26 Nov 2005 03:11:57 +0100 Subject: [quake3] OpenAL doesn't work in the latest build of Quake 3 In-Reply-To: <1f4734040511241622vd180973y4650427c90ac8120@mail.gmail.com> References: <1f4734040511241622vd180973y4650427c90ac8120@mail.gmail.com> Message-ID: <4387C46D.2090804@gsklan.de> Richard Bromley wrote: > OpenAL doesn't work for me in the latest build of Quake 3 (1.33 alpha > 364). SDL sound still works as normal. Here's the relevent console > output: Current revision is 372. Do you get the same result with that version? > Code: > ------ Initializing Sound ------ > Failed to open OpenAL device. > Initializing SDL audio driver... > SDL audio driver is "dsp". > SDL_AudioSpec: > Format: AUDIO_S16LSB > Freq: 22050 > Samples: 512 > Channels: 2 > Starting SDL audio callback... > SDL audio initialized. > ----- Sound Info ----- > 1 stereo > 16384 samples > 16 samplebits > 1 submission_chunk > 22050 speed > 0x8b065c0 dma buffer > No background file. > ---------------------- > Sound intialization successful. > -------------------------------- > > > I have a number of other games that work with OpenAL, both games that > have their own openal.so and ones that don't. Just in case it matters, > here's my .openalrc: > > Code: > # Contains user settings for OpenAL > # Goes in ~/.openalrc > > # Use ALSA (also valid: sdl, native) > (define devices '(alsa)) > > # Four speaker surround with ALSA > (define speaker-num 4) > (define alsa-out-device "surround40:0,0") > > Is this solely my problem or a bug that can be fixed? Thanks for your reply. > What version of OpenAL do you have installed? Especially if you are using ALSA's dmix plugin with OpenAL you should try the latest CVS ( http://www.openal.org ). Don't forget that ALSA uses dmix by default if your soundcard isn't capable of hardware audio mixing. fatty From icculus at icculus.org Sat Nov 26 02:56:07 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Sat, 26 Nov 2005 02:56:07 -0500 Subject: quake3 MacOS hackery, continued... Message-ID: <43881517.2000609@icculus.org> Okay, Subversion Revision #373 adds my Mac OS X work. I fully expect things to be broken in small ways on other platforms...just commit fixes if something is obviously hosed. I corrected the x86 assembly thing that was mentioned previously. Notes and current bugs: - Mac-specific codepaths are gone...it uses SDL and OpenAL in the Unix codepaths now. Less code to maintain. - Compiled VM seems to work, but keeps writing to console: Com_sprintf: overflow of 247 in 144 ...so I guess the calling convention isn't totally right? I think this is a PowerPC Linux issue, too, at the moment. I tried vm_ppc_new. - Interpreted VM works, but eats like 30% of the CPU! - dlopen() is (mostly) gone...uses SDL_LoadObject() and SDL_GL_LoadLibrary() now. The dedicated server still uses dlopen, to avoid the dependency. - Rendering issues: mirror in Q3DM0 doesn't render, water in Q3DM2 doesn't render. Looks like health orbs lose their orbs (just crosses rendering) with some frequency. All of these issues appear to affect PowerPC Linux, too, and did before I started my Mac work...some byteswap issue? - Haven't tried multiplayer at all. - Need OpenAL library...doesn't ship as a system framework until 10.4. SDL Audio and OpenAL output both appear to be fine on Mac OS now. - OpenAL dynamic loading doesn't compile. Should make default and let people choose a specific openal.dylib via a cvar. - Altivec code is split out now, since you can't call a function with Altivec code on a G3, even if the code is protected with an "if (altivec)" check...GCC adds prologue code to these functions to protect the vector registers, so G3 systems crash when you call the function at all. This is fixed now. Also, there is now a "com_altivec" cvar which uses can use to toggle Altivec support on/off at runtime (if com_altivec is 1 at startup, we'll detect support and set it appropriately...if it's 0, we won't detect it and just leave support off). - Altivec code is broken in _debug_ mode...see nastiness in LerpMeshVertexes() in renderer/tr_surface.c ... Works fine in release mode. Don't know why. Someone should fix that, I won't be doing so. - Altivec detection relies on Carbon framework...not sure how to do this on PowerPC Linux. For now, we assume it's available (and G3 Linux users will have to disable it from the command line). - I have (re)added the SDL and OpenAL headers to Subversion. I don't care what the Linux users say, this is non-negotiable. You can choose to use the system versions if you want for Linux, but on Windows and MacOS, these need to be included so we don't have to hunt down the external dependency. As such, there will also be Mac and Windows precompiled versions of the libraries added, too. Again, Linux users can use a system version if they want. By default, they still do this; I haven't changed the Makefile to do otherwise. Seriously, leave these headers in there. - Added support for ccache...please build with USE_CCACHE=0 if you don't want it. - The XCode project files went away with the rest of the Mac-specific bits...this could be reconstructed, though, but I don't personally use it, so I don't personally care. :) - Haven't tried x86 Mac (will do so this upcoming week if there's time). I think most of the remaining issues affect PowerPC Linux, too, so I'm leaving this to the pros from here out unless there's a specific issue for me to tackle. --ryan. From a2daj at mac.com Sat Nov 26 05:58:45 2005 From: a2daj at mac.com (Avelino Santa Ana Jr.) Date: Sat, 26 Nov 2005 02:58:45 -0800 Subject: [quake3] quake3 MacOS hackery, continued... In-Reply-To: <43881517.2000609@icculus.org> References: <43881517.2000609@icculus.org> Message-ID: <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> I didn't see it listed, but it looks like the large increase when using r_smp was enabled disappeared along with the Mac specific code. "Trying SMP acceleration... ^3ERROR: SMP support was disabled at compile time ...failed. " On Nov 25, 2005, at 11:56 PM, Ryan C. Gordon wrote: > > Okay, > > Subversion Revision #373 adds my Mac OS X work. I fully expect > things to be broken in small ways on other platforms...just commit > fixes if something is obviously hosed. I corrected the x86 assembly > thing that was mentioned previously. > > > Notes and current bugs: > > - Mac-specific codepaths are gone...it uses SDL and OpenAL in the Unix > codepaths now. Less code to maintain. > > - Compiled VM seems to work, but keeps writing to console: > > Com_sprintf: overflow of 247 in 144 > > ...so I guess the calling convention isn't totally right? I > think this is a PowerPC Linux issue, too, at the moment. I tried > vm_ppc_new. > > - Interpreted VM works, but eats like 30% of the CPU! > > - dlopen() is (mostly) gone...uses SDL_LoadObject() and > SDL_GL_LoadLibrary() now. The dedicated server still uses dlopen, to > avoid the dependency. > > - Rendering issues: mirror in Q3DM0 doesn't render, water in Q3DM2 > doesn't render. Looks like health orbs lose their orbs (just crosses > rendering) with some frequency. All of these issues appear to affect > PowerPC Linux, too, and did before I started my Mac work...some > byteswap > issue? > > - Haven't tried multiplayer at all. > > - Need OpenAL library...doesn't ship as a system framework until 10.4. > SDL Audio and OpenAL output both appear to be fine on Mac OS now. > > - OpenAL dynamic loading doesn't compile. Should make default and let > people choose a specific openal.dylib via a cvar. > > - Altivec code is split out now, since you can't call a function with > Altivec code on a G3, even if the code is protected with an "if > (altivec)" check...GCC adds prologue code to these functions to > protect > the vector registers, so G3 systems crash when you call the > function at > all. This is fixed now. Also, there is now a "com_altivec" cvar which > uses can use to toggle Altivec support on/off at runtime (if > com_altivec > is 1 at startup, we'll detect support and set it appropriately...if > it's > 0, we won't detect it and just leave support off). > > - Altivec code is broken in _debug_ mode...see nastiness in > LerpMeshVertexes() in renderer/tr_surface.c ... Works fine in > release mode. Don't know why. Someone should fix that, I won't be > doing so. > > - Altivec detection relies on Carbon framework...not sure how to do > this > on PowerPC Linux. For now, we assume it's available (and G3 Linux > users > will have to disable it from the command line). > > - I have (re)added the SDL and OpenAL headers to Subversion. I don't > care what the Linux users say, this is non-negotiable. You can > choose to > use the system versions if you want for Linux, but on Windows and > MacOS, > these need to be included so we don't have to hunt down the external > dependency. As such, there will also be Mac and Windows precompiled > versions of the libraries added, too. Again, Linux users can use a > system version if they want. By default, they still do this; I > haven't changed the Makefile to do otherwise. Seriously, leave > these headers in there. > > - Added support for ccache...please build with USE_CCACHE=0 if you > don't > want it. > > - The XCode project files went away with the rest of the Mac-specific > bits...this could be reconstructed, though, but I don't personally use > it, so I don't personally care. :) > > - Haven't tried x86 Mac (will do so this upcoming week if there's > time). > > I think most of the remaining issues affect PowerPC Linux, too, so > I'm leaving this to the pros from here out unless there's a > specific issue for me to tackle. > > --ryan. > > > From tim at ngus.net Sat Nov 26 13:13:14 2005 From: tim at ngus.net (Tim Angus) Date: Sat, 26 Nov 2005 18:13:14 +0000 Subject: quake3 MacOS hackery, continued... In-Reply-To: <43881517.2000609@icculus.org> References: <43881517.2000609@icculus.org> Message-ID: <20051126181314.5b67fed9.tim@ngus.net> On Sat, 26 Nov 2005 02:56:07 -0500 Ryan wrote: > - I have (re)added the SDL and OpenAL headers to Subversion. I don't > care what the Linux users say, this is non-negotiable. You can choose > to use the system versions if you want for Linux, but on Windows and > MacOS, these need to be included so we don't have to hunt down the > external dependency. I don't agree with this. I don't understand the thinking behind it. I seriously doubt there are proportionately that many Windows or OS X users who actually build ioq3, and those that do surely have enough understanding to locate the required headers. Having system headers in the q3 source adds a maintenance headache we don't need. As it stands the Windows (MinGW) build is now broken as the AL headers in SVN are too old. This is probably also the reason for: > - OpenAL dynamic loading doesn't compile. Should make default and let > people choose a specific openal.dylib via a cvar. > As such, there will also be Mac and Windows > precompiled versions of the libraries added, too. Libraries are something to be considered when compiling binaries for distribution. It makes no sense at all to have them in /source/ control. From ludwig.nussel at suse.de Sat Nov 26 21:03:58 2005 From: ludwig.nussel at suse.de (Ludwig Nussel) Date: Sun, 27 Nov 2005 03:03:58 +0100 Subject: [quake3] Re: quake3 MacOS hackery, continued... In-Reply-To: <20051126181314.5b67fed9.tim@ngus.net> References: <43881517.2000609@icculus.org> <20051126181314.5b67fed9.tim@ngus.net> Message-ID: <200511270303.58911.ludwig.nussel@suse.de> Tim Angus wrote: > On Sat, 26 Nov 2005 02:56:07 -0500 Ryan wrote: > > - I have (re)added the SDL and OpenAL headers to Subversion. I don't > > care what the Linux users say, this is non-negotiable. You can choose > > to use the system versions if you want for Linux, but on Windows and > > MacOS, these need to be included so we don't have to hunt down the > > external dependency. > > I don't agree with this. I don't understand the thinking behind it. I > seriously doubt there are proportionately that many Windows or OS X > users who actually build ioq3, and those that do surely have enough > understanding to locate the required headers. Having system headers in > the q3 source adds a maintenance headache we don't need. > > As it stands the Windows (MinGW) build is now broken as the AL headers > in SVN are too old. This is probably also the reason for: > > > - OpenAL dynamic loading doesn't compile. Should make default and let > > people choose a specific openal.dylib via a cvar. > > > As such, there will also be Mac and Windows > > precompiled versions of the libraries added, too. > > Libraries are something to be considered when compiling binaries for > distribution. It makes no sense at all to have them in /source/ control. I can only add add an ACK to this mail. Those files are part of your private build environment, they don't belong to Q3. Put them in a separate tar.gz on the web site if you want to offer it to the world. cu Ludwig -- (o_ Ludwig Nussel //\ SUSE LINUX Products GmbH, Development V_/_ http://www.suse.de/ From icculus at icculus.org Sun Nov 27 20:25:16 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Sun, 27 Nov 2005 20:25:16 -0500 Subject: [quake3] Re: quake3 MacOS hackery, continued... In-Reply-To: <20051126181314.5b67fed9.tim@ngus.net> References: <43881517.2000609@icculus.org> <20051126181314.5b67fed9.tim@ngus.net> Message-ID: <438A5C7C.5070407@icculus.org> > Libraries are something to be considered when compiling binaries for > distribution. It makes no sense at all to have them in /source/ > control. It's not source control, it's a revision control system. We have the web site in here, too, for example. Semantics aside, you need these libraries when building from source as much as you need them in an installer or at runtime. As I said, it makes sense for Linux distros to manage these dependencies (or Linux users to hunt them down, although frankly I don't like the assumption that they should do more work, but binary incompatibility demands it)...as such I haven't added them to svn for Linux. But on the Mac and Windows, you shouldn't be telling people "oh, go find this, possibly build it, also do this workaround to SDL's build system to make sure it builds correctly on the Mac, and copy it to the following place, which may be just about anywhere on Windows" OpenAL and SDL are not "system headers" ... I'm not advocating adding stdio.h to subversion here. These are third-party external dependencies that we rely upon. You can call it a benefit of Linux development that you can avoid this, but on other platforms, this makes a lot of sense. Being able to pull down the latest revision and just build it, regardless of the system, is a feature, not a bug. That being said, the OpenAL headers I included were from AL1.0...if this broke something, let's update them...but it should also demonstrate why relying on the end-user to supply these is dangerous, since there are probably users out there with the same headers, and they're either going to write a bug report which we would have to flag INVALID or they'll go "this project sucks" and not use it. Or both. --ryan. From icculus at icculus.org Mon Nov 28 04:31:41 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Mon, 28 Nov 2005 04:31:41 -0500 Subject: [quake3] quake3 MacOS hackery, continued... In-Reply-To: <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> Message-ID: <438ACE7D.7090303@icculus.org> > "Trying SMP acceleration... > ^3ERROR: SMP support was disabled at compile time > ...failed. That's my bug...r_smp was, at best, a performance loss on Linux, at least when we were experimenting at Loki in 2000, so I assumed it sucked on every platform. I'll see what it takes to reenable this on the Mac. --ryan. From icculus at icculus.org Mon Nov 28 06:13:15 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Mon, 28 Nov 2005 06:13:15 -0500 Subject: SDL vs SMP... Message-ID: <438AE64B.8030806@icculus.org> Svn revision #385 adds the SMP code to sdl_glimp.c...however it does not work yet. It shouldn't affect non-SMP builds, or even the SMP binary if r_smp is set to 0. I put this in the Makefile for the Mac. To add it to Linux, etc, you'll want to add "sdl_glimp_smp.o" to your platform's Q3POBJ_SMP variable in the Makefile, and remove "linux_glimp_smp.o" if it exists. There isn't a matching sdl_glimp_smp.c file, but the Makefile uses this to know to rebuild sdl_glimp.c with -DSMP on the command line. The problem is that most platforms do not have thread-safe OpenGL libraries, including MacOS and (I think!) Linux. The existing Linux code was just setting the current GL context to NULL in one thread and resetting it to the previous value in the other thread as it switches between them, which apparently helps the GL work out threading issues. SDL, however, wants you to make all video and OpenGL calls from the main thread, mostly for these sorts of reasons, and does not expose a MakeContextCurrent() entry point abstraction (partially because it only allows one GL context at this point), so this is naturally a problem. On the Mac, this blows up inside the first glViewport() call, and I bet Linux won't fare much better. So, if anyone understands the SMP code at all, there are two questions: 1) How hard would it be for the renderer thread to be the intial thread, and instead spin a thread for the game logic? In other words, we need to swap the two threads and otherwise function as before if possible. 2) Does the non-renderer thread ever make any GL calls at all? Even state queries? Ideally, we make the renderer thread the main thread and never call into OpenGL from the other thread. This would fix the problem and also speed up the existing code, since there wouldn't be GL context-switch overhead every frame. Is this even possible? Or sane? --ryan. From zakk at timedoctor.org Mon Nov 28 13:51:05 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 28 Nov 2005 10:51:05 -0800 Subject: [quake3] quake3 MacOS hackery, continued... In-Reply-To: <438ACE7D.7090303@icculus.org> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> Message-ID: <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> On Nov 28, 2005, at 1:31 AM, Ryan C. Gordon wrote: > > That's my bug...r_smp was, at best, a performance loss on Linux, at > least when we were experimenting at Loki in 2000, so I assumed it > sucked > on every platform. I'll see what it takes to reenable this on the Mac. > > --ryan. > SMP support was removed from the makefile some time ago intentionally. I'd prefer it gone on every platform. From davebytes at comcast.net Mon Nov 28 14:50:21 2005 From: davebytes at comcast.net (David Chait) Date: Mon, 28 Nov 2005 14:50:21 -0500 Subject: [quake3] quake3 MacOS hackery, continued... References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> Message-ID: <005001c5f454$f9f09150$6e01a8c0@sixfour> In this case, does "SMP" == "running rendering in a separate thread from main processing thread"? And if so, why wouldn't you want that as an option? Especially in this age of hyperthreading on Intel, and dual-core/-cpu boxes not far off from the mainstream, and the mac being a huge multi-cpu platform... ----- Original Message ----- From: "Zachary J. Slater" To: Sent: Monday, November 28, 2005 1:51 PM Subject: Re: [quake3] quake3 MacOS hackery, continued... | | On Nov 28, 2005, at 1:31 AM, Ryan C. Gordon wrote: | | > | > That's my bug...r_smp was, at best, a performance loss on Linux, at | > least when we were experimenting at Loki in 2000, so I assumed it | > sucked | > on every platform. I'll see what it takes to reenable this on the Mac. | > | > --ryan. | > | | SMP support was removed from the makefile some time ago intentionally. | I'd prefer it gone on every platform. | | From a2daj at mac.com Mon Nov 28 16:48:43 2005 From: a2daj at mac.com (Avelino Santa Ana Jr.) Date: Mon, 28 Nov 2005 13:48:43 -0800 Subject: [quake3] quake3 MacOS hackery, continued... In-Reply-To: <005001c5f454$f9f09150$6e01a8c0@sixfour> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> Message-ID: <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> I just ran a quick benchmark with the 1.32 id release on my Dual 2.5 with an X800. With everything maximized and at 1024x768 I received these scores. r_smp 0 = 228 r_smp 1 = 385 That seems like a significant boost to me. I'm pretty sure a number of Mac users would be very disappointed in Icculus Q3 if it didn't support SMP. On Nov 28, 2005, at 11:50 AM, David Chait wrote: > In this case, does "SMP" == "running rendering in a separate thread > from > main processing thread"? And if so, why wouldn't you want that as an > option? Especially in this age of hyperthreading on Intel, and > dual-core/-cpu boxes not far off from the mainstream, and the mac > being a > huge multi-cpu platform... > > ----- Original Message ----- > From: "Zachary J. Slater" > To: > Sent: Monday, November 28, 2005 1:51 PM > Subject: Re: [quake3] quake3 MacOS hackery, continued... > > > | > | On Nov 28, 2005, at 1:31 AM, Ryan C. Gordon wrote: > | > | > > | > That's my bug...r_smp was, at best, a performance loss on > Linux, at > | > least when we were experimenting at Loki in 2000, so I assumed it > | > sucked > | > on every platform. I'll see what it takes to reenable this on > the Mac. > | > > | > --ryan. > | > > | > | SMP support was removed from the makefile some time ago > intentionally. > | I'd prefer it gone on every platform. > | > | > From jesusclownfox at gmail.com Mon Nov 28 17:34:24 2005 From: jesusclownfox at gmail.com (Luciano Medori) Date: Mon, 28 Nov 2005 17:34:24 -0500 Subject: Binaries for Linux x86? Message-ID: <43c75a590511281434u5f5dc3f9mc458a0d41326416a@mail.gmail.com> Hey there. I'm a recent ML lurker who's very much looking forward to the progress of this project. I just wanted to ask, when it comes time for the first public release, why not at least provide one Linux binary (more specifically, x86)? I get the feeling that a lot of users will be turned off either because they run into compile-time issues, or they're just afraid of using gcc. Also, the ioq3 page states that there's an OS 9 section under q_platform.h. Should there really be a question whether or not to support any pre-10 Mac OS? -------------- next part -------------- An HTML attachment was scrubbed... URL: From jesusclownfox at gmail.com Mon Nov 28 17:37:00 2005 From: jesusclownfox at gmail.com (Luciano Medori) Date: Mon, 28 Nov 2005 17:37:00 -0500 Subject: Binaries for Linux x86? Message-ID: <43c75a590511281437i6f29ff86od9b046b8ca511154@mail.gmail.com> Oh, I'm very sorry about the HTML. I only realized after that Gmail enables it by default. :\ From tim at ngus.net Mon Nov 28 19:33:11 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 29 Nov 2005 00:33:11 +0000 Subject: quake3 MacOS hackery, continued... In-Reply-To: <438A5C7C.5070407@icculus.org> References: <43881517.2000609@icculus.org> <20051126181314.5b67fed9.tim@ngus.net> <438A5C7C.5070407@icculus.org> Message-ID: <20051129003311.43a5e268.tim@ngus.net> On Sun, 27 Nov 2005 20:25:16 -0500 Ryan wrote: > It's not source control, it's a revision control system. We have the > web site in here, too, for example. It would be nice if that were separate to be honest. It really screws with the revision numbers. > Semantics aside, you need these libraries when building from source as > much as you need them in an installer or at runtime. As I said, it > makes sense for Linux distros to manage these dependencies (or Linux > users to hunt them down, although frankly I don't like the assumption > that they should do more work, but binary incompatibility demands > it)...as such I haven't added them to svn for Linux. But on the Mac > and Windows, you shouldn't be telling people "oh, go find this, > possibly build it, also do this workaround to SDL's build system to > make sure it builds correctly on the Mac, and copy it to the following > place, which may be just about anywhere on Windows" This is really reiterating what I said before... I'm not sure there is really any point in making the Windows and Mac builds "easy". I fully expect (at least my experience indicates) that Windows and Mac users couldn't give a damn about how easy the project is to compile. Your average user is going to want a nice installer package that just works. Windows and Mac users that are intrepid enough to actually want to compile the thing are by virtue going to have at least a minor understanding of building software. This is not the case on Linux (yet), where knowing how to compile something from source is regrettably still relatively important. What I'm getting at is that ease of compilation doesn't seem to me to be particularly important on these platforms. Case in point: on the IRC channel there are regularly Linux people having trouble building ioq3. I am aware of a grand total of one user building the MinGW build, and he has yet to ask any questions regarding the build process (before the recent breakage anyway). > OpenAL and SDL are not "system headers" ... I'm not advocating adding > stdio.h to subversion here. These are third-party external > dependencies that we rely upon. Should we include OpenGL and DirectX headers too? > That being said, the OpenAL headers I included were from AL1.0...if > this broke something, let's update them... This is really my major gripe. By having the headers in SVN, it is now up to us to ensure the headers are up to date and appropriate for the platform we're using. I don't think this is something we should really need to do. The current AL headers definitely need to be upgraded. > but it should also > demonstrate why relying on the end-user to supply these is dangerous, When a release is actually made, we don't NEED to rely on the end-user since we'll give them nice binary packages that "just work". > since there are probably users out there with the same headers, and > they're either going to write a bug report which we would have to flag > INVALID or they'll go "this project sucks" and not use it. Or both. Pleasingly, I'm aware of very few bugs involving build problems. Most people seem to attack the mailing list or the IRC channel, and they usually get a prompt response. If we do stick with having the headers in SVN, at the very least it needs to be a build option to use system headers instead. From tim at ngus.net Mon Nov 28 19:34:36 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 29 Nov 2005 00:34:36 +0000 Subject: r383 - trunk/code/tools/lcc In-Reply-To: <20051128095055.7876.qmail@icculus.org> References: <20051128095055.7876.qmail@icculus.org> Message-ID: <20051129003436.6c5021fc.tim@ngus.net> On 28 Nov 2005 04:50:55 -0500 DONOTREPLY at icculus.org wrote: > Author: icculus > Date: 2005-11-28 04:50:55 -0500 (Mon, 28 Nov 2005) > New Revision: 383 > > Modified: > trunk/code/tools/lcc/Makefile > Log: > > Must explicitly define MACOS_X here or lcc won't build on the Mac. Oops. Sorry about that. I grepped for MAXOS_X in the lcc source as I don't remember there being a reason for it being there, but it's needed for q_platform.h. This is doubly embarassing as it was me that introduced the dependance. From tim at ngus.net Mon Nov 28 19:38:54 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 29 Nov 2005 00:38:54 +0000 Subject: quake3 MacOS hackery, continued... In-Reply-To: <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> Message-ID: <20051129003854.4b9f4960.tim@ngus.net> On Mon, 28 Nov 2005 13:48:43 -0800 Avelino wrote: > I just ran a quick benchmark with the 1.32 id release on my Dual 2.5 > with an X800. With everything maximized and at 1024x768 I received > these scores. > > r_smp 0 = 228 > > r_smp 1 = 385 > > That seems like a significant boost to me. I'm pretty sure a number > of Mac users would be very disappointed in Icculus Q3 if it didn't > support SMP. It IS a significant boost, but relatively meaningless one. It seems like an enourmous waste of effort to implement for what is essentially zero real world gain. I'm actually pretty surprised there is such a large difference. On x86 at least, Q3 is a long way off CPU bound and as such parallelising it doesn't really result in a much of a gain. I guess this goes to show that the OS X build is somewhat less efficient CPU wise. It will be interesting to see how OS X/x86 performs. From tim at ngus.net Mon Nov 28 19:42:55 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 29 Nov 2005 00:42:55 +0000 Subject: quake3 MacOS hackery, continued... In-Reply-To: <005001c5f454$f9f09150$6e01a8c0@sixfour> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> Message-ID: <20051129004255.528c45e3.tim@ngus.net> On Mon, 28 Nov 2005 14:50:21 -0500 David wrote: > In this case, does "SMP" == "running rendering in a separate thread > from main processing thread"? And if so, why wouldn't you want that > as an option? Especially in this age of hyperthreading on Intel, and > dual-core/-cpu boxes not far off from the mainstream, and the mac > being a huge multi-cpu platform... Multi-threaded code is an order of magnitude harder to write and debug than single-threaded code. Avoiding it altogether is preferable if possible, and since in the general case for us it does not represent a significant performace gain, it makes little sense to maintain it. Disclaimer: I've run SMP boxes as my main development box for at least the past 5 years and I see very little value to the SMP support in Q3. As an aside, if you use OpenAL, you automatically get some parallelism since it (necessarilly) runs in it's own thread. For software OpenAL, this is potentially quite significant. From tim at ngus.net Mon Nov 28 19:45:30 2005 From: tim at ngus.net (Tim Angus) Date: Tue, 29 Nov 2005 00:45:30 +0000 Subject: r383 - trunk/code/tools/lcc In-Reply-To: <20051129003436.6c5021fc.tim@ngus.net> References: <20051128095055.7876.qmail@icculus.org> <20051129003436.6c5021fc.tim@ngus.net> Message-ID: <20051129004530.16114031.tim@ngus.net> On Tue, 29 Nov 2005 00:34:36 +0000 Tim wrote: > Oops. Sorry about that. I grepped for MAXOS_X in the lcc source as I > don't remember there being a reason for it being there, but it's > needed for q_platform.h. This is doubly embarassing as it was me that > introduced the dependance. I should add, this seems like a good case for not relying on the build system to identify the platform. https://bugzilla.icculus.org/show_bug.cgi?id=2510 seems to suggest __APPLE_CC__ is a candidate. From icculus at icculus.org Mon Nov 28 22:20:40 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Mon, 28 Nov 2005 22:20:40 -0500 Subject: [quake3] Re: quake3 MacOS hackery, continued... In-Reply-To: <20051129003854.4b9f4960.tim@ngus.net> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> <20051129003854.4b9f4960.tim@ngus.net> Message-ID: <438BC908.40105@icculus.org> > I'm actually pretty surprised there is such a large difference. On x86 > at least, Q3 is a long way off CPU bound and as such parallelising it > doesn't really result in a much of a gain. I guess this goes to show > that the OS X build is somewhat less efficient CPU wise. It will be > interesting to see how OS X/x86 performs. I'd be interested to see the performance difference once we get the compiled VM details worked out...the interpreted VM is a total CPU killer on Mac OS. That being said, the profiles show quake 3 eating ~97% of the CPU, which means it's not waiting on the GL at all. --ryan. From zakk at timedoctor.org Tue Nov 29 00:32:39 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 28 Nov 2005 21:32:39 -0800 Subject: [quake3] quake3 MacOS hackery, continued... In-Reply-To: <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> Message-ID: A) This isn't "Icculus Q3" this is either icculus.org/quake3 or ioQ3 as a shorter form B) SMP support was dropped months ago. This was never supposed to get added back, and I'm not sure if it has been (fully or at all). It is a nightmare to maintain and the gains are really minimal Just to be sure, os x isn't being dropped and neither is windows. They will both be supported by release time. SMP binaries of quake 3 were slower than their uniprocessor counterparts when id's last binary was released. I don't consider one benchmark to be demonstrative of SMP's gains on modern computers. Feel free to discuss smp support on this list, however support of it at this time would hinder release considerably. From zakk at timedoctor.org Tue Nov 29 00:37:50 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 28 Nov 2005 21:37:50 -0800 Subject: OS X Fun Message-ID: Just so people know we've currently still got a few problems on OS X: 1) Mouse cursor is always displayed 2) Sound is played too fast with the openal included in SVN 3) Those possibly byteswapping issues with the health orb and the water 4) Player model in the players choice option menu is displayed very strangely (especially on the larger models), like the fov is wrong or something. Ryan suggested some of these problems might be fixed by switching to gcc 3 from 4, they weren't as far as I can tell. More to come. From zakk at timedoctor.org Tue Nov 29 01:28:03 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 28 Nov 2005 22:28:03 -0800 Subject: [quake3] OS X Fun In-Reply-To: References: Message-ID: Actually I'm wrong, it isn't building properly on OS X. There isn't an SDL or openal lib building. The binary I was using was old. The player model bug exists on x86 Linux too, I just checked. I'd like someone else to confirm it too. None of the other issues seem to be present on Linux. From zakk at timedoctor.org Tue Nov 29 02:27:47 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Mon, 28 Nov 2005 23:27:47 -0800 Subject: [quake3] OS X Fun In-Reply-To: References: Message-ID: After replacing the quake 3 binary in the bundle and adding a little bit of SDL, I got it running. However, at 1600x1200 it doesn't display properly at all (Green main menu, random lines, etc). At any lower resolution it seems to run fine. However that /might/ be related to the fact that it is running on a powerbook outputting to DVI to a separate monitor, which worked in the older (ioQ3) binary. So I would suppose SDL is doing something wrong. The model bug is there in the modern build on mac/linux it seems. The water/health bubble thing is still there. The cursor being displayed on the screen is not OpenAL is rendering the sound fine. From vincent at cojot.name Tue Nov 29 04:30:48 2005 From: vincent at cojot.name (vincent at cojot.name) Date: Tue, 29 Nov 2005 10:30:48 +0100 (CET) Subject: [quake3] Re: quake3 MacOS hackery, continued... In-Reply-To: <20051129003854.4b9f4960.tim@ngus.net> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> <20051129003854.4b9f4960.tim@ngus.net> Message-ID: On Tue, 29 Nov 2005, Tim Angus wrote: > On Mon, 28 Nov 2005 13:48:43 -0800 Avelino wrote: >> I just ran a quick benchmark with the 1.32 id release on my Dual 2.5 >> with an X800. With everything maximized and at 1024x768 I received >> these scores. >> >> r_smp 0 = 228 >> >> r_smp 1 = 385 >> >> That seems like a significant boost to me. I'm pretty sure a number >> of Mac users would be very disappointed in Icculus Q3 if it didn't >> support SMP. > > It IS a significant boost, but relatively meaningless one. It seems like > an enourmous waste of effort to implement for what is essentially zero > real world gain. Just my 2c but since q3 runs slowler on Solaris/Sparc, perhaps smp acceleration would be relevant/not meaningless there.. Does smp work in the current ioq3 svn? If so, I'll try it some time soon.. -- ,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~'`^`'~*-, Vincent S. Cojot, Computer Engineering. STEP project. _.,-*~'`^`'~*-,._.,-*~ Ecole Polytechnique de Montreal, Comite Micro-Informatique. _.,-*~'`^`'~*-,. Linux Xview/OpenLook resources page _.,-*~'`^`'~*-,._.,-*~'`^`'~*-,._.,-*~' http://step.polymtl.ca/~coyote _.,-*~'`^`'~*-,._ coyote at NOSPAM4cojot.name They cannot scare me with their empty spaces Between stars - on stars where no human race is I have it in me so much nearer home To scare myself with my own desert places. - Robert Frost From benh at kernel.crashing.org Tue Nov 29 16:21:50 2005 From: benh at kernel.crashing.org (Benjamin Herrenschmidt) Date: Wed, 30 Nov 2005 08:21:50 +1100 Subject: [quake3] quake3 MacOS hackery, continued... In-Reply-To: <43881517.2000609@icculus.org> References: <43881517.2000609@icculus.org> Message-ID: <1133299311.16726.14.camel@gaston> > - Compiled VM seems to work, but keeps writing to console: > > Com_sprintf: overflow of 247 in 144 > > ...so I guess the calling convention isn't totally right? I think > this is a PowerPC Linux issue, too, at the moment. I tried vm_ppc_new. Hrm... I got it approx. to the same point in linux after fixing the trampoline assembly... We may need 2 different versions for OS X vs. linux/ppc here though. I haven't looked at the Com_sprintf problem yet. Ben. From tim at ngus.net Tue Nov 29 21:03:44 2005 From: tim at ngus.net (Tim Angus) Date: Wed, 30 Nov 2005 02:03:44 +0000 Subject: OS X Fun In-Reply-To: References: Message-ID: <20051130020344.159fe4bf.tim@ngus.net> On Mon, 28 Nov 2005 22:28:03 -0800 Zachary wrote: > The player model bug exists on x86 Linux too, I just checked. I'd > like someone else to confirm it too. I see no problem on x86 Linux. From icculus at icculus.org Tue Nov 29 22:23:31 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Tue, 29 Nov 2005 22:23:31 -0500 Subject: [quake3] Re: OS X Fun In-Reply-To: <20051130020344.159fe4bf.tim@ngus.net> References: <20051130020344.159fe4bf.tim@ngus.net> Message-ID: <438D1B33.3030009@icculus.org> > I see no problem on x86 Linux. I'm not seeing this on amd64 Linux, either (haven't updated the x86 box yet). Ubuntu, gcc 4.0.2. I did have issues with gcc4 on the Mac, but not sure exactly what now. --ryan. From RoyLeongYW at ocbc.com.my Tue Nov 29 22:33:52 2005 From: RoyLeongYW at ocbc.com.my (Leong Yew Wah) Date: Wed, 30 Nov 2005 11:33:52 +0800 Subject: [quake3] Newbie on Linux Message-ID: <1A8A208A19C9F8469CB30DEDB8DE5DC7A23D71@PWEXCLUMY51.OACENTRE.OCBC.COM> Greetings mr Ryan, Small matter, thought I'd write in to get help. I tried installing the ol basic Q3A demo as well as Q3A full game (windows via the Linux PR 1.32) onto Fedora Core 4, and got a funny msg - ...somthg somthg glibc somthg somthg ..... I did not have any problems loading them (Q3A demo & fullgame) onto RedHat 9.0 What must I do to get it to install onto Fedora Core 4 ? (no-one I know of uses linux for games...) Rgds, Roy. (subscribing the list to learn) -----Original Message----- From: Ryan C. Gordon [mailto:icculus at icculus.org] Sent: Wednesday, November 30, 2005 11:24 AM To: quake3 at icculus.org Subject: Re: [quake3] Re: OS X Fun > I see no problem on x86 Linux. I'm not seeing this on amd64 Linux, either (haven't updated the x86 box yet). Ubuntu, gcc 4.0.2. I did have issues with gcc4 on the Mac, but not sure exactly what now. --ryan. Information in this message is confidential. It is intended solely for the person or the entity to whom it is addressed. If you are not the intended recipient, you are not to disseminate,distribute or copy this communication. Please notify the sender and delete the message and any other record of it from your system immediately. From icculus at icculus.org Tue Nov 29 22:58:02 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Tue, 29 Nov 2005 22:58:02 -0500 Subject: [quake3] Newbie on Linux In-Reply-To: <1A8A208A19C9F8469CB30DEDB8DE5DC7A23D71@PWEXCLUMY51.OACENTRE.OCBC.COM> References: <1A8A208A19C9F8469CB30DEDB8DE5DC7A23D71@PWEXCLUMY51.OACENTRE.OCBC.COM> Message-ID: <438D234A.7020409@icculus.org> > What must I do to get it to install onto Fedora Core 4 ? (no-one I know > of uses linux for games...) I wonder if the old binaries have glibc conflicts now...you can build it yourself from source, which would solve the problem. --ryan. From a2daj at mac.com Tue Nov 29 23:09:02 2005 From: a2daj at mac.com (Avelino Santa Ana Jr.) Date: Tue, 29 Nov 2005 20:09:02 -0800 Subject: [quake3] Re: quake3 MacOS hackery, continued... In-Reply-To: <20051129003854.4b9f4960.tim@ngus.net> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> <20051129003854.4b9f4960.tim@ngus.net> Message-ID: <94E003E7-B683-4726-A8F0-5C6E848243BD@mac.com> On Nov 28, 2005, at 4:38 PM, Tim Angus wrote: > On Mon, 28 Nov 2005 13:48:43 -0800 Avelino wrote: >> I just ran a quick benchmark with the 1.32 id release on my Dual 2.5 >> with an X800. With everything maximized and at 1024x768 I received >> these scores. >> >> r_smp 0 = 228 >> >> r_smp 1 = 385 >> >> That seems like a significant boost to me. I'm pretty sure a number >> of Mac users would be very disappointed in Icculus Q3 if it didn't >> support SMP. > > It IS a significant boost, but relatively meaningless one. It seems > like > an enourmous waste of effort to implement for what is essentially zero > real world gain. I disagree. The SMP support in the Mac client has been around since OmniGroup ported it to OS X. It has always seen a boost as long as the video card has not been the limiting factor. For example: http://www.xlr8yourmac.com/Graphics/geforce3/GeForce3vs2MXvsRadeon.html Going from 57.2 fps to 95.0 fps average was a very noticeable difference back then. It has also been the only way the Mac client has been able to keep up with the PC benchmarks (I know, I know... benchmarks...). It's your guys project, so if you guys want to dump SMP support, that's your decision. But Mac users have grown accustomed to the support being their so if it's not their anymore, well, you know Mac users... From zakk at timedoctor.org Tue Nov 29 23:28:30 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Tue, 29 Nov 2005 20:28:30 -0800 Subject: [quake3] Re: OS X Fun In-Reply-To: <438D1B33.3030009@icculus.org> References: <20051130020344.159fe4bf.tim@ngus.net> <438D1B33.3030009@icculus.org> Message-ID: On Nov 29, 2005, at 7:23 PM, Ryan C. Gordon wrote: > >> I see no problem on x86 Linux. > > I'm not seeing this on amd64 Linux, either (haven't updated the x86 > box > yet). Ubuntu, gcc 4.0.2. > > I did have issues with gcc4 on the Mac, but not sure exactly what now. > > --ryan. > I am using a 64bit processor, but with a 32bit gentoo install... From zakk at timedoctor.org Tue Nov 29 23:58:54 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Tue, 29 Nov 2005 20:58:54 -0800 Subject: [quake3] Re: quake3 MacOS hackery, continued... In-Reply-To: <94E003E7-B683-4726-A8F0-5C6E848243BD@mac.com> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> <20051129003854.4b9f4960.tim@ngus.net> <94E003E7-B683-4726-A8F0-5C6E848243BD@mac.com> Message-ID: Please don't make vague threats on this list, only relevant discussion. From icculus at icculus.org Wed Nov 30 01:54:37 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Wed, 30 Nov 2005 01:54:37 -0500 Subject: [quake3] Re: quake3 MacOS hackery, continued... In-Reply-To: <94E003E7-B683-4726-A8F0-5C6E848243BD@mac.com> References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> <20051129003854.4b9f4960.tim@ngus.net> <94E003E7-B683-4726-A8F0-5C6E848243BD@mac.com> Message-ID: <438D4CAD.7070100@icculus.org> > It's your guys project, so if you guys want to dump SMP support, > that's your decision. But Mac users have grown accustomed to the > support being their so if it's not their anymore, well, you know Mac > users... I'm not sure I do know Mac users in the way you're insinuating, but no matter. For what it's worth, I can duplicate the results on the Mac...a timedemo of demos/FOUR.DM_68, with no SMP, ran at 218fps on a dual G5/2.0GHz with an NV40...with r_smp 1, it got 288. This is with the SDL code and some Mac-specifics to make it SMP-friendly. Granted, I'm way past the default value of com_maxfps here either way, but that is a significant boost. I was misunderstanding the code as something that renders in one thread and does game logic in the other, but that's not the case; it looks like it renders in both, so you're issuing GL commands on one CPU while the GL is processing the other CPU's frame. Frankly, I think it's a miracle that this works at all, and the fact that it does is probably a testament to how much engineering Apple was willing to throw at this specific problem at the time to improve benchmarks. On linux, the SMP code is a HUGE performance _loss_...same timedemo, dual Opteron 240 system with a Quadro FX 1100, yields 206fps with r_smp 1, and 248fps _without_ it. So I propose we add a few lines to sdl_glimp.c to allow Mac OS X to use SMP, and leave it disabled everywhere else. I think that should make everyone happy. --ryan. From a2daj at mac.com Wed Nov 30 10:51:57 2005 From: a2daj at mac.com (Avelino Santa Ana Jr.) Date: Wed, 30 Nov 2005 07:51:57 -0800 Subject: [quake3] Re: quake3 MacOS hackery, continued... In-Reply-To: References: <43881517.2000609@icculus.org> <3B4FECB4-97F0-4440-B16B-7C3B1374E62E@mac.com> <438ACE7D.7090303@icculus.org> <8F245D49-CAE7-4A50-BABB-E021DB39FABA@timedoctor.org> <005001c5f454$f9f09150$6e01a8c0@sixfour> <92CCC5AF-5D1B-438E-A0BE-A79F44D2B062@mac.com> <20051129003854.4b9f4960.tim@ngus.net> <94E003E7-B683-4726-A8F0-5C6E848243BD@mac.com> Message-ID: On Nov 29, 2005, at 8:58 PM, Zachary J. Slater wrote: > Please don't make vague threats on this list, only relevant > discussion. I'm not making vague threats. I was insinuating that they'll just complain heavily, maybe throwing in a few conspiracy theories, etc. Why does it seem like people take my messages as a personal insult? I'm just providing feedback on what I've seen with ioQ3 on my Macs. From oleroessler at compuserve.de Wed Nov 30 12:29:33 2005 From: oleroessler at compuserve.de (Ole =?iso-8859-1?q?Roe=DFler?=) Date: Wed, 30 Nov 2005 18:29:33 +0100 Subject: [quake3] No dynamic GL support in video driver Message-ID: <200511301829.33614.oleroessler@compuserve.de> hello @ list! i downloaded the latest svn-revision (389). compiling went well, but when i try to start quake3 i get this: bash-2.05b$ ./quake3 ioQ3 1.33 linux-i386 Nov 28 2005 ----- FS_Startup ----- Current search path: /home/ole/.q3a/baseq3 /usr/local/games/quake3/baseq3/pak8.pk3 (9 files) /usr/local/games/quake3/baseq3/pak7.pk3 (4 files) /usr/local/games/quake3/baseq3/pak6.pk3 (64 files) /usr/local/games/quake3/baseq3/pak5.pk3 (7 files) /usr/local/games/quake3/baseq3/pak4.pk3 (272 files) /usr/local/games/quake3/baseq3/pak3.pk3 (4 files) /usr/local/games/quake3/baseq3/pak2.pk3 (148 files) /usr/local/games/quake3/baseq3/pak1.pk3 (26 files) /usr/local/games/quake3/baseq3/pak0.pk3 (3539 files) /usr/local/games/quake3/baseq3/Dangercity.pk3 (198 files) /usr/local/games/quake3/baseq3 ./baseq3/pak8.pk3 (9 files) ./baseq3/pak7.pk3 (4 files) ./baseq3/pak6.pk3 (64 files) ./baseq3/pak5.pk3 (7 files) ./baseq3/pak4.pk3 (272 files) ./baseq3/pak3.pk3 (4 files) ./baseq3/pak2.pk3 (148 files) ./baseq3/pak1.pk3 (26 files) ./baseq3/pak0.pk3 (3539 files) ./baseq3/Dangercity.pk3 (198 files) ./baseq3 ---------------------- 8542 files in pk3 files execing default.cfg execing q3config.cfg couldn't exec autoexec.cfg Hunk_Clear: reset the hunk ok ----- Client Initialization ----- ----- Initializing Renderer ---- ------------------------------- ----- Client Initialization Complete ----- ----- R_Init ----- ------- Input Initialization ------- Joystick is not active. ------------------------------------ ...loading libGL.so.1: Calling SDL_Init(SDL_INIT_VIDEO)... SDL_Init(SDL_INIT_VIDEO) passed. QGL_Init: Can't load libGL.so.1 from /etc/ld.so.conf or current dir: No dynamic GL support in video driver failed ----- CL_Shutdown ----- RE_Shutdown( 1 ) ----------------------- ----- CL_Shutdown ----- ----------------------- Sys_Error: GLimp_Init() - could not load OpenGL subsystem i don't know what the "No dynamic GL support in video driver" means, can anyone help me with this? i'm using the latest nvidia driver, other opengl-apps - especially the id q3a binaries - do work well... thank you greetings ole From icculus at icculus.org Wed Nov 30 14:58:11 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Wed, 30 Nov 2005 14:58:11 -0500 Subject: [quake3] No dynamic GL support in video driver In-Reply-To: <200511301829.33614.oleroessler@compuserve.de> References: <200511301829.33614.oleroessler@compuserve.de> Message-ID: <438E0453.3060400@icculus.org> > i'm using the latest nvidia driver, other opengl-apps - especially the id q3a > binaries - do work well... It means SDL was built without OpenGL support on your system. --ryan. From zakk at timedoctor.org Wed Nov 30 22:02:38 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 30 Nov 2005 19:02:38 -0800 Subject: ibook G3 Quake 3 Message-ID: Got Quake 3 running on my ibook tonight, 1) Model bug wasn't there... 2) Other rendering errors (missing mirror, health biscuits missing the crunchy outer shell, etc) were there 3) The binary was compiled on a G4, so that is why the binary crashed when it got to the menus if I didn't run it with +set com_altivec 0 I was able to run around in single player, and it performed like a champ, so that is good. -- - Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From zakk at timedoctor.org Wed Nov 30 22:16:18 2005 From: zakk at timedoctor.org (Zachary J. Slater) Date: Wed, 30 Nov 2005 19:16:18 -0800 Subject: [quake3] ibook G3 Quake 3 In-Reply-To: References: Message-ID: Ugh, crashed on trying to search for network servers. - Zachary J. Slater zakk at timedoctor.org zacharyslater at gmail.com From icculus at icculus.org Wed Nov 30 23:07:49 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Wed, 30 Nov 2005 23:07:49 -0500 Subject: [quake3] ibook G3 Quake 3 In-Reply-To: References: Message-ID: <438E7715.9010907@icculus.org> > 1) Model bug wasn't there... Pretty sure you were seeing the Altivec bug on the G4, then. > 3) The binary was compiled on a G4, so that is why the binary crashed > when it got > to the menus if I didn't run it with +set com_altivec 0 This is bug...it shouldn't matter what system it is built on, and it should force com_altivec to 0 for you on a G3 (even though the Altivec code still exists in the binary). Please send me a backtrace of where this dies on a G3 in this condition so I can fix it. --ryan. From icculus at icculus.org Wed Nov 30 23:09:10 2005 From: icculus at icculus.org (Ryan C. Gordon) Date: Wed, 30 Nov 2005 23:09:10 -0500 Subject: [quake3] ibook G3 Quake 3 In-Reply-To: References: Message-ID: <438E7766.1030005@icculus.org> > Ugh, crashed on trying to search for network servers. Yeah, no one's tried that yet. :) Backtraces are helpful in these cases, so please stuff them in Bugzilla...the "report to apple" crash dialog has the backtrace, so you don't have to run under gdb to get it on Mac OS X. --ryan.