Message-ID: <001501bfcbd4$e759f1c0$020aa8c0@cervantes> From: "Tilman Sauerbeck" To: References: Subject: Re: problem using "make" Date: Thu, 1 Jun 2000 16:22:27 +0200 MIME-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 8bit X-Priority: 3 X-MSMail-Priority: Normal X-Mailer: Microsoft Outlook Express 5.00.2615.200 X-MimeOLE: Produced By Microsoft MimeOLE V5.00.2615.200 Reply-To: djgpp AT delorie DOT com > Please post the full message printed by Make in this case. Please also > post the Makefile and the file `envlist.txt' produced if you type this: Okay, here's the output of Make: C:\WINDOWS\COMMAND.COM -ec 'gcc -M -DHAVEMPGLIB -DLAMEPARSE -DNDEBUG -D__NO_MATH_INLINES -O -DLAMESNDF ILE mpglib/main.c | sed '\''s;mpglib/main.o;& mpglib/main.d;g'\'' > mpglib/main.d' Angegebenes COMMAND-Verzeichnis ist falsch # This means: false COMMAND directory Zu viele Parameter # = Too many parameters Zu viele Parameter Zu viele Parameter Zu viele Parameter Zu viele Parameter Zu viele Parameter Zu viele Parameter Zu viele Parameter Zu viele Parameter Ungltige Option # = invalid option Zu viele Parameter Zu viele Parameter Zu viele Parameter Ungltige Option Microsoft(R) Windows 98 (C)Copyright Microsoft Corp 1981-1999. C> C>^C C>exit ------ After Make is run and the errors have been printed on the screen, a new shell is started (as you can see). I tried to run "c:\windows\command.com -ec" manually and it said the same as above: "false COMMAND directory" Here's the "set" output: TMP=C:\WINDOWS\TEMP TEMP=C:\WINDOWS\TEMP PROMPT=$p$g winbootdir=C:\WINDOWS COMSPEC=C:\WINDOWS\COMMAND.COM DJGPP=E:\Develop\djgpp\djgpp.env PATH=E:\DEVELOP\DJGPP\BIN;C:\WINDOWS;C:\WINDOWS\COMMAND windir=C:\WINDOWS BLASTER=A220 I5 D1 T4 P330 CMDLINE=make Setting COMSPEC didn't change anything. CMDLINE was set when I called make I think. Finally here's the Makefile: # Makefile for LAME 3.xx # # LAME is reported to work under: # Linux (i86), NetBSD 1.3.2 (StrongARM), FreeBSD (i86) # Compaq Alpha(OSF, Linux, Tru64 Unix), Sun Solaris, SGI IRIX, # OS2 Warp, Macintosh PPC, BeOS, Amiga and even VC++ # UNAME = $(shell uname) ARCH = $(shell uname -m) # defaults: PGM = lame CC = gcc CC_OPTS = -O GTK = GTKLIBS = SNDLIB = -DLAMESNDFILE LIBSNDFILE = LIBS = -lm MAKEDEP = -M BRHIST_SWITCH = LIBTERMCAP = RM = rm -f CPP_OPTS = -DHAVEMPGLIB -DLAMEPARSE ########################################################################## # To remove support for mp3 *decoding*, # remove -DHAVEMPGLIB to CPP_OPTS ########################################################################## ########################################################################## # Define these to compile in code for the optional VBR bitrate histogram. # Requires ncurses, but libtermcap also works. # If you have any trouble, just dont define these ########################################################################## #BRHIST_SWITCH = -DBRHIST #LIBTERMCAP = -lncurses #LIBTERMCAP = -ltermcap ########################################################################## # SNDLIB = no file i/o # SNDLIB = -DLAMESNDFILE to use internal LAME soundfile routines # SNDLIB = -DLIBSNDFILE to use Erik de Castro Lopo's libsndfile # http://www.zip.com.au/~erikd/libsndfile/ # otherwise LAME can only read 16bit wav, aiff and pcm inputfiles. # Note: at present, libsndfile does not support input from stdin. ########################################################################## #SNDLIB = -DLIBSNDFILE #LIBSNDFILE=-lsndfile # if libsndfile is in a custom location, try: #LIBSNDFILE=-L $(LIBSNDHOME) -lsndfile -I $(LIBSNDHOME) ########################################################################## # define these to use compile in support for the GTK mp3 frame analyzer # Requires -DHAVEMPGLIB # and SNDLIB = -DLAME or -DLIBSNDFILE ########################################################################## #GTK = -DHAVEGTK `gtk-config --cflags` #GTKLIBS = `gtk-config --libs` ########################################################################## # LINUX ########################################################################## ifeq ($(UNAME),Linux) # remove these lines if you dont have GTK, or dont want the GTK frame analyzer GTK = -DHAVEGTK `gtk-config --cflags` GTKLIBS = `gtk-config --libs` # Comment out next 2 lines if you want to remove VBR histogram capability BRHIST_SWITCH = -DBRHIST LIBTERMCAP = -lncurses # suggested for gcc-2.7.x CC_OPTS -O3 -fomit-frame-pointer -funroll-loops -ffast-math -finline-functions -W all # CC_OPTS = -O9 -fomit-frame-pointer -fno-strength-reduce -mpentiumpro -ffast-math -f inline-functions -funroll-loops -Wall -malign-double -g -march=pentiumpro -m fancy-math-387 -pipe # for debugging: # CC_OPTS = -UNDEBUG -O -Wall -g -DABORTFP # for lots of debugging: # CC_OPTS = -DDEBUG -UNDEBUG -O -Wall -g -DABORTFP # The following features are experimental code, # so don't use them if you don't know exactly what you do! (RH 2000-01-30) # FEATURES = -DRH_QUALITY_CONTROL # tuned VBR quality control # FEATURES += -DRH_SIDE_VBR # turns side channel reduction off (VBR) # FEATURES += -DRH_SIDE_CBR # turns side channel reduction off (CBR) # FEATURES += -DRH_ATH # max noise instead of average noise # these options for gcc-2.95.2 to produce fast code # CC_OPTS = $(FEATURES)\ # -Wall -O9 -fomit-frame-pointer -march=pentium \ # -finline-functions -fexpensive-optimizations \ # -ffast-math -malign-double -mfancy-math-387 \ # -funroll-loops -funroll-all-loops -pipe -fschedule-insns2 \ # -fno-strength-reduce ########################################################################## # LINUX on Digital/Compaq Alpha CPUs ########################################################################## ifeq ($(ARCH),alpha) # double is faster than float on Alpha CC_OPTS = -O4 -Wall -fomit-frame-pointer -ffast-math -funroll-loops \ -mfp-regs -fschedule-insns -fschedule-insns2 \ -finline-functions \ # -DFLOAT=double # add "-mcpu=21164a -Wa,-m21164a" to optimize for 21164a (ev56) CPU # Compaq's C Compiler #CC = ccc # Options for Compaq's C Compiler #CC_OPTS = -fast -Wall # standard Linux libm #LIBS = -lm # optimized libffm (free fast math library) #LIBS = -lffm # Compaq's fast math library LIBS = -lcpml endif endif ########################################################################## # FreeBSD ########################################################################## ifeq ($(UNAME),FreeBSD) # remove if you do not have GTK or do not want the GTK frame analyzer GTK = -DHAVEGTK `gtk12-config --cflags` GTKLIBS = `gtk12-config --libs` # Comment out next 2 lines if you want to remove VBR histogram capability BRHIST_SWITCH = -DBRHIST LIBTERMCAP = -lncurses endif ########################################################################## # SunOS ########################################################################## ifeq ($(UNAME),SunOS) CC = cc CC_OPTS = -O -xCC MAKEDEP = -xM endif ########################################################################## # SGI ########################################################################## ifeq ($(UNAME),IRIX64) CC = cc endif ########################################################################## # Compaq Alpha running Dec Unix (OSF) ########################################################################## ifeq ($(UNAME),OSF1) CC = cc CC_OPTS = -fast -O3 -std -g3 -non_shared endif ########################################################################## # BeOS ########################################################################## ifeq ($(UNAME),BeOS) CC = $(BE_C_COMPILER) LIBS = ifeq ($(ARCH),BePC) CC_OPTS = -O9 -fomit-frame-pointer -march=pentium \ -mcpu=pentium -ffast-math -funroll-loops \ -fprofile-arcs -fbranch-probabilities else CC_OPTS = -opt all MAKEDEP = -make endif endif ########################################################################### # MOSXS (Rhapsody PPC) ########################################################################### ifeq ($(UNAME),Rhapsody) CC = cc LIBS = CC_OPTS = -O9 -ffast-math -funroll-loops -fomit-frame-pointer MAKEDEP = -make endif ########################################################################## # OS/2 ########################################################################## # Properly installed EMX runtime & development package is a prerequisite. # tools I used: make 3.76.1, uname 1.12, sed 2.05, PD-ksh 5.2.13 # ########################################################################## ifeq ($(UNAME),OS/2) SHELL=sh CC = gcc CC_OPTS = -O3 PGM = lame.exe LIBS = # I use the following for slightly better performance on my Pentium-II # using pgcc-2.91.66: # CC_OPTS = -O6 -ffast-math -funroll-loops -mpentiumpro -march=pentiumpro # Comment out next 2 lines if you want to remove VBR histogram capability BRHIST_SWITCH = -DBRHIST LIBTERMCAP = -ltermcap # Uncomment & inspect the 2 GTK lines to use MP3x GTK frame analyzer. # Properly installed XFree86/devlibs & GTK+ is a prerequisite. # The following works for me using Xfree86/OS2 3.3.5 and GTK+ 1.2.3: # GTK = -DHAVEGTK -IC:/XFree86/include/gtk12 -Zmt -D__ST_MT_ERRNO__ -IC:/XFree86/i nclude/glib12 -IC:/XFree86/include # GTKLIBS = -LC:/XFree86/lib -Zmtd -Zsysv-signals -Zbin-files -lgtk12 -lgdk12 -lgmodul e -lglib12 -lXext -lX11 -lshm -lbsd -lsocket -lm endif # 10/99 added -D__NO_MATH_INLINES to fix a bug in *all* versions of # gcc 2.8+ as of 10/99. CC_SWITCHES = -DNDEBUG -D__NO_MATH_INLINES $(CC_OPTS) $(SNDLIB) $(GTK) $(BRHIST_SWITCH) c_sources = \ brhist.c \ formatBitstream.c \ fft.c \ get_audio.c \ l3bitstream.c \ id3tag.c \ ieeefloat.c \ lame.c \ newmdct.c \ parse.c \ portableio.c \ psymodel.c \ quantize.c \ quantize-pvt.c \ vbrquantize.c \ reservoir.c \ tables.c \ takehiro.c \ timestatus.c \ util.c \ VbrTag.c \ version.c \ gtkanal.c \ gpkplotting.c \ mpglib/common.c \ mpglib/dct64_i386.c \ mpglib/decode_i386.c \ mpglib/layer3.c \ mpglib/tabinit.c \ mpglib/interface.c \ mpglib/main.c OBJ = $(c_sources:.c=.o) DEP = $(c_sources:.c=.d) %.o: %.c $(CC) $(CPP_OPTS) $(CC_SWITCHES) -c $< -o $@ %.d: %.c $(SHELL) -ec '$(CC) $(MAKEDEP) $(CPP_OPTS) $(CC_SWITCHES) $< | sed '\''s;$*.o;& $@;g'\'' > $@' #$(PGM): main.o $(OBJ) Makefile # $(CC) -o $(PGM) main.o $(OBJ) $(LIBS) $(CPP_OPTS) $(LIBSNDFILE) $(GTKLIBS) $(LIBTERMCAP) $(PGM): main.o libmp3lame.a $(CC) -o $(PGM) main.o -L. -lmp3lame $(LIBS) $(LIBSNDFILE) $(GTKLIBS) $(LIBTERMCAP) mp3x: mp3x.o libmp3lame.a $(CC) -o mp3x mp3x.o $(OBJ) $(LIBS) $(LIBSNDFILE) $(GTKLIBS) $(LIBTERMCAP) mp3rtp: rtp.o mp3rtp.o libmp3lame.a $(CC) -o mp3rtp mp3rtp.o rtp.o $(OBJ) $(LIBS) $(LIBSNDFILE) $(GTKLIBS) $(LIBTERMCAP) libmp3lame.a: $(OBJ) Makefile # cd libmp3lame # make libmp3lame ar cr libmp3lame.a $(OBJ) clean: -$(RM) $(OBJ) $(DEP) $(PGM) main.o rtp.o mp3rtp mp3rtp.o \ mp3x.o mp3x libmp3lame.a tags: TAGS TAGS: ${c_sources} etags -T ${c_sources} ifneq ($(MAKECMDGOALS),clean) -include $(DEP) endif # # testcase.mp3 is a 2926 byte file. The first number output by # wc is the number of bytes which differ between new output # and 'official' results. # # Because of compilier options and effects of roundoff, the # number of bytes which are different may not be zero, but # should be at most 30. # test: $(PGM) ./lame --nores -h testcase.wav testcase.new.mp3 cmp -l testcase.new.mp3 testcase.mp3 | wc testg: $(PGM) ./lame -g -h ../test/castanets.wav ---------------- Thanks for answering to my question, Tilman Sauerbeck