mad-dev

mad-dev@lists.mars.org

682 discussions

Re: Re: [mad-dev] how build a standalone mad for arm?
by acen10 18 Mar '02

18 Mar '02

On 2002-03-16 12:22:00 Gareth Bransby wrote: >> the lartware's compile-tool is arm-linux ,the executable >>can >> run without linux? i think it should be arm-elf if i wanna build >> standalone mad,that is right? > >The arm-linux compiler should be fine as far as i know. As >long as your code has no system calls it should run without >linux. For example minimad.c has things like that mmap >call, which i think requires an os. > i dont know how to initiate arm and set the start address using arm-linux.whether in ctr0.s ? >> minimad can >> be compiled and ran no complaint,but it look like doing nothing. > >Use minimad like this : ./minimad < file.mp3 > file.wav >It reads and writes from stdin and stdout. > what do the stdin and stdout refer to ? keyboard and screen? in minimad.c : int main(int argc, char *argv[]) { struct stat stat; void *fdm; if (argc != 1) return 1; .. if you run minimad with any parameters it will return doing nothing. > acen10 acen10(a)263.net

2 1

Re: [mad-dev] how build a standalone mad for arm?
by acen10＠263.net 16 Mar '02

16 Mar '02

Hi, On 2002-03-15 10:29:00 Gareth Bransby wrote: >Hi > >I am attempting the same thing as you. I have compiled libmad using the cross compiler from the Lart website : > >www.lart.tudelft.nl/lartware/compile-tools/ the lartware's compile-tool is arm-linux ,the executable can run without linux? i think it should be arm-elf if i wanna build standalone mad,that is right? >I think minimad.c can be modified fairly easily so it is >independent of any os or system calls. I am having trouble >cross compiling it though so i can start changing it. My >board only has an ARM7td - no enhanced multiply. i am using cygwin and i configured and built successfully mad-0.13.0b on my host.the madplay can produce wave file which can be playback .minimad can be compiled and ran no complaint,but it look like doing nothing. would you like tell me any idea to modify minimad.c? acen10 acen10(a)263.net

2 1

(no subject)
by Gareth 16 Mar '02

16 Mar '02

> You should be able to build minimad with just "make minimad". This > will link with the library in the libmad subdirectory, which is > usually more appropriate than the installed /usr/local/lib/libmad.a > especially when cross-compiling. I had a go at doing 'make minimad' like you suggested. For some reason it seems to be looking for files in tmp/ of my home directory. /home/gar/tm/ccPMTjTj.o: In function 'decode': /home/gar/tm/ccPMTjTj.o: undefined reference to 'mad_decoder_init' /home/gar/tm/ccPMTjTj.o: relocation truncated to fit: R_ARM_PC24 mad_decoder_run I get about 5 of these errors and then it exits with collect2: ld returned 1 exit status I had a look in tmp/ for the .o files it is refering to but they arn't there.

1 0

Re: [mad-dev] MIPS VR4111 implementation of MAD
by Jeroen T 15 Mar '02

15 Mar '02

I just start at the beginning. I have a small MP3 file from 1 second stored in the RAM of the MIPS. It is a one second 440 Hz sinus. Mad decodes it OK on a PC, but not on the MIPS..... Jeroen >From: Guenther Sohler <guenther.sohler(a)newlogic.com> >Reply-To: Guenther Sohler <guenther.sohler(a)newlogic.com> >To: Jeroen T <jeroen_t(a)hotmail.com> >CC: mad-dev(a)lists.mars.org >Subject: Re: [mad-dev] MIPS VR4111 implementation of MAD >Date: Wed, 13 Mar 2002 15:27:51 +0100 (MET) > >I dont have an idea, whether mad is designed to run on big endian machines. >Do you know of any success story ? >I dont know mad too much. i just know, that it works in my amd athlon box. >By seeking i mean, immediatly starting to play an mp3 at a virtual >wave-file-position _________________________________________________________________ Chat with friends online, try MSN Messenger: http://messenger.msn.com

5 7

how build a standalone mad for arm?
by acen10＠263.net 15 Mar '02

15 Mar '02

hi all, how can i make a standalone player on arm7tdmi without OS? i wanna play mp3 on arm cpu,and porting a uclinux to it is too complex for me.furthermore,os itself consume resources and i am afraided the cpu cannt deal with mp3 decoding.but how do it ?i am newbie and any advice and help will be appreciated acen10(a)263.net

3 2

Re: [mad-dev] how build a standalone mad for arm?
by acen10＠263.net 15 Mar '02

15 Mar '02

hi, Guenther Sohler,did you have any experience on it ? first step,i wanna modify the source code ,make mad can run and decode mp3 whithout system support like calling system function . then next step i will build executable for arm. that is appropriate and feasible? on 2002-03-15 08:17:00 Guenther Sohler wrote： >I presume, you'll gonna construct an car mp3 player. this is a good idea. >Please inform me about any details. i am also interested >On 15-Mar-2002 acen10(a)263.net wrote: >> hi all, >> how can i make a standalone player on arm7tdmi without OS? >> i wanna play mp3 on arm cpu,and porting a uclinux to it is too complex for >> me.furthermore,os itself consume resources and i am afraided the cpu cannt >> deal with mp3 decoding.but how do it >> ?i am newbie and any advice and help will be appreciated >> >> acen10(a)263.net >> > >Guenther Sohler > >NewLogic Technologies AG >Millennium Park 6 >A-6890 Lustenau > Phone: +43-5577-62000-507 >E-Mail: guenther.sohler(a)newlogic.com Fax: +43-5577-62000-988 acen10(a)263.net

1 0

Re: [mad-dev] Compiling for ARM
by Gareth Bransby 14 Mar '02

14 Mar '02

> 1) > When compiling minimad normally i do "gcc minmad.c -o mimimad -lmad". -lmad > being the bit that tells gcc to use libmad dunring linking. How does -lmad > relate to /usr/local/lib/libmad.a? When i try this using the cross compiler > it fails. More specicfcally I get : /data/lart/cross/arm-linux/bin/ld: cannot find -lmad collect2: ld returned 1 exit status

1 0

Re: [mad-dev] Compiling for ARM
by Jeroen T 14 Mar '02

14 Mar '02

Hi. A general question about the ARM. What byte order is used in memory for 16 and 32 bit value? Is it high-lowbyte or low-highbyte. I have problems with a MIPS. Decoding works, huffman doesn't give any errors, but the output is not what I expect... Jeroen >From: Guenther Sohler <guenther.sohler(a)newlogic.com> >Reply-To: Guenther Sohler <guenther.sohler(a)newlogic.com> >To: Gareth <the_african(a)hotmail.com> >CC: mad-dev(a)lists.mars.org >Subject: Re: [mad-dev] Compiling for ARM >Date: Thu, 14 Mar 2002 07:48:11 +0100 (MET) > > >On 14-Mar-2002 Gareth wrote: > > Hi, > > > > I have succesfully cross compiled libmad for ARM and now it is time for > > compiling minimad for ARM as well. I have 2 question really > > > > 1) > > When compiling minimad normally i do "gcc minmad.c -o mimimad -lmad". >-lmad > > being the bit that tells gcc to use libmad dunring linking. How does >-lmad > > relate to /usr/local/lib/libmad.a? When i try this using the cross >compiler > > it fails. >Hmm - In My system there does not exist libmad.a in the libmad directory >There just exist a libmad.la >Wha I did, I gave the linker a file in .libs. i dont remember the name, >but i >gave it without the -l option(for library) >Is there a way to make a shared library versin of libmad ? > >Do you need the capability to start from a defined position in the mp3 file >? > _________________________________________________________________ Send and receive Hotmail on your mobile device: http://mobile.msn.com

1 0

Re: [mad-dev] MIPS VR4111 implementation of MAD
by Jeroen T 13 Mar '02

13 Mar '02

I don't know.... The only thing I know is that 16 and 32 bits values on the MIPS are stored using high to low byte in memory and the Intel low to high byte in memory. Can this be a problem? Does MAD use direct way to get bytes from a 16 bits or 32 bits value? Jeroen >From: Guenther Sohler <guenther.sohler(a)newlogic.com> >Reply-To: Guenther Sohler <guenther.sohler(a)newlogic.com> >To: Jeroen <jeroen_t(a)hotmail.com> >CC: mad-dev(a)lists.mars.org >Subject: Re: [mad-dev] MIPS VR4111 implementation of MAD >Date: Wed, 13 Mar 2002 15:12:23 +0100 (MET) > >There might be a problem with the big endian of the machine ? >Do you use seeking in your app ? > >On 13-Mar-2002 Jeroen wrote: > > Hi. > > > > I asked before, and thanks for all the anwers. I still didn't have any > > success on porting MAD to a MIPS VR4111. It compiles perfectly, bus the > > output of the decoder looks like some audio, but it isn't what I >expected... > > Anyone here who did porting to a MIPS before? > > > > Thanks, > > > > Jeroen > > > >Guenther Sohler > >NewLogic Technologies AG >Millennium Park 6 >A-6890 Lustenau > Phone: +43-5577-62000-507 >E-Mail: guenther.sohler(a)newlogic.com Fax: +43-5577-62000-988 > _________________________________________________________________ Send and receive Hotmail on your mobile device: http://mobile.msn.com

2 1

Re: [mad-dev] MIPS VR4111 implementation of MAD
by Jeroen T 13 Mar '02

13 Mar '02

And seeking? What do you mean with that? I just decode a complete MP3 and write it back to a file. Jeroen >From: Guenther Sohler <guenther.sohler(a)newlogic.com> >Reply-To: Guenther Sohler <guenther.sohler(a)newlogic.com> >To: Jeroen <jeroen_t(a)hotmail.com> >CC: mad-dev(a)lists.mars.org >Subject: Re: [mad-dev] MIPS VR4111 implementation of MAD >Date: Wed, 13 Mar 2002 15:12:23 +0100 (MET) > >There might be a problem with the big endian of the machine ? >Do you use seeking in your app ? > >On 13-Mar-2002 Jeroen wrote: > > Hi. > > > > I asked before, and thanks for all the anwers. I still didn't have any > > success on porting MAD to a MIPS VR4111. It compiles perfectly, bus the > > output of the decoder looks like some audio, but it isn't what I >expected... > > Anyone here who did porting to a MIPS before? > > > > Thanks, > > > > Jeroen > > > >Guenther Sohler > >NewLogic Technologies AG >Millennium Park 6 >A-6890 Lustenau > Phone: +43-5577-62000-507 >E-Mail: guenther.sohler(a)newlogic.com Fax: +43-5577-62000-988 > _________________________________________________________________ Join the worlds largest e-mail service with MSN Hotmail. http://www.hotmail.com

1 0

MIPS VR4111 implementation of MAD
by Jeroen 13 Mar '02

13 Mar '02

Hi. I asked before, and thanks for all the anwers. I still didn't have any success on porting MAD to a MIPS VR4111. It compiles perfectly, bus the output of the decoder looks like some audio, but it isn't what I expected... Anyone here who did porting to a MIPS before? Thanks, Jeroen

2 1

Problem with MADPLAYER
by Guenther Sohler 13 Mar '02

13 Mar '02

Hallo List, i have written a mid level access class for mad mp3 decoder engine. It can open, close, read and seek within a mp3 file. But unconfortunately the seek function does not work. The seek function just reads the header of each frame to be faster. Somehow it seeks 19.8x more far than specified. Could anybode look into the code and see whats wrong. thanx rds guenther

1 0

Re: [mad-dev] Cross compiling libmad for ARM
by Gareth Bransby 08 Mar '02

08 Mar '02

I had a go at what you suggested but this had no effect, it came up with exactly the same error. From what i can see from the error, it is looking in the linux/ directory in the libmad/ source directory, not in /usr/src/linux. I tried doing what you suggested but making the link in the libmad directory rather than in /usr/src, so now i have a link mad-0.13.0b/libmad/linux -> /usr/local/arm/gnu/release/arm-linux/include. Now it finds errno.h ok but generates different errors decoder.c: In function 'mad_decoder_finish': decoder.c:91 'EINTR' undeclared (first use in this function) decoder.c:91 (Each undeclared identifier is reported only once for each function it apears in.) decoder.c In function'send_io': decoder.c:120 'EINTR' undeclared (first use in this function) decoder.c In function'recieve_io': decoder.c:142 'EINTR' undeclared (first use in this function) decoder.c:145 'EAGAIN' undeclared (first use in this function) make: *** [decoder.lo] error 1 I am wondering are there some required files that are needed for cross compiling that i have not installed? Thanks On Tue, 5 Mar 2002 13:45:55 +0800 "davidlee" <davidlee(a)gv.com.tw> wrote: > Dear Gareth: > Maybe do could do this > 1. logon as root > 2. cd /usr/src > 3. rm linux > 4. ln -s arm-linux(or any path that you installed Arm Linux enviroment) ./linux > > Try again , Good lucky. > > > ----- Original Message ----- > From: "Gareth" <g.c.bransby-99(a)student.lboro.ac.uk> > To: "Gad Hayisraeli" <gad(a)syete.co.il> > Cc: "Mad Dev List" <mad-dev(a)lists.mars.org> > Sent: Thursday, February 28, 2002 7:56 PM > Subject: RE: [mad-dev] Cross compiling libmad for ARM > > > > Thanks for the help! > > > > I had a go at doing the CC=... and RANLIB=... on the same line as the > > ./configure as suggested. This seems to have the dedsired effect, the > > configure script reports that gcc is a cross compiler and when i look at > > the Makefile all the paths are pointing to the right place. i.e. the > > cross compiler i installed. > > > > Now when i come to make, it runs for a bit but then reports the error : > > > > /usr/local/arm/gnu/release/lib/gcc-lib/arm-linux/2.95.3/../../../../arm-linux/include/bits/errno.h:25: linux/errno.h: No such file or directory > > > > Why is the file bits/errno.h looking for another file of the same name? > > Is this meant to be an ARM specific errno.h? I did a search for this > > file and came up with loads : > > > > /usr/include/asm/errno.h > > /usr/include/linux/errno.h > > /usr/include/bits/errno.h > > /usr/include/errno.h > > /usr/include/sys/errno.h > > /usr/local/arm/gnu/release/arm-linux/include/bits/errno.h > > /usr/local/arm/gnu/release/arm-linux/include/errno.h > > /usr/local/arm/gnu/release/arm-linux/include/sys/errno.h > > /usr/src/linux-2.4.3/include/asm-i386/errno.h > > /usr/src/linux-2.4.3/include/linux/errno.h > > > > Which one is it looking for and how do i tell it to use that specific > > one? > > > > > > On 27 Feb 2002 21:24:37 +0200, Gad Hayisraeli wrote: > > > Try this > > > Shell> rm -f config.cache > > > Shell> CC=<path>/arm-linux-gcc ./configure <options> > > > (CC + configure - all in one line) > > > > > > Gad > > > > > > > > > > -----Original Message----- > > > > From: mad-dev-admin(a)lists.mars.org > > > [mailto:mad-dev-admin@lists.mars.org] On Behalf Of Gareth > > > > Sent: Wednesday, February 27, 2002 9:02 PM > > > > To: mad-dev(a)lists.mars.org > > > > Subject: [mad-dev] Cross compiling libmad for ARM > > > > > > > > Hi, > > > > I am wanting to compile libmad for ARM. I have installed : > > > > > > > > arm-linux-binutils-2.11-1 > > > > arm-linux-gcc-2.95-1 > > > > arm-linux-glibc-2.1.3-1 > > > > > > > > To cross compile i do : > > > > > > > > export CC=/usr/local/arm/gnu/release/bin/arm-linux-gcc > > > > export RANLIB=/usr/local/arm/gnu/release/bin/arm-linux-ranlib > > > > export LD=/usr/local/arm/gnu/release/bin/arm-linux-ld > > > > export NM=/usr/local/arm/gnu/release/bin/arm-linux-nm > > > > > > > > ./configure --host=arm-linux --enable-speed --enable-fpm=arm > > > > > > > > The configure script completes fine but in the messages it prints i > > > see > > > > the lines : > > > > > > > > checking weather the C compiler (gcc ) is a cross-compiler ... no > > > > . > > > > . > > > > cheching for ld used by GCC... (cached) /usr/bin/ld > > > > > > > > so it is not using the cross compiler that i installed. I read in > > > > another mail that i may have to configure manually. Is this the case > > > or > > > > is there somthing i am missing out? > > > > > > > > Thanks in advance. > > > > > >

3 8

Porting on TOSHIBA TX3911
by davidlee 06 Mar '02

06 Mar '02

Dear All: I'm porting MAD to TOSHIBA TX3911, but it seems my cross compiler mipsel-linux-gcc can't compile the asm instruction "madd" that surpported by TX3911 hardware. I wand if anybody can tell me eithe I should change compiler or I shoud change compiler's parameter and how. Thanks in advance. 李大冀 Email: davidlee(a)gv.com.tw

2 2

How can I use configure tools to create a makefile suitable my targer board.
by davidlee 04 Mar '02

04 Mar '02

Dear All: My host is PC, and I'm porting MAD to my Target Board with Toshiba TX3911 CPU. There is many configure tools ,but I don't how to use them to create proper Makefile that can compile TX3911's "madd" instruction. BEST REGARD 李大冀 Email: davidlee(a)gv.com.tw

1 0

What problems can I expect when porting MAD to a 32 bits MIPS VR4111
by Jeroen T 02 Mar '02

02 Mar '02

Hi. I am developing software to play MP3 files from a MIPS VR4111. I connected a IDE harddrive to it to store MP3 files. The software for the harddrive is almost finished and it supports FAT32 (root directory only). Now I still got some problems with porting MAD to the MIPS. It's compiling, and it seems to decode, but the outputdata is not what I expected. I decode a 1 second 440 HZ sinus, and when I decode it on a PC the outout data is OK. On the MIPS the outputdata looks like multiple sinusses with multiple frequencies added to each other. I looked for problems in the multiply functions and probable little/big endian issues. I don't know what to look further to. What can be the problem? When I have MAD running on the MIPS, I will have a stand alone MP3 jukebox with large harddrive! Hope someone can give some further tips. Jeroen _________________________________________________________________ Chat with friends online, try MSN Messenger: http://messenger.msn.com

1 0

Delphi & MAD ?
by Nicola Programmer 02 Mar '02

02 Mar '02

Anybody know how make MAD work from Delphi on Windows platform ? Maybe someone have sources or components wich can use MAD library in Delphi. Thanks. __________________________________________________ Do You Yahoo!? Yahoo! Sports - sign up for Fantasy Baseball http://sports.yahoo.com

1 0

tics
by Mathieu Dube 02 Mar '02

02 Mar '02

Hi, Im decoding mp3s with a program I wrote using libmad. With most mp3s it works fine but with some it does some tics that it doesnt do if I play the mp3 in xmms. Maybe thats a coincidence but all the mp3s that have tics have a 128kbps bitrate. Did anybody have that type of problems?? Would it be in the reading/decoding or the writing that the problem would be? Thanks -Mat -- Mathieu Dube -- mdube(a)8d.com 8D Technologies Inc. http://www.8d.com

2 3

Re: [mad-user] Porting MAD to MIPSIII
by davidlee 02 Mar '02

02 Mar '02

Thank you suggestion , It's my First time deal with MP3 and MAD. I could feel grateful for any help. ----- Original Message ----- From: "Rob Leslie" <rob(a)mars.org> To: <mad-user(a)lists.mars.org> Sent: Wednesday, February 27, 2002 6:23 AM Subject: Re: [mad-user] Porting MAD to MIPSIII > Please note: developer-oriented messages (porting, compiling, > optimizing, etc.) are more appropriate to post to mad-dev, not this > list. The mad-user list is intended for user-oriented discussions of > capabilities, features, usage, and relevant audio topics of general > interest. > > I have set the Reply-To: of this message accordingly. Note that you will > not be able to reply to this message unless you are also subscribed to > mad-dev. (You may wish to review your subscriptions as appropriate.) > > On Thursday, February 28, 2002, at 07:05 PM, davidlee wrote: > > I have a develope board with MipsIII running at 55.8MHz. I wonder > > if I can port MAD on it. Or if there is anyon had done it can give me > > some sugguestion. > > I have ported MAD to a couple of MIPS platforms. It really helps if your > CPU has a 32-bit multiply/accumulate instruction, otherwise you will not > be able to get the desired performance or audio quality. 56MHz may be on > the slow side to get real-time performance from MAD, but much depends on > the availability of mul/add and the quality of code generated by your > compiler. > > -- > Rob Leslie > rob(a)mars.org

1 0

Need some help on usage of the 'low level' interface to libmad
by Lionel Ulmer 28 Feb '02

28 Feb '02

Hi all, For fun, I was using MAD to 'MP3ize' ScummVM (I will let people unaware of what it is to research it on Google :-) ) to reduce the size used by the sound effect data files. For that, I am using the code attached to this mail (extracted from the ScummVM code for ease of debug). If you want to test this on a Linux box, just compile it and do NOT forget to redirect STDERR to a file (as it will output the 'raw' samples there). When playing the MP3 you can find here : http://www.bbrox.org/debug_002.mp3 with MadPlay, it works perfectly. Whereas with my small tool, it almost works but there are a lot of garbade noises added (some sort of clicks, but at a low frequency). I have already used MAD in my movie player, and (as far as I know), I did exactly the same sort of processing there (and in my movie player it works). So what did I do wrong :-) ? I need another pair (pairs ?) of eyes because it's now been one day that I look at this and I still cannot find what I did wrong. Lionel (dreading to find THE obvious bug 2 seconds after pressing the 's'(end) key :-) ) -- Lionel Ulmer - http://www.bbrox.org/

1 2

Cross compiling libmad for ARM
by Gareth 28 Feb '02

28 Feb '02

Hi, I am wanting to compile libmad for ARM. I have installed : arm-linux-binutils-2.11-1 arm-linux-gcc-2.95-1 arm-linux-glibc-2.1.3-1 To cross compile i do : export CC=/usr/local/arm/gnu/release/bin/arm-linux-gcc export RANLIB=/usr/local/arm/gnu/release/bin/arm-linux-ranlib export LD=/usr/local/arm/gnu/release/bin/arm-linux-ld export NM=/usr/local/arm/gnu/release/bin/arm-linux-nm ./configure --host=arm-linux --enable-speed --enable-fpm=arm The configure script completes fine but in the messages it prints i see the lines : checking weather the C compiler (gcc ) is a cross-compiler ... no . . cheching for ld used by GCC... (cached) /usr/bin/ld so it is not using the cross compiler that i installed. I read in another mail that i may have to configure manually. Is this the case or is there somthing i am missing out? Thanks in advance.

4 4

Re: Problems compiling . soft-float
by Jay Carlson 26 Feb '02

26 Feb '02

On Tuesday, February 26, 2002, at 12:10 AM, Daniel Jacobowitz wrote: > On Tue, Feb 26, 2002 at 06:02:37AM +0100, Ralf Baechle wrote: >> On Mon, Feb 25, 2002 at 11:47:44PM -0500, Jay Carlson wrote: >> >>> Ralf is right that the kernel emulator is the supported route. But if >>> you're willing to go to the trouble of building everything from >>> scratch, >>> this does work. >> >> It's really a major pain. Softfp isn't defined in the ABI which >> assumes >> an FPU is available. As the result there is no provision for mixing >> softfp and fp-less code. >> >> Something for the binutils to-do list - ld should make mixing hard-fp >> and soft-fp binaries impossible. > > Or we could see if it is possible to define the ABIs in such a way that > they can call each other... I don't immediately see a problem. The > only code that will clobber FP registers is FP code. Oh, they can intercall without any error until you call sin(1.0), at which point one side's calling convention places 1.0 in integer registers and the other places it in fp registers. Lots of fun to debug this one, especially once you have a kernel emulator that won't SIGILL for the hardfp ops :-) Seriously, this was a common screwup in the Agenda world until the snow specs file. Jay

1 0

Re: Problems compiling . soft-float
by Jay Carlson 26 Feb '02

26 Feb '02

On Monday, February 25, 2002, at 07:25 AM, Ralf Baechle wrote: > On Mon, Feb 25, 2002 at 08:19:36AM +0100, Carlo Agostini wrote: [link fails due to missing "dpmul", "dpadd", etc.] > >> Then, I tried to pass explicitly -msoft-float to gcc as an argument >> ..... > > Not supported. Use the kernel fp emulation. This works fine if you undo the mistake in gcc/config/mips/t-linux, or in elf.h, your choice. The default on MIPS is for gcc to use the GOFAST style of soft-float calls. gcc/config/mips/elf.h sets this up via #include "gofast.h". Now, t-linux *does* include the right goo to get all the software floating point emulation code compiled into libgcc.a. But it's defaulting not to the GNU style but to the GOFAST style calls: echo '#undef US_SOFTWARE_GOFAST' >> dp-bit.c So everything works if we just make the two consistent. The old(er) Agenda toolchain takes the position that "every other gcc MIPS target uses GOFAST, so should we"; you can just change that #undef to #define. But I (now) think the better path is to stick to the GNU-style softfloat calls, Unless you want to modify elf.h, the best you can do is undo the #include "gofast.h" damage in linux.h. Near the end: +/* undo the effects of elf.h including gofast.h */ +#undef US_SOFTWARE_GOFAST +#undef INIT_GOFAST_OPTABS +#define INIT_GOFAST_OPTABS + +#undef GOFAST_CLEAR_NEG_FLOAT_OPTAB +#undef GOFAST_RENAME_LIBCALLS This of course requires knowledge of what exactly gofast.h did, so it's not optimal either. Ralf is right that the kernel emulator is the supported route. But if you're willing to go to the trouble of building everything from scratch, this does work. Jay

1 0

alternative idct implementation question
by Yuxuan Mao 24 Feb '02

24 Feb '02

Hi, I'm doing a school project on mp3 decoders. Other teams in my class is proposing to do an 8-bit cosine lookup table implementation of the idct. Does anyone know if this will work? Yuxuan

1 0

Re: Welcome to the "mad-dev" mailing list (Digest mode)
by Yuxuan Mao 22 Feb '02

22 Feb '02

Hi, We are doing a mp3 decoder school project (building a decoder on a Motorola microcontroller and an altera fpga). We're using MAD because it's an all integer implementation. But we're only interested in mpeg1 layer 3 decoding. We're kinda lost as to which files from the distribution we need just for this. We want to get rid of anything that's not mpeg1 layer 3...

2 1

← Newer
1
...
17
18
19
20
21
22
23
...
28
Older →

Jump to page:

2025

2024

2023

2022

2021

2020

2019

2018

2017

2016

2015

2014

2013

2012

2011

2010

2009

2008

2007

2006

2005

2004

2003

2002

2001

2000

mad-dev