I'm compiling mplayer on the Linux playstation 2 dev-kit. The PS2 has a mips based cpu running in little endian mode and the actual configure failed to configure it properly. I changed the configure script to autodetect the byte ordering. Actually, I don't know if the test works for all cpus so if people can test it on different big endian and little endian cpus I will appreciate. Bertrand
Nobody had a chance to look at this? Any comments are welcomed! Bertrand PS: see previous message fot the patch. On Monday 15 July 2002 11:24 pm, Bertrand Baudet wrote:
I'm compiling mplayer on the Linux playstation 2 dev-kit. The PS2 has a mips based cpu running in little endian mode and the actual configure failed to configure it properly.
I changed the configure script to autodetect the byte ordering. Actually, I don't know if the test works for all cpus so if people can test it on different big endian and little endian cpus I will appreciate.
Bertrand
On Wed, Jul 17, 2002 at 05:42:00PM -0700, Bertrand Baudet wrote:
On Monday 15 July 2002 11:24 pm, Bertrand Baudet wrote:
I'm compiling mplayer on the Linux playstation 2 dev-kit. The PS2 has a mips based cpu running in little endian mode and the actual configure failed to configure it properly.
I changed the configure script to autodetect the byte ordering. Actually, I don't know if the test works for all cpus so if people can test it on different big endian and little endian cpus I will appreciate.
Bertrand Nobody had a chance to look at this? Any comments are welcomed!
Bertrand
I tried it out, but all I have is an AMD box. So I don't know if it works, but at least nothing broke. --Joey -- "Living in the complex world of the future is somewhat like having bees live in your head. But, there they are."
Nobody had a chance to look at this? Any comments are welcomed! it works on my x86 system, I think we should probably push it into cvs and we can still revert (or find different detection) if it happens not to work eg. on 64bit cpus. Btw. in ffmpeg I saw this check, what do you think?
On Thursday 18 July 2002 02:42, Bertrand Baudet wrote: + +# --- +# big/little endian test +cat > $TMPC << EOF +#include <inttypes.h> +int main(int argc, char ** argv){ + volatile uint32_t i=0x01234567; + return (*((uint8_t*)(&i))) == 0x67; +} +EOF + +if $cc -o $TMPO $TMPC 2>/dev/null ; then +$TMPO && bigendian="yes" +else +echo big/little test failed +fi
Bertrand PS: see previous message fot the patch.
On Monday 15 July 2002 11:24 pm, Bertrand Baudet wrote:
I'm compiling mplayer on the Linux playstation 2 dev-kit. The PS2 has a mips based cpu running in little endian mode and the actual configure failed to configure it properly.
I changed the configure script to autodetect the byte ordering. Actually, I don't know if the test works for all cpus so if people can test it on different big endian and little endian cpus I will appreciate.
Bertrand
-- Best Regards, Atmos ____________________________________________ - MPlayer Developer - http://mplayerhq.hu/ - ____________________________________________
On Tuesday 30 July 2002 04:35 pm, Felix Buenemann wrote:
On Thursday 18 July 2002 02:42, Bertrand Baudet wrote:
Nobody had a chance to look at this? Any comments are welcomed!
it works on my x86 system, I think we should probably push it into cvs and we can still revert (or find different detection) if it happens not to work eg. on 64bit cpus. Btw. in ffmpeg I saw this check, what do you think? + +# --- +# big/little endian test +cat > $TMPC << EOF +#include <inttypes.h> +int main(int argc, char ** argv){ + volatile uint32_t i=0x01234567; + return (*((uint8_t*)(&i))) == 0x67; +} +EOF + +if $cc -o $TMPO $TMPC 2>/dev/null ; then +$TMPO && bigendian="yes" +else +echo big/little test failed +fi
Yes, the C code kinda do the same test... I'm gonna use their test, so if it fail we can blame ffmpeg :p I redone my patch since it was broken. I tested it on x86 and PPC, little endian MIPS and it worked. I'm waiting for the go ahead to commit it. Bertrand
If the patch looks ok for you, please commit it, I'm going in vacation and will be away for a least a week. Bertrand On Wednesday 31 July 2002 12:07 am, Bertrand Baudet wrote:
On Tuesday 30 July 2002 04:35 pm, Felix Buenemann wrote:
On Thursday 18 July 2002 02:42, Bertrand Baudet wrote:
Nobody had a chance to look at this? Any comments are welcomed!
it works on my x86 system, I think we should probably push it into cvs and we can still revert (or find different detection) if it happens not to work eg. on 64bit cpus. Btw. in ffmpeg I saw this check, what do you think? + +# --- +# big/little endian test +cat > $TMPC << EOF +#include <inttypes.h> +int main(int argc, char ** argv){ + volatile uint32_t i=0x01234567; + return (*((uint8_t*)(&i))) == 0x67; +} +EOF + +if $cc -o $TMPO $TMPC 2>/dev/null ; then +$TMPO && bigendian="yes" +else +echo big/little test failed +fi
Yes, the C code kinda do the same test... I'm gonna use their test, so if it fail we can blame ffmpeg :p
I redone my patch since it was broken. I tested it on x86 and PPC, little endian MIPS and it worked.
I'm waiting for the go ahead to commit it.
Bertrand
On Friday 02 August 2002 09:41, Bertrand Baudet wrote:
If the patch looks ok for you, please commit it, I'm going in vacation and will be away for a least a week. oh, sorry I've overseen your mail, I'll be away until sunday or monday, then I'll look at it.
Bertrand
On Wednesday 31 July 2002 12:07 am, Bertrand Baudet wrote:
On Tuesday 30 July 2002 04:35 pm, Felix Buenemann wrote:
On Thursday 18 July 2002 02:42, Bertrand Baudet wrote:
Nobody had a chance to look at this? Any comments are welcomed!
it works on my x86 system, I think we should probably push it into cvs and we can still revert (or find different detection) if it happens not to work eg. on 64bit cpus. Btw. in ffmpeg I saw this check, what do you think? + +# --- +# big/little endian test +cat > $TMPC << EOF +#include <inttypes.h> +int main(int argc, char ** argv){ + volatile uint32_t i=0x01234567; + return (*((uint8_t*)(&i))) == 0x67; +} +EOF + +if $cc -o $TMPO $TMPC 2>/dev/null ; then +$TMPO && bigendian="yes" +else +echo big/little test failed +fi
Yes, the C code kinda do the same test... I'm gonna use their test, so if it fail we can blame ffmpeg :p
I redone my patch since it was broken. I tested it on x86 and PPC, little endian MIPS and it worked.
I'm waiting for the go ahead to commit it.
Bertrand
_______________________________________________ MPlayer-dev-eng mailing list MPlayer-dev-eng@mplayerhq.hu http://mplayerhq.hu/mailman/listinfo/mplayer-dev-eng
-- Best Regards, Atmos ____________________________________________ - MPlayer Developer - http://mplayerhq.hu/ - ____________________________________________
participants (3)
-
Bertrand Baudet -
Felix Buenemann -
Joey Parrish