[sane-devel] genesys endianess

2006-03-14 Thread Stéphane VOLTZ
Hello, I tested it and has expected it works. So I removed the idefed part and only keep your code. regards, Stef

[sane-devel] genesys endianess

2006-03-12 Thread Stéphane VOLTZ
Hello, I had to put an ifdef WORDS_BIGENDIAN (like for slope tables), so that my MD6471 could work again after the endianess changes you done. Regards, Stef

[sane-devel] genesys endianess

2006-03-12 Thread Pierre Willenbrock
St?phane VOLTZ schrieb: I had to put an ifdef WORDS_BIGENDIAN (like for slope tables), so that my MD6471 could work again after the endianess changes you done. To be exact, the code i used for gamma was meant to work for both endianesses, but does not work for either.. I forgot to

[sane-devel] genesys endianess

2006-03-08 Thread Jon Chambers
Hi Pierre, On Tue, 7 Mar 2006, Pierre Willenbrock wrote: [...] I am planning to add a function that swaps the bytes in place if needed which should be called on the pixel data. You may find the byte ordering calls built into the socket library (htons, htonl, etc) useful. cheers, Jon

[sane-devel] genesys endianess

2006-03-08 Thread Pierre Willenbrock
Hi Jon, Jon Chambers schrieb: On Tue, 7 Mar 2006, Pierre Willenbrock wrote: [...] I am planning to add a function that swaps the bytes in place if needed which should be called on the pixel data. You may find the byte ordering calls built into the socket library (htons, htonl, etc) useful.

[sane-devel] genesys endianess

2006-03-07 Thread Pierre Willenbrock
Hi all, i am planning to complete the endian conversion code in the genesys backend. I identified the following fields where endian conversion is neccessary: Going down to scanner: * slope tables * shading tables * gamma tables The registers need to be addressed bytewise anyways, so they don't