Am 21.04.2017 um 13:49 schrieb Ville Syrjälä: > On Fri, Apr 21, 2017 at 02:40:18PM +0300, Pekka Paalanen wrote: >> On Fri, 21 Apr 2017 14:08:04 +0300 >> Ville Syrjälä <ville.syrjala at linux.intel.com> wrote: >> >>> On Fri, Apr 21, 2017 at 11:50:18AM +0200, Gerd Hoffmann wrote: >>>> On Fr, 2017-04-21 at 12:25 +0300, Ville Syrjälä wrote: >>>>> On Fri, Apr 21, 2017 at 09:58:24AM +0200, Gerd Hoffmann wrote: >>>>>> While working on graphics support for virtual machines on ppc64 (which >>>>>> exists in both little and big endian variants) I've figured the comments >>>>>> for various drm fourcc formats in the header file don't match reality. >>>>>> >>>>>> Comments says the RGB formats are little endian, but in practice they >>>>>> are native endian. Look at the drm_mode_legacy_fb_format() helper. It >>>>>> maps -- for example -- bpp/depth 32/24 to DRM_FORMAT_XRGB8888, no matter >>>>>> whenever the machine is little endian or big endian. The users of this >>>>>> function (fbdev emulation, DRM_IOCTL_MODE_ADDFB) expect the framebuffer >>>>>> is native endian, not little endian. Most userspace also operates on >>>>>> native endian only. >>>>> I'm not a fan of "native". Native to what? "CPU" or "host" is what I'd >>>>> call it. >>>> native == whatever the cpu is using. >>>> >>>> I personally find "native" more intuitive, but at the end of the day I >>>> don't mind much. If people prefer "host" over "native" I'll change it. >>> "native" to me feels more like "native to the GPU" since these things >>> really are tied to the GPU not the CPU. That's also why I went with the >>> explicit endianness originally so that the driver could properly declare >>> what the GPU supports. >> Hi, >> >> yeah, one should really be explicit on which component's endianess does >> "native" refer to. I just can't imagine "GPU native" to ever be an >> option, because then userspace needs a way to discover what the >> GPU endianess is, > It has to know that. How else would it know how to write the bytes into > memory in the right order for the GPU to consume, or read the stuff the > GPU produced? > >> and I believe that would only deepen the swamp, not >> drain it, because suddenly you need two enums to describe one format. >> >> Ville, wording aside, what do think about changing the endianess >> definition? Is it going in the right direction? > I don't think so, but I guess I'm in the minority. I don't think your are in the minority. At least I would clearly say those formats should be in a fixed byte order and don't care about the CPU in the system. What I need from the driver side is a consistent description of how the bytes in memory map to my hardware. What CPU is in use in the system is completely irrelevant for that. Regards, Christian.
