On 01/13/2014 03:56 PM, Kenneth Graunke wrote: > The existing 'offset' field is unfortunately typed as 'unsigned long', > which is unfortunately only 4 bytes with a 32-bit userspace. > > Traditionally, the hardware has only supported 32-bit virtual addresses, > so even though the kernel uses a __u64, the value would always fit. > > However, Broadwell supports 48-bit addressing. So with a 64-bit kernel, > the card virtual address may be too large to fit in the 'offset' field. > > Ideally, we would change the type of 'offset' to be a uint64_t---but > this would break the libdrm ABI. Instead, we create a new 'offset64' > field to hold the full 64-bit value from the kernel, and store the > 32-bit truncation in the existing 'offset' field, for compatibility. > > Cc: Eric Anholt <e...@anholt.net> > Cc: Daniel Vetter <daniel.vet...@ffwll.ch> > Cc: Ben Widawsky <b...@bwidawsk.net> > Signed-off-by: Kenneth Graunke <kenn...@whitecape.org> > --- > intel/intel_bufmgr.h | 12 +++++++++--- > intel/intel_bufmgr_gem.c | 16 ++++++++++------ > 2 files changed, 19 insertions(+), 9 deletions(-) > > I didn't update the bufmgr_fake stuff. Do I need to...?
Here's the testing I've done: - Piglit with old libdrm and old Mesa - Piglit with libdrm patch, but old Mesa (proves offset still works) - Piglit with libdrm patch and Mesa patches (proves offset64 also works) I have not tested the Broadwell patch because the kernel doesn't actually ever put objects > 4G today. _______________________________________________ mesa-dev mailing list mesa-dev@lists.freedesktop.org http://lists.freedesktop.org/mailman/listinfo/mesa-dev
