3350add5cc
Gen8+ supports 48-bit virtual addresses, but some objects must always be
allocated inside the 32-bit address range.
In specific, any resource used with flat/heapless (0x00000000-0xfffff000)
General State Heap (GSH) or Instruction State Heap (ISH) must be in a
32-bit range, because the General State Offset and Instruction State Offset
are limited to 32-bits.
The i915 driver has been modified to provide a flag to set when the 4GB
limit is not necessary in a given bo (EXEC_OBJECT_SUPPORTS_48B_ADDRESS).
48-bit range will only be used when explicitly requested.
Callers to the existing drm_intel_bo_emit_reloc function should set the
use_48b_address_range flag beforehand, in order to use full ppgtt range.
v2: Make set/clear functions nops on pre-gen8 platforms, and use them
internally in emit_reloc functions (Ben)
s/48BADDRESS/48B_ADDRESS/ (Dave)
v3: Keep set/clear functions internal, no-one needs to use them directly.
v4: Don't set 48bit-support flag in emit reloc, check for ppgtt type
before enabling set/clear function, print full offsets in debug
statements, using port of lower_32_bits and upper_32_bits from linux
kernel (Michał)
References: http://lists.freedesktop.org/archives/intel-gfx/2015-July/072612.html
Cc: Ben Widawsky <ben@bwidawsk.net>
Cc: Michał Winiarski <michal.winiarski@intel.com>
Signed-off-by: Michel Thierry <michel.thierry@intel.com>
Reviewed-by: Kristian Høgsberg Kristensen <kristian.h.kristensen@intel.com>
Signed-off-by: Kristian Høgsberg Kristensen <kristian.h.kristensen@intel.com>
|
||
|---|---|---|
| amdgpu | ||
| exynos | ||
| freedreno | ||
| include/drm | ||
| intel | ||
| libkms | ||
| m4 | ||
| man | ||
| nouveau | ||
| omap | ||
| radeon | ||
| tegra | ||
| tests | ||
| .gitignore | ||
| Android.mk | ||
| CleanSpec.mk | ||
| Makefile.am | ||
| Makefile.sources | ||
| README | ||
| RELEASING | ||
| autogen.sh | ||
| configure.ac | ||
| libdrm.pc.in | ||
| libdrm_lists.h | ||
| libdrm_macros.h | ||
| util_double_list.h | ||
| util_math.h | ||
| xf86atomic.h | ||
| xf86drm.c | ||
| xf86drm.h | ||
| xf86drmHash.c | ||
| xf86drmHash.h | ||
| xf86drmMode.c | ||
| xf86drmMode.h | ||
| xf86drmRandom.c | ||
| xf86drmRandom.h | ||
| xf86drmSL.c | ||
README
libdrm - userspace library for drm This is libdrm, a userspace library for accessing the DRM, direct rendering manager, on Linux, BSD and other operating systes that support the ioctl interface. The library provides wrapper functions for the ioctls to avoid exposing the kernel interface directly, and for chipsets with drm memory manager, support for tracking relocations and buffers. libdrm is a low-level library, typically used by graphics drivers such as the Mesa DRI drivers, the X drivers, libva and similar projects. New functionality in the kernel DRM drivers typically requires a new libdrm, but a new libdrm will always work with an older kernel. Compiling --------- libdrm is a standard autotools packages and follows the normal configure, build and install steps. The first step is to configure the package, which is done by running the configure shell script: ./configure By default, libdrm will install into the /usr/local/ prefix. If you want to install this DRM to replace your system copy, pass --prefix=/usr and --exec-prefix=/ to configure. If you are building libdrm from a git checkout, you first need to run the autogen.sh script. You can pass any options to autogen.sh that you would other wise pass to configure, or you can just re-run configure with the options you need once autogen.sh finishes. Next step is to build libdrm: make and once make finishes successfully, install the package using make install If you are install into a system location, you will need to be root to perform the install step.