diff options
author | John Hubbard <jhubbard@nvidia.com> | 2019-04-21 17:09:24 -0700 |
---|---|---|
committer | John Hubbard <jhubbard@nvidia.com> | 2019-04-24 11:13:31 -0700 |
commit | 60b67c31fcad6d0dc9603a034994a12efe3d5202 (patch) | |
tree | c71d32e4b2a0d2a9770ac6b9add6ed49b0320e51 /pascal/gp100-fbpa.txt | |
parent | d1be635fac4adad238916d28dde406c81ba564a8 (diff) | |
download | open-gpu-doc-60b67c31fcad6d0dc9603a034994a12efe3d5202.tar.xz |
Open doc files, version 1 of everything
This copies over files from:
http://download.nvidia.com/open-gpu-doc/
with a few overrides from local, as-yet-to-be-published
files (generally just a file suffix and licensing tweak).
Here's the script that was used to create this commit:
wget -r http://download.nvidia.com/open-gpu-doc/
mv download.nvidia.com/open-gpu-doc/* .
rm -rf download.nvidia.com/
find . -name index.html | xargs rm -f
find . -name Thumbs.db | xargs rm -f
find . -type f | grep -v \.git | xargs file | grep CRLF | cut -f1 -d: | xargs dos2unix
for f in `find . -name 1 | cut -f2 -d "/"`; do pushd $f; mv 1/* .; popd; done
find . -name 1 | xargs rmdir
rm -rf Host-Fifo/volta/gv100
mkdir -p Host-Fifo/volta/gv100
cp $sw/docs/Public_Devel_Docs/release/Host-Fifo/volta/gv100/* Host-Fifo/volta/gv100/
rm Display-Ref-Manuals/gv100/dev_display.ref
cp $sw/docs/Public_Devel_Docs/release/Display-Ref-Manuals/1/gv100/dev_display.ref \
Display-Ref-Manuals/gv100/dev_display.ref.txt
git add .
git mv DCB/DCB-4.0-Specification.html DCB/DCB-4.x-Specification.html
git reset HEAD Display-Class-Methods/2
git reset HEAD DCB/2
git commit
Reviewed-by: Andy Ritger
Diffstat (limited to 'pascal/gp100-fbpa.txt')
-rw-r--r-- | pascal/gp100-fbpa.txt | 92 |
1 files changed, 92 insertions, 0 deletions
diff --git a/pascal/gp100-fbpa.txt b/pascal/gp100-fbpa.txt new file mode 100644 index 0000000..6258f06 --- /dev/null +++ b/pascal/gp100-fbpa.txt @@ -0,0 +1,92 @@ +-------------------------------------------------------------------------------- + +Synopsis: + +Description of changes made to the framebuffer partition addressing (FBPA) in +Pascal and later NVIDIA architectures. + +-------------------------------------------------------------------------------- + +Description: + +NVIDIA moved and expanded the MMIO space used for accessing the per-partition +information on Pascal and later architectures. Relative offsets to specific +controls remain the same as for prior architectures, for the most part. + +-------------------------------------------------------------------------------- + +Summary: + +These MMIO ranges have been moved and expanded from 0x1000 to 0x4000 in size: + +Name Old Range New Range +NV_PFB_FBPA 0x10F000 (0x1000) 0x9A0000 (0x4000) +NV_PFB_FBPA[i] 0x110000+(i * 0x1000) 0x900000+(i * 0x4000) +NV_PFB_FBPA_MC[i] 0x11D000+(i * 0x1000) 0x980000+(i * 0x4000) + +The number of NV_PFB_FBPA[i] ranges is increased to a maximum of 16. +The number of NV_PFB_FBPA_MC[i] ranges remains 3. + +Memory partition sizing and programming is the same as in prior NVIDIA +architectures. + +FBPAs are grouped into logical FBP units. In most prior NVIDIA architectures +(except GF108) each logical FBP mapped to one FBPA. To support High Bandwidth +Memory (HBM) GP100 groups 2 FBPAs into each logical FBP. This is noted here to +be clear that the number of logical FBPs does not necessarily equate to the number +of physical FBPAs. The register NV_PTOP_SCAL_NUM_FBPA_PER_FBP (0x22458) defines +this relationship in Pascal and later architectures. + +Per-partition memory size detection works similarly to the way it has in prior +NVIDIA architectures: + +1) Determine maximum number of possible FBPAs by reading + NV_PTOP_SCAL_NUM_FBPAS (0x2243C) +2) Determine number of FBPAs per FBP by reading + NV_PTOP_SCAL_NUM_FBPA_PER_FBP (0x22458) +3) Determine maximum number of LTCs per FBP by reading + NV_FUSE_STATUS_OPT_ROP_L2_FBP(i) (0x21d70+(i)*4) +4) For each bit not set in NV_FUSE_STATUS_OPT_FBIO (0x21C14) + a. Read the partition memory size from + NV_PFB_FBPA[i] + _CSTATUS_RAMAMOUNT (0x20C) + b. Up to the number of possible FBPAs determined in #1 +5) Any difference in the per-partition memory size indicates a "mixed memory" + configuration (Fermi & Kepler). +6) Any difference in the per-partition LTC coverage effectively indicates a + "mixed memory" configuration (Maxwell and later). +7) For "mixed memory" configurations: + a. Set NV_PFB_FBHUB_NUM_ACTIVE_FBPS (0x100800) bit 4 to 1. + b. Treat GPU FB address space as split into lower and upper sections. + The lower size is common partition size * FBPA count and is + based at 0. The upper section starts at either 0x2'00000000 + (Fermi/Kepler) or 0x10'00000000 (Maxwell & later) PLUS the + common partition size. Its size is the remaining GPU FB not + already mapped in the lower section. + c. The upper section of memory should not be used for displayable or + compression-related surfaces. + +-------------------------------------------------------------------------------- + +Definitions: + +#define NV_PTOP_SCAL_NUM_FBPAS 0x0002243C /* R--4R */ + +#define NV_PTOP_SCAL_NUM_FBPAS_VALUE 4:0 /* R-IVF */ + + +#define NV_PTOP_SCAL_NUM_FBPA_PER_FBP 0x00022458 /* R--4R */ + +#define NV_PTOP_SCAL_NUM_FBPA_PER_FBP_VALUE 4:0 /* R-IVF */ + + +#define NV_FUSE_STATUS_OPT_ROP_L2_FBP(i) (0x00021d70+(i)*4) /* R-I4A */ +#define NV_FUSE_STATUS_OPT_ROP_L2_FBP__SIZE_1 16 /* */ +#define NV_FUSE_STATUS_OPT_ROP_L2_FBP_DATA 31:0 /* R-IVF */ + +#define NV_FUSE_STATUS_OPT_FBIO 0x00021C14 /* R-I4R */ +#define NV_FUSE_STATUS_OPT_FBIO_DATA 15:0 /* R-IVF */ + +#define NV_PFB_FBHUB_NUM_ACTIVE_FBPS 0x00100800 /* RW-4R */ +#define NV_PFB_FBHUB_NUM_ACTIVE_FBPS_MIXED_MEM_DENSITY 4:4 /* */ + +-------------------------------------------------------------------------------- |