linux/arch/powerpc
Mark Hairgrove 3689c37d23 powerpc/powernv/npu: Use size-based ATSD invalidates
Prior to this change only two types of ATSDs were issued to the NPU:
invalidates targeting a single page and invalidates targeting the whole
address space. The crossover point happened at the configurable
atsd_threshold which defaulted to 2M. Invalidates that size or smaller
would issue per-page invalidates for the whole range.

The NPU supports more invalidation sizes however: 64K, 2M, 1G, and all.
These invalidates target addresses aligned to their size. 2M is a common
invalidation size for GPU-enabled applications because that is a GPU
page size, so reducing the number of invalidates by 32x in that case is a
clear improvement.

ATSD latency is high in general so now we always issue a single invalidate
rather than multiple. This will over-invalidate in some cases, but for any
invalidation size over 2M it matches or improves the prior behavior.
There's also an improvement for single-page invalidates since the prior
version issued two invalidates for that case instead of one.

With this change all issued ATSDs now perform a flush, so the flush
parameter has been removed from all the helpers.

To show the benefit here are some performance numbers from a
microbenchmark which creates a 1G allocation then uses mprotect with
PROT_NONE to trigger invalidates in strides across the allocation.

One NPU (1 GPU):

         mprotect rate (GB/s)
Stride   Before      After      Speedup
64K         5.3        5.6           5%
1M         39.3       57.4          46%
2M         49.7       82.6          66%
4M        286.6      285.7           0%

Two NPUs (6 GPUs):

         mprotect rate (GB/s)
Stride   Before      After      Speedup
64K         6.5        7.4          13%
1M         33.4       67.9         103%
2M         38.7       93.1         141%
4M        356.7      354.6          -1%

Anything over 2M is roughly the same as before since both cases issue a
single ATSD.

Signed-off-by: Mark Hairgrove <mhairgrove@nvidia.com>
Reviewed-By: Alistair Popple <alistair@popple.id.au>
Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2018-10-04 16:55:53 +10:00
..
boot powerpc/boot: Ensure _zimage_start is a weak symbol 2018-09-19 22:08:12 +10:00
configs powerpc/config: Enable CONFIG_PRINTK_TIME 2018-10-03 15:40:06 +10:00
crypto powerpc updates for 4.19 2018-08-17 11:32:50 -07:00
include powerpc/tm: Remove msr_tm_active() 2018-10-03 15:40:05 +10:00
kernel powerpc: Wire up memtest 2018-10-03 16:12:47 +10:00
kvm PPC KVM fixes for 4.19 2018-09-04 21:12:46 +02:00
lib powerpc/64: Remove static branch hints from memset() 2018-09-17 21:17:25 +10:00
math-emu
mm powerpc/nohash: fix undefined behaviour when testing page size support 2018-10-04 16:54:41 +10:00
net powerpc updates for 4.19 2018-08-17 11:32:50 -07:00
oprofile treewide: kzalloc() -> kcalloc() 2018-06-12 16:19:22 -07:00
perf powerpc/perf: Add missing break in power7_marked_instr_event() 2018-10-03 15:39:45 +10:00
platforms powerpc/powernv/npu: Use size-based ATSD invalidates 2018-10-04 16:55:53 +10:00
purgatory powerpc updates for 4.19 2018-08-17 11:32:50 -07:00
sysdev powerpc/xive: Move a dereference below a NULL test 2018-10-03 15:40:02 +10:00
tools powerpc: Add a checkpatch wrapper with our preferred settings 2018-08-07 21:49:25 +10:00
xmon Revert "convert SLB miss handlers to C" and subsequent commits 2018-10-03 15:32:49 +10:00
Kconfig powerpc/64: add stack protector support 2018-10-03 15:40:03 +10:00
Kconfig.debug Kconfig: consolidate the "Kernel hacking" menu 2018-08-02 08:06:48 +09:00
Makefile powerpc: remove leftover code of old GCC version checks 2018-10-04 16:54:51 +10:00
Makefile.postlink