2339fc779c
Updated obsolete comment.
4 years ago
b393849bbd
Updated obsolete comment.
4 years ago
ebe53f8590
Fixed compile error.
4 years ago
b37f82b58b
Fixed compile error.
4 years ago
71749b7caf
Implemented inline array allocation, and moved type->lg2 map to reflection.
4 years ago
9557b97acc
Implemented inline array allocation, and moved type->lg2 map to reflection.
4 years ago
b58d2a0ee6
Shrink overhead of message representation.
4 years ago
0bf063a2ca
Shrink overhead of message representation.
4 years ago
d87ceeacab
Shave off one more store.
4 years ago
ddc52ab9d6
Shave off one more store.
4 years ago
c25d895adf
Shrunk the arena state that needs to be synced.
4 years ago
7f67f68c1c
Shrunk the arena state that needs to be synced.
4 years ago
ff40dd6ea9
Added new internal header.
4 years ago
85a43e5461
Added new internal header.
4 years ago
36662b3735
Refactor some code. I extracted some common code from all message field
...
parsers, to a tail recursive function. Removed the varint jmp table for
a simple varint parse loop, that removes the stack frames. Also careful
with not losing information in repeated message tag check. When written
mindful the checks and loads that happen can be reused for tag dispatch
if not the expected tag.
4 years ago
cbcd635917
Fixed memory leak.
4 years ago
bcbcdadbd2
Fixed memory leak.
4 years ago
746f64692c
Moved arena inline for decoder.
4 years ago
7363b91ac3
Moved arena inline for decoder.
4 years ago
b8ef1dcc57
Removed C++-style comments.
4 years ago
575acd85bd
Re-added const for all of the pointer wrapper types.
4 years ago
5aa5b77b41
Added simple offset-based accessors for defs, and deprecated old iterators.
4 years ago
9938cf8f27
Put submsg_index directly in table data. Drop oneof support for now to focus.
4 years ago
d87179501d
Another build fix.
4 years ago
89bd8b87e1
Fixed a few more C89 compat issues.
4 years ago
64d293894a
Fixed bug introduced by last optimization.
4 years ago
ff957b996c
Fixed C89 compat issues.
4 years ago
537b6f42c2
A few updates to the benchamrk and minor implementation changes.
4 years ago
0dcc5641eb
Replicated dispatch and implemeted array resizing logic. Up to 2.67GB/s.
4 years ago
526e430794
I think this may have reached the optimization limit.
...
-------------------------------------------------------------------------
Benchmark Time CPU Iterations
-------------------------------------------------------------------------
BM_ArenaOneAlloc 21 ns 21 ns 32994231
BM_ArenaInitialBlockOneAlloc 6 ns 6 ns 116318005
BM_ParseDescriptorNoHeap 3028 ns 3028 ns 231138 2.34354GB/s
BM_ParseDescriptor 3557 ns 3557 ns 196583 1.99498GB/s
BM_ParseDescriptorProto2NoArena 33228 ns 33226 ns 21196 218.688MB/s
BM_ParseDescriptorProto2WithArena 22863 ns 22861 ns 30666 317.831MB/s
BM_SerializeDescriptorProto2 5444 ns 5444 ns 127368 1.30348GB/s
BM_SerializeDescriptor 12509 ns 12508 ns 55816 580.914MB/s
$ perf stat bazel-bin/benchmark --benchmark_filter=BM_ParseDescriptorNoHeap
2020-10-08 14:07:06
Running bazel-bin/benchmark
Run on (72 X 3700 MHz CPU s)
CPU Caches:
L1 Data 32K (x36)
L1 Instruction 32K (x36)
L2 Unified 1024K (x36)
L3 Unified 25344K (x2)
----------------------------------------------------------------
Benchmark Time CPU Iterations
----------------------------------------------------------------
BM_ParseDescriptorNoHeap 3071 ns 3071 ns 227743 2.31094GB/s
Performance counter stats for 'bazel-bin/benchmark --benchmark_filter=BM_ParseDescriptorNoHeap':
1,050.22 msec task-clock # 0.978 CPUs utilized
4 context-switches # 0.004 K/sec
0 cpu-migrations # 0.000 K/sec
179 page-faults # 0.170 K/sec
3,875,796,334 cycles # 3.690 GHz
13,282,835,967 instructions # 3.43 insn per cycle
2,887,725,848 branches # 2749.627 M/sec
8,324,912 branch-misses # 0.29% of all branches
1.073924364 seconds time elapsed
1.042806000 seconds user
0.008021000 seconds sys
Profile:
23.96% benchmark benchmark [.] upb_prm_1bt_max192b
22.44% benchmark benchmark [.] fastdecode_dispatch
18.96% benchmark benchmark [.] upb_pss_1bt
14.20% benchmark benchmark [.] upb_psv4_1bt
8.33% benchmark benchmark [.] upb_prm_1bt_max64b
6.66% benchmark benchmark [.] upb_prm_1bt_max128b
1.29% benchmark benchmark [.] upb_psm_1bt_max64b
0.77% benchmark benchmark [.] fastdecode_generic
0.55% benchmark [kernel.kallsyms] [k] smp_call_function_single
0.42% benchmark [kernel.kallsyms] [k] _raw_spin_lock_irqsave
0.42% benchmark benchmark [.] upb_psm_1bt_max256b
0.31% benchmark benchmark [.] upb_psb1_1bt
0.21% benchmark benchmark [.] upb_plv4_5bv
0.14% benchmark benchmark [.] upb_psb1_2bt
0.12% benchmark benchmark [.] decode_longvarint64
0.08% benchmark [kernel.kallsyms] [k] vsnprintf
0.07% benchmark [kernel.kallsyms] [k] _raw_spin_lock
0.07% benchmark benchmark [.] _upb_msg_new
0.06% benchmark ld-2.31.so [.] check_match
4 years ago
4c65b25daf
Handle long varints, now 2GB/s!
4 years ago
e39ec95ca2
Hoisted updates to limits and depth out of the loop.
4 years ago
388b6f64eb
A small optimization: don't increment array length every iteration.
4 years ago
9e5c5ce089
Optimized memset() with cutoff and fixed group & unknown message bugs.
4 years ago
8dd7b5a2ca
A bunch more optimization.
4 years ago
405e7934b1
Handle 2-byte submessage lengths.
4 years ago
88b1ec7784
Table-driven supports repeated sub-messages.
4 years ago
f173642db4
Handle non-repeated submessages.
4 years ago
7ec2c52346
Donate/steal from arena to accelerate decoding.
4 years ago
fac992db83
Cleanup for showing.
4 years ago
3937874a85
We have a properly structured algorithm, but perf regresses by 20%.
4 years ago
438ecaeb5a
Give all field parsers a generic table entry.
4 years ago
383ae5293e
WIP.
4 years ago
26abaa2345
WIP.
4 years ago
34b98bc030
Avoid passing too many params to fallback.
4 years ago
763a3f6293
WIP.
4 years ago
a202ce9629
Add UPB_FORCEINLINE for varint32 decoding.
...
This speeds up the decoder by >20% and also reduces code size slightly!
name old time/op new time/op delta
ArenaOneAlloc 20.4ns ± 0% 20.2ns ± 0% -1.10% (p=0.000 n=12+11)
ArenaInitialBlockOneAlloc 5.25ns ± 0% 5.25ns ± 0% ~ (p=0.786 n=11+12)
ParseDescriptorNoHeap 17.1µs ± 0% 13.1µs ± 0% -23.29% (p=0.000 n=11+12)
ParseDescriptor 17.4µs ± 1% 13.5µs ± 1% -22.51% (p=0.000 n=12+12)
SerializeDescriptor 10.7µs ± 0% 10.9µs ± 0% +1.95% (p=0.000 n=12+12)
FILE SIZE VM SIZE
-------------- --------------
+2.7% +16 +2.7% +16 [LOAD #2 [RX]]
+0.5% +16 [ = ] 0 [Unmapped]
-1.4% -72 -0.7% -32 upb/decode.c
+3.1% +98 +3.1% +98 decode_msg
[DEL] -170 [DEL] -130 decode_varint32
-0.0% -40 -0.0% -16 TOTAL
4 years ago
5741eb9ad7
Expanded benchmarking script and added one size opt to the encoder.
4 years ago
0135399e60
Fixed bug introduced in refactoring.
4 years ago
df3438222b
Notated impossible branch as unreachable.
4 years ago
Joshua Haberman
Gerben Stavenga