 andrew
|
3898cb596a
|
Use _mm512_loadu_epi32 instead of memcpy
The latter somehow generated a vmovdqa64 instruction that segfaulted on
unaligned memory, with gcc and --coverage. Hmm.
|
2024-08-01 15:24:09 -07:00 |
|
|
andrew
|
b8edd92698
|
Fix warning
|
2024-08-01 15:23:43 -07:00 |
|
|
andrew
|
8e480528d5
|
Update corpus
|
2024-08-01 14:56:36 -07:00 |
|
|
andrew
|
4113183155
|
Attempt to fix likely arm bug
|
2024-08-01 14:01:41 -07:00 |
|
|
andrew
|
adb8fdc5e9
|
Simplify nextPhysical
|
2024-08-01 14:00:02 -07:00 |
|
|
andrew
|
c86e407985
|
Return Node from getChildGeq
It seems all callers ultimately want this
|
2024-08-01 13:53:18 -07:00 |
|
|
andrew
|
71a84057cb
|
Find two more call sites for getFirstChildExists
|
2024-08-01 13:37:44 -07:00 |
|
|
andrew
|
9c5e5863c2
|
Simplify remaining "down left spine" loops
|
2024-08-01 13:17:33 -07:00 |
|
|
andrew
|
be67555756
|
Simplify more "down left spine" loops
|
2024-08-01 13:13:55 -07:00 |
|
|
andrew
|
988ec5ce69
|
Add getFirstChildExists
|
2024-08-01 13:05:40 -07:00 |
|
|
andrew
|
f5a0d81c52
|
Remove some redundant branches
I think they were getting optimized out, but still
|
2024-08-01 12:37:57 -07:00 |
|
|
andrew
|
3b2bd16cd1
|
Add overloads of getChild for each type
|
2024-08-01 11:42:55 -07:00 |
|
|
andrew
|
4b3df0a426
|
Avoid dispatching on node type twice in nextPhysical
|
2024-08-01 10:46:08 -07:00 |
|
|
andrew
|
4cdf6deb50
|
Remove a branch in happy path in addWrites
|
2024-08-01 10:28:50 -07:00 |
|
|
andrew
|
f21dde06d3
|
Group write metric accumulators into a struct
|
2024-07-31 21:45:20 -07:00 |
|
|
andrew
|
2b11650589
|
Update aarch64 import symbols
|
2024-07-31 17:34:30 -07:00 |
|
|
andrew
|
fce998460f
|
Use int64_t internally for metrics
So we can use fetch_add
|
2024-07-31 14:41:42 -07:00 |
|
|
andrew
|
6da3125719
|
Remove everything wasm-related
I'm not interested in supporting wasm at this time
|
2024-07-31 14:04:10 -07:00 |
|
|
andrew
|
79410d071f
|
Add accidentally-deleted corpus back
|
2024-07-26 14:57:04 -07:00 |
|
|
andrew
|
1fcca6450d
|
Fix point writes accounting
Previously it wouldn't count a singleton range write
|
2024-07-26 14:41:54 -07:00 |
|
|
andrew
|
55271ad06c
|
Update corpus
To match new "interleaving" test setup
|
2024-07-23 16:54:03 -07:00 |
|
|
andrew
|
e675612599
|
Suppress some warnings in test code
|
2024-07-23 16:50:02 -07:00 |
|
|
andrew
|
42b5d50492
|
Update DEBUG_VERBOSE for interleaving conflict sets in test
|
2024-07-23 15:44:29 -07:00 |
|
|
andrew
|
6394995def
|
Add a test where the "zero" bug causes an incorrect commit
|
2024-07-23 14:39:45 -07:00 |
|
|
andrew
|
c649bc7964
|
Interleave calls for two conflict sets in tests
|
2024-07-23 11:12:02 -07:00 |
|
|
andrew
|
ec85a06d01
|
Bump version
|
2024-07-22 17:25:38 -07:00 |
|
|
andrew
|
fb9f5ce6f4
|
Update InternalVersionT::zero in addWrite call
There can be other conflict sets in the same thread
v0.0.9
|
2024-07-22 16:14:25 -07:00 |
|
|
andrew
|
2b1c710953
|
Add noop getMetricsV1 to HashTable.cpp
|
2024-07-19 11:25:12 -07:00 |
|
|
andrew
|
ebf281220b
|
Add NOLINT for new clangd warning about sizeof pointer
|
2024-07-19 11:24:37 -07:00 |
|
|
andrew
|
6051b2fb2e
|
Remove benchMetrics
It's not all that interesting in the end
|
2024-07-19 11:24:01 -07:00 |
|
|
andrew
|
11c3ca6766
|
Add oldest_extant_version metric
|
2024-07-18 14:19:11 -07:00 |
|
|
andrew
|
b45dec2f1f
|
Add point_writes_total and range_writes_total
|
2024-07-18 14:00:57 -07:00 |
|
|
andrew
|
c5e9f18c47
|
Remove some unnecessary std::exchange's
|
2024-07-17 18:57:12 -07:00 |
|
|
andrew
|
cebbf89cbe
|
Fix build for x86
|
2024-07-17 18:47:12 -07:00 |
|
|
andrew
|
abb791d86b
|
Just use a normal && in a place where being branch-free isn't a big deal
|
2024-07-17 18:21:33 -07:00 |
|
|
andrew
|
12f361f33a
|
Don't plumb impl and ReadContext
|
2024-07-17 18:20:08 -07:00 |
|
|
andrew
|
640c1ca9dd
|
Fix issue with thread_local performance
|
2024-07-17 17:54:39 -07:00 |
|
|
andrew
|
b7d54d44e1
|
Add oldest version and newest version metrics
|
2024-07-17 16:42:46 -07:00 |
|
|
andrew
|
95596f831f
|
Add some metrics for addWrites and setOldestVersion
|
2024-07-17 16:35:29 -07:00 |
|
|
andrew
|
542371d562
|
Build TestDriver for perf testing too
|
2024-07-17 15:48:32 -07:00 |
|
|
andrew
|
958a4e2d0e
|
Strengthen metric name structure according to Prometheus guidance
|
2024-07-16 18:59:42 -07:00 |
|
|
andrew
|
8ce14c58a4
|
Print metrics for blackbox tests
|
2024-07-16 16:46:57 -07:00 |
|
|
andrew
|
56e847b63c
|
Experiment with CLOCK_THREAD_CPUTIME_ID
|
2024-07-15 21:34:22 -07:00 |
|
|
andrew
|
7fd1c9e140
|
Make range_read_iterations_total consistent with others
|
2024-07-15 17:23:48 -07:00 |
|
|
andrew
|
ebaac253e2
|
Don't count loop entry as an "iteration"
|
2024-07-15 14:46:57 -07:00 |
|
|
andrew
|
9b470a367c
|
Add check_bytes_total counter
|
2024-07-15 14:40:22 -07:00 |
|
|
andrew
|
e7806a36d1
|
Fix range_read_total counter
It was overcounting
|
2024-07-15 14:09:31 -07:00 |
|
|
andrew
|
ffd1dfe74d
|
Rework check metrics
{point,prefix,range} x {count,iterations,short_circuits} + node scans
|
2024-07-15 14:02:39 -07:00 |
|
|
andrew
|
c39af9117f
|
Remove nodes allocated/released metrics
|
2024-07-15 13:34:13 -07:00 |
|
|
andrew
|
ed274c24d7
|
Just output metrics in the order declared
|
2024-07-12 16:10:28 -07:00 |
|