Commit Graph

8 Commits

Author SHA1 Message Date
Martin Möhrmann 365594ad59 runtime: simplify memory capacity check in growslice
Instead of comparing if the number of elements will
not fit into memory check if the memory size of the
slices backing memory is higher then the memory limit.

This avoids a division or maxElems lookup.

With et.size > 0:
   uintptr(newcap)                > maxSliceCap(et.size)
-> uintptr(int(capmem / et.size)) > _MaxMem  /  et.size
->             capmem / et.size   > _MaxMem  /  et.size
->             capmem             > _MaxMem

Note that due to integer division from capmem > _MaxMem
it does not follow that uintptr(newcap) > maxSliceCap(et.size).

Consolidated runtime GrowSlice benchmarks by using sub-benchmarks and
added more struct sizes to show performance improvement when division
is avoided for element sizes larger than 32 bytes.

AMD64:
GrowSlice/Byte       38.9ns ± 2%  38.9ns ± 1%    ~     (p=0.974 n=20+20)
GrowSlice/Int        58.3ns ± 3%  58.0ns ± 2%    ~     (p=0.154 n=20+19)
GrowSlice/Ptr        95.7ns ± 2%  95.1ns ± 2%  -0.60%  (p=0.034 n=20+20)
GrowSlice/Struct/24  95.4ns ± 1%  93.9ns ± 1%  -1.54%  (p=0.000 n=19+19)
GrowSlice/Struct/32   110ns ± 1%   108ns ± 1%  -1.76%  (p=0.000 n=19+20)
GrowSlice/Struct/40   138ns ± 1%   128ns ± 1%  -7.09%  (p=0.000 n=20+20)

Change-Id: I1c37857c74ea809da373e668791caffb6a5cbbd3
Reviewed-on: https://go-review.googlesource.com/53471
Run-TryBot: Martin Möhrmann <moehrmann@google.com>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
2017-08-15 04:23:11 +00:00
Carl Mastrangelo c1e267cc73 runtime: make append only clear uncopied memory
Also add a benchmark that shows off the new behavior.  The
existing benchmarks reuse the same slice, and thus don't ever have
to clear memory.  Running the Append|Grow benchmarks in runtime:

name                              old time/op  new time/op  delta
AppendSliceLarge/1024Bytes-12      265ns ± 1%   265ns ± 3%     ~     (p=0.524 n=17+20)
AppendSliceLarge/4096Bytes-12      807ns ± 3%   772ns ± 1%   -4.38%  (p=0.000 n=20+20)
AppendSliceLarge/16384Bytes-12    3.20µs ± 4%  2.82µs ± 4%  -11.93%  (p=0.000 n=19+20)
AppendSliceLarge/65536Bytes-12    13.0µs ± 4%  11.0µs ± 3%  -15.22%  (p=0.000 n=20+20)
AppendSliceLarge/262144Bytes-12   62.7µs ± 1%  51.6µs ± 1%  -17.67%  (p=0.000 n=19+20)
AppendSliceLarge/1048576Bytes-12   337µs ± 3%   289µs ± 3%  -14.36%  (p=0.000 n=20+20)
GrowSliceBytes-12                 31.2ns ± 4%  31.4ns ±11%     ~     (p=0.308 n=19+18)
GrowSliceInts-12                  53.4ns ±14%  45.0ns ± 6%  -15.74%  (p=0.000 n=20+19)
GrowSlicePtr-12                   87.0ns ± 3%  83.3ns ± 3%   -4.26%  (p=0.000 n=18+17)
GrowSliceStruct24Bytes-12         88.9ns ± 5%  77.8ns ± 2%  -12.45%  (p=0.000 n=20+19)
Append-12                         17.2ns ± 1%  17.3ns ± 2%     ~     (p=0.464 n=18+17)
AppendGrowByte-12                 2.28ms ± 1%  1.92ms ± 2%  -15.65%  (p=0.000 n=20+18)
AppendGrowString-12                255ms ± 3%   253ms ± 4%     ~     (p=0.065 n=19+19)
AppendSlice/1Bytes-12             3.13ns ± 0%  3.11ns ± 1%   -0.65%  (p=0.000 n=17+18)
AppendSlice/4Bytes-12             3.02ns ± 2%  3.11ns ± 1%   +3.27%  (p=0.000 n=18+17)
AppendSlice/7Bytes-12             4.14ns ± 3%  4.13ns ± 2%     ~     (p=0.380 n=19+18)
AppendSlice/8Bytes-12             3.74ns ± 3%  3.68ns ± 1%   -1.76%  (p=0.000 n=19+18)
AppendSlice/15Bytes-12            4.03ns ± 2%  4.04ns ± 2%     ~     (p=0.261 n=19+20)
AppendSlice/16Bytes-12            4.03ns ± 2%  4.03ns ± 0%     ~     (p=0.062 n=18+17)
AppendSlice/32Bytes-12            3.23ns ± 4%  3.43ns ± 1%   +6.10%  (p=0.000 n=17+18)
AppendStr/1Bytes-12               3.51ns ± 1%  3.52ns ± 1%     ~     (p=0.321 n=18+19)
AppendStr/4Bytes-12               3.46ns ± 1%  3.46ns ± 1%     ~     (p=0.977 n=18+20)
AppendStr/8Bytes-12               3.18ns ± 1%  3.19ns ± 1%     ~     (p=0.650 n=16+17)
AppendStr/16Bytes-12              6.08ns ±27%  5.52ns ± 3%   -9.16%  (p=0.002 n=18+19)
AppendStr/32Bytes-12              3.71ns ± 1%  3.53ns ± 1%   -4.73%  (p=0.000 n=20+19)
AppendSpecialCase-12              17.7ns ± 1%  17.8ns ± 3%   +0.86%  (p=0.045 n=17+18)
AppendInPlace/NoGrow/Byte-12       375ns ± 1%   376ns ± 1%   +0.35%  (p=0.021 n=20+18)
AppendInPlace/NoGrow/1Ptr-12      1.01µs ± 1%  1.10µs ± 1%   +9.28%  (p=0.000 n=18+20)
AppendInPlace/NoGrow/2Ptr-12      1.85µs ± 2%  1.71µs ± 1%   -7.51%  (p=0.000 n=19+18)
AppendInPlace/NoGrow/3Ptr-12      2.57µs ± 2%  2.44µs ± 1%   -5.08%  (p=0.000 n=19+19)
AppendInPlace/NoGrow/4Ptr-12      3.52µs ± 2%  3.35µs ± 2%   -4.70%  (p=0.000 n=20+19)
AppendInPlace/Grow/Byte-12         212ns ± 1%   217ns ± 8%   +2.57%  (p=0.000 n=20+20)
AppendInPlace/Grow/1Ptr-12         214ns ± 2%   217ns ± 3%   +1.23%  (p=0.001 n=18+19)
AppendInPlace/Grow/2Ptr-12         298ns ± 2%   300ns ± 2%   +0.55%  (p=0.038 n=19+20)
AppendInPlace/Grow/3Ptr-12         367ns ± 2%   366ns ± 2%     ~     (p=0.452 n=20+18)
AppendInPlace/Grow/4Ptr-12         416ns ± 2%   411ns ± 2%   -1.18%  (p=0.000 n=20+19)
StackGrowth-12                    43.4ns ± 1%  43.4ns ± 0%     ~     (p=1.000 n=16+16)
StackGrowthDeep-12                11.4µs ± 4%  10.3µs ± 4%   -9.65%  (p=0.000 n=20+19)

Change-Id: I69a8afbd942c787c591d95b9d9439bd6db4d1e49
Reviewed-on: https://go-review.googlesource.com/30192
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
2016-10-04 22:40:20 +00:00
Marcel van Lohuizen 23cb8864b5 runtime: use Run for more benchmarks
Names for Append?Bytes are slightly changed in addition to adding a slash.

Change-Id: I0291aa29c693f9040fd01368eaad9766259677df
Reviewed-on: https://go-review.googlesource.com/23426
Run-TryBot: Marcel van Lohuizen <mpvl@golang.org>
Reviewed-by: Russ Cox <rsc@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
2016-05-25 17:46:50 +00:00
Josh Bleecher Snyder 411a0adc9b runtime: add benchmarks for in-place append
Change-Id: I2b43cc976d2efbf8b41170be536fdd10364b65e5
Reviewed-on: https://go-review.googlesource.com/22190
Reviewed-by: Keith Randall <khr@golang.org>
2016-04-18 19:08:39 +00:00
Martin Möhrmann ad7448fe98 runtime: speed up makeslice by avoiding divisions
Only compute the number of maximum allowed elements per slice once.

name         old time/op  new time/op  delta
MakeSlice-2  55.5ns ± 1%  45.6ns ± 2%  -17.88%  (p=0.000 n=99+100)

Change-Id: I951feffda5d11910a75e55d7e978d306d14da2c5
Reviewed-on: https://go-review.googlesource.com/21801
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
Reviewed-by: Ian Lance Taylor <iant@golang.org>
2016-04-10 23:39:46 +00:00
Marvin Stenger 6b0688f742 runtime: speed up growslice by avoiding divisions 2
This is a follow-up of https://go-review.googlesource.com/#/c/20653/

Special case computation for slices with elements of byte size or
pointer size.

name                      old time/op  new time/op  delta
GrowSliceBytes-4          86.2ns ± 3%  75.4ns ± 2%  -12.50%  (p=0.000 n=20+20)
GrowSliceInts-4            161ns ± 3%   136ns ± 3%  -15.59%  (p=0.000 n=19+19)
GrowSlicePtr-4             239ns ± 2%   233ns ± 2%   -2.52%  (p=0.000 n=20+20)
GrowSliceStruct24Bytes-4   258ns ± 3%   256ns ± 3%     ~     (p=0.134 n=20+20)

Change-Id: Ice5fa648058fe9d7fa89dee97ca359966f671128
Reviewed-on: https://go-review.googlesource.com/21101
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
2016-03-25 16:47:56 +00:00
Martin Möhrmann 43ed65f869 runtime: speed up growslice by avoiding divisions
Only compute the number of maximum allowed elements per slice once.
Special case newcap computation for slices with byte sized elements.

name              old time/op  new time/op  delta
GrowSliceBytes-2  61.1ns ± 1%  43.4ns ± 1%  -29.00%  (p=0.000 n=20+20)
GrowSliceInts-2   85.9ns ± 1%  75.7ns ± 1%  -11.80%  (p=0.000 n=20+20)

Change-Id: I5d9c0d5987cdd108ac29dc32e31912dcefa2324d
Reviewed-on: https://go-review.googlesource.com/20653
Reviewed-by: Ian Lance Taylor <iant@golang.org>
Run-TryBot: Ian Lance Taylor <iant@golang.org>
Reviewed-by: Keith Randall <khr@golang.org>
TryBot-Result: Gobot Gobot <gobot@golang.org>
2016-03-14 00:45:57 +00:00
Russ Cox c007ce824d build: move package sources from src/pkg to src
Preparation was in CL 134570043.
This CL contains only the effect of 'hg mv src/pkg/* src'.
For more about the move, see golang.org/s/go14nopkg.
2014-09-08 00:08:51 -04:00