Age | Commit message (Collapse) | Author |
|
|
|
|
|
Previously vcpkg was relying on https://repo.msys2.org/mingw/i686/mingw-w64-i686-pkg-config-0.29.2-1-any.pkg.tar.xz which is no longer available. The vcpkg source has been updated to use http://repo.msys2.org/mingw/i686/mingw-w64-i686-pkg-config-0.29.2-2-any.pkg.tar.zst. This PR updates the commit ID used to checkout vcpkg for the updated URL.
|
|
|
|
causing sporadic ABI issues on Visual Studio updates.
Tidied up debug and release configuration blocks in common project file to avoid duplication.
Updated appveyor config to use latest Visual Studio 2019 image.
Changed appveyor config file hash to use a new version of Qt pre-compiled binaries built for Visual Studio 2019 v16.8.1.
Bumped vcpkg version to tag '2020.11-1' for binary caching feature.
See #20392 for related discussion.
|
|
introduced in #19960. It re-applies a setting to inform vcpkg to only build release vesions of the dependencies rather than the default of debug and release.
It had been expected that the vcpkg manifest mechanism introduced in #19960 would do this automatically but it turns out not to be the case.
|
|
This is a workaround as the recent Visual Studio 2019 image update
breaks our builds.
|
|
This fixes the appveyor CI job, see #20066.
Currently the job fails because some of the vcpkg dependencies need to install msys2 and the hardcoded mirror in the vcpkg config is down.
Vcpkg commit 76a7e9248fb3c57350b559966dcaa2d52a5e4458 adds new mirrors to the hardcoded list.
|
|
The vcpkg tool has introduced a proper way to use manifests, https://devblogs.microsoft.com/cppblog/vcpkg-accelerate-your-team-development-environment-with-binary-caching-and-manifests/. This PR replaces the custom text file mechanism with the new manifest approach.
It is planned that vckpg manifests will include the ability to version dependencies in the future. Dependency versions would solve a number of issues that currently require workarounds with the appveyor CI.
Set vcpkg manifest version to 1 to avoid any perception it's related to any release or other version numbering.
|
|
|
|
Latest vm version has updated cmake and the berkeleydb vcpkg now fails to build.
|
|
78c312c983255e15fc274de2368a2ec13ce81cbf Replace current benchmarking framework with nanobench (Martin Ankerl)
Pull request description:
Replace current benchmarking framework with nanobench
This replaces the current benchmarking framework with nanobench [1], an
MIT licensed single-header benchmarking library, of which I am the
autor. This has in my opinion several advantages, especially on Linux:
* fast: Running all benchmarks takes ~6 seconds instead of 4m13s on
an Intel i7-8700 CPU @ 3.20GHz.
* accurate: I ran e.g. the benchmark for SipHash_32b 10 times and
calculate standard deviation / mean = coefficient of variation:
* 0.57% CV for old benchmarking framework
* 0.20% CV for nanobench
So the benchmark results with nanobench seem to vary less than with
the old framework.
* It automatically determines runtime based on clock precision, no need
to specify number of evaluations.
* measure instructions, cycles, branches, instructions per cycle,
branch misses (only Linux, when performance counters are available)
* output in markdown table format.
* Warn about unstable environment (frequency scaling, turbo, ...)
* For better profiling, it is possible to set the environment variable
NANOBENCH_ENDLESS to force endless running of a particular benchmark
without the need to recompile. This makes it to e.g. run "perf top"
and look at hotspots.
Here is an example copy & pasted from the terminal output:
| ns/byte | byte/s | err% | ins/byte | cyc/byte | IPC | bra/byte | miss% | total | benchmark
|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|---------------:|--------:|----------:|:----------
| 2.52 | 396,529,415.94 | 0.6% | 25.42 | 8.02 | 3.169 | 0.06 | 0.0% | 0.03 | `bench/crypto_hash.cpp RIPEMD160`
| 1.87 | 535,161,444.83 | 0.3% | 21.36 | 5.95 | 3.589 | 0.06 | 0.0% | 0.02 | `bench/crypto_hash.cpp SHA1`
| 3.22 | 310,344,174.79 | 1.1% | 36.80 | 10.22 | 3.601 | 0.09 | 0.0% | 0.04 | `bench/crypto_hash.cpp SHA256`
| 2.01 | 496,375,796.23 | 0.0% | 18.72 | 6.43 | 2.911 | 0.01 | 1.0% | 0.00 | `bench/crypto_hash.cpp SHA256D64_1024`
| 7.23 | 138,263,519.35 | 0.1% | 82.66 | 23.11 | 3.577 | 1.63 | 0.1% | 0.00 | `bench/crypto_hash.cpp SHA256_32b`
| 3.04 | 328,780,166.40 | 0.3% | 35.82 | 9.69 | 3.696 | 0.03 | 0.0% | 0.03 | `bench/crypto_hash.cpp SHA512`
[1] https://github.com/martinus/nanobench
ACKs for top commit:
laanwj:
ACK 78c312c983255e15fc274de2368a2ec13ce81cbf
Tree-SHA512: 9e18770b18b6f95a7d0105a4a5497d31cf4eb5efe6574f4482f6f1b4c88d7e0946b9a4a1e9e8e6ecbf41a3f2d7571240677dcb45af29a6f0584e89b25f32e49e
|
|
configuration.
|
|
This replaces the current benchmarking framework with nanobench [1], an
MIT licensed single-header benchmarking library, of which I am the
autor. This has in my opinion several advantages, especially on Linux:
* fast: Running all benchmarks takes ~6 seconds instead of 4m13s on
an Intel i7-8700 CPU @ 3.20GHz.
* accurate: I ran e.g. the benchmark for SipHash_32b 10 times and
calculate standard deviation / mean = coefficient of variation:
* 0.57% CV for old benchmarking framework
* 0.20% CV for nanobench
So the benchmark results with nanobench seem to vary less than with
the old framework.
* It automatically determines runtime based on clock precision, no need
to specify number of evaluations.
* measure instructions, cycles, branches, instructions per cycle,
branch misses (only Linux, when performance counters are available)
* output in markdown table format.
* Warn about unstable environment (frequency scaling, turbo, ...)
* For better profiling, it is possible to set the environment variable
NANOBENCH_ENDLESS to force endless running of a particular benchmark
without the need to recompile. This makes it to e.g. run "perf top"
and look at hotspots.
Here is an example copy & pasted from the terminal output:
| ns/byte | byte/s | err% | ins/byte | cyc/byte | IPC | bra/byte | miss% | total | benchmark
|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|---------------:|--------:|----------:|:----------
| 2.52 | 396,529,415.94 | 0.6% | 25.42 | 8.02 | 3.169 | 0.06 | 0.0% | 0.03 | `bench/crypto_hash.cpp RIPEMD160`
| 1.87 | 535,161,444.83 | 0.3% | 21.36 | 5.95 | 3.589 | 0.06 | 0.0% | 0.02 | `bench/crypto_hash.cpp SHA1`
| 3.22 | 310,344,174.79 | 1.1% | 36.80 | 10.22 | 3.601 | 0.09 | 0.0% | 0.04 | `bench/crypto_hash.cpp SHA256`
| 2.01 | 496,375,796.23 | 0.0% | 18.72 | 6.43 | 2.911 | 0.01 | 1.0% | 0.00 | `bench/crypto_hash.cpp SHA256D64_1024`
| 7.23 | 138,263,519.35 | 0.1% | 82.66 | 23.11 | 3.577 | 1.63 | 0.1% | 0.00 | `bench/crypto_hash.cpp SHA256_32b`
| 3.04 | 328,780,166.40 | 0.3% | 35.82 | 9.69 | 3.696 | 0.03 | 0.0% | 0.03 | `bench/crypto_hash.cpp SHA512`
[1] https://github.com/martinus/nanobench
* Adds support for asymptotes
This adds support to calculate asymptotic complexity of a benchmark.
This is similar to #17375, but currently only one asymptote is
supported, and I have added support in the benchmark `ComplexMemPool`
as an example.
Usage is e.g. like this:
```
./bench_bitcoin -filter=ComplexMemPool -asymptote=25,50,100,200,400,600,800
```
This runs the benchmark `ComplexMemPool` several times but with
different complexityN settings. The benchmark can extract that number
and use it accordingly. Here, it's used for `childTxs`. The output is
this:
| complexityN | ns/op | op/s | err% | ins/op | cyc/op | IPC | total | benchmark
|------------:|--------------------:|--------------------:|--------:|----------------:|----------------:|-------:|----------:|:----------
| 25 | 1,064,241.00 | 939.64 | 1.4% | 3,960,279.00 | 2,829,708.00 | 1.400 | 0.01 | `ComplexMemPool`
| 50 | 1,579,530.00 | 633.10 | 1.0% | 6,231,810.00 | 4,412,674.00 | 1.412 | 0.02 | `ComplexMemPool`
| 100 | 4,022,774.00 | 248.58 | 0.6% | 16,544,406.00 | 11,889,535.00 | 1.392 | 0.04 | `ComplexMemPool`
| 200 | 15,390,986.00 | 64.97 | 0.2% | 63,904,254.00 | 47,731,705.00 | 1.339 | 0.17 | `ComplexMemPool`
| 400 | 69,394,711.00 | 14.41 | 0.1% | 272,602,461.00 | 219,014,691.00 | 1.245 | 0.76 | `ComplexMemPool`
| 600 | 168,977,165.00 | 5.92 | 0.1% | 639,108,082.00 | 535,316,887.00 | 1.194 | 1.86 | `ComplexMemPool`
| 800 | 310,109,077.00 | 3.22 | 0.1% |1,149,134,246.00 | 984,620,812.00 | 1.167 | 3.41 | `ComplexMemPool`
| coefficient | err% | complexity
|--------------:|-------:|------------
| 4.78486e-07 | 4.5% | O(n^2)
| 6.38557e-10 | 21.7% | O(n^3)
| 3.42338e-05 | 38.0% | O(n log n)
| 0.000313914 | 46.9% | O(n)
| 0.0129823 | 114.4% | O(log n)
| 0.0815055 | 133.8% | O(1)
The best fitting curve is O(n^2), so the algorithm seems to scale
quadratic with `childTxs` in the range 25 to 800.
|
|
fac0c8db9f634ea7231fb0b6dc139ae0bda4a051 appveyor: Remove clcache (MarcoFalke)
Pull request description:
The build time without cache seems to be 47 minutes and with cache 46 minutes. Maybe we can save more time by not installing clcache.
Top commit has no ACKs.
Tree-SHA512: ce503641a465f5e49deb7aa6b566aaab5e567148c001704451891c49ba7ade3219ca788bc0d8e55565febb9aadd579e26894f6c03b99755eb8b5d1498acf4dc9
|
|
|
|
Also add a draft for a Windows build on cirrus. The draft can be
finished later.
|
|
|
|
|
|
msvc warning C4834 for the Bitcoin Core build was introduced by Visual Studio 16.4.0. This PR adds an ignore rule for the warning (it's related to the nodiscard attribute and is not considered relevant).
An additional side effect of the msvc compiler update is the prebuilt Qt5.9.8 libraries cannot be linked due to being built with an earlier version of the compiler. To fix this a new Qt5.9.8 version has been compiled and the appveyor job updated to use them. The GitHub Actions job needs to continue to use the original Qt5.9.8 libraries until the latest GitHub Windows image also updates to >= Visual Studio 2019 v16.4.
|
|
use it.
|
|
- Update build image from Visual Studio 2017 to Visual Studio 2019.
- Updated Qt static library from Qt5.9.7 to Qt5.9.8.
- Added commands to update vcpkg port files (this does not update already installed packages).
- Updated vcpkg package list as per #17309.
- Removed commands setting common project file options. Now done via common.init.vcxproj include.
- Changed msbuild verbosity from normal to quiet. Normal rights a LOT of logs and impacts appveyor job duration.
Updated msvc project configs:
- Updated platform toolset from v141 to v142.
- Updated Qt static library from Qt5.9.7 to Qt5.9.8.
- Added ignore for linker warning building bitcoin-qt program.
- Added missing util/str.cpp class file to test_bitcoin project file.
|
|
|
|
script. Increasing the verbosity helps to identify the cause of build errors which is the main purpose of the appveyor script.
|
|
|
|
|
|
|
|
|
|
|
|
d067e81dcf574b6504813bbf742cd5af057e8e9b msvc: add rapid check property tests (Chun Kuan Lee)
Pull request description:
This PR add the property tests into the binaries built by MSVC.
And another trivial change is that I reordered the appveyor package list.
Tree-SHA512: 25d66db464beb7b512cc1f88d8557d6a047000a97d78f49884bb91a65ec142e0458039c919f51bf73413359fcf3e63e1ea4d76586b862f1c140d2ca05ee8b23d
|
|
|
|
|
|
|
|
|
|
|
|
|
|
from /nowarn to project file.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
ff40357da1 AppVeyor: Move AppVeyor YAML to dot-file-style YAML (Mitchell Cash)
Pull request description:
AppVeyor supports dot-file-style YAML named `.appveyor.yml` as is. This helps keep the root of the repository clean(ish) and readable by having the CI files as dot-files.
Source: https://www.appveyor.com/docs/build-configuration/#yaml-file-alternative-naming
Tree-SHA512: da2a78aff775e5e146f0784b1a6617d0371a5821da8a53be9e4aa57409cb16360f43d0afa5745f81f776599950cab4219a5d7ee7247d42e25861963ea487db66
|
|
AppVeyor supports dot-file-style YAML named .appveyor.yml as is. This
helps keep the root of the repository clean(ish) and readable by having
the CI files as dot-files.
|