mirror of https://github.com/pulumi/pulumi.git
9 Commits
Author | SHA1 | Message | Date |
---|---|---|---|
Fraser Waters |
3560333ae6
|
Clean up uses of .Error() (#14965)
Combination of a few cleanups. 1. Don't call .Error() on errors that are being passed to "%s" format functions. Format will call `Error()` itself. 2. Don't call assert.Error then assert.Equal/Contains, just use assert.ErrorEqual/ErrorContains instead. 3. Use "%w" if appropriate, instead of "%v"/"%s". |
|
Fraser Waters |
16d9f4c167
|
Enable perfsprint linter (#14813)
<!--- Thanks so much for your contribution! If this is your first time contributing, please ensure that you have read the [CONTRIBUTING](https://github.com/pulumi/pulumi/blob/master/CONTRIBUTING.md) documentation. --> # Description <!--- Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. --> Prompted by a comment in another review: https://github.com/pulumi/pulumi/pull/14654#discussion_r1419995945 This lints that we don't use `fmt.Errorf` when `errors.New` will suffice, it also covers a load of other cases where `Sprintf` is sub-optimal. Most of these edits were made by running `perfsprint --fix`. ## Checklist - [x] I have run `make tidy` to update any new dependencies - [x] I have run `make lint` to verify my code passes the lint check - [x] I have formatted my code using `gofumpt` <!--- Please provide details if the checkbox below is to be left unchecked. --> - [ ] I have added tests that prove my fix is effective or that my feature works <!--- User-facing changes require a CHANGELOG entry. --> - [ ] I have run `make changelog` and committed the `changelog/pending/<file>` documenting my change <!-- If the change(s) in this PR is a modification of an existing call to the Pulumi Cloud, then the service should honor older versions of the CLI where this change would not exist. You must then bump the API version in /pkg/backend/httpstate/client/api.go, as well as add it to the service. --> - [ ] Yes, there are changes in this PR that warrants bumping the Pulumi Cloud API version <!-- @Pulumi employees: If yes, you must submit corresponding changes in the service repo. --> |
|
Fraser Waters |
478c29908f
|
Add a strict opinionated promise library (#14552)
<!--- Thanks so much for your contribution! If this is your first time contributing, please ensure that you have read the [CONTRIBUTING](https://github.com/pulumi/pulumi/blob/master/CONTRIBUTING.md) documentation. --> # Description <!--- Please include a summary of the change and which issue is fixed. Please also include relevant motivation and context. --> Prompted by having another case in https://github.com/pulumi/pulumi/pull/14548 that looked like "this is just a promise". We already had a mini-promise library for provider config in sdk/go/common/resource/plugin/provider_plugin.go. Plus a couple of places where we did a poor mans Go promise by having a WaitGroup plus a variable to set (see the two changes in /pkg). This adds a little promise library to safely cover all three of these cases plus the case I'll be adding in #14548. It is _minimal_ adding just the features of promises needed for these initial cases. We can add to it as needed. I suspect we've got a number of places in test code that could probably use this as well, but I haven't gone through that. I also suspect that having this type available will result in more places in the future being simpler because "its just a promise" is a fairly common scenario in async systems. In fact Output's internally _are just a promise_ so we could probably rewrite their internals using this, add a `Then()` like method for apply and all the async complexity gets handled in the promise layer while the output layer just cares about unknowns/secrets/etc. ## Checklist - [x] I have run `make tidy` to update any new dependencies - [x] I have run `make lint` to verify my code passes the lint check - [ ] I have formatted my code using `gofumpt` <!--- Please provide details if the checkbox below is to be left unchecked. --> - [x] I have added tests that prove my fix is effective or that my feature works <!--- User-facing changes require a CHANGELOG entry. --> - [ ] I have run `make changelog` and committed the `changelog/pending/<file>` documenting my change - This is in sdk/go/common and I'm considering it an non-public api for now. <!-- If the change(s) in this PR is a modification of an existing call to the Pulumi Cloud, then the service should honor older versions of the CLI where this change would not exist. You must then bump the API version in /pkg/backend/httpstate/client/api.go, as well as add it to the service. --> - [ ] Yes, there are changes in this PR that warrants bumping the Pulumi Cloud API version <!-- @Pulumi employees: If yes, you must submit corresponding changes in the service repo. --> |
|
Pat Gavlin | 1bc94a9d2a |
[snapshot] Use a newer diff package
The version of gopls that gotextdiff is derived from is over three years old. In the intervening time, the differ used by gopls has improved tremendously, and the language has grown new features like generics. These changes use a different diff package that is derived from an up-to-date version of gopls and exposes an API that is parameterized over the storage of the text. Taken together, this allows us to use a much faster and more memory-efficient algorithm for diffing while safely avoiding string copies when invoking the differ. The more efficient algorithm operates on arbitrary slices whose elements are comparable. Naive use of this algorithm--passing byte slices in--can cause increased time spent diffing, as the sequences being compared are longer. In order to avoid this, these changes record spans of marshaled deployments as the deployments are marshaled. These spans are strings that cover the diffable elements of the deployment: its header, its resources, and its pending operations. The output of the newer algorithm is converible to the format we currently send to the service, so these changes should not require service-side changes. In order to avoid bumping the minimum Go version required to build the `github.com/pulumi/pulumi/pkg/v3" and therefore break backwards compatibility with consumers using older toolchains (e.g. providers), these changes use the existing differ when built with a pre-1.20 toolchain. Official builds of the CLI are already using Go 1.20+ and should use the new differ without additional changes. These changes improve CPU time, allocation volume, and the compression ratio by 57%, 87%, and 52%, respectively. Benchmark results and analysis: httpstate ❯ go test -count=10 -run none -benchmem -bench . | tee spanned.txt httpstate ❯ benchstat buffer-pool.txt spanned.txt >stat.txt ``` │ buffer-pool.txt │ spanned.txt │ │ sec/op │ sec/op vs base │ DiffStack/1_x_2_B-10 45.09µ ± 1% 22.81µ ± 1% -49.41% (p=0.000 n=10) DiffStack/2_x_2_B-10 79.49µ ± 1% 40.60µ ± 1% -48.93% (p=0.000 n=10) DiffStack/4_x_2_B-10 172.54µ ± 2% 81.57µ ± 0% -52.72% (p=0.000 n=10) DiffStack/8_x_2_B-10 455.4µ ± 1% 200.4µ ± 0% -56.00% (p=0.000 n=10) DiffStack/16_x_2_B-10 1388.2µ ± 1% 575.1µ ± 1% -58.57% (p=0.000 n=10) DiffStack/32_x_2_B-10 4.409m ± 1% 1.866m ± 0% -57.67% (p=0.000 n=10) DiffStack/48_x_2_B-10 9.275m ± 0% 3.484m ± 0% -62.44% (p=0.000 n=10) DiffStack/64_x_2_B-10 14.564m ± 0% 5.945m ± 1% -59.18% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 124.81µ ± 1% 40.14µ ± 1% -67.84% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 299.8µ ± 0% 121.8µ ± 1% -59.38% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 846.2µ ± 1% 355.8µ ± 0% -57.96% (p=0.000 n=10) DiffStack/8_x_8.2_kB-10 2.743m ± 1% 1.150m ± 0% -58.06% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 9.271m ± 1% 4.041m ± 0% -56.41% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 31.78m ± 0% 14.97m ± 0% -52.88% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 66.14m ± 1% 30.66m ± 0% -53.63% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 107.50m ± 0% 53.44m ± 0% -50.29% (p=0.000 n=10) DiffStack/1_x_33_kB-10 240.51µ ± 1% 76.95µ ± 0% -68.01% (p=0.000 n=10) DiffStack/2_x_33_kB-10 798.5µ ± 0% 310.6µ ± 0% -61.10% (p=0.000 n=10) DiffStack/4_x_33_kB-10 2451.9µ ± 1% 978.9µ ± 3% -60.08% (p=0.000 n=10) DiffStack/8_x_33_kB-10 8.258m ± 0% 3.731m ± 0% -54.82% (p=0.000 n=10) DiffStack/16_x_33_kB-10 28.29m ± 1% 12.77m ± 0% -54.85% (p=0.000 n=10) DiffStack/32_x_33_kB-10 104.11m ± 1% 44.84m ± 1% -56.93% (p=0.000 n=10) DiffStack/48_x_33_kB-10 199.73m ± 1% 95.87m ± 1% -52.00% (p=0.000 n=10) DiffStack/64_x_33_kB-10 339.6m ± 2% 159.2m ± 2% -53.12% (p=0.000 n=10) DiffStack/2_x_131_kB-10 2721.7µ ± 0% 984.3µ ± 0% -63.84% (p=0.000 n=10) DiffStack/4_x_131_kB-10 8.334m ± 1% 3.654m ± 2% -56.15% (p=0.000 n=10) DiffStack/8_x_131_kB-10 25.89m ± 0% 12.23m ± 0% -52.77% (p=0.000 n=10) DiffStack/16_x_131_kB-10 87.49m ± 1% 40.26m ± 0% -53.98% (p=0.000 n=10) DiffStack/32_x_131_kB-10 317.0m ± 5% 150.4m ± 0% -52.57% (p=0.000 n=10) DiffStack/48_x_131_kB-10 681.8m ± 1% 332.1m ± 0% -51.28% (p=0.000 n=10) DiffStack/64_x_131_kB-10 1190.4m ± 1% 583.1m ± 0% -51.01% (p=0.000 n=10) DiffStack/1_x_524_kB-10 3.215m ± 0% 1.086m ± 0% -66.21% (p=0.000 n=10) DiffStack/2_x_524_kB-10 9.676m ± 1% 3.994m ± 0% -58.72% (p=0.000 n=10) DiffStack/4_x_524_kB-10 29.84m ± 0% 12.88m ± 1% -56.84% (p=0.000 n=10) DiffStack/8_x_524_kB-10 92.11m ± 1% 42.35m ± 0% -54.02% (p=0.000 n=10) DiffStack/16_x_524_kB-10 322.1m ± 0% 154.0m ± 0% -52.19% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 54.73m ± 1% 22.63m ± 0% -58.65% (p=0.000 n=10) DiffStackRecorded/checkpoints.json-10 815.1m ± 0% 395.5m ± 2% -51.48% (p=0.000 n=10) geomean 10.45m 4.530m -56.65% │ buffer-pool.txt │ spanned.txt │ │ checkpoint_bytes │ checkpoint_bytes vs base │ DiffStack/1_x_2_B-10 2.539k ± 0% 1.950k ± 0% -23.20% (p=0.000 n=10) DiffStack/2_x_2_B-10 5.201k ± 0% 4.548k ± 0% -12.56% (p=0.000 n=10) DiffStack/4_x_2_B-10 13.63k ± 0% 12.30k ± 0% -9.77% (p=0.000 n=10) DiffStack/8_x_2_B-10 45.09k ± 0% 42.86k ± 0% -4.95% (p=0.000 n=10) DiffStack/16_x_2_B-10 161.9k ± 0% 156.4k ± 0% -3.42% (p=0.000 n=10) DiffStack/32_x_2_B-10 579.8k ± 0% 604.2k ± 0% +4.21% (p=0.000 n=10) DiffStack/48_x_2_B-10 1.306M ± 0% 1.180M ± 0% -9.68% (p=0.000 n=10) DiffStack/64_x_2_B-10 2.164M ± 0% 2.134M ± 0% -1.36% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 27.109k ± 0% 9.856k ± 0% -63.64% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 78.91k ± 0% 61.59k ± 0% -21.94% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 284.2k ± 0% 258.3k ± 0% -9.13% (p=0.000 n=10) DiffStack/8_x_8.2_kB-10 1.119M ± 0% 1.041M ± 0% -6.89% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 4.389M ± 0% 4.201M ± 0% -4.29% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 16.74M ± 0% 17.06M ± 0% +1.95% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 36.30M ± 0% 35.93M ± 0% -1.00% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 60.98M ± 0% 64.06M ± 0% +5.05% (p=0.000 n=10) DiffStack/1_x_33_kB-10 67.79k ± 0% 34.43k ± 0% -49.20% (p=0.000 n=10) DiffStack/2_x_33_kB-10 300.1k ± 0% 233.6k ± 0% -22.15% (p=0.000 n=10) DiffStack/4_x_33_kB-10 1128.3k ± 0% 929.5k ± 0% -17.62% (p=0.000 n=10) DiffStack/8_x_33_kB-10 4.437M ± 0% 4.238M ± 0% -4.49% (p=0.000 n=10) DiffStack/16_x_33_kB-10 16.57M ± 0% 15.78M ± 0% -4.79% (p=0.000 n=10) DiffStack/32_x_33_kB-10 64.35M ± 0% 58.66M ± 0% -8.84% (p=0.000 n=10) DiffStack/48_x_33_kB-10 127.0M ± 0% 128.0M ± 0% +0.80% (p=0.000 n=10) DiffStack/64_x_33_kB-10 217.7M ± 0% 216.5M ± 0% -0.54% (p=0.000 n=10) DiffStack/2_x_131_kB-10 1316.2k ± 0% 921.8k ± 0% -29.97% (p=0.000 n=10) DiffStack/4_x_131_kB-10 4.733M ± 0% 4.207M ± 0% -11.11% (p=0.000 n=10) DiffStack/8_x_131_kB-10 16.03M ± 0% 15.77M ± 0% -1.64% (p=0.000 n=10) DiffStack/16_x_131_kB-10 57.42M ± 0% 54.66M ± 0% -4.81% (p=0.000 n=10) DiffStack/32_x_131_kB-10 214.8M ± 0% 210.7M ± 0% -1.90% (p=0.000 n=10) DiffStack/48_x_131_kB-10 472.9M ± 0% 466.9M ± 0% -1.28% (p=0.000 n=10) DiffStack/64_x_131_kB-10 832.3M ± 0% 824.7M ± 0% -0.92% (p=0.000 n=10) DiffStack/1_x_524_kB-10 1.575M ± 0% 1.051M ± 0% -33.32% (p=0.000 n=10) DiffStack/2_x_524_kB-10 5.248M ± 0% 4.199M ± 0% -20.00% (p=0.000 n=10) DiffStack/4_x_524_kB-10 18.36M ± 0% 15.74M ± 0% -14.28% (p=0.000 n=10) DiffStack/8_x_524_kB-10 60.86M ± 0% 56.66M ± 0% -6.90% (p=0.000 n=10) DiffStack/16_x_524_kB-10 223.5M ± 0% 214.0M ± 0% -4.23% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 13.21M ± 0% 13.20M ± 0% -0.01% (p=0.000 n=10) DiffStackRecorded/checkpoints.json-10 155.9M ± 0% 154.9M ± 0% -0.60% (p=0.000 n=10) geomean 3.606M 3.173M -12.01% │ buffer-pool.txt │ spanned.txt │ │ ratio │ ratio vs base │ DiffStack/1_x_2_B-10 789.5m ± 0% 1641.0m ± 0% +107.85% (p=0.000 n=10) DiffStack/2_x_2_B-10 1.034 ± 0% 2.353 ± 0% +127.56% (p=0.000 n=10) DiffStack/4_x_2_B-10 1.575 ± 0% 3.592 ± 0% +128.06% (p=0.000 n=10) DiffStack/8_x_2_B-10 2.829 ± 0% 6.684 ± 0% +136.27% (p=0.000 n=10) DiffStack/16_x_2_B-10 5.296 ± 0% 12.610 ± 0% +138.10% (p=0.000 n=10) DiffStack/32_x_2_B-10 9.681 ± 0% 24.790 ± 0% +156.07% (p=0.000 n=10) DiffStack/48_x_2_B-10 14.65 ± 0% 32.44 ± 0% +121.43% (p=0.000 n=10) DiffStack/64_x_2_B-10 18.26 ± 0% 44.11 ± 0% +141.57% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 1.383 ± 0% 1.051 ± 0% -24.01% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 1.716 ± 0% 2.324 ± 0% +35.43% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 2.878 ± 0% 4.251 ± 0% +47.71% (p=0.000 n=10) DiffStack/8_x_8.2_kB-10 5.475 ± 0% 8.056 ± 0% +47.14% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 10.56 ± 0% 15.77 ± 0% +49.34% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 19.97 ± 0% 31.57 ± 0% +58.09% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 28.80 ± 0% 44.11 ± 0% +53.16% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 36.23 ± 0% 58.84 ± 0% +62.41% (p=0.000 n=10) DiffStack/1_x_33_kB-10 986.0m ± 0% 1014.0m ± 0% +2.84% (p=0.000 n=10) DiffStack/2_x_33_kB-10 1.777 ± 0% 2.331 ± 0% +31.18% (p=0.000 n=10) DiffStack/4_x_33_kB-10 3.057 ± 0% 3.993 ± 0% +30.62% (p=0.000 n=10) DiffStack/8_x_33_kB-10 5.766 ± 0% 8.511 ± 0% +47.61% (p=0.000 n=10) DiffStack/16_x_33_kB-10 10.55 ± 0% 15.35 ± 0% +45.50% (p=0.000 n=10) DiffStack/32_x_33_kB-10 20.28 ± 0% 28.08 ± 0% +38.46% (p=0.000 n=10) DiffStack/48_x_33_kB-10 26.59 ± 0% 40.64 ± 0% +52.84% (p=0.000 n=10) DiffStack/64_x_33_kB-10 34.14 ± 0% 51.44 ± 0% +50.67% (p=0.000 n=10) DiffStack/2_x_131_kB-10 1.993 ± 0% 2.333 ± 0% +17.06% (p=0.000 n=10) DiffStack/4_x_131_kB-10 3.263 ± 0% 4.569 ± 0% +40.02% (p=0.000 n=10) DiffStack/8_x_131_kB-10 5.291 ± 0% 7.995 ± 0% +51.11% (p=0.000 n=10) DiffStack/16_x_131_kB-10 9.275 ± 0% 13.410 ± 0% +44.58% (p=0.000 n=10) DiffStack/32_x_131_kB-10 17.17 ± 0% 25.44 ± 0% +48.17% (p=0.000 n=10) DiffStack/48_x_131_kB-10 25.11 ± 0% 37.39 ± 0% +48.90% (p=0.000 n=10) DiffStack/64_x_131_kB-10 33.09 ± 0% 49.40 ± 0% +49.29% (p=0.000 n=10) DiffStack/1_x_524_kB-10 1.498 ± 0% 1.999 ± 0% +33.44% (p=0.000 n=10) DiffStack/2_x_524_kB-10 1.998 ± 0% 2.666 ± 0% +33.43% (p=0.000 n=10) DiffStack/4_x_524_kB-10 3.180 ± 0% 4.285 ± 0% +34.75% (p=0.000 n=10) DiffStack/8_x_524_kB-10 5.040 ± 0% 7.199 ± 0% +42.84% (p=0.000 n=10) DiffStack/16_x_524_kB-10 9.058 ± 0% 13.160 ± 0% +45.29% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 1.997 ± 0% 1.999 ± 0% +0.10% (p=0.000 n=10) DiffStackRecorded/checkpoints.json-10 36.60 ± 0% 40.42 ± 0% +10.44% (p=0.000 n=10) geomean 5.993 9.122 +52.20% │ buffer-pool.txt │ spanned.txt │ │ wire_bytes │ wire_bytes vs base │ DiffStack/1_x_2_B-10 3.216k ± 0% 1.188k ± 0% -63.06% (p=0.000 n=10) DiffStack/2_x_2_B-10 5.031k ± 0% 1.933k ± 0% -61.58% (p=0.000 n=10) DiffStack/4_x_2_B-10 8.655k ± 0% 3.424k ± 0% -60.44% (p=0.000 n=10) DiffStack/8_x_2_B-10 15.938k ± 0% 6.412k ± 0% -59.77% (p=0.000 n=10) DiffStack/16_x_2_B-10 30.58k ± 0% 12.40k ± 0% -59.45% (p=0.000 n=10) DiffStack/32_x_2_B-10 59.89k ± 0% 24.37k ± 0% -59.31% (p=0.000 n=10) DiffStack/48_x_2_B-10 89.17k ± 0% 36.38k ± 0% -59.21% (p=0.000 n=10) DiffStack/64_x_2_B-10 118.48k ± 0% 48.39k ± 0% -59.16% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 19.596k ± 0% 9.378k ± 0% -52.14% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 45.98k ± 0% 26.50k ± 0% -42.36% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 98.75k ± 0% 60.76k ± 0% -38.47% (p=0.000 n=10) DiffStack/8_x_8.2_kB-10 204.3k ± 0% 129.3k ± 0% -36.72% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 415.5k ± 0% 266.3k ± 0% -35.91% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 837.9k ± 0% 540.4k ± 0% -35.50% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 1260.4k ± 0% 814.6k ± 0% -35.37% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 1.683M ± 0% 1.089M ± 0% -35.31% (p=0.000 n=10) DiffStack/1_x_33_kB-10 68.75k ± 0% 33.95k ± 0% -50.61% (p=0.000 n=10) DiffStack/2_x_33_kB-10 168.9k ± 0% 100.2k ± 0% -40.64% (p=0.000 n=10) DiffStack/4_x_33_kB-10 369.1k ± 0% 232.8k ± 0% -36.93% (p=0.000 n=10) DiffStack/8_x_33_kB-10 769.6k ± 0% 497.9k ± 0% -35.30% (p=0.000 n=10) DiffStack/16_x_33_kB-10 1.571M ± 0% 1.028M ± 0% -34.53% (p=0.000 n=10) DiffStack/32_x_33_kB-10 3.173M ± 0% 2.089M ± 0% -34.16% (p=0.000 n=10) DiffStack/48_x_33_kB-10 4.775M ± 0% 3.149M ± 0% -34.04% (p=0.000 n=10) DiffStack/64_x_33_kB-10 6.377M ± 0% 4.210M ± 0% -33.98% (p=0.000 n=10) DiffStack/2_x_131_kB-10 660.4k ± 0% 395.1k ± 0% -40.16% (p=0.000 n=10) DiffStack/4_x_131_kB-10 1450.4k ± 0% 920.9k ± 0% -36.51% (p=0.000 n=10) DiffStack/8_x_131_kB-10 3.031M ± 0% 1.972M ± 0% -34.91% (p=0.000 n=10) DiffStack/16_x_131_kB-10 6.191M ± 0% 4.076M ± 0% -34.17% (p=0.000 n=10) DiffStack/32_x_131_kB-10 12.512M ± 0% 8.282M ± 0% -33.81% (p=0.000 n=10) DiffStack/48_x_131_kB-10 18.83M ± 0% 12.49M ± 0% -33.69% (p=0.000 n=10) DiffStack/64_x_131_kB-10 25.15M ± 0% 16.69M ± 0% -33.63% (p=0.000 n=10) DiffStack/1_x_524_kB-10 1051.8k ± 0% 525.5k ± 0% -50.04% (p=0.000 n=10) DiffStack/2_x_524_kB-10 2.626M ± 0% 1.575M ± 0% -40.04% (p=0.000 n=10) DiffStack/4_x_524_kB-10 5.776M ± 0% 3.673M ± 0% -36.40% (p=0.000 n=10) DiffStack/8_x_524_kB-10 12.075M ± 0% 7.871M ± 0% -34.82% (p=0.000 n=10) DiffStack/16_x_524_kB-10 24.67M ± 0% 16.27M ± 0% -34.07% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 6.612M ± 0% 6.606M ± 0% -0.10% (p=0.000 n=10) DiffStackRecorded/checkpoints.json-10 4.259M ± 0% 3.833M ± 0% -10.00% (p=0.000 n=10) geomean 601.7k 347.8k -42.19% │ buffer-pool.txt │ spanned.txt │ │ B/op │ B/op vs base │ DiffStack/1_x_2_B-10 31.85Ki ± 0% 15.79Ki ± 0% -50.42% (p=0.000 n=10) DiffStack/2_x_2_B-10 61.81Ki ± 0% 28.01Ki ± 0% -54.68% (p=0.000 n=10) DiffStack/4_x_2_B-10 147.58Ki ± 0% 55.60Ki ± 0% -62.33% (p=0.000 n=10) DiffStack/8_x_2_B-10 425.9Ki ± 0% 116.9Ki ± 0% -72.55% (p=0.000 n=10) DiffStack/16_x_2_B-10 1429.9Ki ± 0% 307.8Ki ± 0% -78.47% (p=0.000 n=10) DiffStack/32_x_2_B-10 4955.7Ki ± 0% 934.1Ki ± 0% -81.15% (p=0.000 n=10) DiffStack/48_x_2_B-10 10.746Mi ± 0% 1.682Mi ± 0% -84.34% (p=0.000 n=10) DiffStack/64_x_2_B-10 17.664Mi ± 0% 2.822Mi ± 0% -84.03% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 202.50Ki ± 0% 41.18Ki ± 0% -79.66% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 511.5Ki ± 0% 134.1Ki ± 0% -73.78% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 1513.5Ki ± 0% 309.4Ki ± 0% -79.56% (p=0.000 n=10) DiffStack/8_x_8.2_kB-10 4886.0Ki ± 0% 674.2Ki ± 0% -86.20% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 16.287Mi ± 0% 1.427Mi ± 0% -91.24% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 57.541Mi ± 0% 3.198Mi ± 0% -94.44% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 121.703Mi ± 0% 5.810Mi ± 0% -95.23% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 200.588Mi ± 0% 7.495Mi ± 0% -96.26% (p=0.000 n=10) DiffStack/1_x_33_kB-10 562.5Ki ± 0% 136.3Ki ± 0% -75.78% (p=0.000 n=10) DiffStack/2_x_33_kB-10 2041.0Ki ± 0% 505.5Ki ± 0% -75.23% (p=0.000 n=10) DiffStack/4_x_33_kB-10 5.874Mi ± 0% 1.129Mi ± 0% -80.78% (p=0.000 n=10) DiffStack/8_x_33_kB-10 18.837Mi ± 0% 2.410Mi ± 0% -87.21% (p=0.000 n=10) DiffStack/16_x_33_kB-10 61.898Mi ± 0% 5.015Mi ± 0% -91.90% (p=0.000 n=10) DiffStack/32_x_33_kB-10 221.69Mi ± 0% 10.36Mi ± 0% -95.32% (p=0.000 n=10) DiffStack/48_x_33_kB-10 430.13Mi ± 0% 18.50Mi ± 0% -95.70% (p=0.000 n=10) DiffStack/64_x_33_kB-10 722.07Mi ± 0% 21.63Mi ± 0% -97.00% (p=0.000 n=10) DiffStack/2_x_131_kB-10 8.098Mi ± 0% 1.656Mi ± 0% -79.55% (p=0.000 n=10) DiffStack/4_x_131_kB-10 22.662Mi ± 0% 3.934Mi ± 0% -82.64% (p=0.000 n=10) DiffStack/8_x_131_kB-10 65.217Mi ± 0% 8.425Mi ± 0% -87.08% (p=0.000 n=10) DiffStack/16_x_131_kB-10 204.90Mi ± 0% 17.22Mi ± 1% -91.60% (p=0.000 n=10) DiffStack/32_x_131_kB-10 703.12Mi ± 0% 34.89Mi ± 0% -95.04% (p=0.000 n=10) DiffStack/48_x_131_kB-10 1505.36Mi ± 0% 61.24Mi ± 0% -95.93% (p=0.000 n=10) DiffStack/64_x_131_kB-10 2586.04Mi ± 0% 70.48Mi ± 1% -97.27% (p=0.000 n=10) DiffStack/1_x_524_kB-10 11.854Mi ± 1% 2.127Mi ± 0% -82.06% (p=0.000 n=10) DiffStack/2_x_524_kB-10 32.932Mi ± 1% 6.558Mi ± 1% -80.09% (p=0.000 n=10) DiffStack/4_x_524_kB-10 90.04Mi ± 0% 15.15Mi ± 0% -83.17% (p=0.000 n=10) DiffStack/8_x_524_kB-10 252.94Mi ± 0% 32.27Mi ± 1% -87.24% (p=0.000 n=10) DiffStack/16_x_524_kB-10 801.77Mi ± 0% 66.46Mi ± 1% -91.71% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 103.67Mi ± 0% 35.52Mi ± 0% -65.74% (p=0.000 n=10) DiffStackRecorded/checkpoints.json-10 913.90Mi ± 0% 53.53Mi ± 1% -94.14% (p=0.000 n=10) geomean 19.16Mi 2.409Mi -87.43% │ buffer-pool.txt │ spanned.txt │ │ allocs/op │ allocs/op vs base │ DiffStack/1_x_2_B-10 344.0 ± 0% 184.0 ± 0% -46.51% (p=0.000 n=10) DiffStack/2_x_2_B-10 616.0 ± 0% 295.0 ± 0% -52.11% (p=0.000 n=10) DiffStack/4_x_2_B-10 1381.0 ± 0% 524.0 ± 0% -62.06% (p=0.000 n=10) DiffStack/8_x_2_B-10 3.922k ± 0% 1.032k ± 0% -73.69% (p=0.000 n=10) DiffStack/16_x_2_B-10 12.791k ± 0% 2.233k ± 0% -82.54% (p=0.000 n=10) DiffStack/32_x_2_B-10 43.416k ± 0% 5.410k ± 0% -87.54% (p=0.000 n=10) DiffStack/48_x_2_B-10 95.803k ± 0% 9.036k ± 0% -90.57% (p=0.000 n=10) DiffStack/64_x_2_B-10 156.99k ± 0% 13.99k ± 0% -91.09% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 335.0 ± 0% 182.0 ± 0% -45.67% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 597.5 ± 0% 295.0 ± 0% -50.63% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 1304.0 ± 0% 527.0 ± 0% -59.59% (p=0.000 n=10) DiffStack/8_x_8.2_kB-10 3.475k ± 0% 1.036k ± 0% -70.19% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 10.677k ± 0% 2.239k ± 0% -79.03% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 35.357k ± 0% 5.414k ± 0% -84.69% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 72.782k ± 0% 9.296k ± 0% -87.23% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 119.24k ± 0% 14.26k ± 0% -88.04% (p=0.000 n=10) DiffStack/1_x_33_kB-10 309.0 ± 0% 182.0 ± 0% -41.10% (p=0.000 n=10) DiffStack/2_x_33_kB-10 594.0 ± 0% 296.0 ± 0% -50.17% (p=0.000 n=10) DiffStack/4_x_33_kB-10 1324.5 ± 0% 528.0 ± 0% -60.14% (p=0.000 n=10) DiffStack/8_x_33_kB-10 3.554k ± 0% 1.047k ± 0% -70.54% (p=0.000 n=10) DiffStack/16_x_33_kB-10 10.511k ± 0% 2.232k ± 0% -78.77% (p=0.000 n=10) DiffStack/32_x_33_kB-10 35.147k ± 0% 5.194k ± 0% -85.22% (p=0.000 n=10) DiffStack/48_x_33_kB-10 66.519k ± 0% 8.958k ± 0% -86.53% (p=0.000 n=10) DiffStack/64_x_33_kB-10 110.71k ± 0% 13.28k ± 0% -88.00% (p=0.000 n=10) DiffStack/2_x_131_kB-10 644.0 ± 0% 296.0 ± 0% -54.04% (p=0.000 n=10) DiffStack/4_x_131_kB-10 1418.5 ± 0% 535.0 ± 0% -62.28% (p=0.000 n=10) DiffStack/8_x_131_kB-10 3.497k ± 0% 1.047k ± 0% -70.06% (p=0.000 n=10) DiffStack/16_x_131_kB-10 9.805k ± 0% 2.182k ± 0% -77.75% (p=0.000 n=10) DiffStack/32_x_131_kB-10 30.996k ± 0% 5.040k ± 0% -83.74% (p=0.000 n=10) DiffStack/48_x_131_kB-10 63.751k ± 0% 8.656k ± 0% -86.42% (p=0.000 n=10) DiffStack/64_x_131_kB-10 107.97k ± 0% 13.03k ± 0% -87.93% (p=0.000 n=10) DiffStack/1_x_524_kB-10 356.0 ± 0% 186.0 ± 0% -47.75% (p=0.000 n=10) DiffStack/2_x_524_kB-10 671.0 ± 0% 300.0 ± 0% -55.29% (p=0.000 n=10) DiffStack/4_x_524_kB-10 1467.5 ± 0% 536.0 ± 0% -63.48% (p=0.000 n=10) DiffStack/8_x_524_kB-10 3.553k ± 0% 1.038k ± 0% -70.80% (p=0.000 n=10) DiffStack/16_x_524_kB-10 9.971k ± 0% 2.182k ± 0% -78.12% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 489.16k ± 0% 73.65k ± 0% -84.94% (p=0.000 n=10) DiffStackRecorded/checkpoints.json-10 7236.9k ± 0% 857.8k ± 0% -88.15% (p=0.000 n=10) geomean 7.989k 1.934k -75.79% ``` |
|
Pat Gavlin | 2c1c4e6ce1 |
[snapshot] Pool checkpoint buffers
Pool the buffers used for marshaling checkpoints. The pooling works by returning the last saved checkpoint's buffer to the diff state when a new checkpoint is saved. If a pooled buffer is available, it is reused when a checkpoint is marshaled. With our current usage patterns, this will cycle through two buffers. Starting from a fresh diff state: - The first call to MarshalDeployment will create a new buffer B1 - The first call to Saved will put B1 into lastSavedDeployment - The second call to MarshalDeployment will create a new buffer B2 - The second call to Saved will put B1 into the pool and B2 into lastSavedDeployment - The third call to MarshalDeployment will get B1 from the pool - The third call to Saved will put B2 into the pool and B1 into lastSavedDeployment These changes reduce CPU time and allocation volume by a geomean of 6% and 27%, respectively. Benchmark results and analysis: httpstate ❯ go test -count=10 -run none -benchmem -bench . | tee buffer-pool.txt httpstate ❯ benchstat base.txt buffer-pool.txt >stat.txt ``` │ base.txt │ buffer-pool.txt │ │ sec/op │ sec/op vs base │ DiffStack/1_x_2_B-10 46.30µ ± 1% 43.17µ ± 0% -6.74% (p=0.000 n=10) DiffStack/2_x_2_B-10 82.68µ ± 1% 80.07µ ± 0% -3.16% (p=0.000 n=10) DiffStack/4_x_2_B-10 191.0µ ± 1% 175.3µ ± 1% -8.22% (p=0.000 n=10) DiffStack/8_x_2_B-10 499.7µ ± 0% 470.6µ ± 1% -5.83% (p=0.000 n=10) DiffStack/16_x_2_B-10 1.489m ± 0% 1.400m ± 1% -5.92% (p=0.000 n=10) DiffStack/32_x_2_B-10 4.895m ± 1% 4.540m ± 0% -7.25% (p=0.000 n=10) DiffStack/48_x_2_B-10 9.651m ± 1% 8.538m ± 1% -11.53% (p=0.000 n=10) DiffStack/64_x_2_B-10 15.11m ± 0% 13.82m ± 0% -8.56% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 103.3µ ± 1% 123.9µ ± 1% +19.93% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 308.8µ ± 0% 317.1µ ± 1% +2.70% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 899.2µ ± 0% 835.6µ ± 0% -7.07% (p=0.000 n=10) DiffStack/8_x_8.2_kB-10 2.989m ± 1% 2.757m ± 1% -7.76% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 10.528m ± 0% 9.305m ± 1% -11.61% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 36.27m ± 1% 30.55m ± 1% -15.77% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 72.53m ± 0% 69.05m ± 0% -4.80% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 125.5m ± 0% 113.5m ± 0% -9.58% (p=0.000 n=10) DiffStack/1_x_33_kB-10 317.4µ ± 1% 314.2µ ± 0% -1.01% (p=0.001 n=10) DiffStack/2_x_33_kB-10 756.8µ ± 1% 797.3µ ± 1% +5.35% (p=0.000 n=10) DiffStack/4_x_33_kB-10 2.605m ± 0% 2.371m ± 0% -8.99% (p=0.000 n=10) DiffStack/8_x_33_kB-10 9.241m ± 2% 8.342m ± 0% -9.73% (p=0.000 n=10) DiffStack/16_x_33_kB-10 32.86m ± 1% 27.11m ± 1% -17.50% (p=0.000 n=10) DiffStack/32_x_33_kB-10 110.2m ± 2% 101.3m ± 1% -8.01% (p=0.000 n=10) DiffStack/48_x_33_kB-10 231.3m ± 0% 202.1m ± 1% -12.64% (p=0.000 n=10) DiffStack/64_x_33_kB-10 383.8m ± 1% 333.4m ± 1% -13.14% (p=0.000 n=10) DiffStack/2_x_131_kB-10 2.748m ± 1% 2.678m ± 0% -2.52% (p=0.000 n=10) DiffStack/4_x_131_kB-10 8.609m ± 0% 8.278m ± 0% -3.85% (p=0.000 n=10) DiffStack/8_x_131_kB-10 28.44m ± 0% 25.46m ± 0% -10.49% (p=0.000 n=10) DiffStack/16_x_131_kB-10 96.98m ± 1% 88.22m ± 0% -9.03% (p=0.000 n=10) DiffStack/32_x_131_kB-10 349.4m ± 1% 315.3m ± 0% -9.74% (p=0.000 n=10) DiffStack/48_x_131_kB-10 761.7m ± 0% 677.9m ± 1% -11.00% (p=0.000 n=10) DiffStack/64_x_131_kB-10 1.312 ± 1% 1.195 ± 0% -8.90% (p=0.000 n=10) DiffStack/1_x_524_kB-10 3.238m ± 1% 3.213m ± 0% -0.78% (p=0.000 n=10) DiffStack/2_x_524_kB-10 9.867m ± 0% 9.613m ± 0% -2.58% (p=0.000 n=10) DiffStack/4_x_524_kB-10 29.65m ± 0% 29.11m ± 0% -1.83% (p=0.000 n=10) DiffStack/8_x_524_kB-10 98.91m ± 1% 92.23m ± 0% -6.76% (p=0.000 n=10) DiffStack/16_x_524_kB-10 348.1m ± 1% 322.5m ± 1% -7.34% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 46.62m ± 0% 46.42m ± 1% -0.42% (p=0.004 n=10) DiffStackRecorded/checkpoints.json-10 835.8m ± 0% 809.5m ± 0% -3.15% (p=0.000 n=10) geomean 11.15m 10.44m -6.39% │ base.txt │ buffer-pool.txt │ │ ratio │ ratio vs base │ DiffStack/1_x_2_B-10 789.5m ± 0% 700.9m ± 0% -11.22% (p=0.000 n=10) DiffStack/2_x_2_B-10 1.034 ± 0% 1.030 ± 0% -0.39% (p=0.000 n=10) DiffStack/4_x_2_B-10 1.676 ± 0% 1.608 ± 0% -4.06% (p=0.000 n=10) DiffStack/8_x_2_B-10 2.883 ± 0% 2.879 ± 0% -0.14% (p=0.000 n=10) DiffStack/16_x_2_B-10 5.270 ± 0% 5.268 ± 0% -0.04% (p=0.000 n=10) DiffStack/32_x_2_B-10 10.20 ± 0% 10.03 ± 0% -1.67% (p=0.000 n=10) DiffStack/48_x_2_B-10 14.24 ± 0% 13.39 ± 0% -5.97% (p=0.000 n=10) DiffStack/64_x_2_B-10 17.51 ± 0% 17.34 ± 0% -0.97% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 950.9m ± 0% 1383.0m ± 0% +45.44% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 1.716 ± 0% 1.900 ± 0% +10.72% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 2.792 ± 0% 2.792 ± 0% ~ (p=1.000 n=10) ¹ DiffStack/8_x_8.2_kB-10 5.350 ± 0% 5.516 ± 0% +3.10% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 10.62 ± 0% 10.60 ± 0% -0.19% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 20.01 ± 0% 19.10 ± 0% -4.55% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 27.48 ± 0% 30.24 ± 0% +10.04% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 36.97 ± 0% 38.22 ± 0% +3.38% (p=0.000 n=10) DiffStack/1_x_33_kB-10 1.467 ± 0% 1.467 ± 0% ~ (p=1.000 n=10) ¹ DiffStack/2_x_33_kB-10 1.581 ± 0% 1.777 ± 0% +12.40% (p=0.000 n=10) DiffStack/4_x_33_kB-10 2.967 ± 0% 2.967 ± 0% ~ (p=1.000 n=10) ¹ DiffStack/8_x_33_kB-10 5.766 ± 0% 5.809 ± 0% +0.75% (p=0.000 n=10) DiffStack/16_x_33_kB-10 10.87 ± 0% 10.05 ± 0% -7.54% (p=0.000 n=10) DiffStack/32_x_33_kB-10 19.00 ± 0% 20.13 ± 0% +5.95% (p=0.000 n=10) DiffStack/48_x_33_kB-10 27.14 ± 0% 27.18 ± 0% +0.15% (p=0.000 n=10) DiffStack/64_x_33_kB-10 34.39 ± 0% 33.97 ± 0% -1.22% (p=0.000 n=10) DiffStack/2_x_131_kB-10 1.993 ± 0% 1.993 ± 0% ~ (p=1.000 n=10) ¹ DiffStack/4_x_131_kB-10 3.173 ± 0% 3.263 ± 0% +2.84% (p=0.000 n=10) DiffStack/8_x_131_kB-10 5.334 ± 0% 5.291 ± 0% -0.81% (p=0.000 n=10) DiffStack/16_x_131_kB-10 9.296 ± 0% 9.423 ± 0% +1.37% (p=0.000 n=10) DiffStack/32_x_131_kB-10 17.10 ± 0% 17.21 ± 0% +0.64% (p=0.000 n=10) DiffStack/48_x_131_kB-10 25.25 ± 0% 25.06 ± 0% -0.75% (p=0.000 n=10) DiffStack/64_x_131_kB-10 33.13 ± 0% 33.46 ± 0% +1.00% (p=0.000 n=10) DiffStack/1_x_524_kB-10 1.498 ± 0% 1.498 ± 0% ~ (p=1.000 n=10) ¹ DiffStack/2_x_524_kB-10 1.998 ± 0% 1.998 ± 0% ~ (p=1.000 n=10) ¹ DiffStack/4_x_524_kB-10 2.998 ± 0% 3.089 ± 0% +3.04% (p=0.000 n=10) DiffStack/8_x_524_kB-10 5.084 ± 0% 5.084 ± 0% ~ (p=1.000 n=10) ¹ DiffStack/16_x_524_kB-10 9.079 ± 0% 9.058 ± 0% -0.23% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 1.997 ± 0% 1.997 ± 0% ~ (p=1.000 n=10) ¹ DiffStackRecorded/checkpoints.json-10 36.60 ± 0% 36.60 ± 0% ~ (p=1.000 n=10) ¹ geomean 5.961 6.039 +1.31% ¹ all samples are equal │ base.txt │ buffer-pool.txt │ │ B/op │ B/op vs base │ DiffStack/1_x_2_B-10 35.17Ki ± 0% 29.56Ki ± 0% -15.97% (p=0.000 n=10) DiffStack/2_x_2_B-10 70.05Ki ± 0% 61.61Ki ± 0% -12.04% (p=0.000 n=10) DiffStack/4_x_2_B-10 186.8Ki ± 0% 149.9Ki ± 0% -19.75% (p=0.000 n=10) DiffStack/8_x_2_B-10 529.0Ki ± 0% 432.0Ki ± 0% -18.33% (p=0.000 n=10) DiffStack/16_x_2_B-10 1.768Mi ± 0% 1.392Mi ± 0% -21.28% (p=0.000 n=10) DiffStack/32_x_2_B-10 6.513Mi ± 0% 4.986Mi ± 0% -23.45% (p=0.000 n=10) DiffStack/48_x_2_B-10 14.054Mi ± 0% 9.885Mi ± 0% -29.66% (p=0.000 n=10) DiffStack/64_x_2_B-10 22.39Mi ± 0% 16.84Mi ± 0% -24.79% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 154.7Ki ± 0% 202.4Ki ± 0% +30.89% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 571.9Ki ± 0% 540.8Ki ± 0% -5.43% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 1.846Mi ± 0% 1.430Mi ± 0% -22.52% (p=0.000 n=10) DiffStack/8_x_8.2_kB-10 6.920Mi ± 0% 4.809Mi ± 0% -30.51% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 26.50Mi ± 0% 16.34Mi ± 0% -38.35% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 98.33Mi ± 0% 55.24Mi ± 0% -43.82% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 209.4Mi ± 0% 127.2Mi ± 0% -39.23% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 365.6Mi ± 0% 211.2Mi ± 0% -42.23% (p=0.000 n=10) DiffStack/1_x_33_kB-10 808.7Ki ± 0% 764.2Ki ± 0% -5.50% (p=0.000 n=10) DiffStack/2_x_33_kB-10 2.013Mi ± 0% 1.992Mi ± 0% -1.03% (p=0.000 n=10) DiffStack/4_x_33_kB-10 7.551Mi ± 0% 5.764Mi ± 0% -23.67% (p=0.000 n=10) DiffStack/8_x_33_kB-10 28.58Mi ± 0% 18.90Mi ± 0% -33.87% (p=0.000 n=10) DiffStack/16_x_33_kB-10 104.40Mi ± 0% 59.42Mi ± 0% -43.08% (p=0.000 n=10) DiffStack/32_x_33_kB-10 365.5Mi ± 0% 219.9Mi ± 0% -39.82% (p=0.000 n=10) DiffStack/48_x_33_kB-10 801.1Mi ± 0% 438.7Mi ± 0% -45.24% (p=0.000 n=10) DiffStack/64_x_33_kB-10 1314.3Mi ± 0% 718.3Mi ± 0% -45.35% (p=0.000 n=10) DiffStack/2_x_131_kB-10 9.179Mi ± 0% 8.063Mi ± 0% -12.16% (p=0.000 n=10) DiffStack/4_x_131_kB-10 29.14Mi ± 0% 22.68Mi ± 0% -22.18% (p=0.000 n=10) DiffStack/8_x_131_kB-10 98.77Mi ± 0% 65.16Mi ± 0% -34.03% (p=0.000 n=10) DiffStack/16_x_131_kB-10 337.1Mi ± 0% 207.8Mi ± 0% -38.36% (p=0.000 n=10) DiffStack/32_x_131_kB-10 1234.6Mi ± 0% 705.3Mi ± 0% -42.87% (p=0.000 n=10) DiffStack/48_x_131_kB-10 2.771Gi ± 0% 1.468Gi ± 0% -47.03% (p=0.000 n=10) DiffStack/64_x_131_kB-10 4.606Gi ± 0% 2.553Gi ± 0% -44.56% (p=0.000 n=10) DiffStack/1_x_524_kB-10 12.41Mi ± 1% 11.85Mi ± 1% -4.50% (p=0.000 n=10) DiffStack/2_x_524_kB-10 37.01Mi ± 1% 32.79Mi ± 1% -11.41% (p=0.000 n=10) DiffStack/4_x_524_kB-10 110.56Mi ± 0% 89.08Mi ± 1% -19.43% (p=0.000 n=10) DiffStack/8_x_524_kB-10 367.3Mi ± 1% 254.3Mi ± 1% -30.76% (p=0.000 n=10) DiffStack/16_x_524_kB-10 1291.7Mi ± 0% 801.6Mi ± 0% -37.95% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 69.25Mi ± 0% 69.25Mi ± 0% ~ (p=0.165 n=10) DiffStackRecorded/checkpoints.json-10 1226.7Mi ± 0% 914.2Mi ± 0% -25.47% (p=0.000 n=10) geomean 26.14Mi 19.06Mi -27.07% │ base.txt │ buffer-pool.txt │ │ allocs/op │ allocs/op vs base │ DiffStack/1_x_2_B-10 361.0 ± 0% 326.0 ± 0% -9.70% (p=0.000 n=10) DiffStack/2_x_2_B-10 649.0 ± 0% 621.0 ± 0% -4.31% (p=0.000 n=10) DiffStack/4_x_2_B-10 1.512k ± 0% 1.399k ± 0% -7.47% (p=0.000 n=10) DiffStack/8_x_2_B-10 4.141k ± 0% 4.006k ± 0% -3.25% (p=0.000 n=10) DiffStack/16_x_2_B-10 13.09k ± 0% 12.73k ± 0% -2.72% (p=0.000 n=10) DiffStack/32_x_2_B-10 46.38k ± 0% 44.88k ± 0% -3.25% (p=0.000 n=10) DiffStack/48_x_2_B-10 94.38k ± 0% 87.94k ± 0% -6.82% (p=0.000 n=10) DiffStack/64_x_2_B-10 152.7k ± 0% 149.3k ± 0% -2.18% (p=0.000 n=10) DiffStack/1_x_8.2_kB-10 323.0 ± 0% 335.0 ± 0% +3.72% (p=0.000 n=10) DiffStack/2_x_8.2_kB-10 630.0 ± 0% 613.0 ± 0% -2.70% (p=0.000 n=10) DiffStack/4_x_8.2_kB-10 1.354k ± 0% 1.279k ± 0% -5.54% (p=0.000 n=10) DiffStack/8_x_8.2_kB-10 3.606k ± 0% 3.490k ± 0% -3.22% (p=0.000 n=10) DiffStack/16_x_8.2_kB-10 11.16k ± 0% 10.71k ± 0% -4.11% (p=0.000 n=10) DiffStack/32_x_8.2_kB-10 36.45k ± 0% 34.06k ± 0% -6.54% (p=0.000 n=10) DiffStack/48_x_8.2_kB-10 71.55k ± 0% 76.00k ± 0% +6.22% (p=0.000 n=10) DiffStack/64_x_8.2_kB-10 123.9k ± 0% 125.2k ± 0% +1.07% (p=0.000 n=10) DiffStack/1_x_33_kB-10 346.0 ± 0% 329.0 ± 0% -4.91% (p=0.000 n=10) DiffStack/2_x_33_kB-10 607.0 ± 0% 594.0 ± 0% -2.14% (p=0.000 n=10) DiffStack/4_x_33_kB-10 1.396k ± 0% 1.310k ± 0% -6.16% (p=0.000 n=10) DiffStack/8_x_33_kB-10 3.764k ± 0% 3.564k ± 0% -5.30% (p=0.000 n=10) DiffStack/16_x_33_kB-10 11.26k ± 0% 10.15k ± 0% -9.88% (p=0.000 n=10) DiffStack/32_x_33_kB-10 34.55k ± 0% 34.91k ± 0% +1.03% (p=0.000 n=10) DiffStack/48_x_33_kB-10 69.72k ± 0% 67.76k ± 0% -2.82% (p=0.000 n=10) DiffStack/64_x_33_kB-10 114.1k ± 0% 110.1k ± 0% -3.46% (p=0.000 n=10) DiffStack/2_x_131_kB-10 678.0 ± 0% 643.0 ± 0% -5.16% (p=0.000 n=10) DiffStack/4_x_131_kB-10 1.494k ± 0% 1.419k ± 0% -4.99% (p=0.000 n=10) DiffStack/8_x_131_kB-10 3.737k ± 0% 3.489k ± 0% -6.64% (p=0.000 n=10) DiffStack/16_x_131_kB-10 10.400k ± 0% 9.917k ± 0% -4.64% (p=0.000 n=10) DiffStack/32_x_131_kB-10 32.24k ± 0% 31.08k ± 0% -3.59% (p=0.000 n=10) DiffStack/48_x_131_kB-10 66.12k ± 0% 63.62k ± 0% -3.79% (p=0.000 n=10) DiffStack/64_x_131_kB-10 110.9k ± 0% 109.1k ± 0% -1.67% (p=0.000 n=10) DiffStack/1_x_524_kB-10 374.0 ± 1% 356.0 ± 0% -4.81% (p=0.000 n=10) DiffStack/2_x_524_kB-10 707.0 ± 0% 668.0 ± 0% -5.52% (p=0.000 n=10) DiffStack/4_x_524_kB-10 1.528k ± 0% 1.460k ± 1% -4.48% (p=0.000 n=10) DiffStack/8_x_524_kB-10 3.797k ± 1% 3.571k ± 0% -5.95% (p=0.000 n=10) DiffStack/16_x_524_kB-10 10.562k ± 1% 9.966k ± 0% -5.65% (p=0.000 n=10) DiffStackRecorded/two-large-checkpoints.json-10 512.1k ± 0% 512.1k ± 0% ~ (p=0.222 n=10) DiffStackRecorded/checkpoints.json-10 7.240M ± 0% 7.237M ± 0% -0.03% (p=0.000 n=10) geomean 8.309k 7.996k -3.77% ``` |
|
Abhinav Gupta |
7aa5b77a0c
|
all: Reformat with gofumpt
Per team discussion, switching to gofumpt. [gofumpt][1] is an alternative, stricter alternative to gofmt. It addresses other stylistic concerns that gofmt doesn't yet cover. [1]: https://github.com/mvdan/gofumpt See the full list of [Added rules][2], but it includes: - Dropping empty lines around function bodies - Dropping unnecessary variable grouping when there's only one variable - Ensuring an empty line between multi-line functions - simplification (`-s` in gofmt) is always enabled - Ensuring multi-line function signatures end with `) {` on a separate line. [2]: https://github.com/mvdan/gofumpt#Added-rules gofumpt is stricter, but there's no lock-in. All gofumpt output is valid gofmt output, so if we decide we don't like it, it's easy to switch back without any code changes. gofumpt support is built into the tooling we use for development so this won't change development workflows. - golangci-lint includes a gofumpt check (enabled in this PR) - gopls, the LSP for Go, includes a gofumpt option (see [installation instrutions][3]) [3]: https://github.com/mvdan/gofumpt#installation This change was generated by running: ```bash gofumpt -w $(rg --files -g '*.go' | rg -v testdata | rg -v compilation_error) ``` The following files were manually tweaked afterwards: - pkg/cmd/pulumi/stack_change_secrets_provider.go: one of the lines overflowed and had comments in an inconvenient place - pkg/cmd/pulumi/destroy.go: `var x T = y` where `T` wasn't necessary - pkg/cmd/pulumi/policy_new.go: long line because of error message - pkg/backend/snapshot_test.go: long line trying to assign three variables in the same assignment I have included mention of gofumpt in the CONTRIBUTING.md. |
|
Anton Tayanovskyy | 6061b4718e | Reduce memory use for PULUMI_OPTIMIZED_CHECKPOINT_PATCH | |
Anton Tayanovskyy | 401c58dd80 | Opt into delta PATCH updates based on capability API | |
Anton Tayanovskyy | 73888c2908 | Support PULUMI_OPTIMIZED_CHECKPOINT_PATCH protocol |