Compare commits

..

14 Commits

Author SHA1 Message Date
36398bc3f3 Only check TRIGGERED_BUILDKITE_TAG 2019-01-07 19:53:52 -08:00
fa0e1ad356 Don't turn the build red if channel cannot be figured (eg, building a tag) 2019-01-07 19:53:39 -08:00
f56c5dacca Remove some metrics datapoint, as it was causing excessive logging (#2287) (#2291)
- 100 nodes test was bringing down the influx DB server
2019-01-03 10:42:13 -08:00
80e0da132a Rename getConfirmation -> getConfirmationTime 2018-12-22 13:11:10 -08:00
f89debdfa6 Document getConfirmationTime 2018-12-22 13:11:10 -08:00
16f7e46fce Ignore error while enabling nvidia persistence mode (#2265) 2018-12-21 12:47:45 -08:00
3a039c8007 Load nvidia drivers on node startup (#2263) (#2264)
* Load nvidia drivers on node startup

* added new script to enable nvidia driver persistent mode

* remove set -ex
2018-12-21 11:58:06 -08:00
56d5324837 Use CUDA for testnet automation performance calculations (#2259) (#2261) 2018-12-21 05:14:08 -08:00
d3bf0fc707 Use SSD for testnet automation (#2257) (#2258) 2018-12-21 04:52:00 -08:00
f9d8a1d6c0 Rename finality to confirmation (#2250)
* Rename finality to confirmation

* fix cargo fmt errors
2018-12-20 16:03:56 -08:00
70559253ee Use newer votes to calculate confirmation time (#2247) 2018-12-20 16:03:56 -08:00
9c61abe468 Reduce ticks per block to increase voting frequency (#2242) 2018-12-20 16:03:56 -08:00
970954ac3b Stable dashboard can now actually come from the stable channel 2018-12-20 08:06:02 -08:00
39d821ead8 Select correct branch for {testnet,-perf} when using a stable channel tag 2018-12-19 17:47:18 -08:00
2516 changed files with 53794 additions and 416948 deletions

View File

@ -1 +0,0 @@
/secrets_unencrypted.ejson

View File

@ -1,7 +1,10 @@
{
"_public_key": "ae29f4f7ad2fc92de70d470e411c8426d5d48db8817c9e3dae574b122192335f",
"_comment": "These credentials are encrypted and pose no risk",
"environment": {
"CODECOV_TOKEN": "EJ[1:Z7OneT3RdJJ0DipCHQ7rC84snQ+FPbgHwZADQiz54wk=:3K68mE38LJ2RB98VWmjuNLFBNn1XTGR4:cR4r05/TOZQKmEZp1v4CSgUJtC6QJiOaL85QjXW0qZ061fMnsBA8AtAPMDoDq4WCGOZM1A==]"
"CODECOV_TOKEN": "EJ[1:Kqnm+k1Z4p8nr7GqMczXnzh6azTk39tj3bAbCKPitUc=:EzVa4Gpj2Qn5OhZQlVfGFchuROgupvnW:CbWc6sNh1GCrAbrncxDjW00zUAD/Sa+ccg7CFSz8Ua6LnCYnSddTBxJWcJEbEs0MrjuZRQ==]",
"CRATES_IO_TOKEN": "EJ[1:Kqnm+k1Z4p8nr7GqMczXnzh6azTk39tj3bAbCKPitUc=:qF7QrUM8j+19mptcE1YS71CqmrCM13Ah:TZCatJeT1egCHiufE6cGFC1VsdJkKaaqV6QKWkEsMPBKvOAdaZbbVz9Kl+lGnIsF]",
"INFLUX_DATABASE": "EJ[1:Kqnm+k1Z4p8nr7GqMczXnzh6azTk39tj3bAbCKPitUc=:PetD/4c/EbkQmFEcK21g3cBBAPwFqHEw:wvYmDZRajy2WngVFs9AlwyHk]",
"INFLUX_USERNAME": "EJ[1:Kqnm+k1Z4p8nr7GqMczXnzh6azTk39tj3bAbCKPitUc=:WcnqZdmDFtJJ01Zu5LbeGgbYGfRzBdFc:a7c5zDDtCOu5L1Qd2NKkxT6kljyBcbck]",
"INFLUX_PASSWORD": "EJ[1:Kqnm+k1Z4p8nr7GqMczXnzh6azTk39tj3bAbCKPitUc=:LIZgP9Tp9yE9OlpV8iogmLOI7iW7SiU3:x0nYdT1A6sxu+O+MMLIN19d2t6rrK1qJ3+HnoWG3PDodsXjz06YJWQKU/mx6saqH+QbGtGV5mk0=]"
}
}

View File

@ -1,42 +1,2 @@
CI_BUILD_START=$(date +%s)
export CI_BUILD_START
source ci/env.sh
#
# Kill any running docker containers, which are potentially left over from the
# previous CI job
#
(
containers=$(docker ps -q)
if [[ $(hostname) != metrics-solana-com && -n $containers ]]; then
echo "+++ Killing stale docker containers"
docker ps
# shellcheck disable=SC2086 # Don't want to double quote $containers
docker kill $containers
fi
)
# Processes from previously aborted CI jobs seem to loiter, unclear why as one
# would expect the buildkite-agent to clean up all child processes of the
# aborted CI job.
# But as a workaround for now manually kill some known loiterers. These
# processes will all have the `init` process as their PPID:
(
victims=
for name in bash cargo docker solana; do
victims="$victims $(pgrep -u "$(id -u)" -P 1 -d \ $name)"
done
for victim in $victims; do
echo "Killing pid $victim"
kill -9 "$victim" || true
done
)
# HACK: These are in our docker images, need to be removed from CARGO_HOME
# because we try to cache downloads across builds with CARGO_HOME
# cargo lacks a facility for "system" tooling, always tries CARGO_HOME first
cargo uninstall cargo-audit || true
cargo uninstall svgbob_cli || true
cargo uninstall mdbook || true

View File

@ -3,19 +3,14 @@
#
# Save target/ for the next CI build on this machine
#
if [[ -z $CARGO_TARGET_CACHE ]]; then
echo "+++ CARGO_TARGET_CACHE not defined" # pre-command should have defined it
else
(
set -x
mkdir -p "$CARGO_TARGET_CACHE"
set -x
rsync -a --delete --link-dest="$PWD" target "$CARGO_TARGET_CACHE"
du -hs "$CARGO_TARGET_CACHE"
read -r cacheSizeInGB _ < <(du -s --block-size=1800000000 "$CARGO_TARGET_CACHE")
echo "--- ${cacheSizeInGB}GB: $CARGO_TARGET_CACHE"
)
fi
(
set -x
d=$HOME/cargo-target-cache/"$BUILDKITE_LABEL"
mkdir -p "$d"
set -x
rsync -a --delete --link-dest="$PWD" target "$d"
du -hs "$d"
)
#
# Add job_stats data point

View File

@ -11,29 +11,19 @@ export PS4="++"
#
# Restore target/ from the previous CI build on this machine
#
eval "$(ci/channel-info.sh)"
export CARGO_TARGET_CACHE=$HOME/cargo-target-cache/"$CHANNEL"-"$BUILDKITE_LABEL"
(
set -x
MAX_CACHE_SIZE=18 # gigabytes
d=$HOME/cargo-target-cache/"$BUILDKITE_LABEL"
if [[ -d $CARGO_TARGET_CACHE ]]; then
du -hs "$CARGO_TARGET_CACHE"
read -r cacheSizeInGB _ < <(du -s --block-size=1800000000 "$CARGO_TARGET_CACHE")
echo "--- ${cacheSizeInGB}GB: $CARGO_TARGET_CACHE"
if [[ $cacheSizeInGB -gt $MAX_CACHE_SIZE ]]; then
echo "--- $CARGO_TARGET_CACHE is too large, removing it"
rm -rf "$CARGO_TARGET_CACHE"
if [[ -d $d ]]; then
du -hs "$d"
read -r cacheSizeInGB _ < <(du -s --block-size=1000000000 "$d")
if [[ $cacheSizeInGB -gt 10 ]]; then
echo "$d has gotten too large, removing it"
rm -rf "$d"
fi
else
echo "--- $CARGO_TARGET_CACHE not present"
fi
mkdir -p "$CARGO_TARGET_CACHE"/target
rsync -a --delete --link-dest="$CARGO_TARGET_CACHE" "$CARGO_TARGET_CACHE"/target .
# Don't reuse BPF target build artifacts due to incremental build issues with
# `std:
# "found possibly newer version of crate `std` which `xyz` depends on
rm -rf target/bpfel-unknown-unknown
mkdir -p "$d"/target
rsync -a --delete --link-dest="$d" "$d"/target .
)

View File

@ -9,10 +9,12 @@
set -e
cd "$(dirname "$0")"/..
source ci/_
_ ci/buildkite-pipeline.sh pipeline.yml
echo +++ pipeline
cat pipeline.yml
buildkite-agent pipeline upload ci/buildkite.yml
if [[ $BUILDKITE_BRANCH =~ ^pull ]]; then
# Add helpful link back to the corresponding Github Pull Request
buildkite-agent annotate --style info --context pr-backlink \
"Github Pull Request: https://github.com/solana-labs/solana/$BUILDKITE_BRANCH"
fi
_ buildkite-agent pipeline upload pipeline.yml

View File

@ -1,12 +1,5 @@
ignore:
- "src/bin"
coverage:
range: 50..100
round: down
precision: 1
status:
project: off
patch: off
comment:
layout: "diff"
behavior: default
require_changes: no

View File

@ -1,41 +0,0 @@
# To get started with Dependabot version updates, you'll need to specify which
# package ecosystems to update and where the package manifests are located.
# Please see the documentation for all configuration options:
# https://help.github.com/github/administering-a-repository/configuration-options-for-dependency-updates
version: 2
updates:
- package-ecosystem: cargo
directory: "/"
schedule:
interval: daily
time: "01:00"
timezone: America/Los_Angeles
#labels:
# - "automerge"
open-pull-requests-limit: 3
- package-ecosystem: npm
directory: "/web3.js"
schedule:
interval: daily
time: "01:00"
timezone: America/Los_Angeles
labels:
- "automerge"
commit-message:
prefix: "chore:"
open-pull-requests-limit: 3
- package-ecosystem: npm
directory: "/explorer"
schedule:
interval: daily
time: "01:00"
timezone: America/Los_Angeles
labels:
- "automerge"
commit-message:
prefix: "chore:"
include: "scope"
open-pull-requests-limit: 3

39
.github/stale.yml vendored
View File

@ -1,39 +0,0 @@
# Issues with these labels will never be considered stale
exemptLabels:
- security
- blocked
# Label to use when marking a pull request as stale
staleLabel: stale
pulls:
# Number of days of inactivity before a pull request becomes stale
daysUntilStale: 7
# Number of days of inactivity before a stale pull request is closed
daysUntilClose: 7
# Comment to post when marking a pull request as stale. Set to `false` to disable
markComment: >
This pull request has been automatically marked as stale because it has not had
recent activity. It will be closed if no further activity occurs.
# Comment to post when closing a stale pull request. Set to `false` to disable
closeComment: >
This stale pull request has been automatically closed.
Thank you for your contributions.
issues:
# Number of days of inactivity before a issue becomes stale
daysUntilStale: 365
# Number of days of inactivity before a stale issue is closed
daysUntilClose: 7
# Comment to post when marking a issue as stale. Set to `false` to disable
markComment: >
This issue has been automatically marked as stale because it has not had
recent activity. It will be closed if no further activity occurs.
# Comment to post when closing a stale issue. Set to `false` to disable
closeComment: >
This stale issue has been automatically closed.
Thank you for your contributions.

25
.gitignore vendored
View File

@ -1,33 +1,20 @@
/docs/html/
/docs/src/tests.ok
/docs/src/cli/usage.md
/docs/src/.gitbook/assets/*.svg
/farf/
/solana-release/
/solana-release.tar.bz2
/solana-metrics/
/solana-metrics.tar.bz2
/target/
/book/html/
/book/src/img/
/book/src/tests.ok
**/*.rs.bk
.cargo
# node config that is rsynced
/config/
.cache
# node config that remains local
/config-local/
# log files
*.log
log-*.txt
log-*/
# intellij files
/.idea/
/solana.iml
/.vscode/
# fetch-spl.sh artifacts
/spl-genesis-args.sh
/spl_*.so
.DS_Store

View File

@ -1,84 +0,0 @@
# Validate your changes with:
#
# $ curl -F 'data=@.mergify.yml' https://gh.mergify.io/validate/
#
# https://doc.mergify.io/
pull_request_rules:
- name: automatic merge (squash) on CI success
conditions:
- status-success=buildkite/solana
- status-success=Travis CI - Pull Request
- status-success=ci-gate
- label=automerge
- author≠@dont-squash-my-commits
actions:
merge:
method: squash
# Join the dont-squash-my-commits group if you won't like your commits squashed
- name: automatic merge (rebase) on CI success
conditions:
- status-success=buildkite/solana
- status-success=Travis CI - Pull Request
- status-success=ci-gate
- label=automerge
- author=@dont-squash-my-commits
actions:
merge:
method: rebase
- name: remove automerge label on CI failure
conditions:
- label=automerge
- "#status-failure!=0"
actions:
label:
remove:
- automerge
comment:
message: automerge label removed due to a CI failure
- name: remove outdated reviews
conditions:
- base=master
actions:
dismiss_reviews:
changes_requested: true
- name: set automerge label on mergify backport PRs
conditions:
- author=mergify[bot]
- head~=^mergify/bp/
- "#status-failure=0"
actions:
label:
add:
- automerge
- name: v1.3 backport
conditions:
- label=v1.3
actions:
backport:
ignore_conflicts: true
branches:
- v1.3
- name: v1.4 backport
conditions:
- label=v1.4
actions:
backport:
ignore_conflicts: true
branches:
- v1.4
- name: v1.5 backport
conditions:
- label=v1.5
actions:
backport:
ignore_conflicts: true
branches:
- v1.5
- name: v1.6 backport
conditions:
- label=v1.6
actions:
backport:
ignore_conflicts: true
branches:
- v1.6

View File

@ -1,136 +0,0 @@
branches:
only:
- master
- /^v\d+\.\d+/
notifications:
email: false
slack:
on_success: change
if: NOT type = pull_request
secure: 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
os: linux
dist: bionic
language: minimal
jobs:
include:
- name: "Export Github Repositories"
if: type IN (push, cron) AND branch = master
language: python
git:
depth: false
script:
- .travis/export-github-repo.sh web3.js/ solana-web3.js
- .travis/export-github-repo.sh explorer/ explorer
- &release-artifacts
if: type IN (api, cron) OR tag IS present
name: "macOS release artifacts"
os: osx
language: rust
rust:
- stable
install:
- source ci/rust-version.sh
- PATH="/usr/local/opt/coreutils/libexec/gnubin:$PATH"
- readlink -f .
script:
- source ci/env.sh
- ci/publish-tarball.sh
deploy:
- provider: s3
access_key_id: $AWS_ACCESS_KEY_ID
secret_access_key: $AWS_SECRET_ACCESS_KEY
bucket: release.solana.com
region: us-west-1
skip_cleanup: true
acl: public_read
local_dir: travis-s3-upload
on:
all_branches: true
- provider: releases
token: $GITHUB_TOKEN
skip_cleanup: true
file_glob: true
file: travis-release-upload/*
on:
tags: true
- <<: *release-artifacts
name: "Windows release artifacts"
os: windows
# Linux release artifacts are still built by ci/buildkite-secondary.yml
#- <<: *release-artifacts
# name: "Linux release artifacts"
# os: linux
# before_install:
# - sudo apt-get install libssl-dev libudev-dev
# explorer pull request
- name: "explorer"
if: type = pull_request AND branch = master
language: node_js
node_js:
- "node"
cache:
directories:
- ~/.npm
before_install:
- .travis/affects.sh explorer/ .travis || travis_terminate 0
- cd explorer
script:
- npm run build
- npm run format
# web3.js pull request
- name: "web3.js"
if: type = pull_request AND branch = master
language: node_js
node_js:
- "lts/*"
services:
- docker
cache:
directories:
- ~/.npm
before_install:
- .travis/affects.sh web3.js/ .travis || travis_terminate 0
- cd web3.js/
- source .travis/before_install.sh
script:
- ../.travis/commitlint.sh
- source .travis/script.sh
# docs pull request
- name: "docs"
if: type IN (push, pull_request) OR tag IS present
language: node_js
node_js:
- "node"
services:
- docker
cache:
directories:
- ~/.npm
before_install:
- source ci/env.sh
- .travis/channel_restriction.sh edge beta || travis_terminate 0
- .travis/affects.sh docs/ .travis || travis_terminate 0
- cd docs/
- source .travis/before_install.sh
script:
- source .travis/script.sh

View File

@ -1,25 +0,0 @@
#!/usr/bin/env bash
#
# Check if files in the commit range match one or more prefixes
#
# Always run the job if we are on a tagged release
if [[ -n "$TRAVIS_TAG" ]]; then
exit 0
fi
(
set -x
git diff --name-only "$TRAVIS_COMMIT_RANGE"
)
for file in $(git diff --name-only "$TRAVIS_COMMIT_RANGE"); do
for prefix in "$@"; do
if [[ $file =~ ^"$prefix" ]]; then
exit 0
fi
done
done
echo "No modifications to $*"
exit 1

View File

@ -1,19 +0,0 @@
#!/usr/bin/env bash
#
# Only proceed if we are on one of the channels passed in, or a tag build
#
set -ex
[[ -n $CI_TAG ]] && exit 0
eval "$(ci/channel-info.sh)"
for acceptable_channel in "$@"; do
if [[ "$CHANNEL" == "$acceptable_channel" ]]; then
exit 0
fi
done
echo "Not running from one of the following channels: $*"
exit 1

View File

@ -1,32 +0,0 @@
#!/usr/bin/env bash
#
# Runs commitlint in the provided subdirectory
#
set -e
basedir=$1
if [[ -z "$basedir" ]]; then
basedir=.
fi
if [[ ! -d "$basedir" ]]; then
echo "Error: not a directory: $basedir"
exit 1
fi
if [[ ! -f "$basedir"/commitlint.config.js ]]; then
echo "Error: No commitlint configuration found"
exit 1
fi
if [[ -z $TRAVIS_COMMIT_RANGE ]]; then
echo "Error: TRAVIS_COMMIT_RANGE not defined"
exit 1
fi
cd "$basedir"
echo "Checking commits in TRAVIS_COMMIT_RANGE: $TRAVIS_COMMIT_RANGE"
while IFS= read -r line; do
echo "$line" | npx commitlint
done < <(git log "$TRAVIS_COMMIT_RANGE" --format=%s -- .)

View File

@ -1,34 +0,0 @@
#!/usr/bin/env bash
#
# Exports a subdirectory into another github repository
#
set -e
if [[ -z $GITHUB_TOKEN ]]; then
echo GITHUB_TOKEN not defined
exit 1
fi
cd "$(dirname "$0")/.."
pip3 install git-filter-repo
declare subdir=$1
declare repo_name=$2
[[ -n "$subdir" ]] || {
echo "Error: subdir not specified"
exit 1
}
[[ -n "$repo_name" ]] || {
echo "Error: repo_name not specified"
exit 1
}
echo "Exporting $subdir"
set -x
rm -rf .github_export/"$repo_name"
git clone https://"$GITHUB_TOKEN"@github.com/solana-labs/"$repo_name" .github_export/"$repo_name"
git filter-repo --subdirectory-filter "$subdir" --target .github_export/"$repo_name"
git -C .github_export/"$repo_name" push https://"$GITHUB_TOKEN"@github.com/solana-labs/"$repo_name"

View File

@ -1,248 +1,69 @@
# Solana Coding Guidelines
Solana Coding Guidelines
===
The goal of these guidelines is to improve developer productivity by allowing
developers to jump into any file in the codebase and not need to adapt to
inconsistencies in how the code is written. The codebase should appear as if it
had been authored by a single developer. If you don't agree with a convention,
submit a PR patching this document and let's discuss! Once the PR is accepted,
*all* code should be updated as soon as possible to reflect the new
The goal of these guidelines is to improve developer productivity by allowing developers to
jump any file in the codebase and not need to adapt to inconsistencies in how the code is
written. The codebase should appear as if it had been authored by a single developer. If you
don't agree with a convention, submit a PR patching this document and let's discuss! Once
the PR is accepted, *all* code should be updated as soon as possible to reflect the new
conventions.
## Pull Requests
Rust coding conventions
---
Small, frequent PRs are much preferred to large, infrequent ones. A large PR is
difficult to review, can block others from making progress, and can quickly get
its author into "rebase hell". A large PR oftentimes arises when one change
requires another, which requires another, and then another. When you notice
those dependencies, put the fix into a commit of its own, then checkout a new
branch, and cherry-pick it.
* All Rust code is formatted using the latest version of `rustfmt`. Once installed, it will be
updated automatically when you update the compiler with `rustup`.
```bash
$ git commit -am "Fix foo, needed by bar"
$ git checkout master
$ git checkout -b fix-foo
$ git cherry-pick fix-bar
$ git push --set-upstream origin fix-foo
```
* All Rust code is linted with Clippy. If you'd prefer to ignore its advice, do so explicitly:
Open a PR to start the review process and then jump back to your original
branch to keep making progress. Consider rebasing to make your fix the first
commit:
```bash
$ git checkout fix-bar
$ git rebase -i master <Move fix-foo to top>
```
Once the commit is merged, rebase the original branch to purge the
cherry-picked commit:
```bash
$ git pull --rebase upstream master
```
### How big is too big?
If there are no functional changes, PRs can be very large and that's no
problem. If, however, your changes are making meaningful changes or additions,
then about 1,000 lines of changes is about the most you should ask a Solana
maintainer to review.
### Should I send small PRs as I develop large, new components?
Add only code to the codebase that is ready to be deployed. If you are building
a large library, consider developing it in a separate git repository. When it
is ready to be integrated, the Solana maintainers will work with you to decide
on a path forward. Smaller libraries may be copied in whereas very large ones
may be pulled in with a package manager.
## Getting Pull Requests Merged
There is no single person assigned to watching GitHub PR queue and ushering you
through the process. Typically, you will ask the person that wrote a component
to review changes to it. You can find the author using `git blame` or asking on
Discord. When working to get your PR merged, it's most important to understand
that changing the code is your priority and not necessarily a priority of the
person you need an approval from. Also, while you may interact the most with
the component author, you should aim to be inclusive of others. Providing a
detailed problem description is the most effective means of engaging both the
component author and other potentially interested parties.
Consider opening all PRs as Draft Pull Requests first. Using a draft PR allows
you to kickstart the CI automation, which typically takes between 10 and 30
minutes to execute. Use that time to write a detailed problem description. Once
the description is written and CI succeeds, click the "Ready to Review" button
and add reviewers. Adding reviewers before CI succeeds is a fast path to losing
reviewer engagement. Not only will they be notified and see the PR is not yet
ready for them, they will also be bombarded them with additional notifications
each time you push a commit to get past CI or until they "mute" the PR. Once
muted, you'll need to reach out over some other medium, such as Discord, to
request they have another look. When you use draft PRs, no notifications are
sent when you push commits and edit the PR description. Use draft PRs
liberally. Don't bug the humans until you have gotten past the bots.
### What should be in my PR description?
Reviewing code is hard work and generally involves an attempt to guess the
author's intent at various levels. Please assume reviewer time is scarce and do
what you can to make your PR as consumable as possible. Inspired by techniques
for writing good whitepapers, the guidance here aims to maximize reviewer
engagement.
Assume the reviewer will spend no more than a few seconds reading the PR title.
If it doesn't describe a noteworthy change, don't expect the reviewer to click
to see more.
Next, like the abstract of a whitepaper, the reviewer will spend ~30 seconds
reading the PR problem description. If what is described there doesn't look
more important than competing issues, don't expect the reviewer to read on.
Next, the reviewer will read the proposed changes. At this point, the reviewer
needs to be convinced the proposed changes are a *good* solution to the problem
described above. If the proposed changes, not the code changes, generates
discussion, consider closing the PR and returning with a design proposal
instead.
Finally, once the reviewer understands the problem and agrees with the approach
to solving it, the reviewer will view the code changes. At this point, the
reviewer is simply looking to see if the implementation actually implements
what was proposed and if that implementation is maintainable. When a concise,
readable test for each new code path is present, the reviewer can safely ignore
the details of its implementation. When those tests are missing, expect to
either lose engagement or get a pile of review comments as the reviewer
attempts to consider every ambiguity in your implementation.
### The PR Title
The PR title should contain a brief summary of the change, from the perspective
of the user. Examples of good titles:
* Add rent to accounts
* Fix out-of-memory error in validator
* Clean up `process_message()` in runtime
The conventions here are all the same as a good git commit title:
* First word capitalized and in the imperative mood, not past tense ("add", not
"added")
* No trailing period
* What was done, whom it was done to, and in what context
### The PR Problem Statement
The git repo implements a product with various features. The problem statement
should describe how the product is missing a feature, how a feature is
incomplete, or how the implementation of a feature is somehow undesirable. If
an issue being fixed already describes the problem, go ahead and copy-paste it.
As mentioned above, reviewer time is scarce. Given a queue of PRs to review,
the reviewer may ignore PRs that expect them to click through links to see if
the PR warrants attention.
### The Proposed Changes
Typically the content under the "Proposed changes" section will be a bulleted
list of steps taken to solve the problem. Oftentimes, the list is identical to
the subject lines of the git commits contained in the PR. It's especially
generous (and not expected) to rebase or reword commits such that each change
matches the logical flow in your PR description.
### When will my PR be reviewed?
PRs are typically reviewed and merged in under 7 days. If your PR has been open
for longer, it's a strong indicator that the reviewers aren't confident the
change meets the quality standards of the codebase. You might consider closing
it and coming back with smaller PRs and longer descriptions detailing what
problem it solves and how it solves it. Old PRs will be marked stale and then
closed automatically 7 days later.
### How to manage review feedback?
After a reviewer provides feedback, you can quickly say "acknowledged, will
fix" using a thumb's up emoji. If you're confident your fix is exactly as
prescribed, add a reply "Fixed in COMMIT\_HASH" and mark the comment as
resolved. If you're not sure, reply "Is this what you had in mind?
COMMIT\_HASH" and if so, the reviewer will reply and mark the conversation as
resolved. Marking conversations as resolved is an excellent way to engage more
reviewers. Leaving conversations open may imply the PR is not yet ready for
additional review.
### When will my PR be re-reviewed?
Recall that once your PR is opened, a notification is sent every time you push
a commit. After a reviewer adds feedback, they won't be checking on the status
of that feedback after every new commit. Instead, directly mention the reviewer
when you feel your PR is ready for another pass.
## Draft Pull Requests
If you want early feedback on your PR, use GitHub's "Draft Pull Request"
mechanism. Draft PRs are a convenient way to collaborate with the Solana
maintainers without triggering notifications as you make changes. When you feel
your PR is ready for a broader audience, you can transition your draft PR to a
standard PR with the click of a button.
Do not add reviewers to draft PRs. GitHub doesn't automatically clear
approvals when you click "Ready for Review", so a review that meant "I approve
of the direction" suddenly has the appearance of "I approve of these changes."
Instead, add a comment that mentions the usernames that you would like a review
from. Ask explicitly what you would like feedback on.
## Rust coding conventions
* All Rust code is formatted using the latest version of `rustfmt`. Once
installed, it will be updated automatically when you update the compiler with
`rustup`.
* All Rust code is linted with Clippy. If you'd prefer to ignore its advice, do
so explicitly:
```rust #[allow(clippy::too_many_arguments)] ```
```rust
#[allow(clippy::too_many_arguments)]
```
Note: Clippy defaults can be overridden in the top-level file `.clippy.toml`.
* For variable names, when in doubt, spell it out. The mapping from type names
to variable names is to lowercase the type name, putting an underscore before
each capital letter. Variable names should *not* be abbreviated unless being
used as closure arguments and the brevity improves readability. When a function
has multiple instances of the same type, qualify each with a prefix and
underscore (i.e. alice\_keypair) or a numeric suffix (i.e. tx0).
* For variable names, when in doubt, spell it out. The mapping from type names to variable names
is to lowercase the type name, putting an underscore before each capital letter. Variable names
should *not* be abbreviated unless being used as closure arguments and the brevity improves
readability. When a function has multiple instances of the same type, qualify each with a
prefix and underscore (i.e. alice_keypair) or a numeric suffix (i.e. tx0).
* For function and method names, use `<verb>_<subject>`. For unit tests, that
verb should always be `test` and for benchmarks the verb should always be
`bench`. Avoid namespacing function names with some arbitrary word. Avoid
abbreviating words in function names.
* For function and method names, use `<verb>_<subject>`. For unit tests, that verb should
always be `test` and for benchmarks the verb should always be `bench`. Avoid namespacing
function names with some arbitrary word. Avoid abbreviating words in function names.
* As they say, "When in Rome, do as the Romans do." A good patch should
acknowledge the coding conventions of the code that surrounds it, even in the
case where that code has not yet been updated to meet the conventions described
here.
* As they say, "When in Rome, do as the Romans do." A good patch should acknowledge the coding
conventions of the code that surrounds it, even in the case where that code has not yet been
updated to meet the conventions described here.
## Terminology
Terminology
---
Inventing new terms is allowed, but should only be done when the term is widely
used and understood. Avoid introducing new 3-letter terms, which can be
confused with 3-letter acronyms.
Inventing new terms is allowed, but should only be done when the term is widely used and
understood. Avoid introducing new 3-letter terms, which can be confused with 3-letter acronyms.
[Terms currently in use](docs/src/terminology.md)
[Terms currently in use](book/src/terminology.md)
## Design Proposals
Proposing architectural changes
---
Solana's architecture is described by docs generated from markdown files in
the `docs/src/` directory, maintained by an *editor* (currently @garious). To
add a design proposal, you'll need to include it in the
[Accepted Design Proposals](https://docs.solana.com/proposals/accepted-design-proposals)
section of the Solana docs. Here's the full process:
Solana's architecture is described by a book generated from markdown files in
the `book/src/` directory, maintained by an *editor* (currently @garious). To
change the architecture, you'll need to at least propose a change the content
under the [Proposed
Changes](https://solana-labs.github.io/solana/proposals.html) chapter. Here's
the full process:
1. Propose a design by creating a PR that adds a markdown document to the
`docs/src/proposals` directory and references it from the [table of
contents](docs/src/SUMMARY.md). Add any relevant *maintainers* to the PR
review.
1. Propose to a change to the architecture by creating a PR that adds a
markdown document to the directory `book/src/` and references it from the
[table of contents](book/src/SUMMARY.md). Add the editor and any relevant
*maintainers* to the PR review.
2. The PR being merged indicates your proposed change was accepted and that the
maintainers support your plan of attack.
editor and maintainers support your plan of attack.
3. Submit PRs that implement the proposal. When the implementation reveals the
need for tweaks to the proposal, be sure to update the proposal and have that
change reviewed by the same people as in step 1.
4. Once the implementation is complete, submit a PR that moves the link from
the Accepted Proposals to the Implemented Proposals section.
need for tweaks to the architecture, be sure to update the proposal and have
that change reviewed by the same people as in step 1.
4. Once the implementation is complete, the editor will then work to integrate
the document into the book.

6947
Cargo.lock generated

File diff suppressed because it is too large Load Diff

View File

@ -1,77 +1,117 @@
[package]
name = "solana"
description = "Blockchain, Rebuilt for Scale"
version = "0.11.0"
documentation = "https://docs.rs/solana"
homepage = "https://solana.com/"
readme = "README.md"
repository = "https://github.com/solana-labs/solana"
authors = ["Solana Maintainers <maintainers@solana.com>"]
license = "Apache-2.0"
edition = "2018"
[badges]
codecov = { repository = "solana-labs/solana", branch = "master", service = "github" }
[features]
bpf_c = ["solana-bpfloader/bpf_c"]
chacha = []
cuda = []
erasure = []
ipv6 = []
test = []
unstable = []
[dependencies]
bincode = "1.0.0"
bs58 = "0.2.0"
bv = { version = "0.10.0", features = ["serde"] }
byteorder = "1.2.1"
chrono = { version = "0.4.0", features = ["serde"] }
hashbrown = "0.1.7"
indexmap = "1.0"
itertools = "0.8.0"
libc = "0.2.45"
log = "0.4.2"
nix = "0.12.0"
rand = "0.6.1"
rand_chacha = "0.1.0"
rayon = "1.0.0"
reqwest = "0.9.0"
ring = "0.13.2"
rocksdb = "0.10.1"
serde = "1.0.82"
serde_derive = "1.0.82"
serde_json = "1.0.10"
solana-bpfloader = { path = "programs/native/bpf_loader", version = "0.11.0" }
solana-drone = { path = "drone", version = "0.11.0" }
solana-jsonrpc-core = "0.4.0"
solana-jsonrpc-http-server = "0.4.0"
solana-jsonrpc-macros = "0.4.0"
solana-jsonrpc-pubsub = "0.4.0"
solana-jsonrpc-ws-server = "0.4.0"
solana-logger = { path = "logger", version = "0.11.0" }
solana-metrics = { path = "metrics", version = "0.11.0" }
solana-native-loader = { path = "programs/native/native_loader", version = "0.11.0" }
solana-netutil = { path = "netutil", version = "0.11.0" }
solana-sdk = { path = "sdk", version = "0.11.0" }
solana-system-program = { path = "programs/native/system", version = "0.11.0" }
tokio = "0.1"
tokio-codec = "0.1"
untrusted = "0.6.2"
[dev-dependencies]
hex-literal = "0.1.1"
matches = "0.1.6"
[[bench]]
name = "bank"
[[bench]]
name = "banking_stage"
[[bench]]
name = "db_ledger"
[[bench]]
name = "ledger"
[[bench]]
name = "signature"
[[bench]]
name = "sigverify"
[[bench]]
required-features = ["chacha"]
name = "chacha"
[workspace]
members = [
"bench-exchange",
".",
"bench-streamer",
"bench-tps",
"accounts-bench",
"banking-bench",
"banks-client",
"banks-interface",
"banks-server",
"clap-utils",
"cli-config",
"cli-output",
"client",
"core",
"dos",
"download-utils",
"faucet",
"frozen-abi",
"perf",
"validator",
"drone",
"fullnode",
"fullnode-config",
"genesis",
"gossip",
"install",
"keygen",
"ledger",
"ledger-tool",
"local-cluster",
"logger",
"log-analyzer",
"merkle-tree",
"stake-o-matic",
"storage-bigtable",
"storage-proto",
"streamer",
"measure",
"metrics",
"net-shaper",
"notifier",
"poh-bench",
"program-test",
"programs/secp256k1",
"programs/bpf_loader",
"programs/budget",
"programs/config",
"programs/exchange",
"programs/failure",
"programs/noop",
"programs/ownable",
"programs/stake",
"programs/vest",
"programs/vote",
"remote-wallet",
"ramp-tps",
"runtime",
"runtime/store-tool",
"programs/bpf/rust/noop",
"programs/native/bpf_loader",
"programs/native/budget",
"programs/native/erc20",
"programs/native/lua_loader",
"programs/native/native_loader",
"programs/native/noop",
"programs/native/storage",
"programs/native/system",
"programs/native/vote",
"replicator",
"sdk",
"sdk/cargo-build-bpf",
"sdk/cargo-test-bpf",
"scripts",
"stake-accounts",
"stake-monitor",
"sys-tuner",
"tokens",
"transaction-status",
"account-decoder",
"upload-perf",
"net-utils",
"version",
"cli",
"rayon-threadlimit",
"watchtower",
]
exclude = [
"programs/bpf",
"vote-signer",
"wallet",
]

View File

@ -1,4 +1,4 @@
Copyright 2020 Solana Foundation.
Copyright 2018 Solana Labs, Inc.
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.

143
README.md
View File

@ -1,25 +1,48 @@
<p align="center">
<a href="https://solana.com">
<img alt="Solana" src="https://i.imgur.com/OMnvVEz.png" width="250" />
</a>
</p>
[![Solana crate](https://img.shields.io/crates/v/solana-core.svg)](https://crates.io/crates/solana-core)
[![Solana documentation](https://docs.rs/solana-core/badge.svg)](https://docs.rs/solana-core)
[![Solana crate](https://img.shields.io/crates/v/solana.svg)](https://crates.io/crates/solana)
[![Solana documentation](https://docs.rs/solana/badge.svg)](https://docs.rs/solana)
[![Build status](https://badge.buildkite.com/8cc350de251d61483db98bdfc895b9ea0ac8ffa4a32ee850ed.svg?branch=master)](https://buildkite.com/solana-labs/solana/builds?branch=master)
[![codecov](https://codecov.io/gh/solana-labs/solana/branch/master/graph/badge.svg)](https://codecov.io/gh/solana-labs/solana)
# Building
Blockchain Rebuilt for Scale
===
## **1. Install rustc, cargo and rustfmt.**
Solana&trade; is a new blockchain architecture built from the ground up for scale. The architecture supports
up to 710 thousand transactions per second on a gigabit network.
Disclaimer
===
All claims, content, designs, algorithms, estimates, roadmaps, specifications, and performance measurements described in this project are done with the author's best effort. It is up to the reader to check and validate their accuracy and truthfulness. Furthermore nothing in this project constitutes a solicitation for investment.
Introduction
===
It's possible for a centralized database to process 710,000 transactions per second on a standard gigabit network if the transactions are, on average, no more than 176 bytes. A centralized database can also replicate itself and maintain high availability without significantly compromising that transaction rate using the distributed system technique known as Optimistic Concurrency Control [\[H.T.Kung, J.T.Robinson (1981)\]](http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.65.4735). At Solana, we're demonstrating that these same theoretical limits apply just as well to blockchain on an adversarial network. The key ingredient? Finding a way to share time when nodes can't trust one-another. Once nodes can trust time, suddenly ~40 years of distributed systems research becomes applicable to blockchain!
> Perhaps the most striking difference between algorithms obtained by our method and ones based upon timeout is that using timeout produces a traditional distributed algorithm in which the processes operate asynchronously, while our method produces a globally synchronous one in which every process does the same thing at (approximately) the same time. Our method seems to contradict the whole purpose of distributed processing, which is to permit different processes to operate independently and perform different functions. However, if a distributed system is really a single system, then the processes must be synchronized in some way. Conceptually, the easiest way to synchronize processes is to get them all to do the same thing at the same time. Therefore, our method is used to implement a kernel that performs the necessary synchronization--for example, making sure that two different processes do not try to modify a file at the same time. Processes might spend only a small fraction of their time executing the synchronizing kernel; the rest of the time, they can operate independently--e.g., accessing different files. This is an approach we have advocated even when fault-tolerance is not required. The method's basic simplicity makes it easier to understand the precise properties of a system, which is crucial if one is to know just how fault-tolerant the system is. [\[L.Lamport (1984)\]](http://citeseerx.ist.psu.edu/viewdoc/summary?doi=10.1.1.71.1078)
Furthermore, and much to our surprise, it can be implemented using a mechanism that has existed in Bitcoin since day one. The Bitcoin feature is called nLocktime and it can be used to postdate transactions using block height instead of a timestamp. As a Bitcoin client, you'd use block height instead of a timestamp if you don't trust the network. Block height turns out to be an instance of what's being called a Verifiable Delay Function in cryptography circles. It's a cryptographically secure way to say time has passed. In Solana, we use a far more granular verifiable delay function, a SHA 256 hash chain, to checkpoint the ledger and coordinate consensus. With it, we implement Optimistic Concurrency Control and are now well en route towards that theoretical limit of 710,000 transactions per second.
Architecture
===
Before you jump into the code, review the online book [Solana: Blockchain Rebuilt for Scale](https://solana-labs.github.io/solana/).
Developing
===
Building
---
Install rustc, cargo and rustfmt:
```bash
$ curl https://sh.rustup.rs -sSf | sh
$ source $HOME/.cargo/env
$ rustup component add rustfmt
$ rustup component add rustfmt-preview
```
Please sure you are always using the latest stable rust version by running:
If your rustc version is lower than 1.31.0, please update it:
```bash
$ rustup update
@ -28,47 +51,86 @@ $ rustup update
On Linux systems you may need to install libssl-dev, pkg-config, zlib1g-dev, etc. On Ubuntu:
```bash
$ sudo apt-get update
$ sudo apt-get install libssl-dev libudev-dev pkg-config zlib1g-dev llvm clang make
$ sudo apt-get install libssl-dev pkg-config zlib1g-dev llvm clang
```
## **2. Download the source code.**
Download the source code:
```bash
$ git clone https://github.com/solana-labs/solana.git
$ cd solana
```
## **3. Build.**
Build
```bash
$ cargo build
$ cargo build --all
```
## **4. Run a minimal local cluster.**
```bash
$ ./run.sh
```
Testing
---
# Testing
**Run the test suite:**
Run the test suite:
```bash
$ cargo test
$ cargo test --all
```
### Starting a local testnet
Start your own testnet locally, instructions are in the [online docs](https://docs.solana.com/cluster/bench-tps).
To emulate all the tests that will run on a Pull Request, run:
### Accessing the remote development cluster
* `devnet` - stable public cluster for development accessible via
devnet.solana.com. Runs 24/7. Learn more about the [public clusters](https://docs.solana.com/clusters)
```bash
$ ./ci/run-local.sh
```
# Benchmarking
Local Testnet
---
First install the nightly build of rustc. `cargo bench` requires use of the
unstable features only available in the nightly build.
Start your own testnet locally, instructions are in the book [Solana: Blockchain Rebuild for Scale: Getting Started](https://solana-labs.github.io/solana/getting-started.html).
Remote Testnets
---
We maintain several testnets:
* `testnet` - public stable testnet accessible via testnet.solana.com, with an https proxy for web apps at api.testnet.solana.com. Runs 24/7
* `testnet-beta` - public beta channel testnet accessible via beta.testnet.solana.com. Runs 24/7
* `testnet-edge` - public edge channel testnet accessible via edge.testnet.solana.com. Runs 24/7
* `testnet-perf` - permissioned stable testnet running a 24/7 soak test
* `testnet-beta-perf` - permissioned beta channel testnet running a multi-hour soak test weekday mornings
* `testnet-edge-perf` - permissioned edge channel testnet running a multi-hour soak test weekday mornings
## Deploy process
They are deployed with the `ci/testnet-manager.sh` script through a list of [scheduled
buildkite jobs](https://buildkite.com/solana-labs/testnet-management/settings/schedules).
Each testnet can be manually manipulated from buildkite as well. The `-perf`
testnets use a release tarball while the non`-perf` builds use the snap build
(we've observed that the snap build runs slower than a tarball but this has yet
to be root caused).
## How do I reset the testnet?
Manually trigger the [testnet-management](https://buildkite.com/solana-labs/testnet-management) pipeline
and when prompted select the desired testnet
## How can I scale the tx generation rate?
Increase the TX rate by increasing the number of cores on the client machine which is running
`bench-tps` or run multiple clients. Decrease by lowering cores or using the rayon env
variable `RAYON_NUM_THREADS=<xx>`
## How can I test a change on the testnet?
Currently, a merged PR is the only way to test a change on the testnet. But you
can run your own testnet using the scripts in the `net/` directory.
## Adjusting the number of clients or validators on the testnet
Edit `ci/testnet-manager.sh`
Benchmarking
---
First install the nightly build of rustc. `cargo bench` requires unstable features:
```bash
$ rustup install nightly
@ -77,14 +139,16 @@ $ rustup install nightly
Run the benchmarks:
```bash
$ cargo +nightly bench
$ cargo +nightly bench --features="unstable"
```
# Release Process
Release Process
---
The release process for this project is described [here](RELEASE.md).
# Code coverage
Code coverage
---
To generate code coverage statistics:
@ -93,6 +157,7 @@ $ scripts/coverage.sh
$ open target/cov/lcov-local/index.html
```
Why coverage? While most see coverage as a code quality metric, we see it primarily as a developer
productivity metric. When a developer makes a change to the codebase, presumably it's a *solution* to
some problem. Our unit-test suite is how we encode the set of *problems* the codebase solves. Running
@ -105,8 +170,4 @@ better way to solve the same problem, a Pull Request with your solution would mo
welcome! Likewise, if rewriting a test can better communicate what code it's protecting, please
send us that patch!
# Disclaimer
All claims, content, designs, algorithms, estimates, roadmaps, specifications, and performance measurements described in this project are done with the author's best effort. It is up to the reader to check and validate their accuracy and truthfulness. Furthermore nothing in this project constitutes a solicitation for investment.
Any content produced by Solana, or developer resources that Solana provides, are for educational and inspiration purposes only. Solana does not encourage, induce or sanction the deployment of any such applications in violation of applicable laws or regulations.

View File

@ -43,7 +43,8 @@ the `master` branch as late as possible prior to the milestone release.
### v*X.Y.Z* release tag
The release tags are created as desired by the owner of the given stabilization
branch, and cause that *X.Y.Z* release to be shipped to https://crates.io
branch, and cause that *X.Y.Z* release to be shipped to https://crates.io,
https://snapcraft.io/, and elsewhere.
Immediately after a new v*X.Y.Z* branch tag has been created, the `Cargo.toml`
patch version number (*Z*) of the stabilization branch is incremented by the
@ -59,98 +60,31 @@ There are three release channels that map to branches as follows:
* beta - tracks the largest (and latest) `vX.Y` stabilization branch, more stable.
* stable - tracks the second largest `vX.Y` stabilization branch, most stable.
## Steps to Create a Branch
## Release Steps
### Create the new branch
1. Check out the latest commit on `master` branch:
```
git fetch --all
git checkout upstream/master
```
1. Determine the new branch name. The name should be "v" + the first 2 version fields
from Cargo.toml. For example, a Cargo.toml with version = "0.9.0" implies
the next branch name is "v0.9".
1. Create the new branch and push this branch to the `solana` repository:
```
git checkout -b <branchname>
git push -u origin <branchname>
```
### Changing channels
Alternatively use the Github UI.
When cutting a new channel branch these pre-steps are required:
### Update master branch to the next release minor version
1. Pick your branch point for release on master.
2. Create the branch. The name should be "v" + the first 2 "version" fields from Cargo.toml. For example, a Cargo.toml with version = "0.9.0" implies the next branch name is "v0.9".
4. Push the new branch to the solana repository
3. Update Cargo.toml on master to the next semantic version (e.g. 0.9.0 -> 0.10.0) by running `./scripts/increment-cargo-version.sh`.
5. Land your Cargo.toml change as a master PR.
1. After the new branch has been created and pushed, update the Cargo.toml files on **master** to the next semantic version (e.g. 0.9.0 -> 0.10.0) with:
```
$ scripts/increment-cargo-version.sh minor
$ ./scripts/cargo-for-all-lock-files.sh update
```
1. Push all the changed Cargo.toml and Cargo.lock files to the `master` branch with something like:
```
git co -b version_update
git ls-files -m | xargs git add
git commit -m 'Bump version to X.Y+1.0'
git push -u origin version_update
```
1. Confirm that your freshly cut release branch is shown as `BETA_CHANNEL` and the previous release branch as `STABLE_CHANNEL`:
```
ci/channel_info.sh
```
At this point, ci/channel-info.sh should show your freshly cut release branch as "BETA_CHANNEL" and the previous release branch as "STABLE_CHANNEL".
## Steps to Create a Release
### Updating channels (i.e. "making a release")
### Create the Release Tag on GitHub
We use [github's Releases UI](https://github.com/solana-labs/solana/releases) for tagging a release.
1. Go to [GitHub Releases](https://github.com/solana-labs/solana/releases) for tagging a release.
1. Click "Draft new release". The release tag must exactly match the `version`
field in `/Cargo.toml` prefixed by `v`.
1. If the Cargo.toml version field is **0.12.3**, then the release tag must be **v0.12.3**
1. Make sure the Target Branch field matches the branch you want to make a release on.
1. If you want to release v0.12.0, the target branch must be v0.12
1. Fill the release notes.
1. If this is the first release on the branch (e.g. v0.13.**0**), paste in [this
template](https://raw.githubusercontent.com/solana-labs/solana/master/.github/RELEASE_TEMPLATE.md). Engineering Lead can provide summary contents for release notes if needed.
1. If this is a patch release, review all the commits since the previous release on this branch and add details as needed.
1. Click "Save Draft", then confirm the release notes look good and the tag name and branch are correct.
1. Ensure all desired commits (usually backports) are landed on the branch by now.
1. Ensure the release is marked **"This is a pre-release"**. This flag will need to be be removed manually after confirming the the Linux binary artifacts appear at a later step.
1. Go back into edit the release and click "Publish release" while being marked as a pre-release.
1. Confirm there is new git tag with intended version number at the intended revision after running `git fetch` locally.
1. Go [there ;)](https://github.com/solana-labs/solana/releases).
2. Click "Draft new release". The release tag must exactly match the `version` field in `/Cargo.toml` prefixed by `v` (ie, `<branchname>.X`).
3. If the first major release on the branch (e.g. v0.8.0), paste in [this template](https://raw.githubusercontent.com/solana-labs/solana/master/.github/RELEASE_TEMPLATE.md) and fill it in.
4. Test the release by generating a tag using semver's rules. First try at a release should be `<branchname>.X-rc.0`.
5. Verify release automation:
1. [Crates.io](https://crates.io/crates/solana) should have an updated Solana version.
2. ...
6. After testnet deployment, verify that testnets are running correct software. http://metrics.solana.com should show testnet running on a hash from your newly created branch.
7. Once the release has been made, update Cargo.toml on release to the next semantic version (e.g. 0.9.0 -> 0.9.1) by running `./scripts/increment-cargo-version.sh patch`.
### Update release branch with the next patch version
1. After the new release has been tagged, update the Cargo.toml files on **release branch** to the next semantic version (e.g. 0.9.0 -> 0.9.1) with:
```
$ scripts/increment-cargo-version.sh patch
$ ./scripts/cargo-for-all-lock-files.sh tree
```
1. Push all the changed Cargo.toml and Cargo.lock files to the **release branch** with something like:
```
git co -b version_update origin/vX.Y
git add -u
git commit -m 'Bump version to X.Y.Z+1'
git push -u <user-remote> version_update
```
1. Open a PR against origin/vX.Y and then merge the PR after passing CI.
### Prepare for the next release
1. Go to [GitHub Releases](https://github.com/solana-labs/solana/releases) and create a new draft release for `X.Y.Z+1` with empty release notes. This allows people to incrementally add new release notes until it's time for the next release
1. Also, point the branch field to the same branch and mark the relese as **"This is a pre-release"**.
1. Go to the [Github Milestones](https://github.com/solana-labs/solana/milestones). Create a new milestone for the `X.Y.Z+1`, move over
unresolved issues still in the `X.Y.Z` milestone, then close the `X.Y.Z` milestone.
### Verify release automation success
Go to [Solana Releases](https://github.com/solana-labs/solana/releases) and click on the latest release that you just published.
Verify that all of the build artifacts are present, then the uncheck **"This is a pre-release"** for the release.
Build artifacts can take up to 60 minutes after creating the tag before
appearing. To check for progress:
* The `solana-secondary` Buildkite pipeline handles creating the Linux release artifacts and updated crates. Look for a job under the tag name of the release: https://buildkite.com/solana-labs/solana-secondary.
* The macOS and Windows release artifacts are produced by Travis CI: https://travis-ci.com/github/solana-labs/solana/branches
[Crates.io](https://crates.io/crates/solana) should have an updated Solana version. This can take 2-3 hours, and sometimes fails in the `solana-secondary` job.
If this happens and the error is non-fatal, click "Retry" on the "publish crate" job
### Update software on devnet.solana.com/testnet.solana.com/mainnet-beta.solana.com
See the documentation at https://github.com/solana-labs/cluster-ops/

View File

@ -1,30 +0,0 @@
[package]
name = "solana-account-decoder"
version = "1.5.4"
description = "Solana account decoder"
authors = ["Solana Maintainers <maintainers@solana.foundation>"]
repository = "https://github.com/solana-labs/solana"
homepage = "https://solana.com/"
license = "Apache-2.0"
edition = "2018"
[dependencies]
base64 = "0.12.3"
bincode = "1.3.1"
bs58 = "0.3.1"
bv = "0.11.1"
Inflector = "0.11.4"
lazy_static = "1.4.0"
serde = "1.0.112"
serde_derive = "1.0.103"
serde_json = "1.0.56"
solana-config-program = { path = "../programs/config", version = "1.5.4" }
solana-sdk = { path = "../sdk", version = "1.5.4" }
solana-stake-program = { path = "../programs/stake", version = "1.5.4" }
solana-vote-program = { path = "../programs/vote", version = "1.5.4" }
spl-token-v2-0 = { package = "spl-token", version = "=3.0.1", features = ["no-entrypoint"] }
thiserror = "1.0"
zstd = "0.5.1"
[package.metadata.docs.rs]
targets = ["x86_64-unknown-linux-gnu"]

View File

@ -1,233 +0,0 @@
#[macro_use]
extern crate lazy_static;
#[macro_use]
extern crate serde_derive;
pub mod parse_account_data;
pub mod parse_config;
pub mod parse_nonce;
pub mod parse_stake;
pub mod parse_sysvar;
pub mod parse_token;
pub mod parse_vote;
pub mod validator_info;
use {
crate::parse_account_data::{parse_account_data, AccountAdditionalData, ParsedAccount},
solana_sdk::{account::Account, clock::Epoch, fee_calculator::FeeCalculator, pubkey::Pubkey},
std::{
io::{Read, Write},
str::FromStr,
},
};
pub type StringAmount = String;
/// A duplicate representation of an Account for pretty JSON serialization
#[derive(Serialize, Deserialize, Clone, Debug)]
#[serde(rename_all = "camelCase")]
pub struct UiAccount {
pub lamports: u64,
pub data: UiAccountData,
pub owner: String,
pub executable: bool,
pub rent_epoch: Epoch,
}
#[derive(Clone, Debug, PartialEq, Serialize, Deserialize)]
#[serde(rename_all = "camelCase", untagged)]
pub enum UiAccountData {
LegacyBinary(String), // Legacy. Retained for RPC backwards compatibility
Json(ParsedAccount),
Binary(String, UiAccountEncoding),
}
#[derive(Serialize, Deserialize, Clone, Debug, PartialEq)]
#[serde(rename_all = "camelCase")]
pub enum UiAccountEncoding {
Binary, // Legacy. Retained for RPC backwards compatibility
Base58,
Base64,
JsonParsed,
#[serde(rename = "base64+zstd")]
Base64Zstd,
}
impl UiAccount {
pub fn encode(
pubkey: &Pubkey,
account: Account,
encoding: UiAccountEncoding,
additional_data: Option<AccountAdditionalData>,
data_slice_config: Option<UiDataSliceConfig>,
) -> Self {
let data = match encoding {
UiAccountEncoding::Binary => UiAccountData::LegacyBinary(
bs58::encode(slice_data(&account.data, data_slice_config)).into_string(),
),
UiAccountEncoding::Base58 => UiAccountData::Binary(
bs58::encode(slice_data(&account.data, data_slice_config)).into_string(),
encoding,
),
UiAccountEncoding::Base64 => UiAccountData::Binary(
base64::encode(slice_data(&account.data, data_slice_config)),
encoding,
),
UiAccountEncoding::Base64Zstd => {
let mut encoder = zstd::stream::write::Encoder::new(Vec::new(), 0).unwrap();
match encoder
.write_all(slice_data(&account.data, data_slice_config))
.and_then(|()| encoder.finish())
{
Ok(zstd_data) => UiAccountData::Binary(base64::encode(zstd_data), encoding),
Err(_) => UiAccountData::Binary(
base64::encode(slice_data(&account.data, data_slice_config)),
UiAccountEncoding::Base64,
),
}
}
UiAccountEncoding::JsonParsed => {
if let Ok(parsed_data) =
parse_account_data(pubkey, &account.owner, &account.data, additional_data)
{
UiAccountData::Json(parsed_data)
} else {
UiAccountData::Binary(base64::encode(&account.data), UiAccountEncoding::Base64)
}
}
};
UiAccount {
lamports: account.lamports,
data,
owner: account.owner.to_string(),
executable: account.executable,
rent_epoch: account.rent_epoch,
}
}
pub fn decode(&self) -> Option<Account> {
let data = match &self.data {
UiAccountData::Json(_) => None,
UiAccountData::LegacyBinary(blob) => bs58::decode(blob).into_vec().ok(),
UiAccountData::Binary(blob, encoding) => match encoding {
UiAccountEncoding::Base58 => bs58::decode(blob).into_vec().ok(),
UiAccountEncoding::Base64 => base64::decode(blob).ok(),
UiAccountEncoding::Base64Zstd => base64::decode(blob)
.ok()
.map(|zstd_data| {
let mut data = vec![];
zstd::stream::read::Decoder::new(zstd_data.as_slice())
.and_then(|mut reader| reader.read_to_end(&mut data))
.map(|_| data)
.ok()
})
.flatten(),
UiAccountEncoding::Binary | UiAccountEncoding::JsonParsed => None,
},
}?;
Some(Account {
lamports: self.lamports,
data,
owner: Pubkey::from_str(&self.owner).ok()?,
executable: self.executable,
rent_epoch: self.rent_epoch,
})
}
}
#[derive(Serialize, Deserialize, Clone, Debug, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiFeeCalculator {
pub lamports_per_signature: StringAmount,
}
impl From<FeeCalculator> for UiFeeCalculator {
fn from(fee_calculator: FeeCalculator) -> Self {
Self {
lamports_per_signature: fee_calculator.lamports_per_signature.to_string(),
}
}
}
impl Default for UiFeeCalculator {
fn default() -> Self {
Self {
lamports_per_signature: "0".to_string(),
}
}
}
#[derive(Clone, Copy, Debug, PartialEq, Serialize, Deserialize)]
#[serde(rename_all = "camelCase")]
pub struct UiDataSliceConfig {
pub offset: usize,
pub length: usize,
}
fn slice_data(data: &[u8], data_slice_config: Option<UiDataSliceConfig>) -> &[u8] {
if let Some(UiDataSliceConfig { offset, length }) = data_slice_config {
if offset >= data.len() {
&[]
} else if length > data.len() - offset {
&data[offset..]
} else {
&data[offset..offset + length]
}
} else {
data
}
}
#[cfg(test)]
mod test {
use super::*;
#[test]
fn test_slice_data() {
let data = vec![1, 2, 3, 4, 5];
let slice_config = Some(UiDataSliceConfig {
offset: 0,
length: 5,
});
assert_eq!(slice_data(&data, slice_config), &data[..]);
let slice_config = Some(UiDataSliceConfig {
offset: 0,
length: 10,
});
assert_eq!(slice_data(&data, slice_config), &data[..]);
let slice_config = Some(UiDataSliceConfig {
offset: 1,
length: 2,
});
assert_eq!(slice_data(&data, slice_config), &data[1..3]);
let slice_config = Some(UiDataSliceConfig {
offset: 10,
length: 2,
});
assert_eq!(slice_data(&data, slice_config), &[] as &[u8]);
}
#[test]
fn test_base64_zstd() {
let encoded_account = UiAccount::encode(
&Pubkey::default(),
Account {
data: vec![0; 1024],
..Account::default()
},
UiAccountEncoding::Base64Zstd,
None,
None,
);
assert!(matches!(
encoded_account.data,
UiAccountData::Binary(_, UiAccountEncoding::Base64Zstd)
));
let decoded_account = encoded_account.decode().unwrap();
assert_eq!(decoded_account.data, vec![0; 1024]);
}
}

View File

@ -1,145 +0,0 @@
use crate::{
parse_config::parse_config,
parse_nonce::parse_nonce,
parse_stake::parse_stake,
parse_sysvar::parse_sysvar,
parse_token::{parse_token, spl_token_id_v2_0},
parse_vote::parse_vote,
};
use inflector::Inflector;
use serde_json::Value;
use solana_sdk::{instruction::InstructionError, pubkey::Pubkey, system_program, sysvar};
use std::collections::HashMap;
use thiserror::Error;
lazy_static! {
static ref CONFIG_PROGRAM_ID: Pubkey = solana_config_program::id();
static ref STAKE_PROGRAM_ID: Pubkey = solana_stake_program::id();
static ref SYSTEM_PROGRAM_ID: Pubkey = system_program::id();
static ref SYSVAR_PROGRAM_ID: Pubkey = sysvar::id();
static ref TOKEN_PROGRAM_ID: Pubkey = spl_token_id_v2_0();
static ref VOTE_PROGRAM_ID: Pubkey = solana_vote_program::id();
pub static ref PARSABLE_PROGRAM_IDS: HashMap<Pubkey, ParsableAccount> = {
let mut m = HashMap::new();
m.insert(*CONFIG_PROGRAM_ID, ParsableAccount::Config);
m.insert(*SYSTEM_PROGRAM_ID, ParsableAccount::Nonce);
m.insert(*TOKEN_PROGRAM_ID, ParsableAccount::SplToken);
m.insert(*STAKE_PROGRAM_ID, ParsableAccount::Stake);
m.insert(*SYSVAR_PROGRAM_ID, ParsableAccount::Sysvar);
m.insert(*VOTE_PROGRAM_ID, ParsableAccount::Vote);
m
};
}
#[derive(Error, Debug)]
pub enum ParseAccountError {
#[error("{0:?} account not parsable")]
AccountNotParsable(ParsableAccount),
#[error("Program not parsable")]
ProgramNotParsable,
#[error("Additional data required to parse: {0}")]
AdditionalDataMissing(String),
#[error("Instruction error")]
InstructionError(#[from] InstructionError),
#[error("Serde json error")]
SerdeJsonError(#[from] serde_json::error::Error),
}
#[derive(Clone, Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct ParsedAccount {
pub program: String,
pub parsed: Value,
pub space: u64,
}
#[derive(Debug, Serialize, Deserialize)]
#[serde(rename_all = "camelCase")]
pub enum ParsableAccount {
Config,
Nonce,
SplToken,
Stake,
Sysvar,
Vote,
}
#[derive(Default)]
pub struct AccountAdditionalData {
pub spl_token_decimals: Option<u8>,
}
pub fn parse_account_data(
pubkey: &Pubkey,
program_id: &Pubkey,
data: &[u8],
additional_data: Option<AccountAdditionalData>,
) -> Result<ParsedAccount, ParseAccountError> {
let program_name = PARSABLE_PROGRAM_IDS
.get(program_id)
.ok_or(ParseAccountError::ProgramNotParsable)?;
let additional_data = additional_data.unwrap_or_default();
let parsed_json = match program_name {
ParsableAccount::Config => serde_json::to_value(parse_config(data, pubkey)?)?,
ParsableAccount::Nonce => serde_json::to_value(parse_nonce(data)?)?,
ParsableAccount::SplToken => {
serde_json::to_value(parse_token(data, additional_data.spl_token_decimals)?)?
}
ParsableAccount::Stake => serde_json::to_value(parse_stake(data)?)?,
ParsableAccount::Sysvar => serde_json::to_value(parse_sysvar(data, pubkey)?)?,
ParsableAccount::Vote => serde_json::to_value(parse_vote(data)?)?,
};
Ok(ParsedAccount {
program: format!("{:?}", program_name).to_kebab_case(),
parsed: parsed_json,
space: data.len() as u64,
})
}
#[cfg(test)]
mod test {
use super::*;
use solana_sdk::nonce::{
state::{Data, Versions},
State,
};
use solana_vote_program::vote_state::{VoteState, VoteStateVersions};
#[test]
fn test_parse_account_data() {
let account_pubkey = solana_sdk::pubkey::new_rand();
let other_program = solana_sdk::pubkey::new_rand();
let data = vec![0; 4];
assert!(parse_account_data(&account_pubkey, &other_program, &data, None).is_err());
let vote_state = VoteState::default();
let mut vote_account_data: Vec<u8> = vec![0; VoteState::size_of()];
let versioned = VoteStateVersions::new_current(vote_state);
VoteState::serialize(&versioned, &mut vote_account_data).unwrap();
let parsed = parse_account_data(
&account_pubkey,
&solana_vote_program::id(),
&vote_account_data,
None,
)
.unwrap();
assert_eq!(parsed.program, "vote".to_string());
assert_eq!(parsed.space, VoteState::size_of() as u64);
let nonce_data = Versions::new_current(State::Initialized(Data::default()));
let nonce_account_data = bincode::serialize(&nonce_data).unwrap();
let parsed = parse_account_data(
&account_pubkey,
&system_program::id(),
&nonce_account_data,
None,
)
.unwrap();
assert_eq!(parsed.program, "nonce".to_string());
assert_eq!(parsed.space, State::size() as u64);
}
}

View File

@ -1,146 +0,0 @@
use crate::{
parse_account_data::{ParsableAccount, ParseAccountError},
validator_info,
};
use bincode::deserialize;
use serde_json::Value;
use solana_config_program::{get_config_data, ConfigKeys};
use solana_sdk::pubkey::Pubkey;
use solana_stake_program::config::Config as StakeConfig;
pub fn parse_config(data: &[u8], pubkey: &Pubkey) -> Result<ConfigAccountType, ParseAccountError> {
let parsed_account = if pubkey == &solana_stake_program::config::id() {
get_config_data(data)
.ok()
.and_then(|data| deserialize::<StakeConfig>(data).ok())
.map(|config| ConfigAccountType::StakeConfig(config.into()))
} else {
deserialize::<ConfigKeys>(data).ok().and_then(|key_list| {
if !key_list.keys.is_empty() && key_list.keys[0].0 == validator_info::id() {
parse_config_data::<String>(data, key_list.keys).and_then(|validator_info| {
Some(ConfigAccountType::ValidatorInfo(UiConfig {
keys: validator_info.keys,
config_data: serde_json::from_str(&validator_info.config_data).ok()?,
}))
})
} else {
None
}
})
};
parsed_account.ok_or(ParseAccountError::AccountNotParsable(
ParsableAccount::Config,
))
}
fn parse_config_data<T>(data: &[u8], keys: Vec<(Pubkey, bool)>) -> Option<UiConfig<T>>
where
T: serde::de::DeserializeOwned,
{
let config_data: T = deserialize(&get_config_data(data).ok()?).ok()?;
let keys = keys
.iter()
.map(|key| UiConfigKey {
pubkey: key.0.to_string(),
signer: key.1,
})
.collect();
Some(UiConfig { keys, config_data })
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase", tag = "type", content = "info")]
pub enum ConfigAccountType {
StakeConfig(UiStakeConfig),
ValidatorInfo(UiConfig<Value>),
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiConfigKey {
pub pubkey: String,
pub signer: bool,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiStakeConfig {
pub warmup_cooldown_rate: f64,
pub slash_penalty: u8,
}
impl From<StakeConfig> for UiStakeConfig {
fn from(config: StakeConfig) -> Self {
Self {
warmup_cooldown_rate: config.warmup_cooldown_rate,
slash_penalty: config.slash_penalty,
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiConfig<T> {
pub keys: Vec<UiConfigKey>,
pub config_data: T,
}
#[cfg(test)]
mod test {
use super::*;
use crate::validator_info::ValidatorInfo;
use serde_json::json;
use solana_config_program::create_config_account;
#[test]
fn test_parse_config() {
let stake_config = StakeConfig {
warmup_cooldown_rate: 0.25,
slash_penalty: 50,
};
let stake_config_account = create_config_account(vec![], &stake_config, 10);
assert_eq!(
parse_config(
&stake_config_account.data,
&solana_stake_program::config::id()
)
.unwrap(),
ConfigAccountType::StakeConfig(UiStakeConfig {
warmup_cooldown_rate: 0.25,
slash_penalty: 50,
}),
);
let validator_info = ValidatorInfo {
info: serde_json::to_string(&json!({
"name": "Solana",
}))
.unwrap(),
};
let info_pubkey = solana_sdk::pubkey::new_rand();
let validator_info_config_account = create_config_account(
vec![(validator_info::id(), false), (info_pubkey, true)],
&validator_info,
10,
);
assert_eq!(
parse_config(&validator_info_config_account.data, &info_pubkey).unwrap(),
ConfigAccountType::ValidatorInfo(UiConfig {
keys: vec![
UiConfigKey {
pubkey: validator_info::id().to_string(),
signer: false,
},
UiConfigKey {
pubkey: info_pubkey.to_string(),
signer: true,
}
],
config_data: serde_json::from_str(r#"{"name":"Solana"}"#).unwrap(),
}),
);
let bad_data = vec![0; 4];
assert!(parse_config(&bad_data, &info_pubkey).is_err());
}
}

View File

@ -1,67 +0,0 @@
use crate::{parse_account_data::ParseAccountError, UiFeeCalculator};
use solana_sdk::{
instruction::InstructionError,
nonce::{state::Versions, State},
};
pub fn parse_nonce(data: &[u8]) -> Result<UiNonceState, ParseAccountError> {
let nonce_state: Versions = bincode::deserialize(data)
.map_err(|_| ParseAccountError::from(InstructionError::InvalidAccountData))?;
let nonce_state = nonce_state.convert_to_current();
match nonce_state {
State::Uninitialized => Ok(UiNonceState::Uninitialized),
State::Initialized(data) => Ok(UiNonceState::Initialized(UiNonceData {
authority: data.authority.to_string(),
blockhash: data.blockhash.to_string(),
fee_calculator: data.fee_calculator.into(),
})),
}
}
/// A duplicate representation of NonceState for pretty JSON serialization
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase", tag = "type", content = "info")]
pub enum UiNonceState {
Uninitialized,
Initialized(UiNonceData),
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiNonceData {
pub authority: String,
pub blockhash: String,
pub fee_calculator: UiFeeCalculator,
}
#[cfg(test)]
mod test {
use super::*;
use solana_sdk::{
hash::Hash,
nonce::{
state::{Data, Versions},
State,
},
pubkey::Pubkey,
};
#[test]
fn test_parse_nonce() {
let nonce_data = Versions::new_current(State::Initialized(Data::default()));
let nonce_account_data = bincode::serialize(&nonce_data).unwrap();
assert_eq!(
parse_nonce(&nonce_account_data).unwrap(),
UiNonceState::Initialized(UiNonceData {
authority: Pubkey::default().to_string(),
blockhash: Hash::default().to_string(),
fee_calculator: UiFeeCalculator {
lamports_per_signature: 0.to_string(),
},
}),
);
let bad_data = vec![0; 4];
assert!(parse_nonce(&bad_data).is_err());
}
}

View File

@ -1,234 +0,0 @@
use crate::{
parse_account_data::{ParsableAccount, ParseAccountError},
StringAmount,
};
use bincode::deserialize;
use solana_sdk::clock::{Epoch, UnixTimestamp};
use solana_stake_program::stake_state::{Authorized, Delegation, Lockup, Meta, Stake, StakeState};
pub fn parse_stake(data: &[u8]) -> Result<StakeAccountType, ParseAccountError> {
let stake_state: StakeState = deserialize(data)
.map_err(|_| ParseAccountError::AccountNotParsable(ParsableAccount::Stake))?;
let parsed_account = match stake_state {
StakeState::Uninitialized => StakeAccountType::Uninitialized,
StakeState::Initialized(meta) => StakeAccountType::Initialized(UiStakeAccount {
meta: meta.into(),
stake: None,
}),
StakeState::Stake(meta, stake) => StakeAccountType::Delegated(UiStakeAccount {
meta: meta.into(),
stake: Some(stake.into()),
}),
StakeState::RewardsPool => StakeAccountType::RewardsPool,
};
Ok(parsed_account)
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase", tag = "type", content = "info")]
pub enum StakeAccountType {
Uninitialized,
Initialized(UiStakeAccount),
Delegated(UiStakeAccount),
RewardsPool,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiStakeAccount {
pub meta: UiMeta,
pub stake: Option<UiStake>,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiMeta {
pub rent_exempt_reserve: StringAmount,
pub authorized: UiAuthorized,
pub lockup: UiLockup,
}
impl From<Meta> for UiMeta {
fn from(meta: Meta) -> Self {
Self {
rent_exempt_reserve: meta.rent_exempt_reserve.to_string(),
authorized: meta.authorized.into(),
lockup: meta.lockup.into(),
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiLockup {
pub unix_timestamp: UnixTimestamp,
pub epoch: Epoch,
pub custodian: String,
}
impl From<Lockup> for UiLockup {
fn from(lockup: Lockup) -> Self {
Self {
unix_timestamp: lockup.unix_timestamp,
epoch: lockup.epoch,
custodian: lockup.custodian.to_string(),
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiAuthorized {
pub staker: String,
pub withdrawer: String,
}
impl From<Authorized> for UiAuthorized {
fn from(authorized: Authorized) -> Self {
Self {
staker: authorized.staker.to_string(),
withdrawer: authorized.withdrawer.to_string(),
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiStake {
pub delegation: UiDelegation,
pub credits_observed: u64,
}
impl From<Stake> for UiStake {
fn from(stake: Stake) -> Self {
Self {
delegation: stake.delegation.into(),
credits_observed: stake.credits_observed,
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiDelegation {
pub voter: String,
pub stake: StringAmount,
pub activation_epoch: StringAmount,
pub deactivation_epoch: StringAmount,
pub warmup_cooldown_rate: f64,
}
impl From<Delegation> for UiDelegation {
fn from(delegation: Delegation) -> Self {
Self {
voter: delegation.voter_pubkey.to_string(),
stake: delegation.stake.to_string(),
activation_epoch: delegation.activation_epoch.to_string(),
deactivation_epoch: delegation.deactivation_epoch.to_string(),
warmup_cooldown_rate: delegation.warmup_cooldown_rate,
}
}
}
#[cfg(test)]
mod test {
use super::*;
use bincode::serialize;
#[test]
fn test_parse_stake() {
let stake_state = StakeState::Uninitialized;
let stake_data = serialize(&stake_state).unwrap();
assert_eq!(
parse_stake(&stake_data).unwrap(),
StakeAccountType::Uninitialized
);
let pubkey = solana_sdk::pubkey::new_rand();
let custodian = solana_sdk::pubkey::new_rand();
let authorized = Authorized::auto(&pubkey);
let lockup = Lockup {
unix_timestamp: 0,
epoch: 1,
custodian,
};
let meta = Meta {
rent_exempt_reserve: 42,
authorized,
lockup,
};
let stake_state = StakeState::Initialized(meta);
let stake_data = serialize(&stake_state).unwrap();
assert_eq!(
parse_stake(&stake_data).unwrap(),
StakeAccountType::Initialized(UiStakeAccount {
meta: UiMeta {
rent_exempt_reserve: 42.to_string(),
authorized: UiAuthorized {
staker: pubkey.to_string(),
withdrawer: pubkey.to_string(),
},
lockup: UiLockup {
unix_timestamp: 0,
epoch: 1,
custodian: custodian.to_string(),
}
},
stake: None,
})
);
let voter_pubkey = solana_sdk::pubkey::new_rand();
let stake = Stake {
delegation: Delegation {
voter_pubkey,
stake: 20,
activation_epoch: 2,
deactivation_epoch: std::u64::MAX,
warmup_cooldown_rate: 0.25,
},
credits_observed: 10,
};
let stake_state = StakeState::Stake(meta, stake);
let stake_data = serialize(&stake_state).unwrap();
assert_eq!(
parse_stake(&stake_data).unwrap(),
StakeAccountType::Delegated(UiStakeAccount {
meta: UiMeta {
rent_exempt_reserve: 42.to_string(),
authorized: UiAuthorized {
staker: pubkey.to_string(),
withdrawer: pubkey.to_string(),
},
lockup: UiLockup {
unix_timestamp: 0,
epoch: 1,
custodian: custodian.to_string(),
}
},
stake: Some(UiStake {
delegation: UiDelegation {
voter: voter_pubkey.to_string(),
stake: 20.to_string(),
activation_epoch: 2.to_string(),
deactivation_epoch: std::u64::MAX.to_string(),
warmup_cooldown_rate: 0.25,
},
credits_observed: 10,
})
})
);
let stake_state = StakeState::RewardsPool;
let stake_data = serialize(&stake_state).unwrap();
assert_eq!(
parse_stake(&stake_data).unwrap(),
StakeAccountType::RewardsPool
);
let bad_data = vec![1, 2, 3, 4];
assert!(parse_stake(&bad_data).is_err());
}
}

View File

@ -1,329 +0,0 @@
use crate::{
parse_account_data::{ParsableAccount, ParseAccountError},
StringAmount, UiFeeCalculator,
};
use bincode::deserialize;
use bv::BitVec;
use solana_sdk::{
clock::{Clock, Epoch, Slot, UnixTimestamp},
epoch_schedule::EpochSchedule,
pubkey::Pubkey,
rent::Rent,
slot_hashes::SlotHashes,
slot_history::{self, SlotHistory},
stake_history::{StakeHistory, StakeHistoryEntry},
sysvar::{self, fees::Fees, recent_blockhashes::RecentBlockhashes, rewards::Rewards},
};
pub fn parse_sysvar(data: &[u8], pubkey: &Pubkey) -> Result<SysvarAccountType, ParseAccountError> {
let parsed_account = {
if pubkey == &sysvar::clock::id() {
deserialize::<Clock>(data)
.ok()
.map(|clock| SysvarAccountType::Clock(clock.into()))
} else if pubkey == &sysvar::epoch_schedule::id() {
deserialize(data).ok().map(SysvarAccountType::EpochSchedule)
} else if pubkey == &sysvar::fees::id() {
deserialize::<Fees>(data)
.ok()
.map(|fees| SysvarAccountType::Fees(fees.into()))
} else if pubkey == &sysvar::recent_blockhashes::id() {
deserialize::<RecentBlockhashes>(data)
.ok()
.map(|recent_blockhashes| {
let recent_blockhashes = recent_blockhashes
.iter()
.map(|entry| UiRecentBlockhashesEntry {
blockhash: entry.blockhash.to_string(),
fee_calculator: entry.fee_calculator.clone().into(),
})
.collect();
SysvarAccountType::RecentBlockhashes(recent_blockhashes)
})
} else if pubkey == &sysvar::rent::id() {
deserialize::<Rent>(data)
.ok()
.map(|rent| SysvarAccountType::Rent(rent.into()))
} else if pubkey == &sysvar::rewards::id() {
deserialize::<Rewards>(data)
.ok()
.map(|rewards| SysvarAccountType::Rewards(rewards.into()))
} else if pubkey == &sysvar::slot_hashes::id() {
deserialize::<SlotHashes>(data).ok().map(|slot_hashes| {
let slot_hashes = slot_hashes
.iter()
.map(|slot_hash| UiSlotHashEntry {
slot: slot_hash.0,
hash: slot_hash.1.to_string(),
})
.collect();
SysvarAccountType::SlotHashes(slot_hashes)
})
} else if pubkey == &sysvar::slot_history::id() {
deserialize::<SlotHistory>(data).ok().map(|slot_history| {
SysvarAccountType::SlotHistory(UiSlotHistory {
next_slot: slot_history.next_slot,
bits: format!("{:?}", SlotHistoryBits(slot_history.bits)),
})
})
} else if pubkey == &sysvar::stake_history::id() {
deserialize::<StakeHistory>(data).ok().map(|stake_history| {
let stake_history = stake_history
.iter()
.map(|entry| UiStakeHistoryEntry {
epoch: entry.0,
stake_history: entry.1.clone(),
})
.collect();
SysvarAccountType::StakeHistory(stake_history)
})
} else {
None
}
};
parsed_account.ok_or(ParseAccountError::AccountNotParsable(
ParsableAccount::Sysvar,
))
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase", tag = "type", content = "info")]
pub enum SysvarAccountType {
Clock(UiClock),
EpochSchedule(EpochSchedule),
Fees(UiFees),
RecentBlockhashes(Vec<UiRecentBlockhashesEntry>),
Rent(UiRent),
Rewards(UiRewards),
SlotHashes(Vec<UiSlotHashEntry>),
SlotHistory(UiSlotHistory),
StakeHistory(Vec<UiStakeHistoryEntry>),
}
#[derive(Debug, Serialize, Deserialize, PartialEq, Default)]
#[serde(rename_all = "camelCase")]
pub struct UiClock {
pub slot: Slot,
pub epoch: Epoch,
pub epoch_start_timestamp: UnixTimestamp,
pub leader_schedule_epoch: Epoch,
pub unix_timestamp: UnixTimestamp,
}
impl From<Clock> for UiClock {
fn from(clock: Clock) -> Self {
Self {
slot: clock.slot,
epoch: clock.epoch,
epoch_start_timestamp: clock.epoch_start_timestamp,
leader_schedule_epoch: clock.leader_schedule_epoch,
unix_timestamp: clock.unix_timestamp,
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq, Default)]
#[serde(rename_all = "camelCase")]
pub struct UiFees {
pub fee_calculator: UiFeeCalculator,
}
impl From<Fees> for UiFees {
fn from(fees: Fees) -> Self {
Self {
fee_calculator: fees.fee_calculator.into(),
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq, Default)]
#[serde(rename_all = "camelCase")]
pub struct UiRent {
pub lamports_per_byte_year: StringAmount,
pub exemption_threshold: f64,
pub burn_percent: u8,
}
impl From<Rent> for UiRent {
fn from(rent: Rent) -> Self {
Self {
lamports_per_byte_year: rent.lamports_per_byte_year.to_string(),
exemption_threshold: rent.exemption_threshold,
burn_percent: rent.burn_percent,
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq, Default)]
#[serde(rename_all = "camelCase")]
pub struct UiRewards {
pub validator_point_value: f64,
}
impl From<Rewards> for UiRewards {
fn from(rewards: Rewards) -> Self {
Self {
validator_point_value: rewards.validator_point_value,
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiRecentBlockhashesEntry {
pub blockhash: String,
pub fee_calculator: UiFeeCalculator,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiSlotHashEntry {
pub slot: Slot,
pub hash: String,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiSlotHistory {
pub next_slot: Slot,
pub bits: String,
}
struct SlotHistoryBits(BitVec<u64>);
impl std::fmt::Debug for SlotHistoryBits {
fn fmt(&self, f: &mut std::fmt::Formatter<'_>) -> std::fmt::Result {
for i in 0..slot_history::MAX_ENTRIES {
if self.0.get(i) {
write!(f, "1")?;
} else {
write!(f, "0")?;
}
}
Ok(())
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiStakeHistoryEntry {
pub epoch: Epoch,
pub stake_history: StakeHistoryEntry,
}
#[cfg(test)]
mod test {
use super::*;
use solana_sdk::{
account::create_account, fee_calculator::FeeCalculator, hash::Hash,
sysvar::recent_blockhashes::IterItem,
};
#[test]
fn test_parse_sysvars() {
let clock_sysvar = create_account(&Clock::default(), 1);
assert_eq!(
parse_sysvar(&clock_sysvar.data, &sysvar::clock::id()).unwrap(),
SysvarAccountType::Clock(UiClock::default()),
);
let epoch_schedule = EpochSchedule {
slots_per_epoch: 12,
leader_schedule_slot_offset: 0,
warmup: false,
first_normal_epoch: 1,
first_normal_slot: 12,
};
let epoch_schedule_sysvar = create_account(&epoch_schedule, 1);
assert_eq!(
parse_sysvar(&epoch_schedule_sysvar.data, &sysvar::epoch_schedule::id()).unwrap(),
SysvarAccountType::EpochSchedule(epoch_schedule),
);
let fees_sysvar = create_account(&Fees::default(), 1);
assert_eq!(
parse_sysvar(&fees_sysvar.data, &sysvar::fees::id()).unwrap(),
SysvarAccountType::Fees(UiFees::default()),
);
let hash = Hash::new(&[1; 32]);
let fee_calculator = FeeCalculator {
lamports_per_signature: 10,
};
let recent_blockhashes: RecentBlockhashes = vec![IterItem(0, &hash, &fee_calculator)]
.into_iter()
.collect();
let recent_blockhashes_sysvar = create_account(&recent_blockhashes, 1);
assert_eq!(
parse_sysvar(
&recent_blockhashes_sysvar.data,
&sysvar::recent_blockhashes::id()
)
.unwrap(),
SysvarAccountType::RecentBlockhashes(vec![UiRecentBlockhashesEntry {
blockhash: hash.to_string(),
fee_calculator: fee_calculator.into(),
}]),
);
let rent = Rent {
lamports_per_byte_year: 10,
exemption_threshold: 2.0,
burn_percent: 5,
};
let rent_sysvar = create_account(&rent, 1);
assert_eq!(
parse_sysvar(&rent_sysvar.data, &sysvar::rent::id()).unwrap(),
SysvarAccountType::Rent(rent.into()),
);
let rewards_sysvar = create_account(&Rewards::default(), 1);
assert_eq!(
parse_sysvar(&rewards_sysvar.data, &sysvar::rewards::id()).unwrap(),
SysvarAccountType::Rewards(UiRewards::default()),
);
let mut slot_hashes = SlotHashes::default();
slot_hashes.add(1, hash);
let slot_hashes_sysvar = create_account(&slot_hashes, 1);
assert_eq!(
parse_sysvar(&slot_hashes_sysvar.data, &sysvar::slot_hashes::id()).unwrap(),
SysvarAccountType::SlotHashes(vec![UiSlotHashEntry {
slot: 1,
hash: hash.to_string(),
}]),
);
let mut slot_history = SlotHistory::default();
slot_history.add(42);
let slot_history_sysvar = create_account(&slot_history, 1);
assert_eq!(
parse_sysvar(&slot_history_sysvar.data, &sysvar::slot_history::id()).unwrap(),
SysvarAccountType::SlotHistory(UiSlotHistory {
next_slot: slot_history.next_slot,
bits: format!("{:?}", SlotHistoryBits(slot_history.bits)),
}),
);
let mut stake_history = StakeHistory::default();
let stake_history_entry = StakeHistoryEntry {
effective: 10,
activating: 2,
deactivating: 3,
};
stake_history.add(1, stake_history_entry.clone());
let stake_history_sysvar = create_account(&stake_history, 1);
assert_eq!(
parse_sysvar(&stake_history_sysvar.data, &sysvar::stake_history::id()).unwrap(),
SysvarAccountType::StakeHistory(vec![UiStakeHistoryEntry {
epoch: 1,
stake_history: stake_history_entry,
}]),
);
let bad_pubkey = solana_sdk::pubkey::new_rand();
assert!(parse_sysvar(&stake_history_sysvar.data, &bad_pubkey).is_err());
let bad_data = vec![0; 4];
assert!(parse_sysvar(&bad_data, &sysvar::stake_history::id()).is_err());
}
}

View File

@ -1,352 +0,0 @@
use crate::{
parse_account_data::{ParsableAccount, ParseAccountError},
StringAmount,
};
use solana_sdk::pubkey::Pubkey;
use spl_token_v2_0::{
solana_program::{
program_option::COption, program_pack::Pack, pubkey::Pubkey as SplTokenPubkey,
},
state::{Account, AccountState, Mint, Multisig},
};
use std::str::FromStr;
// A helper function to convert spl_token_v2_0::id() as spl_sdk::pubkey::Pubkey to
// solana_sdk::pubkey::Pubkey
pub fn spl_token_id_v2_0() -> Pubkey {
Pubkey::from_str(&spl_token_v2_0::id().to_string()).unwrap()
}
// A helper function to convert spl_token_v2_0::native_mint::id() as spl_sdk::pubkey::Pubkey to
// solana_sdk::pubkey::Pubkey
pub fn spl_token_v2_0_native_mint() -> Pubkey {
Pubkey::from_str(&spl_token_v2_0::native_mint::id().to_string()).unwrap()
}
// A helper function to convert a solana_sdk::pubkey::Pubkey to spl_sdk::pubkey::Pubkey
pub fn spl_token_v2_0_pubkey(pubkey: &Pubkey) -> SplTokenPubkey {
SplTokenPubkey::from_str(&pubkey.to_string()).unwrap()
}
// A helper function to convert a spl_sdk::pubkey::Pubkey to solana_sdk::pubkey::Pubkey
pub fn pubkey_from_spl_token_v2_0(pubkey: &SplTokenPubkey) -> Pubkey {
Pubkey::from_str(&pubkey.to_string()).unwrap()
}
pub fn parse_token(
data: &[u8],
mint_decimals: Option<u8>,
) -> Result<TokenAccountType, ParseAccountError> {
if data.len() == Account::get_packed_len() {
let account = Account::unpack(data)
.map_err(|_| ParseAccountError::AccountNotParsable(ParsableAccount::SplToken))?;
let decimals = mint_decimals.ok_or_else(|| {
ParseAccountError::AdditionalDataMissing(
"no mint_decimals provided to parse spl-token account".to_string(),
)
})?;
Ok(TokenAccountType::Account(UiTokenAccount {
mint: account.mint.to_string(),
owner: account.owner.to_string(),
token_amount: token_amount_to_ui_amount(account.amount, decimals),
delegate: match account.delegate {
COption::Some(pubkey) => Some(pubkey.to_string()),
COption::None => None,
},
state: account.state.into(),
is_native: account.is_native(),
rent_exempt_reserve: match account.is_native {
COption::Some(reserve) => Some(token_amount_to_ui_amount(reserve, decimals)),
COption::None => None,
},
delegated_amount: if account.delegate.is_none() {
None
} else {
Some(token_amount_to_ui_amount(
account.delegated_amount,
decimals,
))
},
close_authority: match account.close_authority {
COption::Some(pubkey) => Some(pubkey.to_string()),
COption::None => None,
},
}))
} else if data.len() == Mint::get_packed_len() {
let mint = Mint::unpack(data)
.map_err(|_| ParseAccountError::AccountNotParsable(ParsableAccount::SplToken))?;
Ok(TokenAccountType::Mint(UiMint {
mint_authority: match mint.mint_authority {
COption::Some(pubkey) => Some(pubkey.to_string()),
COption::None => None,
},
supply: mint.supply.to_string(),
decimals: mint.decimals,
is_initialized: mint.is_initialized,
freeze_authority: match mint.freeze_authority {
COption::Some(pubkey) => Some(pubkey.to_string()),
COption::None => None,
},
}))
} else if data.len() == Multisig::get_packed_len() {
let multisig = Multisig::unpack(data)
.map_err(|_| ParseAccountError::AccountNotParsable(ParsableAccount::SplToken))?;
Ok(TokenAccountType::Multisig(UiMultisig {
num_required_signers: multisig.m,
num_valid_signers: multisig.n,
is_initialized: multisig.is_initialized,
signers: multisig
.signers
.iter()
.filter_map(|pubkey| {
if pubkey != &SplTokenPubkey::default() {
Some(pubkey.to_string())
} else {
None
}
})
.collect(),
}))
} else {
Err(ParseAccountError::AccountNotParsable(
ParsableAccount::SplToken,
))
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase", tag = "type", content = "info")]
pub enum TokenAccountType {
Account(UiTokenAccount),
Mint(UiMint),
Multisig(UiMultisig),
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiTokenAccount {
pub mint: String,
pub owner: String,
pub token_amount: UiTokenAmount,
#[serde(skip_serializing_if = "Option::is_none")]
pub delegate: Option<String>,
pub state: UiAccountState,
pub is_native: bool,
#[serde(skip_serializing_if = "Option::is_none")]
pub rent_exempt_reserve: Option<UiTokenAmount>,
#[serde(skip_serializing_if = "Option::is_none")]
pub delegated_amount: Option<UiTokenAmount>,
#[serde(skip_serializing_if = "Option::is_none")]
pub close_authority: Option<String>,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub enum UiAccountState {
Uninitialized,
Initialized,
Frozen,
}
impl From<AccountState> for UiAccountState {
fn from(state: AccountState) -> Self {
match state {
AccountState::Uninitialized => UiAccountState::Uninitialized,
AccountState::Initialized => UiAccountState::Initialized,
AccountState::Frozen => UiAccountState::Frozen,
}
}
}
#[derive(Serialize, Deserialize, Clone, Debug, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiTokenAmount {
pub ui_amount: f64,
pub decimals: u8,
pub amount: StringAmount,
}
impl UiTokenAmount {
pub fn real_number_string(&self) -> String {
let decimals = self.decimals as usize;
if decimals > 0 {
let amount = u64::from_str(&self.amount).unwrap_or(0);
// Left-pad zeros to decimals + 1, so we at least have an integer zero
let mut s = format!("{:01$}", amount, decimals + 1);
// Add the decimal point (Sorry, "," locales!)
s.insert(s.len() - decimals, '.');
s
} else {
self.amount.clone()
}
}
pub fn real_number_string_trimmed(&self) -> String {
let s = self.real_number_string();
let zeros_trimmed = s.trim_end_matches('0');
let decimal_trimmed = zeros_trimmed.trim_end_matches('.');
decimal_trimmed.to_string()
}
}
pub fn token_amount_to_ui_amount(amount: u64, decimals: u8) -> UiTokenAmount {
// Use `amount_to_ui_amount()` once spl_token is bumped to a version that supports it: https://github.com/solana-labs/solana-program-library/pull/211
let amount_decimals = amount as f64 / 10_usize.pow(decimals as u32) as f64;
UiTokenAmount {
ui_amount: amount_decimals,
decimals,
amount: amount.to_string(),
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiMint {
pub mint_authority: Option<String>,
pub supply: StringAmount,
pub decimals: u8,
pub is_initialized: bool,
pub freeze_authority: Option<String>,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiMultisig {
pub num_required_signers: u8,
pub num_valid_signers: u8,
pub is_initialized: bool,
pub signers: Vec<String>,
}
pub fn get_token_account_mint(data: &[u8]) -> Option<Pubkey> {
if data.len() == Account::get_packed_len() {
Some(Pubkey::new(&data[0..32]))
} else {
None
}
}
#[cfg(test)]
mod test {
use super::*;
#[test]
fn test_parse_token() {
let mint_pubkey = SplTokenPubkey::new(&[2; 32]);
let owner_pubkey = SplTokenPubkey::new(&[3; 32]);
let mut account_data = vec![0; Account::get_packed_len()];
let mut account = Account::unpack_unchecked(&account_data).unwrap();
account.mint = mint_pubkey;
account.owner = owner_pubkey;
account.amount = 42;
account.state = AccountState::Initialized;
account.is_native = COption::None;
account.close_authority = COption::Some(owner_pubkey);
Account::pack(account, &mut account_data).unwrap();
assert!(parse_token(&account_data, None).is_err());
assert_eq!(
parse_token(&account_data, Some(2)).unwrap(),
TokenAccountType::Account(UiTokenAccount {
mint: mint_pubkey.to_string(),
owner: owner_pubkey.to_string(),
token_amount: UiTokenAmount {
ui_amount: 0.42,
decimals: 2,
amount: "42".to_string()
},
delegate: None,
state: UiAccountState::Initialized,
is_native: false,
rent_exempt_reserve: None,
delegated_amount: None,
close_authority: Some(owner_pubkey.to_string()),
}),
);
let mut mint_data = vec![0; Mint::get_packed_len()];
let mut mint = Mint::unpack_unchecked(&mint_data).unwrap();
mint.mint_authority = COption::Some(owner_pubkey);
mint.supply = 42;
mint.decimals = 3;
mint.is_initialized = true;
mint.freeze_authority = COption::Some(owner_pubkey);
Mint::pack(mint, &mut mint_data).unwrap();
assert_eq!(
parse_token(&mint_data, None).unwrap(),
TokenAccountType::Mint(UiMint {
mint_authority: Some(owner_pubkey.to_string()),
supply: 42.to_string(),
decimals: 3,
is_initialized: true,
freeze_authority: Some(owner_pubkey.to_string()),
}),
);
let signer1 = SplTokenPubkey::new(&[1; 32]);
let signer2 = SplTokenPubkey::new(&[2; 32]);
let signer3 = SplTokenPubkey::new(&[3; 32]);
let mut multisig_data = vec![0; Multisig::get_packed_len()];
let mut signers = [SplTokenPubkey::default(); 11];
signers[0] = signer1;
signers[1] = signer2;
signers[2] = signer3;
let mut multisig = Multisig::unpack_unchecked(&multisig_data).unwrap();
multisig.m = 2;
multisig.n = 3;
multisig.is_initialized = true;
multisig.signers = signers;
Multisig::pack(multisig, &mut multisig_data).unwrap();
assert_eq!(
parse_token(&multisig_data, None).unwrap(),
TokenAccountType::Multisig(UiMultisig {
num_required_signers: 2,
num_valid_signers: 3,
is_initialized: true,
signers: vec![
signer1.to_string(),
signer2.to_string(),
signer3.to_string()
],
}),
);
let bad_data = vec![0; 4];
assert!(parse_token(&bad_data, None).is_err());
}
#[test]
fn test_get_token_account_mint() {
let mint_pubkey = SplTokenPubkey::new(&[2; 32]);
let mut account_data = vec![0; Account::get_packed_len()];
let mut account = Account::unpack_unchecked(&account_data).unwrap();
account.mint = mint_pubkey;
Account::pack(account, &mut account_data).unwrap();
let expected_mint_pubkey = Pubkey::new(&[2; 32]);
assert_eq!(
get_token_account_mint(&account_data),
Some(expected_mint_pubkey)
);
}
#[test]
fn test_ui_token_amount_real_string() {
let token_amount = token_amount_to_ui_amount(1, 0);
assert_eq!(&token_amount.real_number_string(), "1");
assert_eq!(&token_amount.real_number_string_trimmed(), "1");
let token_amount = token_amount_to_ui_amount(1, 9);
assert_eq!(&token_amount.real_number_string(), "0.000000001");
assert_eq!(&token_amount.real_number_string_trimmed(), "0.000000001");
let token_amount = token_amount_to_ui_amount(1_000_000_000, 9);
assert_eq!(&token_amount.real_number_string(), "1.000000000");
assert_eq!(&token_amount.real_number_string_trimmed(), "1");
let token_amount = token_amount_to_ui_amount(1_234_567_890, 3);
assert_eq!(&token_amount.real_number_string(), "1234567.890");
assert_eq!(&token_amount.real_number_string_trimmed(), "1234567.89");
}
}

View File

@ -1,146 +0,0 @@
use crate::{parse_account_data::ParseAccountError, StringAmount};
use solana_sdk::{
clock::{Epoch, Slot},
pubkey::Pubkey,
};
use solana_vote_program::vote_state::{BlockTimestamp, Lockout, VoteState};
pub fn parse_vote(data: &[u8]) -> Result<VoteAccountType, ParseAccountError> {
let mut vote_state = VoteState::deserialize(data).map_err(ParseAccountError::from)?;
let epoch_credits = vote_state
.epoch_credits()
.iter()
.map(|(epoch, credits, previous_credits)| UiEpochCredits {
epoch: *epoch,
credits: credits.to_string(),
previous_credits: previous_credits.to_string(),
})
.collect();
let votes = vote_state
.votes
.iter()
.map(|lockout| UiLockout {
slot: lockout.slot,
confirmation_count: lockout.confirmation_count,
})
.collect();
let authorized_voters = vote_state
.authorized_voters()
.iter()
.map(|(epoch, authorized_voter)| UiAuthorizedVoters {
epoch: *epoch,
authorized_voter: authorized_voter.to_string(),
})
.collect();
let prior_voters = vote_state
.prior_voters()
.buf()
.iter()
.filter(|(pubkey, _, _)| pubkey != &Pubkey::default())
.map(
|(authorized_pubkey, epoch_of_last_authorized_switch, target_epoch)| UiPriorVoters {
authorized_pubkey: authorized_pubkey.to_string(),
epoch_of_last_authorized_switch: *epoch_of_last_authorized_switch,
target_epoch: *target_epoch,
},
)
.collect();
Ok(VoteAccountType::Vote(UiVoteState {
node_pubkey: vote_state.node_pubkey.to_string(),
authorized_withdrawer: vote_state.authorized_withdrawer.to_string(),
commission: vote_state.commission,
votes,
root_slot: vote_state.root_slot,
authorized_voters,
prior_voters,
epoch_credits,
last_timestamp: vote_state.last_timestamp,
}))
}
/// A wrapper enum for consistency across programs
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase", tag = "type", content = "info")]
pub enum VoteAccountType {
Vote(UiVoteState),
}
/// A duplicate representation of VoteState for pretty JSON serialization
#[derive(Debug, Serialize, Deserialize, Default, PartialEq)]
#[serde(rename_all = "camelCase")]
pub struct UiVoteState {
node_pubkey: String,
authorized_withdrawer: String,
commission: u8,
votes: Vec<UiLockout>,
root_slot: Option<Slot>,
authorized_voters: Vec<UiAuthorizedVoters>,
prior_voters: Vec<UiPriorVoters>,
epoch_credits: Vec<UiEpochCredits>,
last_timestamp: BlockTimestamp,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
struct UiLockout {
slot: Slot,
confirmation_count: u32,
}
impl From<&Lockout> for UiLockout {
fn from(lockout: &Lockout) -> Self {
Self {
slot: lockout.slot,
confirmation_count: lockout.confirmation_count,
}
}
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
struct UiAuthorizedVoters {
epoch: Epoch,
authorized_voter: String,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
struct UiPriorVoters {
authorized_pubkey: String,
epoch_of_last_authorized_switch: Epoch,
target_epoch: Epoch,
}
#[derive(Debug, Serialize, Deserialize, PartialEq)]
#[serde(rename_all = "camelCase")]
struct UiEpochCredits {
epoch: Epoch,
credits: StringAmount,
previous_credits: StringAmount,
}
#[cfg(test)]
mod test {
use super::*;
use solana_vote_program::vote_state::VoteStateVersions;
#[test]
fn test_parse_vote() {
let vote_state = VoteState::default();
let mut vote_account_data: Vec<u8> = vec![0; VoteState::size_of()];
let versioned = VoteStateVersions::new_current(vote_state);
VoteState::serialize(&versioned, &mut vote_account_data).unwrap();
let expected_vote_state = UiVoteState {
node_pubkey: Pubkey::default().to_string(),
authorized_withdrawer: Pubkey::default().to_string(),
..UiVoteState::default()
};
assert_eq!(
parse_vote(&vote_account_data).unwrap(),
VoteAccountType::Vote(expected_vote_state)
);
let bad_data = vec![0; 4];
assert!(parse_vote(&bad_data).is_err());
}
}

View File

@ -1,18 +0,0 @@
use solana_config_program::ConfigState;
pub const MAX_SHORT_FIELD_LENGTH: usize = 70;
pub const MAX_LONG_FIELD_LENGTH: usize = 300;
pub const MAX_VALIDATOR_INFO: u64 = 576;
solana_sdk::declare_id!("Va1idator1nfo111111111111111111111111111111");
#[derive(Debug, Deserialize, PartialEq, Serialize, Default)]
pub struct ValidatorInfo {
pub info: String,
}
impl ConfigState for ValidatorInfo {
fn max_space() -> u64 {
MAX_VALIDATOR_INFO
}
}

View File

@ -1,24 +0,0 @@
[package]
authors = ["Solana Maintainers <maintainers@solana.foundation>"]
edition = "2018"
name = "solana-accounts-bench"
version = "1.5.4"
repository = "https://github.com/solana-labs/solana"
license = "Apache-2.0"
homepage = "https://solana.com/"
publish = false
[dependencies]
log = "0.4.11"
rayon = "1.4.0"
solana-logger = { path = "../logger", version = "1.5.4" }
solana-runtime = { path = "../runtime", version = "1.5.4" }
solana-measure = { path = "../measure", version = "1.5.4" }
solana-sdk = { path = "../sdk", version = "1.5.4" }
solana-version = { path = "../version", version = "1.5.4" }
rand = "0.7.0"
clap = "2.33.1"
crossbeam-channel = "0.4"
[package.metadata.docs.rs]
targets = ["x86_64-unknown-linux-gnu"]

View File

@ -1,109 +0,0 @@
use clap::{crate_description, crate_name, value_t, App, Arg};
use rayon::prelude::*;
use solana_measure::measure::Measure;
use solana_runtime::{
accounts::{create_test_accounts, update_accounts_bench, Accounts},
accounts_index::Ancestors,
};
use solana_sdk::{genesis_config::ClusterType, pubkey::Pubkey};
use std::{collections::HashSet, env, fs, path::PathBuf};
fn main() {
solana_logger::setup();
let matches = App::new(crate_name!())
.about(crate_description!())
.version(solana_version::version!())
.arg(
Arg::with_name("num_slots")
.long("num_slots")
.takes_value(true)
.value_name("SLOTS")
.help("Number of slots to store to."),
)
.arg(
Arg::with_name("num_accounts")
.long("num_accounts")
.takes_value(true)
.value_name("NUM_ACCOUNTS")
.help("Total number of accounts"),
)
.arg(
Arg::with_name("iterations")
.long("iterations")
.takes_value(true)
.value_name("ITERATIONS")
.help("Number of bench iterations"),
)
.arg(
Arg::with_name("clean")
.long("clean")
.takes_value(false)
.help("Run clean"),
)
.get_matches();
let num_slots = value_t!(matches, "num_slots", usize).unwrap_or(4);
let num_accounts = value_t!(matches, "num_accounts", usize).unwrap_or(10_000);
let iterations = value_t!(matches, "iterations", usize).unwrap_or(20);
let clean = matches.is_present("clean");
println!("clean: {:?}", clean);
let path = PathBuf::from(env::var("FARF_DIR").unwrap_or_else(|_| "farf".to_owned()))
.join("accounts-bench");
if fs::remove_dir_all(path.clone()).is_err() {
println!("Warning: Couldn't remove {:?}", path);
}
let accounts =
Accounts::new_with_config(vec![path], &ClusterType::Testnet, HashSet::new(), false);
println!("Creating {} accounts", num_accounts);
let mut create_time = Measure::start("create accounts");
let pubkeys: Vec<_> = (0..num_slots)
.into_par_iter()
.map(|slot| {
let mut pubkeys: Vec<Pubkey> = vec![];
create_test_accounts(
&accounts,
&mut pubkeys,
num_accounts / num_slots,
slot as u64,
);
pubkeys
})
.collect();
let pubkeys: Vec<_> = pubkeys.into_iter().flatten().collect();
create_time.stop();
println!(
"created {} accounts in {} slots {}",
(num_accounts / num_slots) * num_slots,
num_slots,
create_time
);
let mut ancestors: Ancestors = vec![(0, 0)].into_iter().collect();
for i in 1..num_slots {
ancestors.insert(i as u64, i - 1);
accounts.add_root(i as u64);
}
for x in 0..iterations {
if clean {
let mut time = Measure::start("clean");
accounts.accounts_db.clean_accounts(None);
time.stop();
println!("{}", time);
for slot in 0..num_slots {
update_accounts_bench(&accounts, &pubkeys, ((x + 1) * num_slots + slot) as u64);
accounts.add_root((x * num_slots + slot) as u64);
}
} else {
let mut pubkeys: Vec<Pubkey> = vec![];
let mut time = Measure::start("hash");
let hash = accounts
.accounts_db
.update_accounts_hash(0, &ancestors, true)
.0;
time.stop();
println!("hash: {} {}", hash, time);
create_test_accounts(&accounts, &mut pubkeys, 1, 0);
}
}
}

View File

@ -1,2 +0,0 @@
/target/
/farf/

View File

@ -1,29 +0,0 @@
[package]
authors = ["Solana Maintainers <maintainers@solana.foundation>"]
edition = "2018"
name = "solana-banking-bench"
version = "1.5.4"
repository = "https://github.com/solana-labs/solana"
license = "Apache-2.0"
homepage = "https://solana.com/"
publish = false
[dependencies]
clap = "2.33.1"
crossbeam-channel = "0.4"
log = "0.4.11"
rand = "0.7.0"
rayon = "1.4.0"
solana-core = { path = "../core", version = "1.5.4" }
solana-clap-utils = { path = "../clap-utils", version = "1.5.4" }
solana-streamer = { path = "../streamer", version = "1.5.4" }
solana-perf = { path = "../perf", version = "1.5.4" }
solana-ledger = { path = "../ledger", version = "1.5.4" }
solana-logger = { path = "../logger", version = "1.5.4" }
solana-runtime = { path = "../runtime", version = "1.5.4" }
solana-measure = { path = "../measure", version = "1.5.4" }
solana-sdk = { path = "../sdk", version = "1.5.4" }
solana-version = { path = "../version", version = "1.5.4" }
[package.metadata.docs.rs]
targets = ["x86_64-unknown-linux-gnu"]

View File

@ -1,391 +0,0 @@
use clap::{crate_description, crate_name, value_t, App, Arg};
use crossbeam_channel::unbounded;
use log::*;
use rand::{thread_rng, Rng};
use rayon::prelude::*;
use solana_core::{
banking_stage::{create_test_recorder, BankingStage},
cluster_info::ClusterInfo,
cluster_info::Node,
poh_recorder::PohRecorder,
poh_recorder::WorkingBankEntry,
};
use solana_ledger::{
blockstore::Blockstore,
genesis_utils::{create_genesis_config, GenesisConfigInfo},
get_tmp_ledger_path,
};
use solana_measure::measure::Measure;
use solana_perf::packet::to_packets_chunked;
use solana_runtime::{
accounts_background_service::ABSRequestSender, bank::Bank, bank_forks::BankForks,
};
use solana_sdk::{
hash::Hash,
signature::Keypair,
signature::Signature,
system_transaction,
timing::{duration_as_us, timestamp},
transaction::Transaction,
};
use std::{
sync::{atomic::Ordering, mpsc::Receiver, Arc, Mutex},
thread::sleep,
time::{Duration, Instant},
};
fn check_txs(
receiver: &Arc<Receiver<WorkingBankEntry>>,
ref_tx_count: usize,
poh_recorder: &Arc<Mutex<PohRecorder>>,
) -> bool {
let mut total = 0;
let now = Instant::now();
let mut no_bank = false;
loop {
if let Ok((_bank, (entry, _tick_height))) = receiver.recv_timeout(Duration::from_millis(10))
{
total += entry.transactions.len();
}
if total >= ref_tx_count {
break;
}
if now.elapsed().as_secs() > 60 {
break;
}
if poh_recorder.lock().unwrap().bank().is_none() {
trace!("no bank");
no_bank = true;
break;
}
}
if !no_bank {
assert!(total >= ref_tx_count);
}
no_bank
}
fn make_accounts_txs(
total_num_transactions: usize,
hash: Hash,
same_payer: bool,
) -> Vec<Transaction> {
let to_pubkey = solana_sdk::pubkey::new_rand();
let payer_key = Keypair::new();
let dummy = system_transaction::transfer(&payer_key, &to_pubkey, 1, hash);
(0..total_num_transactions)
.into_par_iter()
.map(|_| {
let mut new = dummy.clone();
let sig: Vec<u8> = (0..64).map(|_| thread_rng().gen()).collect();
if !same_payer {
new.message.account_keys[0] = solana_sdk::pubkey::new_rand();
}
new.message.account_keys[1] = solana_sdk::pubkey::new_rand();
new.signatures = vec![Signature::new(&sig[0..64])];
new
})
.collect()
}
struct Config {
packets_per_batch: usize,
chunk_len: usize,
num_threads: usize,
}
impl Config {
fn get_transactions_index(&self, chunk_index: usize) -> usize {
chunk_index * (self.chunk_len / self.num_threads) * self.packets_per_batch
}
}
fn bytes_as_usize(bytes: &[u8]) -> usize {
bytes[0] as usize | (bytes[1] as usize) << 8
}
#[allow(clippy::cognitive_complexity)]
fn main() {
solana_logger::setup();
let matches = App::new(crate_name!())
.about(crate_description!())
.version(solana_version::version!())
.arg(
Arg::with_name("num_chunks")
.long("num-chunks")
.takes_value(true)
.value_name("SIZE")
.help("Number of transaction chunks."),
)
.arg(
Arg::with_name("packets_per_chunk")
.long("packets-per-chunk")
.takes_value(true)
.value_name("SIZE")
.help("Packets per chunk"),
)
.arg(
Arg::with_name("skip_sanity")
.long("skip-sanity")
.takes_value(false)
.help("Skip transaction sanity execution"),
)
.arg(
Arg::with_name("same_payer")
.long("same-payer")
.takes_value(false)
.help("Use the same payer for transfers"),
)
.arg(
Arg::with_name("iterations")
.long("iterations")
.takes_value(true)
.help("Number of iterations"),
)
.arg(
Arg::with_name("num_threads")
.long("num-threads")
.takes_value(true)
.help("Number of iterations"),
)
.get_matches();
let num_threads =
value_t!(matches, "num_threads", usize).unwrap_or(BankingStage::num_threads() as usize);
// a multiple of packet chunk duplicates to avoid races
let num_chunks = value_t!(matches, "num_chunks", usize).unwrap_or(16);
let packets_per_chunk = value_t!(matches, "packets_per_chunk", usize).unwrap_or(192);
let iterations = value_t!(matches, "iterations", usize).unwrap_or(1000);
let total_num_transactions = num_chunks * num_threads * packets_per_chunk;
let mint_total = 1_000_000_000_000;
let GenesisConfigInfo {
genesis_config,
mint_keypair,
..
} = create_genesis_config(mint_total);
let (verified_sender, verified_receiver) = unbounded();
let (vote_sender, vote_receiver) = unbounded();
let (replay_vote_sender, _replay_vote_receiver) = unbounded();
let bank0 = Bank::new(&genesis_config);
let mut bank_forks = BankForks::new(bank0);
let mut bank = bank_forks.working_bank();
info!("threads: {} txs: {}", num_threads, total_num_transactions);
let same_payer = matches.is_present("same_payer");
let mut transactions =
make_accounts_txs(total_num_transactions, genesis_config.hash(), same_payer);
// fund all the accounts
transactions.iter().for_each(|tx| {
let mut fund = system_transaction::transfer(
&mint_keypair,
&tx.message.account_keys[0],
mint_total / total_num_transactions as u64,
genesis_config.hash(),
);
// Ignore any pesky duplicate signature errors in the case we are using single-payer
let sig: Vec<u8> = (0..64).map(|_| thread_rng().gen()).collect();
fund.signatures = vec![Signature::new(&sig[0..64])];
let x = bank.process_transaction(&fund);
x.unwrap();
});
let skip_sanity = matches.is_present("skip_sanity");
if !skip_sanity {
//sanity check, make sure all the transactions can execute sequentially
transactions.iter().for_each(|tx| {
let res = bank.process_transaction(&tx);
assert!(res.is_ok(), "sanity test transactions error: {:?}", res);
});
bank.clear_signatures();
//sanity check, make sure all the transactions can execute in parallel
let res = bank.process_transactions(&transactions);
for r in res {
assert!(r.is_ok(), "sanity parallel execution error: {:?}", r);
}
bank.clear_signatures();
}
let mut verified: Vec<_> = to_packets_chunked(&transactions, packets_per_chunk);
let ledger_path = get_tmp_ledger_path!();
{
let blockstore = Arc::new(
Blockstore::open(&ledger_path).expect("Expected to be able to open database ledger"),
);
let (exit, poh_recorder, poh_service, signal_receiver) =
create_test_recorder(&bank, &blockstore, None);
let cluster_info = ClusterInfo::new_with_invalid_keypair(Node::new_localhost().info);
let cluster_info = Arc::new(cluster_info);
let banking_stage = BankingStage::new(
&cluster_info,
&poh_recorder,
verified_receiver,
vote_receiver,
None,
replay_vote_sender,
);
poh_recorder.lock().unwrap().set_bank(&bank);
let chunk_len = verified.len() / num_chunks;
let mut start = 0;
// This is so that the signal_receiver does not go out of scope after the closure.
// If it is dropped before poh_service, then poh_service will error when
// calling send() on the channel.
let signal_receiver = Arc::new(signal_receiver);
let mut total_us = 0;
let mut tx_total_us = 0;
let base_tx_count = bank.transaction_count();
let mut txs_processed = 0;
let mut root = 1;
let collector = solana_sdk::pubkey::new_rand();
let config = Config {
packets_per_batch: packets_per_chunk,
chunk_len,
num_threads,
};
let mut total_sent = 0;
for _ in 0..iterations {
let now = Instant::now();
let mut sent = 0;
for (i, v) in verified[start..start + chunk_len]
.chunks(chunk_len / num_threads)
.enumerate()
{
let mut byte = 0;
let index = config.get_transactions_index(start + i);
if index < transactions.len() {
byte = bytes_as_usize(transactions[index].signatures[0].as_ref());
}
trace!(
"sending... {}..{} {} v.len: {} sig: {} transactions.len: {} index: {}",
start + i,
start + chunk_len,
timestamp(),
v.len(),
byte,
transactions.len(),
index,
);
for xv in v {
sent += xv.packets.len();
}
verified_sender.send(v.to_vec()).unwrap();
}
let start_tx_index = config.get_transactions_index(start);
let end_tx_index = config.get_transactions_index(start + chunk_len);
for tx in &transactions[start_tx_index..end_tx_index] {
loop {
if bank.get_signature_status(&tx.signatures[0]).is_some() {
break;
}
if poh_recorder.lock().unwrap().bank().is_none() {
break;
}
sleep(Duration::from_millis(5));
}
}
if check_txs(
&signal_receiver,
total_num_transactions / num_chunks,
&poh_recorder,
) {
debug!(
"resetting bank {} tx count: {} txs_proc: {}",
bank.slot(),
bank.transaction_count(),
txs_processed
);
assert!(txs_processed < bank.transaction_count());
txs_processed = bank.transaction_count();
tx_total_us += duration_as_us(&now.elapsed());
let mut poh_time = Measure::start("poh_time");
poh_recorder.lock().unwrap().reset(
bank.last_blockhash(),
bank.slot(),
Some((bank.slot(), bank.slot() + 1)),
);
poh_time.stop();
let mut new_bank_time = Measure::start("new_bank");
let new_bank = Bank::new_from_parent(&bank, &collector, bank.slot() + 1);
new_bank_time.stop();
let mut insert_time = Measure::start("insert_time");
bank_forks.insert(new_bank);
bank = bank_forks.working_bank();
insert_time.stop();
poh_recorder.lock().unwrap().set_bank(&bank);
assert!(poh_recorder.lock().unwrap().bank().is_some());
if bank.slot() > 32 {
bank_forks.set_root(root, &ABSRequestSender::default(), None);
root += 1;
}
debug!(
"new_bank_time: {}us insert_time: {}us poh_time: {}us",
new_bank_time.as_us(),
insert_time.as_us(),
poh_time.as_us(),
);
} else {
tx_total_us += duration_as_us(&now.elapsed());
}
// This signature clear may not actually clear the signatures
// in this chunk, but since we rotate between CHUNKS then
// we should clear them by the time we come around again to re-use that chunk.
bank.clear_signatures();
total_us += duration_as_us(&now.elapsed());
debug!(
"time: {} us checked: {} sent: {}",
duration_as_us(&now.elapsed()),
total_num_transactions / num_chunks,
sent,
);
total_sent += sent;
if bank.slot() > 0 && bank.slot() % 16 == 0 {
for tx in transactions.iter_mut() {
tx.message.recent_blockhash = bank.last_blockhash();
let sig: Vec<u8> = (0..64).map(|_| thread_rng().gen()).collect();
tx.signatures[0] = Signature::new(&sig[0..64]);
}
verified = to_packets_chunked(&transactions.clone(), packets_per_chunk);
}
start += chunk_len;
start %= verified.len();
}
let txs_processed = bank_forks.working_bank().transaction_count();
debug!("processed: {} base: {}", txs_processed, base_tx_count);
eprintln!(
"{{'name': 'banking_bench_total', 'median': '{:.2}'}}",
(1000.0 * 1000.0 * total_sent as f64) / (total_us as f64),
);
eprintln!(
"{{'name': 'banking_bench_tx_total', 'median': '{:.2}'}}",
(1000.0 * 1000.0 * total_sent as f64) / (tx_total_us as f64),
);
eprintln!(
"{{'name': 'banking_bench_success_tx_total', 'median': '{:.2}'}}",
(1000.0 * 1000.0 * (txs_processed - base_tx_count) as f64) / (total_us as f64),
);
drop(verified_sender);
drop(vote_sender);
exit.store(true, Ordering::Relaxed);
banking_stage.join().unwrap();
debug!("waited for banking_stage");
poh_service.join().unwrap();
sleep(Duration::from_secs(1));
debug!("waited for poh_service");
}
let _unused = Blockstore::destroy(&ledger_path);
}

View File

@ -1,30 +0,0 @@
[package]
name = "solana-banks-client"
version = "1.5.4"
description = "Solana banks client"
authors = ["Solana Maintainers <maintainers@solana.foundation>"]
repository = "https://github.com/solana-labs/solana"
license = "Apache-2.0"
homepage = "https://solana.com/"
edition = "2018"
[dependencies]
bincode = "1.3.1"
futures = "0.3"
mio = "0.7.6"
solana-banks-interface = { path = "../banks-interface", version = "1.5.4" }
solana-sdk = { path = "../sdk", version = "1.5.4" }
tarpc = { version = "0.23.0", features = ["full"] }
tokio = { version = "0.3.5", features = ["full"] }
tokio-serde = { version = "0.6", features = ["bincode"] }
[dev-dependencies]
solana-runtime = { path = "../runtime", version = "1.5.4" }
solana-banks-server = { path = "../banks-server", version = "1.5.4" }
[lib]
crate-type = ["lib"]
name = "solana_banks_client"
[package.metadata.docs.rs]
targets = ["x86_64-unknown-linux-gnu"]

View File

@ -1,371 +0,0 @@
//! A client for the ledger state, from the perspective of an arbitrary validator.
//!
//! Use start_tcp_client() to create a client and then import BanksClientExt to
//! access its methods. Additional "*_with_context" methods are also available,
//! but they are undocumented, may change over time, and are generally more
//! cumbersome to use.
use futures::{future::join_all, Future, FutureExt};
pub use solana_banks_interface::{BanksClient as TarpcClient, TransactionStatus};
use solana_banks_interface::{BanksRequest, BanksResponse};
use solana_sdk::{
account::{from_account, Account},
clock::Slot,
commitment_config::CommitmentLevel,
fee_calculator::FeeCalculator,
hash::Hash,
pubkey::Pubkey,
rent::Rent,
signature::Signature,
sysvar,
transaction::{self, Transaction},
transport,
};
use std::io::{self, Error, ErrorKind};
use tarpc::{
client::{self, channel::RequestDispatch, NewClient},
context::{self, Context},
rpc::{ClientMessage, Response},
serde_transport::tcp,
Transport,
};
use tokio::{net::ToSocketAddrs, time::Duration};
use tokio_serde::formats::Bincode;
// This exists only for backward compatibility
pub trait BanksClientExt {}
#[derive(Clone)]
pub struct BanksClient {
inner: TarpcClient,
}
impl BanksClient {
#[allow(clippy::new_ret_no_self)]
pub fn new<C>(
config: client::Config,
transport: C,
) -> NewClient<TarpcClient, RequestDispatch<BanksRequest, BanksResponse, C>>
where
C: Transport<ClientMessage<BanksRequest>, Response<BanksResponse>>,
{
TarpcClient::new(config, transport)
}
pub fn send_transaction_with_context(
&mut self,
ctx: Context,
transaction: Transaction,
) -> impl Future<Output = io::Result<()>> + '_ {
self.inner.send_transaction_with_context(ctx, transaction)
}
pub fn get_fees_with_commitment_and_context(
&mut self,
ctx: Context,
commitment: CommitmentLevel,
) -> impl Future<Output = io::Result<(FeeCalculator, Hash, Slot)>> + '_ {
self.inner
.get_fees_with_commitment_and_context(ctx, commitment)
}
pub fn get_transaction_status_with_context(
&mut self,
ctx: Context,
signature: Signature,
) -> impl Future<Output = io::Result<Option<TransactionStatus>>> + '_ {
self.inner
.get_transaction_status_with_context(ctx, signature)
}
pub fn get_slot_with_context(
&mut self,
ctx: Context,
commitment: CommitmentLevel,
) -> impl Future<Output = io::Result<Slot>> + '_ {
self.inner.get_slot_with_context(ctx, commitment)
}
pub fn process_transaction_with_commitment_and_context(
&mut self,
ctx: Context,
transaction: Transaction,
commitment: CommitmentLevel,
) -> impl Future<Output = io::Result<Option<transaction::Result<()>>>> + '_ {
self.inner
.process_transaction_with_commitment_and_context(ctx, transaction, commitment)
}
pub fn get_account_with_commitment_and_context(
&mut self,
ctx: Context,
address: Pubkey,
commitment: CommitmentLevel,
) -> impl Future<Output = io::Result<Option<Account>>> + '_ {
self.inner
.get_account_with_commitment_and_context(ctx, address, commitment)
}
/// Send a transaction and return immediately. The server will resend the
/// transaction until either it is accepted by the cluster or the transaction's
/// blockhash expires.
pub fn send_transaction(
&mut self,
transaction: Transaction,
) -> impl Future<Output = io::Result<()>> + '_ {
self.send_transaction_with_context(context::current(), transaction)
}
/// Return the fee parameters associated with a recent, rooted blockhash. The cluster
/// will use the transaction's blockhash to look up these same fee parameters and
/// use them to calculate the transaction fee.
pub fn get_fees(
&mut self,
) -> impl Future<Output = io::Result<(FeeCalculator, Hash, Slot)>> + '_ {
self.get_fees_with_commitment_and_context(context::current(), CommitmentLevel::Root)
}
/// Return the cluster rent
pub fn get_rent(&mut self) -> impl Future<Output = io::Result<Rent>> + '_ {
self.get_account(sysvar::rent::id()).map(|result| {
let rent_sysvar = result?
.ok_or_else(|| io::Error::new(io::ErrorKind::Other, "Rent sysvar not present"))?;
from_account::<Rent>(&rent_sysvar).ok_or_else(|| {
io::Error::new(io::ErrorKind::Other, "Failed to deserialize Rent sysvar")
})
})
}
/// Return a recent, rooted blockhash from the server. The cluster will only accept
/// transactions with a blockhash that has not yet expired. Use the `get_fees`
/// method to get both a blockhash and the blockhash's last valid slot.
pub fn get_recent_blockhash(&mut self) -> impl Future<Output = io::Result<Hash>> + '_ {
self.get_fees().map(|result| Ok(result?.1))
}
/// Send a transaction and return after the transaction has been rejected or
/// reached the given level of commitment.
pub fn process_transaction_with_commitment(
&mut self,
transaction: Transaction,
commitment: CommitmentLevel,
) -> impl Future<Output = transport::Result<()>> + '_ {
let mut ctx = context::current();
ctx.deadline += Duration::from_secs(50);
self.process_transaction_with_commitment_and_context(ctx, transaction, commitment)
.map(|result| match result? {
None => {
Err(Error::new(ErrorKind::TimedOut, "invalid blockhash or fee-payer").into())
}
Some(transaction_result) => Ok(transaction_result?),
})
}
/// Send a transaction and return until the transaction has been finalized or rejected.
pub fn process_transaction(
&mut self,
transaction: Transaction,
) -> impl Future<Output = transport::Result<()>> + '_ {
self.process_transaction_with_commitment(transaction, CommitmentLevel::default())
}
pub async fn process_transactions_with_commitment(
&mut self,
transactions: Vec<Transaction>,
commitment: CommitmentLevel,
) -> transport::Result<()> {
let mut clients: Vec<_> = transactions.iter().map(|_| self.clone()).collect();
let futures = clients
.iter_mut()
.zip(transactions)
.map(|(client, transaction)| {
client.process_transaction_with_commitment(transaction, commitment)
});
let statuses = join_all(futures).await;
statuses.into_iter().collect() // Convert Vec<Result<_, _>> to Result<Vec<_>>
}
/// Send transactions and return until the transaction has been finalized or rejected.
pub fn process_transactions(
&mut self,
transactions: Vec<Transaction>,
) -> impl Future<Output = transport::Result<()>> + '_ {
self.process_transactions_with_commitment(transactions, CommitmentLevel::default())
}
/// Return the most recent rooted slot height. All transactions at or below this height
/// are said to be finalized. The cluster will not fork to a higher slot height.
pub fn get_root_slot(&mut self) -> impl Future<Output = io::Result<Slot>> + '_ {
self.get_slot_with_context(context::current(), CommitmentLevel::Root)
}
/// Return the account at the given address at the slot corresponding to the given
/// commitment level. If the account is not found, None is returned.
pub fn get_account_with_commitment(
&mut self,
address: Pubkey,
commitment: CommitmentLevel,
) -> impl Future<Output = io::Result<Option<Account>>> + '_ {
self.get_account_with_commitment_and_context(context::current(), address, commitment)
}
/// Return the account at the given address at the time of the most recent root slot.
/// If the account is not found, None is returned.
pub fn get_account(
&mut self,
address: Pubkey,
) -> impl Future<Output = io::Result<Option<Account>>> + '_ {
self.get_account_with_commitment(address, CommitmentLevel::default())
}
/// Return the balance in lamports of an account at the given address at the slot
/// corresponding to the given commitment level.
pub fn get_balance_with_commitment(
&mut self,
address: Pubkey,
commitment: CommitmentLevel,
) -> impl Future<Output = io::Result<u64>> + '_ {
self.get_account_with_commitment_and_context(context::current(), address, commitment)
.map(|result| Ok(result?.map(|x| x.lamports).unwrap_or(0)))
}
/// Return the balance in lamports of an account at the given address at the time
/// of the most recent root slot.
pub fn get_balance(&mut self, address: Pubkey) -> impl Future<Output = io::Result<u64>> + '_ {
self.get_balance_with_commitment(address, CommitmentLevel::default())
}
/// Return the status of a transaction with a signature matching the transaction's first
/// signature. Return None if the transaction is not found, which may be because the
/// blockhash was expired or the fee-paying account had insufficient funds to pay the
/// transaction fee. Note that servers rarely store the full transaction history. This
/// method may return None if the transaction status has been discarded.
pub fn get_transaction_status(
&mut self,
signature: Signature,
) -> impl Future<Output = io::Result<Option<TransactionStatus>>> + '_ {
self.get_transaction_status_with_context(context::current(), signature)
}
/// Same as get_transaction_status, but for multiple transactions.
pub async fn get_transaction_statuses(
&mut self,
signatures: Vec<Signature>,
) -> io::Result<Vec<Option<TransactionStatus>>> {
// tarpc futures oddly hold a mutable reference back to the client so clone the client upfront
let mut clients_and_signatures: Vec<_> = signatures
.into_iter()
.map(|signature| (self.clone(), signature))
.collect();
let futs = clients_and_signatures
.iter_mut()
.map(|(client, signature)| client.get_transaction_status(*signature));
let statuses = join_all(futs).await;
// Convert Vec<Result<_, _>> to Result<Vec<_>>
statuses.into_iter().collect()
}
}
pub async fn start_client<C>(transport: C) -> io::Result<BanksClient>
where
C: Transport<ClientMessage<BanksRequest>, Response<BanksResponse>> + Send + 'static,
{
Ok(BanksClient {
inner: TarpcClient::new(client::Config::default(), transport).spawn()?,
})
}
pub async fn start_tcp_client<T: ToSocketAddrs>(addr: T) -> io::Result<BanksClient> {
let transport = tcp::connect(addr, Bincode::default).await?;
Ok(BanksClient {
inner: TarpcClient::new(client::Config::default(), transport).spawn()?,
})
}
#[cfg(test)]
mod tests {
use super::*;
use solana_banks_server::banks_server::start_local_server;
use solana_runtime::{bank::Bank, bank_forks::BankForks, genesis_utils::create_genesis_config};
use solana_sdk::{message::Message, signature::Signer, system_instruction};
use std::sync::{Arc, RwLock};
use tarpc::transport;
use tokio::{runtime::Runtime, time::sleep};
#[test]
fn test_banks_client_new() {
let (client_transport, _server_transport) = transport::channel::unbounded();
BanksClient::new(client::Config::default(), client_transport);
}
#[test]
fn test_banks_server_transfer_via_server() -> io::Result<()> {
// This test shows the preferred way to interact with BanksServer.
// It creates a runtime explicitly (no globals via tokio macros) and calls
// `runtime.block_on()` just once, to run all the async code.
let genesis = create_genesis_config(10);
let bank_forks = Arc::new(RwLock::new(BankForks::new(Bank::new(
&genesis.genesis_config,
))));
let bob_pubkey = solana_sdk::pubkey::new_rand();
let mint_pubkey = genesis.mint_keypair.pubkey();
let instruction = system_instruction::transfer(&mint_pubkey, &bob_pubkey, 1);
let message = Message::new(&[instruction], Some(&mint_pubkey));
Runtime::new()?.block_on(async {
let client_transport = start_local_server(&bank_forks).await;
let mut banks_client = start_client(client_transport).await?;
let recent_blockhash = banks_client.get_recent_blockhash().await?;
let transaction = Transaction::new(&[&genesis.mint_keypair], message, recent_blockhash);
banks_client.process_transaction(transaction).await.unwrap();
assert_eq!(banks_client.get_balance(bob_pubkey).await?, 1);
Ok(())
})
}
#[test]
fn test_banks_server_transfer_via_client() -> io::Result<()> {
// The caller may not want to hold the connection open until the transaction
// is processed (or blockhash expires). In this test, we verify the
// server-side functionality is available to the client.
let genesis = create_genesis_config(10);
let bank_forks = Arc::new(RwLock::new(BankForks::new(Bank::new(
&genesis.genesis_config,
))));
let mint_pubkey = &genesis.mint_keypair.pubkey();
let bob_pubkey = solana_sdk::pubkey::new_rand();
let instruction = system_instruction::transfer(&mint_pubkey, &bob_pubkey, 1);
let message = Message::new(&[instruction], Some(&mint_pubkey));
Runtime::new()?.block_on(async {
let client_transport = start_local_server(&bank_forks).await;
let mut banks_client = start_client(client_transport).await?;
let (_, recent_blockhash, last_valid_slot) = banks_client.get_fees().await?;
let transaction = Transaction::new(&[&genesis.mint_keypair], message, recent_blockhash);
let signature = transaction.signatures[0];
banks_client.send_transaction(transaction).await?;
let mut status = banks_client.get_transaction_status(signature).await?;
while status.is_none() {
let root_slot = banks_client.get_root_slot().await?;
if root_slot > last_valid_slot {
break;
}
sleep(Duration::from_millis(100)).await;
status = banks_client.get_transaction_status(signature).await?;
}
assert!(status.unwrap().err.is_none());
assert_eq!(banks_client.get_balance(bob_pubkey).await?, 1);
Ok(())
})
}
}

View File

@ -1,25 +0,0 @@
[package]
name = "solana-banks-interface"
version = "1.5.4"
description = "Solana banks RPC interface"
authors = ["Solana Maintainers <maintainers@solana.foundation>"]
repository = "https://github.com/solana-labs/solana"
license = "Apache-2.0"
homepage = "https://solana.com/"
edition = "2018"
[dependencies]
mio = "0.7.6"
serde = { version = "1.0.112", features = ["derive"] }
solana-sdk = { path = "../sdk", version = "1.5.4" }
tarpc = { version = "0.23.0", features = ["full"] }
[dev-dependencies]
tokio = { version = "0.3.5", features = ["full"] }
[lib]
crate-type = ["lib"]
name = "solana_banks_interface"
[package.metadata.docs.rs]
targets = ["x86_64-unknown-linux-gnu"]

View File

@ -1,57 +0,0 @@
use serde::{Deserialize, Serialize};
use solana_sdk::{
account::Account,
clock::Slot,
commitment_config::CommitmentLevel,
fee_calculator::FeeCalculator,
hash::Hash,
pubkey::Pubkey,
signature::Signature,
transaction::{self, Transaction, TransactionError},
};
#[derive(Clone, Debug, PartialEq, Serialize, Deserialize)]
pub enum TransactionConfirmationStatus {
Processed,
Confirmed,
Finalized,
}
#[derive(Clone, Debug, PartialEq, Serialize, Deserialize)]
pub struct TransactionStatus {
pub slot: Slot,
pub confirmations: Option<usize>, // None = rooted
pub err: Option<TransactionError>,
pub confirmation_status: Option<TransactionConfirmationStatus>,
}
#[tarpc::service]
pub trait Banks {
async fn send_transaction_with_context(transaction: Transaction);
async fn get_fees_with_commitment_and_context(
commitment: CommitmentLevel,
) -> (FeeCalculator, Hash, Slot);
async fn get_transaction_status_with_context(signature: Signature)
-> Option<TransactionStatus>;
async fn get_slot_with_context(commitment: CommitmentLevel) -> Slot;
async fn process_transaction_with_commitment_and_context(
transaction: Transaction,
commitment: CommitmentLevel,
) -> Option<transaction::Result<()>>;
async fn get_account_with_commitment_and_context(
address: Pubkey,
commitment: CommitmentLevel,
) -> Option<Account>;
}
#[cfg(test)]
mod tests {
use super::*;
use tarpc::{client, transport};
#[test]
fn test_banks_client_new() {
let (client_transport, _server_transport) = transport::channel::unbounded();
BanksClient::new(client::Config::default(), client_transport);
}
}

View File

@ -1,29 +0,0 @@
[package]
name = "solana-banks-server"
version = "1.5.4"
description = "Solana banks server"
authors = ["Solana Maintainers <maintainers@solana.foundation>"]
repository = "https://github.com/solana-labs/solana"
license = "Apache-2.0"
homepage = "https://solana.com/"
edition = "2018"
[dependencies]
bincode = "1.3.1"
futures = "0.3"
log = "0.4.11"
mio = "0.7.6"
solana-banks-interface = { path = "../banks-interface", version = "1.5.4" }
solana-runtime = { path = "../runtime", version = "1.5.4" }
solana-sdk = { path = "../sdk", version = "1.5.4" }
solana-metrics = { path = "../metrics", version = "1.5.4" }
tarpc = { version = "0.23.0", features = ["full"] }
tokio = { version = "0.3", features = ["full"] }
tokio-serde = { version = "0.6", features = ["bincode"] }
[lib]
crate-type = ["lib"]
name = "solana_banks_server"
[package.metadata.docs.rs]
targets = ["x86_64-unknown-linux-gnu"]

View File

@ -1,290 +0,0 @@
use crate::send_transaction_service::{SendTransactionService, TransactionInfo};
use bincode::{deserialize, serialize};
use futures::{
future,
prelude::stream::{self, StreamExt},
};
use solana_banks_interface::{
Banks, BanksRequest, BanksResponse, TransactionConfirmationStatus, TransactionStatus,
};
use solana_runtime::{bank::Bank, bank_forks::BankForks, commitment::BlockCommitmentCache};
use solana_sdk::{
account::Account,
clock::Slot,
commitment_config::CommitmentLevel,
fee_calculator::FeeCalculator,
hash::Hash,
pubkey::Pubkey,
signature::Signature,
transaction::{self, Transaction},
};
use std::{
io,
net::{Ipv4Addr, SocketAddr},
sync::{
mpsc::{channel, Receiver, Sender},
Arc, RwLock,
},
thread::Builder,
time::Duration,
};
use tarpc::{
context::Context,
rpc::{transport::channel::UnboundedChannel, ClientMessage, Response},
serde_transport::tcp,
server::{self, Channel, Handler},
transport,
};
use tokio::time::sleep;
use tokio_serde::formats::Bincode;
#[derive(Clone)]
struct BanksServer {
bank_forks: Arc<RwLock<BankForks>>,
block_commitment_cache: Arc<RwLock<BlockCommitmentCache>>,
transaction_sender: Sender<TransactionInfo>,
}
impl BanksServer {
/// Return a BanksServer that forwards transactions to the
/// given sender. If unit-testing, those transactions can go to
/// a bank in the given BankForks. Otherwise, the receiver should
/// forward them to a validator in the leader schedule.
fn new(
bank_forks: Arc<RwLock<BankForks>>,
block_commitment_cache: Arc<RwLock<BlockCommitmentCache>>,
transaction_sender: Sender<TransactionInfo>,
) -> Self {
Self {
bank_forks,
block_commitment_cache,
transaction_sender,
}
}
fn run(bank: &Bank, transaction_receiver: Receiver<TransactionInfo>) {
while let Ok(info) = transaction_receiver.recv() {
let mut transaction_infos = vec![info];
while let Ok(info) = transaction_receiver.try_recv() {
transaction_infos.push(info);
}
let transactions: Vec<_> = transaction_infos
.into_iter()
.map(|info| deserialize(&info.wire_transaction).unwrap())
.collect();
let _ = bank.process_transactions(&transactions);
}
}
/// Useful for unit-testing
fn new_loopback(bank_forks: Arc<RwLock<BankForks>>) -> Self {
let (transaction_sender, transaction_receiver) = channel();
let bank = bank_forks.read().unwrap().working_bank();
let slot = bank.slot();
let block_commitment_cache = Arc::new(RwLock::new(
BlockCommitmentCache::new_for_tests_with_slots(slot, slot),
));
Builder::new()
.name("solana-bank-forks-client".to_string())
.spawn(move || Self::run(&bank, transaction_receiver))
.unwrap();
Self::new(bank_forks, block_commitment_cache, transaction_sender)
}
fn slot(&self, commitment: CommitmentLevel) -> Slot {
self.block_commitment_cache
.read()
.unwrap()
.slot_with_commitment(commitment)
}
fn bank(&self, commitment: CommitmentLevel) -> Arc<Bank> {
self.bank_forks.read().unwrap()[self.slot(commitment)].clone()
}
async fn poll_signature_status(
self,
signature: &Signature,
blockhash: &Hash,
last_valid_slot: Slot,
commitment: CommitmentLevel,
) -> Option<transaction::Result<()>> {
let mut status = self
.bank(commitment)
.get_signature_status_with_blockhash(signature, blockhash);
while status.is_none() {
sleep(Duration::from_millis(200)).await;
let bank = self.bank(commitment);
if bank.slot() > last_valid_slot {
break;
}
status = bank.get_signature_status_with_blockhash(signature, blockhash);
}
status
}
}
fn verify_transaction(transaction: &Transaction) -> transaction::Result<()> {
if let Err(err) = transaction.verify() {
Err(err)
} else if let Err(err) = transaction.verify_precompiles() {
Err(err)
} else {
Ok(())
}
}
#[tarpc::server]
impl Banks for BanksServer {
async fn send_transaction_with_context(self, _: Context, transaction: Transaction) {
let blockhash = &transaction.message.recent_blockhash;
let last_valid_slot = self
.bank_forks
.read()
.unwrap()
.root_bank()
.get_blockhash_last_valid_slot(&blockhash)
.unwrap();
let signature = transaction.signatures.get(0).cloned().unwrap_or_default();
let info =
TransactionInfo::new(signature, serialize(&transaction).unwrap(), last_valid_slot);
self.transaction_sender.send(info).unwrap();
}
async fn get_fees_with_commitment_and_context(
self,
_: Context,
commitment: CommitmentLevel,
) -> (FeeCalculator, Hash, Slot) {
let bank = self.bank(commitment);
let (blockhash, fee_calculator) = bank.last_blockhash_with_fee_calculator();
let last_valid_slot = bank.get_blockhash_last_valid_slot(&blockhash).unwrap();
(fee_calculator, blockhash, last_valid_slot)
}
async fn get_transaction_status_with_context(
self,
_: Context,
signature: Signature,
) -> Option<TransactionStatus> {
let bank = self.bank(CommitmentLevel::Recent);
let (slot, status) = bank.get_signature_status_slot(&signature)?;
let r_block_commitment_cache = self.block_commitment_cache.read().unwrap();
let optimistically_confirmed_bank = self.bank(CommitmentLevel::SingleGossip);
let optimistically_confirmed =
optimistically_confirmed_bank.get_signature_status_slot(&signature);
let confirmations = if r_block_commitment_cache.root() >= slot
&& r_block_commitment_cache.highest_confirmed_root() >= slot
{
None
} else {
r_block_commitment_cache
.get_confirmation_count(slot)
.or(Some(0))
};
Some(TransactionStatus {
slot,
confirmations,
err: status.err(),
confirmation_status: if confirmations.is_none() {
Some(TransactionConfirmationStatus::Finalized)
} else if optimistically_confirmed.is_some() {
Some(TransactionConfirmationStatus::Confirmed)
} else {
Some(TransactionConfirmationStatus::Processed)
},
})
}
async fn get_slot_with_context(self, _: Context, commitment: CommitmentLevel) -> Slot {
self.slot(commitment)
}
async fn process_transaction_with_commitment_and_context(
self,
_: Context,
transaction: Transaction,
commitment: CommitmentLevel,
) -> Option<transaction::Result<()>> {
if let Err(err) = verify_transaction(&transaction) {
return Some(Err(err));
}
let blockhash = &transaction.message.recent_blockhash;
let last_valid_slot = self
.bank_forks
.read()
.unwrap()
.root_bank()
.get_blockhash_last_valid_slot(blockhash)
.unwrap();
let signature = transaction.signatures.get(0).cloned().unwrap_or_default();
let info =
TransactionInfo::new(signature, serialize(&transaction).unwrap(), last_valid_slot);
self.transaction_sender.send(info).unwrap();
self.poll_signature_status(&signature, blockhash, last_valid_slot, commitment)
.await
}
async fn get_account_with_commitment_and_context(
self,
_: Context,
address: Pubkey,
commitment: CommitmentLevel,
) -> Option<Account> {
let bank = self.bank(commitment);
bank.get_account(&address)
}
}
pub async fn start_local_server(
bank_forks: &Arc<RwLock<BankForks>>,
) -> UnboundedChannel<Response<BanksResponse>, ClientMessage<BanksRequest>> {
let banks_server = BanksServer::new_loopback(bank_forks.clone());
let (client_transport, server_transport) = transport::channel::unbounded();
let server = server::new(server::Config::default())
.incoming(stream::once(future::ready(server_transport)))
.respond_with(banks_server.serve());
tokio::spawn(server);
client_transport
}
pub async fn start_tcp_server(
listen_addr: SocketAddr,
tpu_addr: SocketAddr,
bank_forks: Arc<RwLock<BankForks>>,
block_commitment_cache: Arc<RwLock<BlockCommitmentCache>>,
) -> io::Result<()> {
// Note: These settings are copied straight from the tarpc example.
let server = tcp::listen(listen_addr, Bincode::default)
.await?
// Ignore accept errors.
.filter_map(|r| future::ready(r.ok()))
.map(server::BaseChannel::with_defaults)
// Limit channels to 1 per IP.
.max_channels_per_key(1, |t| {
t.as_ref()
.peer_addr()
.map(|x| x.ip())
.unwrap_or_else(|_| Ipv4Addr::new(0, 0, 0, 0).into())
})
// serve is generated by the service attribute. It takes as input any type implementing
// the generated Banks trait.
.map(move |chan| {
let (sender, receiver) = channel();
SendTransactionService::new(tpu_addr, &bank_forks, receiver);
let server =
BanksServer::new(bank_forks.clone(), block_commitment_cache.clone(), sender);
chan.respond_with(server.serve()).execute()
})
// Max 10 channels.
.buffer_unordered(10)
.for_each(|_| async {});
server.await;
Ok(())
}

View File

@ -1,6 +0,0 @@
pub mod banks_server;
pub mod rpc_banks_service;
pub mod send_transaction_service;
#[macro_use]
extern crate solana_metrics;

View File

@ -1,116 +0,0 @@
//! The `rpc_banks_service` module implements the Solana Banks RPC API.
use crate::banks_server::start_tcp_server;
use futures::{future::FutureExt, pin_mut, prelude::stream::StreamExt, select};
use solana_runtime::{bank_forks::BankForks, commitment::BlockCommitmentCache};
use std::{
net::SocketAddr,
sync::{
atomic::{AtomicBool, Ordering},
Arc, RwLock,
},
thread::{self, Builder, JoinHandle},
};
use tokio::{
runtime::Runtime,
time::{self, Duration},
};
pub struct RpcBanksService {
thread_hdl: JoinHandle<()>,
}
/// Run the TCP service until `exit` is set to true
async fn start_abortable_tcp_server(
listen_addr: SocketAddr,
tpu_addr: SocketAddr,
bank_forks: Arc<RwLock<BankForks>>,
block_commitment_cache: Arc<RwLock<BlockCommitmentCache>>,
exit: Arc<AtomicBool>,
) {
let server = start_tcp_server(
listen_addr,
tpu_addr,
bank_forks.clone(),
block_commitment_cache.clone(),
)
.fuse();
let interval = time::interval(Duration::from_millis(100)).fuse();
pin_mut!(server, interval);
loop {
select! {
_ = server => {},
_ = interval.select_next_some() => {
if exit.load(Ordering::Relaxed) {
break;
}
}
}
}
}
impl RpcBanksService {
fn run(
listen_addr: SocketAddr,
tpu_addr: SocketAddr,
bank_forks: Arc<RwLock<BankForks>>,
block_commitment_cache: Arc<RwLock<BlockCommitmentCache>>,
exit: Arc<AtomicBool>,
) {
let server = start_abortable_tcp_server(
listen_addr,
tpu_addr,
bank_forks,
block_commitment_cache,
exit,
);
Runtime::new().unwrap().block_on(server);
}
pub fn new(
listen_addr: SocketAddr,
tpu_addr: SocketAddr,
bank_forks: &Arc<RwLock<BankForks>>,
block_commitment_cache: &Arc<RwLock<BlockCommitmentCache>>,
exit: &Arc<AtomicBool>,
) -> Self {
let bank_forks = bank_forks.clone();
let block_commitment_cache = block_commitment_cache.clone();
let exit = exit.clone();
let thread_hdl = Builder::new()
.name("solana-rpc-banks".to_string())
.spawn(move || {
Self::run(
listen_addr,
tpu_addr,
bank_forks,
block_commitment_cache,
exit,
)
})
.unwrap();
Self { thread_hdl }
}
pub fn join(self) -> thread::Result<()> {
self.thread_hdl.join()
}
}
#[cfg(test)]
mod tests {
use super::*;
use solana_runtime::bank::Bank;
#[test]
fn test_rpc_banks_server_exit() {
let bank_forks = Arc::new(RwLock::new(BankForks::new(Bank::default())));
let block_commitment_cache = Arc::new(RwLock::new(BlockCommitmentCache::default()));
let exit = Arc::new(AtomicBool::new(false));
let addr = "127.0.0.1:0".parse().unwrap();
let service = RpcBanksService::new(addr, addr, &bank_forks, &block_commitment_cache, &exit);
exit.store(true, Ordering::Relaxed);
service.join().unwrap();
}
}

View File

@ -1,343 +0,0 @@
// TODO: Merge this implementation with the one at `core/src/send_transaction_service.rs`
use log::*;
use solana_metrics::{datapoint_warn, inc_new_counter_info};
use solana_runtime::{bank::Bank, bank_forks::BankForks};
use solana_sdk::{clock::Slot, signature::Signature};
use std::{
collections::HashMap,
net::{SocketAddr, UdpSocket},
sync::{
mpsc::{Receiver, RecvTimeoutError},
Arc, RwLock,
},
thread::{self, Builder, JoinHandle},
time::{Duration, Instant},
};
/// Maximum size of the transaction queue
const MAX_TRANSACTION_QUEUE_SIZE: usize = 10_000; // This seems like a lot but maybe it needs to be bigger one day
pub struct SendTransactionService {
thread: JoinHandle<()>,
}
pub struct TransactionInfo {
pub signature: Signature,
pub wire_transaction: Vec<u8>,
pub last_valid_slot: Slot,
}
impl TransactionInfo {
pub fn new(signature: Signature, wire_transaction: Vec<u8>, last_valid_slot: Slot) -> Self {
Self {
signature,
wire_transaction,
last_valid_slot,
}
}
}
#[derive(Default, Debug, PartialEq)]
struct ProcessTransactionsResult {
rooted: u64,
expired: u64,
retried: u64,
failed: u64,
retained: u64,
}
impl SendTransactionService {
pub fn new(
tpu_address: SocketAddr,
bank_forks: &Arc<RwLock<BankForks>>,
receiver: Receiver<TransactionInfo>,
) -> Self {
let thread = Self::retry_thread(receiver, bank_forks.clone(), tpu_address);
Self { thread }
}
fn retry_thread(
receiver: Receiver<TransactionInfo>,
bank_forks: Arc<RwLock<BankForks>>,
tpu_address: SocketAddr,
) -> JoinHandle<()> {
let mut last_status_check = Instant::now();
let mut transactions = HashMap::new();
let send_socket = UdpSocket::bind("0.0.0.0:0").unwrap();
Builder::new()
.name("send-tx-svc".to_string())
.spawn(move || loop {
match receiver.recv_timeout(Duration::from_secs(1)) {
Err(RecvTimeoutError::Disconnected) => break,
Err(RecvTimeoutError::Timeout) => {}
Ok(transaction_info) => {
Self::send_transaction(
&send_socket,
&tpu_address,
&transaction_info.wire_transaction,
);
if transactions.len() < MAX_TRANSACTION_QUEUE_SIZE {
transactions.insert(transaction_info.signature, transaction_info);
} else {
datapoint_warn!("send_transaction_service-queue-overflow");
}
}
}
if Instant::now().duration_since(last_status_check).as_secs() >= 5 {
if !transactions.is_empty() {
datapoint_info!(
"send_transaction_service-queue-size",
("len", transactions.len(), i64)
);
let bank_forks = bank_forks.read().unwrap();
let root_bank = bank_forks.root_bank();
let working_bank = bank_forks.working_bank();
let _result = Self::process_transactions(
&working_bank,
&root_bank,
&send_socket,
&tpu_address,
&mut transactions,
);
}
last_status_check = Instant::now();
}
})
.unwrap()
}
fn process_transactions(
working_bank: &Arc<Bank>,
root_bank: &Arc<Bank>,
send_socket: &UdpSocket,
tpu_address: &SocketAddr,
transactions: &mut HashMap<Signature, TransactionInfo>,
) -> ProcessTransactionsResult {
let mut result = ProcessTransactionsResult::default();
transactions.retain(|signature, transaction_info| {
if root_bank.has_signature(signature) {
info!("Transaction is rooted: {}", signature);
result.rooted += 1;
inc_new_counter_info!("send_transaction_service-rooted", 1);
false
} else if transaction_info.last_valid_slot < root_bank.slot() {
info!("Dropping expired transaction: {}", signature);
result.expired += 1;
inc_new_counter_info!("send_transaction_service-expired", 1);
false
} else {
match working_bank.get_signature_status_slot(signature) {
None => {
// Transaction is unknown to the working bank, it might have been
// dropped or landed in another fork. Re-send it
info!("Retrying transaction: {}", signature);
result.retried += 1;
inc_new_counter_info!("send_transaction_service-retry", 1);
Self::send_transaction(
&send_socket,
&tpu_address,
&transaction_info.wire_transaction,
);
true
}
Some((_slot, status)) => {
if status.is_err() {
info!("Dropping failed transaction: {}", signature);
result.failed += 1;
inc_new_counter_info!("send_transaction_service-failed", 1);
false
} else {
result.retained += 1;
true
}
}
}
}
});
result
}
fn send_transaction(
send_socket: &UdpSocket,
tpu_address: &SocketAddr,
wire_transaction: &[u8],
) {
if let Err(err) = send_socket.send_to(wire_transaction, tpu_address) {
warn!("Failed to send transaction to {}: {:?}", tpu_address, err);
}
}
pub fn join(self) -> thread::Result<()> {
self.thread.join()
}
}
#[cfg(test)]
mod test {
use super::*;
use solana_sdk::{
genesis_config::create_genesis_config, pubkey::Pubkey, signature::Signer,
system_transaction,
};
use std::sync::mpsc::channel;
#[test]
fn service_exit() {
let tpu_address = "127.0.0.1:0".parse().unwrap();
let bank = Bank::default();
let bank_forks = Arc::new(RwLock::new(BankForks::new(bank)));
let (sender, receiver) = channel();
let send_tranaction_service =
SendTransactionService::new(tpu_address, &bank_forks, receiver);
drop(sender);
send_tranaction_service.join().unwrap();
}
#[test]
fn process_transactions() {
let (genesis_config, mint_keypair) = create_genesis_config(4);
let bank = Bank::new(&genesis_config);
let bank_forks = Arc::new(RwLock::new(BankForks::new(bank)));
let send_socket = UdpSocket::bind("0.0.0.0:0").unwrap();
let tpu_address = "127.0.0.1:0".parse().unwrap();
let root_bank = Arc::new(Bank::new_from_parent(
&bank_forks.read().unwrap().working_bank(),
&Pubkey::default(),
1,
));
let rooted_signature = root_bank
.transfer(1, &mint_keypair, &mint_keypair.pubkey())
.unwrap();
let working_bank = Arc::new(Bank::new_from_parent(&root_bank, &Pubkey::default(), 2));
let non_rooted_signature = working_bank
.transfer(2, &mint_keypair, &mint_keypair.pubkey())
.unwrap();
let failed_signature = {
let blockhash = working_bank.last_blockhash();
let transaction =
system_transaction::transfer(&mint_keypair, &Pubkey::default(), 1, blockhash);
let signature = transaction.signatures[0];
working_bank.process_transaction(&transaction).unwrap_err();
signature
};
let mut transactions = HashMap::new();
info!("Expired transactions are dropped..");
transactions.insert(
Signature::default(),
TransactionInfo::new(Signature::default(), vec![], root_bank.slot() - 1),
);
let result = SendTransactionService::process_transactions(
&working_bank,
&root_bank,
&send_socket,
&tpu_address,
&mut transactions,
);
assert!(transactions.is_empty());
assert_eq!(
result,
ProcessTransactionsResult {
expired: 1,
..ProcessTransactionsResult::default()
}
);
info!("Rooted transactions are dropped...");
transactions.insert(
rooted_signature,
TransactionInfo::new(rooted_signature, vec![], working_bank.slot()),
);
let result = SendTransactionService::process_transactions(
&working_bank,
&root_bank,
&send_socket,
&tpu_address,
&mut transactions,
);
assert!(transactions.is_empty());
assert_eq!(
result,
ProcessTransactionsResult {
rooted: 1,
..ProcessTransactionsResult::default()
}
);
info!("Failed transactions are dropped...");
transactions.insert(
failed_signature,
TransactionInfo::new(failed_signature, vec![], working_bank.slot()),
);
let result = SendTransactionService::process_transactions(
&working_bank,
&root_bank,
&send_socket,
&tpu_address,
&mut transactions,
);
assert!(transactions.is_empty());
assert_eq!(
result,
ProcessTransactionsResult {
failed: 1,
..ProcessTransactionsResult::default()
}
);
info!("Non-rooted transactions are kept...");
transactions.insert(
non_rooted_signature,
TransactionInfo::new(non_rooted_signature, vec![], working_bank.slot()),
);
let result = SendTransactionService::process_transactions(
&working_bank,
&root_bank,
&send_socket,
&tpu_address,
&mut transactions,
);
assert_eq!(transactions.len(), 1);
assert_eq!(
result,
ProcessTransactionsResult {
retained: 1,
..ProcessTransactionsResult::default()
}
);
transactions.clear();
info!("Unknown transactions are retried...");
transactions.insert(
Signature::default(),
TransactionInfo::new(Signature::default(), vec![], working_bank.slot()),
);
let result = SendTransactionService::process_transactions(
&working_bank,
&root_bank,
&send_socket,
&tpu_address,
&mut transactions,
);
assert_eq!(transactions.len(), 1);
assert_eq!(
result,
ProcessTransactionsResult {
retried: 1,
..ProcessTransactionsResult::default()
}
);
}
}

View File

@ -1,4 +0,0 @@
/target/
/config/
/config-local/
/farf/

View File

@ -1,38 +0,0 @@
[package]
authors = ["Solana Maintainers <maintainers@solana.foundation>"]
edition = "2018"
name = "solana-bench-exchange"
version = "1.5.4"
repository = "https://github.com/solana-labs/solana"
license = "Apache-2.0"
homepage = "https://solana.com/"
publish = false
[dependencies]
clap = "2.33.1"
itertools = "0.9.0"
log = "0.4.11"
num-derive = "0.3"
num-traits = "0.2"
rand = "0.7.0"
rayon = "1.4.0"
serde_json = "1.0.56"
serde_yaml = "0.8.13"
solana-clap-utils = { path = "../clap-utils", version = "1.5.4" }
solana-core = { path = "../core", version = "1.5.4" }
solana-genesis = { path = "../genesis", version = "1.5.4" }
solana-client = { path = "../client", version = "1.5.4" }
solana-faucet = { path = "../faucet", version = "1.5.4" }
solana-exchange-program = { path = "../programs/exchange", version = "1.5.4" }
solana-logger = { path = "../logger", version = "1.5.4" }
solana-metrics = { path = "../metrics", version = "1.5.4" }
solana-net-utils = { path = "../net-utils", version = "1.5.4" }
solana-runtime = { path = "../runtime", version = "1.5.4" }
solana-sdk = { path = "../sdk", version = "1.5.4" }
solana-version = { path = "../version", version = "1.5.4" }
[dev-dependencies]
solana-local-cluster = { path = "../local-cluster", version = "1.5.4" }
[package.metadata.docs.rs]
targets = ["x86_64-unknown-linux-gnu"]

View File

@ -1,479 +0,0 @@
# token-exchange
Solana Token Exchange Bench
If you can't wait; jump to [Running the exchange](#Running-the-exchange) to
learn how to start and interact with the exchange.
### Table of Contents
[Overview](#Overview)<br>
[Premise](#Premise)<br>
[Exchange startup](#Exchange-startup)<br>
[Order Requests](#Trade-requests)<br>
[Order Cancellations](#Trade-cancellations)<br>
[Trade swap](#Trade-swap)<br>
[Exchange program operations](#Exchange-program-operations)<br>
[Quotes and OHLCV](#Quotes-and-OHLCV)<br>
[Investor strategies](#Investor-strategies)<br>
[Running the exchange](#Running-the-exchange)<br>
## Overview
An exchange is a marketplace where one asset can be traded for another. This
demo demonstrates one way to host an exchange on the Solana blockchain by
emulating a currency exchange.
The assets are virtual tokens held by investors who may post order requests to
the exchange. A Matcher monitors the exchange and posts swap requests for
matching orders. All the transactions can execute concurrently.
## Premise
- Exchange
- An exchange is a marketplace where one asset can be traded for another.
The exchange in this demo is the on-chain program that implements the
tokens and the policies for trading those tokens.
- Token
- A virtual asset that can be owned, traded, and holds virtual intrinsic value
compared to other assets. There are four types of tokens in this demo, A,
B, C, D. Each one may be traded for another.
- Token account
- An account owned by the exchange that holds a quantity of one type of token.
- Account request
- A request to create a token account
- Token request
- A request to deposit tokens of a particular type into a token account.
- Asset pair
- A struct with fields Base and Quote, representing the two assets which make up a
trading pair, which themselves are Tokens. The Base or 'primary' asset is the
numerator and the Quote is the denominator for pricing purposes.
- Order side
- Describes which side of the market an investor wants to place a trade on. Options
are "Bid" or "Ask", where a bid represents an offer to purchase the Base asset of
the AssetPair for a sum of the Quote Asset and an Ask is an offer to sell Base asset
for the Quote asset.
- Price ratio
- An expression of the relative prices of two tokens. Calculated with the Base
Asset as the numerator and the Quote Asset as the denominator. Ratios are
represented as fixed point numbers. The fixed point scaler is defined in
[exchange_state.rs](https://github.com/solana-labs/solana/blob/c2fdd1362a029dcf89c8907c562d2079d977df11/programs/exchange_api/src/exchange_state.rs#L7)
- Order request
- A Solana transaction sent by a trader to the exchange to submit an order.
Order requests are made up of the token pair, the order side (bid or ask),
quantity of the primary token, the price ratio, and the two token accounts
to be credited/deducted. An example trade request looks like "T AB 5 2"
which reads "Exchange 5 A tokens to B tokens at a price ratio of 1:2" A fulfilled trade would result in 5 A tokens
deducted and 10 B tokens credited to the trade initiator's token accounts.
Successful order requests result in an order.
- Order
- The result of a successful order request. orders are stored in
accounts owned by the submitter of the order request. They can only be
canceled by their owner but can be used by anyone in a trade swap. They
contain the same information as the order request.
- Price spread
- The difference between the two matching orders. The spread is the
profit of the Matcher initiating the swap request.
- Match requirements
- Policies that result in a successful trade swap.
- Match request
- A request to fill two complementary orders (bid/ask), resulting if successful,
in a trade being created.
- Trade
- A successful trade is created from two matching orders that meet
swap requirements which are submitted in a Match Request by a Matcher and
executed by the exchange. A trade may not wholly satisfy one or both of the
orders in which case the orders are adjusted appropriately. Upon execution,
tokens are distributed to the traders' accounts and any overlap or
"negative spread" between orders is deposited into the Matcher's profit
account. All successful trades are recorded in the data of a new solana
account for posterity.
- Investor
- Individual investors who hold a number of tokens and wish to trade them on
the exchange. Investors operate as Solana thin clients who own a set of
accounts containing tokens and/or order requests. Investors post
transactions to the exchange in order to request tokens and post or cancel
order requests.
- Matcher
- An agent who facilitates trading between investors. Matchers operate as
Solana thin clients who monitor all the orders looking for a trade
match. Once found, the Matcher issues a swap request to the exchange.
Matchers are the engine of the exchange and are rewarded for their efforts by
accumulating the price spreads of the swaps they initiate. Matchers also
provide current bid/ask price and OHLCV (Open, High, Low, Close, Volume)
information on demand via a public network port.
- Transaction fees
- Solana transaction fees are paid for by the transaction submitters who are
the Investors and Matchers.
## Exchange startup
The exchange is up and running when it reaches a state where it can take
investors' trades and Matchers' match requests. To achieve this state the
following must occur in order:
- Start the Solana blockchain
- Start the thin-client
- The Matcher subscribes to change notifications for all the accounts owned by
the exchange program id. The subscription is managed via Solana's JSON RPC
interface.
- The Matcher starts responding to queries for bid/ask price and OHLCV
The Matcher responding successfully to price and OHLCV requests is the signal to
the investors that trades submitted after that point will be analyzed. <!--This
is not ideal, and instead investors should be able to submit trades at any time,
and the Matcher could come and go without missing a trade. One way to achieve
this is for the Matcher to read the current state of all accounts looking for all
open orders.-->
Investors will initially query the exchange to discover their current balance
for each type of token. If the investor does not already have an account for
each type of token, they will submit account requests. Matcher as well will
request accounts to hold the tokens they earn by initiating trade swaps.
```rust
/// Supported token types
pub enum Token {
A,
B,
C,
D,
}
/// Supported token pairs
pub enum TokenPair {
AB,
AC,
AD,
BC,
BD,
CD,
}
pub enum ExchangeInstruction {
/// New token account
/// key 0 - Signer
/// key 1 - New token account
AccountRequest,
}
/// Token accounts are populated with this structure
pub struct TokenAccountInfo {
/// Investor who owns this account
pub owner: Pubkey,
/// Current number of tokens this account holds
pub tokens: Tokens,
}
```
For this demo investors or Matcher can request more tokens from the exchange at
any time by submitting token requests. In non-demos, an exchange of this type
would provide another way to exchange a 3rd party asset into tokens.
To request tokens, investors submit transfer requests:
```rust
pub enum ExchangeInstruction {
/// Transfer tokens between two accounts
/// key 0 - Account to transfer tokens to
/// key 1 - Account to transfer tokens from. This can be the exchange program itself,
/// the exchange has a limitless number of tokens it can transfer.
TransferRequest(Token, u64),
}
```
## Order Requests
When an investor decides to exchange a token of one type for another, they
submit a transaction to the Solana Blockchain containing an order request, which,
if successful, is turned into an order. orders do not expire but are
cancellable. <!-- orders should have a timestamp to enable trade
expiration --> When an order is created, tokens are deducted from a token
account and the order acts as an escrow. The tokens are held until the
order is fulfilled or canceled. If the direction is `To`, then the number
of `tokens` are deducted from the primary account, if `From` then `tokens`
multiplied by `price` are deducted from the secondary account. orders are
no longer valid when the number of `tokens` goes to zero, at which point they
can no longer be used. <!-- Could support refilling orders, so order
accounts are refilled rather than accumulating -->
```rust
/// Direction of the exchange between two tokens in a pair
pub enum Direction {
/// Trade first token type (primary) in the pair 'To' the second
To,
/// Trade first token type in the pair 'From' the second (secondary)
From,
}
pub struct OrderRequestInfo {
/// Direction of trade
pub direction: Direction,
/// Token pair to trade
pub pair: TokenPair,
/// Number of tokens to exchange; refers to the primary or the secondary depending on the direction
pub tokens: u64,
/// The price ratio the primary price over the secondary price. The primary price is fixed
/// and equal to the variable `SCALER`.
pub price: u64,
/// Token account to deposit tokens on successful swap
pub dst_account: Pubkey,
}
pub enum ExchangeInstruction {
/// order request
/// key 0 - Signer
/// key 1 - Account in which to record the swap
/// key 2 - Token account associated with this trade
TradeRequest(TradeRequestInfo),
}
/// Trade accounts are populated with this structure
pub struct TradeOrderInfo {
/// Owner of the order
pub owner: Pubkey,
/// Direction of the exchange
pub direction: Direction,
/// Token pair indicating two tokens to exchange, first is primary
pub pair: TokenPair,
/// Number of tokens to exchange; primary or secondary depending on direction
pub tokens: u64,
/// Scaled price of the secondary token given the primary is equal to the scale value
/// If scale is 1 and price is 2 then ratio is 1:2 or 1 primary token for 2 secondary tokens
pub price: u64,
/// account which the tokens were source from. The trade account holds the tokens in escrow
/// until either one or more part of a swap or the trade is canceled.
pub src_account: Pubkey,
/// account which the tokens the tokens will be deposited into on a successful trade
pub dst_account: Pubkey,
}
```
## Order cancellations
An investor may cancel a trade at anytime, but only trades they own. If the
cancellation is successful, any tokens held in escrow are returned to the
account from which they came.
```rust
pub enum ExchangeInstruction {
/// order cancellation
/// key 0 - Signer
/// key 1 -order to cancel
TradeCancellation,
}
```
## Trade swaps
The Matcher is monitoring the accounts assigned to the exchange program and
building a trade-order table. The order table is used to identify
matching orders which could be fulfilled. When a match is found the
Matcher should issue a swap request. Swap requests may not satisfy the entirety
of either order, but the exchange will greedily fulfill it. Any leftover tokens
in either account will keep the order valid for further swap requests in
the future.
Matching orders are defined by the following swap requirements:
- Opposite polarity (one `To` and one `From`)
- Operate on the same token pair
- The price ratio of the `From` order is greater than or equal to the `To` order
- There are sufficient tokens to perform the trade
Orders can be written in the following format:
`investor direction pair quantity price-ratio`
For example:
- `1 T AB 2 1`
- Investor 1 wishes to exchange 2 A tokens to B tokens at a ratio of 1 A to 1
B
- `2 F AC 6 1.2`
- Investor 2 wishes to exchange A tokens from 6 B tokens at a ratio of 1 A
from 1.2 B
An order table could look something like the following. Notice how the columns
are sorted low to high and high to low, respectively. Prices are dramatic and
whole for clarity.
|Row| To | From |
|---|-------------|------------|
| 1 | 1 T AB 2 4 | 2 F AB 2 8 |
| 2 | 1 T AB 1 4 | 2 F AB 2 8 |
| 3 | 1 T AB 6 6 | 2 F AB 2 7 |
| 4 | 1 T AB 2 8 | 2 F AB 3 6 |
| 5 | 1 T AB 2 10 | 2 F AB 1 5 |
As part of a successful swap request, the exchange will credit tokens to the
Matcher's account equal to the difference in the price ratios or the two orders.
These tokens are considered the Matcher's profit for initiating the trade.
The Matcher would initiate the following swap on the order table above:
- Row 1, To: Investor 1 trades 2 A tokens to 8 B tokens
- Row 1, From: Investor 2 trades 2 A tokens from 8 B tokens
- Matcher takes 8 B tokens as profit
Both row 1 trades are fully realized, table becomes:
|Row| To | From |
|---|-------------|------------|
| 1 | 1 T AB 1 4 | 2 F AB 2 8 |
| 2 | 1 T AB 6 6 | 2 F AB 2 7 |
| 3 | 1 T AB 2 8 | 2 F AB 3 6 |
| 4 | 1 T AB 2 10 | 2 F AB 1 5 |
The Matcher would initiate the following swap:
- Row 1, To: Investor 1 trades 1 A token to 4 B tokens
- Row 1, From: Investor 2 trades 1 A token from 4 B tokens
- Matcher takes 4 B tokens as profit
Row 1 From is not fully realized, table becomes:
|Row| To | From |
|---|-------------|------------|
| 1 | 1 T AB 6 6 | 2 F AB 1 8 |
| 2 | 1 T AB 2 8 | 2 F AB 2 7 |
| 3 | 1 T AB 2 10 | 2 F AB 3 6 |
| 4 | | 2 F AB 1 5 |
The Matcher would initiate the following swap:
- Row 1, To: Investor 1 trades 1 A token to 6 B tokens
- Row 1, From: Investor 2 trades 1 A token from 6 B tokens
- Matcher takes 2 B tokens as profit
Row 1 To is now fully realized, table becomes:
|Row| To | From |
|---|-------------|------------|
| 1 | 1 T AB 5 6 | 2 F AB 2 7 |
| 2 | 1 T AB 2 8 | 2 F AB 3 5 |
| 3 | 1 T AB 2 10 | 2 F AB 1 5 |
The Matcher would initiate the following last swap:
- Row 1, To: Investor 1 trades 2 A token to 12 B tokens
- Row 1, From: Investor 2 trades 2 A token from 12 B tokens
- Matcher takes 2 B tokens as profit
Table becomes:
|Row| To | From |
|---|-------------|------------|
| 1 | 1 T AB 3 6 | 2 F AB 3 5 |
| 2 | 1 T AB 2 8 | 2 F AB 1 5 |
| 3 | 1 T AB 2 10 | |
At this point the lowest To's price is larger than the largest From's price so
no more swaps would be initiated until new orders came in.
```rust
pub enum ExchangeInstruction {
/// Trade swap request
/// key 0 - Signer
/// key 1 - Account in which to record the swap
/// key 2 - 'To' order
/// key 3 - `From` order
/// key 4 - Token account associated with the To Trade
/// key 5 - Token account associated with From trade
/// key 6 - Token account in which to deposit the Matcher profit from the swap.
SwapRequest,
}
/// Swap accounts are populated with this structure
pub struct TradeSwapInfo {
/// Pair swapped
pub pair: TokenPair,
/// `To` order
pub to_trade_order: Pubkey,
/// `From` order
pub from_trade_order: Pubkey,
/// Number of primary tokens exchanged
pub primary_tokens: u64,
/// Price the primary tokens were exchanged for
pub primary_price: u64,
/// Number of secondary tokens exchanged
pub secondary_tokens: u64,
/// Price the secondary tokens were exchanged for
pub secondary_price: u64,
}
```
## Exchange program operations
Putting all the commands together from above, the following operations will be
supported by the on-chain exchange program:
```rust
pub enum ExchangeInstruction {
/// New token account
/// key 0 - Signer
/// key 1 - New token account
AccountRequest,
/// Transfer tokens between two accounts
/// key 0 - Account to transfer tokens to
/// key 1 - Account to transfer tokens from. This can be the exchange program itself,
/// the exchange has a limitless number of tokens it can transfer.
TransferRequest(Token, u64),
/// order request
/// key 0 - Signer
/// key 1 - Account in which to record the swap
/// key 2 - Token account associated with this trade
TradeRequest(TradeRequestInfo),
/// order cancellation
/// key 0 - Signer
/// key 1 -order to cancel
TradeCancellation,
/// Trade swap request
/// key 0 - Signer
/// key 1 - Account in which to record the swap
/// key 2 - 'To' order
/// key 3 - `From` order
/// key 4 - Token account associated with the To Trade
/// key 5 - Token account associated with From trade
/// key 6 - Token account in which to deposit the Matcher profit from the swap.
SwapRequest,
}
```
## Quotes and OHLCV
The Matcher will provide current bid/ask price quotes based on trade actively and
also provide OHLCV based on some time window. The details of how the bid/ask
price quotes are calculated are yet to be decided.
## Investor strategies
To make a compelling demo, the investors needs to provide interesting trade
behavior. Something as simple as a randomly twiddled baseline would be a
minimum starting point.
## Running the exchange
The exchange bench posts trades and swaps matches as fast as it can.
You might want to bump the duration up
to 60 seconds and the batch size to 1000 for better numbers. You can modify those
in client_demo/src/demo.rs::test_exchange_local_cluster.
The following command runs the bench:
```bash
$ RUST_LOG=solana_bench_exchange=info cargo test --release -- --nocapture test_exchange_local_cluster
```
To also see the cluster messages:
```bash
$ RUST_LOG=solana_bench_exchange=info,solana=info cargo test --release -- --nocapture test_exchange_local_cluster
```

File diff suppressed because it is too large Load Diff

View File

@ -1,221 +0,0 @@
use clap::{crate_description, crate_name, value_t, App, Arg, ArgMatches};
use solana_core::gen_keys::GenKeys;
use solana_faucet::faucet::FAUCET_PORT;
use solana_sdk::signature::{read_keypair_file, Keypair};
use std::net::SocketAddr;
use std::process::exit;
use std::time::Duration;
pub struct Config {
pub entrypoint_addr: SocketAddr,
pub faucet_addr: SocketAddr,
pub identity: Keypair,
pub threads: usize,
pub num_nodes: usize,
pub duration: Duration,
pub transfer_delay: u64,
pub fund_amount: u64,
pub batch_size: usize,
pub chunk_size: usize,
pub account_groups: usize,
pub client_ids_and_stake_file: String,
pub write_to_client_file: bool,
pub read_from_client_file: bool,
}
impl Default for Config {
fn default() -> Self {
Self {
entrypoint_addr: SocketAddr::from(([127, 0, 0, 1], 8001)),
faucet_addr: SocketAddr::from(([127, 0, 0, 1], FAUCET_PORT)),
identity: Keypair::new(),
num_nodes: 1,
threads: 4,
duration: Duration::new(u64::max_value(), 0),
transfer_delay: 0,
fund_amount: 100_000,
batch_size: 100,
chunk_size: 100,
account_groups: 100,
client_ids_and_stake_file: String::new(),
write_to_client_file: false,
read_from_client_file: false,
}
}
}
pub fn build_args<'a, 'b>(version: &'b str) -> App<'a, 'b> {
App::new(crate_name!())
.about(crate_description!())
.version(version)
.arg(
Arg::with_name("entrypoint")
.short("n")
.long("entrypoint")
.value_name("HOST:PORT")
.takes_value(true)
.required(false)
.default_value("127.0.0.1:8001")
.help("Cluster entry point; defaults to 127.0.0.1:8001"),
)
.arg(
Arg::with_name("faucet")
.short("d")
.long("faucet")
.value_name("HOST:PORT")
.takes_value(true)
.required(false)
.default_value("127.0.0.1:9900")
.help("Location of the faucet; defaults to 127.0.0.1:9900"),
)
.arg(
Arg::with_name("identity")
.short("i")
.long("identity")
.value_name("PATH")
.takes_value(true)
.help("File containing a client identity (keypair)"),
)
.arg(
Arg::with_name("threads")
.long("threads")
.value_name("<threads>")
.takes_value(true)
.required(false)
.default_value("1")
.help("Number of threads submitting transactions"),
)
.arg(
Arg::with_name("num-nodes")
.long("num-nodes")
.value_name("NUM")
.takes_value(true)
.required(false)
.default_value("1")
.help("Wait for NUM nodes to converge"),
)
.arg(
Arg::with_name("duration")
.long("duration")
.value_name("SECS")
.takes_value(true)
.default_value("60")
.help("Seconds to run benchmark, then exit; default is forever"),
)
.arg(
Arg::with_name("transfer-delay")
.long("transfer-delay")
.value_name("<delay>")
.takes_value(true)
.required(false)
.default_value("0")
.help("Delay between each chunk"),
)
.arg(
Arg::with_name("fund-amount")
.long("fund-amount")
.value_name("<fund>")
.takes_value(true)
.required(false)
.default_value("100000")
.help("Number of lamports to fund to each signer"),
)
.arg(
Arg::with_name("batch-size")
.long("batch-size")
.value_name("<batch>")
.takes_value(true)
.required(false)
.default_value("1000")
.help("Number of transactions before the signer rolls over"),
)
.arg(
Arg::with_name("chunk-size")
.long("chunk-size")
.value_name("<cunk>")
.takes_value(true)
.required(false)
.default_value("500")
.help("Number of transactions to generate and send at a time"),
)
.arg(
Arg::with_name("account-groups")
.long("account-groups")
.value_name("<groups>")
.takes_value(true)
.required(false)
.default_value("10")
.help("Number of account groups to cycle for each batch"),
)
.arg(
Arg::with_name("write-client-keys")
.long("write-client-keys")
.value_name("FILENAME")
.takes_value(true)
.help("Generate client keys and stakes and write the list to YAML file"),
)
.arg(
Arg::with_name("read-client-keys")
.long("read-client-keys")
.value_name("FILENAME")
.takes_value(true)
.help("Read client keys and stakes from the YAML file"),
)
}
#[allow(clippy::field_reassign_with_default)]
pub fn extract_args(matches: &ArgMatches) -> Config {
let mut args = Config::default();
args.entrypoint_addr = solana_net_utils::parse_host_port(
matches.value_of("entrypoint").unwrap(),
)
.unwrap_or_else(|e| {
eprintln!("failed to parse entrypoint address: {}", e);
exit(1)
});
args.faucet_addr = solana_net_utils::parse_host_port(matches.value_of("faucet").unwrap())
.unwrap_or_else(|e| {
eprintln!("failed to parse faucet address: {}", e);
exit(1)
});
if matches.is_present("identity") {
args.identity = read_keypair_file(matches.value_of("identity").unwrap())
.expect("can't read client identity");
} else {
args.identity = {
let seed = [42_u8; 32];
let mut rnd = GenKeys::new(seed);
rnd.gen_keypair()
};
}
args.threads = value_t!(matches.value_of("threads"), usize).expect("Failed to parse threads");
args.num_nodes =
value_t!(matches.value_of("num-nodes"), usize).expect("Failed to parse num-nodes");
let duration = value_t!(matches.value_of("duration"), u64).expect("Failed to parse duration");
args.duration = Duration::from_secs(duration);
args.transfer_delay =
value_t!(matches.value_of("transfer-delay"), u64).expect("Failed to parse transfer-delay");
args.fund_amount =
value_t!(matches.value_of("fund-amount"), u64).expect("Failed to parse fund-amount");
args.batch_size =
value_t!(matches.value_of("batch-size"), usize).expect("Failed to parse batch-size");
args.chunk_size =
value_t!(matches.value_of("chunk-size"), usize).expect("Failed to parse chunk-size");
args.account_groups = value_t!(matches.value_of("account-groups"), usize)
.expect("Failed to parse account-groups");
if let Some(s) = matches.value_of("write-client-keys") {
args.write_to_client_file = true;
args.client_ids_and_stake_file = s.to_string();
}
if let Some(s) = matches.value_of("read-client-keys") {
assert!(!args.write_to_client_file);
args.read_from_client_file = true;
args.client_ids_and_stake_file = s.to_string();
}
args
}

View File

@ -1,3 +0,0 @@
pub mod bench;
pub mod cli;
mod order_book;

View File

@ -1,82 +0,0 @@
pub mod bench;
mod cli;
pub mod order_book;
use crate::bench::{airdrop_lamports, create_client_accounts_file, do_bench_exchange, Config};
use log::*;
use solana_core::gossip_service::{discover_cluster, get_multi_client};
use solana_sdk::signature::Signer;
fn main() {
solana_logger::setup();
solana_metrics::set_panic_hook("bench-exchange");
let matches = cli::build_args(solana_version::version!()).get_matches();
let cli_config = cli::extract_args(&matches);
let cli::Config {
entrypoint_addr,
faucet_addr,
identity,
threads,
num_nodes,
duration,
transfer_delay,
fund_amount,
batch_size,
chunk_size,
account_groups,
client_ids_and_stake_file,
write_to_client_file,
read_from_client_file,
..
} = cli_config;
let config = Config {
identity,
threads,
duration,
transfer_delay,
fund_amount,
batch_size,
chunk_size,
account_groups,
client_ids_and_stake_file,
read_from_client_file,
};
if write_to_client_file {
create_client_accounts_file(
&config.client_ids_and_stake_file,
config.batch_size,
config.account_groups,
config.fund_amount,
);
} else {
info!("Connecting to the cluster");
let nodes = discover_cluster(&entrypoint_addr, num_nodes).unwrap_or_else(|_| {
panic!("Failed to discover nodes");
});
let (client, num_clients) = get_multi_client(&nodes);
info!("{} nodes found", num_clients);
if num_clients < num_nodes {
panic!("Error: Insufficient nodes discovered");
}
if !read_from_client_file {
info!("Funding keypair: {}", config.identity.pubkey());
let accounts_in_groups = batch_size * account_groups;
const NUM_SIGNERS: u64 = 2;
airdrop_lamports(
&client,
&faucet_addr,
&config.identity,
fund_amount * (accounts_in_groups + 1) as u64 * NUM_SIGNERS,
);
}
do_bench_exchange(vec![client], config);
}
}

View File

@ -1,134 +0,0 @@
use itertools::EitherOrBoth::{Both, Left, Right};
use itertools::Itertools;
use log::*;
use solana_exchange_program::exchange_state::*;
use solana_sdk::pubkey::Pubkey;
use std::cmp::Ordering;
use std::collections::BinaryHeap;
use std::{error, fmt};
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct ToOrder {
pub pubkey: Pubkey,
pub info: OrderInfo,
}
impl Ord for ToOrder {
fn cmp(&self, other: &Self) -> Ordering {
other.info.price.cmp(&self.info.price)
}
}
impl PartialOrd for ToOrder {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
Some(self.cmp(other))
}
}
#[derive(Clone, Debug, Eq, PartialEq)]
pub struct FromOrder {
pub pubkey: Pubkey,
pub info: OrderInfo,
}
impl Ord for FromOrder {
fn cmp(&self, other: &Self) -> Ordering {
self.info.price.cmp(&other.info.price)
}
}
impl PartialOrd for FromOrder {
fn partial_cmp(&self, other: &Self) -> Option<Ordering> {
Some(self.cmp(other))
}
}
#[derive(Default)]
pub struct OrderBook {
// TODO scale to x token types
to_ab: BinaryHeap<ToOrder>,
from_ab: BinaryHeap<FromOrder>,
}
impl fmt::Display for OrderBook {
fn fmt(&self, f: &mut fmt::Formatter) -> fmt::Result {
writeln!(
f,
"+-Order Book--------------------------+-------------------------------------+"
)?;
for (i, it) in self
.to_ab
.iter()
.zip_longest(self.from_ab.iter())
.enumerate()
{
match it {
Both(to, from) => writeln!(
f,
"| T AB {:8} for {:8}/{:8} | F AB {:8} for {:8}/{:8} |{}",
to.info.tokens,
SCALER,
to.info.price,
from.info.tokens,
SCALER,
from.info.price,
i
)?,
Left(to) => writeln!(
f,
"| T AB {:8} for {:8}/{:8} | |{}",
to.info.tokens, SCALER, to.info.price, i
)?,
Right(from) => writeln!(
f,
"| | F AB {:8} for {:8}/{:8} |{}",
from.info.tokens, SCALER, from.info.price, i
)?,
}
}
write!(
f,
"+-------------------------------------+-------------------------------------+"
)?;
Ok(())
}
}
impl OrderBook {
// TODO
// pub fn cancel(&mut self, pubkey: Pubkey) -> Result<(), Box<dyn error::Error>> {
// Ok(())
// }
pub fn push(&mut self, pubkey: Pubkey, info: OrderInfo) -> Result<(), Box<dyn error::Error>> {
check_trade(info.side, info.tokens, info.price)?;
match info.side {
OrderSide::Ask => {
self.to_ab.push(ToOrder { pubkey, info });
}
OrderSide::Bid => {
self.from_ab.push(FromOrder { pubkey, info });
}
}
Ok(())
}
pub fn pop(&mut self) -> Option<(ToOrder, FromOrder)> {
if let Some(pair) = Self::pop_pair(&mut self.to_ab, &mut self.from_ab) {
return Some(pair);
}
None
}
pub fn get_num_outstanding(&self) -> (usize, usize) {
(self.to_ab.len(), self.from_ab.len())
}
fn pop_pair(
to_ab: &mut BinaryHeap<ToOrder>,
from_ab: &mut BinaryHeap<FromOrder>,
) -> Option<(ToOrder, FromOrder)> {
let to = to_ab.peek()?;
let from = from_ab.peek()?;
if from.info.price < to.info.price {
debug!("Trade not viable");
return None;
}
let to = to_ab.pop()?;
let from = from_ab.pop()?;
Some((to, from))
}
}

View File

@ -1,108 +0,0 @@
use log::*;
use solana_bench_exchange::bench::{airdrop_lamports, do_bench_exchange, Config};
use solana_core::gossip_service::{discover_cluster, get_multi_client};
use solana_core::validator::ValidatorConfig;
use solana_exchange_program::exchange_processor::process_instruction;
use solana_exchange_program::id;
use solana_exchange_program::solana_exchange_program;
use solana_faucet::faucet::run_local_faucet;
use solana_local_cluster::local_cluster::{ClusterConfig, LocalCluster};
use solana_runtime::bank::Bank;
use solana_runtime::bank_client::BankClient;
use solana_sdk::genesis_config::create_genesis_config;
use solana_sdk::signature::{Keypair, Signer};
use std::process::exit;
use std::sync::mpsc::channel;
use std::time::Duration;
#[test]
#[ignore]
fn test_exchange_local_cluster() {
solana_logger::setup();
const NUM_NODES: usize = 1;
let config = Config {
identity: Keypair::new(),
duration: Duration::from_secs(1),
fund_amount: 100_000,
threads: 1,
transfer_delay: 20, // 15
batch_size: 100, // 1000
chunk_size: 10, // 200
account_groups: 1, // 10
..Config::default()
};
let Config {
fund_amount,
batch_size,
account_groups,
..
} = config;
let accounts_in_groups = batch_size * account_groups;
let cluster = LocalCluster::new(&mut ClusterConfig {
node_stakes: vec![100_000; NUM_NODES],
cluster_lamports: 100_000_000_000_000,
validator_configs: vec![ValidatorConfig::default(); NUM_NODES],
native_instruction_processors: [solana_exchange_program!()].to_vec(),
..ClusterConfig::default()
});
let faucet_keypair = Keypair::new();
cluster.transfer(
&cluster.funding_keypair,
&faucet_keypair.pubkey(),
2_000_000_000_000,
);
let (addr_sender, addr_receiver) = channel();
run_local_faucet(faucet_keypair, addr_sender, Some(1_000_000_000_000));
let faucet_addr = addr_receiver.recv_timeout(Duration::from_secs(2)).unwrap();
info!("Connecting to the cluster");
let nodes =
discover_cluster(&cluster.entry_point_info.gossip, NUM_NODES).unwrap_or_else(|err| {
error!("Failed to discover {} nodes: {:?}", NUM_NODES, err);
exit(1);
});
let (client, num_clients) = get_multi_client(&nodes);
info!("clients: {}", num_clients);
assert!(num_clients >= NUM_NODES);
const NUM_SIGNERS: u64 = 2;
airdrop_lamports(
&client,
&faucet_addr,
&config.identity,
fund_amount * (accounts_in_groups + 1) as u64 * NUM_SIGNERS,
);
do_bench_exchange(vec![client], config);
}
#[test]
fn test_exchange_bank_client() {
solana_logger::setup();
let (genesis_config, identity) = create_genesis_config(100_000_000_000_000);
let mut bank = Bank::new(&genesis_config);
bank.add_builtin("exchange_program", id(), process_instruction);
let clients = vec![BankClient::new(bank)];
do_bench_exchange(
clients,
Config {
identity,
duration: Duration::from_secs(1),
fund_amount: 100_000,
threads: 1,
transfer_delay: 20, // 0;
batch_size: 100, // 1500;
chunk_size: 10, // 1500;
account_groups: 1, // 50;
..Config::default()
},
);
}

View File

@ -1,2 +0,0 @@
/target/
/farf/

View File

@ -1,20 +1,17 @@
[package]
authors = ["Solana Maintainers <maintainers@solana.foundation>"]
authors = ["Solana Maintainers <maintainers@solana.com>"]
edition = "2018"
name = "solana-bench-streamer"
version = "1.5.4"
version = "0.11.0"
repository = "https://github.com/solana-labs/solana"
license = "Apache-2.0"
homepage = "https://solana.com/"
publish = false
[dependencies]
clap = "2.33.1"
solana-clap-utils = { path = "../clap-utils", version = "1.5.4" }
solana-streamer = { path = "../streamer", version = "1.5.4" }
solana-logger = { path = "../logger", version = "1.5.4" }
solana-net-utils = { path = "../net-utils", version = "1.5.4" }
solana-version = { path = "../version", version = "1.5.4" }
clap = "2.32.0"
solana = { path = "..", version = "0.11.0" }
solana-logger = { path = "../logger", version = "0.11.0" }
solana-netutil = { path = "../netutil", version = "0.11.0" }
[package.metadata.docs.rs]
targets = ["x86_64-unknown-linux-gnu"]
[features]
cuda = []

View File

@ -1,33 +1,34 @@
use clap::{crate_description, crate_name, App, Arg};
use solana_streamer::packet::{Packet, Packets, PacketsRecycler, PACKET_DATA_SIZE};
use solana_streamer::streamer::{receiver, PacketReceiver};
use clap::{App, Arg};
use solana::packet::{Packet, SharedPackets, BLOB_SIZE, PACKET_DATA_SIZE};
use solana::result::Result;
use solana::streamer::{receiver, PacketReceiver};
use std::cmp::max;
use std::net::{IpAddr, Ipv4Addr, SocketAddr, UdpSocket};
use std::sync::atomic::{AtomicBool, AtomicUsize, Ordering};
use std::sync::mpsc::channel;
use std::sync::Arc;
use std::thread::sleep;
use std::thread::{spawn, JoinHandle, Result};
use std::thread::{spawn, JoinHandle};
use std::time::Duration;
use std::time::SystemTime;
fn producer(addr: &SocketAddr, exit: Arc<AtomicBool>) -> JoinHandle<()> {
let send = UdpSocket::bind("0.0.0.0:0").unwrap();
let mut msgs = Packets::default();
msgs.packets.resize(10, Packet::default());
for w in msgs.packets.iter_mut() {
let msgs = SharedPackets::default();
let msgs_ = msgs.clone();
msgs.write().unwrap().packets.resize(10, Packet::default());
for w in &mut msgs.write().unwrap().packets {
w.meta.size = PACKET_DATA_SIZE;
w.meta.set_addr(&addr);
}
let msgs = Arc::new(msgs);
spawn(move || loop {
if exit.load(Ordering::Relaxed) {
return;
}
let mut num = 0;
for p in &msgs.packets {
for p in &msgs_.read().unwrap().packets {
let a = p.meta.addr();
assert!(p.meta.size <= PACKET_DATA_SIZE);
assert!(p.meta.size < BLOB_SIZE);
send.send_to(&p.data[..p.meta.size], &a).unwrap();
num += 1;
}
@ -42,7 +43,7 @@ fn sink(exit: Arc<AtomicBool>, rvs: Arc<AtomicUsize>, r: PacketReceiver) -> Join
}
let timer = Duration::new(1, 0);
if let Ok(msgs) = r.recv_timeout(timer) {
rvs.fetch_add(msgs.packets.len(), Ordering::Relaxed);
rvs.fetch_add(msgs.read().unwrap().packets.len(), Ordering::Relaxed);
}
})
}
@ -50,9 +51,7 @@ fn sink(exit: Arc<AtomicBool>, rvs: Arc<AtomicUsize>, r: PacketReceiver) -> Join
fn main() -> Result<()> {
let mut num_sockets = 1usize;
let matches = App::new(crate_name!())
.about(crate_description!())
.version(solana_version::version!())
let matches = App::new("solana-bench-streamer")
.arg(
Arg::with_name("num-recv-sockets")
.long("num-recv-sockets")
@ -67,16 +66,14 @@ fn main() -> Result<()> {
}
let mut port = 0;
let ip_addr = IpAddr::V4(Ipv4Addr::new(0, 0, 0, 0));
let mut addr = SocketAddr::new(ip_addr, 0);
let mut addr = SocketAddr::new(IpAddr::V4(Ipv4Addr::new(0, 0, 0, 0)), 0);
let exit = Arc::new(AtomicBool::new(false));
let mut read_channels = Vec::new();
let mut read_threads = Vec::new();
let recycler = PacketsRecycler::default();
for _ in 0..num_sockets {
let read = solana_net_utils::bind_to(ip_addr, port, false).unwrap();
let read = solana_netutil::bind_to(port, false).unwrap();
read.set_read_timeout(Some(Duration::new(1, 0))).unwrap();
addr = read.local_addr().unwrap();
@ -86,10 +83,9 @@ fn main() -> Result<()> {
read_channels.push(r_reader);
read_threads.push(receiver(
Arc::new(read),
&exit,
exit.clone(),
s_reader,
recycler.clone(),
"bench-streamer-test",
"bench-streamer",
));
}

View File

@ -1,4 +0,0 @@
/target/
/config/
/config-local/
/farf/

View File

@ -1,37 +1,21 @@
[package]
authors = ["Solana Maintainers <maintainers@solana.foundation>"]
authors = ["Solana Maintainers <maintainers@solana.com>"]
edition = "2018"
name = "solana-bench-tps"
version = "1.5.4"
version = "0.11.0"
repository = "https://github.com/solana-labs/solana"
license = "Apache-2.0"
homepage = "https://solana.com/"
publish = false
[dependencies]
bincode = "1.3.1"
clap = "2.33.1"
log = "0.4.11"
rayon = "1.4.0"
serde_json = "1.0.56"
serde_yaml = "0.8.13"
solana-clap-utils = { path = "../clap-utils", version = "1.5.4" }
solana-core = { path = "../core", version = "1.5.4" }
solana-genesis = { path = "../genesis", version = "1.5.4" }
solana-client = { path = "../client", version = "1.5.4" }
solana-faucet = { path = "../faucet", version = "1.5.4" }
solana-logger = { path = "../logger", version = "1.5.4" }
solana-metrics = { path = "../metrics", version = "1.5.4" }
solana-measure = { path = "../measure", version = "1.5.4" }
solana-net-utils = { path = "../net-utils", version = "1.5.4" }
solana-runtime = { path = "../runtime", version = "1.5.4" }
solana-sdk = { path = "../sdk", version = "1.5.4" }
solana-version = { path = "../version", version = "1.5.4" }
clap = "2.32.0"
rayon = "1.0.3"
serde_json = "1.0.10"
solana = { path = "..", version = "0.11.0" }
solana-drone = { path = "../drone", version = "0.11.0" }
solana-logger = { path = "../logger", version = "0.11.0" }
solana-metrics = { path = "../metrics", version = "0.11.0" }
solana-sdk = { path = "../sdk", version = "0.11.0" }
[dev-dependencies]
serial_test = "0.4.0"
serial_test_derive = "0.4.0"
solana-local-cluster = { path = "../local-cluster", version = "1.5.4" }
[package.metadata.docs.rs]
targets = ["x86_64-unknown-linux-gnu"]
[features]
cuda = []

File diff suppressed because it is too large Load Diff

View File

@ -1,80 +1,61 @@
use clap::{crate_description, crate_name, App, Arg, ArgMatches};
use solana_faucet::faucet::FAUCET_PORT;
use solana_sdk::fee_calculator::FeeRateGovernor;
use solana_sdk::{
pubkey::Pubkey,
signature::{read_keypair_file, Keypair},
};
use std::{net::SocketAddr, process::exit, time::Duration};
use std::net::SocketAddr;
use std::process::exit;
use std::time::Duration;
const NUM_LAMPORTS_PER_ACCOUNT_DEFAULT: u64 = solana_sdk::native_token::LAMPORTS_PER_SOL;
use clap::{crate_version, App, Arg, ArgMatches};
use solana_drone::drone::DRONE_PORT;
use solana_sdk::signature::{read_keypair, Keypair, KeypairUtil};
/// Holds the configuration for a single run of the benchmark
pub struct Config {
pub entrypoint_addr: SocketAddr,
pub faucet_addr: SocketAddr,
pub network_addr: SocketAddr,
pub drone_addr: SocketAddr,
pub id: Keypair,
pub threads: usize,
pub num_nodes: usize,
pub duration: Duration,
pub tx_count: usize,
pub keypair_multiplier: usize,
pub thread_batch_sleep_ms: usize,
pub sustained: bool,
pub client_ids_and_stake_file: String,
pub write_to_client_file: bool,
pub read_from_client_file: bool,
pub target_lamports_per_signature: u64,
pub multi_client: bool,
pub num_lamports_per_account: u64,
pub target_slots_per_epoch: u64,
pub target_node: Option<Pubkey>,
pub reject_extra_nodes: bool,
pub converge_only: bool,
}
impl Default for Config {
fn default() -> Config {
Config {
entrypoint_addr: SocketAddr::from(([127, 0, 0, 1], 8001)),
faucet_addr: SocketAddr::from(([127, 0, 0, 1], FAUCET_PORT)),
network_addr: SocketAddr::from(([127, 0, 0, 1], 8001)),
drone_addr: SocketAddr::from(([127, 0, 0, 1], DRONE_PORT)),
id: Keypair::new(),
threads: 4,
num_nodes: 1,
duration: Duration::new(std::u64::MAX, 0),
tx_count: 50_000,
keypair_multiplier: 8,
thread_batch_sleep_ms: 1000,
tx_count: 500_000,
sustained: false,
client_ids_and_stake_file: String::new(),
write_to_client_file: false,
read_from_client_file: false,
target_lamports_per_signature: FeeRateGovernor::default().target_lamports_per_signature,
multi_client: true,
num_lamports_per_account: NUM_LAMPORTS_PER_ACCOUNT_DEFAULT,
target_slots_per_epoch: 0,
target_node: None,
reject_extra_nodes: false,
converge_only: false,
}
}
}
/// Defines and builds the CLI args for a run of the benchmark
pub fn build_args<'a, 'b>(version: &'b str) -> App<'a, 'b> {
App::new(crate_name!()).about(crate_description!())
.version(version)
pub fn build_args<'a, 'b>() -> App<'a, 'b> {
App::new("solana-bench-tps")
.version(crate_version!())
.arg(
Arg::with_name("entrypoint")
Arg::with_name("network")
.short("n")
.long("entrypoint")
.long("network")
.value_name("HOST:PORT")
.takes_value(true)
.help("Rendezvous with the cluster at this entry point; defaults to 127.0.0.1:8001"),
.help("Rendezvous with the network at this gossip entry point; defaults to 127.0.0.1:8001"),
)
.arg(
Arg::with_name("faucet")
Arg::with_name("drone")
.short("d")
.long("faucet")
.long("drone")
.value_name("HOST:PORT")
.takes_value(true)
.help("Location of the faucet; defaults to entrypoint:FAUCET_PORT"),
.help("Location of the drone; defaults to network:DRONE_PORT"),
)
.arg(
Arg::with_name("identity")
@ -92,6 +73,11 @@ pub fn build_args<'a, 'b>(version: &'b str) -> App<'a, 'b> {
.takes_value(true)
.help("Wait for NUM nodes to converge"),
)
.arg(
Arg::with_name("reject-extra-nodes")
.long("reject-extra-nodes")
.help("Require exactly `num-nodes` on convergence. Appropriate only for internal networks"),
)
.arg(
Arg::with_name("threads")
.short("t")
@ -107,24 +93,16 @@ pub fn build_args<'a, 'b>(version: &'b str) -> App<'a, 'b> {
.takes_value(true)
.help("Seconds to run benchmark, then exit; default is forever"),
)
.arg(
Arg::with_name("converge-only")
.long("converge-only")
.help("Exit immediately after converging"),
)
.arg(
Arg::with_name("sustained")
.long("sustained")
.help("Use sustained performance mode vs. peak mode. This overlaps the tx generation with transfers."),
)
.arg(
Arg::with_name("no-multi-client")
.long("no-multi-client")
.help("Disable multi-client support, only transact with the entrypoint."),
)
.arg(
Arg::with_name("target_node")
.long("target-node")
.requires("no-multi-client")
.takes_value(true)
.value_name("PUBKEY")
.help("Specify an exact node to send transactions to."),
)
.arg(
Arg::with_name("tx_count")
.long("tx_count")
@ -132,63 +110,6 @@ pub fn build_args<'a, 'b>(version: &'b str) -> App<'a, 'b> {
.takes_value(true)
.help("Number of transactions to send per batch")
)
.arg(
Arg::with_name("keypair_multiplier")
.long("keypair-multiplier")
.value_name("NUM")
.takes_value(true)
.help("Multiply by transaction count to determine number of keypairs to create")
)
.arg(
Arg::with_name("thread-batch-sleep-ms")
.short("z")
.long("thread-batch-sleep-ms")
.value_name("NUM")
.takes_value(true)
.help("Per-thread-per-iteration sleep in ms"),
)
.arg(
Arg::with_name("write-client-keys")
.long("write-client-keys")
.value_name("FILENAME")
.takes_value(true)
.help("Generate client keys and stakes and write the list to YAML file"),
)
.arg(
Arg::with_name("read-client-keys")
.long("read-client-keys")
.value_name("FILENAME")
.takes_value(true)
.help("Read client keys and stakes from the YAML file"),
)
.arg(
Arg::with_name("target_lamports_per_signature")
.long("target-lamports-per-signature")
.value_name("LAMPORTS")
.takes_value(true)
.help(
"The cost in lamports that the cluster will charge for signature \
verification when the cluster is operating at target-signatures-per-slot",
),
)
.arg(
Arg::with_name("num_lamports_per_account")
.long("num-lamports-per-account")
.value_name("LAMPORTS")
.takes_value(true)
.help(
"Number of lamports per account.",
),
)
.arg(
Arg::with_name("target_slots_per_epoch")
.long("target-slots-per-epoch")
.value_name("SLOTS")
.takes_value(true)
.help(
"Wait until epochs are this many slots long.",
),
)
}
/// Parses a clap `ArgMatches` structure into a `Config`
@ -196,25 +117,25 @@ pub fn build_args<'a, 'b>(version: &'b str) -> App<'a, 'b> {
/// * `matches` - command line arguments parsed by clap
/// # Panics
/// Panics if there is trouble parsing any of the arguments
pub fn extract_args(matches: &ArgMatches) -> Config {
pub fn extract_args<'a>(matches: &ArgMatches<'a>) -> Config {
let mut args = Config::default();
if let Some(addr) = matches.value_of("entrypoint") {
args.entrypoint_addr = solana_net_utils::parse_host_port(addr).unwrap_or_else(|e| {
eprintln!("failed to parse entrypoint address: {}", e);
if let Some(addr) = matches.value_of("network") {
args.network_addr = addr.parse().unwrap_or_else(|e| {
eprintln!("failed to parse network: {}", e);
exit(1)
});
}
if let Some(addr) = matches.value_of("faucet") {
args.faucet_addr = solana_net_utils::parse_host_port(addr).unwrap_or_else(|e| {
eprintln!("failed to parse faucet address: {}", e);
if let Some(addr) = matches.value_of("drone") {
args.drone_addr = addr.parse().unwrap_or_else(|e| {
eprintln!("failed to parse drone address: {}", e);
exit(1)
});
}
if matches.is_present("identity") {
args.id = read_keypair_file(matches.value_of("identity").unwrap())
args.id = read_keypair(matches.value_of("identity").unwrap())
.expect("can't read client identity");
}
@ -234,56 +155,12 @@ pub fn extract_args(matches: &ArgMatches) -> Config {
}
if let Some(s) = matches.value_of("tx_count") {
args.tx_count = s.to_string().parse().expect("can't parse tx_count");
}
if let Some(s) = matches.value_of("keypair_multiplier") {
args.keypair_multiplier = s
.to_string()
.parse()
.expect("can't parse keypair-multiplier");
assert!(args.keypair_multiplier >= 2);
}
if let Some(t) = matches.value_of("thread-batch-sleep-ms") {
args.thread_batch_sleep_ms = t
.to_string()
.parse()
.expect("can't parse thread-batch-sleep-ms");
args.tx_count = s.to_string().parse().expect("can't parse tx_account");
}
args.sustained = matches.is_present("sustained");
if let Some(s) = matches.value_of("write-client-keys") {
args.write_to_client_file = true;
args.client_ids_and_stake_file = s.to_string();
}
if let Some(s) = matches.value_of("read-client-keys") {
assert!(!args.write_to_client_file);
args.read_from_client_file = true;
args.client_ids_and_stake_file = s.to_string();
}
if let Some(v) = matches.value_of("target_lamports_per_signature") {
args.target_lamports_per_signature = v.to_string().parse().expect("can't parse lamports");
}
args.multi_client = !matches.is_present("no-multi-client");
args.target_node = matches
.value_of("target_node")
.map(|target_str| target_str.parse().unwrap());
if let Some(v) = matches.value_of("num_lamports_per_account") {
args.num_lamports_per_account = v.to_string().parse().expect("can't parse lamports");
}
if let Some(t) = matches.value_of("target_slots_per_epoch") {
args.target_slots_per_epoch = t
.to_string()
.parse()
.expect("can't parse target slots per epoch");
}
args.converge_only = matches.is_present("converge-only");
args.reject_extra_nodes = matches.is_present("reject-extra-nodes");
args
}

View File

@ -1,2 +0,0 @@
pub mod bench;
pub mod cli;

View File

@ -1,148 +1,304 @@
use log::*;
use solana_bench_tps::bench::{do_bench_tps, generate_and_fund_keypairs, generate_keypairs};
use solana_bench_tps::cli;
use solana_core::gossip_service::{discover_cluster, get_client, get_multi_client};
use solana_genesis::Base64Account;
use solana_sdk::fee_calculator::FeeRateGovernor;
use solana_sdk::signature::{Keypair, Signer};
use solana_sdk::system_program;
use std::{collections::HashMap, fs::File, io::prelude::*, path::Path, process::exit, sync::Arc};
mod bench;
mod cli;
/// Number of signatures for all transactions in ~1 week at ~100K TPS
pub const NUM_SIGNATURES_FOR_TXS: u64 = 100_000 * 60 * 60 * 24 * 7;
use solana::client::mk_client;
use solana::cluster_info::{ClusterInfo, NodeInfo};
use solana::gossip_service::GossipService;
use solana::service::Service;
use solana::signature::GenKeys;
use solana::thin_client::poll_gossip_for_leader;
use solana_metrics;
use solana_sdk::signature::KeypairUtil;
use std::collections::VecDeque;
use std::process::exit;
use std::sync::atomic::{AtomicBool, AtomicIsize, AtomicUsize, Ordering};
use std::sync::{Arc, RwLock};
use std::thread::sleep;
use std::thread::Builder;
use std::time::Duration;
use std::time::Instant;
use crate::bench::*;
/// Creates a cluster and waits for the network to converge, returning the peers, leader, and gossip service
/// # Arguments
/// `leader` - the input leader node
/// `exit_signal` - atomic bool used to signal early exit to cluster
/// `num_nodes` - the number of nodes
/// # Panics
/// Panics if the spy node `RwLock` somehow ends up unreadable
fn converge(
leader: &NodeInfo,
exit_signal: &Arc<AtomicBool>,
num_nodes: usize,
) -> (Vec<NodeInfo>, Option<NodeInfo>, GossipService) {
//lets spy on the network
let (node, gossip_socket) = ClusterInfo::spy_node();
let mut spy_cluster_info = ClusterInfo::new(node);
spy_cluster_info.insert_info(leader.clone());
spy_cluster_info.set_leader(leader.id);
let spy_ref = Arc::new(RwLock::new(spy_cluster_info));
let gossip_service = GossipService::new(&spy_ref, None, gossip_socket, exit_signal.clone());
let mut v: Vec<NodeInfo> = vec![];
// wait for the network to converge, 30 seconds should be plenty
for _ in 0..30 {
{
let spy_ref = spy_ref.read().unwrap();
println!("{}", spy_ref.node_info_trace());
if spy_ref.leader_data().is_some() {
v = spy_ref.rpc_peers();
if v.len() >= num_nodes {
println!("CONVERGED!");
break;
} else {
println!(
"{} node(s) discovered (looking for {} or more)",
v.len(),
num_nodes
);
}
}
}
sleep(Duration::new(1, 0));
}
let leader = spy_ref.read().unwrap().leader_data().cloned();
(v, leader, gossip_service)
}
fn main() {
solana_logger::setup_with_default("solana=info");
solana_logger::setup();
solana_metrics::set_panic_hook("bench-tps");
let matches = cli::build_args(solana_version::version!()).get_matches();
let cli_config = cli::extract_args(&matches);
let matches = cli::build_args().get_matches();
let cfg = cli::extract_args(&matches);
let cli::Config {
entrypoint_addr,
faucet_addr,
network_addr: network,
drone_addr,
id,
threads,
num_nodes,
duration,
tx_count,
keypair_multiplier,
client_ids_and_stake_file,
write_to_client_file,
read_from_client_file,
target_lamports_per_signature,
multi_client,
num_lamports_per_account,
target_node,
..
} = &cli_config;
sustained,
reject_extra_nodes,
converge_only,
} = cfg;
let keypair_count = *tx_count * keypair_multiplier;
if *write_to_client_file {
info!("Generating {} keypairs", keypair_count);
let (keypairs, _) = generate_keypairs(&id, keypair_count as u64);
let num_accounts = keypairs.len() as u64;
let max_fee =
FeeRateGovernor::new(*target_lamports_per_signature, 0).max_lamports_per_signature;
let num_lamports_per_account = (num_accounts - 1 + NUM_SIGNATURES_FOR_TXS * max_fee)
/ num_accounts
+ num_lamports_per_account;
let mut accounts = HashMap::new();
keypairs.iter().for_each(|keypair| {
accounts.insert(
serde_json::to_string(&keypair.to_bytes().to_vec()).unwrap(),
Base64Account {
balance: num_lamports_per_account,
executable: false,
owner: system_program::id().to_string(),
data: String::new(),
},
);
});
println!("Looking for leader at {:?}", network);
let leader = poll_gossip_for_leader(network, None).expect("unable to find leader on network");
info!("Writing {}", client_ids_and_stake_file);
let serialized = serde_yaml::to_string(&accounts).unwrap();
let path = Path::new(&client_ids_and_stake_file);
let mut file = File::create(path).unwrap();
file.write_all(&serialized.into_bytes()).unwrap();
let exit_signal = Arc::new(AtomicBool::new(false));
let (nodes, leader, gossip_service) = converge(&leader, &exit_signal, num_nodes);
if nodes.len() < num_nodes {
println!(
"Error: Insufficient nodes discovered. Expecting {} or more",
num_nodes
);
exit(1);
}
if reject_extra_nodes && nodes.len() > num_nodes {
println!(
"Error: Extra nodes discovered. Expecting exactly {}",
num_nodes
);
exit(1);
}
if leader.is_none() {
println!("no leader");
exit(1);
}
if converge_only {
return;
}
info!("Connecting to the cluster");
let nodes = discover_cluster(&entrypoint_addr, *num_nodes).unwrap_or_else(|err| {
eprintln!("Failed to discover {} nodes: {:?}", num_nodes, err);
exit(1);
});
let leader = leader.unwrap();
let client = if *multi_client {
let (client, num_clients) = get_multi_client(&nodes);
if nodes.len() < num_clients {
eprintln!(
"Error: Insufficient nodes discovered. Expecting {} or more",
num_nodes
);
exit(1);
}
Arc::new(client)
} else if let Some(target_node) = target_node {
info!("Searching for target_node: {:?}", target_node);
let mut target_client = None;
for node in nodes {
if node.id == *target_node {
target_client = Some(Arc::new(get_client(&[node])));
break;
println!("leader RPC is at {} {}", leader.rpc, leader.id);
let mut client = mk_client(&leader);
let mut barrier_client = mk_client(&leader);
let mut seed = [0u8; 32];
seed.copy_from_slice(&id.public_key_bytes()[..32]);
let mut rnd = GenKeys::new(seed);
println!("Creating {} keypairs...", tx_count * 2);
let mut total_keys = 0;
let mut target = tx_count * 2;
while target > 0 {
total_keys += target;
target /= MAX_SPENDS_PER_TX;
}
let gen_keypairs = rnd.gen_n_keypairs(total_keys as u64);
let barrier_id = rnd.gen_n_keypairs(1).pop().unwrap();
println!("Get tokens...");
let num_tokens_per_account = 20;
// Sample the first keypair, see if it has tokens, if so then resume
// to avoid token loss
let keypair0_balance = client
.poll_get_balance(&gen_keypairs.last().unwrap().pubkey())
.unwrap_or(0);
if num_tokens_per_account > keypair0_balance {
let extra = num_tokens_per_account - keypair0_balance;
let total = extra * (gen_keypairs.len() as u64);
airdrop_tokens(&mut client, &drone_addr, &id, total);
println!("adding more tokens {}", extra);
fund_keys(&mut client, &id, &gen_keypairs, extra);
}
let start = gen_keypairs.len() - (tx_count * 2) as usize;
let keypairs = &gen_keypairs[start..];
airdrop_tokens(&mut barrier_client, &drone_addr, &barrier_id, 1);
println!("Get last ID...");
let mut last_id = client.get_last_id();
println!("Got last ID {:?}", last_id);
let first_tx_count = client.transaction_count();
println!("Initial transaction count {}", first_tx_count);
// Setup a thread per validator to sample every period
// collect the max transaction rate and total tx count seen
let maxes = Arc::new(RwLock::new(Vec::new()));
let sample_period = 1; // in seconds
println!("Sampling TPS every {} second...", sample_period);
let v_threads: Vec<_> = nodes
.into_iter()
.map(|v| {
let exit_signal = exit_signal.clone();
let maxes = maxes.clone();
Builder::new()
.name("solana-client-sample".to_string())
.spawn(move || {
sample_tx_count(&exit_signal, &maxes, first_tx_count, &v, sample_period);
})
.unwrap()
})
.collect();
let shared_txs: SharedTransactions = Arc::new(RwLock::new(VecDeque::new()));
let shared_tx_active_thread_count = Arc::new(AtomicIsize::new(0));
let total_tx_sent_count = Arc::new(AtomicUsize::new(0));
let s_threads: Vec<_> = (0..threads)
.map(|_| {
let exit_signal = exit_signal.clone();
let shared_txs = shared_txs.clone();
let leader = leader.clone();
let shared_tx_active_thread_count = shared_tx_active_thread_count.clone();
let total_tx_sent_count = total_tx_sent_count.clone();
Builder::new()
.name("solana-client-sender".to_string())
.spawn(move || {
do_tx_transfers(
&exit_signal,
&shared_txs,
&leader,
&shared_tx_active_thread_count,
&total_tx_sent_count,
);
})
.unwrap()
})
.collect();
// generate and send transactions for the specified duration
let start = Instant::now();
let mut reclaim_tokens_back_to_source_account = false;
let mut i = keypair0_balance;
while start.elapsed() < duration {
let balance = client.poll_get_balance(&id.pubkey()).unwrap_or(0);
metrics_submit_token_balance(balance);
// ping-pong between source and destination accounts for each loop iteration
// this seems to be faster than trying to determine the balance of individual
// accounts
let len = tx_count as usize;
generate_txs(
&shared_txs,
&keypairs[..len],
&keypairs[len..],
threads,
reclaim_tokens_back_to_source_account,
&leader,
);
// In sustained mode overlap the transfers with generation
// this has higher average performance but lower peak performance
// in tested environments.
if !sustained {
while shared_tx_active_thread_count.load(Ordering::Relaxed) > 0 {
sleep(Duration::from_millis(100));
}
}
target_client.unwrap_or_else(|| {
eprintln!("Target node {} not found", target_node);
exit(1);
})
} else {
Arc::new(get_client(&nodes))
};
// It's not feasible (would take too much time) to confirm each of the `tx_count / 2`
// transactions sent by `generate_txs()` so instead send and confirm a single transaction
// to validate the network is still functional.
send_barrier_transaction(&mut barrier_client, &mut last_id, &barrier_id);
let keypairs = if *read_from_client_file {
let path = Path::new(&client_ids_and_stake_file);
let file = File::open(path).unwrap();
info!("Reading {}", client_ids_and_stake_file);
let accounts: HashMap<String, Base64Account> = serde_yaml::from_reader(file).unwrap();
let mut keypairs = vec![];
let mut last_balance = 0;
accounts
.into_iter()
.for_each(|(keypair, primordial_account)| {
let bytes: Vec<u8> = serde_json::from_str(keypair.as_str()).unwrap();
keypairs.push(Keypair::from_bytes(&bytes).unwrap());
last_balance = primordial_account.balance;
});
if keypairs.len() < keypair_count {
eprintln!(
"Expected {} accounts in {}, only received {} (--tx_count mismatch?)",
keypair_count,
client_ids_and_stake_file,
keypairs.len(),
);
exit(1);
i += 1;
if should_switch_directions(num_tokens_per_account, i) {
reclaim_tokens_back_to_source_account = !reclaim_tokens_back_to_source_account;
}
// Sort keypairs so that do_bench_tps() uses the same subset of accounts for each run.
// This prevents the amount of storage needed for bench-tps accounts from creeping up
// across multiple runs.
keypairs.sort_by_key(|x| x.pubkey().to_string());
keypairs
} else {
generate_and_fund_keypairs(
client.clone(),
Some(*faucet_addr),
&id,
keypair_count,
*num_lamports_per_account,
)
.unwrap_or_else(|e| {
eprintln!("Error could not fund keys: {:?}", e);
exit(1);
})
};
}
do_bench_tps(client, cli_config, keypairs);
// Stop the sampling threads so it will collect the stats
exit_signal.store(true, Ordering::Relaxed);
println!("Waiting for validator threads...");
for t in v_threads {
if let Err(err) = t.join() {
println!(" join() failed with: {:?}", err);
}
}
// join the tx send threads
println!("Waiting for transmit threads...");
for t in s_threads {
if let Err(err) = t.join() {
println!(" join() failed with: {:?}", err);
}
}
let balance = client.poll_get_balance(&id.pubkey()).unwrap_or(0);
metrics_submit_token_balance(balance);
compute_and_report_stats(
&maxes,
sample_period,
&start.elapsed(),
total_tx_sent_count.load(Ordering::Relaxed),
);
// join the cluster_info client threads
gossip_service.join().unwrap();
}
#[cfg(test)]
mod tests {
use super::*;
#[test]
fn test_switch_directions() {
assert_eq!(should_switch_directions(20, 0), false);
assert_eq!(should_switch_directions(20, 1), false);
assert_eq!(should_switch_directions(20, 14), false);
assert_eq!(should_switch_directions(20, 15), true);
assert_eq!(should_switch_directions(20, 16), false);
assert_eq!(should_switch_directions(20, 19), false);
assert_eq!(should_switch_directions(20, 20), true);
assert_eq!(should_switch_directions(20, 21), false);
assert_eq!(should_switch_directions(20, 99), false);
assert_eq!(should_switch_directions(20, 100), true);
assert_eq!(should_switch_directions(20, 101), false);
}
}

View File

@ -1,68 +0,0 @@
use serial_test_derive::serial;
use solana_bench_tps::bench::{do_bench_tps, generate_and_fund_keypairs};
use solana_bench_tps::cli::Config;
use solana_client::thin_client::create_client;
use solana_core::cluster_info::VALIDATOR_PORT_RANGE;
use solana_core::validator::ValidatorConfig;
use solana_faucet::faucet::run_local_faucet;
use solana_local_cluster::local_cluster::{ClusterConfig, LocalCluster};
use solana_sdk::signature::{Keypair, Signer};
use std::sync::{mpsc::channel, Arc};
use std::time::Duration;
fn test_bench_tps_local_cluster(config: Config) {
let native_instruction_processors = vec![];
solana_logger::setup();
const NUM_NODES: usize = 1;
let cluster = LocalCluster::new(&mut ClusterConfig {
node_stakes: vec![999_990; NUM_NODES],
cluster_lamports: 200_000_000,
validator_configs: vec![ValidatorConfig::default(); NUM_NODES],
native_instruction_processors,
..ClusterConfig::default()
});
let faucet_keypair = Keypair::new();
cluster.transfer(
&cluster.funding_keypair,
&faucet_keypair.pubkey(),
100_000_000,
);
let client = Arc::new(create_client(
(cluster.entry_point_info.rpc, cluster.entry_point_info.tpu),
VALIDATOR_PORT_RANGE,
));
let (addr_sender, addr_receiver) = channel();
run_local_faucet(faucet_keypair, addr_sender, None);
let faucet_addr = addr_receiver.recv_timeout(Duration::from_secs(2)).unwrap();
let lamports_per_account = 100;
let keypair_count = config.tx_count * config.keypair_multiplier;
let keypairs = generate_and_fund_keypairs(
client.clone(),
Some(faucet_addr),
&config.id,
keypair_count,
lamports_per_account,
)
.unwrap();
let _total = do_bench_tps(client, config, keypairs);
#[cfg(not(debug_assertions))]
assert!(_total > 100);
}
#[test]
#[serial]
fn test_bench_tps_local_cluster_solana() {
test_bench_tps_local_cluster(Config {
tx_count: 100,
duration: Duration::from_secs(10),
..Config::default()
});
}

57
benches/bank.rs Normal file
View File

@ -0,0 +1,57 @@
#![feature(test)]
extern crate test;
use solana::bank::*;
use solana::mint::Mint;
use solana::status_deque::MAX_ENTRY_IDS;
use solana_sdk::hash::hash;
use solana_sdk::signature::{Keypair, KeypairUtil};
use solana_sdk::system_transaction::SystemTransaction;
use solana_sdk::transaction::Transaction;
use test::Bencher;
#[bench]
fn bench_process_transaction(bencher: &mut Bencher) {
let mint = Mint::new(100_000_000);
let bank = Bank::new(&mint);
// Create transactions between unrelated parties.
let transactions: Vec<_> = (0..4096)
.into_iter()
.map(|_| {
// Seed the 'from' account.
let rando0 = Keypair::new();
let tx = Transaction::system_move(
&mint.keypair(),
rando0.pubkey(),
10_000,
bank.last_id(),
0,
);
assert_eq!(bank.process_transaction(&tx), Ok(()));
// Seed the 'to' account and a cell for its signature.
let rando1 = Keypair::new();
let tx = Transaction::system_move(&rando0, rando1.pubkey(), 1, bank.last_id(), 0);
assert_eq!(bank.process_transaction(&tx), Ok(()));
// Finally, return the transaction to the benchmark.
tx
})
.collect();
let mut id = bank.last_id();
for _ in 0..(MAX_ENTRY_IDS - 1) {
bank.register_tick(&id);
id = hash(&id.as_ref())
}
bencher.iter(|| {
// Since benchmarker runs this multiple times, we need to clear the signatures.
bank.clear_signatures();
let results = bank.process_transactions(&transactions);
assert!(results.iter().all(Result::is_ok));
})
}

232
benches/banking_stage.rs Normal file
View File

@ -0,0 +1,232 @@
#![feature(test)]
extern crate test;
use rand::{thread_rng, Rng};
use rayon::prelude::*;
use solana::bank::Bank;
use solana::banking_stage::{BankingStage, NUM_THREADS};
use solana::entry::Entry;
use solana::mint::Mint;
use solana::packet::to_packets_chunked;
use solana::status_deque::MAX_ENTRY_IDS;
use solana_sdk::hash::hash;
use solana_sdk::pubkey::Pubkey;
use solana_sdk::signature::{Keypair, KeypairUtil, Signature};
use solana_sdk::system_transaction::SystemTransaction;
use solana_sdk::transaction::Transaction;
use std::iter;
use std::sync::mpsc::{channel, Receiver};
use std::sync::Arc;
use std::time::Duration;
use test::Bencher;
fn check_txs(receiver: &Receiver<Vec<Entry>>, ref_tx_count: usize) {
let mut total = 0;
loop {
let entries = receiver.recv_timeout(Duration::new(1, 0));
if let Ok(entries) = entries {
for entry in &entries {
total += entry.transactions.len();
}
} else {
break;
}
if total >= ref_tx_count {
break;
}
}
assert_eq!(total, ref_tx_count);
}
#[bench]
fn bench_banking_stage_multi_accounts(bencher: &mut Bencher) {
let txes = 1000 * NUM_THREADS;
let mint_total = 1_000_000_000_000;
let mint = Mint::new(mint_total);
let (verified_sender, verified_receiver) = channel();
let bank = Arc::new(Bank::new(&mint));
let dummy_leader_id = Keypair::new().pubkey();
let dummy = Transaction::system_move(
&mint.keypair(),
mint.keypair().pubkey(),
1,
mint.last_id(),
0,
);
let transactions: Vec<_> = (0..txes)
.into_par_iter()
.map(|_| {
let mut new = dummy.clone();
let from: Vec<u8> = (0..64).map(|_| thread_rng().gen()).collect();
let to: Vec<u8> = (0..64).map(|_| thread_rng().gen()).collect();
let sig: Vec<u8> = (0..64).map(|_| thread_rng().gen()).collect();
new.account_keys[0] = Pubkey::new(&from[0..32]);
new.account_keys[1] = Pubkey::new(&to[0..32]);
new.signatures = vec![Signature::new(&sig[0..64])];
new
})
.collect();
// fund all the accounts
transactions.iter().for_each(|tx| {
let fund = Transaction::system_move(
&mint.keypair(),
tx.account_keys[0],
mint_total / txes as u64,
mint.last_id(),
0,
);
let x = bank.process_transaction(&fund);
assert!(x.is_ok());
});
//sanity check, make sure all the transactions can execute sequentially
transactions.iter().for_each(|tx| {
let res = bank.process_transaction(&tx);
assert!(res.is_ok(), "sanity test transactions");
});
bank.clear_signatures();
//sanity check, make sure all the transactions can execute in parallel
let res = bank.process_transactions(&transactions);
for r in res {
assert!(r.is_ok(), "sanity parallel execution");
}
bank.clear_signatures();
let verified: Vec<_> = to_packets_chunked(&transactions.clone(), 192)
.into_iter()
.map(|x| {
let len = x.read().unwrap().packets.len();
(x, iter::repeat(1).take(len).collect())
})
.collect();
let (_stage, signal_receiver) = BankingStage::new(
&bank,
verified_receiver,
Default::default(),
&mint.last_id(),
None,
dummy_leader_id,
);
let mut id = mint.last_id();
for _ in 0..MAX_ENTRY_IDS {
id = hash(&id.as_ref());
bank.register_tick(&id);
}
bencher.iter(move || {
// make sure the tx last id is still registered
if bank.count_valid_ids(&[mint.last_id()]).len() == 0 {
bank.register_tick(&mint.last_id());
}
for v in verified.chunks(verified.len() / NUM_THREADS) {
verified_sender.send(v.to_vec()).unwrap();
}
check_txs(&signal_receiver, txes);
bank.clear_signatures();
});
}
#[bench]
fn bench_banking_stage_multi_programs(bencher: &mut Bencher) {
let progs = 4;
let txes = 1000 * NUM_THREADS;
let mint_total = 1_000_000_000_000;
let mint = Mint::new(mint_total);
let (verified_sender, verified_receiver) = channel();
let bank = Arc::new(Bank::new(&mint));
let dummy_leader_id = Keypair::new().pubkey();
let dummy = Transaction::system_move(
&mint.keypair(),
mint.keypair().pubkey(),
1,
mint.last_id(),
0,
);
let transactions: Vec<_> = (0..txes)
.into_par_iter()
.map(|_| {
let mut new = dummy.clone();
let from: Vec<u8> = (0..32).map(|_| thread_rng().gen()).collect();
let sig: Vec<u8> = (0..64).map(|_| thread_rng().gen()).collect();
let to: Vec<u8> = (0..32).map(|_| thread_rng().gen()).collect();
new.account_keys[0] = Pubkey::new(&from[0..32]);
new.account_keys[1] = Pubkey::new(&to[0..32]);
let prog = new.instructions[0].clone();
for i in 1..progs {
//generate programs that spend to random keys
let to: Vec<u8> = (0..32).map(|_| thread_rng().gen()).collect();
let to_key = Pubkey::new(&to[0..32]);
new.account_keys.push(to_key);
assert_eq!(new.account_keys.len(), i + 2);
new.instructions.push(prog.clone());
assert_eq!(new.instructions.len(), i + 1);
new.instructions[i].accounts[1] = 1 + i as u8;
assert_eq!(new.key(i, 1), Some(&to_key));
assert_eq!(
new.account_keys[new.instructions[i].accounts[1] as usize],
to_key
);
}
assert_eq!(new.instructions.len(), progs);
new.signatures = vec![Signature::new(&sig[0..64])];
new
})
.collect();
transactions.iter().for_each(|tx| {
let fund = Transaction::system_move(
&mint.keypair(),
tx.account_keys[0],
mint_total / txes as u64,
mint.last_id(),
0,
);
assert!(bank.process_transaction(&fund).is_ok());
});
//sanity check, make sure all the transactions can execute sequentially
transactions.iter().for_each(|tx| {
let res = bank.process_transaction(&tx);
assert!(res.is_ok(), "sanity test transactions");
});
bank.clear_signatures();
//sanity check, make sure all the transactions can execute in parallel
let res = bank.process_transactions(&transactions);
for r in res {
assert!(r.is_ok(), "sanity parallel execution");
}
bank.clear_signatures();
let verified: Vec<_> = to_packets_chunked(&transactions.clone(), 96)
.into_iter()
.map(|x| {
let len = x.read().unwrap().packets.len();
(x, iter::repeat(1).take(len).collect())
})
.collect();
let (_stage, signal_receiver) = BankingStage::new(
&bank,
verified_receiver,
Default::default(),
&mint.last_id(),
None,
dummy_leader_id,
);
let mut id = mint.last_id();
for _ in 0..MAX_ENTRY_IDS {
id = hash(&id.as_ref());
bank.register_tick(&id);
}
bencher.iter(move || {
// make sure the transactions are still valid
if bank.count_valid_ids(&[mint.last_id()]).len() == 0 {
bank.register_tick(&mint.last_id());
}
for v in verified.chunks(verified.len() / NUM_THREADS) {
verified_sender.send(v.to_vec()).unwrap();
}
check_txs(&signal_receiver, txes);
bank.clear_signatures();
});
}

29
benches/chacha.rs Normal file
View File

@ -0,0 +1,29 @@
//#![feature(test)]
//
//extern crate solana;
//extern crate test;
//
//use solana::chacha::chacha_cbc_encrypt_files;
//use std::fs::remove_file;
//use std::fs::File;
//use std::io::Write;
//use std::path::Path;
//use test::Bencher;
//
//#[bench]
//fn bench_chacha_encrypt(bench: &mut Bencher) {
// let in_path = Path::new("bench_chacha_encrypt_file_input.txt");
// let out_path = Path::new("bench_chacha_encrypt_file_output.txt.enc");
// {
// let mut in_file = File::create(in_path).unwrap();
// for _ in 0..1024 {
// in_file.write("123456foobar".as_bytes()).unwrap();
// }
// }
// bench.iter(move || {
// chacha_cbc_encrypt_files(in_path, out_path, "thetestkey".to_string()).unwrap();
// });
//
// remove_file(in_path).unwrap();
// remove_file(out_path).unwrap();
//}

199
benches/db_ledger.rs Normal file
View File

@ -0,0 +1,199 @@
#![feature(test)]
use rand;
extern crate test;
use rand::seq::SliceRandom;
use rand::{thread_rng, Rng};
use rocksdb::{Options, DB};
use solana::db_ledger::{DataCf, DbLedger, LedgerColumnFamilyRaw};
use solana::ledger::{get_tmp_ledger_path, make_large_test_entries, make_tiny_test_entries, Block};
use solana::packet::{Blob, BLOB_HEADER_SIZE};
use test::Bencher;
// Given some blobs and a ledger at ledger_path, benchmark writing the blobs to the ledger
fn bench_write_blobs(bench: &mut Bencher, blobs: &mut [&mut Blob], ledger_path: &str) {
let db_ledger =
DbLedger::open(&ledger_path).expect("Expected to be able to open database ledger");
let slot = 0;
let num_blobs = blobs.len();
bench.iter(move || {
for blob in blobs.iter_mut() {
let index = blob.index().unwrap();
let key = DataCf::key(slot, index);
let size = blob.size().unwrap();
db_ledger
.data_cf
.put(&db_ledger.db, &key, &blob.data[..BLOB_HEADER_SIZE + size])
.unwrap();
blob.set_index(index + num_blobs as u64).unwrap();
}
});
DB::destroy(&Options::default(), &ledger_path)
.expect("Expected successful database destruction");
}
// Insert some blobs into the ledger in preparation for read benchmarks
fn setup_read_bench(
db_ledger: &mut DbLedger,
num_small_blobs: u64,
num_large_blobs: u64,
slot: u64,
) {
// Make some big and small entries
let mut entries = make_large_test_entries(num_large_blobs as usize);
entries.extend(make_tiny_test_entries(num_small_blobs as usize));
// Convert the entries to blobs, write the blobs to the ledger
let shared_blobs = entries.to_blobs();
for b in shared_blobs.iter() {
b.write().unwrap().set_slot(slot).unwrap();
}
db_ledger
.write_shared_blobs(&shared_blobs)
.expect("Expectd successful insertion of blobs into ledger");
}
// Write small blobs to the ledger
#[bench]
#[ignore]
fn bench_write_small(bench: &mut Bencher) {
let ledger_path = get_tmp_ledger_path("bench_write_small");
let num_entries = 32 * 1024;
let entries = make_tiny_test_entries(num_entries);
let shared_blobs = entries.to_blobs();
let mut blob_locks: Vec<_> = shared_blobs.iter().map(|b| b.write().unwrap()).collect();
let mut blobs: Vec<&mut Blob> = blob_locks.iter_mut().map(|b| &mut **b).collect();
bench_write_blobs(bench, &mut blobs, &ledger_path);
}
// Write big blobs to the ledger
#[bench]
#[ignore]
fn bench_write_big(bench: &mut Bencher) {
let ledger_path = get_tmp_ledger_path("bench_write_big");
let num_entries = 32 * 1024;
let entries = make_tiny_test_entries(num_entries);
let shared_blobs = entries.to_blobs();
let mut blob_locks: Vec<_> = shared_blobs.iter().map(|b| b.write().unwrap()).collect();
let mut blobs: Vec<&mut Blob> = blob_locks.iter_mut().map(|b| &mut **b).collect();
bench_write_blobs(bench, &mut blobs, &ledger_path);
}
#[bench]
#[ignore]
fn bench_read_sequential(bench: &mut Bencher) {
let ledger_path = get_tmp_ledger_path("bench_read_sequential");
let mut db_ledger =
DbLedger::open(&ledger_path).expect("Expected to be able to open database ledger");
// Insert some big and small blobs into the ledger
let num_small_blobs = 32 * 1024;
let num_large_blobs = 32 * 1024;
let total_blobs = num_small_blobs + num_large_blobs;
let slot = 0;
setup_read_bench(&mut db_ledger, num_small_blobs, num_large_blobs, slot);
let num_reads = total_blobs / 15;
let mut rng = rand::thread_rng();
bench.iter(move || {
// Generate random starting point in the range [0, total_blobs - 1], read num_reads blobs sequentially
let start_index = rng.gen_range(0, num_small_blobs + num_large_blobs);
for i in start_index..start_index + num_reads {
let _ =
db_ledger
.data_cf
.get_by_slot_index(&db_ledger.db, slot, i as u64 % total_blobs);
}
});
DB::destroy(&Options::default(), &ledger_path)
.expect("Expected successful database destruction");
}
#[bench]
#[ignore]
fn bench_read_random(bench: &mut Bencher) {
let ledger_path = get_tmp_ledger_path("bench_read_random");
let mut db_ledger =
DbLedger::open(&ledger_path).expect("Expected to be able to open database ledger");
// Insert some big and small blobs into the ledger
let num_small_blobs = 32 * 1024;
let num_large_blobs = 32 * 1024;
let total_blobs = num_small_blobs + num_large_blobs;
let slot = 0;
setup_read_bench(&mut db_ledger, num_small_blobs, num_large_blobs, slot);
let num_reads = total_blobs / 15;
// Generate a num_reads sized random sample of indexes in range [0, total_blobs - 1],
// simulating random reads
let mut rng = rand::thread_rng();
let indexes: Vec<usize> = (0..num_reads)
.map(|_| rng.gen_range(0, total_blobs) as usize)
.collect();
bench.iter(move || {
for i in indexes.iter() {
let _ = db_ledger
.data_cf
.get_by_slot_index(&db_ledger.db, slot, *i as u64);
}
});
DB::destroy(&Options::default(), &ledger_path)
.expect("Expected successful database destruction");
}
#[bench]
#[ignore]
fn bench_insert_data_blob_small(bench: &mut Bencher) {
let ledger_path = get_tmp_ledger_path("bench_insert_data_blob_small");
let db_ledger =
DbLedger::open(&ledger_path).expect("Expected to be able to open database ledger");
let num_entries = 32 * 1024;
let entries = make_tiny_test_entries(num_entries);
let mut shared_blobs = entries.to_blobs();
shared_blobs.shuffle(&mut thread_rng());
bench.iter(move || {
for blob in shared_blobs.iter_mut() {
let index = blob.read().unwrap().index().unwrap();
db_ledger.write_shared_blobs(vec![blob.clone()]).unwrap();
blob.write()
.unwrap()
.set_index(index + num_entries as u64)
.unwrap();
}
});
DB::destroy(&Options::default(), &ledger_path)
.expect("Expected successful database destruction");
}
#[bench]
#[ignore]
fn bench_insert_data_blob_big(bench: &mut Bencher) {
let ledger_path = get_tmp_ledger_path("bench_insert_data_blob_big");
let db_ledger =
DbLedger::open(&ledger_path).expect("Expected to be able to open database ledger");
let num_entries = 32 * 1024;
let entries = make_large_test_entries(num_entries);
let mut shared_blobs = entries.to_blobs();
shared_blobs.shuffle(&mut thread_rng());
bench.iter(move || {
for blob in shared_blobs.iter_mut() {
let index = blob.read().unwrap().index().unwrap();
db_ledger.write_shared_blobs(vec![blob.clone()]).unwrap();
blob.write()
.unwrap()
.set_index(index + num_entries as u64)
.unwrap();
}
});
DB::destroy(&Options::default(), &ledger_path)
.expect("Expected successful database destruction");
}

26
benches/ledger.rs Normal file
View File

@ -0,0 +1,26 @@
#![feature(test)]
extern crate test;
use solana::entry::reconstruct_entries_from_blobs;
use solana::ledger::{next_entries, Block};
use solana_sdk::hash::{hash, Hash};
use solana_sdk::signature::{Keypair, KeypairUtil};
use solana_sdk::system_transaction::SystemTransaction;
use solana_sdk::transaction::Transaction;
use test::Bencher;
#[bench]
fn bench_block_to_blobs_to_block(bencher: &mut Bencher) {
let zero = Hash::default();
let one = hash(&zero.as_ref());
let keypair = Keypair::new();
let tx0 = Transaction::system_move(&keypair, keypair.pubkey(), 1, one, 0);
let transactions = vec![tx0; 10];
let entries = next_entries(&zero, 1, transactions);
bencher.iter(|| {
let blobs = entries.to_blobs();
assert_eq!(reconstruct_entries_from_blobs(blobs).unwrap().0, entries);
});
}

View File

@ -2,7 +2,7 @@
extern crate test;
use solana_core::gen_keys::GenKeys;
use solana::signature::GenKeys;
use test::Bencher;
#[bench]

21
benches/sigverify.rs Normal file
View File

@ -0,0 +1,21 @@
#![feature(test)]
extern crate test;
use solana::packet::to_packets;
use solana::sigverify;
use solana::test_tx::test_tx;
use test::Bencher;
#[bench]
fn bench_sigverify(bencher: &mut Bencher) {
let tx = test_tx();
// generate packet vector
let batches = to_packets(&vec![tx; 128]);
// verify packets
bencher.iter(|| {
let _ans = sigverify::ed25519_verify(&batches);
})
}

26
book/README.md Normal file
View File

@ -0,0 +1,26 @@
Building the Solana book
---
Install the book's dependnecies, build, and test the book:
```bash
$ ./build.sh
```
Run any Rust tests in the markdown:
```bash
$ make test
```
Render markdown as HTML:
```bash
$ make build
```
Render and view the book:
```bash
$ make open
```

15
book/art/consensus.msc Normal file
View File

@ -0,0 +1,15 @@
msc {
client,leader,verifier_a,verifier_b,verifier_c;
client=>leader [ label = "SUBMIT" ] ;
leader=>client [ label = "CONFIRMED" ] ;
leader=>verifier_a [ label = "CONFIRMED" ] ;
leader=>verifier_b [ label = "CONFIRMED" ] ;
leader=>verifier_c [ label = "CONFIRMED" ] ;
verifier_a=>leader [ label = "VERIFIED" ] ;
verifier_b=>leader [ label = "VERIFIED" ] ;
leader=>client [ label = "FINALIZED" ] ;
leader=>verifier_a [ label = "FINALIZED" ] ;
leader=>verifier_b [ label = "FINALIZED" ] ;
leader=>verifier_c [ label = "FINALIZED" ] ;
}

25
book/art/data-plane.bob Normal file
View File

@ -0,0 +1,25 @@
.-------------.
| |
.-------------+ Leader +══════════════╗
| | | ║
| `-------------` ║
v v
.-------------. .-------------.
| +--------------------------->| |
.----+ Validator 1 | | Validator 2 +═══╗
| | |<═══════════════════════════+ | ║
| `------+------` `------+------` ║
| | ║ ║
| `------------------------------. ║ ║
| | ║ ║
| ╔════════════════════════════════╝ ║
| ║ | ║
V v V v
.-------------. .-------------. .-------------. .-------------.
| | | | | | | |
| Validator 3 +------>| Validator 4 +══════>| Validator 5 +------>| Validator 6 |
| | | | | | | |
`-------------` `-------------` `-------------` `------+------`
^ ║
║ ║
╚═════════════════════════════════════════════════════════════════╝

27
book/art/fullnode.bob Normal file
View File

@ -0,0 +1,27 @@
.---------------------------.
| Fullnode |
| |
.--------. | .------------------. |
| |---->| | |
| Client | | | JSON RPC Service | |
| |<----| | |
`----+---` | `------------------` |
| | ^ | .------------------.
| | | .----------------. | | Validators |
| | | | Gossip Service +----->| |
| | | `--------+-------` | | .------------. |
| | | ^ | | | | | |
| | | | v | | | Upstream | |
| | .--+---. .-+---. | | | Validators | |
| | | Bank |<--| TVU |<--------------+ | |
| | `------` `-----` | | `------------` |
| | ^ | | |
| | | | | .------------. |
| | .--+--. .-----------. | | | | |
`-------->| TPU +-->| Broadcast +--------->| Downstream | |
| `-----` | Service | | | | Validators | |
| `-----------` | | | | |
| | | `------------` |
`---------------------------` | |
`------------------`

9
book/art/runtime.bob Normal file
View File

@ -0,0 +1,9 @@
.-----------. .-------------. .--------------. .--------------------.
| sigverify +--->| lock memory +--->| validate fee +--->| allocate accounts +--->
`-----------` `-------------` `--------------` `--------------------`
.------------. .---------. .--------------. .--------------.
--->| load data +--->| execute +--->| commit data +-->|unlock memory |
`------------` `---------` `--------------` `--------------`

20
book/art/sdk-tools.bob Normal file
View File

@ -0,0 +1,20 @@
.----------------------------------------.
| Solana Runtime |
| |
| .------------. .------------. |
| | | | | |
.-------->| Verifier +-->| Accounts | |
| | | | | | |
.----------. | | `------------` `------------` |
| +--------` | ^ |
| Client | | LoadAccounts | |
| +--------. | .----------------` |
`----------` | | | |
| | .------+-----. .-------------. |
| | | | | | |
`-------->| Loader +-->| Interpreter | |
| | | | | |
| `------------` `-------------` |
| |
`----------------------------------------`

18
book/art/tpu.bob Normal file
View File

@ -0,0 +1,18 @@
.------------------------------------------------------.
| TPU .-------------. |
| | PoH Service | |
| `--------+----` |
| ^ | |
| | v |
| .-------. .-----------. .-+-------. .--------. | .------------.
.---------. | | Fetch | | SigVerify | | Banking | | Ledger | | | Broadcast |
| Clients |--->| Stage |->| Stage |->| Stage |-->| Write +---->| Service |
`---------` | | | | | | | | Stage | | | |
| `-------` `-----------` `----+----` `--------` | `------------`
| | |
`---------------------------------|--------------------`
|
v
.------.
| Bank |
`------`

22
book/art/tvu.bob Normal file
View File

@ -0,0 +1,22 @@
.--------.
| Leader |
`--------`
^
|
.------------------------------------|---------------------------------.
| TVU | |
| | |
| .-------. .------------. .----+---. .--------. .---------. |
.------------. | | Blob | | Retransmit | | Replay | | Ledger | | Storage | |
| Upstream +----->| Fetch |-->| Stage |-->| Stage |-->| Write |-->| Stage | |
| Validators | | | Stage | | | | | | Stage | | | |
`------------` | `-------` `----+-------` `----+---` `--------` `---------` |
| ^ | | |
| | | | |
`--------|----------|----------------|---------------------------------`
| | |
| V v
.+-----------. .------.
| Gossip | | Bank |
| Service | `------`
`------------`

10
book/book.toml Normal file
View File

@ -0,0 +1,10 @@
[book]
title = "Solana: Blockchain Rebuilt for Scale"
authors = ["The Solana Team"]
[build]
build-dir = "html"
create-missing = false
[output.html]
theme = "theme"

18
book/build.sh Executable file
View File

@ -0,0 +1,18 @@
#!/usr/bin/env bash
set -e
cd "$(dirname "$0")"
cargo_install_unless() {
declare crate=$1
shift
"$@" > /dev/null 2>&1 || \
cargo install "$crate"
}
export PATH=$CARGO_HOME/bin:$PATH
cargo_install_unless mdbook mdbook --help
cargo_install_unless svgbob_cli svgbob --help
make -j"$(nproc)"

33
book/makefile Normal file
View File

@ -0,0 +1,33 @@
BOB_SRCS=$(wildcard art/*.bob)
MD_SRCS=$(wildcard src/*.md)
SVG_IMGS=$(BOB_SRCS:art/%.bob=src/img/%.svg)
all: html/index.html
test: src/tests.ok
open: all
mdbook build --open
watch: $(SVG_IMGS)
mdbook watch
src/img/%.svg: art/%.bob
@mkdir -p $(@D)
svgbob < $< > $@
src/%.md: %.md
@mkdir -p $(@D)
@cp $< $@
src/tests.ok: $(SVG_IMGS) $(MD_SRCS)
mdbook test
touch $@
html/index.html: src/tests.ok
mdbook build
clean:
rm -f $(SVG_IMGS) src/tests.ok
rm -rf html

37
book/src/SUMMARY.md Normal file
View File

@ -0,0 +1,37 @@
# Solana Architecture
- [Introduction](introduction.md)
- [Terminology](terminology.md)
- [Getting Started](getting-started.md)
- [Example: Web Wallet](webwallet.md)
- [Programming Model](programs.md)
- [Example: Tic-Tac-Toe](tictactoe.md)
- [Drones](drones.md)
- [A Solana Cluster](cluster.md)
- [Synchronization](synchronization.md)
- [Leader Rotation](leader-rotation.md)
- [Fork Generation](fork-generation.md)
- [Anatomy of a Fullnode](fullnode.md)
- [TPU](tpu.md)
- [TVU](tvu.md)
- [Gossip Service](gossip.md)
- [The Runtime](runtime.md)
- [Proposed Architectural Changes](proposals.md)
- [Ledger Replication](ledger-replication.md)
- [Secure Enclave](enclave.md)
- [Staking Rewards](staking-rewards.md)
- [Fork Selection](fork-selection.md)
- [Entry Tree](entry-tree.md)
## Appendix
- [Appendix](appendix.md)
- [JSON RPC API](jsonrpc-api.md)
- [JavaScript API](javascript-api.md)
- [solana-wallet CLI](wallet.md)

4
book/src/appendix.md Normal file
View File

@ -0,0 +1,4 @@
# Appendix
The following sections contain reference material you may find useful in your
Solana journey.

99
book/src/cluster.md Normal file
View File

@ -0,0 +1,99 @@
# A Solana Cluster
A Solana cluster is a set of fullnodes working together to serve client
transactions and maintain the integrity of the ledger. Many clusters may
coexist. When two clusters share a common genesis block, they attempt to
converge. Otherwise, they simply ignore the existence of the other.
Transactions sent to the wrong one are quietly rejected. In this chapter, we'll
discuss how a cluster is created, how nodes join the cluster, how they share
the ledger, how they ensure the ledger is replicated, and how they cope with
buggy and malicious nodes.
## Creating a Cluster
Before starting any fullnodes, one first needs to create a *genesis block*.
The block contains entries referencing two public keys, a *mint* and a
*bootstrap leader*. The fullnode holding the bootstrap leader's secret key is
responsible for appending the first entries to the ledger. It initializes its
internal state with the mint's account. That account will hold the number of
native tokens defined by the genesis block. The second fullnode then contacts
the bootstrap leader to register as a *validator* or *replicator*. Additional
fullnodes then register with any registered member of the cluster.
A validator receives all entries from the leader and submits votes confirming
those entries are valid. After voting, the validator is expected to store those
entries until replicator nodes submit proofs that they have stored copies of
it. Once the validator observes a sufficient number of copies exist, it deletes
its copy.
## Joining a Cluster
Fullnodes and replicators enter the cluster via registration messages sent to
its *control plane*. The control plane is implemented using a *gossip*
protocol, meaning that a node may register with any existing node, and expect
its registration to propagate to all nodes in the cluster. The time it takes
for all nodes to synchronize is proportional to the square of the number of
nodes participating in the cluster. Algorithmically, that's considered very
slow, but in exchange for that time, a node is assured that it eventually has
all the same information as every other node, and that that information cannot
be censored by any one node.
## Sending Transactions to a Cluster
Clients send transactions to any fullnode's Transaction Processing Unit (TPU)
port. If the node is in the validator role, it forwards the transaction to the
designated leader. If in the leader role, the node bundles incoming
transactions, timestamps them creating an *entry*, and pushes them onto the
cluster's *data plane*. Once on the data plane, the transactions are validated
by validator nodes and replicated by replicator nodes, effectively appending
them to the ledger.
## Confirming Transactions
A Solana cluster is capable of subsecond *confirmation* for up to 150 nodes
with plans to scale up to hundreds of thousands of nodes. Once fully
implemented, confirmation times are expected to increase only with the
logarithm of the number of validators, where the logarithm's base is very high.
If the base is one thousand, for example, it means that for the first thousand
nodes, confirmation will be the duration of three network hops plus the time it
takes the slowest validator of a supermajority to vote. For the next million
nodes, confirmation increases by only one network hop.
Solana defines confirmation as the duration of time from when the leader
timestamps a new entry to the moment when it recognizes a supermajority of
ledger votes.
A gossip network is much too slow to achieve subsecond confirmation once the
network grows beyond a certain size. The time it takes to send messages to all
nodes is proportional to the square of the number of nodes. If a blockchain
wants to achieve low confirmation and attempts to do it using a gossip network,
it will be forced to centralize to just a handful of nodes.
Scalable confirmation can be achieved using the follow combination of
techniques:
1. Timestamp transactions with a VDF sample and sign the timestamp.
2. Split the transactions into batches, send each to separate nodes and have
each node share its batch with its peers.
3. Repeat the previous step recursively until all nodes have all batches.
Solana rotates leaders at fixed intervals, called *slots*. Each leader may only
produce entries during its allotted slot. The leader therefore timestamps
transactions so that validators may lookup the public key of the designated
leader. The leader then signs the timestamp so that a validator may verify the
signature, proving the signer is owner of the designated leader's public key.
Next, transactions are broken into batches so that a node can send transactions
to multiple parties without making multiple copies. If, for example, the leader
needed to send 60 transactions to 6 nodes, it would break that collection of 60
into batches of 10 transactions and send one to each node. This allows the
leader to put 60 transactions on the wire, not 60 transactions for each node.
Each node then shares its batch with its peers. Once the node has collected all
6 batches, it reconstructs the original set of 60 transactions.
A batch of transactions can only be split so many times before it is so small
that header information becomes the primary consumer of network bandwidth. At
the time of this writing, the approach is scaling well up to about 150
validators. To scale up to hundreds of thousands of validators, each node can
apply the same technique as the leader node to another set of nodes of equal
size. We call the technique *data plane fanout*, but it is not yet implemented.

86
book/src/drones.md Normal file
View File

@ -0,0 +1,86 @@
# Creating Signing Services with Drones
This chapter defines an off-chain service called a *drone*, which acts as
custodian of a user's private key. In its simplest form, it can be used to
create *airdrop* transactions, a token transfer from the drone's account to a
client's account.
## Signing Service
A drone is a simple signing service. It listens for requests to sign
*transaction data*. Once received, the drone validates the request however it
sees fit. It may, for example, only accept transaction data with a
`SystemInstruction::Move` instruction transferring only up to a certain amount
of tokens. If the drone accepts the transaction, it returns an `Ok(Signature)`
where `Signature` is a signature of the transaction data using the drone's
private key. If it rejects the transaction data, it returns a `DroneError`
describing why.
## Examples
### Granting access to an on-chain game
Creator of on-chain game tic-tac-toe hosts a drone that responds to airdrop
requests containing an `InitGame` instruction. The drone signs the transaction
data in the request and returns it, thereby authorizing its account to pay the
transaction fee and as well as seeding the game's account with enough tokens to
play it. The user then creates a transaction for its transaction data and the
drones signature and submits it to the Solana cluster. Each time the user
interacts with the game, the game pays the user enough tokens to pay the next
transaction fee to advance the game. At that point, the user may choose to keep
the tokens instead of advancing the game. If the creator wants to defend
against that case, they could require the user to return to the drone to sign
each instruction.
### Worldwide airdrop of a new token
Creator of a new on-chain token (ERC-20 interface), may wish to do a worldwide
airdrop to distribute its tokens to millions of users over just a few seconds.
That drone cannot spend resources interacting with the Solana cluster. Instead,
the drone should only verify the client is unique and human, and then return
the signature. It may also want to listen to the Solana cluster for recent
entry IDs to support client retries and to ensure the airdrop is targeting the
desired cluster.
## Attack vectors
### Invalid last_id
The drone may prefer its airdrops only target a particular Solana cluster. To
do that, it listens to the cluster for new entry IDs and ensure any requests
reference a recent one.
Note: to listen for new entry IDs assumes the drone is either a fullnode or a
*light* client. At the time of this writing, light clients have not been
implemented and no proposal describes them. This document assumes one of the
following approaches be taken:
1. Define and implement a light client
2. Embed a fullnode
3. Query the jsonrpc API for the latest last id at a rate slightly faster than
ticks are produced.
### Double spends
A client may request multiple airdrops before the first has been submitted to
the ledger. The client may do this maliciously or simply because it thinks the
first request was dropped. The drone should not simply query the cluster to
ensure the client has not already received an airdrop. Instead, it should use
`last_id` to ensure the previous request is expired before signing another.
Note that the Solana cluster will reject any transaction with a `last_id`
beyond a certain *age*.
### Denial of Service
If the transaction data size is smaller than the size of the returned signature
(or descriptive error), a single client can flood the network. Considering
that a simple `Move` operation requires two public keys (each 32 bytes) and a
`fee` field, and that the returned signature is 64 bytes (and a byte to
indicate `Ok`), consideration for this attack may not be required.
In the current design, the drone accepts TCP connections. This allows clients
to DoS the service by simply opening lots of idle connections. Switching to UDP
may be preferred. The transaction data will be smaller than a UDP packet since
the transaction sent to the Solana cluster is already pinned to using UDP.

181
book/src/enclave.md Normal file
View File

@ -0,0 +1,181 @@
# Signing using Secure Enclave
This document defines the security mechanism of signing keys used by the
fullnodes. Every node contains an asymmetric key that's used for signing
and verifying the votes. The node signs the vote transactions using its private
key. Other entities can verify the signature using the node's public key.
The node's stake or its resources could be compromised if its private key is
used to sign incorrect data (e.g. voting on multiple forks of the ledger). So,
it's important to safeguard the private key.
Secure Enclaves (such as SGX) provide a layer of memory and computation
protection. An enclave can be used to generate an asymmetric key and keep the
private key in its protected memory. It can expose an API that user (untrusted)
code can use for signing the transactions.
## Message Flow
1. The node initializes the enclave at startup
* The enclave generates an asymmetric key and returns the public key to the
node
* The keypair is ephemeral. A new keypair is generated on node bootup. A
new keypair might also be generated at runtime based on some TBD
criteria.
* The enclave returns its attestation report to the node
2. The node performs attestation of the enclave (e.g using Intel's IAS APIs)
* The node ensures that the Secure Enclave is running on a TPM and is
signed by a trusted party
3. The owner of the node grants ephemeral key permission to use its stake. This
process is TBD.
4. The node's untrusted, non-enclave software calls trusted enclave software
using its interface to sign transactions and other data.
* In case of vote signing, the node needs to verify the PoH. The PoH
verification is an integral part of signing. The enclave would be
presented with some verifiable data that it'll check before signing the vote.
* The process of generating the verifiable data in untrusted space is TBD
## PoH Verification
1. When the node votes on an en entry `X`, there's a lockout period `N`, for
which it cannot vote on a fork that does not contain `X` in its history.
2. Every time the node votes on the derivative of `X`, say `X+y`, the lockout
period for `X` increases by a factor `F` (i.e. the duration node cannot vote on
a fork that does not contain `X` increases).
* The lockout period for `X+y` is still `N` until the node votes again.
3. The lockout period increment is capped (e.g. factor `F` applies maximum 32
times).
4. The signing enclave must not sign a vote that violates this policy. This
means
* Enclave is initialized with `N`, `F` and `Factor cap`
* Enclave stores `Factor cap` number of entry IDs on which the node had
previously voted
* The sign request contains the entry ID for the new vote
* Enclave verifies that new vote's entry ID is on the correct fork
(following the rules #1 and #2 above)
## Ancestor Verification
This is alternate, albeit, less certain approach to verifying voting fork.
1. The validator maintains an active set of nodes in the cluster
2. It observes the votes from the active set in the last voting period
3. It stores the ancestor/last_tick at which each node voted
4. It sends new vote request to vote-signing service
* It includes previous votes from nodes in the active set, and their
corresponding ancestors
5. The signer checks if the previous votes contains a vote from the validator,
and the vote ancestor matches with majority of the nodes
* It signs the new vote if the check is successful
* It asserts (raises an alarm of some sort) if the check is unsuccessful
The premise is that the validator can be spoofed at most once to vote on
incorrect data. If someone hijacks the validator and submits a vote request for
bogus data, that vote will not be included in the PoH (as it'll be rejected by
the cluster). The next time the validator sends a request to sign the vote, the
signing service will detect that validator's last vote is missing (as part of
#5 above).
## Fork determination
Due to the fact that the enclave cannot process PoH, it has no direct knowledge
of fork history of a submitted validator vote. Each enclave should be initiated
with the current *active set* of public keys. A validator should submit its
current vote along with the votes of the active set (including itself) that it
observed in the slot of its previous vote. In this way, the enclave can surmise
the votes accompanying the validator's previous vote and thus the fork being
voted on. This is not possible for the validator's initial submitted vote, as
it will not have a 'previous' slot to reference. To account for this, a short
voting freeze should apply until the second vote is submitted containing the
votes within the active set, along with it's own vote, at the height of the
initial vote.
## Enclave configuration
A staking client should be configurable to prevent voting on inactive forks.
This mechanism should use the client's known active set `N_active` along with a
threshold vote `N_vote` and a threshold depth `N_depth` to determine whether or
not to continue voting on a submitted fork. This configuration should take the
form of a rule such that the client will only vote on a fork if it observes
more than `N_vote` at `N_depth`. Practically, this represents the client from
confirming that it has observed some probability of economic finality of the
submitted fork at a depth where an additional vote would create a lockout for
an undesirable amount of time if that fork turns out not to be live.
## Signing service
The signing service consists of a a JSON RPC server, and a request processor.
At startup, it starts the RPC server at a configured port and waits for
client/validator requests. It expects the following type of requests.
1. Register a new validator node
* The request contains validator's identity (public key)
* The request is signed with validator's private key
* The service will drop the request if signature of the request cannot be
verified
* The service will create a new voting asymmetric key for the validator,
and return the public key as a response
* If a validator retries to register, it'll return the public key from the
pre-existing keypair
2. Sign a vote
* The request contains voting transaction, and all verification data (as
described in Ancestor Verification)
* The request is signed with validator's private key
* The service will drop the request if signature of the request cannot be
verified
* The service will verify the voting data
* The service will return a signed transaction (or signature for the
transaction)
The service could potentially have different variations, depending on the
hardware platform capabilities. For example, if the hardware supports a secure
enclave, the service can offload asymmetric key generation, and private key
protection to the enclave. A less secure implementation of the service could
simply carry the keypair in the process memory.
## Validator voting
A validator node, at startup, creates a new vote account and registers it with
the cluster. This is done by submitting a new "vote register" transaction. The
transaction contains validator's keypair, it's vote signing public key, and
some additional information. The other nodes on the cluster process this
transaction and include the new validator in the active set.
Subsequently, the validator submits a "new vote" transaction on a voting event.
This vote is signed with validator's voting private key.
The validator code will change to interface with Signing service for "vote
register" and "new vote" use cases.
### Configuration
The validator node will be configured with Signing service's network endpoint
(IP/Port).
### Register
At startup, the validator will call Signing service using JSON RPC to register
itself. The RPC call will return the voting public key for the validator node.
The validator will create a new "vote register" transaction including this
public key in it, and submit it to the cluster.
### Collect votes for last period
The validator will look up the votes submitted by all the nodes in the cluster
for the last voting period. This information will be submitted to signing
service with new vote signing request.
### New Vote Signing
The validator will create a "new vote" transaction and send it to the signing
service using JSON RPC. The RPC request will also include the vote verification
data. On success, RPC call will return the signature for the vote. On failure,
RPC call will return the failure code.
## Challenges
1. The nodes are currently being configured with asymmetric keys that are
generated and stored in PKCS8 files.
2. The genesis block contains an entry that's signed with leader's private key.
This entry is used to identify the primordial leader.
3. Generation of verifiable data in untrusted space for PoH verification in the
enclave.
4. Need infrastructure for granting stake to an ephemeral key.

121
book/src/entry-tree.md Normal file
View File

@ -0,0 +1,121 @@
# Entry Tree
This document proposes a change to ledger and window to support Solana's [fork
generation](fork-generation.md) behavior.
## Current Design
### Functionality of Window And Ledger
The basic responsibilities of the window and the ledger in a Solana fullnode
are:
1. Window: serve as a temporary, RAM-backed store of blobs of the PoH chain
for re-ordering and assembly into contiguous blocks to be sent to the bank
for verification.
2. Window: serve as a RAM-backed repair facility for other validator nodes,
which may query the network for as-yet unreceived blobs.
3. Ledger: provide disk-based storage of the PoH chain in case of node
restart.
4. Ledger: provide disk-backed repair facility for when the (smaller)
RAM-backed window doesn't cover the repair request.
The window is at the front of a validator node's processing pipeline, blobs are
received, cached, re-ordered before being deserialized into Entries, passed to
the bank for verification, and finally on to the ledger, which is at the back
of a validator node's pipeline.
The window holds blobs (the over-the-air format, serialized Entries,
one-per-blob). The ledger holds serialized Entries without any blob
information.
### Limitations
#### One-dimensional key space
The window and the ledger are indexed by ledger height, which is number of
Entries ever generated in the PoH chain until the current blob. This
limitation prevents the window and the ledger from storing the overlapping
histories possible in Solana's consensus protocol.
#### Limited caching
The window is a circular buffer. It cannot accept blobs that are farther in
the future than the window is currently working. If a blob arrives that is too
far ahead, it is dropped and will subsequently need to be repaired, incurring
further delay for the node.
#### Loss of blob signatures
Because the blob signatures are stripped before being stored by the ledger,
repair requests served from the ledger can't be verified to the original
leader.
#### Rollback and checkpoint, switching forks, separate functions
The window and the ledger can't handle replay of alternate forks. Once a Blob
has passed through the window, it's in the past. The replay stage of a
validator will need to roll back to a previous checkpoint and decode an
alternate set of Blobs to the Bank. The separated and one-way nature of window
and ledger makes this hard.
## New Design
A unified window and ledger allows a validator to record every blob it observes
on the network, in any order, as long as the blob is consistent with the
network's leader schedule.
Blobs are moved to a fork-able key space the tuple of `leader slot` + `blob
index` (within the slot). This permits the skip-list structure of the Solana
protocol to be stored in its entirety, without a-priori choosing which fork to
follow, which Entries to persist or when to persist them.
Repair requests for recent blobs are served out of RAM or recent files and out
of deeper storage for less recent blobs, as implemented by the store backing
EntryTree.
### Functionalities of EntryTree
1. Persistence: the EntryTree lives in the front of the nodes verification
pipeline, right behind network receive and signature verification. If the
blob received is consistent with the leader schedule (i.e. was signed by the
leader for the indicated slot), it is immediately stored.
2. Repair: repair is the same as window repair above, but able to serve any
blob that's been received. EntryTree stores blobs with signatures,
preserving the chain of origination.
3. Forks: EntryTree supports random access of blobs, so can support a
validator's need to rollback and replay from a Bank checkpoint.
4. Restart: with proper pruning/culling, the EntryTree can be replayed by
ordered enumeration of entries from slot 0. The logic of the replay stage
(i.e. dealing with forks) will have to be used for the most recent entries in
the EntryTree.
### Interfacing with Bank
The bank exposes to replay stage:
1. prev_id: which PoH chain it's working on as indicated by the id of the last
entry it processed
2. tick_height: the ticks in the PoH chain currently being verified by this
bank
3. votes: a stack of records that contain
1. prev_ids: what anything after this vote must chain to in PoH
2. tick height: the tick_height at which this vote was cast
3. lockout period: how long a chain must be observed to be in the ledger to
be able to be chained below this vote
Replay stage uses EntryTree APIs to find the longest chain of entries it can
hang off a previous vote. If that chain of entries does not hang off the
latest vote, the replay stage rolls back the bank to that vote and replays the
chain from there.
### Pruning EntryTree
Once EntryTree entries are old enough, representing all the possible forks
becomes less useful, perhaps even problematic for replay upon restart. Once a
validator's votes have reached max lockout, however, any EntryTree contents
that are not on the PoH chain for that vote for can be pruned, expunged.
Replicator nodes will be responsible for storing really old ledger contents,
and validators need only persist their bank periodically.

102
book/src/fork-generation.md Normal file
View File

@ -0,0 +1,102 @@
# Fork Generation
The chapter describes how forks naturally occur as a consequence of [leader
rotation](leader-rotation.md).
## Overview
Nodes take turns being leader and generating the PoH that encodes state
changes. The cluster can tolerate loss of connection to any leader by
synthesizing what the leader ***would*** have generated had it been connected
but not ingesting any state changes. The possible number of forks is thereby
limited to a "there/not-there" skip list of forks that may arise on leader
rotation slot boundaries. At any given slot, only a single leader's
transactions will be accepted.
## Message Flow
1. Transactions are ingested by the current leader.
2. Leader filters valid transactions.
3. Leader executes valid transactions updating its state.
4. Leader packages transactions into entries based off its current PoH slot.
5. Leader transmits the entries to validator nodes (in signed blobs)
1. The PoH stream includes ticks; empty entries that indicate liveness of
the leader and the passage of time on the cluster.
2. A leader's stream begins with the tick entries necessary complete the PoH
back to the leaders most recently observed prior leader slot.
6. Validators retransmit entries to peers in their set and to further
downstream nodes.
7. Validators validate the transactions and execute them on their state.
8. Validators compute the hash of the state.
9. At specific times, i.e. specific PoH tick counts, validators transmit votes
to the leader.
1. Votes are signatures of the hash of the computed state at that PoH tick
count
2. Votes are also propagated via gossip
10. Leader executes the votes as any other transaction and broadcasts them to
the cluster.
11. Validators observe their votes and all the votes from the cluster.
## Partitions, Forks
Forks can arise at PoH tick counts that correspond to a vote. The next leader
may not have observed the last vote slot and may start their slot with
generated virtual PoH entries. These empty ticks are generated by all nodes in
the cluster at a cluster-configured rate for hashes/per/tick `Z`.
There are only two possible versions of the PoH during a voting slot: PoH with
`T` ticks and entries generated by the current leader, or PoH with just ticks.
The "just ticks" version of the PoH can be thought of as a virtual ledger, one
that all nodes in the cluster can derive from the last tick in the previous
slot.
Validators can ignore forks at other points (e.g. from the wrong leader), or
slash the leader responsible for the fork.
Validators vote based on a greedy choice to maximize their reward described in
[forks selection](fork-selection.md).
### Validator's View
#### Time Progression The diagram below represents a validator's view of the
PoH stream with possible forks over time. L1, L2, etc. are leader slot, and
`E`s represent entries from that leader during that leader's slot. The 'x's
represent ticks only, and time flows downwards in the diagram.
<img alt="Fork generation" src="img/fork-generation.svg" class="center"/>
Note that an `E` appearing on 2 forks at the same slot is a slashable
condition, so a validator observing `E3` and `E3'` can slash L3 and safely
choose `x` for that slot. Once a validator commits to a forks, other forks can
be discarded below that tick count. For any slot, validators need only
consider a single "has entries" chain or a "ticks only" chain to be proposed by
a leader. But multiple virtual entries may overlap as they link back to the a
previous slot.
#### Time Division
It's useful to consider leader rotation over PoH tick count as time division of
the job of encoding state for the cluster. The following table presents the
above tree of forks as a time-divided ledger.
leader slot | L1 | L2 | L3 | L4 | L5
-------|----|----|----|----|----
data | E1| E2 | E3 | E4 | E5
ticks since prev | | | | x | xx
Note that only data from leader L3 will be accepted during leader slot L3.
Data from L3 may include "catchup" ticks back to a slot other than L2 if L3 did
not observe L2's data. L4 and L5's transmissions include the "ticks to prev"
PoH entries.
This arrangement of the network data streams permits nodes to save exactly this
to the ledger for replay, restart, and checkpoints.
### Leader's View
When a new leader begins a slot, it must first transmit any PoH (ticks)
required to link the new slot with the most recently observed and voted slot.
The fork the leader proposes would link the current slot to a previous fork
that the leader has voted on with virtual ticks.

155
book/src/fork-selection.md Normal file
View File

@ -0,0 +1,155 @@
# Fork Selection
This article describes Solana's *Nakomoto Fork Selection* algorithm based on time
locks. It satisfies the following properties:
* A voter can eventually recover from voting on a fork that doesn't become the
fork with the desired network finality.
* If the voters share a common ancestor then they will converge to a fork
containing that ancestor no matter how they are partitioned. The converged
ancestor may not be the latest possible ancestor at the start of the fork.
* Rollback requires exponentially more time for older votes than for newer
votes.
* Voters have the freedom to set a minimum network confirmation threshold
before committing a vote to a higher lockout. This allows each voter to make
a trade-off between risk and reward. See [cost of rollback](#cost-of-rollback).
## Time
For networks like Solana, time can be the PoH hash count, which is a VDF that
provides a source of time before consensus. Other networks adopting this
approach would need to consider a global source of time.
For Solana, time uniquely identifies a specific leader for fork generation. At
any given time only 1 leader, which can be computed from the ledger itself, can
propose a fork. For more details, see [fork generation](fork-generation.md)
and [leader rotation](leader-rotation.md).
## Algorithm
The basic idea to this approach is to stack consensus votes. Each vote in the
stack is a confirmation of a fork. Each confirmed fork is an ancestor of the
fork above it. Each consensus vote has a `lockout` in units of time before the
validator can submit a vote that does not contain the confirmed fork as an
ancestor.
When a vote is added to the stack, the lockouts of all the previous votes in
the stack are doubled (more on this in [Rollback](#Rollback)). With each new
vote, a voter commits the previous votes to an ever-increasing lockout. At 32
votes we can consider the vote to be at `max lockout` any votes with a lockout
equal to or above `1<<32` are dequeued (FIFO). Dequeuing a vote is the trigger
for a reward. If a vote expires before it is dequeued, it and all the votes
above it are popped (LIFO) from the vote stack. The voter needs to start
rebuilding the stack from that point.
### Rollback
Before a vote is pushed to the stack, all the votes leading up to vote with a
lower lock time than the new vote are popped. After rollback lockouts are not
doubled until the voter catches up to the rollback height of votes.
For example, a vote stack with the following state:
| vote | vote time | lockout | lock expiration time |
|-----:|----------:|--------:|---------------------:|
| 4 | 4 | 2 | 6 |
| 3 | 3 | 4 | 7 |
| 2 | 2 | 8 | 10 |
| 1 | 1 | 16 | 17 |
*Vote 5* is at time 9, and the resulting state is
| vote | vote time | lockout | lock expiration time |
|-----:|----------:|--------:|---------------------:|
| 5 | 9 | 2 | 11 |
| 2 | 2 | 8 | 10 |
| 1 | 1 | 16 | 17 |
*Vote 6* is at time 10
| vote | vote time | lockout | lock expiration time |
|-----:|----------:|--------:|---------------------:|
| 6 | 10 | 2 | 12 |
| 5 | 9 | 4 | 13 |
| 2 | 2 | 8 | 10 |
| 1 | 1 | 16 | 17 |
At time 10 the new votes caught up to the previous votes. But *vote 2* expires
at 10, so the when *vote 7* at time 11 is applied the votes including and above
*vote 2* will be popped.
| vote | vote time | lockout | lock expiration time |
|-----:|----------:|--------:|---------------------:|
| 7 | 11 | 2 | 13 |
| 1 | 1 | 16 | 17 |
The lockout for vote 1 will not increase from 16 until the stack contains 5
votes.
### Slashing and Rewards
The purpose of the lockout is to force a voter to commit opportunity cost to a
specific fork. Voters that violate the lockouts and vote for a diverging fork
within the lockout should be punished. Slashing or simply freezing the voter
from rewards for a long period of time can be used as punishment.
Voters should be rewarded for selecting the fork that the rest of the network
selected as often as possible. This is well-aligned with generating a reward
when the vote stack is full and the oldest vote needs to be dequeued. Thus a
reward should be generated for each successful dequeue.
### Cost of Rollback
Cost of rollback of *fork A* is defined as the cost in terms of lockout time to
the validators to confirm any other fork that does not include *fork A* as an
ancestor.
The **Economic Finality** of *fork A* can be calculated as the loss of all the
rewards from rollback of *fork A* and its descendants, plus the opportunity
cost of reward due to the exponentially growing lockout of the votes that have
confirmed *fork A*.
### Thresholds
Each voter can independently set a threshold of network commitment to a fork
before that voter commits to a fork. For example, at vote stack index 7, the
lockout is 256 time units. A voter may withhold votes and let votes 0-7 expire
unless the vote at index 7 has at greater than 50% commitment in the network.
This allows each voter to independently control how much risk to commit to a
fork. Committing to forks at a higher frequency would allow the voter to earn
more rewards.
### Algorithm parameters
These parameters need to be tuned.
* Number of votes in the stack before dequeue occurs (32).
* Rate of growth for lockouts in the stack (2x).
* Starting default lockout (2).
* Threshold depth for minimum network commitment before committing to the fork
(8).
* Minimum network commitment size at threshold depth (50%+).
### Free Choice
A "Free Choice" is an unenforcible voter action. A voter that maximizes
self-reward over all possible futures should behave in such a way that the
system is stable, and the local greedy choice should result in a greedy choice
over all possible futures. A set of voter that are engaging in choices to
disrupt the protocol should be bound by their stake weight to the denial of
service. Two options exits for voter:
* a voter can outrun previous voters in virtual generation and submit a
concurrent fork
* a voter can withhold a vote to observe multiple forks before voting
In both cases, the voters in the network have several forks to pick from
concurrently, even though each fork represents a different height. In both
cases it is impossible for the protocol to detect if the voter behavior is
intentional or not.
### Greedy Choice for Concurrent Forks
When evaluating multiple forks, each voter should pick the fork that will
maximize economic finality for the network, or the latest fork if all are equal.

29
book/src/fullnode.md Normal file
View File

@ -0,0 +1,29 @@
# Anatomy of a Fullnode
<img alt="Fullnode block diagrams" src="img/fullnode.svg" class="center"/>
## Pipelining
The fullnodes make extensive use of an optimization common in CPU design,
called *pipelining*. Pipelining is the right tool for the job when there's a
stream of input data that needs to be processed by a sequence of steps, and
there's different hardware responsible for each. The quintessential example is
using a washer and dryer to wash/dry/fold several loads of laundry. Washing
must occur before drying and drying before folding, but each of the three
operations is performed by a separate unit. To maximize efficiency, one creates
a pipeline of *stages*. We'll call the washer one stage, the dryer another, and
the folding process a third. To run the pipeline, one adds a second load of
laundry to the washer just after the first load is added to the dryer.
Likewise, the third load is added to the washer after the second is in the
dryer and the first is being folded. In this way, one can make progress on
three loads of laundry simultaneously. Given infinite loads, the pipeline will
consistently complete a load at the rate of the slowest stage in the pipeline.
## Pipelining in the Fullnode
The fullnode contains two pipelined processes, one used in leader mode called
the TPU and one used in validator mode called the TVU. In both cases, the
hardware being pipelined is the same, the network input, the GPU cards, the CPU
cores, writes to disk, and the network output. What it does with that hardware
is different. The TPU exists to create ledger entries whereas the TVU exists
to validate them.

254
book/src/getting-started.md Normal file
View File

@ -0,0 +1,254 @@
# Getting Started
The Solana git repository contains all the scripts you might need to spin up your
own local testnet. Depending on what you're looking to achieve, you may want to
run a different variation, as the full-fledged, performance-enhanced
multinode testnet is considerably more complex to set up than a Rust-only,
singlenode testnode. If you are looking to develop high-level features, such
as experimenting with smart contracts, save yourself some setup headaches and
stick to the Rust-only singlenode demo. If you're doing performance optimization
of the transaction pipeline, consider the enhanced singlenode demo. If you're
doing consensus work, you'll need at least a Rust-only multinode demo. If you want
to reproduce our TPS metrics, run the enhanced multinode demo.
For all four variations, you'd need the latest Rust toolchain and the Solana
source code:
First, install Rust's package manager Cargo.
```bash
$ curl https://sh.rustup.rs -sSf | sh
$ source $HOME/.cargo/env
```
Now checkout the code from github:
```bash
$ git clone https://github.com/solana-labs/solana.git
$ cd solana
```
The demo code is sometimes broken between releases as we add new low-level
features, so if this is your first time running the demo, you'll improve
your odds of success if you check out the
[latest release](https://github.com/solana-labs/solana/releases)
before proceeding:
```bash
$ TAG=$(git describe --tags $(git rev-list --tags --max-count=1))
$ git checkout $TAG
```
### Configuration Setup
The network is initialized with a genesis ledger and fullnode configuration files.
These files can be generated by running the following script.
```bash
$ ./multinode-demo/setup.sh
```
### Drone
In order for the fullnodes and clients to work, we'll need to
spin up a drone to give out some test tokens. The drone delivers Milton
Friedman-style "air drops" (free tokens to requesting clients) to be used in
test transactions.
Start the drone with:
```bash
$ ./multinode-demo/drone.sh
```
### Singlenode Testnet
Before you start a fullnode, make sure you know the IP address of the machine you
want to be the bootstrap leader for the demo, and make sure that udp ports 8000-10000 are
open on all the machines you want to test with.
Now start the bootstrap leader in a separate shell:
```bash
$ ./multinode-demo/bootstrap-leader.sh
```
Wait a few seconds for the server to initialize. It will print "leader ready..." when it's ready to
receive transactions. The leader will request some tokens from the drone if it doesn't have any.
The drone does not need to be running for subsequent leader starts.
### Multinode Testnet
To run a multinode testnet, after starting a leader node, spin up some
additional full nodes in separate shells:
```bash
$ ./multinode-demo/fullnode-x.sh
```
To run a performance-enhanced full node on Linux,
[CUDA 10.0](https://developer.nvidia.com/cuda-downloads) must be installed on
your system:
```bash
$ ./fetch-perf-libs.sh
$ SOLANA_CUDA=1 ./multinode-demo/bootstrap-leader.sh
$ SOLANA_CUDA=1 ./multinode-demo/fullnode-x.sh
```
### Testnet Client Demo
Now that your singlenode or multinode testnet is up and running let's send it
some transactions!
In a separate shell start the client:
```bash
$ ./multinode-demo/client.sh # runs against localhost by default
```
What just happened? The client demo spins up several threads to send 500,000 transactions
to the testnet as quickly as it can. The client then pings the testnet periodically to see
how many transactions it processed in that time. Take note that the demo intentionally
floods the network with UDP packets, such that the network will almost certainly drop a
bunch of them. This ensures the testnet has an opportunity to reach 710k TPS. The client
demo completes after it has convinced itself the testnet won't process any additional
transactions. You should see several TPS measurements printed to the screen. In the
multinode variation, you'll see TPS measurements for each validator node as well.
### Testnet Debugging
There are some useful debug messages in the code, you can enable them on a per-module and per-level
basis. Before running a leader or validator set the normal RUST\_LOG environment variable.
For example
* To enable `info` everywhere and `debug` only in the solana::banking_stage module:
```bash
$ export RUST_LOG=solana=info,solana::banking_stage=debug
```
* To enable BPF program logging:
```bash
$ export RUST_LOG=solana_bpf_loader=trace
```
Generally we are using `debug` for infrequent debug messages, `trace` for potentially frequent
messages and `info` for performance-related logging.
You can also attach to a running process with GDB. The leader's process is named
_solana-fullnode_:
```bash
$ sudo gdb
attach <PID>
set logging on
thread apply all bt
```
This will dump all the threads stack traces into gdb.txt
## Public Testnet
In this example the client connects to our public testnet. To run validators on the testnet you would need to open udp ports `8000-10000`.
```bash
$ ./multinode-demo/client.sh --network $(dig +short testnet.solana.com):8001 --duration 60
```
You can observe the effects of your client's transactions on our [dashboard](https://metrics.solana.com:3000/d/testnet/testnet-hud?orgId=2&from=now-30m&to=now&refresh=5s&var-testnet=testnet)
## Linux Snap
A Linux [Snap](https://snapcraft.io/) is available, which can be used to easily
get Solana running on supported Linux systems without building anything from
source for evaluation. Note that CUDA is not supported by the Snap so
performance will be limited.
The `edge` Snap channel is updated daily with the latest
development from the `master` branch. To install:
```bash
$ sudo snap install solana --edge --devmode
```
Once installed the usual Solana programs will be available as `solona.*` instead
of `solana-*`. For example, `solana.fullnode` instead of `solana-fullnode`.
Update to the latest version at any time with:
```bash
$ snap info solana
$ sudo snap refresh solana --devmode
```
### Daemon Support
The snap supports running fullnodes and a drone as system daemons.
Run `sudo snap get solana` to view the current daemon configuration. To view
daemon logs:
1. Run `sudo snap logs -n=all solana` to view the daemon initialization log
2. Runtime logging can be found under `/var/snap/solana/current/bootstrap-leader/`,
`/var/snap/solana/current/fullnode/`, or `/var/snap/solana/current/drone/` depending
on which `mode=` was selected. Within each log directory the file `current`
contains the latest log, and the files `*.s` (if present) contain older rotated
logs.
Disable the daemon at any time by running:
```bash
$ sudo snap set solana mode=
```
Runtime configuration files for the daemon can be found in
`/var/snap/solana/current/config`.
#### Leader Daemon
```bash
$ sudo snap set solana mode=bootstrap-leader
```
`rsync` must be configured and running on the leader.
1. Ensure rsync is installed with `sudo apt-get -y install rsync`
2. Edit `/etc/rsyncd.conf` to include the following
```ini
[config]
path = /var/snap/solana/current/config
hosts allow = *
read only = true
```
3. Run `sudo systemctl enable rsync; sudo systemctl start rsync`
4. Test by running `rsync -Pzravv rsync://<ip-address-of-leader>/config
solana-config` from another machine. **If the leader is running on a cloud
provider it may be necessary to configure the Firewall rules to permit ingress
to port tcp:873, tcp:9900 and the port range udp:8000-udp:10000**
To run both the Leader and Drone:
```bash
$ sudo snap set solana mode=bootstrap-leader+drone
```
#### Validator daemon
```bash
$ sudo snap set solana mode=fullnode
```
By default the node will attempt to connect to **testnet.solana.com**, override the
cluster entrypoint IP address by running:
```bash
$ sudo snap set solana mode=fullnode entrypoint-ip=127.0.0.1 #<-- change IP address
```
It's assumed that the node at the entrypoint IP will be running `rsync`
configured as described in the previous **Leader daemon** section.

79
book/src/gossip.md Normal file
View File

@ -0,0 +1,79 @@
# Gossip Service
The Gossip Service acts as a gateway to nodes in the control plane. Fullnodes
use the service to ensure information is available to all other nodes in a cluster.
The service broadcasts information using a gossip protocol.
## Gossip Overview
Nodes continuously share signed data objects among themselves in order to
manage a cluster. For example, they share their contact information, ledger
height, and votes.
Every tenth of a second, each node sends a "push" message and/or a "pull"
message. Push and pull messages may elicit responses, and push messages may be
forwarded on to others in the cluster.
Gossip runs on a well-known UDP/IP port or a port in a well-known range. Once
a cluster is bootstrapped, nodes advertise to each other where to find their
gossip endpoint (a socket address).
## Gossip Records
Records shared over gossip are arbitrary, but signed and versioned (with a
timestamp) as needed to make sense to the node receiving them. If a node
recieves two records from the same source, it it updates its own copy with the
record with the most recent timestamp.
## Gossip Service Interface
### Push Message
A node sends a push message to tells the cluster it has information to share.
Nodes send push messages to `PUSH_FANOUT` push peers.
Upon receiving a push message, a node examines the message for:
1. Duplication: if the message has been seen before, the node responds with
`PushMessagePrune` and drops the message
2. New data: if the message is new to the node
* Stores the new information with an updated version in its cluster info and
purges any previous older value
* Stores the message in `pushed_once` (used for detecting duplicates,
purged after `PUSH_MSG_TIMEOUT * 5` ms)
* Retransmits the messages to its own push peers
3. Expiration: nodes drop push messages that are older than `PUSH_MSG_TIMEOUT`
### Push Peers, Prune Message
A nodes selects its push peers at random from the active set of known peers.
The node keeps this selection for a relatively long time. When a prune message
is received, the node drops the push peer that sent the prune. Prune is an
indication that there is another, faster path to that node than direct push.
The set of push peers is kept fresh by rotating a new node into the set every
`PUSH_MSG_TIMEOUT/2` milliseconds.
### Pull Message
A node sends a pull message to ask the cluster if there is any new information.
A pull message is sent to a single peer at random and comprises a Bloom filter
that represents things it already has. A node receiving a pull message
iterates over its values and constructs a pull response of things that miss the
filter and would fit in a message.
A node constructs the pull Bloom filter by iterating over current values and
recently purged values.
A node handles items in a pull response the same way it handles new data in a
push message.
## Purging
Nodes retain prior versions of values (those updated by a pull or push) and
expired values (those older than `GOSSIP_PULL_CRDS_TIMEOUT_MS`) in
`purged_values` (things I recently had). Nodes purge `purged_values` that are
older than `5 * GOSSIP_PULL_CRDS_TIMEOUT_MS`.

117
book/src/introduction.md Normal file
View File

@ -0,0 +1,117 @@
# What is Solana?
Solana is the name of an open source project that is implementing a new
high-performance, permissionless blockchain. Solana is also the name of a
company headquartered in San Francisco that maintains the open source project.
# About this Book
This book describes the Solana open source project, a blockchain built from the
ground up for scale. The book covers why to use it, how to use it, how it
works, and why it will continue to work long after the company Solana closes
its doors. The goal of the Solana architecture is to demonstrate there exists a
set of software algorithms that when used in combination to implement a
blockchain, removes software as a performance bottleneck, allowing transaction
throughput to scale proportionally with network bandwidth. The architecture
goes on to satisfy all three desirable properties of a proper blockchain: that
it not only be scalable, but that it is also secure and decentralized.
The architecture describes a theoretical upper bound of 710 thousand
transactions per second (tps) on a standard gigabit network and 28.4 million
tps on 40 gigabit. Furthermore, the architecture supports safe, concurrent
execution of programs authored in general purpose programming languages such as
C or Rust.
# Disclaimer
All claims, content, designs, algorithms, estimates, roadmaps, specifications,
and performance measurements described in this project are done with the
author's best effort. It is up to the reader to check and validate their
accuracy and truthfulness. Furthermore, nothing in this project constitutes a
solicitation for investment.
# History of the Solana Codebase
In November of 2017 Anatoly Yakovenko published a whitepaper describing Proof
of History, a technique for keeping time between computers that do not trust
one another. From Anatoly's previous experience designing distributed systems
at Qualcomm, Mesosphere and Dropbox, he knew that a reliable clock makes
network synchronization very simple. When synchronization is simple the
resulting network can be blazing fast, bound only by network bandwidth.
Anatoly watched as blockchain systems without clocks, such as Bitcoin and
Ethereum, struggled to scale beyond 15 transactions per second worldwide when
centralized payment systems such as Visa required peaks of 65,000. Without a
clock, it was clear they'd never graduate to being the global payment system or
global supercomputer they had dreamed to be. When Anatoly solved the problem of
getting computers that dont trust each other to agree on time, he knew he had
the key to bring 40 years of distributed systems research to the world of
blockchain. The resulting cluster wouldn't be just 10 times faster, or a 100
times, or a 1,000 times, but 10,000 times faster right out of the gate!
Anatoly's implementation began in a private codebase and was implemented in the
C programming language. Greg Fitzgerald, who had previously worked with Anatoly
at semiconductor giant Qualcomm Incorporated, encouraged him to reimplement the
project in the Rust programming language. Greg had worked on the LLVM compiler
infrastructure, which underlies both the Clang C/C++ compiler as well as the
Rust compiler. Greg claimed that the language's safety guarantees would improve
software productivity and that its lack of a garbage collector would allow
programs to perform as well as those written in C. Anatoly gave it a shot and
just two weeks later, had migrated his entire codebase to Rust. Sold. With
plans to weave all the world's transactions together on a single, scalable
blockchain, Anatoly called the project Loom.
On February 13th of 2018, Greg began prototyping the first open source
implementation of Anatoly's whitepaper. The project was published to GitHub
under the name Silk in the loomprotocol organization. On February 28th, Greg
made his first release, demonstrating 10 thousand signed transactions could be
verified and processed in just over half a second. Shortly after, another
former Qualcomm cohort, Stephen Akridge, demonstrated throughput could be
massively improved by offloading signature verification to graphics processors.
Anatoly recruited Greg, Stephen and three others to co-found a company, then
called Loom.
Around the same time, Ethereum-based project Loom Network sprung up and many
people were confused if they were the same project. The Loom team decided it
would rebrand. They chose the name Solana, a nod to a small beach town North of
San Diego called Solana Beach, where Anatoly, Greg and Stephen lived and surfed
for three years when they worked for Qualcomm. On March 28th, the team created
the Solana Labs GitHub organization and renamed Greg's prototype Silk to
Solana.
In June of 2018, the team scaled up the technology to run on cloud-based
networks and on July 19th, published a 50-node, permissioned, public testnet
consistently supporting bursts of 250,000 transactions per second. In the most
recent release, v0.10 Pillbox, the team published a permissioned testnet
running 150 nodes on a gigabit network and demonstrated soak tests processing
an *average* of 200 thousand transactions per second with bursts over 500
thousand. The project was also extended to support on-chain programs written in
the C programming language and run concurrently in a safe execution environment
called BPF. Next step: going permissionless.
# What is a Solana Cluster?
A cluster is a set of computers that work together and can be viewed from the
outside as a single system. A Solana cluster is a set of independently owned
computers working together (and sometimes against each other) to verify the
output of untrusted, user-submitted programs. A Solana cluster can be utilized
any time a user wants to preserve an immutable record of events in time or
programmatic interpretations of those events. One use is to track which of the
computers did meaningful work to keep the cluster running. Another use might be
to track the possession of real-world assets. In each case, the cluster
produces a record of events called the ledger. It will be preserved for the
lifetime of the cluster. As long as someone somewhere in the world maintains a
copy of the ledger, the output of its programs (which may contain a record of
who possesses what) will forever be reproducible, independent of the
organization that launched it.
# What are Sols?
A sol is the name of Solana's native token, which can be passed to nodes in a
Solana cluster in exchange for running an on-chain program or validating its
output. The Solana protocol defines that only 1 billion sols will ever exist,
but that the system may perform micropayments of fractional sols and that a sol
may be split as many as 34 times. The fractional sol is called a *lamport*. It
is named in honor of Solana's biggest technical influence, [Leslie
Lamport](https://en.wikipedia.org/wiki/Leslie_Lamport). A lamport has a value
of approximately 0.0000000000582 sol (2^-34).

View File

@ -1,5 +1,3 @@
---
title: Web3 JavaScript API
---
# JavaScript API
See [solana-web3](https://solana-labs.github.io/solana-web3.js/).

361
book/src/jsonrpc-api.md Normal file
View File

@ -0,0 +1,361 @@
JSON RPC API
===
Solana nodes accept HTTP requests using the [JSON-RPC 2.0](https://www.jsonrpc.org/specification) specification.
To interact with a Solana node inside a JavaScript application, use the [solana-web3.js](https://github.com/solana-labs/solana-web3.js) library, which gives a convenient interface for the RPC methods.
RPC HTTP Endpoint
---
**Default port:** 8899
eg. http://localhost:8899, http://192.168.1.88:8899
RPC PubSub WebSocket Endpoint
---
**Default port:** 8900
eg. ws://localhost:8900, http://192.168.1.88:8900
Methods
---
* [confirmTransaction](#confirmtransaction)
* [getAccountInfo](#getaccountinfo)
* [getBalance](#getbalance)
* [getConfirmationTime](#getconfirmationTime)
* [getLastId](#getlastid)
* [getSignatureStatus](#getsignaturestatus)
* [getTransactionCount](#gettransactioncount)
* [requestAirdrop](#requestairdrop)
* [sendTransaction](#sendtransaction)
* [startSubscriptionChannel](#startsubscriptionchannel)
* [Subscription Websocket](#subscription-websocket)
* [accountSubscribe](#accountsubscribe)
* [accountUnsubscribe](#accountunsubscribe)
* [signatureSubscribe](#signaturesubscribe)
* [signatureUnsubscribe](#signatureunsubscribe)
Request Formatting
---
To make a JSON-RPC request, send an HTTP POST request with a `Content-Type: application/json` header. The JSON request data should contain 4 fields:
* `jsonrpc`, set to `"2.0"`
* `id`, a unique client-generated identifying integer
* `method`, a string containing the method to be invoked
* `params`, a JSON array of ordered parameter values
Example using curl:
```bash
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0", "id":1, "method":"getBalance", "params":["83astBRguLMdt2h5U1Tpdq5tjFoJ6noeGwaY3mDLVcri"]}' 192.168.1.88:8899
```
The response output will be a JSON object with the following fields:
* `jsonrpc`, matching the request specification
* `id`, matching the request identifier
* `result`, requested data or success confirmation
Requests can be sent in batches by sending an array of JSON-RPC request objects as the data for a single POST.
Definitions
---
* Hash: A SHA-256 hash of a chunk of data.
* Pubkey: The public key of a Ed25519 key-pair.
* Signature: An Ed25519 signature of a chunk of data.
* Transaction: A Solana instruction signed by a client key-pair.
JSON RPC API Reference
---
### confirmTransaction
Returns a transaction receipt
##### Parameters:
* `string` - Signature of Transaction to confirm, as base-58 encoded string
##### Results:
* `boolean` - Transaction status, true if Transaction is confirmed
##### Example:
```bash
// Request
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0", "id":1, "method":"confirmTransaction", "params":["5VERv8NMvzbJMEkV8xnrLkEaWRtSz9CosKDYjCJjBRnbJLgp8uirBgmQpjKhoR4tjF3ZpRzrFmBV6UjKdiSZkQUW"]}' http://localhost:8899
// Result
{"jsonrpc":"2.0","result":true,"id":1}
```
---
### getBalance
Returns the balance of the account of provided Pubkey
##### Parameters:
* `string` - Pubkey of account to query, as base-58 encoded string
##### Results:
* `integer` - quantity, as a signed 64-bit integer
##### Example:
```bash
// Request
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0", "id":1, "method":"getBalance", "params":["83astBRguLMdt2h5U1Tpdq5tjFoJ6noeGwaY3mDLVcri"]}' http://localhost:8899
// Result
{"jsonrpc":"2.0","result":0,"id":1}
```
---
### getAccountInfo
Returns all information associated with the account of provided Pubkey
##### Parameters:
* `string` - Pubkey of account to query, as base-58 encoded string
##### Results:
The result field will be a JSON object with the following sub fields:
* `tokens`, number of tokens assigned to this account, as a signed 64-bit integer
* `owner`, array of 32 bytes representing the program this account has been assigned to
* `userdata`, array of bytes representing any userdata associated with the account
* `executable`, boolean indicating if the account contains a program (and is strictly read-only)
* `loader`, array of 32 bytes representing the loader for this program (if `executable`), otherwise all
##### Example:
```bash
// Request
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0", "id":1, "method":"getAccountInfo", "params":["2gVkYWexTHR5Hb2aLeQN3tnngvWzisFKXDUPrgMHpdST"]}' http://localhost:8899
// Result
{"jsonrpc":"2.0","result":{"executable":false,"loader":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"owner":[1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"tokens":1,"userdata":[3,0,0,0,0,0,0,0,1,0,0,0,0,0,1,0,0,0,0,0,0,0,20,0,0,0,0,0,0,0,50,48,53,48,45,48,49,45,48,49,84,48,48,58,48,48,58,48,48,90,252,10,7,28,246,140,88,177,98,82,10,227,89,81,18,30,194,101,199,16,11,73,133,20,246,62,114,39,20,113,189,32,50,0,0,0,0,0,0,0,247,15,36,102,167,83,225,42,133,127,82,34,36,224,207,130,109,230,224,188,163,33,213,13,5,117,211,251,65,159,197,51,0,0,0,0,0,0]},"id":1}
```
---
### getLastId
Returns the last entry ID from the ledger
##### Parameters:
None
##### Results:
* `string` - the ID of last entry, a Hash as base-58 encoded string
##### Example:
```bash
// Request
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0","id":1, "method":"getLastId"}' http://localhost:8899
// Result
{"jsonrpc":"2.0","result":"GH7ome3EiwEr7tu9JuTh2dpYWBJK3z69Xm1ZE3MEE6JC","id":1}
```
---
### getSignatureStatus
Returns the status of a given signature. This method is similar to
[confirmTransaction](#confirmtransaction) but provides more resolution for error
events.
##### Parameters:
* `string` - Signature of Transaction to confirm, as base-58 encoded string
##### Results:
* `string` - Transaction status:
* `Confirmed` - Transaction was successful
* `SignatureNotFound` - Unknown transaction
* `ProgramRuntimeError` - An error occurred in the program that processed this Transaction
* `AccountInUse` - Another Transaction had a write lock one of the Accounts specified in this Transaction. The Transaction may succeed if retried
* `GenericFailure` - Some other error occurred. **Note**: In the future new Transaction statuses may be added to this list. It's safe to assume that all new statuses will be more specific error conditions that previously presented as `GenericFailure`
##### Example:
```bash
// Request
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0", "id":1, "method":"getSignatureStatus", "params":["5VERv8NMvzbJMEkV8xnrLkEaWRtSz9CosKDYjCJjBRnbJLgp8uirBgmQpjKhoR4tjF3ZpRzrFmBV6UjKdiSZkQUW"]}' http://localhost:8899
// Result
{"jsonrpc":"2.0","result":"SignatureNotFound","id":1}
```
---
### getTransactionCount
Returns the current Transaction count from the ledger
##### Parameters:
None
##### Results:
* `integer` - count, as unsigned 64-bit integer
##### Example:
```bash
// Request
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0","id":1, "method":"getTransactionCount"}' http://localhost:8899
// Result
{"jsonrpc":"2.0","result":268,"id":1}
```
---
### getConfirmationTime
Returns the current cluster confirmation time in milliseconds
##### Parameters:
None
##### Results:
* `integer` - confirmation time in milliseconds, as unsigned 64-bit integer
##### Example:
```bash
// Request
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0","id":1, "method":"getConfirmationTime"}' http://localhost:8899
// Result
{"jsonrpc":"2.0","result":500,"id":1}
```
---
### requestAirdrop
Requests an airdrop of tokens to a Pubkey
##### Parameters:
* `string` - Pubkey of account to receive tokens, as base-58 encoded string
* `integer` - token quantity, as a signed 64-bit integer
##### Results:
* `string` - Transaction Signature of airdrop, as base-58 encoded string
##### Example:
```bash
// Request
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0","id":1, "method":"requestAirdrop", "params":["83astBRguLMdt2h5U1Tpdq5tjFoJ6noeGwaY3mDLVcri", 50]}' http://localhost:8899
// Result
{"jsonrpc":"2.0","result":"5VERv8NMvzbJMEkV8xnrLkEaWRtSz9CosKDYjCJjBRnbJLgp8uirBgmQpjKhoR4tjF3ZpRzrFmBV6UjKdiSZkQUW","id":1}
```
---
### sendTransaction
Creates new transaction
##### Parameters:
* `array` - array of octets containing a fully-signed Transaction
##### Results:
* `string` - Transaction Signature, as base-58 encoded string
##### Example:
```bash
// Request
curl -X POST -H "Content-Type: application/json" -d '{"jsonrpc":"2.0","id":1, "method":"sendTransaction", "params":[[61, 98, 55, 49, 15, 187, 41, 215, 176, 49, 234, 229, 228, 77, 129, 221, 239, 88, 145, 227, 81, 158, 223, 123, 14, 229, 235, 247, 191, 115, 199, 71, 121, 17, 32, 67, 63, 209, 239, 160, 161, 2, 94, 105, 48, 159, 235, 235, 93, 98, 172, 97, 63, 197, 160, 164, 192, 20, 92, 111, 57, 145, 251, 6, 40, 240, 124, 194, 149, 155, 16, 138, 31, 113, 119, 101, 212, 128, 103, 78, 191, 80, 182, 234, 216, 21, 121, 243, 35, 100, 122, 68, 47, 57, 13, 39, 0, 0, 0, 0, 50, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 50, 0, 0, 0, 0, 0, 0, 0, 40, 240, 124, 194, 149, 155, 16, 138, 31, 113, 119, 101, 212, 128, 103, 78, 191, 80, 182, 234, 216, 21, 121, 243, 35, 100, 122, 68, 47, 57, 11, 12, 106, 49, 74, 226, 201, 16, 161, 192, 28, 84, 124, 97, 190, 201, 171, 186, 6, 18, 70, 142, 89, 185, 176, 154, 115, 61, 26, 163, 77, 1, 88, 98, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0]]}' http://localhost:8899
// Result
{"jsonrpc":"2.0","result":"2EBVM6cB8vAAD93Ktr6Vd8p67XPbQzCJX47MpReuiCXJAtcjaxpvWpcg9Ege1Nr5Tk3a2GFrByT7WPBjdsTycY9b","id":1}
```
---
### Subscription Websocket
After connect to the RPC PubSub websocket at `ws://<ADDRESS>/`:
- Submit subscription requests to the websocket using the methods below
- Multiple subscriptions may be active at once
---
### accountSubscribe
Subscribe to an account to receive notifications when the userdata for a given account public key changes
##### Parameters:
* `string` - account Pubkey, as base-58 encoded string
##### Results:
* `integer` - Subscription id (needed to unsubscribe)
##### Example:
```bash
// Request
{"jsonrpc":"2.0", "id":1, "method":"accountSubscribe", "params":["CM78CPUeXjn8o3yroDHxUtKsZZgoy4GPkPPXfouKNH12"]}
// Result
{"jsonrpc": "2.0","result": 0,"id": 1}
```
##### Notification Format:
```bash
{"jsonrpc": "2.0","method": "accountNotification", "params": {"result": {"executable":false,"loader":[0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"owner":[1,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0,0],"tokens":1,"userdata":[3,0,0,0,0,0,0,0,1,0,0,0,0,0,1,0,0,0,0,0,0,0,20,0,0,0,0,0,0,0,50,48,53,48,45,48,49,45,48,49,84,48,48,58,48,48,58,48,48,90,252,10,7,28,246,140,88,177,98,82,10,227,89,81,18,30,194,101,199,16,11,73,133,20,246,62,114,39,20,113,189,32,50,0,0,0,0,0,0,0,247,15,36,102,167,83,225,42,133,127,82,34,36,224,207,130,109,230,224,188,163,33,213,13,5,117,211,251,65,159,197,51,0,0,0,0,0,0]},"subscription":0}}
```
---
### accountUnsubscribe
Unsubscribe from account userdata change notifications
##### Parameters:
* `integer` - id of account Subscription to cancel
##### Results:
* `bool` - unsubscribe success message
##### Example:
```bash
// Request
{"jsonrpc":"2.0", "id":1, "method":"accountUnsubscribe", "params":[0]}
// Result
{"jsonrpc": "2.0","result": true,"id": 1}
```
---
### signatureSubscribe
Subscribe to a transaction signature to receive notification when the transaction is confirmed
On `signatureNotification`, the subscription is automatically cancelled
##### Parameters:
* `string` - Transaction Signature, as base-58 encoded string
##### Results:
* `integer` - subscription id (needed to unsubscribe)
##### Example:
```bash
// Request
{"jsonrpc":"2.0", "id":1, "method":"signatureSubscribe", "params":["2EBVM6cB8vAAD93Ktr6Vd8p67XPbQzCJX47MpReuiCXJAtcjaxpvWpcg9Ege1Nr5Tk3a2GFrByT7WPBjdsTycY9b"]}
// Result
{"jsonrpc": "2.0","result": 0,"id": 1}
```
##### Notification Format:
```bash
{"jsonrpc": "2.0","method": "signatureNotification", "params": {"result": "Confirmed","subscription":0}}
```
---
### signatureUnsubscribe
Unsubscribe from account userdata change notifications
##### Parameters:
* `integer` - id of account subscription to cancel
##### Results:
* `bool` - unsubscribe success message
##### Example:
```bash
// Request
{"jsonrpc":"2.0", "id":1, "method":"signatureUnsubscribe", "params":[0]}
// Result
{"jsonrpc": "2.0","result": true,"id": 1}
```

View File

@ -0,0 +1,52 @@
# Leader Rotation
At any given moment, a cluster expects only one fullnode to produce ledger
entries. By having only one leader at a time, all validators are able to replay
identical copies of the ledger. The drawback of only one leader at a time,
however, is that a malicious leader is cabable of censoring votes and
transactions. Since censoring cannot be distinguished from the network dropping
packets, the cluster cannot simply elect a single node to hold the leader role
indefinitely. Instead, the cluster minimizes the influence of a malcioius
leader by rotating which node takes the lead.
Each validator selects the expected leader using the same algorithm, described
below. When the validator receives a new signed ledger entry, it can be certain
that entry was produced by the expected leader.
## Leader Schedule Generation
Leader schedule is generated using a predefined seed. The process is as follows:
1. Periodically use the PoH tick height (a monotonically increasing counter) to
seed a stable pseudo-random algorithm.
2. At that height, sample the bank for all the staked accounts with leader
identities that have voted within a cluster-configured number of ticks. The
sample is called the *active set*.
3. Sort the active set by stake weight.
4. Use the random seed to select nodes weighted by stake to create a
stake-weighted ordering.
5. This ordering becomes valid after a cluster-configured number of ticks.
The seed that is selected is predictable but unbiasable. There is no grinding
attack to influence its outcome. The active set, however, can be biased by a
leader by censoring validator votes. To reduce the likelihood of censorship,
the active set is sampled many slots in advance, such that votes will have been
collected by multiple leaders. If even one node is honest, the malicious
leaders will not be able to use censorship to influence the leader schedule.
## Appending Entries
The lifetime of a leader schedule is called an *epoch*. The epoch is split into
*slots*, where each slot has a duration of `T` PoH ticks.
A leader transmits entries during its slot. After `T` ticks, all the
validators switch to the next scheduled leader. Validators must ignore entries
sent outside a leader's assigned slot.
All `T` ticks must be observed by the next leader for it to build its own
entries on. If entries are not observed (leader is down) or entries are invalid
(leader is buggy or malicious), the next leader must produce ticks to fill the
previous leader's slot. Note that the next leader should do repair requests in
parallel, and postpone sending ticks until it is confident other validators
also failed to observe the previous leader's entries. If a leader incorrectly
builds on its own ticks, the leader following it must replace all its ticks.

View File

@ -0,0 +1,266 @@
# Ledger Replication
At full capacity on a 1gbps network solana will generate 4 petabytes of data
per year. To prevent the network from centralizing around full nodes that have
to store the full data set this protocol proposes a way for mining nodes to
provide storage capacity for pieces of the network.
The basic idea to Proof of Replication is encrypting a dataset with a public
symmetric key using CBC encryption, then hash the encrypted dataset. The main
problem with the naive approach is that a dishonest storage node can stream the
encryption and delete the data as its hashed. The simple solution is to force
the hash to be done on the reverse of the encryption, or perhaps with a random
order. This ensures that all the data is present during the generation of the
proof and it also requires the validator to have the entirety of the encrypted
data present for verification of every proof of every identity. So the space
required to validate is `number_of_proofs * data_size`
## Terminology
#### replicator
Storage mining client, stores some part of the ledger enumerated in blocks and
submits storage proofs to the chain. Not a full-node.
#### ledger segment
Portion of the ledger which is downloaded by the replicator where storage proof
data is derived.
#### CBC block
Smallest encrypted chunk of ledger, an encrypted ledger segment would be made of
many CBC blocks. `ledger_segment_size / cbc_block_size` to be exact.
#### storage proof
A set of sha hash state which is constructed by sampling the encrypted version
of the stored ledger segment at certain offsets.
#### fake storage proof
A proof which has the same format as a storage proof, but the sha state is
actually from hashing a known ledger value which the storage client can reveal
and is also easily verifiable by the network on-chain.
#### storage proof confirmation
A transaction by a validator which indicates the set of real and fake proofs
submitted by a storage miner. The transaction would contain a list of proof
hash values and a bit which says if this hash is valid or fake.
#### storage proof challenge
A transaction from a replicator that verifiably proves that a validator
confirmed a fake proof.
#### storage proof claim
A transaction from a validator which is after the timeout period given from the
storage proof confirmation and which no successful challenges have been
observed which rewards the parties of the storage proofs and confirmations.
#### storage validation capacity
The number of keys and samples that a validator can verify each storage epoch.
## Optimization with PoH
Our improvement on this approach is to randomly sample the encrypted segments
faster than it takes to encrypt, and record the hash of those samples into the
PoH ledger. Thus the segments stay in the exact same order for every PoRep and
verification can stream the data and verify all the proofs in a single batch.
This way we can verify multiple proofs concurrently, each one on its own CUDA
core. The total space required for verification is `1_ledger_segment +
2_cbc_blocks * number_of_identities` with core count of equal to
`number_of_identities`. We use a 64-byte chacha CBC block size.
## Network
Validators for PoRep are the same validators that are verifying transactions.
They have some stake that they have put up as collateral that ensures that
their work is honest. If you can prove that a validator verified a fake PoRep,
then the validators stake can be slashed.
Replicators are specialized *light clients*. They download a part of the ledger
and store it, and provide PoReps of storing the ledger. For each verified PoRep
replicators earn a reward of sol from the mining pool.
## Constraints
We have the following constraints:
* Verification requires generating the CBC blocks. That requires space of 2
blocks per identity, and 1 CUDA core per identity for the same dataset. So as
many identities at once should be batched with as many proofs for those
identities verified concurrently for the same dataset.
* Validators will randomly sample the set of storage proofs to the set that
they can handle, and only the creators of those chosen proofs will be
rewarded. The validator can run a benchmark whenever its hardware configuration
changes to determine what rate it can validate storage proofs.
## Validation and Replication Protocol
### Constants
1. NUM\_STORAGE\_ENTRIES: Number of entries in a segment of ledger data. The
unit of storage for a replicator.
2. NUM\_KEY\_ROTATION\_TICKS: Number of ticks to save a PoH value and cause a
key generation for the section of ledger just generated and the rotation of
another key in the set.
3. NUM\_STORAGE\_PROOFS: Number of storage proofs required for a storage proof
claim to be successfully rewarded.
4. RATIO\_OF\_FAKE\_PROOFS: Ratio of fake proofs to real proofs that a storage
mining proof claim has to contain to be valid for a reward.
5. NUM\_STORAGE\_SAMPLES: Number of samples required for a storage mining
proof.
6. NUM\_CHACHA\_ROUNDS: Number of encryption rounds performed to generate
encrypted state.
### Validator behavior
1. Validator joins the network and submits a storage validation capacity
transaction which tells the network how many proofs it can process in a given
period defined by NUM\_KEY\_ROTATION\_TICKS.
2. Every NUM\_KEY\_ROTATION\_TICKS the validator stores the PoH value at that
height.
3. Every NUM\_KEY\_ROTATION\_TICKS it also validates samples received from
replicators. It signs the PoH hash at that point and uses the following
algorithm with the signature as the input:
- The low 5 bits of the first byte of the signature creates an index into
another starting byte of the signature.
- The validator then looks at the set of storage proofs where the byte of
the proof's sha state vector starting from the low byte matches exactly
with the chosen byte(s) of the signature.
- If the set of proofs is larger than the validator can handle, then it
increases to matching 2 bytes in the signature.
- Validator continues to increase the number of matching bytes until a
workable set is found.
- It then creates a mask of valid proofs and fake proofs and sends it to
the leader. This is a storage proof confirmation transaction.
4. The storage proof confirmation transaction is integrated into the ledger.
5. After a lockout period of NUM\_SECONDS\_STORAGE\_LOCKOUT seconds, the
validator then submits a storage proof claim transaction which then causes the
distribution of the storage reward if no challenges were seen for the proof to
the validators and replicators party to the proofs.
6. Validator responds to RPC interfaces for what the last storage epoch PoH
value is and its entry\_height.
### Replicator behavior
1. Since a replicator is somewhat of a light client and not downloading all the
ledger data, they have to rely on other full nodes (validators) for
information. Any given validator may or may not be malicious and give incorrect
information, although there are not any obvious attack vectors that this could
accomplish besides having the replicator do extra wasted work. For many of the
operations there are number of options depending on how paranoid a replicator
is:
- (a) replicator can ask a validator
- (b) replicator can ask multiple validators
- (c) replicator can subscribe to the full transaction stream and generate
the information itself
- (d) replicator can subscribe to an abbreviated transaction stream to
generate the information itself
2. A replicator obtains the PoH hash corresponding to the last key rotation
along with its entry\_height.
3. The replicator signs the PoH hash with its keypair. That signature is the
seed used to pick the segment to replicate and also the encryption key. The
replicator mods the signature with the entry\_height to get which segment to
replicate.
4. The replicator retrives the ledger by asking peer validators and
replicators. See 6.5.
5. The replicator then encrypts that segment with the key with chacha algorithm
in CBC mode with NUM\_CHACHA\_ROUNDS of encryption.
6. The replicator initializes a chacha rng with the signature from step 2 as
the seed.
7. The replicator generates NUM\_STORAGE\_SAMPLES samples in the range of the
entry size and samples the encrypted segment with sha256 for 32-bytes at each
offset value. Sampling the state should be faster than generating the encrypted
segment.
8. The replicator sends a PoRep proof transaction which contains its sha state
at the end of the sampling operation, its seed and the samples it used to the
current leader and it is put onto the ledger.
9. The replicator then generates another set of offsets which it submits a fake
proof with an incorrect sha state. It can be proven to be fake by providing the
seed for the hash result.
- A fake proof should consist of a replicator hash of a signature of a PoH
value. That way when the replicator reveals the fake proof, it can be
verified on chain.
10. The replicator monitors the ledger, if it sees a fake proof integrated, it
creates a challenge transaction and submits it to the current leader. The
transacation proves the validator incorrectly validated a fake storage proof.
The replicator is rewarded and the validator's staking balance is slashed or
frozen.
### Finding who has a given block of ledger
1. Validators monitor the transaction stream for storage mining proofs, and
keep a mapping of ledger segments by entry\_height to public keys. When it sees
a storage mining proof it updates this mapping and provides an RPC interface
which takes an entry\_height and hands back a list of public keys. The client
then looks up in their cluster\_info table to see which network address that
corresponds to and sends a repair request to retrieve the necessary blocks of
ledger.
2. Validators would need to prune this list which it could do by periodically
looking at the oldest entries in its mappings and doing a network query to see
if the storage host is still serving the first entry.
## Sybil attacks
For any random seed, we force everyone to use a signature that is derived from
a PoH hash. Everyone must use the same count, so the same PoH hash is signed by
every participant. The signatures are then each cryptographically tied to the
keypair, which prevents a leader from grinding on the resulting value for more
than 1 identity.
Since there are many more client identities then encryption identities, we need
to split the reward for multiple clients, and prevent Sybil attacks from
generating many clients to acquire the same block of data. To remain BFT we
want to avoid a single human entity from storing all the replications of a
single chunk of the ledger.
Our solution to this is to force the clients to continue using the same
identity. If the first round is used to acquire the same block for many client
identities, the second round for the same client identities will force a
redistribution of the signatures, and therefore PoRep identities and blocks.
Thus to get a reward for replicators need to store the first block for free and
the network can reward long lived client identities more than new ones.
## Validator attacks
- If a validator approves fake proofs, replicator can easily out them by
showing the initial state for the hash.
- If a validator marks real proofs as fake, no on-chain computation can be done
to distinguish who is correct. Rewards would have to rely on the results from
multiple validators in a stake-weighted fashion to catch bad actors and
replicators from being locked out of the network.
- Validator stealing mining proof results for itself. The proofs are derived
from a signature from a replicator, since the validator does not know the
private key used to generate the encryption key, it cannot be the generator of
the proof.
## Reward incentives
Fake proofs are easy to generate but difficult to verify. For this reason,
PoRep proof transactions generated by replicators may require a higher fee than
a normal transaction to represent the computational cost required by
validators.
Some percentage of fake proofs are also necessary to receive a reward from
storage mining.
## Notes
* We can reduce the costs of verification of PoRep by using PoH, and actually
make it feasible to verify a large number of proofs for a global dataset.
* We can eliminate grinding by forcing everyone to sign the same PoH hash and
use the signatures as the seed
* The game between validators and replicators is over random blocks and random
encryption identities and random data samples. The goal of randomization is
to prevent colluding groups from having overlap on data or validation.
* Replicator clients fish for lazy validators by submitting fake proofs that
they can prove are fake.
* To defend against Sybil client identities that try to store the same block we
force the clients to store for multiple rounds before receiving a reward.
* Validators should also get rewarded for validating submitted storage proofs
as incentive for storing the ledger. They can only validate proofs if they
are storing that slice of the ledger.

Some files were not shown because too many files have changed in this diff Show More