Compare commits

..

58 Commits

Author SHA1 Message Date
Michael Legleux
2751498cea Set version to 2.4.0-rc2 2025-02-24 20:52:41 +00:00
Ed Hennis
c4d9635f22 Add logging and improve counting of amendment votes from UNL (#5173)
* Add logging for amendment voting decision process
* When counting "received validations" to determine quorum, count the number of validators actually voting, not the total number of possible votes.
2025-02-24 20:52:41 +00:00
Ed Hennis
8f6208eb66 Revert "Reduce duplicate peer traffic for ledger data (#5126)" (#5300)
This reverts commit dd5e6559dd. It has
introduced a regression causing slow close times and syncing issues.
A fix will be attempted later.
2025-02-24 20:52:41 +00:00
Bart
efa1255386 docs: Revert peer port to 51235 (#5299)
Reverts the [port_peer] back to the legacy port 51235 rather than to the default port 2459, to avoid potentially inconveniencing existing operators.
2025-02-24 20:52:41 +00:00
Michael Legleux
6e198078a9 Set version to 2.4.0-rc1 2025-02-24 20:52:41 +00:00
Olek
8d55192f0c fix: Switch Permissioned Domain to Supported::yes (#5287)
Switch Permissioned Domain feature's supported flag from Supported::no to Supported::yes for it to be votable.
2025-02-24 20:52:41 +00:00
Bart
2de83f36ae docs: Clarifies default port of hosts (#5290)
The current comment in the example cfg file incorrectly mentions both "may" and "must". This change fixes this comment to clarify that the default port of hosts is 2459 and that specifying it is therefore optional. It further sets the default port to 2459 instead of the legacy 51235.
2025-02-24 20:52:41 +00:00
Mark Travis
87ba6ba2d8 Log proposals and validations (#5291)
Adds detailed log messages for each validation and proposal received from the network.
2025-02-24 20:52:41 +00:00
Ed Hennis
51f427c4e8 Reduce duplicate peer traffic for ledger data (#5126)
- Drop duplicate outgoing TMGetLedger messages per peer
  - Allow a retry after 30s in case of peer or network congestion.
  - Addresses RIPD-1870
  - (Changes levelization. That is not desirable, and will need to be fixed.)
- Drop duplicate incoming TMGetLedger messages per peer
  - Allow a retry after 15s in case of peer or network congestion.
  - The requestCookie is ignored when computing the hash, thus increasing
    the chances of detecting duplicate messages.
  - With duplicate messages, keep track of the different requestCookies
    (or lack of cookie). When work is finally done for a given request,
    send the response to all the peers that are waiting on the request,
    sending one message per peer, including all the cookies and
    a "directResponse" flag indicating the data is intended for the
    sender, too.
  - Addresses RIPD-1871
- Drop duplicate incoming TMLedgerData messages
  - Addresses RIPD-1869
- Improve logging related to ledger acquisition
- Class "CanProcess" to keep track of processing of distinct items

---------

Co-authored-by: Valentin Balaschenko <13349202+vlntb@users.noreply.github.com>
2025-02-24 20:52:41 +00:00
Bart
6a3c6dd8a3 Support canonical ledger entry names (#5271)
This change enhances the filtering in the ledger, ledger_data, and account_objects methods by also supporting filtering by the canonical name of the LedgerEntryType using case-insensitive matching.
2025-02-24 20:52:41 +00:00
Darius Tumas
f82b56605e chore: Update XRPL Foundation public key (#5289)
Following the XRPL Foundation UNL migration a new set of keys was generated.
2025-02-24 20:52:41 +00:00
Ed Hennis
aa0aa44cf4 refactor: Change recursive_mutex to mutex in DatabaseRotatingImp (#5276)
Rewrites the code so that the lock is not held during the callback. Instead it locks twice, once before, and once after. This is safe due to the structure of the code, but is checked after the second lock. This allows mutex_ to be changed back to a regular mutex.
2025-02-24 20:52:41 +00:00
Bart
4a1ef3d42f fix: Replace charge() by fee_.update() in OnMessage functions (#5269)
In PeerImpl.cpp, if the function is a message handler (onMessage) or called directly from a message handler, then it should use fee_, since when the handler returns (OnMessageEnd) then the charge function is called. If the function is not a message handler, such as a job queue item, it should remain charge.
2025-02-24 20:52:41 +00:00
Elliot Lee
87b13184b1 docs: ensure build_type and CMAKE_BUILD_TYPE match (#5274) 2025-02-24 20:52:41 +00:00
code0xff
cdae348a4b chore: Fix small typos in protocol files (#5279) 2025-02-24 20:52:41 +00:00
Ed Hennis
b7057b5fba chore: Rename missing-commits job, and combine nix job files (#5268)
- Rename the job in missing-commits.yml from "check" to "up_to_date",
  because other jobs named "check" prevent merges, but this one should
  not prevent merges. How else are branches going to get caught up?
- Move the job in instrumentation.yml to nix.yml, but keep it entirely
  independent.
2025-02-24 20:52:41 +00:00
Ed Hennis
e96fb0ce55 docs: Add a summary of the git commit message rules (#5283) 2025-02-24 20:52:41 +00:00
Olek
359565b377 fix: Amendment to add transaction flag checking functionality for Credentials (#5250)
CredentialCreate / CredentialAccept / CredentialDelete transactions will check sfFlags field in preflight() when the amendment is enabled.
2025-02-24 20:52:41 +00:00
Donovan Hide
0376c7af60 fix: Omit superfluous setCurrentThreadName call in GRPCServer.cpp (#5280) 2025-02-24 20:52:41 +00:00
Bronek Kozicki
f5dac74afb fix: Do not allow creating Permissioned Domains if credentials are not enabled (#5275)
If the permissioned domains amendment XLS-80 is enabled before credentials XLS-70, then the permissioned domain users will not be able to match any credentials. The changes here prevent the creation of any permissioned domain objects if credentials are not enabled.
2025-02-24 20:52:41 +00:00
Mayukha Vadari
c92bd4a817 fix: issues in simulate RPC (#5265)
Make `simulate` RPC easier to use:
* Prevent the use of `seed`, `secret`, `seed_hex`, and `passphrase` fields (to avoid confusing with the signing methods).
* Add autofilling of the `NetworkID` field.
2025-02-24 20:52:41 +00:00
Bart
6b16569273 Updates Conan dependencies (#5256)
This PR updates several Conan dependencies:
* boost
* date
* libarchive
* libmysqlclient
* libpq
* lz4
* onetbb
* openssl
* sqlite3
* zlib
* zstd
2025-02-24 20:52:41 +00:00
Shawn Xie
7fdfbc7b87 Amendment fixFrozenLPTokenTransfer (#5227)
Prohibits LPToken holders from sending LPToken to others if they have been frozen by one of the assets in AMM pool.
2025-02-24 20:52:41 +00:00
Ed Hennis
463e62e64b Improve git commit hash lookup (#5225)
- Also get the branch name.
- Use rev-parse instead of describe to get a clean hash.
- Return the git hash and branch name in server_info for admin
  connections.
- Include git hash and branch name on separate lines in --version.
2025-02-24 20:52:41 +00:00
Vlad
ffc88d703b Add deep freeze feature (XLS-77d) (#5187)
- spec: XRPLF/XRPL-Standards#220
- amendment: "DeepFreeze"
- implemented deep freeze spec to allow token issuers to prevent currency holders from being able to acquire more of these tokens.
- in combination with normal freeze, deep freeze effectively prevents any balance trust line balance change of a currency holder (except direct issuer <-> holder payments).
- added 2 new invariant checks to verify that deep freeze cannot be enacted without normal freeze and transfer is not frozen.
- made some fixes to existing freeze handling.

Co-authored-by: Ed Hennis <ed@ripple.com>
Co-authored-by: Howard Hinnant <howard.hinnant@gmail.com>
2025-02-24 20:52:41 +00:00
Ed Hennis
0613024aac Set version to 2.4.0-b3 2025-02-24 20:52:41 +00:00
Mayukha Vadari
b991d24070 Add RPC "simulate" to execute a dry run of a transaction (#5069)
- Spec: https://github.com/XRPLF/XRPL-Standards/tree/master/XLS-0069d-simulate
- Also update signing methods to autofill fees better and properly handle transactions that require a non-standard fee.
2025-02-24 20:52:41 +00:00
Olek
f3d39ec680 Fix CI unit tests (#5196)
- Add retries for rpc client
- Add dynamic port allocation for rpc servers
2025-02-24 20:52:41 +00:00
Michael Legleux
1d0acf1490 Update secp256k1 library to 0.6.0 (#5254) 2025-02-24 20:52:41 +00:00
Ed Hennis
a331a172c6 chore: Update Visual Studio CI to VS 2022, and add VS Debug builds (#5240)
* Debug builds do not run tests, because they take too long.
2025-02-24 20:52:40 +00:00
Bronek Kozicki
1a341cb9cb Add [validator_list_threshold] to validators.txt to improve UNL security (#5112) 2025-02-24 20:52:40 +00:00
Bronek Kozicki
1b043fcc7b Switch from assert to XRPL_ASSERT (#5245) 2025-02-24 20:52:40 +00:00
tequ
e742a6f9fc Add missing space character to a log message (#5251) 2025-02-24 20:52:40 +00:00
Bronek Kozicki
4d837884cf Cleanup API-CHANGELOG.md (#5207) 2025-02-24 20:52:40 +00:00
Ed Hennis
7f55b38377 test: Unit tests to recreate invalid index logic error (#5242)
* One hits the global cache, one does not.
* Also some extra checking.

Co-authored-by: Bronek Kozicki <brok@incorrekt.com>
2025-02-24 20:52:40 +00:00
Ed Hennis
cf96bd6058 Update branch management and merge / release processes (#5215)
* Has more steps, but allows merges to develop to continue when a
  beta / RC is pending, increasing developer velocity.
* Add a CI job to check that no reverse merges have been missed.
* Add some useful scripts in bin/git:
  * Set up upstreams as expected for safer pushes
  * Squash a bunch of branches
  * Set the version number
2025-02-24 20:52:40 +00:00
Sergey Kuznetsov
1718790820 fix: Error consistency in LedgerEntry::parsePermissionedDomains() (#5252)
Update errors for parsing permissioned domains in the LedgerEntry handler to make them consistent with other parsers.
2025-02-24 20:52:40 +00:00
Ed Hennis
7c8ead9c3c Set version to 2.4.0-b2 2025-02-24 20:52:40 +00:00
Ed Hennis
6bc78370ba fix: Use consistent CMake settings for all modules (#5228)
* Resolves an issue introduced in #5111, which inadvertently removed the
  -Wno-maybe-uninitialized compiler option from some xrpl.libxrpl
  modules. This resulted in new "may be used uninitialized" build
  warnings, first noticed in the "protocol" module. When compiling with
  derr=TRUE, those warnings became errors, which made the build fail.
* Github CI actions will build with the assert and werr options turned
  on. This will cause CI jobs to fail if a developer introduces a new
  compiler warning, or causes an assert to fail in release builds.
* Includes the OS and compiler version in the linux dependencies jobs in
  the "check environment" step.
* Translates the `unity` build option into `CMAKE_UNITY_BUILD` setting.
2025-02-24 20:52:40 +00:00
Valentin Balaschenko
afc9a42ff5 Fix levelization script to ignore commented includes (#5194)
Check to ignore single-line comments during dependency analysis.
2025-02-24 20:52:40 +00:00
tequ
090393930f Fix the flag processing of NFTokenModify (#5246)
Adds checks for invalid flags.
2025-02-24 20:52:40 +00:00
Mayukha Vadari
c36daa7857 Fix failing assert in connect RPC (#5235) 2025-02-24 20:52:40 +00:00
Olek
9069e430ad Permissioned Domains (XLS-80d) (#5161) 2025-02-24 20:52:40 +00:00
Mayukha Vadari
6c66b7448d chore: update deprecated Github Actions (#5241) 2025-02-24 20:52:40 +00:00
tequ
1bb3c9db58 XLS-46: DynamicNFT (#5048)
This Amendment adds functionality to update the URI of NFToken objects as described in the XLS-46d: Dynamic Non Fungible Tokens (dNFTs) spec.
2025-02-24 20:52:40 +00:00
rrmanukyan
63cbd84487 chore: add macos dependency installation (#5233)
* python (3.13) and cmake (latest)
2025-02-24 20:52:40 +00:00
Shawn Xie
f6d74d8c29 prefix Uint384 and Uint512 with Hash in server_definitions (#5231) 2025-02-24 20:52:40 +00:00
Mayukha Vadari
117306f8ae refactor: add rpcName to LEDGER_ENTRY macro (#5202)
The LEDGER_ENTRY macro now takes an additional parameter, which makes it easier to avoid missing including the new field in jss.h and to the list of account_objects/ledger_data filters.
2025-02-24 20:52:40 +00:00
Ed Hennis
7717f08a05 Set version to 2.4.0-b1 2025-02-24 20:52:40 +00:00
Michael Legleux
0633844187 fix: Add header for set_difference (#5197)
Fix `error C2039: 'set_difference': is not a member of 'std'`
2025-02-24 20:51:54 +00:00
Mayukha Vadari
dc0608825b fix: allow overlapping types in Expected (#5218)
For example, Expected<std::uint32_t, Json::Value>, will now build even though there is animplicit conversion from unsigned int to Json::Value.
2025-02-24 20:51:54 +00:00
Gregory Tsipenyuk
49f35dc4d5 Add MPTIssue to STIssue (#5200)
Replace Issue in STIssue with Asset. STIssue with MPTIssue is only used in MPT tests.
Will be used in Vault and in transactions with STIssue fields once MPT is integrated into DEX.
2025-02-24 20:51:54 +00:00
Bronek Kozicki
8d78701c96 Antithesis instrumentation improvements (#5213)
* Rename ASSERT to XRPL_ASSERT
* Upgrade to Anthithesis SDK 0.4.4, and use new 0.4.4 features
  * automatic cast to bool, like assert
* Add instrumentation workflow to verify build with instrumentation enabled
2025-02-24 20:51:54 +00:00
John Freeman
a1e3009bff Enforce levelization in libxrpl with CMake (#5111)
Adds two CMake functions:

* add_module(library subdirectory): Declares an OBJECT "library" (a CMake abstraction for a collection of object files) with sources from the given subdirectory of the given library, representing a module. Isolates the module's headers by creating a subdirectory in the build directory, e.g. .build/tmp123, that contains just a symlink, e.g. .build/tmp123/basics, to the module's header directory, e.g. include/xrpl/basics, in the source directory, and putting .build/tmp123 (but not include/xrpl) on the include path of the module sources. This prevents the module sources from including headers not explicitly linked to the module in CMake with target_link_libraries.
* target_link_modules(library scope modules...): Links the library target to each of the module targets, and removes their sources from its source list (so they are not compiled and linked twice).

Uses these functions to separate and explicitly link modules in libxrpl:

    Level 01: beast
    Level 02: basics
    Level 03: json, crypto
    Level 04: protocol
    Level 05: resource, server
2025-02-24 20:51:54 +00:00
Mayukha Vadari
8632aa8d3d refactor: clean up LedgerEntry.cpp (#5199)
Refactors LedgerEntry to make it easier to read and understand.
2025-02-24 20:51:54 +00:00
Ed Hennis
984e489a56 test: Add more test cases for Base58 parser (#5174)
---------
Co-authored-by: John Freeman <jfreeman08@gmail.com>
2025-02-24 20:51:54 +00:00
Ed Hennis
adc30e6fff test: Check for some unlikely null dereferences in tests (#5004) 2025-02-24 20:51:54 +00:00
Bronek Kozicki
6704194941 Add Antithesis intrumentation (#5042)
* Copy Antithesis SDK version 0.4.0 to directory external/
* Add build option `voidstar` to enable instrumentation with Antithesis SDK
* Define instrumentation macros ASSERT and UNREACHABLE in terms of regular C assert
* Replace asserts with named ASSERT or UNREACHABLE
* Add UNREACHABLE to LogicError
* Document instrumentation macros in CONTRIBUTING.md
2025-02-24 20:51:54 +00:00
1133 changed files with 10019 additions and 47341 deletions

View File

@@ -44,7 +44,6 @@ DerivePointerAlignment: false
DisableFormat: false
ExperimentalAutoDetectBinPacking: false
ForEachMacros: [ Q_FOREACH, BOOST_FOREACH ]
IncludeBlocks: Regroup
IncludeCategories:
- Regex: '^<(test)/'
Priority: 0
@@ -54,12 +53,8 @@ IncludeCategories:
Priority: 2
- Regex: '^<(boost)/'
Priority: 3
- Regex: '^.*/'
Priority: 4
- Regex: '^.*\.h'
Priority: 5
- Regex: '.*'
Priority: 6
Priority: 4
IncludeIsMainRegex: '$'
IndentCaseLabels: true
IndentFunctionDeclarationAfterType: false
@@ -94,4 +89,3 @@ SpacesInSquareBrackets: false
Standard: Cpp11
TabWidth: 8
UseTab: Never
QualifierAlignment: Right

View File

@@ -7,13 +7,13 @@ comment:
show_carryforward_flags: false
coverage:
range: "70..85"
range: "60..80"
precision: 1
round: nearest
status:
project:
default:
target: 75%
target: 60%
threshold: 2%
patch:
default:

8
.github/CODEOWNERS vendored
View File

@@ -1,8 +0,0 @@
# Allow anyone to review any change by default.
*
# Require the rpc-reviewers team to review changes to the rpc code.
include/xrpl/protocol/ @xrplf/rpc-reviewers
src/libxrpl/protocol/ @xrplf/rpc-reviewers
src/xrpld/rpc/ @xrplf/rpc-reviewers
src/xrpld/app/misc/ @xrplf/rpc-reviewers

View File

@@ -20,6 +20,8 @@ runs:
${{ inputs.generator && format('-G "{0}"', inputs.generator) || '' }} \
-DCMAKE_TOOLCHAIN_FILE:FILEPATH=build/generators/conan_toolchain.cmake \
-DCMAKE_BUILD_TYPE=${{ inputs.configuration }} \
-Dassert=TRUE \
-Dwerr=TRUE \
-Dtests=TRUE \
-Dxrpld=TRUE \
${{ inputs.cmake-args }} \

View File

@@ -14,7 +14,7 @@ runs:
run: |
conan config set general.revisions_enabled=1
conan export external/snappy snappy/1.1.10@
conan export external/rocksdb rocksdb/9.7.3@
conan export external/rocksdb rocksdb/6.29.5@
conan export external/soci soci/4.0.3@
conan export external/nudb nudb/2.0.8@
- name: add Ripple Conan remote
@@ -55,3 +55,7 @@ runs:
--options xrpld=True \
--settings build_type=${{ inputs.configuration }} \
..
- name: upload dependencies to remote
if: (steps.binaries.outputs.missing != '[]') && (steps.remote.outputs.outcome == 'success')
shell: bash
run: conan upload --remote ripple '*' --all --parallel --confirm

View File

@@ -1,13 +1,9 @@
name: clang-format
on:
push:
pull_request:
types: [opened, reopened, synchronize, ready_for_review]
on: [push, pull_request]
jobs:
check:
if: ${{ github.event_name == 'push' || github.event.pull_request.draft != true || contains(github.event.pull_request.labels.*.name, 'DraftRunCI') }}
runs-on: ubuntu-24.04
env:
CLANG_VERSION: 18
@@ -24,7 +20,7 @@ jobs:
sudo apt-get update
sudo apt-get install clang-format-${CLANG_VERSION}
- name: Format first-party sources
run: find include src tests -type f \( -name '*.cpp' -o -name '*.hpp' -o -name '*.h' -o -name '*.ipp' \) -exec clang-format-${CLANG_VERSION} -i {} +
run: find include src -type f \( -name '*.cpp' -o -name '*.hpp' -o -name '*.h' -o -name '*.ipp' \) -exec clang-format-${CLANG_VERSION} -i {} +
- name: Check for differences
id: assert
run: |

View File

@@ -10,11 +10,11 @@ concurrency:
cancel-in-progress: true
jobs:
documentation:
job:
runs-on: ubuntu-latest
permissions:
contents: write
container: ghcr.io/xrplf/rippled-build-ubuntu:aaf5e3e
container: rippleci/rippled-build-ubuntu:aaf5e3e
steps:
- name: checkout
uses: actions/checkout@v4

View File

@@ -1,13 +1,9 @@
name: levelization
on:
push:
pull_request:
types: [opened, reopened, synchronize, ready_for_review]
on: [push, pull_request]
jobs:
check:
if: ${{ github.event_name == 'push' || github.event.pull_request.draft != true || contains(github.event.pull_request.labels.*.name, 'DraftRunCI') }}
runs-on: ubuntu-latest
env:
CLANG_VERSION: 10

View File

@@ -1,6 +1,6 @@
name: Check libXRPL compatibility with Clio
env:
CONAN_URL: http://18.143.149.228:8081/artifactory/api/conan/dev
CONAN_URL: http://18.143.149.228:8081/artifactory/api/conan/conan-non-prod
CONAN_LOGIN_USERNAME_RIPPLE: ${{ secrets.CONAN_USERNAME }}
CONAN_PASSWORD_RIPPLE: ${{ secrets.CONAN_TOKEN }}
on:
@@ -8,21 +8,19 @@ on:
paths:
- 'src/libxrpl/protocol/BuildInfo.cpp'
- '.github/workflows/libxrpl.yml'
types: [opened, reopened, synchronize, ready_for_review]
concurrency:
group: ${{ github.workflow }}-${{ github.ref }}
cancel-in-progress: true
jobs:
publish:
if: ${{ github.event_name == 'push' || github.event.pull_request.draft != true || contains(github.event.pull_request.labels.*.name, 'DraftRunCI') }}
name: Publish libXRPL
outputs:
outcome: ${{ steps.upload.outputs.outcome }}
version: ${{ steps.version.outputs.version }}
channel: ${{ steps.channel.outputs.channel }}
runs-on: [self-hosted, heavy]
container: ghcr.io/xrplf/rippled-build-ubuntu:aaf5e3e
container: rippleci/rippled-build-ubuntu:aaf5e3e
steps:
- name: Wait for essential checks to succeed
uses: lewagon/wait-on-check-action@v1.3.4
@@ -87,5 +85,4 @@ jobs:
run: |
gh api --method POST -H "Accept: application/vnd.github+json" -H "X-GitHub-Api-Version: 2022-11-28" \
/repos/xrplf/clio/dispatches -f "event_type=check_libxrpl" \
-F "client_payload[version]=${{ needs.publish.outputs.version }}@${{ needs.publish.outputs.channel }}" \
-F "client_payload[pr]=${{ github.event.pull_request.number }}"
-F "client_payload[version]=${{ needs.publish.outputs.version }}@${{ needs.publish.outputs.channel }}"

View File

@@ -1,7 +1,6 @@
name: macos
on:
pull_request:
types: [opened, reopened, synchronize, ready_for_review]
push:
# If the branches list is ever changed, be sure to change it on all
# build/test jobs (nix, macos, windows, instrumentation)
@@ -19,7 +18,6 @@ concurrency:
jobs:
test:
if: ${{ github.event_name == 'push' || github.event.pull_request.draft != true || contains(github.event.pull_request.labels.*.name, 'DraftRunCI') }}
strategy:
matrix:
platform:
@@ -71,13 +69,11 @@ jobs:
nproc --version
echo -n "nproc returns: "
nproc
system_profiler SPHardwareDataType
sysctl -n hw.logicalcpu
clang --version
- name: configure Conan
run : |
conan profile new default --detect || true
conan profile update settings.compiler.cppstd=20 default
conan profile update 'conf.tools.build:cxxflags+=["-DBOOST_ASIO_DISABLE_CONCEPTS"]' default
- name: build dependencies
uses: ./.github/actions/dependencies
env:
@@ -91,7 +87,7 @@ jobs:
with:
generator: ${{ matrix.generator }}
configuration: ${{ matrix.configuration }}
cmake-args: "-Dassert=TRUE -Dwerr=TRUE ${{ matrix.cmake-args }}"
cmake-args: ${{ matrix.cmake-args }}
- name: test
run: |
n=$(nproc)

View File

@@ -1,7 +1,6 @@
name: nix
on:
pull_request:
types: [opened, reopened, synchronize, ready_for_review]
push:
# If the branches list is ever changed, be sure to change it on all
# build/test jobs (nix, macos, windows)
@@ -40,7 +39,6 @@ concurrency:
jobs:
dependencies:
if: ${{ github.event_name == 'push' || github.event.pull_request.draft != true || contains(github.event.pull_request.labels.*.name, 'DraftRunCI') }}
strategy:
fail-fast: false
matrix:
@@ -64,7 +62,7 @@ jobs:
cc: /usr/bin/clang-14
cxx: /usr/bin/clang++-14
runs-on: [self-hosted, heavy]
container: ghcr.io/xrplf/rippled-build-ubuntu:aaf5e3e
container: rippleci/rippled-build-ubuntu:aaf5e3e
env:
build_dir: .build
steps:
@@ -126,7 +124,7 @@ jobs:
- "-Dunity=ON"
needs: dependencies
runs-on: [self-hosted, heavy]
container: ghcr.io/xrplf/rippled-build-ubuntu:aaf5e3e
container: rippleci/rippled-build-ubuntu:aaf5e3e
env:
build_dir: .build
steps:
@@ -160,61 +158,7 @@ jobs:
with:
generator: Ninja
configuration: ${{ matrix.configuration }}
cmake-args: "-Dassert=TRUE -Dwerr=TRUE ${{ matrix.cmake-args }}"
- name: test
run: |
${build_dir}/rippled --unittest --unittest-jobs $(nproc)
reference-fee-test:
strategy:
fail-fast: false
matrix:
platform:
- linux
compiler:
- gcc
configuration:
- Debug
cmake-args:
- "-DUNIT_TEST_REFERENCE_FEE=200"
- "-DUNIT_TEST_REFERENCE_FEE=1000"
needs: dependencies
runs-on: [self-hosted, heavy]
container: ghcr.io/xrplf/rippled-build-ubuntu:aaf5e3e
env:
build_dir: .build
steps:
- name: upgrade conan
run: |
pip install --upgrade "conan<2"
- name: download cache
uses: actions/download-artifact@v4
with:
name: ${{ matrix.platform }}-${{ matrix.compiler }}-${{ matrix.configuration }}
- name: extract cache
run: |
mkdir -p ~/.conan
tar -xzf conan.tar -C ~/.conan
- name: check environment
run: |
env | sort
echo ${PATH} | tr ':' '\n'
conan --version
cmake --version
- name: checkout
uses: actions/checkout@v4
- name: dependencies
uses: ./.github/actions/dependencies
env:
CONAN_URL: http://18.143.149.228:8081/artifactory/api/conan/conan-non-prod
with:
configuration: ${{ matrix.configuration }}
- name: build
uses: ./.github/actions/build
with:
generator: Ninja
configuration: ${{ matrix.configuration }}
cmake-args: "-Dassert=TRUE -Dwerr=TRUE ${{ matrix.cmake-args }}"
cmake-args: ${{ matrix.cmake-args }}
- name: test
run: |
${build_dir}/rippled --unittest --unittest-jobs $(nproc)
@@ -231,7 +175,7 @@ jobs:
- Debug
needs: dependencies
runs-on: [self-hosted, heavy]
container: ghcr.io/xrplf/rippled-build-ubuntu:aaf5e3e
container: rippleci/rippled-build-ubuntu:aaf5e3e
env:
build_dir: .build
steps:
@@ -247,7 +191,7 @@ jobs:
mkdir -p ~/.conan
tar -xzf conan.tar -C ~/.conan
- name: install gcovr
run: pip install "gcovr>=7,<9"
run: pip install "gcovr>=7,<8"
- name: check environment
run: |
echo ${PATH} | tr ':' '\n'
@@ -270,8 +214,6 @@ jobs:
generator: Ninja
configuration: ${{ matrix.configuration }}
cmake-args: >-
-Dassert=TRUE
-Dwerr=TRUE
-Dcoverage=ON
-Dcoverage_format=xml
-DCODE_COVERAGE_VERBOSE=ON
@@ -305,7 +247,7 @@ jobs:
conan:
needs: dependencies
runs-on: [self-hosted, heavy]
container: ghcr.io/xrplf/rippled-build-ubuntu:aaf5e3e
container: rippleci/rippled-build-ubuntu:aaf5e3e
env:
build_dir: .build
configuration: Release
@@ -344,7 +286,7 @@ jobs:
echo "reference=${reference}" >> "${GITHUB_ENV}"
- name: build
run: |
cd tests/conan
cd examples/example
mkdir ${build_dir}
cd ${build_dir}
conan install .. --output-folder . \
@@ -360,7 +302,6 @@ jobs:
# later
instrumentation-build:
if: ${{ github.event_name == 'push' || github.event.pull_request.draft != true || contains(github.event.pull_request.labels.*.name, 'DraftRunCI') }}
env:
CLANG_RELEASE: 16
strategy:

View File

@@ -2,7 +2,6 @@ name: windows
on:
pull_request:
types: [opened, reopened, synchronize, ready_for_review]
push:
# If the branches list is ever changed, be sure to change it on all
# build/test jobs (nix, macos, windows, instrumentation)
@@ -22,7 +21,6 @@ concurrency:
jobs:
test:
if: ${{ github.event_name == 'push' || github.event.pull_request.draft != true || contains(github.event.pull_request.labels.*.name, 'DraftRunCI') }}
strategy:
fail-fast: false
matrix:
@@ -89,7 +87,7 @@ jobs:
generator: '${{ matrix.version.generator }}'
configuration: ${{ matrix.configuration.type }}
# Hard code for now. Move to the matrix if varied options are needed
cmake-args: '-Dassert=TRUE -Dwerr=TRUE -Dreporting=OFF -Dunity=ON'
cmake-args: '-Dassert=ON -Dreporting=OFF -Dunity=ON'
cmake-target: install
- name: test
shell: bash

View File

@@ -83,17 +83,9 @@ The [commandline](https://xrpl.org/docs/references/http-websocket-apis/api-conve
The `network_id` field was added in the `server_info` response in version 1.5.0 (2019), but it is not returned in [reporting mode](https://xrpl.org/rippled-server-modes.html#reporting-mode). However, use of reporting mode is now discouraged, in favor of using [Clio](https://github.com/XRPLF/clio) instead.
## XRP Ledger server version 2.5.0
As of 2025-04-04, version 2.5.0 is in development. You can use a pre-release version by building from source or [using the `nightly` package](https://xrpl.org/docs/infrastructure/installation/install-rippled-on-ubuntu).
### Additions and bugfixes in 2.5.0
- `channel_authorize`: If `signing_support` is not enabled in the config, the RPC is disabled.
## XRP Ledger server version 2.4.0
[Version 2.4.0](https://github.com/XRPLF/rippled/releases/tag/2.4.0) was released on March 4, 2025.
As of 2025-01-28, version 2.4.0 is in development. You can use a pre-release version by building from source or [using the `nightly` package](https://xrpl.org/docs/infrastructure/installation/install-rippled-on-ubuntu).
### Additions and bugfixes in 2.4.0

View File

@@ -178,9 +178,9 @@ It does not override paths to dependencies when building with Visual Studio.
```
# Conan 1.x
conan export external/rocksdb rocksdb/9.7.3@
conan export external/rocksdb rocksdb/6.29.5@
# Conan 2.x
conan export --version 9.7.3 external/rocksdb
conan export --version 6.29.5 external/rocksdb
```
Export our [Conan recipe for SOCI](./external/soci).
@@ -288,7 +288,7 @@ It fixes some source files to add missing `#include`s.
Single-config generators:
```
cmake --build . -j $(nproc)
cmake --build .
```
Multi-config generators:

View File

@@ -14,10 +14,10 @@ Loop: xrpld.app xrpld.net
xrpld.app > xrpld.net
Loop: xrpld.app xrpld.overlay
xrpld.overlay > xrpld.app
xrpld.overlay == xrpld.app
Loop: xrpld.app xrpld.peerfinder
xrpld.peerfinder ~= xrpld.app
xrpld.app > xrpld.peerfinder
Loop: xrpld.app xrpld.rpc
xrpld.rpc > xrpld.app

View File

@@ -6,7 +6,6 @@ libxrpl.protocol > xrpl.basics
libxrpl.protocol > xrpl.json
libxrpl.protocol > xrpl.protocol
libxrpl.resource > xrpl.basics
libxrpl.resource > xrpl.json
libxrpl.resource > xrpl.resource
libxrpl.server > xrpl.basics
libxrpl.server > xrpl.json
@@ -43,7 +42,6 @@ test.consensus > xrpl.basics
test.consensus > xrpld.app
test.consensus > xrpld.consensus
test.consensus > xrpld.ledger
test.consensus > xrpl.json
test.core > test.jtx
test.core > test.toplevel
test.core > test.unit_test
@@ -60,6 +58,7 @@ test.json > test.jtx
test.json > xrpl.json
test.jtx > xrpl.basics
test.jtx > xrpld.app
test.jtx > xrpld.consensus
test.jtx > xrpld.core
test.jtx > xrpld.ledger
test.jtx > xrpld.net
@@ -159,6 +158,7 @@ xrpld.core > xrpl.basics
xrpld.core > xrpl.json
xrpld.core > xrpl.protocol
xrpld.ledger > xrpl.basics
xrpld.ledger > xrpld.core
xrpld.ledger > xrpl.json
xrpld.ledger > xrpl.protocol
xrpld.net > xrpl.basics
@@ -183,6 +183,7 @@ xrpld.peerfinder > xrpld.core
xrpld.peerfinder > xrpl.protocol
xrpld.perflog > xrpl.basics
xrpld.perflog > xrpl.json
xrpld.perflog > xrpl.protocol
xrpld.rpc > xrpl.basics
xrpld.rpc > xrpld.core
xrpld.rpc > xrpld.ledger

View File

@@ -16,18 +16,6 @@ set(CMAKE_CXX_EXTENSIONS OFF)
set(CMAKE_CXX_STANDARD 20)
set(CMAKE_CXX_STANDARD_REQUIRED ON)
if(CMAKE_CXX_COMPILER_ID MATCHES "GNU")
# GCC-specific fixes
add_compile_options(-Wno-unknown-pragmas -Wno-subobject-linkage)
# -Wno-subobject-linkage can be removed when we upgrade GCC version to at least 13.3
elseif(CMAKE_CXX_COMPILER_ID MATCHES "Clang")
# Clang-specific fixes
add_compile_options(-Wno-unknown-warning-option) # Ignore unknown warning options
elseif(MSVC)
# MSVC-specific fixes
add_compile_options(/wd4068) # Ignore unknown pragmas
endif()
# make GIT_COMMIT_HASH define available to all sources
find_package(Git)
if(Git_FOUND)

View File

@@ -1,5 +1,3 @@
[![codecov](https://codecov.io/gh/XRPLF/rippled/graph/badge.svg?token=WyFr5ajq3O)](https://codecov.io/gh/XRPLF/rippled)
# The XRP Ledger
The [XRP Ledger](https://xrpl.org/) is a decentralized cryptographic ledger powered by a network of peer-to-peer nodes. The XRP Ledger uses a novel Byzantine Fault Tolerant consensus algorithm to settle and record transactions in a secure distributed database without a central operator.

View File

@@ -5,108 +5,6 @@
This document contains the release notes for `rippled`, the reference server implementation of the XRP Ledger protocol. To learn more about how to build, run or update a `rippled` server, visit https://xrpl.org/install-rippled.html
Have new ideas? Need help with setting up your node? [Please open an issue here](https://github.com/xrplf/rippled/issues/new/choose).
## Full Changelog
### Amendments
The following amendments are open for voting with this release:
- **DynamicNFT (XLS-46)** - Adds the ability to mint mutable `NFToken` objects whose URI can be changed. ([#5048](https://github.com/XRPLF/rippled/pull/5048))
- **PermissionedDomains (XLS-80)** - Adds Permissioned Domains, which act as part of broader systems on the XRP Ledger to restrict access to satisfy compliance rules. ([#5161](https://github.com/XRPLF/rippled/pull/5161))
- **DeepFreeze (XLS-77)** - Adds the ability to deep freeze trust lines, enabling token issuers to block the transfer of assets for holders who have been deep frozen. ([#5187](https://github.com/XRPLF/rippled/pull/5187))
- **fixFrozenLPTokenTransfer** - Prohibits the transfer of LP tokens when the associated liquidity pool contains at least one frozen asset. ([#5227](https://github.com/XRPLF/rippled/pull/5227))
- **fixInvalidTxFlags** - Adds transaction flag checking for `CredentialCreate`, `CredentialAccept`, and `CredentialDelete` transactions. ([#5250](https://github.com/XRPLF/rippled/pull/5250))
### New Features
- Added a new `simulate` API method to execute dry runs of transactions and see the simulated metadata. ([#5069](https://github.com/XRPLF/rippled/pull/5069), [#5265](https://github.com/XRPLF/rippled/pull/5265))
- Added the ability to specify MPTs when defining assets in transactions. ([#5200](https://github.com/XRPLF/rippled/pull/5200))
- Added a `state` alias for `ripple_state` in the `ledger_entry` API method. Also refactored `LedgerEntry.cpp` to make it easier to read. ([#5199](https://github.com/XRPLF/rippled/pull/5199))
- Improved UNL security by enabling validators to set a minimum number of UNL publishers to agree on validators. ([#5112](https://github.com/XRPLF/rippled/pull/5112))
- Updated the XRPL Foundation UNL keys. ([#5289](https://github.com/XRPLF/rippled/pull/5289))
- Added a new XRPL Foundation subdomain to enable a staged migration without modifying the key for the current UNL list. ([#5326](https://github.com/XRPLF/rippled/pull/5326))
- Added support to filter ledger entry types by their canonical names in the `ledger`, `ledger_data`, and `account_objects` API methods. ([#5271](https://github.com/XRPLF/rippled/pull/5271))
- Added detailed logging for each validation and proposal received from the network. ([#5291](https://github.com/XRPLF/rippled/pull/5291))
- Improved git commit hash lookups when checking the version of a `rippled` debug build. Also added git commit hash info when using the `server_info` API method on an admin connection. ([#5225](https://github.com/XRPLF/rippled/pull/5225))
### Bug fixes
- Fixed an issue with overlapping data types in the `Expected` class. ([#5218](https://github.com/XRPLF/rippled/pull/5218))
- Fixed an issue that prevented `rippled` from building on Windows with VS2022. ([#5197](https://github.com/XRPLF/rippled/pull/5197))
- Fixed `server_definitions` prefixes. ([#5231](https://github.com/XRPLF/rippled/pull/5231))
- Added missing dependency installations for generic MasOS runners. ([#5233](https://github.com/XRPLF/rippled/pull/5233))
- Updated deprecated Github actions. ([#5241](https://github.com/XRPLF/rippled/pull/5241))
- Fixed a failing assert scenario when submitting the `connect` admin RPC. ([#5235](https://github.com/XRPLF/rippled/pull/5235))
- Fixed the levelization script to ignore single-line comments during dependency analysis. ([#5194](https://github.com/XRPLF/rippled/pull/5194))
- Fixed the assert name used in `PermissionedDomainDelete`. ([#5245](https://github.com/XRPLF/rippled/pull/5245))
- Fixed macOS unit tests. ([#5196](https://github.com/XRPLF/rippled/pull/5196))
- Fixed an issue with validators not accurately reflecting amendment votes. Also added debug logging of amendment votes. ([#5173](https://github.com/XRPLF/rippled/pull/5173), [#5312](https://github.com/XRPLF/rippled/pull/5312))
- Fixed a potential issue with double-charging fees. ([#5269](https://github.com/XRPLF/rippled/pull/5269))
- Removed the `new parent hash` assert and replaced it with a log message. ([#5313](https://github.com/XRPLF/rippled/pull/5313))
- Fixed an issue that prevented previously-failed inbound ledgers to not be acquired if a new trusted proposal arrived. ([#5318](https://github.com/XRPLF/rippled/pull/5318))
### Other Improvements
- Added unit tests for `AccountID` handling. ([#5174](https://github.com/XRPLF/rippled/pull/5174))
- Added enforced levelization in `libxrpl` with CMake. ([#5199](https://github.com/XRPLF/rippled/pull/5111))
- Updated `libxrpl` and all submodules to use the same compiler options. ([#5228](https://github.com/XRPLF/rippled/pull/5228))
- Added Antithesis instrumentation. ([#5042](https://github.com/XRPLF/rippled/pull/5042), [#5213](https://github.com/XRPLF/rippled/pull/5213))
- Added `rpcName` to the `LEDGER_ENTRY` macro to help prevent future bugs. ([#5202](https://github.com/XRPLF/rippled/pull/5202))
- Updated the contribution guidelines to introduce a new workflow that avoids code freezes. Also added scripts that can be used by maintainers in branch management, and a CI job to check that code is consistent across the three main branches: `master`, `release`, and `develop`. ([#5215](https://github.com/XRPLF/rippled/pull/5215))
- Added unit tests to check for caching issues fixed in `rippled 2.3.0`. ([#5242](https://github.com/XRPLF/rippled/pull/5242))
- Cleaned up the API changelog. ([#5207](https://github.com/XRPLF/rippled/pull/5207))
- Improved logs readability. ([#5251](https://github.com/XRPLF/rippled/pull/5251))
- Updated Visual Studio CI to VS 2022, and added VS Debug builds. ([#5240](https://github.com/XRPLF/rippled/pull/5240))
- Updated the `secp256k1` library to version 0.6.0. ([#5254](https://github.com/XRPLF/rippled/pull/5254))
- Changed the `[port_peer]` parameter in `rippled` example config back to `51235`; also added the recommendation to use the default port of `2459` for new deployments. ([#5290](https://github.com/XRPLF/rippled/pull/5290), [#5299](https://github.com/XRPLF/rippled/pull/5299))
- Improved CI management. ([#5268](https://github.com/XRPLF/rippled/pull/5268))
- Updated the git commit message rules for contributors. ([#5283](https://github.com/XRPLF/rippled/pull/5283))
- Fixed unnecessary `setCurrentThreadName` calls. ([#5280](https://github.com/XRPLF/rippled/pull/5280))
- Added a check to prevent permissioned domains from being created in the event the Permissioned Domains amendement is enabled before the Credentials amendement. ([#5275](https://github.com/XRPLF/rippled/pull/5275))
- Updated Conan dependencies. ([#5256](https://github.com/XRPLF/rippled/pull/5256))
- Fixed minor typos in code comments. ([#5279](https://github.com/XRPLF/rippled/pull/5279))
- Fixed incorrect build instructions. ([#5274](https://github.com/XRPLF/rippled/pull/5274))
- Refactored `rotateWithLock()` to not hold a lock during callbacks. ([#5276](https://github.com/XRPLF/rippled/pull/5276))
- Cleaned up debug logging by combining multiple data points into a single message. ([#5302](https://github.com/XRPLF/rippled/pull/5302))
- Updated build flags to fix performance regressions. ([#5325](https://github.com/XRPLF/rippled/pull/5325))
## Credits
The following people contributed directly to this release:
- Aanchal Malhotra <amalhotra@ripple.com>
- Bart Thomee <11445373+bthomee@users.noreply.github.com>
- Bronek Kozicki <brok@incorrekt.com>
- code0xff <ian.jungyong.um@gmail.com>
- Darius Tumas <tokeiito@tokeiito.eu>
- David Fuelling <fuelling@ripple.com>
- Donovan Hide <donovanhide@gmail.com>
- Ed Hennis <ed@ripple.com>
- Elliot Lee <github.public@intelliot.com>
- Javier Romero <jromero@ripple.com>
- Kenny Lei <klei@ripple.com>
- Mark Travis <7728157+mtrippled@users.noreply.github.com>
- Mayukha Vadari <mvadari@gmail.com>
- Michael Legleux <mlegleux@ripple.com>
- Oleksandr <115580134+oleks-rip@users.noreply.github.com>
- Qi Zhao <qzhao@ripple.com>
- Ramkumar Srirengaram Gunasegharan <rgunasegharan@ripple.com>
- Shae Wang <swang@ripple.com>
- Shawn Xie <shawnxie920@gmail.com>
- Sophia Xie <sxie@ripple.com>
- Vijay Khanna Raviraj <vraviraj@ripple.com>
- Vladislav Vysokikh <vvysokikh@gmail.com>
- Xun Zhao <xzhao@ripple.com>
## Bug Bounties and Responsible Disclosures
We welcome reviews of the `rippled` code and urge researchers to responsibly disclose any issues they may find.
To report a bug, please send a detailed report to: <bugs@xrpl.org>
# Version 2.3.1

View File

@@ -83,7 +83,7 @@ To report a qualifying bug, please send a detailed report to:
|Long Key ID | `0xCD49A0AFC57929BE` |
|Fingerprint | `24E6 3B02 37E0 FA9C 5E96 8974 CD49 A0AF C579 29BE` |
The full PGP key for this address, which is also available on several key servers (e.g. on [keyserver.ubuntu.com](https://keyserver.ubuntu.com)), is:
The full PGP key for this address, which is also available on several key servers (e.g. on [keys.gnupg.net](https://keys.gnupg.net)), is:
```
-----BEGIN PGP PUBLIC KEY BLOCK-----
mQINBFUwGHYBEAC0wpGpBPkd8W1UdQjg9+cEFzeIEJRaoZoeuJD8mofwI5Ejnjdt

View File

@@ -420,7 +420,6 @@
# - r.ripple.com 51235
# - sahyadri.isrdc.in 51235
# - hubs.xrpkuwait.com 51235
# - hub.xrpl-commons.org 51235
#
# Examples:
#

View File

@@ -26,7 +26,7 @@
#
# Examples:
# https://vl.ripple.com
# https://unl.xrplf.org
# https://vl.xrplf.org
# http://127.0.0.1:8000
# file:///etc/opt/ripple/vl.txt
#
@@ -54,12 +54,12 @@
[validator_list_sites]
https://vl.ripple.com
https://unl.xrplf.org
https://vl.xrplf.org
[validator_list_keys]
#vl.ripple.com
ED2677ABFFD1B33AC6FBC3062B71F1E8397C1505E1C42C64D11AD1B28FF73F4734
#unl.xrplf.org
# vl.xrplf.org
ED42AEC58B701EEBB77356FFFEC26F83C1F0407263530F068C7C73D392C7E06FD1
# To use the test network (see https://xrpl.org/connect-your-rippled-to-the-xrp-test-net.html),

View File

@@ -98,9 +98,6 @@
# 2024-04-03, Bronek Kozicki
# - add support for output formats: jacoco, clover, lcov
#
# 2025-05-12, Jingchen Wu
# - add -fprofile-update=atomic to ensure atomic profile generation
#
# USAGE:
#
# 1. Copy this file into your cmake modules path.
@@ -203,27 +200,15 @@ set(COVERAGE_COMPILER_FLAGS "-g --coverage"
CACHE INTERNAL "")
if(CMAKE_CXX_COMPILER_ID MATCHES "(GNU|Clang)")
include(CheckCXXCompilerFlag)
include(CheckCCompilerFlag)
check_cxx_compiler_flag(-fprofile-abs-path HAVE_cxx_fprofile_abs_path)
if(HAVE_cxx_fprofile_abs_path)
set(COVERAGE_CXX_COMPILER_FLAGS "${COVERAGE_COMPILER_FLAGS} -fprofile-abs-path")
endif()
include(CheckCCompilerFlag)
check_c_compiler_flag(-fprofile-abs-path HAVE_c_fprofile_abs_path)
if(HAVE_c_fprofile_abs_path)
set(COVERAGE_C_COMPILER_FLAGS "${COVERAGE_COMPILER_FLAGS} -fprofile-abs-path")
endif()
check_cxx_compiler_flag(-fprofile-update HAVE_cxx_fprofile_update)
if(HAVE_cxx_fprofile_update)
set(COVERAGE_CXX_COMPILER_FLAGS "${COVERAGE_COMPILER_FLAGS} -fprofile-update=atomic")
endif()
check_c_compiler_flag(-fprofile-update HAVE_c_fprofile_update)
if(HAVE_c_fprofile_update)
set(COVERAGE_C_COMPILER_FLAGS "${COVERAGE_COMPILER_FLAGS} -fprofile-update=atomic")
endif()
endif()
set(CMAKE_Fortran_FLAGS_COVERAGE

View File

@@ -136,9 +136,6 @@ if(xrpld)
add_executable(rippled)
if(tests)
target_compile_definitions(rippled PUBLIC ENABLE_TESTS)
target_compile_definitions(rippled PRIVATE
UNIT_TEST_REFERENCE_FEE=${UNIT_TEST_REFERENCE_FEE}
)
endif()
target_include_directories(rippled
PRIVATE

View File

@@ -53,9 +53,9 @@ set(download_script "${CMAKE_BINARY_DIR}/docs/download-cppreference.cmake")
file(WRITE
"${download_script}"
"file(DOWNLOAD \
https://github.com/PeterFeicht/cppreference-doc/releases/download/v20250209/html-book-20250209.zip \
http://upload.cppreference.com/mwiki/images/b/b2/html_book_20190607.zip \
${CMAKE_BINARY_DIR}/docs/cppreference.zip \
EXPECTED_HASH MD5=bda585f72fbca4b817b29a3d5746567b \
EXPECTED_HASH MD5=82b3a612d7d35a83e3cb1195a63689ab \
)\n \
execute_process( \
COMMAND \"${CMAKE_COMMAND}\" -E tar -xf cppreference.zip \

View File

@@ -2,6 +2,16 @@
convenience variables and sanity checks
#]===================================================================]
include(ProcessorCount)
if (NOT ep_procs)
ProcessorCount(ep_procs)
if (ep_procs GREATER 1)
# never use more than half of cores for EP builds
math (EXPR ep_procs "${ep_procs} / 2")
message (STATUS "Using ${ep_procs} cores for ExternalProject builds.")
endif ()
endif ()
get_property(is_multiconfig GLOBAL PROPERTY GENERATOR_IS_MULTI_CONFIG)
set (CMAKE_CONFIGURATION_TYPES "Debug;Release" CACHE STRING "" FORCE)

View File

@@ -11,12 +11,6 @@ option(assert "Enables asserts, even in release builds" OFF)
option(xrpld "Build xrpld" ON)
option(tests "Build tests" ON)
if(tests)
# This setting allows making a separate workflow to test fees other than default 10
if(NOT UNIT_TEST_REFERENCE_FEE)
set(UNIT_TEST_REFERENCE_FEE "10" CACHE STRING "")
endif()
endif()
option(unity "Creates a build using UNITY support in cmake. This is the default" ON)
if(unity)

View File

@@ -1,4 +1,4 @@
from conan import ConanFile, __version__ as conan_version
from conan import ConanFile
from conan.tools.cmake import CMake, CMakeToolchain, cmake_layout
import re
@@ -24,11 +24,13 @@ class Xrpl(ConanFile):
}
requires = [
'date/3.0.3',
'grpc/1.50.1',
'libarchive/3.7.6',
'nudb/2.0.8',
'openssl/1.1.1v',
'soci/4.0.3',
'xxhash/0.8.2',
'zlib/1.3.1',
]
@@ -97,18 +99,14 @@ class Xrpl(ConanFile):
self.options['boost'].visibility = 'global'
def requirements(self):
# Conan 2 requires transitive headers to be specified
transitive_headers_opt = {'transitive_headers': True} if conan_version.split('.')[0] == '2' else {}
self.requires('boost/1.83.0', force=True, **transitive_headers_opt)
self.requires('date/3.0.3', **transitive_headers_opt)
self.requires('boost/1.83.0', force=True)
self.requires('lz4/1.10.0', force=True)
self.requires('protobuf/3.21.9', force=True)
self.requires('sqlite3/3.47.0', force=True)
if self.options.jemalloc:
self.requires('jemalloc/5.3.0')
if self.options.rocksdb:
self.requires('rocksdb/9.7.3')
self.requires('xxhash/0.8.2', **transitive_headers_opt)
self.requires('rocksdb/6.29.5')
exports_sources = (
'CMakeLists.txt',

View File

@@ -23,7 +23,7 @@ direction.
```
apt update
apt install --yes curl git libssl-dev pipx python3.10-dev python3-pip make g++-11 libprotobuf-dev protobuf-compiler
apt install --yes curl git libssl-dev python3.10-dev python3-pip make g++-11 libprotobuf-dev protobuf-compiler
curl --location --remote-name \
"https://github.com/Kitware/CMake/releases/download/v3.25.1/cmake-3.25.1.tar.gz"
@@ -35,8 +35,7 @@ make --jobs $(nproc)
make install
cd ..
pipx install 'conan<2'
pipx ensurepath
pip3 install 'conan<2'
```
[1]: https://github.com/thejohnfreeman/rippled-docker/blob/master/ubuntu-22.04/install.sh

View File

@@ -558,7 +558,7 @@ struct ConsensusResult
ConsensusTimer roundTime;
// Indicates state in which consensus ended. Once in the accept phase
// will be either Yes or MovedOn or Expired
// will be either Yes or MovedOn
ConsensusState state = ConsensusState::No;
};

View File

@@ -1,10 +1,8 @@
#include <xrpl/protocol/BuildInfo.h>
#include <cstdio>
int
main(int argc, char const** argv)
{
#include <xrpl/protocol/BuildInfo.h>
int main(int argc, char const** argv) {
std::printf("%s\n", ripple::BuildInfo::getVersionString().c_str());
return 0;
}

View File

@@ -1,12 +1,27 @@
sources:
"9.7.3":
url: "https://github.com/facebook/rocksdb/archive/refs/tags/v9.7.3.tar.gz"
sha256: "acfabb989cbfb5b5c4d23214819b059638193ec33dad2d88373c46448d16d38b"
"6.29.5":
url: "https://github.com/facebook/rocksdb/archive/refs/tags/v6.29.5.tar.gz"
sha256: "ddbf84791f0980c0bbce3902feb93a2c7006f6f53bfd798926143e31d4d756f0"
"6.27.3":
url: "https://github.com/facebook/rocksdb/archive/refs/tags/v6.27.3.tar.gz"
sha256: "ee29901749b9132692b26f0a6c1d693f47d1a9ed8e3771e60556afe80282bf58"
"6.20.3":
url: "https://github.com/facebook/rocksdb/archive/refs/tags/v6.20.3.tar.gz"
sha256: "c6502c7aae641b7e20fafa6c2b92273d935d2b7b2707135ebd9a67b092169dca"
"8.8.1":
url: "https://github.com/facebook/rocksdb/archive/refs/tags/v8.8.1.tar.gz"
sha256: "056c7e21ad8ae36b026ac3b94b9d6e0fcc60e1d937fc80330921e4181be5c36e"
patches:
"9.7.3":
- patch_file: "patches/9.x.x-0001-exclude-thirdparty.patch"
"6.29.5":
- patch_file: "patches/6.29.5-0001-add-include-cstdint-for-gcc-13.patch"
patch_description: "Fix build with gcc 13 by including cstdint"
patch_type: "portability"
patch_source: "https://github.com/facebook/rocksdb/pull/11118"
- patch_file: "patches/6.29.5-0002-exclude-thirdparty.patch"
patch_description: "Do not include thirdparty.inc"
patch_type: "portability"
- patch_file: "patches/9.7.3-0001-memory-leak.patch"
patch_description: "Fix a leak of obsolete blob files left open until DB::Close()"
"6.27.3":
- patch_file: "patches/6.27.3-0001-add-include-cstdint-for-gcc-13.patch"
patch_description: "Fix build with gcc 13 by including cstdint"
patch_type: "portability"
patch_source: "https://github.com/facebook/rocksdb/pull/11118"

View File

@@ -15,10 +15,10 @@ required_conan_version = ">=1.53.0"
class RocksDBConan(ConanFile):
name = "rocksdb"
description = "A library that provides an embeddable, persistent key-value store for fast storage"
homepage = "https://github.com/facebook/rocksdb"
license = ("GPL-2.0-only", "Apache-2.0")
url = "https://github.com/conan-io/conan-center-index"
homepage = "https://github.com/facebook/rocksdb"
description = "A library that provides an embeddable, persistent key-value store for fast storage"
topics = ("database", "leveldb", "facebook", "key-value")
package_type = "library"
settings = "os", "arch", "compiler", "build_type"
@@ -58,12 +58,12 @@ class RocksDBConan(ConanFile):
@property
def _compilers_minimum_version(self):
return {} if self._min_cppstd == "11" else {
"apple-clang": "10",
"clang": "7",
"gcc": "7",
"msvc": "191",
"Visual Studio": "15",
}
"apple-clang": "10",
"clang": "7",
"gcc": "7",
"msvc": "191",
"Visual Studio": "15",
}
def export_sources(self):
export_conandata_patches(self)
@@ -115,9 +115,9 @@ class RocksDBConan(ConanFile):
check_min_vs(self, "191")
if self.version == "6.20.3" and \
self.settings.os == "Linux" and \
self.settings.compiler == "gcc" and \
Version(self.settings.compiler.version) < "5":
self.settings.os == "Linux" and \
self.settings.compiler == "gcc" and \
Version(self.settings.compiler.version) < "5":
raise ConanInvalidConfiguration("Rocksdb 6.20.3 is not compilable with gcc <5.") # See https://github.com/facebook/rocksdb/issues/3522
def source(self):
@@ -163,8 +163,6 @@ class RocksDBConan(ConanFile):
if self.options.with_jemalloc:
deps.set_property("jemalloc", "cmake_file_name", "JeMalloc")
deps.set_property("jemalloc", "cmake_target_name", "JeMalloc::JeMalloc")
if self.options.with_zstd:
deps.set_property("zstd", "cmake_target_name", "zstd::zstd")
deps.generate()
def build(self):

View File

@@ -0,0 +1,30 @@
--- a/include/rocksdb/utilities/checkpoint.h
+++ b/include/rocksdb/utilities/checkpoint.h
@@ -8,6 +8,7 @@
#pragma once
#ifndef ROCKSDB_LITE
+#include <cstdint>
#include <string>
#include <vector>
#include "rocksdb/status.h"
--- a/table/block_based/data_block_hash_index.h
+++ b/table/block_based/data_block_hash_index.h
@@ -5,6 +5,7 @@
#pragma once
+#include <cstdint>
#include <string>
#include <vector>
--- a/util/string_util.h
+++ b/util/string_util.h
@@ -6,6 +6,7 @@
#pragma once
+#include <cstdint>
#include <sstream>
#include <string>
#include <unordered_map>

View File

@@ -0,0 +1,16 @@
diff --git a/CMakeLists.txt b/CMakeLists.txt
index ec59d4491..35577c998 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -101 +100,0 @@ if(MSVC)
- option(WITH_GFLAGS "build with GFlags" OFF)
@@ -103,2 +102,2 @@ if(MSVC)
- include(${CMAKE_CURRENT_SOURCE_DIR}/thirdparty.inc)
-else()
+endif()
+
@@ -117 +116 @@ else()
- if(MINGW)
+ if(MINGW OR MSVC)
@@ -183 +181,0 @@ else()
-endif()

View File

@@ -1,319 +0,0 @@
diff --git a/HISTORY.md b/HISTORY.md
index 36d472229..05ad1a202 100644
--- a/HISTORY.md
+++ b/HISTORY.md
@@ -1,6 +1,10 @@
# Rocksdb Change Log
> NOTE: Entries for next release do not go here. Follow instructions in `unreleased_history/README.txt`
+## 9.7.4 (10/31/2024)
+### Bug Fixes
+* Fix a leak of obsolete blob files left open until DB::Close(). This bug was introduced in version 9.4.0.
+
## 9.7.3 (10/16/2024)
### Behavior Changes
* OPTIONS file to be loaded by remote worker is now preserved so that it does not get purged by the primary host. A similar technique as how we are preserving new SST files from getting purged is used for this. min_options_file_numbers_ is tracked like pending_outputs_ is tracked.
diff --git a/db/blob/blob_file_cache.cc b/db/blob/blob_file_cache.cc
index 5f340aadf..1b9faa238 100644
--- a/db/blob/blob_file_cache.cc
+++ b/db/blob/blob_file_cache.cc
@@ -42,6 +42,7 @@ Status BlobFileCache::GetBlobFileReader(
assert(blob_file_reader);
assert(blob_file_reader->IsEmpty());
+ // NOTE: sharing same Cache with table_cache
const Slice key = GetSliceForKey(&blob_file_number);
assert(cache_);
@@ -98,4 +99,13 @@ Status BlobFileCache::GetBlobFileReader(
return Status::OK();
}
+void BlobFileCache::Evict(uint64_t blob_file_number) {
+ // NOTE: sharing same Cache with table_cache
+ const Slice key = GetSliceForKey(&blob_file_number);
+
+ assert(cache_);
+
+ cache_.get()->Erase(key);
+}
+
} // namespace ROCKSDB_NAMESPACE
diff --git a/db/blob/blob_file_cache.h b/db/blob/blob_file_cache.h
index 740e67ada..6858d012b 100644
--- a/db/blob/blob_file_cache.h
+++ b/db/blob/blob_file_cache.h
@@ -36,6 +36,15 @@ class BlobFileCache {
uint64_t blob_file_number,
CacheHandleGuard<BlobFileReader>* blob_file_reader);
+ // Called when a blob file is obsolete to ensure it is removed from the cache
+ // to avoid effectively leaking the open file and assicated memory
+ void Evict(uint64_t blob_file_number);
+
+ // Used to identify cache entries for blob files (not normally useful)
+ static const Cache::CacheItemHelper* GetHelper() {
+ return CacheInterface::GetBasicHelper();
+ }
+
private:
using CacheInterface =
BasicTypedCacheInterface<BlobFileReader, CacheEntryRole::kMisc>;
diff --git a/db/column_family.h b/db/column_family.h
index e4b7adde8..86637736a 100644
--- a/db/column_family.h
+++ b/db/column_family.h
@@ -401,6 +401,7 @@ class ColumnFamilyData {
SequenceNumber earliest_seq);
TableCache* table_cache() const { return table_cache_.get(); }
+ BlobFileCache* blob_file_cache() const { return blob_file_cache_.get(); }
BlobSource* blob_source() const { return blob_source_.get(); }
// See documentation in compaction_picker.h
diff --git a/db/db_impl/db_impl.cc b/db/db_impl/db_impl.cc
index 261593423..06573ac2e 100644
--- a/db/db_impl/db_impl.cc
+++ b/db/db_impl/db_impl.cc
@@ -659,8 +659,9 @@ Status DBImpl::CloseHelper() {
// We need to release them before the block cache is destroyed. The block
// cache may be destroyed inside versions_.reset(), when column family data
// list is destroyed, so leaving handles in table cache after
- // versions_.reset() may cause issues.
- // Here we clean all unreferenced handles in table cache.
+ // versions_.reset() may cause issues. Here we clean all unreferenced handles
+ // in table cache, and (for certain builds/conditions) assert that no obsolete
+ // files are hanging around unreferenced (leak) in the table/blob file cache.
// Now we assume all user queries have finished, so only version set itself
// can possibly hold the blocks from block cache. After releasing unreferenced
// handles here, only handles held by version set left and inside
@@ -668,6 +669,9 @@ Status DBImpl::CloseHelper() {
// time a handle is released, we erase it from the cache too. By doing that,
// we can guarantee that after versions_.reset(), table cache is empty
// so the cache can be safely destroyed.
+#ifndef NDEBUG
+ TEST_VerifyNoObsoleteFilesCached(/*db_mutex_already_held=*/true);
+#endif // !NDEBUG
table_cache_->EraseUnRefEntries();
for (auto& txn_entry : recovered_transactions_) {
@@ -3227,6 +3231,8 @@ Status DBImpl::MultiGetImpl(
s = Status::Aborted();
break;
}
+ // This could be a long-running operation
+ ROCKSDB_THREAD_YIELD_HOOK();
}
// Post processing (decrement reference counts and record statistics)
diff --git a/db/db_impl/db_impl.h b/db/db_impl/db_impl.h
index 5e4fa310b..ccc0abfa7 100644
--- a/db/db_impl/db_impl.h
+++ b/db/db_impl/db_impl.h
@@ -1241,9 +1241,14 @@ class DBImpl : public DB {
static Status TEST_ValidateOptions(const DBOptions& db_options) {
return ValidateOptions(db_options);
}
-
#endif // NDEBUG
+ // In certain configurations, verify that the table/blob file cache only
+ // contains entries for live files, to check for effective leaks of open
+ // files. This can only be called when purging of obsolete files has
+ // "settled," such as during parts of DB Close().
+ void TEST_VerifyNoObsoleteFilesCached(bool db_mutex_already_held) const;
+
// persist stats to column family "_persistent_stats"
void PersistStats();
diff --git a/db/db_impl/db_impl_debug.cc b/db/db_impl/db_impl_debug.cc
index 790a50d7a..67f5b4aaf 100644
--- a/db/db_impl/db_impl_debug.cc
+++ b/db/db_impl/db_impl_debug.cc
@@ -9,6 +9,7 @@
#ifndef NDEBUG
+#include "db/blob/blob_file_cache.h"
#include "db/column_family.h"
#include "db/db_impl/db_impl.h"
#include "db/error_handler.h"
@@ -328,5 +329,49 @@ size_t DBImpl::TEST_EstimateInMemoryStatsHistorySize() const {
InstrumentedMutexLock l(&const_cast<DBImpl*>(this)->stats_history_mutex_);
return EstimateInMemoryStatsHistorySize();
}
+
+void DBImpl::TEST_VerifyNoObsoleteFilesCached(
+ bool db_mutex_already_held) const {
+ // This check is somewhat expensive and obscure to make a part of every
+ // unit test in every build variety. Thus, we only enable it for ASAN builds.
+ if (!kMustFreeHeapAllocations) {
+ return;
+ }
+
+ std::optional<InstrumentedMutexLock> l;
+ if (db_mutex_already_held) {
+ mutex_.AssertHeld();
+ } else {
+ l.emplace(&mutex_);
+ }
+
+ std::vector<uint64_t> live_files;
+ for (auto cfd : *versions_->GetColumnFamilySet()) {
+ if (cfd->IsDropped()) {
+ continue;
+ }
+ // Sneakily add both SST and blob files to the same list
+ cfd->current()->AddLiveFiles(&live_files, &live_files);
+ }
+ std::sort(live_files.begin(), live_files.end());
+
+ auto fn = [&live_files](const Slice& key, Cache::ObjectPtr, size_t,
+ const Cache::CacheItemHelper* helper) {
+ if (helper != BlobFileCache::GetHelper()) {
+ // Skip non-blob files for now
+ // FIXME: diagnose and fix the leaks of obsolete SST files revealed in
+ // unit tests.
+ return;
+ }
+ // See TableCache and BlobFileCache
+ assert(key.size() == sizeof(uint64_t));
+ uint64_t file_number;
+ GetUnaligned(reinterpret_cast<const uint64_t*>(key.data()), &file_number);
+ // Assert file is in sorted live_files
+ assert(
+ std::binary_search(live_files.begin(), live_files.end(), file_number));
+ };
+ table_cache_->ApplyToAllEntries(fn, {});
+}
} // namespace ROCKSDB_NAMESPACE
#endif // NDEBUG
diff --git a/db/db_iter.cc b/db/db_iter.cc
index e02586377..bf4749eb9 100644
--- a/db/db_iter.cc
+++ b/db/db_iter.cc
@@ -540,6 +540,8 @@ bool DBIter::FindNextUserEntryInternal(bool skipping_saved_key,
} else {
iter_.Next();
}
+ // This could be a long-running operation due to tombstones, etc.
+ ROCKSDB_THREAD_YIELD_HOOK();
} while (iter_.Valid());
valid_ = false;
diff --git a/db/table_cache.cc b/db/table_cache.cc
index 71fc29c32..8a5be75e8 100644
--- a/db/table_cache.cc
+++ b/db/table_cache.cc
@@ -164,6 +164,7 @@ Status TableCache::GetTableReader(
}
Cache::Handle* TableCache::Lookup(Cache* cache, uint64_t file_number) {
+ // NOTE: sharing same Cache with BlobFileCache
Slice key = GetSliceForFileNumber(&file_number);
return cache->Lookup(key);
}
@@ -179,6 +180,7 @@ Status TableCache::FindTable(
size_t max_file_size_for_l0_meta_pin, Temperature file_temperature) {
PERF_TIMER_GUARD_WITH_CLOCK(find_table_nanos, ioptions_.clock);
uint64_t number = file_meta.fd.GetNumber();
+ // NOTE: sharing same Cache with BlobFileCache
Slice key = GetSliceForFileNumber(&number);
*handle = cache_.Lookup(key);
TEST_SYNC_POINT_CALLBACK("TableCache::FindTable:0",
diff --git a/db/version_builder.cc b/db/version_builder.cc
index ed8ab8214..c98f53f42 100644
--- a/db/version_builder.cc
+++ b/db/version_builder.cc
@@ -24,6 +24,7 @@
#include <vector>
#include "cache/cache_reservation_manager.h"
+#include "db/blob/blob_file_cache.h"
#include "db/blob/blob_file_meta.h"
#include "db/dbformat.h"
#include "db/internal_stats.h"
@@ -744,12 +745,9 @@ class VersionBuilder::Rep {
return Status::Corruption("VersionBuilder", oss.str());
}
- // Note: we use C++11 for now but in C++14, this could be done in a more
- // elegant way using generalized lambda capture.
- VersionSet* const vs = version_set_;
- const ImmutableCFOptions* const ioptions = ioptions_;
-
- auto deleter = [vs, ioptions](SharedBlobFileMetaData* shared_meta) {
+ auto deleter = [vs = version_set_, ioptions = ioptions_,
+ bc = cfd_ ? cfd_->blob_file_cache()
+ : nullptr](SharedBlobFileMetaData* shared_meta) {
if (vs) {
assert(ioptions);
assert(!ioptions->cf_paths.empty());
@@ -758,6 +756,9 @@ class VersionBuilder::Rep {
vs->AddObsoleteBlobFile(shared_meta->GetBlobFileNumber(),
ioptions->cf_paths.front().path);
}
+ if (bc) {
+ bc->Evict(shared_meta->GetBlobFileNumber());
+ }
delete shared_meta;
};
@@ -766,7 +767,7 @@ class VersionBuilder::Rep {
blob_file_number, blob_file_addition.GetTotalBlobCount(),
blob_file_addition.GetTotalBlobBytes(),
blob_file_addition.GetChecksumMethod(),
- blob_file_addition.GetChecksumValue(), deleter);
+ blob_file_addition.GetChecksumValue(), std::move(deleter));
mutable_blob_file_metas_.emplace(
blob_file_number, MutableBlobFileMetaData(std::move(shared_meta)));
diff --git a/db/version_set.h b/db/version_set.h
index 9336782b1..024f869e7 100644
--- a/db/version_set.h
+++ b/db/version_set.h
@@ -1514,7 +1514,6 @@ class VersionSet {
void GetLiveFilesMetaData(std::vector<LiveFileMetaData>* metadata);
void AddObsoleteBlobFile(uint64_t blob_file_number, std::string path) {
- // TODO: Erase file from BlobFileCache?
obsolete_blob_files_.emplace_back(blob_file_number, std::move(path));
}
diff --git a/include/rocksdb/version.h b/include/rocksdb/version.h
index 2a19796b8..0afa2cab1 100644
--- a/include/rocksdb/version.h
+++ b/include/rocksdb/version.h
@@ -13,7 +13,7 @@
// minor or major version number planned for release.
#define ROCKSDB_MAJOR 9
#define ROCKSDB_MINOR 7
-#define ROCKSDB_PATCH 3
+#define ROCKSDB_PATCH 4
// Do not use these. We made the mistake of declaring macros starting with
// double underscore. Now we have to live with our choice. We'll deprecate these
diff --git a/port/port.h b/port/port.h
index 13aa56d47..141716e5b 100644
--- a/port/port.h
+++ b/port/port.h
@@ -19,3 +19,19 @@
#elif defined(OS_WIN)
#include "port/win/port_win.h"
#endif
+
+#ifdef OS_LINUX
+// A temporary hook into long-running RocksDB threads to support modifying their
+// priority etc. This should become a public API hook once the requirements
+// are better understood.
+extern "C" void RocksDbThreadYield() __attribute__((__weak__));
+#define ROCKSDB_THREAD_YIELD_HOOK() \
+ { \
+ if (RocksDbThreadYield) { \
+ RocksDbThreadYield(); \
+ } \
+ }
+#else
+#define ROCKSDB_THREAD_YIELD_HOOK() \
+ {}
+#endif

View File

@@ -1,30 +0,0 @@
diff --git a/CMakeLists.txt b/CMakeLists.txt
index 93b884d..b715cb6 100644
--- a/CMakeLists.txt
+++ b/CMakeLists.txt
@@ -106,14 +106,9 @@ endif()
include(CMakeDependentOption)
if(MSVC)
- option(WITH_GFLAGS "build with GFlags" OFF)
option(WITH_XPRESS "build with windows built in compression" OFF)
- option(ROCKSDB_SKIP_THIRDPARTY "skip thirdparty.inc" OFF)
-
- if(NOT ROCKSDB_SKIP_THIRDPARTY)
- include(${CMAKE_CURRENT_SOURCE_DIR}/thirdparty.inc)
- endif()
-else()
+endif()
+if(TRUE)
if(CMAKE_SYSTEM_NAME MATCHES "FreeBSD" AND NOT CMAKE_SYSTEM_NAME MATCHES "kFreeBSD")
# FreeBSD has jemalloc as default malloc
# but it does not have all the jemalloc files in include/...
@@ -126,7 +121,7 @@ else()
endif()
endif()
- if(MINGW)
+ if(MSVC OR MINGW)
option(WITH_GFLAGS "build with GFlags" OFF)
else()
option(WITH_GFLAGS "build with GFlags" ON)

View File

@@ -27,6 +27,7 @@
#include <algorithm>
#include <optional>
#include <ostream>
#include <string>
#include <unordered_map>
#include <vector>
@@ -367,7 +368,7 @@ get(Section const& section,
}
inline std::string
get(Section const& section, std::string const& name, char const* defaultValue)
get(Section const& section, std::string const& name, const char* defaultValue)
{
try
{

View File

@@ -22,9 +22,10 @@
#include <xrpl/basics/Slice.h>
#include <xrpl/beast/utility/instrumentation.h>
#include <cstdint>
#include <cstring>
#include <memory>
#include <utility>
namespace ripple {

View File

@@ -21,11 +21,9 @@
#define RIPPLED_COMPRESSIONALGORITHMS_H_INCLUDED
#include <xrpl/basics/contract.h>
#include <lz4.h>
#include <algorithm>
#include <cstdint>
#include <lz4.h>
#include <stdexcept>
#include <vector>
@@ -55,7 +53,7 @@ lz4Compress(void const* in, std::size_t inSize, BufferFactory&& bf)
auto compressed = bf(outCapacity);
auto compressedSize = LZ4_compress_default(
reinterpret_cast<char const*>(in),
reinterpret_cast<const char*>(in),
reinterpret_cast<char*>(compressed),
inSize,
outCapacity);
@@ -89,7 +87,7 @@ lz4Decompress(
Throw<std::runtime_error>("lz4Decompress: integer overflow (output)");
if (LZ4_decompress_safe(
reinterpret_cast<char const*>(in),
reinterpret_cast<const char*>(in),
reinterpret_cast<char*>(decompressed),
inSize,
decompressedSize) != decompressedSize)

View File

@@ -21,7 +21,6 @@
#define RIPPLE_BASICS_COUNTEDOBJECT_H_INCLUDED
#include <xrpl/beast/type_name.h>
#include <atomic>
#include <string>
#include <utility>

View File

@@ -24,7 +24,9 @@
#include <boost/outcome.hpp>
#include <concepts>
#include <stdexcept>
#include <type_traits>
namespace ripple {
@@ -93,7 +95,7 @@ public:
{
}
constexpr E const&
constexpr const E&
value() const&
{
return val_;
@@ -111,7 +113,7 @@ public:
return std::move(val_);
}
constexpr E const&&
constexpr const E&&
value() const&&
{
return std::move(val_);

View File

@@ -1,515 +0,0 @@
//------------------------------------------------------------------------------
/*
This file is part of rippled: https://github.com/ripple/rippled
Copyright (c) 2023 Ripple Labs Inc.
Permission to use, copy, modify, and/or distribute this software for any
purpose with or without fee is hereby granted, provided that the above
copyright notice and this permission notice appear in all copies.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
ANY SPECIAL , DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
//==============================================================================
#ifndef RIPPLE_BASICS_INTRUSIVEPOINTER_H_INCLUDED
#define RIPPLE_BASICS_INTRUSIVEPOINTER_H_INCLUDED
#include <concepts>
#include <cstdint>
#include <type_traits>
#include <utility>
namespace ripple {
//------------------------------------------------------------------------------
/** Tag to create an intrusive pointer from another intrusive pointer by using a
static cast. This is useful to create an intrusive pointer to a derived
class from an intrusive pointer to a base class.
*/
struct StaticCastTagSharedIntrusive
{
};
/** Tag to create an intrusive pointer from another intrusive pointer by using a
dynamic cast. This is useful to create an intrusive pointer to a derived
class from an intrusive pointer to a base class. If the cast fails an empty
(null) intrusive pointer is created.
*/
struct DynamicCastTagSharedIntrusive
{
};
/** When creating or adopting a raw pointer, controls whether the strong count
is incremented or not. Use this tag to increment the strong count.
*/
struct SharedIntrusiveAdoptIncrementStrongTag
{
};
/** When creating or adopting a raw pointer, controls whether the strong count
is incremented or not. Use this tag to leave the strong count unchanged.
*/
struct SharedIntrusiveAdoptNoIncrementTag
{
};
//------------------------------------------------------------------------------
//
template <class T>
concept CAdoptTag = std::is_same_v<T, SharedIntrusiveAdoptIncrementStrongTag> ||
std::is_same_v<T, SharedIntrusiveAdoptNoIncrementTag>;
//------------------------------------------------------------------------------
/** A shared intrusive pointer class that supports weak pointers.
This is meant to be used for SHAMapInnerNodes, but may be useful for other
cases. Since the reference counts are stored on the pointee, the pointee is
not destroyed until both the strong _and_ weak pointer counts go to zero.
When the strong pointer count goes to zero, the "partialDestructor" is
called. This can be used to destroy as much of the object as possible while
still retaining the reference counts. For example, for SHAMapInnerNodes the
children may be reset in that function. Note that std::shared_poiner WILL
run the destructor when the strong count reaches zero, but may not free the
memory used by the object until the weak count reaches zero. In rippled, we
typically allocate shared pointers with the `make_shared` function. When
that is used, the memory is not reclaimed until the weak count reaches zero.
*/
template <class T>
class SharedIntrusive
{
public:
SharedIntrusive() = default;
template <CAdoptTag TAdoptTag>
SharedIntrusive(T* p, TAdoptTag) noexcept;
SharedIntrusive(SharedIntrusive const& rhs);
template <class TT>
// TODO: convertible_to isn't quite right. That include a static castable.
// Find the right concept.
requires std::convertible_to<TT*, T*>
SharedIntrusive(SharedIntrusive<TT> const& rhs);
SharedIntrusive(SharedIntrusive&& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedIntrusive(SharedIntrusive<TT>&& rhs);
SharedIntrusive&
operator=(SharedIntrusive const& rhs);
bool
operator!=(std::nullptr_t) const;
bool
operator==(std::nullptr_t) const;
template <class TT>
requires std::convertible_to<TT*, T*>
SharedIntrusive&
operator=(SharedIntrusive<TT> const& rhs);
SharedIntrusive&
operator=(SharedIntrusive&& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedIntrusive&
operator=(SharedIntrusive<TT>&& rhs);
/** Adopt the raw pointer. The strong reference may or may not be
incremented, depending on the TAdoptTag
*/
template <CAdoptTag TAdoptTag = SharedIntrusiveAdoptIncrementStrongTag>
void
adopt(T* p);
~SharedIntrusive();
/** Create a new SharedIntrusive by statically casting the pointer
controlled by the rhs param.
*/
template <class TT>
SharedIntrusive(
StaticCastTagSharedIntrusive,
SharedIntrusive<TT> const& rhs);
/** Create a new SharedIntrusive by statically casting the pointer
controlled by the rhs param.
*/
template <class TT>
SharedIntrusive(StaticCastTagSharedIntrusive, SharedIntrusive<TT>&& rhs);
/** Create a new SharedIntrusive by dynamically casting the pointer
controlled by the rhs param.
*/
template <class TT>
SharedIntrusive(
DynamicCastTagSharedIntrusive,
SharedIntrusive<TT> const& rhs);
/** Create a new SharedIntrusive by dynamically casting the pointer
controlled by the rhs param.
*/
template <class TT>
SharedIntrusive(DynamicCastTagSharedIntrusive, SharedIntrusive<TT>&& rhs);
T&
operator*() const noexcept;
T*
operator->() const noexcept;
explicit
operator bool() const noexcept;
/** Set the pointer to null, decrement the strong count, and run the
appropriate release action.
*/
void
reset();
/** Get the raw pointer */
T*
get() const;
/** Return the strong count */
std::size_t
use_count() const;
template <class TT, class... Args>
friend SharedIntrusive<TT>
make_SharedIntrusive(Args&&... args);
template <class TT>
friend class SharedIntrusive;
template <class TT>
friend class SharedWeakUnion;
template <class TT>
friend class WeakIntrusive;
private:
/** Return the raw pointer held by this object. */
T*
unsafeGetRawPtr() const;
/** Exchange the current raw pointer held by this object with the given
pointer. Decrement the strong count of the raw pointer previously held
by this object and run the appropriate release action.
*/
void
unsafeReleaseAndStore(T* next);
/** Set the raw pointer directly. This is wrapped in a function so the class
can support both atomic and non-atomic pointers in a future patch.
*/
void
unsafeSetRawPtr(T* p);
/** Exchange the raw pointer directly.
This sets the raw pointer to the given value and returns the previous
value. This is wrapped in a function so the class can support both
atomic and non-atomic pointers in a future patch.
*/
T*
unsafeExchange(T* p);
/** pointer to the type with an intrusive count */
T* ptr_{nullptr};
};
//------------------------------------------------------------------------------
/** A weak intrusive pointer class for the SharedIntrusive pointer class.
Note that this weak pointer class asks differently from normal weak pointer
classes. When the strong pointer count goes to zero, the "partialDestructor"
is called. See the comment on SharedIntrusive for a fuller explanation.
*/
template <class T>
class WeakIntrusive
{
public:
WeakIntrusive() = default;
WeakIntrusive(WeakIntrusive const& rhs);
WeakIntrusive(WeakIntrusive&& rhs);
WeakIntrusive(SharedIntrusive<T> const& rhs);
// There is no move constructor from a strong intrusive ptr because
// moving would be move expensive than copying in this case (the strong
// ref would need to be decremented)
WeakIntrusive(SharedIntrusive<T> const&& rhs) = delete;
// Since there are no current use cases for copy assignment in
// WeakIntrusive, we delete this operator to simplify the implementation. If
// a need arises in the future, we can reintroduce it with proper
// consideration."
WeakIntrusive&
operator=(WeakIntrusive const&) = delete;
template <class TT>
requires std::convertible_to<TT*, T*>
WeakIntrusive&
operator=(SharedIntrusive<TT> const& rhs);
/** Adopt the raw pointer and increment the weak count. */
void
adopt(T* ptr);
~WeakIntrusive();
/** Get a strong pointer from the weak pointer, if possible. This will
only return a seated pointer if the strong count on the raw pointer
is non-zero before locking.
*/
SharedIntrusive<T>
lock() const;
/** Return true if the strong count is zero. */
bool
expired() const;
/** Set the pointer to null and decrement the weak count.
Note: This may run the destructor if the strong count is zero.
*/
void
reset();
private:
T* ptr_ = nullptr;
/** Decrement the weak count. This does _not_ set the raw pointer to
null.
Note: This may run the destructor if the strong count is zero.
*/
void
unsafeReleaseNoStore();
};
//------------------------------------------------------------------------------
/** A combination of a strong and a weak intrusive pointer stored in the
space of a single pointer.
This class is similar to a `std::variant<SharedIntrusive,WeakIntrusive>`
with some optimizations. In particular, it uses a low-order bit to
determine if the raw pointer represents a strong pointer or a weak
pointer. It can also be quickly switched between its strong pointer and
weak pointer representations. This class is useful for storing intrusive
pointers in tagged caches.
*/
template <class T>
class SharedWeakUnion
{
// Tagged pointer. Low bit determines if this is a strong or a weak
// pointer. The low bit must be masked to zero when converting back to a
// pointer. If the low bit is '1', this is a weak pointer.
static_assert(
alignof(T) >= 2,
"Bad alignment: Combo pointer requires low bit to be zero");
public:
SharedWeakUnion() = default;
SharedWeakUnion(SharedWeakUnion const& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakUnion(SharedIntrusive<TT> const& rhs);
SharedWeakUnion(SharedWeakUnion&& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakUnion(SharedIntrusive<TT>&& rhs);
SharedWeakUnion&
operator=(SharedWeakUnion const& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakUnion&
operator=(SharedIntrusive<TT> const& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakUnion&
operator=(SharedIntrusive<TT>&& rhs);
~SharedWeakUnion();
/** Return a strong pointer if this is already a strong pointer (i.e.
don't lock the weak pointer. Use the `lock` method if that's what's
needed)
*/
SharedIntrusive<T>
getStrong() const;
/** Return true if this is a strong pointer and the strong pointer is
seated.
*/
explicit
operator bool() const noexcept;
/** Set the pointer to null, decrement the appropriate ref count, and
run the appropriate release action.
*/
void
reset();
/** If this is a strong pointer, return the raw pointer. Otherwise
return null.
*/
T*
get() const;
/** If this is a strong pointer, return the strong count. Otherwise
* return 0
*/
std::size_t
use_count() const;
/** Return true if there is a non-zero strong count. */
bool
expired() const;
/** If this is a strong pointer, return the strong pointer. Otherwise
attempt to lock the weak pointer.
*/
SharedIntrusive<T>
lock() const;
/** Return true is this represents a strong pointer. */
bool
isStrong() const;
/** Return true is this represents a weak pointer. */
bool
isWeak() const;
/** If this is a weak pointer, attempt to convert it to a strong
pointer.
@return true if successfully converted to a strong pointer (or was
already a strong pointer). Otherwise false.
*/
bool
convertToStrong();
/** If this is a strong pointer, attempt to convert it to a weak
pointer.
@return false if the pointer is null. Otherwise return true.
*/
bool
convertToWeak();
private:
// Tagged pointer. Low bit determines if this is a strong or a weak
// pointer. The low bit must be masked to zero when converting back to a
// pointer. If the low bit is '1', this is a weak pointer.
std::uintptr_t tp_{0};
static constexpr std::uintptr_t tagMask = 1;
static constexpr std::uintptr_t ptrMask = ~tagMask;
private:
/** Return the raw pointer held by this object.
*/
T*
unsafeGetRawPtr() const;
enum class RefStrength { strong, weak };
/** Set the raw pointer and tag bit directly.
*/
void
unsafeSetRawPtr(T* p, RefStrength rs);
/** Set the raw pointer and tag bit to all zeros (strong null pointer).
*/
void unsafeSetRawPtr(std::nullptr_t);
/** Decrement the appropriate ref count, and run the appropriate release
action. Note: this does _not_ set the raw pointer to null.
*/
void
unsafeReleaseNoStore();
};
//------------------------------------------------------------------------------
/** Create a shared intrusive pointer.
Note: unlike std::shared_ptr, where there is an advantage of allocating
the pointer and control block together, there is no benefit for intrusive
pointers.
*/
template <class TT, class... Args>
SharedIntrusive<TT>
make_SharedIntrusive(Args&&... args)
{
auto p = new TT(std::forward<Args>(args)...);
static_assert(
noexcept(SharedIntrusive<TT>(
std::declval<TT*>(),
std::declval<SharedIntrusiveAdoptNoIncrementTag>())),
"SharedIntrusive constructor should not throw or this can leak "
"memory");
return SharedIntrusive<TT>(p, SharedIntrusiveAdoptNoIncrementTag{});
}
//------------------------------------------------------------------------------
namespace intr_ptr {
template <class T>
using SharedPtr = SharedIntrusive<T>;
template <class T>
using WeakPtr = WeakIntrusive<T>;
template <class T>
using SharedWeakUnionPtr = SharedWeakUnion<T>;
template <class T, class... A>
SharedPtr<T>
make_shared(A&&... args)
{
return make_SharedIntrusive<T>(std::forward<A>(args)...);
}
template <class T, class TT>
SharedPtr<T>
static_pointer_cast(TT const& v)
{
return SharedPtr<T>(StaticCastTagSharedIntrusive{}, v);
}
template <class T, class TT>
SharedPtr<T>
dynamic_pointer_cast(TT const& v)
{
return SharedPtr<T>(DynamicCastTagSharedIntrusive{}, v);
}
} // namespace intr_ptr
} // namespace ripple
#endif

View File

@@ -1,740 +0,0 @@
//------------------------------------------------------------------------------
/*
This file is part of rippled: https://github.com/ripple/rippled
Copyright (c) 2023 Ripple Labs Inc.
Permission to use, copy, modify, and/or distribute this software for any
purpose with or without fee is hereby granted, provided that the above
copyright notice and this permission notice appear in all copies.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
ANY SPECIAL , DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
//==============================================================================
#ifndef RIPPLE_BASICS_INTRUSIVEPOINTER_IPP_INCLUDED
#define RIPPLE_BASICS_INTRUSIVEPOINTER_IPP_INCLUDED
#include <xrpl/basics/IntrusivePointer.h>
#include <xrpl/basics/IntrusiveRefCounts.h>
#include <utility>
namespace ripple {
template <class T>
template <CAdoptTag TAdoptTag>
SharedIntrusive<T>::SharedIntrusive(T* p, TAdoptTag) noexcept : ptr_{p}
{
if constexpr (std::is_same_v<
TAdoptTag,
SharedIntrusiveAdoptIncrementStrongTag>)
{
if (p)
p->addStrongRef();
}
}
template <class T>
SharedIntrusive<T>::SharedIntrusive(SharedIntrusive const& rhs)
: ptr_{[&] {
auto p = rhs.unsafeGetRawPtr();
if (p)
p->addStrongRef();
return p;
}()}
{
}
template <class T>
template <class TT>
requires std::convertible_to<TT*, T*>
SharedIntrusive<T>::SharedIntrusive(SharedIntrusive<TT> const& rhs)
: ptr_{[&] {
auto p = rhs.unsafeGetRawPtr();
if (p)
p->addStrongRef();
return p;
}()}
{
}
template <class T>
SharedIntrusive<T>::SharedIntrusive(SharedIntrusive&& rhs)
: ptr_{rhs.unsafeExchange(nullptr)}
{
}
template <class T>
template <class TT>
requires std::convertible_to<TT*, T*>
SharedIntrusive<T>::SharedIntrusive(SharedIntrusive<TT>&& rhs)
: ptr_{rhs.unsafeExchange(nullptr)}
{
}
template <class T>
SharedIntrusive<T>&
SharedIntrusive<T>::operator=(SharedIntrusive const& rhs)
{
if (this == &rhs)
return *this;
auto p = rhs.unsafeGetRawPtr();
if (p)
p->addStrongRef();
unsafeReleaseAndStore(p);
return *this;
}
template <class T>
template <class TT>
// clang-format off
requires std::convertible_to<TT*, T*>
// clang-format on
SharedIntrusive<T>&
SharedIntrusive<T>::operator=(SharedIntrusive<TT> const& rhs)
{
if constexpr (std::is_same_v<T, TT>)
{
// This case should never be hit. The operator above will run instead.
// (The normal operator= is needed or it will be marked `deleted`)
if (this == &rhs)
return *this;
}
auto p = rhs.unsafeGetRawPtr();
if (p)
p->addStrongRef();
unsafeReleaseAndStore(p);
return *this;
}
template <class T>
SharedIntrusive<T>&
SharedIntrusive<T>::operator=(SharedIntrusive&& rhs)
{
if (this == &rhs)
return *this;
unsafeReleaseAndStore(rhs.unsafeExchange(nullptr));
return *this;
}
template <class T>
template <class TT>
// clang-format off
requires std::convertible_to<TT*, T*>
// clang-format on
SharedIntrusive<T>&
SharedIntrusive<T>::operator=(SharedIntrusive<TT>&& rhs)
{
static_assert(
!std::is_same_v<T, TT>,
"This overload should not be instantiated for T == TT");
unsafeReleaseAndStore(rhs.unsafeExchange(nullptr));
return *this;
}
template <class T>
bool
SharedIntrusive<T>::operator!=(std::nullptr_t) const
{
return this->get() != nullptr;
}
template <class T>
bool
SharedIntrusive<T>::operator==(std::nullptr_t) const
{
return this->get() == nullptr;
}
template <class T>
template <CAdoptTag TAdoptTag>
void
SharedIntrusive<T>::adopt(T* p)
{
if constexpr (std::is_same_v<
TAdoptTag,
SharedIntrusiveAdoptIncrementStrongTag>)
{
if (p)
p->addStrongRef();
}
unsafeReleaseAndStore(p);
}
template <class T>
SharedIntrusive<T>::~SharedIntrusive()
{
unsafeReleaseAndStore(nullptr);
};
template <class T>
template <class TT>
SharedIntrusive<T>::SharedIntrusive(
StaticCastTagSharedIntrusive,
SharedIntrusive<TT> const& rhs)
: ptr_{[&] {
auto p = static_cast<T*>(rhs.unsafeGetRawPtr());
if (p)
p->addStrongRef();
return p;
}()}
{
}
template <class T>
template <class TT>
SharedIntrusive<T>::SharedIntrusive(
StaticCastTagSharedIntrusive,
SharedIntrusive<TT>&& rhs)
: ptr_{static_cast<T*>(rhs.unsafeExchange(nullptr))}
{
}
template <class T>
template <class TT>
SharedIntrusive<T>::SharedIntrusive(
DynamicCastTagSharedIntrusive,
SharedIntrusive<TT> const& rhs)
: ptr_{[&] {
auto p = dynamic_cast<T*>(rhs.unsafeGetRawPtr());
if (p)
p->addStrongRef();
return p;
}()}
{
}
template <class T>
template <class TT>
SharedIntrusive<T>::SharedIntrusive(
DynamicCastTagSharedIntrusive,
SharedIntrusive<TT>&& rhs)
{
// This can be simplified without the `exchange`, but the `exchange` is kept
// in anticipation of supporting atomic operations.
auto toSet = rhs.unsafeExchange(nullptr);
if (toSet)
{
ptr_ = dynamic_cast<T*>(toSet);
if (!ptr_)
// need to set the pointer back or will leak
rhs.unsafeExchange(toSet);
}
}
template <class T>
T&
SharedIntrusive<T>::operator*() const noexcept
{
return *unsafeGetRawPtr();
}
template <class T>
T*
SharedIntrusive<T>::operator->() const noexcept
{
return unsafeGetRawPtr();
}
template <class T>
SharedIntrusive<T>::operator bool() const noexcept
{
return bool(unsafeGetRawPtr());
}
template <class T>
void
SharedIntrusive<T>::reset()
{
unsafeReleaseAndStore(nullptr);
}
template <class T>
T*
SharedIntrusive<T>::get() const
{
return unsafeGetRawPtr();
}
template <class T>
std::size_t
SharedIntrusive<T>::use_count() const
{
if (auto p = unsafeGetRawPtr())
return p->use_count();
return 0;
}
template <class T>
T*
SharedIntrusive<T>::unsafeGetRawPtr() const
{
return ptr_;
}
template <class T>
void
SharedIntrusive<T>::unsafeSetRawPtr(T* p)
{
ptr_ = p;
}
template <class T>
T*
SharedIntrusive<T>::unsafeExchange(T* p)
{
return std::exchange(ptr_, p);
}
template <class T>
void
SharedIntrusive<T>::unsafeReleaseAndStore(T* next)
{
auto prev = unsafeExchange(next);
if (!prev)
return;
using enum ReleaseStrongRefAction;
auto action = prev->releaseStrongRef();
switch (action)
{
case noop:
break;
case destroy:
delete prev;
break;
case partialDestroy:
prev->partialDestructor();
partialDestructorFinished(&prev);
// prev is null and may no longer be used
break;
}
}
//------------------------------------------------------------------------------
template <class T>
WeakIntrusive<T>::WeakIntrusive(WeakIntrusive const& rhs) : ptr_{rhs.ptr_}
{
if (ptr_)
ptr_->addWeakRef();
}
template <class T>
WeakIntrusive<T>::WeakIntrusive(WeakIntrusive&& rhs) : ptr_{rhs.ptr_}
{
rhs.ptr_ = nullptr;
}
template <class T>
WeakIntrusive<T>::WeakIntrusive(SharedIntrusive<T> const& rhs)
: ptr_{rhs.unsafeGetRawPtr()}
{
if (ptr_)
ptr_->addWeakRef();
}
template <class T>
template <class TT>
// clang-format off
requires std::convertible_to<TT*, T*>
// clang-format on
WeakIntrusive<T>&
WeakIntrusive<T>::operator=(SharedIntrusive<TT> const& rhs)
{
unsafeReleaseNoStore();
auto p = rhs.unsafeGetRawPtr();
if (p)
p->addWeakRef();
return *this;
}
template <class T>
void
WeakIntrusive<T>::adopt(T* ptr)
{
unsafeReleaseNoStore();
if (ptr)
ptr->addWeakRef();
ptr_ = ptr;
}
template <class T>
WeakIntrusive<T>::~WeakIntrusive()
{
unsafeReleaseNoStore();
}
template <class T>
SharedIntrusive<T>
WeakIntrusive<T>::lock() const
{
if (ptr_ && ptr_->checkoutStrongRefFromWeak())
{
return SharedIntrusive<T>{ptr_, SharedIntrusiveAdoptNoIncrementTag{}};
}
return {};
}
template <class T>
bool
WeakIntrusive<T>::expired() const
{
return (!ptr_ || ptr_->expired());
}
template <class T>
void
WeakIntrusive<T>::reset()
{
unsafeReleaseNoStore();
ptr_ = nullptr;
}
template <class T>
void
WeakIntrusive<T>::unsafeReleaseNoStore()
{
if (!ptr_)
return;
using enum ReleaseWeakRefAction;
auto action = ptr_->releaseWeakRef();
switch (action)
{
case noop:
break;
case destroy:
delete ptr_;
break;
}
}
//------------------------------------------------------------------------------
template <class T>
SharedWeakUnion<T>::SharedWeakUnion(SharedWeakUnion const& rhs) : tp_{rhs.tp_}
{
auto p = rhs.unsafeGetRawPtr();
if (!p)
return;
if (rhs.isStrong())
p->addStrongRef();
else
p->addWeakRef();
}
template <class T>
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakUnion<T>::SharedWeakUnion(SharedIntrusive<TT> const& rhs)
{
auto p = rhs.unsafeGetRawPtr();
if (p)
p->addStrongRef();
unsafeSetRawPtr(p, RefStrength::strong);
}
template <class T>
SharedWeakUnion<T>::SharedWeakUnion(SharedWeakUnion&& rhs) : tp_{rhs.tp_}
{
rhs.unsafeSetRawPtr(nullptr);
}
template <class T>
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakUnion<T>::SharedWeakUnion(SharedIntrusive<TT>&& rhs)
{
auto p = rhs.unsafeGetRawPtr();
if (p)
unsafeSetRawPtr(p, RefStrength::strong);
rhs.unsafeSetRawPtr(nullptr);
}
template <class T>
SharedWeakUnion<T>&
SharedWeakUnion<T>::operator=(SharedWeakUnion const& rhs)
{
if (this == &rhs)
return *this;
unsafeReleaseNoStore();
if (auto p = rhs.unsafeGetRawPtr())
{
if (rhs.isStrong())
{
p->addStrongRef();
unsafeSetRawPtr(p, RefStrength::strong);
}
else
{
p->addWeakRef();
unsafeSetRawPtr(p, RefStrength::weak);
}
}
else
{
unsafeSetRawPtr(nullptr);
}
return *this;
}
template <class T>
template <class TT>
// clang-format off
requires std::convertible_to<TT*, T*>
// clang-format on
SharedWeakUnion<T>&
SharedWeakUnion<T>::operator=(SharedIntrusive<TT> const& rhs)
{
unsafeReleaseNoStore();
auto p = rhs.unsafeGetRawPtr();
if (p)
p->addStrongRef();
unsafeSetRawPtr(p, RefStrength::strong);
return *this;
}
template <class T>
template <class TT>
// clang-format off
requires std::convertible_to<TT*, T*>
// clang-format on
SharedWeakUnion<T>&
SharedWeakUnion<T>::operator=(SharedIntrusive<TT>&& rhs)
{
unsafeReleaseNoStore();
unsafeSetRawPtr(rhs.unsafeGetRawPtr(), RefStrength::strong);
rhs.unsafeSetRawPtr(nullptr);
return *this;
}
template <class T>
SharedWeakUnion<T>::~SharedWeakUnion()
{
unsafeReleaseNoStore();
};
// Return a strong pointer if this is already a strong pointer (i.e. don't
// lock the weak pointer. Use the `lock` method if that's what's needed)
template <class T>
SharedIntrusive<T>
SharedWeakUnion<T>::getStrong() const
{
SharedIntrusive<T> result;
auto p = unsafeGetRawPtr();
if (p && isStrong())
{
result.template adopt<SharedIntrusiveAdoptIncrementStrongTag>(p);
}
return result;
}
template <class T>
SharedWeakUnion<T>::operator bool() const noexcept
{
return bool(get());
}
template <class T>
void
SharedWeakUnion<T>::reset()
{
unsafeReleaseNoStore();
unsafeSetRawPtr(nullptr);
}
template <class T>
T*
SharedWeakUnion<T>::get() const
{
return isStrong() ? unsafeGetRawPtr() : nullptr;
}
template <class T>
std::size_t
SharedWeakUnion<T>::use_count() const
{
if (auto p = get())
return p->use_count();
return 0;
}
template <class T>
bool
SharedWeakUnion<T>::expired() const
{
auto p = unsafeGetRawPtr();
return (!p || p->expired());
}
template <class T>
SharedIntrusive<T>
SharedWeakUnion<T>::lock() const
{
SharedIntrusive<T> result;
auto p = unsafeGetRawPtr();
if (!p)
return result;
if (isStrong())
{
result.template adopt<SharedIntrusiveAdoptIncrementStrongTag>(p);
return result;
}
if (p->checkoutStrongRefFromWeak())
{
result.template adopt<SharedIntrusiveAdoptNoIncrementTag>(p);
return result;
}
return result;
}
template <class T>
bool
SharedWeakUnion<T>::isStrong() const
{
return !(tp_ & tagMask);
}
template <class T>
bool
SharedWeakUnion<T>::isWeak() const
{
return tp_ & tagMask;
}
template <class T>
bool
SharedWeakUnion<T>::convertToStrong()
{
if (isStrong())
return true;
auto p = unsafeGetRawPtr();
if (p && p->checkoutStrongRefFromWeak())
{
[[maybe_unused]] auto action = p->releaseWeakRef();
XRPL_ASSERT(
(action == ReleaseWeakRefAction::noop),
"ripple::SharedWeakUnion::convertToStrong : "
"action is noop");
unsafeSetRawPtr(p, RefStrength::strong);
return true;
}
return false;
}
template <class T>
bool
SharedWeakUnion<T>::convertToWeak()
{
if (isWeak())
return true;
auto p = unsafeGetRawPtr();
if (!p)
return false;
using enum ReleaseStrongRefAction;
auto action = p->addWeakReleaseStrongRef();
switch (action)
{
case noop:
break;
case destroy:
// We just added a weak ref. How could we destroy?
UNREACHABLE(
"ripple::SharedWeakUnion::convertToWeak : destroying freshly "
"added ref");
delete p;
unsafeSetRawPtr(nullptr);
return true; // Should never happen
case partialDestroy:
// This is a weird case. We just converted the last strong
// pointer to a weak pointer.
p->partialDestructor();
partialDestructorFinished(&p);
// p is null and may no longer be used
break;
}
unsafeSetRawPtr(p, RefStrength::weak);
return true;
}
template <class T>
T*
SharedWeakUnion<T>::unsafeGetRawPtr() const
{
return reinterpret_cast<T*>(tp_ & ptrMask);
}
template <class T>
void
SharedWeakUnion<T>::unsafeSetRawPtr(T* p, RefStrength rs)
{
tp_ = reinterpret_cast<std::uintptr_t>(p);
if (tp_ && rs == RefStrength::weak)
tp_ |= tagMask;
}
template <class T>
void
SharedWeakUnion<T>::unsafeSetRawPtr(std::nullptr_t)
{
tp_ = 0;
}
template <class T>
void
SharedWeakUnion<T>::unsafeReleaseNoStore()
{
auto p = unsafeGetRawPtr();
if (!p)
return;
if (isStrong())
{
using enum ReleaseStrongRefAction;
auto strongAction = p->releaseStrongRef();
switch (strongAction)
{
case noop:
break;
case destroy:
delete p;
break;
case partialDestroy:
p->partialDestructor();
partialDestructorFinished(&p);
// p is null and may no longer be used
break;
}
}
else
{
using enum ReleaseWeakRefAction;
auto weakAction = p->releaseWeakRef();
switch (weakAction)
{
case noop:
break;
case destroy:
delete p;
break;
}
}
}
} // namespace ripple
#endif

View File

@@ -1,502 +0,0 @@
//------------------------------------------------------------------------------
/*
This file is part of rippled: https://github.com/ripple/rippled
Copyright (c) 2023 Ripple Labs Inc.
Permission to use, copy, modify, and/or distribute this software for any
purpose with or without fee is hereby granted, provided that the above
copyright notice and this permission notice appear in all copies.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
ANY SPECIAL , DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
//==============================================================================
#ifndef RIPPLE_BASICS_INTRUSIVEREFCOUNTS_H_INCLUDED
#define RIPPLE_BASICS_INTRUSIVEREFCOUNTS_H_INCLUDED
#include <xrpl/beast/utility/instrumentation.h>
#include <atomic>
#include <cstdint>
namespace ripple {
/** Action to perform when releasing a strong pointer.
noop: Do nothing. For example, a `noop` action will occur when a count is
decremented to a non-zero value.
partialDestroy: Run the `partialDestructor`. This action will happen when a
strong count is decremented to zero and the weak count is non-zero.
destroy: Run the destructor. This action will occur when either the strong
count or weak count is decremented and the other count is also zero.
*/
enum class ReleaseStrongRefAction { noop, partialDestroy, destroy };
/** Action to perform when releasing a weak pointer.
noop: Do nothing. For example, a `noop` action will occur when a count is
decremented to a non-zero value.
destroy: Run the destructor. This action will occur when either the strong
count or weak count is decremented and the other count is also zero.
*/
enum class ReleaseWeakRefAction { noop, destroy };
/** Implement the strong count, weak count, and bit flags for an intrusive
pointer.
A class can satisfy the requirements of a ripple::IntrusivePointer by
inheriting from this class.
*/
struct IntrusiveRefCounts
{
virtual ~IntrusiveRefCounts() noexcept;
// This must be `noexcept` or the make_SharedIntrusive function could leak
// memory.
void
addStrongRef() const noexcept;
void
addWeakRef() const noexcept;
ReleaseStrongRefAction
releaseStrongRef() const;
// Same as:
// {
// addWeakRef();
// return releaseStrongRef;
// }
// done as one atomic operation
ReleaseStrongRefAction
addWeakReleaseStrongRef() const;
ReleaseWeakRefAction
releaseWeakRef() const;
// Returns true is able to checkout a strong ref. False otherwise
bool
checkoutStrongRefFromWeak() const noexcept;
bool
expired() const noexcept;
std::size_t
use_count() const noexcept;
// This function MUST be called after a partial destructor finishes running.
// Calling this function may cause other threads to delete the object
// pointed to by `o`, so `o` should never be used after calling this
// function. The parameter will be set to a `nullptr` after calling this
// function to emphasize that it should not be used.
// Note: This is intentionally NOT called at the end of `partialDestructor`.
// The reason for this is if new classes are written to support this smart
// pointer class, they need to write their own `partialDestructor` function
// and ensure `partialDestructorFinished` is called at the end. Putting this
// call inside the smart pointer class itself is expected to be less error
// prone.
// Note: The "two-star" programming is intentional. It emphasizes that `o`
// may be deleted and the unergonomic API is meant to signal the special
// nature of this function call to callers.
// Note: This is a template to support incompletely defined classes.
template <class T>
friend void
partialDestructorFinished(T** o);
private:
// TODO: We may need to use a uint64_t for both counts. This will reduce the
// memory savings. We need to audit the code to make sure 16 bit counts are
// enough for strong pointers and 14 bit counts are enough for weak
// pointers. Use type aliases to make it easy to switch types.
using CountType = std::uint16_t;
static constexpr size_t StrongCountNumBits = sizeof(CountType) * 8;
static constexpr size_t WeakCountNumBits = StrongCountNumBits - 2;
using FieldType = std::uint32_t;
static constexpr size_t FieldTypeBits = sizeof(FieldType) * 8;
static constexpr FieldType one = 1;
/** `refCounts` consists of four fields that are treated atomically:
1. Strong count. This is a count of the number of shared pointers that
hold a reference to this object. When the strong counts goes to zero,
if the weak count is zero, the destructor is run. If the weak count is
non-zero when the strong count goes to zero then the partialDestructor
is run.
2. Weak count. This is a count of the number of weak pointer that hold
a reference to this object. When the weak count goes to zero and the
strong count is also zero, then the destructor is run.
3. Partial destroy started bit. This bit is set if the
`partialDestructor` function has been started (or is about to be
started). This is used to prevent the destructor from running
concurrently with the partial destructor. This can easily happen when
the last strong pointer release its reference in one thread and starts
the partialDestructor, while in another thread the last weak pointer
goes out of scope and starts the destructor while the partialDestructor
is still running. Both a start and finished bit is needed to handle a
corner-case where the last strong pointer goes out of scope, then then
last `weakPointer` goes out of scope, but this happens before the
`partialDestructor` bit is set. It would be possible to use a single
bit if it could also be set atomically when the strong count goes to
zero and the weak count is non-zero, but that would add complexity (and
likely slow down common cases as well).
4. Partial destroy finished bit. This bit is set when the
`partialDestructor` has finished running. See (3) above for more
information.
*/
mutable std::atomic<FieldType> refCounts{strongDelta};
/** Amount to change the strong count when adding or releasing a reference
Note: The strong count is stored in the low `StrongCountNumBits` bits
of refCounts
*/
static constexpr FieldType strongDelta = 1;
/** Amount to change the weak count when adding or releasing a reference
Note: The weak count is stored in the high `WeakCountNumBits` bits of
refCounts
*/
static constexpr FieldType weakDelta = (one << StrongCountNumBits);
/** Flag that is set when the partialDestroy function has started running
(or is about to start running).
See description of the `refCounts` field for a fuller description of
this field.
*/
static constexpr FieldType partialDestroyStartedMask =
(one << (FieldTypeBits - 1));
/** Flag that is set when the partialDestroy function has finished running
See description of the `refCounts` field for a fuller description of
this field.
*/
static constexpr FieldType partialDestroyFinishedMask =
(one << (FieldTypeBits - 2));
/** Mask that will zero out all the `count` bits and leave the tag bits
unchanged.
*/
static constexpr FieldType tagMask =
partialDestroyStartedMask | partialDestroyFinishedMask;
/** Mask that will zero out the `tag` bits and leave the count bits
unchanged.
*/
static constexpr FieldType valueMask = ~tagMask;
/** Mask that will zero out everything except the strong count.
*/
static constexpr FieldType strongMask =
((one << StrongCountNumBits) - 1) & valueMask;
/** Mask that will zero out everything except the weak count.
*/
static constexpr FieldType weakMask =
(((one << WeakCountNumBits) - 1) << StrongCountNumBits) & valueMask;
/** Unpack the count and tag fields from the packed atomic integer form. */
struct RefCountPair
{
CountType strong;
CountType weak;
/** The `partialDestroyStartedBit` is set to on when the partial
destroy function is started. It is not a boolean; it is a uint32
with all bits zero with the possible exception of the
`partialDestroyStartedMask` bit. This is done so it can be directly
masked into the `combinedValue`.
*/
FieldType partialDestroyStartedBit{0};
/** The `partialDestroyFinishedBit` is set to on when the partial
destroy function has finished.
*/
FieldType partialDestroyFinishedBit{0};
RefCountPair(FieldType v) noexcept;
RefCountPair(CountType s, CountType w) noexcept;
/** Convert back to the packed integer form. */
FieldType
combinedValue() const noexcept;
static constexpr CountType maxStrongValue =
static_cast<CountType>((one << StrongCountNumBits) - 1);
static constexpr CountType maxWeakValue =
static_cast<CountType>((one << WeakCountNumBits) - 1);
/** Put an extra margin to detect when running up against limits.
This is only used in debug code, and is useful if we reduce the
number of bits in the strong and weak counts (to 16 and 14 bits).
*/
static constexpr CountType checkStrongMaxValue = maxStrongValue - 32;
static constexpr CountType checkWeakMaxValue = maxWeakValue - 32;
};
};
inline void
IntrusiveRefCounts::addStrongRef() const noexcept
{
refCounts.fetch_add(strongDelta, std::memory_order_acq_rel);
}
inline void
IntrusiveRefCounts::addWeakRef() const noexcept
{
refCounts.fetch_add(weakDelta, std::memory_order_acq_rel);
}
inline ReleaseStrongRefAction
IntrusiveRefCounts::releaseStrongRef() const
{
// Subtract `strongDelta` from refCounts. If this releases the last strong
// ref, set the `partialDestroyStarted` bit. It is important that the ref
// count and the `partialDestroyStartedBit` are changed atomically (hence
// the loop and `compare_exchange` op). If this didn't need to be done
// atomically, the loop could be replaced with a `fetch_sub` and a
// conditional `fetch_or`. This loop will almost always run once.
using enum ReleaseStrongRefAction;
auto prevIntVal = refCounts.load(std::memory_order_acquire);
while (1)
{
RefCountPair const prevVal{prevIntVal};
XRPL_ASSERT(
(prevVal.strong >= strongDelta),
"ripple::IntrusiveRefCounts::releaseStrongRef : previous ref "
"higher than new");
auto nextIntVal = prevIntVal - strongDelta;
ReleaseStrongRefAction action = noop;
if (prevVal.strong == 1)
{
if (prevVal.weak == 0)
{
action = destroy;
}
else
{
nextIntVal |= partialDestroyStartedMask;
action = partialDestroy;
}
}
if (refCounts.compare_exchange_weak(
prevIntVal, nextIntVal, std::memory_order_acq_rel))
{
// Can't be in partial destroy because only decrementing the strong
// count to zero can start a partial destroy, and that can't happen
// twice.
XRPL_ASSERT(
(action == noop) || !(prevIntVal & partialDestroyStartedMask),
"ripple::IntrusiveRefCounts::releaseStrongRef : not in partial "
"destroy");
return action;
}
}
}
inline ReleaseStrongRefAction
IntrusiveRefCounts::addWeakReleaseStrongRef() const
{
using enum ReleaseStrongRefAction;
static_assert(weakDelta > strongDelta);
auto constexpr delta = weakDelta - strongDelta;
auto prevIntVal = refCounts.load(std::memory_order_acquire);
// This loop will almost always run once. The loop is needed to atomically
// change the counts and flags (the count could be atomically changed, but
// the flags depend on the current value of the counts).
//
// Note: If this becomes a perf bottleneck, the `partialDestoryStartedMask`
// may be able to be set non-atomically. But it is easier to reason about
// the code if the flag is set atomically.
while (1)
{
RefCountPair const prevVal{prevIntVal};
// Converted the last strong pointer to a weak pointer.
//
// Can't be in partial destroy because only decrementing the
// strong count to zero can start a partial destroy, and that
// can't happen twice.
XRPL_ASSERT(
(!prevVal.partialDestroyStartedBit),
"ripple::IntrusiveRefCounts::addWeakReleaseStrongRef : not in "
"partial destroy");
auto nextIntVal = prevIntVal + delta;
ReleaseStrongRefAction action = noop;
if (prevVal.strong == 1)
{
if (prevVal.weak == 0)
{
action = noop;
}
else
{
nextIntVal |= partialDestroyStartedMask;
action = partialDestroy;
}
}
if (refCounts.compare_exchange_weak(
prevIntVal, nextIntVal, std::memory_order_acq_rel))
{
XRPL_ASSERT(
(!(prevIntVal & partialDestroyStartedMask)),
"ripple::IntrusiveRefCounts::addWeakReleaseStrongRef : not "
"started partial destroy");
return action;
}
}
}
inline ReleaseWeakRefAction
IntrusiveRefCounts::releaseWeakRef() const
{
auto prevIntVal = refCounts.fetch_sub(weakDelta, std::memory_order_acq_rel);
RefCountPair prev = prevIntVal;
if (prev.weak == 1 && prev.strong == 0)
{
if (!prev.partialDestroyStartedBit)
{
// This case should only be hit if the partialDestroyStartedBit is
// set non-atomically (and even then very rarely). The code is kept
// in case we need to set the flag non-atomically for perf reasons.
refCounts.wait(prevIntVal, std::memory_order_acquire);
prevIntVal = refCounts.load(std::memory_order_acquire);
prev = RefCountPair{prevIntVal};
}
if (!prev.partialDestroyFinishedBit)
{
// partial destroy MUST finish before running a full destroy (when
// using weak pointers)
refCounts.wait(prevIntVal - weakDelta, std::memory_order_acquire);
}
return ReleaseWeakRefAction::destroy;
}
return ReleaseWeakRefAction::noop;
}
inline bool
IntrusiveRefCounts::checkoutStrongRefFromWeak() const noexcept
{
auto curValue = RefCountPair{1, 1}.combinedValue();
auto desiredValue = RefCountPair{2, 1}.combinedValue();
while (!refCounts.compare_exchange_weak(
curValue, desiredValue, std::memory_order_acq_rel))
{
RefCountPair const prev{curValue};
if (!prev.strong)
return false;
desiredValue = curValue + strongDelta;
}
return true;
}
inline bool
IntrusiveRefCounts::expired() const noexcept
{
RefCountPair const val = refCounts.load(std::memory_order_acquire);
return val.strong == 0;
}
inline std::size_t
IntrusiveRefCounts::use_count() const noexcept
{
RefCountPair const val = refCounts.load(std::memory_order_acquire);
return val.strong;
}
inline IntrusiveRefCounts::~IntrusiveRefCounts() noexcept
{
#ifndef NDEBUG
auto v = refCounts.load(std::memory_order_acquire);
XRPL_ASSERT(
(!(v & valueMask)),
"ripple::IntrusiveRefCounts::~IntrusiveRefCounts : count must be zero");
auto t = v & tagMask;
XRPL_ASSERT(
(!t || t == tagMask),
"ripple::IntrusiveRefCounts::~IntrusiveRefCounts : valid tag");
#endif
}
//------------------------------------------------------------------------------
inline IntrusiveRefCounts::RefCountPair::RefCountPair(
IntrusiveRefCounts::FieldType v) noexcept
: strong{static_cast<CountType>(v & strongMask)}
, weak{static_cast<CountType>((v & weakMask) >> StrongCountNumBits)}
, partialDestroyStartedBit{v & partialDestroyStartedMask}
, partialDestroyFinishedBit{v & partialDestroyFinishedMask}
{
XRPL_ASSERT(
(strong < checkStrongMaxValue && weak < checkWeakMaxValue),
"ripple::IntrusiveRefCounts::RefCountPair(FieldType) : inputs inside "
"range");
}
inline IntrusiveRefCounts::RefCountPair::RefCountPair(
IntrusiveRefCounts::CountType s,
IntrusiveRefCounts::CountType w) noexcept
: strong{s}, weak{w}
{
XRPL_ASSERT(
(strong < checkStrongMaxValue && weak < checkWeakMaxValue),
"ripple::IntrusiveRefCounts::RefCountPair(CountType, CountType) : "
"inputs inside range");
}
inline IntrusiveRefCounts::FieldType
IntrusiveRefCounts::RefCountPair::combinedValue() const noexcept
{
XRPL_ASSERT(
(strong < checkStrongMaxValue && weak < checkWeakMaxValue),
"ripple::IntrusiveRefCounts::RefCountPair::combinedValue : inputs "
"inside range");
return (static_cast<IntrusiveRefCounts::FieldType>(weak)
<< IntrusiveRefCounts::StrongCountNumBits) |
static_cast<IntrusiveRefCounts::FieldType>(strong) |
partialDestroyStartedBit | partialDestroyFinishedBit;
}
template <class T>
inline void
partialDestructorFinished(T** o)
{
T& self = **o;
IntrusiveRefCounts::RefCountPair p =
self.refCounts.fetch_or(IntrusiveRefCounts::partialDestroyFinishedMask);
XRPL_ASSERT(
(!p.partialDestroyFinishedBit && p.partialDestroyStartedBit &&
!p.strong),
"ripple::partialDestructorFinished : not a weak ref");
if (!p.weak)
{
// There was a weak count before the partial destructor ran (or we would
// have run the full destructor) and now there isn't a weak count. Some
// thread is waiting to run the destructor.
self.refCounts.notify_one();
}
// Set the pointer to null to emphasize that the object shouldn't be used
// after calling this function as it may be destroyed in another thread.
*o = nullptr;
}
//------------------------------------------------------------------------------
} // namespace ripple
#endif

View File

@@ -21,7 +21,6 @@
#define RIPPLE_BASICS_LOCALVALUE_H_INCLUDED
#include <boost/thread/tss.hpp>
#include <memory>
#include <unordered_map>

View File

@@ -22,10 +22,8 @@
#include <xrpl/basics/UnorderedContainers.h>
#include <xrpl/beast/utility/Journal.h>
#include <boost/beast/core/string.hpp>
#include <boost/filesystem.hpp>
#include <map>
#include <memory>
#include <mutex>
@@ -69,10 +67,6 @@ private:
void
write(beast::severities::Severity level, std::string const& text)
override;
void
writeAlways(beast::severities::Severity level, std::string const& text)
override;
};
/** Manages a system file containing logged output.
@@ -262,14 +256,6 @@ private:
x
#endif
#ifndef CLOG
#define CLOG(ss) \
if (!ss) \
; \
else \
*ss
#endif
//------------------------------------------------------------------------------
// Debug logging:

View File

@@ -20,11 +20,11 @@
#ifndef RIPPLE_BASICS_RESOLVER_H_INCLUDED
#define RIPPLE_BASICS_RESOLVER_H_INCLUDED
#include <xrpl/beast/net/IPEndpoint.h>
#include <functional>
#include <vector>
#include <xrpl/beast/net/IPEndpoint.h>
namespace ripple {
class Resolver

View File

@@ -22,7 +22,6 @@
#include <xrpl/basics/Resolver.h>
#include <xrpl/beast/utility/Journal.h>
#include <boost/asio/io_service.hpp>
namespace ripple {

View File

@@ -1,135 +0,0 @@
//------------------------------------------------------------------------------
/*
This file is part of rippled: https://github.com/ripple/rippled
Copyright (c) 2023 Ripple Labs Inc.
Permission to use, copy, modify, and/or distribute this software for any
purpose with or without fee is hereby granted, provided that the above
copyright notice and this permission notice appear in all copies.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
ANY SPECIAL , DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
//==============================================================================
#ifndef RIPPLE_BASICS_SHAREDWEAKCACHEPOINTER_H_INCLUDED
#define RIPPLE_BASICS_SHAREDWEAKCACHEPOINTER_H_INCLUDED
#include <memory>
#include <variant>
namespace ripple {
/** A combination of a std::shared_ptr and a std::weak_pointer.
This class is a wrapper to a `std::variant<std::shared_ptr,std::weak_ptr>`
This class is useful for storing intrusive pointers in tagged caches using less
memory than storing both pointers directly.
*/
template <class T>
class SharedWeakCachePointer
{
public:
SharedWeakCachePointer() = default;
SharedWeakCachePointer(SharedWeakCachePointer const& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakCachePointer(std::shared_ptr<TT> const& rhs);
SharedWeakCachePointer(SharedWeakCachePointer&& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakCachePointer(std::shared_ptr<TT>&& rhs);
SharedWeakCachePointer&
operator=(SharedWeakCachePointer const& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakCachePointer&
operator=(std::shared_ptr<TT> const& rhs);
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakCachePointer&
operator=(std::shared_ptr<TT>&& rhs);
~SharedWeakCachePointer();
/** Return a strong pointer if this is already a strong pointer (i.e. don't
lock the weak pointer. Use the `lock` method if that's what's needed)
*/
std::shared_ptr<T> const&
getStrong() const;
/** Return true if this is a strong pointer and the strong pointer is
seated.
*/
explicit
operator bool() const noexcept;
/** Set the pointer to null, decrement the appropriate ref count, and run
the appropriate release action.
*/
void
reset();
/** If this is a strong pointer, return the raw pointer. Otherwise return
null.
*/
T*
get() const;
/** If this is a strong pointer, return the strong count. Otherwise return 0
*/
std::size_t
use_count() const;
/** Return true if there is a non-zero strong count. */
bool
expired() const;
/** If this is a strong pointer, return the strong pointer. Otherwise
attempt to lock the weak pointer.
*/
std::shared_ptr<T>
lock() const;
/** Return true is this represents a strong pointer. */
bool
isStrong() const;
/** Return true is this represents a weak pointer. */
bool
isWeak() const;
/** If this is a weak pointer, attempt to convert it to a strong pointer.
@return true if successfully converted to a strong pointer (or was
already a strong pointer). Otherwise false.
*/
bool
convertToStrong();
/** If this is a strong pointer, attempt to convert it to a weak pointer.
@return false if the pointer is null. Otherwise return true.
*/
bool
convertToWeak();
private:
std::variant<std::shared_ptr<T>, std::weak_ptr<T>> combo_;
};
} // namespace ripple
#endif

View File

@@ -1,192 +0,0 @@
//------------------------------------------------------------------------------
/*
This file is part of rippled: https://github.com/ripple/rippled
Copyright (c) 2023 Ripple Labs Inc.
Permission to use, copy, modify, and/or distribute this software for any
purpose with or without fee is hereby granted, provided that the above
copyright notice and this permission notice appear in all copies.
THE SOFTWARE IS PROVIDED "AS IS" AND THE AUTHOR DISCLAIMS ALL WARRANTIES
WITH REGARD TO THIS SOFTWARE INCLUDING ALL IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS. IN NO EVENT SHALL THE AUTHOR BE LIABLE FOR
ANY SPECIAL , DIRECT, INDIRECT, OR CONSEQUENTIAL DAMAGES OR ANY DAMAGES
WHATSOEVER RESULTING FROM LOSS OF USE, DATA OR PROFITS, WHETHER IN AN
ACTION OF CONTRACT, NEGLIGENCE OR OTHER TORTIOUS ACTION, ARISING OUT OF
OR IN CONNECTION WITH THE USE OR PERFORMANCE OF THIS SOFTWARE.
*/
//==============================================================================
#ifndef RIPPLE_BASICS_SHAREDWEAKCACHEPOINTER_IPP_INCLUDED
#define RIPPLE_BASICS_SHAREDWEAKCACHEPOINTER_IPP_INCLUDED
#include <xrpl/basics/SharedWeakCachePointer.h>
namespace ripple {
template <class T>
SharedWeakCachePointer<T>::SharedWeakCachePointer(
SharedWeakCachePointer const& rhs) = default;
template <class T>
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakCachePointer<T>::SharedWeakCachePointer(
std::shared_ptr<TT> const& rhs)
: combo_{rhs}
{
}
template <class T>
SharedWeakCachePointer<T>::SharedWeakCachePointer(
SharedWeakCachePointer&& rhs) = default;
template <class T>
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakCachePointer<T>::SharedWeakCachePointer(std::shared_ptr<TT>&& rhs)
: combo_{std::move(rhs)}
{
}
template <class T>
SharedWeakCachePointer<T>&
SharedWeakCachePointer<T>::operator=(SharedWeakCachePointer const& rhs) =
default;
template <class T>
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakCachePointer<T>&
SharedWeakCachePointer<T>::operator=(std::shared_ptr<TT> const& rhs)
{
combo_ = rhs;
return *this;
}
template <class T>
template <class TT>
requires std::convertible_to<TT*, T*>
SharedWeakCachePointer<T>&
SharedWeakCachePointer<T>::operator=(std::shared_ptr<TT>&& rhs)
{
combo_ = std::move(rhs);
return *this;
}
template <class T>
SharedWeakCachePointer<T>::~SharedWeakCachePointer() = default;
// Return a strong pointer if this is already a strong pointer (i.e. don't
// lock the weak pointer. Use the `lock` method if that's what's needed)
template <class T>
std::shared_ptr<T> const&
SharedWeakCachePointer<T>::getStrong() const
{
static std::shared_ptr<T> const empty;
if (auto p = std::get_if<std::shared_ptr<T>>(&combo_))
return *p;
return empty;
}
template <class T>
SharedWeakCachePointer<T>::operator bool() const noexcept
{
return !!std::get_if<std::shared_ptr<T>>(&combo_);
}
template <class T>
void
SharedWeakCachePointer<T>::reset()
{
combo_ = std::shared_ptr<T>{};
}
template <class T>
T*
SharedWeakCachePointer<T>::get() const
{
return std::get_if<std::shared_ptr<T>>(&combo_).get();
}
template <class T>
std::size_t
SharedWeakCachePointer<T>::use_count() const
{
if (auto p = std::get_if<std::shared_ptr<T>>(&combo_))
return p->use_count();
return 0;
}
template <class T>
bool
SharedWeakCachePointer<T>::expired() const
{
if (auto p = std::get_if<std::weak_ptr<T>>(&combo_))
return p->expired();
return !std::get_if<std::shared_ptr<T>>(&combo_);
}
template <class T>
std::shared_ptr<T>
SharedWeakCachePointer<T>::lock() const
{
if (auto p = std::get_if<std::shared_ptr<T>>(&combo_))
return *p;
if (auto p = std::get_if<std::weak_ptr<T>>(&combo_))
return p->lock();
return {};
}
template <class T>
bool
SharedWeakCachePointer<T>::isStrong() const
{
if (auto p = std::get_if<std::shared_ptr<T>>(&combo_))
return !!p->get();
return false;
}
template <class T>
bool
SharedWeakCachePointer<T>::isWeak() const
{
return !isStrong();
}
template <class T>
bool
SharedWeakCachePointer<T>::convertToStrong()
{
if (isStrong())
return true;
if (auto p = std::get_if<std::weak_ptr<T>>(&combo_))
{
if (auto s = p->lock())
{
combo_ = std::move(s);
return true;
}
}
return false;
}
template <class T>
bool
SharedWeakCachePointer<T>::convertToWeak()
{
if (isWeak())
return true;
if (auto p = std::get_if<std::shared_ptr<T>>(&combo_))
{
combo_ = std::weak_ptr<T>(*p);
return true;
}
return false;
}
} // namespace ripple
#endif

View File

@@ -23,7 +23,6 @@
#include <xrpl/basics/contract.h>
#include <xrpl/basics/strHex.h>
#include <xrpl/beast/utility/instrumentation.h>
#include <algorithm>
#include <array>
#include <cstdint>

View File

@@ -20,14 +20,11 @@
#ifndef RIPPLE_BASICS_TAGGEDCACHE_H_INCLUDED
#define RIPPLE_BASICS_TAGGEDCACHE_H_INCLUDED
#include <xrpl/basics/IntrusivePointer.h>
#include <xrpl/basics/Log.h>
#include <xrpl/basics/SharedWeakCachePointer.ipp>
#include <xrpl/basics/UnorderedContainers.h>
#include <xrpl/basics/hardened_hash.h>
#include <xrpl/beast/clock/abstract_clock.h>
#include <xrpl/beast/insight/Insight.h>
#include <atomic>
#include <functional>
#include <mutex>
@@ -53,8 +50,6 @@ template <
class Key,
class T,
bool IsKeyCache = false,
class SharedWeakUnionPointerType = SharedWeakCachePointer<T>,
class SharedPointerType = std::shared_ptr<T>,
class Hash = hardened_hash<>,
class KeyEqual = std::equal_to<Key>,
class Mutex = std::recursive_mutex>
@@ -65,8 +60,6 @@ public:
using key_type = Key;
using mapped_type = T;
using clock_type = beast::abstract_clock<std::chrono::steady_clock>;
using shared_weak_combo_pointer_type = SharedWeakUnionPointerType;
using shared_pointer_type = SharedPointerType;
public:
TaggedCache(
@@ -76,48 +69,231 @@ public:
clock_type& clock,
beast::Journal journal,
beast::insight::Collector::ptr const& collector =
beast::insight::NullCollector::New());
beast::insight::NullCollector::New())
: m_journal(journal)
, m_clock(clock)
, m_stats(
name,
std::bind(&TaggedCache::collect_metrics, this),
collector)
, m_name(name)
, m_target_size(size)
, m_target_age(expiration)
, m_cache_count(0)
, m_hits(0)
, m_misses(0)
{
}
public:
/** Return the clock associated with the cache. */
clock_type&
clock();
clock()
{
return m_clock;
}
/** Returns the number of items in the container. */
std::size_t
size() const;
size() const
{
std::lock_guard lock(m_mutex);
return m_cache.size();
}
void
setTargetSize(int s)
{
std::lock_guard lock(m_mutex);
m_target_size = s;
if (s > 0)
{
for (auto& partition : m_cache.map())
{
partition.rehash(static_cast<std::size_t>(
(s + (s >> 2)) /
(partition.max_load_factor() * m_cache.partitions()) +
1));
}
}
JLOG(m_journal.debug()) << m_name << " target size set to " << s;
}
clock_type::duration
getTargetAge() const
{
std::lock_guard lock(m_mutex);
return m_target_age;
}
void
setTargetAge(clock_type::duration s)
{
std::lock_guard lock(m_mutex);
m_target_age = s;
JLOG(m_journal.debug())
<< m_name << " target age set to " << m_target_age.count();
}
int
getCacheSize() const;
getCacheSize() const
{
std::lock_guard lock(m_mutex);
return m_cache_count;
}
int
getTrackSize() const;
getTrackSize() const
{
std::lock_guard lock(m_mutex);
return m_cache.size();
}
float
getHitRate();
getHitRate()
{
std::lock_guard lock(m_mutex);
auto const total = static_cast<float>(m_hits + m_misses);
return m_hits * (100.0f / std::max(1.0f, total));
}
void
clear();
clear()
{
std::lock_guard lock(m_mutex);
m_cache.clear();
m_cache_count = 0;
}
void
reset();
reset()
{
std::lock_guard lock(m_mutex);
m_cache.clear();
m_cache_count = 0;
m_hits = 0;
m_misses = 0;
}
/** Refresh the last access time on a key if present.
@return `true` If the key was found.
*/
template <class KeyComparable>
bool
touch_if_exists(KeyComparable const& key);
touch_if_exists(KeyComparable const& key)
{
std::lock_guard lock(m_mutex);
auto const iter(m_cache.find(key));
if (iter == m_cache.end())
{
++m_stats.misses;
return false;
}
iter->second.touch(m_clock.now());
++m_stats.hits;
return true;
}
using SweptPointersVector = std::vector<SharedWeakUnionPointerType>;
using SweptPointersVector = std::pair<
std::vector<std::shared_ptr<mapped_type>>,
std::vector<std::weak_ptr<mapped_type>>>;
void
sweep();
sweep()
{
// Keep references to all the stuff we sweep
// For performance, each worker thread should exit before the swept data
// is destroyed but still within the main cache lock.
std::vector<SweptPointersVector> allStuffToSweep(m_cache.partitions());
clock_type::time_point const now(m_clock.now());
clock_type::time_point when_expire;
auto const start = std::chrono::steady_clock::now();
{
std::lock_guard lock(m_mutex);
if (m_target_size == 0 ||
(static_cast<int>(m_cache.size()) <= m_target_size))
{
when_expire = now - m_target_age;
}
else
{
when_expire =
now - m_target_age * m_target_size / m_cache.size();
clock_type::duration const minimumAge(std::chrono::seconds(1));
if (when_expire > (now - minimumAge))
when_expire = now - minimumAge;
JLOG(m_journal.trace())
<< m_name << " is growing fast " << m_cache.size() << " of "
<< m_target_size << " aging at "
<< (now - when_expire).count() << " of "
<< m_target_age.count();
}
std::vector<std::thread> workers;
workers.reserve(m_cache.partitions());
std::atomic<int> allRemovals = 0;
for (std::size_t p = 0; p < m_cache.partitions(); ++p)
{
workers.push_back(sweepHelper(
when_expire,
now,
m_cache.map()[p],
allStuffToSweep[p],
allRemovals,
lock));
}
for (std::thread& worker : workers)
worker.join();
m_cache_count -= allRemovals;
}
// At this point allStuffToSweep will go out of scope outside the lock
// and decrement the reference count on each strong pointer.
JLOG(m_journal.debug())
<< m_name << " TaggedCache sweep lock duration "
<< std::chrono::duration_cast<std::chrono::milliseconds>(
std::chrono::steady_clock::now() - start)
.count()
<< "ms";
}
bool
del(key_type const& key, bool valid);
del(const key_type& key, bool valid)
{
// Remove from cache, if !valid, remove from map too. Returns true if
// removed from cache
std::lock_guard lock(m_mutex);
auto cit = m_cache.find(key);
if (cit == m_cache.end())
return false;
Entry& entry = cit->second;
bool ret = false;
if (entry.isCached())
{
--m_cache_count;
entry.ptr.reset();
ret = true;
}
if (!valid || entry.isExpired())
m_cache.erase(cit);
return ret;
}
public:
/** Replace aliased objects with originals.
Due to concurrency it is possible for two separate objects with
@@ -131,23 +307,100 @@ public:
@return `true` If the key already existed.
*/
template <class R>
public:
bool
canonicalize(
key_type const& key,
SharedPointerType& data,
R&& replaceCallback);
const key_type& key,
std::shared_ptr<T>& data,
std::function<bool(std::shared_ptr<T> const&)>&& replace)
{
// Return canonical value, store if needed, refresh in cache
// Return values: true=we had the data already
std::lock_guard lock(m_mutex);
auto cit = m_cache.find(key);
if (cit == m_cache.end())
{
m_cache.emplace(
std::piecewise_construct,
std::forward_as_tuple(key),
std::forward_as_tuple(m_clock.now(), data));
++m_cache_count;
return false;
}
Entry& entry = cit->second;
entry.touch(m_clock.now());
if (entry.isCached())
{
if (replace(entry.ptr))
{
entry.ptr = data;
entry.weak_ptr = data;
}
else
{
data = entry.ptr;
}
return true;
}
auto cachedData = entry.lock();
if (cachedData)
{
if (replace(entry.ptr))
{
entry.ptr = data;
entry.weak_ptr = data;
}
else
{
entry.ptr = cachedData;
data = cachedData;
}
++m_cache_count;
return true;
}
entry.ptr = data;
entry.weak_ptr = data;
++m_cache_count;
return false;
}
bool
canonicalize_replace_cache(
key_type const& key,
SharedPointerType const& data);
const key_type& key,
std::shared_ptr<T> const& data)
{
return canonicalize(
key,
const_cast<std::shared_ptr<T>&>(data),
[](std::shared_ptr<T> const&) { return true; });
}
bool
canonicalize_replace_client(key_type const& key, SharedPointerType& data);
canonicalize_replace_client(const key_type& key, std::shared_ptr<T>& data)
{
return canonicalize(
key, data, [](std::shared_ptr<T> const&) { return false; });
}
SharedPointerType
fetch(key_type const& key);
std::shared_ptr<T>
fetch(const key_type& key)
{
std::lock_guard<mutex_type> l(m_mutex);
auto ret = initialFetch(key, l);
if (!ret)
++m_misses;
return ret;
}
/** Insert the element into the container.
If the key already exists, nothing happens.
@@ -156,11 +409,26 @@ public:
template <class ReturnType = bool>
auto
insert(key_type const& key, T const& value)
-> std::enable_if_t<!IsKeyCache, ReturnType>;
-> std::enable_if_t<!IsKeyCache, ReturnType>
{
auto p = std::make_shared<T>(std::cref(value));
return canonicalize_replace_client(key, p);
}
template <class ReturnType = bool>
auto
insert(key_type const& key) -> std::enable_if_t<IsKeyCache, ReturnType>;
insert(key_type const& key) -> std::enable_if_t<IsKeyCache, ReturnType>
{
std::lock_guard lock(m_mutex);
clock_type::time_point const now(m_clock.now());
auto [it, inserted] = m_cache.emplace(
std::piecewise_construct,
std::forward_as_tuple(key),
std::forward_as_tuple(now));
if (!inserted)
it->second.last_access = now;
return inserted;
}
// VFALCO NOTE It looks like this returns a copy of the data in
// the output parameter 'data'. This could be expensive.
@@ -168,18 +436,50 @@ public:
// simply return an iterator.
//
bool
retrieve(key_type const& key, T& data);
retrieve(const key_type& key, T& data)
{
// retrieve the value of the stored data
auto entry = fetch(key);
if (!entry)
return false;
data = *entry;
return true;
}
mutex_type&
peekMutex();
peekMutex()
{
return m_mutex;
}
std::vector<key_type>
getKeys() const;
getKeys() const
{
std::vector<key_type> v;
{
std::lock_guard lock(m_mutex);
v.reserve(m_cache.size());
for (auto const& _ : m_cache)
v.push_back(_.first);
}
return v;
}
// CachedSLEs functions.
/** Returns the fraction of cache hits. */
double
rate() const;
rate() const
{
std::lock_guard lock(m_mutex);
auto const tot = m_hits + m_misses;
if (tot == 0)
return 0;
return double(m_hits) / tot;
}
/** Fetch an item from the cache.
If the digest was not found, Handler
@@ -187,16 +487,73 @@ public:
std::shared_ptr<SLE const>(void)
*/
template <class Handler>
SharedPointerType
fetch(key_type const& digest, Handler const& h);
std::shared_ptr<T>
fetch(key_type const& digest, Handler const& h)
{
{
std::lock_guard l(m_mutex);
if (auto ret = initialFetch(digest, l))
return ret;
}
auto sle = h();
if (!sle)
return {};
std::lock_guard l(m_mutex);
++m_misses;
auto const [it, inserted] =
m_cache.emplace(digest, Entry(m_clock.now(), std::move(sle)));
if (!inserted)
it->second.touch(m_clock.now());
return it->second.ptr;
}
// End CachedSLEs functions.
private:
SharedPointerType
initialFetch(key_type const& key, std::lock_guard<mutex_type> const& l);
std::shared_ptr<T>
initialFetch(key_type const& key, std::lock_guard<mutex_type> const& l)
{
auto cit = m_cache.find(key);
if (cit == m_cache.end())
return {};
Entry& entry = cit->second;
if (entry.isCached())
{
++m_hits;
entry.touch(m_clock.now());
return entry.ptr;
}
entry.ptr = entry.lock();
if (entry.isCached())
{
// independent of cache size, so not counted as a hit
++m_cache_count;
entry.touch(m_clock.now());
return entry.ptr;
}
m_cache.erase(cit);
return {};
}
void
collect_metrics();
collect_metrics()
{
m_stats.size.set(getCacheSize());
{
beast::insight::Gauge::value_type hit_rate(0);
{
std::lock_guard lock(m_mutex);
auto const total(m_hits + m_misses);
if (total != 0)
hit_rate = (m_hits * 100) / total;
}
m_stats.hit_rate.set(hit_rate);
}
}
private:
struct Stats
@@ -242,37 +599,36 @@ private:
class ValueEntry
{
public:
shared_weak_combo_pointer_type ptr;
std::shared_ptr<mapped_type> ptr;
std::weak_ptr<mapped_type> weak_ptr;
clock_type::time_point last_access;
ValueEntry(
clock_type::time_point const& last_access_,
shared_pointer_type const& ptr_)
: ptr(ptr_), last_access(last_access_)
std::shared_ptr<mapped_type> const& ptr_)
: ptr(ptr_), weak_ptr(ptr_), last_access(last_access_)
{
}
bool
isWeak() const
{
if (!ptr)
return true;
return ptr.isWeak();
return ptr == nullptr;
}
bool
isCached() const
{
return ptr && ptr.isStrong();
return ptr != nullptr;
}
bool
isExpired() const
{
return ptr.expired();
return weak_ptr.expired();
}
SharedPointerType
std::shared_ptr<mapped_type>
lock()
{
return ptr.lock();
return weak_ptr.lock();
}
void
touch(clock_type::time_point const& now)
@@ -301,7 +657,72 @@ private:
typename KeyValueCacheType::map_type& partition,
SweptPointersVector& stuffToSweep,
std::atomic<int>& allRemovals,
std::lock_guard<std::recursive_mutex> const&);
std::lock_guard<std::recursive_mutex> const&)
{
return std::thread([&, this]() {
int cacheRemovals = 0;
int mapRemovals = 0;
// Keep references to all the stuff we sweep
// so that we can destroy them outside the lock.
stuffToSweep.first.reserve(partition.size());
stuffToSweep.second.reserve(partition.size());
{
auto cit = partition.begin();
while (cit != partition.end())
{
if (cit->second.isWeak())
{
// weak
if (cit->second.isExpired())
{
stuffToSweep.second.push_back(
std::move(cit->second.weak_ptr));
++mapRemovals;
cit = partition.erase(cit);
}
else
{
++cit;
}
}
else if (cit->second.last_access <= when_expire)
{
// strong, expired
++cacheRemovals;
if (cit->second.ptr.use_count() == 1)
{
stuffToSweep.first.push_back(
std::move(cit->second.ptr));
++mapRemovals;
cit = partition.erase(cit);
}
else
{
// remains weakly cached
cit->second.ptr.reset();
++cit;
}
}
else
{
// strong, not expired
++cit;
}
}
}
if (mapRemovals || cacheRemovals)
{
JLOG(m_journal.debug())
<< "TaggedCache partition sweep " << m_name
<< ": cache = " << partition.size() << "-" << cacheRemovals
<< ", map-=" << mapRemovals;
}
allRemovals += cacheRemovals;
});
}
[[nodiscard]] std::thread
sweepHelper(
@@ -310,7 +731,45 @@ private:
typename KeyOnlyCacheType::map_type& partition,
SweptPointersVector&,
std::atomic<int>& allRemovals,
std::lock_guard<std::recursive_mutex> const&);
std::lock_guard<std::recursive_mutex> const&)
{
return std::thread([&, this]() {
int cacheRemovals = 0;
int mapRemovals = 0;
// Keep references to all the stuff we sweep
// so that we can destroy them outside the lock.
{
auto cit = partition.begin();
while (cit != partition.end())
{
if (cit->second.last_access > now)
{
cit->second.last_access = now;
++cit;
}
else if (cit->second.last_access <= when_expire)
{
cit = partition.erase(cit);
}
else
{
++cit;
}
}
}
if (mapRemovals || cacheRemovals)
{
JLOG(m_journal.debug())
<< "TaggedCache partition sweep " << m_name
<< ": cache = " << partition.size() << "-" << cacheRemovals
<< ", map-=" << mapRemovals;
}
allRemovals += cacheRemovals;
});
};
beast::Journal m_journal;
clock_type& m_clock;
@@ -322,10 +781,10 @@ private:
std::string m_name;
// Desired number of cache entries (0 = ignore)
int const m_target_size;
int m_target_size;
// Desired maximum cache age
clock_type::duration const m_target_age;
clock_type::duration m_target_age;
// Number of items cached
int m_cache_count;

File diff suppressed because it is too large Load Diff

View File

@@ -25,7 +25,6 @@
#include <xrpl/beast/hash/hash_append.h>
#include <xrpl/beast/hash/uhash.h>
#include <xrpl/beast/hash/xxhasher.h>
#include <unordered_map>
#include <unordered_set>

View File

@@ -33,13 +33,12 @@
#include <xrpl/basics/strHex.h>
#include <xrpl/beast/utility/Zero.h>
#include <xrpl/beast/utility/instrumentation.h>
#include <boost/endian/conversion.hpp>
#include <boost/functional/hash.hpp>
#include <algorithm>
#include <array>
#include <cstring>
#include <functional>
#include <type_traits>
namespace ripple {
@@ -374,7 +373,7 @@ public:
}
base_uint&
operator^=(base_uint const& b)
operator^=(const base_uint& b)
{
for (int i = 0; i < WIDTH; i++)
data_[i] ^= b.data_[i];
@@ -383,7 +382,7 @@ public:
}
base_uint&
operator&=(base_uint const& b)
operator&=(const base_uint& b)
{
for (int i = 0; i < WIDTH; i++)
data_[i] &= b.data_[i];
@@ -392,7 +391,7 @@ public:
}
base_uint&
operator|=(base_uint const& b)
operator|=(const base_uint& b)
{
for (int i = 0; i < WIDTH; i++)
data_[i] |= b.data_[i];
@@ -415,11 +414,11 @@ public:
return *this;
}
base_uint const
const base_uint
operator++(int)
{
// postfix operator
base_uint const ret = *this;
const base_uint ret = *this;
++(*this);
return ret;
@@ -441,11 +440,11 @@ public:
return *this;
}
base_uint const
const base_uint
operator--(int)
{
// postfix operator
base_uint const ret = *this;
const base_uint ret = *this;
--(*this);
return ret;
@@ -466,7 +465,7 @@ public:
}
base_uint&
operator+=(base_uint const& b)
operator+=(const base_uint& b)
{
std::uint64_t carry = 0;
@@ -511,7 +510,7 @@ public:
}
[[nodiscard]] constexpr bool
parseHex(char const* str)
parseHex(const char* str)
{
return parseHex(std::string_view{str});
}

View File

@@ -20,16 +20,17 @@
#ifndef RIPPLE_BASICS_CHRONO_H_INCLUDED
#define RIPPLE_BASICS_CHRONO_H_INCLUDED
#include <date/date.h>
#include <xrpl/beast/clock/abstract_clock.h>
#include <xrpl/beast/clock/basic_seconds_clock.h>
#include <xrpl/beast/clock/manual_clock.h>
#include <date/date.h>
#include <chrono>
#include <cstdint>
#include <ratio>
#include <string>
#include <type_traits>
namespace ripple {

View File

@@ -43,7 +43,7 @@ struct less
using result_type = bool;
constexpr bool
operator()(T const& left, T const& right) const
operator()(const T& left, const T& right) const
{
return std::less<T>()(left, right);
}
@@ -55,7 +55,7 @@ struct equal_to
using result_type = bool;
constexpr bool
operator()(T const& left, T const& right) const
operator()(const T& left, const T& right) const
{
return std::equal_to<T>()(left, right);
}

View File

@@ -21,9 +21,9 @@
#define RIPPLE_BASICS_CONTRACT_H_INCLUDED
#include <xrpl/beast/type_name.h>
#include <exception>
#include <string>
#include <typeinfo>
#include <utility>
namespace ripple {

View File

@@ -21,7 +21,6 @@
#define RIPPLE_BASICS_MAKE_SSLCONTEXT_H_INCLUDED
#include <boost/asio/ssl/context.hpp>
#include <string>
namespace ripple {

View File

@@ -22,7 +22,6 @@
#include <xrpl/beast/hash/uhash.h>
#include <xrpl/beast/utility/instrumentation.h>
#include <functional>
#include <optional>
#include <string>
@@ -52,7 +51,7 @@ template <
typename Value,
typename Hash,
typename Pred = std::equal_to<Key>,
typename Alloc = std::allocator<std::pair<Key const, Value>>>
typename Alloc = std::allocator<std::pair<const Key, Value>>>
class partitioned_unordered_map
{
std::size_t partitions_;

View File

@@ -22,9 +22,9 @@
#include <xrpl/beast/utility/instrumentation.h>
#include <xrpl/beast/xor_shift_engine.h>
#include <cstddef>
#include <cstdint>
#include <cstring>
#include <limits>
#include <mutex>
#include <random>

View File

@@ -19,7 +19,6 @@
#define RIPPLE_BASICS_SPINLOCK_H_INCLUDED
#include <xrpl/beast/utility/instrumentation.h>
#include <atomic>
#include <limits>
#include <type_traits>

View File

@@ -21,9 +21,7 @@
#define BEAST_UTILITY_TAGGED_INTEGER_H_INCLUDED
#include <xrpl/beast/hash/hash_append.h>
#include <boost/operators.hpp>
#include <functional>
#include <iostream>
#include <type_traits>
@@ -76,13 +74,13 @@ public:
}
bool
operator<(tagged_integer const& rhs) const noexcept
operator<(const tagged_integer& rhs) const noexcept
{
return m_value < rhs.m_value;
}
bool
operator==(tagged_integer const& rhs) const noexcept
operator==(const tagged_integer& rhs) const noexcept
{
return m_value == rhs.m_value;
}
@@ -144,14 +142,14 @@ public:
}
tagged_integer&
operator<<=(tagged_integer const& rhs) noexcept
operator<<=(const tagged_integer& rhs) noexcept
{
m_value <<= rhs.m_value;
return *this;
}
tagged_integer&
operator>>=(tagged_integer const& rhs) noexcept
operator>>=(const tagged_integer& rhs) noexcept
{
m_value >>= rhs.m_value;
return *this;

View File

@@ -21,7 +21,6 @@
#define BEAST_ASIO_IO_LATENCY_PROBE_H_INCLUDED
#include <xrpl/beast/utility/instrumentation.h>
#include <boost/asio/basic_waitable_timer.hpp>
#include <boost/asio/io_service.hpp>

View File

@@ -25,11 +25,9 @@
#include <xrpl/beast/container/detail/aged_associative_container.h>
#include <xrpl/beast/container/detail/aged_container_iterator.h>
#include <xrpl/beast/container/detail/empty_base_optimization.h>
#include <boost/intrusive/list.hpp>
#include <boost/intrusive/set.hpp>
#include <boost/version.hpp>
#include <algorithm>
#include <functional>
#include <initializer_list>

View File

@@ -25,10 +25,8 @@
#include <xrpl/beast/container/detail/aged_associative_container.h>
#include <xrpl/beast/container/detail/aged_container_iterator.h>
#include <xrpl/beast/container/detail/empty_base_optimization.h>
#include <boost/intrusive/list.hpp>
#include <boost/intrusive/unordered_set.hpp>
#include <algorithm>
#include <cmath>
#include <functional>

View File

@@ -11,7 +11,6 @@
#define BEAST_CONTAINER_DETAIL_EMPTY_BASE_OPTIMIZATION_H_INCLUDED
#include <boost/type_traits/is_final.hpp>
#include <type_traits>
#include <utility>

View File

@@ -23,7 +23,6 @@
#include <xrpl/beast/utility/instrumentation.h>
#include <boost/core/detail/string_view.hpp>
#include <algorithm>
#include <cerrno>
#include <charconv>

View File

@@ -21,6 +21,7 @@
#define BEAST_INTRUSIVE_LIST_H_INCLUDED
#include <iterator>
#include <type_traits>
namespace beast {

View File

@@ -23,7 +23,6 @@
#include <boost/container/flat_set.hpp>
#include <boost/endian/conversion.hpp>
#include <array>
#include <chrono>
#include <cstdint>

View File

@@ -30,7 +30,7 @@ namespace beast {
template <class Hasher = xxhasher>
struct uhash
{
uhash() = default;
explicit uhash() = default;
using result_type = typename Hasher::result_type;

View File

@@ -21,7 +21,6 @@
#define BEAST_HASH_XXHASHER_H_INCLUDED
#include <boost/endian/conversion.hpp>
#include <xxhash.h>
#include <cstddef>

View File

@@ -20,10 +20,10 @@
#ifndef BEAST_INSIGHT_METER_H_INCLUDED
#define BEAST_INSIGHT_METER_H_INCLUDED
#include <xrpl/beast/insight/MeterImpl.h>
#include <memory>
#include <xrpl/beast/insight/MeterImpl.h>
namespace beast {
namespace insight {

View File

@@ -21,6 +21,7 @@
#define BEAST_INSIGHT_STATSDCOLLECTOR_H_INCLUDED
#include <xrpl/beast/insight/Collector.h>
#include <xrpl/beast/net/IPEndpoint.h>
#include <xrpl/beast/utility/Journal.h>

View File

@@ -25,10 +25,8 @@
#include <xrpl/beast/net/IPAddressV4.h>
#include <xrpl/beast/net/IPAddressV6.h>
#include <xrpl/beast/utility/instrumentation.h>
#include <boost/asio/ip/address.hpp>
#include <boost/functional/hash.hpp>
#include <cstdint>
#include <ios>
#include <sstream>

View File

@@ -22,6 +22,8 @@
#include <xrpl/beast/net/IPEndpoint.h>
#include <sstream>
#include <boost/asio.hpp>
namespace beast {

View File

@@ -21,9 +21,7 @@
#define BEAST_NET_IPADDRESSV4_H_INCLUDED
#include <xrpl/beast/hash/hash_append.h>
#include <boost/asio/ip/address_v4.hpp>
#include <cstdint>
#include <functional>
#include <ios>

View File

@@ -21,9 +21,7 @@
#define BEAST_NET_IPADDRESSV6_H_INCLUDED
#include <xrpl/beast/utility/instrumentation.h>
#include <boost/asio/ip/address_v6.hpp>
#include <cstdint>
#include <functional>
#include <ios>

View File

@@ -215,7 +215,7 @@ namespace std {
template <>
struct hash<::beast::IP::Endpoint>
{
hash() = default;
explicit hash() = default;
std::size_t
operator()(::beast::IP::Endpoint const& endpoint) const
@@ -230,7 +230,7 @@ namespace boost {
template <>
struct hash<::beast::IP::Endpoint>
{
hash() = default;
explicit hash() = default;
std::size_t
operator()(::beast::IP::Endpoint const& endpoint) const

View File

@@ -20,6 +20,10 @@
#ifndef BEAST_RFC2616_HPP
#define BEAST_RFC2616_HPP
// TODO: This include <cstdint> is a workaround for beast compilation bug.
// Remove when fix https://github.com/boostorg/beast/pull/2682/ is available.
#include <cstdint>
#include <boost/beast/http/message.hpp>
#include <boost/beast/http/rfc7230.hpp>
#include <boost/range/algorithm/equal.hpp>
@@ -28,7 +32,6 @@
#include <algorithm>
#include <cctype>
#include <cstdint>
#include <iterator>
#include <string>
#include <utility>

View File

@@ -11,7 +11,6 @@
#include <boost/asio/io_service.hpp>
#include <boost/asio/spawn.hpp>
#include <boost/optional.hpp>
#include <condition_variable>
#include <functional>
#include <mutex>

View File

@@ -9,7 +9,6 @@
#define BEAST_UNIT_TEST_MATCH_HPP
#include <xrpl/beast/unit_test/suite_info.h>
#include <string>
namespace beast {

View File

@@ -10,10 +10,8 @@
#include <xrpl/beast/unit_test/amount.h>
#include <xrpl/beast/unit_test/recorder.h>
#include <boost/lexical_cast.hpp>
#include <boost/optional.hpp>
#include <algorithm>
#include <chrono>
#include <functional>

View File

@@ -9,9 +9,7 @@
#define BEAST_UNIT_TEST_RUNNER_H_INCLUDED
#include <xrpl/beast/unit_test/suite_info.h>
#include <boost/assert.hpp>
#include <mutex>
#include <ostream>
#include <string>

View File

@@ -9,11 +9,9 @@
#define BEAST_UNIT_TEST_SUITE_HPP
#include <xrpl/beast/unit_test/runner.h>
#include <boost/filesystem.hpp>
#include <boost/lexical_cast.hpp>
#include <boost/throw_exception.hpp>
#include <ostream>
#include <sstream>
#include <string>

View File

@@ -10,9 +10,7 @@
#include <xrpl/beast/unit_test/detail/const_container.h>
#include <xrpl/beast/unit_test/suite_info.h>
#include <boost/assert.hpp>
#include <set>
#include <typeindex>
#include <unordered_set>

View File

@@ -9,7 +9,6 @@
#define BEAST_UNIT_TEST_THREAD_HPP
#include <xrpl/beast/unit_test/suite.h>
#include <functional>
#include <thread>
#include <utility>

View File

@@ -21,7 +21,6 @@
#define BEAST_UTILITY_JOURNAL_H_INCLUDED
#include <xrpl/beast/utility/instrumentation.h>
#include <sstream>
namespace beast {
@@ -113,16 +112,6 @@ public:
virtual void
write(Severity level, std::string const& text) = 0;
/** Bypass filter and write text to the sink at the specified severity.
* Always write the message, but maintain the same formatting as if
* it passed through a level filter.
*
* @param level Level to display in message.
* @param text Text to write to sink.
*/
virtual void
writeAlways(Severity level, std::string const& text) = 0;
private:
Severity thresh_;
bool m_console;

View File

@@ -22,9 +22,11 @@
#include <xrpl/beast/core/List.h>
#include <cstdint>
#include <mutex>
#include <sstream>
#include <string>
#include <utility>
namespace beast {

View File

@@ -93,13 +93,6 @@ public:
using beast::Journal;
sink_.write(level, prefix_ + text);
}
void
writeAlways(severities::Severity level, std::string const& text) override
{
using beast::Journal;
sink_.writeAlways(level, prefix_ + text);
}
};
} // namespace beast

View File

@@ -21,7 +21,6 @@
#define BEAST_RANDOM_RNGFILL_H_INCLUDED
#include <xrpl/beast/utility/instrumentation.h>
#include <array>
#include <cstdint>
#include <cstring>

View File

@@ -21,7 +21,6 @@
#define BEAST_UTILITY_TEMP_DIR_H_INCLUDED
#include <boost/filesystem.hpp>
#include <string>
namespace beast {
@@ -37,9 +36,9 @@ class temp_dir
public:
#if !GENERATING_DOCS
temp_dir(temp_dir const&) = delete;
temp_dir(const temp_dir&) = delete;
temp_dir&
operator=(temp_dir const&) = delete;
operator=(const temp_dir&) = delete;
#endif
/// Construct a temporary directory.

View File

@@ -21,6 +21,8 @@
#define RIPPLE_CRYPTO_RANDOM_H_INCLUDED
#include <mutex>
#include <string>
#include <type_traits>
namespace ripple {

View File

@@ -21,7 +21,6 @@
#define RIPPLE_JSON_OBJECT_H_INCLUDED
#include <xrpl/json/Writer.h>
#include <memory>
namespace Json {

View File

@@ -21,7 +21,6 @@
#define RIPPLE_JSON_OUTPUT_H_INCLUDED
#include <boost/beast/core/string.hpp>
#include <functional>
#include <string>

View File

@@ -24,7 +24,6 @@
#include <xrpl/basics/contract.h>
#include <xrpl/json/Output.h>
#include <xrpl/json/json_value.h>
#include <memory>
namespace Json {

View File

@@ -20,7 +20,6 @@
#ifndef RIPPLE_JSON_JSON_ASSERT_H_INCLUDED
#define RIPPLE_JSON_JSON_ASSERT_H_INCLUDED
#include <xrpl/basics/contract.h>
#include <xrpl/json/json_errors.h>
#define JSON_ASSERT_MESSAGE(condition, message) \

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