| Age | Commit message (Collapse) | Author |
|---|
|
|
|
|
|
* test for global modifications by various inner apps
* CR adjustments to feeCredit
* Test local state sharing too.
* explain clear state details
|
|
|
|
* Test current state of inner clear state behavior
* First half of clear state fixes. Prevents inner calls in CSP
Should also implement the "CSP requires 700 budget and can't spend
more than 700 budget" rule, but tests still required.
* Match program version. Don't allow downgrade < 6.
* partition test
* typo
Co-authored-by: Michael Diamant <michaeldiamant@users.noreply.github.com>
* typo
Co-authored-by: Michael Diamant <michaeldiamant@users.noreply.github.com>
* Make the default clear_state program match versions to approval
* Correct some tests that now need programs present
* Fix test to actually test INNER clear state
* Add tests for opcode budget and CSPs
* Improve tracing for inner app calls
* Error now reports what the cost *was* before the attempted inst.
* Simplify type switch and maybe kick CI?
* Fewer copies during inner processing
Co-authored-by: Michael Diamant <michaeldiamant@users.noreply.github.com>
|
|
## Summary
Added a new option `--exclude-fields` to `file` subcommand.
This simplifies large dumps comparison by excluding all exclude some header fields.
Also moved some shared cmd code into a new cmdutil package.
## Test Plan
Tested manually
|
|
## Summary
Remove a hardcoded constant in sortition.go which was
used as the denominator in determining the selection ratio.
This clarifies what the maximum possible output size is of
the output VRF based on the SHA algorithm used to generate it.
## Purpose
This is simply a code quality change to make it easier to read and
understand without any focus on new functionality or performance.
It is unlikely that the VRF SHA length used by Algorand will changed,
so the generalization to a dynamic length based on the length of the
Crypto digest is only for the reader.
The use of an `init` function to initialize the `maxFloat` once rather than
anytime `sortition` runs is a minor optimization, but likely to be neglibile
in the grand scheme of things.
## Test Plan
Due to the usage of C src code, a simple `go test -v ./data/committee/sortition/`
cannot be executed without the proper configurations. However, circle CI
should be able to execute the unit tests automatically and verify the changes.
Recommendation: Running the recommended make targets to test the changes
did not automatically due to missing dependencies, so potentially a Dockerfile
could help resolve this for future contributors.
|
|
Our nightly pipeline broke because of centos:8 reaching end of life and deprecating their package repo AppStream. Switching to official new version Centos Stream 8.
|
|
## Summary
* Do not load everything into memory but write directly in into a file
This creates extremely high number of reallocation and CPU and RAM wasting.
* Add cmd options for downloading only (no loading), and for downloading
from a single relay
# Perf data:
| param | current | new |
|-----------|--------|---------|
| RAM | 5 GB | 200 MB |
| Total time | 823 s | 441 s |
Downloading time went down from minutes to tens of seconds.
|
|
## Summary
The TestSyncFromClient was randomly failing. This is a real issue in production code, and the fix it pretty trivial.
## Test Plan
Use existing tests to verify test.
|
|
## Summary
Add a warning in case the VotingKeys method surpasses the expected execution time limits.
## Test Plan
Tested manually.
|
|
|
|
## Summary
In this change, we plan to integrate the ed25519 batch verification algorithm. This will speed up the process of verifying the digital signature of several transactions by roughly 2.4X.
This PR implements the algorithm according to the following paper:
https://eprint.iacr.org/2020/1244.pdf
This PR fixes a **bug** in the paper (Listing 1.2) that didn't reject the points (-0,-1) and (-0,1)
Another blog article with similar background: https://hdevalence.ca/blog/2020-10-04-its-25519am
**Changes Overview**
Changes made on the single ed255519 verification :
1. Signatures that contain small order group R value will now be accepted.
2. use the cofactor version of the equation to validate the signature. (multiple the result by the cofactor and compare to natural element)
3. add ge25519_is_canonical_vartime function (check if a given y point has a canonical or non-canonical encoding ):
- optimize this function using the the quick check.
- reject two non-canonical points by comparison (points (-0,-1) and (-0,1), this resolves a bug in the paper).
Add the batch verification implementation from https://github.com/floodyberry/ed25519-donna to our fork of libsodium.
Changes made on the batch verification (to match the single verification):
1. reject small order A (PK)
2. reject non-canonical A (PK)
3. reject non-canonical R
4. reject non-canonical S
5. use the cofactor equation
## Test Plan
The following tests were added to the libsodium's test suite.
1. Test against the White Paper's test vector
2. Test against all non-canonical options for R
3. Test against all non-canonical options for S
|
|
## Summary
Updates ledger README reference to tracker source code.
Since `trackers.go` does _not_ exist, I _assume_ the appropriate reference is `tracker.go`.
## Test Plan
Since the change _only_ impacts docs, I visually confirmed Markdown rendering.
|
|
* Only allow access to txn effects for previous transactions
* JP code review
* Ok, dawg
|
|
|
|
* txn LastLog
* Insulate test from the type of field
|
|
Fix pooled costs during app to app recursion
Also adds divw, itxnas, gitxnas
Makes accessing 0 accounts in dry-run ok.
|
|
To support performance testing with bandwidth collection, updated Hello World recipe so that it works, and specify UTF-8 encoding when processing metrics files.
|
|
|
|
## Summary
Our SDK client generator was unable to handle this empty type. The type doesn't add anything so I'm removing it instead of adding a new feature to the generator.
## Test Plan
N/A
|
|
* revert fillbytes method for lower version of golang
* awww shut up dude
* use copy
* Use DEPRECATED
Co-authored-by: Will Winder <wwinder.unh@gmail.com>
Co-authored-by: Will Winder <wwinder.unh@gmail.com>
|
|
|
|
* Three new globals for to help contract-to-contract usability
* detritis
* Check error
* doc comments
* Impose limits on the entire "tree" of inner calls.
This also increases the realism of testing of multiple app calls in a
group by creating the EvalParams with the real constructor, thus
getting the pooling stuff tested here without playing games
manipulating the ep after construction.
* Move appID tracking into EvalContext, out of LedgerForLogic
This change increases the seperation between AVM execution and the
ledger being used to lookup resources. Previously, the ledger kept
track of the appID being executed, to offer a narrower interface to
those resources. But now, with app-to-app calls, the appID being
executed must change, and the AVM needs to maintain the current appID.
* Stupid linter
* Fix unit tests error messages
* Allow access to resources created in the same transaction group
The method will be reworked, but the tests are correct and want to get
them visible to team.
* Access to apps created in group
Also adds some tests that are currently skipped for testing
- access to addresses of newly created apps
- use of gaid in inner transactions
Both require some work to implement the thing being tested.
* Remove tracked created mechanism in favor of examining applydata.
* Allow v6 AVM code to use in-group created asas, apps (& their accts)
One exception - apps can not mutate (put or del) keys from the app
accounts, because EvalDelta cannot encode such changes.
* lint docs
* typo
* The review dog needs obedience training.
* Use one EvalParams for logic evals, another for apps in dry run
We used to use one ep per transaction, shared between sig and and
app. But the new model of ep usage is to keep using one while
evaluating an entire group.
The app ep is now built logic.NewAppEvalParams which, hopefully, will
prevent some bugs when we change something in the EvalParams and don't
reflect it in what was a "raw" EvalParams construction in debugger and
dry run.
* Use logic.NewAppEvalParams to decrease copying and bugs in debugger
* Simplify use of NewEvalParams. No more nil return when no apps.
This way, NewEvalParams can be used for all creations of EvalParams,
whether they are intended for logicsig or app use, greatly simplifying
the way we make them for use by dry run or debugger (where they serve
double duty).
* Remove explicit PastSideEffects handling in tealdbg
* Always create EvalParams to evaluate a transaction group.
We used to have an optimization to avoid creating EvalParams unless
there was an app call in the transaction group. But the interface to
allow transaction processing to communicate changes into the
EvalParams is complicated by that (we must only do it if there is
one!)
This also allows us to use the same construction function for eps
created for app and logic evaluation, simplifying dry-run and
debugger.
The optimization is less needed now anyway:
1) The ep is now shared for the whole group, so it's only one.
2) The ep is smaller now, as we only store nil pointers instead of
larger scratch space objects for non-app calls.
* Correct mistaken commit
* Spec improvments
* More spec improvments, including resource "availability"
* Recursively return inner transaction tree
* Lint
* No need for ConfirmedRound, so don't deref a nil pointer!
* license check
* Shut up, dawg.
* base64 merge cleanup
* Remove the extraneous field type arrays.
* bsqrt
* acct_holding_get, a unified opcode for account field access
* Thanks, dawg
* CR and more spec simplification
* e2e test for inner transaction appls
* Give max group size * 16 inner txns, regardless of apps present
* Adjust test for allowing 256 inners
* NCC audit (except double credit) plus app depth limit
* Fix for double credit from inner groups, found by NCC
* Spec updates
* Count inners properly, increase inner txn allocbound
|
|
* ledger: fix `NextRewardsState()` (#3403)
## Summary
A modification of #3336. Added a new test where the rewards pool overspends and proposed a fix in `NextRewardsState()` requiring a consensus upgrade.
## Test Plan
This is mostly tests.
* Fix a potential problem of committing non-uniform consensus versions (#3453)
If accountdb accumulates a large backlog, it is possible catchpoint tracker would attempt to commit a wider range than account updates tracker expects.
* avoid generating log error on EnsureValidatedBlock / EnsureBlock (#3424)
In EnsureBlock,, do not log as error message if the error is ledgercore.ErrNonSequentialBlockEval and the block round is in the past (i.e. already in the ledger).
* Fix typo Fulll to Full (#3456)
Fix typo
* Fix worng message on restore crash db. (#3455)
When crash state is found but could not be restored, noCrashState variable is used to report a warning.
However, this variable was set to false in a case where there was no crash state, and the wrong warning was reported.
* Adding new scenario for feature networks (#3451)
* Fixing telemetry ports (#3497)
Co-authored-by: Tolik Zinovyev <tolik@algorand.com>
Co-authored-by: Pavel Zbitskiy <65323360+algorandskiy@users.noreply.github.com>
Co-authored-by: Shant Karakashian <55754073+algonautshant@users.noreply.github.com>
Co-authored-by: Jack <87339414+algojack@users.noreply.github.com>
|
|
|
|
* ledger: fix `NextRewardsState()` (#3403)
## Summary
A modification of #3336. Added a new test where the rewards pool overspends and proposed a fix in `NextRewardsState()` requiring a consensus upgrade.
## Test Plan
This is mostly tests.
* Fix a potential problem of committing non-uniform consensus versions (#3453)
If accountdb accumulates a large backlog, it is possible catchpoint tracker would attempt to commit a wider range than account updates tracker expects.
* avoid generating log error on EnsureValidatedBlock / EnsureBlock (#3424)
In EnsureBlock,, do not log as error message if the error is ledgercore.ErrNonSequentialBlockEval and the block round is in the past (i.e. already in the ledger).
* Fix typo Fulll to Full (#3456)
Fix typo
* Fix worng message on restore crash db. (#3455)
When crash state is found but could not be restored, noCrashState variable is used to report a warning.
However, this variable was set to false in a case where there was no crash state, and the wrong warning was reported.
* Adding new scenario for feature networks (#3451)
Co-authored-by: Tolik Zinovyev <tolik@algorand.com>
Co-authored-by: Pavel Zbitskiy <65323360+algorandskiy@users.noreply.github.com>
Co-authored-by: Shant Karakashian <55754073+algonautshant@users.noreply.github.com>
|
|
|
|
When crash state is found but could not be restored, noCrashState variable is used to report a warning.
However, this variable was set to false in a case where there was no crash state, and the wrong warning was reported.
|
|
Fix typo
|
|
In EnsureBlock,, do not log as error message if the error is ledgercore.ErrNonSequentialBlockEval and the block round is in the past (i.e. already in the ledger).
|
|
If accountdb accumulates a large backlog, it is possible catchpoint tracker would attempt to commit a wider range than account updates tracker expects.
|
|
## Summary
A modification of #3336. Added a new test where the rewards pool overspends and proposed a fix in `NextRewardsState()` requiring a consensus upgrade.
## Test Plan
This is mostly tests.
|
|
github.com:algorand/go-algorand into feature/contract-to-contract
|
|
* Three new globals for to help contract-to-contract usability
* detritis
* Check error
* doc comments
* Impose limits on the entire "tree" of inner calls.
This also increases the realism of testing of multiple app calls in a
group by creating the EvalParams with the real constructor, thus
getting the pooling stuff tested here without playing games
manipulating the ep after construction.
* Move appID tracking into EvalContext, out of LedgerForLogic
This change increases the seperation between AVM execution and the
ledger being used to lookup resources. Previously, the ledger kept
track of the appID being executed, to offer a narrower interface to
those resources. But now, with app-to-app calls, the appID being
executed must change, and the AVM needs to maintain the current appID.
* Stupid linter
* Fix unit tests error messages
* Allow access to resources created in the same transaction group
The method will be reworked, but the tests are correct and want to get
them visible to team.
* Access to apps created in group
Also adds some tests that are currently skipped for testing
- access to addresses of newly created apps
- use of gaid in inner transactions
Both require some work to implement the thing being tested.
* Remove tracked created mechanism in favor of examining applydata.
* Allow v6 AVM code to use in-group created asas, apps (& their accts)
One exception - apps can not mutate (put or del) keys from the app
accounts, because EvalDelta cannot encode such changes.
* lint docs
* typo
* The review dog needs obedience training.
* Use one EvalParams for logic evals, another for apps in dry run
We used to use one ep per transaction, shared between sig and and
app. But the new model of ep usage is to keep using one while
evaluating an entire group.
The app ep is now built logic.NewAppEvalParams which, hopefully, will
prevent some bugs when we change something in the EvalParams and don't
reflect it in what was a "raw" EvalParams construction in debugger and
dry run.
* Use logic.NewAppEvalParams to decrease copying and bugs in debugger
* Simplify use of NewEvalParams. No more nil return when no apps.
This way, NewEvalParams can be used for all creations of EvalParams,
whether they are intended for logicsig or app use, greatly simplifying
the way we make them for use by dry run or debugger (where they serve
double duty).
* Remove explicit PastSideEffects handling in tealdbg
* Always create EvalParams to evaluate a transaction group.
We used to have an optimization to avoid creating EvalParams unless
there was an app call in the transaction group. But the interface to
allow transaction processing to communicate changes into the
EvalParams is complicated by that (we must only do it if there is
one!)
This also allows us to use the same construction function for eps
created for app and logic evaluation, simplifying dry-run and
debugger.
The optimization is less needed now anyway:
1) The ep is now shared for the whole group, so it's only one.
2) The ep is smaller now, as we only store nil pointers instead of
larger scratch space objects for non-app calls.
* Correct mistaken commit
* Spec improvments
* More spec improvments, including resource "availability"
* Recursively return inner transaction tree
* Lint
* No need for ConfirmedRound, so don't deref a nil pointer!
* license check
* Shut up, dawg.
* testing: Fix unit test TestAsyncTelemetryHook_QueueDepth (#2685)
Fix the unit test TestAsyncTelemetryHook_QueueDepth
* Deprecate `FastPartitionRecovery` from `ConsensusParams` (#3386)
## Summary
This PR removes `FastPartitionRecovery` option from consensus parameters. The code now acts as if this value is set to true.
Closes https://github.com/algorand/go-algorand-internal/issues/1830.
## Test Plan
None.
* base64 merge cleanup
* Remaking a PR for CI (#3398)
* Allow setting manager, reserve, freeze, and clawback at goal asset create
* Add e2e tests
* Add more tests for goal asset create flags
Co-authored-by: Fionna <fionnacst@gmail.com>
* Remove the extraneous field type arrays.
* bsqrt
* acct_holding_get, a unified opcode for account field access
* Thanks, dawg
* [Other] CircleCI pipeline change for binary uploads (#3381)
For nightly builds ("rel/nightly"), we want to have deadlock enabled.
For rel/beta and rel/stable, we want to make sure we can build and upload a binary with deadlock disabled so that it can be used for release testing and validation purposes.
* signer.KeyDilution need not depend on config package (#3265)
crypto package need not depend on config.
There is an unnecessary dependency on config.
signer.KeyDilution takes the `config.ConsensusParams` as argument to pick the DefaultKeyDilution from it.
This introduces dependency from the crypto package to config package.
Instead, only the DefaultKeyDilution value can be passed to signer.KeyDilution.
* CR and more spec simplification
* algodump is a tcpdump-like tool for algod's network protocol (#3166)
This PR introduces algodump, a tcpdump-like tool for monitoring algod network messages.
* Removing C/crypto dependencies from `data/abi` package (#3375)
* Feature Networks pipeline related changes (#3393)
Added support for not having certain files in signing script
* e2e test for inner transaction appls
* testing: Add slightly more coverage to TestAcctUpdatesLookupRetry (#3384)
Add slightly more coverage to TestAcctUpdatesLookupRetry
* add context to (most) agreement logged writes (#3411)
Current agreement code only writes a `context : agreement` to a subset of the logged messages.
This change extends the said entry, which would make it easier to pre-process logs entries by their corresponding component. The change in this PR is focused on:
1. make sure that the "root" agreement logger always injects the `context : agreement` argument.
2. change the various locations in the agreement code to use the root agreement logger instead of referring to the application-global instance (`logging.Base()`).
* network: faster node shutdown (#3416)
During the node shutdown, all the current outgoing connections are being disconnected.
Since these connections are web sockets, they require a close connection message to be sent.
However, sending this message can take a while, and in situations where the other party has already shut down, we might never get a response. That, in turn, would lead the node waiting until the deadline is reached.
The current deadline was 5 seconds. This PR changes the deadline during shutdown to be 50ms.
* Give max group size * 16 inner txns, regardless of apps present
* Adjust test for allowing 256 inners
Co-authored-by: Tsachi Herman <tsachi.herman@algorand.com>
Co-authored-by: Tolik Zinovyev <tolik@algorand.com>
Co-authored-by: Jack <87339414+algojack@users.noreply.github.com>
Co-authored-by: Fionna <fionnacst@gmail.com>
Co-authored-by: algobarb <78746954+algobarb@users.noreply.github.com>
Co-authored-by: Shant Karakashian <55754073+algonautshant@users.noreply.github.com>
Co-authored-by: Nickolai Zeldovich <nickolai@csail.mit.edu>
Co-authored-by: Hang Su <87964331+ahangsu@users.noreply.github.com>
Co-authored-by: chris erway <51567+cce@users.noreply.github.com>
|
|
|
|
* Three new globals for to help contract-to-contract usability
* detritis
* Check error
* doc comments
* Impose limits on the entire "tree" of inner calls.
This also increases the realism of testing of multiple app calls in a
group by creating the EvalParams with the real constructor, thus
getting the pooling stuff tested here without playing games
manipulating the ep after construction.
* Move appID tracking into EvalContext, out of LedgerForLogic
This change increases the seperation between AVM execution and the
ledger being used to lookup resources. Previously, the ledger kept
track of the appID being executed, to offer a narrower interface to
those resources. But now, with app-to-app calls, the appID being
executed must change, and the AVM needs to maintain the current appID.
* Stupid linter
* Fix unit tests error messages
* Allow access to resources created in the same transaction group
The method will be reworked, but the tests are correct and want to get
them visible to team.
* Access to apps created in group
Also adds some tests that are currently skipped for testing
- access to addresses of newly created apps
- use of gaid in inner transactions
Both require some work to implement the thing being tested.
* Remove tracked created mechanism in favor of examining applydata.
* Allow v6 AVM code to use in-group created asas, apps (& their accts)
One exception - apps can not mutate (put or del) keys from the app
accounts, because EvalDelta cannot encode such changes.
* lint docs
* typo
* The review dog needs obedience training.
* Use one EvalParams for logic evals, another for apps in dry run
We used to use one ep per transaction, shared between sig and and
app. But the new model of ep usage is to keep using one while
evaluating an entire group.
The app ep is now built logic.NewAppEvalParams which, hopefully, will
prevent some bugs when we change something in the EvalParams and don't
reflect it in what was a "raw" EvalParams construction in debugger and
dry run.
* Use logic.NewAppEvalParams to decrease copying and bugs in debugger
* Simplify use of NewEvalParams. No more nil return when no apps.
This way, NewEvalParams can be used for all creations of EvalParams,
whether they are intended for logicsig or app use, greatly simplifying
the way we make them for use by dry run or debugger (where they serve
double duty).
* Remove explicit PastSideEffects handling in tealdbg
* Always create EvalParams to evaluate a transaction group.
We used to have an optimization to avoid creating EvalParams unless
there was an app call in the transaction group. But the interface to
allow transaction processing to communicate changes into the
EvalParams is complicated by that (we must only do it if there is
one!)
This also allows us to use the same construction function for eps
created for app and logic evaluation, simplifying dry-run and
debugger.
The optimization is less needed now anyway:
1) The ep is now shared for the whole group, so it's only one.
2) The ep is smaller now, as we only store nil pointers instead of
larger scratch space objects for non-app calls.
* Correct mistaken commit
* Spec improvments
* More spec improvments, including resource "availability"
* Recursively return inner transaction tree
* Lint
* No need for ConfirmedRound, so don't deref a nil pointer!
* license check
* Shut up, dawg.
* base64 merge cleanup
* Remove the extraneous field type arrays.
* bsqrt
* acct_holding_get, a unified opcode for account field access
* Thanks, dawg
* CR and more spec simplification
* e2e test for inner transaction appls
* Give max group size * 16 inner txns, regardless of apps present
* Adjust test for allowing 256 inners
|
|
* Integration tests for C2C
* added stress test
* undid incorrect autoformat
* remove autogenerated catchpoints
* Improve benchmarking
|
|
During the node shutdown, all the current outgoing connections are being disconnected.
Since these connections are web sockets, they require a close connection message to be sent.
However, sending this message can take a while, and in situations where the other party has already shut down, we might never get a response. That, in turn, would lead the node waiting until the deadline is reached.
The current deadline was 5 seconds. This PR changes the deadline during shutdown to be 50ms.
|
|
Current agreement code only writes a `context : agreement` to a subset of the logged messages.
This change extends the said entry, which would make it easier to pre-process logs entries by their corresponding component. The change in this PR is focused on:
1. make sure that the "root" agreement logger always injects the `context : agreement` argument.
2. change the various locations in the agreement code to use the root agreement logger instead of referring to the application-global instance (`logging.Base()`).
|
|
Add slightly more coverage to TestAcctUpdatesLookupRetry
|
|
also an e2e inner appl test
|
|
* testing: Fix unit test TestAsyncTelemetryHook_QueueDepth (#2685)
Fix the unit test TestAsyncTelemetryHook_QueueDepth
* Deprecate `FastPartitionRecovery` from `ConsensusParams` (#3386)
## Summary
This PR removes `FastPartitionRecovery` option from consensus parameters. The code now acts as if this value is set to true.
Closes https://github.com/algorand/go-algorand-internal/issues/1830.
## Test Plan
None.
* Remaking a PR for CI (#3398)
* Allow setting manager, reserve, freeze, and clawback at goal asset create
* Add e2e tests
* Add more tests for goal asset create flags
Co-authored-by: Fionna <fionnacst@gmail.com>
* [Other] CircleCI pipeline change for binary uploads (#3381)
For nightly builds ("rel/nightly"), we want to have deadlock enabled.
For rel/beta and rel/stable, we want to make sure we can build and upload a binary with deadlock disabled so that it can be used for release testing and validation purposes.
* signer.KeyDilution need not depend on config package (#3265)
crypto package need not depend on config.
There is an unnecessary dependency on config.
signer.KeyDilution takes the `config.ConsensusParams` as argument to pick the DefaultKeyDilution from it.
This introduces dependency from the crypto package to config package.
Instead, only the DefaultKeyDilution value can be passed to signer.KeyDilution.
* algodump is a tcpdump-like tool for algod's network protocol (#3166)
This PR introduces algodump, a tcpdump-like tool for monitoring algod network messages.
* Removing C/crypto dependencies from `data/abi` package (#3375)
* Feature Networks pipeline related changes (#3393)
Added support for not having certain files in signing script
Co-authored-by: Tsachi Herman <tsachi.herman@algorand.com>
Co-authored-by: Tolik Zinovyev <tolik@algorand.com>
Co-authored-by: Jack <87339414+algojack@users.noreply.github.com>
Co-authored-by: Fionna <fionnacst@gmail.com>
Co-authored-by: algobarb <78746954+algobarb@users.noreply.github.com>
Co-authored-by: Shant Karakashian <55754073+algonautshant@users.noreply.github.com>
Co-authored-by: Nickolai Zeldovich <nickolai@csail.mit.edu>
Co-authored-by: Hang Su <87964331+ahangsu@users.noreply.github.com>
|
|
Added support for not having certain files in signing script
|
|
|
|
This PR introduces algodump, a tcpdump-like tool for monitoring algod network messages.
|
|
crypto package need not depend on config.
There is an unnecessary dependency on config.
signer.KeyDilution takes the `config.ConsensusParams` as argument to pick the DefaultKeyDilution from it.
This introduces dependency from the crypto package to config package.
Instead, only the DefaultKeyDilution value can be passed to signer.KeyDilution.
|
|
For nightly builds ("rel/nightly"), we want to have deadlock enabled.
For rel/beta and rel/stable, we want to make sure we can build and upload a binary with deadlock disabled so that it can be used for release testing and validation purposes.
|
|
* Three new globals for to help contract-to-contract usability
* detritis
* Check error
* doc comments
* Impose limits on the entire "tree" of inner calls.
This also increases the realism of testing of multiple app calls in a
group by creating the EvalParams with the real constructor, thus
getting the pooling stuff tested here without playing games
manipulating the ep after construction.
* Move appID tracking into EvalContext, out of LedgerForLogic
This change increases the seperation between AVM execution and the
ledger being used to lookup resources. Previously, the ledger kept
track of the appID being executed, to offer a narrower interface to
those resources. But now, with app-to-app calls, the appID being
executed must change, and the AVM needs to maintain the current appID.
* Stupid linter
* Fix unit tests error messages
* Allow access to resources created in the same transaction group
The method will be reworked, but the tests are correct and want to get
them visible to team.
* Access to apps created in group
Also adds some tests that are currently skipped for testing
- access to addresses of newly created apps
- use of gaid in inner transactions
Both require some work to implement the thing being tested.
* Remove tracked created mechanism in favor of examining applydata.
* Allow v6 AVM code to use in-group created asas, apps (& their accts)
One exception - apps can not mutate (put or del) keys from the app
accounts, because EvalDelta cannot encode such changes.
* lint docs
* typo
* The review dog needs obedience training.
* Use one EvalParams for logic evals, another for apps in dry run
We used to use one ep per transaction, shared between sig and and
app. But the new model of ep usage is to keep using one while
evaluating an entire group.
The app ep is now built logic.NewAppEvalParams which, hopefully, will
prevent some bugs when we change something in the EvalParams and don't
reflect it in what was a "raw" EvalParams construction in debugger and
dry run.
* Use logic.NewAppEvalParams to decrease copying and bugs in debugger
* Simplify use of NewEvalParams. No more nil return when no apps.
This way, NewEvalParams can be used for all creations of EvalParams,
whether they are intended for logicsig or app use, greatly simplifying
the way we make them for use by dry run or debugger (where they serve
double duty).
* Remove explicit PastSideEffects handling in tealdbg
* Always create EvalParams to evaluate a transaction group.
We used to have an optimization to avoid creating EvalParams unless
there was an app call in the transaction group. But the interface to
allow transaction processing to communicate changes into the
EvalParams is complicated by that (we must only do it if there is
one!)
This also allows us to use the same construction function for eps
created for app and logic evaluation, simplifying dry-run and
debugger.
The optimization is less needed now anyway:
1) The ep is now shared for the whole group, so it's only one.
2) The ep is smaller now, as we only store nil pointers instead of
larger scratch space objects for non-app calls.
* Correct mistaken commit
* Spec improvments
* More spec improvments, including resource "availability"
* Recursively return inner transaction tree
* Lint
* No need for ConfirmedRound, so don't deref a nil pointer!
* license check
* Shut up, dawg.
* base64 merge cleanup
|
|
* Three new globals for to help contract-to-contract usability
* detritis
* Check error
* doc comments
* Impose limits on the entire "tree" of inner calls.
This also increases the realism of testing of multiple app calls in a
group by creating the EvalParams with the real constructor, thus
getting the pooling stuff tested here without playing games
manipulating the ep after construction.
* Move appID tracking into EvalContext, out of LedgerForLogic
This change increases the seperation between AVM execution and the
ledger being used to lookup resources. Previously, the ledger kept
track of the appID being executed, to offer a narrower interface to
those resources. But now, with app-to-app calls, the appID being
executed must change, and the AVM needs to maintain the current appID.
* Stupid linter
* Fix unit tests error messages
* Allow access to resources created in the same transaction group
The method will be reworked, but the tests are correct and want to get
them visible to team.
* Access to apps created in group
Also adds some tests that are currently skipped for testing
- access to addresses of newly created apps
- use of gaid in inner transactions
Both require some work to implement the thing being tested.
* Remove tracked created mechanism in favor of examining applydata.
* Allow v6 AVM code to use in-group created asas, apps (& their accts)
One exception - apps can not mutate (put or del) keys from the app
accounts, because EvalDelta cannot encode such changes.
* lint docs
* typo
* The review dog needs obedience training.
* Use one EvalParams for logic evals, another for apps in dry run
We used to use one ep per transaction, shared between sig and and
app. But the new model of ep usage is to keep using one while
evaluating an entire group.
The app ep is now built logic.NewAppEvalParams which, hopefully, will
prevent some bugs when we change something in the EvalParams and don't
reflect it in what was a "raw" EvalParams construction in debugger and
dry run.
* Use logic.NewAppEvalParams to decrease copying and bugs in debugger
* Simplify use of NewEvalParams. No more nil return when no apps.
This way, NewEvalParams can be used for all creations of EvalParams,
whether they are intended for logicsig or app use, greatly simplifying
the way we make them for use by dry run or debugger (where they serve
double duty).
* Remove explicit PastSideEffects handling in tealdbg
* Always create EvalParams to evaluate a transaction group.
We used to have an optimization to avoid creating EvalParams unless
there was an app call in the transaction group. But the interface to
allow transaction processing to communicate changes into the
EvalParams is complicated by that (we must only do it if there is
one!)
This also allows us to use the same construction function for eps
created for app and logic evaluation, simplifying dry-run and
debugger.
The optimization is less needed now anyway:
1) The ep is now shared for the whole group, so it's only one.
2) The ep is smaller now, as we only store nil pointers instead of
larger scratch space objects for non-app calls.
* Correct mistaken commit
* Spec improvments
* More spec improvments, including resource "availability"
* Recursively return inner transaction tree
* Lint
* No need for ConfirmedRound, so don't deref a nil pointer!
* license check
* Shut up, dawg.
* base64 merge cleanup
|
