[EIP2315] Add tests for subroutines #685

matkt · 2020-04-27T15:33:57Z

Addition of tests for the https://eips.ethereum.org/EIPS/eip-2315. Tests generated with the current Besu implementation of the EIP2315.

The tests are those present in the description of the EIP.

Need update of retesteth ethereum/retesteth#90

holiman · 2020-04-28T07:14:40Z

Awesome, well done!
I will test the tests and report back. One more thing:

There's been some feedback that may change the EIP slightly, by preventing "walk in" into a subroutine. It's not certain the change will make it in, but it's at least possible, which would mean that the tests would have to be regenerated.
There's one more test that I didn't write to the EIP, because the trace didn't fit into markdown: https://github.com/ethereum/go-ethereum/pull/20619/files#diff-2e8d7ba640acef6aec90e2012851771fR375 .

	// The code recursively calls itself. It should error when the returns-stack
	// grows above 1023
	state.SetCode(address, []byte{
		byte(vm.BEGINSUB),
		byte(vm.PUSH1), 0,
		byte(vm.JUMPSUB),
	})

This case is a bit problematic to test -- because since it exits with error, it will consume all gas regardless of whether it exits at step 3072 or 3075 or 3069. So in order to actually produce a difference in the poststate, we'd have to store a counter at e.g. slot zero, and increment every loop.

holiman · 2020-04-28T07:38:57Z

I can verify that the state-tests pass seems to be correct, at least according to the geth-implementation:

[user@work evm]$ bash testsubroutines.sh 
Testing /home/user/workspace/tests/GeneralStateTests/stSubroutine/simpleSubroutine.json
{"pc":0,"op":96,"gas":"0x5c878","gasCost":"0x3","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"PUSH1","error":""}
{"pc":2,"op":179,"gas":"0x5c875","gasCost":"0x8","memory":"0x","memSize":0,"stack":["0x4"],"returnStack":[],"depth":1,"refund":0,"opName":"JUMPSUB","error":""}
{"pc":4,"op":178,"gas":"0x5c86d","gasCost":"0x1","memory":"0x","memSize":0,"stack":[],"returnStack":["0x2"],"depth":1,"refund":0,"opName":"BEGINSUB","error":""}
{"pc":5,"op":183,"gas":"0x5c86c","gasCost":"0x2","memory":"0x","memSize":0,"stack":[],"returnStack":["0x2"],"depth":1,"refund":0,"opName":"RETURNSUB","error":""}
{"pc":3,"op":0,"gas":"0x5c86a","gasCost":"0x0","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"STOP","error":""}
{"output":"","gasUsed":"0xe","time":116215}
{"stateRoot": "8d2ab4d286b975b1695fd508b2a693a0c145b18ed3556af671d1c9854a68948f"}
[
  {
    "name": "simpleSubroutine",
    "pass": true,
    "fork": "Berlin"
  }
]
Testing /home/user/workspace/tests/GeneralStateTests/stSubroutine/subroutineAtEndOfCode.json
{"pc":0,"op":96,"gas":"0x5c878","gasCost":"0x3","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"PUSH1","error":""}
{"pc":2,"op":86,"gas":"0x5c875","gasCost":"0x8","memory":"0x","memSize":0,"stack":["0x5"],"returnStack":[],"depth":1,"refund":0,"opName":"JUMP","error":""}
{"pc":5,"op":91,"gas":"0x5c86d","gasCost":"0x1","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"JUMPDEST","error":""}
{"pc":6,"op":96,"gas":"0x5c86c","gasCost":"0x3","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"PUSH1","error":""}
{"pc":8,"op":179,"gas":"0x5c869","gasCost":"0x8","memory":"0x","memSize":0,"stack":["0x3"],"returnStack":[],"depth":1,"refund":0,"opName":"JUMPSUB","error":""}
{"pc":3,"op":178,"gas":"0x5c861","gasCost":"0x1","memory":"0x","memSize":0,"stack":[],"returnStack":["0x8"],"depth":1,"refund":0,"opName":"BEGINSUB","error":""}
{"pc":4,"op":183,"gas":"0x5c860","gasCost":"0x2","memory":"0x","memSize":0,"stack":[],"returnStack":["0x8"],"depth":1,"refund":0,"opName":"RETURNSUB","error":""}
{"pc":9,"op":0,"gas":"0x5c85e","gasCost":"0x0","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"STOP","error":""}
{"output":"","gasUsed":"0x1a","time":336982}
{"stateRoot": "805db559e0a7b054b323ca8f4b9f449d79a444bc1d5cac379bae9a242dde0b2a"}
[
  {
    "name": "subroutineAtEndOfCode",
    "pass": true,
    "fork": "Berlin"
  }
]
Testing /home/user/workspace/tests/GeneralStateTests/stSubroutine/subroutineInvalidJump.json
{"pc":0,"op":104,"gas":"0x5c878","gasCost":"0x3","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"PUSH9","error":""}
{"pc":10,"op":179,"gas":"0x5c875","gasCost":"0x8","memory":"0x","memSize":0,"stack":["0x1000000000000000c"],"returnStack":[],"depth":1,"refund":0,"opName":"JUMPSUB","error":""}
{"output":"","gasUsed":"0x5c878","time":1199045,"error":"evm: invalid jump destination"}
{"stateRoot": "b192e50b963cb534265da00cca90f272cd2c258b95d5c49edd37005307ade18f"}
[
  {
    "name": "subroutineInvalidJump",
    "pass": true,
    "fork": "Berlin"
  }
]
Testing /home/user/workspace/tests/GeneralStateTests/stSubroutine/subroutineShallowReturnStack.json
{"pc":0,"op":183,"gas":"0x5c878","gasCost":"0x2","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"RETURNSUB","error":""}
{"output":"","gasUsed":"0x5c878","time":92488,"error":"evm: invalid retsub"}
{"stateRoot": "19f388de55450373529c40d705d958602672f6dc33a0df093c17159942247de2"}
[
  {
    "name": "subroutineShallowReturnStack",
    "pass": true,
    "fork": "Berlin"
  }
]
Testing /home/user/workspace/tests/GeneralStateTests/stSubroutine/twoLevelsSubroutines.json
{"pc":0,"op":104,"gas":"0x5c878","gasCost":"0x3","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"PUSH9","error":""}
{"pc":10,"op":179,"gas":"0x5c875","gasCost":"0x8","memory":"0x","memSize":0,"stack":["0xc"],"returnStack":[],"depth":1,"refund":0,"opName":"JUMPSUB","error":""}
{"pc":12,"op":178,"gas":"0x5c86d","gasCost":"0x1","memory":"0x","memSize":0,"stack":[],"returnStack":["0xa"],"depth":1,"refund":0,"opName":"BEGINSUB","error":""}
{"pc":13,"op":96,"gas":"0x5c86c","gasCost":"0x3","memory":"0x","memSize":0,"stack":[],"returnStack":["0xa"],"depth":1,"refund":0,"opName":"PUSH1","error":""}
{"pc":15,"op":179,"gas":"0x5c869","gasCost":"0x8","memory":"0x","memSize":0,"stack":["0x11"],"returnStack":["0xa"],"depth":1,"refund":0,"opName":"JUMPSUB","error":""}
{"pc":17,"op":178,"gas":"0x5c861","gasCost":"0x1","memory":"0x","memSize":0,"stack":[],"returnStack":["0xa","0xf"],"depth":1,"refund":0,"opName":"BEGINSUB","error":""}
{"pc":18,"op":183,"gas":"0x5c860","gasCost":"0x2","memory":"0x","memSize":0,"stack":[],"returnStack":["0xa","0xf"],"depth":1,"refund":0,"opName":"RETURNSUB","error":""}
{"pc":16,"op":183,"gas":"0x5c85e","gasCost":"0x2","memory":"0x","memSize":0,"stack":[],"returnStack":["0xa"],"depth":1,"refund":0,"opName":"RETURNSUB","error":""}
{"pc":11,"op":0,"gas":"0x5c85c","gasCost":"0x0","memory":"0x","memSize":0,"stack":[],"returnStack":[],"depth":1,"refund":0,"opName":"STOP","error":""}
{"output":"","gasUsed":"0x1c","time":147835}
{"stateRoot": "0e2de1dc4001ee621577748b389f923822b0a9bb0208dc3f77164dc6b92c080a"}
[
  {
    "name": "twoLevelsSubroutines",
    "pass": true,
    "fork": "Berlin"
  }
]

holiman

LGTM, but let's wait a bit more and see if the EIP will change

holiman · 2020-04-28T07:44:14Z

Worth noting, though, is that two of the tests suffer from the same problem that I described regarding that one that exits with error.
If I run the tests without Eip-2315 enabled, two of them still pass - but for the wrong reason:

Testing /home/user/workspace/tests/GeneralStateTests/stSubroutine/subroutineInvalidJump.json
{"pc":0,"op":104,"gas":"0x5c878","gasCost":"0x3","memory":"0x","memSize":0,"stack":[],"depth":1,"refund":0,"opName":"PUSH9","error":""}
{"pc":10,"op":179,"gas":"0x5c875","gasCost":"0x3","memory":"0x","memSize":0,"stack":["0x1000000000000000c"],"depth":1,"refund":0,"opName":"opcode 0xb3 not defined","error":"invalid opcode: opcode 0xb3 not defined"}
{"output":"","gasUsed":"0x5c878","time":121676,"error":"invalid opcode: opcode 0xb3 not defined"}
{"stateRoot": "b192e50b963cb534265da00cca90f272cd2c258b95d5c49edd37005307ade18f"}
[
  {
    "name": "subroutineInvalidJump",
    "pass": true,
    "fork": "Berlin"
  }
]
Testing /home/user/workspace/tests/GeneralStateTests/stSubroutine/subroutineShallowReturnStack.json
{"pc":0,"op":183,"gas":"0x5c878","gasCost":"0x0","memory":"0x","memSize":0,"stack":[],"depth":1,"refund":0,"opName":"opcode 0xb7 not defined","error":"invalid opcode: opcode 0xb7 not defined"}
{"output":"","gasUsed":"0x5c878","time":143542,"error":"invalid opcode: opcode 0xb7 not defined"}
{"stateRoot": "19f388de55450373529c40d705d958602672f6dc33a0df093c17159942247de2"}
[
  {
    "name": "subroutineShallowReturnStack",
    "pass": true,
    "fork": "Berlin"
  }
]

I don't have any splendid ideas immediately on how to improve the accuracy of those tests

matkt · 2020-04-28T12:43:56Z

There's one more test that I didn't write to the EIP, because the trace didn't fit into markdown

I tried your bytecode which allows you to reach the limit of the rstack. But when we add a SSTORE to save a value, the storage in the postState remains empty in case of error.

A simple test without counter

0x6001600055b26005b3 (store a simple value before the recursive call) : '0x00' does not exist
0x6001600055 (store a simple value without the recursive call) : '0x00' exist

holiman · 2020-04-28T13:14:03Z

But when we add a SSTORE to save a value, the storage in the postState remains empty in case of error.

Ah right (doh!)
I guess we could do one test that loops + stores up to "one below the limit", and check that we can get that far. Then another test where we try to loop one more time, and expect that one to fail.

holiman · 2020-04-28T13:37:45Z

Here's an example that does it five times:

The code recursively calls into a subroutine, N times. It should succeeed if N is below the threshold for rstack size
Bytecode: 0x610005600055b260016000540380156018576000556006b35b
From: 0x0000000000000000000000000000000000000000
To: 0x000000000000000000000000636F6E7472616374
Data: 0x
Gas: 50000
Value 0 wei

Pc	Op	Cost	Stack	RStack
0	PUSH2	3	[]	[]
3	PUSH1	3	[5]	[]
5	SSTORE	20000	[5,0]	[]
6	BEGINSUB	1	[]	[]
7	PUSH1	3	[]	[]
9	PUSH1	3	[1]	[]
11	SLOAD	200	[1,0]	[]
12	SUB	3	[1,5]	[]
13	DUP1	3	[4]	[]
14	ISZERO	3	[4,4]	[]
15	PUSH1	3	[4,0]	[]
17	JUMPI	10	[4,0,24]	[]
18	PUSH1	3	[4]	[]
20	SSTORE	5000	[4,0]	[]
21	PUSH1	3	[]	[]
23	JUMPSUB	8	[6]	[]
6	BEGINSUB	1	[]	[23]
7	PUSH1	3	[]	[23]
9	PUSH1	3	[1]	[23]
11	SLOAD	200	[1,0]	[23]
12	SUB	3	[1,4]	[23]
13	DUP1	3	[3]	[23]
14	ISZERO	3	[3,3]	[23]
15	PUSH1	3	[3,0]	[23]
17	JUMPI	10	[3,0,24]	[23]
18	PUSH1	3	[3]	[23]
20	SSTORE	5000	[3,0]	[23]
21	PUSH1	3	[]	[23]
23	JUMPSUB	8	[6]	[23]
6	BEGINSUB	1	[]	[23,23]
7	PUSH1	3	[]	[23,23]
9	PUSH1	3	[1]	[23,23]
11	SLOAD	200	[1,0]	[23,23]
12	SUB	3	[1,3]	[23,23]
13	DUP1	3	[2]	[23,23]
14	ISZERO	3	[2,2]	[23,23]
15	PUSH1	3	[2,0]	[23,23]
17	JUMPI	10	[2,0,24]	[23,23]
18	PUSH1	3	[2]	[23,23]
20	SSTORE	5000	[2,0]	[23,23]
21	PUSH1	3	[]	[23,23]
23	JUMPSUB	8	[6]	[23,23]
6	BEGINSUB	1	[]	[23,23,23]
7	PUSH1	3	[]	[23,23,23]
9	PUSH1	3	[1]	[23,23,23]
11	SLOAD	200	[1,0]	[23,23,23]
12	SUB	3	[1,2]	[23,23,23]
13	DUP1	3	[1]	[23,23,23]
14	ISZERO	3	[1,1]	[23,23,23]
15	PUSH1	3	[1,0]	[23,23,23]
17	JUMPI	10	[1,0,24]	[23,23,23]
18	PUSH1	3	[1]	[23,23,23]
20	SSTORE	5000	[1,0]	[23,23,23]
21	PUSH1	3	[]	[23,23,23]
23	JUMPSUB	8	[6]	[23,23,23]
6	BEGINSUB	1	[]	[23,23,23,23]
7	PUSH1	3	[]	[23,23,23,23]
9	PUSH1	3	[1]	[23,23,23,23]
11	SLOAD	200	[1,0]	[23,23,23,23]
12	SUB	3	[1,1]	[23,23,23,23]
13	DUP1	3	[0]	[23,23,23,23]
14	ISZERO	3	[0,0]	[23,23,23,23]
15	PUSH1	3	[0,1]	[23,23,23,23]
17	JUMPI	10	[0,1,24]	[23,23,23,23]
24	JUMPDEST	1	[0]	[23,23,23,23]
25	STOP	0	[0]	[23,23,23,23]

holiman · 2020-04-28T13:44:32Z

0x610400600055b260016000540380156018576000556006b35b sets N to 1024 (succeeds on geth)
0x610401600055b260016000540380156018576000556006b35b sets N to 1025 (fails on geth)

This is the underlying instrutions:

		code := []byte{
			byte(vm.PUSH2), 0x04, 0x01, //value : 0x400 = 1024
			byte(vm.PUSH1), 0, // location
			byte(vm.SSTORE), // set slot 0 to 1024
			byte(vm.BEGINSUB),
			// if slot 0 == 0 then break
			byte(vm.PUSH1),1, // push the 1 for the sub
			byte(vm.PUSH1), 0, // slot number
			byte(vm.SLOAD), // slot 0 on stack
			byte(vm.SUB), // slot 0 minus one
			byte(vm.DUP1), // dup it
			byte(vm.ISZERO), 
			byte(vm.PUSH1), 24, // jumpdestination if we want to exit
			byte(vm.JUMPI), // jump if the slot reached zero
			// store it again (the new value we dup:ed earlier)
			byte(vm.PUSH1), 0, // location
			byte(vm.SSTORE), // set slot 0 to new number
			// jump into subroutine again
			byte(vm.PUSH1), 6,
			byte(vm.JUMPSUB),
			// exit label
			byte(vm.JUMPDEST),
		}

holiman · 2020-04-28T13:45:40Z

Oh, and the code below only "stores back" the counter after the zero-check, so it should end up with 1 on slot 0 in the success-case, because if it's zero it jumps out instead

matkt · 2020-04-28T15:35:11Z

Oh, and the code below only "stores back" the counter after the zero-check, so it should end up with 1 on slot 0 in the success-case, because if it's zero it jumps out instead

I had not seen your message and I added my own implementation which is very close to yours. As you have already documented yours I will update my tests

matkt · 2020-04-28T15:58:30Z

0x610400600055b260016000540380156018576000556006b35b sets N to 1024 (succeeds on geth)
0x610401600055b260016000540380156018576000556006b35b sets N to 1025 (fails on geth)

This is the underlying instrutions:

		code := []byte{
			byte(vm.PUSH2), 0x04, 0x01, //value : 0x400 = 1024
			byte(vm.PUSH1), 0, // location
			byte(vm.SSTORE), // set slot 0 to 1024
			byte(vm.BEGINSUB),
			// if slot 0 == 0 then break
			byte(vm.PUSH1),1, // push the 1 for the sub
			byte(vm.PUSH1), 0, // slot number
			byte(vm.SLOAD), // slot 0 on stack
			byte(vm.SUB), // slot 0 minus one
			byte(vm.DUP1), // dup it
			byte(vm.ISZERO), 
			byte(vm.PUSH1), 24, // jumpdestination if we want to exit
			byte(vm.JUMPI), // jump if the slot reached zero
			// store it again (the new value we dup:ed earlier)
			byte(vm.PUSH1), 0, // location
			byte(vm.SSTORE), // set slot 0 to new number
			// jump into subroutine again
			byte(vm.PUSH1), 6,
			byte(vm.JUMPSUB),
			// exit label
			byte(vm.JUMPDEST),
		}

I just put your implementation. tests work well on Besu

winsvega

There are some tests with 1023/24/25 subcalls already. You can also peek if there are some ways to organize stack depth calls for tests

src/GeneralStateTestsFiller/stSubroutine/twoLevelsSubroutinesFiller.json

src/GeneralStateTestsFiller/stSubroutine/subroutineShallowReturnStackFiller.json

src/GeneralStateTestsFiller/stSubroutine/subroutineInvalidJumpFiller.json

src/GeneralStateTestsFiller/stSubroutine/subroutineAtEndOfCodeFiller.json

src/GeneralStateTestsFiller/stSubroutine/simpleSubroutineFiller.json

src/GeneralStateTestsFiller/stSubroutine/shouldSucceedWhenReturnStackGrowsUntil1023Filler.json

src/GeneralStateTestsFiller/stSubroutine/shouldErrorWhenReturnStackGrowsAbove1023Filler.json

matkt · 2020-05-18T13:03:02Z

I just updated the tests to manage the last PR (enter only via jumpsub) ethereum/EIPs#2646

I updated shouldSucceedWhenReturnStackGrowsUntil1023 and shouldErrorWhenReturnStackGrowsAbove1023 by adding a JUMPSUB at the start to not go through the BEGINSUB and to not break the test with the last update

Before : 0x610400600055b260016000540380156018576000556006b35b 
After :  0x61[03FF]600055[6009b3]b2600160005403801560[1B]5760005560[09]b35b

I added three new tests for Subroutines can only be entered via JUMPSUB and not BEGINSUB. Attempted execution of a BEGINSUB causes an abort: terminate execution with an OOG (Out Of Gas) exception

shouldErrorWhenExecuteBeginSub

Bytecode: `0x610400b2`
LLL :  (asm 0x0400 BEGINSUB)

PC	Opcode	Cost	Stack	RStack
0	PUSH2	3	[]	[]
3	BEGINSUB	0	[1024]	[]

Error: at pc=3, op=BEGINSUB: evm: out of gas

shouldErrorWhenSubroutineEnteredViaBeginSub

Bytecode: `0x6004b300b2b2b7`
LLL :  (asm 0x04 JUMPSUB STOP BEGINSUB BEGINSUB RETURNSUB)

PC	Opcode	Cost	Stack	RStack
0	PUSH1	3	[]	[]
2	JUMPSUB	8	[4]	[]
5	BEGINSUB	0	[]	[2]

Error: at pc=5, op=BEGINSUB: evm: out of gas

beginSubAtEndOfCode

Bytecode: `0x6003b3b2`
LLL : (asm 0x03 JUMPSUB BEGINSUB)

PC	Opcode	Cost	Stack	RStack
0	PUSH1	3	[]	[]
2	JUMPSUB	8	[3]	[]
4	STOP	0	[]	[]

holiman · 2020-05-19T14:00:18Z

Hm, I'm actually getting an error now

    --- FAIL: TestState/shouldSucceedWhenReturnStackGrowsUntil1023.json (0.09s)
            state_test.go:96: post state root mismatch: got abe6b4c387d5363f896de0c5c3aa9f2762f1e4931975b83dbecdcea962ec5695, want 4b4ded39ed6ea35bbc1b1581b1f657ccb15a834fb6ce641c2280456a9fd4178b

matkt · 2020-05-19T14:07:33Z

Hm, I'm actually getting an error now

    --- FAIL: TestState/shouldSucceedWhenReturnStackGrowsUntil1023.json (0.09s)
            state_test.go:96: post state root mismatch: got abe6b4c387d5363f896de0c5c3aa9f2762f1e4931975b83dbecdcea962ec5695, want 4b4ded39ed6ea35bbc1b1581b1f657ccb15a834fb6ce641c2280456a9fd4178b

are you testing with the latest updates?

holiman · 2020-05-19T14:09:45Z

Sorry, my bad. We haven't enabled 'Berlin', so I added a hack to redefine "Berlin" as "Byzantium+2315", but the correct thing was Istanbul+2315. With that, the tests matches geth! Well done!

holiman · 2020-05-19T14:11:45Z

I guess besu doesn't use the base+EIP notation? It would be pretty handy if other clients supported that, because it makes things easier -- in the case where we have multiple eips going into a fork eventually, and it's not certain which ones were in when a specific test was generated.

matkt · 2020-05-19T18:06:46Z

I guess besu doesn't use the base+EIP notation? It would be pretty handy if other clients supported that, because it makes things easier -- in the case where we have multiple eips going into a fork eventually, and it's not certain which ones were in when a specific test was generated.

Yes, for the moment it is not managed. we will try to manage this feature as soon as possible

matkt · 2020-05-25T16:51:20Z

@holiman thank you for your review on this PR. As the tests are pushed in develop does that mean that it confirms a JUMPSUB which costs 8?

holiman · 2020-05-25T17:48:36Z

I can't really make that call, but it's definitely my opinion. I will make a PR to update the eip and reach out to other implementors

Add tests for subroutines

0957b61

matkt mentioned this pull request Apr 27, 2020

[EIP2315] Add subroutine tests ethereum/retesteth#90

Closed

matkt marked this pull request as ready for review April 27, 2020 15:40

holiman approved these changes Apr 28, 2020

View reviewed changes

Add new tests in order to check the limits of the return stack

c975ec6

Change implementation of recursive call

7a33ba8

holgerd77 mentioned this pull request May 11, 2020

[Berlin] EIP-2315: Simple Subroutines for the EVM ethereumjs/ethereumjs-monorepo#737

Closed

winsvega reviewed May 11, 2020

View reviewed changes

matkt added 2 commits May 12, 2020 14:18

add LLL code for subroutine tests

8efc9b2

fix filler format

96ec2d3

holgerd77 mentioned this pull request May 14, 2020

v7.0.0 pre-Berlin HF Release #687

Merged

add and update tests for PR (enter only via jumpsub) ethereum/EIPs#2646

a1558e8

matkt mentioned this pull request May 18, 2020

Update EIP-2315 implementation (enter only via jumpsub) hyperledger/besu#941

Merged

add test when BEGINSUB is the last instruction in code

4f553e4

holiman merged commit 66b1043 into ethereum:develop May 24, 2020

holiman mentioned this pull request May 25, 2020

Update simple subroutines (EIP 2315) to the latest spec openethereum/parity-ethereum#11731

Merged

3 tasks

This was referenced May 28, 2020

New v8.0.0-beta.1 Berlin HF release #692

Closed

EIP-2315 changes in gascosts and update opcode #693

Merged

shemnon mentioned this pull request May 31, 2020

Reduce balances in SHA3 tests #688

Closed

winsvega mentioned this pull request Nov 27, 2020

The recent tests changes #763

Closed

Provide feedback

Saved searches

Use saved searches to filter your results more quickly

[EIP2315] Add tests for subroutines #685

[EIP2315] Add tests for subroutines #685

matkt commented Apr 27, 2020 •

edited

Loading

holiman commented Apr 28, 2020

holiman commented Apr 28, 2020

holiman left a comment

holiman commented Apr 28, 2020

matkt commented Apr 28, 2020 •

edited

Loading

holiman commented Apr 28, 2020

holiman commented Apr 28, 2020

holiman commented Apr 28, 2020 •

edited

Loading

holiman commented Apr 28, 2020 •

edited

Loading

matkt commented Apr 28, 2020

matkt commented Apr 28, 2020

winsvega left a comment

matkt commented May 18, 2020 •

edited

Loading

holiman commented May 19, 2020 •

edited

Loading

matkt commented May 19, 2020

holiman commented May 19, 2020

holiman commented May 19, 2020

matkt commented May 19, 2020

matkt commented May 25, 2020

holiman commented May 25, 2020 via email

[EIP2315] Add tests for subroutines #685

[EIP2315] Add tests for subroutines #685

Conversation

matkt commented Apr 27, 2020 • edited Loading

holiman commented Apr 28, 2020

holiman commented Apr 28, 2020

holiman left a comment

Choose a reason for hiding this comment

holiman commented Apr 28, 2020

matkt commented Apr 28, 2020 • edited Loading

holiman commented Apr 28, 2020

holiman commented Apr 28, 2020

holiman commented Apr 28, 2020 • edited Loading

holiman commented Apr 28, 2020 • edited Loading

matkt commented Apr 28, 2020

matkt commented Apr 28, 2020

winsvega left a comment

Choose a reason for hiding this comment

matkt commented May 18, 2020 • edited Loading

holiman commented May 19, 2020 • edited Loading

matkt commented May 19, 2020

holiman commented May 19, 2020

holiman commented May 19, 2020

matkt commented May 19, 2020

matkt commented May 25, 2020

holiman commented May 25, 2020 via email

matkt commented Apr 27, 2020 •

edited

Loading

matkt commented Apr 28, 2020 •

edited

Loading

holiman commented Apr 28, 2020 •

edited

Loading

holiman commented Apr 28, 2020 •

edited

Loading

matkt commented May 18, 2020 •

edited

Loading

holiman commented May 19, 2020 •

edited

Loading