This repository was archived by the owner on Nov 6, 2020. It is now read-only.
Change test miner max memory to support malloc allocator#10024
Merged
Conversation
|
It looks like @cheme signed our Contributor License Agreement. 👍 Many thanks, Parity Technologies CLA Bot |
cheme
added
A0-pleasereview 🤓
ordian
approved these changes
Dec 6, 2018
Member
ordian
left a comment
ordian
left a comment
There was a problem hiding this comment.
For the record, heapsize actually works with malloc, it's just malloc can consume more memory than jemalloc in some cases , so that's why our tests in ethore-miner, which depended on max_mem, failed. This PR fixes that for the case of malloc allocator. We still need to replace heapsize because of issues on macOS like #9167. And malloc_size_of could be a viable candidate for the replacement.
ordian
added
A8-looksgood 🦄
ascjones
approved these changes
Dec 6, 2018
tomaka
approved these changes
Dec 6, 2018
5chdn
added
B1-patch-beta 🕷🕷
B0-patch-stable 🕷
niklasad1
approved these changes
Dec 7, 2018
This was referenced Dec 11, 2018
niklasad1
pushed a commit
that referenced
this pull request
Dec 12, 2018
niklasad1
pushed a commit
that referenced
this pull request
Dec 12, 2018
Tbaut
added a commit
that referenced
this pull request
Dec 13, 2018
* bump beta to 2.2.5 * Fix empty steps (#9939) * Don't send empty step twice or empty step then block. * Perform basic validation of locally sealed blocks. * Don't include empty step twice. * Strict empty steps validation (#10041) * Add two failings tests for strict empty steps. * Implement strict validation of empty steps. * ethcore: enable constantinople on ethereum (#10031) * ethcore: change blockreward to 2e18 for foundation after constantinople * ethcore: delay diff bomb by 2e6 blocks for foundation after constantinople * ethcore: enable eip-{145,1014,1052,1283} for foundation after constantinople * Change test miner max memory to malloc reports. (#10024) * Fix: test corpus_inaccessible panic (#10019) If system uptime is less than the duration in the test, thread will panic due to: 'overflow when subtracting duration from instant'. Changed duration to 20 seconds to make it unlikely the test will fail due to this condition. Since no operations that carry a similar risk of panic occur outside of the tests, it doesn't seem warranted to depend on an external crate to fix this. * Bump crossbeam. (#10048)
Tbaut
added a commit
that referenced
this pull request
Dec 13, 2018
* bump stable to 2.1.10 * RPC: parity_getBlockReceipts (#9527) * Block receipts RPC. * Use lazy evaluation of block receipts (ecrecover). * Optimize transaction_receipt to prevent performance regression. * Fix RPC grumbles. * Add block & transaction receipt tests. * Fix conversion to block id. * Update a few parity-common dependencies (#9663) * Update a few parity-common dependencies * cleanup * cleanup * revert update of ethereum/tests * better reporting of network rlp errors * Use rlp 0.3.0-beta.1 * fix util function get_dummy_blocks * Already a Vec * encode_list returns vec already * Address grumble * No need for betas * Fix double spaces * Fix empty steps (#9939) * Don't send empty step twice or empty step then block. * Perform basic validation of locally sealed blocks. * Don't include empty step twice. * Strict empty steps validation (#10041) * Add two failings tests for strict empty steps. * Implement strict validation of empty steps. * ethcore: enable constantinople on ethereum (#10031) * ethcore: change blockreward to 2e18 for foundation after constantinople * ethcore: delay diff bomb by 2e6 blocks for foundation after constantinople * ethcore: enable eip-{145,1014,1052,1283} for foundation after constantinople * Change test miner max memory to malloc reports. (#10024) * Bump crossbeam. (#10048) * Revert "Bump crossbeam. (#10048)" This reverts commit ed1db0c.
niklasad1
pushed a commit
that referenced
this pull request
Dec 16, 2018
insipx
pushed a commit
to insipx/parity-ethereum
that referenced
this pull request
Dec 17, 2018
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to subscribe to this conversation on GitHub.
Already have an account?
Sign in.
7 participants
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Malloc uses a bit more memory for its allocation on small objects (thanks @ordian for noticing).
It breaks some of the mining tests.
This PR changes the test transaction queue memory limit to reflect this. This is a short term solution to #9953.
Ideally the limit could be set depending on the global allocator definition, I am not sure it is worth the additional complexity but is totally doable (I am currently working on swapping heapsize with malloc_size_of and we need at some point to get a handle other the allocator).