Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

perf: use jemalloc as the global allocator on unix (try 2) #7448

Merged
merged 1 commit into from
Mar 28, 2024

Conversation

DaniPopes
Copy link
Member

@DaniPopes DaniPopes commented Mar 19, 2024

We should be able to enable this on unix and x64. Cannot enable on:

@DaniPopes DaniPopes requested a review from onbjerg March 19, 2024 17:09
@DaniPopes DaniPopes requested a review from mattsse as a code owner March 19, 2024 17:09
@onbjerg
Copy link
Member

onbjerg commented Mar 19, 2024

linux arm: incompatible page sizes

this is not strictly true. where do we depend on the page size?

also: "The architecture allows up to 4 levels of translation tables with a 4KB page size and up to 3 levels with a 64KB page size." (4kb 4 layers is the same as x86) https://docs.kernel.org/arch/arm64/memory.html

@DaniPopes
Copy link
Member Author

I don't know, I just know that aarch64 mostly breaks if we enable jemalloc on it, as per many reth issues and the linked issue

@onbjerg
Copy link
Member

onbjerg commented Mar 20, 2024

as per many reth issues

this is a special case for reth because we exhaust the virtual address space on some arm architectures, this should not be the case for foundry

@DaniPopes
Copy link
Member Author

DaniPopes commented Mar 22, 2024

It's not about exhausting any address space, it straight up just does not work, see

@DaniPopes DaniPopes merged commit 369597f into master Mar 28, 2024
20 checks passed
@DaniPopes DaniPopes deleted the dani/jemalloc branch March 28, 2024 02:41
ZhangZhuoSJTU pushed a commit to MEDGA-eth/foundry that referenced this pull request Mar 29, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants