Stream: git-wasmtime

Topic: wasmtime / PR #2358 Use the `psm` crate to figure out the...

Wasmtime GitHub notifications bot (Nov 04 2020 at 15:28):

alexcrichton opened PR #2358 from use-psm to main:

Currently the runtime needs to acquire the current stack pointer so it
can set a limit for where if the wasm stack goes below that point it
will abort the wasm code. Acquiring the stack pointer is done in a
brittle way right now which involves looking at the address of what we
hope is an on-stack structure. This turns out to not work at all with
ASan as well.

Instead this commit switches to the psm crate which is used by the
Rust compiler team for stack manipulation, namely a coarse version of
segmented stacks to avoid stack overflow in the compiler. We don't need
most of the implementation of psm, just the stack_pointer function,
but it shouldn't be a burden to bring in!

Closes #2344

Wasmtime GitHub notifications bot (Nov 04 2020 at 16:21):

tschneidereit submitted PR Review.

Wasmtime GitHub notifications bot (Nov 04 2020 at 16:21):

tschneidereit created PR Review Comment:

Since I just looked at this: "educated guess" makes me a little bit nervous, given the importance of this :) Can you talk a bit about the implications of this, both in terms of security and, if any, resource use limits?

If no implications for either exists, then I think it'd at least be good to update the comment and explain why this is all fine.

Wasmtime GitHub notifications bot (Nov 04 2020 at 16:29):

alexcrichton updated PR #2358 from use-psm to main:

Currently the runtime needs to acquire the current stack pointer so it
can set a limit for where if the wasm stack goes below that point it
will abort the wasm code. Acquiring the stack pointer is done in a
brittle way right now which involves looking at the address of what we
hope is an on-stack structure. This turns out to not work at all with
ASan as well.

Instead this commit switches to the psm crate which is used by the
Rust compiler team for stack manipulation, namely a coarse version of
segmented stacks to avoid stack overflow in the compiler. We don't need
most of the implementation of psm, just the stack_pointer function,
but it shouldn't be a burden to bring in!

Closes #2344

Wasmtime GitHub notifications bot (Nov 04 2020 at 16:30):

alexcrichton submitted PR Review.

Wasmtime GitHub notifications bot (Nov 04 2020 at 16:30):

alexcrichton created PR Review Comment:

Ah good point! I've updated the comment. Now that we're using something like psm it's more official and less educated. Otherwise this just refers to the imprecision about accounting for wasm stacks. If you ask for wasm to have a megabyte of stack, we may end up giving it a megabyte plus or minus a few hundred bytes.

Wasmtime GitHub notifications bot (Nov 04 2020 at 22:33):

tschneidereit submitted PR Review.

Wasmtime GitHub notifications bot (Nov 04 2020 at 22:33):

tschneidereit created PR Review Comment:

Great comment, thank you!

Wasmtime GitHub notifications bot (Nov 04 2020 at 22:34):

tschneidereit requested yurydelendik for a review on PR #2358.

Wasmtime GitHub notifications bot (Nov 04 2020 at 22:34):

tschneidereit submitted PR Review.

Wasmtime GitHub notifications bot (Nov 05 2020 at 13:28):

yurydelendik submitted PR Review.

Wasmtime GitHub notifications bot (Nov 05 2020 at 13:29):

yurydelendik merged PR #2358.

Last updated: Nov 22 2024 at 16:03 UTC