-
Notifications
You must be signed in to change notification settings - Fork 30k
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
vm: memory leak on vm.compileFunction with importModuleDynamically #44211
Comments
In case anyone who is confused, the dilemma we have now:
The fix I proposed above works for either of these two issues. |
I would like to hear your ideas :) @legendecas |
@ywave620 hey, thank you for digging into this. The second argument of the As for the segfaults of issues like #43205, I believe a V8 refactoring is the best effort to settle a solution that fits our use case on host defined options: https://chromium-review.googlesource.com/c/v8/v8/+/3172764. I'm drafting a PR based on the V8 refactoring to show how the crash can be fixed: #44923. |
@legendecas
Discussion on the I'm afraid that you and node/vm group have made a mistake in this point from the very beginning. Please correct me if I'm wrong!
|
The term "context" used in v8 and the Node.js VM module refers to different entities. It is important to disambiguate those "contexts" here. When the dynamic import call is evaluated, the |
In the latest node, the actual object passed as the second argument of |
The value passed to the userland hooks is not the exact same value defined in the ecma262 host hooks. Node.js passes its own corresponding implementation-defined values to userland hooks. |
Then this design does not make sense to me. It definitely leads to a memory leak. We have to keep track of every |
Once the |
So this is a contradictory design, and that is why we have a crash in #43205 (comment). We need to pass the |
That is the problem that moving away from the id-based reference table and saving the wrap object in the script/module context is trying to resolve -- the solution illustrated at #44923 :) |
I have a proposed fix for #44923 doesn't fix the leak of Here is my take of the whole issue of lifetime management around these wrappers. What we need to tell V8 is that:
But how Node.js traces the native objects makes it difficult to inform V8's GC about this. To achieve (1), Things are a bit more complicated for Currently to achieve (2), In the case of |
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously when managing the importModuleDynamically callback of vm.compileFunction(), we use an ID number as the host defined option and maintain a per-Environment ID -> CompiledFnEntry map to retain the top-level referrer function returned by vm.compileFunction() in order to pass it back to the callback, but it would leak because with how we used v8::Persistent to maintain this reference, V8 would not be able to understand the cycle and would just think that the CompiledFnEntry was supposed to live forever. We made an attempt to make that reference known to V8 by making the CompiledFnEntry weak and using a private symbol to make CompiledFnEntry strongly references the top-level referrer function in nodejs#46785, but that turned out to be unsound, because the there's no guarantee that the top-level function must be alive while import() can still be initiated from that function, since V8 could discard the top-level function and only keep inner functions alive, so relying on the top-level function to keep the CompiledFnEntry alive could result in use-after-free which caused a revert of that fix. With this patch we use a symbol in the host defined options instead of a number, because with the stage-3 symbol-as-weakmap-keys proposal we could directly use that symbol to keep the referrer alive using a WeakMap. As a bonus this also keeps the other kinds of referrers alive as long as import() can still be initiated from that Script/Module, so this also fixes the long-standing crash caused by vm.Script being GC'ed too early when its importModuleDynamically callback still needs it. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously we maintain a strong persistent reference to the ModuleWrap to retrieve the ID-to-ModuleWrap mapping from the HostImportModuleDynamicallyCallback using the number ID stored in the host-defined options. As a result the ModuleWrap would be kept alive until the Environment is shut down, which would be a leak for user code. With the new symbol-based host-defined option we can just get the ModuleWrap from the JS-land WeakMap so there's now no need to maintain this strong reference. This would at least fix the leak for vm.SyntheticModule. vm.SourceTextModule is still leaking due to the strong persistent reference to the v8::Module. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously when managing the importModuleDynamically callback of vm.compileFunction(), we use an ID number as the host defined option and maintain a per-Environment ID -> CompiledFnEntry map to retain the top-level referrer function returned by vm.compileFunction() in order to pass it back to the callback, but it would leak because with how we used v8::Persistent to maintain this reference, V8 would not be able to understand the cycle and would just think that the CompiledFnEntry was supposed to live forever. We made an attempt to make that reference known to V8 by making the CompiledFnEntry weak and using a private symbol to make CompiledFnEntry strongly references the top-level referrer function in nodejs#46785, but that turned out to be unsound, because the there's no guarantee that the top-level function must be alive while import() can still be initiated from that function, since V8 could discard the top-level function and only keep inner functions alive, so relying on the top-level function to keep the CompiledFnEntry alive could result in use-after-free which caused a revert of that fix. With this patch we use a symbol in the host defined options instead of a number, because with the stage-3 symbol-as-weakmap-keys proposal we could directly use that symbol to keep the referrer alive using a WeakMap. As a bonus this also keeps the other kinds of referrers alive as long as import() can still be initiated from that Script/Module, so this also fixes the long-standing crash caused by vm.Script being GC'ed too early when its importModuleDynamically callback still needs it. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously we maintain a strong persistent reference to the ModuleWrap to retrieve the ID-to-ModuleWrap mapping from the HostImportModuleDynamicallyCallback using the number ID stored in the host-defined options. As a result the ModuleWrap would be kept alive until the Environment is shut down, which would be a leak for user code. With the new symbol-based host-defined option we can just get the ModuleWrap from the JS-land WeakMap so there's now no need to maintain this strong reference. This would at least fix the leak for vm.SyntheticModule. vm.SourceTextModule is still leaking due to the strong persistent reference to the v8::Module. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
hi @joyeecheung , come here because we are experiencing exactly the same memory leak issue, but i see it is closed and also referenced #48510 closed as well...so what are the plans for the fix here or for a workaround? we are on node v18...thanks! |
@stas-sisense #48510 has landed on the main branch, you can try it out using the nightlies (e.g. see https://gist.github.com/joyeecheung/083f6b6d601e64382121001ab87bbed2 on how to install nightlies) or wait for a v20.x release #49874 or the v21.x release next month (I don't think we can backport this to v18.x though, weakmap with symbol keys isn't shipped there) |
Previously when managing the importModuleDynamically callback of vm.compileFunction(), we use an ID number as the host defined option and maintain a per-Environment ID -> CompiledFnEntry map to retain the top-level referrer function returned by vm.compileFunction() in order to pass it back to the callback, but it would leak because with how we used v8::Persistent to maintain this reference, V8 would not be able to understand the cycle and would just think that the CompiledFnEntry was supposed to live forever. We made an attempt to make that reference known to V8 by making the CompiledFnEntry weak and using a private symbol to make CompiledFnEntry strongly references the top-level referrer function in nodejs#46785, but that turned out to be unsound, because the there's no guarantee that the top-level function must be alive while import() can still be initiated from that function, since V8 could discard the top-level function and only keep inner functions alive, so relying on the top-level function to keep the CompiledFnEntry alive could result in use-after-free which caused a revert of that fix. With this patch we use a symbol in the host defined options instead of a number, because with the stage-3 symbol-as-weakmap-keys proposal we could directly use that symbol to keep the referrer alive using a WeakMap. As a bonus this also keeps the other kinds of referrers alive as long as import() can still be initiated from that Script/Module, so this also fixes the long-standing crash caused by vm.Script being GC'ed too early when its importModuleDynamically callback still needs it. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously we maintain a strong persistent reference to the ModuleWrap to retrieve the ID-to-ModuleWrap mapping from the HostImportModuleDynamicallyCallback using the number ID stored in the host-defined options. As a result the ModuleWrap would be kept alive until the Environment is shut down, which would be a leak for user code. With the new symbol-based host-defined option we can just get the ModuleWrap from the JS-land WeakMap so there's now no need to maintain this strong reference. This would at least fix the leak for vm.SyntheticModule. vm.SourceTextModule is still leaking due to the strong persistent reference to the v8::Module. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously when managing the importModuleDynamically callback of vm.compileFunction(), we use an ID number as the host defined option and maintain a per-Environment ID -> CompiledFnEntry map to retain the top-level referrer function returned by vm.compileFunction() in order to pass it back to the callback, but it would leak because with how we used v8::Persistent to maintain this reference, V8 would not be able to understand the cycle and would just think that the CompiledFnEntry was supposed to live forever. We made an attempt to make that reference known to V8 by making the CompiledFnEntry weak and using a private symbol to make CompiledFnEntry strongly references the top-level referrer function in #46785, but that turned out to be unsound, because the there's no guarantee that the top-level function must be alive while import() can still be initiated from that function, since V8 could discard the top-level function and only keep inner functions alive, so relying on the top-level function to keep the CompiledFnEntry alive could result in use-after-free which caused a revert of that fix. With this patch we use a symbol in the host defined options instead of a number, because with the stage-3 symbol-as-weakmap-keys proposal we could directly use that symbol to keep the referrer alive using a WeakMap. As a bonus this also keeps the other kinds of referrers alive as long as import() can still be initiated from that Script/Module, so this also fixes the long-standing crash caused by vm.Script being GC'ed too early when its importModuleDynamically callback still needs it. Backport-PR-URL: #49874 PR-URL: #48510 Refs: #44211 Refs: #42080 Refs: #47096 Refs: #43205 Refs: #38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously we maintain a strong persistent reference to the ModuleWrap to retrieve the ID-to-ModuleWrap mapping from the HostImportModuleDynamicallyCallback using the number ID stored in the host-defined options. As a result the ModuleWrap would be kept alive until the Environment is shut down, which would be a leak for user code. With the new symbol-based host-defined option we can just get the ModuleWrap from the JS-land WeakMap so there's now no need to maintain this strong reference. This would at least fix the leak for vm.SyntheticModule. vm.SourceTextModule is still leaking due to the strong persistent reference to the v8::Module. Backport-PR-URL: #49874 PR-URL: #48510 Refs: #44211 Refs: #42080 Refs: #47096 Refs: #43205 Refs: #38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. Backport-PR-URL: #49874 PR-URL: #48510 Refs: #44211 Refs: #42080 Refs: #47096 Refs: #43205 Refs: #38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously when managing the importModuleDynamically callback of vm.compileFunction(), we use an ID number as the host defined option and maintain a per-Environment ID -> CompiledFnEntry map to retain the top-level referrer function returned by vm.compileFunction() in order to pass it back to the callback, but it would leak because with how we used v8::Persistent to maintain this reference, V8 would not be able to understand the cycle and would just think that the CompiledFnEntry was supposed to live forever. We made an attempt to make that reference known to V8 by making the CompiledFnEntry weak and using a private symbol to make CompiledFnEntry strongly references the top-level referrer function in nodejs#46785, but that turned out to be unsound, because the there's no guarantee that the top-level function must be alive while import() can still be initiated from that function, since V8 could discard the top-level function and only keep inner functions alive, so relying on the top-level function to keep the CompiledFnEntry alive could result in use-after-free which caused a revert of that fix. With this patch we use a symbol in the host defined options instead of a number, because with the stage-3 symbol-as-weakmap-keys proposal we could directly use that symbol to keep the referrer alive using a WeakMap. As a bonus this also keeps the other kinds of referrers alive as long as import() can still be initiated from that Script/Module, so this also fixes the long-standing crash caused by vm.Script being GC'ed too early when its importModuleDynamically callback still needs it. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously we maintain a strong persistent reference to the ModuleWrap to retrieve the ID-to-ModuleWrap mapping from the HostImportModuleDynamicallyCallback using the number ID stored in the host-defined options. As a result the ModuleWrap would be kept alive until the Environment is shut down, which would be a leak for user code. With the new symbol-based host-defined option we can just get the ModuleWrap from the JS-land WeakMap so there's now no need to maintain this strong reference. This would at least fix the leak for vm.SyntheticModule. vm.SourceTextModule is still leaking due to the strong persistent reference to the v8::Module. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously when managing the importModuleDynamically callback of vm.compileFunction(), we use an ID number as the host defined option and maintain a per-Environment ID -> CompiledFnEntry map to retain the top-level referrer function returned by vm.compileFunction() in order to pass it back to the callback, but it would leak because with how we used v8::Persistent to maintain this reference, V8 would not be able to understand the cycle and would just think that the CompiledFnEntry was supposed to live forever. We made an attempt to make that reference known to V8 by making the CompiledFnEntry weak and using a private symbol to make CompiledFnEntry strongly references the top-level referrer function in nodejs#46785, but that turned out to be unsound, because the there's no guarantee that the top-level function must be alive while import() can still be initiated from that function, since V8 could discard the top-level function and only keep inner functions alive, so relying on the top-level function to keep the CompiledFnEntry alive could result in use-after-free which caused a revert of that fix. With this patch we use a symbol in the host defined options instead of a number, because with the stage-3 symbol-as-weakmap-keys proposal we could directly use that symbol to keep the referrer alive using a WeakMap. As a bonus this also keeps the other kinds of referrers alive as long as import() can still be initiated from that Script/Module, so this also fixes the long-standing crash caused by vm.Script being GC'ed too early when its importModuleDynamically callback still needs it. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously we maintain a strong persistent reference to the ModuleWrap to retrieve the ID-to-ModuleWrap mapping from the HostImportModuleDynamicallyCallback using the number ID stored in the host-defined options. As a result the ModuleWrap would be kept alive until the Environment is shut down, which would be a leak for user code. With the new symbol-based host-defined option we can just get the ModuleWrap from the JS-land WeakMap so there's now no need to maintain this strong reference. This would at least fix the leak for vm.SyntheticModule. vm.SourceTextModule is still leaking due to the strong persistent reference to the v8::Module. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously when managing the importModuleDynamically callback of vm.compileFunction(), we use an ID number as the host defined option and maintain a per-Environment ID -> CompiledFnEntry map to retain the top-level referrer function returned by vm.compileFunction() in order to pass it back to the callback, but it would leak because with how we used v8::Persistent to maintain this reference, V8 would not be able to understand the cycle and would just think that the CompiledFnEntry was supposed to live forever. We made an attempt to make that reference known to V8 by making the CompiledFnEntry weak and using a private symbol to make CompiledFnEntry strongly references the top-level referrer function in nodejs#46785, but that turned out to be unsound, because the there's no guarantee that the top-level function must be alive while import() can still be initiated from that function, since V8 could discard the top-level function and only keep inner functions alive, so relying on the top-level function to keep the CompiledFnEntry alive could result in use-after-free which caused a revert of that fix. With this patch we use a symbol in the host defined options instead of a number, because with the stage-3 symbol-as-weakmap-keys proposal we could directly use that symbol to keep the referrer alive using a WeakMap. As a bonus this also keeps the other kinds of referrers alive as long as import() can still be initiated from that Script/Module, so this also fixes the long-standing crash caused by vm.Script being GC'ed too early when its importModuleDynamically callback still needs it. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously we maintain a strong persistent reference to the ModuleWrap to retrieve the ID-to-ModuleWrap mapping from the HostImportModuleDynamicallyCallback using the number ID stored in the host-defined options. As a result the ModuleWrap would be kept alive until the Environment is shut down, which would be a leak for user code. With the new symbol-based host-defined option we can just get the ModuleWrap from the JS-land WeakMap so there's now no need to maintain this strong reference. This would at least fix the leak for vm.SyntheticModule. vm.SourceTextModule is still leaking due to the strong persistent reference to the v8::Module. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously when managing the importModuleDynamically callback of vm.compileFunction(), we use an ID number as the host defined option and maintain a per-Environment ID -> CompiledFnEntry map to retain the top-level referrer function returned by vm.compileFunction() in order to pass it back to the callback, but it would leak because with how we used v8::Persistent to maintain this reference, V8 would not be able to understand the cycle and would just think that the CompiledFnEntry was supposed to live forever. We made an attempt to make that reference known to V8 by making the CompiledFnEntry weak and using a private symbol to make CompiledFnEntry strongly references the top-level referrer function in nodejs#46785, but that turned out to be unsound, because the there's no guarantee that the top-level function must be alive while import() can still be initiated from that function, since V8 could discard the top-level function and only keep inner functions alive, so relying on the top-level function to keep the CompiledFnEntry alive could result in use-after-free which caused a revert of that fix. With this patch we use a symbol in the host defined options instead of a number, because with the stage-3 symbol-as-weakmap-keys proposal we could directly use that symbol to keep the referrer alive using a WeakMap. As a bonus this also keeps the other kinds of referrers alive as long as import() can still be initiated from that Script/Module, so this also fixes the long-standing crash caused by vm.Script being GC'ed too early when its importModuleDynamically callback still needs it. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously we maintain a strong persistent reference to the ModuleWrap to retrieve the ID-to-ModuleWrap mapping from the HostImportModuleDynamicallyCallback using the number ID stored in the host-defined options. As a result the ModuleWrap would be kept alive until the Environment is shut down, which would be a leak for user code. With the new symbol-based host-defined option we can just get the ModuleWrap from the JS-land WeakMap so there's now no need to maintain this strong reference. This would at least fix the leak for vm.SyntheticModule. vm.SourceTextModule is still leaking due to the strong persistent reference to the v8::Module. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. PR-URL: nodejs#48510 Refs: nodejs#44211 Refs: nodejs#42080 Refs: nodejs#47096 Refs: nodejs#43205 Refs: nodejs#38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously when managing the importModuleDynamically callback of vm.compileFunction(), we use an ID number as the host defined option and maintain a per-Environment ID -> CompiledFnEntry map to retain the top-level referrer function returned by vm.compileFunction() in order to pass it back to the callback, but it would leak because with how we used v8::Persistent to maintain this reference, V8 would not be able to understand the cycle and would just think that the CompiledFnEntry was supposed to live forever. We made an attempt to make that reference known to V8 by making the CompiledFnEntry weak and using a private symbol to make CompiledFnEntry strongly references the top-level referrer function in #46785, but that turned out to be unsound, because the there's no guarantee that the top-level function must be alive while import() can still be initiated from that function, since V8 could discard the top-level function and only keep inner functions alive, so relying on the top-level function to keep the CompiledFnEntry alive could result in use-after-free which caused a revert of that fix. With this patch we use a symbol in the host defined options instead of a number, because with the stage-3 symbol-as-weakmap-keys proposal we could directly use that symbol to keep the referrer alive using a WeakMap. As a bonus this also keeps the other kinds of referrers alive as long as import() can still be initiated from that Script/Module, so this also fixes the long-standing crash caused by vm.Script being GC'ed too early when its importModuleDynamically callback still needs it. PR-URL: #48510 Backport-PR-URL: #51004 Refs: #44211 Refs: #42080 Refs: #47096 Refs: #43205 Refs: #38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Previously we maintain a strong persistent reference to the ModuleWrap to retrieve the ID-to-ModuleWrap mapping from the HostImportModuleDynamicallyCallback using the number ID stored in the host-defined options. As a result the ModuleWrap would be kept alive until the Environment is shut down, which would be a leak for user code. With the new symbol-based host-defined option we can just get the ModuleWrap from the JS-land WeakMap so there's now no need to maintain this strong reference. This would at least fix the leak for vm.SyntheticModule. vm.SourceTextModule is still leaking due to the strong persistent reference to the v8::Module. PR-URL: #48510 Backport-PR-URL: #51004 Refs: #44211 Refs: #42080 Refs: #47096 Refs: #43205 Refs: #38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Replace the persistent handles to v8::Module and v8::UnboundScript with an internal reference that V8's GC is aware of to fix the leaks. PR-URL: #48510 Backport-PR-URL: #51004 Refs: #44211 Refs: #42080 Refs: #47096 Refs: #43205 Refs: #38695 Reviewed-By: Antoine du Hamel <duhamelantoine1995@gmail.com> Reviewed-By: Benjamin Gruenbaum <benjamingr@gmail.com> Reviewed-By: Stephen Belanger <admin@stephenbelanger.com>
Version
v18.6.0
Platform
Darwin
Subsystem
vm
What steps will reproduce the bug?
Run node with
node --experimental-vm-modules --max-heap-size=128 test.js
, and "test.js" as:How often does it reproduce? Is there a required condition?
always
What is the expected behavior?
Should not crash as OOM.
What do you see instead?
The program crashed with OOM shortly.
Additional information
Opening this issue to track the problem #44198 (comment).
The text was updated successfully, but these errors were encountered: