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

src: fix async hooks crashing when there is no node context #19134

Closed
wants to merge 42 commits into from

Conversation

xaviergonz
Copy link
Contributor

@xaviergonz xaviergonz commented Mar 4, 2018

This commits basically skips any promise env hooks when there's no node context currently active, which would otherwise end up in a crash.

Fixes: #19104

Checklist
  • make -j4 test (UNIX), or vcbuild test (Windows) passes
  • tests and/or benchmarks are included
  • documentation is changed or added
  • commit message follows commit guidelines
Affected core subsystem(s)

src (async_hooks)

@nodejs-github-bot nodejs-github-bot added the c++ Issues and PRs that require attention from people who are familiar with C++. label Mar 4, 2018
Copy link
Member

@addaleax addaleax left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM modulo style nit(s), thanks!

src/env.cc Outdated
Environment* env = Environment::GetCurrent(promise->CreationContext());
auto context = promise->CreationContext();
// if the context is undefined (not a node context) then skip
if (context->GetEmbedderData(node::Environment::kContextEmbedderDataIndex)->IsUndefined()) {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I think make lint should give you an error, because this line is longer than 80 characters?

Also, maybe look at the C++ style guide, e.g. comments should ideally be capitalized and punctuated. :)

Copy link
Contributor Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

both done :)

src/env.cc Outdated
auto context = promise->CreationContext();
auto dataIndex = node::Environment::kContextEmbedderDataIndex;
// If the context is undefined (not a node context) then skip.
if (context->GetEmbedderData(dataIndex)->IsUndefined()) {
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@bnoordhuis is this still undefined behavior?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Hm. I’m reading up on the V8 code, and it seems like a real issue here is that this might actually perform an out-of-bounds read if the context has less than dataIndex fields…

I assume the check is left out more or less intentionally on the V8 side, so I guess the options here are:

  • Add a method to v8::Context that tells us how many embedder data fields there are
  • Let Node attach some kind of property (e.g. a private symbol) on a context-specific object, like the global object or the ExtrasBindingObject, then use that to tell whether an Environment has been assigned to a context or not … this might be a bit slower but wouldn’t require changes to V8

/cc @nodejs/v8

Copy link
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

/cc @hashseed many a years ago I ran into similar trouble with the debug context and had similar thoughts to @addaleax. WDYT?

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

how about HasEmbedderData(index)

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@xaviergonz Do you have any thoughts on this? In the end, it’s your PR, so you obviously have some say in what happens here.

In particular, wrt the second option, that should be doable without waiting for anything else to happen, and it would be nice if you could indicate whether you’d want to try it yourself or prefer for somebody else to take a stab.

(If you do: We’d be really, really glad if we can help in any way. Don’t be shy to ask questions, here or on IRC!)

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I'm somewhat certain that Isolate::GetCurrentContext should give you the same context, since we run the microtask in the context of the Promise. If Object::CreationContext and Isolate::GetCurrentContext returns different contexts, you got a problem anyways, since Environment::GetCurrent uses the latter.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

@hashseed How do we get the isolate without Isolate::GetCurrent() or promise->GetIsolate()? Both seem worse than what we have now…

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

promise->GetIsolate get's it by masking off the object address to get to the heap page, where the isolate address is stored. That's not too bad.
Isolate::GetCurrent reads it from TLS, which can be slow.

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

promise->GetIsolate get's it by masking off the object address to get to the heap page, where the isolate address is stored. That's not too bad.

V8_DEPRECATE_SOON("Keep track of isolate correctly", Isolate* GetIsolate());

😢

Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Yeah about that...

@xaviergonz
Copy link
Contributor Author

Just updated with all the comments

@xaviergonz
Copy link
Contributor Author

Btw, the magic number is the hex number for ascii "node" :)

@hashseed
Copy link
Member

hashseed commented Mar 7, 2018

So, I discussed this with @bmeurer a bit.

  • Scratch my comment about not using CreationContext. In this case you actually need to use it since the Isolate::GetCurrentContext gives you the context in which you resolved the promise, which may differ from the context where the promise was created.
  • This approach probably going to kill promise hook performance.
  • How about we get v8::Context::GetNumberOfDataSlots landed upstream and then float the patch on node, then use it to fix this issue?

@xaviergonz
Copy link
Contributor Author

xaviergonz commented Mar 7, 2018

Still, even if that api is available, wouldn't we need two embedded data indexes to make sure that it is a node context?

Something like:
if

  • we have >= 32 embedded data indexes
  • and number 31 is set to a magic number
  • then we are pretty sure 32 is a pointer to a node env class instance

vs
if

  • we have >= 32 embedded data indexes
  • then we are sure 32 is a pointer to a node env class instance (are we? it could be anything else as long as the embedder used it for something else)

@xaviergonz
Copy link
Contributor Author

xaviergonz commented Mar 7, 2018

Also I was wondering, (about the other approach that does not rely on v8 api changing) is it ok to set the embedder index with the magic number inside the promise template itself? that way you wouldn't need to access isolate/global.process at all I think, just:

if promise->getinternalfield(32 or whatever) is the magic number, we are on a node context

@hashseed
Copy link
Member

hashseed commented Mar 7, 2018

Still, even if that api is available, wouldn't we need two embedded data indexes to make sure that it is a node context?

To be extra sure, with a clean solution, yes.

But we know that Chromium contexts only uses two slots. And assuming this does not change dramatically, we can just check the number of slots against 33 for Node.js.

@xaviergonz
Copy link
Contributor Author

For now I have just switched back to promise->GetCreationContext()

@hashseed @addaleax what should I do with the pull request then? wait for v8 to change? :)

@xaviergonz
Copy link
Contributor Author

Another thing I thought about is having a cache, a map<context address, boolean> that for each EnvPromiseHook does:

if (map does not have an entry for the context address then) {
  bool onNodeContext = process && process first internal field is magic number;
  map[promise creation context address] = onNodeContext;
}
if (!map[promise creation context address]) { return; }
else we are on a node context

pros: only need to make the expensive operation once per context, no v8 changes needed
cons: you waste 1 byte (maybe 4?) of memory per context creation

@hashseed
Copy link
Member

hashseed commented Mar 8, 2018

Contexts live on V8's heap and can change their addresses. Also, where would the map live? As a static global value?

@xaviergonz
Copy link
Contributor Author

xaviergonz commented Mar 8, 2018

Yes, a static global value, so you'd be leaking 1-4 byte per context that creates a promise.
Is there a way to uniquely identify a context then?

Though even when it changes the address the only thing that'd happen is that there would be a cache miss and it would need to check for the magic number again (I don't know how often a context changes its address).

@hashseed
Copy link
Member

hashseed commented Mar 8, 2018

I'd just wait for the change in V8 and float it to use it in this fix.

I'm actually wondering whether this should be fixed in Electron instead. That's the only case where embedder data in contexts are used in two different ways.

@xaviergonz
Copy link
Contributor Author

xaviergonz commented Mar 8, 2018

Would need to be changed in chromium then, since it is basically using the chromium generated context in those cases.

@hashseed
Copy link
Member

hashseed commented Mar 8, 2018

Neither Chromium nor Node.js run into this issue on their own because embedder data are not designed to be used for two different embedders in the same isolate. I'd expect Electron to float a fix because it uses Node.js and Chromium contexts side by side.

@xaviergonz
Copy link
Contributor Author

Is it possible for electron to change the embedder data of a chromium context?

@xaviergonz
Copy link
Contributor Author

There would be a case where it would fail though now that I think about it, if two different contexts (one node and another chromium) eventually swapped places in memory and ended up being in the same address, but I don't know if that's possible.

@xaviergonz
Copy link
Contributor Author

xaviergonz commented Mar 8, 2018

Ok, better idea :D

on env start:

  // Used by EnvPromiseHook to know that we are on a node context.
  isolate()->GetCurrentContext()->SetAlignedPointerInEmbedderData(kContextEmbedderDataIndex + 1, (void *)kNodeContextTag);

then on env promise hook

void Environment::EnvPromiseHook(v8::PromiseHookType type,
                                 v8::Local<v8::Promise> promise,
                                 v8::Local<v8::Value> parent) {
  Local<v8::Context> context = promise->CreationContext();

  // Grow the embedder data if necessary to make sure we are not out of bounds
  // when reading the magic number.
  context->SetAlignedPointerInEmbedderData(kContextEmbedderDataIndex + 2, nullptr);
  int magicNumber = (int)context->GetAlignedPointerFromEmbedderData(kContextEmbedderDataIndex + 1);
  if (magicNumber != kNodeContextTag) {
    return;
  }

  Environment* env = Environment::GetCurrent(context);
  for (const PromiseHookCallback& hook : env->promise_hooks_) {
    hook.cb_(type, promise, parent, hook.arg_);
  }
}

pros: no changes to v8, fast
cons: storing a number instead of a pointer looks weird I guess? using embedder fields 32, 33 and 34 instead of just 32

@xaviergonz
Copy link
Contributor Author

Updated to pull request to the latest idea. Confirmed to be working with electron.

@xaviergonz
Copy link
Contributor Author

@addaleax any thoughts?

@xaviergonz
Copy link
Contributor Author

didn't work, same unit test failing, so reverting

@xaviergonz
Copy link
Contributor Author

xaviergonz commented Jun 27, 2018

Ok, I found the offending test:

{
  const d = domain.create();

  d.run(common.mustCall(() => {
    vm.runInNewContext(`
      const promise = Promise.resolve();
      assert.strictEqual(promise.domain, undefined);
      promise.then(common.mustCall(() => {
        assert.strictEqual(process.domain, d); // process.domain is actually undefined
      }));
    `, { common, assert, process, d });
  }));
}

apparently
assert.strictEqual(process.domain, d);

process.domain becomes undefined, since the context is actually not a node context (due to vm.runInNewContext)

now, what I don't know is if this is something correct (vm.runInNewContext creates a non node context) or it is wrong and because vm.runInNewContext should actually create an environment.

@addaleax @apapirovski Any thoughts?

@xaviergonz
Copy link
Contributor Author

Found the issue, moving the setAlignedPointerInEmbedderData to AssignToContext fixed it :)

Copy link
Member

@addaleax addaleax left a comment

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

Sorry for being so absent – thanks, this definitely still looks good!

Let’s see that we can get this landed, and thanks for your patience.

CI: https://ci.nodejs.org/job/node-test-pull-request/15849/

@addaleax addaleax added the author ready PRs that have at least one approval, no pending requests for changes, and a CI started. label Jul 13, 2018
@addaleax
Copy link
Member

Looks like the number of commits in here is making trouble for the CI …

CI without rebase: https://ci.nodejs.org/job/node-test-pull-request/15851/

@addaleax
Copy link
Member

addaleax commented Jul 15, 2018

Thanks for all the hard work!

Landed in fb87d8a 🎉

@addaleax addaleax closed this Jul 15, 2018
addaleax pushed a commit that referenced this pull request Jul 15, 2018
PR-URL: #19134
Fixes: #19104
Reviewed-By: Anna Henningsen <[email protected]>
Reviewed-By: James M Snell <[email protected]>
Reviewed-By: Tiancheng "Timothy" Gu <[email protected]>
targos pushed a commit that referenced this pull request Jul 16, 2018
PR-URL: #19134
Fixes: #19104
Reviewed-By: Anna Henningsen <[email protected]>
Reviewed-By: James M Snell <[email protected]>
Reviewed-By: Tiancheng "Timothy" Gu <[email protected]>
@xaviergonz
Copy link
Contributor Author

just wondering, in which version of node will it land? a pull request for electron depends on this one, that's why I ask 😁

@addaleax
Copy link
Member

@xaviergonz It’s on the v10.x-staging branch, so it will most likely be in the next v10.x release (that is, most likely this or next week).

Since you’re particularly interested in having this in Electron, /cc @codebytere – I think Electron might be able to just cherry-pick this commit?

@targos targos mentioned this pull request Jul 17, 2018
@codebytere
Copy link
Member

@xaviergonz yes, we should be able to cherry-pick this back into our next node upgrade

@xaviergonz
Copy link
Contributor Author

@codebytere nice! the relevant pull request for electron is electron/electron#12109 (comment)

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
c++ Issues and PRs that require attention from people who are familiar with C++.
Projects
None yet
Development

Successfully merging this pull request may close these issues.

new internalNextTick cannot be wrapped by electron