-
Notifications
You must be signed in to change notification settings - Fork 29.8k
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
Forward port from v1.x #1559
Forward port from v1.x #1559
Conversation
If I could get some quick review here that'd be great. It'd be good to get these into 2.0.0 so that we don't have divergence. |
Rubber-stamp LGTM. Maybe run the CI just in case. |
It would be nice to have some sort of metadata to point to PRs for porting patches. Hmm. |
@Fishrock123 So, in this case I believe the appropriate set of operations would be:
|
Buffer#copy() immediately does a ToObject() on the first argument before it checks if it's even an Object. This causes Object::HasIndexedPropertiesInExternalArrayData() to be run on nothing, triggering the segfault. Instead run HasInstance() on the args Value. Which will check if it's actually an Object, before checking if it contains data. Fixes: nodejs#1519 PR-URL: nodejs#1520 PORT-PR-URL: nodejs#1559 Reviewed-by: Evan Lucas <[email protected]>
PR-URL: nodejs#1517 PORT-PR-URL: nodejs#1559 Reviewed-By: Brian White <[email protected]>
074ed63
to
9f45e5e
Compare
@chrisdickinson Some things just got back-ported into v1.x though, won't that cause conflicts now? :/ Or do I just merge with the "ours" rule? |
Ideally this would have gone in first so that the backport commits wouldn't get duplicated. |
I'm going to take a look and see if git chokes on this – I suspect it won't, but it's surprised me before :) |
@chrisdickinson any update? Either way the commits should really make it into 2.0.0 |
|
@chrisdickinson would it work if we did a merge from a branch or something from a point where only these two commits existed in v1.x? Or is a cherry-pick still cleaner? If cherry-pick is cleaner, is this meta-data reasonable? |
The metadata looks reasonable. Try a merge from $(git merge-base master v1.x)+onlythose2changes, but if it doesn't work out, or otherwise produces a complicated history – see |
@chrisdickinson the result of that is this: https://github.com/Fishrock123/io.js/commits/master-with-v1.x-merge I'd personally prefer to cherry-pick, but that may not be the best option. |
The commit history from
|
Looking over it, the merge commit should work. Thanks for catching these commits! |
replaced by #1582 |
Related PRs: #1520, #1517
I was under the impression we were going to be landing everything in
master
and then backport any fixes, unless something doesn't apply to master./cc @iojs/tc, esp @trevnorris & @indutny