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.

Sign up for GitHub

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

buffer: don't set kNoZeroFill flag in allocUnsafe #6007

Closed
wants to merge 1 commit into from
Closed

buffer: don't set kNoZeroFill flag in allocUnsafe #6007

Show file tree
Hide file tree
Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
2 changes: 0 additions & 2 deletions lib/buffer.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -163,8 +163,6 @@ Buffer.alloc = function(size, fill, encoding) {
Buffer.allocUnsafe = function(size) {
if (typeof size !== 'number')
throw new TypeError('"size" argument must be a number');
if (size > 0)
flags[kNoZeroFill] = 1;
return allocate(size);
};

Expand Down
12 changes: 11 additions & 1 deletion test/parallel/test-buffer-safe-unsafe.js
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -7,8 +7,18 @@ const safe = Buffer.alloc(10);

function isZeroFilled(buf) {
for (let n = 0; n < buf.length; n++)
if (buf[n] > 0) return false;
if (buf[n] !== 0) return false;
Copy link
Member

Choose a reason for hiding this comment

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

ha! sigh... I had uint8 on the brain I guess. good catch.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Also -0 could be a factor, but I'm not sure. Any way, tests are a bit non-deterministic, since even if there is no actual zero filling, there is still a chance that some allocations would be zeros

Copy link
Member

Choose a reason for hiding this comment

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

Yeah, that came up in the original review. It's not a great test. The plan was to revisit to see if the test can be made more robust.

Copy link
Contributor Author

Choose a reason for hiding this comment

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

Also for some reason this bug is hard to reproduce with small Uint8Array.

Copy link
Member

Choose a reason for hiding this comment

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

Looking at it now this one should have been obvious :-/. The bug only shows up after doing a pooled unsafeAlloc allocation because the zero fill flag was never being reset (because a real allocation wasn't being done).

(update: ha! I see you noted that in your commit log... it's definitely a friday)

return true;
}

assert(isZeroFilled(safe));

// Test that unsafe allocations doesn't affect subsequent safe allocations
Buffer.allocUnsafe(10);
assert(isZeroFilled(new Float64Array(10)));

new Buffer(10);
assert(isZeroFilled(new Float64Array(10)));

Buffer.allocUnsafe(10);
assert(isZeroFilled(Buffer.alloc(10)));
Copy link
Contributor

Choose a reason for hiding this comment

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

style: missing newline

Copy link
Contributor Author

Choose a reason for hiding this comment

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

fixed!