<html>
<head>
<base href="https://bugs.webkit.org/" />
</head>
<body>
<p>
<div>
<b><a class="bz_bug_link
bz_status_NEW "
title="NEW - Errors in read() are not handled in WTF::cryptographicallyRandomValuesFromOS."
href="https://bugs.webkit.org/show_bug.cgi?id=146473#c5">Comment # 5</a>
on <a class="bz_bug_link
bz_status_NEW "
title="NEW - Errors in read() are not handled in WTF::cryptographicallyRandomValuesFromOS."
href="https://bugs.webkit.org/show_bug.cgi?id=146473">bug 146473</a>
from <span class="vcard"><a class="email" href="mailto:msaboff@apple.com" title="Michael Saboff <msaboff@apple.com>"> <span class="fn">Michael Saboff</span></a>
</span></b>
<pre>Comment on <span class=""><a href="attachment.cgi?id=255860&action=diff" name="attach_255860" title="Patch">attachment 255860</a> <a href="attachment.cgi?id=255860&action=edit" title="Patch">[details]</a></span>
Patch
View in context: <a href="https://bugs.webkit.org/attachment.cgi?id=255860&action=review">https://bugs.webkit.org/attachment.cgi?id=255860&action=review</a>
<span class="quote">> Source/WTF/wtf/OSRandomSource.cpp:63
> + int fd;
> + // We need to check for both EAGAIN and EINTR since on some systems /dev/urandom
> + // is blocking and on others it is non-blocking.
> + do {
> + fd = open("/dev/urandom", O_RDONLY, 0);
> + } while (fd < 0 && (errno == EAGAIN || errno == EINTR));</span >
I don't think there is a need to check for the EAGAIN and EINTR cases for open(). EAGAIN is documented to only happen for slave ptys and EINTR due to a signal.</pre>
</div>
</p>
<hr>
<span>You are receiving this mail because:</span>
<ul>
<li>You are the assignee for the bug.</li>
</ul>
</body>
</html>