diff options
| author | Chris Kuethe <chris.kuethe@gmail.com> | 2006-12-12 18:39:08 +0000 |
|---|---|---|
| committer | Chris Kuethe <chris.kuethe@gmail.com> | 2006-12-12 18:39:08 +0000 |
| commit | 769e6567fba20de0a662f0045b008c7f38bb3d33 (patch) | |
| tree | 8f15a811f9afcece0070721d58c5c39bbbd96120 /www | |
| parent | e80a3cd4ebd4ef4e3ee6c425b07158a645c6f460 (diff) | |
| download | gpsd-769e6567fba20de0a662f0045b008c7f38bb3d33.tar.gz | |
Select() on OS X really is broken
Diffstat (limited to 'www')
| -rw-r--r-- | www/upstream-bugs.html | 11 |
1 files changed, 6 insertions, 5 deletions
diff --git a/www/upstream-bugs.html b/www/upstream-bugs.html index 1a5fe4d5..a05488c2 100644 --- a/www/upstream-bugs.html +++ b/www/upstream-bugs.html @@ -78,12 +78,13 @@ USB devices.</p> sockets), select(2) can return 0 and shows no fd with input waiting even when data is in fact ready to be read from some of the devices.</p> -<p>This happens often enough to make gpsd unreliable. Unfortunately, -it's an unusual enough use case of select(2) that we may be the only -victims.</p> +<p>This happens reliably and makes gpsd unusable. Apple has known about +this for some time now: <a +href='http://lists.apple.com/archives/darwin-development/2001/Oct/msg00716.html'> +http://lists.apple.com/archives/darwin-development/2001/Oct/msg00716.html</a> +</a></p> -<p>Observed by Chris Kuethe under Darwin kernel 8.8.1 on an x86 -Mac mini.</p> +<p>Observed by Chris Kuethe under Darwin kernel 8.8.1 on an x86 Mac mini.</p> <h2>Python socket library barfs on IPV6 notation in the /etc/hosts file</h2> |