Thread (52 messages) 52 messages, 11 authors, 2019-12-08

Re: [RFC PATCH 04/10] pipe: Use head and tail pointers for the ring, not cursor and length [ver #2]

From: Johannes Hirte <hidden>
Date: 2019-12-07 00:00:45
Also in: keyrings, linux-api, linux-block, linux-fsdevel, linux-usb, lkml

On 2019 Dez 06, Linus Torvalds wrote:
On Fri, Dec 6, 2019 at 1:47 PM Johannes Hirte
[off-list ref] wrote:
quoted
This change breaks firefox on my system. I've noticed that some pages
doesn't load correctly anymore (e.g. facebook, spiegel.de). The pages
start loading and than stop. Looks like firefox is waiting for some
dynamic loading content. I've bisected to this commit, but can't revert
because of conflicts.
Can you check the current git tree, and see if we've fixed it for you.
There are several fixes there, one being the (currently) topmost
commit 76f6777c9cc0 ("pipe: Fix iteration end check in
fuse_dev_splice_write()").

I _just_ pushed out that one, so check that you get it - it sometimes
takes a couple of minutes for the public-facing git servers to mirror
out. I doubt that's the one that would fix firefox, but still..

               Linus
Tested with 5.4.0-11505-g347f56fb3890 and still the same wrong behavior.
Reliable testcase is facebook, where timeline isn't updated with firefox.

-- 
Regards,
  Johannes Hirte
Keyboard shortcuts
hback out one level
jnext message in thread
kprevious message in thread
ldrill in
Escclose help / fold thread tree
?toggle this help