[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Gnash-dev] Gnash 0.8.9 status

From: John Gilmore
Subject: Re: [Gnash-dev] Gnash 0.8.9 status
Date: Wed, 02 Mar 2011 16:56:57 -0800

> In the last two weeks we've closed 29 bugs (at time of writing), 14 of
> them serious enough to be considered release blockers: that is bugs that
> cause an abort, a segfault, or an illegal memory access.

That's really great!

> All fixed bugs:

I looked in that set of bugs for the "Gnash consumes 100% of CPU
uselessly" bug but didn't see it.  That's the one I tried to fix
before the August release, but my patch had other bad effects.  You
can see the bug in Youtube after you play a clip, if you have a CPU
meter on your screen somewhere.  It starts burning 100% of one CPU
core while Youtube idly shows possible future video titles to click on.

This may be:

  "CPU busied in low-bandwidth situation"

but we have six open bugs with "CPU" in their summary, all indicating
very high CPU usage:

This one, for example, has 50 votes on Savannah after Fedora commenters
urged others to bump its priority upstream:

  "gnash in firefox high CPU"

I've been meaning to take another shot at fixing this, but other work 
intervened.  Does anyone else want to look at fixing this one?

My patch, which fixed this problem but made Youtube video not start
playing immediately, is in this thread; perhaps it will point someone
more experienced at the real fix:

When considering the importance of this bug, I note that Gnash's
Wikipedia page mentions "Gnash still suffers from high CPU usage.",
citing this review from O'Reilly:

Indeed, I still regularly find gtk-gnash processes lying around
burning up my whole CPU, from when I leave Firefox tabs lying around
that happen to include flash ads or navigation.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]