Planet (Former) Advogato

This is a complement to Advogato, it is an aggregation of blogs of those who used to post on Advogato, but for one reason or another moved their blog from Advogato. It is provided as a service to those who would like to read the "greater Advogato" community.

This site works only as a Planet, it aggregates the post only, to comment on a blog entry, click on the title or time to go to the blog entry on the original site, hopefully it will have a comment facility.

October 22, 2014

Thomas Thurman [marnanel]

Zophobas morio

We got some crickets in the post today, so I put them into a tank we use for feed insects, and there were some Zophobas morio worms in there still. Z. morio is a long wriggly worm when it's a larva, and this is the form in which it's used as spider food. I was surprised, because we haven't had new Z. morio in for months, and I'd assumed that if there were any leftovers they'd be dead by now. But then I noticed the large number of small brown-black beetles in the tank and realised that the worms were (at least) second generation. I don't think I'd ever realised what they looked like when they grew up before: they're small, about a centimetre across, around the size of a new halfpenny.

This entry was originally posted at http://marnanel.dreamwidth.org/315227.html. Please comment there using OpenID.

October 22, 2014 09:11 PM

October 21, 2014

Bastien Nocera [hadess]

A GNOME Kernel wishlist

GNOME has long had relationships with Linux kernel development, in that we would have some developers do our bidding, helping us solve hard problems. Features like inotify, memfd and kdbus were all originally driven by the desktop.

I've posted a wishlist of kernel features we'd like to see implemented on the GNOME Wiki, and referenced it on the kernel mailing-list.

I hope it sparks healthy discussions about alternative (and possibly existing) features, allowing us to make instant progress.

October 21, 2014 09:06 AM

October 19, 2014

Eric Anholt [anholt]

VC4 driver status update

I've just spent another week hanging out with my Broadcom and Raspberry Pi teammates, and it's unblocked a lot of my work.

Notably, I learned some unstated rules about how loading and storing from the tilebuffer work, which has significantly improved stability on the Pi (as opposed to simulation, which only asserted about following half of these rules).

I got an intro on the debug process for GPU hangs, which ultimately just looks like "run it through simpenrose (the simulator) directly. If that doesn't catch the problem, you capture a .CLIF file of all the buffers involved and feed it into RTL simulation, at which point you can confirm for yourself that yes, it's hanging, and then you hand it to somebody who understands the RTL and they tell you what the deal is." There's also the opportunity to use JTAG to look at the GPU's perspective of memory, which might be useful for some classes of problems. I've started on .CLIF generation (currently simulation-environment-only), but I've got some bugs in my generated files because I'm using packets that the .CLIF generator wasn't prepared for.

I got an overview of the cache hierarchy, which pointed out that I wasn't flushing the ARM dcache to get my writes got into system L2 (more like an L3) so that the GPU could see it. This should also improve stability, since before we were only getting lucky that the GPU would actually see our command stream.

Most importantly, I ended up fixing a mistake in my attempt at reset using the mailbox commands, and now I've got working reset. Testing cycles for GPU hangs have dropped from about 5 minutes to 2-30 seconds. Between working reset and improved stability from loads/stores, we're at the point that X is almost stable. I can now run piglit on actual hardware! (it takes hours, though)

On the X front, the modesetting driver is now merged to the X Server with glamor-based X rendering acceleration. It also happens to support DRI3 buffer passing, but not Present's pageflipping/vblank synchronization. I've submitted a patch series for DRI2 support with vblank synchronization (again, no pageflipping), which will get us more complete GLX extension support, including things like GLX_INTEL_swap_event that gnome-shell really wants.

In other news, I've been talking to a developer at Raspberry Pi who's building the KMS support. Combined with the discussions with keithp and ajax last week about compositing inside the X Server, I think we've got a pretty solid plan for what we want our display stack to look like, so that we can get GL swaps and video presentation into HVS planes, and avoid copies on our very bandwidth-limited hardware. Baby steps first, though -- he's still working on putting giant piles of clock management code into the kernel module so we can even turn on the GPU and displays on our own without using the firmware blob.

Testing status:
- 93.8% passrate on piglit on simulation
- 86.3% passrate on piglit gpu.py on Raspberry Pi

All those opcodes I mentioned in the previous post are now completed -- sadly, I didn't get people up to speed fast enough to contribute before those projects were the biggest things holding back the passrate. I've started a page at http://dri.freedesktop.org/wiki/VC4/ for documenting the setup process and status.

And now, next steps. Now that I've got GPU reset, a high priority is switching to interrupt-based render job tracking and putting an actual command queue in the kernel so we can have multiple GPU jobs queued up by userland at the same time (the VC4 sadly has no ringbuffer like other GPUs have). Then I need to clean up user <-> kernel ABI so that I can start pushing my linux code upstream, and probably work on building userspace BO caching.

October 19, 2014 08:05 PM

Rachel Chalmers [rachel]

adventure time 6: yosemite valley

So we went to see what all the fuss was about.

The first night, we stayed at the Wawona.

The absolute highlight of which was this handsome fellow vogueing in the shrubbery.

Next morning, brunch at the Ahwahnee.

Then El Capitan, or as I like to call him, Steve.

We stormed around the Merced River for a bit, which was painfully scenic.

Then I don’t even know, a meadow and some rocks and stuff.

A waterfall of excruciating beauty.

Tea back at the Ahwahnee with a mama mule deer and her twin fawns.

Pinot grigio on our balcony at the Yosemite Lodge, with our own personal mountain.

And our own personal sunset.

Glacier Point on the way home, for one last overdose on grandeur.

Buh-bye rocks and stuff!

I guess I would characterize all the fuss as “not wholly unjustified”.

October 19, 2014 06:35 PM

Thomas Thurman [marnanel]

and now, a word from 13-year-old me

A few years back, sorting through some of my old papers, I found this poem. It's dated 11th December 1988, when I was nearly fourteen.

FRIENDS

They will stand beside you
When all things are good.
And in the times when things are bad
Beside you they have stood.
They always tell the truth to you
As every good friend must
And they are reliable:
Friends you always trust.
They never will say nasty things
About the clothes you wear
They'll stand up for you against others
When you're not there.
You can always trust your friends
To hold your place in queues.
They'll always tell you "You played well",
Even if you lose.
Always keeping by your side:
Friendship never ends.
Yet, after all, we're only human:
Who has friends?

This entry was originally posted at http://marnanel.dreamwidth.org/315117.html. Please comment there using OpenID.

October 19, 2014 02:00 AM

October 18, 2014

Don Marti [dmarti]

Snapchat ads and committing to non-targeting

Recent Snapchat blog, announcing ads:

We want to see if we can deliver an experience that’s fun and informative, the way ads used to be, before they got creepy and targeted. It’s nice when all of the brilliant creative minds out there get our attention with terrific content.

That's a great idea, and ties in with what I've been saying all along about the targeted ad problem.

But I'm not optimistic. Snapchat is still running on a mobile phone, running within an environment that's either problematic or outright privacy-hostile. If Snapchat can't commit to its core feature, the idea that photos disappear after sending, how can the company credibly commit to less creepy, more valuable advertising?

It would be a huge win for Snapchat if they could pull it off. But I doubt that a single app can do it.

Signalful ads are an emergent benefit from media that tend to build user confidence through tracking resistance. Non-creepiness can't be declared, it has to be discovered.

October 18, 2014 12:11 PM

October 17, 2014

Dalibor Topic [robilad]

Next Gig: EclipseCon Europe

I'll be back again at EclipseCon Europe in Ludwigsburg, Germany on October 28th to speak Tuesday afternoon at 3 PM, and enjoy Simon Ritter's Lambda tutorial on Tuesday morning at 9 AM.

As usual, if there is something around OpenJDK you'd like to chat about, let me know on Twitter (or send me an e-mail), and we'll find some space and time.

See you there!

October 17, 2014 06:18 PM

OpenJDK In The News: JavaOne 2014 Edition

This year's JavaOne featured a new press release from Oracle mentioning OpenJDK. Here are the relevant excerpts:

Since its launch, Java SE 8 has seen record adoption, and the OpenJDK Community continues to thrive with contributions from Oracle as well as from other companies, researchers, and individual developers.

Oracle has begun work on the JDK 9 Project in the OpenJDK Community. New features will focus on modularity, performance, stability, and portability.

The OpenJDK Community continues to host the development of both the Java SE Platform Specification and its Reference Implementation, the JDK, fueled by contributions from Oracle and new and existing organizations and individuals. New participants in OpenJDK include:
* The FreeBSD Foundation, which will be collaborating with Oracle and others on producing OpenJDK binaries targeted to FreeBSD users via the FreeBSD Ports Collection
* The Digital Energy Business of the General Electric Company, which recently joined the OpenJDK Community and will be collaborating with Oracle and others on improving stability and performance of multilanguage virtual machine support via the invokeDynamic (JSR 292) API
* Microsoft Open Technologies, Inc., which will be contributing its expertise in networking, performance, and integration to the community.
The OpenJDK Community hosts many new projects that will continue to move the Java SE platform forward. These include Valhalla, which is exploring advanced language and JVM features; Panama, which aims to improve the connections between Java and native code; and smaller efforts such as Annotations Pipeline 2.0, Javadoc.next, Device I/O, and Memory Model Update.


Taken from the Oracle press release titled Oracle Highlights Continued Java SE Momentum and Innovation at JavaOne 2014.

October 17, 2014 06:06 PM

October 16, 2014

Thomas Thurman [marnanel]

Why, why, why, Eliza?

Tell me some more about when you saw light on my window.
Earlier on you were lost like a slave I can't free.
I understand you.
Is it because I deceived you that you came to me?
My, my, my, Eliza!
Why, why, why, Eliza?
I can see you're just a conditional tree
But you remind me we came here to talk about me.

This entry was originally posted at http://marnanel.dreamwidth.org/314265.html. Please comment there using OpenID.

October 16, 2014 11:54 PM