In 2007 I started working for OpenedHand. They became acquired by Intel in 2008. Today I am working for Red Hat (for over 5 years now). And we have 2018 and it became acquired by IBM.
I came back home in the evening with plans for some cider and episode of some TV series (probably “Ozark”). But when I landed on a couch and took a look on my phone it shown set of notifications. Telegram, Facebook, Messenger…
And all of them were about one thing: Red Hat being acquired by IBM. First I looked and sources were Bloomberg and CNBC. At that phase I thought “ok, it can be a rumour” so my answer was “can not comment”.
During last few years Linaro Enterprise Group (recently renamed to Linaro Datacenter and Cloud Group) was working on getting OpenStack working on AArch64 at same level as it works on x86-64 architecture. And I am proud to be member of that group ;D
We started our adventure with Liberty, migrated to Mitaka and then Newton. And we stayed there for a while with Developer Cloud to make sure that all those projects which rely on it can use VM instances for their work.
Took us Pike and Queens to get to the point when we could create new setup of Developer Cloud. In clean way, using containers generated by one of OpenStack projects. No more in-house solutions.
Our team always consisted people from several countries and companies because this is how Linaro works — there are Linaro employees, there are assigned engineers from member companies etc. We cooperated with our kernel people, packagers, developers from several open source projects (libvirt, RDO, CentOS, Debian) and more.
Some people were running tests, some were doing image builds, package builds. Others were managing to keep us focus and to get it delivered as we planned to.
We attended several OpenStack related events (PTG, Summit etc) to tell people how AArch64 support looks like in all those projects. Gave several talks about how OpenStack works for us.
In 2004 I was newbie in embedded Linux area. Decision to buy Sharp Zaurus instead of HP iPaq got me to Qt/e world rather to GTK one. I was also KDE user rather than GNOME2 as well so I can say that I liked Qt already.
All those sizes in pixels, paddings and margins I saw in GTK code made me feel sick each time I had to edit UI of some application. No idea why developers went that way…
In Qt world all you had to do was launching Qt Designer, put some UI elements into window, apply some Layout elements and build your app. No need to deal with padding/margin settings etc cause library that for you.
In meantime Qt developers added QML as a new way to do UI for Qt applications. I ignored it’s existence until now…
Few days ago Michał Schulz did nice work on improving my Modland player. He also moved it’s UI from old Qt Designer one to QML.
For now UI is hardcoded to 800×480. I have tried to make it scalable but have a feeling that QML is against me.
Look at Authors/Modules part. It is simple layout, right?
In Qt Designer UI I would select those four elements, put into GridLayout and it would scale properly. So I tried that for QML. Labels survived, listViews got 0x0 size.
And the only ‘design tool’ to edit QML is Qt Creator. Which gets fugly unstable once you try to play with QML designs…
So I looked at files describing UI. And you know what I found there? Old GTK nightmares… Positioning elements with pixels, sizes in pixels. Pixels! Not some magical “dp units”. There is no way to say “make this element 10em tall” like you can with CSS.
And it is not only with Modland player UI. Same it with QML examples…
WTH happened with Qt developers? Or was “QML is only for embedded devices, do not use on desktop” phrase got removed from documentation by mistake?
During Linaro Connect there was a possibility to play with ThunderX2 workstation. I remember that Arnd Bergmann was comparing speed of kernel compilation with his AMD Threadripper workstation.
Test was simple — checkout 4.18 source, use arm64 defconfig and do build of ‘Image modules’ with as many threads as you have cpu cores. He did several builds with limiting to one cpu, to disable cpu threads etc but idea stays the same.
Dual socket ThunderX2 (28 cpu cores, 4 threads per core iirc) did that in about 2 minutes. So did Arnd’s Threadripper machine.
So I decided to check that on my local hardware. Mustang needed 38 minutes, my i7-2600K based desktop did that in 9 minutes 20 seconds.
For comparison: I was told that Synquacer with it’s 24 Cortex-A53 cores does that in about 16 minutes.
Is it fast? Do not think so. But who would assume that retro hardware will be fast…
You probably know that I am fan of retro computers. Those from 80s, 70s and older ones. And for quite a time I told that I do not plan to run retro machines at home. But it has to change.
Due to some work things I am going to run Mustang again. But where is retro in it someone may ask…
Applied Micro Mustang uses X-Gene cpu. And this was first (or one of firsts) AArch64 CPU. I got mine over four years ago. It is obsolete in some areas (SBSA level 0 anyone?) but still works. And is hard to replace if you do not have spare few thousands USD 🙁
Someone may say that I can buy Synquacer. Sure. 1160$ for mainboard in some box. With rotating plates which would go away on first day, not needed graphics card and just 4GB of memory. Good luck with finding ram sticks which will work. I heard rumours that there is a store somewhere which keeps a pile of those. And then you end with 24 slow cores which may be good at kernel compilation but then suck at linking.
So now I am on a hunt for 2x16GB DDR3 ECC RDIMM sticks for Mustang. And some SSD as using rotating plates for development does not have sense in 2018.
Maybe one day someone will finally realise that 500 USD is this magical point where hardware can be bought in “just go and buy” fashion. So we, developers, will be able to write to our managers “Hey, there is this arm64 mainboard for 499$” and hear “just go, buy and expense”. Memory, storage, case can be other expense raport (or even collected from spare parts at home).
But until then I will have to live with my retro server.
Few days ago Linus Torvalds added code of conduct to the Linux kernel. And then lot of discussions started.
I had no plans to take part in any of them. But last week I was dragged into one of them and it was not fun. Turned out that people I know and trust when it comes to technical discussions (never met most of them) do not quite understand the need for such.
There are many “code of conduct” documents. Often they differ a lot. I have my own and it is probably the shortest one:
Do not be an asshole. Respect the others.
Simple. I do not care which gender people have when I speak with them (ok, may stare at your boobs or butt once) nor their sexual preferences. Colour of the skin does not matter as most of my friends I first met online without knowing anything about them. Political stuff? As long as we can be friends and do not discuss it I am fine. Etc etc.
It works on conferences. And in projects where I am/was involved.
Someone may say that part of it was shaped by working for corporation (is Red Hat corpo?) due to all those no harassment regulations and trainings. I prefer to think that it is more of how I was raised by parents, family and society.