Posts by Roo

Fujitsu: Why we chose 64-bit ARM over SPARC for our exascale super

"If an HPC application spends much time waiting on IO then someone needs to call in a real HPC expert to give the setup a once-over, because that's a total waste of time (as you rightly point out)."

Define "much" ! :)

How does a "real HPC expert" magic up no mem-waits on a 16 core Xeon running sparse matrix code with a 16 way set-associative shared L3 cache ? The killer micros have taken over, they are a lot faster than the beasts that came before them - but equally it's also much harder to extract peak performance from them with apps that feature large memory footprints. I'm not having a dig, just pointing out that some problems are inherently awkward. :)

"Supercomputer applications are designed to scale across thousand of cores - so unlike PCs those cores are not idle!"

They still wait on I/O like any other CPU, the speed of light still has an impact on how code is written and networks are built. ;)

Re: "ARM's larger and healthier software ecosystem?"

"People are compiling Fortran (and to a lesser extent C) to run on supercomputers, they're not using assembly. I guess he must have been talking about support from the Linux kernel community?"

Not just kernels, compilers, profilers and debuggers too. The CPUs shipped argument is pretty one-sided - and it is very unlikely to get better for SPARC because the players with mindshare (ie: Oracle) view SPARC as a cash cow, and they have a track record of actively fighting and sabotaging open source. None of that makes SPARC inherently bad but it does make SPARC harder to use.

Your wget is broken and should DIE, dev tells Microsoft

Re: Reminds me of a very very old joke from the 90's...

"My favorite "Dark" by Microsoft: letting the type "long" be 32 bit on 64 bit systems.

Yes, the C-standard allows it. It's still insane for a general purpose computer."

I can understand the antipathy - the 32bit long/64bit pointer model was actually employed on 64bit RISC boxes before MS had got around to using 32bits properly. The aim was to reduce the memory footprint of apps - and thereby get less cache misses and increase performance. Believe it or not it did actually work in some cases. Personally I found the existence of long long more irritating, and refused to play the game by using things like int64_t instead. :)

"1) How often did the shell move from /usr/bin/bash and /bin/bash lately ?"

I'll grant you that one, and even if the shell does move it's not a show stopper - easily fixed/worked around/bodged etc.

"2) Code "conservatively" is actually an alias "not using features that make shell scripting less horribad"

I think you're being a bit hard on folks here. I use ksh & bash on a daily basis, so I tend to restrict myself to using features common to both simply because I c.b.a with writing a script twice. Besides if I need the stuff bash brings to the table (as handy as they may be) the chances are I should be working in Python instead. :)

"but when scripting shouldn't one test and declare exactly which executable you want to be running as opposed to relying on a user shell environment to be set up correctly?"

In most cases I would say "no" because the users may well have their shell env setup with the intention of using non-standard executables (eg: if they are cross compiling) and that kind of environment testing code renders scripts pretty much unreadable. If you really want that kind of thing I think it should be put into a dedicated environment setup+validation script.

"This seems like just an anti-Microsoft gripe from the Linux fundamentalists."

We see the same complaints from MS Office lovers every time folks suggest LibreOffice can be used in place of MS Office. Plus in this case MS are intentionally ripping off brand names with the intention of fooling people into thinking they are using the real deal. I'm pretty sure the MS community at large wouldn't react any better to LibreOffice renaming their products Excel, Word, Access and Powerpoint.

Windows Update borks PowerShell – Microsoft won't fix it for a week

Re: Embarrassing

"If you were asked to deploy a Linux desktop across your enterprise, would you run for the hills? I would."

Linux desktops have already happened in some few big corps by stealth in the form of Linux powered thin clients replacing desktops connecting to massive Linux servers hosting Windows on VMs.

Beardy Ed Vaizey: 'I can't let go. I like the tech sector'

"Of course there is. It's working via one's own limited company and being treated by HMRC as a real business."

No problem with that, however some interpretations* of IR35 require you to buy tools from your own pocket rather than the company account, adding a ~40% premium to the cost of doing business...

* = Depends on who answers the call at HMRC + phase of the moon.

Windows 10 Anniversary Update completely borks USB webcams. Yay.

"Well, having the OS do the decoding of the video stream on behalf of the multiple applications likely using it to me sounds like a good idea…"

I sincerely hope MS isn't doing the decoding in the kernel. They are still shipping fixes for kernel rendering code vulns they introduced with NT 4.0 (20 years ago). :(

New UK trade deals would not compensate for loss of single market membership

Re: The big picture

"As I wrote, our emphasis has been and should be on the RoW regardless of Brexit."

Fair point, I think most people can also agree that it would have been better for the trade figures to be skewed towards the RoW before lighting the fire under the pan. :)

Re: Really ....

"FTSE 100 up

FTSE 250 up"

My guess is that folks are moving money into shares because the pound is taking a beating on the currency markets, the prospect of negative interest rates will tend to do that.

Re: Really ....

"You lost. It's happening."

Everyone lost, including folks who got the result they wanted. Savings and assets are all worth a lot less, rent will go up to compensate, tax receipts have already gone down so all that "extra" money will be used to fill the widening hole in the balance sheet. The only folks "getting over it" are leaving the country and taking their money with them.

The calm before the storm: AMD's Zen bears down on Intel CPUs

Re: "competitive TDP."

"We've been getting a steady stream of complaints that "the new server is no faster (or slightly slower) than the old one" - and invariably the culprit is badly written, singlethreaded code that simply doesn't know how to run in a multicore system."

We have a similar problem, but the root cause is PHBs thinking that more cores on the same memory + cache config = more speed. They are finding out that more cores is f.all use when memory is the bottleneck. With respect to threads, they tend to make the cache/memory contention problem *worse*, the ideal is a bunch of loosely couple processes that share as little memory as possible. :)

Microsoft has open-sourced PowerShell for Linux, Macs. Repeat, Microsoft has open-sourced PowerShell

Re: Have you used Powershell?

"Can I ask an honest question? How many of the Bash people who are on here bashing Powershell have actually used it?"

I am not a "Bash" person, but I use it daily... The shell + 'standard' UNIX utilities have ~40 years worth of effort & usage invested into them across all kinds of OSes and hardware ranging from -11's all the way up top 10 HPC clusters. They have proven themselves over and over again, Powershell has to appear to be a lot better than the incumbent to win folks over, Microsoft's entire business is built on this concept.

From my point of view (which doesn't count for a great deal in the scheme of things), Powershell just isn't better. I found it was actually *harder* to use - more verbose, a bit jarring on the eyeball and obviously a lot less familiar than my comfy awk slippers and sed pipe. I'm not saying Powershell is all wrong or fundamentally broken, it's just ugly, ungainly, weird and unattractive to my eyes. By the same token countless "MS People" asserted the UNIX "standard" utilities are also ugly, ungainly weird and unattractive to their eyes. It's the vi/emacs war all over again. :)

In the long run I think a bit of cross-pollination of ecosystems is usually a good thing and this is no exception. I won't be unhappy if Powershell unseats Bourne shell *if* it really is a better option, I just want to get the job done without having to make a drama out of it.

"I find it's use of words makes it more easily readable"

I'll let you have that point, but my first impression of Powershell was that it looked like someone had decided to marry the readability of COBOL with the simplicity, elegance, portability and flexibility of DCL. ;)

Personally I found PS pretty awkward to use - but I've been mucking about with tcsh, ksh and bash for a couple of decades - so I am probably incapable of giving it a fair shake. I can accept that some folks like PS - fair play to them, but I don't understand *why* they like it !

Snowden latest: NSA targets Gaza, pumps intelligence to Israel

Re: Roo Roo Jimal Gis Bun Hmmmm

I'll give you a hint: Try searching for "Israel acknowledges it is helping Syrian rebel fighters", it features "Defense Minister Moshe Ya’alon". One of the results should be to an article hosted by the "Times of Israel", the article was published on the 29th of June, 2015.

It has a link to an earlier report about Druze lynching an Ambulance, which states the IDF "has insisted it does not offer medical treatment to Islamist rebels.". It could well be a case of the right hand not knowing what the left was up to.

With respect to Hezbollah not shooting up people treated by the IDF in 2015 there is plenty of material out there - easy to find, plenty of grist for the mill. This particular dimension to the Syrian conflict doesn't get much as much attention as it deserves in my view. YMMV.

Re: Roo Jimal Gis Bun Hmmmm

1. Quite correct w.r.t the refugees, however your entire points are totally irrelevant to the point at hand. So the answer to my question is : No, you don't think at'webs.bout it.

2. Again, you don't think about it.

3. The world is well aware of this irrelevant point. Again, you don't think about it.

4. Unsupportable and irrelevant supposition. Again, you don't think about it.

I've got my answer loud and clear Matt. The Fail is for you.

"It is certainly PORTRAYED as committing atrocities. But "atrocities" implies intent"

Firing 155mm HE shells onto a crowded beach on a summer's day was done with intent to kill and maim, that the expected result of 155mm HE shells fired into crowded areas.

Most folks living within the borders of Israel would bite someone's arm off if they were offered the chance to live in peace and prosperity. Some folks achieve that, but by and large the Palestinians continue to have their land, livelihoods and homes taken away from them and while that process continues they really have no option but to lay down and die or fight. Reverse that process and they have the option to live in peace.

Re: Jimal Gis Bun Hmmmm

"BTW, you may want to read up on the Beirut Marines barracks bombing to get an idea of why Hezbollah is designated a terrorist organisation by the US."

I am genuinely curious Matt, how do you feel about the IDF supplying material, intelligence and medical assistance to ISIS folks who are shot up by Hezbollah ?

Intel overhyping flash-killer XPoint? Shocked, we're totally shocked

Re: Quite good

""Quite Good is nothing to sneeze at, when most things are, ipso facto, Fairly Average. "

The question is "Will you pay ten times more for it?" and the answer in 90% of cases is "NO""

Agreeing violently !

They've done pretty well already with Gen.1, and they are at the start of XPoint's development curve so there is likely a lot of room for improvement on price and performance, I think there is reason to be optimistic - particularly if other big vendors license it. On the other hand Flash has had a couple of decades of competitive development invested in it, there is much less margin for improvement with Flash and much thinner margins.

Linux security backfires: Flaw lets hackers inject malware into downloads, disrupt Tor users, etc

Nice hack

Neat hack. Slightly relieved that HTTPS & SSH still work. :)

Oracle to shutter License Services division – source

BuSab

It appears that Oracle have chosen to retain the services of BuSab. :)

Reactive? Serverless? Put to bed? What's next for Java. Speak up, Oracle

Re: Hasn't Java EE been long dead?

Nah, it's just having a long nap in the tarpit.

Web pages, Word docs, PDF files, fonts – behold your latest keys to infecting Windows PCs

Multiple drive by root exploits...

Yet more ways to exploit rendering code running in ring 0. It's getting dull watching MS punch themselves in the balls, it would be nice to see them admit defeat & take on an idea that originated outside of Redmond. Sadly I suspect that may be a step too far for them.

Of course It is technically *possible* that they may have already taken the lesson, but it would be impossible to tell that from the security bulletin or patch release notes - both seem to have been redacted to uselessness. They are more of a hinderance than a help. :(

Joking apart, it is clear that Redmond's is just going through the motions and their heart is no longer in it. It would be best for everyone to simply disconnect the life support machine from Redmond and use the talent, time & cash freed up to do something more productive for everyone.

Windows 10 Anniversary Update is borking boxen everywhere

Re: Boot scoot

"You know the drill: Clean install!

You say you're afraid you have a virus? Clean install!

You have a txt file that won't open? Clean bloody install!"

The has been the norm since day one - including DOS, the question I have is : Why do MS still store user data on the same partition as the system guff given that users are expected to rebuild their OS as a matter of routine ?

UK employers still reluctant to hire recent CompSci grads

Re: @Roo Interesting...

"I would hope that a CS course would cover something a bit more scientific and a lot broader than your point 2" developing useful tests"

Testing is hard to get right and you're right, it's a very broad field. I chose to phrase #2 in that way because I often I see tests that simply bump up the code-coverage percentage but contribute little or nothing towards validating the system under test. I know I'm not alone in that because folks have developed entire methodologies to address that problem - but I don't want a methodology I want folks who can determine whether a test is *useful* rather than blindly follow a recipe book.

"I have great respect for Tony Hoare and his CSP I would prefer people to be familiar with Rob Milner's work instead (CCS, ACCS, SCCS, and Pi-calculus"

Fair comment, my reason for wanting folks to grok CSP is that it's a very straight forward model that is fairly easy to understand and apply to pretty much anything hardware or software. If they knew Pii-calculus that would be great, the others I can't comment on because I know precious little about them. :)

I'm happy to agree with you that most of this stuff wouldn't necessarily be appropriate for a pure Computer Science course, not sure what the correct course title would be though. :)

"I had an email from this guy who wanted to use it as his final year project for some US University, and he wanted me to convert it to NTSC *and* also, would I very kindly design a printed circuit board for it for him....."

I trust that you replied with your daily rate and NDA agreement for him to sign. :)

Re: Interesting...

"and they need to know about the full software lifecycle"

There are a lot of talent out there amongst the dross, but even the gems tend to lack the following essentials:

1) Knowing what a source control system is and how to use it.

2) Developing *useful* tests.

3) Communicating Sequential Processes (threading knowledge is fine - but it doesn't help folks develop scalable distributed applications at all).

4) Understanding bandwidth & latency - and how it applies to things like CPUs/memory/networks & storage.

5) make (not because we use it, just the principles and how you manage dependencies between components).

YMMV :)

Explo-Xen! Bunker buster bug breaks out guests from hypervisor

"Not that I'm an expert, but as a general principle I would expect that the "hardware" features used to implement hypervisors are likely to have at least as many bugs."

Few people appear to pay attention to the Errata sheets published by CPU vendors, you could become an expert on the topic if you read a couple of them. :)

I suspect folks who have read those errata sheets and who are serious *serious* about securing their hardware they would give things like hypervisors and x86 hardware a miss and maybe looked for something a bit easier to lock down.

Rip up your AMD obits: Gaming, VR, embedded chips to lift biz out of the red by 2016, allegedly

Re: Wait and see

"C'mon man, that post reads like all the hand wringing hopeful posts I read from AMD fans 24 hours before the NDAs finished at the Bulldozer release."

Spot on. :)

However ...

Granted AMD really have no credibility left, but given how heavily optimised for low-end late to market processes their gear is, their CPUs do have plenty of room for improvement. The low-latency cache caught my attention - but I would be surprised if they deliver something radically different/better than the current competition...

That said I'll suspend my disbelief until I see some SPEC results, which is more than I did for the SPARC M7. :)

Mellanox plans to SoC it to storage speed with Multi-ARM BlueField

Interesting beast, I wonder if it has enough FP grunt for the HFT folks to get some joy out of it. If nothing else it would be interesting to have a few dozen of these things in a box to play with. :)

Softbank promises stronger ARM: Greater overseas reach and double the UK jobs

Re: Is this the double size already announced or double the double size...

"Something similar was said about Inmos. Have you seen Aztec West lately?"

*Sniff*

Re: Is this the double size already announced or double the double size...

"The engineers with the skills (which are NOT easily transferable) are not going to just up and move somewhere else."

All the chip-design folks I knew at British chip design firms that got bought, either moved abroad or moved to another sector. As charming as Cambridge may be, I don't see any reason to believe ARM will buck that trend. :(

Chipzilla veteran joins IBM's OpenPOWER

It's hard to argue with sales...

There are quite a few *technical* folks who would prefer *less* cores, more cache per core and more memory bandwidth than Xeons offer right now. The problem is convincing the PHBs that sacrificing cores for cache & memory bandwidth is a win - they have a habit of thinking more cores = more speed.

Here's how police arrested Lauri Love – and what happened next

Re: Flocke Kroes AC Although the burden of proof lies with Love

"they have no grounds to suspect me of any crime"

If you have nothing to hide, why do you post under a pseudonym to slander folks on a tech website ?

Re: AC Although the burden of proof lies with Love

"That is why the NCA chose not to prosecute when he didn't supply his encryption keys, because to do so would hold up his extradition, and they would rather send him off to the States."

The NCA have a duty of care towards UK citizens regardless of their feelings and the orders given to them from across the pond or what the accused has done. If the NCA were to behave in a lawful manner they would be doing everything they could to ensure Love was rehabilitated and returned to society rather than sucking up to their buddies across the pond.

Re: The law only exists for the wealthy

"The law only exists for the wealthy

For the rest of us we take our (rather thin) chances...."

The law exists for everyone, the wealthy can afford a better outcome. ;)

Re: So the Rozzers went "Softly, Softly" on them then

"My barrister said she had never seen such a mishandled case before, or such deliberate stonewalling by the Police."

Presumably that was your Barrister's first ever case. :)

Bank tech boss: Where we're going, we don't need mainframes

"This was not because “a mainframe in itself is a bad technology. It’s maybe one of the most virtualised environments ever invented, even before the whole hypervisor was there.”"

So he hasn't read the history books and is currently repeating history. The new money for old rope game never grows old.

InfiniBand-on-die MIA in Oracle's new 'Sonoma' Sparc S7 processor

Re: Agreed!

"More competition for Intel's dominance in the data center is good, looking forward to benchmarks as with that clock speed they should be good."

I agree, competition would be good and I really want the left-field stuff to thrive (variety is good !) but I doubt the S7 will be anywhere near Intel chips single-threaded performance. The M7's SPEC_rate figures look competitive against Intel v3s, but Oracle won't publish non-rate figures so you can't compare apples with apples for single threaded performance. Dividing the _rate figures by the # of physical cores on both Xeon v3's & M7 doesn't paint a pretty picture for the M7, so I wouldn't expect the S7 to turn the tables on single-thread performance either.

I want to be proven wrong by Oracle, but they are very reluctant to do so. ;)

Re: Silicon wasteland

The infiniband bits should not really take up much die area in comparison to cache etc. What isn't clear to me is whether the package has "Infiniband pins" that are not connected - hopefully that is the case so they can follow up on their promise to make Infiniband work without forcing a board redesign...

"Therefore, verifying that Sonoma's InfiniBand controller works with all the various InfiniBand adapters on the market was deemed a distraction."

If that were the case I would have expected them to ship it with the pins connected but disabled by firmware so they can verify it *after* the shipping date. On the other hand shipping with the pads disconnected would make sense if they didn't actually implement the interfaces or discovered some basic design flaw...

No means no: Windows 10 nagware's red X will stop update – Microsoft

"There is nothing wrong with Linux, but if you can't help them and they need support they are snookered."

The same applies to Windows. I see plenty of snookered people running XP, Vista, 7, 8.1 and 10. PC World can't really fix stuff like drivers not existing for their old peripherals for example.

"I have one windows 10 dual boot system for gaming - and I've removed everything except the games. Runs much faster now. Most of the time I'm on linux mint 18 which although only in beta, looks more polished and stable than windows 10."

I upgraded the kid's boxes from Win 8.1 to 10 over the weekend. It was a long & awkward process due to having to run the Windows Update Troubleshooter multiple times before and after the upgrade on all three machines.

Our eldest asked "Why is it different ?" and then asked "Can I have Linux Mint like your machine because it's easier to use ?". As it turns out all her favourite games now run under Steam on Mint just fine, so Win 10 will be gone at the first sign of trouble in this house. I lie, 10's Windows Update has already failed three times resulting in a few lost hours... Win 10 will be gone at the second sign of trouble... :)

NVMe SSDs tormented for months in some kind of sick review game

"Mr Pott, I salute you for your dedication to your craft. :)"

I second that, fair play Trevor. :)

Even in remotest Africa, Windows 10 nagware ruins your day: Update burns satellite link cash

Re: Don't use Microsoft products ....

"Any bright ideas as to how this behaviour can be stopped before somebody gets killed?"

I suggest sending MS big shots to the game reserve to do a bit of game keeping wearing nice bright visivests and toting slingshots would be the best solution. I wouldn't expect sociopathic MBAs to "get it" until their own lives & money are on the line.

More than half of people on UK counter-terror biometrics databases are innocent

"Only yesterday, I was reading of a "crackdown on legal highs"."

I was able to buy a couple of beers yesterday without any problem from a shop on the high street, so it looks like there is no effective crack-down on "legal highs".

The underbelly of simulation science: replicating the results

Re: Here's a simple experiment...

"> Almost certainly the two binaries produced will be different,...

If you really see this happening, file a bug report."

Before filing that bug report I would do a quick double check to see if it was using some form of profile driven optimisation. The EPIC (aka Itanic) folks used that approach to help them achieve parity and eventually beat much smaller, faster, simpler and easier to code super-scalar RISC cores (which were fabricated on fraction of the die area using a process that was at least 4 years out of date).

Computer architecture folks had already worked out that you needed dynamic instruction scheduling to get the most out of the hardware decades earlier, Intel *and* HP had already learnt that lesson so you have to wonder how they convinced themselves that an entire ISA explicitly designed for statically scheduled instructions would help.

"Wouldn't it be of scientific interest to develop specialized CPU's with much larger floating point units for simulation."

That's been done repeatedly, the hardware often lies idle. There are various libraries that offer arbitrary precision, interval arithmetic etc, etc etc, but these tools mostly lie idle mostly through ignorance of their existence and applicability.

If folks are made aware of these tools there are additional barriers:

1) Performance will take a hit. Folks tend to want a credible but wrong answer faster than the correct answer slower (see the background to Seymour Cray's "Parity is for farmers" quote).

2) It will be "too hard" to validate the results (which they know will be different so they can't simply run diff).

3) It will be "too hard" to get downstream systems/regulators/phbs to accept the new results because they are different.

Oracle drops 248 – count 'em – 248 patches, to fix ... something

"If you think that about a microkernel OS like Windows, what about Linux?! It's way way worse in that regard."

I agree a microkernel OS *should* be smaller and therefore easier to validate but Windows is not a microkernel OS because it runs stuff like font file parsing & rendering in ring 0. By contrast Linux does not render fonts in ring 0 - so that's *less* code to validate in the Linux case.

A lot of this is a moot point anyway - because the modern Wintel hardware has a metric shit load of protection domains that overlap, there are at least two *more* privileged layers above classic "ring 0" these days. Folks need to audit the firmware on the processors and motherboard these days, folks on old school RISC platforms have life a bit easier. :)