Feeds

back to article Linux boot doesn't smash Samsung laptops any more

Samsung laptops will no longer be irreparably destroyed when their users try to boot Linux on them, kernel chieftain Linus Torvalds made certain today. The brainboxes down at Ubuntu-maker Canonical penguinery issued a warning over the incompatibiity and have been tackling the nuclear bug that destroyed the whole Samusung laptop …

COMMENTS

This topic is closed for new posts.

Page:

This post has been deleted by a moderator

Silver badge

Re: Sorry, wrong.

Samsung (Or whoever makes the firmware for them) wrote an off-spec implementation and probably tested it only using Windows. Determining that it booted Windows fine, they apparently decided that it was done.

We've been here before. Exactly the same issue was had years back with the early days of ACPI: Many mainboards used not-fully-compliant implementations that worked for Windows, and didn't bother to test for any other OS. It's the unending curse of linux. Outside of the server space, it is such a niche OS still that there is little to no incentive for manufacturers to even test if their hardware works, much less go to the expense of releasing drivers.

29
0
Anonymous Coward

Re: Sorry, wrong.

Linux code isn't always that bad. There are a lot of morons who commit bad code and think the community will fix it. But Linus keeps telling them off.

14
0
Black Helicopters

Re: Sorry, wrong.

"It seems to be the common practise."

Welcome to the world of Windows-only hardware.

13
2
FAIL

Re: Sorry, wrong.

Any respectable specification *should* have a conformance test suite. Googling finds me a UEFI Self Certification Test specification, so it looks like the spec writers did their job. However, we all know what 'Self Certification' means (think CE marks). Looks like the only test Samsung did, as you say, is whether Windows boots ok.

13
0
Linux

Re: Sorry, wrong.

There may be a minority who want the effectiveness Linux can give to a system admin's, developer's or ISP operator's desktop, but Linux is still used on more desktop hardware than it is installed on. If a Windows using friend asks for help with a broken system I'm likely to boot it from a Linux USB or CD just in order to test that all the hardware works. If it does, it generally indicates a software problem, and if it doesn't it generally means hardware is at fault. Saves much wasted time. How long would it otherwise take to do a full reinstallation of Windows + applications if you needed to do this just to find out whether it's a hardware or software problem ?

27
0
Silver badge
Devil

Re: Sorry, wrong.

Fixed that for ya: "Worked" for windows. Quotes intended.

About 50% of what was erroneously attributed to "bad drivers" was actually bad acpi in the early days. Windows was suffering same as linux if not worse - at least on linux you could easily turn off acpi altogether.

We now have the "history repeating". I am typing this on an Asus Motherboard whose crippled UEFI implementation erroneously initializes the AMD Fusion GPU portion every 3rd or fourth boot. And that is a so called "reputable manufacturer". Everything is just like in the worst days of ACPI 10+ years ago.

13
0

Re: Sorry, wrong.

There is one. It is called the Self Certification (SCT) for the UEFI Specification Version 2.3. It is available at http:://uefi.org.

0
0

Re: Sorry, wrong.

Actually, no. The real bug is that efivars does not use QueryVariableInfo() (Section 7.2.2 UEFI Spec 2.3.1c) to check maximum size of storage for a variable. It simply assumes space is available.

4
1
Silver badge

Re: Sorry, wrong.

I gained about 50 downvotes for suggesting that Linux was partly to blame..

Thanks for posting........I run both Windows and Linux and as a programmer I will never consider any OS or application as being perfect.

Shit happens in IT because code is written by humans and none of us are perfect.......

We learn from our mistakes not from our success.

12
4
Bronze badge
Linux

Uhh, well then...

"The error probably has its origin in mixing up UEFI version 1 and 2. In version 1 variable names still had a fixed size of 128 Byte. Version 2 changed this, but Samsung didn't. "

Oh, no...that means something sicking.

Would I be completely wrong stating that Samsung wrote a buggy bios, and Windows 8 only works on it because it is written to handle buggy bioses?

Sorry, I'm the biggest promoter of Linux within all my friends, and even I think that Windows is doing a better job than Ubuntu with UEFI. So, +1 to Windows 8 in the game of Windows 8 VS. Ubuntu. Sadly, that will be the last time they score.

I never read, but is other Linux distro's affecting these laptops? Is Debian affected? I only care about Debian, it's descendants are dead to me.

1
7
Bronze badge
Happy

Re: Sorry, wrong.

@Khaptain: "I gained about 50 downvotes..."

Yeh, but what does that matter to you? For the hell of it I down voted you, I never pay attention to the voting process of this or any other site, but I noticed this text on this particular site when I down voted you...

"We’re sorry you didn’t like this post."

From that line, you immediately know it's bullshit. I mean, how sorry are they? On their knees sorry?

4
12
Anonymous Coward

Microsoft UEFI?

They might as well name it Microsoft UEFI ..

"There does not seem to be any way to contact Samsung for reporting hard- or software related bugs"

5
1
Silver badge

Re: Sorry, wrong.

Why is it important.

To me personally it is not important. I have a life outside of El Reg, lol,

What is important however is the general consesus, many people vote without thinking beforehand. In the world of IT, pre-thought is vital in everything that we do. Our decision as IT Admins, Programmers or Managers, really does make a difference, we provide the tools that help or hinder our companies.

As professionals we shouold remove ourselves our Golden thrones, jump back down to earth, and consider that we are "paid hansomely" to provide solutions... When we "pretend" to know it all or pretend that one sytem is far superior to another , we are merely fooling ourselves.

Neither Windows, Apple IOS, Linux or any oher sytem holds the "complete" solution. As IT professionals we hold the solution by providing the best that we can with what we have available.

I very seldom agree with some of the "Gold Badged hacks" sense of "journalism", I won't name anyone in particular, but most of what they have to say is valid, albeit a bit theoretical and too much SMB based but it remains honest.. This is contrary to what I understand from the downvotes of the"pretend" Linux crowd, most of these downvotes probably come from neophytes hoping to make their place within the industry.. They still have a "lot" to learn".

Do down votes really mean much, of course they don't, but it does give a certain insight into the readership and their level of experience.

IT is a much as philosophy as it is a science...... We deal with people more than we deal with systems.

When we learn to become honest about what we reallly know; then we start to become usefull.

9
2
Thumb Up

Re: Sorry, wrong.

Me, too. First test, a live linux cd, second test, UBCD for Memtest86+ and the hdd manufacturer's test utility.

2
0
Anonymous Coward

Re: Sorry, wrong.

"... I mean, how sorry are they? On their knees sorry?..."

You do realise that, every time someone hits that downvote button, El Reg's masters launch a full internal enquiry, inevitably resulting in some poor trainee reporter being scapegoated as the cause of the "reader disatisfaction".

Subsequently manhandled from the building and thrown into the street, said hacklette will be lucky to ever work in journalism copy/pasting from other websites again and often ends up descending into a world of crime, prostitution and premature death.

This carnage has taken such a toll that El Reg has recently begun sending civilians into the frontline, in its increasingly bitter war against the forces of balanced reporting.

So think, before idly hitting that button. OK?

22
1
Linux

Re: Uhh, well then...

Nope, it's the case that the buggy BIOSes are tested with Windows and made to "work" just enough to pass QA at the time. Bad coding is bad. Windows 8 does not handle buggy BIOSes better, in fact, buggy BIOSes are often so bad that OEMs release BIOS updates.

You'll often find once a workaround is created, Linux handles known bugs better.

4
0
Big Brother

Re: Sorry, wrong.

I'm sorry, I thought all journalists lived in a world of crime, prostitution and premature death. Because that's what sells advertising.

Big Brother, as a tribute to those fine journalists of the British Tabloid Press - especially the one who recently wrote that journalists shouldn't nark on bent coppers.

1
0

@DanceMan Re: Sorry, wrong.

Last month I had 2 infected machines in my office so I downloaded a bootable Anti Virus CD from one of the more popular companies and was shocked that it was a bootable Linux CD.

0
0
Anonymous Coward

Re: Sorry, wrong.

"In the world of IT..."

You are assuming people that work in IT read this site. Judging by the comments, there are also ignorant people reading this site. It's still better than Wired though.

2
0
Silver badge
Pint

Re: Sorry, wrong.

I thought the majority worked in IT the rest worked at being trolls. There is also a lot of overlap too.

0
0
Silver badge

Re: early days of ACPI

It's not just Samsung, and it's not just Linux. Hardware vendors write a lot of stuff that doesn't quite meet published spec, but when released with a current popular OS works.

Back in the early Pentium days, before Linus had even started work on Linux there was a company that released a multiple CD drive that ran on the IDE channel. Beautiful part about was, under Windows 95, you could share each of the drives in the Caddy, and computers linked on a work group could read what was in the CD caddy. We had a client running some library software in a Windows workgroup configuration (connections to remote PCs were pretty much by modem in those days) and they depended upon this functionality. One day they had problem with an internal modem that fried the motherboard. CPU was good, so we tried to replace the MD and memory. Win95 would install just fine, find the device and initially you would see all the drives. Then you installed the Intel chipset for the MB to make all the devices accessible and all of a sudden the device stopped working. We spent hours trying to figure it out before calling the manufacturer. The driver didn't work with the new Intel chipset, and the company had no intention of updating the driver. They had even discontinued making the device, so we couldn't even just buy a replacement. Instead we had to install a SCSI card and a SCSI caddy, about tripling the bill to the customer (we discounted the labor since we should have called the vendor sooner than we did).

Years later I was in a different environment where they'd been using an IDE "tape drive" for ages to backup data from local PCs (don't ask what they were backing up, I thought it was epic fail at the time), but still in a 95 environment. But drive sizes were finally getting large enough to force us to install 98 on some systems, and the tape drives started failing on the 98 systems. Because apparently the manufacturer didn't fully meet the IDE spec for the drive.

Now, I recognize that sometimes specs come out with things that people don't really have a way to test at the time the spec was written, and that there are monetary limits on what a company can reasonably be expected to test. But cases like these always leave me scratching my head.

0
0
Silver badge

Re: remove ourselves our Golden thrones,

That's what we do on our jobs.

This is El Reg.

It's where we come to not have to be above suspicion, let our hair down, and talk smack about any damn thing we want to without worry about real world consequences.

0
0
Silver badge

Re: Sorry, wrong.

"It seems to be the common practise."

And Microsoft are delighted about this, because it effectively locks competitors out.

Side question: To what extent do the various layers of bodgery needed in Windows to correct various broken implementations of standards end up impacting on its performance?

0
0
Silver badge

Re: early days of ACPI

"early Pentium days, before Linus had even started work on Linux" ..... "under Windows 95"

Um. You do realise that the first Linux kernel release (5 October 1991) was made on an 80386 and predated even Windows 3.1 (6 April 1992), let alone Windows 95 (24 August 1995)?

Not that this invalidates your argument in any way. This, incidentally, is why I support giving hardware manufacturers a simple choice between releasing full annotated Source Code for drivers, or having their products banned from the market.

0
0
Bronze badge

A hardware manufacturer changes its design without telling those who need to know.

Whodathunk?

I'm glad the real problem has been discovered, I really don't like the idea of hardware going into a coma because you put a different OS on it.

7
0
IT Angle

chRe: A hardware manufacturer blah blah blah...

I'm glad the real problem has been discovered

The real problem(s) has not been discovered. Tis but a scratch, not so deep as a well, nor so wide as church-door; but 'tis enough.

See James Bottomley's UEFI notes here --> http://blog.hansenpartnership.com/

7
0
Silver badge
Paris Hilton

Random Poking

Who would have thought that just randomly poking memory of a laptop would brick it.

Has randomly poking anything every worked out well?

16
1

Re: Random Poking

no, and writing code that assumes every error code you don't know about should be treated as "successful" doesn't work that great either.

4
1
Silver badge
Coffee/keyboard

Re: Random Poking

Yes. It has. For a teenager, causing a disk drive to seek-until-death was about as fun as it gets*.

* - provided you don't do it on your dad's new 5.25" drive.

6
0
Anonymous Coward

Re: Random Poking

"Has randomly poking anything every worked out well?"

I believe it to be "HOW BABBY IS FORMED". As to the desirability of this outcome, that can remain as an exercise for the reader.

11
0
Anonymous Coward

Re: Random Poking

As to the desirability of this outcome, refer to theodore's post (immediately above)

1
0
DJV
Mushroom

Re: Random Poking

There's some nice killer POKEs here:

http://en.wikipedia.org/wiki/Killer_poke

The Commdore PET one was the first one I ever heard of that would destroy hardware.

3
0
Go

Re: Random Poking

I suppose the moral here is PEEK before you POKE, it saves a lot of embarrassment the following morning.

7
1
Silver badge
Coffee/keyboard

Re: Random Poking

"Yes. It has. For a teenager"

Really thought that comment was going somewhere other than disk drives...

4
0
Bronze badge
Facepalm

Re: Random Poking - A million Pis waiting to die?

The Commdore PET one was the first one I ever heard of that would destroy hardware.

Looks like the Raspberry Pi could have the same feature -

http://www.raspberrypi.org/phpBB3/viewtopic.php?t=5339&p=71145

0
0
Silver badge

Re: every worked out well?

Back in the day it was how you found out how things worked.

Peeking and poking on a C64 provided days of amusement and learning.

0
0
Bronze badge

Speedy fix!

It seems that the Linux community is pretty quick to respond to problems. At times I wish other vendors were as good, but it will take a while.

18
1

Re: Speedy fix!

As much as I'd like to agree with this, the problem has been known about for several months, though not the exact cause. I recall reading release notes back in October warning not to install Ubuntu on certain models of Samsung laptops with UEFI firmwares, that several had been bricked.

5
0
Anonymous Coward

Re: Speedy fix!

You're having a laugh right, Herby? Bug fixes take an *age* on Linux. Just spend a couple of minutes browsing through a few bug reports if you want proof.

That's what happens when you rely on people that are not paid to do a job. They work to their own schedule, not that of the long suffering users'.

5
17
Silver badge

Re: Speedy fix!

>not that of the long suffering cheap tightwad whiny users'.

The code is right there for download. Its easy enough to hire someone to fix it. Plenty of other companies do so. Support isn't free for anything but the Linux community still gives more than most.

10
3
Anonymous Coward

Re: Speedy fix!

"ou're having a laugh right, Herby?"

Evening RICHTO !

4
2
Gold badge

Re: Bug fixes take an *age* on Linux

Bug fixes take an age on any platform where the maintainer doesn't think it is a serious bug.

I'm sure you can peruse Linux bug trackers and find bugs that have been sitting around for 5 or 6 years with full proof and repro cases but no fix, just as you can find "knowledgebase" articles detailing failures in Microsoft products that are acknowledged as a bug with no workaround and no fix forthcoming.

I'm *not* sure this proves anything.

1
0
Silver badge

Re: Speedy fix!

"That's what happens when you rely on people that are not paid to do a job. They work to their own schedule, not that of the long suffering users'."

Funny, the same thing happens most of the time when people are paid to do a job - bugfixes don't bring in extra income. New features do.

0
0
Anonymous Coward

Are you listening Eadon?

People like to slag off Windows, but for all its faults I can't remember a time when simply booting up bricked the machine!

3
17
Silver badge
Trollface

Re: Are you listening Eadon?

I sure can but that had more to do with overclocking programs than windows itself per se.

2
0
Anonymous Coward

Re: Are you listening Eadon?

I can't remember a time when simply booting up bricked the machine

No, that usually requires Windows to be fully online first :)

12
2

This post has been deleted by its author

Mushroom

Re: Are you listening Eadon?

Chernobyl

0
0
Anonymous Coward

Re: Are you listening Eadon?

"but for all its faults I can't remember a time when simply booting up bricked the machine!"

You must be still in nappies (diapers) or shorts then

8
0

Page:

This topic is closed for new posts.