Quantcast
[ 3 / biz / cgl / ck / diy / fa / g / ic / jp / lit / sci / tg / vr / vt ] [ index / top / reports / report a bug ] [ 4plebs / archived.moe / rbt ]

Due to resource constraints, /g/ and /tg/ will no longer be archived or available. Other archivers continue to archive these boards.Become a Patron!

/g/ - Technology


View post   

[ Toggle deleted replies ]
File: 98 KB, 650x421, image.php.jpg [View same] [iqdb] [saucenao] [google] [report]
61786420 No.61786420 [Reply] [Original] [archived.moe] [rbt]

>This morning I was on a call with AMD and they are now able to confirm they have reproduced the Ryzen "segmentation fault issue" and are working with affected customers.

>AMD engineers found the problem to be very complex and characterize it as a performance marginality problem exclusive to certain workloads on Linux.

>AMD was also able to confirm this issue is not present with AMD Epyc or AMD ThreadRipper processors, but isolated to these early Ryzen processors under Linux.

http://phoronix.com/scan.php?page=news_item&px=Ryzen-Segv-Response

>> No.61786433

intel fags on suicide watch

>> No.61786551
File: 85 KB, 653x668, 1501926861131.png [View same] [iqdb] [saucenao] [google] [report]
61786551

>> No.61786570

AYYMD fags on suicide watch!

>> No.61786577

>>61786551
http://www.phoronix.com/scan.php?page=article&item=ryzen-segv-continues&num=1

>Update [5 August]: As a result of feedback, currently working on some updated results. As some have pointed out, the conftest segmentation faults aren't specific to Ryzen, so updating the tests to avoid confusion.

Kill yourself, retard.

>> No.61786591

>>61786433
>>61786570
Yes I definitely think everyone involved in the shitposting wars should slip into a noose.

>> No.61786604

>>61786420
>isolated to these early Ryzen processors under Linux

Open sores trash strikes again.

>> No.61786611

>>61786604
Linus to write angry message on mailing list and flip off AMD at a conference somewhere.

>> No.61786614

>AMD was also able to confirm this issue is not present with AMD Epyc or AMD ThreadRipper processors
This kill the macacos

>> No.61786701

>>61786420
We'll at least it's not frying SATA ports like Sandy Bridge launch

>> No.61786763

So newer stuff isn't affected? Did something change between early chips and latter chips in the production?

>> No.61786781

>>61786420
AMD is DOA

>> No.61786812

>>61786604
It mean AMD will not fix.
They barely have money to pay the poojets to glue their dies. Expect when the dies start to die due to how shitty they were made.

>> No.61786824

>>61786420
so people get a newer ryzen or are they just openly admitting that ryzen is their beta test?

>> No.61786836 [DELETED] 
File: 94 KB, 748x584, ded35f84023c11c685127f6118a24b5fcf48a04b7fe11fb9d42089e86fdc1442.jpg [View same] [iqdb] [saucenao] [google] [report]
61786836

>>61786812

>> No.61786840

>>61786420
does this mean intel shills will start spamming linux benchmarks here now?

>> No.61786852

>>61786420
Similar kind of shit happened with Phenom TLB bug. This time they were wise enough to push first silicon to consumers though, if EPYC had this bug it would be a disaster

>>61786611
>>61786812
Linus will throw some f-bombs and pull the workaround

>> No.61787478

>>61786812
>Ryzen customers believe to be affected by the problem can contact AMD Customer Care. Some of those who have contacted customer care about the segmentation faults have in turn been affected by thermal, power, or other problems, but AMD says they are committed to working with those encountering this performance marginality issue under Linux.

>> No.61787521
File: 387 KB, 1500x1164, india inside.jpg [View same] [iqdb] [saucenao] [google] [report]
61787521

>>61786812
>They barely have money to pay the poojets to glue their dies

>> No.61787535

>>61786420
TFW I EARLY ADAPTED AND AM LOONIX FAG

NOOOOOOOOOOOOOOO

oh wait I haven't crashed in months

tfw intel has had more issues than amd has for faults

>> No.61787573

>>61787535
You might want to contact AMD about it seems like they are receptive.

>> No.61787729

>>61786812
SOPA

>> No.61787894

>>61786420
Well, this is why you don't become an early adopter.

>> No.61787995

>>61787573
you think they'll give me a new stepping?>>61787894
fuck you

>> No.61788003 [DELETED] 

>pedophile OS
AND nothing of value was lost

>> No.61788140
File: 156 KB, 633x758, 1487789493624.jpg [View same] [iqdb] [saucenao] [google] [report]
61788140

>>61786420

>> No.61788369 [DELETED] 
File: 647 KB, 605x603, lisa-monkey.png [View same] [iqdb] [saucenao] [google] [report]
61788369

>>61786836

>> No.61788394
File: 450 KB, 812x720, 1430460027689.png [View same] [iqdb] [saucenao] [google] [report]
61788394

>>61788369
Did you make that yourself, little BR?

>> No.61788403

>>61787995
I don't know what they'll do, maybe send you a new CPU. Can't hurt to try.

>> No.61788413

>Freetards who always chocked on Amd's cock are attacked by brand royalists from /v/ who ironically shill for the same company.
Just writing this gave me a semi.

>> No.61788415
File: 137 KB, 340x340, 729.gif [View same] [iqdb] [saucenao] [google] [report]
61788415

>>61788394
He didn't even use a png with transparent background

>> No.61788432
File: 395 KB, 858x797, wat.png [View same] [iqdb] [saucenao] [google] [report]
61788432

>>61788415
Ehm, what?

>> No.61788460

>>61786781
Someone say dead CPUs?

https://www.extremetech.com/computing/244074-intel-atom-c2000-bug-killing-products-multiple-manufacturers

>> No.61788467
File: 92 KB, 640x640, poop-emoji-png-42512.png [View same] [iqdb] [saucenao] [google] [report]
61788467

>>61788432
the emoji, it looks like he used a thumbnail or something. very low effort 3/10

>> No.61788480

>linux

>> No.61788493

>>61788460
Don't forget fried SATA ports
https://www.cnet.com/news/intels-sandy-bridge-chipset-flaw-the-fallout/

>> No.61788511

>>61788493
And disabled hardware features. http://techreport.com/news/26911/errata-prompts-intel-to-disable-tsx-in-haswell-early-broadwell-cpus

>> No.61788541

>>61788369
>literally only on early Ryzen chips and not on later Steppings wich include Epyc and Threadripper

>> No.61788571

>>61787521
I want to run emulators on that celeron.

>> No.61788614
File: 16 KB, 320x320, 1502135145903.jpg [View same] [iqdb] [saucenao] [google] [report]
61788614

>>61786824
>>61786812
>>61786781
>>61786551
>>61786570
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/2nd-gen-core-desktop-specification-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/3rd-gen-core-desktop-specification-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/4th-gen-core-family-desktop-specification-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/desktop-mobile-5th-gen-core-family-spec-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/desktop-6th-gen-core-family-spec-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/7th-gen-core-family-spec-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-5200-series-spec-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-5300-spec-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-5400-spec-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-5500-specification-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-5600-specification-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-e5-v2-spec-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-e5-v3-spec-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-e5-v4-spec-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-e7-8800-4800-2800-families-specification-update.pdf
http://www.intel.com/content/dam/www/public/us/en/documents/specification-updates/xeon-e7-v2-spec-update.pdf

oops

>> No.61788667

>>61786763
as far as we know, no. there were a few users on amd's official forums who multiple new rma processors from amd over the past few months said their latest ones from amd no longer have the problem, but the stepping on the processor was the same. only the build date was different.

so for all we can know it could have been a random manufacturing defect that's only able to appear when running on linux. so this also means there could be thousands more running windows with processors that could have this random, strange defect.

why only linux though is odd. maybe linux just puts a harder load on it.

>> No.61788918

>>61786812
T. Brian Krzanich

>> No.61789256
File: 602 KB, 1440x2560, 1482020398238.png [View same] [iqdb] [saucenao] [google] [report]
61789256

>>61786824
>>61786812
>>61786781
>>61786551
>>61786570

>> No.61789280
File: 559 KB, 1440x2560, 1479891250696.png [View same] [iqdb] [saucenao] [google] [report]
61789280

>> No.61789295
File: 600 KB, 1440x2560, 1474373022446.png [View same] [iqdb] [saucenao] [google] [report]
61789295

>> No.61789871

>>61787521
what's with that cpu? can someone give me a rundown?

>> No.61789900

>>61789871
The Celeron-X? It's a joke based on how stupid and pointless Kaby Lake-X is.

>> No.61789915

>>61788667
What build dates are defective?

>> No.61790083

>AMD was also able to confirm this issue is not present with AMD Epyc or AMD ThreadRipper processors, but isolated to these early Ryzen processors under Linux.
I knew early Ryzen adopters were basically beta testers for the server market.

>> No.61790103

>>61787521
>Internet of The Toilet

>> No.61790177

Typical AMD. You never know what problems you get into when buying an AMD product.

>> No.61790253

>>61786420
>Running a commie OS
>surprised it's shit

lel. You should have ran a Free as in burger land OS like *BSD.

>> No.61790255

>>61790177
>Typical Intel. You never know what problems you get into when buying an Intel product.
>>61788460
>>61788493
>>61788511

>> No.61790281
File: 98 KB, 361x453, UMA_MACACO_STRIKE_AGAIN.png [View same] [iqdb] [saucenao] [google] [report]
61790281

>>61786551
>>61786570

>> No.61790286

>>61790255
>current ryzen, and every amd product ever

>> No.61790297

at least it's not as bad as amd gpus.

>> No.61790325

>>61790286
[Citation Needed]

>> No.61790339
File: 87 KB, 927x527, intel battle tested.png [View same] [iqdb] [saucenao] [google] [report]
61790339

>>61790286

>> No.61790365

>>61790083
>Only happens on Linux
>Doesn't happen on new CPUs

FOSS can't FLOSS

>> No.61790388

>>61790365
>only happens on linus
Recreated on every os

>> No.61790420

>>61790388
>problem exclusive to certain workloads on Linux.
FOSS can't FLOSS

>> No.61790461
File: 211 KB, 659x609, amdposter.jpg [View same] [iqdb] [saucenao] [google] [report]
61790461

>AMD is this full of shit
>Blame OS
>Blame Nvidia
>Blame Game
>Blame Buyer
>Blame Software
>Blame Intel
>Blame ________

>> No.61790661

>>61790461
t. Linus

>> No.61791052

>>61790661
are you brain damaged or something? The issue is reproducible on Windows 10 (WSL), FreeBSD, and DragonflyBSD.

>> No.61791335
File: 297 KB, 1080x1080, 1502093084167.jpg [View same] [iqdb] [saucenao] [google] [report]
61791335

>>61786420
>workloads on Linux.
who cares

>> No.61793469

it happens ONLY on early Ryzen, ONLY on linux and some unix-like like freebsd, and ONLY when you do really heavy multithreading work like compiling gentoo.

It's literally nothing and will be fixed.

>> No.61793535

>>61793469
>it happens ONLY on early Ryzen
Does This Mean That AMD Knew Internally That There Was An Issue, And They Corrected It Between Batches While Still Selling The Flawed Processors? Please Tell Me This Is Not What Happened.

>> No.61793661

>>61793469
>it happens ONLY on early Ryzen, ONLY on linux and some unix-like like freebsd, and ONLY when you do really heavy multithreading work like compiling gentoo.
We all know that it happens on windows as well. Who is paying you to omit that?

>> No.61793736

>>61793535
>camelcase paragraph
Are you 10 or just fucking retarded?

>> No.61793747

>>61793661
>hurr durr if you run linux programs on windows then it also happens there

>> No.61793762

>>61793736
>Are you 10 or just fucking retarded?
Did You Have Trouble Reading It, Or Are You Just Attempting To Establish Some Sort Of Dominance Over An Individual That Is Not Near Enough To Beat You Physically?

>> No.61793780

>>61793762
Oh, I see, you're actually retarded.

>> No.61793817

>>61793780
>Oh, I see, you're actually retarded.
I'm Guessing That You're From A Place That Fancies Upvotes, And This Is What You Consider Humour.

>> No.61793853

>>61793469
>it only happens with every operating system, but I'm paid to downplay that fact.
There's nothing to see here people.

>> No.61793908

>>61793762
Yeah, going to go out on a limb and assume you not only physically couldn't beat anyone's ass, but would sit quietly like a little bitch if confronted IRL

>> No.61794608

I have early ryzen 5(ordered next day they became available for order), i run arch, i made some compilations (php 5.6 from aur, fo example) and i had 0 problems. Also tried to get segfaults with ryzen_segv_test and srill no segfaults. I don't really think that segfault problem is widely spread.

>> No.61794766

Working here on TempleOS. Linuxcux BTFO

>> No.61794880

>>61788614
OYYYYYYYYYYY VEEEEEEEEEEEEEY
That's one antisemitic animu girl.

>> No.61795055

>>61788614
>>61789256
You're not funny, also it is ok when Intel does it. Remember how much the Jews suffered during WW2. Cut them some slack.

>> No.61795068

>>61787729
DE

>> No.61795119

>>61788614
Repeat after me

ITS OK WHEN INTEL DOES IT

>> No.61795121

>>61795068
MACACO

>> No.61795198

Here are the facts:
1. Only present on linux
2. Only under specific workloads
3. Doesn't exist in threadripper or Epyc, Ryzen only.

>> No.61795211

Wait a minute, only on early Ryzens? Would that possibly mean my poorfag Ryzen 3 choice might be unaffected being such a late release?

>> No.61795381

>>61786420
>it's a "Linux ruins everything again" episode

Also they sent the guy who reported it a newer CPU and he had zero issues, so it's clearly fixable in microcode.

>> No.61795395

>>61789900
B-but that Celeron-X would be the best way to play Dwarf Fortress.

>> No.61795638

Is this one of those episodes where someone tries to run new hardware on an ancient kernel then whines when they run into issues?

>> No.61795707

>>61793908
I Bet That You Would Not Say That To My Face.

>> No.61795814

>>61795198
Are the people that have claimed to replicate this on windows, and bsd lying?
Will amd issue a recall for the affected cpus since this was a manufacturing defect that self corrected without amd's intervention?

>> No.61795834

>>61786551
kek

no but for real tho they need to fix this shit i'm tired of getting dust on my 1800x

>> No.61795875

>>61786763
Aren't Threadripper and EPYC on the B2 stepping while R7 & R5 were on B1 (pretty much Beta-test) stepping?

>> No.61796099

>>61795381
>having problems like this in the first place

This just proves that amd are operating at an amateur level.

>> No.61796120

>>61788614
>small issues with hardware that is pushing boundaries and is basically bleeding edge all the time without competition
vs.
>company that literally copy pastes a cpu architecture without having to invest any money/ressources into researches and yet still manages to fuck up on a major level like that

>> No.61796690

>>61796099
See
>>61788614
And all the dead Atoms.

>> No.61796786

>>61796690
>bleeding edge technology can't have small problems

Go to bed, kid.

>> No.61796835

>>61795814
>Will amd issue a recall for the affected cpus since this was a manufacturing defect that self corrected without amd's intervention?
It's up to the consumer to rma, but there isn't a recall.

>> No.61796919

>"WORK"loads
>Linux
>work on Linux

If its a server its meaningless anyway because as they said, this doesn't affect actual EPYC server chips. This is just more shit with Loonex. Linux is shit with GPUs and CPUs it seems now. Is there anything that dogshit OS is good for besides servers? It can't even be given away for free.

>> No.61796984

>>61796919
>Is there anything that dogshit OS is good for besides servers?
Literally everything that isn't muh gaymes.

>> No.61797018

>>61796919
> Is there anything that dogshit OS is good for besides servers?

Hacking wifi networks.
Running on old machines that would otherwise require XP.
Compiling and testing software for Linux.

>> No.61797035

>>61796919
>Is there anything that dogshit OS is good for besides servers
It's actually a pretty decent OS to consume multimedia with. Also you could argue that it is a pretty good OS for older family members because they can't fuck up too much and has all the essential stuff that is needed.

>> No.61797049 [DELETED] 
File: 92 KB, 800x500, War_aecf66_5844082.jpg [View same] [iqdb] [saucenao] [google] [report]
61797049

Test test

>> No.61797098

>>61796984
You mean anything that doesn't use a graphics or sound API. 99% of content creators use Windows or Mac for photo, video and music editing and the vast majority of CAD/3d design and engineering simulation software is on Windows.

>> No.61797121

>>61796984
Oh, Solidworks is now on linux?

>> No.61797142

>>61795814
There's been no reliable confirmation of the bug being replicated on other operating systems.
BSD I'd believe since it's UNIX so it has a lot of similarities, but I wouldn't trust the windows one until AMD admits it.
I know
>trusting the manufacturer
But even INTEL tells people about faults in their CPUs. After fixing them, and not publishing anything on them that is.

>> No.61797156

>>61797098
>content creators
no one cares about that bunch of useless sjw faggots

>> No.61797158

>>61797098
>99% of content creators use Windows or Mac for photo

Okay, but "content creators" ( I assume you unironically talk about the uneducated people who "work" as youtubers) don't know any better.

>> No.61797204

>>61797098
>3d design
>99% of content creators use Windows
thats why theres a linux version of maya for like 15 years, a linux version of cinema4d, ... pixar using linux excessively, etc

>> No.61797216

>>61795834
You should contact AMD's customer support sounds like you can get a replacement.

>> No.61797221

>>61796786
Right, because all of those errata are for """"bleeding edge"""" tech

>> No.61797312

>>61797158
You realize there are people that edit video for other purposes besides Youtube right? As in the movie and TV industry?

>> No.61797340

>>61797312
And they funnily don't solely use MS/macOS

>> No.61797428

>AMD was also able to confirm this issue is not present with AMD Epyc or AMD ThreadRipper processors, but isolated to these early Ryzen processors under Linux

So are later released 1200, 1300, 1400 safe from this? What about Ryzen 1600?

>> No.61797534

>>61797428
We don't know how "early" it goes, the ryzen CPUs have been made since February, who knows after what week the bug wasn't there anymore.

>> No.61798310
File: 112 KB, 691x771, 1497565895125.jpg [View same] [iqdb] [saucenao] [google] [report]
61798310

>>61786420
>this issue is not present with AMD Epyc or AMD ThreadRipper processors
NOOOOOOOOOOOOOOOO

>> No.61798459
File: 115 KB, 879x758, 1497163995691.jpg [View same] [iqdb] [saucenao] [google] [report]
61798459

>>61786420
AYYMD BTFO

>> No.61798530

>>61798310
>>61798459
This shit is so stupid.

>> No.61798531

>>61795121
UMA DELICIA

>> No.61798561

>Linux

>> No.61798781

>>61798561
I agree, it should be GNU/Linux.

>> No.61799141

>>61790461
Whatever. Name one other time that amd tried to pass the blame.

>> No.61799925

>>61794608
Because it's not, but Intel is in deep shit and needs to distract people from that.

>> No.61800003

>>61786551
kill yourself

>> No.61800256

>>61793469
I compiled Gentoo without any problems on my 1800X. Neato.

>> No.61801403

>>61795814
It was replicated on Windows 10 using the Linux subsystem so it doesn't really count as "Windows".

>> No.61801460

>>61797142
>There's been no reliable confirmation of the bug being replicated on other operating systems.
>BSD I'd believe since it's UNIX so it has a lot of similarities, but I wouldn't trust the windows one until AMD admits it.
Don't trust the windows claims, because you decide what other people should believe? Got it.

>> No.61802509

>>61797216
i don't really think it's a hardware issue, all bullshit and memes aside

I've done heavy multi-core parallel compiling on windows and it works just fine.

>> No.61803115

>>61796120
>bleeding edge
AMD hasn't really been bleeding edge that much, maybe a few months at a time until intel releases something wildly overpriced that will attract all the fanboys, crushing AMD.

>> No.61803146
File: 94 KB, 799x720, [HorribleSubs] Eromanga-sensei - 12 [720p].mkv_snapshot_08.50_[2017.06.24_14.32.00].jpg [View same] [iqdb] [saucenao] [google] [report]
61803146

>>61803115
>maybe a few months at a time until intel releases something wildly overpriced that will attract all the fanboys, crushing AMD.
Just like Skylake-X did, huh?

>> No.61803226

>>61803146
It surely crushed the r7 line, but this time threadripper came about, which may be able to withstand the intel fanboying.

>> No.61803277

>>61803226
>It surely crushed the r7 line
If by crushed you mean self-immolated, sure.

>> No.61803428
File: 160 KB, 1376x668, 1499615706143.jpg [View same] [iqdb] [saucenao] [google] [report]
61803428

>>61803226
>It surely crushed the r7 line
In power usage and heat output, it did indeed.

>> No.61803592

>>61803428
I meant that intel fanboying will crush it, while threadripper is just better than the intel equivalent.

>> No.61804198
File: 53 KB, 480x480, banana.jpg [View same] [iqdb] [saucenao] [google] [report]
61804198

good thing that i didnt upgrade from intel i5 2500k yet.

>> No.61804387

>>61793762
I did in fact have problems reading it faggot

>> No.61804664

>>61804387
>I did in fact have problems reading it faggot
What The Fuck Is Your Problem?
>Has Problems Reading Words
>Acts Like A Babby And Makes It Everyone's Problem
Grow Up Anon.

>>
Name (leave empty)
Comment (leave empty)
Name
E-mail
Subject
Comment
Password [?]Password used for file deletion.
Captcha
Action