Gentoo Forums
Gentoo Forums
Gentoo Forums
Quick Search: in
AMD Zen/Ryzen thread (part 2)
View unanswered posts
View posts from last 24 hours

Goto page Previous  1, 2, 3, 4, 5, 6, 7  Next  
Reply to topic    Gentoo Forums Forum Index Gentoo Chat
View previous topic :: View next topic  
Author Message
nasaiya
Tux's lil' helper
Tux's lil' helper


Joined: 17 May 2007
Posts: 142

PostPosted: Wed Oct 04, 2017 3:40 am    Post subject: Reply with quote

Chewi wrote:

...
@nasaiya, what motherboard do you have?


I have the ASRock AB350 K4 http://www.asrock.com/mb/AMD/Fatal1ty%20AB350%20Gaming%20K4/ -- upgraded to the latest bios, with ram that is on their support list and a 1600X.
I didn't care enough to research what c6 was or what 'cool and quiet' did so I disabled them and still so far so good... I have not tried gcc 7.2 yet,

EDIT: I'm also using march=znver1 for both kernel and everything else...
_________________
If it ain't broke - fix it till it is!
Back to top
View user's profile Send private message
mir3x
Guru
Guru


Joined: 02 Jun 2012
Posts: 343

PostPosted: Thu Oct 05, 2017 4:10 pm    Post subject: Reply with quote

Ok i got Ryzen back after 13 days. It came with DHL express, but there was no mail or sms notification at morning, courier only called 1h earlier.
Its 33 week SUS.

Happily I put ryzen back.
Started comp and ... no signal ... comp seems to be working.
switching all 4 hdmi outputs, connecting to tv ... no freaking signal.

Removing fan again, taking cpu out , Im slumsy like hell :D cooling paste everywhere now. Im so pissed so I put even that paste on about 20 cpu pins :D. I used brush from electric shaver to clean that mess. Putting cpu back. Then I see that nasty shining Made in Japan thing known as bios battery.
Removed it and put back. Everyhting works finally!!! ( my previous setup was all auto except virtualization )

I've read that core voltage is lower on newer Ryzens - but not in my case ... its still 1.3.
Everything works, I've set ram to 2666 and it seems its segfault free too (auto is on 2.4k).
_________________
Installation aborted to prevent system self-destruction
Back to top
View user's profile Send private message
Chewi
Developer
Developer


Joined: 01 Sep 2003
Posts: 856
Location: Edinburgh, Scotland

PostPosted: Thu Oct 05, 2017 4:24 pm    Post subject: Reply with quote

mir3x wrote:
Happily I put ryzen back.
Started comp and ... no signal ... comp seems to be working.
switching all 4 hdmi outputs, connecting to tv ... no freaking signal.

Removing fan again, taking cpu out , Im slumsy like hell :D cooling paste everywhere now. Im so pissed so I put even that paste on about 20 cpu pins :D. I used brush from electric shaver to clean that mess. Putting cpu back. Then I see that nasty shining Made in Japan thing known as bios battery.
Removed it and put back. Everyhting works finally!!! ( my previous setup was all auto except virtualization )

Funny, I had a similar experience. The fans were spinning quite fast but there was nothing on screen and I think even the power LED was off. Eventually I forced it off, tried again, and it worked but I swear it was on the older F8 BIOS to start with. I didn't notice until I went back into the BIOS later and saw F9a. My board has a dual BIOS switch but I didn't think there was any automatic failover. It's fine now anyway.
Back to top
View user's profile Send private message
krinn
Watchman
Watchman


Joined: 02 May 2003
Posts: 5966

PostPosted: Thu Oct 05, 2017 4:42 pm    Post subject: Reply with quote

Quote:
cooling paste everywhere now

it's not a good idea actually, paste is not doing any cooling, paste is use to transfer heat to another component, and too much paste mean the paste will transfer some heat to something that is not suppose to get heat.
as surprising as it is, too much paste lower transfer and cooling efficiency.
Back to top
View user's profile Send private message
mir3x
Guru
Guru


Joined: 02 Jun 2012
Posts: 343

PostPosted: Thu Oct 05, 2017 6:18 pm    Post subject: Reply with quote

krinn wrote:
as surprising as it is, too much paste lower transfer and cooling efficiency.

I had to remove fan again, I noticed max temp is 55, with previous cpu was 49.
Seems I've put screws too hard ... I barely took fan out .. along with ryzen :D. Those 2 bastards glued together.
I screwed it lighly now and max temp is 50.
_________________
Installation aborted to prevent system self-destruction
Back to top
View user's profile Send private message
Tony0945
Advocate
Advocate


Joined: 25 Jul 2006
Posts: 2038
Location: Illinois, USA

PostPosted: Thu Oct 05, 2017 6:43 pm    Post subject: Reply with quote

With all these problems and AMD trying to sweep them under the rug and horror stories about the RMA process. I'm wondering if an FX-8350 isn't a better upgrade path?
https://www.cpubenchmark.net/compare.php?cmp[]=1780&cmp[]=157&cmp[]=2966
Ryzen 7 is supposed to be a 95W processor but I see posts saying liquid cooling is neccesary and one professional review that cited 400W measured!

Meanwhile, my inquiries to AMD and Newegg result in, "No, we can't guarantee a post-week 25 CPU. They are 'probably' out of the queue and they only affected a very small number of users running Linux and only with unusual workloads (like using all the cores for compiling?)."

The Athlon II X3 is fine for e-mail and web browsing but emerges are increasingly painful. I thought of Intel but they can't be be bothered to ship a decent cooler with the CPU and when they do it is puny. Maybe bulldozer? A Newegg user user suggested Threadripper 8-core but omigod, I'd need a three ton A/C to cool it.
Back to top
View user's profile Send private message
mir3x
Guru
Guru


Joined: 02 Jun 2012
Posts: 343

PostPosted: Thu Oct 05, 2017 7:19 pm    Post subject: Reply with quote

Tony0945 wrote:
With all these problems and AMD trying to sweep them under the rug and horror stories about the RMA process. I'm wondering if an FX-8350 isn't a better upgrade path?.

Bloot is selling RMAed ryzen :D, talk to him.

Tony0945 wrote:

Ryzen 7 is supposed to be a 95W processor but I see posts saying liquid cooling is neccesary and one professional review that cited 400W measured!

No, he has overclocked with GTX1080 or so, check normal videos/articles. I saw on youtube thats with games and similar FPS ryzen takes less watts than intel.

EDIT: check intel 8700K here, 8700 without K will cost the same as 1700X, and except eating a lot of Watts under load should be faster than 1700X
https://www.youtube.com/watch?v=EJOnwF8mgXc
_________________
Installation aborted to prevent system self-destruction


Last edited by mir3x on Thu Oct 05, 2017 7:52 pm; edited 1 time in total
Back to top
View user's profile Send private message
Chewi
Developer
Developer


Joined: 01 Sep 2003
Posts: 856
Location: Edinburgh, Scotland

PostPosted: Thu Oct 05, 2017 7:50 pm    Post subject: Reply with quote

Tony0945 wrote:
With all these problems and AMD trying to sweep them under the rug and horror stories about the RMA process.

I haven't heard anything that bad. I said it took longer than expected but I can't complain other than that. You're in the US so it would probably be much quicker. The tracking showed my replacement went from Canada, to the US (Indiana), to France, to the Netherlands (redirected from their warehouse), to Germany, and finally to Scotland!

nasaiya wrote:
I have the ASRock AB350 K4 http://www.asrock.com/mb/AMD/Fatal1ty%20AB350%20Gaming%20K4/ -- upgraded to the latest bios, with ram that is on their support list and a 1600X.
I didn't care enough to research what c6 was or what 'cool and quiet' did so I disabled them and still so far so good... I have not tried gcc 7.2 yet,

EDIT: I'm also using march=znver1 for both kernel and everything else...

I think it's time to stop talking about GCC being a potential culprit. The RMA replacements have proven the segfaults were a hardware issue. Disabling ASLR in the kernel helped to avoid them but I think any reports around GCC versions or CFLAGS were largely anecdotal. I did see one bug in 7.1.0 only that was fixed in 7.2.0 but that just affected a test case in media-libs/leptonica and it wasn't even strictly Ryzen-specific.

I don't yet know whether the freezing is a hardware or software issue but if it is software then the kernel seems a much more likely culprit. Disabling C6 in the BIOS has worked here. Don't disable Cool and Quiet as that's unnecessary and, if I understand correctly, that controls CPU frequency scaling, which has a much bigger impact on heat and power usage. I'm going to check whether there are any similar reports on bugzilla.kernel.org and file one if not.
Back to top
View user's profile Send private message
Bloot
Tux's lil' helper
Tux's lil' helper


Joined: 10 Mar 2006
Posts: 90
Location: Barcelona

PostPosted: Fri Oct 06, 2017 12:19 am    Post subject: Reply with quote

mir3x wrote:
Tony0945 wrote:
With all these problems and AMD trying to sweep them under the rug and horror stories about the RMA process. I'm wondering if an FX-8350 isn't a better upgrade path?.

Bloot is selling RMAed ryzen :D, talk to him.

Tony0945 wrote:

Ryzen 7 is supposed to be a 95W processor but I see posts saying liquid cooling is neccesary and one professional review that cited 400W measured!

No, he has overclocked with GTX1080 or so, check normal videos/articles. I saw on youtube thats with games and similar FPS ryzen takes less watts than intel.

EDIT: check intel 8700K here, 8700 without K will cost the same as 1700X, and except eating a lot of Watts under load should be faster than 1700X
https://www.youtube.com/watch?v=EJOnwF8mgXc


Well, I received my replacement cpu yesterday, finally. After three long weeks waiting.

It's a 1733SUS and it does not segfault.

I don't know if I'll keep it or sell it, yet.
_________________
Portage 2.3.6 (python 3.4.5-final-0, default/linux/amd64/13.0/desktop/plasma, gcc-6.4.0, glibc-2.23-r4, 4.12.5-gentoo x86_64)
Back to top
View user's profile Send private message
nasaiya
Tux's lil' helper
Tux's lil' helper


Joined: 17 May 2007
Posts: 142

PostPosted: Fri Oct 06, 2017 7:07 am    Post subject: Reply with quote

Chewi wrote:

I think it's time to stop talking about GCC being a potential culprit. The RMA replacements have proven the segfaults were a hardware issue.

I wasn't implying gcc was responsible for the segfaults, only confirming that for whatever reason 7.1 doesn't seem to trigger them, obviously the problem is the chip.
_________________
If it ain't broke - fix it till it is!
Back to top
View user's profile Send private message
krinn
Watchman
Watchman


Joined: 02 May 2003
Posts: 5966

PostPosted: Fri Oct 06, 2017 8:49 am    Post subject: Reply with quote

Tony0945 wrote:
Meanwhile, my inquiries to AMD and Newegg result in, "No, we can't guarantee a post-week 25 CPU. They are 'probably' out of the queue and they only affected a very small number of users running Linux and only with unusual workloads (like using all the cores for compiling?)."

"No we cannot guarantee your car won't explode at 140km/h, it only affect people that do not respect the 130km/h max limit anyway".
:lol:
Back to top
View user's profile Send private message
Chiitoo
Moderator
Moderator


Joined: 28 Feb 2010
Posts: 1399
Location: Here and Away Again

PostPosted: Fri Oct 06, 2017 1:03 pm    Post subject: ><)))°€ Reply with quote

I've been quite eager to get new hardware for some time now, having issues with the motherboard (I believe) that actually hinder my audio set-up and stuff.

Most recently, I was waiting for certain bundles I heard (or read) of, where one could save some money getting a Vega, Ryzen, and a display of sorts, but since Vega is basically non-existent still, I'm probably not going to wait for that any longer (been using nVidia forever now, and apparently Vega is pretty good for gaming even in Linux use, so it would be very interesting to try it out).

The other day, I queried my supplier if they are able to tell if they have post 25-week Ryzens or not, and they actually only have pre 25... so here I am, waiting still. A Zen 2 (or Zen+) would be nice, but I probably can't wait for /that/ long...

Thank you all for the reports and discussion here. It has been a very interesting read indeed. :]
_________________
Kind Regards,
~ The Noob Unlimited ~

Sore wa sore, kore wa kore.
Back to top
View user's profile Send private message
Tony0945
Advocate
Advocate


Joined: 25 Jul 2006
Posts: 2038
Location: Illinois, USA

PostPosted: Fri Oct 06, 2017 1:13 pm    Post subject: Reply with quote

krinn wrote:
"No we cannot guarantee your car won't explode at 140km/h, it only affect people that do not respect the 130km/h max limit anyway".
:lol:
Not an apt analogy.

More like "No, we can't guarantee that your Pinto's gas tank hasn't been fixed so won't explode at a 5mph collision, but it only affects people that are involved in rear end collisions and those are rare.

They seem to be dismissing Linux users and making light of their problem and not explaining. Like polticians that don't understand that the cover up is worse than the original error.
Back to top
View user's profile Send private message
krinn
Watchman
Watchman


Joined: 02 May 2003
Posts: 5966

PostPosted: Fri Oct 06, 2017 1:44 pm    Post subject: Reply with quote

Tony0945 wrote:
krinn wrote:
"No we cannot guarantee your car won't explode at 140km/h, it only affect people that do not respect the 130km/h max limit anyway".
:lol:
Not an apt analogy.

More like "No, we can't guarantee that your Pinto's gas tank hasn't been fixed so won't explode at a 5mph collision, but it only affects people that are involved in rear end collisions and those are rare.

They seem to be dismissing Linux users and making light of their problem and not explaining. Like polticians that don't understand that the cover up is worse than the original error.


I don't like yours too :lol:
With mine, i point that you are not using the car in special conditions, the car is sold to goes at least 140 or more km/h.
While with yours, it would mean an exception (a collision).
The cpu is not faulty because you have explode your m/b or something bad, the cpu is faulty by just using it in normal conditions, for a cpu, using all cores and busy them to 100% is not something unusual.

It's clear for everyone that linux and gcc just emphasis the problem, in no way it create the problem.

I was thinking amd was doing it right (replacing faulty cpu).
I was more thinking you had that stupid answer from newegg than amd, i was wrong?
Back to top
View user's profile Send private message
Chiitoo
Moderator
Moderator


Joined: 28 Feb 2010
Posts: 1399
Location: Here and Away Again

PostPosted: Fri Oct 06, 2017 4:44 pm    Post subject: Reply with quote

krinn wrote:
It's clear for everyone that linux and gcc just emphasis the problem, in no way it create the problem.

Not everyone...

My “supplier” for example mentioned that the problem is specific to Linux, though where they got that idea from, I don't know. >.<
_________________
Kind Regards,
~ The Noob Unlimited ~

Sore wa sore, kore wa kore.
Back to top
View user's profile Send private message
Tony0945
Advocate
Advocate


Joined: 25 Jul 2006
Posts: 2038
Location: Illinois, USA

PostPosted: Fri Oct 06, 2017 5:27 pm    Post subject: Reply with quote

krinn wrote:
I was more thinking you had that stupid answer from newegg than amd, i was wrong?

Newegg and the AMD forum. I think that the AMD website downplayed it too.
Back to top
View user's profile Send private message
lonex
Tux's lil' helper
Tux's lil' helper


Joined: 22 Aug 2004
Posts: 105
Location: Karlsruhe, Germany

PostPosted: Fri Oct 06, 2017 7:21 pm    Post subject: Reply with quote

With all the tips that I've found in this thread, I now can compile anything but one package version without any segfaults appearing. They never happen, unless I try to compile gcc-7.2.0 - and there with a probability of 100% at a random part during the compilation.
Any other gcc version works fine. Anyone else experiencing the same thing?
_________________
There are no dumb questions, just dumb answers. So please help users that are new to linux/gentoo by answering unanswered questions.
Back to top
View user's profile Send private message
thumper
Guru
Guru


Joined: 06 Dec 2002
Posts: 508
Location: Venice FL

PostPosted: Fri Oct 06, 2017 8:33 pm    Post subject: Reply with quote

lonex wrote:
With all the tips that I've found in this thread, I now can compile anything but one package version without any segfaults appearing. They never happen, unless I try to compile gcc-7.2.0 - and there with a probability of 100% at a random part during the compilation.
Any other gcc version works fine. Anyone else experiencing the same thing?


You would not by chance be using dev-libs/boost-1.63.0?

I tried it on Wednesday ( not sure why now ) and Thunderbird would randomly segfault during the compile 100% of the time, I went back to dev-libs/boost-1.62.0-r1 today and it worked fine again.

George
Back to top
View user's profile Send private message
Naib
Watchman
Watchman


Joined: 21 May 2004
Posts: 5078
Location: Removed by Neddy

PostPosted: Fri Oct 06, 2017 8:42 pm    Post subject: Reply with quote

Well I have a system that builds everything and is stable so I am going to start trying to break it. I have just enabled LTO and recompiling gcc before @system --> @world
_________________
The best argument against democracy is a five-minute conversation with the average voter
Great Britain is a republic, with a hereditary president, while the United States is a monarchy with an elective king
Back to top
View user's profile Send private message
stephan-t
Tux's lil' helper
Tux's lil' helper


Joined: 12 May 2014
Posts: 118

PostPosted: Sat Oct 07, 2017 8:08 am    Post subject: Reply with quote

Dear Topic population :)


I need a good, rock stable, very good equipped and excellent built quality (PCB, capacitors, audio, etc.) motherboard for R5 1500x to 24/7 without failure..

Where I find the recommended MB list?
Back to top
View user's profile Send private message
lonex
Tux's lil' helper
Tux's lil' helper


Joined: 22 Aug 2004
Posts: 105
Location: Karlsruhe, Germany

PostPosted: Sat Oct 07, 2017 11:17 am    Post subject: Reply with quote

thumper wrote:
You would not by chance be using dev-libs/boost-1.63.0?

I tried it on Wednesday ( not sure why now ) and Thunderbird would randomly segfault during the compile 100% of the time, I went back to dev-libs/boost-1.62.0-r1 today and it worked fine again.

George


True, I actually was on boost-1.65.0 ... tried reverting it to 1.62.0-r1 but to no avail, still not able to finish gcc-7.2.0.
Thanks for the tip, though.
_________________
There are no dumb questions, just dumb answers. So please help users that are new to linux/gentoo by answering unanswered questions.
Back to top
View user's profile Send private message
mattcode
n00b
n00b


Joined: 06 Oct 2017
Posts: 3

PostPosted: Sat Oct 07, 2017 11:38 am    Post subject: Reply with quote

Hi all, I'm using the Ryzen 1700 in a Gigabyte GA-AB350-Gaming motherboard with the latest BIOS revision (AGESA 1.0.0.6b) and I'm still hitting segfaults all the time whenever I try to compile anything big :(

I'm using kernel 4.12.12-gentoo and GCC 7.2.0 with -march=znver1. Happens with my small overclock and also at the stock clocks/voltage.

I've read a few different posts here and on other sites and I'm still confused: will AMD swap this CPU for one that doesn't segfault every 5 minutes? Thanks
Back to top
View user's profile Send private message
NeddySeagoon
Administrator
Administrator


Joined: 05 Jul 2003
Posts: 39294
Location: 56N 3W

PostPosted: Sat Oct 07, 2017 11:43 am    Post subject: Reply with quote

mattcode,

After you jump through the AMD RMA hoops, since they want to make sure its not something else in your system ... probably.
_________________
Regards,

NeddySeagoon

Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
Back to top
View user's profile Send private message
mattcode
n00b
n00b


Joined: 06 Oct 2017
Posts: 3

PostPosted: Sat Oct 07, 2017 12:05 pm    Post subject: Reply with quote

NeddySeagoon wrote:
mattcode,

After you jump through the AMD RMA hoops, since they want to make sure its not something else in your system ... probably.


Thanks, I'll fill in the RMA form and see what happens. Hopefully doesn't take too long.
Back to top
View user's profile Send private message
wrc1944
Advocate
Advocate


Joined: 15 Aug 2002
Posts: 3105
Location: Gainesville, Florida

PostPosted: Sat Oct 07, 2017 7:28 pm    Post subject: Reply with quote

FWIW, here's my RMA experiences:
My first email form request was sent Sept.18, but got no response (not sure why, but suspect it was that I foolishly didn't include my actual Ryzen 1700 serial number, but just mentioned UA 1707SUT), so I resubmitted. Here's my basic input to the AMD email form RMA request:
Quote:
I apparently need an RMA for my Ryzen 1700 (UA 1707SUT and serial number 9GU4895N70046) which has segfaults, random reboots & lockups.
No thermal issues, 16GB G.Skill FlareX 3200Mhz (samsung B-Die) RAM. ASrock x370 Gaming K4 MOBO. Multiboot system (two Gentoo testing installations), with Linux Mint, Arch, ubuntus, and windows 7 on different partitions. I've run Gentoo testing for 15 years exclusively on AMD systems. No bios updates or settings modifications have eliminated segfaults.

To hopefully save some time, please find an attached detailed PDF summary of my extensive testing.

I've run extensive weeks-long very informed testing on this system with no resolution for the problems, and unfortunately have concluded my only option is to request an RMA replacement.
(I submitted an RMA request on Sept.18, but never heard back by email, so am re-sending.)

Timeline so far:
Sept. 30 (a saturday): Sent the new request, along with detailed 1 page testing summary pdf attachment. 60 seconds later I received 2 Noreply@amd.com emails, the first said "We have created a Customer Care Support account based on the details you submitted with your service request". The second one contained my service request reference number.

Oct. 4 (wed.) 2 more emails, the first from TECH.SUPPORT that said: "Please find the attached warranty claim form and fill all the mandatory fields, get back to so that I can process your warranty request." It was an excel file, into which I entered the required info (just simple name, address, etc. stuff- no technical questions), and sent back via email reply with the updated excel file attachment. Then, the second email was simply an acknowledgement they received my reply.

Oct. 5: # more emails: Email 1 from AMD Support said: "The following processor(s) have been authorized for return to AMD
against the RMA number specified above. Please note that we cannot process any parts other than those marked as approved in this RMA confirmation email."

Product Serial# Result
YD1700BBAEBOX 9GU4895N70046 APPROVED

Email 2 said: "Your RMA has been processed and the confirmation number is: RMA# etc...

Please securely package your CPU in box with the RMA confirmation number (RMA# etc.... and return shipping address clearly written on the outside of the box.

The shipping address is:
AMD/ ModusLink Global Solution Centre
11010 NW 92nd Terrace
Miami, Florida 33178
Tel: 1-305-803-0916

Note: Please keep the heatsink and fan as it is not required to send this back to AMD.
I am currently arranging a pre-paid FedEx return label to return your CPU back to AMD free of charge - It will be emailed to you shortly.
Please print the label and attach it to the return package and arrange pick up or drop off through your local FedEx service centre. As soon as I receive your confirmation that your CPU is in transit, I will have the replacement part shipped out immediately. Once you have received your replacement CPU, please update your motherboard BIOS to the latest version with AGESA 1.0.0.6b after installing the CPU.

Email 3 was the FedEx link to print the prepaid label. "AMD C/O ModusLink has provided you with a return label for the shipment below." Also provided was a user ID and password in case of any problems with fedex. When I printed it out, it generated a FedEx tracking number.

I needed to finish up some work, so didn't ship it out Oct. 5, but could have.

Oct.6: Shipped it out early morning at the FedEx shipping center, and then emailed AMD tech.support the tracking number along with a scanned copy of the fedex receipt for confirmation I had shipped the processor. They immediately acknowledged my email. I checked fedex tracking online later that night, and my package was already in transit to Miami.

Oct.7: Checked fedex tracking again mid morning, and my package was already in Miami on the truck out for delivery. WoW! Fast service.

Apparenty, since my RMA was approved "no questions asked" and no more testing my cpu, sending them a detailed summary of what I had already done was enough for them to go ahead and issue an approved RMA for that exact processor. Guess I'll be getting my new RMA cpu next week sometime, but I must say the AMD RMA process was very satisfactory, at least in my case, and I have no complaints. It is a weekend, so hopefully they will ship it out Monday. Anyway, it's hard to see how it could have gone any better. Of course the proof will be in if the new cpu solves my problems. I'll post back with results.

Oct.11: My package actually arrived and was signed for by AMD-Miami on Oct.9 (Mon.). It's looking like AMD RMA service takes weekends off, but today I got an email response, which I guess is pretty good turn-around, considering. :
Quote:
This is confirmation notice from AMD Global Customer Care with regards
to Warranty Request RMA.

Product Serial# Result
______________________________________________________________

YD1700BBAEBOX 9GU4895N70046 RETURN RECEIVED

This is to confirm receipt of your defective product which you shipped
back to us as part of the Warranty Replacement.

Next Steps

Your returned product will now undergo Visual Mechanical inspection and
you will receive an email confirming the results shortly.

Best Regards,
AMD Global Customer Care

RMA Process Information

Visual Mechanical Inspection:
Please note that your processor(s) will be subject to V/M Inspection
upon arrival at the AMD Returns Center, and must pass this inspection
before your warranty claim can be approved. Should your processor fail
V/M inspection, a Customer Support Analyst will be in contact with you
by telephone or email with this information. You will receive your
original processor approximately 3-5 business days from the ship date
from the AMD Returns Center, depending upon your location.
For information on AMDs “Processor in a Box” limited warranty please
click http://www.amd.com/us/warrantyinfo.

Once you have received this notification email and your processor(s)
pass the VM inspection, it usually takes approximately 2 business days
for your replacement part to ship. You will receive your
replacement(s) approximately 3-5 business days from the ship date from
the AMD Returns Center, depending upon your location.


For reference, here's a copy of my one page summary.
Quote:
Summary of my extensive testing on my Ryzen 1700 (UA 1707SUT) which has segfaults while compiling, and random reboots & lockups even while in idle. Ryzen was from Newegg on 4/20/17, but had to wait till July for my preferred RAM and MOBO BIOS updates to build the system.

1. No thermal issues. Three 120mm case fans. System is located in AC environment, case side is aways off, with small desktop fan providing constant fresh cool air. My board uses the nct6775 sensor module, which works well on linux with gkrellm. Temps during heavy compiling never exceed 42-48C. Idle is around 35-38C. This is similar to my AM3+ FX 8320 system, with the it87 sensor chip.

2. Being a Gentoo testing user, my main tests were of course extensive normal compiling, and centering on a fresh compiling of GCC and Mesa for every setting combination which was tried, which all resulted in segfaults. All tests have been run with many kernel versions (4.11.x, 4.12.x, 4.13-rc’s, 4.13.x). Also, several BIOS updates, currently using AGESA 1006a.

3. Additionally, I’ve disabled/enabled ASLR, Opcache, SMT, either individually or all at once, or in various combination. Also tried different make_opts settings including -j1, -j2. -j8. or -j16 with no improvement. Tried carefully boosting RAM and /or CPU voltages as recommended by AMD, but still got the segfaults while compiling gcc or mesa.

4. All tests were run with default BIOS settings with or without XMP enabled. I’ve tried the recommended AMD requested tests reported by those RMA posters on the extensive community.amd.com Ryzen segfault thread, some reddit AMD segfault threads, and the Gentoo Ryzen thread where I post. This also included the “cpu Vcore voltage increase in .0.05 increments” routine, and SOC set to fixed 1.1v, which did not help the segfault issues. Also have tried kernel options CONFIG_RCU_NOCB_CPU - y and RCU_NOCB_CPU_ALL, plus march=native, znver1, or even generic x86-64, all of which made no difference.

Unfortunately, I can’t run my high-end Ryzen Samsung B-Die G.Skill RAM over 2400mhz, and worse yet must disable ASLR, Opcache, SMT, and use at most -j8 to run the system without random rebooting or lock-ups. However, I do still have reboots and lockups even on Mint, Arch, and Ubuntu, and no bios tweaking eliminates the segfaults while compiling on my main Gentoo installs.

I built this system to provide a rock solid Gentoo testing/compiling platform, and for Professional Audio muti-track recording and composing production (requiring low-latency), so reliable operation and data integrity are essential.

I must say, as a 15 year exclusively AMD system user and system builder plus a huge AMD fan, I’m very surprised after weeks of unsuccessfully troubleshooting this problem that it appears I must RMA my Ryzen 1700, as my current UA 1707SUT presents various unacceptable problems.

_________________
Main box- AsRock x370 Gaming K4
Ryzen 1700, 3.0GHz, 16GB GSkill Flare DDR4 3200mhz
Samsung SATA 1000GB, Radeon HD R7 350 2GB DDR5
Gentoo ~amd64 plasma, glibc-2.26-r3, gcc-7.2.0- kernel-4.14.5 gentoo USE=experimental
Back to top
View user's profile Send private message
Display posts from previous:   
Reply to topic    Gentoo Forums Forum Index Gentoo Chat All times are GMT
Goto page Previous  1, 2, 3, 4, 5, 6, 7  Next
Page 5 of 7

 
Jump to:  
You cannot post new topics in this forum
You cannot reply to topics in this forum
You cannot edit your posts in this forum
You cannot delete your posts in this forum
You cannot vote in polls in this forum