sciff
Member
- Mar 6, 2017
- 136
- 52
- 71
After you subtracted 4 there (changing 0C to 08), it's also important to step back 3 bytes and add 4 there. That way the checksum stays the same. It's the method suggested at Win-raid forums, I've used it numerous times for different people, it works fine!
Understood...
I did that and UBU ran without issue:
Original DWORD:
93 AA 03 0C
NEW:
97 AA 03 08
Again, that got me past the "error writing file" as far as UBU was concerned, everything went great.
Trouble is the Z10PE-D16-WS is spitting out all attempts to use flashback to load bios images that report "failed integrity check" in the normal EZflash tool. That is, the one I edited and the 2 posted tonight all produce the same result:
1. plug into flashback port
2. press flashback button
3. 5 blinks
4. solid blue... (indicating it is NOT loading the flash as I understand it).
I've now successfully flashed an unedited 3305 and 3405 image tonight (though through ez-flash not flashback)...
I used flashback with this board when I first got it without issue (to load BW support to boot 2960v4 - those are out right now and 2696v3 in their place). So, I think I'm doing it right, but...
EDIT: Ok, it looks like it is a USB compatibility issue. I copied the UBU output (APITO work-around + UBU) to a new USB drive and it appears to be working. Just a normal fat32 4G drive (that isn't working)...
Last edited:
sciff
Member
- Mar 6, 2017
- 136
- 52
- 71
That's correct
Why not? That way you would know if it's your USB stick that is the problem or something else.
Great!
Z10PE-D16-WS 3403 + APTIO work-around now results in N/A uCode in bios.
50mv 0x39 .EFI hung during linux boot... trying another...
50mv 0x39 .EFI hung during linux boot... trying another...
Hangs hard at the same spot (setting host-name) every time with
v3x2_50_x39_vcc.efi
V3x2_cup27.efi
Keyboard unresponsive, have to hold power button to turn it off. You guys with this board setting anything in the BIOS? Everything is default.
Last edited:
This is pretty similar to what I saw on windows boot when I tried (same results with no micrcode efi driver also)
I've tried all the various .EFI files throughout this thread (all the V3x2_xxxxx.EFI anyway). I also tried no .EFI file and as you say, it behaves the same whether I have one or not.
I can confirm that the various .EFI load and print their messages (without error). I can also confirm that the uCode reported in the bios is N/A as it should be and 0x38 with the "stock" BIOS.
So, unlike the single processor boards, I'm completely unable to boot at this point. Guess, I'll try more iterations this weekend.
What I have working on the single taichi boards is an older version of v3_1.efi with the linux kernel loading 0x39 via initrd. That's been very stable.
Did you disable C3/C6 states in BIOS ?
My PC (EP2C612WS 2x2686v3) works fine in Win/Lin with v3x2_cp39.efi only when C6 disabled.
But when uncore is forced ("cup" version of efi driver) C3 must be disabled too.
New EFI driver (without -50mV offset) is strange: was x30 after first reboot, no uCode, new record for 2686 v3. Uncore drops from x30 to x20, and after injecting 039 uCode VMWare driver max multiplier drop to x28‒29, and stay there after cold reboot. 027 uCode does the same. Reverting to driver with offset.
On one hand, I'm sorry to hear that you were having trouble. On the other hand, I'm relieved to know I'm not the only one. I'll keep you updated on any attempts I might make (I use my system for work, so I don't mess with it unless I can afford an hour to reload a virgin bios)
I didn't do this, but perhaps that's part of the issue.
What downstream effect does disabling C6 and/or C3 have?
I did (had been following the thread on that issue) but I'll retry the experiment this weekend.
sciff
Member
- Mar 6, 2017
- 136
- 52
- 71
Guys, on dual-CPU motherboards, disabling either C6 or both C6 and C3 is essential, as noted above by Welsper.
timk1980, C3/C6 are deep sleep states, allowing the CPU to run on very low power. Since sleep and hybernation modes have to be disabled anyway in order for the EFI exploit work correctly, there's no much point in C3/C6 being enabled.
At least that's how I understand the issue.
timk1980, C3/C6 are deep sleep states, allowing the CPU to run on very low power. Since sleep and hybernation modes have to be disabled anyway in order for the EFI exploit work correctly, there's no much point in C3/C6 being enabled.
At least that's how I understand the issue.
@Welsper :
@randir :
I would be highly interested in your sources as a starting point, however the link posted earlier is not working anymore.
I would gladly participate in your development and github project.
I own a 2696v3 R2 and Asrock X99E ITX... so far I got 3.2 Ghz allcore, however only light loads. Even non AVX loads that are a bit more intense and my multi drops, due to I believe power restrictions.
I need to find out more.
Anyway if you would reupload your sources i would be very happy indeed.
Chris
Hi,
I also have a X99E-ITX/ac with a E5-2683 v3.
Can someone share the best and most stable steps for this configuration incl. sharing the required files?
Thank you kindly,
Ray
Your CPU is more expensive, than mine, yet you have only 3.2 GHz (I have stable 3.0 with 3.5 for up to 10 cores with balanced power plan, 2.7 under very heavy AVX2). Seems to be motherboard restriction, really... Right now using new EFI driver with -50 mV, uCode 039 through VMWare, i believe this is the best combination for my MSI X99A Raider (but no uCode may give you better results, or you can use this version with built-in 039 uCode). Pinned GPU driver restart bug - MSI Afterburner somehow forgot +24 mV setting, so now i'm 100% stable.
Hi everybody!
And great thanks for a lot of useful info!
My first post on this forum)
I own and tested both Asus Z10PE-D8 WS + pair of 2683v3 OEM (results)+ pair of 2695v3 Pre/QS (results) and
Z10PE-D16 WS with pair of 2696v3 ES (C0/C1) (results)
(win 10 x64 Pro)
Z10PE-D8 WS +2683v3 and 2695v3 Pre/QS - works like a charm - stable and similar with results mentioned here
Z10PE-D16 WS + 2696v3 ES - works only with V3x2_cp.efi and V3x2_cup.efi - stable and similar with results mentioned here. Other efi drivers freezes during loading (it could set +38 for all cores for 1 CPU and freeze during setting second... etc...)
But my main problem and strangeness is - TDP.
It can't rise above Nominal (120W for 2683 and 2695 and 145W for 2696)
Of course I did try almost everything with BIOS settings and EFI driver MC combinations.
Respectively the performance is not as would expected/wanted.
I guess it could be some reasons for it -
1. MB is locked on nominal TDP
2. Efi driver does set the ratio but can't set power limit.
3. Some BIOS settings are missed by me.
And great thanks for a lot of useful info!
My first post on this forum)
I own and tested both Asus Z10PE-D8 WS + pair of 2683v3 OEM (results)+ pair of 2695v3 Pre/QS (results) and
Z10PE-D16 WS with pair of 2696v3 ES (C0/C1) (results)
(win 10 x64 Pro)
Z10PE-D8 WS +2683v3 and 2695v3 Pre/QS - works like a charm - stable and similar with results mentioned here
Z10PE-D16 WS + 2696v3 ES - works only with V3x2_cp.efi and V3x2_cup.efi - stable and similar with results mentioned here. Other efi drivers freezes during loading (it could set +38 for all cores for 1 CPU and freeze during setting second... etc...)
But my main problem and strangeness is - TDP.
It can't rise above Nominal (120W for 2683 and 2695 and 145W for 2696)
Of course I did try almost everything with BIOS settings and EFI driver MC combinations.
Respectively the performance is not as would expected/wanted.
I guess it could be some reasons for it -
1. MB is locked on nominal TDP
2. Efi driver does set the ratio but can't set power limit.
3. Some BIOS settings are missed by me.
I have a dual boot system Windows/Ubuntu, is the EFI driver also working on Ubuntu then? (might be a stupid question)
Non-WS version does not have this problem, i don't have one, but owners of Z10PE-D16 reporting no such troubles. Very strange.
Yes.
I've Win/Lin too: Win is "linked" to UEFI, Ubuntu isn't. Efi-driver is loaded from win-efi partition.
Just use EFI-driver with microcode update or disable microcode setting as in lin as in win if you prefer _no_microcode version.
Same problem. On the load: CPU has about 136W during 56 sec then falls to 120W.
A higher value (144W) is set into "Limit 1" in the EFI-driver, but looks like it is ignored. And I've no idea yet how to come round main TDP value (which one isn't limited by time).
Last edited:
Can you link to specifically which BIOS and EFI you used for the D16-WS? also, what bios settings did you tweak? Thanks!
For my part, I started with 3403 - made an edit to work around the APTIO issue (described in this thread) and then ran ubu. This succeeded and produced ucode version "N/A" in the bios.
I disabled C3 and C6 - made no other bios change
I then tried all the V3x2_xxx.efi I could find in this thread (including _cp, _cup, _vup32, _cup27, cp27, cp39, and the 50/vcc versions as well)
I haven't tried the other 2 BIOS posted to this thread recently yet. Result was always the same and the same as if I did not run the .efi at all... linux gets to "setting hostname" and hangs. power/reset unresponsive. Same exact place each time.
In all cases, the banner of the .efi file appeared and reported "success" in the various ways that each of them do...
You said that v3x2_cp and v3x2_cup both worked, though? I'm debating giving this another try.
No, none have worked for me at all on the Z10PE-D16-WS... all hang the same way including no .efi at all...
Ah, bummer, I misunderstood then. If I should mess with it and get something working, I'll definitely report. But, sounds like it may just be a no-go.
TRENDING THREADS
-
Discussion Intel current and future Lakes & Rapids thread- Started by TheF34RChannel
- Replies: 23K
-
Discussion Zen 5 Speculation (EPYC Turin and Strix Point/Granite Ridge - Ryzen 9000)- Started by DisEnchantment
- Replies: 9K
-
-
Discussion Intel Meteor, Arrow, Lunar & Panther Lakes Discussion Threads
- Started by Tigerick
- Replies: 7K
-
Facebook
Twitter