Has anyone tested this to verify booting without nag?
1.) How was address 98B5A determined to be correct to edit the Platform from 11 to 44?
I see references to 1067A (7A 06 01 00 in Little Endian) at B2F80-B2F83 and 1E72B2-1E72B5
2.) Has anyone tried this same method to change Platform for the 10676 microcode to enable 5400 series Xeons?
I compared the original BIOS to this modded one and other than the one byte edit for Platform, I only see changes at 1F0054-1F0057 which I assume is just the CRC fix to the HDR so that DCCU can make a flashable BIOS
Thanks in advance for any help.
Posts: 20
Threads: 1
Joined: Oct 2014
Reputation:
5
(10-29-2014, 07:39 AM)doyet Wrote: Has anyone tested this to verify booting without nag?
Yes, personally flashed on my own machine.
(10-29-2014, 07:39 AM)doyet Wrote: 1.) How was address 98B5A determined to be correct to edit the Platform from 11 to 44?
I see references to 1067A (7A 06 01 00 in Little Endian) at B2F80-B2F83 and 1E72B2-1E72B5
That area of binary contains all microcode headers in Dell's format. The start to finish of each header was noted, and the byte storing 01067A platform was modified.
(10-29-2014, 07:39 AM)doyet Wrote: 2.) Has anyone tried this same method to change Platform for the 10676 microcode to enable 5400 series Xeons?
This is a chipset (hardware) limitation. 5400 E0 use 1067A platform 2,6 microcode, the same as 3300 E0. There are earlier C0 processors that use 10676, but again, Q45 chipsets do not support multiprocessor Xeons, even in single processor configurations. O780 A15 already contained 10676 microcode for platform 0, platform 2, platform 4, & platform 6 and should natively support single processor C0 Xeons.
(10-29-2014, 07:39 AM)doyet Wrote: I compared the original BIOS to this modded one and other than the one byte edit for Platform, I only see changes at 1F0054-1F0057 which I assume is just the CRC fix to the HDR so that DCCU can make a flashable BIOS
Yes, it was really that easy. 1067A microcode is identical for platforms 0,4 (11000000) and platform 2,6 (44000000).
Posts: 8
Threads: 0
Joined: Oct 2013
Reputation:
0
DELL header format- ugh. I have pulled my hair out trying to figure that out. Hours comparing 771 Dells BIOS to the related 775 DELL BIOS. (Precision 390 to 490, T3400 to T5400) Is it still a full 48 bytes as in the original microcode?
As for the 5400 series Xeons not booting on X and Q chipsets-
I gather that people believe that VT-D is the problem. If you managed to turn off VT-D in Setup then BIOS still goes to DEFAULT when it detects new Processor... which re-enables VT-D.
1.) I was curious if there was a way to maybe disable VT-D by default with a hex edit of the BIOS.
Of course it may be as simple as "The Hardware Won't Do It" and I can give up.
I tried to get DELL Precision 490 and 690 to use 5400 series Xeons.... they both had Dual Processors and 5000x chipsets (Non-DELL 5000x machines released Xeon 5400 support)
Thanks again for the help. HAPPY HALLOWEEN!!
Posts: 20
Threads: 1
Joined: Oct 2014
Reputation:
5
(10-31-2014, 08:51 AM)doyet Wrote: DELL header format- ugh. I have pulled my hair out trying to figure that out. Hours comparing 771 Dells BIOS to the related 775 DELL BIOS. (Precision 390 to 490, T3400 to T5400) Is it still a full 48 bytes as in the original microcode?
Dell's is much shorter, and its length depends on the bios. Intel's platform is 4 bytes, where Dell's in 1 byte.
(10-31-2014, 08:51 AM)doyet Wrote: As for the 5400 series Xeons not booting on X and Q chipsets-
I gather that people believe that VT-D is the problem. If you managed to turn off VT-D in Setup then BIOS still goes to DEFAULT when it detects new Processor... which re-enables VT-D.
VT-d is disabled by default on the two OptiPlex computers I have here. I don’t have the 780 here anymore, but I’d bet it’s the same.
(10-31-2014, 08:51 AM)doyet Wrote: 1.) I was curious if there was a way to maybe disable VT-D by default with a hex edit of the BIOS.
The VT-d setting should just be stored as a single bit, but if it is off by default, this isn’t the issue.
(10-31-2014, 08:51 AM)doyet Wrote: I tried to get DELL Precision 490 and 690 to use 5400 series Xeons.... they both had Dual Processors and 5000x chipsets (Non-DELL 5000x machines released Xeon 5400 support)
These Precision models are probably limited to 65nm CPUs. I was going to inspect the microcode, but apparently it won’t extract on 64-bit.
(10-31-2014, 08:51 AM)doyet Wrote: Thanks again for the help. HAPPY HALLOWEEN!!
No problem. Happy Halloween!
Posts: 8
Threads: 0
Joined: Oct 2013
Reputation:
0
cerverappz ,
You said microcodes won't extract for Precision 490/690... are you working with individual modules? Please let me know if there are better tools than the old HxD to use.
I just use microdecode on the intel .tgz to get microcode. Then skip first 48 bytes of whichever I am looking for (1067Ah for example) then do a search of the BIOS.HDR for the microcode. If the existing code is an older version then I can update it by replacing microcode and fixing CRC.
I still had no luck when trying to insert an entire family of processors into BIOS. Such as when 10676h is not present in BIOS.
Also, do you have any links to info on DELL BIOS editing?? ( still don't know how you found the address for Platform ) I may want to try a Xeon X3363 in a Precision T3400. They already support the lga775 variants like x3380 Xeons.
Posts: 20
Threads: 1
Joined: Oct 2014
Reputation:
5
(11-09-2014, 10:44 PM)doyet Wrote: You said microcodes won't extract for Precision 490/690... are you working with individual modules?
-writehdrfile wouldn't work on x64 for that bios. If it is too old for a 64-bit flash utility, it's likely too old for 45nm.
(11-09-2014, 10:44 PM)doyet Wrote: Please let me know if there are better tools than the old HxD to use.
I almost exclusively use a hex editor, especially on Dell bioses.
(11-09-2014, 10:44 PM)doyet Wrote: I just use microdecode on the intel .tgz to get microcode. Then skip first 48 bytes of whichever I am looking for (1067Ah for example) then do a search of the BIOS.HDR for the microcode. If the existing code is an older version then I can update it by replacing microcode and fixing CRC.
I still had no luck when trying to insert an entire family of processors into BIOS. Such as when 10676h is not present in BIOS.
I wouldn't update the microcode contents unless it's from another Dell bios with the same header structure so the header can be updated as well.
(11-09-2014, 10:44 PM)doyet Wrote: Also, do you have any links to info on DELL BIOS editing??
Unfortunately, most of what I've done has been on my own.
(11-09-2014, 10:44 PM)doyet Wrote: ( still don't know how you found the address for Platform )
Go to the byte I edited, and mess with the columns in your hex editor to see the pattern of headers. For example, 0-9 or 0-F and any other ranges. Hopefully that will help a little.
(11-09-2014, 10:44 PM)doyet Wrote: I may want to try a Xeon X3363 in a Precision T3400. They already support the lga775 variants like x3380 Xeons.
You should be fine since the T3400 bios natively contains microcode for E0 Xeon processors. 3300s should work great, but 5400 likely won't due to the X38 chipset.
Posts: 1,822
Threads: 1
Joined: Dec 2013
Reputation:
60
(11-09-2014, 10:44 PM)doyet Wrote: cerverappz ,
You said microcodes won't extract for Precision 490/690... are you working with individual modules? Please let me know if there are better tools than the old HxD to use.
I just use microdecode on the intel .tgz to get microcode. Then skip first 48 bytes of whichever I am looking for (1067Ah for example) then do a search of the BIOS.HDR for the microcode. If the existing code is an older version then I can update it by replacing microcode and fixing CRC.
I still had no luck when trying to insert an entire family of processors into BIOS. Such as when 10676h is not present in BIOS.
Also, do you have any links to info on DELL BIOS editing?? ( still don't know how you found the address for Platform ) I may want to try a Xeon X3363 in a Precision T3400. They already support the lga775 variants like x3380 Xeons.
hi~~guy
The latest version bios a14 for dell t3400,it already contain stepping e0 (both 771 and 775 as date 2010),and it also include fully micro code for c0(0676) so you don't need to modified micro code for it.
01-22-2017, 11:59 AM (This post was last modified: 01-07-2018, 08:31 AM by DeathBringer.)
Hey everybody,
I have here an Dell Optiplex 780 in the smallest version (USFF). I want to upgrade it to an Quad-Core Processor, which is compatible to the LGA and chipset. It's the Intel Core 2 Quad Q9400, which is also supported by the big brothers from the USFF model (shown here). When I turn on my upgraded system I get the following error "Alert! Incompatible processor detected. System halted!" My friend got the same error. He tried to upgrade the same Optiplex with the Intel Core 2 Quad Q6600. Our current processor is the Intel Core 2 Duo E7500.
Is here an BIOS-Mod possible to support Quad-Core-CPUs? Can anybody help us? Many thanks in advance!