Hi Andy & all members: First of all thanks to all that make possible this amazing forum. I succeeded in modding an Asus Mobo P8P67 LE Bios 3602 with Phoenix Tool v. 2.04, but having a look to the ACPI tables with Aida64, something strange appears as in the image attached you can see. View attachment 15558 Any ideas?
Are u sure this isn't present in the unmodified EFI Could you post the same screenshot from the unmodded EFI? Andy
Actually my Asus X53SC, which uses AMI Aptio EFI, also have some strange entries in ACPI tab in AIDA64 (for both modded & unmodded). But if viewed by RW Everything, then all things are normal. @andyp Just out of curiosity, I want to modify Virtualbox EFI (present in the installation folder in *.fd form). But it failed; most probably no compression method found in it & all methods result errors. My laptop isn't with me, so I can't tell you the exact error message.
Its not an error. Look at the structure, it parses some of it But doesn't parse the last FV thats embedded - just says unknown section Will look Andy
qtm mentioned to flash the bios with USB flashback. As far I know in flashback, there is no security checking.
I really appreciate this. I have been working on this all day and night. My last effort has been comparing the original 02_2.rom and the modified one from the tool. I don't know much about hex code, but I can pick out differences and trial by error switching things back one peace at a time to see if it fixed it. This is infinity complicated by the fact that switching some options in the tool will still break Speed Step, but will also no longer validate W7. And I'm not sure which those are yet. And I haven't even taken a look at your custom mod yet... not that I would really understand it. But I am in a very tired state and that is an endeavor to be done while awake. I can confirm that it is not the SLIC table insertion that is causing the issue... but changing the table ID(s)... but I'm guessing you already knew that. --- One thing I did notice...just now, that may be of help. In the RWeverything when speed step is enabled there are multiple SSDT tables in the ACPI menu. But, when it is disabled, there is only one. So, I made for you two RW reports with the unmoded bios. One with speed step enabled and another with speed step disabled. And as you requested here is also the RW report following your mod. It is with speed step disabled, as that is the only way the machine will boot. --- Thank you again, for taking the time to look into this. View attachment 15591
I'm pretty sure its the missing SSDT tables that are causing the issue. Edit: Puzzled as to why when speedstep off the SSDT aren't there. The mod did what I wanted - replace the MCFG table with SLIC. I will upload some test BIOSes tonight to see if we can pin down the issue. Thanks Andy
I don't think the table is the issue. I think it is changing of table IDs. I noticed some changes in the binary of 02_2.rom that were after the 13 character bios tables OEM IDs. I do not know if this is normal operation. If it is, I wonder if this gets into coded areas.
Also, I have noticed that if you select "no slic" the program does not make any changes (such as changing table OEM/Table IDs to match selected slic. This would be useful if you had to manual add the slic with the manual module option selected). It seems to just puts the bios back together. I do not know if this is intentional behavior.
After "much" trial and error I have found the issue. In the 02_2.rom at address 0x00004B3F There is a character. In the stock bios it is an @ (hex value 40.) Changing the OEM/Table ID the tool changes this to D (hex value 44). This change enables activation... but breaks speed step. EDIT: After further investigation it seems the RSDT table ID changes when speed step is enabled. And is either RSDT@ with no speed step or RSDTd (with an additional section of the table appearing) with speed step... It seems windows requires it to be RSDTD to activate. Also, after using a totally different SLIC and only changing the RSDT OEM ID (and manually hex editing it back to RSDT@ I enabled speed step. In RW it reports the table as beginning with RSDTd with the totally new OEM ID. So I think it is actually patching the table in memory somehow. Now to just figure out where... I've really no idea how to do that and only very basic hex editing skills... This may require more skills than I have.