(UEFI) Dell XPS 15z L511z modded BIOS - and HOWTO

[TimeWalker]

uh oh ... I totally forgot to check the bad dump you posted. :O

---

Well it appears to have written the ME in the correct region, the firmware however is bananas ..
I guess using FITC wasn't a good idea after all as pointed out by @CodeRush ... there is some random data in place of FFs in the original good dumps and there sectors with the opposite - FFs in place of data .. so your laptop got totally confused.
Also it appears to have updated a part of the BIOS regions for some reason ? This may have been due to saving some settings (boot order and such ?) in BIOS setup.. I don't really know where is the BBL part in the bios region (the boot block which is used for start up and recovery initialization... all the boot mode switching technically) but it sure became confused (note the "Note Installed" strings) after this and saved some bs into the bios region.

[kasar]

well, that dump was before my laptop got phisically screwed (I think) because when I plugged back the restored chip after the bad ME update, the laptop powered on, once, then it didnt did anymore.

and while trying to access the third time to the chip, it started acting weird, unstable, sometimes detected as unknown chip, some others as 1mb or 8 mb chip ..., well, it was messed up, and it wasnt the wires, they were exactly the same and as short as when I did the good dumps and unlocked the descriptor settings.

like you said, probably the chip is fried, thats why I want to get another one, plus the chip sockets for the motherboard and programer.

I noticed you talk a lot about @coderush, I though you were talking about the guy who posted this topic

http://forum.techinferno.com/general-not...ptops.html

but he isnt I think, do you have the topic link where you are reading @coderush posts?

[TimeWalker]

Nah, not the same person.

I'm talking to him via private messages. He's active on overclockers.ru and hardforums.com .. but I prefer no to interfere in his threads directly because they are asus-related mostly.
http://forums.overclockers.ru/viewtopic....5&t=447045
http://hardforum.com/showthread.php?t=1726429

Here's his programmer project:
http://we.easyelectronics.ru/electro-and...t232h.html

[kasar]

thanks

uh, russian! O.o

I will have a long time with the translator

I checked some of it, that usb programer is really interesting, and really small.

also have no need of external power suply and have the small size or a small usb stick!

well, since I already built my lpt programmer and I am happy with it, I think I will save the money to other stuff, starting by replace my motherboard if the bios chip is not the only damaged part ^^

but well, every modern computer have usb, that isnt the same as LPT, wich is getting very rare to find,

I used a old pentium 3 desktop machine to flash the chip image from the LPT port.


question.

do another chip with the same size in size and in Mb and same pinout work as replacement for my motherboard?

or it have to be the same one?

[TimeWalker]

There're some checks inside firmware to identify a proper chip... in fact there are even modules for it embedded in the firmware.

W25Q32FlashPartSmm.efi F8CA70B4-6A85-4616-8BDE-3EC569644AC3_1_xxx.ROM
W25Q32FlashPartDxe.efi 68BC3095-8839-4A97-8FEA-BEE06ECA7BBB_1_xxx.ROM

W25Q32BVSIG is Vendor/Device ID: 0xEF4016
Found in your L502X firmware at the offset 0xE00 which is your flash descriptor (hence the name hehe)

[kasar]

then I guess I have to get the same chip then

else I would sit like a dumb thing in front of the machine expecting it boot or do something ^^

I ask because in some cases when I asked for a electronic component on a store,they said they didnt had the specific part, but they had alternatives.

thanks for the info

[nebster]

Since we're flashing the chip directly, can't we just modify the Vendor/Device ID?

[jkbuha]

Morning all

@TheWiz - thank you for your kind offer! We've collected Eur57 so far - any spare cash is much appreciated!
@kasar - let me know when to close the pledge so I can send over the donations to your account. Have you established if it's the mobo or the chip?

[TimeWalker]

Since we're flashing the chip directly, can't we just modify the Vendor/Device ID?

Support for the chip comes inside the firmware. I believe the modules (Smm and Dxe) are coded for SPI BUS communications with PCH (Cougar Point chipset) to use this particular chip.
Same vid/pid of 0xEF4016 can be found inside both of the modules mentioned above (offset 0x2D0h).

Runnin FPT -i on the system identifies the flashrom's structure:

Code:

Platform: Intel(R) HM67 Express Chipset
Reading HSFSTS register... Flash Descriptor: Valid

    --- Flash Devices Found ---
    W25Q32BV    ID:0xEF4016    Size: 4096KB (32768Kb)

    --- Flash Image Information --
    Signature: VALID
    Number of Flash Components: 1
        Component 1 - 4096KB (32768Kb)
    Regions:
        Descriptor - Base: 0x000000, Limit: 0x000FFF
        ME         - Base: 0x001000, Limit: 0x17FFFF
        BIOS       - Base: 0x180000, Limit: 0x3FFFFF
        GbE        - Not present
        PDR        - Not present
    Master Region Access:
        CPU/BIOS - ID: 0x0000, Read: 0x0B, Write: 0x0A
        ME       - ID: 0x0000, Read: 0x0D, Write: 0x0C
        GbE      - ID: 0x0118, Read: 0x08, Write: 0x08

Total Accessable SPI Memory: 4096KB, Total Installed SPI Memory : 4096KB

[kasar]

@jkbuha

well, I have to determine if the chip replacement will fix the motherboard.


today I called my local electronic store, they dont have the winbond chips.

so I ordered them via the link timewalker gave to me.

http://www.ebay.com/itm/5pcs-WINBOND-W25...0668544992

the good point is that I will have 5 of them if for some reason I make another mistake.

the bad point is that it will take its time to arrive.

probably more than 1 or two weeks at least ):

well, anyway I will also go today to the local electronic store and will pick the DIP 8 socket, 100 ohm resistors, and another parallel ports to solder,

this will allow make another parallel port programmer, so I will be able to confirm the programmer is definitly ok and the chip is fried.

anyway since I didnt changed anything from the programer, I supose the chip is already broken.



@timewalker

mmm, so, if we modify the firmware and add another chip ID, it would be posible to replace the chip with another one, well, anyway i see no point doing this because the machine would be fucked on every bios update, making necesary to patch all updates with the another chip.

I also ordered the ones you said, so no issues from that way.