Hi there GN, any news regarding your progress? btw please CC flashrom
mailing list for your future replies, so that someone else could be
able to help you as well
On Sun, Mar 28, 2021 at 7:49 PM GN wrote:
>
> Dear MIke,
>
> Thank you for you detailed answer, very much appreciated.
>
> Thank you for the clarification about the chips, I was getting crazy finding
> the right information about it. I will report the difference across all the
> posts I wrote on the issue :)
>
> The procedure in [1] is for a BBB. I will try to understand what can be
> transferred to the CH341a that I am using. For it, I found this interesting
> DIY, although according to the author the outputted voltage cannot be
> carefully controlled.
>
> I have tried to connect the power supplier to the naked motherboard while
> reading the Winbond chip. As predicted, the power LED on the CH341a lights on
> dimmed and it gets full-on when connected to the USB 3 port. Less obviously,
> flashrom cannot recognize the chip when the power supplier is connected. As I
> disconnect the power supplier, flashrom correctly recognizes the chip again,
> but the writing problems persist.
>
>
> Bests regards,
>
> GN
>
>
> On 25.03.21 14:39, Mike Banon wrote:
>
> I think the only way to prevent the current leaking to the surrounding
> elements of a chip - is desoldering. So, if we are trying to avoid the
> desoldering - yes, you could try reducing the length of wires between
> a programmer and a test clip - i.e. to 10cm. That certainly wouldn't
> hurt, and sometimes it really helps to achieve the successful flashing
> by ISP (in-system programming) method.
>
> Please note: while there are some successful examples of using the
> external PSU for an ISP flashing (i.e. this [1] BBB example at
> libreboot wiki), depending on a motherboard design this could be
> dangerous according to Angel Pons comments under [2] - so using the
> external PSU should be avoided unless all the other options have been
> tried and you still don't want to solder/desolder.
>
> While looking for the examples of L530 flashing I found a thread with
> your comments [3]. To answer your question why the other people were
> also targeting a ps08a chip: this is CMOS, and clearing a CMOS memory
> sometimes is also important to ensure the successful boot of a
> proprietary BIOS - but we haven't reached this stage yet, since we're
> just trying to flash a BIOS into a SPI flash chip.
>
> Have you ever tested your CH341A before in other setups? If not, do
> you have a spare CH341A to try it with a different one? And could you
> try a higher powered USB port to ensure that your CH341A itself is
> sufficiently powered? (i.e. you may be using a USB extender for your
> convenience, which is fine - but that extender also shouldn't be too
> long)
>
> [1] https://libreboot.org/docs/install/bbb_setup.html
> [2] https://review.coreboot.org/c/flashrom/+/31830
> (this was my patch for the older version of flashrom useful in a
> unreliable flashing setup, but I only successfully read with it and
> never successfully wrote).
> [3] https://www.badcaps.net/forum/showthread.php?t=61517
>
> P.S. I'm not sure how you can measure the leakage with a multimeter,
> and in any case knowing the exact values isn't necessary - when we
> already know the cause of unsuccessful flashing and just need to find
> out how to resolve it.
>
>
> On Wed, Mar 24, 2021 at 8:46 PM G. Nalin wrote:
>
> HI,
>
> I have been recommended to desolder the BIOS chip to use the CH341a reliably,
> but I would like to avoid it and use the clip instead! Should I simply cut
> shorter the clip wires? Is there something I can do to ground the motherboard
> or avoid the leaking? How can I measure it (I have just a multimeter here
> atm).
>
> I don´t know how to supply the power externally while the EEPROM is connected
> to the USB. Could you indicate me a reference?
>
> I tried to verify the dumo, but if fails because there is some writing
> involved.
> $ sudo flashrom -V -p ch341a_spi -v bios1b.img
>
> I don´t have a reference so I can´t say if the HEX or BIN dump is actually at
> least partially valid. It is not all FF and 00.
>
> Cheers,
>
> Giammarco Nalin
>
> Handy: +4915252667614
> Adresse: Hausener Weg 96, 60489, Frankfurt am Main
>
>
>
> From: Mike Banon
> Sent: 24 March 2021 13:40
> To: G. Nalin
> Cc: flashrom@flashrom.org
> Subject: Re: [flashrom] Fail flashing
>
> Could you please verify that at least a part of the binary file got
> flashed correctly? If yes, these reliability problems could be caused
> by your setup - i.e. too long wires or the surrounding elements of the
> BIOS chip are drawing too much current - causing a voltage drop on the
> chip itself and the remaining is insufficient for the reliable
> flashing (in this case I could recommend using the external power
> supply for 3.3V line)
>
> On Wed, Mar 24, 2021 at 9:50 AM G. Nalin wrote:
>
> Hi all,
> I am trying to flash a