I have three workstations with the same motherboards, all identical (same revision, chips, etc.). Motherboards have a Winbond W25Q128 flash chip.
I have flashed two of those successfully several times with Buspirate (internal programmer is a no-go on these boards).
On the third motherboard flashrom fails to identify the flash chip 8 out of 10 times and terminates. Then on the few occasions when it does identify the flash chip it fails to complete the flashing process successfully (errors erasing).
I have tried two different Buspirate flashed with firmware 5.1, 6.0, 6.1, and 6.2 (same results regardless of firmware). All the hardware connections are fine - I have checked continuity from the flash chip legs to the Pomona clip to Buspirate. My wire leads from Buspirate to the Pomona clip are short, 10 cm. I am certain this is not a hardware issue since Buspirate/flashrom work on two other motherboards.
I just do not know what other test should be pursued to get to a resolution. I have attached the –V logs for a successful flash from one of the two boards flashrom works, log where flash chip found but aborts on errors, log where flash chip not found, and a –VVV log for the latter. I note that the ‘probing’ results are very different but do not know what that may lead to conclude.
Any pointers, comments, help welcome.
On Mon, 12 May 2014 13:40:42 -0700 Roberto Denis roberto@godenis.com wrote:
I have three workstations with the same motherboards, all identical (same revision, chips, etc.). Motherboards have a Winbond W25Q128 flash chip.
I have flashed two of those successfully several times with Buspirate (internal programmer is a no-go on these boards).
On the third motherboard flashrom fails to identify the flash chip 8 out of 10 times and terminates. Then on the few occasions when it does identify the flash chip it fails to complete the flashing process successfully (errors erasing).
I have tried two different Buspirate flashed with firmware 5.1, 6.0, 6.1, and 6.2 (same results regardless of firmware). All the hardware connections are fine - I have checked continuity from the flash chip legs to the Pomona clip to Buspirate. My wire leads from Buspirate to the Pomona clip are short, 10 cm. I am certain this is not a hardware issue since Buspirate/flashrom work on two other motherboards.
I just do not know what other test should be pursued to get to a resolution. I have attached the –V logs for a successful flash from one of the two boards flashrom works, log where flash chip found but aborts on errors, log where flash chip not found, and a –VVV log for the latter. I note that the ‘probing’ results are very different but do not know what that may lead to conclude.
Any pointers, comments, help welcome.
Intel-based boards? The following page was not written with external programmers in mind, but it is probably the underlying cause... http://flashrom.org/ME see also the last point in the problems section here: http://flashrom.org/ISP
Yes it is an Intel based board (internal programmer is a no-go) and yes the ME code is in the flash chip along with the BIOS code. The ME code portion is indeed signed and changing any part of it renders the ME/ATM feature 'disabled'.
I thought I explained that I am using the ISP approach and have successfully flashed (using the layout file feature in flashrom) the BIOS portion two other identical boards. I am using a Pomona 16-pin SOIC clip and a buspirate. What is stumping me is the third board where the same ISP procedure results in the flash not being found. I attached the logs in hope someone could see something I am missing.
Thanks for the quick reply.
-----Original Message----- From: Stefan Tauner Sent: Monday, May 12, 2014 3:31 PM To: Roberto Denis Cc: flashrom@flashrom.org Subject: Re: [flashrom] Buspirate Erratic Results : WInbond W25Q128.V
On Mon, 12 May 2014 13:40:42 -0700 Roberto Denis roberto@godenis.com wrote:
I have three workstations with the same motherboards, all identical (same revision, chips, etc.). Motherboards have a Winbond W25Q128 flash chip.
I have flashed two of those successfully several times with Buspirate (internal programmer is a no-go on these boards).
On the third motherboard flashrom fails to identify the flash chip 8 out of 10 times and terminates. Then on the few occasions when it does identify the flash chip it fails to complete the flashing process successfully (errors erasing).
I have tried two different Buspirate flashed with firmware 5.1, 6.0, 6.1, and 6.2 (same results regardless of firmware). All the hardware connections are fine - I have checked continuity from the flash chip legs to the Pomona clip to Buspirate. My wire leads from Buspirate to the Pomona clip are short, 10 cm. I am certain this is not a hardware issue since Buspirate/flashrom work on two other motherboards.
I just do not know what other test should be pursued to get to a resolution. I have attached the –V logs for a successful flash from one of the two boards flashrom works, log where flash chip found but aborts on errors, log where flash chip not found, and a –VVV log for the latter. I note that the ‘probing’ results are very different but do not know what that may lead to conclude.
Any pointers, comments, help welcome.
Intel-based boards? The following page was not written with external programmers in mind, but it is probably the underlying cause... http://flashrom.org/ME see also the last point in the problems section here: http://flashrom.org/ISP
Not the most elegant way of doing it, but since ISP programming failed and it had bricked the motherboard I pulled (de-soldered) the flash chip off the board and used buspirate to flash it. Off the board flashing with buspirate worked on the same chip that it would not while installed on the board. Soldered the flash chip back on the board and it is now un-bricked.
So, on some of the boards (all identical) ISP worked but on this particular one it would not. Surely it is all impedance/signal bandpass filtering related just like explained in your http://flashrom.org/ISP. One recommendation would be that when attempting ISP to first read the flash several times (or even run flashrom/buspirate with no arguments) to make sure flashrom/buspirate is consistently working finding and then reading the chip before attempting to write to it.
Thanks for all the help.
-----Original Message----- From: Stefan Tauner Sent: Monday, May 12, 2014 3:31 PM To: Roberto Denis Cc: flashrom@flashrom.org Subject: Re: [flashrom] Buspirate Erratic Results : WInbond W25Q128.V
On Mon, 12 May 2014 13:40:42 -0700 Roberto Denis roberto@godenis.com wrote:
I have three workstations with the same motherboards, all identical (same revision, chips, etc.). Motherboards have a Winbond W25Q128 flash chip.
I have flashed two of those successfully several times with Buspirate (internal programmer is a no-go on these boards).
On the third motherboard flashrom fails to identify the flash chip 8 out of 10 times and terminates. Then on the few occasions when it does identify the flash chip it fails to complete the flashing process successfully (errors erasing).
I have tried two different Buspirate flashed with firmware 5.1, 6.0, 6.1, and 6.2 (same results regardless of firmware). All the hardware connections are fine - I have checked continuity from the flash chip legs to the Pomona clip to Buspirate. My wire leads from Buspirate to the Pomona clip are short, 10 cm. I am certain this is not a hardware issue since Buspirate/flashrom work on two other motherboards.
I just do not know what other test should be pursued to get to a resolution. I have attached the –V logs for a successful flash from one of the two boards flashrom works, log where flash chip found but aborts on errors, log where flash chip not found, and a –VVV log for the latter. I note that the ‘probing’ results are very different but do not know what that may lead to conclude.
Any pointers, comments, help welcome.
Intel-based boards? The following page was not written with external programmers in mind, but it is probably the underlying cause... http://flashrom.org/ME see also the last point in the problems section here: http://flashrom.org/ISP
-
Roberto Denis -
Stefan Tauner