Modify

Opened 6 years ago

Closed 5 years ago

#119 closed enhancement (worksforme)

Winbond W39V040FBPZ is not written correctly by flashrom

Reported by: charles.herndon@… Owned by: somebody
Priority: minor Milestone: flashrom v1.1
Component: flashrom (please use trac on flashrom.org) Keywords: W39V040FBPZ
Cc: Dependencies:
Patch Status: there is no patch

Description

Flash device is detected as a Winbond W39V040B device. Flashrom attempts to flash device, but verification fails. No actual writing to the chip appears to be done. Tried changing write and erase in flashchips.c to: jedec, winbond_fwhub and 49lfxxxc.

./flashrom -wv xxxx.bin Calibrating delay loop... OK. No coreboot table found. Found chipset "Intel ICH7/ICH7R", enabling flash write... OK. Found chip "Winbond W39V040B" (512 KB) at physical address 0xfff80000. Flash image seems to be a legacy BIOS. Disabling checks. Programming page: 0007 at address: 0x00070000 Verifying flash... FAILED! Expected=0xc7, Read=0x49

Attachments (0)

Change History (7)

comment:1 Changed 6 years ago by coreboot01@…

I got a similar problem using W39V040FBPZ.

After starting flashrom, output stopped at "disabling checks" for a minute or so, then the first "Programming page" appeared and nothing more happened. Looking with strace at the process id showed that the process was doing nothing at all. I left the system at this state for a whole day, but still nothing happened.

Then I tried again using a SST 49LF040 chip on the board (VIA PC2500e) and flashing worked without any problem.

Andreas Wanske

comment:2 follow-up: Changed 6 years ago by hailfinger

strace is unable to see any flashrom activity because flashrom does not use syscalls. (There are no syscalls for flashing.)

So if you think flashrom does nothing at all, it is still doing something, but no utility like strace can see it.

The Winbond chips are really problematic. If we try to program them according to their data sheets, they fail sometimes. Any insight is appreciated. No other chip vendor has these problems.

comment:3 in reply to: ↑ 2 Changed 6 years ago by charles.herndon@…

Replying to hailfinger:

strace is unable to see any flashrom activity because flashrom does not use syscalls. (There are no syscalls for flashing.)

So if you think flashrom does nothing at all, it is still doing something, but no utility like strace can see it.

The Winbond chips are really problematic. If we try to program them according to their data sheets, they fail sometimes. Any insight is appreciated. No other chip vendor has these problems.

I'm willing to provide any information you need. The only thing I do know about this chip is that by mistake I was successful programming it with an EEPROM programmer using the settings for SST49LF004 (I forgot to change models to the Winbond part)

comment:4 Changed 6 years ago by stepan

  • Milestone set to flashrom v1.1

comment:5 Changed 6 years ago by hailfinger

The Winbond W39V040B is one of the chips which sometimes totally deviates from the bhaviour specified in the datasheet. We have no idea why, and it may even be possible that the problematic chips are counterfeit (but given that they appear on brand name mainboards, that's unlikely).

Progress printing in flashrom is on the TODO list.

comment:6 Changed 5 years ago by hailfinger

Something I noticed just now:

Charles, you mentioned W39V040FB (FWH) in the summary, but your log says W39V040B (LPC). We did fix one critical timing error for the W39V040B, so this may be solved in current flashrom from svn.

A retest would be appreciated.

comment:7 Changed 5 years ago by hailfinger

  • Resolution set to worksforme
  • Status changed from new to closed

We fixed a few bugs:

  • Erase is now an auto-verified operation (it wasn't in your flashrom version)
  • W39* toggle timings were fixed (those resulted in too early false completion reports)
  • More paranoia and better diagnostics in all steps of flashrom operation

Charles, I will close this bug for now. Feel free to reopen it if you still own the hardware and are willing to test.

Add Comment

Modify Ticket

Action
as closed The owner will remain somebody.
The resolution will be deleted. Next status will be 'reopened'.
Author


E-mail address and user name can be saved in the Preferences.

 
Note: See TracTickets for help on using tickets.