Hello,
I'd like to report a couple of mixed cases for several nodes having a "Winbond W39V080FA" chip.
First, I was using flashrom-v0.9.1. I had success with about 6 nodes, then the flash failed on the next two. The ERASE step succeeded, but the Programming step hung at address 0c0000. So far, I could recover one of the chips in another node (PLCC32 socket). I booted normally, then hot-swapped in the failed chip and was quite happy when a read succeeded using a newly installed flashrom v0.9.3-r1299 (from Fedora EPEL). Re-flash then succeeded for this chip as well. I could not find a significant difference between the nodes that succeeded and those that did not. They started with the same BIOS, but the CMOS may have differed. I then continued with flashrom v0.9.3-r1299 for more nodes and did not have a failure so far (ran on 8 nodes).
I realize that the thin statistics alone do not support declaring full success, but hope that the evidently changed write regime (erase/write is now interleaved) has done the trick.
Thank you! Michael
=====================================================================================
0.9.1 SUCCESS LOG
... Probing for Winbond W39V080FA, 1024 KB: Chip lacks correct probe timing information, using default 10mS/40uS. probe_jedec: id1 0xda, id2 0xd3 Found chip "Winbond W39V080FA" (1024 KB, FWH) at physical address 0xfff00000. Probing for Winbond W39V080FA (dual mode), 512 KB: Chip lacks correct probe timing information, using default 10mS/40uS. probe_jedec: id1 0xda, id2 0xd3 Probing for Atmel unknown Atmel SPI chip, 0 KB: skipped. Host bus type Non-SPI and chip bus type SPI are incompatible. Probing for EON unknown EON SPI chip, 0 KB: skipped. Host bus type Non-SPI and chip bus type SPI are incompatible. Probing for Macronix unknown Macronix SPI chip, 0 KB: skipped. Host bus type Non-SPI and chip bus type SPI are incompatible. Probing for PMC unknown PMC SPI chip, 0 KB: skipped. Host bus type Non-SPI and chip bus type SPI are incompatible. Probing for SST unknown SST SPI chip, 0 KB: skipped. Host bus type Non-SPI and chip bus type SPI are incompatible. Probing for ST unknown ST SPI chip, 0 KB: skipped. Host bus type Non-SPI and chip bus type SPI are incompatible. Flash image seems to be a legacy BIOS. Disabling checks. Writing flash chip... Lockout bits: No hardware bootblock locking (good!) No hardware block locking (good!) Trying to unlock block @0x00000000 = 0x00 Full Access. Trying to unlock block @0x00010000 = 0x00 Full Access. Trying to unlock block @0x00020000 = 0x00 Full Access. Trying to unlock block @0x00030000 = 0x00 Full Access. Trying to unlock block @0x00040000 = 0x00 Full Access. Trying to unlock block @0x00050000 = 0x00 Full Access. Trying to unlock block @0x00060000 = 0x00 Full Access. Trying to unlock block @0x00070000 = 0x00 Full Access. Trying to unlock block @0x00080000 = 0x00 Full Access. Trying to unlock block @0x00090000 = 0x00 Full Access. Trying to unlock block @0x000a0000 = 0x00 Full Access. Trying to unlock block @0x000b0000 = 0x00 Full Access. Trying to unlock block @0x000c0000 = 0x00 Full Access. Trying to unlock block @0x000d0000 = 0x00 Full Access. Trying to unlock block @0x000e0000 = 0x00 Full Access. Trying to unlock block @0x000f0000 = 0x00 Full Access. Erasing: Programming: COMPLETE. Verifying flash... VERIFIED.
=====================================================================================
0.9.1 FAIL LOG
(same as above, except hangs at the Programming step at 0c0000)
=====================================================================================
0.9.3 SUCCESS LOG
flashrom v0.9.3-r1299 on Linux 2.6.18-164.11.1.el5 (x86_64), built with libpci 3.1.7, GCC 4.1.2 20080704 (Red Hat 4.1.2-50), little endian flashrom is free software, get the source code at http://www.flashrom.org
Calibrating delay loop... OS timer resolution is 1 usecs, 2664M loops per second, 10 myus = 10 us, 100 myus = 101 us, 1000 myus = 1000 us, 10000 myus = 10001 us, 4 myus = 4 us, OK. Initializing internal programmer No coreboot table found. DMI string system-manufacturer: "Supermicro" DMI string system-product-name: "X7DBT" DMI string system-version: "0123456789" DMI string baseboard-manufacturer: "Supermicro" DMI string baseboard-product-name: "X7DBT" DMI string baseboard-version: "PCB Version" DMI string chassis-type: "Other" Found chipset "Intel 631xESB/632xESB/3100", enabling flash write... chipset PCI ID is 8086:2670, 0x7fffffff/0x7fffffff FWH IDSEL: 0x0 0x7fffffff/0x7fffffff FWH IDSEL: 0x0 0x7fffffff/0x7fffffff FWH IDSEL: 0x1 0x7fffffff/0x7fffffff FWH IDSEL: 0x1 0x7fffffff/0x7fffffff FWH IDSEL: 0x2 0x7fffffff/0x7fffffff FWH IDSEL: 0x2 0x7fffffff/0x7fffffff FWH IDSEL: 0x3 0x7fffffff/0x7fffffff FWH IDSEL: 0x3 0x7fffffff/0x7fffffff FWH IDSEL: 0x4 0x7fffffff/0x7fffffff FWH IDSEL: 0x5 0x7fffffff/0x7fffffff FWH IDSEL: 0x6 0x7fffffff/0x7fffffff FWH IDSEL: 0x7 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled 0x7fffffff/0x7fffffff FWH decode enabled Maximum FWH chip size: 0x100000 bytes BIOS Lock Enable: disabled, BIOS Write Enable: enabled, BIOS_CNTL is 0x1 OK. This chipset supports the following protocols: FWH. Probing for AMD Am29F010A/B, 128 KB: skipped. Probing for AMD Am29F002(N)BB, 256 KB: skipped.
...
Probing for Winbond W49V002FA, 256 KB: probe_jedec_common: id1 0xda, id2 0xd3 Probing for Winbond W39V080FA, 1024 KB: probe_jedec_common: id1 0xda, id2 0xd3 Found chip "Winbond W39V080FA" (1024 KB, FWH) at physical address 0xfff00000. Lockout bits: Hardware bootblock locking (#TBL) is not active. Hardware remaining chip locking (#WP) is not active.. Lock status of block at 0x00000000 is Write Lock (Default State). Lock status of block at 0x00010000 is Write Lock (Default State). Lock status of block at 0x00020000 is Write Lock (Default State). Lock status of block at 0x00030000 is Write Lock (Default State). Lock status of block at 0x00040000 is Write Lock (Default State). Lock status of block at 0x00050000 is Write Lock (Default State). Lock status of block at 0x00060000 is Write Lock (Default State). Lock status of block at 0x00070000 is Write Lock (Default State). Lock status of block at 0x00080000 is Write Lock (Default State). Lock status of block at 0x00090000 is Write Lock (Default State). Lock status of block at 0x000a0000 is Write Lock (Default State). Lock status of block at 0x000b0000 is Write Lock (Default State). Lock status of block at 0x000c0000 is Write Lock (Default State). Lock status of block at 0x000d0000 is Write Lock (Default State). Lock status of block at 0x000e0000 is Write Lock (Default State). Lock status of block at 0x000f0000 is Write Lock (Default State). Probing for Winbond W39V080FA (dual mode), 512 KB: probe_jedec_common: id1 0xda, id2 0xd3 Probing for AMIC unknown AMIC SPI chip, 0 KB: skipped. Probing for Atmel unknown Atmel SPI chip, 0 KB: skipped. Probing for Eon unknown Eon SPI chip, 0 KB: skipped. Probing for Macronix unknown Macronix SPI chip, 0 KB: skipped. Probing for PMC unknown PMC SPI chip, 0 KB: skipped. Probing for SST unknown SST SPI chip, 0 KB: skipped.Probing for ST unknown ST SPI chip, 0 KB: skipped. Probing for Sanyo unknown Sanyo SPI chip, 0 KB: skipped. Probing for Generic unknown SPI chip (RDID), 0 KB: skipped. Probing for Generic unknown SPI chip (REMS), 0 KB: skipped. === This flash part has status UNTESTED for operations: WRITE The test status of this chip may have been updated in the latest development version of flashrom. If you are running the latest development version, please email a report to flashrom@flashrom.org if any of the above operations work correctly for you with this flash part. Please include the flashrom output with the additional -V option for all operations you tested (-V, -Vr, -Vw, -VE), and mention which mainboard or programmer you tested. Please mention your board in the subject line. Thanks for your help! Unlocking block at 0x0 Unlocking block at 0x10000 Unlocking block at 0x20000 Unlocking block at 0x30000 Unlocking block at 0x40000 Unlocking block at 0x50000 Unlocking block at 0x60000 Unlocking block at 0x70000 Unlocking block at 0x80000 Unlocking block at 0x90000 Unlocking block at 0xa0000 Unlocking block at 0xb0000 Unlocking block at 0xc0000 Unlocking block at 0xd0000 Unlocking block at 0xe0000 Unlocking block at 0xf0000 Lockout bits: Hardware bootblock locking (#TBL) is not active. Hardware remaining chip locking (#WP) is not active.. Flash image seems to be a legacy BIOS. Disabling checks. Reading old flash chip contents... Erasing and writing flash chip... Looking at blockwise erase function 0... trying... 0x000000-0x00ffff:EW, 0x010000-0x01ffff:S, 0x020000-0x02ffff:S, 0x030000-0x03ffff:S, 0x040000-0x04ffff:W, 0x050000-0x05ffff:EW, 0x060000-0x06ffff:EW, 0x070000-0x07ffff:EW, 0x080000-0x08ffff:EW, 0x090000-0x09ffff:EW, 0x0a0000-0x0aff ff:EW, 0x0b0000-0x0bffff:EW, 0x0c0000-0x0cffff:EW, 0x0d0000-0x0dffff:EW, 0x0e0000-0x0effff:EW, 0x0f0000-0x0fffff:EW
Done. Verifying flash... VERIFIED. =====================================================================================
hello michael and thanks for your report!
On Sat, 16 Jul 2011 19:08:08 -0500 Michael Sternberg sternberg@anl.gov wrote:
I'd like to report a couple of mixed cases for several nodes having a "Winbond W39V080FA" chip.
First, I was using flashrom-v0.9.1. I had success with about 6 nodes, then the flash failed on the next two. The ERASE step succeeded, but the Programming step hung at address 0c0000. So far, I could recover one of the chips in another node (PLCC32 socket). I booted normally, then hot-swapped in the failed chip and was quite happy when a read succeeded using a newly installed flashrom v0.9.3-r1299 (from Fedora EPEL). Re-flash then succeeded for this chip as well. I could not find a significant difference between the nodes that succeeded and those that did not. They started with the same BIOS, but the CMOS may have differed. I then continued with flashrom v0.9.3-r1299 for more nodes and did not have a failure so far (ran on 8 nodes).
I realize that the thin statistics alone do not support declaring full success, but hope that the evidently changed write regime (erase/write is now interleaved) has done the trick.
since none failed with the newer version, i think it is safe to declare it as working until further notice.
0.9.1 SUCCESS LOG
wow, that's ancient! :)
flashrom v0.9.3-r1299 on Linux 2.6.18-164.11.1.el5 (x86_64), built with libpci 3.1.7, GCC 4.1.2 20080704 (Red Hat 4.1.2-50), little endian flashrom is free software, get the source code at http://www.flashrom.org
Calibrating delay loop... OS timer resolution is 1 usecs, 2664M loops per second, 10 myus = 10 us, 100 myus = 101 us, 1000 myus = 1000 us, 10000 myus = 10001 us, 4 myus = 4 us, OK. Initializing internal programmer No coreboot table found. DMI string system-manufacturer: "Supermicro" DMI string system-product-name: "X7DBT" DMI string system-version: "0123456789" DMI string baseboard-manufacturer: "Supermicro" DMI string baseboard-product-name: "X7DBT" DMI string baseboard-version: "PCB Version" DMI string chassis-type: "Other" Found chipset "Intel 631xESB/632xESB/3100", enabling flash write... chipset PCI ID is 8086:2670,
i would like to mark your board as supported. the dmi data indicates that is a this board: http://www.supermicro.com/products/motherboard/Xeon1333/5000P/X7DBT.cfm there are similar boards which may have the same signature, especially this one: http://www.supermicro.com/products/motherboard/Xeon1333/5000P/X7DBT-INF.cfm
can you please state the exact model of the board(s) you have tested?
This flash part has status UNTESTED for operations: WRITE
i have marked it as fully tested now.
Unlocking block at 0x0 Unlocking block at 0x10000 Unlocking block at 0x20000 Unlocking block at 0x30000 Unlocking block at 0x40000 Unlocking block at 0x50000 Unlocking block at 0x60000 Unlocking block at 0x70000 Unlocking block at 0x80000 Unlocking block at 0x90000 Unlocking block at 0xa0000 Unlocking block at 0xb0000 Unlocking block at 0xc0000 Unlocking block at 0xd0000 Unlocking block at 0xe0000 Unlocking block at 0xf0000
i have also changed the format of this messages to use 32 bits/8 characters.
Hi Stefan,
On Jul 16, 2011, at 20:24 , Stefan Tauner wrote:
since none failed with the newer version, i think it is safe to declare it as working until further notice.
OK.
0.9.1 SUCCESS LOG
wow, that's ancient! :)
Well, that's what I started with in the flashing business :-)
DMI string system-manufacturer: "Supermicro" DMI string system-product-name: "X7DBT" DMI string system-version: "0123456789" DMI string baseboard-manufacturer: "Supermicro" DMI string baseboard-product-name: "X7DBT" DMI string baseboard-version: "PCB Version" DMI string chassis-type: "Other" Found chipset "Intel 631xESB/632xESB/3100", enabling flash write... chipset PCI ID is 8086:2670,
i would like to mark your board as supported. the dmi data indicates that is a this board: http://www.supermicro.com/products/motherboard/Xeon1333/5000P/X7DBT.cfm there are similar boards which may have the same signature, especially this one: http://www.supermicro.com/products/motherboard/Xeon1333/5000P/X7DBT-INF.cfm
can you please state the exact model of the board(s) you have tested?
Yes - I actually used the X7DBT-INF variant, which has a Mellanox MT25204AD-FCC-D chip on-board. This InfiniBand chip is evidently outside the purview of the BIOS. There is just one manual for {X7DBT,X7DGT}{,-INF}. The BIOS reports just the base variant (as shown), and is the same for these four boards.
http://www.supermicro.com/support/bios/ "7DBT" http://www.supermicro.com/support/manuals/index.cfm
"The X7DBT/X7DBT-INF supports the 5000P Chipset, and the X7DGT/X7DGT-INF supports the 5000X chipset"
This flash part has status UNTESTED for operations: WRITE
i have marked it as fully tested now.
OK.
With best regards, Michael
On Sat, 16 Jul 2011 20:44:40 -0500 Michael Sternberg sternberg@anl.gov wrote:
On Jul 16, 2011, at 20:24 , Stefan Tauner wrote:
i would like to mark your board as supported. the dmi data indicates that is a this board: http://www.supermicro.com/products/motherboard/Xeon1333/5000P/X7DBT.cfm there are similar boards which may have the same signature, especially this one: http://www.supermicro.com/products/motherboard/Xeon1333/5000P/X7DBT-INF.cfm
can you please state the exact model of the board(s) you have tested?
Yes - I actually used the X7DBT-INF variant, which has a Mellanox MT25204AD-FCC-D chip on-board. This InfiniBand chip is evidently outside the purview of the BIOS. There is just one manual for {X7DBT,X7DGT}{,-INF}. The BIOS reports just the base variant (as shown), and is the same for these four boards.
http://www.supermicro.com/support/bios/ "7DBT" http://www.supermicro.com/support/manuals/index.cfm
"The X7DBT/X7DBT-INF supports the 5000P Chipset, and the X7DGT/X7DGT-INF supports the 5000X chipset"
i am not so sure that the infiniband controller or the chipset version is "outside the purview" of the bios, but it is not noted in the DMI data apparently. this is expected and that's the reason i have asked for the exact motherboard model ;) other versions will probably also just work, but we (try to) note the exact board versions in our supported hardware database to be on the safe side. if you test the other variants i am happy to add those too. i have added the X7DBT-INF version to the database and will commit that later. thanks again!
-
Michael Sternberg -
Stefan Tauner