flashrom not working - asus p5q (ICH10R)

[Thomas Prochaska]

hi!

today i tried the flashrom utility from coreboot-v2 svn trunk.
this is the output:
-----
fin@xiix:~$ sudo flashrom
Calibrating delay loop... OK.
Can't mmap memory using /dev/mem: Invalid argument
-----

therefore i want to ask you if my chipset is unsupported.
maybe you can shed some light on this.

my system:
asus p5q
debian/gnu linux (sid) (debian standard kernel 2.6.26.1, customized 
vanilla 2.6.28-rc8)

i tested it with both kernels and got same results.

also i wanted to ask if it's possible to support this motherboard if i 
help you and test stuff and so on. i have some knownledge in c/c++ and 
very little knownledge in programing drivers (linux). also i'm eager to 
learn.

regards
thomas prochaska

[Peter Stuge]

Hi Thomas!

Thomas Prochaska wrote:
> today i tried the flashrom utility from coreboot-v2 svn trunk.
> this is the output:
>
> fin@xiix:~$ sudo flashrom
> Calibrating delay loop... OK.
> Can't mmap memory using /dev/mem: Invalid argument
>
> therefore i want to ask you if my chipset is unsupported.
> maybe you can shed some light on this.
>
> my system:
> asus p5q
> debian/gnu linux (sid) (debian standard kernel 2.6.26.1, customized
> vanilla 2.6.28-rc8)
>
> i tested it with both kernels and got same results.

Please check that CONFIG_STRICT_DEVMEM is not set in your kernel
configuration. If it is set (I believe at least debian has it set)
flashrom can not work properly.
> also i wanted to ask if it's possible to support this motherboard
> if i help you and test stuff and so on.

Yes, it should be possible to support your board. ASUS has had tricky
board enables in the past, but hopefully yours isn't one of them.

//Peter

[Corey Osgood]

On Sun, Dec 21, 2008 at 7:52 PM, Peter Stuge  wrote:
> Hi Thomas!
>
> Thomas Prochaska wrote:
> > today i tried the flashrom utility from coreboot-v2 svn trunk.
> > this is the output:
> >
> > fin@xiix:~$ sudo flashrom
> > Calibrating delay loop... OK.
> > Can't mmap memory using /dev/mem: Invalid argument
> >
> > therefore i want to ask you if my chipset is unsupported.
> > maybe you can shed some light on this.
> >
> > my system:
> > asus p5q
> > debian/gnu linux (sid) (debian standard kernel 2.6.26.1, customized
> > vanilla 2.6.28-rc8)
> >
> > i tested it with both kernels and got same results.
>
> Please check that CONFIG_STRICT_DEVMEM is not set in your kernel
> configuration. If it is set (I believe at least debian has it set)
> flashrom can not work properly.

Debian testing at least does not, but his customized vanilla kernel might ;)

-Corey
> > also i wanted to ask if it's possible to support this motherboard
> > if i help you and test stuff and so on.
>
> Yes, it should be possible to support your board. ASUS has had tricky
> board enables in the past, but hopefully yours isn't one of them.
>
> //Peter

[Thomas Prochaska]

Peter Stuge wrote:
> Hi Thomas!
>
> Thomas Prochaska wrote:
> > today i tried the flashrom utility from coreboot-v2 svn trunk.
> > this is the output:
> >
> > fin@xiix:~$ sudo flashrom
> > Calibrating delay loop... OK.
> > Can't mmap memory using /dev/mem: Invalid argument
> >
> > therefore i want to ask you if my chipset is unsupported.
> > maybe you can shed some light on this.
> >
> > my system:
> > asus p5q
> > debian/gnu linux (sid) (debian standard kernel 2.6.26.1, customized
> > vanilla 2.6.28-rc8)
> >
> > i tested it with both kernels and got same results.
>
> Please check that CONFIG_STRICT_DEVMEM is not set in your kernel
> configuration. If it is set (I believe at least debian has it set)
> flashrom can not work properly.
> > also i wanted to ask if it's possible to support this motherboard
> > if i help you and test stuff and so on.
>
> Yes, it should be possible to support your board. ASUS has had tricky
> board enables in the past, but hopefully yours isn't one of them.
>
> //Peter

hi!

ok, i found the problem. i hadn't set CONFIG_STRICT_DEVMEM but i had 
x86-pat enabled. this was the problem. with the nopat kernel boot option 
it produces this output:
-----
xiix:~# flashrom -V
Calibrating delay loop... 855M loops per second, 100 myus = 200 us. OK.
No coreboot table found.
Found chipset "Intel ICH10R", enabling flash write...
BIOS Lock Enable: disabled, BIOS Write Enable: enabled, BIOS_CNTL is 0x1

Root Complex Register Block address = 0xfed1c000
GCS = 0x1464: BIOS Interface Lock-Down: disabled, BOOT BIOS Straps: 0x1 
(SPI)
Top Swap : not enabled
SPIBAR = 0xfed1c000 + 0x3800
0x00: 0x1fff (HSFS)
FLOCKDN 1, FDV 1, FDOPSS 0, SCIP 0, BERASE 0, AEL 0, FCERR 0, FDONE 0
0x50: 0x00000202 (FRAP)
BMWAG 0, BMRAG 0, BRWA 2, BRRA 2
0x54: 0x00001fff (FREG0)
0x58: 0x00001fff (FREG1)
0x5C: 0x00001fff (FREG2)
0x60: 0x00001fff (FREG3)
0x64: 0x00001fff (FREG4)
0x74: 0x00000000 (PR0)
0x78: 0x00000000 (PR1)
0x7C: 0x00000000 (PR2)
0x80: 0x00000000 (PR3)
0x84: 0x00000000 (PR4)
0xB0: 0x00000000 (FDOC)
SPI Read Configuration: prefetching disabled, caching enabled, OK.
Probing for AMD Am29F002(N)BB, 256 KB: probe_jedec: id1 0x4d, id2 0x5a, 
id1 parity violation
Probing for AMD Am29F002(N)BT, 256 KB: probe_jedec: id1 0x4d, id2 0x5a, 
id1 parity violation
Probing for AMD Am29F016D, 2048 KB: probe_29f040b: id1 0xff, id2 0xff
Probing for AMD Am29F040B, 512 KB: probe_29f040b: id1 0xa3, id2 0x51
Probing for AMD Am29LV040B, 512 KB: probe_29f040b: id1 0xa3, id2 0x51
Probing for ASD AE49F2008, 256 KB: probe_jedec: id1 0x4d, id2 0x5a, id1 
parity violation
Probing for Atmel AT25DF021, 256 KB: Segmentation fault
-----

is that a problem or just because i haven't provided an image to write.

thomas

[Peter Stuge]

Thomas Prochaska wrote:
> ok, i found the problem. i hadn't set CONFIG_STRICT_DEVMEM but i had 
> x86-pat enabled. this was the problem.

Aha! Good find.
> with the nopat kernel boot option it produces this output:
>
> xiix:~# flashrom -V

..
> Probing for Atmel AT25DF021, 256 KB: Segmentation fault
>
> is that a problem or just because i haven't provided an image to
> write.

This is a bug in the ICH SPI driver that must have been introduced in
the last few commits. Unfortunately I'm not completely familiar with
that code.

It would be helpful if you could provide a backtrace, please compile
flashrom with debugging enabled (add -g to CFLAGS and LDFLAGS, and
comment out the STRIP on line 43 in Makefile) and then run flashrom
in gdb:

# gdb ./flashrom
..
(gdb) r -V
..
crash
(gdb) bt

and post the output.

It would also be helpful if you could update (downgrade really) to
revisions 3809 and 3804 in turn, to see if they work better. Run:

svn up -r 3809

Thanks!

//Peter

[Thomas Prochaska]

Peter Stuge wrote:
> Thomas Prochaska wrote:
> > ok, i found the problem. i hadn't set CONFIG_STRICT_DEVMEM but i had
> > x86-pat enabled. this was the problem.
>
> Aha! Good find.
> > with the nopat kernel boot option it produces this output:
> >
> > xiix:~# flashrom -V
>
> ..
> > Probing for Atmel AT25DF021, 256 KB: Segmentation fault
> >
> > is that a problem or just because i haven't provided an image to
> > write.
>
> This is a bug in the ICH SPI driver that must have been introduced in
> the last few commits. Unfortunately I'm not completely familiar with
> that code.
>
> It would be helpful if you could provide a backtrace, please compile
> flashrom with debugging enabled (add -g to CFLAGS and LDFLAGS, and
> comment out the STRIP on line 43 in Makefile) and then run flashrom
> in gdb:
>
> # gdb ./flashrom
> ..
> (gdb) r -V
> ..
> crash
> (gdb) bt
>
> and post the output.
>
> It would also be helpful if you could update (downgrade really) to
> revisions 3809 and 3804 in turn, to see if they work better. Run:
>
> svn up -r 3809
>
> Thanks!
>
> //Peter

so,
thanks for your advice. gdb produced following output:
----
fin@xiix:~$ sudo gdb flashrom
GNU gdb 6.8-debian
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later 
http://gnu.org/licenses/gpl.html
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu"...
(gdb) r -V
Starting program: /usr/local/sbin/flashrom -V
Calibrating delay loop... 855M loops per second, 100 myus = 200 us. OK.
No coreboot table found.
Found chipset "Intel ICH10R", enabling flash write...
BIOS Lock Enable: disabled, BIOS Write Enable: enabled, BIOS_CNTL is 0x1

Root Complex Register Block address = 0xfed1c000
GCS = 0x1464: BIOS Interface Lock-Down: disabled, BOOT BIOS Straps: 0x1 
(SPI)
Top Swap : not enabled
SPIBAR = 0xfed1c000 + 0x3800
0x00: 0x1fff (HSFS)
FLOCKDN 1, FDV 1, FDOPSS 0, SCIP 0, BERASE 0, AEL 0, FCERR 0, FDONE 0
0x50: 0x00000202 (FRAP)
BMWAG 0, BMRAG 0, BRWA 2, BRRA 2
0x54: 0x00001fff (FREG0)
0x58: 0x00001fff (FREG1)
0x5C: 0x00001fff (FREG2)
0x60: 0x00001fff (FREG3)
0x64: 0x00001fff (FREG4)
0x74: 0x00000000 (PR0)
0x78: 0x00000000 (PR1)
0x7C: 0x00000000 (PR2)
0x80: 0x00000000 (PR3)
0x84: 0x00000000 (PR4)
0xB0: 0x00000000 (FDOC)
SPI Read Configuration: prefetching disabled, caching enabled, OK.
Probing for AMD Am29F002(N)BB, 256 KB: probe_jedec: id1 0x4d, id2 0x5a, 
id1 parity violation
Probing for AMD Am29F002(N)BT, 256 KB: probe_jedec: id1 0x4d, id2 0x5a, 
id1 parity violation
Probing for AMD Am29F016D, 2048 KB: probe_29f040b: id1 0xff, id2 0xff
Probing for AMD Am29F040B, 512 KB: probe_29f040b: id1 0xa3, id2 0x51
Probing for AMD Am29LV040B, 512 KB: probe_29f040b: id1 0xa3, id2 0x51
Probing for ASD AE49F2008, 256 KB: probe_jedec: id1 0x4d, id2 0x5a, id1 
parity violation
Probing for Atmel AT25DF021, 256 KB:
Program received signal SIGSEGV, Segmentation fault.
0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3, 
writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
     at ichspi.c:763
763			if ((curopcodes->opcode[a]).opcode == cmd) {
(gdb) bt
#0  0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3, 
writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
     at ichspi.c:763
#1  0x0000000000407fdc in spi_rdid (readarr=0x7fff772399e0 "M", 
bytes=159) at spi.c:59
#2  0x0000000000408027 in probe_spi_rdid_generic (flash=0x610a70, 
bytes=159) at spi.c:108
#3  0x00000000004066b8 in probe_flash (first_flash=0x610860, force=0) at 
flashrom.c:149
#4  0x0000000000406aec in main (argc=2, argv=0x7fff77239c58) at 
flashrom.c:468
(gdb)
----

i also tried revision 3809 and 3804, both worked and produced following 
output:
----
Calibrating delay loop... OK.
No coreboot table found.
Found chipset "Intel ICH10R", enabling flash write... OK.
Found chip "Macronix MX25L8005" (1024 KB) at physical address 0xfff00000.
No operations were specified.
----

[Thomas Prochaska]

Peter Stuge wrote:
> Thomas Prochaska wrote:
> > ok, i found the problem. i hadn't set CONFIG_STRICT_DEVMEM but i had
> > x86-pat enabled. this was the problem.
>
> Aha! Good find.
> > with the nopat kernel boot option it produces this output:
> >
> > xiix:~# flashrom -V
>
> ..
> > Probing for Atmel AT25DF021, 256 KB: Segmentation fault
> >
> > is that a problem or just because i haven't provided an image to
> > write.
>
> This is a bug in the ICH SPI driver that must have been introduced in
> the last few commits. Unfortunately I'm not completely familiar with
> that code.
>
> It would be helpful if you could provide a backtrace, please compile
> flashrom with debugging enabled (add -g to CFLAGS and LDFLAGS, and
> comment out the STRIP on line 43 in Makefile) and then run flashrom
> in gdb:
>
> # gdb ./flashrom
> ..
> (gdb) r -V
> ..
> crash
> (gdb) bt
>
> and post the output.
>
> It would also be helpful if you could update (downgrade really) to
> revisions 3809 and 3804 in turn, to see if they work better. Run:
>
> svn up -r 3809
>
> Thanks!
>
> //Peter

ok,
thanks for your advice.
here's the gdb output:
----
fin@xiix:~$ sudo gdb flashrom
GNU gdb 6.8-debian
Copyright (C) 2008 Free Software Foundation, Inc.
License GPLv3+: GNU GPL version 3 or later 
http://gnu.org/licenses/gpl.html
This is free software: you are free to change and redistribute it.
There is NO WARRANTY, to the extent permitted by law.  Type "show copying"
and "show warranty" for details.
This GDB was configured as "x86_64-linux-gnu"...
(gdb) r -V
Starting program: /usr/local/sbin/flashrom -V
Calibrating delay loop... 855M loops per second, 100 myus = 200 us. OK.
No coreboot table found.
Found chipset "Intel ICH10R", enabling flash write...
BIOS Lock Enable: disabled, BIOS Write Enable: enabled, BIOS_CNTL is 0x1

Root Complex Register Block address = 0xfed1c000
GCS = 0x1464: BIOS Interface Lock-Down: disabled, BOOT BIOS Straps: 0x1 
(SPI)
Top Swap : not enabled
SPIBAR = 0xfed1c000 + 0x3800
0x00: 0x1fff (HSFS)
FLOCKDN 1, FDV 1, FDOPSS 0, SCIP 0, BERASE 0, AEL 0, FCERR 0, FDONE 0
0x50: 0x00000202 (FRAP)
BMWAG 0, BMRAG 0, BRWA 2, BRRA 2
0x54: 0x00001fff (FREG0)
0x58: 0x00001fff (FREG1)
0x5C: 0x00001fff (FREG2)
0x60: 0x00001fff (FREG3)
0x64: 0x00001fff (FREG4)
0x74: 0x00000000 (PR0)
0x78: 0x00000000 (PR1)
0x7C: 0x00000000 (PR2)
0x80: 0x00000000 (PR3)
0x84: 0x00000000 (PR4)
0xB0: 0x00000000 (FDOC)
SPI Read Configuration: prefetching disabled, caching enabled, OK.
Probing for AMD Am29F002(N)BB, 256 KB: probe_jedec: id1 0x4d, id2 0x5a, 
id1 parity violation
Probing for AMD Am29F002(N)BT, 256 KB: probe_jedec: id1 0x4d, id2 0x5a, 
id1 parity violation
Probing for AMD Am29F016D, 2048 KB: probe_29f040b: id1 0xff, id2 0xff
Probing for AMD Am29F040B, 512 KB: probe_29f040b: id1 0xa3, id2 0x51
Probing for AMD Am29LV040B, 512 KB: probe_29f040b: id1 0xa3, id2 0x51
Probing for ASD AE49F2008, 256 KB: probe_jedec: id1 0x4d, id2 0x5a, id1 
parity violation
Probing for Atmel AT25DF021, 256 KB:
Program received signal SIGSEGV, Segmentation fault.
0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3, 
writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
     at ichspi.c:763
763			if ((curopcodes->opcode[a]).opcode == cmd) {
(gdb) bt
#0  0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3, 
writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
     at ichspi.c:763
#1  0x0000000000407fdc in spi_rdid (readarr=0x7fff772399e0 "M", 
bytes=159) at spi.c:59
#2  0x0000000000408027 in probe_spi_rdid_generic (flash=0x610a70, 
bytes=159) at spi.c:108
#3  0x00000000004066b8 in probe_flash (first_flash=0x610860, force=0) at 
flashrom.c:149
#4  0x0000000000406aec in main (argc=2, argv=0x7fff77239c58) at 
flashrom.c:468
(gdb)
----

i checked with revision 3809 and 3804 and both should work. both 
produced following output:
----
Calibrating delay loop... OK.
No coreboot table found.
Found chipset "Intel ICH10R", enabling flash write... OK.
Found chip "Macronix MX25L8005" (1024 KB) at physical address 0xfff00000.
No operations were specified.
----

[Peter Stuge]

Thomas Prochaska wrote:
> Program received signal SIGSEGV, Segmentation fault.
> 0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3, 
> writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
>     at ichspi.c:763
> 763			if ((curopcodes->opcode[a]).opcode == cmd) {
> (gdb) bt
> #0  0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3, 
> writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
>     at ichspi.c:763

Ok.
> i checked with revision 3809 and 3804 and both should work.

Looks good. Ok. Please go back to trunk HEAD (svn up -r HEAD) and
try applying this patch.

//Peter

[Thomas Prochaska]

Peter Stuge wrote:
> Thomas Prochaska wrote:
> > Program received signal SIGSEGV, Segmentation fault.
> > 0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3,
> > writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
> >      at ichspi.c:763
> > 763			if ((curopcodes->opcode[a]).opcode == cmd) {
> > (gdb) bt
> > #0  0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3,
> > writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
> >      at ichspi.c:763
>
> Ok.
> > i checked with revision 3809 and 3804 and both should work.
>
> Looks good. Ok. Please go back to trunk HEAD (svn up -r HEAD) and
> try applying this patch.
>
> //Peter

k,
i checked, that did it. thanks.

[Peter Stuge]

Thomas Prochaska wrote:
> > Looks good. Ok. Please go back to trunk HEAD (svn up -r HEAD) and
> > try applying this patch.
>
> i checked, that did it. thanks.

Thanks. I just committed it, r3830.

//Peter

[FENG Yu Ning]

On Mon, Dec 22, 2008 at 12:10 PM, Peter Stuge  wrote:
> Thomas Prochaska wrote:
> > Program received signal SIGSEGV, Segmentation fault.
> > 0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3,
> > writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
> >     at ichspi.c:763
> > 763                   if ((curopcodes->opcode[a]).opcode == cmd) {
> > (gdb) bt
> > #0  0x0000000000408fae in ich_spi_command (writecnt=1, readcnt=3,
> > writearr=0x7fff772399cf "\237", readarr=0x7fff772399e0 "M")
> >     at ichspi.c:763
>
> Ok.

I felt nervous when I read here.
> > i checked with revision 3809 and 3804 and both should work.
>
> Looks good. Ok. Please go back to trunk HEAD (svn up -r HEAD) and
> try applying this patch.

Having read the patch, I realized that my deletion of opcode
programming affected both ICH7 and 9, but I only initialized opcodes
for ICH7.

On Mon, Dec 22, 2008 at 12:43 PM, Thomas Prochaska  wrote:
> k,
> i checked, that did it. thanks.

Luckily it is a minor mistake and easy to fix.

Thanks for the patch, Peter.

yu ning