Skip to contentSkip to conversation

Is my flash being detected?

DISCUSSION [email protected] 3 participants 6 messages Jun 22, 2001 Memory technology devices (MTD)

lore
Share & tools
lore

[email protected]

author
#0 lore raw

Hi freinds,
I am trying to use mtd on L7205sdb.The flash chip used is AM29LV160DB.
The flash size is 16 MB in byte mode and processor data bus being 32 bits wide. The .config for MTD is as below
CONFIG_MTD=y
CONFIG_MTD_DEBUG=y
CONFIG_MTD_DEBUG_VERBOSE=3
CONFIG_MTD_CFI=y
CONFIG_MTD_CFI_AMDEXT=y
CONFIG_MTD_ROM=y
CONFIG_MTD_PHYSMAP=y
CONFIG_MTD_PHYSMAP_START=000000
CONFIG_MTD_PHYSMAP_LEN=10000
CONFIG_MTD_PHYSMAP_BUSWIDTH=4

It does say that AMd flash found but at the end it complains that "cfi_probe: No supported Vendor Command Set found"

Given below are my bootup messages

Linux version 2.2.1-rmk2 ([email protected]) (libraries: /usr/local/arm/lib/gcc-lib/arm-linux/egcs-2.91.66/:/usr/lib/gcc/arm-linux/egcs-2.91.66/:/usr/local/arm/arm-linux/lib/arm-linux/egcs-2.91.66/:/usr/local/arm/arm-linux/lib/) #916 Fri Jun 22 17:51:46 IST 2001
start_kernl: Before setup_arch, 00000000-00000000

#1 lore raw

Make sure that you have included the cfi_cmdset_002.c module.

Vipin

Quoted 26 lines · [email protected] wrote:

Hi freinds,
I am trying to use mtd on L7205sdb.The flash chip used is AM29LV160DB.
The flash size is 16 MB in byte mode and processor data bus being 32 bits wide. The .config for MTD is as below
CONFIG_MTD=y
CONFIG_MTD_DEBUG=y
CONFIG_MTD_DEBUG_VERBOSE=3
CONFIG_MTD_CFI=y
CONFIG_MTD_CFI_AMDEXT=y
CONFIG_MTD_ROM=y
CONFIG_MTD_PHYSMAP=y
CONFIG_MTD_PHYSMAP_START=000000
CONFIG_MTD_PHYSMAP_LEN=10000
CONFIG_MTD_PHYSMAP_BUSWIDTH=4

It does say that AMd flash found but at the end it complains that "cfi_probe: No supported Vendor Command Set found"

Given below are my bootup messages

Linux version 2.2.1-rmk2 ([email protected]) (libraries: /usr/local/arm/lib/gcc-lib/arm-linux/egcs-2.91.66/:/usr/lib/gcc/arm-linux/egcs-2.91.66/:/usr/local/arm/arm-linux/lib/arm-linux/egcs-2.91.66/:/usr/local/arm/arm-linux/lib/) #916 Fri Jun 22 17:51:46 IST 2001
start_kernl: Before setup_arch, 00000000-00000000

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

author
#2 lore raw

Hi ,
I got the mistake the fucntion check_cmd_set in cfi_probe.c assigns the probe function using inter_module_get_request .Since my kernel does not have module support the function returns null.For the time being I have assigned the address of the function cfi_cmdset_0002 to probe

Now I get the message
AMD/Fujitsu Extended Query table at 0x0040 number of CFI chips 1
mtd:Giving out device 0 to Physically mapped flash :)

so far so good but if I try to read from the flash uisng cp /dev/mtd0 filename Iget the following messages
MTD_OPEN
MTD_READ
unable to handle kernel paging request at virtual address 8f80000

I think I know the reason this may be the readb from _memcpy_form_io may be trying to read from a wrong addres as the readb reads form PCMCIA_IO and I think I know the way to fix this .
I can't use the latest kernel so I am forced to used the mtd on a older kernel.

thanks for the suggestions,

Regards,
Ashwin

#3 lore raw
Quoted 5 lines · [email protected] said:

I got the mistake the fucntion check_cmd_set in cfi_probe.c assigns
the probe function using inter_module_get_request .Since my kernel does not have module support the function returns null.For the time being I have assigned the address of the function cfi_cmdset_0002 to probe

Fixed in CVS. If the command set code is built-in, it get called directly. The inter_module_xxx crap in 2.4 is, well, crap. :)

Quoted 4 lines · [email protected] said:

unable to handle kernel paging request at virtual address 8f80000 I think I know the reason this may be the readb from _memcpy_form_io may be trying to read from a wrong addres as the readb reads form PCMCIA_IO and I think I know the way to fix this .

Is the flash paged - allowing the CPU to see only a 0x80000-byte region at a time? There are examples (octagon-5066.c) of how to deal with this.

Quoted 2 lines

I can't use the latest kernel so I am forced to used the mtd on a
older kernel.

The code in CVS should work on older kernels. Please use it.

--
dwmw2

#4 lore raw
In reply to [email protected]

Ok, I hit send too quickly. If you flash is not CFI, then make sure that the JDEC module is included.

Vipin

Quoted 26 lines · [email protected] wrote:

Hi freinds,
I am trying to use mtd on L7205sdb.The flash chip used is AM29LV160DB.
The flash size is 16 MB in byte mode and processor data bus being 32 bits wide. The .config for MTD is as below
CONFIG_MTD=y
CONFIG_MTD_DEBUG=y
CONFIG_MTD_DEBUG_VERBOSE=3
CONFIG_MTD_CFI=y
CONFIG_MTD_CFI_AMDEXT=y
CONFIG_MTD_ROM=y
CONFIG_MTD_PHYSMAP=y
CONFIG_MTD_PHYSMAP_START=000000
CONFIG_MTD_PHYSMAP_LEN=10000
CONFIG_MTD_PHYSMAP_BUSWIDTH=4

It does say that AMd flash found but at the end it complains that "cfi_probe: No supported Vendor Command Set found"

Given below are my bootup messages

Linux version 2.2.1-rmk2 ([email protected]) (libraries: /usr/local/arm/lib/gcc-lib/arm-linux/egcs-2.91.66/:/usr/lib/gcc/arm-linux/egcs-2.91.66/:/usr/local/arm/arm-linux/lib/arm-linux/egcs-2.91.66/:/usr/local/arm/arm-linux/lib/) #916 Fri Jun 22 17:51:46 IST 2001
start_kernl: Before setup_arch, 00000000-00000000

______________________________________________________
Linux MTD discussion mailing list
http://lists.infradead.org/mailman/listinfo/linux-mtd/

#5 lore raw
Quoted 5 lines · [email protected] said:

Ok, I hit send too quickly. If you flash is not CFI, then make sure that the JDEC module is included.

Quoted 2 lines

It does say that AMd flash found but at the end it complains that "cfi_probe: No supported Vendor Command Set found"

It looks like it's CFI. The code being used is very old though - CONFIG_MTD_CFI_AMDEXT was corrected to CONFIG_MTD_CFI_CMDSTD a long time ago. Please update to the latest code and send the full output.

--
dwmw2