I am developing a USB IP core, and I am now at the stage where I am attempting to create a uClinux driver. With the new driver compiled into the kernel, and debug messages enabled, I see the following when I insert a flash stick. console text (http://www.base2designs.com/console_text.txt) As far as I can tell, when the flash stick is inserted, it is correctly detected, but then again I&#39;m not too sure what all the messages mean. Could somebody take a look at the console messages, and tell me if there are any problems that I should be investigating. If all is well, and the flash stick is indeed correctly attached, how do I get read write access. I&#39;ve read in places that I should add a line to my /etc/fstab file, but I cannot find the fstab file. I tried; mount -n /dev/sda /mnt/ide0 but no success. cat /proc/partitions major minor # blocks name 3 0 62976 hda 3 1 62960 hda1 31 0 4096 mtdblock0 31 1 2048 mtdblock1 31 2 1024 mtdblock2 31 3 1024 mtdblock3 8 0 63488 sda 8 1 63457 sda1# mount -n /dev/sda /mnt/ide0 mount -n /dev/sda /mnt/ide0 mount: special device /dev/sda does not exist

--- Quote Start --- originally posted by sfielding@Jun 22 2005, 03:38 PM # mount -n /dev/sda /mnt/ide0 mount -n /dev/sda /mnt/ide0 mount: special device /dev/sda does not exist --- Quote End --- You need to mount the partition, not the device, try mount -n -t vfat /dev/sda1 /mnt/usb

Thanks marc, that helped. It turned out that I also needed to add some nodes to the filesystem; @sda,b,8,0 @sda1,b,8,1 Hopefully I got this correct? So, now when I mount the USB partition I get mount -n -t vfat /dev/sda1 /mnt/usb FAT: codepage cp437 not found mount: wrong fs type, bad option, bad superblock on /dev/sda1, or too many mounted file systems# What could cause "FAT: codepage cp437 not found" ? I believe that the formatting is correct; Disk /dev/sda: 65 MB, 65011712 bytes 2 heads, 62 sectors/track, 1024 cylinders Units = cylinders of 124 * 512 = 63488 bytes Device Boot Start End Blocks Id System /dev/sda1 1 1024 63457 b Win95 FAT32

--- Quote Start --- originally posted by sfielding+jun 23 2005, 05:49 am--><div class='quotetop'>quote (sfielding @ jun 23 2005, 05:49 am)</div> --- quote start --- thanks marc, that helped. it turned out that i also needed to add some nodes to the filesystem; @sda,b,8,0 @sda1,b,8,1 hopefully i got this correct?[/b] --- quote end --- looks right. $ ls -al /dev/sda* brw-rw---- 1 root disk 8, 0 2002-03-14 22:51 /dev/sda brw-rw---- 1 root disk 8, 1 2002-03-14 22:51 /dev/sda1 <!--quotebegin-sfielding@Jun 23 2005, 05:49 AM so, now when i mount the usb partition i get mount -n -t vfat /dev/sda1 /mnt/usb fat: codepage cp437 not found mount: wrong fs type, bad option, bad superblock on /dev/sda1, or too many mounted file systems# what could cause "fat: codepage cp437 not found" ? --- Quote End --- This looks to my like you either did not compile vfat support in your kernel $ cat /usr/src/linux-2.6.12/.config |grep VFAT CONFIG_VFAT_FS=m (or =y in case you want it compiled in instead of a module), or some NLS settings: $ cat /usr/src/linux-2.6.12/.config |grep 437 CONFIG_FAT_DEFAULT_CODEPAGE=437 CONFIG_NLS_DEFAULT="cp437" CONFIG_NLS_CODEPAGE_437=y but since I never actively remember setting them myself, I would assume they are set by selecting the VFAT configuration. --- Quote Start --- originally posted by sfielding@Jun 23 2005, 05:49 AM i believe that the formatting is correct; disk /dev/sda: 65 mb, 65011712 bytes 2 heads, 62 sectors/track, 1024 cylinders units = cylinders of 124 * 512 = 63488 bytes device boot start end blocks id system /dev/sda1 1 1024 63457 b win95 fat32 --- Quote End --- looks right, at least the default crappy formatting of USB sticks seems already cleared http://forum.niosforum.com/work2/style_emoticons/<#EMO_DIR#>/laugh.gif

You can find the codepage stuff in the kernel configuration under File systems ---> Native Language Support ---> Dennis Scott Microtronix Datacom

It works now! I added the approprite character set support to the kernel, and all is well. Thanks for your help marc and dennis. It is much appreciated.

Mounting usb flash stick | Altera Community

14 Replies

Altera_Forum
Honored Contributor
21 years ago
........
scsi0 : SCSI emulation for USB Mass Storage devices
Vendor: F3D6368? Model: ?Fash Disk 1.6? Rev: !?
Type: Direct-Access ANSI SCSI revision: 02
SCSI device sda: 129281 512-byte hdwr sectors (66 MB)
sda: assuming Write Enabled
sda: assuming drive cache: write through
/dev/scsi/host0/bus0/target0/lun0:<6>check_part[0]=17241628:<7>usb-storage: que
uecommand called
res=0:<6>check_part[1]=17248964:<6>res=0: unknown partition table
Attached scsi removable disk sda at scsi0, channel 0, id 0, lun 0

hi! i added sl811 to altera cyc board, and run uclinux .
the"unknown partition table" have troubled me for a few days.
do u know how to solve it???
thank u!!
Altera_Forum
Honored Contributor
21 years ago
# fdisk -l
/dev/ide/host0/bus0/target0/lun0:<6>check_part[0]=17241628:<6>res=0:<6>check_pa
rt[1]=17248964: p1
res=1:
Disk /dev/hda: 4 heads, 32 sectors, 246 cylinders
Units = <7>usb-storage: queuecommand called
cylinders of 128 * 512 bytes

Device Boot Start End Blocks Id System
/dev/hda1 1 246 15728 83 Linux

Disk /dev/sda: 3 heads, 43 sectors, 1002 cylinders
Units = cylinders of 129 * 512 bytes

Disk /dev/sda doesn't contain a valid partition table#
Altera_Forum
Honored Contributor
21 years ago
Hello cetic,

<div class='quotetop'>QUOTE </div>
--- Quote Start ---
you say in the opencores project that the 50MHz oscillator on the Altera Nios evalboard must be changed by a 48MHz oscillator. I suppose this is because the USB ip in the FPGA must run at a specific frequency that cannot be generated by a PLL from the original 50MHz. Is this true ? If yes, what is that frequency ?[/b]
--- Quote End ---

The FPGA must run at 48MHz. If you only want to use low speed mode, then you can get away with using the PLL to generate 47.62MHz, but for full speed mode the USB 1.1 spec requires 0.05% clock tolerance.
Altera_Forum
Honored Contributor
21 years ago
Hello sfielding,

thanks for your answer. On my custom board, I have a 20MHz oscillator, and with the M/N ratio of a PLL, I cannot generate 48MHz. Do you think that it is possible to put 2 PLLs in series to get the 48MHz ? Is this achievable and stable ? I am using a cyclone C6.

Thanks

Best regards

pod

Forum Discussion

Mounting usb flash stick

14 Replies

Recent Discussions

Nios V/g load-writeback hazard

Run-time errors using EMIF in Agilex 5 NIOS-V design

Ashling RISC Free IDE fails to download ELF file

Dual Port Nios V BSP error

set_output_delay explained for dummies