Skip to contentSkip to conversation

Loopback mounting jffs2 under UML?

uml-devel Rob Landley 4 participants 5 messages Dec 18, 2007 – Dec 19, 2007

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

I want to loopback mount a jffs2 filesystem, and this filesystem has incestuous knowledge of flash devices for wear levelling and stuff, so it will only mount an actual flash device and not a normal block device.

There's a way to make a loopback device look like a flash device, letting you loopback mount jffs2, but it involves recompiling the kerenel to add support for it, and of course I went "User Mode Linux"... Except that UML disables the flash menu entirely.

Basically, I need to be able to enable MTD_BLOCK so I can go:

losetup /dev/loop0 /path/to/imagefile.jffs2
insmod blkmtd erasesz=256 device=/dev/loop0
mount -t jffs2 /dev/mtdblock0 /mnt

This shouldn't need any actual flash hardware. It's just layering another translation layer on top of loopback.

Suggestions?

Rob

Signature · 13 lines
"One of my most productive days was throwing away 1000 lines of code."
  - Ken Thompson.

-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
#1 lore raw
Quoted 6 lines · Rob Landley

There's a way to make a loopback device look like a flash device, letting you loopback mount jffs2, but it involves recompiling the kerenel to add support for it, and of course I went "User Mode Linux"... Except that UML disables the flash menu entirely.

Suggestions?

I see this in menuconfig:

Symbol: MTD [=n]
Prompt: Memory Technology Device (MTD) support
Defined at drivers/mtd/Kconfig:3
Depends on: BROKEN && HAS_IOMEM

You could try removing the dependencies on BROKEN and HAS_IOMEM and see what happens...

Jeff

Signature · 12 lines
Work email - jdike at linux dot intel dot com

-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
#4 lore raw
In reply to Rob Landley

I did try UML + blkmtd and jffs statically compiled in a couple of years ago and it then almost worked out of the box. It failed then for me due to a very small issue: the initializers in the different drivers initialized in the wrong order, so I got a panic due to a null-dereference. (driver initialization does/didn't have depencies).

Just me five cents :-)

Stian

Quoted 15 lines · Rob Landley

I want to loopback mount a jffs2 filesystem, and this filesystem has incestuous knowledge of flash devices for wear levelling and stuff, so it will only mount an actual flash device and not a normal block device.

There's a way to make a loopback device look like a flash device, letting you loopback mount jffs2, but it involves recompiling the kerenel to add support for it, and of course I went "User Mode Linux"... Except that UML disables the flash menu entirely.

Basically, I need to be able to enable MTD_BLOCK so I can go:

losetup /dev/loop0 /path/to/imagefile.jffs2
insmod blkmtd erasesz=256 device=/dev/loop0
mount -t jffs2 /dev/mtdblock0 /mnt

Quoted 6 lines

This shouldn't need any actual flash hardware. It's just layering another translation layer on top of loopback.

Suggestions?

Rob

-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel

#2 lore raw
In reply to Jeff Dike
Quoted 10 lines · On Tuesday 18 December 2007 13:29:11 Jeff Dike wrote:
Quoted 6 lines · Rob Landley

There's a way to make a loopback device look like a flash device, letting you loopback mount jffs2, but it involves recompiling the kerenel to add support for it, and of course I went "User Mode Linux"... Except that UML disables the flash menu entirely.

Suggestions?

I see this in menuconfig:

Quoted 7 lines

Symbol: MTD [=n]
Prompt: Memory Technology Device (MTD) support
Defined at drivers/mtd/Kconfig:3
Depends on: BROKEN && HAS_IOMEM

You could try removing the dependencies on BROKEN and HAS_IOMEM and see what happens...

Actually, in -rc4 (what I have lying around), the depends is just on HAS_IOMEM, and the include in arch/um/Kconfig is guarded by "if BROKEN".

Building it, I hit the same darn error I've been patching my way around for months:

Quoted 10 lines

CC arch/um/kernel/smp.o
In file included from include/asm/arch/tlb.h:9,
from include/asm/tlb.h:4,
from arch/um/kernel/smp.c:8:
include/asm-generic/tlb.h: In function ‘tlb_flush_mmu’:
include/asm-generic/tlb.h:76: error: implicit declaration of function ‘release_pages’ include/asm-generic/tlb.h: In function ‘tlb_remove_page’:
include/asm-generic/tlb.h:105: error: implicit declaration of function ‘page_cache_release’ make[1]: *** [arch/um/kernel/smp.o] Error 1
make: *** [arch/um/kernel] Error 2

Patch that, and then it breaks with:

Quoted 6 lines

CC drivers/mtd/chips/chipreg.o
In file included from drivers/mtd/chips/chipreg.c:13:
include/linux/mtd/map.h: In function ‘inline_map_read’:
include/linux/mtd/map.h:378: error: implicit declaration of function ‘__raw_readb’ include/linux/mtd/map.h:380: error: implicit declaration of function ‘__raw_readw’ include/linux/mtd/map.h:382: error: implicit

Quoted 8 lines

declaration of function ‘__raw_readl’ include/linux/mtd/map.h:385: error:
implicit declaration of function ‘__raw_readq’ include/linux/mtd/map.h:388:
error: implicit declaration of function ‘memcpy_fromio’ include/linux/mtd/map.h: In function ‘inline_map_write’:
include/linux/mtd/map.h:406: error: implicit declaration of function ‘memcpy_toio’ make[3]: *** [drivers/mtd/chips/chipreg.o] Error 1 make[2]: *** [drivers/mtd/chips] Error 2
make[1]: *** [drivers/mtd] Error 2

All I enabled was CONFIG_MTD_BLOCK.

To reproduce what I did, yank these lines:

arch/um/Kconfig +0 -2
diff -ru linux-2.6.24-rc4/arch/um/Kconfig linux-temp/arch/um/Kconfig
@@ -280,9 +280,7 @@
 
 source "drivers/md/Kconfig"
 
-if BROKEN
 	source "drivers/mtd/Kconfig"
-endif
 
 #This is just to shut up some Kconfig warnings, so no prompt.
 config INPUT
arch/um/kernel/smp.c +1 -0
diff -ru linux-2.6.24-rc4/arch/um/kernel/smp.c linux-temp/arch/um/kernel/smp.c
@@ -5,6 +5,7 @@
 
 #include "linux/percpu.h"
 #include "asm/pgalloc.h"
+#include "linux/pagemap.h"
 #include "asm/tlb.h"
 
 /* For some reason, mmu_gathers are referenced when CONFIG_SMP is off. */
drivers/mtd/Kconfig +0 -1
diff -ru linux-2.6.24-rc4/drivers/mtd/Kconfig linux-temp/drivers/mtd/Kconfig
@@ -2,7 +2,6 @@
 
 menuconfig MTD
 	tristate "Memory Technology Device (MTD) support"
-	depends on HAS_IOMEM
 	help
 	  Memory Technology Devices are flash, RAM and similar chips, often
 	  used for solid state file systems on embedded devices. This option

And then configure and build like so:

cat > mini.conf << EOF
CONFIG_BINFMT_ELF=y
CONFIG_HOSTFS=y
CONFIG_BLK_DEV=y
CONFIG_BLK_DEV_LOOP=y CONFIG_STDERR_CONSOLE=y
CONFIG_UNIX98_PTYS=y
CONFIG_EXT2_FS=y
CONFIG_MTD=y
CONFIG_MTD_BLOCK=y
EOF
make ARCH=um allnoconfig KCONFIG_ALLCONFIG=mini.conf make ARCH=um

Rob

Signature · 13 lines
"One of my most productive days was throwing away 1000 lines of code."
  - Ken Thompson.

-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace
_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel
#3 lore raw
Quoted 7 lines · Rob Landley
Quoted 6 lines · On Tuesday 18 December 2007 13:29:11 Jeff Dike wrote:
Quoted 5 lines · Rob Landley

There's a way to make a loopback device look like a flash device, letting you loopback mount jffs2, but it involves recompiling the kerenel to add support for it, and of course I went "User Mode Linux"... Except that UML disables the flash menu entirely.

Quoted 22 lines
Quoted 5 lines
Quoted 1 line

Suggestions?

I see this in menuconfig:

Symbol: MTD [=n]

Actually, in -rc4 (what I have lying around), the depends is just on HAS_IOMEM, and the include in arch/um/Kconfig is guarded by "if BROKEN".

Quoted 14 lines

CC drivers/mtd/chips/chipreg.o
In file included from drivers/mtd/chips/chipreg.c:13:
include/linux/mtd/map.h: In function ‘inline_map_read’:
include/linux/mtd/map.h:378: error: implicit declaration of function ‘__raw_readb’ include/linux/mtd/map.h:380: error: implicit declaration of function ‘__raw_readw’ include/linux/mtd/map.h:382: error: implicit declaration of function ‘__raw_readl’ include/linux/mtd/map.h:385: error:
implicit declaration of function ‘__raw_readq’ include/linux/mtd/map.h:388:
error: implicit declaration of function ‘memcpy_fromio’ include/linux/mtd/map.h: In function ‘inline_map_write’:
include/linux/mtd/map.h:406: error: implicit declaration of function ‘memcpy_toio’ make[3]: *** [drivers/mtd/chips/chipreg.o] Error 1 make[2]: *** [drivers/mtd/chips] Error 2
make[1]: *** [drivers/mtd] Error 2

Protect all static inline routines in <linux/mtd/map.h> that use readb() and friends by #ifdef CONFIG_HAS_IOMEM?

Gr{oetje,eeting}s,

Geert

--
Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- [email protected]

In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that.
-- Linus Torvalds

-------------------------------------------------------------------------
SF.Net email is sponsored by:
Check out the new SourceForge.net Marketplace.
It's the best place to buy or sell services
for just about anything Open Source.
http://ad.doubleclick.net/clk;164216239;13503038;w?http://sf.net/marketplace

_______________________________________________
User-mode-linux-devel mailing list
[email protected]
https://lists.sourceforge.net/lists/listinfo/user-mode-linux-devel