[U-Boot] Problem with SD card and ext2ls

I have an SD card with an ext2 partition and I am trying to access it through ext2ls command but this is what I get:
# ext2ls mmc 0:1 / ** Bad partition - mmc 0:1 **
I have read some previous mails of some people complaining about a similar problem: http://www.mail-archive.com/u-boot@lists.denx.de/msg12776.html
Here also Wolfgang Denk says that ext2 is currently broken in u-boot for newer ext2 images.
Is this still true or has it been fixed since then?
Thank you.

Also if I issue a "mmc rescan 0" I get the following:
# mmc rescan 0 # ext2ls mmc 0:1 / get partition info called raise: Signal # 8 caught raise: Signal # 8 caught raise: Signal # 8 caught
What could be happening here?
Thank you.

Looking deeply in the list I found someone that had the same problem with an AT91: http://www.mail-archive.com/u-boot@lists.denx.de/msg23442.html
I submit here all useful information about my case.
Board: imx27lite Toolchains: eldk-4.2 GIT commit: f67066b6b0740b826ed862615c5ab022aaf4779a SD card formatted in ext2 with gparted.
How to reproduce:
=> mmc rescan 0 => ext2ls mmc 0:1 / <2, 0, 204> ext2fs read inode 2 ext2fs read 0 group descriptor (blkno 2 blkoff 0) <4, 0, 32> raise: Signal # 8 caught ext2fs read inode blkno 0 blkoff 0 raise: Signal # 8 caught raise: Signal # 8 caught ext2fs read inode blkno 6 blkoff 0 <12, 0, 128>

Dear javier Martin,
In message eedb5540911200403r5a4dfb8el7fd73965f33a7c4f@mail.gmail.com you wrote:
Also if I issue a "mmc rescan 0" I get the following:
# mmc rescan 0 # ext2ls mmc 0:1 / get partition info called raise: Signal # 8 caught raise: Signal # 8 caught raise: Signal # 8 caught
What could be happening here?
Sorry, the batteries of my crystal ball are running low...
It _might_ help if you provided a _little_ information, at least the most vital pieces like which architecture / board you are speaking of, and which exact version (which git commit id) you are running.
Best regards,
Wolfgang Denk

Sorry, the batteries of my crystal ball are running low...
It _might_ help if you provided a _little_ information, at least the most vital pieces like which architecture / board you are speaking of, and which exact version (which git commit id) you are running.
Please see my last mail. I sent three mails without reply and maybe this lead to some confusion, sorry. In that, my previous mail, you have all the information: Board, SD card, how to reproduce...
Thank you.

Dear javier Martin,
In message eedb5540911230109k9b3b6f7ibc261b6f6f4c838f@mail.gmail.com you wrote:
Sorry, the batteries of my crystal ball are running low...
It _might_ help if you provided a _little_ information, at least the most vital pieces like which architecture / board you are speaking of, and which exact version (which git commit id) you are running.
Please see my last mail. I sent three mails without reply and maybe this lead to some confusion, sorry. In that, my previous mail, you have all the information: Board, SD card, how to reproduce...
Found it, indeed. Thanks.
Best regards,
Wolfgang Denk

Dear javier Martin,
In message eedb5540911200347t5f11bdcat9206e2d9f1a7f971@mail.gmail.com you wrote:
I have an SD card with an ext2 partition and I am trying to access it through ext2ls command but this is what I get:
# ext2ls mmc 0:1 / ** Bad partition - mmc 0:1 **
I have read some previous mails of some people complaining about a similar problem: http://www.mail-archive.com/u-boot@lists.denx.de/msg12776.html
Here also Wolfgang Denk says that ext2 is currently broken in u-boot for newer ext2 images.
Well, that was a long, long time ago. And the problem reported there (and fixed since) was about newer (bigger) inode sizes - this has nothing to do with "Bad partition" errors.
Is this still true or has it been fixed since then?
The problems referred to in the posting you quote have long been fixed.
Best regards,
Wolfgang Denk

Hello,
Here also Wolfgang Denk says that ext2 is currently broken in u-boot for newer ext2 images.
I rather would say it is broken for rev 0, i.e. old ext2 images and images generated with genext2fs
Is this still true or has it been fixed since then?
I just posted a patch: [PATCH] EXT2FS: fix inode size for ext2fs rev#0.
Best regards,
Michael.
participants (3)
-
javier Martin
-
Michael Brandt
-
Wolfgang Denk