[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: I/O error copying file to jffs, no obvous problem in jffs log?



Ok, here it is. It seems that it fails inside "jffs_do_readpage_nolock()"
(with an error of -5 (EIO, I think))
which is called from jffs_prepare_write().

Vipin





Apr 15 16:25:44 localhost kernel: jffs_write_node(): filename = "umount", ino
= 213, total_size = 4164
Apr 15 16:25:44 localhost kernel: jffs_fmalloc(): fmc = 0xc15ad110, size =
4164, node = 0xc3cb8db0
Apr 15 16:25:44 localhost kernel: jffs_write_node(): setting version of umount
to 5
Apr 15 16:25:44 localhost kernel: jffs_insert_node(): ino = 213, version = 5,
name = "", deleted = 0
Apr 15 16:25:44 localhost kernel: jffs_commit_Write(): Leaving with return
value=0
Apr 15 16:25:44 localhost kernel: jffs_prepare_write():Coming in.
filep=c301eed0, page=c109be8c, from=0, to=4096
Apr 15 16:25:44 localhost kernel: jffs_prepare_write(): ...Leaving, with
return value 0.
Apr 15 16:25:44 localhost kernel: jffs_commit_Write():
Entering:filep=c301eed0, page=c109be8c, from=0, to=4096
Apr 15 16:25:44 localhost kernel: ***jffs_file_write(): inode: 0xc3977830
(ino: 213), filp: 0xc301eed0, buf: 0xc24af000, count: 4
096
Apr 15 16:25:44 localhost kernel: jffs_write_node(): filename = "umount", ino
= 213, total_size = 4164
Apr 15 16:25:44 localhost kernel: jffs_fmalloc(): fmc = 0xc15ad110, size =
4164, node = 0xc3cb8d80
Apr 15 16:25:44 localhost kernel: jffs_write_node(): setting version of umount
to 6
Apr 15 16:25:44 localhost kernel: jffs_insert_node(): ino = 213, version = 6,
name = "", deleted = 0
Apr 15 16:25:44 localhost kernel: jffs_commit_Write(): Leaving with return
value=0
Apr 15 16:25:44 localhost kernel: jffs_prepare_write():Coming in.
filep=c301eed0, page=c106d8a8, from=0, to=4096
Apr 15 16:25:44 localhost kernel: jffs_prepare_write(): ...Leaving, with
return value 0.
Apr 15 16:25:44 localhost kernel: jffs_commit_Write():
Entering:filep=c301eed0, page=c106d8a8, from=0, to=4096
Apr 15 16:25:44 localhost kernel: ***jffs_file_write(): inode: 0xc3977830
(ino: 213), filp: 0xc301eed0, buf: 0xc19c6000, count: 4
096
Apr 15 16:25:44 localhost kernel: jffs_write_node(): filename = "umount", ino
= 213, total_size = 4164
Apr 15 16:25:44 localhost kernel: jffs_fmalloc(): fmc = 0xc15ad110, size =
4164, node = 0xc3cb8d50
Apr 15 16:25:44 localhost kernel: jffs_write_node(): setting version of umount
to 7
Apr 15 16:25:44 localhost kernel: jffs_insert_node(): ino = 213, version = 7,
name = "", deleted = 0
Apr 15 16:25:44 localhost kernel: jffs_commit_Write(): Leaving with return
value=0
Apr 15 16:25:44 localhost kernel: jffs_prepare_write():Coming in.
filep=c301eed0, page=c109be8c, from=0, to=2124
Apr 15 16:25:44 localhost kernel: ***jffs_readpage(): file = "umount",
page->index = 6
Apr 15 16:25:44 localhost kernel: jffs_prepare_write(): about to return back
from inside if() with return val=-5





David Woodhouse wrote:

> vmalik@xxxxxxx.com said:
> >  Ok, here it is. It seems (from the log) that only intergral number of
> > 4096 bytes get written. Any data write less than this number (which
> > would be the last write) fails.
>
> On an arch with 4KiB pages, yes? Looks like commit_write() is failing if it
> doesn't write a complete page. Can you add debugging printks to the start
> and end of readpage, prepare_write and commit_write()?
>
> --
> dwmw2
>
> To unsubscribe from this list: send the line "unsubscribe jffs-dev" in
> the body of a message to majordomo@xxxxxxx.com


To unsubscribe from this list: send the line "unsubscribe jffs-dev" in
the body of a message to majordomo@xxxxxxx.com