Re-2: Why erase_pending_blocks in write_super? (was: [PATCH] JFFS2: Erase max 3 EB's in ...)

[Kai-Uwe Bloem]

> Just a note here to you, I only see these delays at umount time. A rm in 
> normal
> operation doesn't block. This is because my NOR flash supports erase 
> suspend.
> If you see this during normal operation it is because your flash/driver 
> doesn't
> support erase suspend. cms_set0002 just recently got support for erase 
> suspend.

No, it uses cfi_cmdset_0001.c. That has cond_resched in cfi_udelay which should prevent this if the process priorities are set accordingly.

I suspect that a sync() operation also triggers a write_super. At least I can see that sync needs a long time after erasing a large file, although the jffs2 sync method only flushes the write buffer.


To: joakim.tjernlund at transmode.se
Cc: linux-mtd at lists.infradead.org