Re: Performance issue when trashing files (backtraced to fsync)

[Alexander Larsson <alexl redhat com>]

On Tue, 2009-08-11 at 16:46 +0200, Alexander Larsson wrote:
> 3. Extend the checking for existing file case to check if the existing
>    file is 0 bytes (would be the case here) and in that case avoid
>    fsync.

I just pushed a fix doing this, so the trash case is fixed. However, the
question still stands about letting other code that is not as simple
avoid the cost of fsync if it wants.