Re: [PATCH] a.out don't exec over NFS

• Messages sorted by: [ date ][ thread ][ subject ][ author ]
• Next message: Doug Ledford: "Re: SCSI access creating lost time"
• Previous message: Alan Cox: "Re: [PATCH] a.out don't exec over NFS"
• In reply to: Andrea Arcangeli: "Re: select()/socket has problems under 2.2.x."

> > My patch reverts NFS client to 2.0.x behaviour. I can provide a patch
> > that changes only s_blocksize on NFS mounted filesystem.
>
> As I said benchmark it.
>
> Right now stat() returns the superblock s_blocksize field and that is used
> by the higher layers to govern the I/O writing size. If you change this
> to a low value you trash your entire performance, especially versus non
> Linux NFS servers.
>
> To get the right effect you need to decouple the stat() returns and the
> a.out file check.
>

I got the point, and try to work on a.out checks instead of NFS client.

Jan

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/

• Next message: Doug Ledford: "Re: SCSI access creating lost time"
• Previous message: Alan Cox: "Re: [PATCH] a.out don't exec over NFS"
• In reply to: Andrea Arcangeli: "Re: select()/socket has problems under 2.2.x."