Messages in this thread | | | Date | Mon, 14 Jul 2008 10:03:29 -0400 (EDT) | From | Mikulas Patocka <> | Subject | Re: [SUGGESTION]: drop virtual merge accounting in I/O requests |
| |
On Mon, 14 Jul 2008, David Miller wrote:
> From: Mikulas Patocka <mpatocka@redhat.com> > Date: Mon, 14 Jul 2008 08:16:13 -0400 (EDT) > >> As you mentioned ESP driver, it declares .sg_tablesize = SG_ALL, so >> BIO_VMERGE_BOUNDARY has no effect on the operation of this driver. Any >> other driver where BIO_VMERGE_BOUNDARY does matter? > > When BIO_VMERGE_BOUNDARY is turned on, requests that would not > otherwise fit into the device's limits, can.
Why would someone want to overshoot SG_ALL? ... and, shouldn't the constant be increased then --- instead of making buggy BIO_VMERGE_BOUNDARY expectations?
Mikulas
> It's the only way to know ahead of time that the IOMMU is going > to merge things for us.
| |