| To: | "David S. Miller" <davem@xxxxxxxxxx> |
|---|---|
| Subject: | Re: Tigon3 5701 PCI-X recv performance problem |
| From: | John Partridge <johnip@xxxxxxx> |
| Date: | Wed, 08 Oct 2003 14:02:57 -0500 |
| Cc: | netdev@xxxxxxxxxxx, jgarzik@xxxxxxxxx, jes@xxxxxxx |
| In-reply-to: | <20031008112657.2adddeb6.davem@xxxxxxxxxx> |
| References: | <3F844578.40306@xxxxxxx> <20031008101046.376abc3b.davem@xxxxxxxxxx> <3F844EF9.9090901@xxxxxxx> <20031008112657.2adddeb6.davem@xxxxxxxxxx> |
| Reply-to: | johnip@xxxxxxx |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mozilla/5.0 (X11; U; Linux i686; en-US; rv:1.5b) Gecko/20030905 Thunderbird/0.2 |
OK, fair enough, you mean like :-
#if defined(__ia64__)
if (len > RX_COPY_THRESHOLD && tp->rx_offset == 2) {
#else
if (len > RX_COPY_THRESHOLD) {
#endif
David S. Miller wrote:
On Wed, 08 Oct 2003 12:52:57 -0500 John Partridge <johnip@xxxxxxx> wrote:Not too much differentThe problem is that your change is arch-dependant yet you make it run on all platforms. On x86 we don't want to do what your change is doing, the unaligned accesses are cheap enough. We need to abstract this, probably in the same way it is done in the Tulip and other drivers which have similar issues. -- John Partridge Silicon Graphics Inc Tel: 651-683-3428 Vnet: 233-3428 E-Mail: johnip@xxxxxxx |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Tigon3 5701 PCI-X recv performance problem, Andi Kleen |
|---|---|
| Next by Date: | Re: Tigon3 5701 PCI-X recv performance problem, Steve Modica |
| Previous by Thread: | Re: Tigon3 5701 PCI-X recv performance problem, David S. Miller |
| Next by Thread: | Re: Tigon3 5701 PCI-X recv performance problem, Steve Modica |
| Indexes: | [Date] [Thread] [Top] [All Lists] |