| To: | Christoph Lameter <christoph@xxxxxxxxxxx> |
|---|---|
| Subject: | Re: [PATCH] NUMA aware allocation of transmit and receive buffers for e1000 |
| From: | Andi Kleen <ak@xxxxxx> |
| Date: | 30 May 2005 11:41:07 +0200 |
| Date: | Mon, 30 May 2005 11:41:07 +0200 |
| Cc: | Eric Dumazet <dada1@xxxxxxxxxxxxx>, netdev@xxxxxxxxxxx, akpm@xxxxxxxx, shai@xxxxxxxxxxxx |
| In-reply-to: | <Pine.LNX.4.62.0505271413560.25484@graphe.net> |
| References: | <Pine.LNX.4.62.0505171854490.20408@graphe.net> <m1sm0l5lwa.fsf@muc.de> <428AF378.5020109@cosmosbay.com> <Pine.LNX.4.62.0505271413560.25484@graphe.net> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mutt/1.4.1i |
On Fri, May 27, 2005 at 02:15:55PM -0700, Christoph Lameter wrote: > On Wed, 18 May 2005, Eric Dumazet wrote: > > > Related question, is pci_alloc_consistent() already NUMA aware ? > > Nope. It also ultimately acquires memory via vmalloc. x86-64 vmalloc has been NUMA aware forever. I believe the HP BigTux patchkit for IA64 added it there too, although it might not have hit mainline. > > pci_alloc_consistent will also need some rework to allocate memory > consistent (well the name is coherent right?) with the node that the > device is on. Second case of a need for a node aware vmalloc? You can already have it today by temporarily changing the process mempolicy. However for pci_alloc_consistent just using alloc_pages_node directly seems to work fine. -Andi |
| Previous by Date: | Re: [PATCH 2/2] Resend: LSM-IPSec Networking Hooks, James Morris |
|---|---|
| Next by Date: | Re: [PATCH] use mmiowb in tg3_poll, Arthur Kepner |
| Previous by Thread: | Re: [PATCH] NUMA aware allocation of transmit and receive buffers for e1000, Christoph Lameter |
| Next by Thread: | [PATCH] Super TSO, David S. Miller |
| Indexes: | [Date] [Thread] [Top] [All Lists] |