| To: | Scott Feldman <sfeldma@xxxxxxxxx> |
|---|---|
| Subject: | Re: Allow IP header alignment to be overriden |
| From: | Anton Blanchard <anton@xxxxxxxxx> |
| Date: | Sat, 12 Jun 2004 00:23:37 +1000 |
| Cc: | "David S. Miller" <davem@xxxxxxxxxx>, netdev@xxxxxxxxxxx |
| In-reply-to: | <1086939562.3657.10.camel@sfeldma-mobl2.dsl-verizon.net> |
| References: | <20040611012727.GA27672@krispykreme> <20040610223549.5e9ad025.davem@redhat.com> <1086939562.3657.10.camel@sfeldma-mobl2.dsl-verizon.net> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mutt/1.5.6+20040523i |
Hi, > Would replacing "2" with a macro that's defined on a per-arch basis > work? Nice idea. This would avoid us adding useless padding in that case, as we currently do: reserve_len = 2; skb = dev_alloc_skb(adapter->rx_buffer_len + reserve_len); skb_reserve(skb, reserve_len); which would be nice if we are using a power of 2 buffer size. Would creating: /* * Network drivers want to align IP headers. Since we have 14 bytes of * ethernet header, adding 2 bytes will align the IP header. However * this will mean we do unaligned DMA so there is a trade off. * * We allow this to be overridden per arch as the unaligned DMA cost may * outweigh the unaligned CPU cost. */ #ifndef NET_IP_ALIGN #define NET_IP_ALIGN 2 #endif Instead of skb_align make more sense? It does have the advantage of removing another magic number. Anton |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Allow IP header alignment to be overriden, Anton Blanchard |
|---|---|
| Next by Date: | confusion about socket buffers, linux lover |
| Previous by Thread: | Re: Allow IP header alignment to be overriden, Scott Feldman |
| Next by Thread: | Re: Allow IP header alignment to be overriden, David S. Miller |
| Indexes: | [Date] [Thread] [Top] [All Lists] |