netdev
[Top] [All Lists]

Re: [PATCH] NUMA aware allocation of transmit and receive buffers for e1

To: christoph@xxxxxxxxxxx
Subject: Re: [PATCH] NUMA aware allocation of transmit and receive buffers for e1000
From: "David S. Miller" <davem@xxxxxxxxxxxxx>
Date: Tue, 17 May 2005 19:57:03 -0700 (PDT)
Cc: akpm@xxxxxxxx, linux-kernel@xxxxxxxxxxxxxxx, netdev@xxxxxxxxxxx
In-reply-to: <Pine.LNX.4.62.0505171941340.21153@xxxxxxxxxx>
References: <Pine.LNX.4.62.0505171854490.20408@xxxxxxxxxx> <20050517190343.2e57fdd7.akpm@xxxxxxxx> <Pine.LNX.4.62.0505171941340.21153@xxxxxxxxxx>
Sender: netdev-bounce@xxxxxxxxxxx
From: Christoph Lameter <christoph@xxxxxxxxxxx>
Subject: Re: [PATCH] NUMA aware allocation of transmit and receive buffers for 
e1000
Date: Tue, 17 May 2005 19:52:38 -0700 (PDT)

> > > - txdr->buffer_info = vmalloc(size);
> > > + txdr->buffer_info = kmalloc_node(size, GFP_KERNEL, node);
> > 
> > How come that this is safe to do
> 
> Because physically contiguous memory is usually better than virtually 
> contiguous memory? Any reason that physically contiguous memory will 
> break the driver?

The issue is whether size can end up being too large for
kmalloc() to satisfy, whereas vmalloc() would be able to
handle it.

<Prev in Thread] Current Thread [Next in Thread>