Re: [PATCH] dev: use name hash for dev_seq_ops

[Stephen Hemminger <shemminger@vyatta.com>]

On Fri, 21 Oct 2011 09:45:10 +0300
Mihai Maruseac <mihai.maruseac@gmail.com> wrote:

> Instead of using the dev->next chain and trying to resync at each call to
> dev_seq_start, use the name hash, keeping the bucket and the offset in
> seq->private field.
> 
> Tests revealed the following results for ifconfig > /dev/null
> 	* 1000 interfaces:
> 		* 0.114s without patch
> 		* 0.089s with patch
> 	* 3000 interfaces:
> 		* 0.489s without patch
> 		* 0.110s with patch
> 	* 5000 interfaces:
> 		* 1.363s without patch
> 		* 0.250s with patch
> 	* 128000 interfaces (other setup):
> 		* ~100s without patch
> 		* ~30s with patch
> 
> Signed-off-by: Mihai Maruseac <mmaruseac@ixiacom.com>
> ---
>  net/core/dev.c |   84 ++++++++++++++++++++++++++++++++++++++++++++++----------
>  1 files changed, 69 insertions(+), 15 deletions(-)
> 
> diff --git a/net/core/dev.c b/net/core/dev.c
> index 70ecb86..6edbcc5 100644
> --- a/net/core/dev.c
> +++ b/net/core/dev.c
> @@ -4041,6 +4041,60 @@ static int dev_ifconf(struct net *net, char __user *arg)
>  }
>  
>  #ifdef CONFIG_PROC_FS
> +
> +#define BUCKET_SPACE (32 - NETDEV_HASHBITS)
> +
> +struct dev_iter_state {
> +	struct seq_net_private p;
> +	unsigned int pos; /* bucket << BUCKET_SPACE + offset */
> +};
> +
> +#define get_bucket(x) ((x) >> BUCKET_SPACE)
> +#define get_offset(x) ((x) & ((1 << BUCKET_SPACE) - 1))
> +#define set_bucket_offset(b, o) ((b) << BUCKET_SPACE | (o))
> +
> +static inline struct net_device *dev_from_same_bucket(struct seq_file *seq)
>

Why are all these function marked inline? They are big, hardly hot path
and better to not continue the bad practice of inlining too much code.