Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1292630

Re: [PATCH 5/6] bpf: hash: avoid to call kmalloc() in eBPF prog

Show all headers | View raw

On Tue, Dec 15, 2015 at 07:21:03PM +0800, Ming Lei wrote:
> kmalloc() is often a bit time-consuming, also
> one atomic counter has to be used to track the total
> allocated elements, which is also not good.
> 
> This patch pre-allocates element pool in htab_map_alloc(),
> then use percpu_ida to allocate one slot from the pool,
> then the runtime allocation/freeing cost can be decreased.
> 
> From my test, at least 10% fio throughput is improved in block
> I/O test when tools/biolatency of bcc(iovisor) is running.
> 
> Signed-off-by: Ming Lei <tom.leiming@gmail.com>

Looks very intersting as well.
Approach looks good.
If you can make a common allocation helper for this map and
for blk-mq would be even better.

> -	htab->elem_size = sizeof(struct htab_elem) +
> -			  round_up(htab->map.key_size, 8) +
> -			  htab->map.value_size;
> +	htab->elem_size = round_up(sizeof(struct htab_elem) +
> +				   round_up(htab->map.key_size, 8) +
> +				   htab->map.value_size,
> +				   cache_line_size());

this rounding to cache line is great for performance, but it's extra
memory upfront which may not be needed. The per-allocation is a classic
performance vs memory trade-off. In other cases it may hurt.
So could you change the patch to do pre-allocation only when
requested by user space via extra flag for hash map or via
new BPF_MAP_TYPE_HASH_PREALLOC type? Not sure yet whether flag or
new type is better. I guess implementation will dictate.

PS
Glad that you found iovisor/tools/biolatency useful.
It's indeed pretty helpful to analyze real-time block io latency.

--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.kernel.org
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[PATCH 0/6] bpf: hash: optimization Ming Lei <tom.leiming@gmail.com> - 2015-12-15 12:30 +0100
  [PATCH 6/6] bpf: hash: reorganize 'struct htab_elem' Ming Lei <tom.leiming@gmail.com> - 2015-12-15 12:30 +0100
  [PATCH 5/6] bpf: hash: avoid to call kmalloc() in eBPF prog Ming Lei <tom.leiming@gmail.com> - 2015-12-15 12:30 +0100
    Re: [PATCH 5/6] bpf: hash: avoid to call kmalloc() in eBPF prog Alexei Starovoitov <alexei.starovoitov@gmail.com> - 2015-12-16 00:20 +0100
      Re: [PATCH 5/6] bpf: hash: avoid to call kmalloc() in eBPF prog Daniel Borkmann <daniel@iogearbox.net> - 2015-12-16 00:50 +0100
      Re: [PATCH 5/6] bpf: hash: avoid to call kmalloc() in eBPF prog Ming Lei <tom.leiming@gmail.com> - 2015-12-16 08:20 +0100
    Re: [PATCH 5/6] bpf: hash: avoid to call kmalloc() in eBPF prog Daniel Borkmann <daniel@iogearbox.net> - 2015-12-16 00:30 +0100
    Re: [PATCH 5/6] bpf: hash: avoid to call kmalloc() in eBPF prog Daniel Borkmann <daniel@iogearbox.net> - 2015-12-16 00:40 +0100
      Re: [PATCH 5/6] bpf: hash: avoid to call kmalloc() in eBPF prog Daniel Borkmann <daniel@iogearbox.net> - 2015-12-16 01:20 +0100
  [PATCH 2/6] hlist: prepare for supporting bit spinlock Ming Lei <tom.leiming@gmail.com> - 2015-12-15 12:30 +0100
  [PATCH 1/6] bpf: hash: use atomic count Ming Lei <tom.leiming@gmail.com> - 2015-12-15 12:30 +0100
  [PATCH 3/6] bpf: hash: move select_bucket() out of htab's spinlock Ming Lei <tom.leiming@gmail.com> - 2015-12-15 12:30 +0100

csiph-web