[<prev] [next>] [thread-next>] [day] [month] [year] [list]
Message-ID: <20200207081810.3918919-1-kafai@fb.com>
Date: Fri, 7 Feb 2020 00:18:10 -0800
From: Martin KaFai Lau <kafai@...com>
To: <bpf@...r.kernel.org>
CC: Alexei Starovoitov <ast@...nel.org>,
Daniel Borkmann <daniel@...earbox.net>,
David Miller <davem@...emloft.net>, <kernel-team@...com>,
Linus Torvalds <torvalds@...ux-foundation.org>,
Linux-Sparse <linux-sparse@...r.kernel.org>,
Luc Van Oostenryck <luc.vanoostenryck@...il.com>,
<netdev@...r.kernel.org>, Randy Dunlap <rdunlap@...radead.org>
Subject: [PATCH bpf] bpf: Improve bucket_log calculation logic
It was reported that the max_t, ilog2, and roundup_pow_of_two macros have
exponential effects on the number of states in the sparse checker.
This patch breaks them up by calculating the "nbuckets" first so
that the "bucket_log" only needs to take ilog2().
Fixes: 6ac99e8f23d4 ("bpf: Introduce bpf sk local storage")
Reported-by: Randy Dunlap <rdunlap@...radead.org>
Reported-by: Luc Van Oostenryck <luc.vanoostenryck@...il.com>
Suggested-by: Linus Torvalds <torvalds@...ux-foundation.org>
Signed-off-by: Martin KaFai Lau <kafai@...com>
---
net/core/bpf_sk_storage.c | 5 +++--
1 file changed, 3 insertions(+), 2 deletions(-)
diff --git a/net/core/bpf_sk_storage.c b/net/core/bpf_sk_storage.c
index 458be6b3eda9..3ab23f698221 100644
--- a/net/core/bpf_sk_storage.c
+++ b/net/core/bpf_sk_storage.c
@@ -643,9 +643,10 @@ static struct bpf_map *bpf_sk_storage_map_alloc(union bpf_attr *attr)
return ERR_PTR(-ENOMEM);
bpf_map_init_from_attr(&smap->map, attr);
+ nbuckets = roundup_pow_of_two(num_possible_cpus());
/* Use at least 2 buckets, select_bucket() is undefined behavior with 1 bucket */
- smap->bucket_log = max_t(u32, 1, ilog2(roundup_pow_of_two(num_possible_cpus())));
- nbuckets = 1U << smap->bucket_log;
+ nbuckets = max_t(u32, 2, nbuckets);
+ smap->bucket_log = ilog2(nbuckets);
cost = sizeof(*smap->buckets) * nbuckets + sizeof(*smap);
ret = bpf_map_charge_init(&smap->map.memory, cost);
--
2.17.1
Powered by blists - more mailing lists