Re: [PATCH bpf-next 3/3] selftests/bpf: Add bloom map success test for userspace calls

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

 



On 10/29/21 3:04 PM, Yonghong Song wrote:

On 10/29/21 10:01 AM, Joanne Koong wrote:
This patch has two changes:
1) Adds a new function "test_success_cases" to test
successfully creating + adding + looking up a value
in a bloom filter map from the userspace side.

2) Use bpf_create_map instead of bpf_create_map_xattr in
the "test_fail_cases" to make the code look cleaner.

Signed-off-by: Joanne Koong <joannekoong@xxxxxx>

LGTM with one minor comment below.

Acked-by: Yonghong Song <yhs@xxxxxx>

---
  .../bpf/prog_tests/bloom_filter_map.c         | 53 ++++++++++++-------
  1 file changed, 33 insertions(+), 20 deletions(-)

diff --git a/tools/testing/selftests/bpf/prog_tests/bloom_filter_map.c b/tools/testing/selftests/bpf/prog_tests/bloom_filter_map.c
index 9aa3fbed918b..dbc0035e43e5 100644
--- a/tools/testing/selftests/bpf/prog_tests/bloom_filter_map.c
+++ b/tools/testing/selftests/bpf/prog_tests/bloom_filter_map.c
@@ -7,44 +7,32 @@
    static void test_fail_cases(void)
  {
-    struct bpf_create_map_attr xattr = {
-        .name = "bloom_filter_map",
-        .map_type = BPF_MAP_TYPE_BLOOM_FILTER,
-        .max_entries = 100,
-        .value_size = 11,
-    };
      __u32 value;
      int fd, err;
        /* Invalid key size */
-    xattr.key_size = 4;
-    fd = bpf_create_map_xattr(&xattr);
+    fd = bpf_create_map(BPF_MAP_TYPE_BLOOM_FILTER, 4, sizeof(value), 100, 0);       if (!ASSERT_LT(fd, 0, "bpf_create_map bloom filter invalid key size"))
          close(fd);
-    xattr.key_size = 0;
        /* Invalid value size */
-    xattr.value_size = 0;
-    fd = bpf_create_map_xattr(&xattr);
+    fd = bpf_create_map(BPF_MAP_TYPE_BLOOM_FILTER, 0, 0, 100, 0);
      if (!ASSERT_LT(fd, 0, "bpf_create_map bloom filter invalid value size 0"))
          close(fd);
-    xattr.value_size = 11;
        /* Invalid max entries size */
-    xattr.max_entries = 0;
-    fd = bpf_create_map_xattr(&xattr);
-    if (!ASSERT_LT(fd, 0, "bpf_create_map bloom filter invalid max entries size")) +    fd = bpf_create_map(BPF_MAP_TYPE_BLOOM_FILTER, 0, sizeof(value), 0, 0);
+    if (!ASSERT_LT(fd, 0,
+               "bpf_create_map bloom filter invalid max entries size"))

It is OK to have "bpf_create_map ..." in the same line as ASSERT_LT
for better readability and consistent with other ASSERT_LT. It is over 80 but less than 100 char's per line.

Great, I will send out v2 of this patchset where this line break is removed.
Thanks for reviewing the patchset!
          close(fd);
-    xattr.max_entries = 100;
        /* Bloom filter maps do not support BPF_F_NO_PREALLOC */
-    xattr.map_flags = BPF_F_NO_PREALLOC;
-    fd = bpf_create_map_xattr(&xattr);
+    fd = bpf_create_map(BPF_MAP_TYPE_BLOOM_FILTER, 0, sizeof(value), 100,
+                BPF_F_NO_PREALLOC);
      if (!ASSERT_LT(fd, 0, "bpf_create_map bloom filter invalid flags"))
          close(fd);
-    xattr.map_flags = 0;
  -    fd = bpf_create_map_xattr(&xattr);
+    fd = bpf_create_map(BPF_MAP_TYPE_BLOOM_FILTER, 0, sizeof(value), 100, 0);
      if (!ASSERT_GE(fd, 0, "bpf_create_map bloom filter"))
          return;
  @@ -67,6 +55,30 @@ static void test_fail_cases(void)
      close(fd);
  }
[...]



[Index of Archives]     [Linux Samsung SoC]     [Linux Rockchip SoC]     [Linux Actions SoC]     [Linux for Synopsys ARC Processors]     [Linux NFS]     [Linux NILFS]     [Linux USB Devel]     [Video for Linux]     [Linux Audio Users]     [Yosemite News]     [Linux Kernel]     [Linux SCSI]


  Powered by Linux