samples: bpf: refactor xdp_sample_pkts_kern with BTF-defined map
Subsystems (4)
PATCH 1/1 samples: bpf: refactor xdp_sample_pkts_kern with BTF-defined map
Most of the samples were converted to use the new BTF-defined MAP as they moved to libbbpf, but some of the samples were missing.
Instead of using the previous BPF MAP definition, this commit refactors xdp_sample_pkts_kern MAP definition with the new BTF-defined MAP format.
--- samples/bpf/xdp_sample_pkts_kern.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)
samples/bpf/xdp_sample_pkts_kern.c +6 -6
@@ -7,12 +7,12 @@#define SAMPLE_SIZE 64ul#define MAX_CPUS 128 -struct bpf_map_def SEC("maps") my_map = { - .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, - .key_size = sizeof(int), - .value_size = sizeof(u32), - .max_entries = MAX_CPUS, -}; +struct { + __uint(type, BPF_MAP_TYPE_PERF_EVENT_ARRAY); + __uint(key_size, sizeof(int)); + __uint(value_size, sizeof(u32)); + __uint(max_entries, MAX_CPUS); +} my_map SEC(".maps"); SEC("xdp_sample")int xdp_sample_prog (struct xdp_md *ctx)
Signature · 1 line
2.25.1
A message that is not in the archive (sent off-list or not yet received): [email protected]
Quoted 8 lines · Michal Rostecki
Quoted 6 lines · Daniel T. Lee
Most of the samples were converted to use the new BTF-defined MAP as they moved to libbbpf, but some of the samples were missing.
Instead of using the previous BPF MAP definition, this commit refactors
xdp_sample_pkts_kernMAP definition with the new BTF-defined MAP format.
Quoted 8 lines
Quoted 4 lines
Signed-off-by Daniel T. Lee <[email protected]>--- samples/bpf/xdp_sample_pkts_kern.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)I see that
samples/bpf/ibumad_kern.candsamples/bpf/xdp_monitor_kern.cstill have old style BPF map definitions. Maybe you could change them as well?
Thanks for the review!
Actually, I'm well aware that there are some samples left with old style map definitions, But those examples should be transferred from bpf_load to libbbpf, not just bpf map changes.
I'm also planning to refactor those patches in the future. For now I've just refactored this file, but if you think this patch size is small,
I'll send it with other changes.
Quoted 21 lines · Daniel T. Lee
Most of the samples were converted to use the new BTF-defined MAP as they moved to libbbpf, but some of the samples were missing.
Instead of using the previous BPF MAP definition, this commit refactors
xdp_sample_pkts_kernMAP definition with the new BTF-defined MAP format.Signed-off-by Daniel T. Lee <[email protected]>--- samples/bpf/xdp_sample_pkts_kern.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)samples/bpf/xdp_sample_pkts_kern.c +0
-1@@ -7,12 +7,12 @@#define SAMPLE_SIZE 64ul#define MAX_CPUS 128 -struct bpf_map_def SEC("maps") my_map = {
samples/bpf/xdp_sample_pkts_kern.c line 14
Quoted 10 lines
samples/bpf/xdp_sample_pkts_kern.c +5
-5- .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, - .key_size = sizeof(int), - .value_size = sizeof(u32), - .max_entries = MAX_CPUS, -}; +struct { + __uint(type, BPF_MAP_TYPE_PERF_EVENT_ARRAY); + __uint(key_size, sizeof(int)); + __uint(value_size, sizeof(u32)); + __uint(max_entries, MAX_CPUS);
if you drop max_entries property, libbpf will set it to the maximum configured number of CPUs on the host, which is what you probably want. Do you might sending v2 without MAX_CPUS (check if macro is still used anywhere else). Thanks!
Quoted 7 lines
samples/bpf/xdp_sample_pkts_kern.c +1
-1+} my_map SEC(".maps"); SEC("xdp_sample")int xdp_sample_prog (struct xdp_md *ctx) --2.25.1
Quoted 22 lines · Andrii Nakryiko
Quoted 20 lines · Daniel T. Lee
Most of the samples were converted to use the new BTF-defined MAP as they moved to libbbpf, but some of the samples were missing.
Instead of using the previous BPF MAP definition, this commit refactors
xdp_sample_pkts_kernMAP definition with the new BTF-defined MAP format.Signed-off-by Daniel T. Lee <[email protected]>--- samples/bpf/xdp_sample_pkts_kern.c | 12 ++++++------ 1 file changed, 6 insertions(+), 6 deletions(-)samples/bpf/xdp_sample_pkts_kern.c +0
-0@@ -7,12 +7,12 @@#define SAMPLE_SIZE 64ul#define MAX_CPUS 128
Quoted 17 lines
samples/bpf/xdp_sample_pkts_kern.c line 14
Quoted 11 lines
samples/bpf/xdp_sample_pkts_kern.c +5
-6-struct bpf_map_def SEC("maps") my_map = { - .type = BPF_MAP_TYPE_PERF_EVENT_ARRAY, - .key_size = sizeof(int), - .value_size = sizeof(u32), - .max_entries = MAX_CPUS, -}; +struct { + __uint(type, BPF_MAP_TYPE_PERF_EVENT_ARRAY); + __uint(key_size, sizeof(int)); + __uint(value_size, sizeof(u32)); + __uint(max_entries, MAX_CPUS);if you drop
max_entriesproperty, libbpf will set it to the maximum configured number of CPUs on the host, which is what you probably want. Do you might sending v2 withoutMAX_CPUS(check if macro is still used anywhere else). Thanks!
Thanks for your time and effort for the review.
I'll check and send the next version of patch.
Quoted 7 lines
Quoted 7 lines
samples/bpf/xdp_sample_pkts_kern.c +1
-1+} my_map SEC(".maps"); SEC("xdp_sample")int xdp_sample_prog (struct xdp_md *ctx) --2.25.1
Signature · 2 lines
Best, Daniel T. Lee