 Cilium v1.5 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and con�nue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execu�on, BPF programs are guaranteed to never0 码力 | 740 页 | 12.52 MB | 1 年前3 Cilium v1.5 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and con�nue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execu�on, BPF programs are guaranteed to never0 码力 | 740 页 | 12.52 MB | 1 年前3
 Cilium v1.6 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never0 码力 | 734 页 | 11.45 MB | 1 年前3 Cilium v1.6 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never0 码力 | 734 页 | 11.45 MB | 1 年前3
 Cilium v1.10 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never significant bit position of ‘n’. In other terms, the internal buffer size is always a power of 2 with 1 slot reserved for the writer. In effect, from a user perspective, the ring buffer capacity is one less0 码力 | 1307 页 | 19.26 MB | 1 年前3 Cilium v1.10 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never significant bit position of ‘n’. In other terms, the internal buffer size is always a power of 2 with 1 slot reserved for the writer. In effect, from a user perspective, the ring buffer capacity is one less0 码力 | 1307 页 | 19.26 MB | 1 年前3
 Cilium v1.9 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never significant bit position of ‘n’. In other terms, the internal buffer size is always a power of 2 with 1 slot reserved for the writer. In effect, from a user perspective, the ring buffer capacity is one less0 码力 | 1263 页 | 18.62 MB | 1 年前3 Cilium v1.9 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never significant bit position of ‘n’. In other terms, the internal buffer size is always a power of 2 with 1 slot reserved for the writer. In effect, from a user perspective, the ring buffer capacity is one less0 码力 | 1263 页 | 18.62 MB | 1 年前3
 Cilium v1.11 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never significant bit position of ‘n’. In other terms, the internal buffer size is always a power of 2 with 1 slot reserved for the writer. In effect, from a user perspective, the ring buffer capacity is one less0 码力 | 1373 页 | 19.37 MB | 1 年前3 Cilium v1.11 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never significant bit position of ‘n’. In other terms, the internal buffer size is always a power of 2 with 1 slot reserved for the writer. In effect, from a user perspective, the ring buffer capacity is one less0 码力 | 1373 页 | 19.37 MB | 1 年前3
 Cilium v1.7 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never0 码力 | 885 页 | 12.41 MB | 1 年前3 Cilium v1.7 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never0 码力 | 885 页 | 12.41 MB | 1 年前3
 Cilium v1.8 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never0 码力 | 1124 页 | 21.33 MB | 1 年前3 Cilium v1.8 Documentationprogram from the passed file descriptor and atomically replaces program pointers at the given map slot. When no map entry has been found at the provided key, the kernel will just “fall through” and continue in that the program can simply use a BPF map of type BPF_MAP_TYPE_PERCPU_ARRAY with just a single slot of arbitrary value size. This works, because during execution, BPF programs are guaranteed to never0 码力 | 1124 页 | 21.33 MB | 1 年前3
共 7 条
- 1














