Merge pull request #10 from bpftools/new/add-maps

Maps!

Author: fntlnz

content/bpf-vm/304-bpfvm-maps.md

@@ -0,0 +1,14 @@
+---
+title:  "bpfvm-bpf-maps"
+weight: 304
+---
+
+# Maps
+
+- BPF Maps data stores that live in the kernel;
+- Can be accessed by any BPF program that knows about them;
+- Programs that run in user-space can also access these maps by using file descriptors;
+- You can store any kind of data in a map, as long as you specify the data size correctly before hand;
+- The kernel treats keys and values as binary blobs and it doesn’t care about what you keep in a map;
+- This is what we use to let userspace programs to extract or feed information into BPF programs running in the kernel!
+

content/bpf-vm/305-bpfvm-maps-types.md

@@ -0,0 +1,19 @@
+---
+title:  "bpfvm-bpf-maps-types"
+weight: 305
+---
+
+**Many different types of maps**
+
+- Hash table: BPF_MAP_TYPE_HASH
+- Array: BPF_MAP_TYPE_ARRAY
+- Program array maps: BPF_MAP_TYPE_PROG_ARRAY, this one is magic, allows you to store references to bpf programs so that you can do jumps between bpf programs;
+- Perf events array maps: BPF_MAP_TYPE_PERF_EVENT_ARRAY
+- Per-CPU hash maps: BPF_MAP_TYPE_PERCPU_HASH
+- Per-CPU array maps: BPF_MAP_TYPE_PERCPU_ARRAY
+- Stack trace maps: BPF_MAP_TYPE_STACK_TRACE
+- Cgroup array maps: BPF_MAP_TYPE_CGROUP_ARRAY
+- Hash and per cpu has with LRU cache: BPF_MAP_TYPE_LRU_PERCPU_HASH, BPF_MAP_TYPE_LRU_HASH
+- Longest Prefix Match(LPM) Trie: BPF_MAP_TYPE_LPM_TRIE
+- Array of maps, and hash of maps, maps: `BPF_MAP_TYPE_ARRAY_OF_MAPS` and `BPF_MAP_TYPE_HASH_OF_MAPS`
+- And many more! Find all of them `man 2 bpf`

content/bpf-vm/306-bpfvm-maps-operations.md

@@ -0,0 +1,14 @@
+---
+title:  "bpfvm-bpf-maps-operations"
+weight: 306
+---
+
+**Maps operations**
+
+- Lookup a single element value, `bpf_map_lookup_elem`
+- Remove an element, `bpf_map_delete_element`
+- Iterating over elements
+- Updating an element, `bpf_map_update_elem`
+- Get the next key in the map, `bpf_map_get_next_key`
+- Search, get the value and delete in a single atomic operation, `bpf_map_lookup_and_delete_element`
+- Concurrent access is regulated using a mechanism called `bpf_spin_lock` that is essentially a semaphore;