Extend the bpf hashmap docs to include a brief description of the internals of the LRU map type (setting appropriate API expectations), including the original commit message from Martin and a variant on the graph that I had presented during my Linux Plumbers Conference 2022 talk on "Pressure feedback for LRU map types"[0]. The node names in the dot file correspond roughly to the functions where the logic for those decisions or steps is defined, to help curious developers to cross-reference and update this logic if the details of the LRU implementation ever differ from this description. [0]: https://lpc.events/event/16/contributions/1368/ Signed-off-by: Joe Stringer <joe@xxxxxxxxxxxxx> --- v2: Fix issue that caused initial email submission to fail --- Documentation/bpf/map_hash.rst | 193 ++++++++++++++++++++++ Documentation/bpf/map_lru_hash_update.dot | 163 ++++++++++++++++++ 2 files changed, 356 insertions(+) create mode 100644 Documentation/bpf/map_lru_hash_update.dot diff --git a/Documentation/bpf/map_hash.rst b/Documentation/bpf/map_hash.rst index e85120878b27..1bf7c497e5fe 100644 --- a/Documentation/bpf/map_hash.rst +++ b/Documentation/bpf/map_hash.rst @@ -1,5 +1,6 @@ .. SPDX-License-Identifier: GPL-2.0-only .. Copyright (C) 2022 Red Hat, Inc. +.. Copyright (C) 2022 Isovalent, Inc. =============================================== BPF_MAP_TYPE_HASH, with PERCPU and LRU Variants @@ -183,3 +184,195 @@ Userspace walking the map elements from the map declared above: cur_key = &next_key; } } + +Internals +========= + +This section of the document is targeted at Linux developers and describes +aspects of the map implementations that are not considered stable ABI. The +following details are subject to change in future versions of the kernel. + +``BPF_MAP_TYPE_LRU_HASH`` and variants +-------------------------------------- + +An LRU hashmap type consists of two properties: Firstly, it is a hash map and +hence is indexable by key for constant time lookups. Secondly, when at map +capacity, map updates will trigger eviction of old entries based on the age of +the elements in a set of lists. Each of these properties may be either global +or per-CPU, depending on the map type and flags used to create the map: + +.. flat-table:: Comparison of map properties by map type (x-axis) and flags + (y-axis) + + * - + - ``BPF_MAP_TYPE_LRU_HASH`` + - ``BPF_MAP_TYPE_LRU_PERCPU_HASH`` + + * - ``BPF_NO_COMMON_LRU`` + - Per-CPU LRU, global map + - Per-CPU LRU, per-cpu map + + * - ``!BPF_NO_COMMON_LRU`` + - Global LRU, global map + - Global LRU, per-cpu map + +The commit message for LRU map support provides a general overview of the +underlying LRU algorithm used for entry eviction when the table is full: + +:: + + commit 3a08c2fd763450a927d1130de078d6f9e74944fb + Author: Martin KaFai Lau <kafai@xxxxxx> + Date: Fri Nov 11 10:55:06 2016 -0800 + + bpf: LRU List + + Introduce bpf_lru_list which will provide LRU capability to + the bpf_htab in the later patch. + + * General Thoughts: + 1. Target use case. Read is more often than update. + (i.e. bpf_lookup_elem() is more often than bpf_update_elem()). + If bpf_prog does a bpf_lookup_elem() first and then an in-place + update, it still counts as a read operation to the LRU list concern. + 2. It may be useful to think of it as a LRU cache + 3. Optimize the read case + 3.1 No lock in read case + 3.2 The LRU maintenance is only done during bpf_update_elem() + 4. If there is a percpu LRU list, it will lose the system-wise LRU + property. A completely isolated percpu LRU list has the best + performance but the memory utilization is not ideal considering + the work load may be imbalance. + 5. Hence, this patch starts the LRU implementation with a global LRU + list with batched operations before accessing the global LRU list. + As a LRU cache, #read >> #update/#insert operations, it will work well. + 6. There is a local list (for each cpu) which is named + 'struct bpf_lru_locallist'. This local list is not used to sort + the LRU property. Instead, the local list is to batch enough + operations before acquiring the lock of the global LRU list. More + details on this later. + 7. In the later patch, it allows a percpu LRU list by specifying a + map-attribute for scalability reason and for use cases that need to + prepare for the worst (and pathological) case like DoS attack. + The percpu LRU list is completely isolated from each other and the + LRU nodes (including free nodes) cannot be moved across the list. The + following description is for the global LRU list but mostly applicable + to the percpu LRU list also. + + * Global LRU List: + 1. It has three sub-lists: active-list, inactive-list and free-list. + 2. The two list idea, active and inactive, is borrowed from the + page cache. + 3. All nodes are pre-allocated and all sit at the free-list (of the + global LRU list) at the beginning. The pre-allocation reasoning + is similar to the existing BPF_MAP_TYPE_HASH. However, + opting-out prealloc (BPF_F_NO_PREALLOC) is not supported in + the LRU map. + + * Active/Inactive List (of the global LRU list): + 1. The active list, as its name says it, maintains the active set of + the nodes. We can think of it as the working set or more frequently + accessed nodes. The access frequency is approximated by a ref-bit. + The ref-bit is set during the bpf_lookup_elem(). + 2. The inactive list, as its name also says it, maintains a less + active set of nodes. They are the candidates to be removed + from the bpf_htab when we are running out of free nodes. + 3. The ordering of these two lists is acting as a rough clock. + The tail of the inactive list is the older nodes and + should be released first if the bpf_htab needs free element. + + * Rotating the Active/Inactive List (of the global LRU list): + 1. It is the basic operation to maintain the LRU property of + the global list. + 2. The active list is only rotated when the inactive list is running + low. This idea is similar to the current page cache. + Inactive running low is currently defined as + "# of inactive < # of active". + 3. The active list rotation always starts from the tail. It moves + node without ref-bit set to the head of the inactive list. + It moves node with ref-bit set back to the head of the active + list and then clears its ref-bit. + 4. The inactive rotation is pretty simply. + It walks the inactive list and moves the nodes back to the head of + active list if its ref-bit is set. The ref-bit is cleared after moving + to the active list. + If the node does not have ref-bit set, it just leave it as it is + because it is already in the inactive list. + + * Shrinking the Inactive List (of the global LRU list): + 1. Shrinking is the operation to get free nodes when the bpf_htab is + full. + 2. It usually only shrinks the inactive list to get free nodes. + 3. During shrinking, it will walk the inactive list from the tail, + delete the nodes without ref-bit set from bpf_htab. + 4. If no free node found after step (3), it will forcefully get + one node from the tail of inactive or active list. Forcefully is + in the sense that it ignores the ref-bit. + + * Local List: + 1. Each CPU has a 'struct bpf_lru_locallist'. The purpose is to + batch enough operations before acquiring the lock of the + global LRU. + 2. A local list has two sub-lists, free-list and pending-list. + 3. During bpf_update_elem(), it will try to get from the free-list + of (the current CPU local list). + 4. If the local free-list is empty, it will acquire from the + global LRU list. The global LRU list can either satisfy it + by its global free-list or by shrinking the global inactive + list. Since we have acquired the global LRU list lock, + it will try to get at most LOCAL_FREE_TARGET elements + to the local free list. + 5. When a new element is added to the bpf_htab, it will + first sit at the pending-list (of the local list) first. + The pending-list will be flushed to the global LRU list + when it needs to acquire free nodes from the global list + next time. + + * Lock Consideration: + The LRU list has a lock (lru_lock). Each bucket of htab has a + lock (buck_lock). If both locks need to be acquired together, + the lock order is always lru_lock -> buck_lock and this only + happens in the bpf_lru_list.c logic. + + In hashtab.c, both locks are not acquired together (i.e. one + lock is always released first before acquiring another lock). + + Signed-off-by: Martin KaFai Lau <kafai@xxxxxx> + Acked-by: Alexei Starovoitov <ast@xxxxxxxxxx> + Signed-off-by: David S. Miller <davem@xxxxxxxxxxxxx> + +Notably, there are various steps that the update algorithm attempts in order to +enforce the LRU property which have increasing impacts on other CPUs involved +in the operations: + +- Attempt to use CPU-local state to batch operations +- Attempt to fetch free nodes from global lists +- Attempt to pull any node from a global list and remove it from the hashmap +- Attempt to pull any node from any CPU's list and remove it from the hashmap + +Even if an LRU node may be acquired, maps of type ``BPF_MAP_TYPE_LRU_HASH`` +may fail to insert the entry into the map if other CPUs are heavily contending +on the global hashmap lock. + +This algorithm is described visually in the following diagram: + +.. kernel-figure:: map_lru_hash_update.dot + :alt: Diagram outlining the LRU eviction steps taken during map update + + LRU hash eviction during map update for ``BPF_MAP_TYPE_LRU_HASH`` and + variants + +Map updates start from the oval in the top right "begin ``bpf_map_update()``" +and progress through the graph towards the bottom where the result may be +either a successful update or a failure with various error codes. The key in +the top right provides indicators for which locks may be involved in specific +operations. This is intended as a visual hint for reasoning about how map +contention may impact update operations, though the map type and flags may +impact the actual contention on those locks, based on the logic described in +the table above. For instance, if the map is created with type +``BPF_MAP_TYPE_LRU_PERCPU_HASH`` and flags ``BPF_NO_COMMON_LRU`` then all map +properties would be per-cpu. + +The dot file source for the above diagram is uses internal kernel function +names for the node names in order to make the corresponding logic easier to +find. See ``Documentation/bpf/map_lru_hash_update.dot`` for more details. diff --git a/Documentation/bpf/map_lru_hash_update.dot b/Documentation/bpf/map_lru_hash_update.dot new file mode 100644 index 000000000000..3839b68f32b7 --- /dev/null +++ b/Documentation/bpf/map_lru_hash_update.dot @@ -0,0 +1,163 @@ +// SPDX-License-Identifier: GPL-2.0-only +// Copyright (C) 2022 Isovalent, Inc. +digraph { + node [colorscheme=accent4,style=filled] # Apply colorscheme to all nodes + graph [splines=ortho, nodesep=1] + + subgraph cluster_key { + label = "Key\n(locks held during operation)"; + rankdir = TB; + + remote_lock [shape=rectangle,fillcolor=4,label="remote CPU LRU lock"] + hash_lock [shape=rectangle,fillcolor=3,label="hashtab lock"] + lru_lock [shape=rectangle,fillcolor=2,label="LRU lock"] + local_lock [shape=rectangle,fillcolor=1,label="local CPU LRU lock"] + no_lock [shape=rectangle,label="no locks held"] + } + + begin [shape=oval,label="begin\nbpf_map_update()"] + + // Nodes below with an 'fn_' prefix are roughly labeled by the C function + // names that initiate the corresponding logic in kernel/bpf/bpf_lru_list.c. + // Number suffixes and errno suffixes handle subsections of the corresponding + // logic in the function as of the writing of this dot. + fn_bpf_lru_pop_free [shape=diamond,fillcolor=1, + label="Local freelist\nnode available?"]; + fn__local_list_pop_free [shape=rectangle, + label="Use node owned\nby this CPU"] + + common_lru_check [shape=diamond, + label="Map created with\nBPF_NO_COMMON_LRU\nflag set?"]; + + fn_bpf_lru_list_pop_free_to_local [shape=rectangle,fillcolor=2, + label="Flush local pending, + Rotate Global list, move + LOCAL_FREE_TARGET + from global -> local"] + // Also corresponds to: + // fn__local_list_flush() + // fn_bpf_lru_list_rotate() + fn___bpf_lru_node_move_to_free[shape=diamond,fillcolor=2, + label="Able to free\nLOCAL_FREE_TARGET\nnodes?"] + + fn___bpf_lru_list_shrink_inactive [shape=rectangle,fillcolor=3, + label="Shrink inactive list + up to remaining + LOCAL_FREE_TARGET + (global LRU -> local)"] + fn___bpf_lru_list_shrink [shape=diamond,fillcolor=2, + label="> 0 entries in\nlocal free list?"] + fn___bpf_lru_list_shrink2 [shape=rectangle,fillcolor=2, + label="Steal one node from + inactive, or if empty, + from active global list"] + fn___bpf_lru_list_shrink3 [shape=rectangle,fillcolor=3, + label="Try to remove\nnode from hashtab"] + + fn_bpf_lru_pop_free2 [shape=diamond,label="Htab removal\nsuccessful?"] + common_lru_check2 [shape=diamond, + label="Map created with\nBPF_NO_COMMON_LRU\nflag set?"] + + subgraph cluster_remote_lock { + label = "Iterate through CPUs\n(start from current)"; + style = dashed; + rankdir=LR; + + fn_bpf_lru_pop_free5 [shape=diamond,fillcolor=4, + label="Steal a node from\nper-cpu freelist?"] + fn_bpf_lru_pop_free6 [shape=rectangle,fillcolor=4, + label="Steal a node from + (1) Unreferenced pending, or + (2) Any pending node"] + fn_bpf_lru_pop_free7 [shape=rectangle,fillcolor=3, + label="Try to remove\nnode from hashtab"] + fn_htab_lru_map_update_elem [shape=diamond, + label="Stole node\nfrom remote\nCPU?"] + fn_htab_lru_map_update_elem2 [shape=diamond,label="Iterated\nall CPUs?"] + // Also corresponds to: + // fn__local_list_pop_free() + // fn__local_list_pop_pending() + } + + fn_bpf_lru_list_pop_free_to_local2 [shape=rectangle, + label="Use node that was\nnot recently referenced"] + fn_bpf_lru_pop_free4 [shape=rectangle, + label="Use node that was\nactively referenced\nin global list"] + fn_htab_lru_map_update_elem_ENOMEM [shape=oval,label="return -ENOMEM"] + fn_htab_lru_map_update_elem3 [shape=rectangle, + label="Use node that was\nactively referenced\nin (another?) CPU's cache"] + fn_htab_lru_map_update_elem4 [shape=diamond, + label="Can lock this\nhashtab bucket?"] + fn_htab_lru_map_update_elem5 [shape=rectangle,fillcolor=3, + label="Update hashmap\nwith new element"] + fn_htab_lru_map_update_elem6 [shape=oval,label="return 0"] + fn_htab_lru_map_update_elem_EBUSY [shape=oval,label="return -EBUSY"] + + begin -> fn_bpf_lru_pop_free + fn_bpf_lru_pop_free -> fn__local_list_pop_free [xlabel="Y"] + fn_bpf_lru_pop_free -> common_lru_check [xlabel="N"] + common_lru_check -> fn_bpf_lru_list_pop_free_to_local [xlabel="Y"] + common_lru_check -> fn___bpf_lru_list_shrink_inactive [xlabel="N"] + fn_bpf_lru_list_pop_free_to_local -> fn___bpf_lru_node_move_to_free + fn___bpf_lru_node_move_to_free -> + fn_bpf_lru_list_pop_free_to_local2 [xlabel="Y"] + fn___bpf_lru_node_move_to_free -> + fn___bpf_lru_list_shrink_inactive [xlabel="N"] + fn___bpf_lru_list_shrink_inactive -> fn___bpf_lru_list_shrink + fn___bpf_lru_list_shrink -> fn_bpf_lru_list_pop_free_to_local2 [xlabel = "Y"] + fn___bpf_lru_list_shrink -> fn___bpf_lru_list_shrink2 [xlabel="N"] + fn___bpf_lru_list_shrink2 -> fn___bpf_lru_list_shrink3 + fn___bpf_lru_list_shrink3 -> fn_bpf_lru_pop_free2 + fn_bpf_lru_pop_free2 -> fn_bpf_lru_pop_free4 [xlabel = "Y"] + fn_bpf_lru_pop_free2 -> common_lru_check2 [xlabel = "N"] + common_lru_check2 -> fn_htab_lru_map_update_elem_ENOMEM [xlabel = "Y"] + common_lru_check2 -> fn_bpf_lru_pop_free5 [xlabel = "N"] + fn_bpf_lru_pop_free5 -> fn_htab_lru_map_update_elem [xlabel = "Y"] + fn_bpf_lru_pop_free5 -> fn_bpf_lru_pop_free6 [xlabel = "N"] + fn_bpf_lru_pop_free6 -> fn_bpf_lru_pop_free7 + fn_bpf_lru_pop_free7 -> fn_htab_lru_map_update_elem + + fn_htab_lru_map_update_elem -> fn_htab_lru_map_update_elem3 [xlabel = "Y"] + fn_htab_lru_map_update_elem -> fn_htab_lru_map_update_elem2 [xlabel = "N"] + fn_htab_lru_map_update_elem2 -> + fn_htab_lru_map_update_elem_ENOMEM [xlabel = "Y"] + fn_htab_lru_map_update_elem2 -> fn_bpf_lru_pop_free5 [xlabel = "N"] + fn_htab_lru_map_update_elem3 -> fn_htab_lru_map_update_elem4 + + fn__local_list_pop_free -> fn_htab_lru_map_update_elem4 + fn_bpf_lru_list_pop_free_to_local2 -> fn_htab_lru_map_update_elem4 + fn_bpf_lru_pop_free4 -> fn_htab_lru_map_update_elem4 + + fn_htab_lru_map_update_elem4 -> fn_htab_lru_map_update_elem5 [xlabel="Y"] + fn_htab_lru_map_update_elem4 -> + fn_htab_lru_map_update_elem_EBUSY [xlabel="N"] + fn_htab_lru_map_update_elem5 -> fn_htab_lru_map_update_elem6 + + // Create invisible pad nodes to line up various nodes + pad0 [style=invis] + pad1 [style=invis] + pad2 [style=invis] + pad3 [style=invis] + pad4 [style=invis] + + // Line up the key with the top of the graph + no_lock -> local_lock [style=invis] + local_lock -> lru_lock [style=invis] + lru_lock -> hash_lock [style=invis] + hash_lock -> remote_lock [style=invis] + remote_lock -> fn_bpf_lru_pop_free5 [style=invis] + remote_lock -> fn___bpf_lru_list_shrink [style=invis] + + // Line up return code nodes at the bottom of the graph + fn_htab_lru_map_update_elem -> pad0 [style=invis] + pad0 -> pad1 [style=invis] + pad1 -> pad2 [style=invis] + pad2-> fn_htab_lru_map_update_elem_ENOMEM [style=invis] + fn_htab_lru_map_update_elem4 -> pad3 [style=invis] + pad3 -> fn_htab_lru_map_update_elem_EBUSY [style=invis] + + // Reduce diagram width by forcing some nodes to appear above others + fn_bpf_lru_pop_free4 -> fn_htab_lru_map_update_elem3 [style=invis] + common_lru_check2 -> pad4 [style=invis] + pad4 -> fn_bpf_lru_pop_free5 [style=invis] +} -- 2.25.1
