selftests/bpf: veristat: Minimize map size during verification
Subsystems (3)
Emil Tsalapatis
reviewerPATCH 1/1 selftests/bpf: veristat: Minimize map size during verification
The veristat tool verifies that BPF objects along with their maps pass verification for a given kernel version. To do so, veristat loads the objects and maps into the kernel in order to pass them through the verifier.
Currently, veristat sizes the maps according to the max_entries field provided by the program author. Depending on the map type this field may be irrelevant to the verification process. However, loading a large map can fail because of -ENOMEM errors.
This is a problem when the map is supposed to run on large machines, but veristat tests it machines with significantly less RAM (e.g., CI). In that case veristat fails even if the program verifies.
Expand veristat to resize maps whose max_entries are not relevant to verification. Set the max_entries value as low as possible to avoid -ENOMEM errors.
--- tools/testing/selftests/bpf/veristat.c | 46 ++++++++++++++++++++++++++ 1 file changed, 46 insertions(+)
tools/testing/selftests/bpf/veristat.c +46 -0
@@ -37,6 +37,10 @@#define min (a, b) ((a) < (b) ? (a) : (b))#endif +#ifndef div_round_up +#define div_round_up (a, b) (((a) + (b) - 1) / (b)) +#endif +enum stat_id {VERDICT ,DURATION ,
@@ -1238,6 +1242,17 @@ static void mask_unrelated_struct_ops_progs(struct bpf_object *obj, } } +/* size of global arena data in pages (minimum 1) */ +static size_t arena_global_pages (struct bpf_map *map) +{ +size_t min_bytes; + +if (!bpf_map__initial_value (map, &min_bytes)) +return 1 ; + +return div_round_up (min_bytes,sysconf (_SC_PAGESIZE)) ?:1 ; +} +static void fixup_obj_maps (struct bpf_object *obj) {struct bpf_map *map;
@@ -1248,6 +1263,37 @@ static void fixup_obj_maps(struct bpf_object *obj)/* fix up map size, if necessary */ switch (bpf_map__type (map)) { +/* +* if the verifier doesn't use max_entries +* then set to 1 to avoid -ENOMEM +*/ +case BPF_MAP_TYPE_HASH : +case BPF_MAP_TYPE_PERCPU_HASH : +case BPF_MAP_TYPE_LRU_HASH : +case BPF_MAP_TYPE_LRU_PERCPU_HASH : +case BPF_MAP_TYPE_HASH_OF_MAPS : +case BPF_MAP_TYPE_RHASH : +case BPF_MAP_TYPE_SOCKHASH : +case BPF_MAP_TYPE_DEVMAP_HASH : +case BPF_MAP_TYPE_LPM_TRIE : +case BPF_MAP_TYPE_QUEUE : +case BPF_MAP_TYPE_STACK : +case BPF_MAP_TYPE_BLOOM_FILTER : +case BPF_MAP_TYPE_STACK_TRACE : +bpf_map__set_max_entries (map,1 ); +break ; + +/* arenas must be big enough to fit all globals */ +case BPF_MAP_TYPE_ARENA : +bpf_map__set_max_entries (map,arena_global_pages (map)); +break ; + +/* ringbufs must be page-aligned */ +case BPF_MAP_TYPE_RINGBUF : +case BPF_MAP_TYPE_USER_RINGBUF : +bpf_map__set_max_entries (map,sysconf (_SC_PAGESIZE)); +break ; +case BPF_MAP_TYPE_SK_STORAGE :case BPF_MAP_TYPE_TASK_STORAGE :case BPF_MAP_TYPE_INODE_STORAGE :
Signature · 1 line
2.54.0
[email protected]
AI reviewThank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider: - [Medium] Setting `max_entries` to 1 for `BPF_MAP_TYPE_RHASH` maps causes map creation to fail if…
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Setting max_entries to 1 for BPF_MAP_TYPE_RHASH maps causes map creation to fail if map_extra > 1. --
--- Patch [1]: [PATCH bpf-next] selftests/bpf: veristat: Minimize map size during verification --- commit a11623d3f77bb05764867d10c3ff357f2a8a1403
Author: Emil Tsalapatis <[email protected]>
selftests/bpf: veristat: Minimize map size during verification
The veristat tool loads objects and maps into the kernel to verify them. This patch expands veristat to resize maps whose max_entries are not relevant to verification to as low as possible to avoid -ENOMEM errors on smaller test machines.
Quoted 4 lines
tools/testing/selftests/bpf/veristat.c +0
-0
[ ... ]
Quoted 13 lines
tools/testing/selftests/bpf/veristat.c +9
-0@@ -1248,6 +1263,37 @@ static void fixup_obj_maps(struct bpf_object *obj)/* fix up map size, if necessary */ switch (bpf_map__type (map)) { +/* +* if the verifier doesn't use max_entries +* then set to 1 to avoid -ENOMEM +*/ +case BPF_MAP_TYPE_HASH : +case BPF_MAP_TYPE_PERCPU_HASH : +case BPF_MAP_TYPE_LRU_HASH : +case BPF_MAP_TYPE_LRU_PERCPU_HASH : +case BPF_MAP_TYPE_HASH_OF_MAPS :
tools/testing/selftests/bpf/veristat.c static void fixup_obj_maps(struct bpf_object *obj) · line 1284
Quoted 10 lines
tools/testing/selftests/bpf/veristat.c +10
-0+case BPF_MAP_TYPE_RHASH : +case BPF_MAP_TYPE_SOCKHASH : +case BPF_MAP_TYPE_DEVMAP_HASH : +case BPF_MAP_TYPE_LPM_TRIE : +case BPF_MAP_TYPE_QUEUE : +case BPF_MAP_TYPE_STACK : +case BPF_MAP_TYPE_BLOOM_FILTER : +case BPF_MAP_TYPE_STACK_TRACE : +bpf_map__set_max_entries (map,1 ); +break ;
[Severity: Medium]
Will setting max_entries to 1 cause map creation to fail for BPF_MAP_TYPE_RHASH maps that have map_extra set to greater than 1?
The kernel's rhtab_map_alloc_check() enforces that map_extra must be less than or equal to max_entries. If a BPF program legitimately uses an RHASH map with a map_extra preallocation hint greater than 1, this unconditional minimization will cause an -EINVAL load failure, making veristat erroneously report a verification failure.
Does veristat also need to adjust map_extra when reducing max_entries for RHASH maps?
Signature · 1 line
Sashiko AI review · https://sashiko.dev/#/patchset/[email protected]?part=1
Emil Tsalapatis
reviewerQuoted 48 lines · sashiko-bot
Thank you for your contribution! Sashiko AI review found 1 potential issue(s) to consider:
- [Medium] Settingmax_entriesto 1 forBPF_MAP_TYPE_RHASHmaps causes map creation to fail ifmap_extra> 1. ----- Patch [1]: [PATCH bpf-next] selftests/bpf: veristat: Minimize map size during verification --- commit
a11623d3f77bb05764867d10c3ff357f2a8a1403
Author: Emil Tsalapatis <[email protected]>selftests/bpf: veristat: Minimize map size during verification
The veristat tool loads objects and maps into the kernel to verify them. This patch expands veristat to resize maps whose
max_entriesare not relevant to verification to as low as possible to avoid-ENOMEMerrors on smaller test machines.Quoted 4 lines
tools/testing/selftests/bpf/veristat.c +0
-0[ ... ]
Quoted 13 lines
tools/testing/selftests/bpf/veristat.c +9
-0@@ -1248,6 +1263,37 @@ static void fixup_obj_maps(struct bpf_object *obj)/* fix up map size, if necessary */ switch (bpf_map__type (map)) { +/* +* if the verifier doesn't use max_entries +* then set to 1 to avoid -ENOMEM +*/ +case BPF_MAP_TYPE_HASH : +case BPF_MAP_TYPE_PERCPU_HASH : +case BPF_MAP_TYPE_LRU_HASH : +case BPF_MAP_TYPE_LRU_PERCPU_HASH : +case BPF_MAP_TYPE_HASH_OF_MAPS :tools/testing/selftests/bpf/veristat.c static void fixup_obj_maps(struct bpf_object *obj) · line 1284
Quoted 10 lines
tools/testing/selftests/bpf/veristat.c +10
-0+case BPF_MAP_TYPE_RHASH : +case BPF_MAP_TYPE_SOCKHASH : +case BPF_MAP_TYPE_DEVMAP_HASH : +case BPF_MAP_TYPE_LPM_TRIE : +case BPF_MAP_TYPE_QUEUE : +case BPF_MAP_TYPE_STACK : +case BPF_MAP_TYPE_BLOOM_FILTER : +case BPF_MAP_TYPE_STACK_TRACE : +bpf_map__set_max_entries (map,1 ); +break ;[Severity: Medium]
Will settingmax_entriesto 1 cause map creation to fail forBPF_MAP_TYPE_RHASHmaps that havemap_extraset to greater than 1?
Quoted 8 lines
The kernel's
rhtab_map_alloc_check()enforces thatmap_extramust be less than or equal tomax_entries. If a BPF program legitimately uses an RHASH map with amap_extrapreallocation hint greater than 1, this unconditional minimization will cause an-EINVALload failure, making veristat erroneously report a verification failure.Does veristat also need to adjust
map_extrawhen reducingmax_entriesfor RHASH maps?
The concern is valid, the fix imo is not. Rhashtable is resizable in the first place, so we shouldn't be trying to reduce its maximum entries. Same holds for LPM_TRIE and for arena (my worry was we do a memcg reservation upfront but that's not the case). We can just remove all three from the list to fix this issue and simplify the patch.