| Message ID | 20230914222542.2986059-1-song@kernel.org (mailing list archive) |
|---|---|
| State | Accepted |
| Commit | 5c04433daf9ed8b28d4900112be1fd19e1786b25 |
| Delegated to: | BPF |
| Headers | show |
| Series | [v2,bpf-next] bpf: Charge modmem for struct_ops trampoline | expand |
| Context | Check | Description |
|---|---|---|
| bpf/vmtest-bpf-next-PR | fail | PR summary |
| bpf/vmtest-bpf-next-VM_Test-2 | success | Logs for build for aarch64 with gcc |
| bpf/vmtest-bpf-next-VM_Test-0 | success | Logs for ${{ matrix.test }} on ${{ matrix.arch }} with ${{ matrix.toolchain_full }} |
| bpf/vmtest-bpf-next-VM_Test-3 | fail | Logs for build for s390x with gcc |
| bpf/vmtest-bpf-next-VM_Test-4 | success | Logs for build for x86_64 with gcc |
| bpf/vmtest-bpf-next-VM_Test-5 | success | Logs for build for x86_64 with llvm-16 |
| bpf/vmtest-bpf-next-VM_Test-7 | success | Logs for veristat |
| bpf/vmtest-bpf-next-VM_Test-1 | success | Logs for ShellCheck |
| bpf/vmtest-bpf-next-VM_Test-6 | success | Logs for set-matrix |
| netdev/tree_selection | success | Clearly marked for bpf-next |
| netdev/apply | success | Patch already applied to bpf-next |
Hello: This patch was applied to bpf/bpf-next.git (master) by Martin KaFai Lau <martin.lau@kernel.org>: On Thu, 14 Sep 2023 15:25:42 -0700 you wrote: > Current code charges modmem for regular trampoline, but not for struct_ops > trampoline. Add bpf_jit_[charge|uncharge]_modmem() to struct_ops so the > trampoline is charged in both cases. > > Signed-off-by: Song Liu <song@kernel.org> > > > [...] Here is the summary with links: - [v2,bpf-next] bpf: Charge modmem for struct_ops trampoline https://git.kernel.org/bpf/bpf-next/c/5c04433daf9e You are awesome, thank you!
diff --git a/kernel/bpf/bpf_struct_ops.c b/kernel/bpf/bpf_struct_ops.c index fdc3e8705a3c..db6176fb64dc 100644 --- a/kernel/bpf/bpf_struct_ops.c +++ b/kernel/bpf/bpf_struct_ops.c @@ -615,7 +615,10 @@ static void __bpf_struct_ops_map_free(struct bpf_map *map) if (st_map->links) bpf_struct_ops_map_put_progs(st_map); bpf_map_area_free(st_map->links); - bpf_jit_free_exec(st_map->image); + if (st_map->image) { + bpf_jit_free_exec(st_map->image); + bpf_jit_uncharge_modmem(PAGE_SIZE); + } bpf_map_area_free(st_map->uvalue); bpf_map_area_free(st_map); } @@ -657,6 +660,7 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr) struct bpf_struct_ops_map *st_map; const struct btf_type *t, *vt; struct bpf_map *map; + int ret; st_ops = bpf_struct_ops_find_value(attr->btf_vmlinux_value_type_id); if (!st_ops) @@ -681,12 +685,27 @@ static struct bpf_map *bpf_struct_ops_map_alloc(union bpf_attr *attr) st_map->st_ops = st_ops; map = &st_map->map; + ret = bpf_jit_charge_modmem(PAGE_SIZE); + if (ret) { + __bpf_struct_ops_map_free(map); + return ERR_PTR(ret); + } + + st_map->image = bpf_jit_alloc_exec(PAGE_SIZE); + if (!st_map->image) { + /* __bpf_struct_ops_map_free() uses st_map->image as flag + * for "charged or not". In this case, we need to unchange + * here. + */ + bpf_jit_uncharge_modmem(PAGE_SIZE); + __bpf_struct_ops_map_free(map); + return ERR_PTR(-ENOMEM); + } st_map->uvalue = bpf_map_area_alloc(vt->size, NUMA_NO_NODE); st_map->links = bpf_map_area_alloc(btf_type_vlen(t) * sizeof(struct bpf_links *), NUMA_NO_NODE); - st_map->image = bpf_jit_alloc_exec(PAGE_SIZE); - if (!st_map->uvalue || !st_map->links || !st_map->image) { + if (!st_map->uvalue || !st_map->links) { __bpf_struct_ops_map_free(map); return ERR_PTR(-ENOMEM); } @@ -907,4 +926,3 @@ int bpf_struct_ops_link_create(union bpf_attr *attr) kfree(link); return err; } -
Current code charges modmem for regular trampoline, but not for struct_ops trampoline. Add bpf_jit_[charge|uncharge]_modmem() to struct_ops so the trampoline is charged in both cases. Signed-off-by: Song Liu <song@kernel.org> --- Changes in v2: 1. Fix error handling path of bpf_struct_ops_map_alloc(). (Martin) --- kernel/bpf/bpf_struct_ops.c | 26 ++++++++++++++++++++++---- 1 file changed, 22 insertions(+), 4 deletions(-)