Merge tag 'bpf-next-6.16' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next

90b83efa · Linus Torvalds <[email protected]> · 2025-05-28 15:52:42 -07:00

Pull bpf updates from Alexei Starovoitov:

 - Fix and improve BTF deduplication of identical BTF types (Alan
   Maguire and Andrii Nakryiko)

 - Support up to 12 arguments in BPF trampoline on arm64 (Xu Kuohai and
   Alexis Lothoré)

 - Support load-acquire and store-release instructions in BPF JIT on
   riscv64 (Andrea Parri)

 - Fix uninitialized values in BPF_{CORE,PROBE}_READ macros (Anton
   Protopopov)

 - Streamline allowed helpers across program types (Feng Yang)

 - Support atomic update for hashtab of BPF maps (Hou Tao)

 - Implement json output for BPF helpers (Ihor Solodrai)

 - Several s390 JIT fixes (Ilya Leoshkevich)

 - Various sockmap fixes (Jiayuan Chen)

 - Support mmap of vmlinux BTF data (Lorenz Bauer)

 - Support BPF rbtree traversal and list peeking (Martin KaFai Lau)

 - Tests for sockmap/sockhash redirection (Michal Luczaj)

 - Introduce kfuncs for memory reads into dynptrs (Mykyta Yatsenko)

 - Add support for dma-buf iterators in BPF (T.J. Mercier)

 - The verifier support for __bpf_trap() (Yonghong Song)

* tag 'bpf-next-6.16' of git://git.kernel.org/pub/scm/linux/kernel/git/bpf/bpf-next: (135 commits)
  bpf, arm64: Remove unused-but-set function and variable.
  selftests/bpf: Add tests with stack ptr register in conditional jmp
  bpf: Do not include stack ptr register in precision backtracking bookkeeping
  selftests/bpf: enable many-args tests for arm64
  bpf, arm64: Support up to 12 function arguments
  bpf: Check rcu_read_lock_trace_held() in bpf_map_lookup_percpu_elem()
  bpf: Avoid __bpf_prog_ret0_warn when jit fails
  bpftool: Add support for custom BTF path in prog load/loadall
  selftests/bpf: Add unit tests with __bpf_trap() kfunc
  bpf: Warn with __bpf_trap() kfunc maybe due to uninitialized variable
  bpf: Remove special_kfunc_set from verifier
  selftests/bpf: Add test for open coded dmabuf_iter
  selftests/bpf: Add test for dmabuf_iter
  bpf: Add open coded dmabuf iterator
  bpf: Add dmabuf iterator
  dma-buf: Rename debugfs symbols
  bpf: Fix error return value in bpf_copy_from_user_dynptr
  libbpf: Use mmap to parse vmlinux BTF from sysfs
  selftests: bpf: Add a test for mmapable vmlinux BTF
  btf: Allow mmap of vmlinux btf
  ...

Commit: 90b83efa6701656e02c86e7df2cb1765ea602d07
Parents: 1b98f357 · c5cebb24

Merge commit. Changes below are compared with the first parent: 1b98f357.

Changes

108 files changed · +5801 −1713

File	Status	Add	Del	Content
Documentation/bpf/bpf_iterators.rst	modified	+112	−5	View file
Documentation/bpf/kfuncs.rst	modified	+17	−0	View file
arch/arm64/net/bpf_jit_comp.c	modified	+171	−71	View file
arch/riscv/net/bpf_jit.h	modified	+15	−0	View file
arch/riscv/net/bpf_jit_comp64.c	modified	+227	−105	View file
arch/riscv/net/bpf_jit_core.c	modified	+1	−2	View file
arch/s390/include/asm/nospec-branch.h	modified	+0	−4	View file
arch/s390/net/bpf_jit_comp.c	modified	+66	−72	View file
drivers/dma-buf/dma-buf.c	modified	+78	−20	View file
include/asm-generic/vmlinux.lds.h	modified	+2	−1	View file
include/linux/bpf-cgroup.h	modified	+0	−8	View file
include/linux/bpf.h	modified	+20	−0	View file
include/linux/bpf_verifier.h	modified	+20	−4	View file
include/linux/dma-buf.h	modified	+2	−2	View file
include/uapi/linux/bpf.h	modified	+12	−7	View file
kernel/bpf/Makefile	modified	+3	−0	View file
kernel/bpf/bpf_struct_ops.c	modified	+1	−1	View file
kernel/bpf/btf.c	modified	+20	−25	View file
kernel/bpf/cgroup.c	modified	+0	−32	View file
kernel/bpf/core.c	modified	+17	−12	View file
kernel/bpf/dmabuf_iter.c	added	+150	−0	View file
kernel/bpf/hashtab.c	modified	+72	−76	View file
kernel/bpf/helpers.c	modified	+118	−15	View file
kernel/bpf/syscall.c	modified	+6	−4	View file
kernel/bpf/sysfs_btf.c	modified	+32	−0	View file
kernel/bpf/verifier.c	modified	+334	−302	View file
kernel/sched/ext.c	modified	+1	−14	View file
kernel/trace/bpf_trace.c	modified	+211	−110	View file
kernel/trace/trace_uprobe.c	modified	+1	−1	View file
net/bpf/test_run.c	modified	+7	−1	View file
net/core/filter.c	modified	+0	−14	View file
net/core/skmsg.c	modified	+35	−21	View file
net/tls/tls_sw.c	modified	+13	−2	View file
scripts/Makefile.btf	modified	+2	−0	View file
scripts/bpf_doc.py	modified	+99	−20	View file
tools/bpf/bpftool/Documentation/bpftool-prog.rst	modified	+8	−2	View file
tools/bpf/bpftool/bash-completion/bpftool	modified	+2	−2	View file
tools/bpf/bpftool/cgroup.c	modified	+7	−7	View file
tools/bpf/bpftool/link.c	modified	+3	−0	View file
tools/bpf/bpftool/prog.c	modified	+11	−1	View file
tools/include/uapi/linux/bpf.h	modified	+12	−7	View file
tools/lib/bpf/bpf_core_read.h	modified	+6	−0	View file
tools/lib/bpf/bpf_helpers.h	modified	+8	−0	View file
tools/lib/bpf/btf.c	modified	+171	−55	View file
tools/lib/bpf/libbpf.c	modified	+48	−39	View file
tools/lib/bpf/libbpf.h	modified	+6	−0	View file
tools/lib/bpf/libbpf.map	modified	+4	−0	View file
tools/lib/bpf/libbpf_internal.h	modified	+9	−0	View file
tools/lib/bpf/linker.c	modified	+3	−3	View file
tools/lib/bpf/nlattr.c	modified	+7	−8	View file
tools/testing/selftests/bpf/DENYLIST	modified	+1	−0	View file
tools/testing/selftests/bpf/DENYLIST.aarch64	modified	+0	−2	View file
tools/testing/selftests/bpf/Makefile	modified	+11	−5	View file
tools/testing/selftests/bpf/bench.c	modified	+4	−0	View file
tools/testing/selftests/bpf/benchs/bench_htab_mem.c	modified	+1	−2	View file
tools/testing/selftests/bpf/benchs/bench_sockmap.c	added	+598	−0	View file
tools/testing/selftests/bpf/bpf_experimental.h	modified	+5	−0	View file
tools/testing/selftests/bpf/config	modified	+3	−0	View file
tools/testing/selftests/bpf/prog_tests/arena_spin_lock.c	modified	+8	−6	View file
tools/testing/selftests/bpf/prog_tests/attach_probe.c	modified	+84	−0	View file
tools/testing/selftests/bpf/prog_tests/bpf_nf.c	modified	+6	−0	View file
tools/testing/selftests/bpf/prog_tests/btf_dedup_split.c	modified	+101	−0	View file
tools/testing/selftests/bpf/prog_tests/btf_split.c	modified	+52	−6	View file
tools/testing/selftests/bpf/prog_tests/btf_sysfs.c	added	+81	−0	View file
tools/testing/selftests/bpf/prog_tests/dmabuf_iter.c	added	+285	−0	View file
tools/testing/selftests/bpf/prog_tests/dynptr.c	modified	+13	−0	View file
tools/testing/selftests/bpf/prog_tests/fd_htab_lookup.c	added	+192	−0	View file
tools/testing/selftests/bpf/prog_tests/fill_link_info.c	modified	+16	−2	View file
tools/testing/selftests/bpf/prog_tests/kmem_cache_iter.c	modified	+1	−1	View file
tools/testing/selftests/bpf/prog_tests/linked_list.c	modified	+6	−0	View file
tools/testing/selftests/bpf/prog_tests/rbtree.c	modified	+6	−0	View file
tools/testing/selftests/bpf/prog_tests/sk_assign.c	modified	+3	−1	View file
tools/testing/selftests/bpf/prog_tests/socket_helpers.h	modified	+79	−5	View file
tools/testing/selftests/bpf/prog_tests/sockmap_helpers.h	modified	+11	−14	View file
tools/testing/selftests/bpf/prog_tests/sockmap_ktls.c	modified	+239	−58	View file
tools/testing/selftests/bpf/prog_tests/sockmap_listen.c	modified	+0	−457	View file
tools/testing/selftests/bpf/prog_tests/sockmap_redir.c	added	+465	−0	View file
tools/testing/selftests/bpf/prog_tests/tc_redirect.c	modified	+6	−5	View file
tools/testing/selftests/bpf/prog_tests/test_btf_ext.c	added	+64	−0	View file
tools/testing/selftests/bpf/prog_tests/test_veristat.c	modified	+5	−0	View file
tools/testing/selftests/bpf/prog_tests/verifier.c	modified	+2	−0	View file
tools/testing/selftests/bpf/prog_tests/xdp_metadata.c	modified	+21	−1	View file
tools/testing/selftests/bpf/progs/bench_sockmap_prog.c	added	+65	−0	View file
tools/testing/selftests/bpf/bpf_arena_spin_lock.h → tools/testing/selftests/bpf/progs/bpf_arena_spin_lock.h	renamed	+12	−3	View file
tools/testing/selftests/bpf/progs/bpf_misc.h	modified	+3	−2	View file
tools/testing/selftests/bpf/progs/dmabuf_iter.c	added	+101	−0	View file
tools/testing/selftests/bpf/progs/dynptr_success.c	modified	+230	−0	View file
tools/testing/selftests/bpf/progs/fd_htab_lookup.c	added	+25	−0	View file
tools/testing/selftests/bpf/progs/iters.c	modified	+0	−2	View file
tools/testing/selftests/bpf/progs/linked_list_peek.c	added	+113	−0	View file
tools/testing/selftests/bpf/progs/prepare.c	modified	+0	−1	View file
tools/testing/selftests/bpf/progs/rbtree_fail.c	modified	+15	−14	View file
tools/testing/selftests/bpf/progs/rbtree_search.c	added	+206	−0	View file
tools/testing/selftests/bpf/progs/set_global_vars.c	modified	+41	−0	View file
tools/testing/selftests/bpf/progs/test_btf_ext.c	added	+22	−0	View file
tools/testing/selftests/bpf/progs/test_sockmap_ktls.c	added	+36	−0	View file
tools/testing/selftests/bpf/progs/test_sockmap_redir.c	added	+68	−0	View file
tools/testing/selftests/bpf/progs/test_tcp_custom_syncookie.c	modified	+3	−1	View file
tools/testing/selftests/bpf/progs/verifier_bpf_trap.c	added	+71	−0	View file
tools/testing/selftests/bpf/progs/verifier_btf_ctx_access.c	modified	+12	−0	View file
tools/testing/selftests/bpf/progs/verifier_load_acquire.c	modified	+32	−16	View file
tools/testing/selftests/bpf/progs/verifier_precision.c	modified	+55	−3	View file
tools/testing/selftests/bpf/progs/verifier_store_release.c	modified	+27	−12	View file
tools/testing/selftests/bpf/progs/xdp_metadata.c	modified	+13	−0	View file
tools/testing/selftests/bpf/test_kmods/bpf_testmod.c	modified	+6	−2	View file
tools/testing/selftests/bpf/test_loader.c	modified	+8	−6	View file
tools/testing/selftests/bpf/test_verifier.c	modified	+4	−4	View file
tools/testing/selftests/bpf/veristat.c	modified	+94	−7	View file

Only the first 600 diff lines are shown. Download the complete patch.

Documentation/bpf/bpf_iterators.rst

View file

diff --git a/Documentation/bpf/bpf_iterators.rst b/Documentation/bpf/bpf_iterators.rst
index 7f514cb..189e3ec 100644
--- a/Documentation/bpf/bpf_iterators.rst
+++ b/Documentation/bpf/bpf_iterators.rst
@@ -2,10 +2,117 @@
 BPF Iterators
 =============
 
+--------
+Overview
+--------
+
+BPF supports two separate entities collectively known as "BPF iterators": BPF
+iterator *program type* and *open-coded* BPF iterators. The former is
+a stand-alone BPF program type which, when attached and activated by user,
+will be called once for each entity (task_struct, cgroup, etc) that is being
+iterated. The latter is a set of BPF-side APIs implementing iterator
+functionality and available across multiple BPF program types. Open-coded
+iterators provide similar functionality to BPF iterator programs, but gives
+more flexibility and control to all other BPF program types. BPF iterator
+programs, on the other hand, can be used to implement anonymous or BPF
+FS-mounted special files, whose contents are generated by attached BPF iterator
+program, backed by seq_file functionality. Both are useful depending on
+specific needs.
+
+When adding a new BPF iterator program, it is expected that similar
+functionality will be added as open-coded iterator for maximum flexibility.
+It's also expected that iteration logic and code will be maximally shared and
+reused between two iterator API surfaces.
 
-----------
-Motivation
-----------
+------------------------
+Open-coded BPF Iterators
+------------------------
+
+Open-coded BPF iterators are implemented as tightly-coupled trios of kfuncs
+(constructor, next element fetch, destructor) and iterator-specific type
+describing on-the-stack iterator state, which is guaranteed by the BPF
+verifier to not be tampered with outside of the corresponding
+constructor/destructor/next APIs.
+
+Each kind of open-coded BPF iterator has its own associated
+struct bpf_iter_<type>, where <type> denotes a specific type of iterator.
+bpf_iter_<type> state needs to live on BPF program stack, so make sure it's
+small enough to fit on BPF stack. For performance reasons its best to avoid
+dynamic memory allocation for iterator state and size the state struct big
+enough to fit everything necessary. But if necessary, dynamic memory
+allocation is a way to bypass BPF stack limitations. Note, state struct size
+is part of iterator's user-visible API, so changing it will break backwards
+compatibility, so be deliberate about designing it.
+
+All kfuncs (constructor, next, destructor) have to be named consistently as
+bpf_iter_<type>_{new,next,destroy}(), respectively. <type> represents iterator
+type, and iterator state should be represented as a matching
+`struct bpf_iter_<type>` state type. Also, all iter kfuncs should have
+a pointer to this `struct bpf_iter_<type>` as the very first argument.
+
+Additionally:
+  - Constructor, i.e., `bpf_iter_<type>_new()`, can have arbitrary extra
+    number of arguments. Return type is not enforced either.
+  - Next method, i.e., `bpf_iter_<type>_next()`, has to return a pointer
+    type and should have exactly one argument: `struct bpf_iter_<type> *`
+    (const/volatile/restrict and typedefs are ignored).
+  - Destructor, i.e., `bpf_iter_<type>_destroy()`, should return void and
+    should have exactly one argument, similar to the next method.
+  - `struct bpf_iter_<type>` size is enforced to be positive and
+    a multiple of 8 bytes (to fit stack slots correctly).
+
+Such strictness and consistency allows to build generic helpers abstracting
+important, but boilerplate, details to be able to use open-coded iterators
+effectively and ergonomically (see libbpf's bpf_for_each() macro). This is
+enforced at kfunc registration point by the kernel.
+
+Constructor/next/destructor implementation contract is as follows:
+  - constructor, `bpf_iter_<type>_new()`, always initializes iterator state on
+    the stack. If any of the input arguments are invalid, constructor should
+    make sure to still initialize it such that subsequent next() calls will
+    return NULL. I.e., on error, *return error and construct empty iterator*.
+    Constructor kfunc is marked with KF_ITER_NEW flag.
+
+  - next method, `bpf_iter_<type>_next()`, accepts pointer to iterator state
+    and produces an element. Next method should always return a pointer. The
+    contract between BPF verifier is that next method *guarantees* that it
+    will eventually return NULL when elements are exhausted. Once NULL is
+    returned, subsequent next calls *should keep returning NULL*. Next method
+    is marked with KF_ITER_NEXT (and should also have KF_RET_NULL as
+    NULL-returning kfunc, of course).
+
+  - destructor, `bpf_iter_<type>_destroy()`, is always called once. Even if
+    constructor failed or next returned nothing.  Destructor frees up any
+    resources and marks stack space used by `struct bpf_iter_<type>` as usable
+    for something else. Destructor is marked with KF_ITER_DESTROY flag.
+
+Any open-coded BPF iterator implementation has to implement at least these
+three methods. It is enforced that for any given type of iterator only
+applicable constructor/destructor/next are callable. I.e., verifier ensures
+you can't pass number iterator state into, say, cgroup iterator's next method.
+
+From a 10,000-feet BPF verification point of view, next methods are the points
+of forking a verification state, which are conceptually similar to what
+verifier is doing when validating conditional jumps. Verifier is branching out
+`call bpf_iter_<type>_next` instruction and simulates two outcomes: NULL
+(iteration is done) and non-NULL (new element is returned). NULL is simulated
+first and is supposed to reach exit without looping. After that non-NULL case
+is validated and it either reaches exit (for trivial examples with no real
+loop), or reaches another `call bpf_iter_<type>_next` instruction with the
+state equivalent to already (partially) validated one. State equivalency at
+that point means we technically are going to be looping forever without
+"breaking out" out of established "state envelope" (i.e., subsequent
+iterations don't add any new knowledge or constraints to the verifier state,
+so running 1, 2, 10, or a million of them doesn't matter). But taking into
+account the contract stating that iterator next method *has to* return NULL
+eventually, we can conclude that loop body is safe and will eventually
+terminate. Given we validated logic outside of the loop (NULL case), and
+concluded that loop body is safe (though potentially looping many times),
+verifier can claim safety of the overall program logic.
+
+------------------------
+BPF Iterators Motivation
+------------------------
 
 There are a few existing ways to dump kernel data into user space. The most
 popular one is the ``/proc`` system. For example, ``cat /proc/net/tcp6`` dumps
@@ -323,8 +430,8 @@ Now, in the userspace program, pass the pointer of struct to the
 
 ::
 
-  link = bpf_program__attach_iter(prog, &opts); iter_fd =
-  bpf_iter_create(bpf_link__fd(link));
+  link = bpf_program__attach_iter(prog, &opts);
+  iter_fd = bpf_iter_create(bpf_link__fd(link));
 
 If both *tid* and *pid* are zero, an iterator created from this struct
 ``bpf_iter_attach_opts`` will include every opened file of every task in the

Documentation/bpf/kfuncs.rst

View file

diff --git a/Documentation/bpf/kfuncs.rst b/Documentation/bpf/kfuncs.rst
index a8f5782..ae468b7 100644
--- a/Documentation/bpf/kfuncs.rst
+++ b/Documentation/bpf/kfuncs.rst
@@ -160,6 +160,23 @@ Or::
                 ...
         }
 
+2.2.6 __prog Annotation
+---------------------------
+This annotation is used to indicate that the argument needs to be fixed up to
+the bpf_prog_aux of the caller BPF program. Any value passed into this argument
+is ignored, and rewritten by the verifier.
+
+An example is given below::
+
+        __bpf_kfunc int bpf_wq_set_callback_impl(struct bpf_wq *wq,
+                                                 int (callback_fn)(void *map, int *key, void *value),
+                                                 unsigned int flags,
+                                                 void *aux__prog)
+         {
+                struct bpf_prog_aux *aux = aux__prog;
+                ...
+         }
+
 .. _BPF_kfunc_nodef:
 
 2.3 Using an existing kernel function

arch/arm64/net/bpf_jit_comp.c

View file

diff --git a/arch/arm64/net/bpf_jit_comp.c b/arch/arm64/net/bpf_jit_comp.c
index 634d784..da8b89d 100644
--- a/arch/arm64/net/bpf_jit_comp.c
+++ b/arch/arm64/net/bpf_jit_comp.c
@@ -2113,7 +2113,7 @@ bool bpf_jit_supports_subprog_tailcalls(void)
 }
 
 static void invoke_bpf_prog(struct jit_ctx *ctx, struct bpf_tramp_link *l,
-			    int args_off, int retval_off, int run_ctx_off,
+			    int bargs_off, int retval_off, int run_ctx_off,
 			    bool save_ret)
 {
 	__le32 *branch;
@@ -2155,7 +2155,7 @@ static void invoke_bpf_prog(struct jit_ctx *ctx, struct bpf_tramp_link *l,
 	branch = ctx->image + ctx->idx;
 	emit(A64_NOP, ctx);
 
-	emit(A64_ADD_I(1, A64_R(0), A64_SP, args_off), ctx);
+	emit(A64_ADD_I(1, A64_R(0), A64_SP, bargs_off), ctx);
 	if (!p->jited)
 		emit_addr_mov_i64(A64_R(1), (const u64)p->insnsi, ctx);
 
@@ -2180,7 +2180,7 @@ static void invoke_bpf_prog(struct jit_ctx *ctx, struct bpf_tramp_link *l,
 }
 
 static void invoke_bpf_mod_ret(struct jit_ctx *ctx, struct bpf_tramp_links *tl,
-			       int args_off, int retval_off, int run_ctx_off,
+			       int bargs_off, int retval_off, int run_ctx_off,
 			       __le32 **branches)
 {
 	int i;
@@ -2190,7 +2190,7 @@ static void invoke_bpf_mod_ret(struct jit_ctx *ctx, struct bpf_tramp_links *tl,
 	 */
 	emit(A64_STR64I(A64_ZR, A64_SP, retval_off), ctx);
 	for (i = 0; i < tl->nr_links; i++) {
-		invoke_bpf_prog(ctx, tl->links[i], args_off, retval_off,
+		invoke_bpf_prog(ctx, tl->links[i], bargs_off, retval_off,
 				run_ctx_off, true);
 		/* if (*(u64 *)(sp + retval_off) !=  0)
 		 *	goto do_fexit;
@@ -2204,23 +2204,125 @@ static void invoke_bpf_mod_ret(struct jit_ctx *ctx, struct bpf_tramp_links *tl,
 	}
 }
 
-static void save_args(struct jit_ctx *ctx, int args_off, int nregs)
+struct arg_aux {
+	/* how many args are passed through registers, the rest of the args are
+	 * passed through stack
+	 */
+	int args_in_regs;
+	/* how many registers are used to pass arguments */
+	int regs_for_args;
+	/* how much stack is used for additional args passed to bpf program
+	 * that did not fit in original function registers
+	 */
+	int bstack_for_args;
+	/* home much stack is used for additional args passed to the
+	 * original function when called from trampoline (this one needs
+	 * arguments to be properly aligned)
+	 */
+	int ostack_for_args;
+};
+
+static int calc_arg_aux(const struct btf_func_model *m,
+			 struct arg_aux *a)
 {
-	int i;
+	int stack_slots, nregs, slots, i;
+
+	/* verifier ensures m->nr_args <= MAX_BPF_FUNC_ARGS */
+	for (i = 0, nregs = 0; i < m->nr_args; i++) {
+		slots = (m->arg_size[i] + 7) / 8;
+		if (nregs + slots <= 8) /* passed through register ? */
+			nregs += slots;
+		else
+			break;
+	}
+
+	a->args_in_regs = i;
+	a->regs_for_args = nregs;
+	a->ostack_for_args = 0;
+	a->bstack_for_args = 0;
+
+	/* the rest arguments are passed through stack */
+	for (; i < m->nr_args; i++) {
+		/* We can not know for sure about exact alignment needs for
+		 * struct passed on stack, so deny those
+		 */
+		if (m->arg_flags[i] & BTF_FMODEL_STRUCT_ARG)
+			return -ENOTSUPP;
+		stack_slots = (m->arg_size[i] + 7) / 8;
+		a->bstack_for_args += stack_slots * 8;
+		a->ostack_for_args = a->ostack_for_args + stack_slots * 8;
+	}
+
+	return 0;
+}
 
-	for (i = 0; i < nregs; i++) {
-		emit(A64_STR64I(i, A64_SP, args_off), ctx);
-		args_off += 8;
+static void clear_garbage(struct jit_ctx *ctx, int reg, int effective_bytes)
+{
+	if (effective_bytes) {
+		int garbage_bits = 64 - 8 * effective_bytes;
+#ifdef CONFIG_CPU_BIG_ENDIAN
+		/* garbage bits are at the right end */
+		emit(A64_LSR(1, reg, reg, garbage_bits), ctx);
+		emit(A64_LSL(1, reg, reg, garbage_bits), ctx);
+#else
+		/* garbage bits are at the left end */
+		emit(A64_LSL(1, reg, reg, garbage_bits), ctx);
+		emit(A64_LSR(1, reg, reg, garbage_bits), ctx);
+#endif
 	}
 }
 
-static void restore_args(struct jit_ctx *ctx, int args_off, int nregs)
+static void save_args(struct jit_ctx *ctx, int bargs_off, int oargs_off,
+		      const struct btf_func_model *m,
+		      const struct arg_aux *a,
+		      bool for_call_origin)
 {
 	int i;
+	int reg;
+	int doff;
+	int soff;
+	int slots;
+	u8 tmp = bpf2a64[TMP_REG_1];
+
+	/* store arguments to the stack for the bpf program, or restore
+	 * arguments from stack for the original function
+	 */
+	for (reg = 0; reg < a->regs_for_args; reg++) {
+		emit(for_call_origin ?
+		     A64_LDR64I(reg, A64_SP, bargs_off) :
+		     A64_STR64I(reg, A64_SP, bargs_off),
+		     ctx);
+		bargs_off += 8;
+	}
+
+	soff = 32; /* on stack arguments start from FP + 32 */
+	doff = (for_call_origin ? oargs_off : bargs_off);
+
+	/* save on stack arguments */
+	for (i = a->args_in_regs; i < m->nr_args; i++) {
+		slots = (m->arg_size[i] + 7) / 8;
+		/* verifier ensures arg_size <= 16, so slots equals 1 or 2 */
+		while (slots-- > 0) {
+			emit(A64_LDR64I(tmp, A64_FP, soff), ctx);
+			/* if there is unused space in the last slot, clear
+			 * the garbage contained in the space.
+			 */
+			if (slots == 0 && !for_call_origin)
+				clear_garbage(ctx, tmp, m->arg_size[i] % 8);
+			emit(A64_STR64I(tmp, A64_SP, doff), ctx);
+			soff += 8;
+			doff += 8;
+		}
+	}
+}
 
-	for (i = 0; i < nregs; i++) {
-		emit(A64_LDR64I(i, A64_SP, args_off), ctx);
-		args_off += 8;
+static void restore_args(struct jit_ctx *ctx, int bargs_off, int nregs)
+{
+	int reg;
+
+	for (reg = 0; reg < nregs; reg++) {
+		emit(A64_LDR64I(reg, A64_SP, bargs_off), ctx);
+		bargs_off += 8;
 	}
 }
 
@@ -2243,17 +2345,21 @@ static bool is_struct_ops_tramp(const struct bpf_tramp_links *fentry_links)
  */
 static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 			      struct bpf_tramp_links *tlinks, void *func_addr,
-			      int nregs, u32 flags)
+			      const struct btf_func_model *m,
+			      const struct arg_aux *a,
+			      u32 flags)
 {
 	int i;
 	int stack_size;
 	int retaddr_off;
 	int regs_off;
 	int retval_off;
-	int args_off;
-	int nregs_off;
+	int bargs_off;
+	int nfuncargs_off;
 	int ip_off;
 	int run_ctx_off;
+	int oargs_off;
+	int nfuncargs;
 	struct bpf_tramp_links *fentry = &tlinks[BPF_TRAMP_FENTRY];
 	struct bpf_tramp_links *fexit = &tlinks[BPF_TRAMP_FEXIT];
 	struct bpf_tramp_links *fmod_ret = &tlinks[BPF_TRAMP_MODIFY_RETURN];
@@ -2262,31 +2368,38 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 	bool is_struct_ops = is_struct_ops_tramp(fentry);
 
 	/* trampoline stack layout:
-	 *                  [ parent ip         ]
-	 *                  [ FP                ]
-	 * SP + retaddr_off [ self ip           ]
-	 *                  [ FP                ]
+	 *                    [ parent ip         ]
+	 *                    [ FP                ]
+	 * SP + retaddr_off   [ self ip           ]
+	 *                    [ FP                ]
 	 *
-	 *                  [ padding           ] align SP to multiples of 16
+	 *                    [ padding           ] align SP to multiples of 16
 	 *
-	 *                  [ x20               ] callee saved reg x20
-	 * SP + regs_off    [ x19               ] callee saved reg x19
+	 *                    [ x20               ] callee saved reg x20
+	 * SP + regs_off      [ x19               ] callee saved reg x19
 	 *
-	 * SP + retval_off  [ return value      ] BPF_TRAMP_F_CALL_ORIG or
-	 *                                        BPF_TRAMP_F_RET_FENTRY_RET
+	 * SP + retval_off    [ return value      ] BPF_TRAMP_F_CALL_ORIG or
+	 *                                          BPF_TRAMP_F_RET_FENTRY_RET
+	 *                    [ arg reg N         ]
+	 *                    [ ...               ]
+	 * SP + bargs_off     [ arg reg 1         ] for bpf
 	 *
-	 *                  [ arg reg N         ]
-	 *                  [ ...               ]
-	 * SP + args_off    [ arg reg 1         ]
+	 * SP + nfuncargs_off [ arg regs count    ]
 	 *
-	 * SP + nregs_off   [ arg regs count    ]
+	 * SP + ip_off        [ traced function   ] BPF_TRAMP_F_IP_ARG flag
 	 *
-	 * SP + ip_off      [ traced function   ] BPF_TRAMP_F_IP_ARG flag
+	 * SP + run_ctx_off   [ bpf_tramp_run_ctx ]
 	 *
-	 * SP + run_ctx_off [ bpf_tramp_run_ctx ]
+	 *                    [ stack arg N       ]
+	 *                    [ ...               ]
+	 * SP + oargs_off     [ stack arg 1       ] for original func
 	 */
 
 	stack_size = 0;
+	oargs_off = stack_size;
+	if (flags & BPF_TRAMP_F_CALL_ORIG)
+		stack_size +=  a->ostack_for_args;
+
 	run_ctx_off = stack_size;
 	/* room for bpf_tramp_run_ctx */
 	stack_size += round_up(sizeof(struct bpf_tramp_run_ctx), 8);
@@ -2296,13 +2409,14 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 	if (flags & BPF_TRAMP_F_IP_ARG)
 		stack_size += 8;
 
-	nregs_off = stack_size;
+	nfuncargs_off = stack_size;
 	/* room for args count */
 	stack_size += 8;
 
-	args_off = stack_size;
+	bargs_off = stack_size;
 	/* room for args */
-	stack_size += nregs * 8;
+	nfuncargs = a->regs_for_args + a->bstack_for_args / 8;
+	stack_size += 8 * nfuncargs;
 
 	/* room for return value */
 	retval_off = stack_size;
@@ -2349,11 +2463,11 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 	}
 
 	/* save arg regs count*/
-	emit(A64_MOVZ(1, A64_R(10), nregs, 0), ctx);
-	emit(A64_STR64I(A64_R(10), A64_SP, nregs_off), ctx);
+	emit(A64_MOVZ(1, A64_R(10), nfuncargs, 0), ctx);
+	emit(A64_STR64I(A64_R(10), A64_SP, nfuncargs_off), ctx);
 
-	/* save arg regs */
-	save_args(ctx, args_off, nregs);
+	/* save args for bpf */
+	save_args(ctx, bargs_off, oargs_off, m, a, false);
 
 	/* save callee saved registers */
 	emit(A64_STR64I(A64_R(19), A64_SP, regs_off), ctx);
@@ -2369,7 +2483,7 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 	}
 
 	for (i = 0; i < fentry->nr_links; i++)
-		invoke_bpf_prog(ctx, fentry->links[i], args_off,
+		invoke_bpf_prog(ctx, fentry->links[i], bargs_off,
 				retval_off, run_ctx_off,
 				flags & BPF_TRAMP_F_RET_FENTRY_RET);
 
@@ -2379,12 +2493,13 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 		if (!branches)
 			return -ENOMEM;
 
-		invoke_bpf_mod_ret(ctx, fmod_ret, args_off, retval_off,
+		invoke_bpf_mod_ret(ctx, fmod_ret, bargs_off, retval_off,
 				   run_ctx_off, branches);
 	}
 
 	if (flags & BPF_TRAMP_F_CALL_ORIG) {
-		restore_args(ctx, args_off, nregs);
+		/* save args for original func */
+		save_args(ctx, bargs_off, oargs_off, m, a, true);
 		/* call original func */
 		emit(A64_LDR64I(A64_R(10), A64_SP, retaddr_off), ctx);
 		emit(A64_ADR(A64_LR, AARCH64_INSN_SIZE * 2), ctx);
@@ -2403,7 +2518,7 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 	}
 
 	for (i = 0; i < fexit->nr_links; i++)
-		invoke_bpf_prog(ctx, fexit->links[i], args_off, retval_off,
+		invoke_bpf_prog(ctx, fexit->links[i], bargs_off, retval_off,
 				run_ctx_off, false);
 
 	if (flags & BPF_TRAMP_F_CALL_ORIG) {
@@ -2417,7 +2532,7 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 	}
 
 	if (flags & BPF_TRAMP_F_RESTORE_REGS)
-		restore_args(ctx, args_off, nregs);
+		restore_args(ctx, bargs_off, a->regs_for_args);
 
 	/* restore callee saved register x19 and x20 */
 	emit(A64_LDR64I(A64_R(19), A64_SP, regs_off), ctx);
@@ -2454,21 +2569,6 @@ static int prepare_trampoline(struct jit_ctx *ctx, struct bpf_tramp_image *im,
 	return ctx->idx;
 }
 
-static int btf_func_model_nregs(const struct btf_func_model *m)
-{
-	int nregs = m->nr_args;
-	int i;
-
-	/* extra registers needed for struct argument */
-	for (i = 0; i < MAX_BPF_FUNC_ARGS; i++) {
-		/* The arg_size is at most 16 bytes, enforced by the verifier. */
-		if (m->arg_flags[i] & BTF_FMODEL_STRUCT_ARG)
-			nregs += (m->arg_size[i] + 7) / 8 - 1;
-	}
-
-	return nregs;
-}
-
 int arch_bpf_trampoline_size(const struct btf_func_model *m, u32 flags,
 			     struct bpf_tramp_links *tlinks, void *func_addr)
 {
@@ -2477,14 +2577,14 @@ int arch_bpf_trampoline_size(const struct btf_func_model *m, u32 flags,
 		.idx = 0,
 	};
 	struct bpf_tramp_image im;
-	int nregs, ret;
+	struct arg_aux aaux;
+	int ret;
 
-	nregs = btf_func_model_nregs(m);
-	/* the first 8 registers are used for arguments */
-	if (nregs > 8)
-		return -ENOTSUPP;
+	ret = calc_arg_aux(m, &aaux);
+	if (ret < 0)
+		return ret;
 
-	ret = prepare_trampoline(&ctx, &im, tlinks, func_addr, nregs, flags);
+	ret = prepare_trampoline(&ctx, &im, tlinks, func_addr, m, &aaux, flags);
 	if (ret < 0)
 		return ret;
 
@@ -2511,9 +2611,10 @@ int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *ro_image,
 				u32 flags, struct bpf_tramp_links *tlinks,
 				void *func_addr)
 {
-	int ret, nregs;
-	void *image, *tmp;
 	u32 size = ro_image_end - ro_image;
+	struct arg_aux aaux;
+	void *image, *tmp;
+	int ret;
 
 	/* image doesn't need to be in module memory range, so we can
 	 * use kvmalloc.
@@ -2529,13 +2630,12 @@ int arch_prepare_bpf_trampoline(struct bpf_tramp_image *im, void *ro_image,
 		.write = true,
 	};
 
-	nregs = btf_func_model_nregs(m);
-	/* the first 8 registers are used for arguments */
-	if (nregs > 8)
-		return -ENOTSUPP;
 
 	jit_fill_hole(image, (unsigned int)(ro_image_end - ro_image));
-	ret = prepare_trampoline(&ctx, im, tlinks, func_addr, nregs, flags);
+	ret = calc_arg_aux(m, &aaux);
+	if (ret)
+		goto out;
+	ret = prepare_trampoline(&ctx, im, tlinks, func_addr, m, &aaux, flags);
 
 	if (ret > 0 && validate_code(&ctx) < 0) {
 		ret = -EINVAL;

arch/riscv/net/bpf_jit.h

View file

diff --git a/arch/riscv/net/bpf_jit.h b/arch/riscv/net/bpf_jit.h
index 1d1c78d..e7b032d 100644
--- a/arch/riscv/net/bpf_jit.h
+++ b/arch/riscv/net/bpf_jit.h
@@ -608,6 +608,21 @@ static inline u32 rv_fence(u8 pred, u8 succ)
 	return rv_i_insn(imm11_0, 0, 0, 0, 0xf);
 }
 
+static inline void emit_fence_r_rw(struct rv_jit_context *ctx)
+{
+	emit(rv_fence(0x2, 0x3), ctx);
+}
+
+static inline void emit_fence_rw_w(struct rv_jit_context *ctx)
+{
+	emit(rv_fence(0x3, 0x1), ctx);
+}
+
+static inline void emit_fence_rw_rw(struct rv_jit_context *ctx)
+{
+	emit(rv_fence(0x3, 0x3), ctx);
+}
+
 static inline u32 rv_nop(void)
 {
 	return rv_i_insn(0, 0, 0, 0, 0x13);

arch/riscv/net/bpf_jit_comp64.c

View file

diff --git a/arch/riscv/net/bpf_jit_comp64.c b/arch/riscv/net/bpf_jit_comp64.c
index ca60db7..10e01ff 100644
--- a/arch/riscv/net/bpf_jit_comp64.c
+++ b/arch/riscv/net/bpf_jit_comp64.c
@@ -473,11 +473,212 @@ static inline void emit_kcfi(u32 hash, struct rv_jit_context *ctx)

This diff is incomplete. Download the complete patch.