ftrace: Synchronize the initialization of ftrace_opsgrafted

There's some internal state that ftrace_ops needs to have set, but since
it can be declared outside of the ftrace.c code, it calls
ftrace_ops_init() on the ops in every global function. The issue is that
if two tasks call it on the same ops at the same time it is possible to
have the initialization of one corrupt the initialization of the other
call.

Create a ops_mutex to use to synchronize every initialization of the
ftrace_ops. The mutex is taken within checking the ftrace_ops flag that
states it was initializied but the flag is checked again after the mutex
has been taken. Checking first outside the mutex allows it to shortcut
having to take the mutex. But then the check needs to be done again after
the mute is taken in case of races.

Cc: stable@vger.kernel.org
Link: https://patch.msgid.link/20260902095501.6b59af20@gandalf.local.home
Fixes: f04f24fb7e48d ("ftrace, kprobes: Fix a deadlock on ftrace_regex_lock")
Reported-by: sashiko-bot@kernel.org
Close: https://lore.kernel.org/all/20260829025528.49A831F000E9@smtp.kernel.org/
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>

1 files changed, 151 insertions, 0 deletions

diff --git a/tools/perf/util/blake2s.c b/tools/perf/util/blake2s.c
new file mode 100644
index 000000000..ce5d89a19
--- /dev/null
+++ b/tools/perf/util/blake2s.c
@@ -0,0 +1,151 @@
+// SPDX-License-Identifier: GPL-2.0 OR MIT
+/*
+ * Copyright (C) 2015-2019 Jason A. Donenfeld <Jason@zx2c4.com>. All Rights Reserved.
+ *
+ * This is an implementation of the BLAKE2s hash and PRF functions.
+ *
+ * Information: https://blake2.net/
+ */
+
+#include "blake2s.h"
+#include <linux/kernel.h>
+
+static inline u32 ror32(u32 v, int n)
+{
+	return (v >> n) | (v << (32 - n));
+}
+
+static inline void le32_to_cpu_array(u32 a[], size_t n)
+{
+	for (size_t i = 0; i < n; i++)
+		a[i] = le32_to_cpu((__force __le32)a[i]);
+}
+
+static inline void cpu_to_le32_array(u32 a[], size_t n)
+{
+	for (size_t i = 0; i < n; i++)
+		a[i] = (__force u32)cpu_to_le32(a[i]);
+}
+
+static const u8 blake2s_sigma[10][16] = {
+	{ 0, 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 },
+	{ 14, 10, 4, 8, 9, 15, 13, 6, 1, 12, 0, 2, 11, 7, 5, 3 },
+	{ 11, 8, 12, 0, 5, 2, 15, 13, 10, 14, 3, 6, 7, 1, 9, 4 },
+	{ 7, 9, 3, 1, 13, 12, 11, 14, 2, 6, 5, 10, 4, 0, 15, 8 },
+	{ 9, 0, 5, 7, 2, 4, 10, 15, 14, 1, 11, 12, 6, 8, 3, 13 },
+	{ 2, 12, 6, 10, 0, 11, 8, 3, 4, 13, 7, 5, 15, 14, 1, 9 },
+	{ 12, 5, 1, 15, 14, 13, 4, 10, 0, 7, 6, 3, 9, 2, 8, 11 },
+	{ 13, 11, 7, 14, 12, 1, 3, 9, 5, 0, 15, 4, 8, 6, 2, 10 },
+	{ 6, 15, 14, 9, 11, 3, 0, 8, 12, 2, 13, 7, 1, 4, 10, 5 },
+	{ 10, 2, 8, 4, 7, 6, 1, 5, 15, 11, 9, 14, 3, 12, 13, 0 },
+};
+
+static inline void blake2s_increment_counter(struct blake2s_ctx *ctx, u32 inc)
+{
+	ctx->t[0] += inc;
+	ctx->t[1] += (ctx->t[0] < inc);
+}
+
+static void blake2s_compress(struct blake2s_ctx *ctx,
+			     const u8 *data, size_t nblocks, u32 inc)
+{
+	u32 m[16];
+	u32 v[16];
+	int i;
+
+	while (nblocks > 0) {
+		blake2s_increment_counter(ctx, inc);
+		memcpy(m, data, BLAKE2S_BLOCK_SIZE);
+		le32_to_cpu_array(m, ARRAY_SIZE(m));
+		memcpy(v, ctx->h, 32);
+		v[ 8] = BLAKE2S_IV0;
+		v[ 9] = BLAKE2S_IV1;
+		v[10] = BLAKE2S_IV2;
+		v[11] = BLAKE2S_IV3;
+		v[12] = BLAKE2S_IV4 ^ ctx->t[0];
+		v[13] = BLAKE2S_IV5 ^ ctx->t[1];
+		v[14] = BLAKE2S_IV6 ^ ctx->f[0];
+		v[15] = BLAKE2S_IV7 ^ ctx->f[1];
+
+#define G(r, i, a, b, c, d) do { \
+	a += b + m[blake2s_sigma[r][2 * i + 0]]; \
+	d = ror32(d ^ a, 16); \
+	c += d; \
+	b = ror32(b ^ c, 12); \
+	a += b + m[blake2s_sigma[r][2 * i + 1]]; \
+	d = ror32(d ^ a, 8); \
+	c += d; \
+	b = ror32(b ^ c, 7); \
+} while (0)
+
+#define ROUND(r) do { \
+	G(r, 0, v[0], v[ 4], v[ 8], v[12]); \
+	G(r, 1, v[1], v[ 5], v[ 9], v[13]); \
+	G(r, 2, v[2], v[ 6], v[10], v[14]); \
+	G(r, 3, v[3], v[ 7], v[11], v[15]); \
+	G(r, 4, v[0], v[ 5], v[10], v[15]); \
+	G(r, 5, v[1], v[ 6], v[11], v[12]); \
+	G(r, 6, v[2], v[ 7], v[ 8], v[13]); \
+	G(r, 7, v[3], v[ 4], v[ 9], v[14]); \
+} while (0)
+		ROUND(0);
+		ROUND(1);
+		ROUND(2);
+		ROUND(3);
+		ROUND(4);
+		ROUND(5);
+		ROUND(6);
+		ROUND(7);
+		ROUND(8);
+		ROUND(9);
+
+#undef G
+#undef ROUND
+
+		for (i = 0; i < 8; ++i)
+			ctx->h[i] ^= v[i] ^ v[i + 8];
+
+		data += BLAKE2S_BLOCK_SIZE;
+		--nblocks;
+	}
+}
+
+static inline void blake2s_set_lastblock(struct blake2s_ctx *ctx)
+{
+	ctx->f[0] = -1;
+}
+
+void blake2s_update(struct blake2s_ctx *ctx, const u8 *in, size_t inlen)
+{
+	const size_t fill = BLAKE2S_BLOCK_SIZE - ctx->buflen;
+
+	if (unlikely(!inlen))
+		return;
+	if (inlen > fill) {
+		memcpy(ctx->buf + ctx->buflen, in, fill);
+		blake2s_compress(ctx, ctx->buf, 1, BLAKE2S_BLOCK_SIZE);
+		ctx->buflen = 0;
+		in += fill;
+		inlen -= fill;
+	}
+	if (inlen > BLAKE2S_BLOCK_SIZE) {
+		const size_t nblocks = DIV_ROUND_UP(inlen, BLAKE2S_BLOCK_SIZE);
+
+		blake2s_compress(ctx, in, nblocks - 1, BLAKE2S_BLOCK_SIZE);
+		in += BLAKE2S_BLOCK_SIZE * (nblocks - 1);
+		inlen -= BLAKE2S_BLOCK_SIZE * (nblocks - 1);
+	}
+	memcpy(ctx->buf + ctx->buflen, in, inlen);
+	ctx->buflen += inlen;
+}
+
+void blake2s_final(struct blake2s_ctx *ctx, u8 *out)
+{
+	blake2s_set_lastblock(ctx);
+	memset(ctx->buf + ctx->buflen, 0,
+	       BLAKE2S_BLOCK_SIZE - ctx->buflen); /* Padding */
+	blake2s_compress(ctx, ctx->buf, 1, ctx->buflen);
+	cpu_to_le32_array(ctx->h, ARRAY_SIZE(ctx->h));
+	memcpy(out, ctx->h, ctx->outlen);
+	memset(ctx, 0, sizeof(*ctx));
+}