Message-ID: <20250430004128.474388-4-irogers@google.com>
Date: Tue, 29 Apr 2025 17:41:25 -0700
From: Ian Rogers <irogers@...gle.com>
To: Peter Zijlstra <peterz@...radead.org>, Ingo Molnar <mingo@...hat.com>, 
	Arnaldo Carvalho de Melo <acme@...nel.org>, Namhyung Kim <namhyung@...nel.org>, 
	Mark Rutland <mark.rutland@....com>, 
	Alexander Shishkin <alexander.shishkin@...ux.intel.com>, Jiri Olsa <jolsa@...nel.org>, 
	Adrian Hunter <adrian.hunter@...el.com>, Kan Liang <kan.liang@...ux.intel.com>, 
	Miguel Ojeda <ojeda@...nel.org>, Alex Gaynor <alex.gaynor@...il.com>, 
	Boqun Feng <boqun.feng@...il.com>, Gary Guo <gary@...yguo.net>, 
	"Björn Roy Baron" <bjorn3_gh@...tonmail.com>, Benno Lossin <benno.lossin@...ton.me>, 
	Andreas Hindborg <a.hindborg@...nel.org>, Alice Ryhl <aliceryhl@...gle.com>, 
	Trevor Gross <tmgross@...ch.edu>, Danilo Krummrich <dakr@...nel.org>, 
	Nathan Chancellor <nathan@...nel.org>, Nick Desaulniers <nick.desaulniers+lkml@...il.com>, 
	Bill Wendling <morbo@...gle.com>, Justin Stitt <justinstitt@...gle.com>, 
	James Clark <james.clark@...aro.org>, Howard Chu <howardchu95@...il.com>, 
	Jiapeng Chong <jiapeng.chong@...ux.alibaba.com>, Ravi Bangoria <ravi.bangoria@....com>, 
	"Masami Hiramatsu (Google)" <mhiramat@...nel.org>, Stephen Brennan <stephen.s.brennan@...cle.com>, 
	linux-kernel@...r.kernel.org, linux-perf-users@...r.kernel.org, 
	rust-for-linux@...r.kernel.org, llvm@...ts.linux.dev, 
	Daniel Xu <dxu@...uu.xyz>, Ariel Ben-Yehuda <ariel.byd@...il.com>
Cc: Ian Rogers <irogers@...gle.com>
Subject: [PATCH v2 3/6] perf demangle-rust: Remove previous legacy rust decoder

Code is unused since the introduction of rustc-demangle demangler.

Reviewed-by: Miguel Ojeda <ojeda@...nel.org>
Signed-off-by: Ian Rogers <irogers@...gle.com>
---
 tools/perf/util/demangle-rust.c | 269 --------------------------------
 tools/perf/util/demangle-rust.h |   8 -
 2 files changed, 277 deletions(-)
 delete mode 100644 tools/perf/util/demangle-rust.c
 delete mode 100644 tools/perf/util/demangle-rust.h

diff --git a/tools/perf/util/demangle-rust.c b/tools/perf/util/demangle-rust.c
deleted file mode 100644
index a659fc69f73a..000000000000
--- a/tools/perf/util/demangle-rust.c
+++ /dev/null
@@ -1,269 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0
-#include <string.h>
-#include "debug.h"
-
-#include "demangle-rust.h"
-
-/*
- * Mangled Rust symbols look like this:
- *
- *     _$LT$std..sys..fd..FileDesc$u20$as$u20$core..ops..Drop$GT$::drop::hc68340e1baa4987a
- *
- * The original symbol is:
- *
- *     <std::sys::fd::FileDesc as core::ops::Drop>::drop
- *
- * The last component of the path is a 64-bit hash in lowercase hex, prefixed
- * with "h". Rust does not have a global namespace between crates, an illusion
- * which Rust maintains by using the hash to distinguish things that would
- * otherwise have the same symbol.
- *
- * Any path component not starting with a XID_Start character is prefixed with
- * "_".
- *
- * The following escape sequences are used:
- *
- *     ","  =>  $C$
- *     "@"  =>  $SP$
- *     "*"  =>  $BP$
- *     "&"  =>  $RF$
- *     "<"  =>  $LT$
- *     ">"  =>  $GT$
- *     "("  =>  $LP$
- *     ")"  =>  $RP$
- *     " "  =>  $u20$
- *     "'"  =>  $u27$
- *     "["  =>  $u5b$
- *     "]"  =>  $u5d$
- *     "~"  =>  $u7e$
- *
- * A double ".." means "::" and a single "." means "-".
- *
- * The only characters allowed in the mangled symbol are a-zA-Z0-9 and _.:$
- */
-
-static const char *hash_prefix = "::h";
-static const size_t hash_prefix_len = 3;
-static const size_t hash_len = 16;
-
-static bool is_prefixed_hash(const char *start);
-static bool looks_like_rust(const char *sym, size_t len);
-static bool unescape(const char **in, char **out, const char *seq, char value);
-
-/*
- * INPUT:
- *     sym: symbol that has been through BFD-demangling
- *
- * This function looks for the following indicators:
- *
- *  1. The hash must consist of "h" followed by 16 lowercase hex digits.
- *
- *  2. As a sanity check, the hash must use between 5 and 15 of the 16 possible
- *     hex digits. This is true of 99.9998% of hashes so once in your life you
- *     may see a false negative. The point is to notice path components that
- *     could be Rust hashes but are probably not, like "haaaaaaaaaaaaaaaa". In
- *     this case a false positive (non-Rust symbol has an important path
- *     component removed because it looks like a Rust hash) is worse than a
- *     false negative (the rare Rust symbol is not demangled) so this sets the
- *     balance in favor of false negatives.
- *
- *  3. There must be no characters other than a-zA-Z0-9 and _.:$
- *
- *  4. There must be no unrecognized $-sign sequences.
- *
- *  5. There must be no sequence of three or more dots in a row ("...").
- */
-bool
-rust_is_mangled(const char *sym)
-{
-	size_t len, len_without_hash;
-
-	if (!sym)
-		return false;
-
-	len = strlen(sym);
-	if (len <= hash_prefix_len + hash_len)
-		/* Not long enough to contain "::h" + hash + something else */
-		return false;
-
-	len_without_hash = len - (hash_prefix_len + hash_len);
-	if (!is_prefixed_hash(sym + len_without_hash))
-		return false;
-
-	return looks_like_rust(sym, len_without_hash);
-}
-
-/*
- * A hash is the prefix "::h" followed by 16 lowercase hex digits. The hex
- * digits must comprise between 5 and 15 (inclusive) distinct digits.
- */
-static bool is_prefixed_hash(const char *str)
-{
-	const char *end;
-	bool seen[16];
-	size_t i;
-	int count;
-
-	if (strncmp(str, hash_prefix, hash_prefix_len))
-		return false;
-	str += hash_prefix_len;
-
-	memset(seen, false, sizeof(seen));
-	for (end = str + hash_len; str < end; str++)
-		if (*str >= '0' && *str <= '9')
-			seen[*str - '0'] = true;
-		else if (*str >= 'a' && *str <= 'f')
-			seen[*str - 'a' + 10] = true;
-		else
-			return false;
-
-	/* Count how many distinct digits seen */
-	count = 0;
-	for (i = 0; i < 16; i++)
-		if (seen[i])
-			count++;
-
-	return count >= 5 && count <= 15;
-}
-
-static bool looks_like_rust(const char *str, size_t len)
-{
-	const char *end = str + len;
-
-	while (str < end)
-		switch (*str) {
-		case '$':
-			if (!strncmp(str, "$C$", 3))
-				str += 3;
-			else if (!strncmp(str, "$SP$", 4)
-					|| !strncmp(str, "$BP$", 4)
-					|| !strncmp(str, "$RF$", 4)
-					|| !strncmp(str, "$LT$", 4)
-					|| !strncmp(str, "$GT$", 4)
-					|| !strncmp(str, "$LP$", 4)
-					|| !strncmp(str, "$RP$", 4))
-				str += 4;
-			else if (!strncmp(str, "$u20$", 5)
-					|| !strncmp(str, "$u27$", 5)
-					|| !strncmp(str, "$u5b$", 5)
-					|| !strncmp(str, "$u5d$", 5)
-					|| !strncmp(str, "$u7e$", 5))
-				str += 5;
-			else
-				return false;
-			break;
-		case '.':
-			/* Do not allow three or more consecutive dots */
-			if (!strncmp(str, "...", 3))
-				return false;
-			/* Fall through */
-		case 'a' ... 'z':
-		case 'A' ... 'Z':
-		case '0' ... '9':
-		case '_':
-		case ':':
-			str++;
-			break;
-		default:
-			return false;
-		}
-
-	return true;
-}
-
-/*
- * INPUT:
- *     sym: symbol for which rust_is_mangled(sym) returns true
- *
- * The input is demangled in-place because the mangled name is always longer
- * than the demangled one.
- */
-void
-rust_demangle_sym(char *sym)
-{
-	const char *in;
-	char *out;
-	const char *end;
-
-	if (!sym)
-		return;
-
-	in = sym;
-	out = sym;
-	end = sym + strlen(sym) - (hash_prefix_len + hash_len);
-
-	while (in < end)
-		switch (*in) {
-		case '$':
-			if (!(unescape(&in, &out, "$C$", ',')
-					|| unescape(&in, &out, "$SP$", '@')
-					|| unescape(&in, &out, "$BP$", '*')
-					|| unescape(&in, &out, "$RF$", '&')
-					|| unescape(&in, &out, "$LT$", '<')
-					|| unescape(&in, &out, "$GT$", '>')
-					|| unescape(&in, &out, "$LP$", '(')
-					|| unescape(&in, &out, "$RP$", ')')
-					|| unescape(&in, &out, "$u20$", ' ')
-					|| unescape(&in, &out, "$u27$", '\'')
-					|| unescape(&in, &out, "$u5b$", '[')
-					|| unescape(&in, &out, "$u5d$", ']')
-					|| unescape(&in, &out, "$u7e$", '~'))) {
-				pr_err("demangle-rust: unexpected escape sequence");
-				goto done;
-			}
-			break;
-		case '_':
-			/*
-			 * If this is the start of a path component and the next
-			 * character is an escape sequence, ignore the
-			 * underscore. The mangler inserts an underscore to make
-			 * sure the path component begins with a XID_Start
-			 * character.
-			 */
-			if ((in == sym || in[-1] == ':') && in[1] == '$')
-				in++;
-			else
-				*out++ = *in++;
-			break;
-		case '.':
-			if (in[1] == '.') {
-				/* ".." becomes "::" */
-				*out++ = ':';
-				*out++ = ':';
-				in += 2;
-			} else {
-				/* "." becomes "-" */
-				*out++ = '-';
-				in++;
-			}
-			break;
-		case 'a' ... 'z':
-		case 'A' ... 'Z':
-		case '0' ... '9':
-		case ':':
-			*out++ = *in++;
-			break;
-		default:
-			pr_err("demangle-rust: unexpected character '%c' in symbol\n",
-				*in);
-			goto done;
-		}
-
-done:
-	*out = '\0';
-}
-
-static bool unescape(const char **in, char **out, const char *seq, char value)
-{
-	size_t len = strlen(seq);
-
-	if (strncmp(*in, seq, len))
-		return false;
-
-	**out = value;
-
-	*in += len;
-	*out += 1;
-
-	return true;
-}
diff --git a/tools/perf/util/demangle-rust.h b/tools/perf/util/demangle-rust.h
deleted file mode 100644
index 2fca618b1aa5..000000000000
--- a/tools/perf/util/demangle-rust.h
+++ /dev/null
@@ -1,8 +0,0 @@
-/* SPDX-License-Identifier: GPL-2.0 */
-#ifndef __PERF_DEMANGLE_RUST
-#define __PERF_DEMANGLE_RUST 1
-
-bool rust_is_mangled(const char *str);
-void rust_demangle_sym(char *str);
-
-#endif /* __PERF_DEMANGLE_RUST */
-- 
2.49.0.901.g37484f566f-goog

Powered by blists - more mailing lists