public inbox for linux-kbuild@vger.kernel.org
 help / color / mirror / Atom feed
From: Sami Tolvanen <samitolvanen@google.com>
To: Masahiro Yamada <masahiroy@kernel.org>,
	Luis Chamberlain <mcgrof@kernel.org>,
	 Miguel Ojeda <ojeda@kernel.org>,
	Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Matthew Maurer <mmaurer@google.com>,
	Alex Gaynor <alex.gaynor@gmail.com>,
	 Wedson Almeida Filho <wedsonaf@gmail.com>,
	Gary Guo <gary@garyguo.net>, Petr Pavlu <petr.pavlu@suse.com>,
	 Neal Gompa <neal@gompa.dev>, Hector Martin <marcan@marcan.st>,
	Janne Grunau <j@jannau.net>,  Asahi Linux <asahi@lists.linux.dev>,
	linux-kbuild@vger.kernel.org,  linux-kernel@vger.kernel.org,
	linux-modules@vger.kernel.org,  rust-for-linux@vger.kernel.org,
	Sami Tolvanen <samitolvanen@google.com>
Subject: [PATCH v2 03/19] gendwarfksyms: Add address matching
Date: Thu, 15 Aug 2024 17:39:07 +0000	[thread overview]
Message-ID: <20240815173903.4172139-24-samitolvanen@google.com> (raw)
In-Reply-To: <20240815173903.4172139-21-samitolvanen@google.com>

The compiler may choose not to emit type information in DWARF for all
aliases, but it's possible for each alias to be exported separately.
To ensure we find type information for the aliases as well, read
{section, address} tuples from the symbol table and match symbols also
by address.

Signed-off-by: Sami Tolvanen <samitolvanen@google.com>
---
 scripts/gendwarfksyms/gendwarfksyms.c |   2 +
 scripts/gendwarfksyms/gendwarfksyms.h |   7 ++
 scripts/gendwarfksyms/symbols.c       | 161 +++++++++++++++++++++++++-
 3 files changed, 165 insertions(+), 5 deletions(-)

diff --git a/scripts/gendwarfksyms/gendwarfksyms.c b/scripts/gendwarfksyms/gendwarfksyms.c
index d209b237766b..e2f8ee5a4bf3 100644
--- a/scripts/gendwarfksyms/gendwarfksyms.c
+++ b/scripts/gendwarfksyms/gendwarfksyms.c
@@ -118,6 +118,8 @@ int main(int argc, const char **argv)
 			return -1;
 		}
 
+		check(symbol_read_symtab(fd));
+
 		dwfl = dwfl_begin(&callbacks);
 		if (!dwfl) {
 			error("dwfl_begin failed for '%s': %s", object_files[n],
diff --git a/scripts/gendwarfksyms/gendwarfksyms.h b/scripts/gendwarfksyms/gendwarfksyms.h
index 03f3e408a839..cb9106dfddb9 100644
--- a/scripts/gendwarfksyms/gendwarfksyms.h
+++ b/scripts/gendwarfksyms/gendwarfksyms.h
@@ -61,6 +61,10 @@ extern bool debug;
 /*
  * symbols.c
  */
+struct symbol_addr {
+	uint32_t section;
+	Elf64_Addr address;
+};
 
 static inline u32 name_hash(const char *name)
 {
@@ -69,10 +73,13 @@ static inline u32 name_hash(const char *name)
 
 struct symbol {
 	const char *name;
+	struct symbol_addr addr;
+	struct hlist_node addr_hash;
 	struct hlist_node name_hash;
 };
 
 extern int symbol_read_exports(FILE *file);
+extern int symbol_read_symtab(int fd);
 extern struct symbol *symbol_get(const char *name);
 
 /*
diff --git a/scripts/gendwarfksyms/symbols.c b/scripts/gendwarfksyms/symbols.c
index 673ad9cf9e77..f96acb941196 100644
--- a/scripts/gendwarfksyms/symbols.c
+++ b/scripts/gendwarfksyms/symbols.c
@@ -6,11 +6,43 @@
 #include "gendwarfksyms.h"
 
 #define SYMBOL_HASH_BITS 15
+
+/* struct symbol_addr -> struct symbol */
+static DEFINE_HASHTABLE(symbol_addrs, SYMBOL_HASH_BITS);
+/* name -> struct symbol */
 static DEFINE_HASHTABLE(symbol_names, SYMBOL_HASH_BITS);
 
+static inline u32 symbol_addr_hash(const struct symbol_addr *addr)
+{
+	return jhash(addr, sizeof(struct symbol_addr), 0);
+}
+
 typedef int (*symbol_callback_t)(struct symbol *, void *arg);
 
-static int for_each(const char *name, symbol_callback_t func, void *data)
+static int __for_each_addr(struct symbol *sym, symbol_callback_t func,
+			   void *data)
+{
+	struct hlist_node *tmp;
+	struct symbol *match = NULL;
+	int processed = 0;
+
+	hash_for_each_possible_safe(symbol_addrs, match, tmp, addr_hash,
+				    symbol_addr_hash(&sym->addr)) {
+		if (match == sym)
+			continue; /* Already processed */
+
+		if (match->addr.section == sym->addr.section &&
+		    match->addr.address == sym->addr.address) {
+			check(func(match, data));
+			++processed;
+		}
+	}
+
+	return processed;
+}
+
+static int for_each(const char *name, bool name_only, symbol_callback_t func,
+		    void *data)
 {
 	struct hlist_node *tmp;
 	struct symbol *match;
@@ -23,9 +55,13 @@ static int for_each(const char *name, symbol_callback_t func, void *data)
 		if (strcmp(match->name, name))
 			continue;
 
+		/* Call func for the match, and all address matches */
 		if (func)
 			check(func(match, data));
 
+		if (!name_only && match->addr.section != SHN_UNDEF)
+			return checkp(__for_each_addr(match, func, data)) + 1;
+
 		return 1;
 	}
 
@@ -34,7 +70,7 @@ static int for_each(const char *name, symbol_callback_t func, void *data)
 
 static bool is_exported(const char *name)
 {
-	return checkp(for_each(name, NULL, NULL)) > 0;
+	return checkp(for_each(name, true, NULL, NULL)) > 0;
 }
 
 int symbol_read_exports(FILE *file)
@@ -57,13 +93,14 @@ int symbol_read_exports(FILE *file)
 		if (is_exported(name))
 			continue; /* Ignore duplicates */
 
-		sym = malloc(sizeof(struct symbol));
+		sym = calloc(1, sizeof(struct symbol));
 		if (!sym) {
-			error("malloc failed");
+			error("calloc failed");
 			return -1;
 		}
 
 		sym->name = name;
+		sym->addr.section = SHN_UNDEF;
 		name = NULL;
 
 		hash_add(symbol_names, &sym->name_hash, name_hash(sym->name));
@@ -91,6 +128,120 @@ struct symbol *symbol_get(const char *name)
 {
 	struct symbol *sym = NULL;
 
-	for_each(name, get_symbol, &sym);
+	for_each(name, false, get_symbol, &sym);
 	return sym;
 }
+
+typedef int (*elf_symbol_callback_t)(const char *name, GElf_Sym *sym,
+				     Elf32_Word xndx, void *arg);
+
+static int elf_for_each_symbol(int fd, elf_symbol_callback_t func, void *arg)
+{
+	size_t sym_size;
+	GElf_Shdr shdr_mem;
+	GElf_Shdr *shdr;
+	Elf_Data *xndx_data = NULL;
+	Elf_Scn *scn;
+	Elf *elf;
+
+	if (elf_version(EV_CURRENT) != EV_CURRENT) {
+		error("elf_version failed: %s", elf_errmsg(-1));
+		return -1;
+	}
+
+	elf = elf_begin(fd, ELF_C_READ_MMAP, NULL);
+	if (!elf) {
+		error("elf_begin failed: %s", elf_errmsg(-1));
+		return -1;
+	}
+
+	sym_size = gelf_getclass(elf) == ELFCLASS32 ? sizeof(Elf32_Sym) :
+						      sizeof(Elf64_Sym);
+
+	scn = elf_nextscn(elf, NULL);
+
+	while (scn) {
+		shdr = gelf_getshdr(scn, &shdr_mem);
+
+		if (shdr && shdr->sh_type == SHT_SYMTAB_SHNDX) {
+			xndx_data = elf_getdata(scn, NULL);
+			break;
+		}
+
+		scn = elf_nextscn(elf, scn);
+	}
+
+	scn = elf_nextscn(elf, NULL);
+
+	while (scn) {
+		shdr = gelf_getshdr(scn, &shdr_mem);
+
+		if (shdr && shdr->sh_type == SHT_SYMTAB) {
+			Elf_Data *data = elf_getdata(scn, NULL);
+			unsigned int nsyms = data->d_size / sym_size;
+			unsigned int n;
+
+			for (n = 0; n < nsyms; ++n) {
+				const char *name = NULL;
+				Elf32_Word xndx = 0;
+				GElf_Sym sym_mem;
+				GElf_Sym *sym;
+
+				sym = gelf_getsymshndx(data, xndx_data, n,
+						       &sym_mem, &xndx);
+
+				if (sym->st_shndx != SHN_XINDEX)
+					xndx = sym->st_shndx;
+
+				name = elf_strptr(elf, shdr->sh_link,
+						  sym->st_name);
+
+				/* Skip empty symbol names */
+				if (name && *name &&
+				    checkp(func(name, sym, xndx, arg)) > 0)
+					break;
+			}
+		}
+
+		scn = elf_nextscn(elf, scn);
+	}
+
+	return check(elf_end(elf));
+}
+
+static int set_symbol_addr(struct symbol *sym, void *arg)
+{
+	struct symbol_addr *addr = arg;
+
+	if (sym->addr.section == SHN_UNDEF) {
+		sym->addr.section = addr->section;
+		sym->addr.address = addr->address;
+		hash_add(symbol_addrs, &sym->addr_hash,
+			 symbol_addr_hash(&sym->addr));
+
+		debug("%s -> { %u, %lx }", sym->name, sym->addr.section,
+		      sym->addr.address);
+	} else {
+		warn("multiple addresses for symbol %s?", sym->name);
+	}
+
+	return 0;
+}
+
+static int process_symbol(const char *name, GElf_Sym *sym, Elf32_Word xndx,
+			  void *arg)
+{
+	struct symbol_addr addr = { .section = xndx, .address = sym->st_value };
+
+	/* Set addresses for exported symbols */
+	if (GELF_ST_BIND(sym->st_info) != STB_LOCAL &&
+	    addr.section != SHN_UNDEF)
+		checkp(for_each(name, true, set_symbol_addr, &addr));
+
+	return 0;
+}
+
+int symbol_read_symtab(int fd)
+{
+	return elf_for_each_symbol(fd, process_symbol, NULL);
+}
-- 
2.46.0.184.g6999bdac58-goog


  parent reply	other threads:[~2024-08-15 17:39 UTC|newest]

Thread overview: 105+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-08-15 17:39 [PATCH v2 00/19] Implement DWARF modversions Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 01/19] tools: Add gendwarfksyms Sami Tolvanen
2024-08-16  7:14   ` Greg Kroah-Hartman
2024-08-27 16:44     ` Sami Tolvanen
2024-08-26 17:41   ` Petr Pavlu
2024-08-26 18:47     ` Sami Tolvanen
2024-08-28 12:31       ` Petr Pavlu
2024-08-28 21:28         ` Sami Tolvanen
2024-08-28 17:45   ` Masahiro Yamada
2024-08-28 21:32     ` Sami Tolvanen
2024-09-05  2:29   ` Masahiro Yamada
2024-09-05 20:52     ` Sami Tolvanen
2024-09-10  9:43       ` Masahiro Yamada
2024-09-10 21:09         ` Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 02/19] gendwarfksyms: Add symbol list handling Sami Tolvanen
2024-08-27  9:16   ` Petr Pavlu
2024-08-27 18:47     ` Sami Tolvanen
2024-08-28 12:35   ` Petr Pavlu
2024-08-28 23:09     ` Sami Tolvanen
2024-09-02  9:52       ` Petr Pavlu
2024-08-28 18:16   ` Masahiro Yamada
2024-08-28 21:50     ` Sami Tolvanen
2024-09-01 10:59   ` Masahiro Yamada
2024-09-04 20:51     ` Sami Tolvanen
2024-08-15 17:39 ` Sami Tolvanen [this message]
2024-08-27 12:40   ` [PATCH v2 03/19] gendwarfksyms: Add address matching Petr Pavlu
2024-08-27 21:28     ` Sami Tolvanen
2024-08-28 18:22   ` Masahiro Yamada
2024-08-28 21:56     ` Sami Tolvanen
2024-09-01 11:10   ` Masahiro Yamada
2024-09-04 20:48     ` Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 04/19] gendwarfksyms: Add support for type pointers Sami Tolvanen
2024-08-28  6:50   ` Masahiro Yamada
2024-08-28  7:15     ` Masahiro Yamada
2024-08-28 21:58       ` Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 05/19] gendwarfksyms: Expand base_type Sami Tolvanen
2024-08-28 12:46   ` Petr Pavlu
2024-08-28 22:19     ` Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 06/19] gendwarfksyms: Add a cache for processed DIEs Sami Tolvanen
2024-08-28 18:15   ` Masahiro Yamada
2024-08-28 22:27     ` Sami Tolvanen
2024-09-02 10:05   ` Petr Pavlu
2024-09-05 17:19     ` Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 07/19] gendwarfksyms: Expand type modifiers and typedefs Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 08/19] gendwarfksyms: Expand subroutine_type Sami Tolvanen
2024-09-03 15:11   ` Petr Pavlu
2024-09-05 17:22     ` Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 09/19] gendwarfksyms: Expand array_type Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 10/19] gendwarfksyms: Expand structure types Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 11/19] gendwarfksyms: Limit structure expansion Sami Tolvanen
2024-09-03 15:15   ` Petr Pavlu
2024-09-05 18:15     ` Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 12/19] gendwarfksyms: Add die_map debugging Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 13/19] gendwarfksyms: Add symtypes output Sami Tolvanen
2024-09-10 14:58   ` Petr Pavlu
2024-09-10 21:15     ` Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 14/19] gendwarfksyms: Add symbol versioning Sami Tolvanen
2024-09-11 10:08   ` Petr Pavlu
2024-09-11 16:03     ` Sami Tolvanen
2024-09-12 10:28       ` Petr Pavlu
2024-08-15 17:39 ` [PATCH v2 15/19] gendwarfksyms: Add support for declaration-only data structures Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 16/19] gendwarfksyms: Add support for reserved structure fields Sami Tolvanen
2024-08-16  7:20   ` Greg Kroah-Hartman
2024-08-16 15:50     ` Sami Tolvanen
2024-08-17  7:41       ` Greg Kroah-Hartman
2024-08-17 13:19         ` Benno Lossin
2024-08-19 18:25           ` Greg Kroah-Hartman
2024-08-19 21:46             ` Benno Lossin
2024-08-19 19:38           ` Sami Tolvanen
2024-08-19 22:16             ` Benno Lossin
2024-08-20 18:47               ` Sami Tolvanen
2024-08-20 20:03                 ` Matthew Maurer
2024-08-21 11:31                   ` Benno Lossin
2024-08-21 23:01                     ` Sami Tolvanen
2024-08-21 23:29                     ` Greg Kroah-Hartman
2024-08-22  5:55                       ` Benno Lossin
2024-08-22  7:29                         ` Greg Kroah-Hartman
2024-08-22 12:00                           ` Benno Lossin
2024-08-22 23:53                             ` Greg Kroah-Hartman
2024-08-23 19:17                               ` Sami Tolvanen
2024-08-24 13:29                                 ` Benno Lossin
2024-08-24 13:27                               ` Benno Lossin
2024-08-30  9:34   ` Miroslav Benes
2024-08-31  0:05     ` Sami Tolvanen
2024-09-11 11:43       ` Petr Pavlu
2024-09-12 16:06         ` Sami Tolvanen
2024-09-12 18:08           ` Benno Lossin
2024-09-12 20:58             ` Sami Tolvanen
2024-09-12 21:58               ` Benno Lossin
2024-09-12 22:37                 ` Sami Tolvanen
2024-09-13  8:00                   ` Benno Lossin
2024-08-15 17:39 ` [PATCH v2 17/19] export: Add __gendwarfksyms_ptr_ references to exported symbols Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 18/19] x86/asm-prototypes: Include <asm/ptrace.h> Sami Tolvanen
2024-09-01 10:50   ` Masahiro Yamada
2024-09-04 20:47     ` Sami Tolvanen
2024-08-15 17:39 ` [PATCH v2 19/19] kbuild: Add gendwarfksyms as an alternative to genksyms Sami Tolvanen
2024-08-15 20:13 ` [PATCH v2 00/19] Implement DWARF modversions Sedat Dilek
2024-08-15 20:47   ` Sami Tolvanen
2024-08-21  0:12     ` Sedat Dilek
2024-08-16  7:15 ` Greg Kroah-Hartman
2024-08-22 16:43 ` Jonathan Corbet
2024-08-22 17:57   ` Sami Tolvanen
2024-08-28  7:04 ` Masahiro Yamada
2024-08-28 22:53   ` Sami Tolvanen
2024-09-02  9:57     ` Petr Pavlu

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=20240815173903.4172139-24-samitolvanen@google.com \
    --to=samitolvanen@google.com \
    --cc=alex.gaynor@gmail.com \
    --cc=asahi@lists.linux.dev \
    --cc=gary@garyguo.net \
    --cc=gregkh@linuxfoundation.org \
    --cc=j@jannau.net \
    --cc=linux-kbuild@vger.kernel.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-modules@vger.kernel.org \
    --cc=marcan@marcan.st \
    --cc=masahiroy@kernel.org \
    --cc=mcgrof@kernel.org \
    --cc=mmaurer@google.com \
    --cc=neal@gompa.dev \
    --cc=ojeda@kernel.org \
    --cc=petr.pavlu@suse.com \
    --cc=rust-for-linux@vger.kernel.org \
    --cc=wedsonaf@gmail.com \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox