From: Peter Zijlstra <peterz@infradead.org>
To: tglx@linutronix.de, jpoimboe@redhat.com
Cc: linux-kernel@vger.kernel.org, x86@kernel.org,
peterz@infradead.org, mhiramat@kernel.org, mbenes@suse.cz,
brgerst@gmail.com
Subject: [PATCH v2 17/19] objtool: Optimize !vmlinux.o again
Date: Tue, 17 Mar 2020 18:02:51 +0100 [thread overview]
Message-ID: <20200317170910.819744197@infradead.org> (raw)
In-Reply-To: 20200317170234.897520633@infradead.org
When doing kbuild tests to see if the objtool changes affected those I
found that there was a measurable regression:
pre post
real 1m13.594 1m16.488s
user 34m58.246s 35m23.947s
sys 4m0.393s 4m27.312s
Perf showed that for small files the increased hash-table sizes were a
measurable difference. Since we already have -l "vmlinux" to
distinguish between the modes, make it also use a smaller portion of
the hash-tables.
This flips it into a small win:
real 1m14.143s
user 34m49.292s
sys 3m44.746s
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
---
tools/objtool/elf.c | 51 ++++++++++++++++++++++++++++++++++-----------------
tools/objtool/elf.h | 4 ++--
2 files changed, 36 insertions(+), 19 deletions(-)
--- a/tools/objtool/elf.c
+++ b/tools/objtool/elf.c
@@ -27,6 +27,22 @@ static inline u32 str_hash(const char *s
return jhash(str, strlen(str), 0);
}
+static inline int elf_hash_bits(void)
+{
+ return vmlinux ? 20 : 16;
+}
+
+#define elf_hash_add(hashtable, node, key) \
+ hlist_add_head(node, &hashtable[hash_min(key, elf_hash_bits())])
+
+static void elf_hash_init(struct hlist_head *table)
+{
+ __hash_init(table, 1U << elf_hash_bits());
+}
+
+#define elf_hash_for_each_possible(name, obj, member, key) \
+ hlist_for_each_entry(obj, &name[hash_min(key, elf_hash_bits())], member)
+
static void rb_add(struct rb_root *tree, struct rb_node *node,
int (*cmp)(struct rb_node *, const struct rb_node *))
{
@@ -115,7 +131,7 @@ struct section *find_section_by_name(str
{
struct section *sec;
- hash_for_each_possible(elf->section_name_hash, sec, name_hash, str_hash(name))
+ elf_hash_for_each_possible(elf->section_name_hash, sec, name_hash, str_hash(name))
if (!strcmp(sec->name, name))
return sec;
@@ -127,7 +143,7 @@ static struct section *find_section_by_i
{
struct section *sec;
- hash_for_each_possible(elf->section_hash, sec, hash, idx)
+ elf_hash_for_each_possible(elf->section_hash, sec, hash, idx)
if (sec->idx == idx)
return sec;
@@ -138,7 +154,7 @@ static struct symbol *find_symbol_by_ind
{
struct symbol *sym;
- hash_for_each_possible(elf->symbol_hash, sym, hash, idx)
+ elf_hash_for_each_possible(elf->symbol_hash, sym, hash, idx)
if (sym->idx == idx)
return sym;
@@ -205,7 +221,7 @@ struct symbol *find_symbol_by_name(struc
{
struct symbol *sym;
- hash_for_each_possible(elf->symbol_name_hash, sym, name_hash, str_hash(name))
+ elf_hash_for_each_possible(elf->symbol_name_hash, sym, name_hash, str_hash(name))
if (!strcmp(sym->name, name))
return sym;
@@ -309,8 +325,8 @@ static int read_sections(struct elf *elf
sec->len = sec->sh.sh_size;
list_add_tail(&sec->list, &elf->sections);
- hash_add(elf->section_hash, &sec->hash, sec->idx);
- hash_add(elf->section_name_hash, &sec->name_hash, str_hash(sec->name));
+ elf_hash_add(elf->section_hash, &sec->hash, sec->idx);
+ elf_hash_add(elf->section_name_hash, &sec->name_hash, str_hash(sec->name));
}
if (stats)
@@ -394,8 +410,8 @@ static int read_symbols(struct elf *elf)
else
entry = &sym->sec->symbol_list;
list_add(&sym->list, entry);
- hash_add(elf->symbol_hash, &sym->hash, sym->idx);
- hash_add(elf->symbol_name_hash, &sym->name_hash, str_hash(sym->name));
+ elf_hash_add(elf->symbol_hash, &sym->hash, sym->idx);
+ elf_hash_add(elf->symbol_name_hash, &sym->name_hash, str_hash(sym->name));
}
if (stats)
@@ -504,7 +520,7 @@ static int read_relas(struct elf *elf)
}
list_add_tail(&rela->list, &sec->rela_list);
- hash_add(elf->rela_hash, &rela->hash, rela_hash(rela));
+ elf_hash_add(elf->rela_hash, &rela->hash, rela_hash(rela));
nr_rela++;
}
max_rela = max(max_rela, nr_rela);
@@ -531,15 +547,16 @@ struct elf *elf_read(const char *name, i
perror("malloc");
return NULL;
}
- memset(elf, 0, sizeof(*elf));
+ memset(elf, 0, offsetof(struct elf, sections));
- hash_init(elf->symbol_hash);
- hash_init(elf->symbol_name_hash);
- hash_init(elf->section_hash);
- hash_init(elf->section_name_hash);
- hash_init(elf->rela_hash);
INIT_LIST_HEAD(&elf->sections);
+ elf_hash_init(elf->symbol_hash);
+ elf_hash_init(elf->symbol_name_hash);
+ elf_hash_init(elf->section_hash);
+ elf_hash_init(elf->section_name_hash);
+ elf_hash_init(elf->rela_hash);
+
elf->fd = open(name, flags);
if (elf->fd == -1) {
fprintf(stderr, "objtool: Can't open '%s': %s\n",
@@ -676,8 +693,8 @@ struct section *elf_create_section(struc
shstrtab->changed = true;
list_add_tail(&sec->list, &elf->sections);
- hash_add(elf->section_hash, &sec->hash, sec->idx);
- hash_add(elf->section_name_hash, &sec->name_hash, str_hash(sec->name));
+ elf_hash_add(elf->section_hash, &sec->hash, sec->idx);
+ elf_hash_add(elf->section_name_hash, &sec->name_hash, str_hash(sec->name));
return sec;
}
--- a/tools/objtool/elf.h
+++ b/tools/objtool/elf.h
@@ -80,8 +80,8 @@ struct elf {
struct list_head sections;
DECLARE_HASHTABLE(symbol_hash, 20);
DECLARE_HASHTABLE(symbol_name_hash, 20);
- DECLARE_HASHTABLE(section_hash, 16);
- DECLARE_HASHTABLE(section_name_hash, 16);
+ DECLARE_HASHTABLE(section_hash, 20);
+ DECLARE_HASHTABLE(section_name_hash, 20);
DECLARE_HASHTABLE(rela_hash, 20);
};
next prev parent reply other threads:[~2020-03-17 17:12 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-03-17 17:02 [PATCH v2 00/19] objtool: vmlinux.o and noinstr validation Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 01/19] objtool: Introduce validate_return() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 02/19] objtool: Rename func_for_each_insn() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 03/19] objtool: Rename func_for_each_insn_all() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 04/19] x86/kexec: Use RIP relative addressing Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 05/19] objtool: Optimize find_symbol_by_index() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 06/19] objtool: Add a statistics mode Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 07/19] objtool: Optimize find_section_by_index() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 08/19] objtool: Optimize find_section_by_name() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 09/19] objtool: Optimize find_symbol_*() and read_symbols() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 10/19] objtool: Rename find_containing_func() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 11/19] objtool: Resize insn_hash Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 12/19] objtool: Optimize find_symbol_by_name() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 13/19] objtool: Optimize read_sections() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 14/19] objtool: Delete cleanup() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 15/19] objtool: Optimize find_rela_by_dest_range() Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 16/19] objtool: Implement noinstr validation Peter Zijlstra
2020-03-17 21:00 ` Josh Poimboeuf
2020-03-18 9:03 ` Peter Zijlstra
2020-03-18 10:06 ` Peter Zijlstra
2020-03-18 13:21 ` Peter Zijlstra
2020-03-17 17:02 ` Peter Zijlstra [this message]
2020-03-18 13:20 ` [PATCH v2 17/19] objtool: Optimize !vmlinux.o again Peter Zijlstra
2020-03-20 16:20 ` Miroslav Benes
2020-03-21 15:14 ` Peter Zijlstra
2020-03-21 16:11 ` Peter Zijlstra
2020-03-23 7:27 ` Miroslav Benes
2020-03-17 17:02 ` [PATCH v2 18/19] objtool: Use sec_offset_hash() for insn_hash Peter Zijlstra
2020-03-17 17:02 ` [PATCH v2 19/19] objtool: Detect loading function pointers across noinstr Peter Zijlstra
2020-03-17 23:39 ` kbuild test robot
2020-03-17 23:43 ` Nick Desaulniers
2020-03-18 11:02 ` Peter Zijlstra
2020-03-18 7:18 ` kbuild test robot
2020-03-17 21:05 ` [PATCH v2 00/19] objtool: vmlinux.o and noinstr validation Josh Poimboeuf
2020-03-18 13:18 ` [RFC][PATCH v2 20/19] kbuild/objtool: Add objtool-vmlinux.o pass Peter Zijlstra
2020-03-18 14:13 ` Peter Zijlstra
2020-03-18 18:34 ` Josh Poimboeuf
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=20200317170910.819744197@infradead.org \
--to=peterz@infradead.org \
--cc=brgerst@gmail.com \
--cc=jpoimboe@redhat.com \
--cc=linux-kernel@vger.kernel.org \
--cc=mbenes@suse.cz \
--cc=mhiramat@kernel.org \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/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;
as well as URLs for NNTP newsgroup(s).