From: Linus Torvalds <torvalds@linux-foundation.org>
To: Peter Anvin <hpa@zytor.com>, Ingo Molnar <mingo@kernel.org>,
Borislav Petkov <bp@alien8.de>,
Thomas Gleixner <tglx@linutronix.de>,
Rasmus Villemoes <linux@rasmusvillemoes.dk>,
Josh Poimboeuf <jpoimboe@kernel.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will@kernel.org>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
the arch/x86 maintainers <x86@kernel.org>,
linux-arm-kernel@lists.infradead.org,
linux-arch <linux-arch@vger.kernel.org>,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: [PATCH 1/7] vfs: dcache: move hashlen_hash() from callers into d_hash()
Date: Mon, 10 Jun 2024 13:48:15 -0700 [thread overview]
Message-ID: <20240610204821.230388-2-torvalds@linux-foundation.org> (raw)
In-Reply-To: <20240610204821.230388-1-torvalds@linux-foundation.org>
Both __d_lookup_rcu() and __d_lookup_rcu_op_compare() have the full
'name_hash' value of the qstr that they want to look up, and mask it off
to just the low 32-bit hash before calling down to d_hash().
Other callers just load the 32-bit hash and pass it as the argument.
If we move the masking into d_hash() itself, it simplifies the two
callers that currently do the masking, and is a no-op for the other
cases. It doesn't actually change the generated code since the compiler
will inline d_hash() and see that the end result is the same.
[ Technically, since the parse tree changes, the code generation may not
be 100% the same, and for me on x86-64, this does result in gcc
switching the operands around for one 'cmpl' instruction. So not
necessarily the exact same code generation, but equivalent ]
However, this does encapsulate the 'd_hash()' operation more, and makes
the shift operation in particular be a "shift 32 bits right, return full
word". Which matches the instruction semantics on both x86-64 and arm64
better, since a 32-bit shift will clear the upper bits.
That makes the next step of introducing a "shift by runtime constant"
more obvious and generates the shift with no extraneous type masking.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
fs/dcache.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/fs/dcache.c b/fs/dcache.c
index 407095188f83..8b4382f5c99d 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -100,9 +100,9 @@ static unsigned int d_hash_shift __ro_after_init;
static struct hlist_bl_head *dentry_hashtable __ro_after_init;
-static inline struct hlist_bl_head *d_hash(unsigned int hash)
+static inline struct hlist_bl_head *d_hash(unsigned long hashlen)
{
- return dentry_hashtable + (hash >> d_hash_shift);
+ return dentry_hashtable + ((u32)hashlen >> d_hash_shift);
}
#define IN_LOOKUP_SHIFT 10
@@ -2104,7 +2104,7 @@ static noinline struct dentry *__d_lookup_rcu_op_compare(
unsigned *seqp)
{
u64 hashlen = name->hash_len;
- struct hlist_bl_head *b = d_hash(hashlen_hash(hashlen));
+ struct hlist_bl_head *b = d_hash(hashlen);
struct hlist_bl_node *node;
struct dentry *dentry;
@@ -2171,7 +2171,7 @@ struct dentry *__d_lookup_rcu(const struct dentry *parent,
{
u64 hashlen = name->hash_len;
const unsigned char *str = name->name;
- struct hlist_bl_head *b = d_hash(hashlen_hash(hashlen));
+ struct hlist_bl_head *b = d_hash(hashlen);
struct hlist_bl_node *node;
struct dentry *dentry;
--
2.45.1.209.gc6f12300df
WARNING: multiple messages have this Message-ID (diff)
From: Linus Torvalds <torvalds@linux-foundation.org>
To: Peter Anvin <hpa@zytor.com>, Ingo Molnar <mingo@kernel.org>,
Borislav Petkov <bp@alien8.de>,
Thomas Gleixner <tglx@linutronix.de>,
Rasmus Villemoes <linux@rasmusvillemoes.dk>,
Josh Poimboeuf <jpoimboe@kernel.org>,
Catalin Marinas <catalin.marinas@arm.com>,
Will Deacon <will@kernel.org>
Cc: Linux Kernel Mailing List <linux-kernel@vger.kernel.org>,
the arch/x86 maintainers <x86@kernel.org>,
linux-arm-kernel@lists.infradead.org,
linux-arch <linux-arch@vger.kernel.org>,
Linus Torvalds <torvalds@linux-foundation.org>
Subject: [PATCH 1/7] vfs: dcache: move hashlen_hash() from callers into d_hash()
Date: Mon, 10 Jun 2024 13:48:15 -0700 [thread overview]
Message-ID: <20240610204821.230388-2-torvalds@linux-foundation.org> (raw)
In-Reply-To: <20240610204821.230388-1-torvalds@linux-foundation.org>
Both __d_lookup_rcu() and __d_lookup_rcu_op_compare() have the full
'name_hash' value of the qstr that they want to look up, and mask it off
to just the low 32-bit hash before calling down to d_hash().
Other callers just load the 32-bit hash and pass it as the argument.
If we move the masking into d_hash() itself, it simplifies the two
callers that currently do the masking, and is a no-op for the other
cases. It doesn't actually change the generated code since the compiler
will inline d_hash() and see that the end result is the same.
[ Technically, since the parse tree changes, the code generation may not
be 100% the same, and for me on x86-64, this does result in gcc
switching the operands around for one 'cmpl' instruction. So not
necessarily the exact same code generation, but equivalent ]
However, this does encapsulate the 'd_hash()' operation more, and makes
the shift operation in particular be a "shift 32 bits right, return full
word". Which matches the instruction semantics on both x86-64 and arm64
better, since a 32-bit shift will clear the upper bits.
That makes the next step of introducing a "shift by runtime constant"
more obvious and generates the shift with no extraneous type masking.
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
---
fs/dcache.c | 8 ++++----
1 file changed, 4 insertions(+), 4 deletions(-)
diff --git a/fs/dcache.c b/fs/dcache.c
index 407095188f83..8b4382f5c99d 100644
--- a/fs/dcache.c
+++ b/fs/dcache.c
@@ -100,9 +100,9 @@ static unsigned int d_hash_shift __ro_after_init;
static struct hlist_bl_head *dentry_hashtable __ro_after_init;
-static inline struct hlist_bl_head *d_hash(unsigned int hash)
+static inline struct hlist_bl_head *d_hash(unsigned long hashlen)
{
- return dentry_hashtable + (hash >> d_hash_shift);
+ return dentry_hashtable + ((u32)hashlen >> d_hash_shift);
}
#define IN_LOOKUP_SHIFT 10
@@ -2104,7 +2104,7 @@ static noinline struct dentry *__d_lookup_rcu_op_compare(
unsigned *seqp)
{
u64 hashlen = name->hash_len;
- struct hlist_bl_head *b = d_hash(hashlen_hash(hashlen));
+ struct hlist_bl_head *b = d_hash(hashlen);
struct hlist_bl_node *node;
struct dentry *dentry;
@@ -2171,7 +2171,7 @@ struct dentry *__d_lookup_rcu(const struct dentry *parent,
{
u64 hashlen = name->hash_len;
const unsigned char *str = name->name;
- struct hlist_bl_head *b = d_hash(hashlen_hash(hashlen));
+ struct hlist_bl_head *b = d_hash(hashlen);
struct hlist_bl_node *node;
struct dentry *dentry;
--
2.45.1.209.gc6f12300df
_______________________________________________
linux-arm-kernel mailing list
linux-arm-kernel@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-arm-kernel
next prev parent reply other threads:[~2024-06-10 20:48 UTC|newest]
Thread overview: 38+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-06-10 20:48 [PATCH 0/7] arm64 / x86-64: low-level code generation issues Linus Torvalds
2024-06-10 20:48 ` Linus Torvalds
2024-06-10 20:48 ` Linus Torvalds [this message]
2024-06-10 20:48 ` [PATCH 1/7] vfs: dcache: move hashlen_hash() from callers into d_hash() Linus Torvalds
2024-06-10 20:48 ` [PATCH 2/7] add default dummy 'runtime constant' infrastructure Linus Torvalds
2024-06-10 20:48 ` Linus Torvalds
2024-06-12 10:04 ` Borislav Petkov
2024-06-10 20:48 ` [PATCH 3/7] x86: add 'runtime constant' support Linus Torvalds
2024-06-10 20:48 ` Linus Torvalds
2024-06-10 20:48 ` [PATCH 4/7] arm64: " Linus Torvalds
2024-06-10 20:48 ` Linus Torvalds
2024-06-11 14:29 ` Mark Rutland
2024-06-11 14:29 ` Mark Rutland
2024-06-11 16:56 ` Linus Torvalds
2024-06-11 16:56 ` Linus Torvalds
2024-06-11 17:20 ` [PATCH 4/7 v2] " Linus Torvalds
2024-06-12 18:42 ` Mark Rutland
2024-06-11 17:48 ` [PATCH 4/7] " Mark Rutland
2024-06-11 17:59 ` Linus Torvalds
2024-06-11 18:59 ` Linus Torvalds
2024-06-11 20:22 ` Mark Rutland
2024-06-11 21:08 ` Linus Torvalds
2024-06-10 20:48 ` [PATCH 5/7] arm64: start using 'asm goto' for get_user() when available Linus Torvalds
2024-06-10 20:48 ` Linus Torvalds
2024-06-10 20:48 ` [PATCH 6/7] arm64: start using 'asm goto' for put_user() " Linus Torvalds
2024-06-10 20:48 ` Linus Torvalds
2024-06-11 21:55 ` Nathan Chancellor
2024-06-11 23:29 ` Linus Torvalds
2024-06-11 23:40 ` [PATCH 6/7 v2] " Linus Torvalds
2024-06-10 20:48 ` [PATCH 7/7] arm64: access_ok() optimization Linus Torvalds
2024-06-10 20:48 ` Linus Torvalds
2024-06-10 21:08 ` [PATCH 0/7] arm64 / x86-64: low-level code generation issues Linus Torvalds
2024-06-11 17:24 ` David Woodhouse
2024-06-11 17:40 ` Linus Torvalds
2024-06-11 17:52 ` Linus Torvalds
2024-06-12 18:41 ` Mark Rutland
2024-06-12 20:02 ` Linus Torvalds
2024-06-12 22:25 ` Linus Torvalds
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=20240610204821.230388-2-torvalds@linux-foundation.org \
--to=torvalds@linux-foundation.org \
--cc=bp@alien8.de \
--cc=catalin.marinas@arm.com \
--cc=hpa@zytor.com \
--cc=jpoimboe@kernel.org \
--cc=linux-arch@vger.kernel.org \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux@rasmusvillemoes.dk \
--cc=mingo@kernel.org \
--cc=tglx@linutronix.de \
--cc=will@kernel.org \
--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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.