From: Marco Elver <elver@google.com>
To: elver@google.com, Peter Zijlstra <peterz@infradead.org>,
Will Deacon <will@kernel.org>
Cc: Ingo Molnar <mingo@kernel.org>,
Thomas Gleixner <tglx@linutronix.de>,
Boqun Feng <boqun.feng@gmail.com>,
Waiman Long <longman@redhat.com>,
Bart Van Assche <bvanassche@acm.org>,
llvm@lists.linux.dev,
David Laight <david.laight.linux@gmail.com>,
Catalin Marinas <catalin.marinas@arm.com>,
Arnd Bergmann <arnd@arndb.de>,
linux-arm-kernel@lists.infradead.org,
linux-kernel@vger.kernel.org
Subject: [PATCH v3 2/3] arm64: Optimize __READ_ONCE() with CONFIG_LTO=y
Date: Fri, 30 Jan 2026 14:28:25 +0100 [thread overview]
Message-ID: <20260130132951.2714396-3-elver@google.com> (raw)
In-Reply-To: <20260130132951.2714396-1-elver@google.com>
Rework arm64 LTO __READ_ONCE() to improve code generation as follows:
1. Replace _Generic-based __unqual_scalar_typeof() with more complete
__rwonce_typeof_unqual(). This strips qualifiers from all types, not
just integer types, which is required to be able to assign (must be
non-const) to __u.__val in the non-atomic case (required for #2).
Once our minimum compiler versions are bumped, this just becomes
TYPEOF_UNQUAL() (or typeof_unqual() should we decide to adopt C23
naming). Sadly the fallback version of __rwonce_typeof_unqual() cannot
be used as a general TYPEOF_UNQUAL() fallback (see code comments).
One subtle point here is that non-integer types of __val could be const
or volatile within the union with the old __unqual_scalar_typeof(), if
the passed variable is const or volatile. This would then result in a
forced load from the stack if __u.__val is volatile; in the case of
const, it does look odd if the underlying storage changes, but the
compiler is told said member is "const" -- it smells like UB.
2. Eliminate the atomic flag and ternary conditional expression. Move
the fallback volatile load into the default case of the switch,
ensuring __u is unconditionally initialized across all paths.
The statement expression now unconditionally returns __u.__val.
This refactoring appears to help the compiler improve (or fix) code
generation.
With a defconfig + LTO + debug options builds, we observe different
codegen for the following functions:
btrfs_reclaim_sweep (708 -> 1032 bytes)
btrfs_sinfo_bg_reclaim_threshold_store (200 -> 204 bytes)
check_mem_access (3652 -> 3692 bytes) [inlined bpf_map_is_rdonly]
console_flush_all (1268 -> 1264 bytes)
console_lock_spinning_disable_and_check (180 -> 176 bytes)
igb_add_filter (640 -> 636 bytes)
igb_config_tx_modes (2404 -> 2400 bytes)
kvm_vcpu_on_spin (480 -> 476 bytes)
map_freeze (376 -> 380 bytes)
netlink_bind (1664 -> 1656 bytes)
nmi_cpu_backtrace (404 -> 400 bytes)
set_rps_cpu (516 -> 520 bytes)
swap_cluster_readahead (944 -> 932 bytes)
tcp_accecn_third_ack (328 -> 336 bytes)
tcp_create_openreq_child (1764 -> 1772 bytes)
tcp_data_queue (5784 -> 5892 bytes)
tcp_ecn_rcv_synack (620 -> 628 bytes)
xen_manage_runstate_time (944 -> 896 bytes)
xen_steal_clock (340 -> 296 bytes)
Increase of some functions are due to more aggressive inlining due to
better codegen (in this build, e.g. bpf_map_is_rdonly is no longer
present due to being inlined completely).
Signed-off-by: Marco Elver <elver@google.com>
---
v3:
* Comment.
v2:
* Add __rwonce_typeof_unqual() as fallback for old compilers.
---
arch/arm64/include/asm/rwonce.h | 21 +++++++++++++++++----
1 file changed, 17 insertions(+), 4 deletions(-)
diff --git a/arch/arm64/include/asm/rwonce.h b/arch/arm64/include/asm/rwonce.h
index fc0fb42b0b64..42c9e8429274 100644
--- a/arch/arm64/include/asm/rwonce.h
+++ b/arch/arm64/include/asm/rwonce.h
@@ -19,6 +19,20 @@
"ldapr" #sfx "\t" #regs, \
ARM64_HAS_LDAPR)
+#ifdef USE_TYPEOF_UNQUAL
+#define __rwonce_typeof_unqual(x) TYPEOF_UNQUAL(x)
+#else
+/*
+ * Fallback for older compilers (Clang < 19).
+ *
+ * Uses the fact that, for all supported Clang versions, 'auto' correctly drops
+ * qualifiers. Unlike typeof_unqual(), the type must be completely defined, i.e.
+ * no forward-declared struct pointer dereferences. The array-to-pointer decay
+ * case does not matter for usage in READ_ONCE() either.
+ */
+#define __rwonce_typeof_unqual(x) typeof(({ auto ____t = (x); ____t; }))
+#endif
+
/*
* When building with LTO, there is an increased risk of the compiler
* converting an address dependency headed by a READ_ONCE() invocation
@@ -32,8 +46,7 @@
#define __READ_ONCE(x) \
({ \
typeof(&(x)) __x = &(x); \
- int atomic = 1; \
- union { __unqual_scalar_typeof(*__x) __val; char __c[1]; } __u; \
+ union { __rwonce_typeof_unqual(*__x) __val; char __c[1]; } __u; \
switch (sizeof(x)) { \
case 1: \
asm volatile(__LOAD_RCPC(b, %w0, %1) \
@@ -56,9 +69,9 @@
: "Q" (*__x) : "memory"); \
break; \
default: \
- atomic = 0; \
+ __u.__val = *(volatile typeof(*__x) *)__x; \
} \
- atomic ? (typeof(*__x))__u.__val : (*(volatile typeof(*__x) *)__x);\
+ __u.__val; \
})
#endif /* !BUILD_VDSO */
--
2.53.0.rc1.225.gd81095ad13-goog
next prev parent reply other threads:[~2026-01-30 13:30 UTC|newest]
Thread overview: 37+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-01-30 13:28 [PATCH v3 0/3] arm64: Fixes for __READ_ONCE() with CONFIG_LTO=y Marco Elver
2026-01-30 13:28 ` [PATCH v3 1/3] arm64: Fix non-atomic " Marco Elver
2026-01-30 15:06 ` David Laight
2026-01-30 13:28 ` Marco Elver [this message]
2026-01-30 15:11 ` [PATCH v3 2/3] arm64: Optimize " David Laight
2026-02-02 15:36 ` Will Deacon
2026-02-02 16:01 ` Peter Zijlstra
2026-02-02 16:05 ` Will Deacon
2026-02-02 17:48 ` Marco Elver
2026-02-02 19:28 ` David Laight
2026-01-30 13:28 ` [PATCH v3 3/3] arm64, compiler-context-analysis: Permit alias analysis through " Marco Elver
2026-01-30 15:13 ` David Laight
2026-02-02 15:39 ` Will Deacon
2026-02-02 19:29 ` David Laight
2026-02-03 11:47 ` Will Deacon
2026-02-04 10:46 ` Marco Elver
2026-02-04 13:14 ` Peter Zijlstra
2026-02-04 14:15 ` Will Deacon
2026-02-06 15:09 ` Marco Elver
2026-02-06 18:26 ` David Laight
2026-02-15 21:55 ` Marco Elver
2026-02-15 22:16 ` David Laight
2026-02-15 22:43 ` Marco Elver
2026-02-15 23:18 ` David Laight
2026-02-15 23:40 ` Linus Torvalds
2026-02-16 11:09 ` David Laight
2026-02-16 15:32 ` Linus Torvalds
2026-02-16 17:43 ` David Laight
2026-02-17 12:16 ` Peter Zijlstra
2026-02-17 14:25 ` David Laight
2026-02-17 16:23 ` Linus Torvalds
2026-02-17 16:32 ` Linus Torvalds
2026-02-18 19:34 ` Boqun Feng
2026-02-18 20:18 ` Linus Torvalds
2026-02-19 15:21 ` Gary Guo
2026-02-19 18:36 ` Linus Torvalds
2026-02-02 19:13 ` [PATCH v3 0/3] arm64: Fixes for " Will Deacon
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=20260130132951.2714396-3-elver@google.com \
--to=elver@google.com \
--cc=arnd@arndb.de \
--cc=boqun.feng@gmail.com \
--cc=bvanassche@acm.org \
--cc=catalin.marinas@arm.com \
--cc=david.laight.linux@gmail.com \
--cc=linux-arm-kernel@lists.infradead.org \
--cc=linux-kernel@vger.kernel.org \
--cc=llvm@lists.linux.dev \
--cc=longman@redhat.com \
--cc=mingo@kernel.org \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=will@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