summaryrefslogtreecommitdiff
diff options
context:
space:
mode:
authorMarco Elver <elver@google.com>2026-02-16 17:16:22 +0300
committerWill Deacon <will@kernel.org>2026-02-26 21:03:07 +0300
commitabf1be684dc270b94b7c8782f562959b33766fc0 (patch)
tree4fd7aa580ec99944b1720bb38cf6fa03349171d8
parent6de23f81a5e08be8fbf5e8d7e9febc72a5b5f27f (diff)
downloadlinux-abf1be684dc270b94b7c8782f562959b33766fc0.tar.xz
arm64: Optimize __READ_ONCE() with CONFIG_LTO=y
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). 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). NOTE: The return-value-of-function-drops-qualifiers hack was first suggested by Al Viro in [1], which notes some of its limitations which make it unsuitable for a general __unqual_scalar_typeof() replacement. Notably, array types are not supported, and GCC 8.1-8.3 still fail. Why should we use it here? READ_ONCE() does not support reading whole arrays, and the GCC version problem only affects 3 minor releases of a very ancient still-supported GCC version; not only that, this arm64 READ_ONCE() version is currently only activated by LTO builds, which to-date are *only supported by Clang*! Link: https://lore.kernel.org/all/20260111182010.GH3634291@ZenIV/ [1] Signed-off-by: Marco Elver <elver@google.com> Signed-off-by: Will Deacon <will@kernel.org>
-rw-r--r--arch/arm64/include/asm/rwonce.h18
1 files changed, 14 insertions, 4 deletions
diff --git a/arch/arm64/include/asm/rwonce.h b/arch/arm64/include/asm/rwonce.h
index fc0fb42b0b64..9fd24cef3376 100644
--- a/arch/arm64/include/asm/rwonce.h
+++ b/arch/arm64/include/asm/rwonce.h
@@ -20,6 +20,17 @@
ARM64_HAS_LDAPR)
/*
+ * Replace this with typeof_unqual() when minimum compiler versions are
+ * increased to GCC 14 and Clang 19. For the time being, we need this
+ * workaround, which relies on function return values dropping qualifiers.
+ */
+#define __rwonce_typeof_unqual(x) typeof(({ \
+ __diag_push() \
+ __diag_ignore_all("-Wignored-qualifiers", "") \
+ ((typeof(x)(*)(void))0)(); \
+ __diag_pop() }))
+
+/*
* When building with LTO, there is an increased risk of the compiler
* converting an address dependency headed by a READ_ONCE() invocation
* into a control dependency and consequently allowing for harmful
@@ -32,8 +43,7 @@
#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 +66,9 @@
: "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 */