arm64: atomics: remove redundant static branch

Due to a historical oversight, we emit a redundant static branch for
each atomic/atomic64 operation when CONFIG_ARM64_LSE_ATOMICS is
selected. We can safely remove this, making the kernel Image reasonably
smaller.

When CONFIG_ARM64_LSE_ATOMICS is selected, every LSE atomic operation
has two preceding static branches with the same target, e.g.

	b	f7c <kernel_init_freeable+0xa4>
	b	f7c <kernel_init_freeable+0xa4>
	mov	w0, #0x1                   	// #1
	ldadd	w0, w0, [x19]

This is because the __lse_ll_sc_body() wrapper uses
system_uses_lse_atomics(), which checks both `arm64_const_caps_ready`
and `cpu_hwcap_keys[ARM64_HAS_LSE_ATOMICS]`, each of which emits a
static branch. This has been the case since commit:

  addfc38672c73efd ("arm64: atomics: avoid out-of-line ll/sc atomics")

However, there was never a need to check `arm64_const_caps_ready`, which
was itself introduced in commit:

  63a1e1c95e60e798 ("arm64/cpufeature: don't use mutex in bringup path")

... so that cpus_have_const_cap() could fall back to checking the
`cpu_hwcaps` bitmap prior to the static keys for individual caps
becoming enabled. As system_uses_lse_atomics() doesn't check
`cpu_hwcaps`, and doesn't need to as we can safely use the LL/SC atomics
prior to enabling the `ARM64_HAS_LSE_ATOMICS` static key, it doesn't
need to check `arm64_const_caps_ready`.

This patch removes the `arm64_const_caps_ready` check from
system_uses_lse_atomics(). As the arch_atomic_* routines are meant to be
safely usable in noinstr code, I've also marked
system_uses_lse_atomics() as __always_inline.

This results in one fewer static branch per atomic operation, with the
prior example becoming:

	b	f78 <kernel_init_freeable+0xa0>
	mov	w0, #0x1                   	// #1
	ldadd	w0, w0, [x19]

Each static branch consists of the branch itself and an associated
__jump_table entry. Removing these has a reasonable impact on the Image
size, with a GCC 11.1.0 defconfig v5.17-rc2 Image being reduced by
128KiB:

| [mark@lakrids:~/src/linux]% ls -al Image*
| -rw-r--r-- 1 mark mark 34619904 Feb  3 18:24 Image.baseline
| -rw-r--r-- 1 mark mark 34488832 Feb  3 18:33 Image.onebranch

Signed-off-by: Mark Rutland <mark.rutland@arm.com>
Cc: Ard Biesheuvel <ardb@kernel.org>
Cc: Catalin Marinas <catalin.marinas@arm.com>
Cc: Suzuki Poulose <suzuki.poulose@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lore.kernel.org/r/20220204104439.270567-1-mark.rutland@arm.com
Signed-off-by: Will Deacon <will@kernel.org>

1 files changed, 2 insertions, 4 deletions

diff --git a/arch/arm64/include/asm/lse.h b/arch/arm64/include/asm/lse.h
index 5d10051c3e62..29c85810ae69 100644
--- a/arch/arm64/include/asm/lse.h
+++ b/arch/arm64/include/asm/lse.h
@@ -17,12 +17,10 @@
 #include <asm/cpucaps.h>
 
 extern struct static_key_false cpu_hwcap_keys[ARM64_NCAPS];
-extern struct static_key_false arm64_const_caps_ready;
 
-static inline bool system_uses_lse_atomics(void)
+static __always_inline bool system_uses_lse_atomics(void)
 {
-	return (static_branch_likely(&arm64_const_caps_ready)) &&
-		static_branch_likely(&cpu_hwcap_keys[ARM64_HAS_LSE_ATOMICS]);
+	return static_branch_likely(&cpu_hwcap_keys[ARM64_HAS_LSE_ATOMICS]);
 }
 
 #define __lse_ll_sc_body(op, ...)					\