[PATCH v3 10/10] compiler: allow all arches to enable CONFIG_OPTIMIZE_INLINING

From: Masahiro Yamada
Date: Mon Apr 22 2019 - 23:23:15 EST


Commit 60a3cdd06394 ("x86: add optimized inlining") introduced
CONFIG_OPTIMIZE_INLINING, but it has been available only for x86.

The idea is obviously arch-agnostic. This commit moves the config
entry from arch/x86/Kconfig.debug to lib/Kconfig.debug so that all
architectures can benefit from it.

This can make a huge difference in kernel image size especially when
CONFIG_OPTIMIZE_FOR_SIZE is enabled.

For example, I got 3.5% smaller arm64 kernel for v5.1-rc1.

dec file
18983424 arch/arm64/boot/Image.before
18321920 arch/arm64/boot/Image.after

This also slightly improves the "Kernel hacking" Kconfig menu as
e61aca5158a8 ("Merge branch 'kconfig-diet' from Dave Hansen') suggested;
this config option would be a good fit in the "compiler option" menu.

Signed-off-by: Masahiro Yamada <yamada.masahiro@xxxxxxxxxxxxx>
Acked-by: Borislav Petkov <bp@xxxxxxx>
---

Changes in v3: None
Changes in v2:
- split into a separate patch

arch/x86/Kconfig | 3 ---
arch/x86/Kconfig.debug | 14 --------------
include/linux/compiler_types.h | 3 +--
lib/Kconfig.debug | 14 ++++++++++++++
4 files changed, 15 insertions(+), 19 deletions(-)

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 62fc3fda1a05..f214bb5d60d8 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -310,9 +310,6 @@ config ZONE_DMA32
config AUDIT_ARCH
def_bool y if X86_64

-config ARCH_SUPPORTS_OPTIMIZED_INLINING
- def_bool y
-
config ARCH_SUPPORTS_DEBUG_PAGEALLOC
def_bool y

diff --git a/arch/x86/Kconfig.debug b/arch/x86/Kconfig.debug
index 15d0fbe27872..f730680dc818 100644
--- a/arch/x86/Kconfig.debug
+++ b/arch/x86/Kconfig.debug
@@ -266,20 +266,6 @@ config CPA_DEBUG
---help---
Do change_page_attr() self-tests every 30 seconds.

-config OPTIMIZE_INLINING
- bool "Allow gcc to uninline functions marked 'inline'"
- ---help---
- This option determines if the kernel forces gcc to inline the functions
- developers have marked 'inline'. Doing so takes away freedom from gcc to
- do what it thinks is best, which is desirable for the gcc 3.x series of
- compilers. The gcc 4.x series have a rewritten inlining algorithm and
- enabling this option will generate a smaller kernel there. Hopefully
- this algorithm is so good that allowing gcc 4.x and above to make the
- decision will become the default in the future. Until then this option
- is there to test gcc for this.
-
- If unsure, say N.
-
config DEBUG_ENTRY
bool "Debug low-level entry code"
depends on DEBUG_KERNEL
diff --git a/include/linux/compiler_types.h b/include/linux/compiler_types.h
index ba814f18cb4c..19e58b9138a0 100644
--- a/include/linux/compiler_types.h
+++ b/include/linux/compiler_types.h
@@ -140,8 +140,7 @@ struct ftrace_likely_data {
* Do not use __always_inline here, since currently it expands to inline again
* (which would break users of __always_inline).
*/
-#if !defined(CONFIG_ARCH_SUPPORTS_OPTIMIZED_INLINING) || \
- !defined(CONFIG_OPTIMIZE_INLINING)
+#if !defined(CONFIG_OPTIMIZE_INLINING)
#define inline inline __attribute__((__always_inline__)) __gnu_inline \
__maybe_unused notrace
#else
diff --git a/lib/Kconfig.debug b/lib/Kconfig.debug
index 00dbcdbc9a0d..37402f210115 100644
--- a/lib/Kconfig.debug
+++ b/lib/Kconfig.debug
@@ -310,6 +310,20 @@ config HEADERS_CHECK
exported to $(INSTALL_HDR_PATH) (usually 'usr/include' in
your build tree), to make sure they're suitable.

+config OPTIMIZE_INLINING
+ bool "Allow compiler to uninline functions marked 'inline'"
+ help
+ This option determines if the kernel forces gcc to inline the functions
+ developers have marked 'inline'. Doing so takes away freedom from gcc to
+ do what it thinks is best, which is desirable for the gcc 3.x series of
+ compilers. The gcc 4.x series have a rewritten inlining algorithm and
+ enabling this option will generate a smaller kernel there. Hopefully
+ this algorithm is so good that allowing gcc 4.x and above to make the
+ decision will become the default in the future. Until then this option
+ is there to test gcc for this.
+
+ If unsure, say N.
+
config DEBUG_SECTION_MISMATCH
bool "Enable full Section mismatch analysis"
help
--
2.17.1