FAILED: patch "[PATCH] kbuild: Remove support for Clang's ThinLTO caching" failed to apply to 6.1-stable tree
Subsystems (2)
[email protected]
maintainerThe patch below does not apply to the 6.1-stable tree.
If someone wants it applied there, or to any other stable or longterm tree, then please email the backport, including the original git commit id to <[email protected]>.
To reproduce the conflict and resubmit, you may use the following commands:
git fetch https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux.git/ linux-6.1.y
git checkout FETCH_HEAD
git cherry-pick -x aba091547ef6159d52471f42a3ef531b7b660ed8
# <resolve conflicts, build, test, etc.>
git commit -s
git send-email --to '<[email protected]>' --in-reply-to '2024061340-troubling-automated-9989@gregkh' --subject-prefix 'PATCH 6.1.y' HEAD^..
Possible dependencies:
aba091547ef6 ("kbuild: Remove support for Clang's ThinLTO caching") 1db773da58df ("kbuild: remove old Rust docs output path") 7ea01d3169a2 ("rust: delete rust-project.json when running make clean")
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From aba091547ef6159d52471f42a3ef531b7b660ed8 Mon Sep 17 00:00:00 2001
From: Nathan Chancellor <[email protected]>
Date: Wed, 1 May 2024 15:55:25 -0700
Subject: [PATCH] kbuild: Remove support for Clang's ThinLTO caching
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
There is an issue in clang's ThinLTO caching (enabled for the kernel via '--thinlto-cache-dir') with .incbin, which the kernel occasionally uses to include data within the kernel, such as the .config file for /proc/config.gz. For example, when changing the .config and rebuilding vmlinux, the copy of .config in vmlinux does not match the copy of .config in the build folder:
$ echo 'CONFIG_LTO_NONE=n CONFIG_LTO_CLANG_THIN=y CONFIG_IKCONFIG=y CONFIG_HEADERS_INSTALL=y' >kernel/configs/repro.config
$ make -skj"$(nproc)" ARCH=x86_64 LLVM=1 clean defconfig repro.config vmlinux ...
$ grep CONFIG_HEADERS_INSTALL .config CONFIG_HEADERS_INSTALL=y
$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=y
$ scripts/config -d HEADERS_INSTALL
$ make -kj"$(nproc)" ARCH=x86_64 LLVM=1 vmlinux
...
UPD kernel/config_data
GZIP kernel/config_data.gz
CC kernel/configs.o
...
LD vmlinux
...$ grep CONFIG_HEADERS_INSTALL .config # CONFIG_HEADERS_INSTALL is not set
$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=y
Without '--thinlto-cache-dir' or when using full LTO, this issue does not occur.
Benchmarking incremental builds on a few different machines with and without the cache shows a 20% increase in incremental build time without the cache when measured by touching init/main.c and running 'make all'.
ARCH=arm64 defconfig + CONFIG_LTO_CLANG_THIN=y on an arm64 host:
Benchmark 1: With ThinLTO cache
Time (mean ± σ): 56.347 s ± 0.163 s [User: 83.768 s, System: 24.661 s]
Range (min … max): 56.109 s … 56.594 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 67.740 s ± 0.479 s [User: 718.458 s, System: 31.797 s]
Range (min … max): 67.059 s … 68.556 s 10 runs Summary
With ThinLTO cache ran
1.20 ± 0.01 times faster than Without ThinLTO cache
ARCH=x86_64 defconfig + CONFIG_LTO_CLANG_THIN=y on an x86_64 host:
Benchmark 1: With ThinLTO cache
Time (mean ± σ): 85.772 s ± 0.252 s [User: 91.505 s, System: 8.408 s]
Range (min … max): 85.447 s … 86.244 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 103.833 s ± 0.288 s [User: 232.058 s, System: 8.569 s]
Range (min … max): 103.286 s … 104.124 s 10 runs Summary
With ThinLTO cache ran
1.21 ± 0.00 times faster than Without ThinLTO cache
While it is unfortunate to take this performance improvement off the table, correctness is more important. If/when this is fixed in LLVM, it can potentially be brought back in a conditional manner. Alternatively, a developer can just disable LTO if doing incremental compiles quickly is important, as a full compile cycle can still take over a minute even with the cache and it is unlikely that LTO will result in functional differences for a kernel change.
Makefile +2 -3
@@ -942,7 +942,6 @@ endif ifdef CONFIG_LTO_CLANG ifdef CONFIG_LTO_CLANG_THIN CC_FLAGS_LTO := -flto=thin -fsplit-lto-unit -KBUILD_LDFLAGS += --thinlto-cache-dir=$(extmod_prefix).thinlto-cache else CC_FLAGS_LTO := -flto endif
@@ -1480,7 +1479,7 @@ endif # CONFIG_MODULES # Directories & files removed with 'make clean' CLEAN_FILES += vmlinux.symvers modules-only.symvers \ modules.builtin modules.builtin.modinfo modules.nsdeps \ - compile_commands.json .thinlto-cache rust/test \ + compile_commands.json rust/test \ rust-project.json .vmlinux.objs .vmlinux.export.c # Directories & files removed with 'make mrproper'
@@ -1787,7 +1786,7 @@ PHONY += compile_commands.json clean-dirs := $(KBUILD_EXTMOD) clean: rm-files := $(KBUILD_EXTMOD)/Module.symvers $(KBUILD_EXTMOD)/modules.nsdeps \ - $(KBUILD_EXTMOD)/compile_commands.json $(KBUILD_EXTMOD)/.thinlto-cache + $(KBUILD_EXTMOD)/compile_commands.json PHONY += prepare # now expand this into a simple variable to reduce the cost of shell evaluations
Nathan Chancellor
maintainerPATCH 1/1 kbuild: Remove support for Clang's ThinLTO caching
commit aba091547ef6159d52471f42a3ef531b7b660ed8 upstream.
There is an issue in clang's ThinLTO caching (enabled for the kernel via '--thinlto-cache-dir') with .incbin, which the kernel occasionally uses to include data within the kernel, such as the .config file for /proc/config.gz. For example, when changing the .config and rebuilding vmlinux, the copy of .config in vmlinux does not match the copy of .config in the build folder:
$ echo 'CONFIG_LTO_NONE=n CONFIG_LTO_CLANG_THIN=y CONFIG_IKCONFIG=y CONFIG_HEADERS_INSTALL=y' >kernel/configs/repro.config
$ make -skj"$(nproc)" ARCH=x86_64 LLVM=1 clean defconfig repro.config vmlinux ...
$ grep CONFIG_HEADERS_INSTALL .config CONFIG_HEADERS_INSTALL=y
$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=y
$ scripts/config -d HEADERS_INSTALL
$ make -kj"$(nproc)" ARCH=x86_64 LLVM=1 vmlinux
...
UPD kernel/config_data
GZIP kernel/config_data.gz
CC kernel/configs.o
...
LD vmlinux
...$ grep CONFIG_HEADERS_INSTALL .config # CONFIG_HEADERS_INSTALL is not set
$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=y
Without '--thinlto-cache-dir' or when using full LTO, this issue does not occur.
Benchmarking incremental builds on a few different machines with and without the cache shows a 20% increase in incremental build time without the cache when measured by touching init/main.c and running 'make all'.
ARCH=arm64 defconfig + CONFIG_LTO_CLANG_THIN=y on an arm64 host:
Benchmark 1: With ThinLTO cache
Time (mean ± σ): 56.347 s ± 0.163 s [User: 83.768 s, System: 24.661 s]
Range (min … max): 56.109 s … 56.594 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 67.740 s ± 0.479 s [User: 718.458 s, System: 31.797 s]
Range (min … max): 67.059 s … 68.556 s 10 runs Summary
With ThinLTO cache ran
1.20 ± 0.01 times faster than Without ThinLTO cache
ARCH=x86_64 defconfig + CONFIG_LTO_CLANG_THIN=y on an x86_64 host:
Benchmark 1: With ThinLTO cache
Time (mean ± σ): 85.772 s ± 0.252 s [User: 91.505 s, System: 8.408 s]
Range (min … max): 85.447 s … 86.244 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 103.833 s ± 0.288 s [User: 232.058 s, System: 8.569 s]
Range (min … max): 103.286 s … 104.124 s 10 runs Summary
With ThinLTO cache ran
1.21 ± 0.00 times faster than Without ThinLTO cache
While it is unfortunate to take this performance improvement off the table, correctness is more important. If/when this is fixed in LLVM, it can potentially be brought back in a conditional manner. Alternatively, a developer can just disable LTO if doing incremental compiles quickly is important, as a full compile cycle can still take over a minute even with the cache and it is unlikely that LTO will result in functional differences for a kernel change.
--- Makefile | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
Makefile +2 -3
@@ -980,7 +980,6 @@ endif ifdef CONFIG_LTO_CLANG ifdef CONFIG_LTO_CLANG_THIN CC_FLAGS_LTO := -flto=thin -fsplit-lto-unit -KBUILD_LDFLAGS += --thinlto-cache-dir=$(extmod_prefix).thinlto-cache else CC_FLAGS_LTO := -flto endif
@@ -1588,7 +1587,7 @@ endif # CONFIG_MODULES # Directories & files removed with 'make clean' CLEAN_FILES += include/ksym vmlinux.symvers modules-only.symvers \ modules.builtin modules.builtin.modinfo modules.nsdeps \ - compile_commands.json .thinlto-cache rust/test rust/doc \ + compile_commands.json rust/test rust/doc \ .vmlinux.objs .vmlinux.export.c # Directories & files removed with 'make mrproper'
@@ -1884,7 +1883,7 @@ PHONY += compile_commands.json clean-dirs := $(KBUILD_EXTMOD) clean: rm-files := $(KBUILD_EXTMOD)/Module.symvers $(KBUILD_EXTMOD)/modules.nsdeps \ - $(KBUILD_EXTMOD)/compile_commands.json $(KBUILD_EXTMOD)/.thinlto-cache + $(KBUILD_EXTMOD)/compile_commands.json PHONY += prepare # now expand this into a simple variable to reduce the cost of shell evaluations
base-commit: ae9f2a70d69e9c840ee1eda201f09662ca7e2038
Signature · 1 line
2.45.2
Greg KH
maintainerRe: [PATCH 6.1.y] kbuild: Remove support for Clang's ThinLTO caching
Quoted 93 lines · Nathan Chancellor
commit
aba091547ef6159d52471f42a3ef531b7b660ed8upstream.There is an issue in clang's ThinLTO caching (enabled for the kernel via '--thinlto-cache-dir') with .incbin, which the kernel occasionally uses to include data within the kernel, such as the .config file for
/proc/config.gz. For example, when changing the .config and rebuilding vmlinux, the copy of .config in vmlinux does not match the copy of .config in the build folder:$ echo 'CONFIG_LTO_NONE=n CONFIG_LTO_CLANG_THIN=y CONFIG_IKCONFIG=y CONFIG_HEADERS_INSTALL=y' >kernel/configs/repro.config$ make -skj"$(nproc)" ARCH=x86_64 LLVM=1 clean defconfig repro.config vmlinux ...$ grep CONFIG_HEADERS_INSTALL .config CONFIG_HEADERS_INSTALL=y$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=y$ scripts/config -d HEADERS_INSTALL$ make -kj"$(nproc)" ARCH=x86_64 LLVM=1 vmlinux ... UPD kernel/config_data GZIP kernel/config_data.gz CC kernel/configs.o ... LD vmlinux ...$ grep CONFIG_HEADERS_INSTALL .config # CONFIG_HEADERS_INSTALL is not set$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=yWithout '--thinlto-cache-dir' or when using full LTO, this issue does not occur.
Benchmarking incremental builds on a few different machines with and without the cache shows a 20% increase in incremental build time without the cache when measured by touching
init/main.cand running 'make all'.ARCH=arm64 defconfig +
CONFIG_LTO_CLANG_THIN=yon an arm64 host:Benchmark 1: With ThinLTO cache
Time (mean ± σ): 56.347 s ± 0.163 s [User: 83.768 s, System: 24.661 s] Range (min … max): 56.109 s … 56.594 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 67.740 s ± 0.479 s [User: 718.458 s, System: 31.797 s] Range (min … max): 67.059 s … 68.556 s 10 runsSummary
With ThinLTO cache ran
1.20 ± 0.01 times faster than Without ThinLTO cacheARCH=x86_64 defconfig +
CONFIG_LTO_CLANG_THIN=yon anx86_64host:Benchmark 1: With ThinLTO cache
Time (mean ± σ): 85.772 s ± 0.252 s [User: 91.505 s, System: 8.408 s] Range (min … max): 85.447 s … 86.244 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 103.833 s ± 0.288 s [User: 232.058 s, System: 8.569 s] Range (min … max): 103.286 s … 104.124 s 10 runsSummary
With ThinLTO cache ran
1.21 ± 0.00 times faster than Without ThinLTO cacheWhile it is unfortunate to take this performance improvement off the table, correctness is more important. If/when this is fixed in LLVM, it can potentially be brought back in a conditional manner. Alternatively, a developer can just disable LTO if doing incremental compiles quickly is important, as a full compile cycle can still take over a minute even with the cache and it is unlikely that LTO will result in functional differences for a kernel change.
Fixes dc5723b02e52 ("kbuild: add support for Clang LTO")Reported-by Yifan Hong <[email protected]>Reported-by Masami Hiramatsu <[email protected]>Signed-off-by Nathan Chancellor <[email protected]>Signed-off-by Masahiro Yamada <[email protected]>[nathan: Address conflict in Makefile]Signed-off-by Nathan Chancellor <[email protected]>
Quoted 3 lines
--- Makefile | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
This applied to 5.15.y, not 6.1.y :(
Can you rebase and resend a fix for 6.1.y?
thanks,
greg k-h
Nathan Chancellor
maintainerRe: [PATCH 6.1.y] kbuild: Remove support for Clang's ThinLTO caching
Quoted 98 lines · Greg KH
Quoted 93 lines · Nathan Chancellor
commit
aba091547ef6159d52471f42a3ef531b7b660ed8upstream.There is an issue in clang's ThinLTO caching (enabled for the kernel via '--thinlto-cache-dir') with .incbin, which the kernel occasionally uses to include data within the kernel, such as the .config file for
/proc/config.gz. For example, when changing the .config and rebuilding vmlinux, the copy of .config in vmlinux does not match the copy of .config in the build folder:$ echo 'CONFIG_LTO_NONE=n CONFIG_LTO_CLANG_THIN=y CONFIG_IKCONFIG=y CONFIG_HEADERS_INSTALL=y' >kernel/configs/repro.config$ make -skj"$(nproc)" ARCH=x86_64 LLVM=1 clean defconfig repro.config vmlinux ...$ grep CONFIG_HEADERS_INSTALL .config CONFIG_HEADERS_INSTALL=y$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=y$ scripts/config -d HEADERS_INSTALL$ make -kj"$(nproc)" ARCH=x86_64 LLVM=1 vmlinux ... UPD kernel/config_data GZIP kernel/config_data.gz CC kernel/configs.o ... LD vmlinux ...$ grep CONFIG_HEADERS_INSTALL .config # CONFIG_HEADERS_INSTALL is not set$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=yWithout '--thinlto-cache-dir' or when using full LTO, this issue does not occur.
Benchmarking incremental builds on a few different machines with and without the cache shows a 20% increase in incremental build time without the cache when measured by touching
init/main.cand running 'make all'.ARCH=arm64 defconfig +
CONFIG_LTO_CLANG_THIN=yon an arm64 host:Benchmark 1: With ThinLTO cache
Time (mean ± σ): 56.347 s ± 0.163 s [User: 83.768 s, System: 24.661 s] Range (min … max): 56.109 s … 56.594 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 67.740 s ± 0.479 s [User: 718.458 s, System: 31.797 s] Range (min … max): 67.059 s … 68.556 s 10 runsSummary
With ThinLTO cache ran
1.20 ± 0.01 times faster than Without ThinLTO cacheARCH=x86_64 defconfig +
CONFIG_LTO_CLANG_THIN=yon anx86_64host:Benchmark 1: With ThinLTO cache
Time (mean ± σ): 85.772 s ± 0.252 s [User: 91.505 s, System: 8.408 s] Range (min … max): 85.447 s … 86.244 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 103.833 s ± 0.288 s [User: 232.058 s, System: 8.569 s] Range (min … max): 103.286 s … 104.124 s 10 runsSummary
With ThinLTO cache ran
1.21 ± 0.00 times faster than Without ThinLTO cacheWhile it is unfortunate to take this performance improvement off the table, correctness is more important. If/when this is fixed in LLVM, it can potentially be brought back in a conditional manner. Alternatively, a developer can just disable LTO if doing incremental compiles quickly is important, as a full compile cycle can still take over a minute even with the cache and it is unlikely that LTO will result in functional differences for a kernel change.
Fixes dc5723b02e52 ("kbuild: add support for Clang LTO")Reported-by Yifan Hong <[email protected]>Reported-by Masami Hiramatsu <[email protected]>Signed-off-by Nathan Chancellor <[email protected]>Signed-off-by Masahiro Yamada <[email protected]>[nathan: Address conflict in Makefile]Signed-off-by Nathan Chancellor <[email protected]>Quoted 3 lines
--- Makefile | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
Quoted 3 lines
This applied to 5.15.y, not 6.1.y :(
Can you rebase and resend a fix for 6.1.y?
I don't understand how that is possible, this was generated directly on top of 6.1.93 (as evidenced by the base commit) and there were no changes to Makefile in 6.1.94. It still applies cleanly for me?
$ curl -LSs https://lore.kernel.org/all/[email protected]/raw | patch -p1 patching file Makefile
Cheers,
Nathan
Greg KH
maintainerRe: [PATCH 6.1.y] kbuild: Remove support for Clang's ThinLTO caching
Quoted 101 lines · Nathan Chancellor
Quoted 100 lines · Greg KH
Quoted 93 lines · Nathan Chancellor
commit
aba091547ef6159d52471f42a3ef531b7b660ed8upstream.There is an issue in clang's ThinLTO caching (enabled for the kernel via '--thinlto-cache-dir') with .incbin, which the kernel occasionally uses to include data within the kernel, such as the .config file for
/proc/config.gz. For example, when changing the .config and rebuilding vmlinux, the copy of .config in vmlinux does not match the copy of .config in the build folder:$ echo 'CONFIG_LTO_NONE=n CONFIG_LTO_CLANG_THIN=y CONFIG_IKCONFIG=y CONFIG_HEADERS_INSTALL=y' >kernel/configs/repro.config$ make -skj"$(nproc)" ARCH=x86_64 LLVM=1 clean defconfig repro.config vmlinux ...$ grep CONFIG_HEADERS_INSTALL .config CONFIG_HEADERS_INSTALL=y$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=y$ scripts/config -d HEADERS_INSTALL$ make -kj"$(nproc)" ARCH=x86_64 LLVM=1 vmlinux ... UPD kernel/config_data GZIP kernel/config_data.gz CC kernel/configs.o ... LD vmlinux ...$ grep CONFIG_HEADERS_INSTALL .config # CONFIG_HEADERS_INSTALL is not set$ scripts/extract-ikconfig vmlinux | grep CONFIG_HEADERS_INSTALL CONFIG_HEADERS_INSTALL=yWithout '--thinlto-cache-dir' or when using full LTO, this issue does not occur.
Benchmarking incremental builds on a few different machines with and without the cache shows a 20% increase in incremental build time without the cache when measured by touching
init/main.cand running 'make all'.ARCH=arm64 defconfig +
CONFIG_LTO_CLANG_THIN=yon an arm64 host:Benchmark 1: With ThinLTO cache
Time (mean ± σ): 56.347 s ± 0.163 s [User: 83.768 s, System: 24.661 s] Range (min … max): 56.109 s … 56.594 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 67.740 s ± 0.479 s [User: 718.458 s, System: 31.797 s] Range (min … max): 67.059 s … 68.556 s 10 runsSummary
With ThinLTO cache ran
1.20 ± 0.01 times faster than Without ThinLTO cacheARCH=x86_64 defconfig +
CONFIG_LTO_CLANG_THIN=yon anx86_64host:Benchmark 1: With ThinLTO cache
Time (mean ± σ): 85.772 s ± 0.252 s [User: 91.505 s, System: 8.408 s] Range (min … max): 85.447 s … 86.244 s 10 runsBenchmark 2: Without ThinLTO cache
Time (mean ± σ): 103.833 s ± 0.288 s [User: 232.058 s, System: 8.569 s] Range (min … max): 103.286 s … 104.124 s 10 runsSummary
With ThinLTO cache ran
1.21 ± 0.00 times faster than Without ThinLTO cacheWhile it is unfortunate to take this performance improvement off the table, correctness is more important. If/when this is fixed in LLVM, it can potentially be brought back in a conditional manner. Alternatively, a developer can just disable LTO if doing incremental compiles quickly is important, as a full compile cycle can still take over a minute even with the cache and it is unlikely that LTO will result in functional differences for a kernel change.
Fixes dc5723b02e52 ("kbuild: add support for Clang LTO")Reported-by Yifan Hong <[email protected]>Reported-by Masami Hiramatsu <[email protected]>Signed-off-by Nathan Chancellor <[email protected]>Signed-off-by Masahiro Yamada <[email protected]>[nathan: Address conflict in Makefile]Signed-off-by Nathan Chancellor <[email protected]>Quoted 3 lines
--- Makefile | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)This applied to 5.15.y, not 6.1.y :(
Quoted 9 lines
Quoted 1 line
Can you rebase and resend a fix for 6.1.y?
I don't understand how that is possible, this was generated directly on top of 6.1.93 (as evidenced by the base commit) and there were no changes to
Makefilein 6.1.94. It still applies cleanly for me?$ curl -LSs https://lore.kernel.org/all/[email protected]/raw | patch -p1 patching file Makefile
Very odd, I just tried it again and it worked, must have been a problem on my side, sorry for the noise.
greg k-h