* [PATCH 1/2] AutoFDO: Remove the architecture specific config
@ 2024-12-11 22:29 Rong Xu
2024-12-11 22:29 ` [PATCH 2/2] Propeller: " Rong Xu
` (2 more replies)
0 siblings, 3 replies; 10+ messages in thread
From: Rong Xu @ 2024-12-11 22:29 UTC (permalink / raw)
To: Masahiro Yamada, Yabin Cui, Will Deacon, Han Shen,
Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen,
H. Peter Anvin, Kees Cook, Andrew Morton, Sami Tolvanen,
Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland,
Paul E. McKenney, Rong Xu, Arnd Bergmann
Cc: x86, linux-kernel
The CONFIG_AUTOFDO_CLANG option currently depends on
ARCH_SUPPORTS_AUTOFDO_CLANG, but this dependency seems unnecessary.
Remove ARCH_SUPPORTS_AUTOFDO_CLANG and allow users to control AutoFDO
builds solely through CONFIG_AUTOFDO_CLANG. This simplifies the kconfig
and avoids potential confusion.
Signed-off-by: Rong Xu <xur@google.com>
Suggested-by: Will Deacon <will@kernel.org>
---
arch/Kconfig | 4 ----
arch/x86/Kconfig | 1 -
2 files changed, 5 deletions(-)
diff --git a/arch/Kconfig b/arch/Kconfig
index 6682b2a53e342..0b36d74d47031 100644
--- a/arch/Kconfig
+++ b/arch/Kconfig
@@ -812,12 +812,8 @@ config LTO_CLANG_THIN
If unsure, say Y.
endchoice
-config ARCH_SUPPORTS_AUTOFDO_CLANG
- bool
-
config AUTOFDO_CLANG
bool "Enable Clang's AutoFDO build (EXPERIMENTAL)"
- depends on ARCH_SUPPORTS_AUTOFDO_CLANG
depends on CC_IS_CLANG && CLANG_VERSION >= 170000
help
This option enables Clang’s AutoFDO build. When
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 9d7bd0ae48c42..9363d9cc9a00a 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -128,7 +128,6 @@ config X86
select ARCH_SUPPORTS_LTO_CLANG
select ARCH_SUPPORTS_LTO_CLANG_THIN
select ARCH_SUPPORTS_RT
- select ARCH_SUPPORTS_AUTOFDO_CLANG
select ARCH_SUPPORTS_PROPELLER_CLANG if X86_64
select ARCH_USE_BUILTIN_BSWAP
select ARCH_USE_CMPXCHG_LOCKREF if X86_CMPXCHG64
base-commit: f92f4749861b06fed908d336b4dee1326003291b
--
2.47.0.338.g60cca15819-goog
^ permalink raw reply related [flat|nested] 10+ messages in thread* [PATCH 2/2] Propeller: Remove the architecture specific config 2024-12-11 22:29 [PATCH 1/2] AutoFDO: Remove the architecture specific config Rong Xu @ 2024-12-11 22:29 ` Rong Xu 2024-12-11 23:05 ` Nathan Chancellor 2025-02-04 20:05 ` Kees Cook 2024-12-19 15:52 ` [PATCH 1/2] AutoFDO: " Will Deacon 2025-02-04 20:04 ` Kees Cook 2 siblings, 2 replies; 10+ messages in thread From: Rong Xu @ 2024-12-11 22:29 UTC (permalink / raw) To: Masahiro Yamada, Yabin Cui, Will Deacon, Han Shen, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, H. Peter Anvin, Kees Cook, Andrew Morton, Sami Tolvanen, Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland, Paul E. McKenney, Rong Xu, Arnd Bergmann Cc: x86, linux-kernel The CONFIG_PROPELLER_CLANG option currently depends on ARCH_SUPPORTS_PROPELLER_CLANG, but this dependency seems unnecessary. Remove ARCH_SUPPORTS_PROPELLER_CLANG and allow users to control Propeller builds solely through CONFIG_PROPELLER_CLANG. This simplifies the kconfig and avoids potential confusion. Signed-off-by: Rong Xu <xur@google.com> Suggested-by: Will Deacon <will@kernel.org> --- arch/Kconfig | 4 ---- arch/x86/Kconfig | 1 - 2 files changed, 5 deletions(-) diff --git a/arch/Kconfig b/arch/Kconfig index 0b36d74d47031..83731b858ba65 100644 --- a/arch/Kconfig +++ b/arch/Kconfig @@ -828,12 +828,8 @@ config AUTOFDO_CLANG If unsure, say N. -config ARCH_SUPPORTS_PROPELLER_CLANG - bool - config PROPELLER_CLANG bool "Enable Clang's Propeller build" - depends on ARCH_SUPPORTS_PROPELLER_CLANG depends on CC_IS_CLANG && CLANG_VERSION >= 190000 help This option enables Clang’s Propeller build. When the Propeller diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig index 9363d9cc9a00a..6c633d93c6390 100644 --- a/arch/x86/Kconfig +++ b/arch/x86/Kconfig @@ -128,7 +128,6 @@ config X86 select ARCH_SUPPORTS_LTO_CLANG select ARCH_SUPPORTS_LTO_CLANG_THIN select ARCH_SUPPORTS_RT - select ARCH_SUPPORTS_PROPELLER_CLANG if X86_64 select ARCH_USE_BUILTIN_BSWAP select ARCH_USE_CMPXCHG_LOCKREF if X86_CMPXCHG64 select ARCH_USE_MEMTEST -- 2.47.0.338.g60cca15819-goog ^ permalink raw reply related [flat|nested] 10+ messages in thread
* Re: [PATCH 2/2] Propeller: Remove the architecture specific config 2024-12-11 22:29 ` [PATCH 2/2] Propeller: " Rong Xu @ 2024-12-11 23:05 ` Nathan Chancellor 2024-12-12 7:48 ` Rong Xu 2025-02-04 20:05 ` Kees Cook 1 sibling, 1 reply; 10+ messages in thread From: Nathan Chancellor @ 2024-12-11 23:05 UTC (permalink / raw) To: Rong Xu Cc: Masahiro Yamada, Yabin Cui, Will Deacon, Han Shen, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, H. Peter Anvin, Kees Cook, Andrew Morton, Sami Tolvanen, Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland, Paul E. McKenney, Arnd Bergmann, x86, linux-kernel Hi Rong, On Wed, Dec 11, 2024 at 02:29:43PM -0800, Rong Xu wrote: > The CONFIG_PROPELLER_CLANG option currently depends on > ARCH_SUPPORTS_PROPELLER_CLANG, but this dependency seems unnecessary. > > Remove ARCH_SUPPORTS_PROPELLER_CLANG and allow users to control > Propeller builds solely through CONFIG_PROPELLER_CLANG. This simplifies > the kconfig and avoids potential confusion. > > Signed-off-by: Rong Xu <xur@google.com> > Suggested-by: Will Deacon <will@kernel.org> In commit d5dc95836147 ("kbuild: Add Propeller configuration for kernel build"), you added .llvm_bb_addr_map to arch/x86/kernel/vmlinux.lds.S. Was this to address a orphan section warning from the linker? Is that same change needed in the linker scripts of the other architectures that clang supports building in the kernel? Cheers, Nathan > --- > arch/Kconfig | 4 ---- > arch/x86/Kconfig | 1 - > 2 files changed, 5 deletions(-) > > diff --git a/arch/Kconfig b/arch/Kconfig > index 0b36d74d47031..83731b858ba65 100644 > --- a/arch/Kconfig > +++ b/arch/Kconfig > @@ -828,12 +828,8 @@ config AUTOFDO_CLANG > > If unsure, say N. > > -config ARCH_SUPPORTS_PROPELLER_CLANG > - bool > - > config PROPELLER_CLANG > bool "Enable Clang's Propeller build" > - depends on ARCH_SUPPORTS_PROPELLER_CLANG > depends on CC_IS_CLANG && CLANG_VERSION >= 190000 > help > This option enables Clang’s Propeller build. When the Propeller > diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig > index 9363d9cc9a00a..6c633d93c6390 100644 > --- a/arch/x86/Kconfig > +++ b/arch/x86/Kconfig > @@ -128,7 +128,6 @@ config X86 > select ARCH_SUPPORTS_LTO_CLANG > select ARCH_SUPPORTS_LTO_CLANG_THIN > select ARCH_SUPPORTS_RT > - select ARCH_SUPPORTS_PROPELLER_CLANG if X86_64 > select ARCH_USE_BUILTIN_BSWAP > select ARCH_USE_CMPXCHG_LOCKREF if X86_CMPXCHG64 > select ARCH_USE_MEMTEST > -- > 2.47.0.338.g60cca15819-goog > ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 2/2] Propeller: Remove the architecture specific config 2024-12-11 23:05 ` Nathan Chancellor @ 2024-12-12 7:48 ` Rong Xu 2025-01-31 22:36 ` Yabin Cui 0 siblings, 1 reply; 10+ messages in thread From: Rong Xu @ 2024-12-12 7:48 UTC (permalink / raw) To: Nathan Chancellor Cc: Masahiro Yamada, Yabin Cui, Will Deacon, Han Shen, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, H. Peter Anvin, Kees Cook, Andrew Morton, Sami Tolvanen, Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland, Paul E. McKenney, Arnd Bergmann, x86, linux-kernel (o On Wed, Dec 11, 2024 at 3:06 PM Nathan Chancellor <nathan@kernel.org> wrote: > > Hi Rong, > > On Wed, Dec 11, 2024 at 02:29:43PM -0800, Rong Xu wrote: > > The CONFIG_PROPELLER_CLANG option currently depends on > > ARCH_SUPPORTS_PROPELLER_CLANG, but this dependency seems unnecessary. > > > > Remove ARCH_SUPPORTS_PROPELLER_CLANG and allow users to control > > Propeller builds solely through CONFIG_PROPELLER_CLANG. This simplifies > > the kconfig and avoids potential confusion. > > > > Signed-off-by: Rong Xu <xur@google.com> > > Suggested-by: Will Deacon <will@kernel.org> > > In commit d5dc95836147 ("kbuild: Add Propeller configuration for kernel > build"), you added .llvm_bb_addr_map to arch/x86/kernel/vmlinux.lds.S. > Was this to address a orphan section warning from the linker? Is that > same change needed in the linker scripts of the other architectures that > clang supports building in the kernel? The .llvm_bb_addr_map section is created when compiling with the -fbasic-block-address-map flag (or the older -fbasic-block-sections=labels option), with LLVM. We need the change in the linker script to group the sections together and emit in the final vmlinux. This applies to all architectures, as LLVM can generate these sections for any architecture when the corresponding option is used. The statement that AutoFDO and Propeller are only supported on certain platforms really refers to the availability of *native* profile generation: the native profile generation is only available on these platforms. Hope this helps, -Rong > Cheers, > Nathan > > > --- > > arch/Kconfig | 4 ---- > > arch/x86/Kconfig | 1 - > > 2 files changed, 5 deletions(-) > > > > diff --git a/arch/Kconfig b/arch/Kconfig > > index 0b36d74d47031..83731b858ba65 100644 > > --- a/arch/Kconfig > > +++ b/arch/Kconfig > > @@ -828,12 +828,8 @@ config AUTOFDO_CLANG > > > > If unsure, say N. > > > > -config ARCH_SUPPORTS_PROPELLER_CLANG > > - bool > > - > > config PROPELLER_CLANG > > bool "Enable Clang's Propeller build" > > - depends on ARCH_SUPPORTS_PROPELLER_CLANG > > depends on CC_IS_CLANG && CLANG_VERSION >= 190000 > > help > > This option enables Clang’s Propeller build. When the Propeller > > diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig > > index 9363d9cc9a00a..6c633d93c6390 100644 > > --- a/arch/x86/Kconfig > > +++ b/arch/x86/Kconfig > > @@ -128,7 +128,6 @@ config X86 > > select ARCH_SUPPORTS_LTO_CLANG > > select ARCH_SUPPORTS_LTO_CLANG_THIN > > select ARCH_SUPPORTS_RT > > - select ARCH_SUPPORTS_PROPELLER_CLANG if X86_64 > > select ARCH_USE_BUILTIN_BSWAP > > select ARCH_USE_CMPXCHG_LOCKREF if X86_CMPXCHG64 > > select ARCH_USE_MEMTEST > > -- > > 2.47.0.338.g60cca15819-goog > > ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 2/2] Propeller: Remove the architecture specific config 2024-12-12 7:48 ` Rong Xu @ 2025-01-31 22:36 ` Yabin Cui 0 siblings, 0 replies; 10+ messages in thread From: Yabin Cui @ 2025-01-31 22:36 UTC (permalink / raw) To: Rong Xu Cc: Nathan Chancellor, Masahiro Yamada, Will Deacon, Han Shen, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, H. Peter Anvin, Kees Cook, Andrew Morton, Sami Tolvanen, Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland, Paul E. McKenney, Arnd Bergmann, x86, linux-kernel Tested-by: Yabin Cui <yabinc@google.com> On Wed, Dec 11, 2024 at 11:48 PM Rong Xu <xur@google.com> wrote: > > (o > > On Wed, Dec 11, 2024 at 3:06 PM Nathan Chancellor <nathan@kernel.org> wrote: > > > > Hi Rong, > > > > On Wed, Dec 11, 2024 at 02:29:43PM -0800, Rong Xu wrote: > > > The CONFIG_PROPELLER_CLANG option currently depends on > > > ARCH_SUPPORTS_PROPELLER_CLANG, but this dependency seems unnecessary. > > > > > > Remove ARCH_SUPPORTS_PROPELLER_CLANG and allow users to control > > > Propeller builds solely through CONFIG_PROPELLER_CLANG. This simplifies > > > the kconfig and avoids potential confusion. > > > > > > Signed-off-by: Rong Xu <xur@google.com> > > > Suggested-by: Will Deacon <will@kernel.org> > > > > In commit d5dc95836147 ("kbuild: Add Propeller configuration for kernel > > build"), you added .llvm_bb_addr_map to arch/x86/kernel/vmlinux.lds.S. > > Was this to address a orphan section warning from the linker? Is that > > same change needed in the linker scripts of the other architectures that > > clang supports building in the kernel? > > The .llvm_bb_addr_map section is created when compiling with the > -fbasic-block-address-map flag (or the older -fbasic-block-sections=labels > option), with LLVM. We need the change in the linker script to group the > sections together and emit in the final vmlinux. > > This applies to all architectures, as LLVM can generate these sections for > any architecture when the corresponding option is used. > > The statement that AutoFDO and Propeller are only supported on certain > platforms really refers to the availability of *native* profile generation: > the native profile generation is only available on these platforms. > > Hope this helps, > > -Rong > > > Cheers, > > Nathan > > > > > --- > > > arch/Kconfig | 4 ---- > > > arch/x86/Kconfig | 1 - > > > 2 files changed, 5 deletions(-) > > > > > > diff --git a/arch/Kconfig b/arch/Kconfig > > > index 0b36d74d47031..83731b858ba65 100644 > > > --- a/arch/Kconfig > > > +++ b/arch/Kconfig > > > @@ -828,12 +828,8 @@ config AUTOFDO_CLANG > > > > > > If unsure, say N. > > > > > > -config ARCH_SUPPORTS_PROPELLER_CLANG > > > - bool > > > - > > > config PROPELLER_CLANG > > > bool "Enable Clang's Propeller build" > > > - depends on ARCH_SUPPORTS_PROPELLER_CLANG > > > depends on CC_IS_CLANG && CLANG_VERSION >= 190000 > > > help > > > This option enables Clang’s Propeller build. When the Propeller > > > diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig > > > index 9363d9cc9a00a..6c633d93c6390 100644 > > > --- a/arch/x86/Kconfig > > > +++ b/arch/x86/Kconfig > > > @@ -128,7 +128,6 @@ config X86 > > > select ARCH_SUPPORTS_LTO_CLANG > > > select ARCH_SUPPORTS_LTO_CLANG_THIN > > > select ARCH_SUPPORTS_RT > > > - select ARCH_SUPPORTS_PROPELLER_CLANG if X86_64 > > > select ARCH_USE_BUILTIN_BSWAP > > > select ARCH_USE_CMPXCHG_LOCKREF if X86_CMPXCHG64 > > > select ARCH_USE_MEMTEST > > > -- > > > 2.47.0.338.g60cca15819-goog > > > ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 2/2] Propeller: Remove the architecture specific config 2024-12-11 22:29 ` [PATCH 2/2] Propeller: " Rong Xu 2024-12-11 23:05 ` Nathan Chancellor @ 2025-02-04 20:05 ` Kees Cook 1 sibling, 0 replies; 10+ messages in thread From: Kees Cook @ 2025-02-04 20:05 UTC (permalink / raw) To: Rong Xu Cc: Masahiro Yamada, Yabin Cui, Will Deacon, Han Shen, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, H. Peter Anvin, Andrew Morton, Sami Tolvanen, Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland, Paul E. McKenney, Arnd Bergmann, x86, linux-kernel On Wed, Dec 11, 2024 at 02:29:43PM -0800, Rong Xu wrote: > The CONFIG_PROPELLER_CLANG option currently depends on > ARCH_SUPPORTS_PROPELLER_CLANG, but this dependency seems unnecessary. > > Remove ARCH_SUPPORTS_PROPELLER_CLANG and allow users to control > Propeller builds solely through CONFIG_PROPELLER_CLANG. This simplifies > the kconfig and avoids potential confusion. > > Signed-off-by: Rong Xu <xur@google.com> With Nathan's question answered, this looks good to me: Reviewed-by: Kees Cook <kees@kernel.org> -- Kees Cook ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] AutoFDO: Remove the architecture specific config 2024-12-11 22:29 [PATCH 1/2] AutoFDO: Remove the architecture specific config Rong Xu 2024-12-11 22:29 ` [PATCH 2/2] Propeller: " Rong Xu @ 2024-12-19 15:52 ` Will Deacon 2024-12-19 17:19 ` Rong Xu 2025-02-04 20:04 ` Kees Cook 2 siblings, 1 reply; 10+ messages in thread From: Will Deacon @ 2024-12-19 15:52 UTC (permalink / raw) To: Rong Xu Cc: Masahiro Yamada, Yabin Cui, Han Shen, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, H. Peter Anvin, Kees Cook, Andrew Morton, Sami Tolvanen, Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland, Paul E. McKenney, Arnd Bergmann, x86, linux-kernel On Wed, Dec 11, 2024 at 02:29:42PM -0800, Rong Xu wrote: > The CONFIG_AUTOFDO_CLANG option currently depends on > ARCH_SUPPORTS_AUTOFDO_CLANG, but this dependency seems unnecessary. > > Remove ARCH_SUPPORTS_AUTOFDO_CLANG and allow users to control AutoFDO > builds solely through CONFIG_AUTOFDO_CLANG. This simplifies the kconfig > and avoids potential confusion. > > Signed-off-by: Rong Xu <xur@google.com> > Suggested-by: Will Deacon <will@kernel.org> > --- > arch/Kconfig | 4 ---- > arch/x86/Kconfig | 1 - > 2 files changed, 5 deletions(-) Acked-by: Will Deacon <will@kernel.org> However, you've dropped the previous hunk you had to Documentation/dev-tools/autofdo.rst which had some text about using ETM to collect fdo data on arm64. It seems like that might still be useful to have? Will ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] AutoFDO: Remove the architecture specific config 2024-12-19 15:52 ` [PATCH 1/2] AutoFDO: " Will Deacon @ 2024-12-19 17:19 ` Rong Xu 2025-01-31 22:35 ` Yabin Cui 0 siblings, 1 reply; 10+ messages in thread From: Rong Xu @ 2024-12-19 17:19 UTC (permalink / raw) To: Will Deacon Cc: Masahiro Yamada, Yabin Cui, Han Shen, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, H. Peter Anvin, Kees Cook, Andrew Morton, Sami Tolvanen, Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland, Paul E. McKenney, Arnd Bergmann, x86, linux-kernel Thanks for the Acked-by. Regarding the ETM documentation change, I haven't forgotten it. I think it's best for Yabin to submit another patch (or revision of his previous patch), as he has done all the ETM side work. Best regards, -Rong On Thu, Dec 19, 2024 at 7:52 AM Will Deacon <will@kernel.org> wrote: > > On Wed, Dec 11, 2024 at 02:29:42PM -0800, Rong Xu wrote: > > The CONFIG_AUTOFDO_CLANG option currently depends on > > ARCH_SUPPORTS_AUTOFDO_CLANG, but this dependency seems unnecessary. > > > > Remove ARCH_SUPPORTS_AUTOFDO_CLANG and allow users to control AutoFDO > > builds solely through CONFIG_AUTOFDO_CLANG. This simplifies the kconfig > > and avoids potential confusion. > > > > Signed-off-by: Rong Xu <xur@google.com> > > Suggested-by: Will Deacon <will@kernel.org> > > --- > > arch/Kconfig | 4 ---- > > arch/x86/Kconfig | 1 - > > 2 files changed, 5 deletions(-) > > Acked-by: Will Deacon <will@kernel.org> > > However, you've dropped the previous hunk you had to > Documentation/dev-tools/autofdo.rst which had some text about using ETM > to collect fdo data on arm64. It seems like that might still be useful > to have? > > Will ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] AutoFDO: Remove the architecture specific config 2024-12-19 17:19 ` Rong Xu @ 2025-01-31 22:35 ` Yabin Cui 0 siblings, 0 replies; 10+ messages in thread From: Yabin Cui @ 2025-01-31 22:35 UTC (permalink / raw) To: Rong Xu Cc: Will Deacon, Masahiro Yamada, Han Shen, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, H. Peter Anvin, Kees Cook, Andrew Morton, Sami Tolvanen, Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland, Paul E. McKenney, Arnd Bergmann, x86, linux-kernel Hi Rong, Thanks for the patch! Sorry for the delay in testing it. I will upload another patch to document steps for using ETM. Tested-by: Yabin Cui <yabinc@google.com> Thanks, Yabin On Thu, Dec 19, 2024 at 9:20 AM Rong Xu <xur@google.com> wrote: > > Thanks for the Acked-by. > > Regarding the ETM documentation change, I haven't forgotten it. I > think it's best for Yabin to submit another patch (or revision of his > previous patch), as he has done all the ETM side work. > > Best regards, > > -Rong > > On Thu, Dec 19, 2024 at 7:52 AM Will Deacon <will@kernel.org> wrote: > > > > On Wed, Dec 11, 2024 at 02:29:42PM -0800, Rong Xu wrote: > > > The CONFIG_AUTOFDO_CLANG option currently depends on > > > ARCH_SUPPORTS_AUTOFDO_CLANG, but this dependency seems unnecessary. > > > > > > Remove ARCH_SUPPORTS_AUTOFDO_CLANG and allow users to control AutoFDO > > > builds solely through CONFIG_AUTOFDO_CLANG. This simplifies the kconfig > > > and avoids potential confusion. > > > > > > Signed-off-by: Rong Xu <xur@google.com> > > > Suggested-by: Will Deacon <will@kernel.org> > > > --- > > > arch/Kconfig | 4 ---- > > > arch/x86/Kconfig | 1 - > > > 2 files changed, 5 deletions(-) > > > > Acked-by: Will Deacon <will@kernel.org> > > > > However, you've dropped the previous hunk you had to > > Documentation/dev-tools/autofdo.rst which had some text about using ETM > > to collect fdo data on arm64. It seems like that might still be useful > > to have? > > > > Will ^ permalink raw reply [flat|nested] 10+ messages in thread
* Re: [PATCH 1/2] AutoFDO: Remove the architecture specific config 2024-12-11 22:29 [PATCH 1/2] AutoFDO: Remove the architecture specific config Rong Xu 2024-12-11 22:29 ` [PATCH 2/2] Propeller: " Rong Xu 2024-12-19 15:52 ` [PATCH 1/2] AutoFDO: " Will Deacon @ 2025-02-04 20:04 ` Kees Cook 2 siblings, 0 replies; 10+ messages in thread From: Kees Cook @ 2025-02-04 20:04 UTC (permalink / raw) To: Rong Xu Cc: Masahiro Yamada, Yabin Cui, Will Deacon, Han Shen, Thomas Gleixner, Ingo Molnar, Borislav Petkov, Dave Hansen, H. Peter Anvin, Andrew Morton, Sami Tolvanen, Alice Ryhl, Mike Rapoport (Microsoft), Samuel Holland, Paul E. McKenney, Arnd Bergmann, x86, linux-kernel On Wed, Dec 11, 2024 at 02:29:42PM -0800, Rong Xu wrote: > The CONFIG_AUTOFDO_CLANG option currently depends on > ARCH_SUPPORTS_AUTOFDO_CLANG, but this dependency seems unnecessary. > > Remove ARCH_SUPPORTS_AUTOFDO_CLANG and allow users to control AutoFDO > builds solely through CONFIG_AUTOFDO_CLANG. This simplifies the kconfig > and avoids potential confusion. > > Signed-off-by: Rong Xu <xur@google.com> Looks right to me! Reviewed-by: Kees Cook <kees@kernel.org> -- Kees Cook ^ permalink raw reply [flat|nested] 10+ messages in thread
end of thread, other threads:[~2025-02-04 20:05 UTC | newest] Thread overview: 10+ messages (download: mbox.gz follow: Atom feed -- links below jump to the message on this page -- 2024-12-11 22:29 [PATCH 1/2] AutoFDO: Remove the architecture specific config Rong Xu 2024-12-11 22:29 ` [PATCH 2/2] Propeller: " Rong Xu 2024-12-11 23:05 ` Nathan Chancellor 2024-12-12 7:48 ` Rong Xu 2025-01-31 22:36 ` Yabin Cui 2025-02-04 20:05 ` Kees Cook 2024-12-19 15:52 ` [PATCH 1/2] AutoFDO: " Will Deacon 2024-12-19 17:19 ` Rong Xu 2025-01-31 22:35 ` Yabin Cui 2025-02-04 20:04 ` Kees Cook
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox