Skip to contentSkip to conversation

Kconfig symbol clean-up on ./arch/x86/

PATCH Applied Lukas Bulwahn 5 participants 25 messages Aug 3, 2021 – Dec 22, 2021

Subsystems (4)
lore
Share & tools
lore
#0 lore raw

cover Kconfig symbol clean-up on ./arch/x86/

Dear x86 maintainers,

Randy Dunlap pointed out that ./scripts/checkkconfigsymbols.py warns on invalid references to Kconfig symbols (often, minor typos, name confusions or outdated references): see Link below.

This is a first patch series addressing the issues reported by ./scripts/checkkconfigsymbols.py in the ./arch/x86/ directories, quickly filtered down with:

./scripts/checkkconfigsymbols.py | grep "arch/x86" -B 1 -A 2

It addresses some issues I considered to be "true positives"; so, issues that should be addressed and cleaned up.

The patches on updating the comments might arguably be considered of low value by itself, but as the other patches show, checkkconfigsymbols does detect some relevant issues in the kernel tree otherwise being unnoticed. So, it might be worth to clean up the kernel tree to have checkkconfigsymbols produce a rather short list of issues and then continuously follow and check what checkkconfigsymbols reports.

Some work from checkkconfigsymbols on arch/x86 is deferred:

- OSNOISE_TRAECR typo in arch/x86/kernel/trace.c
This will added to a separate janitor cleanup series with other topics on this file.

- a reference to STRICT_IOMEM in arch/x86/mm/init.c unclear to me: which exact config this refers to

- Given some updates of the minimal version for binutils (now >= 2.23), the whole special detection of binutils support for CONFIG_X86_X32 and its speciality with CONFIG_X86_X32_ABI might be unneeded. I might look into that in a separate task.

The remaining issues that checkkconfigsymbols reports are (as far as I understand it now) "false positives" (the tool misinterprets the actual content) or "intentional", e.g., they refer to historic configs for the purpose of documentation (e.g., in changelogs) or are in various kernel configs that are not continuously updated to match the current set of kernel configs.

Please pick this series of minor clean-up patches on ./arch/x86/.

It applies cleanly on next-20210802.

Best regards,

Lukas

Lukas Bulwahn (9):
x86/entry: correct reference to intended CONFIG_64_BIT x86, olpc: correct condition to intended CONFIG_OLPC_XO15_SCI x86/Kconfig: correct reference to MWINCHIP3D
x86/Kconfig: remove reference to obsolete MFD_INTEL_MSIC config x86/Kconfig: remove reference to obsolete APB_TIMER config x86/irq: fix slightly wrong reference in comment x86/uaccess: adjust comment for endif of CONFIG_CC_HAS_ASM_GOTO_OUTPUT
x86: page_32.h: adjust comment for endif of CONFIG_X86_USE_3DNOW
x86: ia32.h: adjust comment for endif of CONFIG_IA32_EMULATION

 arch/x86/Kconfig                    | 4 +---
 arch/x86/include/asm/entry-common.h | 2 +-
 arch/x86/include/asm/ia32.h         | 2 +-
 arch/x86/include/asm/irq_stack.h    | 2 +-
 arch/x86/include/asm/page_32.h      | 2 +-
 arch/x86/include/asm/uaccess.h      | 2 +-
 arch/x86/platform/olpc/olpc.c       | 2 +-
 7 files changed, 7 insertions(+), 9 deletions(-)
Signature · 1 line
2.17.1
#1 lore raw

PATCH 1/9 x86/entry: correct reference to intended CONFIG_64_BIT

Commit 662a0221893a ("x86/entry: Fix AC assertion") adds a condition with IS_ENABLED(CONFIG_64_BIT), but the intended config is called CONFIG_64BIT, as defined in ./arch/x86/Kconfig.

Fortunately, ./scripts/checkkconfigsymbols.py warns:

64_BIT
Referencing files: arch/x86/include/asm/entry-common.h

Correct the reference to the intended config symbol.

Fixes 662a0221893a ("x86/entry: Fix AC assertion")
---
 arch/x86/include/asm/entry-common.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/entry-common.h +1 -1
index 14ebd2196569..43184640b579 100644
@@ -25,7 +25,7 @@ static __always_inline void arch_check_user_regs(struct pt_regs *regs)
 		 * For !SMAP hardware we patch out CLAC on entry.
 		 */
 		if (boot_cpu_has(X86_FEATURE_SMAP) ||
-		    (IS_ENABLED(CONFIG_64_BIT) && boot_cpu_has(X86_FEATURE_XENPV)))
+		    (IS_ENABLED(CONFIG_64BIT) && boot_cpu_has(X86_FEATURE_XENPV)))
 			mask |= X86_EFLAGS_AC;
 
 		WARN_ON_ONCE(flags & mask);
Signature · 1 line
2.17.1
merge bot Applied
#2 lore raw

[tip: x86/urgent] x86/entry: Correct reference to intended CONFIG_64_BIT

The following commit has been merged into the x86/urgent branch of tip:

The following commit has been merged into the x86/urgent branch of tip:

Commit-ID:     b2c238397b8f737590a7f1b4e8f9dbbd5043e340
Gitweb:        https://git.kernel.org/tip/b2c238397b8f737590a7f1b4e8f9dbbd5043e340
Author:        Lukas Bulwahn <[email protected]>
AuthorDate:    Tue, 03 Aug 2021 13:35:23 +02:00
Committer:     Borislav Petkov <[email protected]>
CommitterDate: Tue, 05 Oct 2021 11:10:21 +02:00

x86/entry: Correct reference to intended CONFIG_64_BIT

Commit in Fixes adds a condition with IS_ENABLED(CONFIG_64_BIT), but the intended config item is called CONFIG_64BIT, as defined in arch/x86/Kconfig.

Fortunately, scripts/checkkconfigsymbols.py warns:

64_BIT
Referencing files: arch/x86/include/asm/entry-common.h

Correct the reference to the intended config symbol.

---
 arch/x86/include/asm/entry-common.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/entry-common.h +1 -1
index 14ebd21..4318464 100644
@@ -25,7 +25,7 @@ static __always_inline void arch_check_user_regs(struct pt_regs *regs)
 		 * For !SMAP hardware we patch out CLAC on entry.
 		 */
 		if (boot_cpu_has(X86_FEATURE_SMAP) ||
-		    (IS_ENABLED(CONFIG_64_BIT) && boot_cpu_has(X86_FEATURE_XENPV)))
+		    (IS_ENABLED(CONFIG_64BIT) && boot_cpu_has(X86_FEATURE_XENPV)))
 			mask |= X86_EFLAGS_AC;
 
 		WARN_ON_ONCE(flags & mask);
merge bot Applied
#3 lore raw

[tip: x86/urgent] x86/entry: Correct reference to intended CONFIG_64_BIT

In reply to Lukas Bulwahn
The following commit has been merged into the x86/urgent branch of tip:

The following commit has been merged into the x86/urgent branch of tip:

Commit-ID:     2c861f2b859385e9eaa6e464a8a7435b5a6bf564
Gitweb:        https://git.kernel.org/tip/2c861f2b859385e9eaa6e464a8a7435b5a6bf564
Author:        Lukas Bulwahn <[email protected]>
AuthorDate:    Tue, 03 Aug 2021 13:35:23 +02:00
Committer:     Borislav Petkov <[email protected]>
CommitterDate: Wed, 06 Oct 2021 18:46:02 +02:00

x86/entry: Correct reference to intended CONFIG_64_BIT

Commit in Fixes adds a condition with IS_ENABLED(CONFIG_64_BIT), but the intended config item is called CONFIG_64BIT, as defined in arch/x86/Kconfig.

Fortunately, scripts/checkkconfigsymbols.py warns:

64_BIT
Referencing files: arch/x86/include/asm/entry-common.h

Correct the reference to the intended config symbol.

---
 arch/x86/include/asm/entry-common.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/entry-common.h +1 -1
index 14ebd21..4318464 100644
@@ -25,7 +25,7 @@ static __always_inline void arch_check_user_regs(struct pt_regs *regs)
 		 * For !SMAP hardware we patch out CLAC on entry.
 		 */
 		if (boot_cpu_has(X86_FEATURE_SMAP) ||
-		    (IS_ENABLED(CONFIG_64_BIT) && boot_cpu_has(X86_FEATURE_XENPV)))
+		    (IS_ENABLED(CONFIG_64BIT) && boot_cpu_has(X86_FEATURE_XENPV)))
 			mask |= X86_EFLAGS_AC;
 
 		WARN_ON_ONCE(flags & mask);
#4 lore raw

PATCH 2/9 x86, olpc: correct condition to intended CONFIG_OLPC_XO15_SCI

In reply to Lukas Bulwahn

The refactoring in commit ec9964b48033 ("Platform: OLPC: Move EC-specific functionality out from x86") introduces an ifdef CONFIG_OLPC_XO1_5_SCI, however the config is actually called "CONFIG_OLPC_XO15_SCI".

Fortunately, ./scripts/checkkconfigsymbols.py warns:

OLPC_XO1_5_SCI
Referencing files: arch/x86/platform/olpc/olpc.c

Correct this ifdef condition to the intended config symbol.

Fixes ec9964b48033 ("Platform: OLPC: Move EC-specific functionality out from x86")
---
 arch/x86/platform/olpc/olpc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/platform/olpc/olpc.c +1 -1
index ee2beda590d0..1d4a00e767ec 100644
@@ -274,7 +274,7 @@ static struct olpc_ec_driver ec_xo1_driver = {
 
 static struct olpc_ec_driver ec_xo1_5_driver = {
 	.ec_cmd = olpc_xo1_ec_cmd,
-#ifdef CONFIG_OLPC_XO1_5_SCI
+#ifdef CONFIG_OLPC_XO15_SCI
 	/*
 	 * XO-1.5 EC wakeups are available when olpc-xo15-sci driver is
 	 * compiled in
Signature · 1 line
2.17.1
merge bot Applied
#5 lore raw

[tip: x86/urgent] x86/platform/olpc: Correct ifdef symbol to intended CONFIG_OLPC_XO15_SCI

The following commit has been merged into the x86/urgent branch of tip:

The following commit has been merged into the x86/urgent branch of tip:

Commit-ID:     4f90c68790aaa260fb5852406c28e25018872e45
Gitweb:        https://git.kernel.org/tip/4f90c68790aaa260fb5852406c28e25018872e45
Author:        Lukas Bulwahn <[email protected]>
AuthorDate:    Tue, 03 Aug 2021 13:35:24 +02:00
Committer:     Borislav Petkov <[email protected]>
CommitterDate: Wed, 06 Oct 2021 00:46:10 +02:00

x86/platform/olpc: Correct ifdef symbol to intended CONFIG_OLPC_XO15_SCI

The refactoring in the commit in Fixes introduced an ifdef CONFIG_OLPC_XO1_5_SCI, however the config symbol is actually called "CONFIG_OLPC_XO15_SCI".

Fortunately, ./scripts/checkkconfigsymbols.py warns:

OLPC_XO1_5_SCI
Referencing files: arch/x86/platform/olpc/olpc.c

Correct this ifdef condition to the intended config symbol.

Fixes ec9964b48033 ("Platform: OLPC: Move EC-specific functionality out from x86")
---
 arch/x86/platform/olpc/olpc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/platform/olpc/olpc.c +1 -1
index ee2beda..1d4a00e 100644
@@ -274,7 +274,7 @@ static struct olpc_ec_driver ec_xo1_driver = {
 
 static struct olpc_ec_driver ec_xo1_5_driver = {
 	.ec_cmd = olpc_xo1_ec_cmd,
-#ifdef CONFIG_OLPC_XO1_5_SCI
+#ifdef CONFIG_OLPC_XO15_SCI
 	/*
 	 * XO-1.5 EC wakeups are available when olpc-xo15-sci driver is
 	 * compiled in
merge bot Applied
#6 lore raw

[tip: x86/urgent] x86/platform/olpc: Correct ifdef symbol to intended CONFIG_OLPC_XO15_SCI

In reply to Lukas Bulwahn
The following commit has been merged into the x86/urgent branch of tip:

The following commit has been merged into the x86/urgent branch of tip:

Commit-ID:     4758fd801f919b8b9acad78d2e49a195ec2be46b
Gitweb:        https://git.kernel.org/tip/4758fd801f919b8b9acad78d2e49a195ec2be46b
Author:        Lukas Bulwahn <[email protected]>
AuthorDate:    Tue, 03 Aug 2021 13:35:24 +02:00
Committer:     Borislav Petkov <[email protected]>
CommitterDate: Wed, 06 Oct 2021 18:46:06 +02:00

x86/platform/olpc: Correct ifdef symbol to intended CONFIG_OLPC_XO15_SCI

The refactoring in the commit in Fixes introduced an ifdef CONFIG_OLPC_XO1_5_SCI, however the config symbol is actually called "CONFIG_OLPC_XO15_SCI".

Fortunately, ./scripts/checkkconfigsymbols.py warns:

OLPC_XO1_5_SCI
Referencing files: arch/x86/platform/olpc/olpc.c

Correct this ifdef condition to the intended config symbol.

Fixes ec9964b48033 ("Platform: OLPC: Move EC-specific functionality out from x86")
---
 arch/x86/platform/olpc/olpc.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/platform/olpc/olpc.c +1 -1
index ee2beda..1d4a00e 100644
@@ -274,7 +274,7 @@ static struct olpc_ec_driver ec_xo1_driver = {
 
 static struct olpc_ec_driver ec_xo1_5_driver = {
 	.ec_cmd = olpc_xo1_ec_cmd,
-#ifdef CONFIG_OLPC_XO1_5_SCI
+#ifdef CONFIG_OLPC_XO15_SCI
 	/*
 	 * XO-1.5 EC wakeups are available when olpc-xo15-sci driver is
 	 * compiled in
#7 lore raw

PATCH 3/9 x86/Kconfig: correct reference to MWINCHIP3D

In reply to Lukas Bulwahn

Commit 69b8d3fcabdc ("x86/Kconfig: Exclude i586-class CPUs lacking PAE support from the HIGHMEM64G Kconfig group") intends to exclude the Winchip architectures and refers to CONFIG_WINCHIP3D, but the config is called CONFIG_MWINCHIP3D.

Hence, ./scripts/checkkconfigsymbols.py warns:

WINCHIP3D
Referencing files: arch/x86/Kconfig

Correct the reference to the intended config symbol.

---
 arch/x86/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/Kconfig +1 -1
index 45962aaf2b2c..79e7457db5c9 100644
@@ -1401,7 +1401,7 @@ config HIGHMEM4G
 
 config HIGHMEM64G
 	bool "64GB"
-	depends on !M486SX && !M486 && !M586 && !M586TSC && !M586MMX && !MGEODE_LX && !MGEODEGX1 && !MCYRIXIII && !MELAN && !MWINCHIPC6 && !WINCHIP3D && !MK6
+	depends on !M486SX && !M486 && !M586 && !M586TSC && !M586MMX && !MGEODE_LX && !MGEODEGX1 && !MCYRIXIII && !MELAN && !MWINCHIPC6 && !MWINCHIP3D && !MK6
 	select X86_PAE
 	help
 	  Select this if you have a 32-bit processor and more than 4
Signature · 1 line
2.17.1
merge bot Applied
#8 lore raw

[tip: x86/urgent] x86/Kconfig: Correct reference to MWINCHIP3D

The following commit has been merged into the x86/urgent branch of tip:

The following commit has been merged into the x86/urgent branch of tip:

Commit-ID:     9608312f03f412eafa1630b134db710b4473d62a
Gitweb:        https://git.kernel.org/tip/9608312f03f412eafa1630b134db710b4473d62a
Author:        Lukas Bulwahn <[email protected]>
AuthorDate:    Tue, 03 Aug 2021 13:35:25 +02:00
Committer:     Borislav Petkov <[email protected]>
CommitterDate: Wed, 06 Oct 2021 00:46:44 +02:00

x86/Kconfig: Correct reference to MWINCHIP3D

Commit in Fixes intended to exclude the Winchip series and referred to CONFIG_WINCHIP3D, but the config symbol is called CONFIG_MWINCHIP3D.

Hence, scripts/checkkconfigsymbols.py warns:

WINCHIP3D
Referencing files: arch/x86/Kconfig

Correct the reference to the intended config symbol.

Fixes 69b8d3fcabdc ("x86/Kconfig: Exclude i586-class CPUs lacking PAE support from the HIGHMEM64G Kconfig group")
---
 arch/x86/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/Kconfig +1 -1
index ab83c22..8055da4 100644
@@ -1405,7 +1405,7 @@ config HIGHMEM4G
 
 config HIGHMEM64G
 	bool "64GB"
-	depends on !M486SX && !M486 && !M586 && !M586TSC && !M586MMX && !MGEODE_LX && !MGEODEGX1 && !MCYRIXIII && !MELAN && !MWINCHIPC6 && !WINCHIP3D && !MK6
+	depends on !M486SX && !M486 && !M586 && !M586TSC && !M586MMX && !MGEODE_LX && !MGEODEGX1 && !MCYRIXIII && !MELAN && !MWINCHIPC6 && !MWINCHIP3D && !MK6
 	select X86_PAE
 	help
 	  Select this if you have a 32-bit processor and more than 4
merge bot Applied
#9 lore raw

[tip: x86/urgent] x86/Kconfig: Correct reference to MWINCHIP3D

In reply to Lukas Bulwahn
The following commit has been merged into the x86/urgent branch of tip:

The following commit has been merged into the x86/urgent branch of tip:

Commit-ID:     225bac2dc5d192e55f2c50123ee539b1edf8a411
Gitweb:        https://git.kernel.org/tip/225bac2dc5d192e55f2c50123ee539b1edf8a411
Author:        Lukas Bulwahn <[email protected]>
AuthorDate:    Tue, 03 Aug 2021 13:35:25 +02:00
Committer:     Borislav Petkov <[email protected]>
CommitterDate: Wed, 06 Oct 2021 18:46:06 +02:00

x86/Kconfig: Correct reference to MWINCHIP3D

Commit in Fixes intended to exclude the Winchip series and referred to CONFIG_WINCHIP3D, but the config symbol is called CONFIG_MWINCHIP3D.

Hence, scripts/checkkconfigsymbols.py warns:

WINCHIP3D
Referencing files: arch/x86/Kconfig

Correct the reference to the intended config symbol.

Fixes 69b8d3fcabdc ("x86/Kconfig: Exclude i586-class CPUs lacking PAE support from the HIGHMEM64G Kconfig group")
---
 arch/x86/Kconfig | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/Kconfig +1 -1
index ab83c22..8055da4 100644
@@ -1405,7 +1405,7 @@ config HIGHMEM4G
 
 config HIGHMEM64G
 	bool "64GB"
-	depends on !M486SX && !M486 && !M586 && !M586TSC && !M586MMX && !MGEODE_LX && !MGEODEGX1 && !MCYRIXIII && !MELAN && !MWINCHIPC6 && !WINCHIP3D && !MK6
+	depends on !M486SX && !M486 && !M586 && !M586TSC && !M586MMX && !MGEODE_LX && !MGEODEGX1 && !MCYRIXIII && !MELAN && !MWINCHIPC6 && !MWINCHIP3D && !MK6
 	select X86_PAE
 	help
 	  Select this if you have a 32-bit processor and more than 4
#10 lore raw

PATCH 4/9 x86/Kconfig: remove reference to obsolete MFD_INTEL_MSIC config

In reply to Lukas Bulwahn

Commit ef3c67b6454b ("mfd: intel_msic: Remove driver for deprecated platform") removes the definition of the config MFD_INTEL_MSIC in ./drivers/mfd/Kconfig, but misses to remove a reference to it in config X86_INTEL_MID in ./arch/x86/Kconfig.

Fortunately, ./scripts/checkkconfigsymbols.py warns:

MFD_INTEL_MSIC
Referencing files: arch/x86/Kconfig

Remove this reference to the obsolete config.

Fixes ef3c67b6454b ("mfd: intel_msic: Remove driver for deprecated platform")
---
 arch/x86/Kconfig | 1 -
 1 file changed, 1 deletion(-)
arch/x86/Kconfig +0 -1
index 79e7457db5c9..b553e8b4df4e 100644
@@ -608,7 +608,6 @@ config X86_INTEL_MID
 	select DW_APB_TIMER
 	select APB_TIMER
 	select INTEL_SCU_PCI
-	select MFD_INTEL_MSIC
 	help
 	  Select to build a kernel capable of supporting Intel MID (Mobile
 	  Internet Device) platform systems which do not have the PCI legacy
Signature · 1 line
2.17.1
merge bot Applied
#11 lore raw

[tip: x86/cleanups] x86/Kconfig: Remove references to obsolete Kconfig symbols

The following commit has been merged into the x86/cleanups branch of tip:

The following commit has been merged into the x86/cleanups branch of tip:

Commit-ID:     3fd3590b53d1462ab534523e8d9ebdebd9b55f79
Gitweb:        https://git.kernel.org/tip/3fd3590b53d1462ab534523e8d9ebdebd9b55f79
Author:        Lukas Bulwahn <[email protected]>
AuthorDate:    Tue, 03 Aug 2021 13:35:26 +02:00
Committer:     Borislav Petkov <[email protected]>
CommitterDate: Tue, 05 Oct 2021 21:36:58 +02:00

x86/Kconfig: Remove references to obsolete Kconfig symbols

Remove two symbols referenced in Kconfig which have been removed previously by:

ef3c67b6454b ("mfd: intel_msic: Remove driver for deprecated platform") 1b79fc4f2bfd ("x86/apb_timer: Remove driver for deprecated platform")

Detected by scripts/checkkconfigsymbols.py.

[ bp: Merge into a single patch. ]

---
 arch/x86/Kconfig | 2 --
 1 file changed, 2 deletions(-)
arch/x86/Kconfig +0 -2
index 4e001bb..b79e88e 100644
@@ -605,9 +605,7 @@ config X86_INTEL_MID
 	depends on X86_IO_APIC
 	select I2C
 	select DW_APB_TIMER
-	select APB_TIMER
 	select INTEL_SCU_PCI
-	select MFD_INTEL_MSIC
 	help
 	  Select to build a kernel capable of supporting Intel MID (Mobile
 	  Internet Device) platform systems which do not have the PCI legacy
#12 lore raw

PATCH 5/9 x86/Kconfig: remove reference to obsolete APB_TIMER config

In reply to Lukas Bulwahn

Commit 1b79fc4f2bfd ("x86/apb_timer: Remove driver for deprecated platform") removes the definition of the config APB_TIMER in ./arch/x86/Kconfig, but misses to remove a reference to it in config X86_INTEL_MID.

Fortunately, ./scripts/checkkconfigsymbols.py warns:

APB_TIMER
Referencing files: arch/x86/Kconfig

Remove this reference to the obsolete config.

Fixes 1b79fc4f2bfd ("x86/apb_timer: Remove driver for deprecated platform")
---
 arch/x86/Kconfig | 1 -
 1 file changed, 1 deletion(-)
arch/x86/Kconfig +0 -1
index b553e8b4df4e..cd466223c557 100644
@@ -606,7 +606,6 @@ config X86_INTEL_MID
 	depends on X86_IO_APIC
 	select I2C
 	select DW_APB_TIMER
-	select APB_TIMER
 	select INTEL_SCU_PCI
 	help
 	  Select to build a kernel capable of supporting Intel MID (Mobile
Signature · 1 line
2.17.1
maintainer
#13 lore raw

Re: [PATCH 5/9] x86/Kconfig: remove reference to obsolete APB_TIMER config

Quoted 10 lines · Lukas Bulwahn

Commit 1b79fc4f2bfd ("x86/apb_timer: Remove driver for deprecated platform") removes the definition of the config APB_TIMER in ./arch/x86/Kconfig, but misses to remove a reference to it in config X86_INTEL_MID.

Fortunately, ./scripts/checkkconfigsymbols.py warns:

APB_TIMER
Referencing files: arch/x86/Kconfig

Quoted 8 lines

Remove this reference to the obsolete config.

Fixes 1b79fc4f2bfd ("x86/apb_timer: Remove driver for deprecated platform")
---
 arch/x86/Kconfig | 1 -
 1 file changed, 1 deletion(-)

I've combined that and the previous one into a single patch since they're pretty trivial:

---
From: Lukas Bulwahn <[email protected]>
Date: Tue, 3 Aug 2021 13:35:26 +0200
Subject: [PATCH] x86/Kconfig: Remove references to obsolete Kconfig symbols

Remove two symbols referenced in Kconfig which have been removed previously by:

ef3c67b6454b ("mfd: intel_msic: Remove driver for deprecated platform") 1b79fc4f2bfd ("x86/apb_timer: Remove driver for deprecated platform")

Detected by scripts/checkkconfigsymbols.py.

[ bp: Merge into a single patch. ]

---
 arch/x86/Kconfig | 2 --
 1 file changed, 2 deletions(-)
arch/x86/Kconfig +0 -2
index 4e001bbbb425..b79e88ee6627 100644
@@ -605,9 +605,7 @@ config X86_INTEL_MID
 	depends on X86_IO_APIC
 	select I2C
 	select DW_APB_TIMER
-	select APB_TIMER
 	select INTEL_SCU_PCI
-	select MFD_INTEL_MSIC
 	help
 	  Select to build a kernel capable of supporting Intel MID (Mobile
 	  Internet Device) platform systems which do not have the PCI legacy
Signature · 8 lines
2.29.2


-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
#14 lore raw

PATCH 6/9 x86/irq: fix slightly wrong reference in comment

In reply to Lukas Bulwahn

Commit a0cfc74d0b00 ("x86/irq: Provide macro for inlining irq stack switching") refers to CONFIG_UNWIND_FRAME_POINTER, but the config is called CONFIG_UNWINDER_FRAME_POINTER.

Hence, ./scripts/checkkconfigsymbols.py warns:

UNWIND_FRAME_POINTER
Referencing files: arch/x86/include/asm/irq_stack.h

Adjust the name in the comment.

---
 arch/x86/include/asm/irq_stack.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/irq_stack.h +1 -1
index 562854c60808..8912492a78f1 100644
@@ -58,7 +58,7 @@
  *     the output constraints to make the compiler aware that R11 cannot be
  *     reused after the asm() statement.
  *
- *     For builds with CONFIG_UNWIND_FRAME_POINTER ASM_CALL_CONSTRAINT is
+ *     For builds with CONFIG_UNWINDER_FRAME_POINTER, ASM_CALL_CONSTRAINT is
  *     required as well as this prevents certain creative GCC variants from
  *     misplacing the ASM code.
  *
Signature · 1 line
2.17.1
merge bot Applied
#15 lore raw

[tip: x86/cleanups] x86: Fix misspelled Kconfig symbols

The following commit has been merged into the x86/cleanups branch of tip:

The following commit has been merged into the x86/cleanups branch of tip:

Commit-ID:     6bf8a55d8344df1f61a29b18c398bcdf3539e163
Gitweb:        https://git.kernel.org/tip/6bf8a55d8344df1f61a29b18c398bcdf3539e163
Author:        Lukas Bulwahn <[email protected]>
AuthorDate:    Tue, 05 Oct 2021 21:48:30 +02:00
Committer:     Borislav Petkov <[email protected]>
CommitterDate: Tue, 05 Oct 2021 21:48:30 +02:00

x86: Fix misspelled Kconfig symbols

Fix misspelled Kconfig symbols as detected by
scripts/checkkconfigsymbols.py.

[ bp: Combine into a single patch. ]

---
 arch/x86/include/asm/ia32.h      | 2 +-
 arch/x86/include/asm/irq_stack.h | 2 +-
 arch/x86/include/asm/page_32.h   | 2 +-
 arch/x86/include/asm/uaccess.h   | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)
arch/x86/include/asm/ia32.h +1 -1
index 2c5f786..fada857 100644
@@ -68,6 +68,6 @@ extern void ia32_pick_mmap_layout(struct mm_struct *mm);
 
 #endif
 
-#endif /* !CONFIG_IA32_SUPPORT */
+#endif /* CONFIG_IA32_EMULATION */
 
 #endif /* _ASM_X86_IA32_H */
arch/x86/include/asm/irq_stack.h +1 -1
index 562854c..8912492 100644
@@ -58,7 +58,7 @@
  *     the output constraints to make the compiler aware that R11 cannot be
  *     reused after the asm() statement.
  *
- *     For builds with CONFIG_UNWIND_FRAME_POINTER ASM_CALL_CONSTRAINT is
+ *     For builds with CONFIG_UNWINDER_FRAME_POINTER, ASM_CALL_CONSTRAINT is
  *     required as well as this prevents certain creative GCC variants from
  *     misplacing the ASM code.
  *
arch/x86/include/asm/page_32.h +1 -1
index 94dbd51..b13f848 100644
@@ -43,7 +43,7 @@ static inline void copy_page(void *to, void *from)
 {
 	memcpy(to, from, PAGE_SIZE);
 }
-#endif	/* CONFIG_X86_3DNOW */
+#endif	/* CONFIG_X86_USE_3DNOW */
 #endif	/* !__ASSEMBLY__ */
 
 #endif /* _ASM_X86_PAGE_32_H */
arch/x86/include/asm/uaccess.h +1 -1
index c9fa7be..e7fc2c5 100644
@@ -411,7 +411,7 @@ do {									\
 		     : [umem] "m" (__m(addr)),				\
 		       [efault] "i" (-EFAULT), "0" (err))
 
-#endif // CONFIG_CC_ASM_GOTO_OUTPUT
+#endif // CONFIG_CC_HAS_ASM_GOTO_OUTPUT
 
 /* FIXME: this hack is definitely wrong -AK */
 struct __large_struct { unsigned long buf[100]; };
#16 lore raw

PATCH 7/9 x86/uaccess: adjust comment for endif of CONFIG_CC_HAS_ASM_GOTO_OUTPUT

In reply to Lukas Bulwahn

Commit 865c50e1d279 ("x86/uaccess: utilize CONFIG_CC_HAS_ASM_GOTO_OUTPUT") adds an ifdef CONFIG_CC_HAS_ASM_GOTO_OUTPUT, and a comment on the corresponding endif. However, it refers slightly mismatching to CONFIG_CC_ASM_GOTO_OUTPUT instead.

Hence, ./scripts/checkkconfigsymbols.py warns:

CC_ASM_GOTO_OUTPUT
Referencing files: arch/x86/include/asm/uaccess.h

Adjust the comment on endif to the actual ifdef condition.

---
 arch/x86/include/asm/uaccess.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/uaccess.h +1 -1
index c9fa7be3df82..e7fc2c515e08 100644
@@ -411,7 +411,7 @@ do {									\
 		     : [umem] "m" (__m(addr)),				\
 		       [efault] "i" (-EFAULT), "0" (err))
 
-#endif // CONFIG_CC_ASM_GOTO_OUTPUT
+#endif // CONFIG_CC_HAS_ASM_GOTO_OUTPUT
 
 /* FIXME: this hack is definitely wrong -AK */
 struct __large_struct { unsigned long buf[100]; };
Signature · 1 line
2.17.1
#17 lore raw

PATCH 8/9 x86: page_32.h: adjust comment for endif of CONFIG_X86_USE_3DNOW

In reply to Lukas Bulwahn

Commit 345b904c3f7c ("x86: page.h: unify page copying and clearing") adds an ifdef CONFIG_X86_USE_3DNOW, and a comment on the corresponding endif. However, it refers slightly mismatching to CONFIG_X86_3DNOW instead.

Hence, ./scripts/checkkconfigsymbols.py warns:

X86_3DNOW
Referencing files: arch/x86/include/asm/page_32.h

Adjust the comment on endif to the actual ifdef condition.

---
 arch/x86/include/asm/page_32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/page_32.h +1 -1
index 94dbd51df58f..b13f8488ac85 100644
@@ -43,7 +43,7 @@ static inline void copy_page(void *to, void *from)
 {
 	memcpy(to, from, PAGE_SIZE);
 }
-#endif	/* CONFIG_X86_3DNOW */
+#endif	/* CONFIG_X86_USE_3DNOW */
 #endif	/* !__ASSEMBLY__ */
 
 #endif /* _ASM_X86_PAGE_32_H */
Signature · 1 line
2.17.1
#18 lore raw

PATCH 9/9 x86: ia32.h: adjust comment for endif of CONFIG_IA32_EMULATION

In reply to Lukas Bulwahn

The content of the ia32 header is guarded by
"ifdef CONFIG_IA32_EMULATION". The comment on the corresponding endif refers slightly mismatching to CONFIG_IA32_SUPPORT instead.

Hence, ./scripts/checkkconfigsymbols.py warns:

IA32_SUPPORT
Referencing files: arch/x86/include/asm/ia32.h

Adjust the comment on endif to the actual ifdef condition.

---
 arch/x86/include/asm/ia32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/include/asm/ia32.h +1 -1
index 2c5f7861d373..fada857f0a1e 100644
@@ -68,6 +68,6 @@ extern void ia32_pick_mmap_layout(struct mm_struct *mm);
 
 #endif
 
-#endif /* !CONFIG_IA32_SUPPORT */
+#endif /* CONFIG_IA32_EMULATION */
 
 #endif /* _ASM_X86_IA32_H */
Signature · 1 line
2.17.1
maintainer
#19 lore raw

Re: [PATCH 9/9] x86: ia32.h: adjust comment for endif of CONFIG_IA32_EMULATION

Quoted 9 lines · Lukas Bulwahn

The content of the ia32 header is guarded by
"ifdef CONFIG_IA32_EMULATION". The comment on the corresponding endif refers slightly mismatching to CONFIG_IA32_SUPPORT instead.

Hence, ./scripts/checkkconfigsymbols.py warns:

IA32_SUPPORT
Referencing files: arch/x86/include/asm/ia32.h

Quoted 6 lines

Adjust the comment on endif to the actual ifdef condition.

---
 arch/x86/include/asm/ia32.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

Merged the last 4 into a single patch because they're trivial:

---
From: Lukas Bulwahn <[email protected]>
Date: Tue, 5 Oct 2021 21:48:30 +0200
Subject: [PATCH] x86: Fix misspelled Kconfig symbols

Fix misspelled Kconfig symbols as detected by scripts/checkkconfigsymbols.py.

[ bp: Combine into a single patch. ]

---
 arch/x86/include/asm/ia32.h      | 2 +-
 arch/x86/include/asm/irq_stack.h | 2 +-
 arch/x86/include/asm/page_32.h   | 2 +-
 arch/x86/include/asm/uaccess.h   | 2 +-
 4 files changed, 4 insertions(+), 4 deletions(-)
arch/x86/include/asm/ia32.h +1 -1
index 2c5f7861d373..fada857f0a1e 100644
@@ -68,6 +68,6 @@ extern void ia32_pick_mmap_layout(struct mm_struct *mm);
 
 #endif
 
-#endif /* !CONFIG_IA32_SUPPORT */
+#endif /* CONFIG_IA32_EMULATION */
 
 #endif /* _ASM_X86_IA32_H */
arch/x86/include/asm/irq_stack.h +1 -1
index 562854c60808..8912492a78f1 100644
@@ -58,7 +58,7 @@
  *     the output constraints to make the compiler aware that R11 cannot be
  *     reused after the asm() statement.
  *
- *     For builds with CONFIG_UNWIND_FRAME_POINTER ASM_CALL_CONSTRAINT is
+ *     For builds with CONFIG_UNWINDER_FRAME_POINTER, ASM_CALL_CONSTRAINT is
  *     required as well as this prevents certain creative GCC variants from
  *     misplacing the ASM code.
  *
arch/x86/include/asm/page_32.h +1 -1
index 94dbd51df58f..b13f8488ac85 100644
@@ -43,7 +43,7 @@ static inline void copy_page(void *to, void *from)
 {
 	memcpy(to, from, PAGE_SIZE);
 }
-#endif	/* CONFIG_X86_3DNOW */
+#endif	/* CONFIG_X86_USE_3DNOW */
 #endif	/* !__ASSEMBLY__ */
 
 #endif /* _ASM_X86_PAGE_32_H */
arch/x86/include/asm/uaccess.h +1 -1
index c9fa7be3df82..e7fc2c515e08 100644
@@ -411,7 +411,7 @@ do {									\
 		     : [umem] "m" (__m(addr)),				\
 		       [efault] "i" (-EFAULT), "0" (err))
 
-#endif // CONFIG_CC_ASM_GOTO_OUTPUT
+#endif // CONFIG_CC_HAS_ASM_GOTO_OUTPUT
 
 /* FIXME: this hack is definitely wrong -AK */
 struct __large_struct { unsigned long buf[100]; };
Signature · 7 lines
2.29.2

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette
#20 lore raw
In reply to Lukas Bulwahn
Quoted 2 lines · Lukas Bulwahn

- a reference to STRICT_IOMEM in arch/x86/mm/init.c unclear to me: which exact config this refers to

Are you referring to the reference in this comment?

Review

Quoted 8 lines
(quoted diff) +8 -0
+       /*
+        * This must follow RAM test, since System RAM is considered a
+        * restricted resource under CONFIG_STRICT_IOMEM.
+        */
+       if (iomem_is_exclusive(pagenr << PAGE_SHIFT)) {
+               /* Low 1MB bypasses iomem restrictions. */
+               if (pagenr < 256)
+                       return 1;

...

That came from here:

Quoted 5 lines

commit a4866aa812518ed1a37d8ea0c881dc946409de94

Author: Kees Cook <[email protected]>
Date:   Wed Apr 5 09:39:08 2017 -0700

mm: Tighten x86 /dev/mem with zeroing reads

Which also added this hunk:

Review

Quoted 5 lines

#ifdef CONFIG_STRICT_DEVMEM

(quoted diff) +4 -0
+static inline int page_is_allowed(unsigned long pfn)
+{
+       return devmem_is_allowed(pfn);
+}

and talks about CONFIG_STRICT_DEVMEM in the changelog:

Quoted 4 lines

mm: Tighten x86 /dev/mem with zeroing reads

Under CONFIG_STRICT_DEVMEM, reading System RAM through /dev/mem is disallowed. However, on x86, the first 1MB was always allowed for BIOS

...

It's a pretty safe guess that STRICT_IOMEM refers to CONFIG_STRICT_DEVMEM.

#21 lore raw
Quoted 15 lines · Dave Hansen
Quoted 2 lines · Lukas Bulwahn

- a reference to STRICT_IOMEM in arch/x86/mm/init.c unclear to me: which exact config this refers to

Are you referring to the reference in this comment?

Review

Quoted 8 lines
(quoted diff) +8 -0
+       /*
+        * This must follow RAM test, since System RAM is considered a
+        * restricted resource under CONFIG_STRICT_IOMEM.
+        */
+       if (iomem_is_exclusive(pagenr << PAGE_SHIFT)) {
+               /* Low 1MB bypasses iomem restrictions. */
+               if (pagenr < 256)
+                       return 1;
Quoted 2 lines

...

Yes, that is what I referred to.

Quoted 18 lines

That came from here:

Quoted 5 lines

commit a4866aa812518ed1a37d8ea0c881dc946409de94

Author: Kees Cook <[email protected]>
Date:   Wed Apr 5 09:39:08 2017 -0700

mm: Tighten x86 /dev/mem with zeroing reads

Which also added this hunk:

Review

Quoted 5 lines

#ifdef CONFIG_STRICT_DEVMEM

(quoted diff) +4 -0
+static inline int page_is_allowed(unsigned long pfn)
+{
+       return devmem_is_allowed(pfn);
+}

and talks about CONFIG_STRICT_DEVMEM in the changelog:

Quoted 7 lines
Quoted 4 lines

mm: Tighten x86 /dev/mem with zeroing reads

Under CONFIG_STRICT_DEVMEM, reading System RAM through /dev/mem is disallowed. However, on x86, the first 1MB was always allowed for BIOS

...

It's a pretty safe guess that STRICT_IOMEM refers to CONFIG_STRICT_DEVMEM.

Thanks, Dave.

If the maintainers consider updates to comments making them consistent with the code as worth being picked, I will turn your analysis into a proper commit message and provide a patch to update that comment.

Lukas

#22 lore raw
Quoted 6 lines · Lukas Bulwahn
Quoted 1 line

It's a pretty safe guess that STRICT_IOMEM refers to CONFIG_STRICT_DEVMEM.

Thanks, Dave.

If the maintainers consider updates to comments making them consistent with the code as worth being picked, I will turn your analysis into a proper commit message and provide a patch to update that comment.

I'd happily ack a patch that does that. Might as well fix that kind of stuff now before more time passes and it gets harder to track down.

#23 lore raw
Quoted 11 lines · Dave Hansen
Quoted 6 lines · Lukas Bulwahn
Quoted 1 line

It's a pretty safe guess that STRICT_IOMEM refers to CONFIG_STRICT_DEVMEM.

Thanks, Dave.

If the maintainers consider updates to comments making them consistent with the code as worth being picked, I will turn your analysis into a proper commit message and provide a patch to update that comment.

I'd happily ack a patch that does that. Might as well fix that kind of stuff now before more time passes and it gets harder to track down.

Sorry, Dave, it took me a bit longer to come back to my second pass on references to non-existing configs in the kernel tree...

I have just sent out a quick fix for this issue, now in your and other x86 maintainers' mailbox:

https://lore.kernel.org/all/[email protected]/

I am happy to get an ack for that patch.

Lukas

Acked-by on the v1 series
#24 lore raw
In reply to Lukas Bulwahn
Quoted 7 lines · Lukas Bulwahn

Dear x86 maintainers,

Randy Dunlap pointed out that ./scripts/checkkconfigsymbols.py warns on invalid references to Kconfig symbols (often, minor typos, name confusions or outdated references): see Link below.

Quoted 9 lines

This is a first patch series addressing the issues reported by ./scripts/checkkconfigsymbols.py in the ./arch/x86/ directories, quickly filtered down with:

./scripts/checkkconfigsymbols.py | grep "arch/x86" -B 1 -A 2

It addresses some issues I considered to be "true positives"; so, issues that should be addressed and cleaned up.

...

for the series:

Thanks.

Quoted 21 lines

Lukas Bulwahn (9):
x86/entry: correct reference to intended CONFIG_64_BIT x86, olpc: correct condition to intended CONFIG_OLPC_XO15_SCI x86/Kconfig: correct reference to MWINCHIP3D
x86/Kconfig: remove reference to obsolete MFD_INTEL_MSIC config x86/Kconfig: remove reference to obsolete APB_TIMER config x86/irq: fix slightly wrong reference in comment x86/uaccess: adjust comment for endif of CONFIG_CC_HAS_ASM_GOTO_OUTPUT
x86: page_32.h: adjust comment for endif of CONFIG_X86_USE_3DNOW
x86: ia32.h: adjust comment for endif of CONFIG_IA32_EMULATION

  arch/x86/Kconfig                    | 4 +---
  arch/x86/include/asm/entry-common.h | 2 +-
  arch/x86/include/asm/ia32.h         | 2 +-
  arch/x86/include/asm/irq_stack.h    | 2 +-
  arch/x86/include/asm/page_32.h      | 2 +-
  arch/x86/include/asm/uaccess.h      | 2 +-
  arch/x86/platform/olpc/olpc.c       | 2 +-
  7 files changed, 7 insertions(+), 9 deletions(-)
Signature · 1 line
~Randy