Skip to contentSkip to conversation

powerpc: Make setjmp/longjump signature standard

PATCH In review Clement Courbet 6 participants 13 messages Mar 27, 2020 – Apr 1, 2020

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 powerpc: Make setjmp/longjump signature standard

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

---
 arch/powerpc/include/asm/setjmp.h | 6 ++++--
 arch/powerpc/kexec/Makefile       | 3 ---
 2 files changed, 4 insertions(+), 5 deletions(-)
arch/powerpc/include/asm/setjmp.h +4 -2
index e9f81bb3f83b..84bb0d140d59 100644
@@ -7,7 +7,9 @@
 
 #define JMP_BUF_LEN    23
 
-extern long setjmp(long *) __attribute__((returns_twice));
-extern void longjmp(long *, long) __attribute__((noreturn));
+typedef long *jmp_buf;
+
+extern int setjmp(jmp_buf env) __attribute__((returns_twice));
+extern void longjmp(jmp_buf env, int val) __attribute__((noreturn));
 
 #endif /* _ASM_POWERPC_SETJMP_H */
arch/powerpc/kexec/Makefile +0 -3
index 378f6108a414..86380c69f5ce 100644
@@ -3,9 +3,6 @@
 # Makefile for the linux kernel.
 #
 
-# Avoid clang warnings around longjmp/setjmp declarations
-CFLAGS_crash.o += -ffreestanding
-
 obj-y				+= core.o crash.o core_$(BITS).o
 
 obj-$(CONFIG_PPC32)		+= relocate_32.o
Signature · 1 line
2.25.1.696.g5e7596f4ac-goog
#1 lore raw
Quoted 9 lines · Clement Courbet

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

Quoted 7 lines

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

Hi Clement, thanks for the patch! Would you mind sending a V2 that included a similar fix to arch/powerpc/xmon/Makefile?

For context, this was the original patch:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=aea447141c7e7824b81b49acd1bc78
which was then modified to:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=c9029ef9c95765e7b63c4d9aa780674447db1ec0

So on your V2, if you include in the commit message, the line:

Fixes c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")

then that will help our LTS branch maintainers back port it to the appropriate branches.

Quoted 39 lines
---
 arch/powerpc/include/asm/setjmp.h | 6 ++++--
 arch/powerpc/kexec/Makefile       | 3 ---
 2 files changed, 4 insertions(+), 5 deletions(-)
arch/powerpc/include/asm/setjmp.h +4 -2
index e9f81bb3f83b..84bb0d140d59 100644
@@ -7,7 +7,9 @@
 
 #define JMP_BUF_LEN    23
 
-extern long setjmp(long *) __attribute__((returns_twice));
-extern void longjmp(long *, long) __attribute__((noreturn));
+typedef long *jmp_buf;
+
+extern int setjmp(jmp_buf env) __attribute__((returns_twice));
+extern void longjmp(jmp_buf env, int val) __attribute__((noreturn));
 
 #endif /* _ASM_POWERPC_SETJMP_H */
arch/powerpc/kexec/Makefile +0 -4
index 378f6108a414..86380c69f5ce 100644
@@ -3,9 +3,6 @@
 # Makefile for the linux kernel.
 #
 
-# Avoid clang warnings around longjmp/setjmp declarations
-CFLAGS_crash.o += -ffreestanding
-
 obj-y                          += core.o crash.o core_$(BITS).o
 
 obj-$(CONFIG_PPC32)            += relocate_32.o
--

2.25.1.696.g5e7596f4ac-goog

Signature · 2 lines
Thanks,
~Nick Desaulniers
Reviewed-by Tested-by
#2 lore raw
Quoted 10 lines · Nick Desaulniers
Quoted 9 lines · Clement Courbet

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

Quoted 7 lines
Quoted 7 lines

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

Thanks for fixing this properly, not really sure why I did not think of this in the first place. I guess my thought was the warning makes it seem like clang is going to ignore the kernel's implementation of setjmp/longjmp but I can't truly remember.

Quoted 2 lines

Hi Clement, thanks for the patch! Would you mind sending a V2 that included a similar fix to arch/powerpc/xmon/Makefile?

Agreed.

Quoted 11 lines

For context, this was the original patch:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=aea447141c7e7824b81b49acd1bc78
which was then modified to:
https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/commit/?id=c9029ef9c95765e7b63c4d9aa780674447db1ec0

So on your V2, if you include in the commit message, the line:

Fixes c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")

then that will help our LTS branch maintainers back port it to the appropriate branches.

The tags should be:

Cc [email protected] # v4.14+
Fixes c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")

that way it explicitly gets picked up for stable, rather than Sasha's AUTOSEL process, which could miss it.

With the xmon/Makefile -ffreestanding removed and the tags updated, consider this:

Cheers,
Nathan

#3 lore raw
In reply to Clement Courbet

Subject line, change longjump to longjmp

Le 27/03/2020 à 11:07, Clement Courbet a écrit :

Quoted 25 lines

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

---
  arch/powerpc/include/asm/setjmp.h | 6 ++++--
  arch/powerpc/kexec/Makefile       | 3 ---
  2 files changed, 4 insertions(+), 5 deletions(-)
arch/powerpc/include/asm/setjmp.h +0 -0
index e9f81bb3f83b..84bb0d140d59 100644

arch/powerpc/include/asm/setjmp.h line 10

Quoted 6 lines
arch/powerpc/include/asm/setjmp.h +1 -2
@@ -7,7 +7,9 @@
 
  #define JMP_BUF_LEN    23
 
-extern long setjmp(long *) __attribute__((returns_twice));
-extern void longjmp(long *, long) __attribute__((noreturn));
+typedef long *jmp_buf;

Do we need that new opaque typedef ? Why not just keep long * ?

Quoted 11 lines
arch/powerpc/include/asm/setjmp.h +3 -0
+
+extern int setjmp(jmp_buf env) __attribute__((returns_twice));
+extern void longjmp(jmp_buf env, int val) __attribute__((noreturn));
 
  #endif /* _ASM_POWERPC_SETJMP_H */
arch/powerpc/kexec/Makefile +0 -0
index 378f6108a414..86380c69f5ce 100644

arch/powerpc/kexec/Makefile line 8

Quoted 9 lines
arch/powerpc/kexec/Makefile +0 -3
@@ -3,9 +3,6 @@
  # Makefile for the linux kernel.
  #
 
-# Avoid clang warnings around longjmp/setjmp declarations
-CFLAGS_crash.o += -ffreestanding
-
  obj-y				+= core.o crash.o core_$(BITS).o
 
  obj-$(CONFIG_PPC32)		+= relocate_32.o

Christophe

#4 lore raw
Quoted 23 lines · Christophe Leroy

Subject line, change longjump to longjmp

Le 27/03/2020 à 11:07, Clement Courbet a écrit :

Quoted 20 lines

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

---
  arch/powerpc/include/asm/setjmp.h | 6 ++++--
  arch/powerpc/kexec/Makefile       | 3 ---
  2 files changed, 4 insertions(+), 5 deletions(-)
Quoted 11 lines

arch/powerpc/include/asm/setjmp.h line 8

Quoted 9 lines
arch/powerpc/include/asm/setjmp.h +1 -2
index e9f81bb3f83b..84bb0d140d59 100644
@@ -7,7 +7,9 @@
  #define JMP_BUF_LEN    23
-extern long setjmp(long *) __attribute__((returns_twice));
-extern void longjmp(long *, long) __attribute__((noreturn));
+typedef long *jmp_buf;

Do we need that new opaque typedef ? Why not just keep long * ?

Yes, otherwise the warning comes back:

In file included from arch/powerpc/kexec/crash.c:25:
arch/powerpc/include/asm/setjmp.h:10:12: error: declaration of built-in function 'setjmp' requires the declaration of the 'jmp_buf' type, commonly provided in the header <setjmp.h>. [-Werror,-Wincomplete-setjmp-declaration]

extern int setjmp(long *env) __attribute__((returns_twice));
           ^
arch/powerpc/include/asm/setjmp.h:11:13: error: declaration of built-in function 'longjmp' requires the declaration of the 'jmp_buf' type, commonly provided in the header <setjmp.h>. [-Werror,-Wincomplete-setjmp-declaration]
extern void longjmp(long *env, int val) __attribute__((noreturn));
            ^

2 errors generated.

Quoted 19 lines
Quoted 9 lines
arch/powerpc/include/asm/setjmp.h +3 -0
+
+extern int setjmp(jmp_buf env) __attribute__((returns_twice));
+extern void longjmp(jmp_buf env, int val) __attribute__((noreturn));
  #endif /* _ASM_POWERPC_SETJMP_H */
arch/powerpc/kexec/Makefile +0 -0
index 378f6108a414..86380c69f5ce 100644

arch/powerpc/kexec/Makefile line 7

Quoted 8 lines
arch/powerpc/kexec/Makefile +0 -3
@@ -3,9 +3,6 @@
  # Makefile for the linux kernel.
  #
-# Avoid clang warnings around longjmp/setjmp declarations
-CFLAGS_crash.o += -ffreestanding
-
  obj-y				+= core.o crash.o core_$(BITS).o
  obj-$(CONFIG_PPC32)		+= relocate_32.o
 

Christophe

#5 lore raw

PATCH 1/1 powerpc: Make setjmp/longjump signature standard

In reply to Clement Courbet

Thanks you all for the comments. Everything addressed, plus the array vs pointer suggestion from Segher Boessenkool on the other thread, which is only cosmetic and does not change anything wrt behaviour.

#6 lore raw

PATCH v2 1/1 powerpc: Make setjmp/longjmp signature standard

In reply to Clement Courbet

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

---

v2:
Use and array type as suggested by Segher Boessenkool

Cc [email protected] # v4.14+
Fixes c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")
---
 arch/powerpc/include/asm/setjmp.h | 6 ++++--
 arch/powerpc/kexec/Makefile       | 3 ---
 arch/powerpc/xmon/Makefile        | 3 ---
 3 files changed, 4 insertions(+), 8 deletions(-)
arch/powerpc/include/asm/setjmp.h +4 -2
index e9f81bb3f83b..f798e80e4106 100644
@@ -7,7 +7,9 @@
 
 #define JMP_BUF_LEN    23
 
-extern long setjmp(long *) __attribute__((returns_twice));
-extern void longjmp(long *, long) __attribute__((noreturn));
+typedef long jmp_buf[JMP_BUF_LEN];
+
+extern int setjmp(jmp_buf env) __attribute__((returns_twice));
+extern void longjmp(jmp_buf env, int val) __attribute__((noreturn));
 
 #endif /* _ASM_POWERPC_SETJMP_H */
arch/powerpc/kexec/Makefile +0 -3
index 378f6108a414..86380c69f5ce 100644
@@ -3,9 +3,6 @@
 # Makefile for the linux kernel.
 #
 
-# Avoid clang warnings around longjmp/setjmp declarations
-CFLAGS_crash.o += -ffreestanding
-
 obj-y				+= core.o crash.o core_$(BITS).o
 
 obj-$(CONFIG_PPC32)		+= relocate_32.o
arch/powerpc/xmon/Makefile +0 -3
index c3842dbeb1b7..6f9cccea54f3 100644
@@ -1,9 +1,6 @@
 # SPDX-License-Identifier: GPL-2.0
 # Makefile for xmon
 
-# Avoid clang warnings around longjmp/setjmp declarations
-subdir-ccflags-y := -ffreestanding
-
 GCOV_PROFILE := n
 KCOV_INSTRUMENT := n
 UBSAN_SANITIZE := n
Signature · 1 line
2.26.0.rc2.310.g2932bb562d-goog
#7 lore raw

Re: [PATCH v2] powerpc: Make setjmp/longjmp signature standard

Quoted 18 lines · Clement Courbet

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

Quoted 6 lines

---

v2:
Use and array type as suggested by Segher Boessenkool

Cc [email protected] # v4.14+
Fixes c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")

Actual diff itself looks good but these two tags need to be above your signoff to be included in the final changelog. Not sure if Michael will want a v3 with that or if it can manually be done when applying.

Cheers,
Nathan

#8 lore raw

PATCH v3 1/1 powerpc: Make setjmp/longjmp signature standard

In reply to Clement Courbet

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

Cc [email protected] # v4.14+
Fixes c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")

---

v2:
Use and array type as suggested by Segher Boessenkool Add fix tags.

v3:
Properly place tags.

---
 arch/powerpc/include/asm/setjmp.h | 6 ++++--
 arch/powerpc/kexec/Makefile       | 3 ---
 arch/powerpc/xmon/Makefile        | 3 ---
 3 files changed, 4 insertions(+), 8 deletions(-)
arch/powerpc/include/asm/setjmp.h +4 -2
index e9f81bb3f83b..f798e80e4106 100644
@@ -7,7 +7,9 @@
 
 #define JMP_BUF_LEN    23
 
-extern long setjmp(long *) __attribute__((returns_twice));
-extern void longjmp(long *, long) __attribute__((noreturn));
+typedef long jmp_buf[JMP_BUF_LEN];
+
+extern int setjmp(jmp_buf env) __attribute__((returns_twice));
+extern void longjmp(jmp_buf env, int val) __attribute__((noreturn));
 
 #endif /* _ASM_POWERPC_SETJMP_H */
arch/powerpc/kexec/Makefile +0 -3
index 378f6108a414..86380c69f5ce 100644
@@ -3,9 +3,6 @@
 # Makefile for the linux kernel.
 #
 
-# Avoid clang warnings around longjmp/setjmp declarations
-CFLAGS_crash.o += -ffreestanding
-
 obj-y				+= core.o crash.o core_$(BITS).o
 
 obj-$(CONFIG_PPC32)		+= relocate_32.o
arch/powerpc/xmon/Makefile +0 -3
index c3842dbeb1b7..6f9cccea54f3 100644
@@ -1,9 +1,6 @@
 # SPDX-License-Identifier: GPL-2.0
 # Makefile for xmon
 
-# Avoid clang warnings around longjmp/setjmp declarations
-subdir-ccflags-y := -ffreestanding
-
 GCOV_PROFILE := n
 KCOV_INSTRUMENT := n
 UBSAN_SANITIZE := n
Signature · 1 line
2.26.0.rc2.310.g2932bb562d-goog
Reviewed-by Tested-by on v3 1/1
#9 lore raw

Re: [PATCH v3] powerpc: Make setjmp/longjmp signature standard

Quoted 15 lines · Clement Courbet

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

Quoted 5 lines
Cc [email protected] # v4.14+
Fixes c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")

Third time's a charm (for patches that tackle this warning). Thanks for following up on this cleanup!

The extent of my testing was compile testing with Clang.

Quoted 62 lines

---

v2:
Use and array type as suggested by Segher Boessenkool Add fix tags.

v3:
Properly place tags.

---
 arch/powerpc/include/asm/setjmp.h | 6 ++++--
 arch/powerpc/kexec/Makefile       | 3 ---
 arch/powerpc/xmon/Makefile        | 3 ---
 3 files changed, 4 insertions(+), 8 deletions(-)
arch/powerpc/include/asm/setjmp.h +4 -2
index e9f81bb3f83b..f798e80e4106 100644
@@ -7,7 +7,9 @@
 
 #define JMP_BUF_LEN    23
 
-extern long setjmp(long *) __attribute__((returns_twice));
-extern void longjmp(long *, long) __attribute__((noreturn));
+typedef long jmp_buf[JMP_BUF_LEN];
+
+extern int setjmp(jmp_buf env) __attribute__((returns_twice));
+extern void longjmp(jmp_buf env, int val) __attribute__((noreturn));
 
 #endif /* _ASM_POWERPC_SETJMP_H */
arch/powerpc/kexec/Makefile +0 -3
index 378f6108a414..86380c69f5ce 100644
@@ -3,9 +3,6 @@
 # Makefile for the linux kernel.
 #
 
-# Avoid clang warnings around longjmp/setjmp declarations
-CFLAGS_crash.o += -ffreestanding
-
 obj-y                          += core.o crash.o core_$(BITS).o
 
 obj-$(CONFIG_PPC32)            += relocate_32.o
arch/powerpc/xmon/Makefile +0 -4
index c3842dbeb1b7..6f9cccea54f3 100644
@@ -1,9 +1,6 @@
 # SPDX-License-Identifier: GPL-2.0
 # Makefile for xmon
 
-# Avoid clang warnings around longjmp/setjmp declarations
-subdir-ccflags-y := -ffreestanding
-
 GCOV_PROFILE := n
 KCOV_INSTRUMENT := n
 UBSAN_SANITIZE := n
--

2.26.0.rc2.310.g2932bb562d-goog

Signature · 2 lines
Thanks,
~Nick Desaulniers
maintainer
#10 lore raw

Re: [PATCH v3] powerpc: Make setjmp/longjmp signature standard

In reply to Clement Courbet

Hi

[This is an automated email]

This commit has been processed because it contains a "Fixes:" tag
fixing commit: c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp").

The bot has tested the following trees: v5.5.13, v5.4.28, v4.19.113, v4.14.174.

v5.5.13: Build OK!
v5.4.28: Failed to apply! Possible dependencies:
74277f00b232 ("powerpc/fsl_booke/kaslr: export offset in VMCOREINFO ELF notes") 793b08e2efff ("powerpc/kexec: Move kexec files into a dedicated subdir.") 9f7bd9201521 ("powerpc/32: Split kexec low level code out of misc_32.S")

v4.19.113: Failed to apply! Possible dependencies:
2874c5fd2842 ("treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152") 40b0b3f8fb2d ("treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 230") 793b08e2efff ("powerpc/kexec: Move kexec files into a dedicated subdir.") 9f7bd9201521 ("powerpc/32: Split kexec low level code out of misc_32.S") aa497d435241 ("powerpc: Add attributes for setjmp/longjmp") c47ca98d32a2 ("powerpc: Move core kernel logic into arch/powerpc/Kbuild") fb0b0a73b223 ("powerpc: Enable kcov")

v4.14.174: Failed to apply! Possible dependencies:
06bb53b33804 ("powerpc: store and restore the pkey state across context switches") 1421dc6d4829 ("powerpc/kbuild: Use flags variables rather than overriding LD/CC/AS") 1b4037deadb6 ("powerpc: helper function to read, write AMR, IAMR, UAMOR registers") 2ddc53f3a751 ("powerpc: implementation for arch_set_user_pkey_access()") 471d7ff8b51b ("powerpc/64s: Remove POWER4 support")
4d70b698f957 ("powerpc: helper functions to initialize AMR, IAMR and UAMOR registers") 4fb158f65ac5 ("powerpc: track allocation status of all pkeys") 793b08e2efff ("powerpc/kexec: Move kexec files into a dedicated subdir.") 92e3da3cf193 ("powerpc: initial pkey plumbing")
a73657ea19ae ("powerpc/64: Add GENERIC_CPU support for little endian") aa497d435241 ("powerpc: Add attributes for setjmp/longjmp")
badf436f6fa5 ("powerpc/Makefiles: Convert ifeq to ifdef where possible") c0d64cf9fefd ("powerpc: Use feature bit for RTC presence rather than timebase presence") c1807e3f8466 ("powerpc/64: Free up CPU_FTR_ICSWX")
c47ca98d32a2 ("powerpc: Move core kernel logic into arch/powerpc/Kbuild") cf43d3b26452 ("powerpc: Enable pkey subsystem")

NOTE: The patch will not be queued to stable trees until it is upstream.

How should we proceed with this patch?

Signature · 2 lines
Thanks
Sasha
#11 lore raw

Re: [PATCH v3] powerpc: Make setjmp/longjmp signature standard

Michael, is this worth us sending manual backports to stable?

Quoted 51 lines · Sasha Levin

Hi

[This is an automated email]

This commit has been processed because it contains a "Fixes:" tag
fixing commit: c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp").

The bot has tested the following trees: v5.5.13, v5.4.28, v4.19.113, v4.14.174.

v5.5.13: Build OK!
v5.4.28: Failed to apply! Possible dependencies:
74277f00b232 ("powerpc/fsl_booke/kaslr: export offset in VMCOREINFO ELF notes") 793b08e2efff ("powerpc/kexec: Move kexec files into a dedicated subdir.") 9f7bd9201521 ("powerpc/32: Split kexec low level code out of misc_32.S")

v4.19.113: Failed to apply! Possible dependencies:
2874c5fd2842 ("treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 152") 40b0b3f8fb2d ("treewide: Replace GPLv2 boilerplate/reference with SPDX - rule 230") 793b08e2efff ("powerpc/kexec: Move kexec files into a dedicated subdir.") 9f7bd9201521 ("powerpc/32: Split kexec low level code out of misc_32.S") aa497d435241 ("powerpc: Add attributes for setjmp/longjmp") c47ca98d32a2 ("powerpc: Move core kernel logic into arch/powerpc/Kbuild") fb0b0a73b223 ("powerpc: Enable kcov")

v4.14.174: Failed to apply! Possible dependencies:
06bb53b33804 ("powerpc: store and restore the pkey state across context switches") 1421dc6d4829 ("powerpc/kbuild: Use flags variables rather than overriding LD/CC/AS") 1b4037deadb6 ("powerpc: helper function to read, write AMR, IAMR, UAMOR registers") 2ddc53f3a751 ("powerpc: implementation for arch_set_user_pkey_access()") 471d7ff8b51b ("powerpc/64s: Remove POWER4 support")
4d70b698f957 ("powerpc: helper functions to initialize AMR, IAMR and UAMOR registers") 4fb158f65ac5 ("powerpc: track allocation status of all pkeys") 793b08e2efff ("powerpc/kexec: Move kexec files into a dedicated subdir.") 92e3da3cf193 ("powerpc: initial pkey plumbing")
a73657ea19ae ("powerpc/64: Add GENERIC_CPU support for little endian") aa497d435241 ("powerpc: Add attributes for setjmp/longjmp")
badf436f6fa5 ("powerpc/Makefiles: Convert ifeq to ifdef where possible") c0d64cf9fefd ("powerpc: Use feature bit for RTC presence rather than timebase presence") c1807e3f8466 ("powerpc/64: Free up CPU_FTR_ICSWX")
c47ca98d32a2 ("powerpc: Move core kernel logic into arch/powerpc/Kbuild") cf43d3b26452 ("powerpc: Enable pkey subsystem")

NOTE: The patch will not be queued to stable trees until it is upstream.

How should we proceed with this patch?

--
Thanks
Sasha

Signature · 2 lines
Thanks,
~Nick Desaulniers
#12 lore raw

Re: [PATCH v3] powerpc: Make setjmp/longjmp signature standard

In reply to Clement Courbet
Quoted 14 lines · Clement Courbet

Declaring setjmp()/longjmp() as taking longs makes the signature non-standard, and makes clang complain. In the past, this has been worked around by adding -ffreestanding to the compile flags.

The implementation looks like it only ever propagates the value (in longjmp) or sets it to 1 (in setjmp), and we only call longjmp with integer parameters.

This allows removing -ffreestanding from the compilation flags.

Context:
https://lore.kernel.org/patchwork/patch/1214060
https://lore.kernel.org/patchwork/patch/1216174

Quoted 5 lines
Cc [email protected] # v4.14+
Fixes c9029ef9c957 ("powerpc: Avoid clang warnings around setjmp and longjmp")

Applied to powerpc next, thanks.

https://git.kernel.org/powerpc/c/c17eb4dca5a353a9dbbb8ad6934fe57af7165e91

cheers