Skip to contentSkip to conversation

s390: only build for new CPUs with clang

PATCH Superseded Arnd Bergmann 4 participants 11 messages Apr 10, 2019 – Apr 15, 2019 s390

v1 of 2 Superseded Latest is v2 →

lore
Share & tools
lore
#0 lore raw

PATCH 1/2 s390: only build for new CPUs with clang

llvm does does not understand -march=z9-109 and older target specifiers, so disable the respective Kconfig settings and the logic to make the boot code work on old systems when building with clang.

---
 arch/s390/Kconfig       | 6 ++++++
 arch/s390/boot/Makefile | 2 ++
 2 files changed, 8 insertions(+)
arch/s390/Kconfig +6 -0
index 8cd860cba4d1..1a2eec61196d 100644
@@ -240,6 +240,7 @@ choice
 
 config MARCH_Z900
 	bool "IBM zSeries model z800 and z900"
+	depends on !CC_IS_CLANG
 	select HAVE_MARCH_Z900_FEATURES
 	help
 	  Select this to enable optimizations for model z800/z900 (2064 and
@@ -248,6 +249,7 @@ config MARCH_Z900
 
 config MARCH_Z990
 	bool "IBM zSeries model z890 and z990"
+	depends on !CC_IS_CLANG
 	select HAVE_MARCH_Z990_FEATURES
 	help
 	  Select this to enable optimizations for model z890/z990 (2084 and
@@ -256,6 +258,7 @@ config MARCH_Z990
 
 config MARCH_Z9_109
 	bool "IBM System z9"
+	depends on !CC_IS_CLANG
 	select HAVE_MARCH_Z9_109_FEATURES
 	help
 	  Select this to enable optimizations for IBM System z9 (2094 and
@@ -347,12 +350,15 @@ config TUNE_DEFAULT
 
 config TUNE_Z900
 	bool "IBM zSeries model z800 and z900"
+	depends on !CC_IS_CLANG
 
 config TUNE_Z990
 	bool "IBM zSeries model z890 and z990"
+	depends on !CC_IS_CLANG
 
 config TUNE_Z9_109
 	bool "IBM System z9"
+	depends on !CC_IS_CLANG
 
 config TUNE_Z10
 	bool "IBM System z10"
arch/s390/boot/Makefile +2 -0
index c844eaf24ed7..953a74d04990 100644
@@ -11,6 +11,7 @@ KASAN_SANITIZE := n
 KBUILD_AFLAGS := $(KBUILD_AFLAGS_DECOMPRESSOR)
 KBUILD_CFLAGS := $(KBUILD_CFLAGS_DECOMPRESSOR)
 
+ifndef CONFIG_CC_IS_CLANG
 #
 # Use -march=z900 for als.c to be able to print an error
 # message if the kernel is started on a machine which is too old
@@ -25,6 +26,7 @@ CFLAGS_als.o			+= -march=z900
 CFLAGS_REMOVE_sclp_early_core.o	+= $(CC_FLAGS_MARCH)
 CFLAGS_sclp_early_core.o	+= -march=z900
 endif
+endif
 
 CFLAGS_sclp_early_core.o += -I$(srctree)/drivers/s390/char
 
Signature · 1 line
2.20.0
#1 lore raw

PATCH 2/2 s390: boot, purgatory: pass $(CLANG_FLAGS) where needed

The purgatory and boot Makefiles do not inherit the original cflags, so clang falls back to the default target architecture when building it, typically this would be x86 when cross-compiling.

Add $(CLANG_FLAGS) everywhere so we pass the correct --target=s390x-linux option when cross-compiling.

---
 arch/s390/Makefile           | 5 +++--
 arch/s390/purgatory/Makefile | 1 +
 2 files changed, 4 insertions(+), 2 deletions(-)
arch/s390/Makefile +3 -2
index 9c079a506325..443990791099 100644
@@ -17,12 +17,13 @@ KBUILD_CFLAGS_MODULE += -fPIC
 KBUILD_AFLAGS	+= -m64
 KBUILD_CFLAGS	+= -m64
 aflags_dwarf	:= -Wa,-gdwarf-2
-KBUILD_AFLAGS_DECOMPRESSOR := -m64 -D__ASSEMBLY__
+KBUILD_AFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -D__ASSEMBLY__
 KBUILD_AFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),$(aflags_dwarf))
-KBUILD_CFLAGS_DECOMPRESSOR := -m64 -O2
+KBUILD_CFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -O2
 KBUILD_CFLAGS_DECOMPRESSOR += -DDISABLE_BRANCH_PROFILING -D__NO_FORTIFY
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-delete-null-pointer-checks -msoft-float
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-asynchronous-unwind-tables
+KBUILD_CFLAGS_DECOMPRESSOR += $(call cc-disable-warning,pointer-sign)
 KBUILD_CFLAGS_DECOMPRESSOR += $(call cc-option,-ffreestanding)
 KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),-g)
 KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO_DWARF4), $(call cc-option, -gdwarf-4,))
arch/s390/purgatory/Makefile +1 -0
index ce6a3f75065b..ecd0b3847fef 100644
@@ -22,6 +22,7 @@ KBUILD_CFLAGS := -fno-strict-aliasing -Wall -Wstrict-prototypes
 KBUILD_CFLAGS += -Wno-pointer-sign -Wno-sign-compare
 KBUILD_CFLAGS += -fno-zero-initialized-in-bss -fno-builtin -ffreestanding
 KBUILD_CFLAGS += -c -MD -Os -m64 -msoft-float -fno-common
+KBUILD_CFLAGS += $(CLANG_FLAGS)
 KBUILD_CFLAGS += $(call cc-option,-fno-PIE)
 KBUILD_AFLAGS := $(filter-out -DCC_USING_EXPOLINE,$(KBUILD_AFLAGS))
 
Signature · 1 line
2.20.0
#2 lore raw

Re: [PATCH 2/2] s390: boot, purgatory: pass $(CLANG_FLAGS) where needed

Quoted 20 lines · Arnd Bergmann

The purgatory and boot Makefiles do not inherit the original cflags, so clang falls back to the default target architecture when building it, typically this would be x86 when cross-compiling.

Add $(CLANG_FLAGS) everywhere so we pass the correct --target=s390x-linux option when cross-compiling.

---
 arch/s390/Makefile           | 5 +++--
 arch/s390/purgatory/Makefile | 1 +
 2 files changed, 4 insertions(+), 2 deletions(-)
arch/s390/Makefile +0 -0
index 9c079a506325..443990791099 100644
@@ -17,12 +17,13 @@ KBUILD_CFLAGS_MODULE += -fPIC
 KBUILD_AFLAGS  += -m64

arch/s390/Makefile KBUILD_CFLAGS_MODULE += -fPIC · line 25

Quoted 10 lines
arch/s390/Makefile +2 -2
 KBUILD_CFLAGS  += -m64
 aflags_dwarf   := -Wa,-gdwarf-2
-KBUILD_AFLAGS_DECOMPRESSOR := -m64 -D__ASSEMBLY__
+KBUILD_AFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -D__ASSEMBLY__
 KBUILD_AFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),$(aflags_dwarf))
-KBUILD_CFLAGS_DECOMPRESSOR := -m64 -O2
+KBUILD_CFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -O2
 KBUILD_CFLAGS_DECOMPRESSOR += -DDISABLE_BRANCH_PROFILING -D__NO_FORTIFY
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-delete-null-pointer-checks -msoft-float
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-asynchronous-unwind-tables

Thanks for the respin with Nathan's suggestion.

arch/s390/Makefile KBUILD_CFLAGS_MODULE += -fPIC · line 26

Quoted 1 line
arch/s390/Makefile +1 -0
+KBUILD_CFLAGS_DECOMPRESSOR += $(call cc-disable-warning,pointer-sign)

What's up with this ^ ? Seems like the top level sets it (without cc-disable-warning :( ), but then KBUILD_CFLAGS_DECOMPRESSOR discards it. Does Clang actually flag code in this arch (that GCC doesn't)?

Quoted 18 lines
arch/s390/Makefile +0 -0
 KBUILD_CFLAGS_DECOMPRESSOR += $(call cc-option,-ffreestanding)
 KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),-g)
 KBUILD_CFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO_DWARF4), $(call cc-option, -gdwarf-4,))
arch/s390/purgatory/Makefile +1 -1
index ce6a3f75065b..ecd0b3847fef 100644
@@ -22,6 +22,7 @@ KBUILD_CFLAGS := -fno-strict-aliasing -Wall -Wstrict-prototypes
 KBUILD_CFLAGS += -Wno-pointer-sign -Wno-sign-compare
 KBUILD_CFLAGS += -fno-zero-initialized-in-bss -fno-builtin -ffreestanding
 KBUILD_CFLAGS += -c -MD -Os -m64 -msoft-float -fno-common
+KBUILD_CFLAGS += $(CLANG_FLAGS)
 KBUILD_CFLAGS += $(call cc-option,-fno-PIE)
 KBUILD_AFLAGS := $(filter-out -DCC_USING_EXPOLINE,$(KBUILD_AFLAGS))
 
--

2.20.0

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

Re: [PATCH 2/2] s390: boot, purgatory: pass $(CLANG_FLAGS) where needed

Quoted 34 lines · Nick Desaulniers' via Clang Built Linux
Quoted 20 lines · Arnd Bergmann

The purgatory and boot Makefiles do not inherit the original cflags, so clang falls back to the default target architecture when building it, typically this would be x86 when cross-compiling.

Add $(CLANG_FLAGS) everywhere so we pass the correct --target=s390x-linux option when cross-compiling.

---
 arch/s390/Makefile           | 5 +++--
 arch/s390/purgatory/Makefile | 1 +
 2 files changed, 4 insertions(+), 2 deletions(-)
arch/s390/Makefile +0 -0
index 9c079a506325..443990791099 100644
@@ -17,12 +17,13 @@ KBUILD_CFLAGS_MODULE += -fPIC
 KBUILD_AFLAGS  += -m64

arch/s390/Makefile KBUILD_CFLAGS_MODULE += -fPIC · line 25

Quoted 10 lines
arch/s390/Makefile +2 -2
 KBUILD_CFLAGS  += -m64
 aflags_dwarf   := -Wa,-gdwarf-2
-KBUILD_AFLAGS_DECOMPRESSOR := -m64 -D__ASSEMBLY__
+KBUILD_AFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -D__ASSEMBLY__
 KBUILD_AFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),$(aflags_dwarf))
-KBUILD_CFLAGS_DECOMPRESSOR := -m64 -O2
+KBUILD_CFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -O2
 KBUILD_CFLAGS_DECOMPRESSOR += -DDISABLE_BRANCH_PROFILING -D__NO_FORTIFY
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-delete-null-pointer-checks -msoft-float
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-asynchronous-unwind-tables

Thanks for the respin with Nathan's suggestion.

Quoted 5 lines

arch/s390/Makefile KBUILD_CFLAGS_MODULE += -fPIC · line 26

Quoted 1 line
arch/s390/Makefile +1 -0
+KBUILD_CFLAGS_DECOMPRESSOR += $(call cc-disable-warning,pointer-sign)

What's up with this ^ ? Seems like the top level sets it (without cc-disable-warning :( ), but then KBUILD_CFLAGS_DECOMPRESSOR discards it. Does Clang actually flag code in this arch (that GCC doesn't)?

Oops, that should have been a separate patch.

I think what happens is that clang warns more aggressively about pointer sign bugs than gcc in some cases, and some of those cases happen in s390 header files that are included by both the kernel and the decompressor.

The full warning log without this change is rather long, see
https://pastebin.com/KG9xaTNB

I also tried patching the code to avoid the warnings, but I'm not entirely happy with that result either, see
https://pastebin.com/pSMz5eZA

Arnd

#4 lore raw

Re: [PATCH 2/2] s390: boot, purgatory: pass $(CLANG_FLAGS) where needed

Quoted 43 lines · Arnd Bergmann
Quoted 34 lines · Nick Desaulniers' via Clang Built Linux
Quoted 20 lines · Arnd Bergmann

The purgatory and boot Makefiles do not inherit the original cflags, so clang falls back to the default target architecture when building it, typically this would be x86 when cross-compiling.

Add $(CLANG_FLAGS) everywhere so we pass the correct --target=s390x-linux option when cross-compiling.

---
 arch/s390/Makefile           | 5 +++--
 arch/s390/purgatory/Makefile | 1 +
 2 files changed, 4 insertions(+), 2 deletions(-)
arch/s390/Makefile +0 -0
index 9c079a506325..443990791099 100644
@@ -17,12 +17,13 @@ KBUILD_CFLAGS_MODULE += -fPIC
 KBUILD_AFLAGS  += -m64

arch/s390/Makefile KBUILD_CFLAGS_MODULE += -fPIC · line 25

Quoted 10 lines
arch/s390/Makefile +2 -2
 KBUILD_CFLAGS  += -m64
 aflags_dwarf   := -Wa,-gdwarf-2
-KBUILD_AFLAGS_DECOMPRESSOR := -m64 -D__ASSEMBLY__
+KBUILD_AFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -D__ASSEMBLY__
 KBUILD_AFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),$(aflags_dwarf))
-KBUILD_CFLAGS_DECOMPRESSOR := -m64 -O2
+KBUILD_CFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -O2
 KBUILD_CFLAGS_DECOMPRESSOR += -DDISABLE_BRANCH_PROFILING -D__NO_FORTIFY
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-delete-null-pointer-checks -msoft-float
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-asynchronous-unwind-tables

Thanks for the respin with Nathan's suggestion.

Quoted 5 lines

arch/s390/Makefile KBUILD_CFLAGS_MODULE += -fPIC · line 26

Quoted 1 line
arch/s390/Makefile +1 -0
+KBUILD_CFLAGS_DECOMPRESSOR += $(call cc-disable-warning,pointer-sign)

What's up with this ^ ? Seems like the top level sets it (without cc-disable-warning :( ), but then KBUILD_CFLAGS_DECOMPRESSOR discards it. Does Clang actually flag code in this arch (that GCC doesn't)?

Quoted 8 lines

Oops, that should have been a separate patch.

I think what happens is that clang warns more aggressively about pointer sign bugs than gcc in some cases, and some of those cases happen in s390 header files that are included by both the kernel and the decompressor.

The full warning log without this change is rather long, see
https://pastebin.com/KG9xaTNB

From this link, it looks like the definitions of:
__atomic64_or
__atomic64_and
__atomic64_xor
and their *_barrier variants are problematic. I think converting those to use unsigned long is the way to go. Shouldn't you be doing bitwise ops on unsigned types anyways?

The warnings with __atomic64_add are tougher to read/understand since at that point the log lines look like they start to mix together.

Quoted 4 lines

I also tried patching the code to avoid the warnings, but I'm not entirely happy with that result either, see
https://pastebin.com/pSMz5eZA

That's no terrible, IMO, particularly with the change I suggest above.

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

Re: [PATCH 2/2] s390: boot, purgatory: pass $(CLANG_FLAGS) where needed

Quoted 53 lines · On Thu, 11 Apr 2019 11:08:31 -0700 Nick Desaulniers <[email protected]>…
Quoted 43 lines · Arnd Bergmann
Quoted 34 lines · Nick Desaulniers' via Clang Built Linux
Quoted 20 lines · Arnd Bergmann

The purgatory and boot Makefiles do not inherit the original cflags, so clang falls back to the default target architecture when building it, typically this would be x86 when cross-compiling.

Add $(CLANG_FLAGS) everywhere so we pass the correct --target=s390x-linux option when cross-compiling.

---
 arch/s390/Makefile           | 5 +++--
 arch/s390/purgatory/Makefile | 1 +
 2 files changed, 4 insertions(+), 2 deletions(-)
arch/s390/Makefile +0 -0
index 9c079a506325..443990791099 100644
@@ -17,12 +17,13 @@ KBUILD_CFLAGS_MODULE += -fPIC
 KBUILD_AFLAGS  += -m64

arch/s390/Makefile KBUILD_CFLAGS_MODULE += -fPIC · line 25

Quoted 10 lines
arch/s390/Makefile +2 -2
 KBUILD_CFLAGS  += -m64
 aflags_dwarf   := -Wa,-gdwarf-2
-KBUILD_AFLAGS_DECOMPRESSOR := -m64 -D__ASSEMBLY__
+KBUILD_AFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -D__ASSEMBLY__
 KBUILD_AFLAGS_DECOMPRESSOR += $(if $(CONFIG_DEBUG_INFO),$(aflags_dwarf))
-KBUILD_CFLAGS_DECOMPRESSOR := -m64 -O2
+KBUILD_CFLAGS_DECOMPRESSOR := $(CLANG_FLAGS) -m64 -O2
 KBUILD_CFLAGS_DECOMPRESSOR += -DDISABLE_BRANCH_PROFILING -D__NO_FORTIFY
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-delete-null-pointer-checks -msoft-float
 KBUILD_CFLAGS_DECOMPRESSOR += -fno-asynchronous-unwind-tables  

Thanks for the respin with Nathan's suggestion.

Quoted 5 lines

arch/s390/Makefile KBUILD_CFLAGS_MODULE += -fPIC · line 26

Quoted 1 line
arch/s390/Makefile +1 -0
+KBUILD_CFLAGS_DECOMPRESSOR += $(call cc-disable-warning,pointer-sign)  

What's up with this ^ ? Seems like the top level sets it (without cc-disable-warning :( ), but then KBUILD_CFLAGS_DECOMPRESSOR discards it. Does Clang actually flag code in this arch (that GCC doesn't)?

Quoted 8 lines

Oops, that should have been a separate patch.

I think what happens is that clang warns more aggressively about pointer sign bugs than gcc in some cases, and some of those cases happen in s390 header files that are included by both the kernel and the decompressor.

The full warning log without this change is rather long, see
https://pastebin.com/KG9xaTNB

Quoted 7 lines

From this link, it looks like the definitions of:
__atomic64_or
__atomic64_and
__atomic64_xor
and their *_barrier variants are problematic. I think converting those to use unsigned long is the way to go. Shouldn't you be doing bitwise ops on unsigned types anyways?

These functions follow the type of atomic64_t which is a "long" wrapped in a structure. We do not want to change that to unsigned long, are we? Then having some of the functions operate on "long" and others on "unsigned long" seem odd.

Quoted 9 lines

The warnings with __atomic64_add are tougher to read/understand since at that point the log lines look like they start to mix together.

Quoted 4 lines

I also tried patching the code to avoid the warnings, but I'm not entirely happy with that result either, see
https://pastebin.com/pSMz5eZA

That's no terrible, IMO, particularly with the change I suggest above.

That is not too bad, the only change I do not like is the s/u8/char/ in struct ipl_block_fcp.

Signature · 4 lines
blue skies,
   Martin.

"Reality continues to ruin my life." - Calvin.
#6 lore raw
In reply to Arnd Bergmann
Quoted 2 lines · Arnd Bergmann

llvm does does not understand -march=z9-109 and older target

Please file bugs for these in LLVM's issue tracker. It might be possible to enable these additional architecture variants if they're similar to existing ones and simply unrecognized. IIRC, we had this issue with armv7 variants.

Quoted 77 lines

specifiers, so disable the respective Kconfig settings and the logic to make the boot code work on old systems when building with clang.

---
 arch/s390/Kconfig       | 6 ++++++
 arch/s390/boot/Makefile | 2 ++
 2 files changed, 8 insertions(+)
arch/s390/Kconfig +6 -0
index 8cd860cba4d1..1a2eec61196d 100644
@@ -240,6 +240,7 @@ choice
 
 config MARCH_Z900
        bool "IBM zSeries model z800 and z900"
+       depends on !CC_IS_CLANG
        select HAVE_MARCH_Z900_FEATURES
        help
          Select this to enable optimizations for model z800/z900 (2064 and
@@ -248,6 +249,7 @@ config MARCH_Z900
 
 config MARCH_Z990
        bool "IBM zSeries model z890 and z990"
+       depends on !CC_IS_CLANG
        select HAVE_MARCH_Z990_FEATURES
        help
          Select this to enable optimizations for model z890/z990 (2084 and
@@ -256,6 +258,7 @@ config MARCH_Z990
 
 config MARCH_Z9_109
        bool "IBM System z9"
+       depends on !CC_IS_CLANG
        select HAVE_MARCH_Z9_109_FEATURES
        help
          Select this to enable optimizations for IBM System z9 (2094 and
@@ -347,12 +350,15 @@ config TUNE_DEFAULT
 
 config TUNE_Z900
        bool "IBM zSeries model z800 and z900"
+       depends on !CC_IS_CLANG
 
 config TUNE_Z990
        bool "IBM zSeries model z890 and z990"
+       depends on !CC_IS_CLANG
 
 config TUNE_Z9_109
        bool "IBM System z9"
+       depends on !CC_IS_CLANG
 
 config TUNE_Z10
        bool "IBM System z10"
arch/s390/boot/Makefile +2 -1
index c844eaf24ed7..953a74d04990 100644
@@ -11,6 +11,7 @@ KASAN_SANITIZE := n
 KBUILD_AFLAGS := $(KBUILD_AFLAGS_DECOMPRESSOR)
 KBUILD_CFLAGS := $(KBUILD_CFLAGS_DECOMPRESSOR)
 
+ifndef CONFIG_CC_IS_CLANG
 #
 # Use -march=z900 for als.c to be able to print an error
 # message if the kernel is started on a machine which is too old
@@ -25,6 +26,7 @@ CFLAGS_als.o                  += -march=z900
 CFLAGS_REMOVE_sclp_early_core.o        += $(CC_FLAGS_MARCH)
 CFLAGS_sclp_early_core.o       += -march=z900
 endif
+endif
 
 CFLAGS_sclp_early_core.o += -I$(srctree)/drivers/s390/char
 
--

2.20.0

Signature · 2 lines
Thanks,
~Nick Desaulniers
#7 lore raw
Quoted 9 lines · Nick Desaulniers' via Clang Built Linux
Quoted 2 lines · Arnd Bergmann

llvm does does not understand -march=z9-109 and older target

Please file bugs for these in LLVM's issue tracker. It might be possible to enable these additional architecture variants if they're similar to existing ones and simply unrecognized. IIRC, we had this issue with armv7 variants.

I really don't see that as necessary here. While generally speaking, it would be nice to support the widest possible range of hardware, and it would not be hard to add this support, it seems highly unlikely that anyone would actually want to use that in case of s390.

The last enterprise distros to support z9 were SLES11 (end of life as of last month) and RHEL6 (ending its 10 year life next year). Anyone who is still on those releases will probably not install a brand new compiler, and anyone who is on newer releases won't run on prehistoric hardware.

Debian still runs on very old hardware in theory, but even there you'd have a hard time finding someone to test the output of the compiler on an old machine.

Arnd

#8 lore raw
Quoted 18 lines · Arnd Bergmann
Quoted 9 lines · Nick Desaulniers' via Clang Built Linux
Quoted 2 lines · Arnd Bergmann

llvm does does not understand -march=z9-109 and older target

Please file bugs for these in LLVM's issue tracker. It might be possible to enable these additional architecture variants if they're similar to existing ones and simply unrecognized. IIRC, we had this issue with armv7 variants.

I really don't see that as necessary here. While generally speaking, it would be nice to support the widest possible range of hardware, and it would not be hard to add this support, it seems highly unlikely that anyone would actually want to use that in case of s390.

Quoted 5 lines

The last enterprise distros to support z9 were SLES11 (end of life as of last month) and RHEL6 (ending its 10 year life next year). Anyone who is still on those releases will probably not install a brand new compiler, and anyone who is on newer releases won't run on prehistoric hardware.

Are those machine variants worth dropping kernel support for then?

Quoted 6 lines

Debian still runs on very old hardware in theory, but even there you'd have a hard time finding someone to test the output of the compiler on an old machine.

Arnd

Signature · 2 lines
Thanks,
~Nick Desaulniers
#9 lore raw
In reply to Arnd Bergmann
Quoted 10 lines · Arnd Bergmann

llvm does does not understand -march=z9-109 and older target specifiers, so disable the respective Kconfig settings and the logic to make the boot code work on old systems when building with clang.

---
 arch/s390/Kconfig       | 6 ++++++
 arch/s390/boot/Makefile | 2 ++
 2 files changed, 8 insertions(+)

...

Quoted 13 lines
arch/s390/boot/Makefile +1 -0
index c844eaf24ed7..953a74d04990 100644
@@ -11,6 +11,7 @@ KASAN_SANITIZE := n
 KBUILD_AFLAGS := $(KBUILD_AFLAGS_DECOMPRESSOR)
 KBUILD_CFLAGS := $(KBUILD_CFLAGS_DECOMPRESSOR)
 
+ifndef CONFIG_CC_IS_CLANG
 #
 # Use -march=z900 for als.c to be able to print an error
 # message if the kernel is started on a machine which is too old

arch/s390/boot/Makefile CFLAGS_als.o += -march=z900 · line 29

Quoted 4 lines
arch/s390/boot/Makefile +1 -0
@@ -25,6 +26,7 @@ CFLAGS_als.o			+= -march=z900
 CFLAGS_REMOVE_sclp_early_core.o	+= $(CC_FLAGS_MARCH)
 CFLAGS_sclp_early_core.o	+= -march=z900
 endif
+endif

This contradicts the whole purpose of als.c - printing an error message to the console if the kernel is compiled for a newer architecture than it is running on (and therefore uses instructions unknown to the current system).
If this can't be fixed/changed in clang, then it should be at least changed to the lowest possible architecture.

#10 lore raw
Quoted 21 lines · Heiko Carstens
Quoted 10 lines · Arnd Bergmann

llvm does does not understand -march=z9-109 and older target specifiers, so disable the respective Kconfig settings and the logic to make the boot code work on old systems when building with clang.

---
 arch/s390/Kconfig       | 6 ++++++
 arch/s390/boot/Makefile | 2 ++
 2 files changed, 8 insertions(+)

...

Quoted 8 lines
arch/s390/boot/Makefile +0 -0
index c844eaf24ed7..953a74d04990 100644
@@ -11,6 +11,7 @@ KASAN_SANITIZE := n
 KBUILD_AFLAGS := $(KBUILD_AFLAGS_DECOMPRESSOR)
 KBUILD_CFLAGS := $(KBUILD_CFLAGS_DECOMPRESSOR)
 
Quoted 16 lines

arch/s390/boot/Makefile CFLAGS_als.o += -march=z900 · line 29

Quoted 9 lines
arch/s390/boot/Makefile +2 -0
+ifndef CONFIG_CC_IS_CLANG
 #
 # Use -march=z900 for als.c to be able to print an error
 # message if the kernel is started on a machine which is too old
@@ -25,6 +26,7 @@ CFLAGS_als.o                        += -march=z900
 CFLAGS_REMOVE_sclp_early_core.o      += $(CC_FLAGS_MARCH)
 CFLAGS_sclp_early_core.o     += -march=z900
 endif
+endif

This contradicts the whole purpose of als.c - printing an error message to the console if the kernel is compiled for a newer architecture than it is running on (and therefore uses instructions unknown to the current system).
If this can't be fixed/changed in clang, then it should be at least changed to the lowest possible architecture.

Ok, I'll do that. I originally did something like it, but then went back to the simpler workaround to just make it compile.

Arnd