| Message ID | 4A170535.3090007@web.de (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
* Jan Kiszka (jan.kiszka@web.de) wrote: > This redundancy breaks subtly when building against recent OpenSUSE > kernels. What breaks? -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Chris Wright wrote: > * Jan Kiszka (jan.kiszka@web.de) wrote: >> This redundancy breaks subtly when building against recent OpenSUSE >> kernels. > > What breaks? When building external-module-compat.c, linux/scripts/basic/hash somehow gets called with a multi-word modname. It complains via its usage, and gcc then tries to compile files it derives from that output: gcc -Wp,-MD,/data/kvm-kmod/x86/../.external-module-compat.o.d -nostdinc -isystem /usr/lib64/gcc/x86_64-suse-linux/4.3/include -D__KERNEL__ -I/data/kvm-kmod/include -Iinclude -Iinclude2 -I/lib/modules/2.6.27.21-0.1-default/source/include -Iarch/x86/include -I/data/kvm-kmod/include-compat -include include/linux/autoconf.h -include /data/kvm-kmod/x86/external-module-compat.h -I/data/kvm-kmod/x86 -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Os -m64 -mtune=generic -mno-red-zone -mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -pipe -Wno-sign-compare -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -I/usr/src/linux-2.6.27.21-0.1/include/asm-x86/mach-default -Iinclude/asm-x86/mach-default -fno-stack-protector -fomit-frame-pointer -fasynchronous-unwind-tables -g -Wdeclaration-after-statement -Wno-pointer-sign -fwrapv -DMODULE -D"KBUILD_STR( s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(external_module_compat)" -D"DEBUG_HASH=Usage: ./scripts/basic/hash <djb2|r5> <modname>" -D"DEBUG_HASH2=Usage: ./scripts/basic/hash <djb2|r5> <modname>" -c -o /data/kvm-kmod/x86/../.tmp_external-module-compat.o /data/kvm-kmod/x86/../external-module-compat.c gcc: <djb2|r5>: Datei oder Verzeichnis nicht gefunden gcc: <modname>: Datei oder Verzeichnis nicht gefunden gcc: <djb2|r5>: Datei oder Verzeichnis nicht gefunden gcc: <modname>: Datei oder Verzeichnis nicht gefunden Jan
Jan Kiszka wrote: > This redundancy breaks subtly when building against recent OpenSUSE > kernels. > > Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> > --- > > x86/Kbuild | 4 ++-- > 1 files changed, 2 insertions(+), 2 deletions(-) > > diff --git a/x86/Kbuild b/x86/Kbuild > index e304c79..ade87fc 100644 > --- a/x86/Kbuild > +++ b/x86/Kbuild > @@ -13,7 +13,7 @@ endif > ifeq ($(CONFIG_IOMMU_API),y) > kvm-objs += iommu.o > endif > -kvm-intel-objs := vmx.o vmx-debug.o ../external-module-compat.o > -kvm-amd-objs := svm.o ../external-module-compat.o > +kvm-intel-objs := vmx.o vmx-debug.o > +kvm-amd-objs := svm.o > > IIRC, for really old kernels, some symbols are defined in external-module-compat.c and needed by kvm-$arch.ko. So we'll have to export those symbols if we drop the double link (which is a good idea regardless of the build problem).
Jan Kiszka wrote: > Chris Wright wrote: > >> * Jan Kiszka (jan.kiszka@web.de) wrote: >> >>> This redundancy breaks subtly when building against recent OpenSUSE >>> kernels. >>> >> What breaks? >> > > When building external-module-compat.c, linux/scripts/basic/hash somehow > gets called with a multi-word modname. It complains via its usage, and > gcc then tries to compile files it derives from that output: > > gcc -Wp,-MD,/data/kvm-kmod/x86/../.external-module-compat.o.d -nostdinc -isystem /usr/lib64/gcc/x86_64-suse-linux/4.3/include -D__KERNEL__ -I/data/kvm-kmod/include -Iinclude -Iinclude2 -I/lib/modules/2.6.27.21-0.1-default/source/include -Iarch/x86/include -I/data/kvm-kmod/include-compat -include include/linux/autoconf.h -include /data/kvm-kmod/x86/external-module-compat.h -I/data/kvm-kmod/x86 -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Os -m64 -mtune=generic -mno-red-zone -mcmodel=kernel -funit-at-a-time -maccumulate-outgoing-args -DCONFIG_AS_CFI=1 -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -pipe -Wno-sign-compare -mno-sse -mno-mmx -mno-sse2 -mno-3dnow -I/usr/src/linux-2.6.27.21-0.1/include/asm-x86/mach-default -Iinclude/asm-x86/mach-default -fno-stack-protector -fomit-frame-pointer -fasynchronous-unwind-tables -g -Wdeclaration-after-statement -Wno-pointer-sign -fwrapv -DMODULE -D"KBUILD_STR( > s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(external_module_compat)" -D"DEBUG_HASH=Usage: ./scripts/basic/hash <djb2|r5> <modname>" -D"DEBUG_HASH2=Usage: ./scripts/basic/hash <djb2|r5> <modname>" -c -o /data/kvm-kmod/x86/../.tmp_external-module-compat.o /data/kvm-kmod/x86/../external-module-compat.c > gcc: <djb2|r5>: Datei oder Verzeichnis nicht gefunden > gcc: <modname>: Datei oder Verzeichnis nicht gefunden > gcc: <djb2|r5>: Datei oder Verzeichnis nicht gefunden > gcc: <modname>: Datei oder Verzeichnis nicht gefunden > > Can you show the call to hash? It's more interesting than the aftermath of its failure.
Avi Kivity wrote: > Jan Kiszka wrote: >> Chris Wright wrote: >> >>> * Jan Kiszka (jan.kiszka@web.de) wrote: >>> >>>> This redundancy breaks subtly when building against recent OpenSUSE >>>> kernels. >>>> >>> What breaks? >>> >> >> When building external-module-compat.c, linux/scripts/basic/hash somehow >> gets called with a multi-word modname. It complains via its usage, and >> gcc then tries to compile files it derives from that output: >> >> gcc -Wp,-MD,/data/kvm-kmod/x86/../.external-module-compat.o.d >> -nostdinc -isystem /usr/lib64/gcc/x86_64-suse-linux/4.3/include >> -D__KERNEL__ -I/data/kvm-kmod/include -Iinclude -Iinclude2 >> -I/lib/modules/2.6.27.21-0.1-default/source/include >> -Iarch/x86/include -I/data/kvm-kmod/include-compat -include >> include/linux/autoconf.h -include >> /data/kvm-kmod/x86/external-module-compat.h -I/data/kvm-kmod/x86 >> -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing >> -fno-common -Werror-implicit-function-declaration -Os -m64 >> -mtune=generic -mno-red-zone -mcmodel=kernel -funit-at-a-time >> -maccumulate-outgoing-args -DCONFIG_AS_CFI=1 >> -DCONFIG_AS_CFI_SIGNAL_FRAME=1 -pipe -Wno-sign-compare -mno-sse >> -mno-mmx -mno-sse2 -mno-3dnow >> -I/usr/src/linux-2.6.27.21-0.1/include/asm-x86/mach-default >> -Iinclude/asm-x86/mach-default -fno-stack-protector >> -fomit-frame-pointer -fasynchronous-unwind-tables -g >> -Wdeclaration-after-statement -Wno-pointer-sign -fwrapv -DMODULE >> -D"KBUILD_STR( >> s)=#s" -D"KBUILD_BASENAME=KBUILD_STR(external_module_compat)" >> -D"DEBUG_HASH=Usage: ./scripts/basic/hash <djb2|r5> <modname>" >> -D"DEBUG_HASH2=Usage: ./scripts/basic/hash <djb2|r5> <modname>" -c -o >> /data/kvm-kmod/x86/../.tmp_external-module-compat.o >> /data/kvm-kmod/x86/../external-module-compat.c >> gcc: <djb2|r5>: Datei oder Verzeichnis nicht gefunden >> gcc: <modname>: Datei oder Verzeichnis nicht gefunden >> gcc: <djb2|r5>: Datei oder Verzeichnis nicht gefunden >> gcc: <modname>: Datei oder Verzeichnis nicht gefunden >> >> > > Can you show the call to hash? It's more interesting than the aftermath > of its failure. > strace said eg.: execve("./scripts/basic/hash", ["./scripts/basic/hash", "djb2", "/data/kvm-kmod/x86/..kvm", "kvm-amd", "kvm-intel"], ... Jan
Jan Kiszka wrote: >> Can you show the call to hash? It's more interesting than the aftermath >> of its failure. >> >> > > strace said eg.: > > execve("./scripts/basic/hash", ["./scripts/basic/hash", "djb2", > "/data/kvm-kmod/x86/..kvm", "kvm-amd", "kvm-intel"], ... > Tried to look what's going on. Never again.
diff --git a/x86/Kbuild b/x86/Kbuild index e304c79..ade87fc 100644 --- a/x86/Kbuild +++ b/x86/Kbuild @@ -13,7 +13,7 @@ endif ifeq ($(CONFIG_IOMMU_API),y) kvm-objs += iommu.o endif -kvm-intel-objs := vmx.o vmx-debug.o ../external-module-compat.o -kvm-amd-objs := svm.o ../external-module-compat.o +kvm-intel-objs := vmx.o vmx-debug.o +kvm-amd-objs := svm.o CFLAGS_kvm_main.o = -DKVM_MAIN
This redundancy breaks subtly when building against recent OpenSUSE kernels. Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com> --- x86/Kbuild | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-)