| Message ID | 20140605155614.GA18376@sepie.suse.cz (mailing list archive) |
|---|---|
| State | New, archived |
| Headers | show |
Dne 5.6.2014 17:56, Michal Marek napsal(a): > From d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 Mon Sep 17 00:00:00 2001 > From: Michal Marek <mmarek@suse.cz> > Date: Thu, 5 Jun 2014 17:44:44 +0200 > Subject: [PATCH] um: Fix for relative objtree when generating x86 headers Oops, this email got trapped in the mail queue on my machine. Can you please check if the patch fixes your problem? I'd like to add it to the kbuild branch and send it to Linus. Thanks, Michal > > In an O= build, rely on the generated Makefile to call the main Makefile > properly. When building in the source tree, we do not need to specify > the -C and O= either. This fixes the problem when $(objtree) is a > relative path and the -C changes the directory. > > Reported-by: Richard Weinberger <richard.weinberger@gmail.com> > Signed-off-by: Michal Marek <mmarek@suse.cz> > --- > arch/um/Makefile | 3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-) > > diff --git a/arch/um/Makefile b/arch/um/Makefile > index 36e658a..e4b1a96 100644 > --- a/arch/um/Makefile > +++ b/arch/um/Makefile > @@ -111,8 +111,7 @@ endef > KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig > > archheaders: > - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ > - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders > + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders > > archprepare: include/generated/user_constants.h > > -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 06/05/14 08:56, Michal Marek wrote: > On Wed, Jun 04, 2014 at 03:12:33PM +0200, Michal Marek wrote: >> On 2014-06-04 11:43, Michal Marek wrote: >>> On 2014-06-04 11:03, Richard Weinberger wrote: >>>> To reproduce run: >>>> make defconfig ARCH=um O=/mnt/o && make linux ARCH=um O=/mnt/ >>>> >>>> If there is anything in UML which needs fixing, please tell. :-) >>> >>> I'll have a look, thanks for the report. >> >> Findings so far: For some reason, syscalls_32.h is generated in the >> source tree (which is wrong) and syscalls_64.h is not generated at all. >> Looking further. > > Can you try the below patch? The same pattern is used in the rules for > tools/ and tools/% in the main Makefile, need to look into that as well. > But UML should work now. > > Michal > > From d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 Mon Sep 17 00:00:00 2001 > From: Michal Marek <mmarek@suse.cz> > Date: Thu, 5 Jun 2014 17:44:44 +0200 > Subject: [PATCH] um: Fix for relative objtree when generating x86 headers > > In an O= build, rely on the generated Makefile to call the main Makefile > properly. When building in the source tree, we do not need to specify > the -C and O= either. This fixes the problem when $(objtree) is a > relative path and the -C changes the directory. > > Reported-by: Richard Weinberger <richard.weinberger@gmail.com> > Signed-off-by: Michal Marek <mmarek@suse.cz> > --- > arch/um/Makefile | 3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-) > > diff --git a/arch/um/Makefile b/arch/um/Makefile > index 36e658a..e4b1a96 100644 > --- a/arch/um/Makefile > +++ b/arch/um/Makefile > @@ -111,8 +111,7 @@ endef > KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig > > archheaders: > - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ > - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders > + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders > > archprepare: include/generated/user_constants.h > > I still get this build error when building uml for i386: CC arch/x86/um/user-offsets.s ../arch/x86/um/user-offsets.c:14:29: fatal error: asm/syscalls_32.h: No such file or directory compilation terminated. make[2]: *** [arch/x86/um/user-offsets.s] Error 1
Dne 9.6.2014 23:23, Randy Dunlap napsal(a): > On 06/05/14 08:56, Michal Marek wrote: >> On Wed, Jun 04, 2014 at 03:12:33PM +0200, Michal Marek wrote: >>> On 2014-06-04 11:43, Michal Marek wrote: >>>> On 2014-06-04 11:03, Richard Weinberger wrote: >>>>> To reproduce run: >>>>> make defconfig ARCH=um O=/mnt/o && make linux ARCH=um O=/mnt/ >>>>> >>>>> If there is anything in UML which needs fixing, please tell. :-) >>>> >>>> I'll have a look, thanks for the report. >>> >>> Findings so far: For some reason, syscalls_32.h is generated in the >>> source tree (which is wrong) and syscalls_64.h is not generated at all. >>> Looking further. >> >> Can you try the below patch? The same pattern is used in the rules for >> tools/ and tools/% in the main Makefile, need to look into that as well. >> But UML should work now. >> >> Michal >> >> From d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 Mon Sep 17 00:00:00 2001 >> From: Michal Marek <mmarek@suse.cz> >> Date: Thu, 5 Jun 2014 17:44:44 +0200 >> Subject: [PATCH] um: Fix for relative objtree when generating x86 headers >> >> In an O= build, rely on the generated Makefile to call the main Makefile >> properly. When building in the source tree, we do not need to specify >> the -C and O= either. This fixes the problem when $(objtree) is a >> relative path and the -C changes the directory. >> >> Reported-by: Richard Weinberger <richard.weinberger@gmail.com> >> Signed-off-by: Michal Marek <mmarek@suse.cz> >> --- >> arch/um/Makefile | 3 +-- >> 1 file changed, 1 insertion(+), 2 deletions(-) >> >> diff --git a/arch/um/Makefile b/arch/um/Makefile >> index 36e658a..e4b1a96 100644 >> --- a/arch/um/Makefile >> +++ b/arch/um/Makefile >> @@ -111,8 +111,7 @@ endef >> KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig >> >> archheaders: >> - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ >> - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders >> + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders >> >> archprepare: include/generated/user_constants.h >> >> > > I still get this build error when building uml for i386: > > CC arch/x86/um/user-offsets.s > ../arch/x86/um/user-offsets.c:14:29: fatal error: asm/syscalls_32.h: No such file or directory > compilation terminated. > make[2]: *** [arch/x86/um/user-offsets.s] Error 1 Thanks for testing the patch. I cannot reproduce it though: make ARCH=um SUBARCH=i386 O=/dev/shm/li defconfig make ARCH=um SUBARCH=i386 O=/dev/shm/li works fine (at least it gets to the point when it starts compiling the actual kernel source). This is the previous kbuild/kbuild branch (commit 9da0763) with the above fix applied. Also, you pointed out i386 -- Does it mean that the fix worked for you on x86_64? Thanks, Michal -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 06/09/14 15:14, Michal Marek wrote: > Dne 9.6.2014 23:23, Randy Dunlap napsal(a): >> On 06/05/14 08:56, Michal Marek wrote: >>> On Wed, Jun 04, 2014 at 03:12:33PM +0200, Michal Marek wrote: >>>> On 2014-06-04 11:43, Michal Marek wrote: >>>>> On 2014-06-04 11:03, Richard Weinberger wrote: >>>>>> To reproduce run: >>>>>> make defconfig ARCH=um O=/mnt/o && make linux ARCH=um O=/mnt/ >>>>>> >>>>>> If there is anything in UML which needs fixing, please tell. :-) >>>>> >>>>> I'll have a look, thanks for the report. >>>> >>>> Findings so far: For some reason, syscalls_32.h is generated in the >>>> source tree (which is wrong) and syscalls_64.h is not generated at all. >>>> Looking further. >>> >>> Can you try the below patch? The same pattern is used in the rules for >>> tools/ and tools/% in the main Makefile, need to look into that as well. >>> But UML should work now. >>> >>> Michal >>> >>> From d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 Mon Sep 17 00:00:00 2001 >>> From: Michal Marek <mmarek@suse.cz> >>> Date: Thu, 5 Jun 2014 17:44:44 +0200 >>> Subject: [PATCH] um: Fix for relative objtree when generating x86 headers >>> >>> In an O= build, rely on the generated Makefile to call the main Makefile >>> properly. When building in the source tree, we do not need to specify >>> the -C and O= either. This fixes the problem when $(objtree) is a >>> relative path and the -C changes the directory. >>> >>> Reported-by: Richard Weinberger <richard.weinberger@gmail.com> >>> Signed-off-by: Michal Marek <mmarek@suse.cz> >>> --- >>> arch/um/Makefile | 3 +-- >>> 1 file changed, 1 insertion(+), 2 deletions(-) >>> >>> diff --git a/arch/um/Makefile b/arch/um/Makefile >>> index 36e658a..e4b1a96 100644 >>> --- a/arch/um/Makefile >>> +++ b/arch/um/Makefile >>> @@ -111,8 +111,7 @@ endef >>> KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig >>> >>> archheaders: >>> - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ >>> - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders >>> + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders >>> >>> archprepare: include/generated/user_constants.h >>> >>> >> >> I still get this build error when building uml for i386: >> >> CC arch/x86/um/user-offsets.s >> ../arch/x86/um/user-offsets.c:14:29: fatal error: asm/syscalls_32.h: No such file or directory >> compilation terminated. >> make[2]: *** [arch/x86/um/user-offsets.s] Error 1 > > Thanks for testing the patch. I cannot reproduce it though: > > make ARCH=um SUBARCH=i386 O=/dev/shm/li defconfig > make ARCH=um SUBARCH=i386 O=/dev/shm/li That doesn't work for me either. Must be a difference somewhere else. I am using linux-next of 20140606 (latest that I know of) with only your recent patch applied to it. > works fine (at least it gets to the point when it starts compiling the > actual kernel source). This is the previous kbuild/kbuild branch (commit > 9da0763) with the above fix applied. > > Also, you pointed out i386 -- Does it mean that the fix worked for you > on x86_64? Yes, it did work on x86_64 for some reason.
On Mon, Jun 09, 2014 at 03:24:11PM -0700, Randy Dunlap wrote: > On 06/09/14 15:14, Michal Marek wrote: > > Dne 9.6.2014 23:23, Randy Dunlap napsal(a): > >> On 06/05/14 08:56, Michal Marek wrote: > >>> From d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 Mon Sep 17 00:00:00 2001 > >>> From: Michal Marek <mmarek@suse.cz> > >>> Date: Thu, 5 Jun 2014 17:44:44 +0200 > >>> Subject: [PATCH] um: Fix for relative objtree when generating x86 headers > >>> > >>> In an O= build, rely on the generated Makefile to call the main Makefile > >>> properly. When building in the source tree, we do not need to specify > >>> the -C and O= either. This fixes the problem when $(objtree) is a > >>> relative path and the -C changes the directory. > >>> > >>> Reported-by: Richard Weinberger <richard.weinberger@gmail.com> > >>> Signed-off-by: Michal Marek <mmarek@suse.cz> > >>> --- > >>> arch/um/Makefile | 3 +-- > >>> 1 file changed, 1 insertion(+), 2 deletions(-) > >>> > >>> diff --git a/arch/um/Makefile b/arch/um/Makefile > >>> index 36e658a..e4b1a96 100644 > >>> --- a/arch/um/Makefile > >>> +++ b/arch/um/Makefile > >>> @@ -111,8 +111,7 @@ endef > >>> KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig > >>> > >>> archheaders: > >>> - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ > >>> - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders > >>> + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders > >>> > >>> archprepare: include/generated/user_constants.h > >>> > >>> > >> > >> I still get this build error when building uml for i386: > >> > >> CC arch/x86/um/user-offsets.s > >> ../arch/x86/um/user-offsets.c:14:29: fatal error: asm/syscalls_32.h: No such file or directory > >> compilation terminated. > >> make[2]: *** [arch/x86/um/user-offsets.s] Error 1 > > > > Thanks for testing the patch. I cannot reproduce it though: > > > > make ARCH=um SUBARCH=i386 O=/dev/shm/li defconfig > > make ARCH=um SUBARCH=i386 O=/dev/shm/li > > That doesn't work for me either. Must be a difference somewhere else. > I am using linux-next of 20140606 (latest that I know of) with only your > recent patch applied to it. I checked out next-20140606 and added the fix on top: $ git cherry-pick d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 [test cfc8100] um: Fix for relative objtree when generating x86 headers 1 file changed, 1 insertion(+), 2 deletions(-) $ git rev-parse HEAD^{tree} 8fad860b10c2abc86b1ca820a1b9f2a9267e669b The resulting tree id should be identical to yours (the commit id depends on time). It still works for me, however: $ make mrproper ... $ make ARCH=um SUBARCH=i386 O=/dev/shm/li defconfig GEN ./Makefile HOSTCC scripts/kconfig/zconf.tab.o HOSTLD scripts/kconfig/conf *** Default configuration is based on 'i386_defconfig' # # configuration written to .config # $ make ARCH=um SUBARCH=i386 O=/dev/shm/li GEN ./Makefile scripts/kconfig/conf --silentoldconfig arch/x86/um/Kconfig CHK include/generated/uapi/linux/version.h WRAP arch/um/include/generated/asm/scatterlist.h SYSTBL arch/x86/syscalls/../include/generated/asm/syscalls_32.h SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_32.h SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_64.h SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_x32.h CC arch/x86/um/user-offsets.s CHK include/generated/user_constants.h UPD include/generated/user_constants.h CHK include/config/kernel.release UPD include/config/kernel.release Using /labs/mmarek/linux-2.6 as source for kernel GEN ./Makefile CHK include/generated/utsrelease.h UPD include/generated/utsrelease.h CC kernel/bounds.s GEN include/generated/bounds.h CC arch/um/kernel/asm-offsets.s GEN include/generated/asm-offsets.h CALL /labs/mmarek/linux-2.6/scripts/checksyscalls.sh HOSTCC scripts/mod/mk_elfconfig MKELF scripts/mod/elfconfig.h CC scripts/mod/devicetable-offsets.s GEN scripts/mod/devicetable-offsets.h HOSTCC scripts/mod/file2alias.o HOSTCC scripts/mod/modpost.o HOSTCC scripts/mod/sumversion.o HOSTLD scripts/mod/modpost CC init/main.o ... and so on. Can you please post the whole make V=1 log leading to the failure? > > Also, you pointed out i386 -- Does it mean that the fix worked for you > > on x86_64? > > Yes, it did work on x86_64 for some reason. That's great. I will push the fix, since at least on scenario is confirmed to work. Thanks, Michal -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 06/09/14 15:39, Michal Marek wrote: > On Mon, Jun 09, 2014 at 03:24:11PM -0700, Randy Dunlap wrote: >> On 06/09/14 15:14, Michal Marek wrote: >>> Dne 9.6.2014 23:23, Randy Dunlap napsal(a): >>>> On 06/05/14 08:56, Michal Marek wrote: >>>>> From d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 Mon Sep 17 00:00:00 2001 >>>>> From: Michal Marek <mmarek@suse.cz> >>>>> Date: Thu, 5 Jun 2014 17:44:44 +0200 >>>>> Subject: [PATCH] um: Fix for relative objtree when generating x86 headers >>>>> >>>>> In an O= build, rely on the generated Makefile to call the main Makefile >>>>> properly. When building in the source tree, we do not need to specify >>>>> the -C and O= either. This fixes the problem when $(objtree) is a >>>>> relative path and the -C changes the directory. >>>>> >>>>> Reported-by: Richard Weinberger <richard.weinberger@gmail.com> >>>>> Signed-off-by: Michal Marek <mmarek@suse.cz> >>>>> --- >>>>> arch/um/Makefile | 3 +-- >>>>> 1 file changed, 1 insertion(+), 2 deletions(-) >>>>> >>>>> diff --git a/arch/um/Makefile b/arch/um/Makefile >>>>> index 36e658a..e4b1a96 100644 >>>>> --- a/arch/um/Makefile >>>>> +++ b/arch/um/Makefile >>>>> @@ -111,8 +111,7 @@ endef >>>>> KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig >>>>> >>>>> archheaders: >>>>> - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ >>>>> - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders >>>>> + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders >>>>> >>>>> archprepare: include/generated/user_constants.h >>>>> >>>>> >>>> >>>> I still get this build error when building uml for i386: >>>> >>>> CC arch/x86/um/user-offsets.s >>>> ../arch/x86/um/user-offsets.c:14:29: fatal error: asm/syscalls_32.h: No such file or directory >>>> compilation terminated. >>>> make[2]: *** [arch/x86/um/user-offsets.s] Error 1 >>> >>> Thanks for testing the patch. I cannot reproduce it though: >>> >>> make ARCH=um SUBARCH=i386 O=/dev/shm/li defconfig >>> make ARCH=um SUBARCH=i386 O=/dev/shm/li >> >> That doesn't work for me either. Must be a difference somewhere else. >> I am using linux-next of 20140606 (latest that I know of) with only your >> recent patch applied to it. > > I checked out next-20140606 and added the fix on top: > > $ git cherry-pick d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 > [test cfc8100] um: Fix for relative objtree when generating x86 headers > 1 file changed, 1 insertion(+), 2 deletions(-) > $ git rev-parse HEAD^{tree} > 8fad860b10c2abc86b1ca820a1b9f2a9267e669b > > The resulting tree id should be identical to yours (the commit id > depends on time). It still works for me, however: > > $ make mrproper > ... > $ make ARCH=um SUBARCH=i386 O=/dev/shm/li defconfig > GEN ./Makefile > HOSTCC scripts/kconfig/zconf.tab.o > HOSTLD scripts/kconfig/conf > *** Default configuration is based on 'i386_defconfig' > # > # configuration written to .config > # > $ make ARCH=um SUBARCH=i386 O=/dev/shm/li > GEN ./Makefile > scripts/kconfig/conf --silentoldconfig arch/x86/um/Kconfig > CHK include/generated/uapi/linux/version.h > WRAP arch/um/include/generated/asm/scatterlist.h > SYSTBL arch/x86/syscalls/../include/generated/asm/syscalls_32.h > SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_32.h > SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_64.h > SYSHDR arch/x86/syscalls/../include/generated/uapi/asm/unistd_x32.h > CC arch/x86/um/user-offsets.s > CHK include/generated/user_constants.h > UPD include/generated/user_constants.h > CHK include/config/kernel.release > UPD include/config/kernel.release > Using /labs/mmarek/linux-2.6 as source for kernel > GEN ./Makefile > CHK include/generated/utsrelease.h > UPD include/generated/utsrelease.h > CC kernel/bounds.s > GEN include/generated/bounds.h > CC arch/um/kernel/asm-offsets.s > GEN include/generated/asm-offsets.h > CALL /labs/mmarek/linux-2.6/scripts/checksyscalls.sh > HOSTCC scripts/mod/mk_elfconfig > MKELF scripts/mod/elfconfig.h > CC scripts/mod/devicetable-offsets.s > GEN scripts/mod/devicetable-offsets.h > HOSTCC scripts/mod/file2alias.o > HOSTCC scripts/mod/modpost.o > HOSTCC scripts/mod/sumversion.o > HOSTLD scripts/mod/modpost > CC init/main.o > > ... and so on. Can you please post the whole make V=1 log leading to the > failure? sure. I did mkdir /tmp/O then make ARCH=um SUBARCH=i386 O=/tmp/O defconfig make ARCH=um SUBARCH=i386 V=1 O=/tmp/O make -C /tmp/O \ KBUILD_SRC=/local/lnx/next/linux-next-20140606 \ KBUILD_EXTMOD="" -f /local/lnx/next/linux-next-20140606/Makefile \ make -f /local/lnx/next/linux-next-20140606/Makefile silentoldconfig make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.build obj=scripts/basic rm -f .tmp_quiet_recordmcount ln -fsn /local/lnx/next/linux-next-20140606 source /bin/sh /local/lnx/next/linux-next-20140606/scripts/mkmakefile \ /local/lnx/next/linux-next-20140606 . 3 15 GEN ./Makefile mkdir -p include/linux include/config make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.build obj=scripts/kconfig silentoldconfig mkdir -p include/generated scripts/kconfig/conf --silentoldconfig arch/x86/um/Kconfig set -e; : ' CHK include/generated/uapi/linux/version.h'; mkdir -p include/generated/uapi/linux/; (echo \#define LINUX_VERSION_CODE 200448; echo '#define KERNEL_VERSION(a,b,c) (((a) << 16) + ((b) << 8) + (c))';) < /local/lnx/next/linux-next-20140606/Makefile > include/generated/uapi/linux/version.h.tmp; if [ -r include/generated/uapi/linux/version.h ] && cmp -s include/generated/uapi/linux/version.h include/generated/uapi/linux/version.h.tmp; then rm -f include/generated/uapi/linux/version.h.tmp; else : ' UPD include/generated/uapi/linux/version.h'; mv -f include/generated/uapi/linux/version.h.tmp include/generated/uapi/linux/version.h; fi make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.build obj=scripts/basic (cat /dev/null; ) > scripts/basic/modules.order rm -f .tmp_quiet_recordmcount make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.asm-generic \ src=asm obj=arch/um/include/generated/asm echo "#include <asm-generic/barrier.h>" >arch/um/include/generated/asm/barrier.h echo "#include <asm-generic/bug.h>" >arch/um/include/generated/asm/bug.h echo "#include <asm-generic/clkdev.h>" >arch/um/include/generated/asm/clkdev.h echo "#include <asm-generic/cputime.h>" >arch/um/include/generated/asm/cputime.h echo "#include <asm-generic/current.h>" >arch/um/include/generated/asm/current.h echo "#include <asm-generic/delay.h>" >arch/um/include/generated/asm/delay.h echo "#include <asm-generic/device.h>" >arch/um/include/generated/asm/device.h echo "#include <asm-generic/emergency-restart.h>" >arch/um/include/generated/asm/emergency-restart.h echo "#include <asm-generic/exec.h>" >arch/um/include/generated/asm/exec.h echo "#include <asm-generic/ftrace.h>" >arch/um/include/generated/asm/ftrace.h echo "#include <asm-generic/futex.h>" >arch/um/include/generated/asm/futex.h echo "#include <asm-generic/hardirq.h>" >arch/um/include/generated/asm/hardirq.h echo "#include <asm-generic/hash.h>" >arch/um/include/generated/asm/hash.h echo "#include <asm-generic/hw_irq.h>" >arch/um/include/generated/asm/hw_irq.h echo "#include <asm-generic/io.h>" >arch/um/include/generated/asm/io.h echo "#include <asm-generic/irq_regs.h>" >arch/um/include/generated/asm/irq_regs.h echo "#include <asm-generic/kdebug.h>" >arch/um/include/generated/asm/kdebug.h echo "#include <asm-generic/mcs_spinlock.h>" >arch/um/include/generated/asm/mcs_spinlock.h echo "#include <asm-generic/mutex.h>" >arch/um/include/generated/asm/mutex.h echo "#include <asm-generic/param.h>" >arch/um/include/generated/asm/param.h echo "#include <asm-generic/pci.h>" >arch/um/include/generated/asm/pci.h echo "#include <asm-generic/percpu.h>" >arch/um/include/generated/asm/percpu.h echo "#include <asm-generic/preempt.h>" >arch/um/include/generated/asm/preempt.h echo "#include <asm-generic/scatterlist.h>" >arch/um/include/generated/asm/scatterlist.h echo "#include <asm-generic/sections.h>" >arch/um/include/generated/asm/sections.h echo "#include <asm-generic/switch_to.h>" >arch/um/include/generated/asm/switch_to.h echo "#include <asm-generic/topology.h>" >arch/um/include/generated/asm/topology.h echo "#include <asm-generic/trace_clock.h>" >arch/um/include/generated/asm/trace_clock.h echo "#include <asm-generic/xor.h>" >arch/um/include/generated/asm/xor.h make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.asm-generic \ src=uapi/asm obj=arch/um/include/generated/uapi/asm make KBUILD_SRC= ARCH=x86 archheaders make -C /local/lnx/next/linux-next-20140606 O=/tmp/O/. archheaders make -C /tmp/O \ KBUILD_SRC=/local/lnx/next/linux-next-20140606 \ KBUILD_EXTMOD="" -f /local/lnx/next/linux-next-20140606/Makefile \ archheaders make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.build obj=arch/x86/syscalls all make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.build obj=scripts build_unifdef gcc -Wp,-MD,scripts/.unifdef.d -Iscripts -Wall -Wmissing-prototypes -Wstrict-prototypes -O2 -fomit-frame-pointer -I/local/lnx/next/linux-next-20140606/tools/include -o scripts/unifdef /local/lnx/next/linux-next-20140606/scripts/unifdef.c make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.build obj=arch/x86/um arch/x86/um/user-offsets.s gcc -Wp,-MD,arch/x86/um/.user-offsets.s.d -Wall -Wundef -Wstrict-prototypes -Wno-trigraphs -fno-strict-aliasing -fno-common -Werror-implicit-function-declaration -Wno-format-security -fno-delete-null-pointer-checks -Os -Wno-maybe-uninitialized -m32 -funit-at-a-time -march=i686 -Wa,-mtune=generic32 -mpreferred-stack-boundary=2 -ffreestanding -D__arch_um__ -Dvmap=kernel_vmap -Din6addr_loopback=kernel_in6addr_loopback -Din6addr_any=kernel_in6addr_any -Dstrrchr=kernel_strrchr -D_LARGEFILE64_SOURCE -Wframe-larger-than=1024 -fno-stack-protector -Wno-unused-but-set-variable -fno-omit-frame-pointer -fno-optimize-sibling-calls -g -Wdeclaration-after-statement -Wno-pointer-sign -fno-strict-overflow -fconserve-stack -Werror=implicit-int -Werror=strict-prototypes -DCC_HAVE_ASM_GOTO -I/local/lnx/next/linux-next-20140606/arch/um/include/shared -I/local/lnx/next/linux-next-20140606/arch/x86/um/shared -I/local/lnx/next/linux-next-20140606/arch/um/include/shared/skas -D_FILE_OFFSET_BITS= 64 -idira fter include -D_GNU_SOURCE -D_LARGEFILE64_SOURCE -Iarch/x86/include/generated -fverbose-asm -S -o arch/x86/um/user-offsets.s /local/lnx/next/linux-next-20140606/arch/x86/um/user-offsets.c /local/lnx/next/linux-next-20140606/arch/x86/um/user-offsets.c:14:29: fatal error: asm/syscalls_32.h: No such file or directory compilation terminated. make[2]: *** [arch/x86/um/user-offsets.s] Error 1 make[1]: *** [arch/x86/um/user-offsets.s] Error 2 make: *** [sub-make] Error 2 > >>> Also, you pointed out i386 -- Does it mean that the fix worked for you >>> on x86_64? >> >> Yes, it did work on x86_64 for some reason. > > That's great. I will push the fix, since at least on scenario is confirmed to > work.
Hi Michael, On Thu, Jun 5, 2014 at 5:56 PM, Michal Marek <mmarek@suse.cz> wrote: > --- a/arch/um/Makefile > +++ b/arch/um/Makefile > @@ -111,8 +111,7 @@ endef > KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig > > archheaders: > - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ > - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders > + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders Should the logic in https://lkml.org/lkml/2014/6/1/17 be changed as well? Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 2014-06-10 09:40, Geert Uytterhoeven wrote: > Hi Michael, > > On Thu, Jun 5, 2014 at 5:56 PM, Michal Marek <mmarek@suse.cz> wrote: >> --- a/arch/um/Makefile >> +++ b/arch/um/Makefile >> @@ -111,8 +111,7 @@ endef >> KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig >> >> archheaders: >> - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ >> - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders >> + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders > > Should the logic in https://lkml.org/lkml/2014/6/1/17 be changed as well? I haven't tried it, but I don't see a problem with the added make invocations: + $(Q)$(MAKE) -f $(srctree)/scripts/Makefile.asm-generic \ + SRCARCH=$(HEADER_ARCH) src=asm \ + obj=arch/$(HEADER_ARCH)/include/generated/asm + $(Q)$(MAKE) -f $(srctree)/scripts/Makefile.asm-generic \ + SRCARCH=$(HEADER_ARCH) src=uapi/asm \ + obj=arch/$(HEADER_ARCH)/include/generated/uapi/asm The problem with the first make was the directory change (-C ...), which broke the relative paths. These two just spawn another make in the current directory, so it should be fine. Michal -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 2014-06-10 01:47, Randy Dunlap wrote: > sure. I did mkdir /tmp/O then > make ARCH=um SUBARCH=i386 O=/tmp/O defconfig > make ARCH=um SUBARCH=i386 V=1 O=/tmp/O So I tried with the exact same paths and it worked for me: $ mkdir -p /local/lnx/next/linux-next-20140606 $ git archive d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 | tar xf - -C /local/lnx/next/linux-next-20140606/ $ cd /local/lnx/next/linux-next-20140606/ $ make ARCH=um SUBARCH=i386 O=/tmp/O defconfig $ make ARCH=um SUBARCH=i386 V=1 O=/tmp/O and it worked. > make KBUILD_SRC= ARCH=x86 archheaders > make -C /local/lnx/next/linux-next-20140606 O=/tmp/O/. archheaders > make -C /tmp/O \ > KBUILD_SRC=/local/lnx/next/linux-next-20140606 \ > KBUILD_EXTMOD="" -f /local/lnx/next/linux-next-20140606/Makefile \ > archheaders > make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.build obj=arch/x86/syscalls all Here, it should do /bin/sh '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscallhdr.sh' '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscall_32.tbl' 'arch/x86/syscalls/../include/generated/uapi/asm/unistd_32.h' 'i386' '' '' /bin/sh '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscallhdr.sh' '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscall_64.tbl' 'arch/x86/syscalls/../include/generated/uapi/asm/unistd_64.h' 'common,64' '' '' /bin/sh '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscallhdr.sh' '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscall_64.tbl' 'arch/x86/syscalls/../include/generated/uapi/asm/unistd_x32.h' 'common,x32' '' '__X32_SYSCALL_BIT' /bin/sh '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscalltbl.sh' /local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscall_32.tbl arch/x86/syscalls/../include/generated/asm/syscalls_32.h which is not happening in your case for some reason. Can you please check if you have some stale files in include/generated in the *source* tree? git ls-file -o --directory It could be that there is some include/generated directory from a previous in-tree build. Michal -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
On 06/10/14 02:02, Michal Marek wrote: > On 2014-06-10 01:47, Randy Dunlap wrote: >> sure. I did mkdir /tmp/O then >> make ARCH=um SUBARCH=i386 O=/tmp/O defconfig >> make ARCH=um SUBARCH=i386 V=1 O=/tmp/O > > So I tried with the exact same paths and it worked for me: > $ mkdir -p /local/lnx/next/linux-next-20140606 > $ git archive d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 | tar xf - -C > /local/lnx/next/linux-next-20140606/ > $ cd /local/lnx/next/linux-next-20140606/ > $ make ARCH=um SUBARCH=i386 O=/tmp/O defconfig > $ make ARCH=um SUBARCH=i386 V=1 O=/tmp/O > > and it worked. > > >> make KBUILD_SRC= ARCH=x86 archheaders >> make -C /local/lnx/next/linux-next-20140606 O=/tmp/O/. archheaders >> make -C /tmp/O \ >> KBUILD_SRC=/local/lnx/next/linux-next-20140606 \ >> KBUILD_EXTMOD="" -f /local/lnx/next/linux-next-20140606/Makefile \ >> archheaders >> make -f /local/lnx/next/linux-next-20140606/scripts/Makefile.build obj=arch/x86/syscalls all > > Here, it should do > > /bin/sh > '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscallhdr.sh' > '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscall_32.tbl' > 'arch/x86/syscalls/../include/generated/uapi/asm/unistd_32.h' 'i386' '' '' > /bin/sh > '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscallhdr.sh' > '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscall_64.tbl' > 'arch/x86/syscalls/../include/generated/uapi/asm/unistd_64.h' > 'common,64' '' '' > /bin/sh > '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscallhdr.sh' > '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscall_64.tbl' > 'arch/x86/syscalls/../include/generated/uapi/asm/unistd_x32.h' > 'common,x32' '' '__X32_SYSCALL_BIT' > /bin/sh > '/local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscalltbl.sh' > /local/lnx/next/linux-next-20140606/arch/x86/syscalls/syscall_32.tbl > arch/x86/syscalls/../include/generated/asm/syscalls_32.h > > which is not happening in your case for some reason. Can you please > check if you have some stale files in include/generated in the *source* > tree? > > git ls-file -o --directory I'm not using git, just tarballs. I don't have an include/generated subdirectory at all in my source tree. However, today's linux-next builds OK for me (both i386 and x86_64), so let's go with your patch. > It could be that there is some include/generated directory from a > previous in-tree build.
On 2014-06-10 16:09, Randy Dunlap wrote: > I'm not using git, just tarballs. I don't have an include/generated > subdirectory at all in my source tree. > > However, today's linux-next builds OK for me (both i386 and x86_64), > so let's go with your patch. OK, then let's hope this does not reoccur. Michal -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Michal, On Thu, Jun 5, 2014 at 5:56 PM, Michal Marek <mmarek@suse.cz> wrote: > On Wed, Jun 04, 2014 at 03:12:33PM +0200, Michal Marek wrote: >> On 2014-06-04 11:43, Michal Marek wrote: >> > On 2014-06-04 11:03, Richard Weinberger wrote: >> >> To reproduce run: >> >> make defconfig ARCH=um O=/mnt/o && make linux ARCH=um O=/mnt/ >> >> >> >> If there is anything in UML which needs fixing, please tell. :-) >> > >> > I'll have a look, thanks for the report. >> >> Findings so far: For some reason, syscalls_32.h is generated in the >> source tree (which is wrong) and syscalls_64.h is not generated at all. >> Looking further. > > Can you try the below patch? The same pattern is used in the rules for > tools/ and tools/% in the main Makefile, need to look into that as well. > But UML should work now. > > Michal > > From d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 Mon Sep 17 00:00:00 2001 > From: Michal Marek <mmarek@suse.cz> > Date: Thu, 5 Jun 2014 17:44:44 +0200 > Subject: [PATCH] um: Fix for relative objtree when generating x86 headers > > In an O= build, rely on the generated Makefile to call the main Makefile > properly. When building in the source tree, we do not need to specify > the -C and O= either. This fixes the problem when $(objtree) is a > relative path and the -C changes the directory. > > Reported-by: Richard Weinberger <richard.weinberger@gmail.com> > Signed-off-by: Michal Marek <mmarek@suse.cz> > --- > arch/um/Makefile | 3 +-- > 1 file changed, 1 insertion(+), 2 deletions(-) > > diff --git a/arch/um/Makefile b/arch/um/Makefile > index 36e658a..e4b1a96 100644 > --- a/arch/um/Makefile > +++ b/arch/um/Makefile > @@ -111,8 +111,7 @@ endef > KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig > > archheaders: > - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ > - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders > + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders > > archprepare: include/generated/user_constants.h This now commit d4bc590f8716f7dd ("um: Fix for relative objtree when generating x86 headers"). When using O=, the build enters an infinite loop on v4.14-rc5. With V=1: make KBUILD_SRC= ARCH=x86 archheaders make -C /scratch/geert/linux/linux O=/scratch/geert/build/linux-uml-amd64/. archheaders make -C /scratch/geert/build/linux-uml-amd64 KBUILD_SRC=/scratch/geert/linux/linux \ -f /scratch/geert/linux/linux/Makefile archheaders make KBUILD_SRC= ARCH=x86 archheaders make -C /scratch/geert/linux/linux O=/scratch/geert/build/linux-uml-amd64/. archheaders make -C /scratch/geert/build/linux-uml-amd64 KBUILD_SRC=/scratch/geert/linux/linux \ -f /scratch/geert/linux/linux/Makefile archheaders ... As it's been a (long) while I buily UML, I started a bisect, which pointed to the commit above. The problem actually already happened on commit d4bc590f8716f7dd, so this has been broken since v3.16: make KBUILD_SRC= ARCH=x86 archheaders make -C /scratch/geert/linux/linux O=/scratch/geert/build/linux-uml-amd64/. archheaders make -C /scratch/geert/build/linux-uml-amd64 \ KBUILD_SRC=/scratch/geert/linux/linux \ KBUILD_EXTMOD="" -f /scratch/geert/linux/linux/Makefile \ archheaders make KBUILD_SRC= ARCH=x86 archheaders make -C /scratch/geert/linux/linux O=/scratch/geert/build/linux-uml-amd64/. archheaders make -C /scratch/geert/build/linux-uml-amd64 \ KBUILD_SRC=/scratch/geert/linux/linux \ KBUILD_EXTMOD="" -f /scratch/geert/linux/linux/Makefile \ archheaders ... Unfortunately reverting that commit doesn't help, as that causes the failure that commit was supposed to fix: /scratch/geert/linux/linux/arch/x86/um/user-offsets.c:19:29: fatal error: asm/syscalls_64.h: No such file or directory #include <asm/syscalls_64.h> ^ compilation terminated. /scratch/geert/linux/linux/scripts/Makefile.build:154: recipe for target 'arch/x86/um/user-offsets.s' failed Do you have any clue? Thanks! Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Michal, (new SuSE address) On Mon, Oct 16, 2017 at 12:26 PM, Geert Uytterhoeven <geert@linux-m68k.org> wrote: > On Thu, Jun 5, 2014 at 5:56 PM, Michal Marek <mmarek@suse.cz> wrote: >> On Wed, Jun 04, 2014 at 03:12:33PM +0200, Michal Marek wrote: >>> On 2014-06-04 11:43, Michal Marek wrote: >>> > On 2014-06-04 11:03, Richard Weinberger wrote: >>> >> To reproduce run: >>> >> make defconfig ARCH=um O=/mnt/o && make linux ARCH=um O=/mnt/ >>> >> >>> >> If there is anything in UML which needs fixing, please tell. :-) >>> > >>> > I'll have a look, thanks for the report. >>> >>> Findings so far: For some reason, syscalls_32.h is generated in the >>> source tree (which is wrong) and syscalls_64.h is not generated at all. >>> Looking further. >> >> Can you try the below patch? The same pattern is used in the rules for >> tools/ and tools/% in the main Makefile, need to look into that as well. >> But UML should work now. >> >> Michal >> >> From d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 Mon Sep 17 00:00:00 2001 >> From: Michal Marek <mmarek@suse.cz> >> Date: Thu, 5 Jun 2014 17:44:44 +0200 >> Subject: [PATCH] um: Fix for relative objtree when generating x86 headers >> >> In an O= build, rely on the generated Makefile to call the main Makefile >> properly. When building in the source tree, we do not need to specify >> the -C and O= either. This fixes the problem when $(objtree) is a >> relative path and the -C changes the directory. >> >> Reported-by: Richard Weinberger <richard.weinberger@gmail.com> >> Signed-off-by: Michal Marek <mmarek@suse.cz> >> --- >> arch/um/Makefile | 3 +-- >> 1 file changed, 1 insertion(+), 2 deletions(-) >> >> diff --git a/arch/um/Makefile b/arch/um/Makefile >> index 36e658a..e4b1a96 100644 >> --- a/arch/um/Makefile >> +++ b/arch/um/Makefile >> @@ -111,8 +111,7 @@ endef >> KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig >> >> archheaders: >> - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ >> - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders >> + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders >> >> archprepare: include/generated/user_constants.h > > This now commit d4bc590f8716f7dd ("um: Fix for relative objtree when > generating x86 headers"). > > When using O=, the build enters an infinite loop on v4.14-rc5. > With V=1: > > make KBUILD_SRC= ARCH=x86 archheaders > make -C /scratch/geert/linux/linux > O=/scratch/geert/build/linux-uml-amd64/. archheaders > make -C /scratch/geert/build/linux-uml-amd64 > KBUILD_SRC=/scratch/geert/linux/linux \ > -f /scratch/geert/linux/linux/Makefile archheaders > make KBUILD_SRC= ARCH=x86 archheaders > make -C /scratch/geert/linux/linux > O=/scratch/geert/build/linux-uml-amd64/. archheaders > make -C /scratch/geert/build/linux-uml-amd64 > KBUILD_SRC=/scratch/geert/linux/linux \ > -f /scratch/geert/linux/linux/Makefile archheaders > ... > > As it's been a (long) while I buily UML, I started a bisect, which pointed to > the commit above. > The problem actually already happened on commit d4bc590f8716f7dd, so this > has been broken since v3.16: > > make KBUILD_SRC= ARCH=x86 archheaders > make -C /scratch/geert/linux/linux > O=/scratch/geert/build/linux-uml-amd64/. archheaders > make -C /scratch/geert/build/linux-uml-amd64 \ > KBUILD_SRC=/scratch/geert/linux/linux \ > KBUILD_EXTMOD="" -f /scratch/geert/linux/linux/Makefile \ > archheaders > make KBUILD_SRC= ARCH=x86 archheaders > make -C /scratch/geert/linux/linux > O=/scratch/geert/build/linux-uml-amd64/. archheaders > make -C /scratch/geert/build/linux-uml-amd64 \ > KBUILD_SRC=/scratch/geert/linux/linux \ > KBUILD_EXTMOD="" -f /scratch/geert/linux/linux/Makefile \ > archheaders > ... > > Unfortunately reverting that commit doesn't help, as that causes the > failure that > commit was supposed to fix: > > /scratch/geert/linux/linux/arch/x86/um/user-offsets.c:19:29: fatal > error: asm/syscalls_64.h: No such file or directory > #include <asm/syscalls_64.h> > ^ > compilation terminated. > /scratch/geert/linux/linux/scripts/Makefile.build:154: recipe for > target 'arch/x86/um/user-offsets.s' failed > > Do you have any clue? Thanks! Gr{oetje,eeting}s, Geert -- Geert Uytterhoeven -- There's lots of Linux beyond ia32 -- geert@linux-m68k.org In personal conversations with technical people, I call myself a hacker. But when I'm talking to journalists I just say "programmer" or something like that. -- Linus Torvalds -- To unsubscribe from this list: send the line "unsubscribe linux-kbuild" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Hi Geert, (+CC Michal's new address) 2017-10-16 19:26 GMT+09:00 Geert Uytterhoeven <geert@linux-m68k.org>: > Hi Michal, > > On Thu, Jun 5, 2014 at 5:56 PM, Michal Marek <mmarek@suse.cz> wrote: >> On Wed, Jun 04, 2014 at 03:12:33PM +0200, Michal Marek wrote: >>> On 2014-06-04 11:43, Michal Marek wrote: >>> > On 2014-06-04 11:03, Richard Weinberger wrote: >>> >> To reproduce run: >>> >> make defconfig ARCH=um O=/mnt/o && make linux ARCH=um O=/mnt/ >>> >> >>> >> If there is anything in UML which needs fixing, please tell. :-) >>> > >>> > I'll have a look, thanks for the report. >>> >>> Findings so far: For some reason, syscalls_32.h is generated in the >>> source tree (which is wrong) and syscalls_64.h is not generated at all. >>> Looking further. >> >> Can you try the below patch? The same pattern is used in the rules for >> tools/ and tools/% in the main Makefile, need to look into that as well. >> But UML should work now. >> >> Michal >> >> From d4bc590f8716f7dde6b7bca319097ac30a8cb0b4 Mon Sep 17 00:00:00 2001 >> From: Michal Marek <mmarek@suse.cz> >> Date: Thu, 5 Jun 2014 17:44:44 +0200 >> Subject: [PATCH] um: Fix for relative objtree when generating x86 headers >> >> In an O= build, rely on the generated Makefile to call the main Makefile >> properly. When building in the source tree, we do not need to specify >> the -C and O= either. This fixes the problem when $(objtree) is a >> relative path and the -C changes the directory. >> >> Reported-by: Richard Weinberger <richard.weinberger@gmail.com> >> Signed-off-by: Michal Marek <mmarek@suse.cz> >> --- >> arch/um/Makefile | 3 +-- >> 1 file changed, 1 insertion(+), 2 deletions(-) >> >> diff --git a/arch/um/Makefile b/arch/um/Makefile >> index 36e658a..e4b1a96 100644 >> --- a/arch/um/Makefile >> +++ b/arch/um/Makefile >> @@ -111,8 +111,7 @@ endef >> KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig >> >> archheaders: >> - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ >> - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders >> + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders >> >> archprepare: include/generated/user_constants.h > > This now commit d4bc590f8716f7dd ("um: Fix for relative objtree when > generating x86 headers"). > > When using O=, the build enters an infinite loop on v4.14-rc5. Strange. I can not reproduce this. For example, the following command works fine for me. make ARCH=um O=foo/bar defconfig all Could you share the command you input? > With V=1: > > make KBUILD_SRC= ARCH=x86 archheaders > make -C /scratch/geert/linux/linux > O=/scratch/geert/build/linux-uml-amd64/. archheaders > make -C /scratch/geert/build/linux-uml-amd64 > KBUILD_SRC=/scratch/geert/linux/linux \ > -f /scratch/geert/linux/linux/Makefile archheaders > make KBUILD_SRC= ARCH=x86 archheaders > make -C /scratch/geert/linux/linux > O=/scratch/geert/build/linux-uml-amd64/. archheaders > make -C /scratch/geert/build/linux-uml-amd64 > KBUILD_SRC=/scratch/geert/linux/linux \ > -f /scratch/geert/linux/linux/Makefile archheaders > ... > > As it's been a (long) while I buily UML, I started a bisect, which pointed to > the commit above. > The problem actually already happened on commit d4bc590f8716f7dd, so this > has been broken since v3.16: > > make KBUILD_SRC= ARCH=x86 archheaders > make -C /scratch/geert/linux/linux > O=/scratch/geert/build/linux-uml-amd64/. archheaders > make -C /scratch/geert/build/linux-uml-amd64 \ > KBUILD_SRC=/scratch/geert/linux/linux \ > KBUILD_EXTMOD="" -f /scratch/geert/linux/linux/Makefile \ > archheaders > make KBUILD_SRC= ARCH=x86 archheaders > make -C /scratch/geert/linux/linux > O=/scratch/geert/build/linux-uml-amd64/. archheaders > make -C /scratch/geert/build/linux-uml-amd64 \ > KBUILD_SRC=/scratch/geert/linux/linux \ > KBUILD_EXTMOD="" -f /scratch/geert/linux/linux/Makefile \ > archheaders > ... > > Unfortunately reverting that commit doesn't help, as that causes the > failure that > commit was supposed to fix: > > /scratch/geert/linux/linux/arch/x86/um/user-offsets.c:19:29: fatal > error: asm/syscalls_64.h: No such file or directory > #include <asm/syscalls_64.h> > ^ > compilation terminated. > /scratch/geert/linux/linux/scripts/Makefile.build:154: recipe for > target 'arch/x86/um/user-offsets.s' failed > > Do you have any clue? Thanks! > > Gr{oetje,eeting}s,
diff --git a/arch/um/Makefile b/arch/um/Makefile index 36e658a..e4b1a96 100644 --- a/arch/um/Makefile +++ b/arch/um/Makefile @@ -111,8 +111,7 @@ endef KBUILD_KCONFIG := $(HOST_DIR)/um/Kconfig archheaders: - $(Q)$(MAKE) -C '$(srctree)' KBUILD_SRC= \ - ARCH=$(HEADER_ARCH) O='$(objtree)' archheaders + $(Q)$(MAKE) KBUILD_SRC= ARCH=$(HEADER_ARCH) archheaders archprepare: include/generated/user_constants.h