April 2018 - kernelnewbies@kernelnewbies.org - Kernelnewbies Mailing Lists

Re: make olddefconfig cmd failing
by Alex Arvelaez April 22, 2018

April 22, 2018

On Apr 21, 2018 7:11 AM, Sumit Kumar <sumit686215(a)gmail.com> wrote: > > Hi, > > I was following instructions given on this web page: https://kernelnewbies.org/OutreachyfirstpatchSetup > > I get errors when I execute make olddefconfig cmd: > > sumitsum@sumit-personal:~/Documents/linux_kernel/git/kernels/staging$ make olddefconfig > HOSTCC scripts/basic/fixdep > HOSTCC scripts/kconfig/conf.o > YACC scripts/kconfig/zconf.tab.c > /bin/sh: 1: bison: not found You need to install bison. Make sure to install all other build dependencies as well (there's a list in the kernel docs). > scripts/Makefile.lib:210: recipe for target 'scripts/kconfig/zconf.tab.c' failed > make[1]: *** [scripts/kconfig/zconf.tab.c] Error 127 > Makefile:525: recipe for target 'olddefconfig' failed > make: *** [olddefconfig] Error 2 Regards, Alex

3 3

How can I disable compile optimization in kernel for friendly debugging, Thanks
by sizel April 21, 2018

April 21, 2018

4 7

make olddefconfig cmd failing
by Sumit Kumar April 21, 2018

April 21, 2018

Hi, I was following instructions given on this web page: https://kernelnewbies.org/OutreachyfirstpatchSetup I get errors when I execute make olddefconfig cmd: sumitsum@sumit-personal:~/Documents/linux_kernel/git/kernels/staging$ make olddefconfig HOSTCC scripts/basic/fixdep HOSTCC scripts/kconfig/conf.o YACC scripts/kconfig/zconf.tab.c /bin/sh: 1: bison: not found scripts/Makefile.lib:210: recipe for target 'scripts/kconfig/zconf.tab.c' failed make[1]: *** [scripts/kconfig/zconf.tab.c] Error 127 Makefile:525: recipe for target 'olddefconfig' failed make: *** [olddefconfig] Error 2 -- Thanks and Regards, Sumit

1 0

Re: [PATCH] sched/fair: Change sched_feat(x) in !CONFIG_SCHED_DEBUG case
by Philipp Klocke April 20, 2018

April 20, 2018

On 20.04.2018 09:57, Peter Zijlstra wrote: > On Mon, Apr 16, 2018 at 10:54:26AM +0200, Philipp Klocke wrote: > >> This patch is motivated by the clang warning Wconstant-logical-operand, >> issued when logically comparing a variable to a constant integer that is >> neither 1 nor 0. It happens for sched_feat(x) when sysctl_sched_features >> is constant, i.e., CONFIG_SCHED_DEBUG is not set. >> >> kernel/sched/fair.c:3927:14: warning: use of logical '&&' with constant operand [-Wconstant-logical-operand] >> if (initial && sched_feat(START_DEBIT)) >> ^ ~~~~~~~~~~~~~~~~~~~~~~~ >> kernel/sched/fair.c:3927:14: note: use '&' for a bitwise operation >> if (initial && sched_feat(START_DEBIT)) >> ^~ >> & >> kernel/sched/fair.c:3927:14: note: remove constant to silence this warning >> if (initial && sched_feat(START_DEBIT)) >> ~^~~~~~~~~~~~~~~~~~~~~~~~~~ >> @@ -1305,7 +1305,11 @@ static const_debug __maybe_unused unsigned int sysctl_sched_features = >> 0; >> #undef SCHED_FEAT >> >> +#ifdef CONFIG_SCHED_DEBUG >> #define sched_feat(x) (sysctl_sched_features & (1UL << __SCHED_FEAT_##x)) >> +#else >> +#define sched_feat(x) ((sysctl_sched_features >> __SCHED_FEAT_##x) & 1UL) >> +#endif > So this is extra ugly, for no gain? The gain is stopping a warning that clutters the output log of clang. To improve readability, one can drop the ifdef-structure and just keep the right shift version, like Nicholas suggested. This will have a (very small) impact on performance in CONFIG_SCHED_DEBUG case, but when debugging, performance is no problem anyways. > WTH does clang complain about a constant? Can't you just disable that > stupid warning? There are 2 ways to disable the warning. Either disable it for this particular occurrence, which clutters the code with #pragma's. THIS is really ugly. Or disable it globally and maybe miss some important/helpful warnings. > Also, if sysctl_sched_features is a constant, the both expressions > _should_ really result in a constant and clang should still warn about > it. No, because clang only warns if the constant is neither 1 nor 0. (These being the 'best' integer representations of booleans) > I'm really not seeing why we'd want to do this. Just fix clang to not be > stupid. >

2 1

Re: [PATCH] sched/fair: Change sched_feat(x) in !CONFIG_SCHED_DEBUG case
by Peter Zijlstra April 20, 2018

April 20, 2018

Also, please don't cross-post with moderated lists, that's just annoying.

1 0

Re: [PATCH] sched/fair: Change sched_feat(x) in !CONFIG_SCHED_DEBUG case
by Peter Zijlstra April 20, 2018

April 20, 2018

On Mon, Apr 16, 2018 at 10:54:26AM +0200, Philipp Klocke wrote: > This patch is motivated by the clang warning Wconstant-logical-operand, > issued when logically comparing a variable to a constant integer that is > neither 1 nor 0. It happens for sched_feat(x) when sysctl_sched_features > is constant, i.e., CONFIG_SCHED_DEBUG is not set. > > kernel/sched/fair.c:3927:14: warning: use of logical '&&' with constant operand [-Wconstant-logical-operand] > if (initial && sched_feat(START_DEBIT)) > ^ ~~~~~~~~~~~~~~~~~~~~~~~ > kernel/sched/fair.c:3927:14: note: use '&' for a bitwise operation > if (initial && sched_feat(START_DEBIT)) > ^~ > & > kernel/sched/fair.c:3927:14: note: remove constant to silence this warning > if (initial && sched_feat(START_DEBIT)) > ~^~~~~~~~~~~~~~~~~~~~~~~~~~ > @@ -1305,7 +1305,11 @@ static const_debug __maybe_unused unsigned int sysctl_sched_features = > 0; > #undef SCHED_FEAT > > +#ifdef CONFIG_SCHED_DEBUG > #define sched_feat(x) (sysctl_sched_features & (1UL << __SCHED_FEAT_##x)) > +#else > +#define sched_feat(x) ((sysctl_sched_features >> __SCHED_FEAT_##x) & 1UL) > +#endif So this is extra ugly, for no gain? WTH does clang complain about a constant? Can't you just disable that stupid warning? Also, if sysctl_sched_features is a constant, the both expressions _should_ really result in a constant and clang should still warn about it. I'm really not seeing why we'd want to do this. Just fix clang to not be stupid.

1 0

Re: Building separate files for a kernel module
by Daniel. April 18, 2018

April 18, 2018

except if it's GPL too* 2018-04-18 12:57 GMT-03:00 Daniel. <danielhilst(a)gmail.com>: > AFAIK if foo is GPL, you can't link bar to it except if it's GPL to. Doing > so would be license breaking. If foo is not GPL, you are tainting your > kernel and would be hard to get help with a tainted kernel. > > 2018-04-09 14:19 GMT-03:00 Martin Galvan <omgalvan.86(a)gmail.com>: > >> 2018-04-05 12:02 GMT-03:00 Greg KH <greg(a)kroah.com>: >> > On Thu, Apr 05, 2018 at 11:49:36AM -0300, Martin Galvan wrote: >> >> I want my module to be DKMS-enabled, but since the 'bar' binaries >> >> don't use anything Linux-specific I don't want to distribute the >> >> sources for it. >> > >> > Hahaha, good luck, please discuss this with a lawyer as to the >> > legalities involved. You are on your own here. >> >> You're right, and I apologize if this practice is frowned upon. I just >> assumed it wasn't since the kbuild docs mention ways for having binary >> blobs included with kernel modules. >> >> Thanks for the answer. >> >> _______________________________________________ >> Kernelnewbies mailing list >> Kernelnewbies(a)kernelnewbies.org >> https://lists.kernelnewbies.org/mailman/listinfo/kernelnewbies >> > > > > -- > “If you're going to try, go all the way. Otherwise, don't even start. ..." > Charles Bukowski > -- “If you're going to try, go all the way. Otherwise, don't even start. ..." Charles Bukowski

1 0

Building separate files for a kernel module
by Martin Galvan April 18, 2018

April 18, 2018

Hi all, I'm trying to build a kernel module by statically linking object files with a library. My source tree looks like this: source/ ├── bar/ | ├── bar1.c | └── bar2.c └── foo.c The relevant parts of my Makefile look like this: mydriver-y := foo.o bar.a obj-m += mydriver.o bar-objs := bar/bar1.o bar/bar2.o # This builds foo.o mydriver: $(MAKE) -C $(kernel-src-dir) M=$(CURDIR) modules $(bar-objs): %.o: %.c $(MAKE) -C $(kernel-src-dir) M=$(CURDIR) $@ bar.a: $(bar-objs) ar rcs $@ $^ Here, the rule for $(bar-objs) should use the kbuild machinery to build bar1.o and bar2.o, as stated in Documentation/kbuld/modules.txt: --- 2.4 Building Separate Files It is possible to build single files that are part of a module. This works equally well for the kernel, a module, and even for external modules. Example (The module foo.ko, consist of bar.o and baz.o): make -C $KDIR M=$PWD bar.lst make -C $KDIR M=$PWD baz.o make -C $KDIR M=$PWD foo.ko make -C $KDIR M=$PWD / However, when I try to build bar.a, I get the following: make -C /lib/modules/$(uname -r)/build M=/home/martin/source/bar/bar1.o make[1]: Entering directory '/usr/src/linux-headers-4.13.0-38-generic' scripts/Makefile.build:44: /home/martin/source/bar/Makefile: No such file or directory make[2]: *** No rule to make target '/home/martin/source/bar/Makefile'. Stop. It's like kbuild expects me to have a Makefile on the 'bar' subdir as well. Am I doing something wrong? PS: Yes, I'm aware I could just add $(bar-objs) to mydriver-y and avoid building bar.a, but I really need to have those files as a separate library. Building them with a separate Makefile (i.e. without the kbuild machinery) isn't desirable either because I want to use the full array of flags and such that kbuild adds when invoking gcc.

4 5

Re: [PATCH] sched/fair: Change sched_feat(x) in !CONFIG_SCHED_DEBUG case
by Nicholas Mc Guire April 18, 2018

April 18, 2018

On Mon, Apr 16, 2018 at 10:54:26AM +0200, Philipp Klocke wrote: > Make sched_feat(x) return 1 or 0 instead of 2**x or 0 when > sysctl_sched_features is constant, by changing the left shift of the > mask-bit to a right shift of the bitmap. The behaviour of the code > remains unchanged. > We prove this by showing that the compiler can evaluate this shift > during compile time, which results in generating the same > machine code as before. > > This patch is motivated by the clang warning Wconstant-logical-operand, > issued when logically comparing a variable to a constant integer that is > neither 1 nor 0. It happens for sched_feat(x) when sysctl_sched_features > is constant, i.e., CONFIG_SCHED_DEBUG is not set. > > kernel/sched/fair.c:3927:14: warning: use of logical '&&' with constant operand [-Wconstant-logical-operand] > if (initial && sched_feat(START_DEBIT)) > ^ ~~~~~~~~~~~~~~~~~~~~~~~ > kernel/sched/fair.c:3927:14: note: use '&' for a bitwise operation > if (initial && sched_feat(START_DEBIT)) > ^~ > & > kernel/sched/fair.c:3927:14: note: remove constant to silence this warning > if (initial && sched_feat(START_DEBIT)) > ~^~~~~~~~~~~~~~~~~~~~~~~~~~ > > This resolves the warning, leaves the code base largely as is. > > Tested with gcc 7.3.1 and clang 6.0.0 on x86_64, comparing resulting > binaries with diff. > Applicable to all modern compilers and architectures > > Signed-off-by: Philipp Klocke <Phil_K97(a)gmx.de> > Suggested-by: Lukas Bulwahn <lukas.bulwahn(a)gmail.com> Reviewed-by: Nicholas Mc Guire <der.herr(a)hofr.at> > --- > kernel/sched/sched.h | 4 ++++ > 1 file changed, 4 insertions(+) > > diff --git a/kernel/sched/sched.h b/kernel/sched/sched.h > index fb5fc458547f..d2aedee6ab0f 100644 > --- a/kernel/sched/sched.h > +++ b/kernel/sched/sched.h > @@ -1305,7 +1305,11 @@ static const_debug __maybe_unused unsigned int sysctl_sched_features = > 0; > #undef SCHED_FEAT > > +#ifdef CONFIG_SCHED_DEBUG > #define sched_feat(x) (sysctl_sched_features & (1UL << __SCHED_FEAT_##x)) > +#else > +#define sched_feat(x) ((sysctl_sched_features >> __SCHED_FEAT_##x) & 1UL) > +#endif > The changed sched_feat(y) line is fine but I do not get/like the of the ifdef - keeping the change minimal is ok if there is a relevant impact but here there is no effective difference (you write the object code is the same for the !CONFIG_SCHED_DEBUG case) So I think the ifdef should be kicked here and the proposed change seems fine to me. > #endif /* SCHED_DEBUG && HAVE_JUMP_LABEL */ > > -- > 2.17.0 >

1 0

Softlockup and Hardlockup sample test module
by Ivid Suvarna April 15, 2018

April 15, 2018

Hi All, Could you please provide some sample test module for causing a Softlockup and Hardlockup in latest kernel for SMP for testing purposes. I assume there are some Kconfigs to be enabled as well for this. I had tried with the module where I put a busy loop inside spinlock but was not able to cause any lockups. Maybe this is because of SMP which schedule the job to other CPU. "How do I make a task to run on single CPU only?" I only get RCU stalls sometimes instead of softlockup when I use busy loop inside a spinlock. Is there any explanation for difference between these stalls and softlockup. Thanks, IVID

1 0