Hi Richard, At Fri, 24 Apr 2015 09:40:32 +0200, Richard Weinberger wrote: > > Hi! > > Am 19.04.2015 um 15:28 schrieb Hajime Tazaki: > > changes from v2: > > - Patch 02/11 ("slab: add private memory allocator header for arch/lib") > > * add new allocator named SLIB (Library Allocator): Patch 04/11 is > > integrated > > to 02 (commented by Christoph Lameter) > > - Overall > > * rewrite commit log messages > > > > changes from v1: > > - Patch 01/11 ("sysctl: make some functions unstatic to access by > > arch/lib"): > > * add prefix ctl_table_ to newly publiced functions (commented by Joe > > Perches) > > - Patch 08/11 ("lib: other kernel glue layer code"): > > * significantly reduce glue codes (stubs) (commented by Richard Weinberger) > > - Others > > * adapt to linux-4.0.0 > > * detect make dependency by Kbuild .cmd files > > I still fail to build it. :-( > > for-asm-upstream-v3 on top of Linus' tree gives: (snip) > arch/lib/Makefile:210: recipe for target 'arch/lib/capability.o' failed > make: *** [arch/lib/capability.o] Error 1
I'm also aware of and already fixed this issue for pre-v4 patch of libos. > And on top of v4.0 it fails too: (snip) > In file included from arch/lib/lib-socket.c:12:0: > ./include/linux/net.h:216:5: note: declared here > int sock_sendmsg(struct socket *sock, struct msghdr *msg, size_t len); > ^ > arch/lib/Makefile:210: recipe for target 'arch/lib/lib-socket.o' failed > make: *** [arch/lib/lib-socket.o] Error 1 since tag v4.0 to libos v3 patch, there is an update on the sock_sendmsg(): v3 patch already followed the change. that's why the patch can't build on top of v4.0. > You *really* need to shape up wrt the build process. at the moment, the implementation of libos can't automate to follow such changes in the build process. but good news is it's a trivial task to follow up the latest function. my observation on this manual follow up since around 3.7 kernel (2.5 yrs ago) is that these changes mostly happened during merge-window of each new version, and the fix only takes a couple of hours at maximum. I think I can survive with these changes but I'd like to ask broader opinions. one more question: I'd really like to have a suggestion on which tree I should base for libos tree. I'm proposing a patchset to arnd/asm-generic tree (which I believe the base tree for new arch/), while the patchset is tested with davem/net-next tree because right now libos is only for net/. shall I propose a patchset based on Linus' tree instead ? thank you for your feedback. -- Hajime -- To unsubscribe from this list: send the line "unsubscribe netdev" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html