On Wed, May 29, 2024 at 8:28 AM Xi Ruoyao <xry111@xxxxxxxxxxx> wrote: > > On Tue, 2024-05-28 at 09:50 -1000, Tejun Heo wrote: > > (cc'ing loongarch folks) > > > > On Mon, May 27, 2024 at 04:14:10PM +0800, kernel test robot wrote: > > > tree/branch: https://git.kernel.org/pub/scm/linux/kernel/git/tj/cgroup.git for-next > > > branch HEAD: a8d55ff5f3acf52e6380976fb5d0a9172032dcb0 kernel/cgroup: cleanup cgroup_base_files when fail to add cgroup_psi_files > > > > > > Error/Warning reports: > > > > > > https://lore.kernel.org/oe-kbuild-all/202405270728.d1SabzhU-lkp@xxxxxxxxx > > > > > > Error/Warning: (recently discovered and may have been fixed) > > > > > > kernel/cgroup/pids.o: warning: objtool: __jump_table+0x0: special: can't find orig instruction > > > > > > Error/Warning ids grouped by kconfigs: > > > > > > gcc_recent_errors > > > `-- loongarch-defconfig > > > `-- kernel-cgroup-pids.o:warning:objtool:__jump_table:special:can-t-find-orig-instruction > > > > I don't know what to make of this build warning. I can't reproduce the > > problem on x86 and the referenced commit doesn't have anything special. It > > *looks* like it could be something specific to loongarch. Can you guys > > please take a look? > > For now on LoongArch objtool does not work well with jump tables. We > already have: > > ifdef CONFIG_OBJTOOL > KBUILD_CFLAGS += -fno-jump-tables > endif > > So why this doesn't stop GCC from producing a jump table? We cannot produce this warning, but currently objtool indeed cannot work with -jump-table, and we are investigate to solve it, which may take some time. Huacai > > -- > Xi Ruoyao <xry111@xxxxxxxxxxx> > School of Aerospace Science and Technology, Xidian University >
