On Wed, Oct 07 2015, kbuild test robot <fengguang.wu@xxxxxxxxx> wrote: > tree: https://git.kernel.org/pub/scm/linux/kernel/git/next/linux-next.git master > head: 15e5b1b566ed0689093effb31f13fe29e4c9905a > commit: 5196456ab776a5974d24994aea1b20afe0c8d020 [5871/6146] slab.h: sprinkle __assume_aligned attributes > config: sparc64-defconfig (attached as .config) > reproduce: > wget https://git.kernel.org/cgit/linux/kernel/git/wfg/lkp-tests.git/plain/sbin/make.cross -O ~/bin/make.cross > chmod +x ~/bin/make.cross > git checkout 5196456ab776a5974d24994aea1b20afe0c8d020 > # save the attached .config to linux build tree > make.cross ARCH=sparc64 > > Note: it may well be a FALSE warning. FWIW you are at least aware of > it now. I'm pretty sure it is a false warning, but just for the record (and so someone can prove me wrong), here's my analysis. Relevant code (drivers/net/ethernet/broadcom/tg3.c): 12030 if ((b_offset = (offset & 3))) { 12031 /* adjustments to start on required 4 byte boundary */ 12032 ret = tg3_nvram_read_be32(tp, offset-b_offset, &start); 12033 if (ret) 12034 return ret; 12035 len += b_offset; 12036 offset &= ~3; 12037 if (len < 4) 12038 len = 4; 12039 } 12052 if (b_offset || odd_len) { 12053 buf = kmalloc(len, GFP_KERNEL); 12054 if (!buf) 12055 return -ENOMEM; 12056 if (b_offset) 12057 memcpy(buf, &start, 4); 12058 if (odd_len) 12059 memcpy(buf+len-4, &end, 4); 12060 memcpy(buf + b_offset, data, eeprom->len); 12061 } So first off, the real problematic line would be 12057. I'm guessing that the now known alignment of buf made gcc replace the memcpy call with effectively "*(__be32*)buf = start;", which then triggered the warning. But the memcpy is guarded by b_offset != 0, and in that case we know start is initialized by tg3_nvram_read_be32 (or that we'd taken the early return). Rasmus > http://gcc.gnu.org/wiki/Better_Uninitialized_Warnings > > All warnings (new ones prefixed by >>): > > In file included from arch/sparc/include/asm/string.h:4:0, > from include/linux/string.h:17, > from include/linux/dynamic_debug.h:111, > from include/linux/printk.h:277, > from include/linux/kernel.h:13, > from include/linux/list.h:8, > from include/linux/module.h:9, > from drivers/net/ethernet/broadcom/tg3.c:19: > drivers/net/ethernet/broadcom/tg3.c: In function 'tg3_set_eeprom': >>> arch/sparc/include/asm/string_64.h:25:25: warning: 'start' may be used uninitialized in this function [-Wmaybe-uninitialized] > #define memcpy(t, f, n) __builtin_memcpy(t, f, n) > ^ > drivers/net/ethernet/broadcom/tg3.c:12021:9: note: 'start' was declared here > __be32 start, end; > ^ > > vim +/start +25 arch/sparc/include/asm/string_64.h > > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 9 #ifndef __SPARC64_STRING_H__ > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 10 #define __SPARC64_STRING_H__ > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 11 > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 12 /* Really, userland/ksyms should not see any of this stuff. */ > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 13 > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 14 #ifdef __KERNEL__ > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 15 > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 16 #include <asm/asi.h> > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 17 > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 18 #ifndef EXPORT_SYMTAB_STROPS > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 19 > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 20 /* First the mem*() things. */ > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 21 #define __HAVE_ARCH_MEMMOVE > f05a6865 arch/sparc/include/asm/string_64.h Sam Ravnborg 2014-05-16 22 void *memmove(void *, const void *, __kernel_size_t); > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 23 > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 24 #define __HAVE_ARCH_MEMCPY > 4d14a459 arch/sparc/include/asm/string_64.h David S. Miller 2009-12-10 @25 #define memcpy(t, f, n) __builtin_memcpy(t, f, n) > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 26 > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 27 #define __HAVE_ARCH_MEMSET > 4d14a459 arch/sparc/include/asm/string_64.h David S. Miller 2009-12-10 28 #define memset(s, c, count) __builtin_memset(s, c, count) > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 29 > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 30 #define __HAVE_ARCH_MEMSCAN > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 31 > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 32 #undef memscan > f5e706ad include/asm-sparc/string_64.h Sam Ravnborg 2008-07-17 33 #define memscan(__arg0, __char, __arg2) \ > > :::::: The code at line 25 was first introduced by commit > :::::: 4d14a459857bd151ecbd14bcd37b4628da00792b sparc: Stop trying to be so fancy and use __builtin_{memcpy,memset}() > > :::::: TO: David S. Miller <davem@xxxxxxxxxxxxx> > :::::: CC: David S. Miller <davem@xxxxxxxxxxxxx> > > --- > 0-DAY kernel test infrastructure Open Source Technology Center > https://lists.01.org/pipermail/kbuild-all Intel Corporation -- To unsubscribe, send a message with 'unsubscribe linux-mm' in the body to majordomo@xxxxxxxxx. For more info on Linux MM, see: http://www.linux-mm.org/ . Don't email: <a href=mailto:"dont@xxxxxxxxx";> email@xxxxxxxxx </a>