include/asm-generic/percpu.h:21:44: warning: array subscript 4 is above array bounds of 'long unsigned int[4]']
PATCH 9/23 include/asm-generic/percpu.h:21:44: warning: array subscript 4 is above array bounds of 'long unsigned int[4]']
Hi Alexander,
I'm working on top of your recent cons bitmap series, and again received a bug report from LKP regarding one of you patches. Can you please look at this report?
I tried reproduce it myself, but the build is failed for me:arch/arc/kernel/entry.S: Assembler messages:arch/arc/kernel/entry.S:280: Error: operand out of range (260 is not between -256 and 255) make[2]: *** [scripts/Makefile.build:322: arch/arc/kernel/entry-compact.o] Error 1 make[2]: *** Waiting for unfinished jobs....
If you think it's mis-attributed, we should let LKP people know that something is wrong on their side.
Thanks,
Yury
----- Forwarded message from kernel test robot <[email protected]> -----
Date: Sat, 2 Jul 2022 08:42:59 +0800
From: kernel test robot <[email protected]>
To: Yury Norov <[email protected]>
Cc: [email protected]
Subject: [norov:fns 9/23] include/asm-generic/percpu.h:21:44: warning: array subscript 4 is above array bounds of
'long unsigned int[4]'
tree: https://github.com/norov/linux fns head: 1df9509fd92a58cb37e9a419235823bbb13367de commit: 3e7e5baaaba78075a7f3a57432609e363bf2a486 [9/23] bitmap: don't assume compiler evaluates small mem*() builtins calls config: arc-randconfig-r035-20220629 (https://download.01.org/0day-ci/archive/20220702/[email protected]/config) compiler: arc-elf-gcc (GCC) 11.3.0
reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# https://github.com/norov/linux/commit/3e7e5baaaba78075a7f3a57432609e363bf2a486
git remote add norov https://github.com/norov/linux
git fetch --no-tags norov fns
git checkout 3e7e5baaaba78075a7f3a57432609e363bf2a486
# save the config file
mkdir build_dir && cp config build_dir/.config
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.3.0 make.cross W=1 O=build_dir ARCH=arc SHELL=/bin/bash kernel/
If you fix the issue, kindly add following tag where applicable
All warnings (new ones prefixed by >>):
In file included from include/linux/compiler_types.h:93,
from <command-line>:
kernel/sched/topology.c: In function 'build_sched_domains':
Quoted 1 line
Quoted 1 line
include/asm-generic/percpu.h:21:44: warning: array subscript 4 is above array bounds of 'long unsigned int[4]' [-Warray-bounds]
21 | #define per_cpu_offset(x) (__per_cpu_offset[x])
include/linux/compiler-gcc.h:35:33: note: in definition of macro 'RELOC_HIDE'
35 | (typeof(ptr)) (__ptr + (off)); \
| ^~~
include/linux/percpu-defs.h:236:9: note: in expansion of macro 'SHIFT_PERCPU_PTR'
236 | SHIFT_PERCPU_PTR((ptr), per_cpu_offset((cpu))); \
| ^~~~~~~~~~~~~~~~
include/linux/percpu-defs.h:236:33: note: in expansion of macro 'per_cpu_offset'
236 | SHIFT_PERCPU_PTR((ptr), per_cpu_offset((cpu))); \
| ^~~~~~~~~~~~~~
kernel/sched/topology.c:969:20: note: in expansion of macro 'per_cpu_ptr'
969 | sg->sgc = *per_cpu_ptr(sdd->sgc, cpu);
| ^~~~~~~~~~~
In file included from ./arch/arc/include/generated/asm/percpu.h:1,
from include/linux/irqflags.h:17,
from include/linux/rcupdate.h:26,
from include/linux/rculist.h:11,
from include/linux/sched/signal.h:5,
from include/linux/sched/cputime.h:5,
from kernel/sched/build_utility.c:13:
include/asm-generic/percpu.h:19:22: note: while referencing '__per_cpu_offset'
19 | extern unsigned long __per_cpu_offset[NR_CPUS];
| ^~~~~~~~~~~~~~~~vim +21 include/asm-generic/percpu.h
^1da177e4c3f415 Linus Torvalds 2005-04-16 20a875a69f8b00a38 Ingo Molnar 2006-07-03 @21 #define per_cpu_offset(x) (__per_cpu_offset[x])acdac87202a4081 [email protected] 2008-01-30 22 #endif acdac87202a4081 [email protected] 2008-01-30 23
:::::: The code at line 21 was first introduced by commit
:::::: a875a69f8b00a38b4f40d9632a4fc71a159f0e0d [PATCH] lockdep: add per_cpu_offset()
:::::: TO: Ingo Molnar <[email protected]>
:::::: CC: Linus Torvalds <[email protected]>
Signature · 4 lines
0-DAY CI Kernel Test Service https://01.org/lkp ----- End forwarded message -----
From: Yury Norov <[email protected]>
Date: Fri, 1 Jul 2022 19:33:24 -0700
Quoted 1 line
Hi Alexander,
Hi!
Quoted 66 lines
I'm working on top of your recent cons bitmap series, and again received a bug report from LKP regarding one of you patches. Can you please look at this report?
I tried reproduce it myself, but the build is failed for me:
arch/arc/kernel/entry.S: Assembler messages:arch/arc/kernel/entry.S:280: Error: operand out of range (260 is not between -256 and 255) make[2]: *** [scripts/Makefile.build:322:arch/arc/kernel/entry-compact.o] Error 1 make[2]: *** Waiting for unfinished jobs....If you think it's mis-attributed, we should let LKP people know that something is wrong on their side.
Thanks,
Yury----- Forwarded message from kernel test robot <[email protected]> -----
Date: Sat, 2 Jul 2022 08:42:59 +0800
From: kernel test robot <[email protected]>
To: Yury Norov <[email protected]>
Cc: [email protected]
Subject: [norov:fns 9/23]include/asm-generic/percpu.h:21:44: warning: array subscript 4 is above array bounds of
'long unsigned int[4]'tree: https://github.com/norov/linux fns head: 1df9509fd92a58cb37e9a419235823bbb13367de commit: 3e7e5baaaba78075a7f3a57432609e363bf2a486 [9/23] bitmap: don't assume compiler evaluates small mem*() builtins calls config: arc-randconfig-r035-20220629 (https://download.01.org/0day-ci/archive/20220702/[email protected]/config) compiler: arc-elf-gcc (GCC) 11.3.0reproduce (this is a W=1 build):
wget https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross -O ~/bin/make.cross chmod +x ~/bin/make.cross
# https://github.com/norov/linux/commit/3e7e5baaaba78075a7f3a57432609e363bf2a486
git remote add norov https://github.com/norov/linux
git fetch --no-tags norov fns
git checkout3e7e5baaaba78075a7f3a57432609e363bf2a486
# save the config file
mkdirbuild_dir&& cp config build_dir/.config
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-11.3.0 make.cross W=1 O=build_dir ARCH=arc SHELL=/bin/bashkernel/If you fix the issue, kindly add following tag where applicable
Reported-by kernel test robot <[email protected]>All warnings (new ones prefixed by >>):
In file included from
include/linux/compiler_types.h:93,
from <command-line>:
kernel/sched/topology.c: In function 'build_sched_domains':Quoted 1 line
Quoted 1 line
include/asm-generic/percpu.h:21:44: warning: array subscript 4 is above array bounds of 'long unsigned int[4]' [-Warray-bounds]21 | #define per_cpu_offset(x) (__per_cpu_offset[x]) include/linux/compiler-gcc.h:35:33: note: in definition of macro 'RELOC_HIDE' 35 | (typeof(ptr)) (__ptr + (off)); \ | ^~~ include/linux/percpu-defs.h:236:9: note: in expansion of macro 'SHIFT_PERCPU_PTR' 236 | SHIFT_PERCPU_PTR((ptr), per_cpu_offset((cpu))); \ | ^~~~~~~~~~~~~~~~ include/linux/percpu-defs.h:236:33: note: in expansion of macro 'per_cpu_offset' 236 | SHIFT_PERCPU_PTR((ptr), per_cpu_offset((cpu))); \ | ^~~~~~~~~~~~~~ kernel/sched/topology.c:969:20: note: in expansion of macro 'per_cpu_ptr' 969 | sg->sgc = *per_cpu_ptr(sdd->sgc, cpu); | ^~~~~~~~~~~In file included from
./arch/arc/include/generated/asm/percpu.h:1,
frominclude/linux/irqflags.h:17,
Quoted 8 lines
from
include/linux/rcupdate.h:26, frominclude/linux/rculist.h:11, frominclude/linux/sched/signal.h:5, frominclude/linux/sched/cputime.h:5, fromkernel/sched/build_utility.c:13:include/asm-generic/percpu.h:19:22: note: while referencing '__per_cpu_offset' 19 | extern unsigned long __per_cpu_offset[NR_CPUS]; | ^~~~~~~~~~~~~~~~
Ah, I've seen that one and tried to reproduce it hardly, but didn't manage to do that. Then I took a look at the code mentioned here, but again couldn't find any code pieces which would go past the array (with CPU ID higher than 3 on a 4-CPU kernel). I've heard that -Warray-bounds is still somewhat noisy and produces a fair amount of false-positives, so that it's been even disabled for 5.19-rc1 after being enabled during the merge window.
I'm super fine with letting lkp guys know that it's an unrelated stuff, whom should I write to?
BTW, I didn't receive any reports from the linux-next bots and haven't found a single over the LKML, so looks like it should be okay in general :)
Quoted 9 lines
vim +21
include/asm-generic/percpu.h^1da177e4c3f415 Linus Torvalds 2005-04-16 20
a875a69f8b00a38Ingo Molnar 2006-07-03 @21 #defineper_cpu_offset(x)(__per_cpu_offset[x])acdac87202a4081[email protected] 2008-01-30 22 #endifacdac87202a4081[email protected] 2008-01-30 23
Quoted 11 lines
:::::: The code at line 21 was first introduced by commit
::::::a875a69f8b00a38b4f40d9632a4fc71a159f0e0d[PATCH] lockdep: addper_cpu_offset():::::: TO: Ingo Molnar <[email protected]>
:::::: CC: Linus Torvalds <[email protected]>Signature · 4 lines
0-DAY CI Kernel Test Service https://01.org/lkp ----- End forwarded message -----
Thanks,
Olek