* [PATCH -next] um: Use ARRAY_SIZE for array length
@ 2025-08-07 2:32 Jiapeng Chong
2025-08-07 5:08 ` Benjamin Berg
2025-08-07 14:44 ` kernel test robot
0 siblings, 2 replies; 3+ messages in thread
From: Jiapeng Chong @ 2025-08-07 2:32 UTC (permalink / raw)
To: richard
Cc: anton.ivanov, johannes, linux-um, linux-kernel, Jiapeng Chong,
Abaci Robot
Use of macro ARRAY_SIZE to calculate array size minimizes
the redundant code and improves code reusability.
./arch/um/kernel/skas/stub_exe.c:196:23-24: WARNING: Use ARRAY_SIZE.
Reported-by: Abaci Robot <abaci@linux.alibaba.com>
Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=23410
Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
---
arch/um/kernel/skas/stub_exe.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/arch/um/kernel/skas/stub_exe.c b/arch/um/kernel/skas/stub_exe.c
index cbafaa684e66..01a938a5bb4f 100644
--- a/arch/um/kernel/skas/stub_exe.c
+++ b/arch/um/kernel/skas/stub_exe.c
@@ -193,7 +193,7 @@ noinline static void real_init(void)
BPF_STMT(BPF_RET | BPF_K, SECCOMP_RET_ALLOW),
};
struct sock_fprog prog = {
- .len = sizeof(filter) / sizeof(filter[0]),
+ .len = ARRAY_SIZE(filter),
.filter = filter,
};
--
2.43.5
^ permalink raw reply related [flat|nested] 3+ messages in thread
* Re: [PATCH -next] um: Use ARRAY_SIZE for array length
2025-08-07 2:32 [PATCH -next] um: Use ARRAY_SIZE for array length Jiapeng Chong
@ 2025-08-07 5:08 ` Benjamin Berg
2025-08-07 14:44 ` kernel test robot
1 sibling, 0 replies; 3+ messages in thread
From: Benjamin Berg @ 2025-08-07 5:08 UTC (permalink / raw)
To: Jiapeng Chong, richard
Cc: anton.ivanov, johannes, linux-um, linux-kernel, Abaci Robot
Hi,
there is a good reason for this. The bug is a false positive as the
relevant include is not (and cannot be) used in the file. This patch
does not even compile.
Benjamin
On Thu, 2025-08-07 at 10:32 +0800, Jiapeng Chong wrote:
> Use of macro ARRAY_SIZE to calculate array size minimizes
> the redundant code and improves code reusability.
>
> ./arch/um/kernel/skas/stub_exe.c:196:23-24: WARNING: Use ARRAY_SIZE.
>
> Reported-by: Abaci Robot <abaci@linux.alibaba.com>
> Closes: https://bugzilla.openanolis.cn/show_bug.cgi?id=23410
> Signed-off-by: Jiapeng Chong <jiapeng.chong@linux.alibaba.com>
> ---
> arch/um/kernel/skas/stub_exe.c | 2 +-
> 1 file changed, 1 insertion(+), 1 deletion(-)
>
> diff --git a/arch/um/kernel/skas/stub_exe.c
> b/arch/um/kernel/skas/stub_exe.c
> index cbafaa684e66..01a938a5bb4f 100644
> --- a/arch/um/kernel/skas/stub_exe.c
> +++ b/arch/um/kernel/skas/stub_exe.c
> @@ -193,7 +193,7 @@ noinline static void real_init(void)
> BPF_STMT(BPF_RET | BPF_K,
> SECCOMP_RET_ALLOW),
> };
> struct sock_fprog prog = {
> - .len = sizeof(filter) / sizeof(filter[0]),
> + .len = ARRAY_SIZE(filter),
> .filter = filter,
> };
>
^ permalink raw reply [flat|nested] 3+ messages in thread
* Re: [PATCH -next] um: Use ARRAY_SIZE for array length
2025-08-07 2:32 [PATCH -next] um: Use ARRAY_SIZE for array length Jiapeng Chong
2025-08-07 5:08 ` Benjamin Berg
@ 2025-08-07 14:44 ` kernel test robot
1 sibling, 0 replies; 3+ messages in thread
From: kernel test robot @ 2025-08-07 14:44 UTC (permalink / raw)
To: Jiapeng Chong, richard
Cc: oe-kbuild-all, anton.ivanov, johannes, linux-um, linux-kernel,
Jiapeng Chong, Abaci Robot
Hi Jiapeng,
kernel test robot noticed the following build errors:
[auto build test ERROR on next-20250806]
url: https://github.com/intel-lab-lkp/linux/commits/Jiapeng-Chong/um-Use-ARRAY_SIZE-for-array-length/20250807-103534
base: next-20250806
patch link: https://lore.kernel.org/r/20250807023227.2443863-1-jiapeng.chong%40linux.alibaba.com
patch subject: [PATCH -next] um: Use ARRAY_SIZE for array length
config: um-randconfig-001-20250807 (https://download.01.org/0day-ci/archive/20250807/202508072253.pVtUjvoV-lkp@intel.com/config)
compiler: gcc-12 (Debian 12.2.0-14+deb12u1) 12.2.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20250807/202508072253.pVtUjvoV-lkp@intel.com/reproduce)
If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202508072253.pVtUjvoV-lkp@intel.com/
All errors (new ones prefixed by >>):
arch/um/kernel/skas/stub_exe.c: In function 'real_init':
>> arch/um/kernel/skas/stub_exe.c:196:32: error: implicit declaration of function 'ARRAY_SIZE' [-Werror=implicit-function-declaration]
196 | .len = ARRAY_SIZE(filter),
| ^~~~~~~~~~
cc1: some warnings being treated as errors
vim +/ARRAY_SIZE +196 arch/um/kernel/skas/stub_exe.c
12
13 noinline static void real_init(void)
14 {
15 struct stub_init_data init_data;
16 unsigned long res;
17 struct {
18 void *ss_sp;
19 int ss_flags;
20 size_t ss_size;
21 } stack = {
22 .ss_size = STUB_DATA_PAGES * UM_KERN_PAGE_SIZE,
23 };
24 struct {
25 void *sa_handler_;
26 unsigned long sa_flags;
27 void *sa_restorer;
28 unsigned long long sa_mask;
29 } sa = {
30 /* Need to set SA_RESTORER (but the handler never returns) */
31 .sa_flags = SA_ONSTACK | SA_NODEFER | SA_SIGINFO | 0x04000000,
32 };
33
34 /* set a nice name */
35 stub_syscall2(__NR_prctl, PR_SET_NAME, (unsigned long)"uml-userspace");
36
37 /* Make sure this process dies if the kernel dies */
38 stub_syscall2(__NR_prctl, PR_SET_PDEATHSIG, SIGKILL);
39
40 /* Needed in SECCOMP mode (and safe to do anyway) */
41 stub_syscall5(__NR_prctl, PR_SET_NO_NEW_PRIVS, 1, 0, 0, 0);
42
43 /* read information from STDIN and close it */
44 res = stub_syscall3(__NR_read, 0,
45 (unsigned long)&init_data, sizeof(init_data));
46 if (res != sizeof(init_data))
47 stub_syscall1(__NR_exit, 10);
48
49 /* In SECCOMP mode, FD 0 is a socket and is later used for FD passing */
50 if (!init_data.seccomp)
51 stub_syscall1(__NR_close, 0);
52 else
53 stub_syscall3(__NR_fcntl, 0, F_SETFL, O_NONBLOCK);
54
55 /* map stub code + data */
56 res = stub_syscall6(STUB_MMAP_NR,
57 init_data.stub_start, UM_KERN_PAGE_SIZE,
58 PROT_READ | PROT_EXEC, MAP_FIXED | MAP_SHARED,
59 init_data.stub_code_fd, init_data.stub_code_offset);
60 if (res != init_data.stub_start)
61 stub_syscall1(__NR_exit, 11);
62
63 res = stub_syscall6(STUB_MMAP_NR,
64 init_data.stub_start + UM_KERN_PAGE_SIZE,
65 STUB_DATA_PAGES * UM_KERN_PAGE_SIZE,
66 PROT_READ | PROT_WRITE, MAP_FIXED | MAP_SHARED,
67 init_data.stub_data_fd, init_data.stub_data_offset);
68 if (res != init_data.stub_start + UM_KERN_PAGE_SIZE)
69 stub_syscall1(__NR_exit, 12);
70
71 /* In SECCOMP mode, we only need the signalling FD from now on */
72 if (init_data.seccomp) {
73 res = stub_syscall3(__NR_close_range, 1, ~0U, 0);
74 if (res != 0)
75 stub_syscall1(__NR_exit, 13);
76 }
77
78 /* setup signal stack inside stub data */
79 stack.ss_sp = (void *)init_data.stub_start + UM_KERN_PAGE_SIZE;
80 stub_syscall2(__NR_sigaltstack, (unsigned long)&stack, 0);
81
82 /* register signal handlers */
83 sa.sa_handler_ = (void *) init_data.signal_handler;
84 sa.sa_restorer = (void *) init_data.signal_restorer;
85 if (!init_data.seccomp) {
86 /* In ptrace mode, the SIGSEGV handler never returns */
87 sa.sa_mask = 0;
88
89 res = stub_syscall4(__NR_rt_sigaction, SIGSEGV,
90 (unsigned long)&sa, 0, sizeof(sa.sa_mask));
91 if (res != 0)
92 stub_syscall1(__NR_exit, 14);
93 } else {
94 /* SECCOMP mode uses rt_sigreturn, need to mask all signals */
95 sa.sa_mask = ~0ULL;
96
97 res = stub_syscall4(__NR_rt_sigaction, SIGSEGV,
98 (unsigned long)&sa, 0, sizeof(sa.sa_mask));
99 if (res != 0)
100 stub_syscall1(__NR_exit, 15);
101
102 res = stub_syscall4(__NR_rt_sigaction, SIGSYS,
103 (unsigned long)&sa, 0, sizeof(sa.sa_mask));
104 if (res != 0)
105 stub_syscall1(__NR_exit, 16);
106
107 res = stub_syscall4(__NR_rt_sigaction, SIGALRM,
108 (unsigned long)&sa, 0, sizeof(sa.sa_mask));
109 if (res != 0)
110 stub_syscall1(__NR_exit, 17);
111
112 res = stub_syscall4(__NR_rt_sigaction, SIGTRAP,
113 (unsigned long)&sa, 0, sizeof(sa.sa_mask));
114 if (res != 0)
115 stub_syscall1(__NR_exit, 18);
116
117 res = stub_syscall4(__NR_rt_sigaction, SIGILL,
118 (unsigned long)&sa, 0, sizeof(sa.sa_mask));
119 if (res != 0)
120 stub_syscall1(__NR_exit, 19);
121
122 res = stub_syscall4(__NR_rt_sigaction, SIGFPE,
123 (unsigned long)&sa, 0, sizeof(sa.sa_mask));
124 if (res != 0)
125 stub_syscall1(__NR_exit, 20);
126 }
127
128 /*
129 * If in seccomp mode, install the SECCOMP filter and trigger a syscall.
130 * Otherwise set PTRACE_TRACEME and do a SIGSTOP.
131 */
132 if (init_data.seccomp) {
133 struct sock_filter filter[] = {
134 #if __BITS_PER_LONG > 32
135 /* [0] Load upper 32bit of instruction pointer from seccomp_data */
136 BPF_STMT(BPF_LD | BPF_W | BPF_ABS,
137 (offsetof(struct seccomp_data, instruction_pointer) + 4)),
138
139 /* [1] Jump forward 3 instructions if the upper address is not identical */
140 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, (init_data.stub_start) >> 32, 0, 3),
141 #endif
142 /* [2] Load lower 32bit of instruction pointer from seccomp_data */
143 BPF_STMT(BPF_LD | BPF_W | BPF_ABS,
144 (offsetof(struct seccomp_data, instruction_pointer))),
145
146 /* [3] Mask out lower bits */
147 BPF_STMT(BPF_ALU | BPF_AND | BPF_K, 0xfffff000),
148
149 /* [4] Jump to [6] if the lower bits are not on the expected page */
150 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, (init_data.stub_start) & 0xfffff000, 1, 0),
151
152 /* [5] Trap call, allow */
153 BPF_STMT(BPF_RET | BPF_K, SECCOMP_RET_TRAP),
154
155 /* [6,7] Check architecture */
156 BPF_STMT(BPF_LD | BPF_W | BPF_ABS,
157 offsetof(struct seccomp_data, arch)),
158 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K,
159 UM_SECCOMP_ARCH_NATIVE, 1, 0),
160
161 /* [8] Kill (for architecture check) */
162 BPF_STMT(BPF_RET | BPF_K, SECCOMP_RET_KILL_PROCESS),
163
164 /* [9] Load syscall number */
165 BPF_STMT(BPF_LD | BPF_W | BPF_ABS,
166 offsetof(struct seccomp_data, nr)),
167
168 /* [10-16] Check against permitted syscalls */
169 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, __NR_futex,
170 7, 0),
171 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K,__NR_recvmsg,
172 6, 0),
173 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K,__NR_close,
174 5, 0),
175 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, STUB_MMAP_NR,
176 4, 0),
177 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, __NR_munmap,
178 3, 0),
179 #ifdef __i386__
180 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, __NR_set_thread_area,
181 2, 0),
182 #else
183 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, __NR_arch_prctl,
184 2, 0),
185 #endif
186 BPF_JUMP(BPF_JMP | BPF_JEQ | BPF_K, __NR_rt_sigreturn,
187 1, 0),
188
189 /* [17] Not one of the permitted syscalls */
190 BPF_STMT(BPF_RET | BPF_K, SECCOMP_RET_KILL_PROCESS),
191
192 /* [18] Permitted call for the stub */
193 BPF_STMT(BPF_RET | BPF_K, SECCOMP_RET_ALLOW),
194 };
195 struct sock_fprog prog = {
> 196 .len = ARRAY_SIZE(filter),
197 .filter = filter,
198 };
199
200 if (stub_syscall3(__NR_seccomp, SECCOMP_SET_MODE_FILTER,
201 SECCOMP_FILTER_FLAG_TSYNC,
202 (unsigned long)&prog) != 0)
203 stub_syscall1(__NR_exit, 21);
204
205 /* Fall through, the exit syscall will cause SIGSYS */
206 } else {
207 stub_syscall4(__NR_ptrace, PTRACE_TRACEME, 0, 0, 0);
208
209 stub_syscall2(__NR_kill, stub_syscall0(__NR_getpid), SIGSTOP);
210 }
211
212 stub_syscall1(__NR_exit, 30);
213
214 __builtin_unreachable();
215 }
216
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
^ permalink raw reply [flat|nested] 3+ messages in thread
end of thread, other threads:[~2025-08-07 14:44 UTC | newest]
Thread overview: 3+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2025-08-07 2:32 [PATCH -next] um: Use ARRAY_SIZE for array length Jiapeng Chong
2025-08-07 5:08 ` Benjamin Berg
2025-08-07 14:44 ` kernel test robot
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).