[<prev] [next>] [day] [month] [year] [list]
Message-ID: <202512151134.zkcBSJgf-lkp@intel.com>
Date: Mon, 15 Dec 2025 11:28:17 +0800
From: kernel test robot <lkp@...el.com>
To: "Masami Hiramatsu (Google)" <mhiramat@...nel.org>
Cc: oe-kbuild-all@...ts.linux.dev, linux-kernel@...r.kernel.org,
"Steven Rostedt (Google)" <rostedt@...dmis.org>
Subject: arch/arm64/kernel/ftrace.c:203:33: sparse: sparse: cast to
restricted __le32
tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
head: 8f0b4cce4481fb22653697cced8d0d04027cb1e8
commit: 2bc56fdae1ba3fc80ee37a648346abc5f152357d ftrace: Add ftrace_get_symaddr to convert fentry_ip to symaddr
date: 12 months ago
config: arm64-randconfig-r113-20251215 (https://download.01.org/0day-ci/archive/20251215/202512151134.zkcBSJgf-lkp@intel.com/config)
compiler: aarch64-linux-gcc (GCC) 14.3.0
reproduce (this is a W=1 build): (https://download.01.org/0day-ci/archive/20251215/202512151134.zkcBSJgf-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@...el.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202512151134.zkcBSJgf-lkp@intel.com/
sparse warnings: (new ones prefixed by >>)
>> arch/arm64/kernel/ftrace.c:203:33: sparse: sparse: cast to restricted __le32
vim +203 arch/arm64/kernel/ftrace.c
145
146 /* Convert fentry_ip to the symbol address without kallsyms */
147 unsigned long arch_ftrace_get_symaddr(unsigned long fentry_ip)
148 {
149 u32 insn;
150
151 /*
152 * When using patchable-function-entry without pre-function NOPS, ftrace
153 * entry is the address of the first NOP after the function entry point.
154 *
155 * The compiler has either generated:
156 *
157 * func+00: func: NOP // To be patched to MOV X9, LR
158 * func+04: NOP // To be patched to BL <caller>
159 *
160 * Or:
161 *
162 * func-04: BTI C
163 * func+00: func: NOP // To be patched to MOV X9, LR
164 * func+04: NOP // To be patched to BL <caller>
165 *
166 * The fentry_ip is the address of `BL <caller>` which is at `func + 4`
167 * bytes in either case.
168 */
169 if (!IS_ENABLED(CONFIG_DYNAMIC_FTRACE_WITH_CALL_OPS))
170 return fentry_ip - AARCH64_INSN_SIZE;
171
172 /*
173 * When using patchable-function-entry with pre-function NOPs, BTI is
174 * a bit different.
175 *
176 * func+00: func: NOP // To be patched to MOV X9, LR
177 * func+04: NOP // To be patched to BL <caller>
178 *
179 * Or:
180 *
181 * func+00: func: BTI C
182 * func+04: NOP // To be patched to MOV X9, LR
183 * func+08: NOP // To be patched to BL <caller>
184 *
185 * The fentry_ip is the address of `BL <caller>` which is at either
186 * `func + 4` or `func + 8` depends on whether there is a BTI.
187 */
188
189 /* If there is no BTI, the func address should be one instruction before. */
190 if (!IS_ENABLED(CONFIG_ARM64_BTI_KERNEL))
191 return fentry_ip - AARCH64_INSN_SIZE;
192
193 /* We want to be extra safe in case entry ip is on the page edge,
194 * but otherwise we need to avoid get_kernel_nofault()'s overhead.
195 */
196 if ((fentry_ip & ~PAGE_MASK) < AARCH64_INSN_SIZE * 2) {
197 if (get_kernel_nofault(insn, (u32 *)(fentry_ip - AARCH64_INSN_SIZE * 2)))
198 return 0;
199 } else {
200 insn = *(u32 *)(fentry_ip - AARCH64_INSN_SIZE * 2);
201 }
202
> 203 if (aarch64_insn_is_bti(le32_to_cpu((__le32)insn)))
204 return fentry_ip - AARCH64_INSN_SIZE * 2;
205
206 return fentry_ip - AARCH64_INSN_SIZE;
207 }
208
--
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki
Powered by blists - more mailing lists