From: kernel test robot <[email protected]>
To: Palmer Dabbelt <[email protected]>
Cc: [email protected], Ammar Faizi <[email protected]>,
GNU/Weeb Mailing List <[email protected]>,
[email protected]
Subject: [ammarfaizi2-block:palmer/linux/riscv-hwprobe 4/4] arch/riscv/kernel/cpufeature.c:233:23: error: implicit declaration of function 'hartid_to_cpuid_map'
Date: Wed, 14 Sep 2022 08:52:38 +0800 [thread overview]
Message-ID: <[email protected]> (raw)
tree: https://github.com/ammarfaizi2/linux-block palmer/linux/riscv-hwprobe
head: 9be297f7ed349945cccc85f8df9d90e5ab68c1d9
commit: 9be297f7ed349945cccc85f8df9d90e5ab68c1d9 [4/4] (WIP) RISC-V: hwprobe: Support probing of misaligned accesss performance
config: riscv-randconfig-r042-20220911 (https://download.01.org/0day-ci/archive/20220914/[email protected]/config)
compiler: riscv64-linux-gcc (GCC) 12.1.0
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/ammarfaizi2/linux-block/commit/9be297f7ed349945cccc85f8df9d90e5ab68c1d9
git remote add ammarfaizi2-block https://github.com/ammarfaizi2/linux-block
git fetch --no-tags ammarfaizi2-block palmer/linux/riscv-hwprobe
git checkout 9be297f7ed349945cccc85f8df9d90e5ab68c1d9
# save the config file
mkdir build_dir && cp config build_dir/.config
COMPILER_INSTALL_PATH=$HOME/0day COMPILER=gcc-12.1.0 make.cross W=1 O=build_dir ARCH=riscv SHELL=/bin/bash arch/riscv/kernel/
If you fix the issue, kindly add following tag where applicable
Reported-by: kernel test robot <[email protected]>
All errors (new ones prefixed by >>):
arch/riscv/kernel/cpufeature.c: In function 'riscv_fill_hwcap':
>> arch/riscv/kernel/cpufeature.c:233:23: error: implicit declaration of function 'hartid_to_cpuid_map' [-Werror=implicit-function-declaration]
233 | cpu = hartid_to_cpuid_map(hartid);
| ^~~~~~~~~~~~~~~~~~~
cc1: some warnings being treated as errors
vim +/hartid_to_cpuid_map +233 arch/riscv/kernel/cpufeature.c
74
75 void __init riscv_fill_hwcap(void)
76 {
77 struct device_node *node;
78 const char *isa, *misaligned;
79 char print_str[NUM_ALPHA_EXTS + 1];
80 int i, j, rc;
81 static unsigned long isa2hwcap[256] = {0};
82 unsigned long hartid, cpu;
83
84 isa2hwcap['i'] = isa2hwcap['I'] = COMPAT_HWCAP_ISA_I;
85 isa2hwcap['m'] = isa2hwcap['M'] = COMPAT_HWCAP_ISA_M;
86 isa2hwcap['a'] = isa2hwcap['A'] = COMPAT_HWCAP_ISA_A;
87 isa2hwcap['f'] = isa2hwcap['F'] = COMPAT_HWCAP_ISA_F;
88 isa2hwcap['d'] = isa2hwcap['D'] = COMPAT_HWCAP_ISA_D;
89 isa2hwcap['c'] = isa2hwcap['C'] = COMPAT_HWCAP_ISA_C;
90
91 elf_hwcap = 0;
92
93 bitmap_zero(riscv_isa, RISCV_ISA_EXT_MAX);
94
95 for_each_of_cpu_node(node) {
96 unsigned long this_hwcap = 0;
97 DECLARE_BITMAP(this_isa, RISCV_ISA_EXT_MAX);
98 const char *temp;
99
100 rc = riscv_of_processor_hartid(node, &hartid);
101 if (rc < 0)
102 continue;
103
104 if (of_property_read_string(node, "riscv,isa", &isa)) {
105 pr_warn("Unable to find \"riscv,isa\" devicetree entry\n");
106 continue;
107 }
108
109 temp = isa;
110 #if IS_ENABLED(CONFIG_32BIT)
111 if (!strncmp(isa, "rv32", 4))
112 isa += 4;
113 #elif IS_ENABLED(CONFIG_64BIT)
114 if (!strncmp(isa, "rv64", 4))
115 isa += 4;
116 #endif
117 /* The riscv,isa DT property must start with rv64 or rv32 */
118 if (temp == isa)
119 continue;
120 bitmap_zero(this_isa, RISCV_ISA_EXT_MAX);
121 for (; *isa; ++isa) {
122 const char *ext = isa++;
123 const char *ext_end = isa;
124 bool ext_long = false, ext_err = false;
125
126 switch (*ext) {
127 case 's':
128 /**
129 * Workaround for invalid single-letter 's' & 'u'(QEMU).
130 * No need to set the bit in riscv_isa as 's' & 'u' are
131 * not valid ISA extensions. It works until multi-letter
132 * extension starting with "Su" appears.
133 */
134 if (ext[-1] != '_' && ext[1] == 'u') {
135 ++isa;
136 ext_err = true;
137 break;
138 }
139 fallthrough;
140 case 'x':
141 case 'z':
142 ext_long = true;
143 /* Multi-letter extension must be delimited */
144 for (; *isa && *isa != '_'; ++isa)
145 if (unlikely(!islower(*isa)
146 && !isdigit(*isa)))
147 ext_err = true;
148 /* Parse backwards */
149 ext_end = isa;
150 if (unlikely(ext_err))
151 break;
152 if (!isdigit(ext_end[-1]))
153 break;
154 /* Skip the minor version */
155 while (isdigit(*--ext_end))
156 ;
157 if (ext_end[0] != 'p'
158 || !isdigit(ext_end[-1])) {
159 /* Advance it to offset the pre-decrement */
160 ++ext_end;
161 break;
162 }
163 /* Skip the major version */
164 while (isdigit(*--ext_end))
165 ;
166 ++ext_end;
167 break;
168 default:
169 if (unlikely(!islower(*ext))) {
170 ext_err = true;
171 break;
172 }
173 /* Find next extension */
174 if (!isdigit(*isa))
175 break;
176 /* Skip the minor version */
177 while (isdigit(*++isa))
178 ;
179 if (*isa != 'p')
180 break;
181 if (!isdigit(*++isa)) {
182 --isa;
183 break;
184 }
185 /* Skip the major version */
186 while (isdigit(*++isa))
187 ;
188 break;
189 }
190 if (*isa != '_')
191 --isa;
192
193 #define SET_ISA_EXT_MAP(name, bit) \
194 do { \
195 if ((ext_end - ext == sizeof(name) - 1) && \
196 !memcmp(ext, name, sizeof(name) - 1)) \
197 set_bit(bit, this_isa); \
198 } while (false) \
199
200 if (unlikely(ext_err))
201 continue;
202 if (!ext_long) {
203 this_hwcap |= isa2hwcap[(unsigned char)(*ext)];
204 set_bit(*ext - 'a', this_isa);
205 } else {
206 SET_ISA_EXT_MAP("sscofpmf", RISCV_ISA_EXT_SSCOFPMF);
207 SET_ISA_EXT_MAP("svpbmt", RISCV_ISA_EXT_SVPBMT);
208 SET_ISA_EXT_MAP("zicbom", RISCV_ISA_EXT_ZICBOM);
209 SET_ISA_EXT_MAP("zihintpause", RISCV_ISA_EXT_ZIHINTPAUSE);
210 SET_ISA_EXT_MAP("sstc", RISCV_ISA_EXT_SSTC);
211 }
212 #undef SET_ISA_EXT_MAP
213 }
214
215 /*
216 * All "okay" hart should have same isa. Set HWCAP based on
217 * common capabilities of every "okay" hart, in case they don't
218 * have.
219 */
220 if (elf_hwcap)
221 elf_hwcap &= this_hwcap;
222 else
223 elf_hwcap = this_hwcap;
224
225 if (bitmap_empty(riscv_isa, RISCV_ISA_EXT_MAX))
226 bitmap_copy(riscv_isa, this_isa, RISCV_ISA_EXT_MAX);
227 else
228 bitmap_and(riscv_isa, riscv_isa, this_isa, RISCV_ISA_EXT_MAX);
229
230 /*
231 * Check for the performance of misaligned accesses.
232 */
> 233 cpu = hartid_to_cpuid_map(hartid);
234 if (cpu < 0)
235 continue;
236
237 if (of_property_read_string(node, "riscv,misaligned-access-performance", &misaligned)) {
238 if (strcmp(misaligned, "emulated") == 0)
239 per_cpu(misaligned_access_speed, cpu) = RISCV_HWPROBE_MISALIGNED_EMULATED;
240 if (strcmp(misaligned, "slow") == 0)
241 per_cpu(misaligned_access_speed, cpu) = RISCV_HWPROBE_MISALIGNED_SLOW;
242 if (strcmp(misaligned, "fast") == 0)
243 per_cpu(misaligned_access_speed, cpu) = RISCV_HWPROBE_MISALIGNED_FAST;
244 }
245 }
246
247 /* We don't support systems with F but without D, so mask those out
248 * here. */
249 if ((elf_hwcap & COMPAT_HWCAP_ISA_F) && !(elf_hwcap & COMPAT_HWCAP_ISA_D)) {
250 pr_info("This kernel does not support systems with F but not D\n");
251 elf_hwcap &= ~COMPAT_HWCAP_ISA_F;
252 }
253
254 memset(print_str, 0, sizeof(print_str));
255 for (i = 0, j = 0; i < NUM_ALPHA_EXTS; i++)
256 if (riscv_isa[0] & BIT_MASK(i))
257 print_str[j++] = (char)('a' + i);
258 pr_info("riscv: base ISA extensions %s\n", print_str);
259
260 memset(print_str, 0, sizeof(print_str));
261 for (i = 0, j = 0; i < NUM_ALPHA_EXTS; i++)
262 if (elf_hwcap & BIT_MASK(i))
263 print_str[j++] = (char)('a' + i);
264 pr_info("riscv: ELF capabilities %s\n", print_str);
265
266 for_each_set_bit(i, riscv_isa, RISCV_ISA_EXT_MAX) {
267 j = riscv_isa_ext2key(i);
268 if (j >= 0)
269 static_branch_enable(&riscv_isa_ext_keys[j]);
270 }
271 }
272
--
0-DAY CI Kernel Test Service
https://01.org/lkp
reply other threads:[~2022-09-14 0:53 UTC|newest]
Thread overview: [no followups] expand[flat|nested] mbox.gz Atom feed
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
[email protected] \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox