* [kbuild] arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
@ 2022-02-04 10:37 ` Dan Carpenter
0 siblings, 0 replies; 11+ messages in thread
From: Dan Carpenter @ 2022-02-04 10:37 UTC (permalink / raw)
To: kbuild, Christophe Leroy; +Cc: lkp, kbuild-all, linux-kernel, Michael Ellerman
tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
head: 88808fbbead481aedb46640a5ace69c58287f56a
commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
config: powerpc-randconfig-m031-20220130 (https://download.01.org/0day-ci/archive/20220204/202202040124.7MMKaXXV-lkp@intel.com/config )
compiler: powerpc-linux-gcc (GCC) 11.2.0
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
smatch warnings:
arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
vim +/pass +225 arch/powerpc/net/bpf_jit_comp.c
4ea76e90a97d22 Christophe Leroy 2021-03-22 88 struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
4ea76e90a97d22 Christophe Leroy 2021-03-22 89 {
4ea76e90a97d22 Christophe Leroy 2021-03-22 90 u32 proglen;
4ea76e90a97d22 Christophe Leroy 2021-03-22 91 u32 alloclen;
4ea76e90a97d22 Christophe Leroy 2021-03-22 92 u8 *image = NULL;
4ea76e90a97d22 Christophe Leroy 2021-03-22 93 u32 *code_base;
4ea76e90a97d22 Christophe Leroy 2021-03-22 94 u32 *addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 95 struct powerpc64_jit_data *jit_data;
4ea76e90a97d22 Christophe Leroy 2021-03-22 96 struct codegen_context cgctx;
4ea76e90a97d22 Christophe Leroy 2021-03-22 97 int pass;
^^^^^^^^
4ea76e90a97d22 Christophe Leroy 2021-03-22 98 int flen;
4ea76e90a97d22 Christophe Leroy 2021-03-22 99 struct bpf_binary_header *bpf_hdr;
4ea76e90a97d22 Christophe Leroy 2021-03-22 100 struct bpf_prog *org_fp = fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 101 struct bpf_prog *tmp_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 102 bool bpf_blinded = false;
4ea76e90a97d22 Christophe Leroy 2021-03-22 103 bool extra_pass = false;
4ea76e90a97d22 Christophe Leroy 2021-03-22 104
4ea76e90a97d22 Christophe Leroy 2021-03-22 105 if (!fp->jit_requested)
4ea76e90a97d22 Christophe Leroy 2021-03-22 106 return org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 107
4ea76e90a97d22 Christophe Leroy 2021-03-22 108 tmp_fp = bpf_jit_blind_constants(org_fp);
4ea76e90a97d22 Christophe Leroy 2021-03-22 109 if (IS_ERR(tmp_fp))
4ea76e90a97d22 Christophe Leroy 2021-03-22 110 return org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 111
4ea76e90a97d22 Christophe Leroy 2021-03-22 112 if (tmp_fp != org_fp) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 113 bpf_blinded = true;
4ea76e90a97d22 Christophe Leroy 2021-03-22 114 fp = tmp_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 115 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 116
4ea76e90a97d22 Christophe Leroy 2021-03-22 117 jit_data = fp->aux->jit_data;
4ea76e90a97d22 Christophe Leroy 2021-03-22 118 if (!jit_data) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 119 jit_data = kzalloc(sizeof(*jit_data), GFP_KERNEL);
4ea76e90a97d22 Christophe Leroy 2021-03-22 120 if (!jit_data) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 121 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 122 goto out;
4ea76e90a97d22 Christophe Leroy 2021-03-22 123 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 124 fp->aux->jit_data = jit_data;
4ea76e90a97d22 Christophe Leroy 2021-03-22 125 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 126
4ea76e90a97d22 Christophe Leroy 2021-03-22 127 flen = fp->len;
4ea76e90a97d22 Christophe Leroy 2021-03-22 128 addrs = jit_data->addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 129 if (addrs) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 130 cgctx = jit_data->ctx;
4ea76e90a97d22 Christophe Leroy 2021-03-22 131 image = jit_data->image;
4ea76e90a97d22 Christophe Leroy 2021-03-22 132 bpf_hdr = jit_data->header;
4ea76e90a97d22 Christophe Leroy 2021-03-22 133 proglen = jit_data->proglen;
4ea76e90a97d22 Christophe Leroy 2021-03-22 134 alloclen = proglen + FUNCTION_DESCR_SIZE;
4ea76e90a97d22 Christophe Leroy 2021-03-22 135 extra_pass = true;
4ea76e90a97d22 Christophe Leroy 2021-03-22 136 goto skip_init_ctx;
Assume we hit this goto
4ea76e90a97d22 Christophe Leroy 2021-03-22 137 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 138
4ea76e90a97d22 Christophe Leroy 2021-03-22 139 addrs = kcalloc(flen + 1, sizeof(*addrs), GFP_KERNEL);
4ea76e90a97d22 Christophe Leroy 2021-03-22 140 if (addrs == NULL) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 141 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 142 goto out_addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 143 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 144
4ea76e90a97d22 Christophe Leroy 2021-03-22 145 memset(&cgctx, 0, sizeof(struct codegen_context));
4ea76e90a97d22 Christophe Leroy 2021-03-22 146
4ea76e90a97d22 Christophe Leroy 2021-03-22 147 /* Make sure that the stack is quadword aligned. */
4ea76e90a97d22 Christophe Leroy 2021-03-22 148 cgctx.stack_size = round_up(fp->aux->stack_depth, 16);
4ea76e90a97d22 Christophe Leroy 2021-03-22 149
4ea76e90a97d22 Christophe Leroy 2021-03-22 150 /* Scouting faux-generate pass 0 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 151 if (bpf_jit_build_body(fp, 0, &cgctx, addrs, false)) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 152 /* We hit something illegal or unsupported. */
4ea76e90a97d22 Christophe Leroy 2021-03-22 153 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 154 goto out_addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 155 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 156
4ea76e90a97d22 Christophe Leroy 2021-03-22 157 /*
4ea76e90a97d22 Christophe Leroy 2021-03-22 158 * If we have seen a tail call, we need a second pass.
4ea76e90a97d22 Christophe Leroy 2021-03-22 159 * This is because bpf_jit_emit_common_epilogue() is called
4ea76e90a97d22 Christophe Leroy 2021-03-22 160 * from bpf_jit_emit_tail_call() with a not yet stable ctx->seen.
4ea76e90a97d22 Christophe Leroy 2021-03-22 161 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 162 if (cgctx.seen & SEEN_TAILCALL) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 163 cgctx.idx = 0;
4ea76e90a97d22 Christophe Leroy 2021-03-22 164 if (bpf_jit_build_body(fp, 0, &cgctx, addrs, false)) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 165 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 166 goto out_addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 167 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 168 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 169
4ea76e90a97d22 Christophe Leroy 2021-03-22 170 /*
4ea76e90a97d22 Christophe Leroy 2021-03-22 171 * Pretend to build prologue, given the features we've seen. This will
4ea76e90a97d22 Christophe Leroy 2021-03-22 172 * update ctgtx.idx as it pretends to output instructions, then we can
4ea76e90a97d22 Christophe Leroy 2021-03-22 173 * calculate total size from idx.
4ea76e90a97d22 Christophe Leroy 2021-03-22 174 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 175 bpf_jit_build_prologue(0, &cgctx);
4ea76e90a97d22 Christophe Leroy 2021-03-22 176 bpf_jit_build_epilogue(0, &cgctx);
4ea76e90a97d22 Christophe Leroy 2021-03-22 177
4ea76e90a97d22 Christophe Leroy 2021-03-22 178 proglen = cgctx.idx * 4;
4ea76e90a97d22 Christophe Leroy 2021-03-22 179 alloclen = proglen + FUNCTION_DESCR_SIZE;
4ea76e90a97d22 Christophe Leroy 2021-03-22 180
4ea76e90a97d22 Christophe Leroy 2021-03-22 181 bpf_hdr = bpf_jit_binary_alloc(alloclen, &image, 4, bpf_jit_fill_ill_insns);
4ea76e90a97d22 Christophe Leroy 2021-03-22 182 if (!bpf_hdr) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 183 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 184 goto out_addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 185 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 186
4ea76e90a97d22 Christophe Leroy 2021-03-22 187 skip_init_ctx:
4ea76e90a97d22 Christophe Leroy 2021-03-22 188 code_base = (u32 *)(image + FUNCTION_DESCR_SIZE);
4ea76e90a97d22 Christophe Leroy 2021-03-22 189
4ea76e90a97d22 Christophe Leroy 2021-03-22 190 if (extra_pass) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 191 /*
4ea76e90a97d22 Christophe Leroy 2021-03-22 192 * Do not touch the prologue and epilogue as they will remain
4ea76e90a97d22 Christophe Leroy 2021-03-22 193 * unchanged. Only fix the branch target address for subprog
4ea76e90a97d22 Christophe Leroy 2021-03-22 194 * calls in the body.
4ea76e90a97d22 Christophe Leroy 2021-03-22 195 *
4ea76e90a97d22 Christophe Leroy 2021-03-22 196 * This does not change the offsets and lengths of the subprog
4ea76e90a97d22 Christophe Leroy 2021-03-22 197 * call instruction sequences and hence, the size of the JITed
4ea76e90a97d22 Christophe Leroy 2021-03-22 198 * image as well.
4ea76e90a97d22 Christophe Leroy 2021-03-22 199 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 200 bpf_jit_fixup_subprog_calls(fp, code_base, &cgctx, addrs);
4ea76e90a97d22 Christophe Leroy 2021-03-22 201
4ea76e90a97d22 Christophe Leroy 2021-03-22 202 /* There is no need to perform the usual passes. */
4ea76e90a97d22 Christophe Leroy 2021-03-22 203 goto skip_codegen_passes;
Goto before pass is inintialized
4ea76e90a97d22 Christophe Leroy 2021-03-22 204 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 205
4ea76e90a97d22 Christophe Leroy 2021-03-22 206 /* Code generation passes 1-2 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 207 for (pass = 1; pass < 3; pass++) {
^^^^^^^^
Here
4ea76e90a97d22 Christophe Leroy 2021-03-22 208 /* Now build the prologue, body code & epilogue for real. */
4ea76e90a97d22 Christophe Leroy 2021-03-22 209 cgctx.idx = 0;
4ea76e90a97d22 Christophe Leroy 2021-03-22 210 bpf_jit_build_prologue(code_base, &cgctx);
4ea76e90a97d22 Christophe Leroy 2021-03-22 211 bpf_jit_build_body(fp, code_base, &cgctx, addrs, extra_pass);
4ea76e90a97d22 Christophe Leroy 2021-03-22 212 bpf_jit_build_epilogue(code_base, &cgctx);
4ea76e90a97d22 Christophe Leroy 2021-03-22 213
4ea76e90a97d22 Christophe Leroy 2021-03-22 214 if (bpf_jit_enable > 1)
4ea76e90a97d22 Christophe Leroy 2021-03-22 215 pr_info("Pass %d: shrink = %d, seen = 0x%x\n", pass,
4ea76e90a97d22 Christophe Leroy 2021-03-22 216 proglen - (cgctx.idx * 4), cgctx.seen);
4ea76e90a97d22 Christophe Leroy 2021-03-22 217 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 218
4ea76e90a97d22 Christophe Leroy 2021-03-22 219 skip_codegen_passes:
4ea76e90a97d22 Christophe Leroy 2021-03-22 220 if (bpf_jit_enable > 1)
4ea76e90a97d22 Christophe Leroy 2021-03-22 221 /*
4ea76e90a97d22 Christophe Leroy 2021-03-22 222 * Note that we output the base address of the code_base
4ea76e90a97d22 Christophe Leroy 2021-03-22 223 * rather than image, since opcodes are in code_base.
4ea76e90a97d22 Christophe Leroy 2021-03-22 224 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 @225 bpf_jit_dump(flen, proglen, pass, code_base);
^^^^
Uninitialized.
4ea76e90a97d22 Christophe Leroy 2021-03-22 226
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
_______________________________________________
kbuild mailing list -- kbuild@lists.01.org
To unsubscribe send an email to kbuild-leave@lists.01.org
^ permalink raw reply [flat|nested] 11+ messages in thread* [kbuild] arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
@ 2022-02-04 10:37 ` Dan Carpenter
0 siblings, 0 replies; 11+ messages in thread
From: Dan Carpenter @ 2022-02-04 10:37 UTC (permalink / raw)
To: kbuild-all
[-- Attachment #1: Type: text/plain, Size: 11796 bytes --]
tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
head: 88808fbbead481aedb46640a5ace69c58287f56a
commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
config: powerpc-randconfig-m031-20220130 (https://download.01.org/0day-ci/archive/20220204/202202040124.7MMKaXXV-lkp(a)intel.com/config )
compiler: powerpc-linux-gcc (GCC) 11.2.0
If you fix the issue, kindly add following tag as appropriate
Reported-by: kernel test robot <lkp@intel.com>
Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
smatch warnings:
arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
vim +/pass +225 arch/powerpc/net/bpf_jit_comp.c
4ea76e90a97d22 Christophe Leroy 2021-03-22 88 struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
4ea76e90a97d22 Christophe Leroy 2021-03-22 89 {
4ea76e90a97d22 Christophe Leroy 2021-03-22 90 u32 proglen;
4ea76e90a97d22 Christophe Leroy 2021-03-22 91 u32 alloclen;
4ea76e90a97d22 Christophe Leroy 2021-03-22 92 u8 *image = NULL;
4ea76e90a97d22 Christophe Leroy 2021-03-22 93 u32 *code_base;
4ea76e90a97d22 Christophe Leroy 2021-03-22 94 u32 *addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 95 struct powerpc64_jit_data *jit_data;
4ea76e90a97d22 Christophe Leroy 2021-03-22 96 struct codegen_context cgctx;
4ea76e90a97d22 Christophe Leroy 2021-03-22 97 int pass;
^^^^^^^^
4ea76e90a97d22 Christophe Leroy 2021-03-22 98 int flen;
4ea76e90a97d22 Christophe Leroy 2021-03-22 99 struct bpf_binary_header *bpf_hdr;
4ea76e90a97d22 Christophe Leroy 2021-03-22 100 struct bpf_prog *org_fp = fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 101 struct bpf_prog *tmp_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 102 bool bpf_blinded = false;
4ea76e90a97d22 Christophe Leroy 2021-03-22 103 bool extra_pass = false;
4ea76e90a97d22 Christophe Leroy 2021-03-22 104
4ea76e90a97d22 Christophe Leroy 2021-03-22 105 if (!fp->jit_requested)
4ea76e90a97d22 Christophe Leroy 2021-03-22 106 return org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 107
4ea76e90a97d22 Christophe Leroy 2021-03-22 108 tmp_fp = bpf_jit_blind_constants(org_fp);
4ea76e90a97d22 Christophe Leroy 2021-03-22 109 if (IS_ERR(tmp_fp))
4ea76e90a97d22 Christophe Leroy 2021-03-22 110 return org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 111
4ea76e90a97d22 Christophe Leroy 2021-03-22 112 if (tmp_fp != org_fp) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 113 bpf_blinded = true;
4ea76e90a97d22 Christophe Leroy 2021-03-22 114 fp = tmp_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 115 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 116
4ea76e90a97d22 Christophe Leroy 2021-03-22 117 jit_data = fp->aux->jit_data;
4ea76e90a97d22 Christophe Leroy 2021-03-22 118 if (!jit_data) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 119 jit_data = kzalloc(sizeof(*jit_data), GFP_KERNEL);
4ea76e90a97d22 Christophe Leroy 2021-03-22 120 if (!jit_data) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 121 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 122 goto out;
4ea76e90a97d22 Christophe Leroy 2021-03-22 123 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 124 fp->aux->jit_data = jit_data;
4ea76e90a97d22 Christophe Leroy 2021-03-22 125 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 126
4ea76e90a97d22 Christophe Leroy 2021-03-22 127 flen = fp->len;
4ea76e90a97d22 Christophe Leroy 2021-03-22 128 addrs = jit_data->addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 129 if (addrs) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 130 cgctx = jit_data->ctx;
4ea76e90a97d22 Christophe Leroy 2021-03-22 131 image = jit_data->image;
4ea76e90a97d22 Christophe Leroy 2021-03-22 132 bpf_hdr = jit_data->header;
4ea76e90a97d22 Christophe Leroy 2021-03-22 133 proglen = jit_data->proglen;
4ea76e90a97d22 Christophe Leroy 2021-03-22 134 alloclen = proglen + FUNCTION_DESCR_SIZE;
4ea76e90a97d22 Christophe Leroy 2021-03-22 135 extra_pass = true;
4ea76e90a97d22 Christophe Leroy 2021-03-22 136 goto skip_init_ctx;
Assume we hit this goto
4ea76e90a97d22 Christophe Leroy 2021-03-22 137 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 138
4ea76e90a97d22 Christophe Leroy 2021-03-22 139 addrs = kcalloc(flen + 1, sizeof(*addrs), GFP_KERNEL);
4ea76e90a97d22 Christophe Leroy 2021-03-22 140 if (addrs == NULL) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 141 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 142 goto out_addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 143 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 144
4ea76e90a97d22 Christophe Leroy 2021-03-22 145 memset(&cgctx, 0, sizeof(struct codegen_context));
4ea76e90a97d22 Christophe Leroy 2021-03-22 146
4ea76e90a97d22 Christophe Leroy 2021-03-22 147 /* Make sure that the stack is quadword aligned. */
4ea76e90a97d22 Christophe Leroy 2021-03-22 148 cgctx.stack_size = round_up(fp->aux->stack_depth, 16);
4ea76e90a97d22 Christophe Leroy 2021-03-22 149
4ea76e90a97d22 Christophe Leroy 2021-03-22 150 /* Scouting faux-generate pass 0 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 151 if (bpf_jit_build_body(fp, 0, &cgctx, addrs, false)) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 152 /* We hit something illegal or unsupported. */
4ea76e90a97d22 Christophe Leroy 2021-03-22 153 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 154 goto out_addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 155 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 156
4ea76e90a97d22 Christophe Leroy 2021-03-22 157 /*
4ea76e90a97d22 Christophe Leroy 2021-03-22 158 * If we have seen a tail call, we need a second pass.
4ea76e90a97d22 Christophe Leroy 2021-03-22 159 * This is because bpf_jit_emit_common_epilogue() is called
4ea76e90a97d22 Christophe Leroy 2021-03-22 160 * from bpf_jit_emit_tail_call() with a not yet stable ctx->seen.
4ea76e90a97d22 Christophe Leroy 2021-03-22 161 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 162 if (cgctx.seen & SEEN_TAILCALL) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 163 cgctx.idx = 0;
4ea76e90a97d22 Christophe Leroy 2021-03-22 164 if (bpf_jit_build_body(fp, 0, &cgctx, addrs, false)) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 165 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 166 goto out_addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 167 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 168 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 169
4ea76e90a97d22 Christophe Leroy 2021-03-22 170 /*
4ea76e90a97d22 Christophe Leroy 2021-03-22 171 * Pretend to build prologue, given the features we've seen. This will
4ea76e90a97d22 Christophe Leroy 2021-03-22 172 * update ctgtx.idx as it pretends to output instructions, then we can
4ea76e90a97d22 Christophe Leroy 2021-03-22 173 * calculate total size from idx.
4ea76e90a97d22 Christophe Leroy 2021-03-22 174 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 175 bpf_jit_build_prologue(0, &cgctx);
4ea76e90a97d22 Christophe Leroy 2021-03-22 176 bpf_jit_build_epilogue(0, &cgctx);
4ea76e90a97d22 Christophe Leroy 2021-03-22 177
4ea76e90a97d22 Christophe Leroy 2021-03-22 178 proglen = cgctx.idx * 4;
4ea76e90a97d22 Christophe Leroy 2021-03-22 179 alloclen = proglen + FUNCTION_DESCR_SIZE;
4ea76e90a97d22 Christophe Leroy 2021-03-22 180
4ea76e90a97d22 Christophe Leroy 2021-03-22 181 bpf_hdr = bpf_jit_binary_alloc(alloclen, &image, 4, bpf_jit_fill_ill_insns);
4ea76e90a97d22 Christophe Leroy 2021-03-22 182 if (!bpf_hdr) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 183 fp = org_fp;
4ea76e90a97d22 Christophe Leroy 2021-03-22 184 goto out_addrs;
4ea76e90a97d22 Christophe Leroy 2021-03-22 185 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 186
4ea76e90a97d22 Christophe Leroy 2021-03-22 187 skip_init_ctx:
4ea76e90a97d22 Christophe Leroy 2021-03-22 188 code_base = (u32 *)(image + FUNCTION_DESCR_SIZE);
4ea76e90a97d22 Christophe Leroy 2021-03-22 189
4ea76e90a97d22 Christophe Leroy 2021-03-22 190 if (extra_pass) {
4ea76e90a97d22 Christophe Leroy 2021-03-22 191 /*
4ea76e90a97d22 Christophe Leroy 2021-03-22 192 * Do not touch the prologue and epilogue as they will remain
4ea76e90a97d22 Christophe Leroy 2021-03-22 193 * unchanged. Only fix the branch target address for subprog
4ea76e90a97d22 Christophe Leroy 2021-03-22 194 * calls in the body.
4ea76e90a97d22 Christophe Leroy 2021-03-22 195 *
4ea76e90a97d22 Christophe Leroy 2021-03-22 196 * This does not change the offsets and lengths of the subprog
4ea76e90a97d22 Christophe Leroy 2021-03-22 197 * call instruction sequences and hence, the size of the JITed
4ea76e90a97d22 Christophe Leroy 2021-03-22 198 * image as well.
4ea76e90a97d22 Christophe Leroy 2021-03-22 199 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 200 bpf_jit_fixup_subprog_calls(fp, code_base, &cgctx, addrs);
4ea76e90a97d22 Christophe Leroy 2021-03-22 201
4ea76e90a97d22 Christophe Leroy 2021-03-22 202 /* There is no need to perform the usual passes. */
4ea76e90a97d22 Christophe Leroy 2021-03-22 203 goto skip_codegen_passes;
Goto before pass is inintialized
4ea76e90a97d22 Christophe Leroy 2021-03-22 204 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 205
4ea76e90a97d22 Christophe Leroy 2021-03-22 206 /* Code generation passes 1-2 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 207 for (pass = 1; pass < 3; pass++) {
^^^^^^^^
Here
4ea76e90a97d22 Christophe Leroy 2021-03-22 208 /* Now build the prologue, body code & epilogue for real. */
4ea76e90a97d22 Christophe Leroy 2021-03-22 209 cgctx.idx = 0;
4ea76e90a97d22 Christophe Leroy 2021-03-22 210 bpf_jit_build_prologue(code_base, &cgctx);
4ea76e90a97d22 Christophe Leroy 2021-03-22 211 bpf_jit_build_body(fp, code_base, &cgctx, addrs, extra_pass);
4ea76e90a97d22 Christophe Leroy 2021-03-22 212 bpf_jit_build_epilogue(code_base, &cgctx);
4ea76e90a97d22 Christophe Leroy 2021-03-22 213
4ea76e90a97d22 Christophe Leroy 2021-03-22 214 if (bpf_jit_enable > 1)
4ea76e90a97d22 Christophe Leroy 2021-03-22 215 pr_info("Pass %d: shrink = %d, seen = 0x%x\n", pass,
4ea76e90a97d22 Christophe Leroy 2021-03-22 216 proglen - (cgctx.idx * 4), cgctx.seen);
4ea76e90a97d22 Christophe Leroy 2021-03-22 217 }
4ea76e90a97d22 Christophe Leroy 2021-03-22 218
4ea76e90a97d22 Christophe Leroy 2021-03-22 219 skip_codegen_passes:
4ea76e90a97d22 Christophe Leroy 2021-03-22 220 if (bpf_jit_enable > 1)
4ea76e90a97d22 Christophe Leroy 2021-03-22 221 /*
4ea76e90a97d22 Christophe Leroy 2021-03-22 222 * Note that we output the base address of the code_base
4ea76e90a97d22 Christophe Leroy 2021-03-22 223 * rather than image, since opcodes are in code_base.
4ea76e90a97d22 Christophe Leroy 2021-03-22 224 */
4ea76e90a97d22 Christophe Leroy 2021-03-22 @225 bpf_jit_dump(flen, proglen, pass, code_base);
^^^^
Uninitialized.
4ea76e90a97d22 Christophe Leroy 2021-03-22 226
---
0-DAY CI Kernel Test Service, Intel Corporation
https://lists.01.org/hyperkitty/list/kbuild-all(a)lists.01.org
_______________________________________________
kbuild mailing list -- kbuild(a)lists.01.org
To unsubscribe send an email to kbuild-leave(a)lists.01.org
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [kbuild] arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
2022-02-04 10:37 ` Dan Carpenter
(?)
(?)
@ 2022-02-04 13:18 ` Christophe Leroy
2022-02-04 14:14 ` Dan Carpenter
-1 siblings, 1 reply; 11+ messages in thread
From: Christophe Leroy @ 2022-02-04 13:18 UTC (permalink / raw)
To: Dan Carpenter, kbuild@lists.01.org
Cc: lkp@intel.com, kbuild-all@lists.01.org,
linux-kernel@vger.kernel.org, Michael Ellerman
Hi Dan,
Le 04/02/2022 à 11:37, Dan Carpenter a écrit :
> tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> head: 88808fbbead481aedb46640a5ace69c58287f56a
> commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
As far as I can see, it's been there long before that.
Seems it comes from 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes
for bpf function calls")
Christophe
> config: powerpc-randconfig-m031-20220130 (https://download.01.org/0day-ci/archive/20220204/202202040124.7MMKaXXV-lkp@intel.com/config )
> compiler: powerpc-linux-gcc (GCC) 11.2.0
>
> If you fix the issue, kindly add following tag as appropriate
> Reported-by: kernel test robot <lkp@intel.com>
> Reported-by: Dan Carpenter <dan.carpenter@oracle.com>
>
> smatch warnings:
> arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
>
> vim +/pass +225 arch/powerpc/net/bpf_jit_comp.c
>
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 88 struct bpf_prog *bpf_int_jit_compile(struct bpf_prog *fp)
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 89 {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 90 u32 proglen;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 91 u32 alloclen;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 92 u8 *image = NULL;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 93 u32 *code_base;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 94 u32 *addrs;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 95 struct powerpc64_jit_data *jit_data;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 96 struct codegen_context cgctx;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 97 int pass;
> ^^^^^^^^
>
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 98 int flen;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 99 struct bpf_binary_header *bpf_hdr;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 100 struct bpf_prog *org_fp = fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 101 struct bpf_prog *tmp_fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 102 bool bpf_blinded = false;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 103 bool extra_pass = false;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 104
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 105 if (!fp->jit_requested)
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 106 return org_fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 107
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 108 tmp_fp = bpf_jit_blind_constants(org_fp);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 109 if (IS_ERR(tmp_fp))
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 110 return org_fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 111
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 112 if (tmp_fp != org_fp) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 113 bpf_blinded = true;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 114 fp = tmp_fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 115 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 116
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 117 jit_data = fp->aux->jit_data;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 118 if (!jit_data) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 119 jit_data = kzalloc(sizeof(*jit_data), GFP_KERNEL);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 120 if (!jit_data) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 121 fp = org_fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 122 goto out;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 123 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 124 fp->aux->jit_data = jit_data;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 125 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 126
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 127 flen = fp->len;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 128 addrs = jit_data->addrs;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 129 if (addrs) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 130 cgctx = jit_data->ctx;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 131 image = jit_data->image;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 132 bpf_hdr = jit_data->header;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 133 proglen = jit_data->proglen;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 134 alloclen = proglen + FUNCTION_DESCR_SIZE;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 135 extra_pass = true;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 136 goto skip_init_ctx;
>
> Assume we hit this goto
>
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 137 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 138
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 139 addrs = kcalloc(flen + 1, sizeof(*addrs), GFP_KERNEL);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 140 if (addrs == NULL) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 141 fp = org_fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 142 goto out_addrs;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 143 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 144
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 145 memset(&cgctx, 0, sizeof(struct codegen_context));
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 146
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 147 /* Make sure that the stack is quadword aligned. */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 148 cgctx.stack_size = round_up(fp->aux->stack_depth, 16);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 149
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 150 /* Scouting faux-generate pass 0 */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 151 if (bpf_jit_build_body(fp, 0, &cgctx, addrs, false)) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 152 /* We hit something illegal or unsupported. */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 153 fp = org_fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 154 goto out_addrs;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 155 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 156
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 157 /*
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 158 * If we have seen a tail call, we need a second pass.
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 159 * This is because bpf_jit_emit_common_epilogue() is called
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 160 * from bpf_jit_emit_tail_call() with a not yet stable ctx->seen.
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 161 */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 162 if (cgctx.seen & SEEN_TAILCALL) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 163 cgctx.idx = 0;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 164 if (bpf_jit_build_body(fp, 0, &cgctx, addrs, false)) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 165 fp = org_fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 166 goto out_addrs;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 167 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 168 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 169
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 170 /*
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 171 * Pretend to build prologue, given the features we've seen. This will
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 172 * update ctgtx.idx as it pretends to output instructions, then we can
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 173 * calculate total size from idx.
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 174 */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 175 bpf_jit_build_prologue(0, &cgctx);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 176 bpf_jit_build_epilogue(0, &cgctx);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 177
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 178 proglen = cgctx.idx * 4;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 179 alloclen = proglen + FUNCTION_DESCR_SIZE;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 180
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 181 bpf_hdr = bpf_jit_binary_alloc(alloclen, &image, 4, bpf_jit_fill_ill_insns);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 182 if (!bpf_hdr) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 183 fp = org_fp;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 184 goto out_addrs;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 185 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 186
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 187 skip_init_ctx:
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 188 code_base = (u32 *)(image + FUNCTION_DESCR_SIZE);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 189
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 190 if (extra_pass) {
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 191 /*
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 192 * Do not touch the prologue and epilogue as they will remain
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 193 * unchanged. Only fix the branch target address for subprog
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 194 * calls in the body.
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 195 *
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 196 * This does not change the offsets and lengths of the subprog
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 197 * call instruction sequences and hence, the size of the JITed
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 198 * image as well.
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 199 */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 200 bpf_jit_fixup_subprog_calls(fp, code_base, &cgctx, addrs);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 201
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 202 /* There is no need to perform the usual passes. */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 203 goto skip_codegen_passes;
>
> Goto before pass is inintialized
>
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 204 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 205
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 206 /* Code generation passes 1-2 */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 207 for (pass = 1; pass < 3; pass++) {
> ^^^^^^^^
> Here
>
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 208 /* Now build the prologue, body code & epilogue for real. */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 209 cgctx.idx = 0;
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 210 bpf_jit_build_prologue(code_base, &cgctx);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 211 bpf_jit_build_body(fp, code_base, &cgctx, addrs, extra_pass);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 212 bpf_jit_build_epilogue(code_base, &cgctx);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 213
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 214 if (bpf_jit_enable > 1)
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 215 pr_info("Pass %d: shrink = %d, seen = 0x%x\n", pass,
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 216 proglen - (cgctx.idx * 4), cgctx.seen);
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 217 }
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 218
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 219 skip_codegen_passes:
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 220 if (bpf_jit_enable > 1)
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 221 /*
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 222 * Note that we output the base address of the code_base
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 223 * rather than image, since opcodes are in code_base.
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 224 */
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 @225 bpf_jit_dump(flen, proglen, pass, code_base);
> ^^^^
> Uninitialized.
>
> 4ea76e90a97d22 Christophe Leroy 2021-03-22 226
>
>
> ---
> 0-DAY CI Kernel Test Service, Intel Corporation
> https://lists.01.org/hyperkitty/list/kbuild-all@lists.01.org
> _______________________________________________
> kbuild mailing list -- kbuild@lists.01.org
> To unsubscribe send an email to kbuild-leave@lists.01.org
>
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
2022-02-04 13:18 ` Christophe Leroy
2022-02-04 14:14 ` Dan Carpenter
@ 2022-02-04 14:14 ` Dan Carpenter
0 siblings, 0 replies; 11+ messages in thread
From: Dan Carpenter @ 2022-02-04 14:14 UTC (permalink / raw)
To: kbuild
[-- Attachment #1: Type: text/plain, Size: 982 bytes --]
On Fri, Feb 04, 2022 at 01:18:24PM +0000, Christophe Leroy wrote:
> Hi Dan,
>
> Le 04/02/2022 à 11:37, Dan Carpenter a écrit :
> > tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> > head: 88808fbbead481aedb46640a5ace69c58287f56a
> > commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
>
> As far as I can see, it's been there long before that.
>
> Seems it comes from 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes
> for bpf function calls")
These emails are auto generated by bot. I just look over the email and
verify it looks reasonable. Moving the code around sometimes makes it
show up as a new bug. Also changes to the Kconfig file can affect which
code is checked.
The commit that you mention does not generate a warning. The warning
started in commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes for
bpf function calls").
regards,
dan carpenter
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [kbuild] arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
@ 2022-02-04 14:14 ` Dan Carpenter
0 siblings, 0 replies; 11+ messages in thread
From: Dan Carpenter @ 2022-02-04 14:14 UTC (permalink / raw)
To: Christophe Leroy
Cc: kbuild@lists.01.org, lkp@intel.com, kbuild-all@lists.01.org,
linux-kernel@vger.kernel.org, Michael Ellerman
On Fri, Feb 04, 2022 at 01:18:24PM +0000, Christophe Leroy wrote:
> Hi Dan,
>
> Le 04/02/2022 à 11:37, Dan Carpenter a écrit :
> > tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> > head: 88808fbbead481aedb46640a5ace69c58287f56a
> > commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
>
> As far as I can see, it's been there long before that.
>
> Seems it comes from 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes
> for bpf function calls")
These emails are auto generated by bot. I just look over the email and
verify it looks reasonable. Moving the code around sometimes makes it
show up as a new bug. Also changes to the Kconfig file can affect which
code is checked.
The commit that you mention does not generate a warning. The warning
started in commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes for
bpf function calls").
regards,
dan carpenter
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [kbuild] arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
@ 2022-02-04 14:14 ` Dan Carpenter
0 siblings, 0 replies; 11+ messages in thread
From: Dan Carpenter @ 2022-02-04 14:14 UTC (permalink / raw)
To: kbuild-all
[-- Attachment #1: Type: text/plain, Size: 982 bytes --]
On Fri, Feb 04, 2022 at 01:18:24PM +0000, Christophe Leroy wrote:
> Hi Dan,
>
> Le 04/02/2022 à 11:37, Dan Carpenter a écrit :
> > tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> > head: 88808fbbead481aedb46640a5ace69c58287f56a
> > commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
>
> As far as I can see, it's been there long before that.
>
> Seems it comes from 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes
> for bpf function calls")
These emails are auto generated by bot. I just look over the email and
verify it looks reasonable. Moving the code around sometimes makes it
show up as a new bug. Also changes to the Kconfig file can affect which
code is checked.
The commit that you mention does not generate a warning. The warning
started in commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes for
bpf function calls").
regards,
dan carpenter
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [kbuild] arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
2022-02-04 14:14 ` Dan Carpenter
(?)
(?)
@ 2022-02-04 14:24 ` Christophe Leroy
2022-02-04 14:30 ` Dan Carpenter
-1 siblings, 1 reply; 11+ messages in thread
From: Christophe Leroy @ 2022-02-04 14:24 UTC (permalink / raw)
To: Dan Carpenter
Cc: kbuild@lists.01.org, lkp@intel.com, kbuild-all@lists.01.org,
linux-kernel@vger.kernel.org, Michael Ellerman
Le 04/02/2022 à 15:14, Dan Carpenter a écrit :
> On Fri, Feb 04, 2022 at 01:18:24PM +0000, Christophe Leroy wrote:
>> Hi Dan,
>>
>> Le 04/02/2022 à 11:37, Dan Carpenter a écrit :
>>> tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
>>> head: 88808fbbead481aedb46640a5ace69c58287f56a
>>> commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
>>
>> As far as I can see, it's been there long before that.
>>
>> Seems it comes from 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes
>> for bpf function calls")
>
> These emails are auto generated by bot. I just look over the email and
> verify it looks reasonable. Moving the code around sometimes makes it
> show up as a new bug. Also changes to the Kconfig file can affect which
> code is checked.
>
> The commit that you mention does not generate a warning. The warning
> started in commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes for
> bpf function calls").
>
???
Didn't I mention commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT
passes for bpf function calls") ?
Christophe
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
2022-02-04 14:24 ` Christophe Leroy
2022-02-04 14:30 ` Dan Carpenter
@ 2022-02-04 14:30 ` Dan Carpenter
0 siblings, 0 replies; 11+ messages in thread
From: Dan Carpenter @ 2022-02-04 14:30 UTC (permalink / raw)
To: kbuild
[-- Attachment #1: Type: text/plain, Size: 1411 bytes --]
On Fri, Feb 04, 2022 at 02:24:50PM +0000, Christophe Leroy wrote:
>
>
> Le 04/02/2022 à 15:14, Dan Carpenter a écrit :
> > On Fri, Feb 04, 2022 at 01:18:24PM +0000, Christophe Leroy wrote:
> >> Hi Dan,
> >>
> >> Le 04/02/2022 à 11:37, Dan Carpenter a écrit :
> >>> tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> >>> head: 88808fbbead481aedb46640a5ace69c58287f56a
> >>> commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
> >>
> >> As far as I can see, it's been there long before that.
> >>
> >> Seems it comes from 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes
> >> for bpf function calls")
> >
> > These emails are auto generated by bot. I just look over the email and
> > verify it looks reasonable. Moving the code around sometimes makes it
> > show up as a new bug. Also changes to the Kconfig file can affect which
> > code is checked.
> >
> > The commit that you mention does not generate a warning. The warning
> > started in commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes for
> > bpf function calls").
> >
>
> ???
>
> Didn't I mention commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT
> passes for bpf function calls") ?
>
Oh... I'm not sure what I was looking at. :P Must be time for me to
knock off for the weekend.
regards,
dan carpenter
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [kbuild] arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
@ 2022-02-04 14:30 ` Dan Carpenter
0 siblings, 0 replies; 11+ messages in thread
From: Dan Carpenter @ 2022-02-04 14:30 UTC (permalink / raw)
To: Christophe Leroy
Cc: kbuild@lists.01.org, lkp@intel.com, kbuild-all@lists.01.org,
linux-kernel@vger.kernel.org, Michael Ellerman
On Fri, Feb 04, 2022 at 02:24:50PM +0000, Christophe Leroy wrote:
>
>
> Le 04/02/2022 à 15:14, Dan Carpenter a écrit :
> > On Fri, Feb 04, 2022 at 01:18:24PM +0000, Christophe Leroy wrote:
> >> Hi Dan,
> >>
> >> Le 04/02/2022 à 11:37, Dan Carpenter a écrit :
> >>> tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> >>> head: 88808fbbead481aedb46640a5ace69c58287f56a
> >>> commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
> >>
> >> As far as I can see, it's been there long before that.
> >>
> >> Seems it comes from 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes
> >> for bpf function calls")
> >
> > These emails are auto generated by bot. I just look over the email and
> > verify it looks reasonable. Moving the code around sometimes makes it
> > show up as a new bug. Also changes to the Kconfig file can affect which
> > code is checked.
> >
> > The commit that you mention does not generate a warning. The warning
> > started in commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes for
> > bpf function calls").
> >
>
> ???
>
> Didn't I mention commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT
> passes for bpf function calls") ?
>
Oh... I'm not sure what I was looking at. :P Must be time for me to
knock off for the weekend.
regards,
dan carpenter
^ permalink raw reply [flat|nested] 11+ messages in thread* Re: [kbuild] arch/powerpc/net/bpf_jit_comp.c:225 bpf_int_jit_compile() error: uninitialized symbol 'pass'.
@ 2022-02-04 14:30 ` Dan Carpenter
0 siblings, 0 replies; 11+ messages in thread
From: Dan Carpenter @ 2022-02-04 14:30 UTC (permalink / raw)
To: kbuild-all
[-- Attachment #1: Type: text/plain, Size: 1411 bytes --]
On Fri, Feb 04, 2022 at 02:24:50PM +0000, Christophe Leroy wrote:
>
>
> Le 04/02/2022 à 15:14, Dan Carpenter a écrit :
> > On Fri, Feb 04, 2022 at 01:18:24PM +0000, Christophe Leroy wrote:
> >> Hi Dan,
> >>
> >> Le 04/02/2022 à 11:37, Dan Carpenter a écrit :
> >>> tree: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git master
> >>> head: 88808fbbead481aedb46640a5ace69c58287f56a
> >>> commit: 51c66ad849a703d9bbfd7704c941827aed0fd9fd powerpc/bpf: Implement extended BPF on PPC32
> >>
> >> As far as I can see, it's been there long before that.
> >>
> >> Seems it comes from 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes
> >> for bpf function calls")
> >
> > These emails are auto generated by bot. I just look over the email and
> > verify it looks reasonable. Moving the code around sometimes makes it
> > show up as a new bug. Also changes to the Kconfig file can affect which
> > code is checked.
> >
> > The commit that you mention does not generate a warning. The warning
> > started in commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT passes for
> > bpf function calls").
> >
>
> ???
>
> Didn't I mention commit 025dceb0fab3 ("bpf: powerpc64: optimize JIT
> passes for bpf function calls") ?
>
Oh... I'm not sure what I was looking at. :P Must be time for me to
knock off for the weekend.
regards,
dan carpenter
^ permalink raw reply [flat|nested] 11+ messages in thread