Re: [RFC PATCH v2 4/6] livepatch: reuse module loader code to write relocations

On Thu, Dec 10, 2015 at 08:56:27PM +0100, Jiri Kosina wrote:

On Thu, 10 Dec 2015, Minfei Huang wrote:

quoted

quoted

+	klp_for_each_reloc_sec(obj, reloc_sec) {
+		relindex = reloc_sec->index;
+		num_relas = pmod->sechdrs[relindex].sh_size / sizeof(Elf_Rela);
+		rela = (Elf_Rela *) pmod->sechdrs[relindex].sh_addr;
+
+		/* For each rela in this __klp_rela section */
+		for (i = 0; i < num_relas; i++, rela++) {
+			sym = symtab + ELF_R_SYM(rela->r_info);
+			symname = pmod->core_strtab + sym->st_name;
+
+			if (sym->st_shndx == SHN_LIVEPATCH) {
+				if (sym->st_info == 'K')
+					ret = klp_find_external_symbol(pmod, symname, &addr);
+				else
+					ret = klp_find_object_symbol(obj->name, symname, &addr);
+				if (ret)
+					return ret;
+				sym->st_value = addr;
+			}
 		}
+		ret = apply_relocate_add(pmod->sechdrs, pmod->core_strtab,
+					 pmod->index.sym, relindex, pmod);

It is more appropiate to test the ret.

And if error is encountered ... then you'd propose to do what?

The code is correct as is; the return value is propagated properly to 
caller, and all the existing callers have proper error handling.

apply_relocate_add() is called from inside a loop, so if there's an
error, it should return immediately instead of continuing the next
iteration.

-- 
Josh