| Summary: | Missing "uninitialized" warning; may involve "return" statements | ||
|---|---|---|---|
| Product: | gcc | Reporter: | Kevin Cernekee <cernekee> |
| Component: | middle-end | Assignee: | Not yet assigned to anyone <unassigned> |
| Status: | RESOLVED DUPLICATE | ||
| Severity: | normal | CC: | manu |
| Priority: | P3 | ||
| Version: | 4.5.3 | ||
| Target Milestone: | --- | ||
| Host: | Target: | ||
| Build: | Known to work: | ||
| Known to fail: | Last reconfirmed: | ||
This is the standard CCP optimization getting in the way of uninitialized variable warnings. (In reply to comment #1) > This is the standard CCP optimization getting in the way of uninitialized > variable warnings. Correct. *** This bug has been marked as a duplicate of bug 18501 *** |
This code does not produce any warnings ("uninitialized" or otherwise) on gcc 4.4.3, 4.4.5, or 4.5.3: int testcase(unsigned int in) { int a = 0, b; a++; if (in == 1) return -1; b++; return a + b; } Command line: gcc -O2 -c test.c -o test.o -Wall Disassembly: 0000000000000000 <testcase>: 0: b8 ff ff ff ff mov $0xffffffff,%eax 5: c3 retq (IOW, it silently changed the entire function into "return -1") Compiler: $ gcc -v Using built-in specs. Target: x86_64-linux-gnu Configured with: ../src/configure -v --with-pkgversion='Ubuntu 4.4.3-4ubuntu5' --with-bugurl=file:///usr/share/doc/gcc-4.4/README.Bugs --enable-languages=c,c++,fortran,objc,obj-c++ --prefix=/usr --enable-shared --enable-multiarch --enable-linker-build-id --with-system-zlib --libexecdir=/usr/lib --without-included-gettext --enable-threads=posix --with-gxx-include-dir=/usr/include/c++/4.4 --program-suffix=-4.4 --enable-nls --enable-clocale=gnu --enable-libstdcxx-debug --enable-plugin --enable-objc-gc --disable-werror --with-arch-32=i486 --with-tune=generic --enable-checking=release --build=x86_64-linux-gnu --host=x86_64-linux-gnu --target=x86_64-linux-gnu Thread model: posix gcc version 4.4.3 (Ubuntu 4.4.3-4ubuntu5) gcc 4.2.0 and 4.3.5 do produce a warning, using the same command lines. -O0 on gcc 4.4.3 / 4.4.5 / 4.5.3 does produce a warning. -O[123s] do not. If I leave "a" uninitialized, gcc will produce a warning. "a" does not seem to be strictly necessary to reproduce the problem involving "b". If I assign the value of "b" to a global variable before returning, gcc will produce a warning.