GCC10 LTO is too smart (and somewhat cryptic):
find-debuginfo.c: In function ‘dwfl_standard_find_debuginfo’:
debuginfod-client.c:85:8: error: ‘bits’ may be used uninitialized
in this function [-Werror=maybe-uninitialized]
find-debuginfo.c:360:24: note: ‘bits’ was declared here
lto1: all warnings being treated as errors
So it inlines __libdwfl_debuginfod_find_debuginfo into
dwfl_standard_find_debuginfo and since it cannot see into the
function pointer (*fp_debuginfod_find_debuginfo), it assumes that
build_id_bit (== bits in dwfl_standard_find_debuginfo) will be used
by the called function and it might not be initialized.
But if you read the code the there is a check for build_id_len > 0
to see whether bits is or isn't initialized before using bits.
But gcc isn't smart enough to figure that out.
Maybe that actually should be an heuristic gcc lto should use.
If the callchain I am inlining is so deep that I cannot figure out
maybe-uninitialized variables anymore I should stop inlining.
For now just help GCC out and initialize bits to NULL.
Signed-off-by: Mark Wielaard <mark@klomp.org>
+2020-04-16 Mark Wielaard <mark@klomp.org>
+
+ * find-debuginfo.c (dwfl_standard_find_debuginfo): Initialize bits
+ to NULL.
+
2020-01-24 Mark Wielaard <mark@klomp.org>
* linux-kernel-modules.c (find_kernel_elf): Check release isn't NULL.
{
/* First try by build ID if we have one. If that succeeds or fails
other than just by finding nothing, that's all we do. */
- const unsigned char *bits;
+ const unsigned char *bits = NULL;
GElf_Addr vaddr;
int bits_len;
if ((bits_len = INTUSE(dwfl_module_build_id) (mod, &bits, &vaddr)) > 0)