Bootstrap of current trunk on powerpc64-linux fails in libstdc++
building system_error.lo. The code that fails was added a few days
ago,
but the failure seems to be the same as the one reported in PR 31490.
I
verified that the patch from comment #10 of that PR allows bootstrap of
c,c++,fortran to succeed and now I'm doing a bootstrap and regtest of
all languages but Ada. I'll report the results in the PR.
Since there seems to be interest in that patch again, I bootstrapped
it on x86-64 (all default languages); no new fails, except for
gcc.dg/20051207-3.c, which is to be expected since the whole goal
of the patch is to change this behaviour (i.e., to allow both writable
and read-only objects in the same section).
I can't bootstrap and test a powerpc64 compiler in less than 48 hours
or so; the patch does fix the original problem though (tested with a
cross-compiler).
My original patch (attached to PR31490) had some little problems,
here's the thing I actually tested.
I know it needs a testsuite change (and a Changelog entry); I'll wait
for Janis' testing for that though (and thanks for that testing, btw
:-) )
Segher
--- a/gcc/varasm.c
+++ b/gcc/varasm.c
@@ -536,10 +536,20 @@ get_section (const char *name, unsigned int
flags, tree de
else
{
sect = *slot;
+
+ /* Sanity check user variables for flag changes. */
if ((sect->common.flags & ~SECTION_DECLARED) != flags
&& ((sect->common.flags | flags) & SECTION_OVERRIDE) == 0)
{
- /* Sanity check user variables for flag changes. */
+ /* Allow mixed writable and read-only objects in named
sections. */
+ if ((sect->common.flags & SECTION_NAMED) != 0
+ && ((sect->common.flags ^ flags) & ~SECTION_DECLARED)
+ == SECTION_WRITE)
+ {
+ sect->common.flags |= SECTION_WRITE;
+ return sect;
+ }
+
if (decl == 0)
decl = sect->named.decl;
gcc_assert (decl);
