https://sourceware.org/bugzilla/show_bug.cgi?id=27217
--- Comment #8 from Kinsey Moore <nyphbl8d+sourceware at gmail dot com> --- (In reply to Nick Clifton from comment #7) > (In reply to Kinsey Moore from comment #6) > > I've tested the new patch and, while everything now compiles and links, > > there seems to be some weird behavior with global symbols. > > Yeah - this is the kind of thing that I was worried about. > > > This bug was > > previously affecting two tests in RTEMS under AArch64 and both tests now > > exhibit global symbols not yielding their assigned values. The remainder of > > the tests still pass appropriately. > > Am I correct in thinking that these assigned values are non-zero ? You are correct, the assigned values are non-zero in both cases. > One effect of removing the relocations entirely is that the effected > symbol's value is always treated as 0... > > Do you have a test case that I can look at to see if I can improve the > patch ? I'll see if I can come up with a reduced test case. Do you prefer an executable output that runs on QEMU or just something that compiles to an object file to pick apart? -- You are receiving this mail because: You are on the CC list for the bug.
