bug-binutils
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[Bug ld/18452] ld allows overlapping sections


From: cristiangavril_olar at yahoo dot com
Subject: [Bug ld/18452] ld allows overlapping sections
Date: Tue, 29 Mar 2016 22:33:35 +0000

https://sourceware.org/bugzilla/show_bug.cgi?id=18452

--- Comment #12 from Cristian Gavril Olar <cristiangavril_olar at yahoo dot 
com> ---
Created attachment 9141
  --> https://sourceware.org/bugzilla/attachment.cgi?id=9141&action=edit
Quick thing to check

I'd like to quickly check something with you all. See this attached modified
version of the first testcase. I just did a quick check, didn't yet make all of
the changes suggested but just plaed some printfs like:

if (dot < last->vma
                    && os->bfd_section->size != 0
                    && dot + os->bfd_section->size <= last->vma)
...
                        printf("Overlays detected!\n");

...
                  }
                else
                  {
                        printf("Overlays not detected!\n");

I notice with this, as was my expectation, there are no overlays detected, so
then, with no overlays detected, the !SEC_LOAD would pass. But the VMA address
check would not. Still this is an intended overlay for my usecase in there. It
is just that the intended overlap is of nolaod sections not load sections.
Because in either way I do not want nor need memory eaten up by neither of the
.bss or .otherbss sections.

This usecase is extremely valid and required for me. I'd like to understand the
position on this. I would have preferred to stay in tune with vanilla binutils
but if it must be, I guess that's the beauty of GPL :) -> we'll just have to
fork binutils into something that accepts this use case.

If this usecase is accepted as needed (as it definitely is for my usecase) then
I would be open to discuss further and help if possible with implementing
support for it. Otherwise I wont' bother anymore and go with forking.

-- 
You are receiving this mail because:
You are on the CC list for the bug.


reply via email to

[Prev in Thread] Current Thread [Next in Thread]