[Bug binutils/25662] objcopy sets invalid sh_offset for the first section in a no_contents segment containing program headers

[jozef.l at mittosystems dot com]

https://sourceware.org/bugzilla/show_bug.cgi?id=25662

--- Comment #10 from Jozef Lawrynowicz <jozef.l at mittosystems dot com> ---
(In reply to Alan Modra from comment #9)
> The testcase patch looks mostly good, thanks.  At first I thought you had
> the remote host file names wrong, but it doesn't seem too important what
> file names are used on the host.
> 
> Some other issues:
> - The new test is using an ELF .type directive.  Either remove that or make
> the test ELF only.
> - The test will error if ld isn't built, as is the case for a few targets,
> ia64-vms being an example.
> 
> Please post to the mailing list for any further discussion.

I posted an updated patch here
https://sourceware.org/pipermail/binutils/2020-March/110408.html.

BTW, the test fails for i386-pe, but as expected it's not fixed by your patch
to elf.c. A quick comparison of objdump -x output shows that the "Time/Date"
field has been reset to epoch 0.

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