This is the mail archive of the gdb-prs@sourceware.org mailing list for the GDB project.


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]
Other format: [Raw text]

[Bug gdb/11786] PIE support may not work for some PIEs


https://sourceware.org/bugzilla/show_bug.cgi?id=11786

--- Comment #8 from dje at google dot com ---
(In reply to Jan Kratochvil from comment #5)
> (In reply to dje from comment #3)
> > suppose we ignore the flags field for all segments: what's the
> > worst that could happen?
> 
> If you use non-matching build of an executable it may get incorrectly
> relocated by some small displacement.  (*1)
> 
> People sometimes try to use non-matching build, IMO it cannot give
> meaningful output but it may make their attempts even more difficult when
> the executable gets "randomly" displaced.  Currently GDB does not display
> any notice it used PIE displacement (and there is also no longer any notice
> it used PIC displacement), unless one has "set verbose" (which nobody has). 
> This may make a falsely-matched displacement a pain.
> 
> (*1) Currently the displacement still has to be PAGE_SIZE aligned, if it is
> not PAGE_SIZE aligned no displacement gets used.  But I have TODOlisted I
> should recheck the PAGE_SIZE requirement as it was removed for shlibs by
> 2da7921acc5c7b327b3619a95ca7ca36a0314dc4 and IMO it should be equally
> removed for compatibility with prelinked/unprelinked PIE executables.

If the addresses/displacements are different, sure.  One recognizes that going
in.
But I asked about the flags field.

btw, 2da7921acc5c7b327b3619a95ca7ca36a0314dc4:  kinda meaningless without some
reference to a repository.  For archeology's sake, IWBN to record that here.

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


Index Nav: [Date Index] [Subject Index] [Author Index] [Thread Index]
Message Nav: [Date Prev] [Date Next] [Thread Prev] [Thread Next]