[Bug c/100618] New: Add a -fno-semantic-interposition variant which allows variable interposition

[Bug c/100618] New: Add a -fno-semantic-interposition variant which allows variable interposition

[i at maskray dot me]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100618

            Bug ID: 100618
           Summary: Add a -fno-semantic-interposition variant which allows
                    variable interposition
           Product: gcc
           Version: 11.0
            Status: UNCONFIRMED
          Severity: normal
          Priority: P3
         Component: c
          Assignee: unassigned at gcc dot gnu.org
          Reporter: i at maskray dot me
  Target Milestone: ---

Extracted from https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100483

The documentation says -fno-semantic-interposition applies to variables.

Having an option which only apply to external linkage function definitions will
be useful. Assuming no-variable-interposition is unfortunately incompatible
with the plethora of copy relocations: -fno-pic emits direct access relocations
referencing a global variable. If the global variable turns out to be defined
in a shared object, there will be a copy relocation in the executable. The
object the shared object sees and the executable sees will be different.

See
https://maskray.me/blog/2021-05-16-elf-interposition-and-bsymbolic#the-last-alliance-of-elf-and-men
for more context.

[Bug c/100618] Add a -fno-semantic-interposition variant which allows variable interposition

[i at maskray dot me]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100618

--- Comment #1 from Fangrui Song <i at maskray dot me> ---
Perhaps

-fsemantic-interposition=function,variable (default -fpic/-fPIC)
-fsemantic-interposition=variable   (compatible with copy relocations but
enable function optimizations)
-fsemantic-interposition=  (alias: -fno-semantic-interposition)

?

[Bug c/100618] Add a -fno-semantic-interposition variant which allows variable interposition

[amonakov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100618

Alexander Monakov <amonakov at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |amonakov at gcc dot gnu.org

--- Comment #2 from Alexander Monakov <amonakov at gcc dot gnu.org> ---
The implied split is 'code,data,tls-data' rather than 'functions,variables'
(there are constants too, which are not variables, but should be treated like
variables here; and TLS data does not rely on copy relocations).

Since the original option was intended to be used mainly in the negative form,
I think it may be less confusing to introduce

-f[no-]semantic-code-interposition

[Bug c/100618] Add a -fno-semantic-interposition variant which allows variable interposition

[amonakov at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100618

--- Comment #3 from Alexander Monakov <amonakov at gcc dot gnu.org> ---
Furthermore as discussed in bug 100483 this request appears based on a
misunderstanding what the 'semantic-' part of the option is about. It does not
affect assembly/linker-level binding mechanism, so things like presence of copy
relocations should not be affected.

[Bug c/100618] Add a -fno-semantic-interposition variant which allows variable interposition

[i at maskray dot me]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100618

Fangrui Song <i at maskray dot me> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|---                         |FIXED
             Status|UNCONFIRMED                 |RESOLVED

--- Comment #4 from Fangrui Song <i at maskray dot me> ---
Clang 13  -fno-semantic-interposition will be mostly consistent with GCC 
-fno-semantic-interposition. It looked like a misunderstand from my side.

[Bug c/100618] Add a -fno-semantic-interposition variant which allows variable interposition

[redi at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100618

Jonathan Wakely <redi at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
         Resolution|FIXED                       |WORKSFORME