This is the mail archive of the
binutils@sourceware.org
mailing list for the binutils project.
[PATCH] Issue an error for GC on __patchable_function_entries section
On Sat, Feb 1, 2020 at 10:21 AM Fangrui Song <i@maskray.me> wrote:
>
> On 2020-02-01, H.J. Lu wrote:
> >On Sat, Feb 1, 2020 at 9:34 AM Fangrui Song <i@maskray.me> wrote:
> >>
> >> On 2020-02-01, H.J. Lu wrote:
> >> >After all text sections have been garbage collected, if a
> >> >__patchable_function_entries section references a section which
> >> >wasn't marked, mark it with SEC_EXCLUDE and return NULL. Otherwise,
> >> >keep it.
> >> >
> >> >Should it be handled in _bfd_elf_gc_mark_extra_sections?
> >>
> >> Thanks for paying attention to these feature requests.
> >>
> >> I referenced GNU as and ld requests at
> >> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93492#c2
> >> If we
> >>
> >> * implement SHF_LINK_ORDER
> >
> >I am not sure if overloading (abusing?) SHF_LINK_ORDER is a good idea.
>
> It is an extension, but it is agreed by multiple parties on
> https://groups.google.com/d/msg/generic-abi/_CbBM6T6WeM/eGF9A0AnAAAJ ,
> including HP-UX and Solaris developers.
>
> >> * allow multiple sections with the same name ("unique")
> >
> >This is orthogonal to this. I have a question on assembly syntax:
> >
> >https://sourceware.org/bugzilla/show_bug.cgi?id=25380#c1
> >
> >> * teach GCC to use SHF_LINK_ORDER and "unique" (see https://gcc.gnu.org/ml/gcc/2020-01/msg00067.html)
> >>
> >> An ad-hoc gc marking will be unnecessary.
> >
> >We need to scan relocations in _patchable_function_entries section for
> >references to garbage collected sections. We can either check section
> >name or a SHF_XXX. But I don't know if SHF_LINK_ORDER is a good
> >approach.
>
> We don't need to if we use multiple __patchable_function_entries
> sections. Multiple sections are a must due to
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93195#c1 (COMDAT)
>
> > A symbol table entry with STB_LOCAL binding that is defined relative
> > to one of a group's sections, and that is contained in a symbol table
> > section that is not part of the group, must be discarded if the group
> > members are discarded. References to this symbol table entry from
> > outside the group are not allowed.
>
> The __patchable_function_entries creation logic I implemented in clang:
>
> foreach function foo
> find the first function label defined in foo's section, name it $associated
> ($associated can have 2 reasonable values, w/ or w/o -ffunction-sections)
> get or create an id for $associated, say, $unique
>
> if foo is in a COMDAT named $comdat
> .section __patchable_function_entries,"awo",@progbits,$associated,comdat,$comdat,unique,$unique
> else
> .section __patchable_function_entries,"awo",@progbits,$associated,unique,$unique
>
> This approach uses feature requests I referenced in *direct* links of
> https://gcc.gnu.org/ml/gcc/2020-01/msg00067.html plus
> https://gcc.gnu.org/bugzilla/show_bug.cgi?id=93492#c2 ,
> and addresses all bugs I filed.
>
Here is a linker patch to issue an error to avoid corrupt
linker output.
--
H.J.
From 27c1af63b6b4a69efbd0dbaeff912d069a1997d4 Mon Sep 17 00:00:00 2001
From: "H.J. Lu" <hjl.tools@gmail.com>
Date: Sun, 2 Feb 2020 15:32:34 -0800
Subject: [PATCH] Issue an error for GC on __patchable_function_entries section
__patchable_function_entries section is generated by a compiler with
-fpatchable-function-entry=XX. The assembly code looks like this:
---
.text
.globl _start
.type _start, %function
_start:
.section __patchable_function_entries,"awo",%progbits,_start
.dc.a .LPFE1
.text
.LPFE1:
.byte 0
---
But --gc-sections will silently remove __patchable_function_entries
section and generate corrupt result. The linker bug will be fixed
by implementing the 'o' section flag with linked-to section:
https://sourceware.org/bugzilla/show_bug.cgi?id=25381
In the meantime, this patch disallows garbage collection on
__patchable_function_entries section without linked-to section.
bfd/
PR ld/25490
* elflink.c (_bfd_elf_gc_mark_extra_sections): Issue an error
for garbage collection on __patchable_function_entries section
without linked-to section.
ld/
PR ld/25490
* testsuite/ld-elf/pr25490-1.d: New file.
* testsuite/ld-elf/pr25490-1.d: Likewise.
---
bfd/elflink.c | 7 +++++++
ld/testsuite/ld-elf/pr25490-1.d | 2 ++
ld/testsuite/ld-elf/pr25490-1.s | 9 +++++++++
3 files changed, 18 insertions(+)
create mode 100644 ld/testsuite/ld-elf/pr25490-1.d
create mode 100644 ld/testsuite/ld-elf/pr25490-1.s
diff --git a/bfd/elflink.c b/bfd/elflink.c
index 5217528a79b..e4d92952aaf 100644
--- a/bfd/elflink.c
+++ b/bfd/elflink.c
@@ -13350,6 +13350,13 @@ _bfd_elf_gc_mark_extra_sections (struct bfd_link_info *info,
&& (isec->flags & SEC_DEBUGGING)
&& CONST_STRNEQ (isec->name, ".debug_line."))
debug_frag_seen = TRUE;
+ else if (strcmp (bfd_section_name (isec),
+ "__patchable_function_entries") == 0
+ && elf_linked_to_section (isec) == NULL)
+ info->callbacks->einfo (_("%F%P: %pB(%pA): error: "
+ "need linked-to section "
+ "for --gc-sections\n"),
+ isec->owner, isec);
}
/* If no non-note alloc section in this file will be kept, then
diff --git a/ld/testsuite/ld-elf/pr25490-1.d b/ld/testsuite/ld-elf/pr25490-1.d
new file mode 100644
index 00000000000..7cc2e6aaa1c
--- /dev/null
+++ b/ld/testsuite/ld-elf/pr25490-1.d
@@ -0,0 +1,2 @@
+#ld: --gc-sections -e _start
+#error: .*\(__patchable_function_entries\): error: need linked-to section for --gc-sections
diff --git a/ld/testsuite/ld-elf/pr25490-1.s b/ld/testsuite/ld-elf/pr25490-1.s
new file mode 100644
index 00000000000..51ba1ea8014
--- /dev/null
+++ b/ld/testsuite/ld-elf/pr25490-1.s
@@ -0,0 +1,9 @@
+ .text
+ .globl _start
+ .type _start, %function
+_start:
+ .section __patchable_function_entries,"aw",%progbits
+ .dc.a .LPFE1
+ .text
+.LPFE1:
+ .byte 0
--
2.24.1