[PATCH] Mark vDSO as not a file

[PATCH] Mark vDSO as not a file

[Tom Tromey]

The vDSO objfile is not a real file, so mark it as such.  I noticed
this because, when playing with debuginfod, I saw:

Downloading 0.01 MB separate debug info for /tmp/system-supplied DSO at 0x7ffff7fc9000

That "/tmp" is wrong -- it's just gdb's cwd.  This patch corrects the
problem.

Regression tested on x86-64 Fedora 34.
---
 gdb/symfile-mem.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/gdb/symfile-mem.c b/gdb/symfile-mem.c
index b4c359ccd31..73c5ee378b8 100644
--- a/gdb/symfile-mem.c
+++ b/gdb/symfile-mem.c
@@ -88,7 +88,7 @@ symbol_file_add_from_memory (struct bfd *templ, CORE_ADDR addr,
   struct bfd *nbfd;
   struct bfd_section *sec;
   bfd_vma loadbase;
-  symfile_add_flags add_flags = 0;
+  symfile_add_flags add_flags = SYMFILE_NOT_FILENAME;
 
   if (bfd_get_flavour (templ) != bfd_target_elf_flavour)
     error (_("add-symbol-file-from-memory not supported for this target"));
-- 
2.34.1

Re: [PATCH] Mark vDSO as not a file

[Simon Marchi]

On 2022-03-04 14:26, Tom Tromey via Gdb-patches wrote:
> The vDSO objfile is not a real file, so mark it as such.  I noticed
> this because, when playing with debuginfod, I saw:
> 
> Downloading 0.01 MB separate debug info for /tmp/system-supplied DSO at 0x7ffff7fc9000

The patch LGTM, just one question.

What's the output with that fixed, the same thing without the "/tmp/"?

Simon

Re: [PATCH] Mark vDSO as not a file

[Tom Tromey]

Simon> What's the output with that fixed, the same thing without the "/tmp/"?

Yes, exactly.

I suspect there are still spots in gdb that use the objfile name as a
filename even when OBJF_NOT_FILENAME is set.  I haven't audited
this... probably worth doing.

Tom