[PATCH] Fix macro info lookup for binaries containing DWARFv5 line table

[Keith Seitz]

On 5/24/21 4:36 AM, Tomar, Sourabh Singh wrote:
> [AMD Public Use]
> 
> Hello Keith,
> 
> Could you please share your plan WRT to this patch.
> Do you want to take it forward ? or you want me to take this forward.
> 

I can pursue this...

In that vein, does anyone (maintainers?) have an input on my "counterpatch"
(reposted below) that removes this IS_ABSOLUTE_PREFIX stuff and copies the
symtab's filename?

I haven't officially submitted this as a patch because I'm curious whether
my reading of this is correct/complete. Maybe the documentation/comments
are incorrect or no longer valid? 

FWIW, I've tested that patch on native x86_64 Fedora 34 with no regressions.

Keith

Patch under discussion:

gdb/ChangeLog

	* dwarf2/line-header.c (line_header::file_file_name): Copy
	the symtab's filename.

diff --git a/gdb/dwarf2/line-header.c b/gdb/dwarf2/line-header.c
index 7575297f966..117c5a42cc5 100644
--- a/gdb/dwarf2/line-header.c
+++ b/gdb/dwarf2/line-header.c
@@ -69,15 +69,10 @@ line_header::file_file_name (int file) const
     {
       const file_entry *fe = file_name_at (file);
 
-      if (!IS_ABSOLUTE_PATH (fe->name))
-	{
-	  const char *dir = fe->include_dir (this);
-	  if (dir != NULL)
-	    return gdb::unique_xmalloc_ptr<char> (concat (dir, SLASH_STRING,
-							  fe->name,
-							  (char *) NULL));
-	}
-      return make_unique_xstrdup (fe->name);
+      /* macro_source_file requires: "This filename is relative to the
+	 compilation directory, it exactly matches the symtab->filename
+	 content."  */
+      return make_unique_xstrdup (fe->symtab->filename);
     }
   else
     {