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/16201] internal error on a cygwin program linked against a DLL with no .data section


https://sourceware.org/bugzilla/show_bug.cgi?id=16201

Pierre Muller <muller at sourceware dot org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
                 CC|                            |muller at sourceware dot org

--- Comment #15 from Pierre Muller <muller at sourceware dot org> ---
  This problem might indeed have been introduced by
my patch to improve reading of export sections in DLLs.

  The problem is that minimal symbols only recognizes
three symbol types in enum ms_type:
  mst_text, mst_data and mst_unknown
but objfile header
itself has four special sections:
sect_index_text, sect_index_data, sect_index_bss and sect_index_rodata

See:
https://sourceware.org/cgi-bin/cvsweb.cgi/src/gdb/coff-pe-read.c?rev=1.19&content-type=text/x-cvsweb-markup&cvsroot=src

The following patch seems to work for me,
but I am not sure it is correct...

diff --git a/gdb/coff-pe-read.c b/gdb/coff-pe-read.c
index 91ee3f6..954c457 100644
--- a/gdb/coff-pe-read.c
+++ b/gdb/coff-pe-read.c
@@ -466,6 +466,13 @@ read_pe_exported_syms (struct objfile *objfile)
        {
          section_data[sectix].rva_start = vaddr;
          section_data[sectix].rva_end = vaddr + vsize;
+         /* Force sect_index, even if it was already set before.  */
+         if (sectix == PE_SECTION_INDEX_TEXT)
+           objfile->sect_index_text = sectix;
+         if (sectix == PE_SECTION_INDEX_DATA)
+           objfile->sect_index_data = sectix;
+         if (sectix == PE_SECTION_INDEX_BSS)
+           objfile->sect_index_bss = sectix;
        }
       else
        {
@@ -480,11 +487,23 @@ read_pe_exported_syms (struct objfile *objfile)
          section_data[otherix].rva_end = vaddr + vsize;
          section_data[otherix].vma_offset = 0;
          if (characteristics & IMAGE_SCN_CNT_CODE)
-           section_data[otherix].ms_type = mst_text;
+           {
+             section_data[otherix].ms_type = mst_text;
+             if (objfile->sect_index_text == -1)
+               objfile->sect_index_text = otherix;
+           }
          else if (characteristics & IMAGE_SCN_CNT_INITIALIZED_DATA)
-           section_data[otherix].ms_type = mst_data;
+           {
+             section_data[otherix].ms_type = mst_data;
+             if (objfile->sect_index_data == -1)
+             objfile->sect_index_data = otherix;
+           }
          else if (characteristics & IMAGE_SCN_CNT_UNINITIALIZED_DATA)
-           section_data[otherix].ms_type = mst_bss;
+           {
+             section_data[otherix].ms_type = mst_bss;
+             if (objfile->sect_index_bss == -1)
+               objfile->sect_index_bss = otherix;
+           }
          else
            section_data[otherix].ms_type = mst_unknown;
          otherix++;


In the hope this helps,

Pierre Muller

PS: Is it "normal" that the program generates
a SIGSEGV?

-- 
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]