Mirror of the gdb-patches mailing list
 help / color / mirror / Atom feed
* Fix `add-symbol-file -o ... -s ...` address mapping bug
@ 2026-09-05 10:16 Dragorn421
  0 siblings, 0 replies; only message in thread
From: Dragorn421 @ 2026-09-05 10:16 UTC (permalink / raw)
  To: gdb-patches

Hello,

I encountered an issue in GDB and am hereby submitting a patch for 
fixing it.

First, let me explain the problem: it has to do with the 
`add-symbol-file` command that is used to map elf files to arbitrary 
addresses.

For example if I want to map the .text section from main.o at 0x1234:

```
(gdb) add-symbol-file main.o -s .text 0x1234
add symbol table from file "main.o" at
         .text_addr = 0x1234
(y or n) y
Reading symbols from main.o...
(No debugging symbols found in main.o)
(gdb) info files
...
         0x0000000000001030 - 0x0000000000001040 is .plt.got
         0x0000000000001234 - 0x000000000000132c is .text
         0x0000000000001138 - 0x0000000000001145 is .fini
...
```

`info files` does then report the expect address for .text

However this breaks when combining with the -o "set offset for other 
unspecified sections" option:

```
(gdb) add-symbol-file main.o -o 0xFF000000 -s .text 0x1234
add symbol table from file "main.o" at
         .text_addr = 0x1234
with other sections offset by 0xff000000
(y or n) y
Reading symbols from main.o...
(No debugging symbols found in main.o)
(gdb) info files
...
         0x00000000ff001030 - 0x00000000ff001040 is .plt.got
         0x0000000000001040 - 0x0000000000001138 is .text
         0x00000000ff001138 - 0x00000000ff001145 is .fini
...
```

We notice here the -o option was correctly used per the addresses of 
e.g. the `.fini` section, but the .text section address is now wrong.

This behavior boils down to the `set_objfile_default_section_offset` 
function assuming it can pass 0 as `objfile_relocate`'s `offsets` 
entries to not modify the `-s` mappings previously set by the 
`symbol_file_add` call in `add_symbol_file_command`. When in fact 
`objfile_relocate` does not check for 0, it only checks for the new 
offset being the same as the current one.

The proposed fix is to change `set_objfile_default_section_offset` to 
pass the current offset instead of 0 for sections that should be unaltered:

```diff
diff --git a/gdb/symfile.c b/gdb/symfile.c
index 017f7a49d8d..5691c2a46d9 100644
--- a/gdb/symfile.c
+++ b/gdb/symfile.c
@@ -2139,8 +2139,8 @@ set_objfile_default_section_offset (struct objfile 
*objf,
      = addrs_section_sort (objf_addrs);

    /* Walk the BFD section list, and if a matching section is found in
-     ADDRS_SORTED_LIST, set its offset to zero to keep its address
-     unchanged.
+     ADDRS_SORTED_LIST, set its offset to its current offset to keep
+     its address unchanged.

       Note that both lists may contain multiple sections with the same
       name, and then the sections from ADDRS are matched in BFD order
@@ -2163,7 +2163,7 @@ set_objfile_default_section_offset (struct objfile 
*objf,
         }

        if (cmp == 0)
-       offsets[objf_sect->sectindex] = 0;
+       offsets[objf_sect->sectindex] = 
objf->section_offsets[objf_sect->sectindex];
      }

    /* Apply the new section offsets.  */
```

This is my first time contributing so my apologies if I'm doing anything 
not properly, please let me know.

Sincerely,
Dragorn421


^ permalink raw reply	[flat|nested] only message in thread

only message in thread, other threads:[~2026-09-05 10:17 UTC | newest]

Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2026-09-05 10:16 Fix `add-symbol-file -o ... -s ...` address mapping bug Dragorn421

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox