Importing binaries

Importing creates a program database from a file. It does not mean the sample has been fully analyzed or that the selected language is correct.

Import one file

1. In the Project window or CodeBrowser, choose File → Import File.
2. Select the file and inspect the importer dialog.
3. Review the detected format and language/compiler selection.
4. Open the loader-specific options when you need to change memory mapping or symbol-loading behavior.
5. Accept the import, then read the result summary before opening the program.

The upstream importer supports formats including ELF, PE, Mach-O, and raw binary. An archive can open the batch importer instead. VERIFY format support and option names in the Opcodex release.

Treat raw binary as a separate task

A raw blob may contain no reliable architecture or load-address metadata. Gather evidence from the device, firmware documentation, executable container, or a known boot address. Select processor, endianness, and base address deliberately. Plausible-looking instructions alone do not establish that the mapping is right.

Check before continuing

• Expected sections or memory blocks are present.
• Architecture and bitness match the sample.
• Entry points and addresses are plausible.
• Import warnings have been recorded.

Keep the original sample unchanged. If you need a substantially different mapping, re-import into a separately named program to compare results.

Next: Auto analysis or Headless import.

---

Source: Importer help.

Adapted from the Ghidra documentation (Apache 2.0).