This reverts commit 01262f75.

This should not adversely affect the rest of the logic as it is only ever entered in cases in which we would have asserted false before

Enables us to instrument x509

Replacing RIP with insn->size() results in a loss of information

The main problem is the use of an addition instead of a multiplication for scale values.
In addition, to make sure the result can be used to create new instructions from assembly,
all generated numbers must be in hex.

* Update 68-sled code to be keystone friendly
* Corrected spelling of getInstruction.  removed extraneous assert.
* Disable should-pin-immediately because it interfers with sleds.
* Remove assert from fix-calls about odd prefixing, just dont fix the call.  More of an issue with FII doing disassembly
* Update switch->code discovery path to only happen if switch table is in text segment
* fixed FII to do what the last commit said -- only disassembly when TABLE is in text.  also patched zipr to handle xbegin gracefully.
* Disable verbose setting
* Create lots of dbs instead of just a few

Daniel Chen authored 4 years ago and Jason Hiser committed 4 years ago

which includes an interface for multi-line assembly where it can fill
in template parameters automatically for the caller.  See
irdb-cookbook/init_stack.

Daniel Chen authored 4 years ago and Jason Hiser committed 4 years ago

The old method involved writing assembly to a file, asking an external assembler to do the conversion, then processing the binary file to read the binary instructions back out.  This method is slow and relies on nasm, which isn't available on all machines or for the full variety of supported architectures.  The replacement is keystone, a 3rd party library that is linked to provide assembly->binary conversion.