Add some more DWARF ARM register numbers

Already filed as Breakpad issue 457, the DWARF parsing code doesn't know about
all valid ARM registers, so it spews warnings when dumping ARM binaries. The ARM
DWARF documentation (
http://infocenter.arm.com/help/topic/com.arm.doc.ihi0040a/IHI0040A_aadwarf.pdf )
lists all the register numbers. This patch doesn't add all of them, but enough
to stop spewing out warnings from the Android libraries I'm dumping.

[jimb]

LGTM

http://breakpad.appspot.com/418002/diff/1/src/common/dwarf_cfi_to_module.cc
File src/common/dwarf_cfi_to_module.cc (right):

http://breakpad.appspot.com/418002/diff/1/src/common/dwarf_cfi_to_module.cc#n...
src/common/dwarf_cfi_to_module.cc:93: "fps", "cpsr", "",   "",    "",    "",   
"",    "",
If we ever get CFI that references some of these unmapped register numbers
(accidentally or otherwise), it's going to be pretty confusing. Would it be
worth it to have DwarfCFIToModule::RegisterName check for empty strings, and
treat it the same way it treats out-of-bounds register numbers?

[Ted Mielczarek]

On 2012/07/18 17:49:49, jimb wrote:
> LGTM
> 
> http://breakpad.appspot.com/418002/diff/1/src/common/dwarf_cfi_to_module.cc
> File src/common/dwarf_cfi_to_module.cc (right):
> 
>
http://breakpad.appspot.com/418002/diff/1/src/common/dwarf_cfi_to_module.cc#n...
> src/common/dwarf_cfi_to_module.cc:93: "fps", "cpsr", "",   "",    "",    "",  

> "",    "",
> If we ever get CFI that references some of these unmapped register numbers
> (accidentally or otherwise), it's going to be pretty confusing. Would it be
> worth it to have DwarfCFIToModule::RegisterName check for empty strings, and
> treat it the same way it treats out-of-bounds register numbers?

That's an excellent idea, I'll add that before landing.