zturner accepted this revision.
This revision is now accepted and ready to land.
================
Comment at:
source/Plugins/Process/minidump/RegisterContextMinidump_x86_64.cpp:50
@@ +49,3 @@
+ size = 8;
+ break;
+ }
----------------
If they are just smaller (but never bigger), you can call `take_front(N)` on
the result. So you could say:
```
writeRegister(source_data,
reg_info[lldb_cs_x86_64].mutable_data(result_base).take_front(2));
```
Now the line is starting to get long so might look ugly if it wraps, so it's up
to you. It is somewhat safer at least since you will assert if you
accidentally specify an invalid size, and you don't have to worry about getting
the computation right.
Perhaps you could combine this with the approach you've already taken here.
Add the functions to `RegisterInfo`, then change `getDestRegister` to look like
this:
```
llvm::MutableArrayRef<uint8_t> getDestRegister(const RegisterInfo ®,
uint8_t *context, uint32_t
lldb_reg_num) {
auto bytes = reg.mutable_data(context);
switch (lldb_reg_num) {
case lldb_cs_x86_64:
case lldb_ds_x86_64:
case lldb_es_x86_64:
case lldb_fs_x86_64:
case lldb_gs_x86_64:
case lldb_ss_x86_64:
return bytes.take_front(2);
break;
case lldb_rflags_x86_64:
return bytes.take_front(4);
break;
default:
return bytes.take_front(8);
}
```
then you could call `writeRegister` like this:
```
writeRegister(source_data, getDestRegister(reg_info, result_base,
lldb_cs_x86_64));
```
At this point though I'm not sure what's better. So don't consider this a
requirement, up to you :)
https://reviews.llvm.org/D24919
_______________________________________________
lldb-commits mailing list
lldb-commits@lists.llvm.org
http://lists.llvm.org/cgi-bin/mailman/listinfo/lldb-commits
