[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH v5 19/76] target/mips: Add emulation of nanoMIPS
|
From: |
Richard Henderson |
|
Subject: |
Re: [Qemu-devel] [PATCH v5 19/76] target/mips: Add emulation of nanoMIPS 16-bit arithmetic instructions |
|
Date: |
Wed, 1 Aug 2018 13:11:00 -0400 |
|
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.9.1 |
On 08/01/2018 12:02 PM, Aleksandar Markovic wrote:
>> I suggest a "target_long imm;" at the top of the function to handle all such
>> that will be required when this function is filled out.
>
> gen_arith_imm() and similar functions have "int" as the type of offset/imm
> parameter (in this case the last parameter) - shouldn't "int" be more
> appropriate type for local variable than "target_long"?
Fair enough. It certainly fits all of the immediates that nanomips can
generate.
If you eventually split up gen_arith_imm etc as previously suggested, I suggest
changing to target_long though. That would allow e.g. the decode for the
MIPS64R6 DATI and DAHI instructions to use the basic gen_daddiu expander with
the immediate already shifted into place in the high 32 bits.
r~