[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [PATCH] exec.c: Replace memory_region_is_unassigned() w
From: |
Richard Henderson |
Subject: |
Re: [Qemu-devel] [PATCH] exec.c: Replace memory_region_is_unassigned() with memory_region_is_ram_backed() |
Date: |
Mon, 9 Jul 2018 10:50:51 -0700 |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.8.0 |
On 07/09/2018 10:33 AM, Peter Maydell wrote:
>> Well... ok. We should document that this, surprisingly, does not include
>> actual ram. Just things that are ram with caveats.
>
> I think it must include actual RAM, or we would not be able to
> execute from actual RAM ? The only way to get to get_page_addr_code()'s
> "here's the ram_addr_t for this" exit path is if memory_region_is_unassigned()
> returns false.
>
>>> - return mr != &io_mem_rom && mr != &io_mem_notdirty && !mr->rom_device
>>> - && mr != &io_mem_watch;
Huh. I don't follow how this old expression excludes ram,
and so assumed that must be checked separately earlier.
There's clearly still something here I don't understand.
r~