qemu-devel
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Qemu-devel] [RFC 0/3] target/m68k: convert to transaction_failed ho


From: Laurent Vivier
Subject: Re: [Qemu-devel] [RFC 0/3] target/m68k: convert to transaction_failed hook
Date: Wed, 12 Dec 2018 21:43:47 +0100
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:52.0) Gecko/20100101 Thunderbird/52.8.0

On 10/12/2018 17:56, Peter Maydell wrote:
> This patchset converts the m68k target from the deprecated
> unassigned_access hook to the new transaction_failed hook.
> It's RFC for a couple of reasons:
>  * it's untested, since I don't have an m68k test image
>  * the second patch just makes "bus error while trying to
>    read page tables" be treated as a page fault, when it
>    should probably cause a fault reporting it as a bus error
>    of some kind
>  * I don't understand why the old unassigned_access hook
>    set the ATC bit in the MMU SSW, since the docs I have say
>    this should be set if the fault happened during a table
>    search, but cleared if it's just an ordinary bus-errored
>    data or insn access. Probably this is a pre-existing bug?
> 
> Anyway, I send it out as a skeleton for comments, because
> it would be nice to get rid of the old unassigned_access
> hook, which is fundamentally broken (it's still used by m68k,
> microblaze, mips and sparc).
> 
> thanks
> -- PMM
> 
> Peter Maydell (3):
>   target/m68k: In dump_address_map() check for memory access failures
>   target/m68k: In get_physical_address() check for memory access
>     failures
>   target/m68k: Switch to transaction_failed hook
> 
>  target/m68k/cpu.h       |  7 ++--
>  target/m68k/cpu.c       |  2 +-
>  target/m68k/helper.c    | 84 ++++++++++++++++++++++++++++++++---------
>  target/m68k/op_helper.c | 20 ++++------
>  4 files changed, 80 insertions(+), 33 deletions(-)
> 

Tested-by: Laurent Vivier <address@hidden>

I'll try to review this later...

Thanks,
Laurent



reply via email to

[Prev in Thread] Current Thread [Next in Thread]