[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
Re: [Qemu-devel] [RFC PATCH] linux-user: Add signal handling for x86_64
From: |
Pranith Kumar |
Subject: |
Re: [Qemu-devel] [RFC PATCH] linux-user: Add signal handling for x86_64 |
Date: |
Fri, 03 Feb 2017 10:55:42 -0500 |
User-agent: |
mu4e 0.9.17; emacs 25.1.1 |
Peter Maydell writes:
> On 25 January 2017 at 00:10, Pranith Kumar <address@hidden> wrote:
>> Adopted from a previous patch posting:
>> https://lists.gnu.org/archive/html/qemu-devel/2016-07/msg02079.html
>>
>> CC: Allan Wirth <address@hidden>
>> CC: Peter Maydell <address@hidden>
>> Signed-off-by: Pranith Kumar <address@hidden>
>
> Thanks for picking this patch up. A nit about commit message format:
> because this is mostly Allan's work you need to add his signed-off-by:
> line (which he provided on his original patch posting), and make
> a brief not of what was changed, so it looks like:
>
> Signed-off-by: Original Author <address@hidden>
> [OP: changed X, Y, Z]
> Signed-off-by: Other Person <address@hidden>
>
> It's also in this kind of situation worth considering whether the
> patch would be better attributed to Allan as the git commit 'author'.
> If I've taken somebody else's work and made mostly minor overhauls
> to it I tend to go for giving them credit in the git commit log.
OK, I'll add these SOB lines and attribute it to Allan as he did most of the
work.
>
>> ---
>> linux-user/signal.c | 264
>> ++++++++++++++++++++++++++++++++++++++++-------
>> target/i386/cpu.h | 2 +
>> target/i386/fpu_helper.c | 12 +++
>> 3 files changed, 242 insertions(+), 36 deletions(-)
>>
>> diff --git a/linux-user/signal.c b/linux-user/signal.c
>> index 0a5bb4e26b..0248621d66 100644
>> --- a/linux-user/signal.c
>> +++ b/linux-user/signal.c
>> @@ -253,8 +253,7 @@ int do_sigprocmask(int how, const sigset_t *set,
>> sigset_t *oldset)
>> return 0;
>> }
>>
>> -#if !defined(TARGET_OPENRISC) && !defined(TARGET_UNICORE32) && \
>> - !defined(TARGET_X86_64)
>> +#if !defined(TARGET_OPENRISC) && !defined(TARGET_UNICORE32)
>> /* Just set the guest's signal mask to the specified value; the
>> * caller is assumed to have called block_signals() already.
>> */
>
> There's a minor conflict here with the Nios2 code that's now
> in master (which added another clause to this #if), but it's
> trivial to resolve.
I'll rebase my patch on master and fix up the conflicts and send a v2.
>
> Otherwise:
>
> Reviewed-by: Peter Maydell <address@hidden>
Thanks for the review!
--
Pranith