mbox series

[for-4.14-to-4.16,0/2] Backports for XSA-409 fixes

Message ID 20221025092112.50168-1-Henry.Wang@arm.com (mailing list archive)
Headers show
Series Backports for XSA-409 fixes | expand

Message

Henry Wang Oct. 25, 2022, 9:21 a.m. UTC
This series is a backported series for XSA-409 fixes from master for
4.14 to 4.16, which cherry-picking commits:
3783e583319f arm/p2m: Rework p2m_init()
c7cff1188802 xen/arm: p2m: Populate pages for GICv2 mapping in p2m_init()

Andrew Cooper (1):
  arm/p2m: Rework p2m_init()

Henry Wang (1):
  xen/arm: p2m: Populate pages for GICv2 mapping in p2m_init()

 xen/arch/arm/domain.c     |  2 +-
 xen/arch/arm/p2m.c        | 58 +++++++++++++++++++++++++++++----------
 xen/include/asm-arm/p2m.h | 14 +++++++---
 3 files changed, 55 insertions(+), 19 deletions(-)

Comments

Julien Grall Oct. 25, 2022, 8:06 p.m. UTC | #1
Hi Henry,

On 25/10/2022 10:21, Henry Wang wrote:
> This series is a backported series for XSA-409 fixes from master for
> 4.14 to 4.16, which cherry-picking commits:
> 3783e583319f arm/p2m: Rework p2m_init()
> c7cff1188802 xen/arm: p2m: Populate pages for GICv2 mapping in p2m_init()

I have merged the 2 patches in Xen 4.16, 4.15 and 4.14. I noticed that 
the "released-acked-by" tags were removed.

We usually keep the commit message as-is (including tags). But I vaguely 
remember that we may have stripped the "released-acked-by" tag in the 
past. So I left it alone.

Also, it looks like the tools I am using to download the patches (b4) 
decided to move the "cherry-picked ..." line before the tags. I am not 
entirely sure why... So I have modified the commit message to re-add the 
line where you initially added (this is the correct place!).

Cheers,
Henry Wang Oct. 26, 2022, 12:24 a.m. UTC | #2
Hi Julien,

> -----Original Message-----
> From: Julien Grall <julien@xen.org>
> Subject: Re: [PATCH for-4.14-to-4.16 0/2] Backports for XSA-409 fixes
> 
> Hi Henry,
> 
> On 25/10/2022 10:21, Henry Wang wrote:
> > This series is a backported series for XSA-409 fixes from master for
> > 4.14 to 4.16, which cherry-picking commits:
> > 3783e583319f arm/p2m: Rework p2m_init()
> > c7cff1188802 xen/arm: p2m: Populate pages for GICv2 mapping in
> p2m_init()
> 
> I have merged the 2 patches in Xen 4.16, 4.15 and 4.14.

Thank you very much!

> I noticed that the "released-acked-by" tags were removed.

Yeah, I cannot find any "Release-acked-by" from all the backported
patches so I removed that tag for convenience of the maintainer doing
the backport.

> 
> We usually keep the commit message as-is (including tags). But I vaguely
> remember that we may have stripped the "released-acked-by" tag in the
> past. So I left it alone.

Thanks.

> 
> Also, it looks like the tools I am using to download the patches (b4)
> decided to move the "cherry-picked ..." line before the tags. I am not
> entirely sure why... So I have modified the commit message to re-add the
> line where you initially added (this is the correct place!).

Hmmm this is strange...probably b4 thinks this "cherry picked commit" is
some kind of unnecessary information which shouldn't appear in the commit
message :)

My initial thought was adding the "cherry picked commit" would reduce
the maintainer's workload so the patch can be applied without any modification
but it seems that you still did some extra work....sorry about that.

Kind regards,
Henry

> 
> Cheers,
> 
> --
> Julien Grall
Jan Beulich Oct. 26, 2022, 9:14 a.m. UTC | #3
On 25.10.2022 22:06, Julien Grall wrote:
> On 25/10/2022 10:21, Henry Wang wrote:
>> This series is a backported series for XSA-409 fixes from master for
>> 4.14 to 4.16, which cherry-picking commits:
>> 3783e583319f arm/p2m: Rework p2m_init()
>> c7cff1188802 xen/arm: p2m: Populate pages for GICv2 mapping in p2m_init()
> 
> I have merged the 2 patches in Xen 4.16, 4.15 and 4.14. I noticed that 
> the "released-acked-by" tags were removed.
> 
> We usually keep the commit message as-is (including tags). But I vaguely 
> remember that we may have stripped the "released-acked-by" tag in the 
> past. So I left it alone.

Indeed I'm trying to remember to strip those when doing backports, as they
aren't meaningful (but potentially confusing) on the stable branches.

Jan
Julien Grall Oct. 27, 2022, 12:16 p.m. UTC | #4
On 26/10/2022 01:24, Henry Wang wrote:
> Hi Julien,
> 
>> -----Original Message-----
>> From: Julien Grall <julien@xen.org>
>> Subject: Re: [PATCH for-4.14-to-4.16 0/2] Backports for XSA-409 fixes
>>
>> Hi Henry,
>>
>> On 25/10/2022 10:21, Henry Wang wrote:
>>> This series is a backported series for XSA-409 fixes from master for
>>> 4.14 to 4.16, which cherry-picking commits:
>>> 3783e583319f arm/p2m: Rework p2m_init()
>>> c7cff1188802 xen/arm: p2m: Populate pages for GICv2 mapping in
>> p2m_init()
>>
>> I have merged the 2 patches in Xen 4.16, 4.15 and 4.14.
> 
> Thank you very much!
> 
>> I noticed that the "released-acked-by" tags were removed.
> 
> Yeah, I cannot find any "Release-acked-by" from all the backported
> patches so I removed that tag for convenience of the maintainer doing
> the backport.
> 
>>
>> We usually keep the commit message as-is (including tags). But I vaguely
>> remember that we may have stripped the "released-acked-by" tag in the
>> past. So I left it alone.
> 
> Thanks.
> 
>>
>> Also, it looks like the tools I am using to download the patches (b4)
>> decided to move the "cherry-picked ..." line before the tags. I am not
>> entirely sure why... So I have modified the commit message to re-add the
>> line where you initially added (this is the correct place!).
> 
> Hmmm this is strange...probably b4 thinks this "cherry picked commit" is
> some kind of unnecessary information which shouldn't appear in the commit
> message :)

I am guessing 'b4' is removing all the tags and then append them at the end.

> 
> My initial thought was adding the "cherry picked commit" would reduce
> the maintainer's workload so the patch can be applied without any modification
> but it seems that you still did some extra work....sorry about that.

You did the right thing to add the cherry-pick tag. This is more my 
workflow that is broken :). I will have a look if I can fix it.

Cheers,