Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT
On Fri, Dec 09, 2022 at 02:54:15PM +, Henry Wang wrote: > Hi, > > > -Original Message- > > From: Demi Marie Obenour > > Subject: Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT > > > > On Fri, Dec 09, 2022 at 07:37:53AM +, Henry Wang wrote: > > > Hi Julien, > > > > > > > -Original Message----- > > > > From: Julien Grall > > > > Subject: Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT > > > > > > > > Hi, > > > > > > > > >>> Signed-off-by: Demi Marie Obenour > > > > >> > > > > >> Acked-by: Jan Beulich > > > > >> > > > > >>> Should this be included in 4.17? It is a bug fix for a feature new > > > > >>> to > > > > >>> 4.17, so I suspect yes, but it is ultimately up to Henry Wang. The > > > > >>> code > > > > >>> is identical to v2, but I have improved the commit message. > > > > >> > > > > >> It may be too late now, looking at the state of the tree. Henry, > > > > >> Julien? > > > > > > > > > > Like I said in v2, I don't object the change if you would like to > > > > > include > > this > > > > patch > > > > > to 4.17, so if you are sure this patch is safe and want to commit it, > > > > > feel > > free > > > > to add: > > > > > > > > > > Release-acked-by: Henry Wang > > > > > > > > > > Since we also need to commit: > > > > > "[PATCH for-4.17] SUPPORT.md: Define support lifetime" so from my > > side > > > > > I am no problem. Julien might have different opinion though, if Julien > > > > object > > > > > the change I would like to respect his opinion and leave this patch > > > > uncommitted. > > > > > > > > I have committed it after SUPPORT.md. So if for some reasons we are > > seen > > > > any issues with Osstest, then I can tag the tree without this patch > > > > > > This is a great solution :) > > > > > > > (that said, I would rather prefer if we have staging-4.17 == > > > > stable-4.17). > > > > > > Looks like now staging-4.17 == stable-4.17 now, with this patch pushed. > > > So we are ready to tag. > > > > And it turns out that I botched the initial patch, sorry. (I forgot to > > handle the multiboot2 case.) > > > > I understand if it is too late for stable-4.17, but it ought to make > > stable 4.17.1 as it was simply omitted from the initial patch series. > > I don't think this patch will make it today so I would suggest we still follow > what Julien planned yesterday. Also I think this is also consistent with the > release management guideline. That’s okay. Qubes can take this as an out of tree patch until it is merged. -- Sincerely, Demi Marie Obenour (she/her/hers) Invisible Things Lab signature.asc Description: PGP signature
RE: [PATCH v3] Use EfiACPIReclaimMemory for ESRT
Hi, > -Original Message- > From: Demi Marie Obenour > Subject: Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT > > On Fri, Dec 09, 2022 at 07:37:53AM +, Henry Wang wrote: > > Hi Julien, > > > > > -Original Message- > > > From: Julien Grall > > > Subject: Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT > > > > > > Hi, > > > > > > >>> Signed-off-by: Demi Marie Obenour > > > >> > > > >> Acked-by: Jan Beulich > > > >> > > > >>> Should this be included in 4.17? It is a bug fix for a feature new to > > > >>> 4.17, so I suspect yes, but it is ultimately up to Henry Wang. The > > > >>> code > > > >>> is identical to v2, but I have improved the commit message. > > > >> > > > >> It may be too late now, looking at the state of the tree. Henry, > > > >> Julien? > > > > > > > > Like I said in v2, I don't object the change if you would like to > > > > include > this > > > patch > > > > to 4.17, so if you are sure this patch is safe and want to commit it, > > > > feel > free > > > to add: > > > > > > > > Release-acked-by: Henry Wang > > > > > > > > Since we also need to commit: > > > > "[PATCH for-4.17] SUPPORT.md: Define support lifetime" so from my > side > > > > I am no problem. Julien might have different opinion though, if Julien > > > object > > > > the change I would like to respect his opinion and leave this patch > > > uncommitted. > > > > > > I have committed it after SUPPORT.md. So if for some reasons we are > seen > > > any issues with Osstest, then I can tag the tree without this patch > > > > This is a great solution :) > > > > > (that said, I would rather prefer if we have staging-4.17 == stable-4.17). > > > > Looks like now staging-4.17 == stable-4.17 now, with this patch pushed. > > So we are ready to tag. > > And it turns out that I botched the initial patch, sorry. (I forgot to > handle the multiboot2 case.) > > I understand if it is too late for stable-4.17, but it ought to make > stable 4.17.1 as it was simply omitted from the initial patch series. I don't think this patch will make it today so I would suggest we still follow what Julien planned yesterday. Also I think this is also consistent with the release management guideline. Kind regards, Henry > -- > Sincerely, > Demi Marie Obenour (she/her/hers) > Invisible Things Lab
Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT
On Fri, Dec 09, 2022 at 07:37:53AM +, Henry Wang wrote: > Hi Julien, > > > -Original Message- > > From: Julien Grall > > Subject: Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT > > > > Hi, > > > > >>> Signed-off-by: Demi Marie Obenour > > >> > > >> Acked-by: Jan Beulich > > >> > > >>> Should this be included in 4.17? It is a bug fix for a feature new to > > >>> 4.17, so I suspect yes, but it is ultimately up to Henry Wang. The code > > >>> is identical to v2, but I have improved the commit message. > > >> > > >> It may be too late now, looking at the state of the tree. Henry, Julien? > > > > > > Like I said in v2, I don't object the change if you would like to include > > > this > > patch > > > to 4.17, so if you are sure this patch is safe and want to commit it, > > > feel free > > to add: > > > > > > Release-acked-by: Henry Wang > > > > > > Since we also need to commit: > > > "[PATCH for-4.17] SUPPORT.md: Define support lifetime" so from my side > > > I am no problem. Julien might have different opinion though, if Julien > > object > > > the change I would like to respect his opinion and leave this patch > > uncommitted. > > > > I have committed it after SUPPORT.md. So if for some reasons we are seen > > any issues with Osstest, then I can tag the tree without this patch > > This is a great solution :) > > > (that said, I would rather prefer if we have staging-4.17 == stable-4.17). > > Looks like now staging-4.17 == stable-4.17 now, with this patch pushed. > So we are ready to tag. And it turns out that I botched the initial patch, sorry. (I forgot to handle the multiboot2 case.) I understand if it is too late for stable-4.17, but it ought to make stable 4.17.1 as it was simply omitted from the initial patch series. -- Sincerely, Demi Marie Obenour (she/her/hers) Invisible Things Lab signature.asc Description: PGP signature
RE: [PATCH v3] Use EfiACPIReclaimMemory for ESRT
Hi Julien, > -Original Message- > From: Julien Grall > Subject: Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT > > Hi, > > >>> Signed-off-by: Demi Marie Obenour > >> > >> Acked-by: Jan Beulich > >> > >>> Should this be included in 4.17? It is a bug fix for a feature new to > >>> 4.17, so I suspect yes, but it is ultimately up to Henry Wang. The code > >>> is identical to v2, but I have improved the commit message. > >> > >> It may be too late now, looking at the state of the tree. Henry, Julien? > > > > Like I said in v2, I don't object the change if you would like to include > > this > patch > > to 4.17, so if you are sure this patch is safe and want to commit it, feel > > free > to add: > > > > Release-acked-by: Henry Wang > > > > Since we also need to commit: > > "[PATCH for-4.17] SUPPORT.md: Define support lifetime" so from my side > > I am no problem. Julien might have different opinion though, if Julien > object > > the change I would like to respect his opinion and leave this patch > uncommitted. > > I have committed it after SUPPORT.md. So if for some reasons we are seen > any issues with Osstest, then I can tag the tree without this patch This is a great solution :) > (that said, I would rather prefer if we have staging-4.17 == stable-4.17). Looks like now staging-4.17 == stable-4.17 now, with this patch pushed. So we are ready to tag. > > My plan is to prepare the tarball tomorrow. Thanks very much. Kind regards, Henry > > Cheers, > > -- > Julien Grall
Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT
Hi,
On 08/12/2022 08:53, Henry Wang wrote:
-Original Message-
From: Jan Beulich
Subject: Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT
On 07.12.2022 23:42, Demi Marie Obenour wrote:
A previous patch tried to get Linux to use the ESRT under Xen if it is
in memory of type EfiRuntimeServicesData. However, EfiRuntimeServices*
memory needs to be included in the EFI page tables, so it is best to
minimize the amount of memory of this type. Since EFI runtime services
do not need access to the ESRT, EfiACPIReclaimMemory is a better choice.
Link: https://lists.xenproject.org/archives/html/xen-devel/2022-
09/msg01365.html
Fixes: dc7da0874ba4 ("EFI: preserve the System Resource Table for dom0")
Signed-off-by: Demi Marie Obenour
Acked-by: Jan Beulich
Should this be included in 4.17? It is a bug fix for a feature new to
4.17, so I suspect yes, but it is ultimately up to Henry Wang. The code
is identical to v2, but I have improved the commit message.
It may be too late now, looking at the state of the tree. Henry, Julien?
Like I said in v2, I don't object the change if you would like to include this
patch
to 4.17, so if you are sure this patch is safe and want to commit it, feel free
to add:
Release-acked-by: Henry Wang
Since we also need to commit:
"[PATCH for-4.17] SUPPORT.md: Define support lifetime" so from my side
I am no problem. Julien might have different opinion though, if Julien object
the change I would like to respect his opinion and leave this patch uncommitted.
I have committed it after SUPPORT.md. So if for some reasons we are seen
any issues with Osstest, then I can tag the tree without this patch
(that said, I would rather prefer if we have staging-4.17 == stable-4.17).
My plan is to prepare the tarball tomorrow.
Cheers,
--
Julien Grall
RE: [PATCH v3] Use EfiACPIReclaimMemory for ESRT
Hi Jan,
> -Original Message-
> From: Jan Beulich
> Subject: Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT
>
> On 07.12.2022 23:42, Demi Marie Obenour wrote:
> > A previous patch tried to get Linux to use the ESRT under Xen if it is
> > in memory of type EfiRuntimeServicesData. However, EfiRuntimeServices*
> > memory needs to be included in the EFI page tables, so it is best to
> > minimize the amount of memory of this type. Since EFI runtime services
> > do not need access to the ESRT, EfiACPIReclaimMemory is a better choice.
> >
> > Link: https://lists.xenproject.org/archives/html/xen-devel/2022-
> 09/msg01365.html
> > Fixes: dc7da0874ba4 ("EFI: preserve the System Resource Table for dom0")
> > Signed-off-by: Demi Marie Obenour
>
> Acked-by: Jan Beulich
>
> > Should this be included in 4.17? It is a bug fix for a feature new to
> > 4.17, so I suspect yes, but it is ultimately up to Henry Wang. The code
> > is identical to v2, but I have improved the commit message.
>
> It may be too late now, looking at the state of the tree. Henry, Julien?
Like I said in v2, I don't object the change if you would like to include this
patch
to 4.17, so if you are sure this patch is safe and want to commit it, feel free
to add:
Release-acked-by: Henry Wang
Since we also need to commit:
"[PATCH for-4.17] SUPPORT.md: Define support lifetime" so from my side
I am no problem. Julien might have different opinion though, if Julien object
the change I would like to respect his opinion and leave this patch uncommitted.
Kind regards,
Henry
>
> Jan
Re: [PATCH v3] Use EfiACPIReclaimMemory for ESRT
On 07.12.2022 23:42, Demi Marie Obenour wrote:
> A previous patch tried to get Linux to use the ESRT under Xen if it is
> in memory of type EfiRuntimeServicesData. However, EfiRuntimeServices*
> memory needs to be included in the EFI page tables, so it is best to
> minimize the amount of memory of this type. Since EFI runtime services
> do not need access to the ESRT, EfiACPIReclaimMemory is a better choice.
>
> Link:
> https://lists.xenproject.org/archives/html/xen-devel/2022-09/msg01365.html
> Fixes: dc7da0874ba4 ("EFI: preserve the System Resource Table for dom0")
> Signed-off-by: Demi Marie Obenour
Acked-by: Jan Beulich
> Should this be included in 4.17? It is a bug fix for a feature new to
> 4.17, so I suspect yes, but it is ultimately up to Henry Wang. The code
> is identical to v2, but I have improved the commit message.
It may be too late now, looking at the state of the tree. Henry, Julien?
Jan
