Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork

[Laszlo Ersek]

On 04/08/16 21:00, Mark wrote:
> Thanks for pointing that out, sorry.
> 
> On branch ide_perf_fix_v2
> Your branch is up-to-date with 'origin/ide_perf_fix_v2'.
> nothing to commit, working directory clean
> 
> Build success, qemu command line:
> qemu-system-x86_64 -enable-kvm -net none -m 1500 -pflash ~/OVMF_CODE.fd
> -cdrom /stor/Downloads/X17-58997.iso
> 
> Back to full performance while loading the win7 ISO, also the archlinux ISO.
> 
> Thanks!

Perfect, thank you very much for the report and the testing!

(Also for the willingness to subscribe to edk2-devel; some users dislike
that. I for one definitely dislike that subscription is a requirement
for posting; no other OSS devel mailing list that I know of demands it.)

Cheers!
Laszlo

> 
> On Fri, Apr 8, 2016 at 11:51 AM, Laszlo Ersek  > wrote:
> 
> On 04/08/16 20:36, Mark wrote:
> > I cloned your github but couldn't find the appropriate branch to
> build.
> >
> > [m@ornage edk2]$ git status
> > On branch remove_old_pci_host
> > Your branch is up-to-date with 'origin/remove_old_pci_host'.
> > nothing to commit, working directory clean
> >
> > If that is the one, I built it successfully but did not solve the
> > performance issue.
> >
> > Should I be in another branch?
> 
> Please see the cover letter of the v2 series that I CC'd you on
> (although I forgot to mark the subject as v2):
> 
> http://thread.gmane.org/gmane.comp.bios.edk2.devel/10511
> 
> It states that the branch name is "ide_perf_fix_v2".
> 
> Thanks
> Laszlo
> 
> >
> > On Fri, Apr 8, 2016 at 2:08 AM, Laszlo Ersek  
> > >> wrote:
> >
> > On 04/08/16 05:09, Mark wrote:
> > > Yes I'm the same person. :)
> > >
> > > Currently I am using /usr/share/ovmf/x64/ovmf_code_x64.bin
> built from
> > > arch user repo package ovmf-git on 5 Apr 2016.  When I was
> > reporting the
> > > issues, I was using
> > > edk2.git-ovmf-x64-0-20160324.b1635.gf0bbcdf.noarch.rpm from the
> > Gerd repo.
> > >
> > > CPU on the host had a single core pegged during the poor
> performance.
> > >
> > > Multiple distros was referring to the host, sorry for the
> > ambiguity.  I
> > > also tried arch linux on a guest just to see if it was a
> win7 iso
> > issue,
> > > the xml dump was dumped at a moment I had that arch ISO loaded.
> > >
> > > I just rebuilt ovmf-git 18308.8207d79-1 (Thu Apr  7 20:04:21
> PDT 2016)
> > > and the issue still exists, I see that it is being addressed.
> > >
> > > Thanks to everyone who contributes to this project, I enjoy the
> > product
> > > very much.  If there is any more info I can provide, please ask.
> >
> > Can you perhaps confirm the fix works? I'll soon send a v2,
> and push it
> > to my personal github repo as well. If you could fetch it,
> build it, and
> > test it, that would be great.
> >
> > Thanks!
> > Laszlo
> >
> > > On Thu, Apr 7, 2016 at 12:36 PM, Laszlo Ersek
> 
> > >
> > > 
>  > >
> > > On 04/07/16 20:24, Jordan Justen wrote:
> > > > On 2016-04-07 08:07:35, Laszlo Ersek wrote:
> > > >> I like the diffstat on this one.
> > > >>
> > > >
> > > > :)
> > > >
> > > > Series Reviewed-by: Jordan Justen
> 
> >  >
> > >  
> >   > >
> > > Thanks. Committed as aa47e5297851..4a392a451d4d.
> > >
> > > > Ray,
> > > >
> > > > I noticed one possible issue. With the new driver, did we
> > lose this
> > > > commit?
> > > >
> > > > commit 1fd376d979221e9765dbec215916f5a6ab92879f
> > > >
> > > > PcAtChipsetPkg/PciHostBridgeDxe: Improve KVM FIFO I/O
> > > read/write performance
> > >
> > > I think so. Apologies for not noticing it. :(
> > >
> > > I hope that Ray can resurrect / reapply the C code directly
> > from the
> > > commit you 

Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork

[Mark]

Thanks for pointing that out, sorry.

On branch ide_perf_fix_v2
Your branch is up-to-date with 'origin/ide_perf_fix_v2'.
nothing to commit, working directory clean

Build success, qemu command line:
qemu-system-x86_64 -enable-kvm -net none -m 1500 -pflash ~/OVMF_CODE.fd
-cdrom /stor/Downloads/X17-58997.iso

Back to full performance while loading the win7 ISO, also the archlinux ISO.

Thanks!

On Fri, Apr 8, 2016 at 11:51 AM, Laszlo Ersek  wrote:

> On 04/08/16 20:36, Mark wrote:
> > I cloned your github but couldn't find the appropriate branch to build.
> >
> > [m@ornage edk2]$ git status
> > On branch remove_old_pci_host
> > Your branch is up-to-date with 'origin/remove_old_pci_host'.
> > nothing to commit, working directory clean
> >
> > If that is the one, I built it successfully but did not solve the
> > performance issue.
> >
> > Should I be in another branch?
>
> Please see the cover letter of the v2 series that I CC'd you on
> (although I forgot to mark the subject as v2):
>
> http://thread.gmane.org/gmane.comp.bios.edk2.devel/10511
>
> It states that the branch name is "ide_perf_fix_v2".
>
> Thanks
> Laszlo
>
> >
> > On Fri, Apr 8, 2016 at 2:08 AM, Laszlo Ersek  > > wrote:
> >
> > On 04/08/16 05:09, Mark wrote:
> > > Yes I'm the same person. :)
> > >
> > > Currently I am using /usr/share/ovmf/x64/ovmf_code_x64.bin built
> from
> > > arch user repo package ovmf-git on 5 Apr 2016.  When I was
> > reporting the
> > > issues, I was using
> > > edk2.git-ovmf-x64-0-20160324.b1635.gf0bbcdf.noarch.rpm from the
> > Gerd repo.
> > >
> > > CPU on the host had a single core pegged during the poor
> performance.
> > >
> > > Multiple distros was referring to the host, sorry for the
> > ambiguity.  I
> > > also tried arch linux on a guest just to see if it was a win7 iso
> > issue,
> > > the xml dump was dumped at a moment I had that arch ISO loaded.
> > >
> > > I just rebuilt ovmf-git 18308.8207d79-1 (Thu Apr  7 20:04:21 PDT
> 2016)
> > > and the issue still exists, I see that it is being addressed.
> > >
> > > Thanks to everyone who contributes to this project, I enjoy the
> > product
> > > very much.  If there is any more info I can provide, please ask.
> >
> > Can you perhaps confirm the fix works? I'll soon send a v2, and push
> it
> > to my personal github repo as well. If you could fetch it, build it,
> and
> > test it, that would be great.
> >
> > Thanks!
> > Laszlo
> >
> > > On Thu, Apr 7, 2016 at 12:36 PM, Laszlo Ersek  > 
> > > >> wrote:
> > >
> > > On 04/07/16 20:24, Jordan Justen wrote:
> > > > On 2016-04-07 08:07:35, Laszlo Ersek wrote:
> > > >> I like the diffstat on this one.
> > > >>
> > > >
> > > > :)
> > > >
> > > > Series Reviewed-by: Jordan Justen  > 
> > >  > >>
> > >
> > > Thanks. Committed as aa47e5297851..4a392a451d4d.
> > >
> > > > Ray,
> > > >
> > > > I noticed one possible issue. With the new driver, did we
> > lose this
> > > > commit?
> > > >
> > > > commit 1fd376d979221e9765dbec215916f5a6ab92879f
> > > >
> > > > PcAtChipsetPkg/PciHostBridgeDxe: Improve KVM FIFO I/O
> > > read/write performance
> > >
> > > I think so. Apologies for not noticing it. :(
> > >
> > > I hope that Ray can resurrect / reapply the C code directly
> > from the
> > > commit you reference above, and the assembly (in NASM syntax)
> from
> > > 6026bf460037 and ace1d0517b65.
> > >
> > > > Laszlo,
> > > >
> > > > Someone was reporting slowness with the Windows 7 ISO on irc
> > a few
> > > > days back. Maybe this is the cause?
> > > >
> > > > I don't think this would impact virtio performance.
> > >
> > > I think you are referring to the following discussion on April
> > 4th:
> > >
> > > >  hello, im having performance issues when using ovmf
> > > >  (OVMF_CODE-pure-efi.fd from gerd hoffman repo) in
> QEMU
> > > >  emulator version 2.3.0 (Debian
> 1:2.3+dfsg-5ubuntu9.2)
> > > >
> > > >  the specific issue im having is slow performance
> > while trying
> > > >  to install win7, the initial screen 'loading
> > windows files'
> > > >  is very slow and im only getting about 1 MB/sec
> disk IO
> > > >  according to iotop
> > > >
> > > >  if i use a VM in bios mode it operates at full
> 

Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork

[Laszlo Ersek]

On 04/08/16 20:36, Mark wrote:
> I cloned your github but couldn't find the appropriate branch to build.
> 
> [m@ornage edk2]$ git status
> On branch remove_old_pci_host
> Your branch is up-to-date with 'origin/remove_old_pci_host'.
> nothing to commit, working directory clean
> 
> If that is the one, I built it successfully but did not solve the
> performance issue.
> 
> Should I be in another branch?

Please see the cover letter of the v2 series that I CC'd you on
(although I forgot to mark the subject as v2):

http://thread.gmane.org/gmane.comp.bios.edk2.devel/10511

It states that the branch name is "ide_perf_fix_v2".

Thanks
Laszlo

> 
> On Fri, Apr 8, 2016 at 2:08 AM, Laszlo Ersek  > wrote:
> 
> On 04/08/16 05:09, Mark wrote:
> > Yes I'm the same person. :)
> >
> > Currently I am using /usr/share/ovmf/x64/ovmf_code_x64.bin built from
> > arch user repo package ovmf-git on 5 Apr 2016.  When I was
> reporting the
> > issues, I was using
> > edk2.git-ovmf-x64-0-20160324.b1635.gf0bbcdf.noarch.rpm from the
> Gerd repo.
> >
> > CPU on the host had a single core pegged during the poor performance.
> >
> > Multiple distros was referring to the host, sorry for the
> ambiguity.  I
> > also tried arch linux on a guest just to see if it was a win7 iso
> issue,
> > the xml dump was dumped at a moment I had that arch ISO loaded.
> >
> > I just rebuilt ovmf-git 18308.8207d79-1 (Thu Apr  7 20:04:21 PDT 2016)
> > and the issue still exists, I see that it is being addressed.
> >
> > Thanks to everyone who contributes to this project, I enjoy the
> product
> > very much.  If there is any more info I can provide, please ask.
> 
> Can you perhaps confirm the fix works? I'll soon send a v2, and push it
> to my personal github repo as well. If you could fetch it, build it, and
> test it, that would be great.
> 
> Thanks!
> Laszlo
> 
> > On Thu, Apr 7, 2016 at 12:36 PM, Laszlo Ersek  
> > >> wrote:
> >
> > On 04/07/16 20:24, Jordan Justen wrote:
> > > On 2016-04-07 08:07:35, Laszlo Ersek wrote:
> > >> I like the diffstat on this one.
> > >>
> > >
> > > :)
> > >
> > > Series Reviewed-by: Jordan Justen  
> >  >>
> >
> > Thanks. Committed as aa47e5297851..4a392a451d4d.
> >
> > > Ray,
> > >
> > > I noticed one possible issue. With the new driver, did we
> lose this
> > > commit?
> > >
> > > commit 1fd376d979221e9765dbec215916f5a6ab92879f
> > >
> > > PcAtChipsetPkg/PciHostBridgeDxe: Improve KVM FIFO I/O
> > read/write performance
> >
> > I think so. Apologies for not noticing it. :(
> >
> > I hope that Ray can resurrect / reapply the C code directly
> from the
> > commit you reference above, and the assembly (in NASM syntax) from
> > 6026bf460037 and ace1d0517b65.
> >
> > > Laszlo,
> > >
> > > Someone was reporting slowness with the Windows 7 ISO on irc
> a few
> > > days back. Maybe this is the cause?
> > >
> > > I don't think this would impact virtio performance.
> >
> > I think you are referring to the following discussion on April
> 4th:
> >
> > >  hello, im having performance issues when using ovmf
> > >  (OVMF_CODE-pure-efi.fd from gerd hoffman repo) in QEMU
> > >  emulator version 2.3.0 (Debian 1:2.3+dfsg-5ubuntu9.2)
> > >
> > >  the specific issue im having is slow performance
> while trying
> > >  to install win7, the initial screen 'loading
> windows files'
> > >  is very slow and im only getting about 1 MB/sec disk IO
> > >  according to iotop
> > >
> > >  if i use a VM in bios mode it operates at full
> performance
> > >
> > >  this behavior is the same across multiple distros
> (ubuntu,
> > >  debian, arch, and gentoo)
> > >
> > >  kvm is enabled and working according to qemu monitor
> > >
> > >  Naruni, what kind of virtual disk are you using?
> Are you
> > >  using virtio-blk or virtio-scsi with OVMF as well?
> > >
> > >  Naruni, also, I guess the best "advice" I can give
> at the
> > >  moment is: "just use the most recent upstream QEMU
> release".
> >
> > Unfortunately, "Naruni" didn't respond (I'll admit that I
> noticed his
> > question only 

Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork

[Mark]

I cloned your github but couldn't find the appropriate branch to build.

[m@ornage edk2]$ git status
On branch remove_old_pci_host
Your branch is up-to-date with 'origin/remove_old_pci_host'.
nothing to commit, working directory clean

If that is the one, I built it successfully but did not solve the
performance issue.

Should I be in another branch?

On Fri, Apr 8, 2016 at 2:08 AM, Laszlo Ersek  wrote:

> On 04/08/16 05:09, Mark wrote:
> > Yes I'm the same person. :)
> >
> > Currently I am using /usr/share/ovmf/x64/ovmf_code_x64.bin built from
> > arch user repo package ovmf-git on 5 Apr 2016.  When I was reporting the
> > issues, I was using
> > edk2.git-ovmf-x64-0-20160324.b1635.gf0bbcdf.noarch.rpm from the Gerd
> repo.
> >
> > CPU on the host had a single core pegged during the poor performance.
> >
> > Multiple distros was referring to the host, sorry for the ambiguity.  I
> > also tried arch linux on a guest just to see if it was a win7 iso issue,
> > the xml dump was dumped at a moment I had that arch ISO loaded.
> >
> > I just rebuilt ovmf-git 18308.8207d79-1 (Thu Apr  7 20:04:21 PDT 2016)
> > and the issue still exists, I see that it is being addressed.
> >
> > Thanks to everyone who contributes to this project, I enjoy the product
> > very much.  If there is any more info I can provide, please ask.
>
> Can you perhaps confirm the fix works? I'll soon send a v2, and push it
> to my personal github repo as well. If you could fetch it, build it, and
> test it, that would be great.
>
> Thanks!
> Laszlo
>
> > On Thu, Apr 7, 2016 at 12:36 PM, Laszlo Ersek  > > wrote:
> >
> > On 04/07/16 20:24, Jordan Justen wrote:
> > > On 2016-04-07 08:07:35, Laszlo Ersek wrote:
> > >> I like the diffstat on this one.
> > >>
> > >
> > > :)
> > >
> > > Series Reviewed-by: Jordan Justen  > >
> >
> > Thanks. Committed as aa47e5297851..4a392a451d4d.
> >
> > > Ray,
> > >
> > > I noticed one possible issue. With the new driver, did we lose this
> > > commit?
> > >
> > > commit 1fd376d979221e9765dbec215916f5a6ab92879f
> > >
> > > PcAtChipsetPkg/PciHostBridgeDxe: Improve KVM FIFO I/O
> > read/write performance
> >
> > I think so. Apologies for not noticing it. :(
> >
> > I hope that Ray can resurrect / reapply the C code directly from the
> > commit you reference above, and the assembly (in NASM syntax) from
> > 6026bf460037 and ace1d0517b65.
> >
> > > Laszlo,
> > >
> > > Someone was reporting slowness with the Windows 7 ISO on irc a few
> > > days back. Maybe this is the cause?
> > >
> > > I don't think this would impact virtio performance.
> >
> > I think you are referring to the following discussion on April 4th:
> >
> > >  hello, im having performance issues when using ovmf
> > >  (OVMF_CODE-pure-efi.fd from gerd hoffman repo) in QEMU
> > >  emulator version 2.3.0 (Debian 1:2.3+dfsg-5ubuntu9.2)
> > >
> > >  the specific issue im having is slow performance while
> trying
> > >  to install win7, the initial screen 'loading windows
> files'
> > >  is very slow and im only getting about 1 MB/sec disk IO
> > >  according to iotop
> > >
> > >  if i use a VM in bios mode it operates at full performance
> > >
> > >  this behavior is the same across multiple distros (ubuntu,
> > >  debian, arch, and gentoo)
> > >
> > >  kvm is enabled and working according to qemu monitor
> > >
> > >  Naruni, what kind of virtual disk are you using? Are you
> > >  using virtio-blk or virtio-scsi with OVMF as well?
> > >
> > >  Naruni, also, I guess the best "advice" I can give at the
> > >  moment is: "just use the most recent upstream QEMU
> release".
> >
> > Unfortunately, "Naruni" didn't respond (I'll admit that I noticed his
> > question only 90 minutes later or so).
> >
> > However, I am now putting this together with an email, on the
> vfio-users
> > list (also April 4th):
> >
> > https://www.redhat.com/archives/vfio-users/2016-April/msg00029.html
> >
> > The description is practically identical, so I'm thinking the same
> user
> > reported it. I'm adding the email address from the latter message to
> the
> > Cc list now. Mark, are you Naruni? :)
> >
> > Unfortunately, the report doesn't name the version of Gerd's OVMF
> RPM in
> > use. It also doesn't say whether the CPU was pegged while
> experiencing
> > the slow performance.
> >
> > The "multiple distros" comment in the IRC report is not clear -- does
> > that mean multiple distros as guests, or hosts? I guess it's the
> former;
> > first, the host is reported as "QEMU emulator version 2.3.0 (Debian
> > 1:2.3+dfsg-5ubuntu9.2)", while 

Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork

[Jordan Justen]

On 2016-04-07 19:53:37, Ni, Ruiyu wrote:
> PciHostBridge driver calls CpuIo to perform IO operation,
> which is different from the previous implementation in
> PcAtChipsetPkg.
> 
> I suggest we put the IO optimization in CpuIo driver.
> It can also benefit all other modules.
> 
> Jeff,
> Do you agree to sync below commit to UefiCpuPkg/CpuIo driver?
> commit 1fd376d979221e9765dbec215916f5a6ab92879f
> PcAtChipsetPkg/PciHostBridgeDxe: Improve KVM FIFO I/O read/write performance
> 

Laszlo ported this in this patch which he posted:

  UefiCpuPkg: CpuIo2Dxe: optimize FIFO reads and writes of IO ports

-Jordan

> 
> >-Original Message-
> >From: Justen, Jordan L
> >Sent: Friday, April 8, 2016 4:50 AM
> >To: Laszlo Ersek <ler...@redhat.com>; Ni, Ruiyu <ruiyu...@intel.com>; Mark 
> ><kram...@gmail.com>
> >Cc: edk2-devel-01 <edk2-devel@lists.01.org>; Gao, Liming 
> ><liming@intel.com>
> >Subject: Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork
> >
> >On 2016-04-07 12:36:36, Laszlo Ersek wrote:
> >> On 04/07/16 20:24, Jordan Justen wrote:
> >> > On 2016-04-07 08:07:35, Laszlo Ersek wrote:
> >> >> I like the diffstat on this one.
> >> >>
> >> >
> >> > :)
> >> >
> >> > Series Reviewed-by: Jordan Justen <jordan.l.jus...@intel.com>
> >>
> >> Thanks. Committed as aa47e5297851..4a392a451d4d.
> >>
> >> > Ray,
> >> >
> >> > I noticed one possible issue. With the new driver, did we lose this
> >> > commit?
> >> >
> >> > commit 1fd376d979221e9765dbec215916f5a6ab92879f
> >> >
> >> > PcAtChipsetPkg/PciHostBridgeDxe: Improve KVM FIFO I/O read/write 
> >> > performance
> >>
> >> I think so. Apologies for not noticing it. :(
> >>
> >> I hope that Ray can resurrect / reapply the C code directly from the
> >> commit you reference above, and the assembly (in NASM syntax) from
> >> 6026bf460037 and ace1d0517b65.
> >>
> >
> >We may have to hold off on the NASM conversion for now, since the
> >driver is in MdeModulePkg. We haven't converted this package to NASM
> >yet.
> >
> >-Jordan
> >
> >> > Laszlo,
> >> >
> >> > Someone was reporting slowness with the Windows 7 ISO on irc a few
> >> > days back. Maybe this is the cause?
> >> >
> >> > I don't think this would impact virtio performance.
> >>
> >> I think you are referring to the following discussion on April 4th:
> >>
> >> >  hello, im having performance issues when using ovmf
> >> >  (OVMF_CODE-pure-efi.fd from gerd hoffman repo) in QEMU
> >> >  emulator version 2.3.0 (Debian 1:2.3+dfsg-5ubuntu9.2)
> >> >
> >> >  the specific issue im having is slow performance while trying
> >> >  to install win7, the initial screen 'loading windows files'
> >> >  is very slow and im only getting about 1 MB/sec disk IO
> >> >  according to iotop
> >> >
> >> >  if i use a VM in bios mode it operates at full performance
> >> >
> >> >  this behavior is the same across multiple distros (ubuntu,
> >> >  debian, arch, and gentoo)
> >> >
> >> >  kvm is enabled and working according to qemu monitor
> >> >
> >> >  Naruni, what kind of virtual disk are you using? Are you
> >> >  using virtio-blk or virtio-scsi with OVMF as well?
> >> >
> >> >  Naruni, also, I guess the best "advice" I can give at the
> >> >  moment is: "just use the most recent upstream QEMU release".
> >>
> >> Unfortunately, "Naruni" didn't respond (I'll admit that I noticed his
> >> question only 90 minutes later or so).
> >>
> >> However, I am now putting this together with an email, on the vfio-users
> >> list (also April 4th):
> >>
> >> https://www.redhat.com/archives/vfio-users/2016-April/msg00029.html
> >>
> >> The description is practically identical, so I'm thinking the same user
> >> reported it. I'm adding the email address from the latter message to the
> >> Cc list now. Mark, are you Naruni? :)
> >>
> >> Unfortunately, the report doesn't name the version of Gerd's OVMF RPM in
> >> use. It also doesn't say whether the CPU was pegged while experiencing
> >> the slow performance.
> >>
> >> The "multiple distros" comment in the IRC report is not clear -- does
> >> that mean multiple distros as guests, or hosts? I guess it's the former;
> >> first, the host is reported as "QEMU emulator version 2.3.0 (Debian
> >> 1:2.3+dfsg-5ubuntu9.2)", while the domain XML pasted into Mark's email
> >> lists an ArchLinux installer ISO (i.e., guest).
> >>
> >> The domain XML clarifies another thing: the CD-ROM is IDE, not
> >> virtio-scsi. Also, there is no other disk drive (of any kind).
> >>
> >> Let me see if I can reproduce this, and perhaps bisect it.
> >>
> >> Mark, if you want to follow up on this list (= edk2-devel), please note
> >> that first you will have to subscribe to the list (and await the
> >> confirmation), at <https://lists.01.org/mailman/listinfo/edk2-devel>.
> >>
> >> Thanks
> >> Laszlo
___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel

Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork

[Jordan Justen]

On 2016-04-07 12:36:36, Laszlo Ersek wrote:
> On 04/07/16 20:24, Jordan Justen wrote:
> > On 2016-04-07 08:07:35, Laszlo Ersek wrote:
> >> I like the diffstat on this one.
> >>
> > 
> > :)
> > 
> > Series Reviewed-by: Jordan Justen 
> 
> Thanks. Committed as aa47e5297851..4a392a451d4d.
> 
> > Ray,
> > 
> > I noticed one possible issue. With the new driver, did we lose this
> > commit?
> > 
> > commit 1fd376d979221e9765dbec215916f5a6ab92879f
> > 
> > PcAtChipsetPkg/PciHostBridgeDxe: Improve KVM FIFO I/O read/write 
> > performance
> 
> I think so. Apologies for not noticing it. :(
> 
> I hope that Ray can resurrect / reapply the C code directly from the
> commit you reference above, and the assembly (in NASM syntax) from
> 6026bf460037 and ace1d0517b65.
> 

We may have to hold off on the NASM conversion for now, since the
driver is in MdeModulePkg. We haven't converted this package to NASM
yet.

-Jordan

> > Laszlo,
> > 
> > Someone was reporting slowness with the Windows 7 ISO on irc a few
> > days back. Maybe this is the cause?
> > 
> > I don't think this would impact virtio performance.
> 
> I think you are referring to the following discussion on April 4th:
> 
> >  hello, im having performance issues when using ovmf
> >  (OVMF_CODE-pure-efi.fd from gerd hoffman repo) in QEMU
> >  emulator version 2.3.0 (Debian 1:2.3+dfsg-5ubuntu9.2)
> >
> >  the specific issue im having is slow performance while trying
> >  to install win7, the initial screen 'loading windows files'
> >  is very slow and im only getting about 1 MB/sec disk IO
> >  according to iotop
> >
> >  if i use a VM in bios mode it operates at full performance
> >
> >  this behavior is the same across multiple distros (ubuntu,
> >  debian, arch, and gentoo)
> >
> >  kvm is enabled and working according to qemu monitor
> >
> >  Naruni, what kind of virtual disk are you using? Are you
> >  using virtio-blk or virtio-scsi with OVMF as well?
> >
> >  Naruni, also, I guess the best "advice" I can give at the
> >  moment is: "just use the most recent upstream QEMU release".
> 
> Unfortunately, "Naruni" didn't respond (I'll admit that I noticed his
> question only 90 minutes later or so).
> 
> However, I am now putting this together with an email, on the vfio-users
> list (also April 4th):
> 
> https://www.redhat.com/archives/vfio-users/2016-April/msg00029.html
> 
> The description is practically identical, so I'm thinking the same user
> reported it. I'm adding the email address from the latter message to the
> Cc list now. Mark, are you Naruni? :)
> 
> Unfortunately, the report doesn't name the version of Gerd's OVMF RPM in
> use. It also doesn't say whether the CPU was pegged while experiencing
> the slow performance.
> 
> The "multiple distros" comment in the IRC report is not clear -- does
> that mean multiple distros as guests, or hosts? I guess it's the former;
> first, the host is reported as "QEMU emulator version 2.3.0 (Debian
> 1:2.3+dfsg-5ubuntu9.2)", while the domain XML pasted into Mark's email
> lists an ArchLinux installer ISO (i.e., guest).
> 
> The domain XML clarifies another thing: the CD-ROM is IDE, not
> virtio-scsi. Also, there is no other disk drive (of any kind).
> 
> Let me see if I can reproduce this, and perhaps bisect it.
> 
> Mark, if you want to follow up on this list (= edk2-devel), please note
> that first you will have to subscribe to the list (and await the
> confirmation), at .
> 
> Thanks
> Laszlo
___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel

Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork

[Laszlo Ersek]

On 04/07/16 21:36, Laszlo Ersek wrote:
> On 04/07/16 20:24, Jordan Justen wrote:

>> Someone was reporting slowness with the Windows 7 ISO on irc a few
>> days back. Maybe this is the cause?

> I think you are referring to the following discussion on April 4th:

>>  if i use a VM in bios mode it operates at full performance

> The domain XML clarifies another thing: the CD-ROM is IDE, not
> virtio-scsi. Also, there is no other disk drive (of any kind).

Sigh, this doesn't look good. If you look at the SeaBIOS source, it has 
ata_pio_transfer() in "src/hw/ata.c", doing:

for (;;) {
if (iswrite) {
// Write data to controller
dprintf(16, "Write sector id=%p dest=%p\n", op->drive_gf, buf_fl);
if (CONFIG_ATA_PIO32)
outsl_fl(iobase1, buf_fl, blocksize / 4);
else
outsw_fl(iobase1, buf_fl, blocksize / 2);
} else {
// Read data from controller
dprintf(16, "Read sector id=%p dest=%p\n", op->drive_gf, buf_fl);
if (CONFIG_ATA_PIO32)
insl_fl(iobase1, buf_fl, blocksize / 4);
else
insw_fl(iobase1, buf_fl, blocksize / 2);
}
buf_fl += blocksize;

The outsl_fl / outsw_fl / insl_fl / insw_fl functions are forwarded by 
"src/farptr.h" to primitives in "src/x86.h" (without the _fl suffix). Those 
primitives *do* use inline assembly with the REP prefix.

... Right, I can reproduce the issue. Booting a Fedora LiveCD from an IDE 
CD-ROM takes ages, and the host CPU is pegged, in system load. :( :( :(

Let me see if I can come up with a patch for this. Good catch, Jordan!

Laszlo
___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel

Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork

[Laszlo Ersek]

On 04/07/16 20:24, Jordan Justen wrote:
> On 2016-04-07 08:07:35, Laszlo Ersek wrote:
>> I like the diffstat on this one.
>>
> 
> :)
> 
> Series Reviewed-by: Jordan Justen 

Thanks. Committed as aa47e5297851..4a392a451d4d.

> Ray,
> 
> I noticed one possible issue. With the new driver, did we lose this
> commit?
> 
> commit 1fd376d979221e9765dbec215916f5a6ab92879f
> 
> PcAtChipsetPkg/PciHostBridgeDxe: Improve KVM FIFO I/O read/write 
> performance

I think so. Apologies for not noticing it. :(

I hope that Ray can resurrect / reapply the C code directly from the
commit you reference above, and the assembly (in NASM syntax) from
6026bf460037 and ace1d0517b65.

> Laszlo,
> 
> Someone was reporting slowness with the Windows 7 ISO on irc a few
> days back. Maybe this is the cause?
> 
> I don't think this would impact virtio performance.

I think you are referring to the following discussion on April 4th:

>  hello, im having performance issues when using ovmf
>  (OVMF_CODE-pure-efi.fd from gerd hoffman repo) in QEMU
>  emulator version 2.3.0 (Debian 1:2.3+dfsg-5ubuntu9.2)
>
>  the specific issue im having is slow performance while trying
>  to install win7, the initial screen 'loading windows files'
>  is very slow and im only getting about 1 MB/sec disk IO
>  according to iotop
>
>  if i use a VM in bios mode it operates at full performance
>
>  this behavior is the same across multiple distros (ubuntu,
>  debian, arch, and gentoo)
>
>  kvm is enabled and working according to qemu monitor
>
>  Naruni, what kind of virtual disk are you using? Are you
>  using virtio-blk or virtio-scsi with OVMF as well?
>
>  Naruni, also, I guess the best "advice" I can give at the
>  moment is: "just use the most recent upstream QEMU release".

Unfortunately, "Naruni" didn't respond (I'll admit that I noticed his
question only 90 minutes later or so).

However, I am now putting this together with an email, on the vfio-users
list (also April 4th):

https://www.redhat.com/archives/vfio-users/2016-April/msg00029.html

The description is practically identical, so I'm thinking the same user
reported it. I'm adding the email address from the latter message to the
Cc list now. Mark, are you Naruni? :)

Unfortunately, the report doesn't name the version of Gerd's OVMF RPM in
use. It also doesn't say whether the CPU was pegged while experiencing
the slow performance.

The "multiple distros" comment in the IRC report is not clear -- does
that mean multiple distros as guests, or hosts? I guess it's the former;
first, the host is reported as "QEMU emulator version 2.3.0 (Debian
1:2.3+dfsg-5ubuntu9.2)", while the domain XML pasted into Mark's email
lists an ArchLinux installer ISO (i.e., guest).

The domain XML clarifies another thing: the CD-ROM is IDE, not
virtio-scsi. Also, there is no other disk drive (of any kind).

Let me see if I can reproduce this, and perhaps bisect it.

Mark, if you want to follow up on this list (= edk2-devel), please note
that first you will have to subscribe to the list (and await the
confirmation), at .

Thanks
Laszlo
___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel

Re: [edk2] [PATCH 0/2] OvmfPkg: drop PciHostBridgeDxe fork

[Jordan Justen]

On 2016-04-07 08:07:35, Laszlo Ersek wrote:
> I like the diffstat on this one.
> 

:)

Series Reviewed-by: Jordan Justen 

Ray,

I noticed one possible issue. With the new driver, did we lose this
commit?

commit 1fd376d979221e9765dbec215916f5a6ab92879f

PcAtChipsetPkg/PciHostBridgeDxe: Improve KVM FIFO I/O read/write performance

Laszlo,

Someone was reporting slowness with the Windows 7 ISO on irc a few
days back. Maybe this is the cause?

I don't think this would impact virtio performance.

-Jordan

> Public branch:
> .
> 
> Cc: Jordan Justen 
> 
> Laszlo Ersek (2):
>   OvmfPkg: remove USE_OLD_PCI_HOST build option
>   OvmfPkg: remove PciHostBridgeDxe fork
> 
>  OvmfPkg/OvmfPkgIa32.dsc   |   19 -
>  OvmfPkg/OvmfPkgIa32X64.dsc|   19 -
>  OvmfPkg/OvmfPkgX64.dsc|   19 -
>  OvmfPkg/OvmfPkgIa32.fdf   |4 -
>  OvmfPkg/OvmfPkgIa32X64.fdf|4 -
>  OvmfPkg/OvmfPkgX64.fdf|4 -
>  OvmfPkg/PciHostBridgeDxe/PciHostBridgeDxe.inf |   64 -
>  OvmfPkg/PciHostBridgeDxe/IoFifo.h |  176 --
>  OvmfPkg/PciHostBridgeDxe/PciHostBridge.h  |  651 -
>  OvmfPkg/PciHostBridgeDxe/PciHostBridge.c  | 1551 
>  OvmfPkg/PciHostBridgeDxe/PciRootBridgeIo.c| 2628 
>  OvmfPkg/PciHostBridgeDxe/Ia32/IoFifo.nasm |  136 -
>  OvmfPkg/PciHostBridgeDxe/X64/IoFifo.nasm  |  125 -
>  13 files changed, 5400 deletions(-)
>  delete mode 100644 OvmfPkg/PciHostBridgeDxe/PciHostBridgeDxe.inf
>  delete mode 100644 OvmfPkg/PciHostBridgeDxe/IoFifo.h
>  delete mode 100644 OvmfPkg/PciHostBridgeDxe/PciHostBridge.h
>  delete mode 100644 OvmfPkg/PciHostBridgeDxe/PciHostBridge.c
>  delete mode 100644 OvmfPkg/PciHostBridgeDxe/PciRootBridgeIo.c
>  delete mode 100644 OvmfPkg/PciHostBridgeDxe/Ia32/IoFifo.nasm
>  delete mode 100644 OvmfPkg/PciHostBridgeDxe/X64/IoFifo.nasm
> 
> -- 
> 1.8.3.1
> 
> ___
> edk2-devel mailing list
> edk2-devel@lists.01.org
> https://lists.01.org/mailman/listinfo/edk2-devel
___
edk2-devel mailing list
edk2-devel@lists.01.org
https://lists.01.org/mailman/listinfo/edk2-devel