Re: [PATCH v2 0/4] SdMmcOverride extension

[Marcin Wojtas <mw@semihalf.com>]

pt., 12 paź 2018 o 14:48 Wu, Hao A <hao.a.wu@intel.com> napisał(a):
>
> > -----Original Message-----
> > From: Marcin Wojtas [mailto:mw@semihalf.com]
> > Sent: Friday, October 12, 2018 1:33 PM
> > To: Wu, Hao A
> > Cc: edk2-devel-01; Tian, Feng; Kinney, Michael D; Gao, Liming; Leif Lindholm;
> > Ard Biesheuvel; nadavh@marvell.com; jsd@semihalf.com; Tomasz Michalec
> > Subject: Re: [PATCH v2 0/4] SdMmcOverride extension
> >
> > Hi Hao,
> >
> > pt., 12 paź 2018 o 07:25 Wu, Hao A <hao.a.wu@intel.com> napisał(a):
> > >
> > > Hi Marcin,
> > >
> > > Please grant me some time for this series.
> > >
> > > Since I found that the extension of the SdMmc override protocol (mainly
> > > the 3rd and 4th patch of the series) may have something overlaps with a
> > > (internal) request to configure the driver strength parameter and operating
> > > clock frequency of the SD/EMMC devices.
> > >
> > > For the (driver strength/operating freq) customize, we already have a
> > > proposal on the way. So I am wondering if you could grant me some time
> > to
> > > investigate whether both the cases can be addressed together based on
> > your
> > > proposed patch.
> > >
> >
> > Sure. I'm only wondering if it's not best to collect all remarks and
> > maybe update to v3 both edk2 and edk2-platforms sides (so far the
> > issues have been not critical, such as typos, parameters' names,
> > etc.). In the meantime you would be able to validate if the solution
> > is sufficient for you as well. What do you think? When do you expect
> > to be able to look at it vs your internal requirements more deeply?
> >
> > Best regards,
> > Marcin
>
> I think you can hold the new version of the patch if the feedbacks do not
> lead to considerable changes. At this moment, I can barely take time for
> the evaluation. I think I will be able to fully shift to this in about 2
> weeks. Does it sound acceptable to you with regard to the urgency level
> for the series?
>
> I will try my best to move up the process. Sorry again for the possible
> delay.

I will proceed with other remaining items for my platforms and allow
myself to ping you about status around end of October :)

Best regards,
Marcin

>
> >
> >
> > > Thanks in advance.
> > >
> > > Best Regards,
> > > Hao Wu
> > >
> > >
> > > > -----Original Message-----
> > > > From: Marcin Wojtas [mailto:mw@semihalf.com]
> > > > Sent: Friday, October 05, 2018 9:25 PM
> > > > To: edk2-devel@lists.01.org
> > > > Cc: Tian, Feng; Kinney, Michael D; Gao, Liming; leif.lindholm@linaro.org;
> > Wu,
> > > > Hao A; ard.biesheuvel@linaro.org; nadavh@marvell.com;
> > > > mw@semihalf.com; jsd@semihalf.com; tm@semihalf.com
> > > > Subject: [PATCH v2 0/4] SdMmcOverride extension
> > > >
> > > > Hi,
> > > >
> > > > This is the second version of the patchset. Initial one was
> > > > interleaved with the fixes, which after split got already merged.
> > > > The biggest change is - resigning from the new callbacks
> > > > and extending parameter lists of both NotifyPhase and Capability
> > > > routines.
> > > >
> > > > Patches are available in the github:
> > > > https://github.com/MarvellEmbeddedProcessors/edk2-open-
> > > > platform/commits/sdmmc-override-upstream-r20181005
> > > >
> > > > Please note that extending SdMmcOverride protocol was impacting
> > > > so far the only user of it (Synquacer controller). In paralel
> > > > edk2-platforms patchset, a patch can be found:
> > > > ("Silicon/SynQuacer/PlatformDxe: adjust to updated SdMmcOverride")
> > > > which immunizes for above and future extensions of the protocol:
> > > > https://github.com/MarvellEmbeddedProcessors/edk2-open-
> > > > platform/commits/xenon-upstream-r20181005
> > > >
> > > > I'm looking forward to the comments and remarks.
> > > >
> > > > Best regards,
> > > > Marcin
> > > >
> > > > Changelog:
> > > > v1 -> v2
> > > > * Rebase onto newest master
> > > > * 1/4 [new patch] - preparation for extending NotifyPhase
> > > > * 2/4 - UhsSignaling as a part of NotifyPhase
> > > > * 3/4 - SwitchClockFreqPost as a part of NotifyPhase
> > > > * 4/4 - Allow updating BaseClkFreq via Capability instead of the
> > > >         independent callback.
> > > >
> > > > Marcin Wojtas (2):
> > > >   MdeModulePkg/SdMmcPciHcDxe: Add an optional parameter in
> > > > NotifyPhase
> > > >   MdeModulePkg/SdMmcPciHcDxe: Allow overriding base clock frequency
> > > >
> > > > Tomasz Michalec (2):
> > > >   MdeModulePkg/SdMmcPciHcDxe: Add UhsSignaling to SdMmcOverride
> > > > protocol
> > > >   MdeModulePkg/SdMmcPciHcDxe: Add SwitchClockFreqPost to
> > > > SdMmcOverride
> > > >
> > > >  MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdMmcPciHcDxe.h |   6 +
> > > >  MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdMmcPciHci.h   |  62 +++++-
> > > >  MdeModulePkg/Include/Protocol/SdMmcOverride.h      |  12 +-
> > > >  MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/EmmcDevice.c    | 215
> > > > ++++++++++++++------
> > > >  MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdDevice.c      |  57 +++++-
> > > >  MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdMmcPciHcDxe.c |  18 +-
> > > >  MdeModulePkg/Bus/Pci/SdMmcPciHcDxe/SdMmcPciHci.c   | 108
> > ++++++++-
> > > > -
> > > >  7 files changed, 383 insertions(+), 95 deletions(-)
> > > >
> > > > --
> > > > 2.7.4
> > >