| Message ID | 1800633.tdWV9SEqCh@kreacher (mailing list archive) |
|---|---|
| Headers | show |
| Series | PCI: ACPI: Get rid of struct pci_platform_pm_ops and clean up code | expand |
[+cc Ferry] On Mon, Sep 20, 2021 at 08:52:19PM +0200, Rafael J. Wysocki wrote: > Hi All, > > As explained in the changelog of patch [2/7], using struct pci_platform_pm_ops > for ACPI is not particularly beneficial, so it is better to get rid of it and > call the functions pointed to by it directly from the PCI core. > > However, struct pci_platform_pm_ops is also used by the Intel MID support code, > but it is actually better to call the MID PM function directly from the PCI > core either, which is done in patch [1/7]. > > After these changes, patch [3/7] removes struct pci_platform_pm_ops and the > rest is just cleanups and some code consolidation on top of that. I like these a lot. Not sure exactly where everything is after the conversation with Ferry. Let me know if I should be doing anything.
On Wed, Sep 29, 2021 at 1:28 AM Bjorn Helgaas <helgaas@kernel.org> wrote: > > [+cc Ferry] > > On Mon, Sep 20, 2021 at 08:52:19PM +0200, Rafael J. Wysocki wrote: > > Hi All, > > > > As explained in the changelog of patch [2/7], using struct pci_platform_pm_ops > > for ACPI is not particularly beneficial, so it is better to get rid of it and > > call the functions pointed to by it directly from the PCI core. > > > > However, struct pci_platform_pm_ops is also used by the Intel MID support code, > > but it is actually better to call the MID PM function directly from the PCI > > core either, which is done in patch [1/7]. > > > > After these changes, patch [3/7] removes struct pci_platform_pm_ops and the > > rest is just cleanups and some code consolidation on top of that. > > I like these a lot. Not sure exactly where everything is after the > conversation with Ferry. It's mostly OK, the problem was in one of the "tail" patches that was not rebased properly. There will be a follow-up series to test for Ferry (later today). > Let me know if I should be doing anything. I'm going to take this lot if that's not a problem. If I need anything from you, I'll let you know.
On Wed, Sep 29, 2021 at 02:00:59PM +0200, Rafael J. Wysocki wrote: > On Wed, Sep 29, 2021 at 1:28 AM Bjorn Helgaas <helgaas@kernel.org> wrote: > > > > [+cc Ferry] > > > > On Mon, Sep 20, 2021 at 08:52:19PM +0200, Rafael J. Wysocki wrote: > > > Hi All, > > > > > > As explained in the changelog of patch [2/7], using struct pci_platform_pm_ops > > > for ACPI is not particularly beneficial, so it is better to get rid of it and > > > call the functions pointed to by it directly from the PCI core. > > > > > > However, struct pci_platform_pm_ops is also used by the Intel MID support code, > > > but it is actually better to call the MID PM function directly from the PCI > > > core either, which is done in patch [1/7]. > > > > > > After these changes, patch [3/7] removes struct pci_platform_pm_ops and the > > > rest is just cleanups and some code consolidation on top of that. > > > > I like these a lot. Not sure exactly where everything is after the > > conversation with Ferry. > > It's mostly OK, the problem was in one of the "tail" patches that was > not rebased properly. > > There will be a follow-up series to test for Ferry (later today). > > > Let me know if I should be doing anything. > > I'm going to take this lot if that's not a problem. If I need > anything from you, I'll let you know. Sounds good, thanks, Rafael! Bjorn
Hi Op 29-09-2021 om 01:28 schreef Bjorn Helgaas: > [+cc Ferry] > > On Mon, Sep 20, 2021 at 08:52:19PM +0200, Rafael J. Wysocki wrote: >> Hi All, >> >> As explained in the changelog of patch [2/7], using struct pci_platform_pm_ops >> for ACPI is not particularly beneficial, so it is better to get rid of it and >> call the functions pointed to by it directly from the PCI core. >> >> However, struct pci_platform_pm_ops is also used by the Intel MID support code, >> but it is actually better to call the MID PM function directly from the PCI >> core either, which is done in patch [1/7]. >> >> After these changes, patch [3/7] removes struct pci_platform_pm_ops and the >> rest is just cleanups and some code consolidation on top of that. > > I like these a lot. Not sure exactly where everything is after the > conversation with Ferry. Let me know if I should be doing anything. > I will happily retest likely on Sunday after I return from short holiday and report back here.