On Thu, Jul 23, 2020 at 1:19 AM Bjorn Helgaas <helgaas@xxxxxxxxxx> wrote: > > On Wed, Jul 22, 2020 at 02:15:13AM +0200, Julia Suvorova wrote: > > Scanning for PCI devices at boot takes a long time for KVM guests. It > > can be reduced if KVM will handle all configuration space accesses for > > non-existent devices without going to userspace [1]. But for this to > > work, all accesses must go through MMCONFIG. > > This change allows to use pci_mmcfg as raw_pci_ops for 64-bit KVM > > guests making MMCONFIG the default access method. > > The above *looks* like it's intended to be two paragraphs, which would > be easier to read with a blank line between. > > The last sentence should say what the patch actually *does*, e.g., > "Use pci_mmcfg as raw_pci_ops ..." > > > [1] https://lkml.org/lkml/2020/5/14/936 > > Please use a lore.kernel.org URL instead because it's more usable and > I'd rather depend on kernel.org than lkml.org. > > > Signed-off-by: Julia Suvorova <jusual@xxxxxxxxxx> > > --- > > arch/x86/pci/direct.c | 5 +++++ > > arch/x86/pci/mmconfig_64.c | 3 +++ > > 2 files changed, 8 insertions(+) > > > > diff --git a/arch/x86/pci/direct.c b/arch/x86/pci/direct.c > > index a51074c55982..8ff6b65d8f48 100644 > > --- a/arch/x86/pci/direct.c > > +++ b/arch/x86/pci/direct.c > > @@ -6,6 +6,7 @@ > > #include <linux/pci.h> > > #include <linux/init.h> > > #include <linux/dmi.h> > > +#include <linux/kvm_para.h> > > #include <asm/pci_x86.h> > > > > /* > > @@ -264,6 +265,10 @@ void __init pci_direct_init(int type) > > { > > if (type == 0) > > return; > > + > > + if (raw_pci_ext_ops && kvm_para_available()) > > + return; > > printk(KERN_INFO "PCI: Using configuration type %d for base access\n", > > type); > > if (type == 1) { > > diff --git a/arch/x86/pci/mmconfig_64.c b/arch/x86/pci/mmconfig_64.c > > index 0c7b6e66c644..9eb772821766 100644 > > --- a/arch/x86/pci/mmconfig_64.c > > +++ b/arch/x86/pci/mmconfig_64.c > > @@ -10,6 +10,7 @@ > > #include <linux/init.h> > > #include <linux/acpi.h> > > #include <linux/bitmap.h> > > +#include <linux/kvm_para.h> > > #include <linux/rcupdate.h> > > #include <asm/e820/api.h> > > #include <asm/pci_x86.h> > > @@ -122,6 +123,8 @@ int __init pci_mmcfg_arch_init(void) > > } > > > > raw_pci_ext_ops = &pci_mmcfg; > > + if (kvm_para_available()) > > + raw_pci_ops = &pci_mmcfg; > > The idea of using MMCONFIG for *all* config space, not just extended > config space, makes sense to me, although the very long discussion at > https://lore.kernel.org/lkml/20071225032605.29147200@xxxxxxxxxxxxxxxxxxxxx/ > makes me wary. Of course I realize you're talking specifically about > KVM, not doing this in general. > > But it doesn't seem right to make this specific to KVM, since it's not > obvious to me that there's a basis in PCI for making this distinction. Bugs that were fixed (or more accurately, avoided) by a0ca99096094 ("PCI x86: always use conf1 to access config space below 256 bytes") are still present. And to enable MMCONFIG for the entire config space, we need to re-introduce all these fixes or at least identify affected devices, which may be impossible. We can avoid KVM-specific changes in the generic PCI code by implementing x86_init.pci.arch_init inside KVM code, as Vitaly suggested. What do you think? Best regards, Julia Suvorova.
