This adds Kconfig option INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON to make it easier for distributions to enable or disable the Intel IOMMU scalable mode by default during kernel build. Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com> --- drivers/iommu/Kconfig | 9 +++++++++ drivers/iommu/intel-iommu.c | 7 ++++++- 2 files changed, 15 insertions(+), 1 deletion(-) diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig index e3842eabcfdd..fbdf3fd291d9 100644 --- a/drivers/iommu/Kconfig +++ b/drivers/iommu/Kconfig @@ -242,6 +242,15 @@ config INTEL_IOMMU_FLOPPY_WA workaround will setup a 1:1 mapping for the first 16MiB to make floppy (an ISA device) work. +config INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON + prompt "Enable Intel IOMMU scalable mode by default" + depends on INTEL_IOMMU + help + Selecting this option will enable the scalable mode if + hardware presents the capability. If this option is not + selected, scalable mode support could also be enabled + by passing intel_iommu=sm_on to the kernel. + config IRQ_REMAP bool "Support for Interrupt Remapping" depends on X86_64 && X86_IO_APIC && PCI_MSI && ACPI diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c index 6db6d969e31c..6051fe790c61 100644 --- a/drivers/iommu/intel-iommu.c +++ b/drivers/iommu/intel-iommu.c @@ -355,9 +355,14 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain, int dmar_disabled = 0; #else int dmar_disabled = 1; -#endif /*CONFIG_INTEL_IOMMU_DEFAULT_ON*/ +#endif /* CONFIG_INTEL_IOMMU_DEFAULT_ON */ +#ifdef INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON +int intel_iommu_sm = 1; +#else int intel_iommu_sm; +#endif /* INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON */ + int intel_iommu_enabled = 0; EXPORT_SYMBOL_GPL(intel_iommu_enabled); -- 2.17.1 _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
> On Nov 8, 2019, at 10:40 PM, Lu Baolu <baolu.lu@linux.intel.com> wrote: > > This adds Kconfig option INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON > to make it easier for distributions to enable or disable the > Intel IOMMU scalable mode by default during kernel build. > > Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com> > --- > drivers/iommu/Kconfig | 9 +++++++++ > drivers/iommu/intel-iommu.c | 7 ++++++- > 2 files changed, 15 insertions(+), 1 deletion(-) > > diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig > index e3842eabcfdd..fbdf3fd291d9 100644 > --- a/drivers/iommu/Kconfig > +++ b/drivers/iommu/Kconfig > @@ -242,6 +242,15 @@ config INTEL_IOMMU_FLOPPY_WA > workaround will setup a 1:1 mapping for the first > 16MiB to make floppy (an ISA device) work. > > +config INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON > + prompt "Enable Intel IOMMU scalable mode by default" > + depends on INTEL_IOMMU > + help > + Selecting this option will enable the scalable mode if > + hardware presents the capability. If this option is not > + selected, scalable mode support could also be enabled > + by passing intel_iommu=sm_on to the kernel. Maybe a sentence or two to describe what the scalable mode is in layman's terms could be useful, so developers don’t need to search around for the Kconfig selection? > + > config IRQ_REMAP > bool "Support for Interrupt Remapping" > depends on X86_64 && X86_IO_APIC && PCI_MSI && ACPI > diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c > index 6db6d969e31c..6051fe790c61 100644 > --- a/drivers/iommu/intel-iommu.c > +++ b/drivers/iommu/intel-iommu.c > @@ -355,9 +355,14 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain, > int dmar_disabled = 0; > #else > int dmar_disabled = 1; > -#endif /*CONFIG_INTEL_IOMMU_DEFAULT_ON*/ > +#endif /* CONFIG_INTEL_IOMMU_DEFAULT_ON */ > > +#ifdef INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON > +int intel_iommu_sm = 1; > +#else > int intel_iommu_sm; > +#endif /* INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON */ > + > int intel_iommu_enabled = 0; > EXPORT_SYMBOL_GPL(intel_iommu_enabled); > > -- > 2.17.1 > > _______________________________________________ > iommu mailing list > iommu@lists.linux-foundation.org > https://lists.linuxfoundation.org/mailman/listinfo/iommu _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
Hi, On 11/9/19 11:59 AM, Qian Cai wrote: > > >> On Nov 8, 2019, at 10:40 PM, Lu Baolu <baolu.lu@linux.intel.com> wrote: >> >> This adds Kconfig option INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON >> to make it easier for distributions to enable or disable the >> Intel IOMMU scalable mode by default during kernel build. >> >> Signed-off-by: Lu Baolu <baolu.lu@linux.intel.com> >> --- >> drivers/iommu/Kconfig | 9 +++++++++ >> drivers/iommu/intel-iommu.c | 7 ++++++- >> 2 files changed, 15 insertions(+), 1 deletion(-) >> >> diff --git a/drivers/iommu/Kconfig b/drivers/iommu/Kconfig >> index e3842eabcfdd..fbdf3fd291d9 100644 >> --- a/drivers/iommu/Kconfig >> +++ b/drivers/iommu/Kconfig >> @@ -242,6 +242,15 @@ config INTEL_IOMMU_FLOPPY_WA >> workaround will setup a 1:1 mapping for the first >> 16MiB to make floppy (an ISA device) work. >> >> +config INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON >> + prompt "Enable Intel IOMMU scalable mode by default" >> + depends on INTEL_IOMMU >> + help >> + Selecting this option will enable the scalable mode if >> + hardware presents the capability. If this option is not >> + selected, scalable mode support could also be enabled >> + by passing intel_iommu=sm_on to the kernel. > > > Maybe a sentence or two to describe what the scalable mode is in layman's > terms could be useful, so developers don’t need to search around for the > Kconfig selection? How about "pasid based multiple stages DMA translation"? Best regards, baolu > >> + >> config IRQ_REMAP >> bool "Support for Interrupt Remapping" >> depends on X86_64 && X86_IO_APIC && PCI_MSI && ACPI >> diff --git a/drivers/iommu/intel-iommu.c b/drivers/iommu/intel-iommu.c >> index 6db6d969e31c..6051fe790c61 100644 >> --- a/drivers/iommu/intel-iommu.c >> +++ b/drivers/iommu/intel-iommu.c >> @@ -355,9 +355,14 @@ static phys_addr_t intel_iommu_iova_to_phys(struct iommu_domain *domain, >> int dmar_disabled = 0; >> #else >> int dmar_disabled = 1; >> -#endif /*CONFIG_INTEL_IOMMU_DEFAULT_ON*/ >> +#endif /* CONFIG_INTEL_IOMMU_DEFAULT_ON */ >> >> +#ifdef INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON >> +int intel_iommu_sm = 1; >> +#else >> int intel_iommu_sm; >> +#endif /* INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON */ >> + >> int intel_iommu_enabled = 0; >> EXPORT_SYMBOL_GPL(intel_iommu_enabled); >> >> -- >> 2.17.1 >> >> _______________________________________________ >> iommu mailing list >> iommu@lists.linux-foundation.org >> https://lists.linuxfoundation.org/mailman/listinfo/iommu > > _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
> On Nov 10, 2019, at 8:30 PM, Lu Baolu <baolu.lu@linux.intel.com> wrote: > > How about "pasid based multiple stages DMA translation"? It is better but I am still not sure how developers should select it or not when asking. Ideally, should it mention pros and cons of this? At minimal, there should be a line said “if not sure what this is, select N”? _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
Hi, On 11/11/19 10:00 AM, Qian Cai wrote: > > >> On Nov 10, 2019, at 8:30 PM, Lu Baolu <baolu.lu@linux.intel.com> wrote: >> >> How about "pasid based multiple stages DMA translation"? > > It is better but I am still not sure how developers should select it or not when asking. Ideally, should it mention pros and cons of this? At minimal, there should be a line said “if not sure what this is, select N”? > Actually, I'd recommend "if not sure, use the default value". :-) Best regards, baolu _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
> On Nov 8, 2019, at 10:43 PM, Lu Baolu <baolu.lu@linux.intel.com> wrote: > > +config INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON > + prompt "Enable Intel IOMMU scalable mode by default" > + depends on INTEL_IOMMU > + help > + Selecting this option will enable the scalable mode if > + hardware presents the capability. If this option is not > + selected, scalable mode support could also be enabled > + by passing intel_iommu=sm_on to the kernel. > + Does it also make sense to mention which hardware presents this capability or how to check it? _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
Hi, On 11/11/19 10:58 AM, Qian Cai wrote: > > >> On Nov 8, 2019, at 10:43 PM, Lu Baolu <baolu.lu@linux.intel.com> wrote: >> >> +config INTEL_IOMMU_SCALABLE_MODE_DEFAULT_ON >> + prompt "Enable Intel IOMMU scalable mode by default" >> + depends on INTEL_IOMMU >> + help >> + Selecting this option will enable the scalable mode if >> + hardware presents the capability. If this option is not >> + selected, scalable mode support could also be enabled >> + by passing intel_iommu=sm_on to the kernel. >> + > > Does it also make sense to mention which hardware presents this capability or how to check it? > The scalable mode is defined in VT-d 3.0. The scalable mode capability could be checked by reading /sys/devices/virtual/iommu/dmar*/intel- iommu/ecap. It's currently not friendly for reading. You need to decode it according to the spec. Best regards, baolu _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
> On Nov 11, 2019, at 12:23 AM, Lu Baolu <baolu.lu@linux.intel.com> wrote: > > The scalable mode is defined in VT-d 3.0. The scalable mode capability > could be checked by reading /sys/devices/virtual/iommu/dmar*/intel- > iommu/ecap. It's currently not friendly for reading. You need to decode > it according to the spec. This looks like some perfect information to put in the Kconfig description. _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu
Hi, On 11/11/19 10:05 PM, Qian Cai wrote: > > >> On Nov 11, 2019, at 12:23 AM, Lu Baolu <baolu.lu@linux.intel.com> wrote: >> >> The scalable mode is defined in VT-d 3.0. The scalable mode capability >> could be checked by reading /sys/devices/virtual/iommu/dmar*/intel- >> iommu/ecap. It's currently not friendly for reading. You need to decode >> it according to the spec. > > This looks like some perfect information to put in the Kconfig description. > Okay. Will add it in the next version. Best regards, baolu _______________________________________________ iommu mailing list iommu@lists.linux-foundation.org https://lists.linuxfoundation.org/mailman/listinfo/iommu