diff mbox

[dpdk-dev] pci: get IOMMU class sPAPR iommu fix

Message ID 1509710719-17741-1-git-send-email-jpf@zurich.ibm.com (mailing list archive)
State Accepted, archived
Headers

Checks

Context Check Description
ci/checkpatch success coding style OK
ci/Intel-compilation success Compilation OK

Commit Message

Jonas Pfefferle1 Nov. 3, 2017, 12:05 p.m. UTC 
  PPC64 sPAPR iommu does not support iova as va.
Use pa mode instead.

Signed-off-by: Jonas Pfefferle <jpf@zurich.ibm.com>
---
 drivers/bus/pci/linux/pci.c | 11 ++++++++++-
 1 file changed, 10 insertions(+), 1 deletion(-)

Comments

Thomas Monjalon Nov. 6, 2017, 11:38 p.m. UTC | #1 
03/11/2017 13:05, Jonas Pfefferle:
> PPC64 sPAPR iommu does not support iova as va.
> Use pa mode instead.
> 
> Signed-off-by: Jonas Pfefferle <jpf@zurich.ibm.com>

Applied, thanks
Jonas Pfefferle1 Nov. 7, 2017, 2:31 p.m. UTC | #2 
Thomas Monjalon <thomas@monjalon.net> wrote on 11/07/2017 12:38:11 AM:

> From: Thomas Monjalon <thomas@monjalon.net>
> To: Jonas Pfefferle <jpf@zurich.ibm.com>
> Cc: dev@dpdk.org, anatoly.burakov@intel.com
> Date: 11/07/2017 12:38 AM
> Subject: Re: [dpdk-dev] [PATCH] pci: get IOMMU class sPAPR iommu fix
>
> 03/11/2017 13:05, Jonas Pfefferle:
> > PPC64 sPAPR iommu does not support iova as va.
> > Use pa mode instead.
> >
> > Signed-off-by: Jonas Pfefferle <jpf@zurich.ibm.com>
>
> Applied, thanks
>

Hi Thomas,

I just noticed that I send in an old patch by mistake.
It uses the wrong define RTE_ARCH_PPC64 instead of RTE_ARCH_PPC_64.
Sorry for this, I will send in a fix for this in the next hour.

Thanks,
Jonas
Thomas Monjalon Nov. 7, 2017, 2:37 p.m. UTC | #3 
07/11/2017 15:31, Jonas Pfefferle1:
> Thomas Monjalon <thomas@monjalon.net> wrote on 11/07/2017 12:38:11 AM:
> 
> > From: Thomas Monjalon <thomas@monjalon.net>
> > To: Jonas Pfefferle <jpf@zurich.ibm.com>
> > Cc: dev@dpdk.org, anatoly.burakov@intel.com
> > Date: 11/07/2017 12:38 AM
> > Subject: Re: [dpdk-dev] [PATCH] pci: get IOMMU class sPAPR iommu fix
> >
> > 03/11/2017 13:05, Jonas Pfefferle:
> > > PPC64 sPAPR iommu does not support iova as va.
> > > Use pa mode instead.
> > >
> > > Signed-off-by: Jonas Pfefferle <jpf@zurich.ibm.com>
> >
> > Applied, thanks
> >
> 
> Hi Thomas,
> 
> I just noticed that I send in an old patch by mistake.
> It uses the wrong define RTE_ARCH_PPC64 instead of RTE_ARCH_PPC_64.
> Sorry for this, I will send in a fix for this in the next hour.

OK, thanks for the notice
diff mbox

Patch

diff --git a/drivers/bus/pci/linux/pci.c b/drivers/bus/pci/linux/pci.c
index cdf8106..6a58c5f 100644
--- a/drivers/bus/pci/linux/pci.c
+++ b/drivers/bus/pci/linux/pci.c
@@ -586,6 +586,12 @@  rte_pci_get_iommu_class(void)
 	bool is_vfio_noiommu_enabled = true;
 	bool has_iova_va;
 	bool is_bound_uio;
+	bool spapr_iommu =
+#if defined(RTE_ARCH_PPC64)
+		true;
+#else
+		false;
+#endif
 
 	is_bound = pci_one_device_is_bound();
 	if (!is_bound)
@@ -598,7 +604,8 @@  rte_pci_get_iommu_class(void)
 					true : false;
 #endif
 
-	if (has_iova_va && !is_bound_uio && !is_vfio_noiommu_enabled)
+	if (has_iova_va && !is_bound_uio && !is_vfio_noiommu_enabled &&
+			!spapr_iommu)
 		return RTE_IOVA_VA;
 
 	if (has_iova_va) {
@@ -607,6 +614,8 @@  rte_pci_get_iommu_class(void)
 			RTE_LOG(WARNING, EAL, "vfio-noiommu mode configured\n");
 		if (is_bound_uio)
 			RTE_LOG(WARNING, EAL, "few device bound to UIO\n");
+		if (spapr_iommu)
+			RTE_LOG(WARNING, EAL, "sPAPR iommu does not support iova as va\n");
 	}
 
 	return RTE_IOVA_PA;