Re: [PATCH v3 6/6] powerpc/powernv: allocate sparse PE# when using M64 BAR in Single PE mode

On Fri, Aug 14, 2015 at 11:03:00AM +1000, Gavin Shan wrote:

On Thu, Aug 13, 2015 at 10:11:11PM +0800, Wei Yang wrote:

quoted

When M64 BAR is set to Single PE mode, the PE# assigned to VF could be
sparse.

This patch restructures the patch to allocate sparse PE# for VFs when M64
BAR is set to Single PE mode.

Signed-off-by: Wei Yang <redacted>
---
arch/powerpc/include/asm/pci-bridge.h     |    2 +-
arch/powerpc/platforms/powernv/pci-ioda.c |   59 +++++++++++++++++++----------
2 files changed, 41 insertions(+), 20 deletions(-)

diff --git a/arch/powerpc/include/asm/pci-bridge.h b/arch/powerpc/include/asm/pci-bridge.h
index 9d33ada..b026ef8 100644
--- a/arch/powerpc/include/asm/pci-bridge.h
+++ b/arch/powerpc/include/asm/pci-bridge.h

@@ -214,7 +214,7 @@ struct pci_dn {

#ifdef CONFIG_PCI_IOV
	u16     vfs_expanded;		/* number of VFs IOV BAR expanded */
	u16     num_vfs;		/* number of VFs enabled*/
-	int     offset;			/* PE# for the first VF PE */
+	int     pe_num_map[MAX_M64_BAR];/* PE# for the first VF PE or array */

Same question as to "m64_map". pdn for non-PF doesn't need it.

The same, I prefer the dynamic version.

quoted

	bool    m64_single_mode;	/* Use M64 BAR in Single Mode */
#define IODA_INVALID_M64        (-1)
	int      m64_map[PCI_SRIOV_NUM_BARS][MAX_M64_BAR];

diff --git a/arch/powerpc/platforms/powernv/pci-ioda.c b/arch/powerpc/platforms/powernv/pci-ioda.c
index 1e6ac86..7633538 100644
--- a/arch/powerpc/platforms/powernv/pci-ioda.c
+++ b/arch/powerpc/platforms/powernv/pci-ioda.c

@@ -1232,7 +1232,7 @@ static int pnv_pci_vf_assign_m64(struct pci_dev *pdev, u16 num_vfs)

			/* Map the M64 here */
			if (pdn->m64_single_mode) {
-				pe_num = pdn->offset + j;
+				pe_num = pdn->pe_num_map[j];
				rc = opal_pci_map_pe_mmio_window(phb->opal_id,
						pe_num, OPAL_M64_WINDOW_TYPE,
						pdn->m64_map[i][j], 0);

@@ -1336,7 +1336,7 @@ void pnv_pci_sriov_disable(struct pci_dev *pdev)

	struct pnv_phb        *phb;
	struct pci_dn         *pdn;
	struct pci_sriov      *iov;
-	u16 num_vfs;
+	u16 num_vfs, i;

	bus = pdev->bus;
	hose = pci_bus_to_host(bus);

@@ -1350,14 +1350,17 @@ void pnv_pci_sriov_disable(struct pci_dev *pdev)

	if (phb->type == PNV_PHB_IODA2) {
		if (!pdn->m64_single_mode)
-			pnv_pci_vf_resource_shift(pdev, -pdn->offset);
+			pnv_pci_vf_resource_shift(pdev, -pdn->pe_num_map[0]);

		/* Release M64 windows */
		pnv_pci_vf_release_m64(pdev);

		/* Release PE numbers */
-		bitmap_clear(phb->ioda.pe_alloc, pdn->offset, num_vfs);
-		pdn->offset = 0;
+		if (pdn->m64_single_mode) {
+			for (i = 0; i < num_vfs; i++)
+				pnv_ioda_free_pe(phb, pdn->pe_num_map[i]);
+		} else
+			bitmap_clear(phb->ioda.pe_alloc, pdn->pe_num_map[0], num_vfs);
	}
}

@@ -1383,7 +1386,10 @@ static void pnv_ioda_setup_vf_PE(struct pci_dev *pdev, u16 num_vfs)

	/* Reserve PE for each VF */
	for (vf_index = 0; vf_index < num_vfs; vf_index++) {
-		pe_num = pdn->offset + vf_index;
+		if (pdn->m64_single_mode)
+			pe_num = pdn->pe_num_map[vf_index];
+		else
+			pe_num = pdn->pe_num_map[0] + vf_index;

		pe = &phb->ioda.pe_array[pe_num];
		pe->pe_number = pe_num;

@@ -1425,6 +1431,7 @@ int pnv_pci_sriov_enable(struct pci_dev *pdev, u16 num_vfs)

	struct pnv_phb        *phb;
	struct pci_dn         *pdn;
	int                    ret;
+	u16                    i;

	bus = pdev->bus;
	hose = pci_bus_to_host(bus);

@@ -1448,19 +1455,30 @@ int pnv_pci_sriov_enable(struct pci_dev *pdev, u16 num_vfs)

		}

		/* Calculate available PE for required VFs */
-		mutex_lock(&phb->ioda.pe_alloc_mutex);
-		pdn->offset = bitmap_find_next_zero_area(
-			phb->ioda.pe_alloc, phb->ioda.total_pe,
-			0, num_vfs, 0);
-		if (pdn->offset >= phb->ioda.total_pe) {
+		if (pdn->m64_single_mode) {
+			for (i = 0; i < num_vfs; i++)
+				pdn->pe_num_map[i] = IODA_INVALID_PE;
+			for (i = 0; i < num_vfs; i++) {
+				pdn->pe_num_map[i] = pnv_ioda_alloc_pe(phb);
+				if (pdn->pe_num_map[i] == IODA_INVALID_PE) {
+					ret = -EBUSY;
+					goto m64_failed;
+				}
+			}
+		} else {
+			mutex_lock(&phb->ioda.pe_alloc_mutex);
+			pdn->pe_num_map[0] = bitmap_find_next_zero_area(
+				phb->ioda.pe_alloc, phb->ioda.total_pe,
+				0, num_vfs, 0);
+			if (pdn->pe_num_map[0] >= phb->ioda.total_pe) {
+				mutex_unlock(&phb->ioda.pe_alloc_mutex);
+				dev_info(&pdev->dev, "Failed to enable VF%d\n", num_vfs);
+				return -EBUSY;
+			}
+			bitmap_set(phb->ioda.pe_alloc, pdn->pe_num_map[0], num_vfs);
			mutex_unlock(&phb->ioda.pe_alloc_mutex);
-			dev_info(&pdev->dev, "Failed to enable VF%d\n", num_vfs);
-			pdn->offset = 0;
-			return -EBUSY;
		}
-		bitmap_set(phb->ioda.pe_alloc, pdn->offset, num_vfs);
		pdn->num_vfs = num_vfs;
-		mutex_unlock(&phb->ioda.pe_alloc_mutex);

		/* Assign M64 window accordingly */
		ret = pnv_pci_vf_assign_m64(pdev, num_vfs);

@@ -1475,7 +1493,7 @@ int pnv_pci_sriov_enable(struct pci_dev *pdev, u16 num_vfs)

		 * Otherwise, the PE# for the VF will conflict with others.
		 */
		if (!pdn->m64_single_mode) {
-			ret = pnv_pci_vf_resource_shift(pdev, pdn->offset);
+			ret = pnv_pci_vf_resource_shift(pdev, pdn->pe_num_map[0]);
			if (ret)
				goto m64_failed;
		}

@@ -1487,8 +1505,11 @@ int pnv_pci_sriov_enable(struct pci_dev *pdev, u16 num_vfs)

	return 0;

m64_failed:
-	bitmap_clear(phb->ioda.pe_alloc, pdn->offset, num_vfs);
-	pdn->offset = 0;
+	if (pdn->m64_single_mode) {
+		for (i = 0; i < num_vfs; i++)
+			pnv_ioda_free_pe(phb, pdn->pe_num_map[i]);

if pdn->pe_num_map[i] isn't valid PE number, what will happen?

You are right, we need to check this.

quoted

+	} else
+		bitmap_clear(phb->ioda.pe_alloc, pdn->pe_num_map[0], num_vfs);

	return ret;
}
-- 
1.7.9.5

-- 
Richard Yang
Help you, Help me