VFIO Updates for Linux v4.15

* Virtualize PCI MPS and MRRS registers
 
  * Avoid soft lockups on SPAPR when clearing TCE
 
  * Broadcom FlexRM platform device support
 
  * Samples driver cleanup & type1 integer overflow fix
 -----BEGIN PGP SIGNATURE-----
 Version: GnuPG v2.0.14 (GNU/Linux)
 
 iQIcBAABAgAGBQJaCgl2AAoJECObm247sIsiT3kP/A7GsPfKMjKwkTFHUzlQ0/gj
 vCaMeQbfrSSmQE4c3dad5X53Y//FPlB9U+4y4eJiF+VwNYy8Emcue9SuMv8gp4WK
 HUONhRYcbfpL/qdYAcWRkdlS5In9fBsWJAauHSBbBW7jfCGmXiSVcHlEXscMfX1O
 VLs5aW9SHwabHmIgIVu3BL8rMXgm/nLeUHy5mjMq9SiBE1Wc69xkEn+gq4rn7Ibd
 JoIZymGpvQ5L4G4L9fZNAOKma9WQgXOvENfNbYqWxSJCm5/8Ke30tmZEljoIPsqE
 O/Twa/KBJ8Ah9FwLheSPmk0+qxq5VPfcprNb3ZlFYBUJl97EihSzEnw6KtiMgOZA
 GPEfzZOTip+lPShIhGfTuE/EwHcubYUXnFt6jLmjZHIFuMrYy/uDsIyNL6hgYjDl
 qJCiRzCyXvD8f2BvLqp7Ar1yd2khqYoLrxEshmxEt3ysDz4iqb3+RcHqVgmV9bP1
 xHip0SY+VrdZBEICCFas8rhS9fhE+8JuAAeMfvNHn30v7lHi3Wgp34tghy3nIikf
 /tb+1N4IRjo0TxTs1an/L1AaM+4J1D8qWvDE8NuiW+reY3oTtq0rmkIkD/HBZRhL
 zm8Y0+fNcfMAPVBd/amXWKuqKyxtKjqtIO5IQEpBB6fV+HThf1/UcguiI0x//ZfW
 TdP6GydKc1oiNsEv7217
 =86A4
 -----END PGP SIGNATURE-----

Merge tag 'vfio-v4.15-rc1' of git://github.com/awilliam/linux-vfio

Pull VFIO updates from Alex Williamson:

 - Virtualize PCI MPS and MRRS registers

 - Avoid soft lockups on SPAPR when clearing TCE

 - Broadcom FlexRM platform device support

 - Samples driver cleanup & type1 integer overflow fix

* tag 'vfio-v4.15-rc1' of git://github.com/awilliam/linux-vfio:
  vfio: platform: reset: Add Broadcom FlexRM reset module
  vfio/type1: silence integer overflow warning
  vfio-mdev/samples: make mdev_fops const and static
  vfio/spapr: Add cond_resched() for huge updates
  vfio/pci: Virtualize Maximum Read Request Size
  vfio/pci: Virtualize Maximum Payload Size

drivers/vfio/pci/vfio_pci_config.c

@@ -808,6 +808,7 @@ static int vfio_exp_config_write(struct vfio_pci_device *vdev, int pos,
 {
 	__le16 *ctrl = (__le16 *)(vdev->vconfig + pos -
 				  offset + PCI_EXP_DEVCTL);
+	int readrq = le16_to_cpu(*ctrl) & PCI_EXP_DEVCTL_READRQ;
 
 	count = vfio_default_config_write(vdev, pos, count, perm, offset, val);
 	if (count < 0)
@@ -833,6 +834,27 @@ static int vfio_exp_config_write(struct vfio_pci_device *vdev, int pos,
 			pci_try_reset_function(vdev->pdev);
 	}
 
+	/*
+	 * MPS is virtualized to the user, writes do not change the physical
+	 * register since determining a proper MPS value requires a system wide
+	 * device view.  The MRRS is largely independent of MPS, but since the
+	 * user does not have that system-wide view, they might set a safe, but
+	 * inefficiently low value.  Here we allow writes through to hardware,
+	 * but we set the floor to the physical device MPS setting, so that
+	 * we can at least use full TLPs, as defined by the MPS value.
+	 *
+	 * NB, if any devices actually depend on an artificially low MRRS
+	 * setting, this will need to be revisited, perhaps with a quirk
+	 * though pcie_set_readrq().
+	 */
+	if (readrq != (le16_to_cpu(*ctrl) & PCI_EXP_DEVCTL_READRQ)) {
+		readrq = 128 <<
+			((le16_to_cpu(*ctrl) & PCI_EXP_DEVCTL_READRQ) >> 12);
+		readrq = max(readrq, pcie_get_mps(vdev->pdev));
+
+		pcie_set_readrq(vdev->pdev, readrq);
+	}
+
 	return count;
 }
 
@@ -849,11 +871,14 @@ static int __init init_pci_cap_exp_perm(struct perm_bits *perm)
 
 	/*
 	 * Allow writes to device control fields, except devctl_phantom,
-	 * which could confuse IOMMU, and the ARI bit in devctl2, which
-	 * is set at probe time.  FLR gets virtualized via our writefn.
+	 * which could confuse IOMMU, MPS, which can break communication
+	 * with other physical devices, and the ARI bit in devctl2, which
+	 * is set at probe time.  FLR and MRRS get virtualized via our
+	 * writefn.
 	 */
 	p_setw(perm, PCI_EXP_DEVCTL,
-	       PCI_EXP_DEVCTL_BCR_FLR, ~PCI_EXP_DEVCTL_PHANTOM);
+	       PCI_EXP_DEVCTL_BCR_FLR | PCI_EXP_DEVCTL_PAYLOAD |
+	       PCI_EXP_DEVCTL_READRQ, ~PCI_EXP_DEVCTL_PHANTOM);
 	p_setw(perm, PCI_EXP_DEVCTL2, NO_VIRT, ~PCI_EXP_DEVCTL2_ARI);
 	return 0;
 }

drivers/vfio/platform/reset/Kconfig

@@ -13,3 +13,12 @@ config VFIO_PLATFORM_AMDXGBE_RESET
 	  Enables the VFIO platform driver to handle reset for AMD XGBE
 
 	  If you don't know what to do here, say N.
+
+config VFIO_PLATFORM_BCMFLEXRM_RESET
+	tristate "VFIO support for Broadcom FlexRM reset"
+	depends on VFIO_PLATFORM && (ARCH_BCM_IPROC || COMPILE_TEST)
+	default ARCH_BCM_IPROC
+	help
+	  Enables the VFIO platform driver to handle reset for Broadcom FlexRM
+
+	  If you don't know what to do here, say N.

drivers/vfio/platform/reset/Makefile

@@ -6,3 +6,4 @@ ccflags-y += -Idrivers/vfio/platform
 
 obj-$(CONFIG_VFIO_PLATFORM_CALXEDAXGMAC_RESET) += vfio-platform-calxedaxgmac.o
 obj-$(CONFIG_VFIO_PLATFORM_AMDXGBE_RESET) += vfio-platform-amdxgbe.o
+obj-$(CONFIG_VFIO_PLATFORM_BCMFLEXRM_RESET) += vfio_platform_bcmflexrm.o

drivers/vfio/platform/reset/vfio_platform_bcmflexrm.c

@@ -0,0 +1,113 @@
+/*
+ * Copyright (C) 2017 Broadcom
+ *
+ * This program is free software; you can redistribute it and/or
+ * modify it under the terms of the GNU General Public License as
+ * published by the Free Software Foundation version 2.
+ *
+ * This program is distributed "as is" WITHOUT ANY WARRANTY of any
+ * kind, whether express or implied; without even the implied warranty
+ * of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
+ * GNU General Public License for more details.
+ */
+
+/*
+ * This driver provides reset support for Broadcom FlexRM ring manager
+ * to VFIO platform.
+ */
+
+#include <linux/delay.h>
+#include <linux/device.h>
+#include <linux/init.h>
+#include <linux/io.h>
+#include <linux/kernel.h>
+#include <linux/module.h>
+
+#include "vfio_platform_private.h"
+
+/* FlexRM configuration */
+#define RING_REGS_SIZE					0x10000
+#define RING_VER_MAGIC					0x76303031
+
+/* Per-Ring register offsets */
+#define RING_VER					0x000
+#define RING_CONTROL					0x034
+#define RING_FLUSH_DONE					0x038
+
+/* Register RING_CONTROL fields */
+#define CONTROL_FLUSH_SHIFT				5
+
+/* Register RING_FLUSH_DONE fields */
+#define FLUSH_DONE_MASK					0x1
+
+static int vfio_platform_bcmflexrm_shutdown(void __iomem *ring)
+{
+	unsigned int timeout;
+
+	/* Disable/inactivate ring */
+	writel_relaxed(0x0, ring + RING_CONTROL);
+
+	/* Set ring flush state */
+	timeout = 1000; /* timeout of 1s */
+	writel_relaxed(BIT(CONTROL_FLUSH_SHIFT), ring + RING_CONTROL);
+	do {
+		if (readl_relaxed(ring + RING_FLUSH_DONE) &
+		    FLUSH_DONE_MASK)
+			break;
+		mdelay(1);
+	} while (--timeout);
+	if (!timeout)
+		return -ETIMEDOUT;
+
+	/* Clear ring flush state */
+	timeout = 1000; /* timeout of 1s */
+	writel_relaxed(0x0, ring + RING_CONTROL);
+	do {
+		if (!(readl_relaxed(ring + RING_FLUSH_DONE) &
+		      FLUSH_DONE_MASK))
+			break;
+		mdelay(1);
+	} while (--timeout);
+	if (!timeout)
+		return -ETIMEDOUT;
+
+	return 0;
+}
+
+static int vfio_platform_bcmflexrm_reset(struct vfio_platform_device *vdev)
+{
+	void __iomem *ring;
+	int rc = 0, ret = 0, ring_num = 0;
+	struct vfio_platform_region *reg = &vdev->regions[0];
+
+	/* Map FlexRM ring registers if not mapped */
+	if (!reg->ioaddr) {
+		reg->ioaddr = ioremap_nocache(reg->addr, reg->size);
+		if (!reg->ioaddr)
+			return -ENOMEM;
+	}
+
+	/* Discover and shutdown each FlexRM ring */
+	for (ring = reg->ioaddr;
+	     ring < (reg->ioaddr + reg->size); ring += RING_REGS_SIZE) {
+		if (readl_relaxed(ring + RING_VER) == RING_VER_MAGIC) {
+			rc = vfio_platform_bcmflexrm_shutdown(ring);
+			if (rc) {
+				dev_warn(vdev->device,
+					 "FlexRM ring%d shutdown error %d\n",
+					 ring_num, rc);
+				ret |= rc;
+			}
+			ring_num++;
+		}
+	}
+
+	return ret;
+}
+
+module_vfio_reset_handler("brcm,iproc-flexrm-mbox",
+			  vfio_platform_bcmflexrm_reset);
+
+MODULE_LICENSE("GPL v2");
+MODULE_AUTHOR("Anup Patel <anup.patel@broadcom.com>");
+MODULE_DESCRIPTION("Reset support for Broadcom FlexRM VFIO platform device");

drivers/vfio/vfio_iommu_spapr_tce.c

@@ -507,6 +507,8 @@ static int tce_iommu_clear(struct tce_container *container,
 	enum dma_data_direction direction;
 
 	for ( ; pages; --pages, ++entry) {
+		cond_resched();
+
 		direction = DMA_NONE;
 		oldhpa = 0;
 		ret = iommu_tce_xchg(tbl, entry, &oldhpa, &direction);

drivers/vfio/vfio_iommu_type1.c

@@ -767,6 +767,9 @@ static int vfio_dma_do_unmap(struct vfio_iommu *iommu,
 		return -EINVAL;
 	if (!unmap->size || unmap->size & mask)
 		return -EINVAL;
+	if (unmap->iova + unmap->size < unmap->iova ||
+	    unmap->size > SIZE_MAX)
+		return -EINVAL;
 
 	WARN_ON(mask & PAGE_MASK);
 again:

samples/vfio-mdev/mtty.c

@@ -1413,7 +1413,7 @@ struct attribute_group *mdev_type_groups[] = {
 	NULL,
 };
 
-struct mdev_parent_ops mdev_fops = {
+static const struct mdev_parent_ops mdev_fops = {
 	.owner                  = THIS_MODULE,
 	.dev_attr_groups        = mtty_dev_groups,
 	.mdev_attr_groups       = mdev_dev_groups,