[PULL] virtio and lguest tree

From: Rusty Russell
Date: Mon Dec 29 2008 - 18:07:01 EST


The following changes since commit 3c92ec8ae91ecf59d88c798301833d7cf83f2179:
Linus Torvalds (1):
Merge branch 'next' of git://git.kernel.org/.../paulus/powerpc

are available in the git repository at:

ssh://master.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus.git master

Christian Borntraeger (2):
virtio_console: support console resizing
kvm-s390: implement config_changed for virtio on s390

Hollis Blanchard (2):
virtio-pci queue allocation not page-aligned
virtio: avoid implicit use of Linux page size in balloon interface

Kay Sievers (1):
virtio: struct device - replace bus_id with dev_name(), dev_set_name()

Mark McLoughlin (2):
virtio: add PCI device release() function
lguest: struct device - replace bus_id with dev_name()

Matias Zabaljauregui (1):
lguest: move the initial guest page table creation code to the host

Randy Dunlap (1):
virtio_blk: fix type warning

Rusty Russell (8):
virtio: Don't use PAGE_SIZE in virtio_pci.c
virtio: rename 'pagesize' arg to vring_init/vring_size
virtio: Don't use PAGE_SIZE for vring alignment in virtio_pci.
virtio: use LGUEST_VRING_ALIGN instead of relying on pagesize
virtio: use KVM_S390_VIRTIO_RING_ALIGN instead of relying on pagesize
virtio: hand virtio ring alignment as argument to vring_new_virtqueue
virtio: set max_segment_size and max_sectors to infinite.
virtio: block: dynamic maximum segments

Documentation/lguest/lguest.c | 66 ++++----------------------------
arch/s390/include/asm/kvm_virtio.h | 4 ++
arch/x86/lguest/i386_head.S | 15 -------
drivers/block/virtio_blk.c | 41 ++++++++++++++------
drivers/char/hvc_console.c | 1 +
drivers/char/virtio_console.c | 30 ++++++++++++++-
drivers/lguest/lg.h | 2 +-
drivers/lguest/lguest_device.c | 8 ++--
drivers/lguest/lguest_user.c | 13 ++----
drivers/lguest/page_tables.c | 72 +++++++++++++++++++++++++++++++++++-
drivers/s390/kvm/kvm_virtio.c | 34 +++++++++++++---
drivers/virtio/virtio.c | 2 +-
drivers/virtio/virtio_balloon.c | 13 +++++-
drivers/virtio/virtio_pci.c | 43 +++++++++++++--------
drivers/virtio/virtio_ring.c | 3 +-
include/linux/lguest_launcher.h | 6 ++-
include/linux/virtio_balloon.h | 3 +
include/linux/virtio_console.h | 11 +++++
include/linux/virtio_pci.h | 8 ++++
include/linux/virtio_ring.h | 13 +++---
20 files changed, 253 insertions(+), 135 deletions(-)

diff --git a/Documentation/lguest/lguest.c b/Documentation/lguest/lguest.c
index 8045206..f2dbbf3 100644
--- a/Documentation/lguest/lguest.c
+++ b/Documentation/lguest/lguest.c
@@ -481,51 +481,6 @@ static unsigned long load_initrd(const char *name, unsigned long mem)
/* We return the initrd size. */
return len;
}
-
-/* Once we know how much memory we have we can construct simple linear page
- * tables which set virtual == physical which will get the Guest far enough
- * into the boot to create its own.
- *
- * We lay them out of the way, just below the initrd (which is why we need to
- * know its size here). */
-static unsigned long setup_pagetables(unsigned long mem,
- unsigned long initrd_size)
-{
- unsigned long *pgdir, *linear;
- unsigned int mapped_pages, i, linear_pages;
- unsigned int ptes_per_page = getpagesize()/sizeof(void *);
-
- mapped_pages = mem/getpagesize();
-
- /* Each PTE page can map ptes_per_page pages: how many do we need? */
- linear_pages = (mapped_pages + ptes_per_page-1)/ptes_per_page;
-
- /* We put the toplevel page directory page at the top of memory. */
- pgdir = from_guest_phys(mem) - initrd_size - getpagesize();
-
- /* Now we use the next linear_pages pages as pte pages */
- linear = (void *)pgdir - linear_pages*getpagesize();
-
- /* Linear mapping is easy: put every page's address into the mapping in
- * order. PAGE_PRESENT contains the flags Present, Writable and
- * Executable. */
- for (i = 0; i < mapped_pages; i++)
- linear[i] = ((i * getpagesize()) | PAGE_PRESENT);
-
- /* The top level points to the linear page table pages above. */
- for (i = 0; i < mapped_pages; i += ptes_per_page) {
- pgdir[i/ptes_per_page]
- = ((to_guest_phys(linear) + i*sizeof(void *))
- | PAGE_PRESENT);
- }
-
- verbose("Linear mapping of %u pages in %u pte pages at %#lx\n",
- mapped_pages, linear_pages, to_guest_phys(linear));
-
- /* We return the top level (guest-physical) address: the kernel needs
- * to know where it is. */
- return to_guest_phys(pgdir);
-}
/*:*/

/* Simple routine to roll all the commandline arguments together with spaces
@@ -548,13 +503,13 @@ static void concat(char *dst, char *args[])

/*L:185 This is where we actually tell the kernel to initialize the Guest. We
* saw the arguments it expects when we looked at initialize() in lguest_user.c:
- * the base of Guest "physical" memory, the top physical page to allow, the
- * top level pagetable and the entry point for the Guest. */
-static int tell_kernel(unsigned long pgdir, unsigned long start)
+ * the base of Guest "physical" memory, the top physical page to allow and the
+ * entry point for the Guest. */
+static int tell_kernel(unsigned long start)
{
unsigned long args[] = { LHREQ_INITIALIZE,
(unsigned long)guest_base,
- guest_limit / getpagesize(), pgdir, start };
+ guest_limit / getpagesize(), start };
int fd;

verbose("Guest: %p - %p (%#lx)\n",
@@ -1030,7 +985,7 @@ static void update_device_status(struct device *dev)
/* Zero out the virtqueues. */
for (vq = dev->vq; vq; vq = vq->next) {
memset(vq->vring.desc, 0,
- vring_size(vq->config.num, getpagesize()));
+ vring_size(vq->config.num, LGUEST_VRING_ALIGN));
lg_last_avail(vq) = 0;
}
} else if (dev->desc->status & VIRTIO_CONFIG_S_FAILED) {
@@ -1211,7 +1166,7 @@ static void add_virtqueue(struct device *dev, unsigned int num_descs,
void *p;

/* First we need some memory for this virtqueue. */
- pages = (vring_size(num_descs, getpagesize()) + getpagesize() - 1)
+ pages = (vring_size(num_descs, LGUEST_VRING_ALIGN) + getpagesize() - 1)
/ getpagesize();
p = get_pages(pages);

@@ -1228,7 +1183,7 @@ static void add_virtqueue(struct device *dev, unsigned int num_descs,
vq->config.pfn = to_guest_phys(p) / getpagesize();

/* Initialize the vring. */
- vring_init(&vq->vring, num_descs, p, getpagesize());
+ vring_init(&vq->vring, num_descs, p, LGUEST_VRING_ALIGN);

/* Append virtqueue to this device's descriptor. We use
* device_config() to get the end of the device's current virtqueues;
@@ -1941,7 +1896,7 @@ int main(int argc, char *argv[])
{
/* Memory, top-level pagetable, code startpoint and size of the
* (optional) initrd. */
- unsigned long mem = 0, pgdir, start, initrd_size = 0;
+ unsigned long mem = 0, start, initrd_size = 0;
/* Two temporaries and the /dev/lguest file descriptor. */
int i, c, lguest_fd;
/* The boot information for the Guest. */
@@ -2040,9 +1995,6 @@ int main(int argc, char *argv[])
boot->hdr.type_of_loader = 0xFF;
}

- /* Set up the initial linear pagetables, starting below the initrd. */
- pgdir = setup_pagetables(mem, initrd_size);
-
/* The Linux boot header contains an "E820" memory map: ours is a
* simple, single region. */
boot->e820_entries = 1;
@@ -2064,7 +2016,7 @@ int main(int argc, char *argv[])

/* We tell the kernel to initialize the Guest: this returns the open
* /dev/lguest file descriptor. */
- lguest_fd = tell_kernel(pgdir, start);
+ lguest_fd = tell_kernel(start);

/* We clone off a thread, which wakes the Launcher whenever one of the
* input file descriptors needs attention. We call this the Waker, and
diff --git a/arch/s390/include/asm/kvm_virtio.h b/arch/s390/include/asm/kvm_virtio.h
index c13568b..0503936 100644
--- a/arch/s390/include/asm/kvm_virtio.h
+++ b/arch/s390/include/asm/kvm_virtio.h
@@ -50,6 +50,10 @@ struct kvm_vqconfig {
#define KVM_S390_VIRTIO_RESET 1
#define KVM_S390_VIRTIO_SET_STATUS 2

+/* The alignment to use between consumer and producer parts of vring.
+ * This is pagesize for historical reasons. */
+#define KVM_S390_VIRTIO_RING_ALIGN 4096
+
#ifdef __KERNEL__
/* early virtio console setup */
#ifdef CONFIG_S390_GUEST
diff --git a/arch/x86/lguest/i386_head.S b/arch/x86/lguest/i386_head.S
index 5c7cef3..10b9bd3 100644
--- a/arch/x86/lguest/i386_head.S
+++ b/arch/x86/lguest/i386_head.S
@@ -30,21 +30,6 @@ ENTRY(lguest_entry)
movl $lguest_data - __PAGE_OFFSET, %edx
int $LGUEST_TRAP_ENTRY

- /* The Host put the toplevel pagetable in lguest_data.pgdir. The movsl
- * instruction uses %esi implicitly as the source for the copy we're
- * about to do. */
- movl lguest_data - __PAGE_OFFSET + LGUEST_DATA_pgdir, %esi
-
- /* Copy first 32 entries of page directory to __PAGE_OFFSET entries.
- * This means the first 128M of kernel memory will be mapped at
- * PAGE_OFFSET where the kernel expects to run. This will get it far
- * enough through boot to switch to its own pagetables. */
- movl $32, %ecx
- movl %esi, %edi
- addl $((__PAGE_OFFSET >> 22) * 4), %edi
- rep
- movsl
-
/* Set up the initial stack so we can run C code. */
movl $(init_thread_union+THREAD_SIZE),%esp

diff --git a/drivers/block/virtio_blk.c b/drivers/block/virtio_blk.c
index 85d79a0..300078b 100644
--- a/drivers/block/virtio_blk.c
+++ b/drivers/block/virtio_blk.c
@@ -6,7 +6,6 @@
#include <linux/virtio_blk.h>
#include <linux/scatterlist.h>

-#define VIRTIO_MAX_SG (3+MAX_PHYS_SEGMENTS)
#define PART_BITS 4

static int major, index;
@@ -26,8 +25,11 @@ struct virtio_blk

mempool_t *pool;

+ /* What host tells us, plus 2 for header & tailer. */
+ unsigned int sg_elems;
+
/* Scatterlist: can be too big for stack. */
- struct scatterlist sg[VIRTIO_MAX_SG];
+ struct scatterlist sg[/*sg_elems*/];
};

struct virtblk_req
@@ -97,8 +99,6 @@ static bool do_req(struct request_queue *q, struct virtio_blk *vblk,
if (blk_barrier_rq(vbr->req))
vbr->out_hdr.type |= VIRTIO_BLK_T_BARRIER;

- /* This init could be done at vblk creation time */
- sg_init_table(vblk->sg, VIRTIO_MAX_SG);
sg_set_buf(&vblk->sg[0], &vbr->out_hdr, sizeof(vbr->out_hdr));
num = blk_rq_map_sg(q, vbr->req, vblk->sg+1);
sg_set_buf(&vblk->sg[num+1], &vbr->status, sizeof(vbr->status));
@@ -130,7 +130,7 @@ static void do_virtblk_request(struct request_queue *q)

while ((req = elv_next_request(q)) != NULL) {
vblk = req->rq_disk->private_data;
- BUG_ON(req->nr_phys_segments > ARRAY_SIZE(vblk->sg));
+ BUG_ON(req->nr_phys_segments + 2 > vblk->sg_elems);

/* If this request fails, stop queue and wait for something to
finish to restart it. */
@@ -196,12 +196,22 @@ static int virtblk_probe(struct virtio_device *vdev)
int err;
u64 cap;
u32 v;
- u32 blk_size;
+ u32 blk_size, sg_elems;

if (index_to_minor(index) >= 1 << MINORBITS)
return -ENOSPC;

- vdev->priv = vblk = kmalloc(sizeof(*vblk), GFP_KERNEL);
+ /* We need to know how many segments before we allocate. */
+ err = virtio_config_val(vdev, VIRTIO_BLK_F_SEG_MAX,
+ offsetof(struct virtio_blk_config, seg_max),
+ &sg_elems);
+ if (err)
+ sg_elems = 1;
+
+ /* We need an extra sg elements at head and tail. */
+ sg_elems += 2;
+ vdev->priv = vblk = kmalloc(sizeof(*vblk) +
+ sizeof(vblk->sg[0]) * sg_elems, GFP_KERNEL);
if (!vblk) {
err = -ENOMEM;
goto out;
@@ -210,6 +220,8 @@ static int virtblk_probe(struct virtio_device *vdev)
INIT_LIST_HEAD(&vblk->reqs);
spin_lock_init(&vblk->lock);
vblk->vdev = vdev;
+ vblk->sg_elems = sg_elems;
+ sg_init_table(vblk->sg, vblk->sg_elems);

/* We expect one virtqueue, for output. */
vblk->vq = vdev->config->find_vq(vdev, 0, blk_done);
@@ -277,6 +289,13 @@ static int virtblk_probe(struct virtio_device *vdev)
}
set_capacity(vblk->disk, cap);

+ /* We can handle whatever the host told us to handle. */
+ blk_queue_max_phys_segments(vblk->disk->queue, vblk->sg_elems-2);
+ blk_queue_max_hw_segments(vblk->disk->queue, vblk->sg_elems-2);
+
+ /* No real sector limit. */
+ blk_queue_max_sectors(vblk->disk->queue, -1U);
+
/* Host can optionally specify maximum segment size and number of
* segments. */
err = virtio_config_val(vdev, VIRTIO_BLK_F_SIZE_MAX,
@@ -284,12 +303,8 @@ static int virtblk_probe(struct virtio_device *vdev)
&v);
if (!err)
blk_queue_max_segment_size(vblk->disk->queue, v);
-
- err = virtio_config_val(vdev, VIRTIO_BLK_F_SEG_MAX,
- offsetof(struct virtio_blk_config, seg_max),
- &v);
- if (!err)
- blk_queue_max_hw_segments(vblk->disk->queue, v);
+ else
+ blk_queue_max_segment_size(vblk->disk->queue, -1U);

/* Host can optionally specify the block size of the device */
err = virtio_config_val(vdev, VIRTIO_BLK_F_BLK_SIZE,
diff --git a/drivers/char/hvc_console.c b/drivers/char/hvc_console.c
index fb57f67..0587b66 100644
--- a/drivers/char/hvc_console.c
+++ b/drivers/char/hvc_console.c
@@ -695,6 +695,7 @@ void hvc_resize(struct hvc_struct *hp, struct winsize ws)
hp->ws = ws;
schedule_work(&hp->tty_resize);
}
+EXPORT_SYMBOL_GPL(hvc_resize);

/*
* This kthread is either polling or interrupt driven. This is determined by
diff --git a/drivers/char/virtio_console.c b/drivers/char/virtio_console.c
index 3fb0d2c..ff6f5a4 100644
--- a/drivers/char/virtio_console.c
+++ b/drivers/char/virtio_console.c
@@ -138,12 +138,33 @@ int __init virtio_cons_early_init(int (*put_chars)(u32, const char *, int))
}

/*
+ * virtio console configuration. This supports:
+ * - console resize
+ */
+static void virtcons_apply_config(struct virtio_device *dev)
+{
+ struct winsize ws;
+
+ if (virtio_has_feature(dev, VIRTIO_CONSOLE_F_SIZE)) {
+ dev->config->get(dev,
+ offsetof(struct virtio_console_config, cols),
+ &ws.ws_col, sizeof(u16));
+ dev->config->get(dev,
+ offsetof(struct virtio_console_config, rows),
+ &ws.ws_row, sizeof(u16));
+ hvc_resize(hvc, ws);
+ }
+}
+
+/*
* we support only one console, the hvc struct is a global var
- * There is no need to do anything
+ * We set the configuration at this point, since we now have a tty
*/
static int notifier_add_vio(struct hvc_struct *hp, int data)
{
hp->irq_requested = 1;
+ virtcons_apply_config(vdev);
+
return 0;
}

@@ -234,11 +255,18 @@ static struct virtio_device_id id_table[] = {
{ 0 },
};

+static unsigned int features[] = {
+ VIRTIO_CONSOLE_F_SIZE,
+};
+
static struct virtio_driver virtio_console = {
+ .feature_table = features,
+ .feature_table_size = ARRAY_SIZE(features),
.driver.name = KBUILD_MODNAME,
.driver.owner = THIS_MODULE,
.id_table = id_table,
.probe = virtcons_probe,
+ .config_changed = virtcons_apply_config,
};

static int __init init(void)
diff --git a/drivers/lguest/lg.h b/drivers/lguest/lg.h
index 5faefea..f2c641e 100644
--- a/drivers/lguest/lg.h
+++ b/drivers/lguest/lg.h
@@ -164,7 +164,7 @@ void copy_gdt(const struct lg_cpu *cpu, struct desc_struct *gdt);
void copy_gdt_tls(const struct lg_cpu *cpu, struct desc_struct *gdt);

/* page_tables.c: */
-int init_guest_pagetable(struct lguest *lg, unsigned long pgtable);
+int init_guest_pagetable(struct lguest *lg);
void free_guest_pagetable(struct lguest *lg);
void guest_new_pagetable(struct lg_cpu *cpu, unsigned long pgtable);
void guest_set_pmd(struct lguest *lg, unsigned long gpgdir, u32 i);
diff --git a/drivers/lguest/lguest_device.c b/drivers/lguest/lguest_device.c
index a661bbd..915da6b 100644
--- a/drivers/lguest/lguest_device.c
+++ b/drivers/lguest/lguest_device.c
@@ -250,7 +250,7 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev,
/* Figure out how many pages the ring will take, and map that memory */
lvq->pages = lguest_map((unsigned long)lvq->config.pfn << PAGE_SHIFT,
DIV_ROUND_UP(vring_size(lvq->config.num,
- PAGE_SIZE),
+ LGUEST_VRING_ALIGN),
PAGE_SIZE));
if (!lvq->pages) {
err = -ENOMEM;
@@ -259,8 +259,8 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev,

/* OK, tell virtio_ring.c to set up a virtqueue now we know its size
* and we've got a pointer to its pages. */
- vq = vring_new_virtqueue(lvq->config.num, vdev, lvq->pages,
- lg_notify, callback);
+ vq = vring_new_virtqueue(lvq->config.num, LGUEST_VRING_ALIGN,
+ vdev, lvq->pages, lg_notify, callback);
if (!vq) {
err = -ENOMEM;
goto unmap;
@@ -272,7 +272,7 @@ static struct virtqueue *lg_find_vq(struct virtio_device *vdev,
* the interrupt as a source of randomness: it'd be nice to have that
* back.. */
err = request_irq(lvq->config.irq, vring_interrupt, IRQF_SHARED,
- vdev->dev.bus_id, vq);
+ dev_name(&vdev->dev), vq);
if (err)
goto destroy_vring;

diff --git a/drivers/lguest/lguest_user.c b/drivers/lguest/lguest_user.c
index e73a000..34bc017 100644
--- a/drivers/lguest/lguest_user.c
+++ b/drivers/lguest/lguest_user.c
@@ -146,7 +146,7 @@ static int lg_cpu_start(struct lg_cpu *cpu, unsigned id, unsigned long start_ip)
return 0;
}

-/*L:020 The initialization write supplies 4 pointer sized (32 or 64 bit)
+/*L:020 The initialization write supplies 3 pointer sized (32 or 64 bit)
* values (in addition to the LHREQ_INITIALIZE value). These are:
*
* base: The start of the Guest-physical memory inside the Launcher memory.
@@ -155,9 +155,6 @@ static int lg_cpu_start(struct lg_cpu *cpu, unsigned id, unsigned long start_ip)
* allowed to access. The Guest memory lives inside the Launcher, so it sets
* this to ensure the Guest can only reach its own memory.
*
- * pgdir: The (Guest-physical) address of the top of the initial Guest
- * pagetables (which are set up by the Launcher).
- *
* start: The first instruction to execute ("eip" in x86-speak).
*/
static int initialize(struct file *file, const unsigned long __user *input)
@@ -166,7 +163,7 @@ static int initialize(struct file *file, const unsigned long __user *input)
* Guest. */
struct lguest *lg;
int err;
- unsigned long args[4];
+ unsigned long args[3];

/* We grab the Big Lguest lock, which protects against multiple
* simultaneous initializations. */
@@ -192,14 +189,14 @@ static int initialize(struct file *file, const unsigned long __user *input)
lg->mem_base = (void __user *)args[0];
lg->pfn_limit = args[1];

- /* This is the first cpu (cpu 0) and it will start booting at args[3] */
- err = lg_cpu_start(&lg->cpus[0], 0, args[3]);
+ /* This is the first cpu (cpu 0) and it will start booting at args[2] */
+ err = lg_cpu_start(&lg->cpus[0], 0, args[2]);
if (err)
goto release_guest;

/* Initialize the Guest's shadow page tables, using the toplevel
* address the Launcher gave us. This allocates memory, so can fail. */
- err = init_guest_pagetable(lg, args[2]);
+ err = init_guest_pagetable(lg);
if (err)
goto free_regs;

diff --git a/drivers/lguest/page_tables.c b/drivers/lguest/page_tables.c
index 81d0c60..576a831 100644
--- a/drivers/lguest/page_tables.c
+++ b/drivers/lguest/page_tables.c
@@ -14,6 +14,7 @@
#include <linux/percpu.h>
#include <asm/tlbflush.h>
#include <asm/uaccess.h>
+#include <asm/bootparam.h>
#include "lg.h"

/*M:008 We hold reference to pages, which prevents them from being swapped.
@@ -581,15 +582,82 @@ void guest_set_pmd(struct lguest *lg, unsigned long gpgdir, u32 idx)
release_pgd(lg, lg->pgdirs[pgdir].pgdir + idx);
}

+/* Once we know how much memory we have we can construct simple identity
+ * (which set virtual == physical) and linear mappings
+ * which will get the Guest far enough into the boot to create its own.
+ *
+ * We lay them out of the way, just below the initrd (which is why we need to
+ * know its size here). */
+static unsigned long setup_pagetables(struct lguest *lg,
+ unsigned long mem,
+ unsigned long initrd_size)
+{
+ pgd_t __user *pgdir;
+ pte_t __user *linear;
+ unsigned int mapped_pages, i, linear_pages, phys_linear;
+ unsigned long mem_base = (unsigned long)lg->mem_base;
+
+ /* We have mapped_pages frames to map, so we need
+ * linear_pages page tables to map them. */
+ mapped_pages = mem / PAGE_SIZE;
+ linear_pages = (mapped_pages + PTRS_PER_PTE - 1) / PTRS_PER_PTE;
+
+ /* We put the toplevel page directory page at the top of memory. */
+ pgdir = (pgd_t *)(mem + mem_base - initrd_size - PAGE_SIZE);
+
+ /* Now we use the next linear_pages pages as pte pages */
+ linear = (void *)pgdir - linear_pages * PAGE_SIZE;
+
+ /* Linear mapping is easy: put every page's address into the
+ * mapping in order. */
+ for (i = 0; i < mapped_pages; i++) {
+ pte_t pte;
+ pte = pfn_pte(i, __pgprot(_PAGE_PRESENT|_PAGE_RW|_PAGE_USER));
+ if (copy_to_user(&linear[i], &pte, sizeof(pte)) != 0)
+ return -EFAULT;
+ }
+
+ /* The top level points to the linear page table pages above.
+ * We setup the identity and linear mappings here. */
+ phys_linear = (unsigned long)linear - mem_base;
+ for (i = 0; i < mapped_pages; i += PTRS_PER_PTE) {
+ pgd_t pgd;
+ pgd = __pgd((phys_linear + i * sizeof(pte_t)) |
+ (_PAGE_PRESENT | _PAGE_RW | _PAGE_USER));
+
+ if (copy_to_user(&pgdir[i / PTRS_PER_PTE], &pgd, sizeof(pgd))
+ || copy_to_user(&pgdir[pgd_index(PAGE_OFFSET)
+ + i / PTRS_PER_PTE],
+ &pgd, sizeof(pgd)))
+ return -EFAULT;
+ }
+
+ /* We return the top level (guest-physical) address: remember where
+ * this is. */
+ return (unsigned long)pgdir - mem_base;
+}
+
/*H:500 (vii) Setting up the page tables initially.
*
* When a Guest is first created, the Launcher tells us where the toplevel of
* its first page table is. We set some things up here: */
-int init_guest_pagetable(struct lguest *lg, unsigned long pgtable)
+int init_guest_pagetable(struct lguest *lg)
{
+ u64 mem;
+ u32 initrd_size;
+ struct boot_params __user *boot = (struct boot_params *)lg->mem_base;
+
+ /* Get the Guest memory size and the ramdisk size from the boot header
+ * located at lg->mem_base (Guest address 0). */
+ if (copy_from_user(&mem, &boot->e820_map[0].size, sizeof(mem))
+ || get_user(initrd_size, &boot->hdr.ramdisk_size))
+ return -EFAULT;
+
/* We start on the first shadow page table, and give it a blank PGD
* page. */
- lg->pgdirs[0].gpgdir = pgtable;
+ lg->pgdirs[0].gpgdir = setup_pagetables(lg, mem, initrd_size);
+ if (IS_ERR_VALUE(lg->pgdirs[0].gpgdir))
+ return lg->pgdirs[0].gpgdir;
lg->pgdirs[0].pgdir = (pgd_t *)get_zeroed_page(GFP_KERNEL);
if (!lg->pgdirs[0].pgdir)
return -ENOMEM;
diff --git a/drivers/s390/kvm/kvm_virtio.c b/drivers/s390/kvm/kvm_virtio.c
index 3d44244..28c90b8 100644
--- a/drivers/s390/kvm/kvm_virtio.c
+++ b/drivers/s390/kvm/kvm_virtio.c
@@ -188,11 +188,13 @@ static struct virtqueue *kvm_find_vq(struct virtio_device *vdev,
config = kvm_vq_config(kdev->desc)+index;

err = vmem_add_mapping(config->address,
- vring_size(config->num, PAGE_SIZE));
+ vring_size(config->num,
+ KVM_S390_VIRTIO_RING_ALIGN));
if (err)
goto out;

- vq = vring_new_virtqueue(config->num, vdev, (void *) config->address,
+ vq = vring_new_virtqueue(config->num, KVM_S390_VIRTIO_RING_ALIGN,
+ vdev, (void *) config->address,
kvm_notify, callback);
if (!vq) {
err = -ENOMEM;
@@ -209,7 +211,8 @@ static struct virtqueue *kvm_find_vq(struct virtio_device *vdev,
return vq;
unmap:
vmem_remove_mapping(config->address,
- vring_size(config->num, PAGE_SIZE));
+ vring_size(config->num,
+ KVM_S390_VIRTIO_RING_ALIGN));
out:
return ERR_PTR(err);
}
@@ -220,7 +223,8 @@ static void kvm_del_vq(struct virtqueue *vq)

vring_del_virtqueue(vq);
vmem_remove_mapping(config->address,
- vring_size(config->num, PAGE_SIZE));
+ vring_size(config->num,
+ KVM_S390_VIRTIO_RING_ALIGN));
}

/*
@@ -295,13 +299,29 @@ static void scan_devices(void)
*/
static void kvm_extint_handler(u16 code)
{
- void *data = (void *) *(long *) __LC_PFAULT_INTPARM;
- u16 subcode = S390_lowcore.cpu_addr;
+ struct virtqueue *vq;
+ u16 subcode;
+ int config_changed;

+ subcode = S390_lowcore.cpu_addr;
if ((subcode & 0xff00) != VIRTIO_SUBCODE_64)
return;

- vring_interrupt(0, data);
+ /* The LSB might be overloaded, we have to mask it */
+ vq = (struct virtqueue *) ((*(long *) __LC_PFAULT_INTPARM) & ~1UL);
+
+ /* We use the LSB of extparam, to decide, if this interrupt is a config
+ * change or a "standard" interrupt */
+ config_changed = (*(int *) __LC_EXT_PARAMS & 1);
+
+ if (config_changed) {
+ struct virtio_driver *drv;
+ drv = container_of(vq->vdev->dev.driver,
+ struct virtio_driver, driver);
+ if (drv->config_changed)
+ drv->config_changed(vq->vdev);
+ } else
+ vring_interrupt(0, vq);
}

/*
diff --git a/drivers/virtio/virtio.c b/drivers/virtio/virtio.c
index 5b78fd0..018c070 100644
--- a/drivers/virtio/virtio.c
+++ b/drivers/virtio/virtio.c
@@ -176,7 +176,7 @@ int register_virtio_device(struct virtio_device *dev)

/* Assign a unique device index and hence name. */
dev->index = dev_index++;
- sprintf(dev->dev.bus_id, "virtio%u", dev->index);
+ dev_set_name(&dev->dev, "virtio%u", dev->index);

/* We always start by resetting the device, in case a previous
* driver messed it up. This also tests that code path a little. */
diff --git a/drivers/virtio/virtio_balloon.c b/drivers/virtio/virtio_balloon.c
index 62eab43..5926826 100644
--- a/drivers/virtio/virtio_balloon.c
+++ b/drivers/virtio/virtio_balloon.c
@@ -56,6 +56,15 @@ static struct virtio_device_id id_table[] = {
{ 0 },
};

+static u32 page_to_balloon_pfn(struct page *page)
+{
+ unsigned long pfn = page_to_pfn(page);
+
+ BUILD_BUG_ON(PAGE_SHIFT < VIRTIO_BALLOON_PFN_SHIFT);
+ /* Convert pfn from Linux page size to balloon page size. */
+ return pfn >> (PAGE_SHIFT - VIRTIO_BALLOON_PFN_SHIFT);
+}
+
static void balloon_ack(struct virtqueue *vq)
{
struct virtio_balloon *vb;
@@ -99,7 +108,7 @@ static void fill_balloon(struct virtio_balloon *vb, size_t num)
msleep(200);
break;
}
- vb->pfns[vb->num_pfns] = page_to_pfn(page);
+ vb->pfns[vb->num_pfns] = page_to_balloon_pfn(page);
totalram_pages--;
vb->num_pages++;
list_add(&page->lru, &vb->pages);
@@ -132,7 +141,7 @@ static void leak_balloon(struct virtio_balloon *vb, size_t num)
for (vb->num_pfns = 0; vb->num_pfns < num; vb->num_pfns++) {
page = list_first_entry(&vb->pages, struct page, lru);
list_del(&page->lru);
- vb->pfns[vb->num_pfns] = page_to_pfn(page);
+ vb->pfns[vb->num_pfns] = page_to_balloon_pfn(page);
vb->num_pages--;
}

diff --git a/drivers/virtio/virtio_pci.c b/drivers/virtio/virtio_pci.c
index c7dc37c..265fdf2 100644
--- a/drivers/virtio/virtio_pci.c
+++ b/drivers/virtio/virtio_pci.c
@@ -75,7 +75,7 @@ MODULE_DEVICE_TABLE(pci, virtio_pci_id_table);
* would make more sense for virtio to not insist on having it's own device. */
static struct device virtio_pci_root = {
.parent = NULL,
- .bus_id = "virtio-pci",
+ .init_name = "virtio-pci",
};

/* Convert a generic virtio device to our structure */
@@ -216,7 +216,7 @@ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index,
struct virtio_pci_device *vp_dev = to_vp_device(vdev);
struct virtio_pci_vq_info *info;
struct virtqueue *vq;
- unsigned long flags;
+ unsigned long flags, size;
u16 num;
int err;

@@ -237,19 +237,20 @@ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index,
info->queue_index = index;
info->num = num;

- info->queue = kzalloc(PAGE_ALIGN(vring_size(num,PAGE_SIZE)), GFP_KERNEL);
+ size = PAGE_ALIGN(vring_size(num, VIRTIO_PCI_VRING_ALIGN));
+ info->queue = alloc_pages_exact(size, GFP_KERNEL|__GFP_ZERO);
if (info->queue == NULL) {
err = -ENOMEM;
goto out_info;
}

/* activate the queue */
- iowrite32(virt_to_phys(info->queue) >> PAGE_SHIFT,
+ iowrite32(virt_to_phys(info->queue) >> VIRTIO_PCI_QUEUE_ADDR_SHIFT,
vp_dev->ioaddr + VIRTIO_PCI_QUEUE_PFN);

/* create the vring */
- vq = vring_new_virtqueue(info->num, vdev, info->queue,
- vp_notify, callback);
+ vq = vring_new_virtqueue(info->num, VIRTIO_PCI_VRING_ALIGN,
+ vdev, info->queue, vp_notify, callback);
if (!vq) {
err = -ENOMEM;
goto out_activate_queue;
@@ -266,7 +267,7 @@ static struct virtqueue *vp_find_vq(struct virtio_device *vdev, unsigned index,

out_activate_queue:
iowrite32(0, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_PFN);
- kfree(info->queue);
+ free_pages_exact(info->queue, size);
out_info:
kfree(info);
return ERR_PTR(err);
@@ -277,7 +278,7 @@ static void vp_del_vq(struct virtqueue *vq)
{
struct virtio_pci_device *vp_dev = to_vp_device(vq->vdev);
struct virtio_pci_vq_info *info = vq->priv;
- unsigned long flags;
+ unsigned long flags, size;

spin_lock_irqsave(&vp_dev->lock, flags);
list_del(&info->node);
@@ -289,7 +290,8 @@ static void vp_del_vq(struct virtqueue *vq)
iowrite16(info->queue_index, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_SEL);
iowrite32(0, vp_dev->ioaddr + VIRTIO_PCI_QUEUE_PFN);

- kfree(info->queue);
+ size = PAGE_ALIGN(vring_size(info->num, VIRTIO_PCI_VRING_ALIGN));
+ free_pages_exact(info->queue, size);
kfree(info);
}

@@ -305,6 +307,20 @@ static struct virtio_config_ops virtio_pci_config_ops = {
.finalize_features = vp_finalize_features,
};

+static void virtio_pci_release_dev(struct device *_d)
+{
+ struct virtio_device *dev = container_of(_d, struct virtio_device, dev);
+ struct virtio_pci_device *vp_dev = to_vp_device(dev);
+ struct pci_dev *pci_dev = vp_dev->pci_dev;
+
+ free_irq(pci_dev->irq, vp_dev);
+ pci_set_drvdata(pci_dev, NULL);
+ pci_iounmap(pci_dev, vp_dev->ioaddr);
+ pci_release_regions(pci_dev);
+ pci_disable_device(pci_dev);
+ kfree(vp_dev);
+}
+
/* the PCI probing function */
static int __devinit virtio_pci_probe(struct pci_dev *pci_dev,
const struct pci_device_id *id)
@@ -328,6 +344,7 @@ static int __devinit virtio_pci_probe(struct pci_dev *pci_dev,
return -ENOMEM;

vp_dev->vdev.dev.parent = &virtio_pci_root;
+ vp_dev->vdev.dev.release = virtio_pci_release_dev;
vp_dev->vdev.config = &virtio_pci_config_ops;
vp_dev->pci_dev = pci_dev;
INIT_LIST_HEAD(&vp_dev->virtqueues);
@@ -357,7 +374,7 @@ static int __devinit virtio_pci_probe(struct pci_dev *pci_dev,

/* register a handler for the queue with the PCI device's interrupt */
err = request_irq(vp_dev->pci_dev->irq, vp_interrupt, IRQF_SHARED,
- vp_dev->vdev.dev.bus_id, vp_dev);
+ dev_name(&vp_dev->vdev.dev), vp_dev);
if (err)
goto out_set_drvdata;

@@ -387,12 +404,6 @@ static void __devexit virtio_pci_remove(struct pci_dev *pci_dev)
struct virtio_pci_device *vp_dev = pci_get_drvdata(pci_dev);

unregister_virtio_device(&vp_dev->vdev);
- free_irq(pci_dev->irq, vp_dev);
- pci_set_drvdata(pci_dev, NULL);
- pci_iounmap(pci_dev, vp_dev->ioaddr);
- pci_release_regions(pci_dev);
- pci_disable_device(pci_dev);
- kfree(vp_dev);
}

#ifdef CONFIG_PM
diff --git a/drivers/virtio/virtio_ring.c b/drivers/virtio/virtio_ring.c
index 6eb5303..5777196 100644
--- a/drivers/virtio/virtio_ring.c
+++ b/drivers/virtio/virtio_ring.c
@@ -274,6 +274,7 @@ static struct virtqueue_ops vring_vq_ops = {
};

struct virtqueue *vring_new_virtqueue(unsigned int num,
+ unsigned int vring_align,
struct virtio_device *vdev,
void *pages,
void (*notify)(struct virtqueue *),
@@ -292,7 +293,7 @@ struct virtqueue *vring_new_virtqueue(unsigned int num,
if (!vq)
return NULL;

- vring_init(&vq->vring, num, pages, PAGE_SIZE);
+ vring_init(&vq->vring, num, pages, vring_align);
vq->vq.callback = callback;
vq->vq.vdev = vdev;
vq->vq.vq_ops = &vring_vq_ops;
diff --git a/include/linux/lguest_launcher.h b/include/linux/lguest_launcher.h
index e7217dc..a53407a 100644
--- a/include/linux/lguest_launcher.h
+++ b/include/linux/lguest_launcher.h
@@ -54,9 +54,13 @@ struct lguest_vqconfig {
/* Write command first word is a request. */
enum lguest_req
{
- LHREQ_INITIALIZE, /* + base, pfnlimit, pgdir, start */
+ LHREQ_INITIALIZE, /* + base, pfnlimit, start */
LHREQ_GETDMA, /* No longer used */
LHREQ_IRQ, /* + irq */
LHREQ_BREAK, /* + on/off flag (on blocks until someone does off) */
};
+
+/* The alignment to use between consumer and producer parts of vring.
+ * x86 pagesize for historical reasons. */
+#define LGUEST_VRING_ALIGN 4096
#endif /* _LINUX_LGUEST_LAUNCHER */
diff --git a/include/linux/virtio_balloon.h b/include/linux/virtio_balloon.h
index c30c7bf..8726ff7 100644
--- a/include/linux/virtio_balloon.h
+++ b/include/linux/virtio_balloon.h
@@ -10,6 +10,9 @@
/* The feature bitmap for virtio balloon */
#define VIRTIO_BALLOON_F_MUST_TELL_HOST 0 /* Tell before reclaiming pages */

+/* Size of a PFN in the balloon interface. */
+#define VIRTIO_BALLOON_PFN_SHIFT 12
+
struct virtio_balloon_config
{
/* Number of pages host wants Guest to give up. */
diff --git a/include/linux/virtio_console.h b/include/linux/virtio_console.h
index 19a0da0..7615ffc 100644
--- a/include/linux/virtio_console.h
+++ b/include/linux/virtio_console.h
@@ -7,6 +7,17 @@
/* The ID for virtio console */
#define VIRTIO_ID_CONSOLE 3

+/* Feature bits */
+#define VIRTIO_CONSOLE_F_SIZE 0 /* Does host provide console size? */
+
+struct virtio_console_config {
+ /* colums of the screens */
+ __u16 cols;
+ /* rows of the screens */
+ __u16 rows;
+} __attribute__((packed));
+
+
#ifdef __KERNEL__
int __init virtio_cons_early_init(int (*put_chars)(u32, const char *, int));
#endif /* __KERNEL__ */
diff --git a/include/linux/virtio_pci.h b/include/linux/virtio_pci.h
index cdef357..cd0fd5d 100644
--- a/include/linux/virtio_pci.h
+++ b/include/linux/virtio_pci.h
@@ -53,4 +53,12 @@

/* Virtio ABI version, this must match exactly */
#define VIRTIO_PCI_ABI_VERSION 0
+
+/* How many bits to shift physical queue address written to QUEUE_PFN.
+ * 12 is historical, and due to x86 page size. */
+#define VIRTIO_PCI_QUEUE_ADDR_SHIFT 12
+
+/* The alignment to use between consumer and producer parts of vring.
+ * x86 pagesize again. */
+#define VIRTIO_PCI_VRING_ALIGN 4096
#endif
diff --git a/include/linux/virtio_ring.h b/include/linux/virtio_ring.h
index c4a598f..71e0372 100644
--- a/include/linux/virtio_ring.h
+++ b/include/linux/virtio_ring.h
@@ -83,7 +83,7 @@ struct vring {
* __u16 avail_idx;
* __u16 available[num];
*
- * // Padding to the next page boundary.
+ * // Padding to the next align boundary.
* char pad[];
*
* // A ring of used descriptor heads with free-running index.
@@ -93,19 +93,19 @@ struct vring {
* };
*/
static inline void vring_init(struct vring *vr, unsigned int num, void *p,
- unsigned long pagesize)
+ unsigned long align)
{
vr->num = num;
vr->desc = p;
vr->avail = p + num*sizeof(struct vring_desc);
- vr->used = (void *)(((unsigned long)&vr->avail->ring[num] + pagesize-1)
- & ~(pagesize - 1));
+ vr->used = (void *)(((unsigned long)&vr->avail->ring[num] + align-1)
+ & ~(align - 1));
}

-static inline unsigned vring_size(unsigned int num, unsigned long pagesize)
+static inline unsigned vring_size(unsigned int num, unsigned long align)
{
return ((sizeof(struct vring_desc) * num + sizeof(__u16) * (2 + num)
- + pagesize - 1) & ~(pagesize - 1))
+ + align - 1) & ~(align - 1))
+ sizeof(__u16) * 2 + sizeof(struct vring_used_elem) * num;
}

@@ -115,6 +115,7 @@ struct virtio_device;
struct virtqueue;

struct virtqueue *vring_new_virtqueue(unsigned int num,
+ unsigned int vring_align,
struct virtio_device *vdev,
void *pages,
void (*notify)(struct virtqueue *vq),
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/