Merge tag 'virtio-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux

Pull virtio & lguest updates from Rusty Russell:
 "Lots of virtio work which wasn't quite ready for last merge window.

  Plus I dived into lguest again, reworking the pagetable code so we can
  move the switcher page: our fixmaps sometimes take more than 2MB now..."

Ugh.  Annoying conflicts with the tcm_vhost -> vhost_scsi rename.
Hopefully correctly resolved.

* tag 'virtio-next-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux: (57 commits)
  caif_virtio: Remove bouncing email addresses
  lguest: improve code readability in lg_cpu_start.
  virtio-net: fill only rx queues which are being used
  lguest: map Switcher below fixmap.
  lguest: cache last cpu we ran on.
  lguest: map Switcher text whenever we allocate a new pagetable.
  lguest: don't share Switcher PTE pages between guests.
  lguest: expost switcher_pages array (as lg_switcher_pages).
  lguest: extract shadow PTE walking / allocating.
  lguest: make check_gpte et. al return bool.
  lguest: assume Switcher text is a single page.
  lguest: rename switcher_page to switcher_pages.
  lguest: remove RESERVE_MEM constant.
  lguest: check vaddr not pgd for Switcher protection.
  lguest: prepare to make SWITCHER_ADDR a variable.
  virtio: console: replace EMFILE with EBUSY for already-open port
  virtio-scsi: reset virtqueue affinity when doing cpu hotplug
  virtio-scsi: introduce multiqueue support
  virtio-scsi: push vq lock/unlock into virtscsi_vq_done
  virtio-scsi: pass struct virtio_scsi to virtqueue completion function
  ...

drivers/lguest/Kconfig

@@ -5,10 +5,9 @@ config LGUEST
 	---help---
 	  This is a very simple module which allows you to run
 	  multiple instances of the same Linux kernel, using the
-	  "lguest" command found in the Documentation/virtual/lguest
-	  directory.
+	  "lguest" command found in the tools/lguest directory.
 
 	  Note that "lguest" is pronounced to rhyme with "fell quest",
-	  not "rustyvisor". See Documentation/virtual/lguest/lguest.txt.
+	  not "rustyvisor". See tools/lguest/lguest.txt.
 
 	  If unsure, say N.  If curious, say M.  If masochistic, say Y.

drivers/lguest/core.c

@@ -20,9 +20,9 @@
 #include <asm/asm-offsets.h>
 #include "lg.h"
 
-
+unsigned long switcher_addr;
+struct page **lg_switcher_pages;
 static struct vm_struct *switcher_vma;
-static struct page **switcher_page;
 
 /* This One Big lock protects all inter-guest data structures. */
 DEFINE_MUTEX(lguest_lock);
@@ -52,13 +52,21 @@ static __init int map_switcher(void)
 	 * easy.
 	 */
 
+	/* We assume Switcher text fits into a single page. */
+	if (end_switcher_text - start_switcher_text > PAGE_SIZE) {
+		printk(KERN_ERR "lguest: switcher text too large (%zu)\n",
+		       end_switcher_text - start_switcher_text);
+		return -EINVAL;
+	}
+
 	/*
 	 * We allocate an array of struct page pointers.  map_vm_area() wants
 	 * this, rather than just an array of pages.
 	 */
-	switcher_page = kmalloc(sizeof(switcher_page[0])*TOTAL_SWITCHER_PAGES,
-				GFP_KERNEL);
-	if (!switcher_page) {
+	lg_switcher_pages = kmalloc(sizeof(lg_switcher_pages[0])
+				    * TOTAL_SWITCHER_PAGES,
+				    GFP_KERNEL);
+	if (!lg_switcher_pages) {
 		err = -ENOMEM;
 		goto out;
 	}
@@ -68,32 +76,29 @@ static __init int map_switcher(void)
 	 * so we make sure they're zeroed.
 	 */
 	for (i = 0; i < TOTAL_SWITCHER_PAGES; i++) {
-		switcher_page[i] = alloc_page(GFP_KERNEL|__GFP_ZERO);
-		if (!switcher_page[i]) {
+		lg_switcher_pages[i] = alloc_page(GFP_KERNEL|__GFP_ZERO);
+		if (!lg_switcher_pages[i]) {
 			err = -ENOMEM;
 			goto free_some_pages;
 		}
 	}
 
 	/*
-	 * First we check that the Switcher won't overlap the fixmap area at
-	 * the top of memory.  It's currently nowhere near, but it could have
-	 * very strange effects if it ever happened.
+	 * We place the Switcher underneath the fixmap area, which is the
+	 * highest virtual address we can get.  This is important, since we
+	 * tell the Guest it can't access this memory, so we want its ceiling
+	 * as high as possible.
 	 */
-	if (SWITCHER_ADDR + (TOTAL_SWITCHER_PAGES+1)*PAGE_SIZE > FIXADDR_START){
-		err = -ENOMEM;
-		printk("lguest: mapping switcher would thwack fixmap\n");
-		goto free_pages;
-	}
+	switcher_addr = FIXADDR_START - (TOTAL_SWITCHER_PAGES+1)*PAGE_SIZE;
 
 	/*
-	 * Now we reserve the "virtual memory area" we want: 0xFFC00000
-	 * (SWITCHER_ADDR).  We might not get it in theory, but in practice
-	 * it's worked so far.  The end address needs +1 because __get_vm_area
-	 * allocates an extra guard page, so we need space for that.
+	 * Now we reserve the "virtual memory area" we want.  We might
+	 * not get it in theory, but in practice it's worked so far.
+	 * The end address needs +1 because __get_vm_area allocates an
+	 * extra guard page, so we need space for that.
 	 */
 	switcher_vma = __get_vm_area(TOTAL_SWITCHER_PAGES * PAGE_SIZE,
-				     VM_ALLOC, SWITCHER_ADDR, SWITCHER_ADDR
+				     VM_ALLOC, switcher_addr, switcher_addr
 				     + (TOTAL_SWITCHER_PAGES+1) * PAGE_SIZE);
 	if (!switcher_vma) {
 		err = -ENOMEM;
@@ -103,12 +108,12 @@ static __init int map_switcher(void)
 
 	/*
 	 * This code actually sets up the pages we've allocated to appear at
-	 * SWITCHER_ADDR.  map_vm_area() takes the vma we allocated above, the
+	 * switcher_addr.  map_vm_area() takes the vma we allocated above, the
 	 * kind of pages we're mapping (kernel pages), and a pointer to our
 	 * array of struct pages.  It increments that pointer, but we don't
 	 * care.
 	 */
-	pagep = switcher_page;
+	pagep = lg_switcher_pages;
 	err = map_vm_area(switcher_vma, PAGE_KERNEL_EXEC, &pagep);
 	if (err) {
 		printk("lguest: map_vm_area failed: %i\n", err);
@@ -133,8 +138,8 @@ free_pages:
 	i = TOTAL_SWITCHER_PAGES;
 free_some_pages:
 	for (--i; i >= 0; i--)
-		__free_pages(switcher_page[i], 0);
-	kfree(switcher_page);
+		__free_pages(lg_switcher_pages[i], 0);
+	kfree(lg_switcher_pages);
 out:
 	return err;
 }
@@ -149,8 +154,8 @@ static void unmap_switcher(void)
 	vunmap(switcher_vma->addr);
 	/* Now we just need to free the pages we copied the switcher into */
 	for (i = 0; i < TOTAL_SWITCHER_PAGES; i++)
-		__free_pages(switcher_page[i], 0);
-	kfree(switcher_page);
+		__free_pages(lg_switcher_pages[i], 0);
+	kfree(lg_switcher_pages);
 }
 
 /*H:032
@@ -323,15 +328,10 @@ static int __init init(void)
 	if (err)
 		goto out;
 
-	/* Now we set up the pagetable implementation for the Guests. */
-	err = init_pagetables(switcher_page, SHARED_SWITCHER_PAGES);
-	if (err)
-		goto unmap;
-
 	/* We might need to reserve an interrupt vector. */
 	err = init_interrupts();
 	if (err)
-		goto free_pgtables;
+		goto unmap;
 
 	/* /dev/lguest needs to be registered. */
 	err = lguest_device_init();
@@ -346,8 +346,6 @@ static int __init init(void)
 
 free_interrupts:
 	free_interrupts();
-free_pgtables:
-	free_pagetables();
 unmap:
 	unmap_switcher();
 out:
@@ -359,7 +357,6 @@ static void __exit fini(void)
 {
 	lguest_device_remove();
 	free_interrupts();
-	free_pagetables();
 	unmap_switcher();
 
 	lguest_arch_host_fini();

drivers/lguest/lg.h

@@ -14,11 +14,10 @@
 
 #include <asm/lguest.h>
 
-void free_pagetables(void);
-int init_pagetables(struct page **switcher_page, unsigned int pages);
-
 struct pgdir {
 	unsigned long gpgdir;
+	bool switcher_mapped;
+	int last_host_cpu;
 	pgd_t *pgdir;
 };
 
@@ -124,6 +123,7 @@ bool lguest_address_ok(const struct lguest *lg,
 		       unsigned long addr, unsigned long len);
 void __lgread(struct lg_cpu *, void *, unsigned long, unsigned);
 void __lgwrite(struct lg_cpu *, unsigned long, const void *, unsigned);
+extern struct page **lg_switcher_pages;
 
 /*H:035
  * Using memory-copy operations like that is usually inconvient, so we

drivers/lguest/lguest_user.c

@@ -250,13 +250,13 @@ static ssize_t read(struct file *file, char __user *user, size_t size,loff_t*o)
  */
 static int lg_cpu_start(struct lg_cpu *cpu, unsigned id, unsigned long start_ip)
 {
-	/* We have a limited number the number of CPUs in the lguest struct. */
+	/* We have a limited number of CPUs in the lguest struct. */
 	if (id >= ARRAY_SIZE(cpu->lg->cpus))
 		return -EINVAL;
 
 	/* Set up this CPU's id, and pointer back to the lguest struct. */
 	cpu->id = id;
-	cpu->lg = container_of((cpu - id), struct lguest, cpus[0]);
+	cpu->lg = container_of(cpu, struct lguest, cpus[id]);
 	cpu->lg->nr_cpus++;
 
 	/* Each CPU has a timer it can set. */
@@ -270,7 +270,7 @@ static int lg_cpu_start(struct lg_cpu *cpu, unsigned id, unsigned long start_ip)
 	if (!cpu->regs_page)
 		return -ENOMEM;
 
-	/* We actually put the registers at the bottom of the page. */
+	/* We actually put the registers at the end of the page. */
 	cpu->regs = (void *)cpu->regs_page + PAGE_SIZE - sizeof(*cpu->regs);
 
 	/*

drivers/lguest/x86/core.c

@@ -59,14 +59,13 @@ static struct {
 /* Offset from where switcher.S was compiled to where we've copied it */
 static unsigned long switcher_offset(void)
 {
-	return SWITCHER_ADDR - (unsigned long)start_switcher_text;
+	return switcher_addr - (unsigned long)start_switcher_text;
 }
 
-/* This cpu's struct lguest_pages. */
+/* This cpu's struct lguest_pages (after the Switcher text page) */
 static struct lguest_pages *lguest_pages(unsigned int cpu)
 {
-	return &(((struct lguest_pages *)
-		  (SWITCHER_ADDR + SHARED_SWITCHER_PAGES*PAGE_SIZE))[cpu]);
+	return &(((struct lguest_pages *)(switcher_addr + PAGE_SIZE))[cpu]);
 }
 
 static DEFINE_PER_CPU(struct lg_cpu *, lg_last_cpu);