Merge git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-f…

…or-linus

* git://git.kernel.org/pub/scm/linux/kernel/git/rusty/linux-2.6-for-linus:
  lguest: don't force VIRTIO_F_NOTIFY_ON_EMPTY
  lguest: cleanup for map_switcher()
  lguest: use PGDIR_SHIFT for PAE code to allow different PAGE_OFFSET
  lguest: use set_pte/set_pmd uniformly for real page table entries
  lguest: move panic notifier registration to its expected place.
  virtio_blk: add support for cache flush
  virtio: add virtio IDs file
  virtio: get rid of redundant VIRTIO_ID_9P definition
  virtio: make add_buf return capacity remaining
  virtio_pci: minor MSI-X cleanups

Documentation/lguest/lguest.c

@@ -42,6 +42,7 @@
 #include <signal.h>
 #include "linux/lguest_launcher.h"
 #include "linux/virtio_config.h"
+#include <linux/virtio_ids.h>
 #include "linux/virtio_net.h"
 #include "linux/virtio_blk.h"
 #include "linux/virtio_console.h"
@@ -133,6 +134,9 @@ struct device {
 	/* Is it operational */
 	bool running;
 
+	/* Does Guest want an intrrupt on empty? */
+	bool irq_on_empty;
+
 	/* Device-specific data. */
 	void *priv;
 };
@@ -623,10 +627,13 @@ static void trigger_irq(struct virtqueue *vq)
 		return;
 	vq->pending_used = 0;
 
-	/* If they don't want an interrupt, don't send one, unless empty. */
-	if ((vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT)
-	    && lg_last_avail(vq) != vq->vring.avail->idx)
-		return;
+	/* If they don't want an interrupt, don't send one... */
+	if (vq->vring.avail->flags & VRING_AVAIL_F_NO_INTERRUPT) {
+		/* ... unless they've asked us to force one on empty. */
+		if (!vq->dev->irq_on_empty
+		    || lg_last_avail(vq) != vq->vring.avail->idx)
+			return;
+	}
 
 	/* Send the Guest an interrupt tell them we used something up. */
 	if (write(lguest_fd, buf, sizeof(buf)) != 0)
@@ -1042,6 +1049,15 @@ static void create_thread(struct virtqueue *vq)
 	close(vq->eventfd);
 }
 
+static bool accepted_feature(struct device *dev, unsigned int bit)
+{
+	const u8 *features = get_feature_bits(dev) + dev->feature_len;
+
+	if (dev->feature_len < bit / CHAR_BIT)
+		return false;
+	return features[bit / CHAR_BIT] & (1 << (bit % CHAR_BIT));
+}
+
 static void start_device(struct device *dev)
 {
 	unsigned int i;
@@ -1055,6 +1071,8 @@ static void start_device(struct device *dev)
 		verbose(" %02x", get_feature_bits(dev)
 			[dev->feature_len+i]);
 
+	dev->irq_on_empty = accepted_feature(dev, VIRTIO_F_NOTIFY_ON_EMPTY);
+
 	for (vq = dev->vq; vq; vq = vq->next) {
 		if (vq->service)
 			create_thread(vq);

arch/x86/lguest/boot.c

@@ -1135,11 +1135,6 @@ static struct notifier_block paniced = {
 /* Setting up memory is fairly easy. */
 static __init char *lguest_memory_setup(void)
 {
-	/* We do this here and not earlier because lockcheck used to barf if we
-	 * did it before start_kernel().  I think we fixed that, so it'd be
-	 * nice to move it back to lguest_init.  Patch welcome... */
-	atomic_notifier_chain_register(&panic_notifier_list, &paniced);
-
 	/*
 	 *The Linux bootloader header contains an "e820" memory map: the
 	 * Launcher populated the first entry with our memory limit.
@@ -1364,10 +1359,13 @@ __init void lguest_init(void)
 
 	/*
 	 * If we don't initialize the lock dependency checker now, it crashes
-	 * paravirt_disable_iospace.
+	 * atomic_notifier_chain_register, then paravirt_disable_iospace.
 	 */
 	lockdep_init();
 
+	/* Hook in our special panic hypercall code. */
+	atomic_notifier_chain_register(&panic_notifier_list, &paniced);
+
 	/*
 	 * The IDE code spends about 3 seconds probing for disks: if we reserve
 	 * all the I/O ports up front it can't get them and so doesn't probe.

drivers/block/virtio_blk.c

@@ -3,6 +3,7 @@
 #include <linux/blkdev.h>
 #include <linux/hdreg.h>
 #include <linux/virtio.h>
+#include <linux/virtio_ids.h>
 #include <linux/virtio_blk.h>
 #include <linux/scatterlist.h>
 
@@ -91,15 +92,26 @@ static bool do_req(struct request_queue *q, struct virtio_blk *vblk,
 		return false;
 
 	vbr->req = req;
-	if (blk_fs_request(vbr->req)) {
+	switch (req->cmd_type) {
+	case REQ_TYPE_FS:
 		vbr->out_hdr.type = 0;
 		vbr->out_hdr.sector = blk_rq_pos(vbr->req);
 		vbr->out_hdr.ioprio = req_get_ioprio(vbr->req);
-	} else if (blk_pc_request(vbr->req)) {
+		break;
+	case REQ_TYPE_BLOCK_PC:
 		vbr->out_hdr.type = VIRTIO_BLK_T_SCSI_CMD;
 		vbr->out_hdr.sector = 0;
 		vbr->out_hdr.ioprio = req_get_ioprio(vbr->req);
-	} else {
+		break;
+	case REQ_TYPE_LINUX_BLOCK:
+		if (req->cmd[0] == REQ_LB_OP_FLUSH) {
+			vbr->out_hdr.type = VIRTIO_BLK_T_FLUSH;
+			vbr->out_hdr.sector = 0;
+			vbr->out_hdr.ioprio = req_get_ioprio(vbr->req);
+			break;
+		}
+		/*FALLTHRU*/
+	default:
 		/* We don't put anything else in the queue. */
 		BUG();
 	}
@@ -139,7 +151,7 @@ static bool do_req(struct request_queue *q, struct virtio_blk *vblk,
 		}
 	}
 
-	if (vblk->vq->vq_ops->add_buf(vblk->vq, vblk->sg, out, in, vbr)) {
+	if (vblk->vq->vq_ops->add_buf(vblk->vq, vblk->sg, out, in, vbr) < 0) {
 		mempool_free(vbr, vblk->pool);
 		return false;
 	}
@@ -199,6 +211,12 @@ static int virtblk_identify(struct gendisk *disk, void *argp)
 	return err;
 }
 
+static void virtblk_prepare_flush(struct request_queue *q, struct request *req)
+{
+	req->cmd_type = REQ_TYPE_LINUX_BLOCK;
+	req->cmd[0] = REQ_LB_OP_FLUSH;
+}
+
 static int virtblk_ioctl(struct block_device *bdev, fmode_t mode,
 			 unsigned cmd, unsigned long data)
 {
@@ -337,7 +355,10 @@ static int __devinit virtblk_probe(struct virtio_device *vdev)
 	index++;
 
 	/* If barriers are supported, tell block layer that queue is ordered */
-	if (virtio_has_feature(vdev, VIRTIO_BLK_F_BARRIER))
+	if (virtio_has_feature(vdev, VIRTIO_BLK_F_FLUSH))
+		blk_queue_ordered(vblk->disk->queue, QUEUE_ORDERED_DRAIN_FLUSH,
+				  virtblk_prepare_flush);
+	else if (virtio_has_feature(vdev, VIRTIO_BLK_F_BARRIER))
 		blk_queue_ordered(vblk->disk->queue, QUEUE_ORDERED_TAG, NULL);
 
 	/* If disk is read-only in the host, the guest should obey */
@@ -424,7 +445,7 @@ static struct virtio_device_id id_table[] = {
 static unsigned int features[] = {
 	VIRTIO_BLK_F_BARRIER, VIRTIO_BLK_F_SEG_MAX, VIRTIO_BLK_F_SIZE_MAX,
 	VIRTIO_BLK_F_GEOMETRY, VIRTIO_BLK_F_RO, VIRTIO_BLK_F_BLK_SIZE,
-	VIRTIO_BLK_F_SCSI, VIRTIO_BLK_F_IDENTIFY
+	VIRTIO_BLK_F_SCSI, VIRTIO_BLK_F_IDENTIFY, VIRTIO_BLK_F_FLUSH
 };
 
 /*

drivers/char/hw_random/virtio-rng.c

@@ -21,6 +21,7 @@
 #include <linux/scatterlist.h>
 #include <linux/spinlock.h>
 #include <linux/virtio.h>
+#include <linux/virtio_ids.h>
 #include <linux/virtio_rng.h>
 
 /* The host will fill any buffer we give it with sweet, sweet randomness.  We
@@ -51,7 +52,7 @@ static void register_buffer(void)
 
 	sg_init_one(&sg, random_data+data_left, RANDOM_DATA_SIZE-data_left);
 	/* There should always be room for one buffer. */
-	if (vq->vq_ops->add_buf(vq, &sg, 0, 1, random_data) != 0)
+	if (vq->vq_ops->add_buf(vq, &sg, 0, 1, random_data) < 0)
 		BUG();
 	vq->vq_ops->kick(vq);
 }

drivers/char/virtio_console.c

@@ -31,6 +31,7 @@
 #include <linux/err.h>
 #include <linux/init.h>
 #include <linux/virtio.h>
+#include <linux/virtio_ids.h>
 #include <linux/virtio_console.h>
 #include "hvc_console.h"
 
@@ -65,7 +66,7 @@ static int put_chars(u32 vtermno, const char *buf, int count)
 
 	/* add_buf wants a token to identify this buffer: we hand it any
 	 * non-NULL pointer, since there's only ever one buffer. */
-	if (out_vq->vq_ops->add_buf(out_vq, sg, 1, 0, (void *)1) == 0) {
+	if (out_vq->vq_ops->add_buf(out_vq, sg, 1, 0, (void *)1) >= 0) {
 		/* Tell Host to go! */
 		out_vq->vq_ops->kick(out_vq);
 		/* Chill out until it's done with the buffer. */
@@ -85,7 +86,7 @@ static void add_inbuf(void)
 	sg_init_one(sg, inbuf, PAGE_SIZE);
 
 	/* We should always be able to add one buffer to an empty queue. */
-	if (in_vq->vq_ops->add_buf(in_vq, sg, 0, 1, inbuf) != 0)
+	if (in_vq->vq_ops->add_buf(in_vq, sg, 0, 1, inbuf) < 0)
 		BUG();
 	in_vq->vq_ops->kick(in_vq);
 }

drivers/lguest/core.c

@@ -67,12 +67,11 @@ static __init int map_switcher(void)
 	 * so we make sure they're zeroed.
 	 */
 	for (i = 0; i < TOTAL_SWITCHER_PAGES; i++) {
-		unsigned long addr = get_zeroed_page(GFP_KERNEL);
-		if (!addr) {
+		switcher_page[i] = alloc_page(GFP_KERNEL|__GFP_ZERO);
+		if (!switcher_page[i]) {
 			err = -ENOMEM;
 			goto free_some_pages;
 		}
-		switcher_page[i] = virt_to_page(addr);
 	}
 
 	/*

drivers/lguest/page_tables.c

@@ -380,7 +380,7 @@ bool demand_page(struct lg_cpu *cpu, unsigned long vaddr, int errcode)
 		 * And we copy the flags to the shadow PMD entry.  The page
 		 * number in the shadow PMD is the page we just allocated.
 		 */
-		native_set_pmd(spmd, __pmd(__pa(ptepage) | pmd_flags(gpmd)));
+		set_pmd(spmd, __pmd(__pa(ptepage) | pmd_flags(gpmd)));
 	}
 
 	/*
@@ -447,7 +447,7 @@ bool demand_page(struct lg_cpu *cpu, unsigned long vaddr, int errcode)
 		 * we will come back here when a write does actually occur, so
 		 * we can update the Guest's _PAGE_DIRTY flag.
 		 */
-		native_set_pte(spte, gpte_to_spte(cpu, pte_wrprotect(gpte), 0));
+		set_pte(spte, gpte_to_spte(cpu, pte_wrprotect(gpte), 0));
 
 	/*
 	 * Finally, we write the Guest PTE entry back: we've set the
@@ -528,7 +528,7 @@ static void release_pmd(pmd_t *spmd)
 		/* Now we can free the page of PTEs */
 		free_page((long)ptepage);
 		/* And zero out the PMD entry so we never release it twice. */
-		native_set_pmd(spmd, __pmd(0));
+		set_pmd(spmd, __pmd(0));
 	}
 }
 
@@ -833,15 +833,15 @@ static void do_set_pte(struct lg_cpu *cpu, int idx,
 			 */
 			if (pte_flags(gpte) & (_PAGE_DIRTY | _PAGE_ACCESSED)) {
 				check_gpte(cpu, gpte);
-				native_set_pte(spte,
-						gpte_to_spte(cpu, gpte,
+				set_pte(spte,
+					gpte_to_spte(cpu, gpte,
 						pte_flags(gpte) & _PAGE_DIRTY));
 			} else {
 				/*
 				 * Otherwise kill it and we can demand_page()
 				 * it in later.
 				 */
-				native_set_pte(spte, __pte(0));
+				set_pte(spte, __pte(0));
 			}
 #ifdef CONFIG_X86_PAE
 		}
@@ -983,25 +983,22 @@ static unsigned long setup_pagetables(struct lguest *lg,
 	 */
 	for (i = j = 0; i < mapped_pages && j < PTRS_PER_PMD;
 	     i += PTRS_PER_PTE, j++) {
-		/* FIXME: native_set_pmd is overkill here. */
-		native_set_pmd(&pmd, __pmd(((unsigned long)(linear + i)
-		- mem_base) | _PAGE_PRESENT | _PAGE_RW | _PAGE_USER));
+		pmd = pfn_pmd(((unsigned long)&linear[i] - mem_base)/PAGE_SIZE,
+			      __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_USER));
 
 		if (copy_to_user(&pmds[j], &pmd, sizeof(pmd)) != 0)
 			return -EFAULT;
 	}
 
 	/* One PGD entry, pointing to that PMD page. */
-	set_pgd(&pgd, __pgd(((u32)pmds - mem_base) | _PAGE_PRESENT));
+	pgd = __pgd(((unsigned long)pmds - mem_base) | _PAGE_PRESENT);
 	/* Copy it in as the first PGD entry (ie. addresses 0-1G). */
 	if (copy_to_user(&pgdir[0], &pgd, sizeof(pgd)) != 0)
 		return -EFAULT;
 	/*
-	 * And the third PGD entry (ie. addresses 3G-4G).
-	 *
-	 * FIXME: This assumes that PAGE_OFFSET for the Guest is 0xC0000000.
+	 * And the other PGD entry to make the linear mapping at PAGE_OFFSET
 	 */
-	if (copy_to_user(&pgdir[3], &pgd, sizeof(pgd)) != 0)
+	if (copy_to_user(&pgdir[KERNEL_PGD_BOUNDARY], &pgd, sizeof(pgd)))
 		return -EFAULT;
 #else
 	/*
@@ -1141,23 +1138,21 @@ void map_switcher_in_guest(struct lg_cpu *cpu, struct lguest_pages *pages)
 {
 	pte_t *switcher_pte_page = __get_cpu_var(switcher_pte_pages);
 	pte_t regs_pte;
-	unsigned long pfn;
 
 #ifdef CONFIG_X86_PAE
 	pmd_t switcher_pmd;
 	pmd_t *pmd_table;
 
-	/* FIXME: native_set_pmd is overkill here. */
-	native_set_pmd(&switcher_pmd, pfn_pmd(__pa(switcher_pte_page) >>
-		       PAGE_SHIFT, PAGE_KERNEL_EXEC));
+	switcher_pmd = pfn_pmd(__pa(switcher_pte_page) >> PAGE_SHIFT,
+			       PAGE_KERNEL_EXEC);
 
 	/* Figure out where the pmd page is, by reading the PGD, and converting
 	 * it to a virtual address. */
 	pmd_table = __va(pgd_pfn(cpu->lg->
 			pgdirs[cpu->cpu_pgd].pgdir[SWITCHER_PGD_INDEX])
 								<< PAGE_SHIFT);
 	/* Now write it into the shadow page table. */
-	native_set_pmd(&pmd_table[SWITCHER_PMD_INDEX], switcher_pmd);
+	set_pmd(&pmd_table[SWITCHER_PMD_INDEX], switcher_pmd);
 #else
 	pgd_t switcher_pgd;
 
@@ -1179,10 +1174,8 @@ void map_switcher_in_guest(struct lg_cpu *cpu, struct lguest_pages *pages)
 	 * page is already mapped there, we don't have to copy them out
 	 * again.
 	 */
-	pfn = __pa(cpu->regs_page) >> PAGE_SHIFT;
-	native_set_pte(&regs_pte, pfn_pte(pfn, PAGE_KERNEL));
-	native_set_pte(&switcher_pte_page[pte_index((unsigned long)pages)],
-			regs_pte);
+	regs_pte = pfn_pte(__pa(cpu->regs_page) >> PAGE_SHIFT, PAGE_KERNEL);
+	set_pte(&switcher_pte_page[pte_index((unsigned long)pages)], regs_pte);
 }
 /*:*/
 
@@ -1209,22 +1202,22 @@ static __init void populate_switcher_pte_page(unsigned int cpu,
 
 	/* The first entries are easy: they map the Switcher code. */
 	for (i = 0; i < pages; i++) {
-		native_set_pte(&pte[i], mk_pte(switcher_page[i],
+		set_pte(&pte[i], mk_pte(switcher_page[i],
 				__pgprot(_PAGE_PRESENT|_PAGE_ACCESSED)));
 	}
 
 	/* The only other thing we map is this CPU's pair of pages. */
 	i = pages + cpu*2;
 
 	/* First page (Guest registers) is writable from the Guest */
-	native_set_pte(&pte[i], pfn_pte(page_to_pfn(switcher_page[i]),
+	set_pte(&pte[i], pfn_pte(page_to_pfn(switcher_page[i]),
 			 __pgprot(_PAGE_PRESENT|_PAGE_ACCESSED|_PAGE_RW)));
 
 	/*
 	 * The second page contains the "struct lguest_ro_state", and is
 	 * read-only.
 	 */
-	native_set_pte(&pte[i+1], pfn_pte(page_to_pfn(switcher_page[i+1]),
+	set_pte(&pte[i+1], pfn_pte(page_to_pfn(switcher_page[i+1]),
 			   __pgprot(_PAGE_PRESENT|_PAGE_ACCESSED)));
 }