Merge branch 'for-rmk' of git://gitorious.org/linux-gemini/mainline

Documentation/kernel-parameters.txt

@@ -1535,6 +1535,10 @@ and is between 256 and 4096 characters. It is defined in the file
 			register save and restore. The kernel will only save
 			legacy floating-point registers on task switch.
 
+	noxsave		[BUGS=X86] Disables x86 extended register state save
+			and restore using xsave. The kernel will fallback to
+			enabling legacy floating-point and sse state.
+
 	nohlt		[BUGS=ARM,SH] Tells the kernel that the sleep(SH) or
 			wfi(ARM) instruction doesn't work correctly and not to
 			use it. This is also useful when using JTAG debugger.

Documentation/sound/alsa/HD-Audio-Models.txt

@@ -334,6 +334,7 @@ STAC9227/9228/9229/927x
   ref-no-jd	Reference board without HP/Mic jack detection
   3stack	D965 3stack
   5stack	D965 5stack + SPDIF
+  5stack-no-fp	D965 5stack without front panel
   dell-3stack	Dell Dimension E520
   dell-bios	Fixes with Dell BIOS setup
   auto		BIOS setup (default)

MAINTAINERS

@@ -624,6 +624,7 @@ M:	paulius.zaleckas@teltonika.lt
 L:	linux-arm-kernel@lists.arm.linux.org.uk (subscribers-only)
 T:	git git://gitorious.org/linux-gemini/mainline.git
 S:	Maintained
+F:	arch/arm/mach-gemini/
 
 ARM/EBSA110 MACHINE SUPPORT
 P:	Russell King
@@ -650,6 +651,7 @@ P:	Paulius Zaleckas
 M:	paulius.zaleckas@teltonika.lt
 L:	linux-arm-kernel@lists.arm.linux.org.uk (subscribers-only)
 S:	Maintained
+F:	arch/arm/mm/*-fa*
 
 ARM/FOOTBRIDGE ARCHITECTURE
 P:	Russell King
@@ -1132,17 +1134,17 @@ F:	fs/bfs/
 F:	include/linux/bfs_fs.h
 
 BLACKFIN ARCHITECTURE
-P:	Bryan Wu
-M:	cooloney@kernel.org
+P:	Mike Frysinger
+M:	vapier@gentoo.org
 L:	uclinux-dist-devel@blackfin.uclinux.org
 W:	http://blackfin.uclinux.org
 S:	Supported
 F:	arch/blackfin/
 
 BLACKFIN EMAC DRIVER
-P:	Bryan Wu
-M:	cooloney@kernel.org
-L:	uclinux-dist-devel@blackfin.uclinux.org (subscribers-only)
+P:	Michael Hennerich
+M:	michael.hennerich@analog.com
+L:	uclinux-dist-devel@blackfin.uclinux.org
 W:	http://blackfin.uclinux.org
 S:	Supported
 F:	drivers/net/bfin_mac.*
@@ -1150,7 +1152,7 @@ F:	drivers/net/bfin_mac.*
 BLACKFIN RTC DRIVER
 P:	Mike Frysinger
 M:	vapier.adi@gmail.com
-L:	uclinux-dist-devel@blackfin.uclinux.org (subscribers-only)
+L:	uclinux-dist-devel@blackfin.uclinux.org
 W:	http://blackfin.uclinux.org
 S:	Supported
 F:	drivers/rtc/rtc-bfin.c
@@ -1158,7 +1160,7 @@ F:	drivers/rtc/rtc-bfin.c
 BLACKFIN SERIAL DRIVER
 P:	Sonic Zhang
 M:	sonic.zhang@analog.com
-L:	uclinux-dist-devel@blackfin.uclinux.org (subscribers-only)
+L:	uclinux-dist-devel@blackfin.uclinux.org
 W:	http://blackfin.uclinux.org
 S:	Supported
 F:	drivers/serial/bfin_5xx.c
@@ -1166,7 +1168,7 @@ F:	drivers/serial/bfin_5xx.c
 BLACKFIN WATCHDOG DRIVER
 P:	Mike Frysinger
 M:	vapier.adi@gmail.com
-L:	uclinux-dist-devel@blackfin.uclinux.org (subscribers-only)
+L:	uclinux-dist-devel@blackfin.uclinux.org
 W:	http://blackfin.uclinux.org
 S:	Supported
 F:	drivers/watchdog/bfin_wdt.c
@@ -1174,7 +1176,7 @@ F:	drivers/watchdog/bfin_wdt.c
 BLACKFIN I2C TWI DRIVER
 P:	Sonic Zhang
 M:	sonic.zhang@analog.com
-L:	uclinux-dist-devel@blackfin.uclinux.org (subscribers-only)
+L:	uclinux-dist-devel@blackfin.uclinux.org
 W:	http://blackfin.uclinux.org/
 S:	Supported
 F:	drivers/i2c/busses/i2c-bfin-twi.c
@@ -1431,6 +1433,14 @@ P:	Russell King
 M:	linux@arm.linux.org.uk
 F:	include/linux/clk.h
 
+CISCO FCOE HBA DRIVER
+P:	Abhijeet Joglekar
+M:	abjoglek@cisco.com
+P:	Joe Eykholt
+M:	jeykholt@cisco.com
+L:	linux-scsi@vger.kernel.org
+S:	Supported
+
 CODA FILE SYSTEM
 P:	Jan Harkes
 M:	jaharkes@cs.cmu.edu

Makefile

@@ -1,8 +1,8 @@
 VERSION = 2
 PATCHLEVEL = 6
 SUBLEVEL = 30
-EXTRAVERSION = -rc6
-NAME = Vindictive Armadillo
+EXTRAVERSION = -rc7
+NAME = Man-Eating Seals of Antiquity
 
 # *DOCUMENTATION*
 # To see a list of typical targets execute "make help"

arch/arm/mach-gemini/include/mach/hardware.h

@@ -15,10 +15,9 @@
 /*
  * Memory Map definitions
  */
-/* FIXME: Does it really swap SRAM like this? */
 #ifdef CONFIG_GEMINI_MEM_SWAP
 # define GEMINI_DRAM_BASE	0x00000000
-# define GEMINI_SRAM_BASE	0x20000000
+# define GEMINI_SRAM_BASE	0x70000000
 #else
 # define GEMINI_SRAM_BASE	0x00000000
 # define GEMINI_DRAM_BASE	0x10000000

arch/blackfin/include/asm/.gitignore

@@ -1 +0,0 @@
-+mach

arch/blackfin/include/asm/unistd.h

@@ -378,8 +378,10 @@
 #define __NR_dup3		363
 #define __NR_pipe2		364
 #define __NR_inotify_init1	365
+#define __NR_preadv		366
+#define __NR_pwritev		367
 
-#define __NR_syscall		366
+#define __NR_syscall		368
 #define NR_syscalls		__NR_syscall
 
 /* Old optional stuff no one actually uses */

arch/blackfin/kernel/.gitignore

@@ -0,0 +1 @@
+vmlinux.lds

arch/blackfin/lib/strncmp.c

@@ -8,9 +8,8 @@
 
 #define strncmp __inline_strncmp
 #include <asm/string.h>
-#undef strncmp
-
 #include <linux/module.h>
+#undef strncmp
 
 int strncmp(const char *cs, const char *ct, size_t count)
 {

arch/blackfin/mach-common/entry.S

@@ -1581,6 +1581,8 @@ ENTRY(_sys_call_table)
 	.long _sys_dup3
 	.long _sys_pipe2
 	.long _sys_inotify_init1	/* 365 */
+	.long _sys_preadv
+	.long _sys_pwritev
 
 	.rept NR_syscalls-(.-_sys_call_table)/4
 	.long _sys_ni_syscall

arch/powerpc/Kconfig

@@ -868,6 +868,18 @@ config TASK_SIZE
 	default "0x80000000" if PPC_PREP || PPC_8xx
 	default "0xc0000000"
 
+config CONSISTENT_SIZE_BOOL
+	bool "Set custom consistent memory pool size"
+	depends on ADVANCED_OPTIONS && NOT_COHERENT_CACHE
+	help
+	  This option allows you to set the size of the
+	  consistent memory pool.  This pool of virtual memory
+	  is used to make consistent memory allocations.
+
+config CONSISTENT_SIZE
+	hex "Size of consistent memory pool" if CONSISTENT_SIZE_BOOL
+	default "0x00200000" if NOT_COHERENT_CACHE
+
 config PIN_TLB
 	bool "Pinned Kernel TLBs (860 ONLY)"
 	depends on ADVANCED_OPTIONS && 8xx

arch/powerpc/include/asm/dma-mapping.h

@@ -26,7 +26,9 @@
  * allocate the space "normally" and use the cache management functions
  * to ensure it is consistent.
  */
-extern void *__dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp);
+struct device;
+extern void *__dma_alloc_coherent(struct device *dev, size_t size,
+				  dma_addr_t *handle, gfp_t gfp);
 extern void __dma_free_coherent(size_t size, void *vaddr);
 extern void __dma_sync(void *vaddr, size_t size, int direction);
 extern void __dma_sync_page(struct page *page, unsigned long offset,
@@ -37,7 +39,7 @@ extern void __dma_sync_page(struct page *page, unsigned long offset,
  * Cache coherent cores.
  */
 
-#define __dma_alloc_coherent(gfp, size, handle)	NULL
+#define __dma_alloc_coherent(dev, gfp, size, handle)	NULL
 #define __dma_free_coherent(size, addr)		((void)0)
 #define __dma_sync(addr, size, rw)		((void)0)
 #define __dma_sync_page(pg, off, sz, rw)	((void)0)

arch/powerpc/include/asm/fixmap.h

@@ -14,8 +14,6 @@
 #ifndef _ASM_FIXMAP_H
 #define _ASM_FIXMAP_H
 
-extern unsigned long FIXADDR_TOP;
-
 #ifndef __ASSEMBLY__
 #include <linux/kernel.h>
 #include <asm/page.h>
@@ -24,6 +22,8 @@ extern unsigned long FIXADDR_TOP;
 #include <asm/kmap_types.h>
 #endif
 
+#define FIXADDR_TOP	((unsigned long)(-PAGE_SIZE))
+
 /*
  * Here we define all the compile-time 'special' virtual
  * addresses. The point is to have a constant address at

arch/powerpc/include/asm/pgtable-ppc32.h

@@ -10,7 +10,7 @@
 
 extern unsigned long va_to_phys(unsigned long address);
 extern pte_t *va_to_pte(unsigned long address);
-extern unsigned long ioremap_bot, ioremap_base;
+extern unsigned long ioremap_bot;
 
 #ifdef CONFIG_44x
 extern int icache_44x_need_flush;
@@ -55,9 +55,31 @@ extern int icache_44x_need_flush;
 #define pgd_ERROR(e) \
 	printk("%s:%d: bad pgd %08lx.\n", __FILE__, __LINE__, pgd_val(e))
 
+/*
+ * This is the bottom of the PKMAP area with HIGHMEM or an arbitrary
+ * value (for now) on others, from where we can start layout kernel
+ * virtual space that goes below PKMAP and FIXMAP
+ */
+#ifdef CONFIG_HIGHMEM
+#define KVIRT_TOP	PKMAP_BASE
+#else
+#define KVIRT_TOP	(0xfe000000UL)	/* for now, could be FIXMAP_BASE ? */
+#endif
+
+/*
+ * ioremap_bot starts at that address. Early ioremaps move down from there,
+ * until mem_init() at which point this becomes the top of the vmalloc
+ * and ioremap space
+ */
+#ifdef CONFIG_NOT_COHERENT_CACHE
+#define IOREMAP_TOP	((KVIRT_TOP - CONFIG_CONSISTENT_SIZE) & PAGE_MASK)
+#else
+#define IOREMAP_TOP	KVIRT_TOP
+#endif
+
 /*
  * Just any arbitrary offset to the start of the vmalloc VM area: the
- * current 64MB value just means that there will be a 64MB "hole" after the
+ * current 16MB value just means that there will be a 64MB "hole" after the
  * physical memory until the kernel virtual memory starts.  That means that
  * any out-of-bounds memory accesses will hopefully be caught.
  * The vmalloc() routines leaves a hole of 4kB between each vmalloced

arch/powerpc/kernel/dma.c

@@ -32,7 +32,7 @@ void *dma_direct_alloc_coherent(struct device *dev, size_t size,
 {
 	void *ret;
 #ifdef CONFIG_NOT_COHERENT_CACHE
-	ret = __dma_alloc_coherent(size, dma_handle, flag);
+	ret = __dma_alloc_coherent(dev, size, dma_handle, flag);
 	if (ret == NULL)
 		return NULL;
 	*dma_handle += get_dma_direct_offset(dev);

arch/powerpc/lib/Makefile

@@ -18,7 +18,6 @@ obj-$(CONFIG_PPC64)	+= copypage_64.o copyuser_64.o \
 			   memcpy_64.o usercopy_64.o mem_64.o string.o
 obj-$(CONFIG_XMON)	+= sstep.o
 obj-$(CONFIG_KPROBES)	+= sstep.o
-obj-$(CONFIG_NOT_COHERENT_CACHE)	+= dma-noncoherent.o
 
 ifeq ($(CONFIG_PPC64),y)
 obj-$(CONFIG_SMP)	+= locks.o

arch/powerpc/lib/dma-noncoherent.c

@@ -1,237 +0,0 @@
-/*
- *  PowerPC version derived from arch/arm/mm/consistent.c
- *    Copyright (C) 2001 Dan Malek (dmalek@jlc.net)
- *
- *  Copyright (C) 2000 Russell King
- *
- * Consistent memory allocators.  Used for DMA devices that want to
- * share uncached memory with the processor core.  The function return
- * is the virtual address and 'dma_handle' is the physical address.
- * Mostly stolen from the ARM port, with some changes for PowerPC.
- *						-- Dan
- *
- * Reorganized to get rid of the arch-specific consistent_* functions
- * and provide non-coherent implementations for the DMA API. -Matt
- *
- * Added in_interrupt() safe dma_alloc_coherent()/dma_free_coherent()
- * implementation. This is pulled straight from ARM and barely
- * modified. -Matt
- *
- * This program is free software; you can redistribute it and/or modify
- * it under the terms of the GNU General Public License version 2 as
- * published by the Free Software Foundation.
- */
-
-#include <linux/sched.h>
-#include <linux/kernel.h>
-#include <linux/errno.h>
-#include <linux/string.h>
-#include <linux/types.h>
-#include <linux/highmem.h>
-#include <linux/dma-mapping.h>
-#include <linux/vmalloc.h>
-
-#include <asm/tlbflush.h>
-
-/*
- * Allocate DMA-coherent memory space and return both the kernel remapped
- * virtual and bus address for that space.
- */
-void *
-__dma_alloc_coherent(size_t size, dma_addr_t *handle, gfp_t gfp)
-{
-	struct page *page;
-	unsigned long order;
-	int i;
-	unsigned int nr_pages = PAGE_ALIGN(size)>>PAGE_SHIFT;
-	unsigned int array_size = nr_pages * sizeof(struct page *);
-	struct page **pages;
-	struct page *end;
-	u64 mask = 0x00ffffff, limit; /* ISA default */
-	struct vm_struct *area;
-
-	BUG_ON(!mem_init_done);
-	size = PAGE_ALIGN(size);
-	limit = (mask + 1) & ~mask;
-	if (limit && size >= limit) {
-		printk(KERN_WARNING "coherent allocation too big (requested "
-				"%#x mask %#Lx)\n", size, mask);
-		return NULL;
-	}
-
-	order = get_order(size);
-
-	if (mask != 0xffffffff)
-		gfp |= GFP_DMA;
-
-	page = alloc_pages(gfp, order);
-	if (!page)
-		goto no_page;
-
-	end = page + (1 << order);
-
-	/*
-	 * Invalidate any data that might be lurking in the
-	 * kernel direct-mapped region for device DMA.
-	 */
-	{
-		unsigned long kaddr = (unsigned long)page_address(page);
-		memset(page_address(page), 0, size);
-		flush_dcache_range(kaddr, kaddr + size);
-	}
-
-	split_page(page, order);
-
-	/*
-	 * Set the "dma handle"
-	 */
-	*handle = page_to_phys(page);
-
-	area = get_vm_area_caller(size, VM_IOREMAP,
-			__builtin_return_address(1));
-	if (!area)
-		goto out_free_pages;
-
-	if (array_size > PAGE_SIZE) {
-		pages = vmalloc(array_size);
-		area->flags |= VM_VPAGES;
-	} else {
-		pages = kmalloc(array_size, GFP_KERNEL);
-	}
-	if (!pages)
-		goto out_free_area;
-
-	area->pages = pages;
-	area->nr_pages = nr_pages;
-
-	for (i = 0; i < nr_pages; i++)
-		pages[i] = page + i;
-
-	if (map_vm_area(area, pgprot_noncached(PAGE_KERNEL), &pages))
-		goto out_unmap;
-
-	/*
-	 * Free the otherwise unused pages.
-	 */
-	page += nr_pages;
-	while (page < end) {
-		__free_page(page);
-		page++;
-	}
-
-	return area->addr;
-out_unmap:
-	vunmap(area->addr);
-	if (array_size > PAGE_SIZE)
-		vfree(pages);
-	else
-		kfree(pages);
-	goto out_free_pages;
-out_free_area:
-	free_vm_area(area);
-out_free_pages:
-	if (page)
-		__free_pages(page, order);
-no_page:
-	return NULL;
-}
-EXPORT_SYMBOL(__dma_alloc_coherent);
-
-/*
- * free a page as defined by the above mapping.
- */
-void __dma_free_coherent(size_t size, void *vaddr)
-{
-	vfree(vaddr);
-
-}
-EXPORT_SYMBOL(__dma_free_coherent);
-
-/*
- * make an area consistent.
- */
-void __dma_sync(void *vaddr, size_t size, int direction)
-{
-	unsigned long start = (unsigned long)vaddr;
-	unsigned long end   = start + size;
-
-	switch (direction) {
-	case DMA_NONE:
-		BUG();
-	case DMA_FROM_DEVICE:
-		/*
-		 * invalidate only when cache-line aligned otherwise there is
-		 * the potential for discarding uncommitted data from the cache
-		 */
-		if ((start & (L1_CACHE_BYTES - 1)) || (size & (L1_CACHE_BYTES - 1)))
-			flush_dcache_range(start, end);
-		else
-			invalidate_dcache_range(start, end);
-		break;
-	case DMA_TO_DEVICE:		/* writeback only */
-		clean_dcache_range(start, end);
-		break;
-	case DMA_BIDIRECTIONAL:	/* writeback and invalidate */
-		flush_dcache_range(start, end);
-		break;
-	}
-}
-EXPORT_SYMBOL(__dma_sync);
-
-#ifdef CONFIG_HIGHMEM
-/*
- * __dma_sync_page() implementation for systems using highmem.
- * In this case, each page of a buffer must be kmapped/kunmapped
- * in order to have a virtual address for __dma_sync(). This must
- * not sleep so kmap_atomic()/kunmap_atomic() are used.
- *
- * Note: yes, it is possible and correct to have a buffer extend
- * beyond the first page.
- */
-static inline void __dma_sync_page_highmem(struct page *page,
-		unsigned long offset, size_t size, int direction)
-{
-	size_t seg_size = min((size_t)(PAGE_SIZE - offset), size);
-	size_t cur_size = seg_size;
-	unsigned long flags, start, seg_offset = offset;
-	int nr_segs = 1 + ((size - seg_size) + PAGE_SIZE - 1)/PAGE_SIZE;
-	int seg_nr = 0;
-
-	local_irq_save(flags);
-
-	do {
-		start = (unsigned long)kmap_atomic(page + seg_nr,
-				KM_PPC_SYNC_PAGE) + seg_offset;
-
-		/* Sync this buffer segment */
-		__dma_sync((void *)start, seg_size, direction);
-		kunmap_atomic((void *)start, KM_PPC_SYNC_PAGE);
-		seg_nr++;
-
-		/* Calculate next buffer segment size */
-		seg_size = min((size_t)PAGE_SIZE, size - cur_size);
-
-		/* Add the segment size to our running total */
-		cur_size += seg_size;
-		seg_offset = 0;
-	} while (seg_nr < nr_segs);
-
-	local_irq_restore(flags);
-}
-#endif /* CONFIG_HIGHMEM */
-
-/*
- * __dma_sync_page makes memory consistent. identical to __dma_sync, but
- * takes a struct page instead of a virtual address
- */
-void __dma_sync_page(struct page *page, unsigned long offset,
-	size_t size, int direction)
-{
-#ifdef CONFIG_HIGHMEM
-	__dma_sync_page_highmem(page, offset, size, direction);
-#else
-	unsigned long start = (unsigned long)page_address(page) + offset;
-	__dma_sync((void *)start, size, direction);
-#endif
-}
-EXPORT_SYMBOL(__dma_sync_page);

arch/powerpc/mm/Makefile

@@ -26,3 +26,4 @@ obj-$(CONFIG_NEED_MULTIPLE_NODES) += numa.o
 obj-$(CONFIG_PPC_MM_SLICES)	+= slice.o
 obj-$(CONFIG_HUGETLB_PAGE)	+= hugetlbpage.o
 obj-$(CONFIG_PPC_SUBPAGE_PROT)	+= subpage-prot.o
+obj-$(CONFIG_NOT_COHERENT_CACHE) += dma-noncoherent.o

arch/powerpc/mm/dma-noncoherent.c

@@ -0,0 +1,400 @@
+/*
+ *  PowerPC version derived from arch/arm/mm/consistent.c
+ *    Copyright (C) 2001 Dan Malek (dmalek@jlc.net)
+ *
+ *  Copyright (C) 2000 Russell King
+ *
+ * Consistent memory allocators.  Used for DMA devices that want to
+ * share uncached memory with the processor core.  The function return
+ * is the virtual address and 'dma_handle' is the physical address.
+ * Mostly stolen from the ARM port, with some changes for PowerPC.
+ *						-- Dan
+ *
+ * Reorganized to get rid of the arch-specific consistent_* functions
+ * and provide non-coherent implementations for the DMA API. -Matt
+ *
+ * Added in_interrupt() safe dma_alloc_coherent()/dma_free_coherent()
+ * implementation. This is pulled straight from ARM and barely
+ * modified. -Matt
+ *
+ * This program is free software; you can redistribute it and/or modify
+ * it under the terms of the GNU General Public License version 2 as
+ * published by the Free Software Foundation.
+ */
+
+#include <linux/sched.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/string.h>
+#include <linux/types.h>
+#include <linux/highmem.h>
+#include <linux/dma-mapping.h>
+
+#include <asm/tlbflush.h>
+
+#include "mmu_decl.h"
+
+/*
+ * This address range defaults to a value that is safe for all
+ * platforms which currently set CONFIG_NOT_COHERENT_CACHE. It
+ * can be further configured for specific applications under
+ * the "Advanced Setup" menu. -Matt
+ */
+#define CONSISTENT_BASE		(IOREMAP_TOP)
+#define CONSISTENT_END 		(CONSISTENT_BASE + CONFIG_CONSISTENT_SIZE)
+#define CONSISTENT_OFFSET(x)	(((unsigned long)(x) - CONSISTENT_BASE) >> PAGE_SHIFT)
+
+/*
+ * This is the page table (2MB) covering uncached, DMA consistent allocations
+ */
+static DEFINE_SPINLOCK(consistent_lock);
+
+/*
+ * VM region handling support.
+ *
+ * This should become something generic, handling VM region allocations for
+ * vmalloc and similar (ioremap, module space, etc).
+ *
+ * I envisage vmalloc()'s supporting vm_struct becoming:
+ *
+ *  struct vm_struct {
+ *    struct vm_region	region;
+ *    unsigned long	flags;
+ *    struct page	**pages;
+ *    unsigned int	nr_pages;
+ *    unsigned long	phys_addr;
+ *  };
+ *
+ * get_vm_area() would then call vm_region_alloc with an appropriate
+ * struct vm_region head (eg):
+ *
+ *  struct vm_region vmalloc_head = {
+ *	.vm_list	= LIST_HEAD_INIT(vmalloc_head.vm_list),
+ *	.vm_start	= VMALLOC_START,
+ *	.vm_end		= VMALLOC_END,
+ *  };
+ *
+ * However, vmalloc_head.vm_start is variable (typically, it is dependent on
+ * the amount of RAM found at boot time.)  I would imagine that get_vm_area()
+ * would have to initialise this each time prior to calling vm_region_alloc().
+ */
+struct ppc_vm_region {
+	struct list_head	vm_list;
+	unsigned long		vm_start;
+	unsigned long		vm_end;
+};
+
+static struct ppc_vm_region consistent_head = {
+	.vm_list	= LIST_HEAD_INIT(consistent_head.vm_list),
+	.vm_start	= CONSISTENT_BASE,
+	.vm_end		= CONSISTENT_END,
+};
+
+static struct ppc_vm_region *
+ppc_vm_region_alloc(struct ppc_vm_region *head, size_t size, gfp_t gfp)
+{
+	unsigned long addr = head->vm_start, end = head->vm_end - size;
+	unsigned long flags;
+	struct ppc_vm_region *c, *new;
+
+	new = kmalloc(sizeof(struct ppc_vm_region), gfp);
+	if (!new)
+		goto out;
+
+	spin_lock_irqsave(&consistent_lock, flags);
+
+	list_for_each_entry(c, &head->vm_list, vm_list) {
+		if ((addr + size) < addr)
+			goto nospc;
+		if ((addr + size) <= c->vm_start)
+			goto found;
+		addr = c->vm_end;
+		if (addr > end)
+			goto nospc;
+	}
+
+ found:
+	/*
+	 * Insert this entry _before_ the one we found.
+	 */
+	list_add_tail(&new->vm_list, &c->vm_list);
+	new->vm_start = addr;
+	new->vm_end = addr + size;
+
+	spin_unlock_irqrestore(&consistent_lock, flags);
+	return new;
+
+ nospc:
+	spin_unlock_irqrestore(&consistent_lock, flags);
+	kfree(new);
+ out:
+	return NULL;
+}
+
+static struct ppc_vm_region *ppc_vm_region_find(struct ppc_vm_region *head, unsigned long addr)
+{
+	struct ppc_vm_region *c;
+
+	list_for_each_entry(c, &head->vm_list, vm_list) {
+		if (c->vm_start == addr)
+			goto out;
+	}
+	c = NULL;
+ out:
+	return c;
+}
+
+/*
+ * Allocate DMA-coherent memory space and return both the kernel remapped
+ * virtual and bus address for that space.
+ */
+void *
+__dma_alloc_coherent(struct device *dev, size_t size, dma_addr_t *handle, gfp_t gfp)
+{
+	struct page *page;
+	struct ppc_vm_region *c;
+	unsigned long order;
+	u64 mask = ISA_DMA_THRESHOLD, limit;
+
+	if (dev) {
+		mask = dev->coherent_dma_mask;
+
+		/*
+		 * Sanity check the DMA mask - it must be non-zero, and
+		 * must be able to be satisfied by a DMA allocation.
+		 */
+		if (mask == 0) {
+			dev_warn(dev, "coherent DMA mask is unset\n");
+			goto no_page;
+		}
+
+		if ((~mask) & ISA_DMA_THRESHOLD) {
+			dev_warn(dev, "coherent DMA mask %#llx is smaller "
+				 "than system GFP_DMA mask %#llx\n",
+				 mask, (unsigned long long)ISA_DMA_THRESHOLD);
+			goto no_page;
+		}
+	}
+
+
+	size = PAGE_ALIGN(size);
+	limit = (mask + 1) & ~mask;
+	if ((limit && size >= limit) ||
+	    size >= (CONSISTENT_END - CONSISTENT_BASE)) {
+		printk(KERN_WARNING "coherent allocation too big (requested %#x mask %#Lx)\n",
+		       size, mask);
+		return NULL;
+	}
+
+	order = get_order(size);
+
+	/* Might be useful if we ever have a real legacy DMA zone... */
+	if (mask != 0xffffffff)
+		gfp |= GFP_DMA;
+
+	page = alloc_pages(gfp, order);
+	if (!page)
+		goto no_page;
+
+	/*
+	 * Invalidate any data that might be lurking in the
+	 * kernel direct-mapped region for device DMA.
+	 */
+	{
+		unsigned long kaddr = (unsigned long)page_address(page);
+		memset(page_address(page), 0, size);
+		flush_dcache_range(kaddr, kaddr + size);
+	}
+
+	/*
+	 * Allocate a virtual address in the consistent mapping region.
+	 */
+	c = ppc_vm_region_alloc(&consistent_head, size,
+			    gfp & ~(__GFP_DMA | __GFP_HIGHMEM));
+	if (c) {
+		unsigned long vaddr = c->vm_start;
+		struct page *end = page + (1 << order);
+
+		split_page(page, order);
+
+		/*
+		 * Set the "dma handle"
+		 */
+		*handle = page_to_phys(page);
+
+		do {
+			SetPageReserved(page);
+			map_page(vaddr, page_to_phys(page),
+				 pgprot_noncached(PAGE_KERNEL));
+			page++;
+			vaddr += PAGE_SIZE;
+		} while (size -= PAGE_SIZE);
+
+		/*
+		 * Free the otherwise unused pages.
+		 */
+		while (page < end) {
+			__free_page(page);
+			page++;
+		}
+
+		return (void *)c->vm_start;
+	}
+
+	if (page)
+		__free_pages(page, order);
+ no_page:
+	return NULL;
+}
+EXPORT_SYMBOL(__dma_alloc_coherent);
+
+/*
+ * free a page as defined by the above mapping.
+ */
+void __dma_free_coherent(size_t size, void *vaddr)
+{
+	struct ppc_vm_region *c;
+	unsigned long flags, addr;
+	
+	size = PAGE_ALIGN(size);
+
+	spin_lock_irqsave(&consistent_lock, flags);
+
+	c = ppc_vm_region_find(&consistent_head, (unsigned long)vaddr);
+	if (!c)
+		goto no_area;
+
+	if ((c->vm_end - c->vm_start) != size) {
+		printk(KERN_ERR "%s: freeing wrong coherent size (%ld != %d)\n",
+		       __func__, c->vm_end - c->vm_start, size);
+		dump_stack();
+		size = c->vm_end - c->vm_start;
+	}
+
+	addr = c->vm_start;
+	do {
+		pte_t *ptep;
+		unsigned long pfn;
+
+		ptep = pte_offset_kernel(pmd_offset(pud_offset(pgd_offset_k(addr),
+							       addr),
+						    addr),
+					 addr);
+		if (!pte_none(*ptep) && pte_present(*ptep)) {
+			pfn = pte_pfn(*ptep);
+			pte_clear(&init_mm, addr, ptep);
+			if (pfn_valid(pfn)) {
+				struct page *page = pfn_to_page(pfn);
+
+				ClearPageReserved(page);
+				__free_page(page);
+			}
+		}
+		addr += PAGE_SIZE;
+	} while (size -= PAGE_SIZE);
+
+	flush_tlb_kernel_range(c->vm_start, c->vm_end);
+
+	list_del(&c->vm_list);
+
+	spin_unlock_irqrestore(&consistent_lock, flags);
+
+	kfree(c);
+	return;
+
+ no_area:
+	spin_unlock_irqrestore(&consistent_lock, flags);
+	printk(KERN_ERR "%s: trying to free invalid coherent area: %p\n",
+	       __func__, vaddr);
+	dump_stack();
+}
+EXPORT_SYMBOL(__dma_free_coherent);
+
+/*
+ * make an area consistent.
+ */
+void __dma_sync(void *vaddr, size_t size, int direction)
+{
+	unsigned long start = (unsigned long)vaddr;
+	unsigned long end   = start + size;
+
+	switch (direction) {
+	case DMA_NONE:
+		BUG();
+	case DMA_FROM_DEVICE:
+		/*
+		 * invalidate only when cache-line aligned otherwise there is
+		 * the potential for discarding uncommitted data from the cache
+		 */
+		if ((start & (L1_CACHE_BYTES - 1)) || (size & (L1_CACHE_BYTES - 1)))
+			flush_dcache_range(start, end);
+		else
+			invalidate_dcache_range(start, end);
+		break;
+	case DMA_TO_DEVICE:		/* writeback only */
+		clean_dcache_range(start, end);
+		break;
+	case DMA_BIDIRECTIONAL:	/* writeback and invalidate */
+		flush_dcache_range(start, end);
+		break;
+	}
+}
+EXPORT_SYMBOL(__dma_sync);
+
+#ifdef CONFIG_HIGHMEM
+/*
+ * __dma_sync_page() implementation for systems using highmem.
+ * In this case, each page of a buffer must be kmapped/kunmapped
+ * in order to have a virtual address for __dma_sync(). This must
+ * not sleep so kmap_atomic()/kunmap_atomic() are used.
+ *
+ * Note: yes, it is possible and correct to have a buffer extend
+ * beyond the first page.
+ */
+static inline void __dma_sync_page_highmem(struct page *page,
+		unsigned long offset, size_t size, int direction)
+{
+	size_t seg_size = min((size_t)(PAGE_SIZE - offset), size);
+	size_t cur_size = seg_size;
+	unsigned long flags, start, seg_offset = offset;
+	int nr_segs = 1 + ((size - seg_size) + PAGE_SIZE - 1)/PAGE_SIZE;
+	int seg_nr = 0;
+
+	local_irq_save(flags);
+
+	do {
+		start = (unsigned long)kmap_atomic(page + seg_nr,
+				KM_PPC_SYNC_PAGE) + seg_offset;
+
+		/* Sync this buffer segment */
+		__dma_sync((void *)start, seg_size, direction);
+		kunmap_atomic((void *)start, KM_PPC_SYNC_PAGE);
+		seg_nr++;
+
+		/* Calculate next buffer segment size */
+		seg_size = min((size_t)PAGE_SIZE, size - cur_size);
+
+		/* Add the segment size to our running total */
+		cur_size += seg_size;
+		seg_offset = 0;
+	} while (seg_nr < nr_segs);
+
+	local_irq_restore(flags);
+}
+#endif /* CONFIG_HIGHMEM */
+
+/*
+ * __dma_sync_page makes memory consistent. identical to __dma_sync, but
+ * takes a struct page instead of a virtual address
+ */
+void __dma_sync_page(struct page *page, unsigned long offset,
+	size_t size, int direction)
+{
+#ifdef CONFIG_HIGHMEM
+	__dma_sync_page_highmem(page, offset, size, direction);
+#else
+	unsigned long start = (unsigned long)page_address(page) + offset;
+	__dma_sync((void *)start, size, direction);
+#endif
+}
+EXPORT_SYMBOL(__dma_sync_page);

arch/powerpc/mm/init_32.c

@@ -168,12 +168,8 @@ void __init MMU_init(void)
 		ppc_md.progress("MMU:mapin", 0x301);
 	mapin_ram();
 
-#ifdef CONFIG_HIGHMEM
-	ioremap_base = PKMAP_BASE;
-#else
-	ioremap_base = 0xfe000000UL;	/* for now, could be 0xfffff000 */
-#endif /* CONFIG_HIGHMEM */
-	ioremap_bot = ioremap_base;
+	/* Initialize early top-down ioremap allocator */
+	ioremap_bot = IOREMAP_TOP;
 
 	/* Map in I/O resources */
 	if (ppc_md.progress)

arch/powerpc/mm/mem.c

@@ -380,6 +380,23 @@ void __init mem_init(void)
 		bsssize >> 10,
 		initsize >> 10);
 
+#ifdef CONFIG_PPC32
+	pr_info("Kernel virtual memory layout:\n");
+	pr_info("  * 0x%08lx..0x%08lx  : fixmap\n", FIXADDR_START, FIXADDR_TOP);
+#ifdef CONFIG_HIGHMEM
+	pr_info("  * 0x%08lx..0x%08lx  : highmem PTEs\n",
+		PKMAP_BASE, PKMAP_ADDR(LAST_PKMAP));
+#endif /* CONFIG_HIGHMEM */
+#ifdef CONFIG_NOT_COHERENT_CACHE
+	pr_info("  * 0x%08lx..0x%08lx  : consistent mem\n",
+		IOREMAP_TOP, IOREMAP_TOP + CONFIG_CONSISTENT_SIZE);
+#endif /* CONFIG_NOT_COHERENT_CACHE */
+	pr_info("  * 0x%08lx..0x%08lx  : early ioremap\n",
+		ioremap_bot, IOREMAP_TOP);
+	pr_info("  * 0x%08lx..0x%08lx  : vmalloc & ioremap\n",
+		VMALLOC_START, VMALLOC_END);
+#endif /* CONFIG_PPC32 */
+
 	mem_init_done = 1;
 }
 

arch/powerpc/mm/mmu_context_nohash.c

@@ -127,12 +127,12 @@ static unsigned int steal_context_up(unsigned int id)
 
 	pr_debug("[%d] steal context %d from mm @%p\n", cpu, id, mm);
 
-	/* Mark this mm has having no context anymore */
-	mm->context.id = MMU_NO_CONTEXT;
-
 	/* Flush the TLB for that context */
 	local_flush_tlb_mm(mm);
 
+	/* Mark this mm has having no context anymore */
+	mm->context.id = MMU_NO_CONTEXT;
+
 	/* XXX This clear should ultimately be part of local_flush_tlb_mm */
 	__clear_bit(id, stale_map[cpu]);
 

arch/powerpc/mm/pgtable_32.c

@@ -399,8 +399,6 @@ void kernel_map_pages(struct page *page, int numpages, int enable)
 #endif /* CONFIG_DEBUG_PAGEALLOC */
 
 static int fixmaps;
-unsigned long FIXADDR_TOP = (-PAGE_SIZE);
-EXPORT_SYMBOL(FIXADDR_TOP);
 
 void __set_fixmap (enum fixed_addresses idx, phys_addr_t phys, pgprot_t flags)
 {

arch/x86/boot/compressed/relocs.c

@@ -504,8 +504,11 @@ static void walk_relocs(void (*visit)(Elf32_Rel *rel, Elf32_Sym *sym))
 			if (sym->st_shndx == SHN_ABS) {
 				continue;
 			}
-			if (r_type == R_386_PC32) {
-				/* PC relative relocations don't need to be adjusted */
+			if (r_type == R_386_NONE || r_type == R_386_PC32) {
+				/*
+				 * NONE can be ignored and and PC relative
+				 * relocations don't need to be adjusted.
+				 */
 			}
 			else if (r_type == R_386_32) {
 				/* Visit relocations that need to be adjusted */

arch/x86/boot/memory.c

@@ -17,11 +17,6 @@
 
 #define SMAP	0x534d4150	/* ASCII "SMAP" */
 
-struct e820_ext_entry {
-	struct e820entry std;
-	u32 ext_flags;
-} __attribute__((packed));
-
 static int detect_memory_e820(void)
 {
 	int count = 0;
@@ -29,13 +24,21 @@ static int detect_memory_e820(void)
 	u32 size, id, edi;
 	u8 err;
 	struct e820entry *desc = boot_params.e820_map;
-	static struct e820_ext_entry buf; /* static so it is zeroed */
+	static struct e820entry buf; /* static so it is zeroed */
 
 	/*
-	 * Set this here so that if the BIOS doesn't change this field
-	 * but still doesn't change %ecx, we're still okay...
+	 * Note: at least one BIOS is known which assumes that the
+	 * buffer pointed to by one e820 call is the same one as
+	 * the previous call, and only changes modified fields.  Therefore,
+	 * we use a temporary buffer and copy the results entry by entry.
+	 *
+	 * This routine deliberately does not try to account for
+	 * ACPI 3+ extended attributes.  This is because there are
+	 * BIOSes in the field which report zero for the valid bit for
+	 * all ranges, and we don't currently make any use of the
+	 * other attribute bits.  Revisit this if we see the extended
+	 * attribute bits deployed in a meaningful way in the future.
 	 */
-	buf.ext_flags = 1;
 
 	do {
 		size = sizeof buf;
@@ -66,13 +69,7 @@ static int detect_memory_e820(void)
 			break;
 		}
 
-		/* ACPI 3.0 added the extended flags support.  If bit 0
-		   in the extended flags is zero, we're supposed to simply
-		   ignore the entry -- a backwards incompatible change! */
-		if (size > 20 && !(buf.ext_flags & 1))
-			continue;
-
-		*desc++ = buf.std;
+		*desc++ = buf;
 		count++;
 	} while (next && count < ARRAY_SIZE(boot_params.e820_map));
 

arch/x86/kernel/cpu/common.c

@@ -114,6 +114,13 @@ DEFINE_PER_CPU_PAGE_ALIGNED(struct gdt_page, gdt_page) = { .gdt = {
 } };
 EXPORT_PER_CPU_SYMBOL_GPL(gdt_page);
 
+static int __init x86_xsave_setup(char *s)
+{
+	setup_clear_cpu_cap(X86_FEATURE_XSAVE);
+	return 1;
+}
+__setup("noxsave", x86_xsave_setup);
+
 #ifdef CONFIG_X86_32
 static int cachesize_override __cpuinitdata = -1;
 static int disable_x86_serial_nr __cpuinitdata = 1;

arch/x86/kernel/cpu/cpufreq/p4-clockmod.c

@@ -168,6 +168,7 @@ static unsigned int cpufreq_p4_get_frequency(struct cpuinfo_x86 *c)
 		case 0x0E: /* Core */
 		case 0x0F: /* Core Duo */
 		case 0x16: /* Celeron Core */
+		case 0x1C: /* Atom */
 			p4clockmod_driver.flags |= CPUFREQ_CONST_LOOPS;
 			return speedstep_get_frequency(SPEEDSTEP_CPU_PCORE);
 		case 0x0D: /* Pentium M (Dothan) */

arch/x86/kernel/cpu/cpufreq/powernow-k7.c

@@ -168,10 +168,12 @@ static int check_powernow(void)
 	return 1;
 }
 
+#ifdef CONFIG_X86_POWERNOW_K7_ACPI
 static void invalidate_entry(unsigned int entry)
 {
 	powernow_table[entry].frequency = CPUFREQ_ENTRY_INVALID;
 }
+#endif
 
 static int get_ranges(unsigned char *pst)
 {

arch/x86/kernel/cpu/cpufreq/powernow-k8.c

@@ -649,6 +649,20 @@ static void print_basics(struct powernow_k8_data *data)
 				data->batps);
 }
 
+static u32 freq_from_fid_did(u32 fid, u32 did)
+{
+	u32 mhz = 0;
+
+	if (boot_cpu_data.x86 == 0x10)
+		mhz = (100 * (fid + 0x10)) >> did;
+	else if (boot_cpu_data.x86 == 0x11)
+		mhz = (100 * (fid + 8)) >> did;
+	else
+		BUG();
+
+	return mhz * 1000;
+}
+
 static int fill_powernow_table(struct powernow_k8_data *data,
 		struct pst_s *pst, u8 maxvid)
 {
@@ -923,8 +937,13 @@ static int fill_powernow_table_pstate(struct powernow_k8_data *data,
 
 		powernow_table[i].index = index;
 
-		powernow_table[i].frequency =
-			data->acpi_data.states[i].core_frequency * 1000;
+		/* Frequency may be rounded for these */
+		if (boot_cpu_data.x86 == 0x10 || boot_cpu_data.x86 == 0x11) {
+			powernow_table[i].frequency =
+				freq_from_fid_did(lo & 0x3f, (lo >> 6) & 7);
+		} else
+			powernow_table[i].frequency =
+				data->acpi_data.states[i].core_frequency * 1000;
 	}
 	return 0;
 }
@@ -1215,13 +1234,16 @@ static int powernowk8_verify(struct cpufreq_policy *pol)
 	return cpufreq_frequency_table_verify(pol, data->powernow_table);
 }
 
+static const char ACPI_PSS_BIOS_BUG_MSG[] =
+	KERN_ERR FW_BUG PFX "No compatible ACPI _PSS objects found.\n"
+	KERN_ERR FW_BUG PFX "Try again with latest BIOS.\n";
+
 /* per CPU init entry point to the driver */
 static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 {
 	struct powernow_k8_data *data;
 	cpumask_t oldmask;
 	int rc;
-	static int print_once;
 
 	if (!cpu_online(pol->cpu))
 		return -ENODEV;
@@ -1244,19 +1266,7 @@ static int __cpuinit powernowk8_cpu_init(struct cpufreq_policy *pol)
 		 * an UP version, and is deprecated by AMD.
 		 */
 		if (num_online_cpus() != 1) {
-			/*
-			 * Replace this one with print_once as soon as such a
-			 * thing gets introduced
-			 */
-			if (!print_once) {
-				WARN_ONCE(1, KERN_ERR FW_BUG PFX "Your BIOS "
-					"does not provide ACPI _PSS objects "
-					"in a way that Linux understands. "
-					"Please report this to the Linux ACPI"
-					" maintainers and complain to your "
-					"BIOS vendor.\n");
-				print_once++;
-			}
+			printk_once(ACPI_PSS_BIOS_BUG_MSG);
 			goto err_out;
 		}
 		if (pol->cpu != 0) {

arch/x86/kernel/reboot.c

@@ -232,6 +232,14 @@ static struct dmi_system_id __initdata reboot_dmi_table[] = {
 			DMI_MATCH(DMI_PRODUCT_NAME, "Dell DXP061"),
 		},
 	},
+	{	/* Handle problems with rebooting on Sony VGN-Z540N */
+		.callback = set_bios_reboot,
+		.ident = "Sony VGN-Z540N",
+		.matches = {
+			DMI_MATCH(DMI_SYS_VENDOR, "Sony Corporation"),
+			DMI_MATCH(DMI_PRODUCT_NAME, "VGN-Z540N"),
+		},
+	},
 	{ }
 };
 

arch/x86/kernel/setup_percpu.c

@@ -160,8 +160,10 @@ static ssize_t __init setup_pcpu_remap(size_t static_size)
 	/*
 	 * If large page isn't supported, there's no benefit in doing
 	 * this.  Also, on non-NUMA, embedding is better.
+	 *
+	 * NOTE: disabled for now.
 	 */
-	if (!cpu_has_pse || !pcpu_need_numa())
+	if (true || !cpu_has_pse || !pcpu_need_numa())
 		return -EINVAL;
 
 	/*

arch/x86/kvm/mmu.c

@@ -2897,8 +2897,7 @@ static int kvm_pv_mmu_write(struct kvm_vcpu *vcpu,
 
 static int kvm_pv_mmu_flush_tlb(struct kvm_vcpu *vcpu)
 {
-	kvm_x86_ops->tlb_flush(vcpu);
-	set_bit(KVM_REQ_MMU_SYNC, &vcpu->requests);
+	kvm_set_cr3(vcpu, vcpu->arch.cr3);
 	return 1;
 }
 

arch/x86/kvm/x86.c

@@ -338,6 +338,9 @@ EXPORT_SYMBOL_GPL(kvm_lmsw);
 
 void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
 {
+	unsigned long old_cr4 = vcpu->arch.cr4;
+	unsigned long pdptr_bits = X86_CR4_PGE | X86_CR4_PSE | X86_CR4_PAE;
+
 	if (cr4 & CR4_RESERVED_BITS) {
 		printk(KERN_DEBUG "set_cr4: #GP, reserved bits\n");
 		kvm_inject_gp(vcpu, 0);
@@ -351,7 +354,8 @@ void kvm_set_cr4(struct kvm_vcpu *vcpu, unsigned long cr4)
 			kvm_inject_gp(vcpu, 0);
 			return;
 		}
-	} else if (is_paging(vcpu) && !is_pae(vcpu) && (cr4 & X86_CR4_PAE)
+	} else if (is_paging(vcpu) && (cr4 & X86_CR4_PAE)
+		   && ((cr4 ^ old_cr4) & pdptr_bits)
 		   && !load_pdptrs(vcpu, vcpu->arch.cr3)) {
 		printk(KERN_DEBUG "set_cr4: #GP, pdptrs reserved bits\n");
 		kvm_inject_gp(vcpu, 0);

arch/x86/mm/pageattr.c

@@ -153,7 +153,7 @@ static void __cpa_flush_all(void *arg)
 	 */
 	__flush_tlb_all();
 
-	if (cache && boot_cpu_data.x86_model >= 4)
+	if (cache && boot_cpu_data.x86 >= 4)
 		wbinvd();
 }
 
@@ -208,20 +208,15 @@ static void cpa_flush_array(unsigned long *start, int numpages, int cache,
 			    int in_flags, struct page **pages)
 {
 	unsigned int i, level;
+	unsigned long do_wbinvd = cache && numpages >= 1024; /* 4M threshold */
 
 	BUG_ON(irqs_disabled());
 
-	on_each_cpu(__cpa_flush_range, NULL, 1);
+	on_each_cpu(__cpa_flush_all, (void *) do_wbinvd, 1);
 
-	if (!cache)
+	if (!cache || do_wbinvd)
 		return;
 
-	/* 4M threshold */
-	if (numpages >= 1024) {
-		if (boot_cpu_data.x86_model >= 4)
-			wbinvd();
-		return;
-	}
 	/*
 	 * We only need to flush on one CPU,
 	 * clflush is a MESI-coherent instruction that

drivers/base/power/main.c

@@ -357,6 +357,7 @@ static void dpm_power_up(pm_message_t state)
 {
 	struct device *dev;
 
+	mutex_lock(&dpm_list_mtx);
 	list_for_each_entry(dev, &dpm_list, power.entry)
 		if (dev->power.status > DPM_OFF) {
 			int error;
@@ -366,6 +367,7 @@ static void dpm_power_up(pm_message_t state)
 			if (error)
 				pm_dev_err(dev, state, " early", error);
 		}
+	mutex_unlock(&dpm_list_mtx);
 }
 
 /**
@@ -614,6 +616,7 @@ int device_power_down(pm_message_t state)
 	int error = 0;
 
 	suspend_device_irqs();
+	mutex_lock(&dpm_list_mtx);
 	list_for_each_entry_reverse(dev, &dpm_list, power.entry) {
 		error = suspend_device_noirq(dev, state);
 		if (error) {
@@ -622,6 +625,7 @@ int device_power_down(pm_message_t state)
 		}
 		dev->power.status = DPM_OFF_IRQ;
 	}
+	mutex_unlock(&dpm_list_mtx);
 	if (error)
 		device_power_up(resume_event(state));
 	return error;

drivers/cpufreq/cpufreq.c

@@ -1070,11 +1070,11 @@ static int __cpufreq_remove_dev(struct sys_device *sys_dev)
 	spin_unlock_irqrestore(&cpufreq_driver_lock, flags);
 #endif
 
+	unlock_policy_rwsem_write(cpu);
+
 	if (cpufreq_driver->target)
 		__cpufreq_governor(data, CPUFREQ_GOV_STOP);
 
-	unlock_policy_rwsem_write(cpu);
-
 	kobject_put(&data->kobj);
 
 	/* we need to make sure that the underlying kobj is actually

drivers/cpufreq/cpufreq_conservative.c

@@ -91,6 +91,9 @@ static unsigned int dbs_enable;	/* number of CPUs using this policy */
  * (like __cpufreq_driver_target()) is being called with dbs_mutex taken, then
  * cpu_hotplug lock should be taken before that. Note that cpu_hotplug lock
  * is recursive for the same process. -Venki
+ * DEADLOCK ALERT! (2) : do_dbs_timer() must not take the dbs_mutex, because it
+ * would deadlock with cancel_delayed_work_sync(), which is needed for proper
+ * raceless workqueue teardown.
  */
 static DEFINE_MUTEX(dbs_mutex);
 
@@ -542,7 +545,7 @@ static inline void dbs_timer_init(struct cpu_dbs_info_s *dbs_info)
 static inline void dbs_timer_exit(struct cpu_dbs_info_s *dbs_info)
 {
 	dbs_info->enable = 0;
-	cancel_delayed_work(&dbs_info->work);
+	cancel_delayed_work_sync(&dbs_info->work);
 }
 
 static int cpufreq_governor_dbs(struct cpufreq_policy *policy,

drivers/cpufreq/cpufreq_ondemand.c

@@ -98,6 +98,9 @@ static unsigned int dbs_enable;	/* number of CPUs using this policy */
  * (like __cpufreq_driver_target()) is being called with dbs_mutex taken, then
  * cpu_hotplug lock should be taken before that. Note that cpu_hotplug lock
  * is recursive for the same process. -Venki
+ * DEADLOCK ALERT! (2) : do_dbs_timer() must not take the dbs_mutex, because it
+ * would deadlock with cancel_delayed_work_sync(), which is needed for proper
+ * raceless workqueue teardown.
  */
 static DEFINE_MUTEX(dbs_mutex);
 
@@ -562,7 +565,7 @@ static inline void dbs_timer_init(struct cpu_dbs_info_s *dbs_info)
 static inline void dbs_timer_exit(struct cpu_dbs_info_s *dbs_info)
 {
 	dbs_info->enable = 0;
-	cancel_delayed_work(&dbs_info->work);
+	cancel_delayed_work_sync(&dbs_info->work);
 }
 
 static int cpufreq_governor_dbs(struct cpufreq_policy *policy,

drivers/gpu/drm/i915/i915_drv.h

@@ -180,7 +180,8 @@ typedef struct drm_i915_private {
 	int backlight_duty_cycle;  /* restore backlight to this value */
 	bool panel_wants_dither;
 	struct drm_display_mode *panel_fixed_mode;
-	struct drm_display_mode *vbt_mode; /* if any */
+	struct drm_display_mode *lfp_lvds_vbt_mode; /* if any */
+	struct drm_display_mode *sdvo_lvds_vbt_mode; /* if any */
 
 	/* Feature bits from the VBIOS */
 	unsigned int int_tv_support:1;

drivers/gpu/drm/i915/i915_gem.c

@@ -349,7 +349,7 @@ i915_gem_shmem_pread_slow(struct drm_device *dev, struct drm_gem_object *obj,
 	last_data_page = (data_ptr + args->size - 1) / PAGE_SIZE;
 	num_pages = last_data_page - first_data_page + 1;
 
-	user_pages = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL);
+	user_pages = drm_calloc_large(num_pages, sizeof(struct page *));
 	if (user_pages == NULL)
 		return -ENOMEM;
 
@@ -429,7 +429,7 @@ i915_gem_shmem_pread_slow(struct drm_device *dev, struct drm_gem_object *obj,
 		SetPageDirty(user_pages[i]);
 		page_cache_release(user_pages[i]);
 	}
-	kfree(user_pages);
+	drm_free_large(user_pages);
 
 	return ret;
 }
@@ -649,7 +649,7 @@ i915_gem_gtt_pwrite_slow(struct drm_device *dev, struct drm_gem_object *obj,
 	last_data_page = (data_ptr + args->size - 1) / PAGE_SIZE;
 	num_pages = last_data_page - first_data_page + 1;
 
-	user_pages = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL);
+	user_pages = drm_calloc_large(num_pages, sizeof(struct page *));
 	if (user_pages == NULL)
 		return -ENOMEM;
 
@@ -719,7 +719,7 @@ i915_gem_gtt_pwrite_slow(struct drm_device *dev, struct drm_gem_object *obj,
 out_unpin_pages:
 	for (i = 0; i < pinned_pages; i++)
 		page_cache_release(user_pages[i]);
-	kfree(user_pages);
+	drm_free_large(user_pages);
 
 	return ret;
 }
@@ -824,7 +824,7 @@ i915_gem_shmem_pwrite_slow(struct drm_device *dev, struct drm_gem_object *obj,
 	last_data_page = (data_ptr + args->size - 1) / PAGE_SIZE;
 	num_pages = last_data_page - first_data_page + 1;
 
-	user_pages = kcalloc(num_pages, sizeof(struct page *), GFP_KERNEL);
+	user_pages = drm_calloc_large(num_pages, sizeof(struct page *));
 	if (user_pages == NULL)
 		return -ENOMEM;
 
@@ -902,7 +902,7 @@ i915_gem_shmem_pwrite_slow(struct drm_device *dev, struct drm_gem_object *obj,
 fail_put_user_pages:
 	for (i = 0; i < pinned_pages; i++)
 		page_cache_release(user_pages[i]);
-	kfree(user_pages);
+	drm_free_large(user_pages);
 
 	return ret;
 }
@@ -1145,7 +1145,7 @@ int i915_gem_fault(struct vm_area_struct *vma, struct vm_fault *vmf)
 			mutex_unlock(&dev->struct_mutex);
 			return VM_FAULT_SIGBUS;
 		}
-		list_add(&obj_priv->list, &dev_priv->mm.inactive_list);
+		list_add_tail(&obj_priv->list, &dev_priv->mm.inactive_list);
 	}
 
 	/* Need a new fence register? */
@@ -1375,7 +1375,7 @@ i915_gem_mmap_gtt_ioctl(struct drm_device *dev, void *data,
 			mutex_unlock(&dev->struct_mutex);
 			return ret;
 		}
-		list_add(&obj_priv->list, &dev_priv->mm.inactive_list);
+		list_add_tail(&obj_priv->list, &dev_priv->mm.inactive_list);
 	}
 
 	drm_gem_object_unreference(obj);
@@ -1408,9 +1408,7 @@ i915_gem_object_put_pages(struct drm_gem_object *obj)
 		}
 	obj_priv->dirty = 0;
 
-	drm_free(obj_priv->pages,
-		 page_count * sizeof(struct page *),
-		 DRM_MEM_DRIVER);
+	drm_free_large(obj_priv->pages);
 	obj_priv->pages = NULL;
 }
 
@@ -2024,8 +2022,7 @@ i915_gem_object_get_pages(struct drm_gem_object *obj)
 	 */
 	page_count = obj->size / PAGE_SIZE;
 	BUG_ON(obj_priv->pages != NULL);
-	obj_priv->pages = drm_calloc(page_count, sizeof(struct page *),
-				     DRM_MEM_DRIVER);
+	obj_priv->pages = drm_calloc_large(page_count, sizeof(struct page *));
 	if (obj_priv->pages == NULL) {
 		DRM_ERROR("Faled to allocate page list\n");
 		obj_priv->pages_refcount--;
@@ -3111,7 +3108,7 @@ i915_gem_get_relocs_from_user(struct drm_i915_gem_exec_object *exec_list,
 		reloc_count += exec_list[i].relocation_count;
 	}
 
-	*relocs = drm_calloc(reloc_count, sizeof(**relocs), DRM_MEM_DRIVER);
+	*relocs = drm_calloc_large(reloc_count, sizeof(**relocs));
 	if (*relocs == NULL)
 		return -ENOMEM;
 
@@ -3125,8 +3122,7 @@ i915_gem_get_relocs_from_user(struct drm_i915_gem_exec_object *exec_list,
 				     exec_list[i].relocation_count *
 				     sizeof(**relocs));
 		if (ret != 0) {
-			drm_free(*relocs, reloc_count * sizeof(**relocs),
-				 DRM_MEM_DRIVER);
+			drm_free_large(*relocs);
 			*relocs = NULL;
 			return -EFAULT;
 		}
@@ -3165,7 +3161,7 @@ i915_gem_put_relocs_to_user(struct drm_i915_gem_exec_object *exec_list,
 	}
 
 err:
-	drm_free(relocs, reloc_count * sizeof(*relocs), DRM_MEM_DRIVER);
+	drm_free_large(relocs);
 
 	return ret;
 }
@@ -3198,10 +3194,8 @@ i915_gem_execbuffer(struct drm_device *dev, void *data,
 		return -EINVAL;
 	}
 	/* Copy in the exec list from userland */
-	exec_list = drm_calloc(sizeof(*exec_list), args->buffer_count,
-			       DRM_MEM_DRIVER);
-	object_list = drm_calloc(sizeof(*object_list), args->buffer_count,
-				 DRM_MEM_DRIVER);
+	exec_list = drm_calloc_large(sizeof(*exec_list), args->buffer_count);
+	object_list = drm_calloc_large(sizeof(*object_list), args->buffer_count);
 	if (exec_list == NULL || object_list == NULL) {
 		DRM_ERROR("Failed to allocate exec or object list "
 			  "for %d buffers\n",
@@ -3462,10 +3456,8 @@ i915_gem_execbuffer(struct drm_device *dev, void *data,
 	}
 
 pre_mutex_err:
-	drm_free(object_list, sizeof(*object_list) * args->buffer_count,
-		 DRM_MEM_DRIVER);
-	drm_free(exec_list, sizeof(*exec_list) * args->buffer_count,
-		 DRM_MEM_DRIVER);
+	drm_free_large(object_list);
+	drm_free_large(exec_list);
 	drm_free(cliprects, sizeof(*cliprects) * args->num_cliprects,
 		 DRM_MEM_DRIVER);
 

drivers/gpu/drm/i915/i915_reg.h

@@ -1410,9 +1410,25 @@
 
 /* Cursor A & B regs */
 #define CURACNTR		0x70080
+/* Old style CUR*CNTR flags (desktop 8xx) */
+#define   CURSOR_ENABLE		0x80000000
+#define   CURSOR_GAMMA_ENABLE	0x40000000
+#define   CURSOR_STRIDE_MASK	0x30000000
+#define   CURSOR_FORMAT_SHIFT	24
+#define   CURSOR_FORMAT_MASK	(0x07 << CURSOR_FORMAT_SHIFT)
+#define   CURSOR_FORMAT_2C	(0x00 << CURSOR_FORMAT_SHIFT)
+#define   CURSOR_FORMAT_3C	(0x01 << CURSOR_FORMAT_SHIFT)
+#define   CURSOR_FORMAT_4C	(0x02 << CURSOR_FORMAT_SHIFT)
+#define   CURSOR_FORMAT_ARGB	(0x04 << CURSOR_FORMAT_SHIFT)
+#define   CURSOR_FORMAT_XRGB	(0x05 << CURSOR_FORMAT_SHIFT)
+/* New style CUR*CNTR flags */
+#define   CURSOR_MODE		0x27
 #define   CURSOR_MODE_DISABLE   0x00
 #define   CURSOR_MODE_64_32B_AX 0x07
 #define   CURSOR_MODE_64_ARGB_AX ((1 << 5) | CURSOR_MODE_64_32B_AX)
+#define   MCURSOR_PIPE_SELECT	(1 << 28)
+#define   MCURSOR_PIPE_A	0x00
+#define   MCURSOR_PIPE_B	(1 << 28)
 #define   MCURSOR_GAMMA_ENABLE  (1 << 26)
 #define CURABASE		0x70084
 #define CURAPOS			0x70088
@@ -1420,6 +1436,7 @@
 #define   CURSOR_POS_SIGN       0x8000
 #define   CURSOR_X_SHIFT        0
 #define   CURSOR_Y_SHIFT        16
+#define CURSIZE			0x700a0
 #define CURBCNTR		0x700c0
 #define CURBBASE		0x700c4
 #define CURBPOS			0x700c8

drivers/gpu/drm/i915/intel_bios.c

@@ -57,9 +57,43 @@ find_section(struct bdb_header *bdb, int section_id)
 	return NULL;
 }
 
-/* Try to find panel data */
 static void
-parse_panel_data(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
+fill_detail_timing_data(struct drm_display_mode *panel_fixed_mode,
+			struct lvds_dvo_timing *dvo_timing)
+{
+	panel_fixed_mode->hdisplay = (dvo_timing->hactive_hi << 8) |
+		dvo_timing->hactive_lo;
+	panel_fixed_mode->hsync_start = panel_fixed_mode->hdisplay +
+		((dvo_timing->hsync_off_hi << 8) | dvo_timing->hsync_off_lo);
+	panel_fixed_mode->hsync_end = panel_fixed_mode->hsync_start +
+		dvo_timing->hsync_pulse_width;
+	panel_fixed_mode->htotal = panel_fixed_mode->hdisplay +
+		((dvo_timing->hblank_hi << 8) | dvo_timing->hblank_lo);
+
+	panel_fixed_mode->vdisplay = (dvo_timing->vactive_hi << 8) |
+		dvo_timing->vactive_lo;
+	panel_fixed_mode->vsync_start = panel_fixed_mode->vdisplay +
+		dvo_timing->vsync_off;
+	panel_fixed_mode->vsync_end = panel_fixed_mode->vsync_start +
+		dvo_timing->vsync_pulse_width;
+	panel_fixed_mode->vtotal = panel_fixed_mode->vdisplay +
+		((dvo_timing->vblank_hi << 8) | dvo_timing->vblank_lo);
+	panel_fixed_mode->clock = dvo_timing->clock * 10;
+	panel_fixed_mode->type = DRM_MODE_TYPE_PREFERRED;
+
+	/* Some VBTs have bogus h/vtotal values */
+	if (panel_fixed_mode->hsync_end > panel_fixed_mode->htotal)
+		panel_fixed_mode->htotal = panel_fixed_mode->hsync_end + 1;
+	if (panel_fixed_mode->vsync_end > panel_fixed_mode->vtotal)
+		panel_fixed_mode->vtotal = panel_fixed_mode->vsync_end + 1;
+
+	drm_mode_set_name(panel_fixed_mode);
+}
+
+/* Try to find integrated panel data */
+static void
+parse_lfp_panel_data(struct drm_i915_private *dev_priv,
+			    struct bdb_header *bdb)
 {
 	struct bdb_lvds_options *lvds_options;
 	struct bdb_lvds_lfp_data *lvds_lfp_data;
@@ -91,38 +125,45 @@ parse_panel_data(struct drm_i915_private *dev_priv, struct bdb_header *bdb)
 	panel_fixed_mode = drm_calloc(1, sizeof(*panel_fixed_mode),
 				      DRM_MEM_DRIVER);
 
-	panel_fixed_mode->hdisplay = (dvo_timing->hactive_hi << 8) |
-		dvo_timing->hactive_lo;
-	panel_fixed_mode->hsync_start = panel_fixed_mode->hdisplay +
-		((dvo_timing->hsync_off_hi << 8) | dvo_timing->hsync_off_lo);
-	panel_fixed_mode->hsync_end = panel_fixed_mode->hsync_start +
-		dvo_timing->hsync_pulse_width;
-	panel_fixed_mode->htotal = panel_fixed_mode->hdisplay +
-		((dvo_timing->hblank_hi << 8) | dvo_timing->hblank_lo);
+	fill_detail_timing_data(panel_fixed_mode, dvo_timing);
 
-	panel_fixed_mode->vdisplay = (dvo_timing->vactive_hi << 8) |
-		dvo_timing->vactive_lo;
-	panel_fixed_mode->vsync_start = panel_fixed_mode->vdisplay +
-		dvo_timing->vsync_off;
-	panel_fixed_mode->vsync_end = panel_fixed_mode->vsync_start +
-		dvo_timing->vsync_pulse_width;
-	panel_fixed_mode->vtotal = panel_fixed_mode->vdisplay +
-		((dvo_timing->vblank_hi << 8) | dvo_timing->vblank_lo);
-	panel_fixed_mode->clock = dvo_timing->clock * 10;
-	panel_fixed_mode->type = DRM_MODE_TYPE_PREFERRED;
+	dev_priv->lfp_lvds_vbt_mode = panel_fixed_mode;
 
-	/* Some VBTs have bogus h/vtotal values */
-	if (panel_fixed_mode->hsync_end > panel_fixed_mode->htotal)
-		panel_fixed_mode->htotal = panel_fixed_mode->hsync_end + 1;
-	if (panel_fixed_mode->vsync_end > panel_fixed_mode->vtotal)
-		panel_fixed_mode->vtotal = panel_fixed_mode->vsync_end + 1;
+	DRM_DEBUG("Found panel mode in BIOS VBT tables:\n");
+	drm_mode_debug_printmodeline(panel_fixed_mode);
 
-	drm_mode_set_name(panel_fixed_mode);
+	return;
+}
+
+/* Try to find sdvo panel data */
+static void
+parse_sdvo_panel_data(struct drm_i915_private *dev_priv,
+		      struct bdb_header *bdb)
+{
+	struct bdb_sdvo_lvds_options *sdvo_lvds_options;
+	struct lvds_dvo_timing *dvo_timing;
+	struct drm_display_mode *panel_fixed_mode;
 
-	dev_priv->vbt_mode = panel_fixed_mode;
+	dev_priv->sdvo_lvds_vbt_mode = NULL;
 
-	DRM_DEBUG("Found panel mode in BIOS VBT tables:\n");
-	drm_mode_debug_printmodeline(panel_fixed_mode);
+	sdvo_lvds_options = find_section(bdb, BDB_SDVO_LVDS_OPTIONS);
+	if (!sdvo_lvds_options)
+		return;
+
+	dvo_timing = find_section(bdb, BDB_SDVO_PANEL_DTDS);
+	if (!dvo_timing)
+		return;
+
+	panel_fixed_mode = drm_calloc(1, sizeof(*panel_fixed_mode),
+				      DRM_MEM_DRIVER);
+
+	if (!panel_fixed_mode)
+		return;
+
+	fill_detail_timing_data(panel_fixed_mode,
+			dvo_timing + sdvo_lvds_options->panel_type);
+
+	dev_priv->sdvo_lvds_vbt_mode = panel_fixed_mode;
 
 	return;
 }
@@ -199,7 +240,8 @@ intel_init_bios(struct drm_device *dev)
 
 	/* Grab useful general definitions */
 	parse_general_features(dev_priv, bdb);
-	parse_panel_data(dev_priv, bdb);
+	parse_lfp_panel_data(dev_priv, bdb);
+	parse_sdvo_panel_data(dev_priv, bdb);
 
 	pci_unmap_rom(pdev, bios);
 

drivers/gpu/drm/i915/intel_bios.h

@@ -279,6 +279,23 @@ struct vch_bdb_22 {
 	struct vch_panel_data panels[16];
 } __attribute__((packed));
 
+struct bdb_sdvo_lvds_options {
+	u8 panel_backlight;
+	u8 h40_set_panel_type;
+	u8 panel_type;
+	u8 ssc_clk_freq;
+	u16 als_low_trip;
+	u16 als_high_trip;
+	u8 sclalarcoeff_tab_row_num;
+	u8 sclalarcoeff_tab_row_size;
+	u8 coefficient[8];
+	u8 panel_misc_bits_1;
+	u8 panel_misc_bits_2;
+	u8 panel_misc_bits_3;
+	u8 panel_misc_bits_4;
+} __attribute__((packed));
+
+
 bool intel_init_bios(struct drm_device *dev);
 
 /*

drivers/gpu/drm/i915/intel_crt.c

@@ -198,9 +198,142 @@ static bool intel_crt_detect_ddc(struct drm_connector *connector)
 	return intel_ddc_probe(intel_output);
 }
 
+static enum drm_connector_status
+intel_crt_load_detect(struct drm_crtc *crtc, struct intel_output *intel_output)
+{
+	struct drm_encoder *encoder = &intel_output->enc;
+	struct drm_device *dev = encoder->dev;
+	struct drm_i915_private *dev_priv = dev->dev_private;
+	struct intel_crtc *intel_crtc = to_intel_crtc(crtc);
+	uint32_t pipe = intel_crtc->pipe;
+	uint32_t save_bclrpat;
+	uint32_t save_vtotal;
+	uint32_t vtotal, vactive;
+	uint32_t vsample;
+	uint32_t vblank, vblank_start, vblank_end;
+	uint32_t dsl;
+	uint32_t bclrpat_reg;
+	uint32_t vtotal_reg;
+	uint32_t vblank_reg;
+	uint32_t vsync_reg;
+	uint32_t pipeconf_reg;
+	uint32_t pipe_dsl_reg;
+	uint8_t	st00;
+	enum drm_connector_status status;
+
+	if (pipe == 0) {
+		bclrpat_reg = BCLRPAT_A;
+		vtotal_reg = VTOTAL_A;
+		vblank_reg = VBLANK_A;
+		vsync_reg = VSYNC_A;
+		pipeconf_reg = PIPEACONF;
+		pipe_dsl_reg = PIPEADSL;
+	} else {
+		bclrpat_reg = BCLRPAT_B;
+		vtotal_reg = VTOTAL_B;
+		vblank_reg = VBLANK_B;
+		vsync_reg = VSYNC_B;
+		pipeconf_reg = PIPEBCONF;
+		pipe_dsl_reg = PIPEBDSL;
+	}
+
+	save_bclrpat = I915_READ(bclrpat_reg);
+	save_vtotal = I915_READ(vtotal_reg);
+	vblank = I915_READ(vblank_reg);
+
+	vtotal = ((save_vtotal >> 16) & 0xfff) + 1;
+	vactive = (save_vtotal & 0x7ff) + 1;
+
+	vblank_start = (vblank & 0xfff) + 1;
+	vblank_end = ((vblank >> 16) & 0xfff) + 1;
+
+	/* Set the border color to purple. */
+	I915_WRITE(bclrpat_reg, 0x500050);
+
+	if (IS_I9XX(dev)) {
+		uint32_t pipeconf = I915_READ(pipeconf_reg);
+		I915_WRITE(pipeconf_reg, pipeconf | PIPECONF_FORCE_BORDER);
+		/* Wait for next Vblank to substitue
+		 * border color for Color info */
+		intel_wait_for_vblank(dev);
+		st00 = I915_READ8(VGA_MSR_WRITE);
+		status = ((st00 & (1 << 4)) != 0) ?
+			connector_status_connected :
+			connector_status_disconnected;
+
+		I915_WRITE(pipeconf_reg, pipeconf);
+	} else {
+		bool restore_vblank = false;
+		int count, detect;
+
+		/*
+		* If there isn't any border, add some.
+		* Yes, this will flicker
+		*/
+		if (vblank_start <= vactive && vblank_end >= vtotal) {
+			uint32_t vsync = I915_READ(vsync_reg);
+			uint32_t vsync_start = (vsync & 0xffff) + 1;
+
+			vblank_start = vsync_start;
+			I915_WRITE(vblank_reg,
+				   (vblank_start - 1) |
+				   ((vblank_end - 1) << 16));
+			restore_vblank = true;
+		}
+		/* sample in the vertical border, selecting the larger one */
+		if (vblank_start - vactive >= vtotal - vblank_end)
+			vsample = (vblank_start + vactive) >> 1;
+		else
+			vsample = (vtotal + vblank_end) >> 1;
+
+		/*
+		 * Wait for the border to be displayed
+		 */
+		while (I915_READ(pipe_dsl_reg) >= vactive)
+			;
+		while ((dsl = I915_READ(pipe_dsl_reg)) <= vsample)
+			;
+		/*
+		 * Watch ST00 for an entire scanline
+		 */
+		detect = 0;
+		count = 0;
+		do {
+			count++;
+			/* Read the ST00 VGA status register */
+			st00 = I915_READ8(VGA_MSR_WRITE);
+			if (st00 & (1 << 4))
+				detect++;
+		} while ((I915_READ(pipe_dsl_reg) == dsl));
+
+		/* restore vblank if necessary */
+		if (restore_vblank)
+			I915_WRITE(vblank_reg, vblank);
+		/*
+		 * If more than 3/4 of the scanline detected a monitor,
+		 * then it is assumed to be present. This works even on i830,
+		 * where there isn't any way to force the border color across
+		 * the screen
+		 */
+		status = detect * 4 > count * 3 ?
+			 connector_status_connected :
+			 connector_status_disconnected;
+	}
+
+	/* Restore previous settings */
+	I915_WRITE(bclrpat_reg, save_bclrpat);
+
+	return status;
+}
+
 static enum drm_connector_status intel_crt_detect(struct drm_connector *connector)
 {
 	struct drm_device *dev = connector->dev;
+	struct intel_output *intel_output = to_intel_output(connector);
+	struct drm_encoder *encoder = &intel_output->enc;
+	struct drm_crtc *crtc;
+	int dpms_mode;
+	enum drm_connector_status status;
 
 	if (IS_I9XX(dev) && !IS_I915G(dev) && !IS_I915GM(dev)) {
 		if (intel_crt_detect_hotplug(connector))
@@ -212,8 +345,20 @@ static enum drm_connector_status intel_crt_detect(struct drm_connector *connecto
 	if (intel_crt_detect_ddc(connector))
 		return connector_status_connected;
 
-	/* TODO use load detect */
-	return connector_status_unknown;
+	/* for pre-945g platforms use load detect */
+	if (encoder->crtc && encoder->crtc->enabled) {
+		status = intel_crt_load_detect(encoder->crtc, intel_output);
+	} else {
+		crtc = intel_get_load_detect_pipe(intel_output,
+						  NULL, &dpms_mode);
+		if (crtc) {
+			status = intel_crt_load_detect(crtc, intel_output);
+			intel_release_load_detect_pipe(intel_output, dpms_mode);
+		} else
+			status = connector_status_unknown;
+	}
+
+	return status;
 }
 
 static void intel_crt_destroy(struct drm_connector *connector)

drivers/gpu/drm/i915/intel_display.c

@@ -1357,7 +1357,7 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc,
 	int pipe = intel_crtc->pipe;
 	uint32_t control = (pipe == 0) ? CURACNTR : CURBCNTR;
 	uint32_t base = (pipe == 0) ? CURABASE : CURBBASE;
-	uint32_t temp;
+	uint32_t temp = I915_READ(control);
 	size_t addr;
 	int ret;
 
@@ -1366,7 +1366,12 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc,
 	/* if we want to turn off the cursor ignore width and height */
 	if (!handle) {
 		DRM_DEBUG("cursor off\n");
-		temp = CURSOR_MODE_DISABLE;
+		if (IS_MOBILE(dev) || IS_I9XX(dev)) {
+			temp &= ~(CURSOR_MODE | MCURSOR_GAMMA_ENABLE);
+			temp |= CURSOR_MODE_DISABLE;
+		} else {
+			temp &= ~(CURSOR_ENABLE | CURSOR_GAMMA_ENABLE);
+		}
 		addr = 0;
 		bo = NULL;
 		mutex_lock(&dev->struct_mutex);
@@ -1409,10 +1414,19 @@ static int intel_crtc_cursor_set(struct drm_crtc *crtc,
 		addr = obj_priv->phys_obj->handle->busaddr;
 	}
 
-	temp = 0;
-	/* set the pipe for the cursor */
-	temp |= (pipe << 28);
-	temp |= CURSOR_MODE_64_ARGB_AX | MCURSOR_GAMMA_ENABLE;
+	if (!IS_I9XX(dev))
+		I915_WRITE(CURSIZE, (height << 12) | width);
+
+	/* Hooray for CUR*CNTR differences */
+	if (IS_MOBILE(dev) || IS_I9XX(dev)) {
+		temp &= ~(CURSOR_MODE | MCURSOR_PIPE_SELECT);
+		temp |= CURSOR_MODE_64_ARGB_AX | MCURSOR_GAMMA_ENABLE;
+		temp |= (pipe << 28); /* Connect to correct pipe */
+	} else {
+		temp &= ~(CURSOR_FORMAT_MASK);
+		temp |= CURSOR_ENABLE;
+		temp |= CURSOR_FORMAT_ARGB | CURSOR_GAMMA_ENABLE;
+	}
 
  finish:
 	I915_WRITE(control, temp);

drivers/gpu/drm/i915/intel_lvds.c

@@ -511,10 +511,10 @@ void intel_lvds_init(struct drm_device *dev)
 	}
 
 	/* Failed to get EDID, what about VBT? */
-	if (dev_priv->vbt_mode) {
+	if (dev_priv->lfp_lvds_vbt_mode) {
 		mutex_lock(&dev->mode_config.mutex);
 		dev_priv->panel_fixed_mode =
-			drm_mode_duplicate(dev, dev_priv->vbt_mode);
+			drm_mode_duplicate(dev, dev_priv->lfp_lvds_vbt_mode);
 		mutex_unlock(&dev->mode_config.mutex);
 		if (dev_priv->panel_fixed_mode) {
 			dev_priv->panel_fixed_mode->type |=

drivers/gpu/drm/i915/intel_sdvo.c

@@ -69,6 +69,10 @@ struct intel_sdvo_priv {
 	 * This is set if we treat the device as HDMI, instead of DVI.
 	 */
 	bool is_hdmi;
+	/**
+	 * This is set if we detect output of sdvo device as LVDS.
+	 */
+	bool is_lvds;
 
 	/**
 	 * Returned SDTV resolutions allowed for the current format, if the
@@ -1398,10 +1402,8 @@ static enum drm_connector_status intel_sdvo_detect(struct drm_connector *connect
 static void intel_sdvo_get_ddc_modes(struct drm_connector *connector)
 {
 	struct intel_output *intel_output = to_intel_output(connector);
-	struct intel_sdvo_priv *sdvo_priv = intel_output->dev_priv;
 
 	/* set the bus switch and get the modes */
-	intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus);
 	intel_ddc_get_modes(intel_output);
 
 #if 0
@@ -1543,6 +1545,37 @@ static void intel_sdvo_get_tv_modes(struct drm_connector *connector)
 		}
 }
 
+static void intel_sdvo_get_lvds_modes(struct drm_connector *connector)
+{
+	struct intel_output *intel_output = to_intel_output(connector);
+	struct intel_sdvo_priv *sdvo_priv = intel_output->dev_priv;
+	struct drm_i915_private *dev_priv = connector->dev->dev_private;
+
+	/*
+	 * Attempt to get the mode list from DDC.
+	 * Assume that the preferred modes are
+	 * arranged in priority order.
+	 */
+	/* set the bus switch and get the modes */
+	intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus);
+	intel_ddc_get_modes(intel_output);
+	if (list_empty(&connector->probed_modes) == false)
+		return;
+
+	/* Fetch modes from VBT */
+	if (dev_priv->sdvo_lvds_vbt_mode != NULL) {
+		struct drm_display_mode *newmode;
+		newmode = drm_mode_duplicate(connector->dev,
+					     dev_priv->sdvo_lvds_vbt_mode);
+		if (newmode != NULL) {
+			/* Guarantee the mode is preferred */
+			newmode->type = (DRM_MODE_TYPE_PREFERRED |
+					 DRM_MODE_TYPE_DRIVER);
+			drm_mode_probed_add(connector, newmode);
+		}
+	}
+}
+
 static int intel_sdvo_get_modes(struct drm_connector *connector)
 {
 	struct intel_output *output = to_intel_output(connector);
@@ -1550,6 +1583,8 @@ static int intel_sdvo_get_modes(struct drm_connector *connector)
 
 	if (sdvo_priv->is_tv)
 		intel_sdvo_get_tv_modes(connector);
+	else if (sdvo_priv->is_lvds == true)
+		intel_sdvo_get_lvds_modes(connector);
 	else
 		intel_sdvo_get_ddc_modes(connector);
 
@@ -1564,6 +1599,9 @@ static void intel_sdvo_destroy(struct drm_connector *connector)
 
 	if (intel_output->i2c_bus)
 		intel_i2c_destroy(intel_output->i2c_bus);
+	if (intel_output->ddc_bus)
+		intel_i2c_destroy(intel_output->ddc_bus);
+
 	drm_sysfs_connector_remove(connector);
 	drm_connector_cleanup(connector);
 	kfree(intel_output);
@@ -1660,12 +1698,56 @@ intel_sdvo_get_digital_encoding_mode(struct intel_output *output)
 	return true;
 }
 
+static struct intel_output *
+intel_sdvo_chan_to_intel_output(struct intel_i2c_chan *chan)
+{
+	struct drm_device *dev = chan->drm_dev;
+	struct drm_connector *connector;
+	struct intel_output *intel_output = NULL;
+
+	list_for_each_entry(connector,
+			&dev->mode_config.connector_list, head) {
+		if (to_intel_output(connector)->ddc_bus == chan) {
+			intel_output = to_intel_output(connector);
+			break;
+		}
+	}
+	return intel_output;
+}
+
+static int intel_sdvo_master_xfer(struct i2c_adapter *i2c_adap,
+				  struct i2c_msg msgs[], int num)
+{
+	struct intel_output *intel_output;
+	struct intel_sdvo_priv *sdvo_priv;
+	struct i2c_algo_bit_data *algo_data;
+	struct i2c_algorithm *algo;
+
+	algo_data = (struct i2c_algo_bit_data *)i2c_adap->algo_data;
+	intel_output =
+		intel_sdvo_chan_to_intel_output(
+				(struct intel_i2c_chan *)(algo_data->data));
+	if (intel_output == NULL)
+		return -EINVAL;
+
+	sdvo_priv = intel_output->dev_priv;
+	algo = (struct i2c_algorithm *)intel_output->i2c_bus->adapter.algo;
+
+	intel_sdvo_set_control_bus_switch(intel_output, sdvo_priv->ddc_bus);
+	return algo->master_xfer(i2c_adap, msgs, num);
+}
+
+static struct i2c_algorithm intel_sdvo_i2c_bit_algo = {
+	.master_xfer	= intel_sdvo_master_xfer,
+};
+
 bool intel_sdvo_init(struct drm_device *dev, int output_device)
 {
 	struct drm_connector *connector;
 	struct intel_output *intel_output;
 	struct intel_sdvo_priv *sdvo_priv;
 	struct intel_i2c_chan *i2cbus = NULL;
+	struct intel_i2c_chan *ddcbus = NULL;
 	int connector_type;
 	u8 ch[0x40];
 	int i;
@@ -1676,17 +1758,9 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 		return false;
 	}
 
-	connector = &intel_output->base;
-
-	drm_connector_init(dev, connector, &intel_sdvo_connector_funcs,
-			   DRM_MODE_CONNECTOR_Unknown);
-	drm_connector_helper_add(connector, &intel_sdvo_connector_helper_funcs);
 	sdvo_priv = (struct intel_sdvo_priv *)(intel_output + 1);
 	intel_output->type = INTEL_OUTPUT_SDVO;
 
-	connector->interlace_allowed = 0;
-	connector->doublescan_allowed = 0;
-
 	/* setup the DDC bus. */
 	if (output_device == SDVOB)
 		i2cbus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOB");
@@ -1694,7 +1768,7 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 		i2cbus = intel_i2c_create(dev, GPIOE, "SDVOCTRL_E for SDVOC");
 
 	if (!i2cbus)
-		goto err_connector;
+		goto err_inteloutput;
 
 	sdvo_priv->i2c_bus = i2cbus;
 
@@ -1710,7 +1784,6 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 	intel_output->i2c_bus = i2cbus;
 	intel_output->dev_priv = sdvo_priv;
 
-
 	/* Read the regs to test if we can talk to the device */
 	for (i = 0; i < 0x40; i++) {
 		if (!intel_sdvo_read_byte(intel_output, i, &ch[i])) {
@@ -1720,6 +1793,22 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 		}
 	}
 
+	/* setup the DDC bus. */
+	if (output_device == SDVOB)
+		ddcbus = intel_i2c_create(dev, GPIOE, "SDVOB DDC BUS");
+	else
+		ddcbus = intel_i2c_create(dev, GPIOE, "SDVOC DDC BUS");
+
+	if (ddcbus == NULL)
+		goto err_i2c;
+
+	intel_sdvo_i2c_bit_algo.functionality =
+		intel_output->i2c_bus->adapter.algo->functionality;
+	ddcbus->adapter.algo = &intel_sdvo_i2c_bit_algo;
+	intel_output->ddc_bus = ddcbus;
+
+	/* In defaut case sdvo lvds is false */
+	sdvo_priv->is_lvds = false;
 	intel_sdvo_get_capabilities(intel_output, &sdvo_priv->caps);
 
 	if (sdvo_priv->caps.output_flags &
@@ -1729,7 +1818,6 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 		else
 			sdvo_priv->controlled_output = SDVO_OUTPUT_TMDS1;
 
-		connector->display_info.subpixel_order = SubPixelHorizontalRGB;
 		encoder_type = DRM_MODE_ENCODER_TMDS;
 		connector_type = DRM_MODE_CONNECTOR_DVID;
 
@@ -1747,7 +1835,6 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 	else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_SVID0)
 	{
 		sdvo_priv->controlled_output = SDVO_OUTPUT_SVID0;
-		connector->display_info.subpixel_order = SubPixelHorizontalRGB;
 		encoder_type = DRM_MODE_ENCODER_TVDAC;
 		connector_type = DRM_MODE_CONNECTOR_SVIDEO;
 		sdvo_priv->is_tv = true;
@@ -1756,30 +1843,28 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 	else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_RGB0)
 	{
 		sdvo_priv->controlled_output = SDVO_OUTPUT_RGB0;
-		connector->display_info.subpixel_order = SubPixelHorizontalRGB;
 		encoder_type = DRM_MODE_ENCODER_DAC;
 		connector_type = DRM_MODE_CONNECTOR_VGA;
 	}
 	else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_RGB1)
 	{
 		sdvo_priv->controlled_output = SDVO_OUTPUT_RGB1;
-		connector->display_info.subpixel_order = SubPixelHorizontalRGB;
 		encoder_type = DRM_MODE_ENCODER_DAC;
 		connector_type = DRM_MODE_CONNECTOR_VGA;
 	}
 	else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_LVDS0)
 	{
 		sdvo_priv->controlled_output = SDVO_OUTPUT_LVDS0;
-		connector->display_info.subpixel_order = SubPixelHorizontalRGB;
 		encoder_type = DRM_MODE_ENCODER_LVDS;
 		connector_type = DRM_MODE_CONNECTOR_LVDS;
+		sdvo_priv->is_lvds = true;
 	}
 	else if (sdvo_priv->caps.output_flags & SDVO_OUTPUT_LVDS1)
 	{
 		sdvo_priv->controlled_output = SDVO_OUTPUT_LVDS1;
-		connector->display_info.subpixel_order = SubPixelHorizontalRGB;
 		encoder_type = DRM_MODE_ENCODER_LVDS;
 		connector_type = DRM_MODE_CONNECTOR_LVDS;
+		sdvo_priv->is_lvds = true;
 	}
 	else
 	{
@@ -1795,9 +1880,16 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 		goto err_i2c;
 	}
 
+	connector = &intel_output->base;
+	drm_connector_init(dev, connector, &intel_sdvo_connector_funcs,
+			   connector_type);
+	drm_connector_helper_add(connector, &intel_sdvo_connector_helper_funcs);
+	connector->interlace_allowed = 0;
+	connector->doublescan_allowed = 0;
+	connector->display_info.subpixel_order = SubPixelHorizontalRGB;
+
 	drm_encoder_init(dev, &intel_output->enc, &intel_sdvo_enc_funcs, encoder_type);
 	drm_encoder_helper_add(&intel_output->enc, &intel_sdvo_helper_funcs);
-	connector->connector_type = connector_type;
 
 	drm_mode_connector_attach_encoder(&intel_output->base, &intel_output->enc);
 	drm_sysfs_connector_add(connector);
@@ -1829,14 +1921,13 @@ bool intel_sdvo_init(struct drm_device *dev, int output_device)
 		  sdvo_priv->caps.output_flags &
 			(SDVO_OUTPUT_TMDS1 | SDVO_OUTPUT_RGB1) ? 'Y' : 'N');
 
-	intel_output->ddc_bus = i2cbus;
-
 	return true;
 
 err_i2c:
+	if (ddcbus != NULL)
+		intel_i2c_destroy(intel_output->ddc_bus);
 	intel_i2c_destroy(intel_output->i2c_bus);
-err_connector:
-	drm_connector_cleanup(connector);
+err_inteloutput:
 	kfree(intel_output);
 
 	return false;

drivers/isdn/gigaset/isocdata.c

@@ -175,7 +175,7 @@ int gigaset_isowbuf_getbytes(struct isowbuf_t *iwb, int size)
 		return -EINVAL;
 	}
 	src = iwb->read;
-	if (unlikely(limit > BAS_OUTBUFSIZE + BAS_OUTBUFPAD ||
+	if (unlikely(limit >= BAS_OUTBUFSIZE + BAS_OUTBUFPAD ||
 		     (read < src && limit >= src))) {
 		pr_err("isoc write buffer frame reservation violated\n");
 		return -EFAULT;

drivers/lguest/x86/core.c

@@ -358,6 +358,16 @@ void lguest_arch_handle_trap(struct lg_cpu *cpu)
 			if (emulate_insn(cpu))
 				return;
 		}
+		/* If KVM is active, the vmcall instruction triggers a
+		 * General Protection Fault.  Normally it triggers an
+		 * invalid opcode fault (6): */
+	case 6:
+		/* We need to check if ring == GUEST_PL and
+		 * faulting instruction == vmcall. */
+		if (is_hypercall(cpu)) {
+			rewrite_hypercall(cpu);
+			return;
+		}
 		break;
 	case 14: /* We've intercepted a Page Fault. */
 		/* The Guest accessed a virtual address that wasn't mapped.
@@ -403,15 +413,6 @@ void lguest_arch_handle_trap(struct lg_cpu *cpu)
 		 * up the pointer now to indicate a hypercall is pending. */
 		cpu->hcall = (struct hcall_args *)cpu->regs;
 		return;
-	case 6:
-		/* kvm hypercalls trigger an invalid opcode fault (6).
-		 * We need to check if ring == GUEST_PL and
-		 * faulting instruction == vmcall. */
-		if (is_hypercall(cpu)) {
-			rewrite_hypercall(cpu);
-			return;
-		}
-		break;
 	}
 
 	/* We didn't handle the trap, so it needs to go to the Guest. */

drivers/md/bitmap.c

@@ -1097,14 +1097,12 @@ void bitmap_daemon_work(struct bitmap *bitmap)
 	}
 	bitmap->allclean = 1;
 
+	spin_lock_irqsave(&bitmap->lock, flags);
 	for (j = 0; j < bitmap->chunks; j++) {
 		bitmap_counter_t *bmc;
-		spin_lock_irqsave(&bitmap->lock, flags);
-		if (!bitmap->filemap) {
+		if (!bitmap->filemap)
 			/* error or shutdown */
-			spin_unlock_irqrestore(&bitmap->lock, flags);
 			break;
-		}
 
 		page = filemap_get_page(bitmap, j);
 
@@ -1121,6 +1119,8 @@ void bitmap_daemon_work(struct bitmap *bitmap)
 					write_page(bitmap, page, 0);
 					bitmap->allclean = 0;
 				}
+				spin_lock_irqsave(&bitmap->lock, flags);
+				j |= (PAGE_BITS - 1);
 				continue;
 			}
 
@@ -1181,9 +1181,10 @@ void bitmap_daemon_work(struct bitmap *bitmap)
 					ext2_clear_bit(file_page_offset(j), paddr);
 				kunmap_atomic(paddr, KM_USER0);
 			}
-		}
-		spin_unlock_irqrestore(&bitmap->lock, flags);
+		} else
+			j |= PAGE_COUNTER_MASK;
 	}
+	spin_unlock_irqrestore(&bitmap->lock, flags);
 
 	/* now sync the final page */
 	if (lastpage != NULL) {

drivers/md/md.c

@@ -1375,6 +1375,9 @@ static void super_1_sync(mddev_t *mddev, mdk_rdev_t *rdev)
 
 	sb->raid_disks = cpu_to_le32(mddev->raid_disks);
 	sb->size = cpu_to_le64(mddev->dev_sectors);
+	sb->chunksize = cpu_to_le32(mddev->chunk_size >> 9);
+	sb->level = cpu_to_le32(mddev->level);
+	sb->layout = cpu_to_le32(mddev->layout);
 
 	if (mddev->bitmap && mddev->bitmap_file == NULL) {
 		sb->bitmap_offset = cpu_to_le32((__u32)mddev->bitmap_offset);
@@ -3303,7 +3306,9 @@ static ssize_t
 action_show(mddev_t *mddev, char *page)
 {
 	char *type = "idle";
-	if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) ||
+	if (test_bit(MD_RECOVERY_FROZEN, &mddev->recovery))
+		type = "frozen";
+	else if (test_bit(MD_RECOVERY_RUNNING, &mddev->recovery) ||
 	    (!mddev->ro && test_bit(MD_RECOVERY_NEEDED, &mddev->recovery))) {
 		if (test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery))
 			type = "reshape";
@@ -3326,7 +3331,12 @@ action_store(mddev_t *mddev, const char *page, size_t len)
 	if (!mddev->pers || !mddev->pers->sync_request)
 		return -EINVAL;
 
-	if (cmd_match(page, "idle")) {
+	if (cmd_match(page, "frozen"))
+		set_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+	else
+		clear_bit(MD_RECOVERY_FROZEN, &mddev->recovery);
+
+	if (cmd_match(page, "idle") || cmd_match(page, "frozen")) {
 		if (mddev->sync_thread) {
 			set_bit(MD_RECOVERY_INTR, &mddev->recovery);
 			md_unregister_thread(mddev->sync_thread);
@@ -3680,7 +3690,7 @@ array_size_store(mddev_t *mddev, const char *buf, size_t len)
 		if (strict_blocks_to_sectors(buf, &sectors) < 0)
 			return -EINVAL;
 		if (mddev->pers && mddev->pers->size(mddev, 0, 0) < sectors)
-			return -EINVAL;
+			return -E2BIG;
 
 		mddev->external_size = 1;
 	}
@@ -5557,7 +5567,7 @@ static struct block_device_operations md_fops =
 	.owner		= THIS_MODULE,
 	.open		= md_open,
 	.release	= md_release,
-	.locked_ioctl	= md_ioctl,
+	.ioctl		= md_ioctl,
 	.getgeo		= md_getgeo,
 	.media_changed	= md_media_changed,
 	.revalidate_disk= md_revalidate,
@@ -6352,12 +6362,13 @@ void md_do_sync(mddev_t *mddev)
 
 		skipped = 0;
 
-		if ((mddev->curr_resync > mddev->curr_resync_completed &&
-		     (mddev->curr_resync - mddev->curr_resync_completed)
-		    > (max_sectors >> 4)) ||
-		    (j - mddev->curr_resync_completed)*2
-		    >= mddev->resync_max - mddev->curr_resync_completed
-			) {
+		if (!test_bit(MD_RECOVERY_RESHAPE, &mddev->recovery) &&
+		    ((mddev->curr_resync > mddev->curr_resync_completed &&
+		      (mddev->curr_resync - mddev->curr_resync_completed)
+		      > (max_sectors >> 4)) ||
+		     (j - mddev->curr_resync_completed)*2
+		     >= mddev->resync_max - mddev->curr_resync_completed
+			    )) {
 			/* time to update curr_resync_completed */
 			blk_unplug(mddev->queue);
 			wait_event(mddev->recovery_wait,

drivers/md/raid5.c

@@ -3811,13 +3811,13 @@ static sector_t reshape_request(mddev_t *mddev, sector_t sector_nr, int *skipped
 	safepos = conf->reshape_safe;
 	sector_div(safepos, data_disks);
 	if (mddev->delta_disks < 0) {
-		writepos -= reshape_sectors;
+		writepos -= min_t(sector_t, reshape_sectors, writepos);
 		readpos += reshape_sectors;
 		safepos += reshape_sectors;
 	} else {
 		writepos += reshape_sectors;
-		readpos -= reshape_sectors;
-		safepos -= reshape_sectors;
+		readpos -= min_t(sector_t, reshape_sectors, readpos);
+		safepos -= min_t(sector_t, reshape_sectors, safepos);
 	}
 
 	/* 'writepos' is the most advanced device address we might write.

drivers/misc/enclosure.c

@@ -119,7 +119,7 @@ enclosure_register(struct device *dev, const char *name, int components,
 	edev->edev.class = &enclosure_class;
 	edev->edev.parent = get_device(dev);
 	edev->cb = cb;
-	dev_set_name(&edev->edev, name);
+	dev_set_name(&edev->edev, "%s", name);
 	err = device_register(&edev->edev);
 	if (err)
 		goto err;
@@ -255,8 +255,8 @@ enclosure_component_register(struct enclosure_device *edev,
 	ecomp->number = number;
 	cdev = &ecomp->cdev;
 	cdev->parent = get_device(&edev->edev);
-	if (name)
-		dev_set_name(cdev, name);
+	if (name && name[0])
+		dev_set_name(cdev, "%s", name);
 	else
 		dev_set_name(cdev, "%u", number);
 

drivers/net/Makefile

@@ -102,7 +102,7 @@ obj-$(CONFIG_HAMACHI) += hamachi.o
 obj-$(CONFIG_NET) += Space.o loopback.o
 obj-$(CONFIG_SEEQ8005) += seeq8005.o
 obj-$(CONFIG_NET_SB1000) += sb1000.o
-obj-$(CONFIG_MAC8390) += mac8390.o 8390.o
+obj-$(CONFIG_MAC8390) += mac8390.o
 obj-$(CONFIG_APNE) += apne.o 8390.o
 obj-$(CONFIG_PCMCIA_PCNET) += 8390.o
 obj-$(CONFIG_HP100) += hp100.o

drivers/net/gianfar.c

@@ -1885,8 +1885,17 @@ int gfar_clean_rx_ring(struct net_device *dev, int rx_work_limit)
 
 			if (unlikely(!newskb))
 				newskb = skb;
-			else if (skb)
+			else if (skb) {
+				/*
+				 * We need to reset ->data to what it
+				 * was before gfar_new_skb() re-aligned
+				 * it to an RXBUF_ALIGNMENT boundary
+				 * before we put the skb back on the
+				 * recycle list.
+				 */
+				skb->data = skb->head + NET_SKB_PAD;
 				__skb_queue_head(&priv->rx_recycle, skb);
+			}
 		} else {
 			/* Increment the number of packets */
 			dev->stats.rx_packets++;

drivers/net/mac8390.c

@@ -304,7 +304,7 @@ struct net_device * __init mac8390_probe(int unit)
 	if (!MACH_IS_MAC)
 		return ERR_PTR(-ENODEV);
 
-	dev = alloc_ei_netdev();
+	dev = ____alloc_ei_netdev(0);
 	if (!dev)
 		return ERR_PTR(-ENOMEM);
 
@@ -481,10 +481,10 @@ void cleanup_module(void)
 static const struct net_device_ops mac8390_netdev_ops = {
 	.ndo_open 		= mac8390_open,
 	.ndo_stop		= mac8390_close,
-	.ndo_start_xmit		= ei_start_xmit,
-	.ndo_tx_timeout		= ei_tx_timeout,
-	.ndo_get_stats		= ei_get_stats,
-	.ndo_set_multicast_list = ei_set_multicast_list,
+	.ndo_start_xmit		= __ei_start_xmit,
+	.ndo_tx_timeout		= __ei_tx_timeout,
+	.ndo_get_stats		= __ei_get_stats,
+	.ndo_set_multicast_list = __ei_set_multicast_list,
 	.ndo_validate_addr	= eth_validate_addr,
 	.ndo_set_mac_address 	= eth_mac_addr,
 	.ndo_change_mtu		= eth_change_mtu,

drivers/net/r8169.c

@@ -3554,54 +3554,64 @@ static irqreturn_t rtl8169_interrupt(int irq, void *dev_instance)
 	int handled = 0;
 	int status;
 
+	/* loop handling interrupts until we have no new ones or
+	 * we hit a invalid/hotplug case.
+	 */
 	status = RTL_R16(IntrStatus);
+	while (status && status != 0xffff) {
+		handled = 1;
 
-	/* hotplug/major error/no more work/shared irq */
-	if ((status == 0xffff) || !status)
-		goto out;
-
-	handled = 1;
+		/* Handle all of the error cases first. These will reset
+		 * the chip, so just exit the loop.
+		 */
+		if (unlikely(!netif_running(dev))) {
+			rtl8169_asic_down(ioaddr);
+			break;
+		}
 
-	if (unlikely(!netif_running(dev))) {
-		rtl8169_asic_down(ioaddr);
-		goto out;
-	}
+		/* Work around for rx fifo overflow */
+		if (unlikely(status & RxFIFOOver) &&
+		(tp->mac_version == RTL_GIGA_MAC_VER_11)) {
+			netif_stop_queue(dev);
+			rtl8169_tx_timeout(dev);
+			break;
+		}
 
-	status &= tp->intr_mask;
-	RTL_W16(IntrStatus,
-		(status & RxFIFOOver) ? (status | RxOverflow) : status);
+		if (unlikely(status & SYSErr)) {
+			rtl8169_pcierr_interrupt(dev);
+			break;
+		}
 
-	if (!(status & tp->intr_event))
-		goto out;
+		if (status & LinkChg)
+			rtl8169_check_link_status(dev, tp, ioaddr);
 
-	/* Work around for rx fifo overflow */
-	if (unlikely(status & RxFIFOOver) &&
-	    (tp->mac_version == RTL_GIGA_MAC_VER_11)) {
-		netif_stop_queue(dev);
-		rtl8169_tx_timeout(dev);
-		goto out;
-	}
+		/* We need to see the lastest version of tp->intr_mask to
+		 * avoid ignoring an MSI interrupt and having to wait for
+		 * another event which may never come.
+		 */
+		smp_rmb();
+		if (status & tp->intr_mask & tp->napi_event) {
+			RTL_W16(IntrMask, tp->intr_event & ~tp->napi_event);
+			tp->intr_mask = ~tp->napi_event;
+
+			if (likely(napi_schedule_prep(&tp->napi)))
+				__napi_schedule(&tp->napi);
+			else if (netif_msg_intr(tp)) {
+				printk(KERN_INFO "%s: interrupt %04x in poll\n",
+				dev->name, status);
+			}
+		}
 
-	if (unlikely(status & SYSErr)) {
-		rtl8169_pcierr_interrupt(dev);
-		goto out;
+		/* We only get a new MSI interrupt when all active irq
+		 * sources on the chip have been acknowledged. So, ack
+		 * everything we've seen and check if new sources have become
+		 * active to avoid blocking all interrupts from the chip.
+		 */
+		RTL_W16(IntrStatus,
+			(status & RxFIFOOver) ? (status | RxOverflow) : status);
+		status = RTL_R16(IntrStatus);
 	}
 
-	if (status & LinkChg)
-		rtl8169_check_link_status(dev, tp, ioaddr);
-
-	if (status & tp->napi_event) {
-		RTL_W16(IntrMask, tp->intr_event & ~tp->napi_event);
-		tp->intr_mask = ~tp->napi_event;
-
-		if (likely(napi_schedule_prep(&tp->napi)))
-			__napi_schedule(&tp->napi);
-		else if (netif_msg_intr(tp)) {
-			printk(KERN_INFO "%s: interrupt %04x in poll\n",
-			       dev->name, status);
-		}
-	}
-out:
 	return IRQ_RETVAL(handled);
 }
 
@@ -3617,13 +3627,15 @@ static int rtl8169_poll(struct napi_struct *napi, int budget)
 
 	if (work_done < budget) {
 		napi_complete(napi);
-		tp->intr_mask = 0xffff;
-		/*
-		 * 20040426: the barrier is not strictly required but the
-		 * behavior of the irq handler could be less predictable
-		 * without it. Btw, the lack of flush for the posted pci
-		 * write is safe - FR
+
+		/* We need for force the visibility of tp->intr_mask
+		 * for other CPUs, as we can loose an MSI interrupt
+		 * and potentially wait for a retransmit timeout if we don't.
+		 * The posted write to IntrMask is safe, as it will
+		 * eventually make it to the chip and we won't loose anything
+		 * until it does.
 		 */
+		tp->intr_mask = 0xffff;
 		smp_wmb();
 		RTL_W16(IntrMask, tp->intr_event);
 	}

drivers/net/wimax/i2400m/usb.c

@@ -505,27 +505,52 @@ int i2400mu_suspend(struct usb_interface *iface, pm_message_t pm_msg)
 #ifdef CONFIG_PM
 	struct usb_device *usb_dev = i2400mu->usb_dev;
 #endif
+	unsigned is_autosuspend = 0;
 	struct i2400m *i2400m = &i2400mu->i2400m;
 
+#ifdef CONFIG_PM
+	if (usb_dev->auto_pm > 0)
+		is_autosuspend = 1;
+#endif
+
 	d_fnstart(3, dev, "(iface %p pm_msg %u)\n", iface, pm_msg.event);
 	if (i2400m->updown == 0)
 		goto no_firmware;
-	d_printf(1, dev, "fw up, requesting standby\n");
+	if (i2400m->state == I2400M_SS_DATA_PATH_CONNECTED && is_autosuspend) {
+		/* ugh -- the device is connected and this suspend
+		 * request is an autosuspend one (not a system standby
+		 * / hibernate).
+		 *
+		 * The only way the device can go to standby is if the
+		 * link with the base station is in IDLE mode; that
+		 * were the case, we'd be in status
+		 * I2400M_SS_CONNECTED_IDLE. But we are not.
+		 *
+		 * If we *tell* him to go power save now, it'll reset
+		 * as a precautionary measure, so if this is an
+		 * autosuspend thing, say no and it'll come back
+		 * later, when the link is IDLE
+		 */
+		result = -EBADF;
+		d_printf(1, dev, "fw up, link up, not-idle, autosuspend: "
+			 "not entering powersave\n");
+		goto error_not_now;
+	}
+	d_printf(1, dev, "fw up: entering powersave\n");
 	atomic_dec(&i2400mu->do_autopm);
 	result = i2400m_cmd_enter_powersave(i2400m);
 	atomic_inc(&i2400mu->do_autopm);
-#ifdef CONFIG_PM
-	if (result < 0 && usb_dev->auto_pm == 0) {
+	if (result < 0 && !is_autosuspend) {
 		/* System suspend, can't fail */
 		dev_err(dev, "failed to suspend, will reset on resume\n");
 		result = 0;
 	}
-#endif
 	if (result < 0)
 		goto error_enter_powersave;
 	i2400mu_notification_release(i2400mu);
-	d_printf(1, dev, "fw up, got standby\n");
+	d_printf(1, dev, "powersave requested\n");
 error_enter_powersave:
+error_not_now:
 no_firmware:
 	d_fnend(3, dev, "(iface %p pm_msg %u) = %d\n",
 		iface, pm_msg.event, result);

drivers/net/wireless/airo.c

@@ -6467,6 +6467,7 @@ static int airo_get_encode(struct net_device *dev,
 {
 	struct airo_info *local = dev->ml_priv;
 	int index = (dwrq->flags & IW_ENCODE_INDEX) - 1;
+	int wep_key_len;
 	u8 buf[16];
 
 	if (!local->wep_capable)
@@ -6500,11 +6501,13 @@ static int airo_get_encode(struct net_device *dev,
 	dwrq->flags |= index + 1;
 
 	/* Copy the key to the user buffer */
-	dwrq->length = get_wep_key(local, index, &buf[0], sizeof(buf));
-	if (dwrq->length != -1)
-		memcpy(extra, buf, dwrq->length);
-	else
+	wep_key_len = get_wep_key(local, index, &buf[0], sizeof(buf));
+	if (wep_key_len < 0) {
 		dwrq->length = 0;
+	} else {
+		dwrq->length = wep_key_len;
+		memcpy(extra, buf, dwrq->length);
+	}
 
 	return 0;
 }
@@ -6617,7 +6620,7 @@ static int airo_get_encodeext(struct net_device *dev,
 	struct airo_info *local = dev->ml_priv;
 	struct iw_point *encoding = &wrqu->encoding;
 	struct iw_encode_ext *ext = (struct iw_encode_ext *)extra;
-	int idx, max_key_len;
+	int idx, max_key_len, wep_key_len;
 	u8 buf[16];
 
 	if (!local->wep_capable)
@@ -6661,11 +6664,13 @@ static int airo_get_encodeext(struct net_device *dev,
 	memset(extra, 0, 16);
 	
 	/* Copy the key to the user buffer */
-	ext->key_len = get_wep_key(local, idx, &buf[0], sizeof(buf));
-	if (ext->key_len != -1)
-		memcpy(extra, buf, ext->key_len);
-	else
+	wep_key_len = get_wep_key(local, idx, &buf[0], sizeof(buf));
+	if (wep_key_len < 0) {
 		ext->key_len = 0;
+	} else {
+		ext->key_len = wep_key_len;
+		memcpy(extra, buf, ext->key_len);
+	}
 
 	return 0;
 }

drivers/net/wireless/ath5k/phy.c

@@ -1487,28 +1487,35 @@ ath5k_get_linear_pcdac_min(const u8 *stepL, const u8 *stepR,
 {
 	s8 tmp;
 	s16 min_pwrL, min_pwrR;
-	s16 pwr_i = pwrL[0];
-
-	do {
-		pwr_i--;
-		tmp = (s8) ath5k_get_interpolated_value(pwr_i,
-						pwrL[0], pwrL[1],
-						stepL[0], stepL[1]);
-
-	} while (tmp > 1);
-
-	min_pwrL = pwr_i;
-
-	pwr_i = pwrR[0];
-	do {
-		pwr_i--;
-		tmp = (s8) ath5k_get_interpolated_value(pwr_i,
-						pwrR[0], pwrR[1],
-						stepR[0], stepR[1]);
-
-	} while (tmp > 1);
+	s16 pwr_i;
+
+	if (pwrL[0] == pwrL[1])
+		min_pwrL = pwrL[0];
+	else {
+		pwr_i = pwrL[0];
+		do {
+			pwr_i--;
+			tmp = (s8) ath5k_get_interpolated_value(pwr_i,
+							pwrL[0], pwrL[1],
+							stepL[0], stepL[1]);
+		} while (tmp > 1);
+
+		min_pwrL = pwr_i;
+	}
 
-	min_pwrR = pwr_i;
+	if (pwrR[0] == pwrR[1])
+		min_pwrR = pwrR[0];
+	else {
+		pwr_i = pwrR[0];
+		do {
+			pwr_i--;
+			tmp = (s8) ath5k_get_interpolated_value(pwr_i,
+							pwrR[0], pwrR[1],
+							stepR[0], stepR[1]);
+		} while (tmp > 1);
+
+		min_pwrR = pwr_i;
+	}
 
 	/* Keep the right boundary so that it works for both curves */
 	return max(min_pwrL, min_pwrR);

drivers/net/wireless/ath5k/reset.c

@@ -26,7 +26,7 @@
 \*****************************/
 
 #include <linux/pci.h> 		/* To determine if a card is pci-e */
-#include <linux/bitops.h>	/* For get_bitmask_order */
+#include <linux/log2.h>
 #include "ath5k.h"
 #include "reg.h"
 #include "base.h"
@@ -69,10 +69,10 @@ static inline int ath5k_hw_write_ofdm_timings(struct ath5k_hw *ah,
 
 	/* Get exponent
 	 * ALGO: coef_exp = 14 - highest set bit position */
-	coef_exp = get_bitmask_order(coef_scaled);
+	coef_exp = ilog2(coef_scaled);
 
 	/* Doesn't make sense if it's zero*/
-	if (!coef_exp)
+	if (!coef_scaled || !coef_exp)
 		return -EINVAL;
 
 	/* Note: we've shifted coef_scaled by 24 */
@@ -359,7 +359,7 @@ int ath5k_hw_nic_wakeup(struct ath5k_hw *ah, int flags, bool initial)
 			mode |= AR5K_PHY_MODE_FREQ_5GHZ;
 
 			if (ah->ah_radio == AR5K_RF5413)
-				clock |= AR5K_PHY_PLL_40MHZ_5413;
+				clock = AR5K_PHY_PLL_40MHZ_5413;
 			else
 				clock |= AR5K_PHY_PLL_40MHZ;
 

drivers/net/wireless/iwlwifi/iwl-5000.c

@@ -46,7 +46,7 @@
 #include "iwl-6000-hw.h"
 
 /* Highest firmware API version supported */
-#define IWL5000_UCODE_API_MAX 1
+#define IWL5000_UCODE_API_MAX 2
 #define IWL5150_UCODE_API_MAX 2
 
 /* Lowest firmware API version supported */

drivers/net/wireless/iwlwifi/iwl-agn.c

@@ -669,13 +669,6 @@ static int iwl_set_mode(struct iwl_priv *priv, int mode)
 	if (!iwl_is_ready_rf(priv))
 		return -EAGAIN;
 
-	cancel_delayed_work(&priv->scan_check);
-	if (iwl_scan_cancel_timeout(priv, 100)) {
-		IWL_WARN(priv, "Aborted scan still in progress after 100ms\n");
-		IWL_DEBUG_MAC80211(priv, "leaving - scan abort failed.\n");
-		return -EAGAIN;
-	}
-
 	iwl_commit_rxon(priv);
 
 	return 0;

drivers/net/wireless/iwlwifi/iwl-scan.c

@@ -227,9 +227,6 @@ static void iwl_rx_scan_complete_notif(struct iwl_priv *priv,
 	/* The HW is no longer scanning */
 	clear_bit(STATUS_SCAN_HW, &priv->status);
 
-	/* The scan completion notification came in, so kill that timer... */
-	cancel_delayed_work(&priv->scan_check);
-
 	IWL_DEBUG_INFO(priv, "Scan pass on %sGHz took %dms\n",
 		       (priv->scan_bands & BIT(IEEE80211_BAND_2GHZ)) ?
 						"2.4" : "5.2",
@@ -712,6 +709,8 @@ static void iwl_bg_request_scan(struct work_struct *data)
 
 	mutex_lock(&priv->mutex);
 
+	cancel_delayed_work(&priv->scan_check);
+
 	if (!iwl_is_ready(priv)) {
 		IWL_WARN(priv, "request scan called when driver not ready.\n");
 		goto done;
@@ -925,6 +924,8 @@ void iwl_bg_scan_completed(struct work_struct *work)
 
 	IWL_DEBUG_SCAN(priv, "SCAN complete scan\n");
 
+	cancel_delayed_work(&priv->scan_check);
+
 	ieee80211_scan_completed(priv->hw, false);
 
 	if (test_bit(STATUS_EXIT_PENDING, &priv->status))

drivers/net/wireless/iwlwifi/iwl3945-base.c

@@ -782,13 +782,6 @@ static int iwl3945_set_mode(struct iwl_priv *priv, int mode)
 	if (!iwl_is_ready_rf(priv))
 		return -EAGAIN;
 
-	cancel_delayed_work(&priv->scan_check);
-	if (iwl_scan_cancel_timeout(priv, 100)) {
-		IWL_WARN(priv, "Aborted scan still in progress after 100ms\n");
-		IWL_DEBUG_MAC80211(priv, "leaving - scan abort failed.\n");
-		return -EAGAIN;
-	}
-
 	iwl3945_commit_rxon(priv);
 
 	return 0;
@@ -3298,6 +3291,8 @@ static void iwl3945_bg_request_scan(struct work_struct *data)
 
 	mutex_lock(&priv->mutex);
 
+	cancel_delayed_work(&priv->scan_check);
+
 	if (!iwl_is_ready(priv)) {
 		IWL_WARN(priv, "request scan called when driver not ready.\n");
 		goto done;

drivers/net/wireless/rt2x00/rt2x00debug.c

@@ -138,7 +138,7 @@ void rt2x00debug_update_crypto(struct rt2x00_dev *rt2x00dev,
 
 	if (cipher == CIPHER_TKIP_NO_MIC)
 		cipher = CIPHER_TKIP;
-	if (cipher == CIPHER_NONE || cipher > CIPHER_MAX)
+	if (cipher == CIPHER_NONE || cipher >= CIPHER_MAX)
 		return;
 
 	/* Remove CIPHER_NONE index */

drivers/oprofile/cpu_buffer.c

@@ -78,16 +78,20 @@ void free_cpu_buffers(void)
 	op_ring_buffer_write = NULL;
 }
 
+#define RB_EVENT_HDR_SIZE 4
+
 int alloc_cpu_buffers(void)
 {
 	int i;
 
 	unsigned long buffer_size = oprofile_cpu_buffer_size;
+	unsigned long byte_size = buffer_size * (sizeof(struct op_sample) +
+						 RB_EVENT_HDR_SIZE);
 
-	op_ring_buffer_read = ring_buffer_alloc(buffer_size, OP_BUFFER_FLAGS);
+	op_ring_buffer_read = ring_buffer_alloc(byte_size, OP_BUFFER_FLAGS);
 	if (!op_ring_buffer_read)
 		goto fail;
-	op_ring_buffer_write = ring_buffer_alloc(buffer_size, OP_BUFFER_FLAGS);
+	op_ring_buffer_write = ring_buffer_alloc(byte_size, OP_BUFFER_FLAGS);
 	if (!op_ring_buffer_write)
 		goto fail;
 

drivers/scsi/3w-9xxx.c

@@ -1978,7 +1978,8 @@ static void twa_unmap_scsi_data(TW_Device_Extension *tw_dev, int request_id)
 {
 	struct scsi_cmnd *cmd = tw_dev->srb[request_id];
 
-	scsi_dma_unmap(cmd);
+	if (cmd->SCp.phase == TW_PHASE_SGLIST)
+		scsi_dma_unmap(cmd);
 } /* End twa_unmap_scsi_data() */
 
 /* scsi_host_template initializer */

drivers/scsi/3w-xxxx.c

@@ -6,7 +6,7 @@
    		     Arnaldo Carvalho de Melo <acme@conectiva.com.br>
                      Brad Strand <linux@3ware.com>
 
-   Copyright (C) 1999-2007 3ware Inc.
+   Copyright (C) 1999-2009 3ware Inc.
 
    Kernel compatiblity By: 	Andre Hedrick <andre@suse.com>
    Non-Copyright (C) 2000	Andre Hedrick <andre@suse.com>
@@ -1294,7 +1294,8 @@ static void tw_unmap_scsi_data(struct pci_dev *pdev, struct scsi_cmnd *cmd)
 {
 	dprintk(KERN_WARNING "3w-xxxx: tw_unmap_scsi_data()\n");
 
-	scsi_dma_unmap(cmd);
+	if (cmd->SCp.phase == TW_PHASE_SGLIST)
+		scsi_dma_unmap(cmd);
 } /* End tw_unmap_scsi_data() */
 
 /* This function will reset a device extension */

drivers/scsi/3w-xxxx.h

@@ -6,7 +6,7 @@
    		     Arnaldo Carvalho de Melo <acme@conectiva.com.br>
                      Brad Strand <linux@3ware.com>
 
-   Copyright (C) 1999-2007 3ware Inc.
+   Copyright (C) 1999-2009 3ware Inc.
 
    Kernel compatiblity By:	Andre Hedrick <andre@suse.com>
    Non-Copyright (C) 2000	Andre Hedrick <andre@suse.com>

drivers/scsi/Kconfig

@@ -628,6 +628,17 @@ config FCOE
 	---help---
 	  Fibre Channel over Ethernet module
 
+config FCOE_FNIC
+	tristate "Cisco FNIC Driver"
+	depends on PCI && X86
+	select LIBFC
+	help
+	  This is support for the Cisco PCI-Express FCoE HBA.
+
+	  To compile this driver as a module, choose M here and read
+	  <file:Documentation/scsi/scsi.txt>.
+	  The module will be called fnic.
+
 config SCSI_DMX3191D
 	tristate "DMX3191D SCSI support"
 	depends on PCI && SCSI

drivers/scsi/Makefile

@@ -39,6 +39,7 @@ obj-$(CONFIG_SCSI_DH)		+= device_handler/
 obj-$(CONFIG_LIBFC)		+= libfc/
 obj-$(CONFIG_LIBFCOE)		+= fcoe/
 obj-$(CONFIG_FCOE)		+= fcoe/
+obj-$(CONFIG_FCOE_FNIC)		+= fnic/
 obj-$(CONFIG_ISCSI_TCP) 	+= libiscsi.o	libiscsi_tcp.o iscsi_tcp.o
 obj-$(CONFIG_INFINIBAND_ISER) 	+= libiscsi.o
 obj-$(CONFIG_SCSI_A4000T)	+= 53c700.o	a4000t.o

drivers/scsi/fnic/Makefile

@@ -0,0 +1,15 @@
+obj-$(CONFIG_FCOE_FNIC) += fnic.o
+
+fnic-y	:= \
+	fnic_attrs.o \
+	fnic_isr.o \
+	fnic_main.o \
+	fnic_res.o \
+	fnic_fcs.o \
+	fnic_scsi.o \
+	vnic_cq.o \
+	vnic_dev.o \
+	vnic_intr.o \
+	vnic_rq.o \
+	vnic_wq_copy.o \
+	vnic_wq.o

drivers/scsi/fnic/cq_desc.h

@@ -0,0 +1,78 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _CQ_DESC_H_
+#define _CQ_DESC_H_
+
+/*
+ * Completion queue descriptor types
+ */
+enum cq_desc_types {
+	CQ_DESC_TYPE_WQ_ENET = 0,
+	CQ_DESC_TYPE_DESC_COPY = 1,
+	CQ_DESC_TYPE_WQ_EXCH = 2,
+	CQ_DESC_TYPE_RQ_ENET = 3,
+	CQ_DESC_TYPE_RQ_FCP = 4,
+};
+
+/* Completion queue descriptor: 16B
+ *
+ * All completion queues have this basic layout.  The
+ * type_specfic area is unique for each completion
+ * queue type.
+ */
+struct cq_desc {
+	__le16 completed_index;
+	__le16 q_number;
+	u8 type_specfic[11];
+	u8 type_color;
+};
+
+#define CQ_DESC_TYPE_BITS        4
+#define CQ_DESC_TYPE_MASK        ((1 << CQ_DESC_TYPE_BITS) - 1)
+#define CQ_DESC_COLOR_MASK       1
+#define CQ_DESC_COLOR_SHIFT      7
+#define CQ_DESC_Q_NUM_BITS       10
+#define CQ_DESC_Q_NUM_MASK       ((1 << CQ_DESC_Q_NUM_BITS) - 1)
+#define CQ_DESC_COMP_NDX_BITS    12
+#define CQ_DESC_COMP_NDX_MASK    ((1 << CQ_DESC_COMP_NDX_BITS) - 1)
+
+static inline void cq_desc_dec(const struct cq_desc *desc_arg,
+	u8 *type, u8 *color, u16 *q_number, u16 *completed_index)
+{
+	const struct cq_desc *desc = desc_arg;
+	const u8 type_color = desc->type_color;
+
+	*color = (type_color >> CQ_DESC_COLOR_SHIFT) & CQ_DESC_COLOR_MASK;
+
+	/*
+	 * Make sure color bit is read from desc *before* other fields
+	 * are read from desc.  Hardware guarantees color bit is last
+	 * bit (byte) written.  Adding the rmb() prevents the compiler
+	 * and/or CPU from reordering the reads which would potentially
+	 * result in reading stale values.
+	 */
+
+	rmb();
+
+	*type = type_color & CQ_DESC_TYPE_MASK;
+	*q_number = le16_to_cpu(desc->q_number) & CQ_DESC_Q_NUM_MASK;
+	*completed_index = le16_to_cpu(desc->completed_index) &
+		CQ_DESC_COMP_NDX_MASK;
+}
+
+#endif /* _CQ_DESC_H_ */

drivers/scsi/fnic/cq_enet_desc.h

@@ -0,0 +1,167 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _CQ_ENET_DESC_H_
+#define _CQ_ENET_DESC_H_
+
+#include "cq_desc.h"
+
+/* Ethernet completion queue descriptor: 16B */
+struct cq_enet_wq_desc {
+	__le16 completed_index;
+	__le16 q_number;
+	u8 reserved[11];
+	u8 type_color;
+};
+
+static inline void cq_enet_wq_desc_dec(struct cq_enet_wq_desc *desc,
+	u8 *type, u8 *color, u16 *q_number, u16 *completed_index)
+{
+	cq_desc_dec((struct cq_desc *)desc, type,
+		color, q_number, completed_index);
+}
+
+/* Completion queue descriptor: Ethernet receive queue, 16B */
+struct cq_enet_rq_desc {
+	__le16 completed_index_flags;
+	__le16 q_number_rss_type_flags;
+	__le32 rss_hash;
+	__le16 bytes_written_flags;
+	__le16 vlan;
+	__le16 checksum_fcoe;
+	u8 flags;
+	u8 type_color;
+};
+
+#define CQ_ENET_RQ_DESC_FLAGS_INGRESS_PORT          (0x1 << 12)
+#define CQ_ENET_RQ_DESC_FLAGS_FCOE                  (0x1 << 13)
+#define CQ_ENET_RQ_DESC_FLAGS_EOP                   (0x1 << 14)
+#define CQ_ENET_RQ_DESC_FLAGS_SOP                   (0x1 << 15)
+
+#define CQ_ENET_RQ_DESC_RSS_TYPE_BITS               4
+#define CQ_ENET_RQ_DESC_RSS_TYPE_MASK \
+	((1 << CQ_ENET_RQ_DESC_RSS_TYPE_BITS) - 1)
+#define CQ_ENET_RQ_DESC_RSS_TYPE_NONE               0
+#define CQ_ENET_RQ_DESC_RSS_TYPE_IPv4               1
+#define CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv4           2
+#define CQ_ENET_RQ_DESC_RSS_TYPE_IPv6               3
+#define CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv6           4
+#define CQ_ENET_RQ_DESC_RSS_TYPE_IPv6_EX            5
+#define CQ_ENET_RQ_DESC_RSS_TYPE_TCP_IPv6_EX        6
+
+#define CQ_ENET_RQ_DESC_FLAGS_CSUM_NOT_CALC         (0x1 << 14)
+
+#define CQ_ENET_RQ_DESC_BYTES_WRITTEN_BITS          14
+#define CQ_ENET_RQ_DESC_BYTES_WRITTEN_MASK \
+	((1 << CQ_ENET_RQ_DESC_BYTES_WRITTEN_BITS) - 1)
+#define CQ_ENET_RQ_DESC_FLAGS_TRUNCATED             (0x1 << 14)
+#define CQ_ENET_RQ_DESC_FLAGS_VLAN_STRIPPED         (0x1 << 15)
+
+#define CQ_ENET_RQ_DESC_FCOE_SOF_BITS               4
+#define CQ_ENET_RQ_DESC_FCOE_SOF_MASK \
+	((1 << CQ_ENET_RQ_DESC_FCOE_SOF_BITS) - 1)
+#define CQ_ENET_RQ_DESC_FCOE_EOF_BITS               8
+#define CQ_ENET_RQ_DESC_FCOE_EOF_MASK \
+	((1 << CQ_ENET_RQ_DESC_FCOE_EOF_BITS) - 1)
+#define CQ_ENET_RQ_DESC_FCOE_EOF_SHIFT              8
+
+#define CQ_ENET_RQ_DESC_FLAGS_TCP_UDP_CSUM_OK       (0x1 << 0)
+#define CQ_ENET_RQ_DESC_FCOE_FC_CRC_OK              (0x1 << 0)
+#define CQ_ENET_RQ_DESC_FLAGS_UDP                   (0x1 << 1)
+#define CQ_ENET_RQ_DESC_FCOE_ENC_ERROR              (0x1 << 1)
+#define CQ_ENET_RQ_DESC_FLAGS_TCP                   (0x1 << 2)
+#define CQ_ENET_RQ_DESC_FLAGS_IPV4_CSUM_OK          (0x1 << 3)
+#define CQ_ENET_RQ_DESC_FLAGS_IPV6                  (0x1 << 4)
+#define CQ_ENET_RQ_DESC_FLAGS_IPV4                  (0x1 << 5)
+#define CQ_ENET_RQ_DESC_FLAGS_IPV4_FRAGMENT         (0x1 << 6)
+#define CQ_ENET_RQ_DESC_FLAGS_FCS_OK                (0x1 << 7)
+
+static inline void cq_enet_rq_desc_dec(struct cq_enet_rq_desc *desc,
+	u8 *type, u8 *color, u16 *q_number, u16 *completed_index,
+	u8 *ingress_port, u8 *fcoe, u8 *eop, u8 *sop, u8 *rss_type,
+	u8 *csum_not_calc, u32 *rss_hash, u16 *bytes_written, u8 *packet_error,
+	u8 *vlan_stripped, u16 *vlan, u16 *checksum, u8 *fcoe_sof,
+	u8 *fcoe_fc_crc_ok, u8 *fcoe_enc_error, u8 *fcoe_eof,
+	u8 *tcp_udp_csum_ok, u8 *udp, u8 *tcp, u8 *ipv4_csum_ok,
+	u8 *ipv6, u8 *ipv4, u8 *ipv4_fragment, u8 *fcs_ok)
+{
+	u16 completed_index_flags = le16_to_cpu(desc->completed_index_flags);
+	u16 q_number_rss_type_flags =
+		le16_to_cpu(desc->q_number_rss_type_flags);
+	u16 bytes_written_flags = le16_to_cpu(desc->bytes_written_flags);
+
+	cq_desc_dec((struct cq_desc *)desc, type,
+		color, q_number, completed_index);
+
+	*ingress_port = (completed_index_flags &
+		CQ_ENET_RQ_DESC_FLAGS_INGRESS_PORT) ? 1 : 0;
+	*fcoe = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_FCOE) ?
+		1 : 0;
+	*eop = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_EOP) ?
+		1 : 0;
+	*sop = (completed_index_flags & CQ_ENET_RQ_DESC_FLAGS_SOP) ?
+		1 : 0;
+
+	*rss_type = (u8)((q_number_rss_type_flags >> CQ_DESC_Q_NUM_BITS) &
+		CQ_ENET_RQ_DESC_RSS_TYPE_MASK);
+	*csum_not_calc = (q_number_rss_type_flags &
+		CQ_ENET_RQ_DESC_FLAGS_CSUM_NOT_CALC) ? 1 : 0;
+
+	*rss_hash = le32_to_cpu(desc->rss_hash);
+
+	*bytes_written = bytes_written_flags &
+		CQ_ENET_RQ_DESC_BYTES_WRITTEN_MASK;
+	*packet_error = (bytes_written_flags &
+		CQ_ENET_RQ_DESC_FLAGS_TRUNCATED) ? 1 : 0;
+	*vlan_stripped = (bytes_written_flags &
+		CQ_ENET_RQ_DESC_FLAGS_VLAN_STRIPPED) ? 1 : 0;
+
+	*vlan = le16_to_cpu(desc->vlan);
+
+	if (*fcoe) {
+		*fcoe_sof = (u8)(le16_to_cpu(desc->checksum_fcoe) &
+			CQ_ENET_RQ_DESC_FCOE_SOF_MASK);
+		*fcoe_fc_crc_ok = (desc->flags &
+			CQ_ENET_RQ_DESC_FCOE_FC_CRC_OK) ? 1 : 0;
+		*fcoe_enc_error = (desc->flags &
+			CQ_ENET_RQ_DESC_FCOE_ENC_ERROR) ? 1 : 0;
+		*fcoe_eof = (u8)((desc->checksum_fcoe >>
+			CQ_ENET_RQ_DESC_FCOE_EOF_SHIFT) &
+			CQ_ENET_RQ_DESC_FCOE_EOF_MASK);
+		*checksum = 0;
+	} else {
+		*fcoe_sof = 0;
+		*fcoe_fc_crc_ok = 0;
+		*fcoe_enc_error = 0;
+		*fcoe_eof = 0;
+		*checksum = le16_to_cpu(desc->checksum_fcoe);
+	}
+
+	*tcp_udp_csum_ok =
+		(desc->flags & CQ_ENET_RQ_DESC_FLAGS_TCP_UDP_CSUM_OK) ? 1 : 0;
+	*udp = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_UDP) ? 1 : 0;
+	*tcp = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_TCP) ? 1 : 0;
+	*ipv4_csum_ok =
+		(desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4_CSUM_OK) ? 1 : 0;
+	*ipv6 = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV6) ? 1 : 0;
+	*ipv4 = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4) ? 1 : 0;
+	*ipv4_fragment =
+		(desc->flags & CQ_ENET_RQ_DESC_FLAGS_IPV4_FRAGMENT) ? 1 : 0;
+	*fcs_ok = (desc->flags & CQ_ENET_RQ_DESC_FLAGS_FCS_OK) ? 1 : 0;
+}
+
+#endif /* _CQ_ENET_DESC_H_ */

drivers/scsi/fnic/cq_exch_desc.h

@@ -0,0 +1,182 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _CQ_EXCH_DESC_H_
+#define _CQ_EXCH_DESC_H_
+
+#include "cq_desc.h"
+
+/* Exchange completion queue descriptor: 16B */
+struct cq_exch_wq_desc {
+	u16 completed_index;
+	u16 q_number;
+	u16 exchange_id;
+	u8  tmpl;
+	u8  reserved0;
+	u32 reserved1;
+	u8  exch_status;
+	u8  reserved2[2];
+	u8  type_color;
+};
+
+#define CQ_EXCH_WQ_STATUS_BITS      2
+#define CQ_EXCH_WQ_STATUS_MASK      ((1 << CQ_EXCH_WQ_STATUS_BITS) - 1)
+
+enum cq_exch_status_types {
+	CQ_EXCH_WQ_STATUS_TYPE_COMPLETE = 0,
+	CQ_EXCH_WQ_STATUS_TYPE_ABORT = 1,
+	CQ_EXCH_WQ_STATUS_TYPE_SGL_EOF = 2,
+	CQ_EXCH_WQ_STATUS_TYPE_TMPL_ERR = 3,
+};
+
+static inline void cq_exch_wq_desc_dec(struct cq_exch_wq_desc *desc_ptr,
+				       u8  *type,
+				       u8  *color,
+				       u16 *q_number,
+				       u16 *completed_index,
+				       u8  *exch_status)
+{
+	cq_desc_dec((struct cq_desc *)desc_ptr, type,
+		    color, q_number, completed_index);
+	*exch_status = desc_ptr->exch_status & CQ_EXCH_WQ_STATUS_MASK;
+}
+
+struct cq_fcp_rq_desc {
+	u16 completed_index_eop_sop_prt;
+	u16 q_number;
+	u16 exchange_id;
+	u16 tmpl;
+	u16 bytes_written;
+	u16 vlan;
+	u8  sof;
+	u8  eof;
+	u8  fcs_fer_fck;
+	u8  type_color;
+};
+
+#define CQ_FCP_RQ_DESC_FLAGS_SOP		(1 << 15)
+#define CQ_FCP_RQ_DESC_FLAGS_EOP		(1 << 14)
+#define CQ_FCP_RQ_DESC_FLAGS_PRT		(1 << 12)
+#define CQ_FCP_RQ_DESC_TMPL_MASK		0x1f
+#define CQ_FCP_RQ_DESC_BYTES_WRITTEN_MASK	0x3fff
+#define CQ_FCP_RQ_DESC_PACKET_ERR_SHIFT		14
+#define CQ_FCP_RQ_DESC_PACKET_ERR_MASK (1 << CQ_FCP_RQ_DESC_PACKET_ERR_SHIFT)
+#define CQ_FCP_RQ_DESC_VS_STRIPPED_SHIFT	15
+#define CQ_FCP_RQ_DESC_VS_STRIPPED_MASK (1 << CQ_FCP_RQ_DESC_VS_STRIPPED_SHIFT)
+#define CQ_FCP_RQ_DESC_FC_CRC_OK_MASK		0x1
+#define CQ_FCP_RQ_DESC_FCOE_ERR_SHIFT		1
+#define CQ_FCP_RQ_DESC_FCOE_ERR_MASK (1 << CQ_FCP_RQ_DESC_FCOE_ERR_SHIFT)
+#define CQ_FCP_RQ_DESC_FCS_OK_SHIFT		7
+#define CQ_FCP_RQ_DESC_FCS_OK_MASK (1 << CQ_FCP_RQ_DESC_FCS_OK_SHIFT)
+
+static inline void cq_fcp_rq_desc_dec(struct cq_fcp_rq_desc *desc_ptr,
+				      u8  *type,
+				      u8  *color,
+				      u16 *q_number,
+				      u16 *completed_index,
+				      u8  *eop,
+				      u8  *sop,
+				      u8  *fck,
+				      u16 *exchange_id,
+				      u16 *tmpl,
+				      u32 *bytes_written,
+				      u8  *sof,
+				      u8  *eof,
+				      u8  *ingress_port,
+				      u8  *packet_err,
+				      u8  *fcoe_err,
+				      u8  *fcs_ok,
+				      u8  *vlan_stripped,
+				      u16 *vlan)
+{
+	cq_desc_dec((struct cq_desc *)desc_ptr, type,
+		    color, q_number, completed_index);
+	*eop = (desc_ptr->completed_index_eop_sop_prt &
+		CQ_FCP_RQ_DESC_FLAGS_EOP) ? 1 : 0;
+	*sop = (desc_ptr->completed_index_eop_sop_prt &
+		CQ_FCP_RQ_DESC_FLAGS_SOP) ? 1 : 0;
+	*ingress_port =
+		(desc_ptr->completed_index_eop_sop_prt &
+		 CQ_FCP_RQ_DESC_FLAGS_PRT) ? 1 : 0;
+	*exchange_id = desc_ptr->exchange_id;
+	*tmpl = desc_ptr->tmpl & CQ_FCP_RQ_DESC_TMPL_MASK;
+	*bytes_written =
+		desc_ptr->bytes_written & CQ_FCP_RQ_DESC_BYTES_WRITTEN_MASK;
+	*packet_err =
+		(desc_ptr->bytes_written & CQ_FCP_RQ_DESC_PACKET_ERR_MASK) >>
+		CQ_FCP_RQ_DESC_PACKET_ERR_SHIFT;
+	*vlan_stripped =
+		(desc_ptr->bytes_written & CQ_FCP_RQ_DESC_VS_STRIPPED_MASK) >>
+		CQ_FCP_RQ_DESC_VS_STRIPPED_SHIFT;
+	*vlan = desc_ptr->vlan;
+	*sof = desc_ptr->sof;
+	*fck = desc_ptr->fcs_fer_fck & CQ_FCP_RQ_DESC_FC_CRC_OK_MASK;
+	*fcoe_err = (desc_ptr->fcs_fer_fck & CQ_FCP_RQ_DESC_FCOE_ERR_MASK) >>
+		CQ_FCP_RQ_DESC_FCOE_ERR_SHIFT;
+	*eof = desc_ptr->eof;
+	*fcs_ok =
+		(desc_ptr->fcs_fer_fck & CQ_FCP_RQ_DESC_FCS_OK_MASK) >>
+		CQ_FCP_RQ_DESC_FCS_OK_SHIFT;
+}
+
+struct cq_sgl_desc {
+	u16 exchange_id;
+	u16 q_number;
+	u32 active_burst_offset;
+	u32 tot_data_bytes;
+	u16 tmpl;
+	u8  sgl_err;
+	u8  type_color;
+};
+
+enum cq_sgl_err_types {
+	CQ_SGL_ERR_NO_ERROR = 0,
+	CQ_SGL_ERR_OVERFLOW,         /* data ran beyond end of SGL */
+	CQ_SGL_ERR_SGL_LCL_ADDR_ERR, /* sgl access to local vnic addr illegal*/
+	CQ_SGL_ERR_ADDR_RSP_ERR,     /* sgl address error */
+	CQ_SGL_ERR_DATA_RSP_ERR,     /* sgl data rsp error */
+	CQ_SGL_ERR_CNT_ZERO_ERR,     /* SGL count is 0 */
+	CQ_SGL_ERR_CNT_MAX_ERR,      /* SGL count is larger than supported */
+	CQ_SGL_ERR_ORDER_ERR,        /* frames recv on both ports, order err */
+	CQ_SGL_ERR_DATA_LCL_ADDR_ERR,/* sgl data buf to local vnic addr ill */
+	CQ_SGL_ERR_HOST_CQ_ERR,      /* host cq entry to local vnic addr ill */
+};
+
+#define CQ_SGL_SGL_ERR_MASK             0x1f
+#define CQ_SGL_TMPL_MASK                0x1f
+
+static inline void cq_sgl_desc_dec(struct cq_sgl_desc *desc_ptr,
+				   u8  *type,
+				   u8  *color,
+				   u16 *q_number,
+				   u16 *exchange_id,
+				   u32 *active_burst_offset,
+				   u32 *tot_data_bytes,
+				   u16 *tmpl,
+				   u8  *sgl_err)
+{
+	/* Cheat a little by assuming exchange_id is the same as completed
+	   index */
+	cq_desc_dec((struct cq_desc *)desc_ptr, type, color, q_number,
+		    exchange_id);
+	*active_burst_offset = desc_ptr->active_burst_offset;
+	*tot_data_bytes = desc_ptr->tot_data_bytes;
+	*tmpl = desc_ptr->tmpl & CQ_SGL_TMPL_MASK;
+	*sgl_err = desc_ptr->sgl_err & CQ_SGL_SGL_ERR_MASK;
+}
+
+#endif /* _CQ_EXCH_DESC_H_ */

drivers/scsi/fnic/fcpio.h

@@ -0,0 +1,780 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _FCPIO_H_
+#define _FCPIO_H_
+
+#include <linux/if_ether.h>
+
+/*
+ * This header file includes all of the data structures used for
+ * communication by the host driver to the fcp firmware.
+ */
+
+/*
+ * Exchange and sequence id space allocated to the host driver
+ */
+#define FCPIO_HOST_EXCH_RANGE_START         0x1000
+#define FCPIO_HOST_EXCH_RANGE_END           0x1fff
+#define FCPIO_HOST_SEQ_ID_RANGE_START       0x80
+#define FCPIO_HOST_SEQ_ID_RANGE_END         0xff
+
+/*
+ * Command entry type
+ */
+enum fcpio_type {
+	/*
+	 * Initiator request types
+	 */
+	FCPIO_ICMND_16 = 0x1,
+	FCPIO_ICMND_32,
+	FCPIO_ICMND_CMPL,
+	FCPIO_ITMF,
+	FCPIO_ITMF_CMPL,
+
+	/*
+	 * Target request types
+	 */
+	FCPIO_TCMND_16 = 0x11,
+	FCPIO_TCMND_32,
+	FCPIO_TDATA,
+	FCPIO_TXRDY,
+	FCPIO_TRSP,
+	FCPIO_TDRSP_CMPL,
+	FCPIO_TTMF,
+	FCPIO_TTMF_ACK,
+	FCPIO_TABORT,
+	FCPIO_TABORT_CMPL,
+
+	/*
+	 * Misc request types
+	 */
+	FCPIO_ACK = 0x20,
+	FCPIO_RESET,
+	FCPIO_RESET_CMPL,
+	FCPIO_FLOGI_REG,
+	FCPIO_FLOGI_REG_CMPL,
+	FCPIO_ECHO,
+	FCPIO_ECHO_CMPL,
+	FCPIO_LUNMAP_CHNG,
+	FCPIO_LUNMAP_REQ,
+	FCPIO_LUNMAP_REQ_CMPL,
+	FCPIO_FLOGI_FIP_REG,
+	FCPIO_FLOGI_FIP_REG_CMPL,
+};
+
+/*
+ * Header status codes from the firmware
+ */
+enum fcpio_status {
+	FCPIO_SUCCESS = 0,              /* request was successful */
+
+	/*
+	 * If a request to the firmware is rejected, the original request
+	 * header will be returned with the status set to one of the following:
+	 */
+	FCPIO_INVALID_HEADER,    /* header contains invalid data */
+	FCPIO_OUT_OF_RESOURCE,   /* out of resources to complete request */
+	FCPIO_INVALID_PARAM,     /* some parameter in request is invalid */
+	FCPIO_REQ_NOT_SUPPORTED, /* request type is not supported */
+	FCPIO_IO_NOT_FOUND,      /* requested I/O was not found */
+
+	/*
+	 * Once a request is processed, the firmware will usually return
+	 * a cmpl message type.  In cases where errors occurred,
+	 * the header status field would be filled in with one of the following:
+	 */
+	FCPIO_ABORTED = 0x41,     /* request was aborted */
+	FCPIO_TIMEOUT,            /* request was timed out */
+	FCPIO_SGL_INVALID,        /* request was aborted due to sgl error */
+	FCPIO_MSS_INVALID,        /* request was aborted due to mss error */
+	FCPIO_DATA_CNT_MISMATCH,  /* recv/sent more/less data than exp. */
+	FCPIO_FW_ERR,             /* request was terminated due to fw error */
+	FCPIO_ITMF_REJECTED,      /* itmf req was rejected by remote node */
+	FCPIO_ITMF_FAILED,        /* itmf req was failed by remote node */
+	FCPIO_ITMF_INCORRECT_LUN, /* itmf req targeted incorrect LUN */
+	FCPIO_CMND_REJECTED,      /* request was invalid and rejected */
+	FCPIO_NO_PATH_AVAIL,      /* no paths to the lun was available */
+	FCPIO_PATH_FAILED,        /* i/o sent to current path failed */
+	FCPIO_LUNMAP_CHNG_PEND,   /* i/o rejected due to lunmap change */
+};
+
+/*
+ * The header command tag.  All host requests will use the "tag" field
+ * to mark commands with a unique tag.  When the firmware responds to
+ * a host request, it will copy the tag field into the response.
+ *
+ * The only firmware requests that will use the rx_id/ox_id fields instead
+ * of the tag field will be the target command and target task management
+ * requests.  These two requests do not have corresponding host requests
+ * since they come directly from the FC initiator on the network.
+ */
+struct fcpio_tag {
+	union {
+		u32 req_id;
+		struct {
+			u16 rx_id;
+			u16 ox_id;
+		} ex_id;
+	} u;
+};
+
+static inline void
+fcpio_tag_id_enc(struct fcpio_tag *tag, u32 id)
+{
+	tag->u.req_id = id;
+}
+
+static inline void
+fcpio_tag_id_dec(struct fcpio_tag *tag, u32 *id)
+{
+	*id = tag->u.req_id;
+}
+
+static inline void
+fcpio_tag_exid_enc(struct fcpio_tag *tag, u16 ox_id, u16 rx_id)
+{
+	tag->u.ex_id.rx_id = rx_id;
+	tag->u.ex_id.ox_id = ox_id;
+}
+
+static inline void
+fcpio_tag_exid_dec(struct fcpio_tag *tag, u16 *ox_id, u16 *rx_id)
+{
+	*rx_id = tag->u.ex_id.rx_id;
+	*ox_id = tag->u.ex_id.ox_id;
+}
+
+/*
+ * The header for an fcpio request, whether from the firmware or from the
+ * host driver
+ */
+struct fcpio_header {
+	u8            type;           /* enum fcpio_type */
+	u8            status;         /* header status entry */
+	u16           _resvd;         /* reserved */
+	struct fcpio_tag    tag;      /* header tag */
+};
+
+static inline void
+fcpio_header_enc(struct fcpio_header *hdr,
+		 u8 type, u8 status,
+		 struct fcpio_tag tag)
+{
+	hdr->type = type;
+	hdr->status = status;
+	hdr->_resvd = 0;
+	hdr->tag = tag;
+}
+
+static inline void
+fcpio_header_dec(struct fcpio_header *hdr,
+		 u8 *type, u8 *status,
+		 struct fcpio_tag *tag)
+{
+	*type = hdr->type;
+	*status = hdr->status;
+	*tag = hdr->tag;
+}
+
+#define CDB_16      16
+#define CDB_32      32
+#define LUN_ADDRESS 8
+
+/*
+ * fcpio_icmnd_16: host -> firmware request
+ *
+ * used for sending out an initiator SCSI 16-byte command
+ */
+struct fcpio_icmnd_16 {
+	u32	  lunmap_id;		/* index into lunmap table */
+	u8	  special_req_flags;	/* special exchange request flags */
+	u8	  _resvd0[3];	        /* reserved */
+	u32	  sgl_cnt;		/* scatter-gather list count */
+	u32	  sense_len;		/* sense buffer length */
+	u64	  sgl_addr;		/* scatter-gather list addr */
+	u64	  sense_addr;		/* sense buffer address */
+	u8	  crn;			/* SCSI Command Reference No. */
+	u8	  pri_ta;		/* SCSI Priority and Task attribute */
+	u8	  _resvd1;		/* reserved: should be 0 */
+	u8	  flags;		/* command flags */
+	u8	  scsi_cdb[CDB_16];	/* SCSI Cmnd Descriptor Block */
+	u32	  data_len;		/* length of data expected */
+	u8	  lun[LUN_ADDRESS];	/* FC vNIC only: LUN address */
+	u8	  _resvd2;		/* reserved */
+	u8	  d_id[3];		/* FC vNIC only: Target D_ID */
+	u16	  mss;			/* FC vNIC only: max burst */
+	u16	  _resvd3;		/* reserved */
+	u32	  r_a_tov;		/* FC vNIC only: Res. Alloc Timeout */
+	u32	  e_d_tov;	        /* FC vNIC only: Err Detect Timeout */
+};
+
+/*
+ * Special request flags
+ */
+#define FCPIO_ICMND_SRFLAG_RETRY 0x01   /* Enable Retry handling on exchange */
+
+/*
+ * Priority/Task Attribute settings
+ */
+#define FCPIO_ICMND_PTA_SIMPLE      0   /* simple task attribute */
+#define FCPIO_ICMND_PTA_HEADQ       1   /* head of queue task attribute */
+#define FCPIO_ICMND_PTA_ORDERED     2   /* ordered task attribute */
+#define FCPIO_ICMND_PTA_ACA         4   /* auto contingent allegiance */
+#define FCPIO_ICMND_PRI_SHIFT       3   /* priority field starts in bit 3 */
+
+/*
+ * Command flags
+ */
+#define FCPIO_ICMND_RDDATA      0x02    /* read data */
+#define FCPIO_ICMND_WRDATA      0x01    /* write data */
+
+/*
+ * fcpio_icmnd_32: host -> firmware request
+ *
+ * used for sending out an initiator SCSI 32-byte command
+ */
+struct fcpio_icmnd_32 {
+	u32   lunmap_id;              /* index into lunmap table */
+	u8    special_req_flags;      /* special exchange request flags */
+	u8    _resvd0[3];             /* reserved */
+	u32   sgl_cnt;                /* scatter-gather list count */
+	u32   sense_len;              /* sense buffer length */
+	u64   sgl_addr;               /* scatter-gather list addr */
+	u64   sense_addr;             /* sense buffer address */
+	u8    crn;                    /* SCSI Command Reference No. */
+	u8    pri_ta;                 /* SCSI Priority and Task attribute */
+	u8    _resvd1;                /* reserved: should be 0 */
+	u8    flags;                  /* command flags */
+	u8    scsi_cdb[CDB_32];       /* SCSI Cmnd Descriptor Block */
+	u32   data_len;               /* length of data expected */
+	u8    lun[LUN_ADDRESS];       /* FC vNIC only: LUN address */
+	u8    _resvd2;                /* reserved */
+	u8    d_id[3];		      /* FC vNIC only: Target D_ID */
+	u16   mss;                    /* FC vNIC only: max burst */
+	u16   _resvd3;                /* reserved */
+	u32   r_a_tov;                /* FC vNIC only: Res. Alloc Timeout */
+	u32   e_d_tov;                /* FC vNIC only: Error Detect Timeout */
+};
+
+/*
+ * fcpio_itmf: host -> firmware request
+ *
+ * used for requesting the firmware to abort a request and/or send out
+ * a task management function
+ *
+ * The t_tag field is only needed when the request type is ABT_TASK.
+ */
+struct fcpio_itmf {
+	u32   lunmap_id;              /* index into lunmap table */
+	u32   tm_req;                 /* SCSI Task Management request */
+	u32   t_tag;                  /* header tag of fcpio to be aborted */
+	u32   _resvd;                 /* _reserved */
+	u8    lun[LUN_ADDRESS];       /* FC vNIC only: LUN address */
+	u8    _resvd1;                /* reserved */
+	u8    d_id[3];		      /* FC vNIC only: Target D_ID */
+	u32   r_a_tov;                /* FC vNIC only: R_A_TOV in msec */
+	u32   e_d_tov;                /* FC vNIC only: E_D_TOV in msec */
+};
+
+/*
+ * Task Management request
+ */
+enum fcpio_itmf_tm_req_type {
+	FCPIO_ITMF_ABT_TASK_TERM = 0x01,    /* abort task and terminate */
+	FCPIO_ITMF_ABT_TASK,                /* abort task and issue abts */
+	FCPIO_ITMF_ABT_TASK_SET,            /* abort task set */
+	FCPIO_ITMF_CLR_TASK_SET,            /* clear task set */
+	FCPIO_ITMF_LUN_RESET,               /* logical unit reset task mgmt */
+	FCPIO_ITMF_CLR_ACA,                 /* Clear ACA condition */
+};
+
+/*
+ * fcpio_tdata: host -> firmware request
+ *
+ * used for requesting the firmware to send out a read data transfer for a
+ * target command
+ */
+struct fcpio_tdata {
+	u16   rx_id;                  /* FC rx_id of target command */
+	u16   flags;                  /* command flags */
+	u32   rel_offset;             /* data sequence relative offset */
+	u32   sgl_cnt;                /* scatter-gather list count */
+	u32   data_len;               /* length of data expected to send */
+	u64   sgl_addr;               /* scatter-gather list address */
+};
+
+/*
+ * Command flags
+ */
+#define FCPIO_TDATA_SCSI_RSP    0x01    /* send a scsi resp. after last frame */
+
+/*
+ * fcpio_txrdy: host -> firmware request
+ *
+ * used for requesting the firmware to send out a write data transfer for a
+ * target command
+ */
+struct fcpio_txrdy {
+	u16   rx_id;                  /* FC rx_id of target command */
+	u16   _resvd0;                /* reserved */
+	u32   rel_offset;             /* data sequence relative offset */
+	u32   sgl_cnt;                /* scatter-gather list count */
+	u32   data_len;               /* length of data expected to send */
+	u64   sgl_addr;               /* scatter-gather list address */
+};
+
+/*
+ * fcpio_trsp: host -> firmware request
+ *
+ * used for requesting the firmware to send out a response for a target
+ * command
+ */
+struct fcpio_trsp {
+	u16   rx_id;                  /* FC rx_id of target command */
+	u16   _resvd0;                /* reserved */
+	u32   sense_len;              /* sense data buffer length */
+	u64   sense_addr;             /* sense data buffer address */
+	u16   _resvd1;                /* reserved */
+	u8    flags;                  /* response request flags */
+	u8    scsi_status;            /* SCSI status */
+	u32   residual;               /* SCSI data residual value of I/O */
+};
+
+/*
+ * resposnse request flags
+ */
+#define FCPIO_TRSP_RESID_UNDER  0x08   /* residual is valid and is underflow */
+#define FCPIO_TRSP_RESID_OVER   0x04   /* residual is valid and is overflow */
+
+/*
+ * fcpio_ttmf_ack: host -> firmware response
+ *
+ * used by the host to indicate to the firmware it has received and processed
+ * the target tmf request
+ */
+struct fcpio_ttmf_ack {
+	u16   rx_id;                  /* FC rx_id of target command */
+	u16   _resvd0;                /* reserved */
+	u32   tmf_status;             /* SCSI task management status */
+};
+
+/*
+ * fcpio_tabort: host -> firmware request
+ *
+ * used by the host to request the firmware to abort a target request that was
+ * received by the firmware
+ */
+struct fcpio_tabort {
+	u16   rx_id;                  /* rx_id of the target request */
+};
+
+/*
+ * fcpio_reset: host -> firmware request
+ *
+ * used by the host to signal a reset of the driver to the firmware
+ * and to request firmware to clean up all outstanding I/O
+ */
+struct fcpio_reset {
+	u32   _resvd;
+};
+
+enum fcpio_flogi_reg_format_type {
+	FCPIO_FLOGI_REG_DEF_DEST = 0,    /* Use the oui | s_id mac format */
+	FCPIO_FLOGI_REG_GW_DEST,         /* Use the fixed gateway mac */
+};
+
+/*
+ * fcpio_flogi_reg: host -> firmware request
+ *
+ * fc vnic only
+ * used by the host to notify the firmware of the lif's s_id
+ * and destination mac address format
+ */
+struct fcpio_flogi_reg {
+	u8 format;
+	u8 s_id[3];			/* FC vNIC only: Source S_ID */
+	u8 gateway_mac[ETH_ALEN];	/* Destination gateway mac */
+	u16 _resvd;
+	u32 r_a_tov;			/* R_A_TOV in msec */
+	u32 e_d_tov;			/* E_D_TOV in msec */
+};
+
+/*
+ * fcpio_echo: host -> firmware request
+ *
+ * sends a heartbeat echo request to the firmware
+ */
+struct fcpio_echo {
+	u32 _resvd;
+};
+
+/*
+ * fcpio_lunmap_req: host -> firmware request
+ *
+ * scsi vnic only
+ * sends a request to retrieve the lunmap table for scsi vnics
+ */
+struct fcpio_lunmap_req {
+	u64 addr;                     /* address of the buffer */
+	u32 len;                      /* len of the buffer */
+};
+
+/*
+ * fcpio_flogi_fip_reg: host -> firmware request
+ *
+ * fc vnic only
+ * used by the host to notify the firmware of the lif's s_id
+ * and destination mac address format
+ */
+struct fcpio_flogi_fip_reg {
+	u8    _resvd0;
+	u8     s_id[3];               /* FC vNIC only: Source S_ID */
+	u8     fcf_mac[ETH_ALEN];     /* FCF Target destination mac */
+	u16   _resvd1;
+	u32   r_a_tov;                /* R_A_TOV in msec */
+	u32   e_d_tov;                /* E_D_TOV in msec */
+	u8    ha_mac[ETH_ALEN];       /* Host adapter source mac */
+	u16   _resvd2;
+};
+
+/*
+ * Basic structure for all fcpio structures that are sent from the host to the
+ * firmware.  They are 128 bytes per structure.
+ */
+#define FCPIO_HOST_REQ_LEN      128     /* expected length of host requests */
+
+struct fcpio_host_req {
+	struct fcpio_header hdr;
+
+	union {
+		/*
+		 * Defines space needed for request
+		 */
+		u8 buf[FCPIO_HOST_REQ_LEN - sizeof(struct fcpio_header)];
+
+		/*
+		 * Initiator host requests
+		 */
+		struct fcpio_icmnd_16               icmnd_16;
+		struct fcpio_icmnd_32               icmnd_32;
+		struct fcpio_itmf                   itmf;
+
+		/*
+		 * Target host requests
+		 */
+		struct fcpio_tdata                  tdata;
+		struct fcpio_txrdy                  txrdy;
+		struct fcpio_trsp                   trsp;
+		struct fcpio_ttmf_ack               ttmf_ack;
+		struct fcpio_tabort                 tabort;
+
+		/*
+		 * Misc requests
+		 */
+		struct fcpio_reset                  reset;
+		struct fcpio_flogi_reg              flogi_reg;
+		struct fcpio_echo                   echo;
+		struct fcpio_lunmap_req             lunmap_req;
+		struct fcpio_flogi_fip_reg          flogi_fip_reg;
+	} u;
+};
+
+/*
+ * fcpio_icmnd_cmpl: firmware -> host response
+ *
+ * used for sending the host a response to an initiator command
+ */
+struct fcpio_icmnd_cmpl {
+	u8    _resvd0[6];             /* reserved */
+	u8    flags;                  /* response flags */
+	u8    scsi_status;            /* SCSI status */
+	u32   residual;               /* SCSI data residual length */
+	u32   sense_len;              /* SCSI sense length */
+};
+
+/*
+ * response flags
+ */
+#define FCPIO_ICMND_CMPL_RESID_UNDER    0x08    /* resid under and valid */
+#define FCPIO_ICMND_CMPL_RESID_OVER     0x04    /* resid over and valid */
+
+/*
+ * fcpio_itmf_cmpl: firmware -> host response
+ *
+ * used for sending the host a response for a itmf request
+ */
+struct fcpio_itmf_cmpl {
+	u32    _resvd;                /* reserved */
+};
+
+/*
+ * fcpio_tcmnd_16: firmware -> host request
+ *
+ * used by the firmware to notify the host of an incoming target SCSI 16-Byte
+ * request
+ */
+struct fcpio_tcmnd_16 {
+	u8    lun[LUN_ADDRESS];       /* FC vNIC only: LUN address */
+	u8    crn;                    /* SCSI Command Reference No. */
+	u8    pri_ta;                 /* SCSI Priority and Task attribute */
+	u8    _resvd2;                /* reserved: should be 0 */
+	u8    flags;                  /* command flags */
+	u8    scsi_cdb[CDB_16];       /* SCSI Cmnd Descriptor Block */
+	u32   data_len;               /* length of data expected */
+	u8    _resvd1;                /* reserved */
+	u8    s_id[3];		      /* FC vNIC only: Source S_ID */
+};
+
+/*
+ * Priority/Task Attribute settings
+ */
+#define FCPIO_TCMND_PTA_SIMPLE      0   /* simple task attribute */
+#define FCPIO_TCMND_PTA_HEADQ       1   /* head of queue task attribute */
+#define FCPIO_TCMND_PTA_ORDERED     2   /* ordered task attribute */
+#define FCPIO_TCMND_PTA_ACA         4   /* auto contingent allegiance */
+#define FCPIO_TCMND_PRI_SHIFT       3   /* priority field starts in bit 3 */
+
+/*
+ * Command flags
+ */
+#define FCPIO_TCMND_RDDATA      0x02    /* read data */
+#define FCPIO_TCMND_WRDATA      0x01    /* write data */
+
+/*
+ * fcpio_tcmnd_32: firmware -> host request
+ *
+ * used by the firmware to notify the host of an incoming target SCSI 32-Byte
+ * request
+ */
+struct fcpio_tcmnd_32 {
+	u8    lun[LUN_ADDRESS];       /* FC vNIC only: LUN address */
+	u8    crn;                    /* SCSI Command Reference No. */
+	u8    pri_ta;                 /* SCSI Priority and Task attribute */
+	u8    _resvd2;                /* reserved: should be 0 */
+	u8    flags;                  /* command flags */
+	u8    scsi_cdb[CDB_32];       /* SCSI Cmnd Descriptor Block */
+	u32   data_len;               /* length of data expected */
+	u8    _resvd0;                /* reserved */
+	u8    s_id[3];		      /* FC vNIC only: Source S_ID */
+};
+
+/*
+ * fcpio_tdrsp_cmpl: firmware -> host response
+ *
+ * used by the firmware to notify the host of a response to a host target
+ * command
+ */
+struct fcpio_tdrsp_cmpl {
+	u16   rx_id;                  /* rx_id of the target request */
+	u16   _resvd0;                /* reserved */
+};
+
+/*
+ * fcpio_ttmf: firmware -> host request
+ *
+ * used by the firmware to notify the host of an incoming task management
+ * function request
+ */
+struct fcpio_ttmf {
+	u8    _resvd0;                /* reserved */
+	u8    s_id[3];		      /* FC vNIC only: Source S_ID */
+	u8    lun[LUN_ADDRESS];       /* FC vNIC only: LUN address */
+	u8    crn;                    /* SCSI Command Reference No. */
+	u8    _resvd2[3];             /* reserved */
+	u32   tmf_type;               /* task management request type */
+};
+
+/*
+ * Task Management request
+ */
+#define FCPIO_TTMF_CLR_ACA      0x40    /* Clear ACA condition */
+#define FCPIO_TTMF_LUN_RESET    0x10    /* logical unit reset task mgmt */
+#define FCPIO_TTMF_CLR_TASK_SET 0x04    /* clear task set */
+#define FCPIO_TTMF_ABT_TASK_SET 0x02    /* abort task set */
+#define FCPIO_TTMF_ABT_TASK     0x01    /* abort task */
+
+/*
+ * fcpio_tabort_cmpl: firmware -> host response
+ *
+ * used by the firmware to respond to a host's tabort request
+ */
+struct fcpio_tabort_cmpl {
+	u16   rx_id;                  /* rx_id of the target request */
+	u16   _resvd0;                /* reserved */
+};
+
+/*
+ * fcpio_ack: firmware -> host response
+ *
+ * used by firmware to notify the host of the last work request received
+ */
+struct fcpio_ack {
+	u16  request_out;             /* last host entry received */
+	u16  _resvd;
+};
+
+/*
+ * fcpio_reset_cmpl: firmware -> host response
+ *
+ * use by firmware to respond to the host's reset request
+ */
+struct fcpio_reset_cmpl {
+	u16   vnic_id;
+};
+
+/*
+ * fcpio_flogi_reg_cmpl: firmware -> host response
+ *
+ * fc vnic only
+ * response to the fcpio_flogi_reg request
+ */
+struct fcpio_flogi_reg_cmpl {
+	u32 _resvd;
+};
+
+/*
+ * fcpio_echo_cmpl: firmware -> host response
+ *
+ * response to the fcpio_echo request
+ */
+struct fcpio_echo_cmpl {
+	u32 _resvd;
+};
+
+/*
+ * fcpio_lunmap_chng: firmware -> host notification
+ *
+ * scsi vnic only
+ * notifies the host that the lunmap tables have changed
+ */
+struct fcpio_lunmap_chng {
+	u32 _resvd;
+};
+
+/*
+ * fcpio_lunmap_req_cmpl: firmware -> host response
+ *
+ * scsi vnic only
+ * response for lunmap table request from the host
+ */
+struct fcpio_lunmap_req_cmpl {
+	u32 _resvd;
+};
+
+/*
+ * Basic structure for all fcpio structures that are sent from the firmware to
+ * the host.  They are 64 bytes per structure.
+ */
+#define FCPIO_FW_REQ_LEN        64      /* expected length of fw requests */
+struct fcpio_fw_req {
+	struct fcpio_header hdr;
+
+	union {
+		/*
+		 * Defines space needed for request
+		 */
+		u8 buf[FCPIO_FW_REQ_LEN - sizeof(struct fcpio_header)];
+
+		/*
+		 * Initiator firmware responses
+		 */
+		struct fcpio_icmnd_cmpl         icmnd_cmpl;
+		struct fcpio_itmf_cmpl          itmf_cmpl;
+
+		/*
+		 * Target firmware new requests
+		 */
+		struct fcpio_tcmnd_16           tcmnd_16;
+		struct fcpio_tcmnd_32           tcmnd_32;
+
+		/*
+		 * Target firmware responses
+		 */
+		struct fcpio_tdrsp_cmpl         tdrsp_cmpl;
+		struct fcpio_ttmf               ttmf;
+		struct fcpio_tabort_cmpl        tabort_cmpl;
+
+		/*
+		 * Firmware response to work received
+		 */
+		struct fcpio_ack                ack;
+
+		/*
+		 * Misc requests
+		 */
+		struct fcpio_reset_cmpl         reset_cmpl;
+		struct fcpio_flogi_reg_cmpl     flogi_reg_cmpl;
+		struct fcpio_echo_cmpl          echo_cmpl;
+		struct fcpio_lunmap_chng        lunmap_chng;
+		struct fcpio_lunmap_req_cmpl    lunmap_req_cmpl;
+	} u;
+};
+
+/*
+ * Access routines to encode and decode the color bit, which is the most
+ * significant bit of the MSB of the structure
+ */
+static inline void fcpio_color_enc(struct fcpio_fw_req *fw_req, u8 color)
+{
+	u8 *c = ((u8 *) fw_req) + sizeof(struct fcpio_fw_req) - 1;
+
+	if (color)
+		*c |= 0x80;
+	else
+		*c &= ~0x80;
+}
+
+static inline void fcpio_color_dec(struct fcpio_fw_req *fw_req, u8 *color)
+{
+	u8 *c = ((u8 *) fw_req) + sizeof(struct fcpio_fw_req) - 1;
+
+	*color = *c >> 7;
+
+	/*
+	 * Make sure color bit is read from desc *before* other fields
+	 * are read from desc.  Hardware guarantees color bit is last
+	 * bit (byte) written.  Adding the rmb() prevents the compiler
+	 * and/or CPU from reordering the reads which would potentially
+	 * result in reading stale values.
+	 */
+
+	rmb();
+
+}
+
+/*
+ * Lunmap table entry for scsi vnics
+ */
+#define FCPIO_LUNMAP_TABLE_SIZE     256
+#define FCPIO_FLAGS_LUNMAP_VALID    0x80
+#define FCPIO_FLAGS_BOOT            0x01
+struct fcpio_lunmap_entry {
+	u8    bus;
+	u8    target;
+	u8    lun;
+	u8    path_cnt;
+	u16   flags;
+	u16   update_cnt;
+};
+
+struct fcpio_lunmap_tbl {
+	u32                   update_cnt;
+	struct fcpio_lunmap_entry   lunmaps[FCPIO_LUNMAP_TABLE_SIZE];
+};
+
+#endif /* _FCPIO_H_ */

drivers/scsi/fnic/fnic.h

@@ -0,0 +1,265 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _FNIC_H_
+#define _FNIC_H_
+
+#include <linux/interrupt.h>
+#include <linux/netdevice.h>
+#include <linux/workqueue.h>
+#include <scsi/libfc.h>
+#include "fnic_io.h"
+#include "fnic_res.h"
+#include "vnic_dev.h"
+#include "vnic_wq.h"
+#include "vnic_rq.h"
+#include "vnic_cq.h"
+#include "vnic_wq_copy.h"
+#include "vnic_intr.h"
+#include "vnic_stats.h"
+#include "vnic_scsi.h"
+
+#define DRV_NAME		"fnic"
+#define DRV_DESCRIPTION		"Cisco FCoE HBA Driver"
+#define DRV_VERSION		"1.0.0.1121"
+#define PFX			DRV_NAME ": "
+#define DFX                     DRV_NAME "%d: "
+
+#define DESC_CLEAN_LOW_WATERMARK 8
+#define FNIC_MAX_IO_REQ		2048 /* scsi_cmnd tag map entries */
+#define	FNIC_IO_LOCKS		64 /* IO locks: power of 2 */
+#define FNIC_DFLT_QUEUE_DEPTH	32
+#define	FNIC_STATS_RATE_LIMIT	4 /* limit rate at which stats are pulled up */
+
+/*
+ * Tag bits used for special requests.
+ */
+#define BIT(nr)			(1UL << (nr))
+#define FNIC_TAG_ABORT		BIT(30)		/* tag bit indicating abort */
+#define FNIC_TAG_DEV_RST	BIT(29)		/* indicates device reset */
+#define FNIC_TAG_MASK		(BIT(24) - 1)	/* mask for lookup */
+#define FNIC_NO_TAG             -1
+
+/*
+ * Usage of the scsi_cmnd scratchpad.
+ * These fields are locked by the hashed io_req_lock.
+ */
+#define CMD_SP(Cmnd)		((Cmnd)->SCp.ptr)
+#define CMD_STATE(Cmnd)		((Cmnd)->SCp.phase)
+#define CMD_ABTS_STATUS(Cmnd)	((Cmnd)->SCp.Message)
+#define CMD_LR_STATUS(Cmnd)	((Cmnd)->SCp.have_data_in)
+#define CMD_TAG(Cmnd)           ((Cmnd)->SCp.sent_command)
+
+#define FCPIO_INVALID_CODE 0x100 /* hdr_status value unused by firmware */
+
+#define FNIC_LUN_RESET_TIMEOUT	     10000	/* mSec */
+#define FNIC_HOST_RESET_TIMEOUT	     10000	/* mSec */
+#define FNIC_RMDEVICE_TIMEOUT        1000       /* mSec */
+#define FNIC_HOST_RESET_SETTLE_TIME  30         /* Sec */
+
+#define FNIC_MAX_FCP_TARGET     256
+
+extern unsigned int fnic_log_level;
+
+#define FNIC_MAIN_LOGGING 0x01
+#define FNIC_FCS_LOGGING 0x02
+#define FNIC_SCSI_LOGGING 0x04
+#define FNIC_ISR_LOGGING 0x08
+
+#define FNIC_CHECK_LOGGING(LEVEL, CMD)				\
+do {								\
+	if (unlikely(fnic_log_level & LEVEL))			\
+		do {						\
+			CMD;					\
+		} while (0);					\
+} while (0)
+
+#define FNIC_MAIN_DBG(kern_level, host, fmt, args...)		\
+	FNIC_CHECK_LOGGING(FNIC_MAIN_LOGGING,			\
+			 shost_printk(kern_level, host, fmt, ##args);)
+
+#define FNIC_FCS_DBG(kern_level, host, fmt, args...)		\
+	FNIC_CHECK_LOGGING(FNIC_FCS_LOGGING,			\
+			 shost_printk(kern_level, host, fmt, ##args);)
+
+#define FNIC_SCSI_DBG(kern_level, host, fmt, args...)		\
+	FNIC_CHECK_LOGGING(FNIC_SCSI_LOGGING,			\
+			 shost_printk(kern_level, host, fmt, ##args);)
+
+#define FNIC_ISR_DBG(kern_level, host, fmt, args...)		\
+	FNIC_CHECK_LOGGING(FNIC_ISR_LOGGING,			\
+			 shost_printk(kern_level, host, fmt, ##args);)
+
+extern const char *fnic_state_str[];
+
+enum fnic_intx_intr_index {
+	FNIC_INTX_WQ_RQ_COPYWQ,
+	FNIC_INTX_ERR,
+	FNIC_INTX_NOTIFY,
+	FNIC_INTX_INTR_MAX,
+};
+
+enum fnic_msix_intr_index {
+	FNIC_MSIX_RQ,
+	FNIC_MSIX_WQ,
+	FNIC_MSIX_WQ_COPY,
+	FNIC_MSIX_ERR_NOTIFY,
+	FNIC_MSIX_INTR_MAX,
+};
+
+struct fnic_msix_entry {
+	int requested;
+	char devname[IFNAMSIZ];
+	irqreturn_t (*isr)(int, void *);
+	void *devid;
+};
+
+enum fnic_state {
+	FNIC_IN_FC_MODE = 0,
+	FNIC_IN_FC_TRANS_ETH_MODE,
+	FNIC_IN_ETH_MODE,
+	FNIC_IN_ETH_TRANS_FC_MODE,
+};
+
+#define FNIC_WQ_COPY_MAX 1
+#define FNIC_WQ_MAX 1
+#define FNIC_RQ_MAX 1
+#define FNIC_CQ_MAX (FNIC_WQ_COPY_MAX + FNIC_WQ_MAX + FNIC_RQ_MAX)
+
+struct mempool;
+
+/* Per-instance private data structure */
+struct fnic {
+	struct fc_lport *lport;
+	struct vnic_dev_bar bar0;
+
+	struct msix_entry msix_entry[FNIC_MSIX_INTR_MAX];
+	struct fnic_msix_entry msix[FNIC_MSIX_INTR_MAX];
+
+	struct vnic_stats *stats;
+	unsigned long stats_time;	/* time of stats update */
+	struct vnic_nic_cfg *nic_cfg;
+	char name[IFNAMSIZ];
+	struct timer_list notify_timer; /* used for MSI interrupts */
+
+	unsigned int err_intr_offset;
+	unsigned int link_intr_offset;
+
+	unsigned int wq_count;
+	unsigned int cq_count;
+
+	u32 fcoui_mode:1;		/* use fcoui address*/
+	u32 vlan_hw_insert:1;	        /* let hw insert the tag */
+	u32 in_remove:1;                /* fnic device in removal */
+	u32 stop_rx_link_events:1;      /* stop proc. rx frames, link events */
+
+	struct completion *remove_wait; /* device remove thread blocks */
+
+	struct fc_frame *flogi;
+	struct fc_frame *flogi_resp;
+	u16 flogi_oxid;
+	unsigned long s_id;
+	enum fnic_state state;
+	spinlock_t fnic_lock;
+
+	u16 vlan_id;	                /* VLAN tag including priority */
+	u8 mac_addr[ETH_ALEN];
+	u8 dest_addr[ETH_ALEN];
+	u8 data_src_addr[ETH_ALEN];
+	u64 fcp_input_bytes;		/* internal statistic */
+	u64 fcp_output_bytes;		/* internal statistic */
+	u32 link_down_cnt;
+	int link_status;
+
+	struct list_head list;
+	struct pci_dev *pdev;
+	struct vnic_fc_config config;
+	struct vnic_dev *vdev;
+	unsigned int raw_wq_count;
+	unsigned int wq_copy_count;
+	unsigned int rq_count;
+	int fw_ack_index[FNIC_WQ_COPY_MAX];
+	unsigned short fw_ack_recd[FNIC_WQ_COPY_MAX];
+	unsigned short wq_copy_desc_low[FNIC_WQ_COPY_MAX];
+	unsigned int intr_count;
+	u32 __iomem *legacy_pba;
+	struct fnic_host_tag *tags;
+	mempool_t *io_req_pool;
+	mempool_t *io_sgl_pool[FNIC_SGL_NUM_CACHES];
+	spinlock_t io_req_lock[FNIC_IO_LOCKS];	/* locks for scsi cmnds */
+
+	struct work_struct link_work;
+	struct work_struct frame_work;
+	struct sk_buff_head frame_queue;
+
+	/* copy work queue cache line section */
+	____cacheline_aligned struct vnic_wq_copy wq_copy[FNIC_WQ_COPY_MAX];
+	/* completion queue cache line section */
+	____cacheline_aligned struct vnic_cq cq[FNIC_CQ_MAX];
+
+	spinlock_t wq_copy_lock[FNIC_WQ_COPY_MAX];
+
+	/* work queue cache line section */
+	____cacheline_aligned struct vnic_wq wq[FNIC_WQ_MAX];
+	spinlock_t wq_lock[FNIC_WQ_MAX];
+
+	/* receive queue cache line section */
+	____cacheline_aligned struct vnic_rq rq[FNIC_RQ_MAX];
+
+	/* interrupt resource cache line section */
+	____cacheline_aligned struct vnic_intr intr[FNIC_MSIX_INTR_MAX];
+};
+
+extern struct workqueue_struct *fnic_event_queue;
+extern struct device_attribute *fnic_attrs[];
+
+void fnic_clear_intr_mode(struct fnic *fnic);
+int fnic_set_intr_mode(struct fnic *fnic);
+void fnic_free_intr(struct fnic *fnic);
+int fnic_request_intr(struct fnic *fnic);
+
+int fnic_send(struct fc_lport *, struct fc_frame *);
+void fnic_free_wq_buf(struct vnic_wq *wq, struct vnic_wq_buf *buf);
+void fnic_handle_frame(struct work_struct *work);
+void fnic_handle_link(struct work_struct *work);
+int fnic_rq_cmpl_handler(struct fnic *fnic, int);
+int fnic_alloc_rq_frame(struct vnic_rq *rq);
+void fnic_free_rq_buf(struct vnic_rq *rq, struct vnic_rq_buf *buf);
+int fnic_send_frame(struct fnic *fnic, struct fc_frame *fp);
+
+int fnic_queuecommand(struct scsi_cmnd *, void (*done)(struct scsi_cmnd *));
+int fnic_abort_cmd(struct scsi_cmnd *);
+int fnic_device_reset(struct scsi_cmnd *);
+int fnic_host_reset(struct scsi_cmnd *);
+int fnic_reset(struct Scsi_Host *);
+void fnic_scsi_cleanup(struct fc_lport *);
+void fnic_scsi_abort_io(struct fc_lport *);
+void fnic_empty_scsi_cleanup(struct fc_lport *);
+void fnic_exch_mgr_reset(struct fc_lport *, u32, u32);
+int fnic_wq_copy_cmpl_handler(struct fnic *fnic, int);
+int fnic_wq_cmpl_handler(struct fnic *fnic, int);
+int fnic_flogi_reg_handler(struct fnic *fnic);
+void fnic_wq_copy_cleanup_handler(struct vnic_wq_copy *wq,
+				  struct fcpio_host_req *desc);
+int fnic_fw_reset_handler(struct fnic *fnic);
+void fnic_terminate_rport_io(struct fc_rport *);
+const char *fnic_state_to_str(unsigned int state);
+
+void fnic_log_q_error(struct fnic *fnic);
+void fnic_handle_link_event(struct fnic *fnic);
+
+#endif /* _FNIC_H_ */

drivers/scsi/fnic/fnic_attrs.c

@@ -0,0 +1,56 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/string.h>
+#include <linux/device.h>
+#include <scsi/scsi_host.h>
+#include "fnic.h"
+
+static ssize_t fnic_show_state(struct device *dev,
+			       struct device_attribute *attr, char *buf)
+{
+	struct fc_lport *lp = shost_priv(class_to_shost(dev));
+	struct fnic *fnic = lport_priv(lp);
+
+	return snprintf(buf, PAGE_SIZE, "%s\n", fnic_state_str[fnic->state]);
+}
+
+static ssize_t fnic_show_drv_version(struct device *dev,
+				     struct device_attribute *attr, char *buf)
+{
+	return snprintf(buf, PAGE_SIZE, "%s\n", DRV_VERSION);
+}
+
+static ssize_t fnic_show_link_state(struct device *dev,
+				    struct device_attribute *attr, char *buf)
+{
+	struct fc_lport *lp = shost_priv(class_to_shost(dev));
+
+	return snprintf(buf, PAGE_SIZE, "%s\n", (lp->link_up)
+			? "Link Up" : "Link Down");
+}
+
+static DEVICE_ATTR(fnic_state, S_IRUGO, fnic_show_state, NULL);
+static DEVICE_ATTR(drv_version, S_IRUGO, fnic_show_drv_version, NULL);
+static DEVICE_ATTR(link_state, S_IRUGO, fnic_show_link_state, NULL);
+
+struct device_attribute *fnic_attrs[] = {
+	&dev_attr_fnic_state,
+	&dev_attr_drv_version,
+	&dev_attr_link_state,
+	NULL,
+};

drivers/scsi/fnic/fnic_fcs.c

@@ -0,0 +1,742 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/skbuff.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
+#include <linux/workqueue.h>
+#include <scsi/fc/fc_els.h>
+#include <scsi/fc/fc_fcoe.h>
+#include <scsi/fc_frame.h>
+#include <scsi/libfc.h>
+#include "fnic_io.h"
+#include "fnic.h"
+#include "cq_enet_desc.h"
+#include "cq_exch_desc.h"
+
+struct workqueue_struct *fnic_event_queue;
+
+void fnic_handle_link(struct work_struct *work)
+{
+	struct fnic *fnic = container_of(work, struct fnic, link_work);
+	unsigned long flags;
+	int old_link_status;
+	u32 old_link_down_cnt;
+
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+
+	if (fnic->stop_rx_link_events) {
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		return;
+	}
+
+	old_link_down_cnt = fnic->link_down_cnt;
+	old_link_status = fnic->link_status;
+	fnic->link_status = vnic_dev_link_status(fnic->vdev);
+	fnic->link_down_cnt = vnic_dev_link_down_cnt(fnic->vdev);
+
+	if (old_link_status == fnic->link_status) {
+		if (!fnic->link_status)
+			/* DOWN -> DOWN */
+			spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		else {
+			if (old_link_down_cnt != fnic->link_down_cnt) {
+				/* UP -> DOWN -> UP */
+				fnic->lport->host_stats.link_failure_count++;
+				spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+				FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
+					     "link down\n");
+				fc_linkdown(fnic->lport);
+				FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
+					     "link up\n");
+				fc_linkup(fnic->lport);
+			} else
+				/* UP -> UP */
+				spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		}
+	} else if (fnic->link_status) {
+		/* DOWN -> UP */
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, "link up\n");
+		fc_linkup(fnic->lport);
+	} else {
+		/* UP -> DOWN */
+		fnic->lport->host_stats.link_failure_count++;
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host, "link down\n");
+		fc_linkdown(fnic->lport);
+	}
+
+}
+
+/*
+ * This function passes incoming fabric frames to libFC
+ */
+void fnic_handle_frame(struct work_struct *work)
+{
+	struct fnic *fnic = container_of(work, struct fnic, frame_work);
+	struct fc_lport *lp = fnic->lport;
+	unsigned long flags;
+	struct sk_buff *skb;
+	struct fc_frame *fp;
+
+	while ((skb = skb_dequeue(&fnic->frame_queue))) {
+
+		spin_lock_irqsave(&fnic->fnic_lock, flags);
+		if (fnic->stop_rx_link_events) {
+			spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+			dev_kfree_skb(skb);
+			return;
+		}
+		fp = (struct fc_frame *)skb;
+		/* if Flogi resp frame, register the address */
+		if (fr_flags(fp)) {
+			vnic_dev_add_addr(fnic->vdev,
+					  fnic->data_src_addr);
+			fr_flags(fp) = 0;
+		}
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+		fc_exch_recv(lp, lp->emp, fp);
+	}
+
+}
+
+static inline void fnic_import_rq_fc_frame(struct sk_buff *skb,
+					   u32 len, u8 sof, u8 eof)
+{
+	struct fc_frame *fp = (struct fc_frame *)skb;
+
+	skb_trim(skb, len);
+	fr_eof(fp) = eof;
+	fr_sof(fp) = sof;
+}
+
+
+static inline int fnic_import_rq_eth_pkt(struct sk_buff *skb, u32 len)
+{
+	struct fc_frame *fp;
+	struct ethhdr *eh;
+	struct vlan_ethhdr *vh;
+	struct fcoe_hdr *fcoe_hdr;
+	struct fcoe_crc_eof *ft;
+	u32    transport_len = 0;
+
+	eh = (struct ethhdr *)skb->data;
+	vh = (struct vlan_ethhdr *)skb->data;
+	if (vh->h_vlan_proto == htons(ETH_P_8021Q) &&
+	    vh->h_vlan_encapsulated_proto == htons(ETH_P_FCOE)) {
+		skb_pull(skb, sizeof(struct vlan_ethhdr));
+		transport_len += sizeof(struct vlan_ethhdr);
+	} else if (eh->h_proto == htons(ETH_P_FCOE)) {
+		transport_len += sizeof(struct ethhdr);
+		skb_pull(skb, sizeof(struct ethhdr));
+	} else
+		return -1;
+
+	fcoe_hdr = (struct fcoe_hdr *)skb->data;
+	if (FC_FCOE_DECAPS_VER(fcoe_hdr) != FC_FCOE_VER)
+		return -1;
+
+	fp = (struct fc_frame *)skb;
+	fc_frame_init(fp);
+	fr_sof(fp) = fcoe_hdr->fcoe_sof;
+	skb_pull(skb, sizeof(struct fcoe_hdr));
+	transport_len += sizeof(struct fcoe_hdr);
+
+	ft = (struct fcoe_crc_eof *)(skb->data + len -
+				     transport_len - sizeof(*ft));
+	fr_eof(fp) = ft->fcoe_eof;
+	skb_trim(skb, len - transport_len - sizeof(*ft));
+	return 0;
+}
+
+static inline int fnic_handle_flogi_resp(struct fnic *fnic,
+					 struct fc_frame *fp)
+{
+	u8 mac[ETH_ALEN] = FC_FCOE_FLOGI_MAC;
+	struct ethhdr *eth_hdr;
+	struct fc_frame_header *fh;
+	int ret = 0;
+	unsigned long flags;
+	struct fc_frame *old_flogi_resp = NULL;
+
+	fh = (struct fc_frame_header *)fr_hdr(fp);
+
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+
+	if (fnic->state == FNIC_IN_ETH_MODE) {
+
+		/*
+		 * Check if oxid matches on taking the lock. A new Flogi
+		 * issued by libFC might have changed the fnic cached oxid
+		 */
+		if (fnic->flogi_oxid != ntohs(fh->fh_ox_id)) {
+			FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
+				     "Flogi response oxid not"
+				     " matching cached oxid, dropping frame"
+				     "\n");
+			ret = -1;
+			spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+			dev_kfree_skb_irq(fp_skb(fp));
+			goto handle_flogi_resp_end;
+		}
+
+		/* Drop older cached flogi response frame, cache this frame */
+		old_flogi_resp = fnic->flogi_resp;
+		fnic->flogi_resp = fp;
+		fnic->flogi_oxid = FC_XID_UNKNOWN;
+
+		/*
+		 * this frame is part of flogi get the src mac addr from this
+		 * frame if the src mac is fcoui based then we mark the
+		 * address mode flag to use fcoui base for dst mac addr
+		 * otherwise we have to store the fcoe gateway addr
+		 */
+		eth_hdr = (struct ethhdr *)skb_mac_header(fp_skb(fp));
+		memcpy(mac, eth_hdr->h_source, ETH_ALEN);
+
+		if (ntoh24(mac) == FC_FCOE_OUI)
+			fnic->fcoui_mode = 1;
+		else {
+			fnic->fcoui_mode = 0;
+			memcpy(fnic->dest_addr, mac, ETH_ALEN);
+		}
+
+		/*
+		 * Except for Flogi frame, all outbound frames from us have the
+		 * Eth Src address as FC_FCOE_OUI"our_sid". Flogi frame uses
+		 * the vnic MAC address as the Eth Src address
+		 */
+		fc_fcoe_set_mac(fnic->data_src_addr, fh->fh_d_id);
+
+		/* We get our s_id from the d_id of the flogi resp frame */
+		fnic->s_id = ntoh24(fh->fh_d_id);
+
+		/* Change state to reflect transition from Eth to FC mode */
+		fnic->state = FNIC_IN_ETH_TRANS_FC_MODE;
+
+	} else {
+		FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
+			     "Unexpected fnic state %s while"
+			     " processing flogi resp\n",
+			     fnic_state_to_str(fnic->state));
+		ret = -1;
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		dev_kfree_skb_irq(fp_skb(fp));
+		goto handle_flogi_resp_end;
+	}
+
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	/* Drop older cached frame */
+	if (old_flogi_resp)
+		dev_kfree_skb_irq(fp_skb(old_flogi_resp));
+
+	/*
+	 * send flogi reg request to firmware, this will put the fnic in
+	 * in FC mode
+	 */
+	ret = fnic_flogi_reg_handler(fnic);
+
+	if (ret < 0) {
+		int free_fp = 1;
+		spin_lock_irqsave(&fnic->fnic_lock, flags);
+		/*
+		 * free the frame is some other thread is not
+		 * pointing to it
+		 */
+		if (fnic->flogi_resp != fp)
+			free_fp = 0;
+		else
+			fnic->flogi_resp = NULL;
+
+		if (fnic->state == FNIC_IN_ETH_TRANS_FC_MODE)
+			fnic->state = FNIC_IN_ETH_MODE;
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		if (free_fp)
+			dev_kfree_skb_irq(fp_skb(fp));
+	}
+
+ handle_flogi_resp_end:
+	return ret;
+}
+
+/* Returns 1 for a response that matches cached flogi oxid */
+static inline int is_matching_flogi_resp_frame(struct fnic *fnic,
+					       struct fc_frame *fp)
+{
+	struct fc_frame_header *fh;
+	int ret = 0;
+	u32 f_ctl;
+
+	fh = fc_frame_header_get(fp);
+	f_ctl = ntoh24(fh->fh_f_ctl);
+
+	if (fnic->flogi_oxid == ntohs(fh->fh_ox_id) &&
+	    fh->fh_r_ctl == FC_RCTL_ELS_REP &&
+	    (f_ctl & (FC_FC_EX_CTX | FC_FC_SEQ_CTX)) == FC_FC_EX_CTX &&
+	    fh->fh_type == FC_TYPE_ELS)
+		ret = 1;
+
+	return ret;
+}
+
+static void fnic_rq_cmpl_frame_recv(struct vnic_rq *rq, struct cq_desc
+				    *cq_desc, struct vnic_rq_buf *buf,
+				    int skipped __attribute__((unused)),
+				    void *opaque)
+{
+	struct fnic *fnic = vnic_dev_priv(rq->vdev);
+	struct sk_buff *skb;
+	struct fc_frame *fp;
+	unsigned int eth_hdrs_stripped;
+	u8 type, color, eop, sop, ingress_port, vlan_stripped;
+	u8 fcoe = 0, fcoe_sof, fcoe_eof;
+	u8 fcoe_fc_crc_ok = 1, fcoe_enc_error = 0;
+	u8 tcp_udp_csum_ok, udp, tcp, ipv4_csum_ok;
+	u8 ipv6, ipv4, ipv4_fragment, rss_type, csum_not_calc;
+	u8 fcs_ok = 1, packet_error = 0;
+	u16 q_number, completed_index, bytes_written = 0, vlan, checksum;
+	u32 rss_hash;
+	u16 exchange_id, tmpl;
+	u8 sof = 0;
+	u8 eof = 0;
+	u32 fcp_bytes_written = 0;
+	unsigned long flags;
+
+	pci_unmap_single(fnic->pdev, buf->dma_addr, buf->len,
+			 PCI_DMA_FROMDEVICE);
+	skb = buf->os_buf;
+	buf->os_buf = NULL;
+
+	cq_desc_dec(cq_desc, &type, &color, &q_number, &completed_index);
+	if (type == CQ_DESC_TYPE_RQ_FCP) {
+		cq_fcp_rq_desc_dec((struct cq_fcp_rq_desc *)cq_desc,
+				   &type, &color, &q_number, &completed_index,
+				   &eop, &sop, &fcoe_fc_crc_ok, &exchange_id,
+				   &tmpl, &fcp_bytes_written, &sof, &eof,
+				   &ingress_port, &packet_error,
+				   &fcoe_enc_error, &fcs_ok, &vlan_stripped,
+				   &vlan);
+		eth_hdrs_stripped = 1;
+
+	} else if (type == CQ_DESC_TYPE_RQ_ENET) {
+		cq_enet_rq_desc_dec((struct cq_enet_rq_desc *)cq_desc,
+				    &type, &color, &q_number, &completed_index,
+				    &ingress_port, &fcoe, &eop, &sop,
+				    &rss_type, &csum_not_calc, &rss_hash,
+				    &bytes_written, &packet_error,
+				    &vlan_stripped, &vlan, &checksum,
+				    &fcoe_sof, &fcoe_fc_crc_ok,
+				    &fcoe_enc_error, &fcoe_eof,
+				    &tcp_udp_csum_ok, &udp, &tcp,
+				    &ipv4_csum_ok, &ipv6, &ipv4,
+				    &ipv4_fragment, &fcs_ok);
+		eth_hdrs_stripped = 0;
+
+	} else {
+		/* wrong CQ type*/
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "fnic rq_cmpl wrong cq type x%x\n", type);
+		goto drop;
+	}
+
+	if (!fcs_ok || packet_error || !fcoe_fc_crc_ok || fcoe_enc_error) {
+		FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
+			     "fnic rq_cmpl fcoe x%x fcsok x%x"
+			     " pkterr x%x fcoe_fc_crc_ok x%x, fcoe_enc_err"
+			     " x%x\n",
+			     fcoe, fcs_ok, packet_error,
+			     fcoe_fc_crc_ok, fcoe_enc_error);
+		goto drop;
+	}
+
+	if (eth_hdrs_stripped)
+		fnic_import_rq_fc_frame(skb, fcp_bytes_written, sof, eof);
+	else if (fnic_import_rq_eth_pkt(skb, bytes_written))
+		goto drop;
+
+	fp = (struct fc_frame *)skb;
+
+	/*
+	 * If frame is an ELS response that matches the cached FLOGI OX_ID,
+	 * and is accept, issue flogi_reg_request copy wq request to firmware
+	 * to register the S_ID and determine whether FC_OUI mode or GW mode.
+	 */
+	if (is_matching_flogi_resp_frame(fnic, fp)) {
+		if (!eth_hdrs_stripped) {
+			if (fc_frame_payload_op(fp) == ELS_LS_ACC) {
+				fnic_handle_flogi_resp(fnic, fp);
+				return;
+			}
+			/*
+			 * Recd. Flogi reject. No point registering
+			 * with fw, but forward to libFC
+			 */
+			goto forward;
+		}
+		goto drop;
+	}
+	if (!eth_hdrs_stripped)
+		goto drop;
+
+forward:
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+	if (fnic->stop_rx_link_events) {
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		goto drop;
+	}
+	/* Use fr_flags to indicate whether succ. flogi resp or not */
+	fr_flags(fp) = 0;
+	fr_dev(fp) = fnic->lport;
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	skb_queue_tail(&fnic->frame_queue, skb);
+	queue_work(fnic_event_queue, &fnic->frame_work);
+
+	return;
+drop:
+	dev_kfree_skb_irq(skb);
+}
+
+static int fnic_rq_cmpl_handler_cont(struct vnic_dev *vdev,
+				     struct cq_desc *cq_desc, u8 type,
+				     u16 q_number, u16 completed_index,
+				     void *opaque)
+{
+	struct fnic *fnic = vnic_dev_priv(vdev);
+
+	vnic_rq_service(&fnic->rq[q_number], cq_desc, completed_index,
+			VNIC_RQ_RETURN_DESC, fnic_rq_cmpl_frame_recv,
+			NULL);
+	return 0;
+}
+
+int fnic_rq_cmpl_handler(struct fnic *fnic, int rq_work_to_do)
+{
+	unsigned int tot_rq_work_done = 0, cur_work_done;
+	unsigned int i;
+	int err;
+
+	for (i = 0; i < fnic->rq_count; i++) {
+		cur_work_done = vnic_cq_service(&fnic->cq[i], rq_work_to_do,
+						fnic_rq_cmpl_handler_cont,
+						NULL);
+		if (cur_work_done) {
+			err = vnic_rq_fill(&fnic->rq[i], fnic_alloc_rq_frame);
+			if (err)
+				shost_printk(KERN_ERR, fnic->lport->host,
+					     "fnic_alloc_rq_frame cant alloc"
+					     " frame\n");
+		}
+		tot_rq_work_done += cur_work_done;
+	}
+
+	return tot_rq_work_done;
+}
+
+/*
+ * This function is called once at init time to allocate and fill RQ
+ * buffers. Subsequently, it is called in the interrupt context after RQ
+ * buffer processing to replenish the buffers in the RQ
+ */
+int fnic_alloc_rq_frame(struct vnic_rq *rq)
+{
+	struct fnic *fnic = vnic_dev_priv(rq->vdev);
+	struct sk_buff *skb;
+	u16 len;
+	dma_addr_t pa;
+
+	len = FC_FRAME_HEADROOM + FC_MAX_FRAME + FC_FRAME_TAILROOM;
+	skb = dev_alloc_skb(len);
+	if (!skb) {
+		FNIC_FCS_DBG(KERN_DEBUG, fnic->lport->host,
+			     "Unable to allocate RQ sk_buff\n");
+		return -ENOMEM;
+	}
+	skb_reset_mac_header(skb);
+	skb_reset_transport_header(skb);
+	skb_reset_network_header(skb);
+	skb_put(skb, len);
+	pa = pci_map_single(fnic->pdev, skb->data, len, PCI_DMA_FROMDEVICE);
+	fnic_queue_rq_desc(rq, skb, pa, len);
+	return 0;
+}
+
+void fnic_free_rq_buf(struct vnic_rq *rq, struct vnic_rq_buf *buf)
+{
+	struct fc_frame *fp = buf->os_buf;
+	struct fnic *fnic = vnic_dev_priv(rq->vdev);
+
+	pci_unmap_single(fnic->pdev, buf->dma_addr, buf->len,
+			 PCI_DMA_FROMDEVICE);
+
+	dev_kfree_skb(fp_skb(fp));
+	buf->os_buf = NULL;
+}
+
+static inline int is_flogi_frame(struct fc_frame_header *fh)
+{
+	return fh->fh_r_ctl == FC_RCTL_ELS_REQ && *(u8 *)(fh + 1) == ELS_FLOGI;
+}
+
+int fnic_send_frame(struct fnic *fnic, struct fc_frame *fp)
+{
+	struct vnic_wq *wq = &fnic->wq[0];
+	struct sk_buff *skb;
+	dma_addr_t pa;
+	struct ethhdr *eth_hdr;
+	struct vlan_ethhdr *vlan_hdr;
+	struct fcoe_hdr *fcoe_hdr;
+	struct fc_frame_header *fh;
+	u32 tot_len, eth_hdr_len;
+	int ret = 0;
+	unsigned long flags;
+
+	fh = fc_frame_header_get(fp);
+	skb = fp_skb(fp);
+
+	if (!fnic->vlan_hw_insert) {
+		eth_hdr_len = sizeof(*vlan_hdr) + sizeof(*fcoe_hdr);
+		vlan_hdr = (struct vlan_ethhdr *)skb_push(skb, eth_hdr_len);
+		eth_hdr = (struct ethhdr *)vlan_hdr;
+		vlan_hdr->h_vlan_proto = htons(ETH_P_8021Q);
+		vlan_hdr->h_vlan_encapsulated_proto = htons(ETH_P_FCOE);
+		vlan_hdr->h_vlan_TCI = htons(fnic->vlan_id);
+		fcoe_hdr = (struct fcoe_hdr *)(vlan_hdr + 1);
+	} else {
+		eth_hdr_len = sizeof(*eth_hdr) + sizeof(*fcoe_hdr);
+		eth_hdr = (struct ethhdr *)skb_push(skb, eth_hdr_len);
+		eth_hdr->h_proto = htons(ETH_P_FCOE);
+		fcoe_hdr = (struct fcoe_hdr *)(eth_hdr + 1);
+	}
+
+	if (is_flogi_frame(fh)) {
+		fc_fcoe_set_mac(eth_hdr->h_dest, fh->fh_d_id);
+		memcpy(eth_hdr->h_source, fnic->mac_addr, ETH_ALEN);
+	} else {
+		if (fnic->fcoui_mode)
+			fc_fcoe_set_mac(eth_hdr->h_dest, fh->fh_d_id);
+		else
+			memcpy(eth_hdr->h_dest, fnic->dest_addr, ETH_ALEN);
+		memcpy(eth_hdr->h_source, fnic->data_src_addr, ETH_ALEN);
+	}
+
+	tot_len = skb->len;
+	BUG_ON(tot_len % 4);
+
+	memset(fcoe_hdr, 0, sizeof(*fcoe_hdr));
+	fcoe_hdr->fcoe_sof = fr_sof(fp);
+	if (FC_FCOE_VER)
+		FC_FCOE_ENCAPS_VER(fcoe_hdr, FC_FCOE_VER);
+
+	pa = pci_map_single(fnic->pdev, eth_hdr, tot_len, PCI_DMA_TODEVICE);
+
+	spin_lock_irqsave(&fnic->wq_lock[0], flags);
+
+	if (!vnic_wq_desc_avail(wq)) {
+		pci_unmap_single(fnic->pdev, pa,
+				 tot_len, PCI_DMA_TODEVICE);
+		ret = -1;
+		goto fnic_send_frame_end;
+	}
+
+	fnic_queue_wq_desc(wq, skb, pa, tot_len, fr_eof(fp),
+			   fnic->vlan_hw_insert, fnic->vlan_id, 1, 1, 1);
+fnic_send_frame_end:
+	spin_unlock_irqrestore(&fnic->wq_lock[0], flags);
+
+	if (ret)
+		dev_kfree_skb_any(fp_skb(fp));
+
+	return ret;
+}
+
+/*
+ * fnic_send
+ * Routine to send a raw frame
+ */
+int fnic_send(struct fc_lport *lp, struct fc_frame *fp)
+{
+	struct fnic *fnic = lport_priv(lp);
+	struct fc_frame_header *fh;
+	int ret = 0;
+	enum fnic_state old_state;
+	unsigned long flags;
+	struct fc_frame *old_flogi = NULL;
+	struct fc_frame *old_flogi_resp = NULL;
+
+	if (fnic->in_remove) {
+		dev_kfree_skb(fp_skb(fp));
+		ret = -1;
+		goto fnic_send_end;
+	}
+
+	fh = fc_frame_header_get(fp);
+	/* if not an Flogi frame, send it out, this is the common case */
+	if (!is_flogi_frame(fh))
+		return fnic_send_frame(fnic, fp);
+
+	/* Flogi frame, now enter the state machine */
+
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+again:
+	/* Get any old cached frames, free them after dropping lock */
+	old_flogi = fnic->flogi;
+	fnic->flogi = NULL;
+	old_flogi_resp = fnic->flogi_resp;
+	fnic->flogi_resp = NULL;
+
+	fnic->flogi_oxid = FC_XID_UNKNOWN;
+
+	old_state = fnic->state;
+	switch (old_state) {
+	case FNIC_IN_FC_MODE:
+	case FNIC_IN_ETH_TRANS_FC_MODE:
+	default:
+		fnic->state = FNIC_IN_FC_TRANS_ETH_MODE;
+		vnic_dev_del_addr(fnic->vdev, fnic->data_src_addr);
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+		if (old_flogi) {
+			dev_kfree_skb(fp_skb(old_flogi));
+			old_flogi = NULL;
+		}
+		if (old_flogi_resp) {
+			dev_kfree_skb(fp_skb(old_flogi_resp));
+			old_flogi_resp = NULL;
+		}
+
+		ret = fnic_fw_reset_handler(fnic);
+
+		spin_lock_irqsave(&fnic->fnic_lock, flags);
+		if (fnic->state != FNIC_IN_FC_TRANS_ETH_MODE)
+			goto again;
+		if (ret) {
+			fnic->state = old_state;
+			spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+			dev_kfree_skb(fp_skb(fp));
+			goto fnic_send_end;
+		}
+		old_flogi = fnic->flogi;
+		fnic->flogi = fp;
+		fnic->flogi_oxid = ntohs(fh->fh_ox_id);
+		old_flogi_resp = fnic->flogi_resp;
+		fnic->flogi_resp = NULL;
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		break;
+
+	case FNIC_IN_FC_TRANS_ETH_MODE:
+		/*
+		 * A reset is pending with the firmware. Store the flogi
+		 * and its oxid. The transition out of this state happens
+		 * only when Firmware completes the reset, either with
+		 * success or failed. If success, transition to
+		 * FNIC_IN_ETH_MODE, if fail, then transition to
+		 * FNIC_IN_FC_MODE
+		 */
+		fnic->flogi = fp;
+		fnic->flogi_oxid = ntohs(fh->fh_ox_id);
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		break;
+
+	case FNIC_IN_ETH_MODE:
+		/*
+		 * The fw/hw is already in eth mode. Store the oxid,
+		 * and send the flogi frame out. The transition out of this
+		 * state happens only we receive flogi response from the
+		 * network, and the oxid matches the cached oxid when the
+		 * flogi frame was sent out. If they match, then we issue
+		 * a flogi_reg request and transition to state
+		 * FNIC_IN_ETH_TRANS_FC_MODE
+		 */
+		fnic->flogi_oxid = ntohs(fh->fh_ox_id);
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		ret = fnic_send_frame(fnic, fp);
+		break;
+	}
+
+fnic_send_end:
+	if (old_flogi)
+		dev_kfree_skb(fp_skb(old_flogi));
+	if (old_flogi_resp)
+		dev_kfree_skb(fp_skb(old_flogi_resp));
+	return ret;
+}
+
+static void fnic_wq_complete_frame_send(struct vnic_wq *wq,
+					struct cq_desc *cq_desc,
+					struct vnic_wq_buf *buf, void *opaque)
+{
+	struct sk_buff *skb = buf->os_buf;
+	struct fc_frame *fp = (struct fc_frame *)skb;
+	struct fnic *fnic = vnic_dev_priv(wq->vdev);
+
+	pci_unmap_single(fnic->pdev, buf->dma_addr,
+			 buf->len, PCI_DMA_TODEVICE);
+	dev_kfree_skb_irq(fp_skb(fp));
+	buf->os_buf = NULL;
+}
+
+static int fnic_wq_cmpl_handler_cont(struct vnic_dev *vdev,
+				     struct cq_desc *cq_desc, u8 type,
+				     u16 q_number, u16 completed_index,
+				     void *opaque)
+{
+	struct fnic *fnic = vnic_dev_priv(vdev);
+	unsigned long flags;
+
+	spin_lock_irqsave(&fnic->wq_lock[q_number], flags);
+	vnic_wq_service(&fnic->wq[q_number], cq_desc, completed_index,
+			fnic_wq_complete_frame_send, NULL);
+	spin_unlock_irqrestore(&fnic->wq_lock[q_number], flags);
+
+	return 0;
+}
+
+int fnic_wq_cmpl_handler(struct fnic *fnic, int work_to_do)
+{
+	unsigned int wq_work_done = 0;
+	unsigned int i;
+
+	for (i = 0; i < fnic->raw_wq_count; i++) {
+		wq_work_done  += vnic_cq_service(&fnic->cq[fnic->rq_count+i],
+						 work_to_do,
+						 fnic_wq_cmpl_handler_cont,
+						 NULL);
+	}
+
+	return wq_work_done;
+}
+
+
+void fnic_free_wq_buf(struct vnic_wq *wq, struct vnic_wq_buf *buf)
+{
+	struct fc_frame *fp = buf->os_buf;
+	struct fnic *fnic = vnic_dev_priv(wq->vdev);
+
+	pci_unmap_single(fnic->pdev, buf->dma_addr,
+			 buf->len, PCI_DMA_TODEVICE);
+
+	dev_kfree_skb(fp_skb(fp));
+	buf->os_buf = NULL;
+}

drivers/scsi/fnic/fnic_io.h

@@ -0,0 +1,67 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _FNIC_IO_H_
+#define _FNIC_IO_H_
+
+#include <scsi/fc/fc_fcp.h>
+
+#define FNIC_DFLT_SG_DESC_CNT  32
+#define FNIC_MAX_SG_DESC_CNT        1024    /* Maximum descriptors per sgl */
+#define FNIC_SG_DESC_ALIGN          16      /* Descriptor address alignment */
+
+struct host_sg_desc {
+	__le64 addr;
+	__le32 len;
+	u32 _resvd;
+};
+
+struct fnic_dflt_sgl_list {
+	struct host_sg_desc sg_desc[FNIC_DFLT_SG_DESC_CNT];
+};
+
+struct fnic_sgl_list {
+	struct host_sg_desc sg_desc[FNIC_MAX_SG_DESC_CNT];
+};
+
+enum fnic_sgl_list_type {
+	FNIC_SGL_CACHE_DFLT = 0,  /* cache with default size sgl */
+	FNIC_SGL_CACHE_MAX,       /* cache with max size sgl */
+	FNIC_SGL_NUM_CACHES       /* number of sgl caches */
+};
+
+enum fnic_ioreq_state {
+	FNIC_IOREQ_CMD_PENDING = 0,
+	FNIC_IOREQ_ABTS_PENDING,
+	FNIC_IOREQ_ABTS_COMPLETE,
+	FNIC_IOREQ_CMD_COMPLETE,
+};
+
+struct fnic_io_req {
+	struct host_sg_desc *sgl_list; /* sgl list */
+	void *sgl_list_alloc; /* sgl list address used for free */
+	dma_addr_t sense_buf_pa; /* dma address for sense buffer*/
+	dma_addr_t sgl_list_pa;	/* dma address for sgl list */
+	u16 sgl_cnt;
+	u8 sgl_type; /* device DMA descriptor list type */
+	u8 io_completed:1; /* set to 1 when fw completes IO */
+	u32 port_id; /* remote port DID */
+	struct completion *abts_done; /* completion for abts */
+	struct completion *dr_done; /* completion for device reset */
+};
+
+#endif /* _FNIC_IO_H_ */

drivers/scsi/fnic/fnic_isr.c

@@ -0,0 +1,332 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/pci.h>
+#include <linux/interrupt.h>
+#include <scsi/libfc.h>
+#include <scsi/fc_frame.h>
+#include "vnic_dev.h"
+#include "vnic_intr.h"
+#include "vnic_stats.h"
+#include "fnic_io.h"
+#include "fnic.h"
+
+static irqreturn_t fnic_isr_legacy(int irq, void *data)
+{
+	struct fnic *fnic = data;
+	u32 pba;
+	unsigned long work_done = 0;
+
+	pba = vnic_intr_legacy_pba(fnic->legacy_pba);
+	if (!pba)
+		return IRQ_NONE;
+
+	if (pba & (1 << FNIC_INTX_NOTIFY)) {
+		vnic_intr_return_all_credits(&fnic->intr[FNIC_INTX_NOTIFY]);
+		fnic_handle_link_event(fnic);
+	}
+
+	if (pba & (1 << FNIC_INTX_ERR)) {
+		vnic_intr_return_all_credits(&fnic->intr[FNIC_INTX_ERR]);
+		fnic_log_q_error(fnic);
+	}
+
+	if (pba & (1 << FNIC_INTX_WQ_RQ_COPYWQ)) {
+		work_done += fnic_wq_copy_cmpl_handler(fnic, 8);
+		work_done += fnic_wq_cmpl_handler(fnic, 4);
+		work_done += fnic_rq_cmpl_handler(fnic, 4);
+
+		vnic_intr_return_credits(&fnic->intr[FNIC_INTX_WQ_RQ_COPYWQ],
+					 work_done,
+					 1 /* unmask intr */,
+					 1 /* reset intr timer */);
+	}
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t fnic_isr_msi(int irq, void *data)
+{
+	struct fnic *fnic = data;
+	unsigned long work_done = 0;
+
+	work_done += fnic_wq_copy_cmpl_handler(fnic, 8);
+	work_done += fnic_wq_cmpl_handler(fnic, 4);
+	work_done += fnic_rq_cmpl_handler(fnic, 4);
+
+	vnic_intr_return_credits(&fnic->intr[0],
+				 work_done,
+				 1 /* unmask intr */,
+				 1 /* reset intr timer */);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t fnic_isr_msix_rq(int irq, void *data)
+{
+	struct fnic *fnic = data;
+	unsigned long rq_work_done = 0;
+
+	rq_work_done = fnic_rq_cmpl_handler(fnic, 4);
+	vnic_intr_return_credits(&fnic->intr[FNIC_MSIX_RQ],
+				 rq_work_done,
+				 1 /* unmask intr */,
+				 1 /* reset intr timer */);
+
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t fnic_isr_msix_wq(int irq, void *data)
+{
+	struct fnic *fnic = data;
+	unsigned long wq_work_done = 0;
+
+	wq_work_done = fnic_wq_cmpl_handler(fnic, 4);
+	vnic_intr_return_credits(&fnic->intr[FNIC_MSIX_WQ],
+				 wq_work_done,
+				 1 /* unmask intr */,
+				 1 /* reset intr timer */);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t fnic_isr_msix_wq_copy(int irq, void *data)
+{
+	struct fnic *fnic = data;
+	unsigned long wq_copy_work_done = 0;
+
+	wq_copy_work_done = fnic_wq_copy_cmpl_handler(fnic, 8);
+	vnic_intr_return_credits(&fnic->intr[FNIC_MSIX_WQ_COPY],
+				 wq_copy_work_done,
+				 1 /* unmask intr */,
+				 1 /* reset intr timer */);
+	return IRQ_HANDLED;
+}
+
+static irqreturn_t fnic_isr_msix_err_notify(int irq, void *data)
+{
+	struct fnic *fnic = data;
+
+	vnic_intr_return_all_credits(&fnic->intr[FNIC_MSIX_ERR_NOTIFY]);
+	fnic_log_q_error(fnic);
+	fnic_handle_link_event(fnic);
+
+	return IRQ_HANDLED;
+}
+
+void fnic_free_intr(struct fnic *fnic)
+{
+	int i;
+
+	switch (vnic_dev_get_intr_mode(fnic->vdev)) {
+	case VNIC_DEV_INTR_MODE_INTX:
+	case VNIC_DEV_INTR_MODE_MSI:
+		free_irq(fnic->pdev->irq, fnic);
+		break;
+
+	case VNIC_DEV_INTR_MODE_MSIX:
+		for (i = 0; i < ARRAY_SIZE(fnic->msix); i++)
+			if (fnic->msix[i].requested)
+				free_irq(fnic->msix_entry[i].vector,
+					 fnic->msix[i].devid);
+		break;
+
+	default:
+		break;
+	}
+}
+
+int fnic_request_intr(struct fnic *fnic)
+{
+	int err = 0;
+	int i;
+
+	switch (vnic_dev_get_intr_mode(fnic->vdev)) {
+
+	case VNIC_DEV_INTR_MODE_INTX:
+		err = request_irq(fnic->pdev->irq, &fnic_isr_legacy,
+				  IRQF_SHARED, DRV_NAME, fnic);
+		break;
+
+	case VNIC_DEV_INTR_MODE_MSI:
+		err = request_irq(fnic->pdev->irq, &fnic_isr_msi,
+				  0, fnic->name, fnic);
+		break;
+
+	case VNIC_DEV_INTR_MODE_MSIX:
+
+		sprintf(fnic->msix[FNIC_MSIX_RQ].devname,
+			"%.11s-fcs-rq", fnic->name);
+		fnic->msix[FNIC_MSIX_RQ].isr = fnic_isr_msix_rq;
+		fnic->msix[FNIC_MSIX_RQ].devid = fnic;
+
+		sprintf(fnic->msix[FNIC_MSIX_WQ].devname,
+			"%.11s-fcs-wq", fnic->name);
+		fnic->msix[FNIC_MSIX_WQ].isr = fnic_isr_msix_wq;
+		fnic->msix[FNIC_MSIX_WQ].devid = fnic;
+
+		sprintf(fnic->msix[FNIC_MSIX_WQ_COPY].devname,
+			"%.11s-scsi-wq", fnic->name);
+		fnic->msix[FNIC_MSIX_WQ_COPY].isr = fnic_isr_msix_wq_copy;
+		fnic->msix[FNIC_MSIX_WQ_COPY].devid = fnic;
+
+		sprintf(fnic->msix[FNIC_MSIX_ERR_NOTIFY].devname,
+			"%.11s-err-notify", fnic->name);
+		fnic->msix[FNIC_MSIX_ERR_NOTIFY].isr =
+			fnic_isr_msix_err_notify;
+		fnic->msix[FNIC_MSIX_ERR_NOTIFY].devid = fnic;
+
+		for (i = 0; i < ARRAY_SIZE(fnic->msix); i++) {
+			err = request_irq(fnic->msix_entry[i].vector,
+					  fnic->msix[i].isr, 0,
+					  fnic->msix[i].devname,
+					  fnic->msix[i].devid);
+			if (err) {
+				shost_printk(KERN_ERR, fnic->lport->host,
+					     "MSIX: request_irq"
+					     " failed %d\n", err);
+				fnic_free_intr(fnic);
+				break;
+			}
+			fnic->msix[i].requested = 1;
+		}
+		break;
+
+	default:
+		break;
+	}
+
+	return err;
+}
+
+int fnic_set_intr_mode(struct fnic *fnic)
+{
+	unsigned int n = ARRAY_SIZE(fnic->rq);
+	unsigned int m = ARRAY_SIZE(fnic->wq);
+	unsigned int o = ARRAY_SIZE(fnic->wq_copy);
+	unsigned int i;
+
+	/*
+	 * Set interrupt mode (INTx, MSI, MSI-X) depending
+	 * system capabilities.
+	 *
+	 * Try MSI-X first
+	 *
+	 * We need n RQs, m WQs, o Copy WQs, n+m+o CQs, and n+m+o+1 INTRs
+	 * (last INTR is used for WQ/RQ errors and notification area)
+	 */
+
+	BUG_ON(ARRAY_SIZE(fnic->msix_entry) < n + m + o + 1);
+	for (i = 0; i < n + m + o + 1; i++)
+		fnic->msix_entry[i].entry = i;
+
+	if (fnic->rq_count >= n &&
+	    fnic->raw_wq_count >= m &&
+	    fnic->wq_copy_count >= o &&
+	    fnic->cq_count >= n + m + o) {
+		if (!pci_enable_msix(fnic->pdev, fnic->msix_entry,
+				    n + m + o + 1)) {
+			fnic->rq_count = n;
+			fnic->raw_wq_count = m;
+			fnic->wq_copy_count = o;
+			fnic->wq_count = m + o;
+			fnic->cq_count = n + m + o;
+			fnic->intr_count = n + m + o + 1;
+			fnic->err_intr_offset = FNIC_MSIX_ERR_NOTIFY;
+
+			FNIC_ISR_DBG(KERN_DEBUG, fnic->lport->host,
+				     "Using MSI-X Interrupts\n");
+			vnic_dev_set_intr_mode(fnic->vdev,
+					       VNIC_DEV_INTR_MODE_MSIX);
+			return 0;
+		}
+	}
+
+	/*
+	 * Next try MSI
+	 * We need 1 RQ, 1 WQ, 1 WQ_COPY, 3 CQs, and 1 INTR
+	 */
+	if (fnic->rq_count >= 1 &&
+	    fnic->raw_wq_count >= 1 &&
+	    fnic->wq_copy_count >= 1 &&
+	    fnic->cq_count >= 3 &&
+	    fnic->intr_count >= 1 &&
+	    !pci_enable_msi(fnic->pdev)) {
+
+		fnic->rq_count = 1;
+		fnic->raw_wq_count = 1;
+		fnic->wq_copy_count = 1;
+		fnic->wq_count = 2;
+		fnic->cq_count = 3;
+		fnic->intr_count = 1;
+		fnic->err_intr_offset = 0;
+
+		FNIC_ISR_DBG(KERN_DEBUG, fnic->lport->host,
+			     "Using MSI Interrupts\n");
+		vnic_dev_set_intr_mode(fnic->vdev, VNIC_DEV_INTR_MODE_MSI);
+
+		return 0;
+	}
+
+	/*
+	 * Next try INTx
+	 * We need 1 RQ, 1 WQ, 1 WQ_COPY, 3 CQs, and 3 INTRs
+	 * 1 INTR is used for all 3 queues, 1 INTR for queue errors
+	 * 1 INTR for notification area
+	 */
+
+	if (fnic->rq_count >= 1 &&
+	    fnic->raw_wq_count >= 1 &&
+	    fnic->wq_copy_count >= 1 &&
+	    fnic->cq_count >= 3 &&
+	    fnic->intr_count >= 3) {
+
+		fnic->rq_count = 1;
+		fnic->raw_wq_count = 1;
+		fnic->wq_copy_count = 1;
+		fnic->cq_count = 3;
+		fnic->intr_count = 3;
+
+		FNIC_ISR_DBG(KERN_DEBUG, fnic->lport->host,
+			     "Using Legacy Interrupts\n");
+		vnic_dev_set_intr_mode(fnic->vdev, VNIC_DEV_INTR_MODE_INTX);
+
+		return 0;
+	}
+
+	vnic_dev_set_intr_mode(fnic->vdev, VNIC_DEV_INTR_MODE_UNKNOWN);
+
+	return -EINVAL;
+}
+
+void fnic_clear_intr_mode(struct fnic *fnic)
+{
+	switch (vnic_dev_get_intr_mode(fnic->vdev)) {
+	case VNIC_DEV_INTR_MODE_MSIX:
+		pci_disable_msix(fnic->pdev);
+		break;
+	case VNIC_DEV_INTR_MODE_MSI:
+		pci_disable_msi(fnic->pdev);
+		break;
+	default:
+		break;
+	}
+
+	vnic_dev_set_intr_mode(fnic->vdev, VNIC_DEV_INTR_MODE_INTX);
+}
+

drivers/scsi/fnic/fnic_main.c

@@ -0,0 +1,942 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/module.h>
+#include <linux/mempool.h>
+#include <linux/string.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/pci.h>
+#include <linux/skbuff.h>
+#include <linux/interrupt.h>
+#include <linux/spinlock.h>
+#include <linux/workqueue.h>
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_transport.h>
+#include <scsi/scsi_transport_fc.h>
+#include <scsi/scsi_tcq.h>
+#include <scsi/libfc.h>
+#include <scsi/fc_frame.h>
+
+#include "vnic_dev.h"
+#include "vnic_intr.h"
+#include "vnic_stats.h"
+#include "fnic_io.h"
+#include "fnic.h"
+
+#define PCI_DEVICE_ID_CISCO_FNIC	0x0045
+
+/* Timer to poll notification area for events. Used for MSI interrupts */
+#define FNIC_NOTIFY_TIMER_PERIOD	(2 * HZ)
+
+static struct kmem_cache *fnic_sgl_cache[FNIC_SGL_NUM_CACHES];
+static struct kmem_cache *fnic_io_req_cache;
+LIST_HEAD(fnic_list);
+DEFINE_SPINLOCK(fnic_list_lock);
+
+/* Supported devices by fnic module */
+static struct pci_device_id fnic_id_table[] = {
+	{ PCI_DEVICE(PCI_VENDOR_ID_CISCO, PCI_DEVICE_ID_CISCO_FNIC) },
+	{ 0, }
+};
+
+MODULE_DESCRIPTION(DRV_DESCRIPTION);
+MODULE_AUTHOR("Abhijeet Joglekar <abjoglek@cisco.com>, "
+	      "Joseph R. Eykholt <jeykholt@cisco.com>");
+MODULE_LICENSE("GPL v2");
+MODULE_VERSION(DRV_VERSION);
+MODULE_DEVICE_TABLE(pci, fnic_id_table);
+
+unsigned int fnic_log_level;
+module_param(fnic_log_level, int, S_IRUGO|S_IWUSR);
+MODULE_PARM_DESC(fnic_log_level, "bit mask of fnic logging levels");
+
+
+static struct libfc_function_template fnic_transport_template = {
+	.frame_send = fnic_send,
+	.fcp_abort_io = fnic_empty_scsi_cleanup,
+	.fcp_cleanup = fnic_empty_scsi_cleanup,
+	.exch_mgr_reset = fnic_exch_mgr_reset
+};
+
+static int fnic_slave_alloc(struct scsi_device *sdev)
+{
+	struct fc_rport *rport = starget_to_rport(scsi_target(sdev));
+	struct fc_lport *lp = shost_priv(sdev->host);
+	struct fnic *fnic = lport_priv(lp);
+
+	sdev->tagged_supported = 1;
+
+	if (!rport || fc_remote_port_chkready(rport))
+		return -ENXIO;
+
+	scsi_activate_tcq(sdev, FNIC_DFLT_QUEUE_DEPTH);
+	rport->dev_loss_tmo = fnic->config.port_down_timeout / 1000;
+
+	return 0;
+}
+
+static struct scsi_host_template fnic_host_template = {
+	.module = THIS_MODULE,
+	.name = DRV_NAME,
+	.queuecommand = fnic_queuecommand,
+	.eh_abort_handler = fnic_abort_cmd,
+	.eh_device_reset_handler = fnic_device_reset,
+	.eh_host_reset_handler = fnic_host_reset,
+	.slave_alloc = fnic_slave_alloc,
+	.change_queue_depth = fc_change_queue_depth,
+	.change_queue_type = fc_change_queue_type,
+	.this_id = -1,
+	.cmd_per_lun = 3,
+	.can_queue = FNIC_MAX_IO_REQ,
+	.use_clustering = ENABLE_CLUSTERING,
+	.sg_tablesize = FNIC_MAX_SG_DESC_CNT,
+	.max_sectors = 0xffff,
+	.shost_attrs = fnic_attrs,
+};
+
+static void fnic_get_host_speed(struct Scsi_Host *shost);
+static struct scsi_transport_template *fnic_fc_transport;
+static struct fc_host_statistics *fnic_get_stats(struct Scsi_Host *);
+
+static struct fc_function_template fnic_fc_functions = {
+
+	.show_host_node_name = 1,
+	.show_host_port_name = 1,
+	.show_host_supported_classes = 1,
+	.show_host_supported_fc4s = 1,
+	.show_host_active_fc4s = 1,
+	.show_host_maxframe_size = 1,
+	.show_host_port_id = 1,
+	.show_host_supported_speeds = 1,
+	.get_host_speed = fnic_get_host_speed,
+	.show_host_speed = 1,
+	.show_host_port_type = 1,
+	.get_host_port_state = fc_get_host_port_state,
+	.show_host_port_state = 1,
+	.show_host_symbolic_name = 1,
+	.show_rport_maxframe_size = 1,
+	.show_rport_supported_classes = 1,
+	.show_host_fabric_name = 1,
+	.show_starget_node_name = 1,
+	.show_starget_port_name = 1,
+	.show_starget_port_id = 1,
+	.show_rport_dev_loss_tmo = 1,
+	.issue_fc_host_lip = fnic_reset,
+	.get_fc_host_stats = fnic_get_stats,
+	.dd_fcrport_size = sizeof(struct fc_rport_libfc_priv),
+	.terminate_rport_io = fnic_terminate_rport_io,
+};
+
+static void fnic_get_host_speed(struct Scsi_Host *shost)
+{
+	struct fc_lport *lp = shost_priv(shost);
+	struct fnic *fnic = lport_priv(lp);
+	u32 port_speed = vnic_dev_port_speed(fnic->vdev);
+
+	/* Add in other values as they get defined in fw */
+	switch (port_speed) {
+	case 10000:
+		fc_host_speed(shost) = FC_PORTSPEED_10GBIT;
+		break;
+	default:
+		fc_host_speed(shost) = FC_PORTSPEED_10GBIT;
+		break;
+	}
+}
+
+static struct fc_host_statistics *fnic_get_stats(struct Scsi_Host *host)
+{
+	int ret;
+	struct fc_lport *lp = shost_priv(host);
+	struct fnic *fnic = lport_priv(lp);
+	struct fc_host_statistics *stats = &lp->host_stats;
+	struct vnic_stats *vs;
+	unsigned long flags;
+
+	if (time_before(jiffies, fnic->stats_time + HZ / FNIC_STATS_RATE_LIMIT))
+		return stats;
+	fnic->stats_time = jiffies;
+
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+	ret = vnic_dev_stats_dump(fnic->vdev, &fnic->stats);
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	if (ret) {
+		FNIC_MAIN_DBG(KERN_DEBUG, fnic->lport->host,
+			      "fnic: Get vnic stats failed"
+			      " 0x%x", ret);
+		return stats;
+	}
+	vs = fnic->stats;
+	stats->tx_frames = vs->tx.tx_unicast_frames_ok;
+	stats->tx_words  = vs->tx.tx_unicast_bytes_ok / 4;
+	stats->rx_frames = vs->rx.rx_unicast_frames_ok;
+	stats->rx_words  = vs->rx.rx_unicast_bytes_ok / 4;
+	stats->error_frames = vs->tx.tx_errors + vs->rx.rx_errors;
+	stats->dumped_frames = vs->tx.tx_drops + vs->rx.rx_drop;
+	stats->invalid_crc_count = vs->rx.rx_crc_errors;
+	stats->seconds_since_last_reset = (jiffies - lp->boot_time) / HZ;
+	stats->fcp_input_megabytes = div_u64(fnic->fcp_input_bytes, 1000000);
+	stats->fcp_output_megabytes = div_u64(fnic->fcp_output_bytes, 1000000);
+
+	return stats;
+}
+
+void fnic_log_q_error(struct fnic *fnic)
+{
+	unsigned int i;
+	u32 error_status;
+
+	for (i = 0; i < fnic->raw_wq_count; i++) {
+		error_status = ioread32(&fnic->wq[i].ctrl->error_status);
+		if (error_status)
+			shost_printk(KERN_ERR, fnic->lport->host,
+				     "WQ[%d] error_status"
+				     " %d\n", i, error_status);
+	}
+
+	for (i = 0; i < fnic->rq_count; i++) {
+		error_status = ioread32(&fnic->rq[i].ctrl->error_status);
+		if (error_status)
+			shost_printk(KERN_ERR, fnic->lport->host,
+				     "RQ[%d] error_status"
+				     " %d\n", i, error_status);
+	}
+
+	for (i = 0; i < fnic->wq_copy_count; i++) {
+		error_status = ioread32(&fnic->wq_copy[i].ctrl->error_status);
+		if (error_status)
+			shost_printk(KERN_ERR, fnic->lport->host,
+				     "CWQ[%d] error_status"
+				     " %d\n", i, error_status);
+	}
+}
+
+void fnic_handle_link_event(struct fnic *fnic)
+{
+	unsigned long flags;
+
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+	if (fnic->stop_rx_link_events) {
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		return;
+	}
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	queue_work(fnic_event_queue, &fnic->link_work);
+
+}
+
+static int fnic_notify_set(struct fnic *fnic)
+{
+	int err;
+
+	switch (vnic_dev_get_intr_mode(fnic->vdev)) {
+	case VNIC_DEV_INTR_MODE_INTX:
+		err = vnic_dev_notify_set(fnic->vdev, FNIC_INTX_NOTIFY);
+		break;
+	case VNIC_DEV_INTR_MODE_MSI:
+		err = vnic_dev_notify_set(fnic->vdev, -1);
+		break;
+	case VNIC_DEV_INTR_MODE_MSIX:
+		err = vnic_dev_notify_set(fnic->vdev, FNIC_MSIX_ERR_NOTIFY);
+		break;
+	default:
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Interrupt mode should be set up"
+			     " before devcmd notify set %d\n",
+			     vnic_dev_get_intr_mode(fnic->vdev));
+		err = -1;
+		break;
+	}
+
+	return err;
+}
+
+static void fnic_notify_timer(unsigned long data)
+{
+	struct fnic *fnic = (struct fnic *)data;
+
+	fnic_handle_link_event(fnic);
+	mod_timer(&fnic->notify_timer,
+		  round_jiffies(jiffies + FNIC_NOTIFY_TIMER_PERIOD));
+}
+
+static void fnic_notify_timer_start(struct fnic *fnic)
+{
+	switch (vnic_dev_get_intr_mode(fnic->vdev)) {
+	case VNIC_DEV_INTR_MODE_MSI:
+		/*
+		 * Schedule first timeout immediately. The driver is
+		 * initiatialized and ready to look for link up notification
+		 */
+		mod_timer(&fnic->notify_timer, jiffies);
+		break;
+	default:
+		/* Using intr for notification for INTx/MSI-X */
+		break;
+	};
+}
+
+static int fnic_dev_wait(struct vnic_dev *vdev,
+			 int (*start)(struct vnic_dev *, int),
+			 int (*finished)(struct vnic_dev *, int *),
+			 int arg)
+{
+	unsigned long time;
+	int done;
+	int err;
+
+	err = start(vdev, arg);
+	if (err)
+		return err;
+
+	/* Wait for func to complete...2 seconds max */
+	time = jiffies + (HZ * 2);
+	do {
+		err = finished(vdev, &done);
+		if (err)
+			return err;
+		if (done)
+			return 0;
+		schedule_timeout_uninterruptible(HZ / 10);
+	} while (time_after(time, jiffies));
+
+	return -ETIMEDOUT;
+}
+
+static int fnic_cleanup(struct fnic *fnic)
+{
+	unsigned int i;
+	int err;
+	unsigned long flags;
+	struct fc_frame *flogi = NULL;
+	struct fc_frame *flogi_resp = NULL;
+
+	vnic_dev_disable(fnic->vdev);
+	for (i = 0; i < fnic->intr_count; i++)
+		vnic_intr_mask(&fnic->intr[i]);
+
+	for (i = 0; i < fnic->rq_count; i++) {
+		err = vnic_rq_disable(&fnic->rq[i]);
+		if (err)
+			return err;
+	}
+	for (i = 0; i < fnic->raw_wq_count; i++) {
+		err = vnic_wq_disable(&fnic->wq[i]);
+		if (err)
+			return err;
+	}
+	for (i = 0; i < fnic->wq_copy_count; i++) {
+		err = vnic_wq_copy_disable(&fnic->wq_copy[i]);
+		if (err)
+			return err;
+	}
+
+	/* Clean up completed IOs and FCS frames */
+	fnic_wq_copy_cmpl_handler(fnic, -1);
+	fnic_wq_cmpl_handler(fnic, -1);
+	fnic_rq_cmpl_handler(fnic, -1);
+
+	/* Clean up the IOs and FCS frames that have not completed */
+	for (i = 0; i < fnic->raw_wq_count; i++)
+		vnic_wq_clean(&fnic->wq[i], fnic_free_wq_buf);
+	for (i = 0; i < fnic->rq_count; i++)
+		vnic_rq_clean(&fnic->rq[i], fnic_free_rq_buf);
+	for (i = 0; i < fnic->wq_copy_count; i++)
+		vnic_wq_copy_clean(&fnic->wq_copy[i],
+				   fnic_wq_copy_cleanup_handler);
+
+	for (i = 0; i < fnic->cq_count; i++)
+		vnic_cq_clean(&fnic->cq[i]);
+	for (i = 0; i < fnic->intr_count; i++)
+		vnic_intr_clean(&fnic->intr[i]);
+
+	/*
+	 * Remove cached flogi and flogi resp frames if any
+	 * These frames are not in any queue, and therefore queue
+	 * cleanup does not clean them. So clean them explicitly
+	 */
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+	flogi = fnic->flogi;
+	fnic->flogi = NULL;
+	flogi_resp = fnic->flogi_resp;
+	fnic->flogi_resp = NULL;
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	if (flogi)
+		dev_kfree_skb(fp_skb(flogi));
+
+	if (flogi_resp)
+		dev_kfree_skb(fp_skb(flogi_resp));
+
+	mempool_destroy(fnic->io_req_pool);
+	for (i = 0; i < FNIC_SGL_NUM_CACHES; i++)
+		mempool_destroy(fnic->io_sgl_pool[i]);
+
+	return 0;
+}
+
+static void fnic_iounmap(struct fnic *fnic)
+{
+	if (fnic->bar0.vaddr)
+		iounmap(fnic->bar0.vaddr);
+}
+
+/*
+ * Allocate element for mempools requiring GFP_DMA flag.
+ * Otherwise, checks in kmem_flagcheck() hit BUG_ON().
+ */
+static void *fnic_alloc_slab_dma(gfp_t gfp_mask, void *pool_data)
+{
+	struct kmem_cache *mem = pool_data;
+
+	return kmem_cache_alloc(mem, gfp_mask | GFP_ATOMIC | GFP_DMA);
+}
+
+static int __devinit fnic_probe(struct pci_dev *pdev,
+				const struct pci_device_id *ent)
+{
+	struct Scsi_Host *host;
+	struct fc_lport *lp;
+	struct fnic *fnic;
+	mempool_t *pool;
+	int err;
+	int i;
+	unsigned long flags;
+
+	/*
+	 * Allocate SCSI Host and set up association between host,
+	 * local port, and fnic
+	 */
+	host = scsi_host_alloc(&fnic_host_template,
+			       sizeof(struct fc_lport) + sizeof(struct fnic));
+	if (!host) {
+		printk(KERN_ERR PFX "Unable to alloc SCSI host\n");
+		err = -ENOMEM;
+		goto err_out;
+	}
+	lp = shost_priv(host);
+	lp->host = host;
+	fnic = lport_priv(lp);
+	fnic->lport = lp;
+
+	snprintf(fnic->name, sizeof(fnic->name) - 1, "%s%d", DRV_NAME,
+		 host->host_no);
+
+	host->transportt = fnic_fc_transport;
+
+	err = scsi_init_shared_tag_map(host, FNIC_MAX_IO_REQ);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Unable to alloc shared tag map\n");
+		goto err_out_free_hba;
+	}
+
+	/* Setup PCI resources */
+	pci_set_drvdata(pdev, fnic);
+
+	fnic->pdev = pdev;
+
+	err = pci_enable_device(pdev);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Cannot enable PCI device, aborting.\n");
+		goto err_out_free_hba;
+	}
+
+	err = pci_request_regions(pdev, DRV_NAME);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Cannot enable PCI resources, aborting\n");
+		goto err_out_disable_device;
+	}
+
+	pci_set_master(pdev);
+
+	/* Query PCI controller on system for DMA addressing
+	 * limitation for the device.  Try 40-bit first, and
+	 * fail to 32-bit.
+	 */
+	err = pci_set_dma_mask(pdev, DMA_40BIT_MASK);
+	if (err) {
+		err = pci_set_dma_mask(pdev, DMA_32BIT_MASK);
+		if (err) {
+			shost_printk(KERN_ERR, fnic->lport->host,
+				     "No usable DMA configuration "
+				     "aborting\n");
+			goto err_out_release_regions;
+		}
+		err = pci_set_consistent_dma_mask(pdev, DMA_32BIT_MASK);
+		if (err) {
+			shost_printk(KERN_ERR, fnic->lport->host,
+				     "Unable to obtain 32-bit DMA "
+				     "for consistent allocations, aborting.\n");
+			goto err_out_release_regions;
+		}
+	} else {
+		err = pci_set_consistent_dma_mask(pdev, DMA_40BIT_MASK);
+		if (err) {
+			shost_printk(KERN_ERR, fnic->lport->host,
+				     "Unable to obtain 40-bit DMA "
+				     "for consistent allocations, aborting.\n");
+			goto err_out_release_regions;
+		}
+	}
+
+	/* Map vNIC resources from BAR0 */
+	if (!(pci_resource_flags(pdev, 0) & IORESOURCE_MEM)) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "BAR0 not memory-map'able, aborting.\n");
+		err = -ENODEV;
+		goto err_out_release_regions;
+	}
+
+	fnic->bar0.vaddr = pci_iomap(pdev, 0, 0);
+	fnic->bar0.bus_addr = pci_resource_start(pdev, 0);
+	fnic->bar0.len = pci_resource_len(pdev, 0);
+
+	if (!fnic->bar0.vaddr) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Cannot memory-map BAR0 res hdr, "
+			     "aborting.\n");
+		err = -ENODEV;
+		goto err_out_release_regions;
+	}
+
+	fnic->vdev = vnic_dev_register(NULL, fnic, pdev, &fnic->bar0);
+	if (!fnic->vdev) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "vNIC registration failed, "
+			     "aborting.\n");
+		err = -ENODEV;
+		goto err_out_iounmap;
+	}
+
+	err = fnic_dev_wait(fnic->vdev, vnic_dev_open,
+			    vnic_dev_open_done, 0);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "vNIC dev open failed, aborting.\n");
+		goto err_out_vnic_unregister;
+	}
+
+	err = vnic_dev_init(fnic->vdev, 0);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "vNIC dev init failed, aborting.\n");
+		goto err_out_dev_close;
+	}
+
+	err = vnic_dev_mac_addr(fnic->vdev, fnic->mac_addr);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "vNIC get MAC addr failed \n");
+		goto err_out_dev_close;
+	}
+
+	/* Get vNIC configuration */
+	err = fnic_get_vnic_config(fnic);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Get vNIC configuration failed, "
+			     "aborting.\n");
+		goto err_out_dev_close;
+	}
+	host->max_lun = fnic->config.luns_per_tgt;
+	host->max_id = FNIC_MAX_FCP_TARGET;
+
+	fnic_get_res_counts(fnic);
+
+	err = fnic_set_intr_mode(fnic);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Failed to set intr mode, "
+			     "aborting.\n");
+		goto err_out_dev_close;
+	}
+
+	err = fnic_request_intr(fnic);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Unable to request irq.\n");
+		goto err_out_clear_intr;
+	}
+
+	err = fnic_alloc_vnic_resources(fnic);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Failed to alloc vNIC resources, "
+			     "aborting.\n");
+		goto err_out_free_intr;
+	}
+
+
+	/* initialize all fnic locks */
+	spin_lock_init(&fnic->fnic_lock);
+
+	for (i = 0; i < FNIC_WQ_MAX; i++)
+		spin_lock_init(&fnic->wq_lock[i]);
+
+	for (i = 0; i < FNIC_WQ_COPY_MAX; i++) {
+		spin_lock_init(&fnic->wq_copy_lock[i]);
+		fnic->wq_copy_desc_low[i] = DESC_CLEAN_LOW_WATERMARK;
+		fnic->fw_ack_recd[i] = 0;
+		fnic->fw_ack_index[i] = -1;
+	}
+
+	for (i = 0; i < FNIC_IO_LOCKS; i++)
+		spin_lock_init(&fnic->io_req_lock[i]);
+
+	fnic->io_req_pool = mempool_create_slab_pool(2, fnic_io_req_cache);
+	if (!fnic->io_req_pool)
+		goto err_out_free_resources;
+
+	pool = mempool_create(2, fnic_alloc_slab_dma, mempool_free_slab,
+			      fnic_sgl_cache[FNIC_SGL_CACHE_DFLT]);
+	if (!pool)
+		goto err_out_free_ioreq_pool;
+	fnic->io_sgl_pool[FNIC_SGL_CACHE_DFLT] = pool;
+
+	pool = mempool_create(2, fnic_alloc_slab_dma, mempool_free_slab,
+			      fnic_sgl_cache[FNIC_SGL_CACHE_MAX]);
+	if (!pool)
+		goto err_out_free_dflt_pool;
+	fnic->io_sgl_pool[FNIC_SGL_CACHE_MAX] = pool;
+
+	/* setup vlan config, hw inserts vlan header */
+	fnic->vlan_hw_insert = 1;
+	fnic->vlan_id = 0;
+
+	fnic->flogi_oxid = FC_XID_UNKNOWN;
+	fnic->flogi = NULL;
+	fnic->flogi_resp = NULL;
+	fnic->state = FNIC_IN_FC_MODE;
+
+	/* Enable hardware stripping of vlan header on ingress */
+	fnic_set_nic_config(fnic, 0, 0, 0, 0, 0, 0, 1);
+
+	/* Setup notification buffer area */
+	err = fnic_notify_set(fnic);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Failed to alloc notify buffer, aborting.\n");
+		goto err_out_free_max_pool;
+	}
+
+	/* Setup notify timer when using MSI interrupts */
+	if (vnic_dev_get_intr_mode(fnic->vdev) == VNIC_DEV_INTR_MODE_MSI)
+		setup_timer(&fnic->notify_timer,
+			    fnic_notify_timer, (unsigned long)fnic);
+
+	/* allocate RQ buffers and post them to RQ*/
+	for (i = 0; i < fnic->rq_count; i++) {
+		err = vnic_rq_fill(&fnic->rq[i], fnic_alloc_rq_frame);
+		if (err) {
+			shost_printk(KERN_ERR, fnic->lport->host,
+				     "fnic_alloc_rq_frame can't alloc "
+				     "frame\n");
+			goto err_out_free_rq_buf;
+		}
+	}
+
+	/*
+	 * Initialization done with PCI system, hardware, firmware.
+	 * Add host to SCSI
+	 */
+	err = scsi_add_host(lp->host, &pdev->dev);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "fnic: scsi_add_host failed...exiting\n");
+		goto err_out_free_rq_buf;
+	}
+
+	/* Start local port initiatialization */
+
+	lp->link_up = 0;
+	lp->tt = fnic_transport_template;
+
+	lp->emp = fc_exch_mgr_alloc(lp, FC_CLASS_3,
+				    FCPIO_HOST_EXCH_RANGE_START,
+				    FCPIO_HOST_EXCH_RANGE_END);
+	if (!lp->emp) {
+		err = -ENOMEM;
+		goto err_out_remove_scsi_host;
+	}
+
+	lp->max_retry_count = fnic->config.flogi_retries;
+	lp->service_params = (FCP_SPPF_INIT_FCN | FCP_SPPF_RD_XRDY_DIS |
+			      FCP_SPPF_CONF_COMPL);
+	if (fnic->config.flags & VFCF_FCP_SEQ_LVL_ERR)
+		lp->service_params |= FCP_SPPF_RETRY;
+
+	lp->boot_time = jiffies;
+	lp->e_d_tov = fnic->config.ed_tov;
+	lp->r_a_tov = fnic->config.ra_tov;
+	lp->link_supported_speeds = FC_PORTSPEED_10GBIT;
+	fc_set_wwnn(lp, fnic->config.node_wwn);
+	fc_set_wwpn(lp, fnic->config.port_wwn);
+
+	fc_exch_init(lp);
+	fc_lport_init(lp);
+	fc_elsct_init(lp);
+	fc_rport_init(lp);
+	fc_disc_init(lp);
+
+	fc_lport_config(lp);
+
+	if (fc_set_mfs(lp, fnic->config.maxdatafieldsize +
+		       sizeof(struct fc_frame_header))) {
+		err = -EINVAL;
+		goto err_out_free_exch_mgr;
+	}
+	fc_host_maxframe_size(lp->host) = lp->mfs;
+
+	sprintf(fc_host_symbolic_name(lp->host),
+		DRV_NAME " v" DRV_VERSION " over %s", fnic->name);
+
+	spin_lock_irqsave(&fnic_list_lock, flags);
+	list_add_tail(&fnic->list, &fnic_list);
+	spin_unlock_irqrestore(&fnic_list_lock, flags);
+
+	INIT_WORK(&fnic->link_work, fnic_handle_link);
+	INIT_WORK(&fnic->frame_work, fnic_handle_frame);
+	skb_queue_head_init(&fnic->frame_queue);
+
+	/* Enable all queues */
+	for (i = 0; i < fnic->raw_wq_count; i++)
+		vnic_wq_enable(&fnic->wq[i]);
+	for (i = 0; i < fnic->rq_count; i++)
+		vnic_rq_enable(&fnic->rq[i]);
+	for (i = 0; i < fnic->wq_copy_count; i++)
+		vnic_wq_copy_enable(&fnic->wq_copy[i]);
+
+	fc_fabric_login(lp);
+
+	vnic_dev_enable(fnic->vdev);
+	for (i = 0; i < fnic->intr_count; i++)
+		vnic_intr_unmask(&fnic->intr[i]);
+
+	fnic_notify_timer_start(fnic);
+
+	return 0;
+
+err_out_free_exch_mgr:
+	fc_exch_mgr_free(lp->emp);
+err_out_remove_scsi_host:
+	fc_remove_host(fnic->lport->host);
+	scsi_remove_host(fnic->lport->host);
+err_out_free_rq_buf:
+	for (i = 0; i < fnic->rq_count; i++)
+		vnic_rq_clean(&fnic->rq[i], fnic_free_rq_buf);
+	vnic_dev_notify_unset(fnic->vdev);
+err_out_free_max_pool:
+	mempool_destroy(fnic->io_sgl_pool[FNIC_SGL_CACHE_MAX]);
+err_out_free_dflt_pool:
+	mempool_destroy(fnic->io_sgl_pool[FNIC_SGL_CACHE_DFLT]);
+err_out_free_ioreq_pool:
+	mempool_destroy(fnic->io_req_pool);
+err_out_free_resources:
+	fnic_free_vnic_resources(fnic);
+err_out_free_intr:
+	fnic_free_intr(fnic);
+err_out_clear_intr:
+	fnic_clear_intr_mode(fnic);
+err_out_dev_close:
+	vnic_dev_close(fnic->vdev);
+err_out_vnic_unregister:
+	vnic_dev_unregister(fnic->vdev);
+err_out_iounmap:
+	fnic_iounmap(fnic);
+err_out_release_regions:
+	pci_release_regions(pdev);
+err_out_disable_device:
+	pci_disable_device(pdev);
+err_out_free_hba:
+	scsi_host_put(lp->host);
+err_out:
+	return err;
+}
+
+static void __devexit fnic_remove(struct pci_dev *pdev)
+{
+	struct fnic *fnic = pci_get_drvdata(pdev);
+	unsigned long flags;
+
+	/*
+	 * Mark state so that the workqueue thread stops forwarding
+	 * received frames and link events to the local port. ISR and
+	 * other threads that can queue work items will also stop
+	 * creating work items on the fnic workqueue
+	 */
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+	fnic->stop_rx_link_events = 1;
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	if (vnic_dev_get_intr_mode(fnic->vdev) == VNIC_DEV_INTR_MODE_MSI)
+		del_timer_sync(&fnic->notify_timer);
+
+	/*
+	 * Flush the fnic event queue. After this call, there should
+	 * be no event queued for this fnic device in the workqueue
+	 */
+	flush_workqueue(fnic_event_queue);
+	skb_queue_purge(&fnic->frame_queue);
+
+	/*
+	 * Log off the fabric. This stops all remote ports, dns port,
+	 * logs off the fabric. This flushes all rport, disc, lport work
+	 * before returning
+	 */
+	fc_fabric_logoff(fnic->lport);
+
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+	fnic->in_remove = 1;
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	fc_lport_destroy(fnic->lport);
+
+	/*
+	 * This stops the fnic device, masks all interrupts. Completed
+	 * CQ entries are drained. Posted WQ/RQ/Copy-WQ entries are
+	 * cleaned up
+	 */
+	fnic_cleanup(fnic);
+
+	BUG_ON(!skb_queue_empty(&fnic->frame_queue));
+
+	spin_lock_irqsave(&fnic_list_lock, flags);
+	list_del(&fnic->list);
+	spin_unlock_irqrestore(&fnic_list_lock, flags);
+
+	fc_remove_host(fnic->lport->host);
+	scsi_remove_host(fnic->lport->host);
+	fc_exch_mgr_free(fnic->lport->emp);
+	vnic_dev_notify_unset(fnic->vdev);
+	fnic_free_vnic_resources(fnic);
+	fnic_free_intr(fnic);
+	fnic_clear_intr_mode(fnic);
+	vnic_dev_close(fnic->vdev);
+	vnic_dev_unregister(fnic->vdev);
+	fnic_iounmap(fnic);
+	pci_release_regions(pdev);
+	pci_disable_device(pdev);
+	pci_set_drvdata(pdev, NULL);
+	scsi_host_put(fnic->lport->host);
+}
+
+static struct pci_driver fnic_driver = {
+	.name = DRV_NAME,
+	.id_table = fnic_id_table,
+	.probe = fnic_probe,
+	.remove = __devexit_p(fnic_remove),
+};
+
+static int __init fnic_init_module(void)
+{
+	size_t len;
+	int err = 0;
+
+	printk(KERN_INFO PFX "%s, ver %s\n", DRV_DESCRIPTION, DRV_VERSION);
+
+	/* Create a cache for allocation of default size sgls */
+	len = sizeof(struct fnic_dflt_sgl_list);
+	fnic_sgl_cache[FNIC_SGL_CACHE_DFLT] = kmem_cache_create
+		("fnic_sgl_dflt", len + FNIC_SG_DESC_ALIGN, FNIC_SG_DESC_ALIGN,
+		 SLAB_HWCACHE_ALIGN | SLAB_CACHE_DMA,
+		 NULL);
+	if (!fnic_sgl_cache[FNIC_SGL_CACHE_DFLT]) {
+		printk(KERN_ERR PFX "failed to create fnic dflt sgl slab\n");
+		err = -ENOMEM;
+		goto err_create_fnic_sgl_slab_dflt;
+	}
+
+	/* Create a cache for allocation of max size sgls*/
+	len = sizeof(struct fnic_sgl_list);
+	fnic_sgl_cache[FNIC_SGL_CACHE_MAX] = kmem_cache_create
+		("fnic_sgl_max", len + FNIC_SG_DESC_ALIGN, FNIC_SG_DESC_ALIGN,
+		 SLAB_HWCACHE_ALIGN | SLAB_CACHE_DMA,
+		 NULL);
+	if (!fnic_sgl_cache[FNIC_SGL_CACHE_MAX]) {
+		printk(KERN_ERR PFX "failed to create fnic max sgl slab\n");
+		err = -ENOMEM;
+		goto err_create_fnic_sgl_slab_max;
+	}
+
+	/* Create a cache of io_req structs for use via mempool */
+	fnic_io_req_cache = kmem_cache_create("fnic_io_req",
+					      sizeof(struct fnic_io_req),
+					      0, SLAB_HWCACHE_ALIGN, NULL);
+	if (!fnic_io_req_cache) {
+		printk(KERN_ERR PFX "failed to create fnic io_req slab\n");
+		err = -ENOMEM;
+		goto err_create_fnic_ioreq_slab;
+	}
+
+	fnic_event_queue = create_singlethread_workqueue("fnic_event_wq");
+	if (!fnic_event_queue) {
+		printk(KERN_ERR PFX "fnic work queue create failed\n");
+		err = -ENOMEM;
+		goto err_create_fnic_workq;
+	}
+
+	spin_lock_init(&fnic_list_lock);
+	INIT_LIST_HEAD(&fnic_list);
+
+	fnic_fc_transport = fc_attach_transport(&fnic_fc_functions);
+	if (!fnic_fc_transport) {
+		printk(KERN_ERR PFX "fc_attach_transport error\n");
+		err = -ENOMEM;
+		goto err_fc_transport;
+	}
+
+	/* register the driver with PCI system */
+	err = pci_register_driver(&fnic_driver);
+	if (err < 0) {
+		printk(KERN_ERR PFX "pci register error\n");
+		goto err_pci_register;
+	}
+	return err;
+
+err_pci_register:
+	fc_release_transport(fnic_fc_transport);
+err_fc_transport:
+	destroy_workqueue(fnic_event_queue);
+err_create_fnic_workq:
+	kmem_cache_destroy(fnic_io_req_cache);
+err_create_fnic_ioreq_slab:
+	kmem_cache_destroy(fnic_sgl_cache[FNIC_SGL_CACHE_MAX]);
+err_create_fnic_sgl_slab_max:
+	kmem_cache_destroy(fnic_sgl_cache[FNIC_SGL_CACHE_DFLT]);
+err_create_fnic_sgl_slab_dflt:
+	return err;
+}
+
+static void __exit fnic_cleanup_module(void)
+{
+	pci_unregister_driver(&fnic_driver);
+	destroy_workqueue(fnic_event_queue);
+	kmem_cache_destroy(fnic_sgl_cache[FNIC_SGL_CACHE_MAX]);
+	kmem_cache_destroy(fnic_sgl_cache[FNIC_SGL_CACHE_DFLT]);
+	kmem_cache_destroy(fnic_io_req_cache);
+	fc_release_transport(fnic_fc_transport);
+}
+
+module_init(fnic_init_module);
+module_exit(fnic_cleanup_module);
+

drivers/scsi/fnic/fnic_res.c

@@ -0,0 +1,444 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include "wq_enet_desc.h"
+#include "rq_enet_desc.h"
+#include "cq_enet_desc.h"
+#include "vnic_resource.h"
+#include "vnic_dev.h"
+#include "vnic_wq.h"
+#include "vnic_rq.h"
+#include "vnic_cq.h"
+#include "vnic_intr.h"
+#include "vnic_stats.h"
+#include "vnic_nic.h"
+#include "fnic.h"
+
+int fnic_get_vnic_config(struct fnic *fnic)
+{
+	struct vnic_fc_config *c = &fnic->config;
+	int err;
+
+#define GET_CONFIG(m) \
+	do { \
+		err = vnic_dev_spec(fnic->vdev, \
+				    offsetof(struct vnic_fc_config, m), \
+				    sizeof(c->m), &c->m); \
+		if (err) { \
+			shost_printk(KERN_ERR, fnic->lport->host, \
+				     "Error getting %s, %d\n", #m, \
+				     err); \
+			return err; \
+		} \
+	} while (0);
+
+	GET_CONFIG(node_wwn);
+	GET_CONFIG(port_wwn);
+	GET_CONFIG(wq_enet_desc_count);
+	GET_CONFIG(wq_copy_desc_count);
+	GET_CONFIG(rq_desc_count);
+	GET_CONFIG(maxdatafieldsize);
+	GET_CONFIG(ed_tov);
+	GET_CONFIG(ra_tov);
+	GET_CONFIG(intr_timer);
+	GET_CONFIG(intr_timer_type);
+	GET_CONFIG(flags);
+	GET_CONFIG(flogi_retries);
+	GET_CONFIG(flogi_timeout);
+	GET_CONFIG(plogi_retries);
+	GET_CONFIG(plogi_timeout);
+	GET_CONFIG(io_throttle_count);
+	GET_CONFIG(link_down_timeout);
+	GET_CONFIG(port_down_timeout);
+	GET_CONFIG(port_down_io_retries);
+	GET_CONFIG(luns_per_tgt);
+
+	c->wq_enet_desc_count =
+		min_t(u32, VNIC_FNIC_WQ_DESCS_MAX,
+		      max_t(u32, VNIC_FNIC_WQ_DESCS_MIN,
+			    c->wq_enet_desc_count));
+	c->wq_enet_desc_count = ALIGN(c->wq_enet_desc_count, 16);
+
+	c->wq_copy_desc_count =
+		min_t(u32, VNIC_FNIC_WQ_COPY_DESCS_MAX,
+		      max_t(u32, VNIC_FNIC_WQ_COPY_DESCS_MIN,
+			    c->wq_copy_desc_count));
+	c->wq_copy_desc_count = ALIGN(c->wq_copy_desc_count, 16);
+
+	c->rq_desc_count =
+		min_t(u32, VNIC_FNIC_RQ_DESCS_MAX,
+		      max_t(u32, VNIC_FNIC_RQ_DESCS_MIN,
+			    c->rq_desc_count));
+	c->rq_desc_count = ALIGN(c->rq_desc_count, 16);
+
+	c->maxdatafieldsize =
+		min_t(u16, VNIC_FNIC_MAXDATAFIELDSIZE_MAX,
+		      max_t(u16, VNIC_FNIC_MAXDATAFIELDSIZE_MIN,
+			    c->maxdatafieldsize));
+	c->ed_tov =
+		min_t(u32, VNIC_FNIC_EDTOV_MAX,
+		      max_t(u32, VNIC_FNIC_EDTOV_MIN,
+			    c->ed_tov));
+
+	c->ra_tov =
+		min_t(u32, VNIC_FNIC_RATOV_MAX,
+		      max_t(u32, VNIC_FNIC_RATOV_MIN,
+			    c->ra_tov));
+
+	c->flogi_retries =
+		min_t(u32, VNIC_FNIC_FLOGI_RETRIES_MAX, c->flogi_retries);
+
+	c->flogi_timeout =
+		min_t(u32, VNIC_FNIC_FLOGI_TIMEOUT_MAX,
+		      max_t(u32, VNIC_FNIC_FLOGI_TIMEOUT_MIN,
+			    c->flogi_timeout));
+
+	c->plogi_retries =
+		min_t(u32, VNIC_FNIC_PLOGI_RETRIES_MAX, c->plogi_retries);
+
+	c->plogi_timeout =
+		min_t(u32, VNIC_FNIC_PLOGI_TIMEOUT_MAX,
+		      max_t(u32, VNIC_FNIC_PLOGI_TIMEOUT_MIN,
+			    c->plogi_timeout));
+
+	c->io_throttle_count =
+		min_t(u32, VNIC_FNIC_IO_THROTTLE_COUNT_MAX,
+		      max_t(u32, VNIC_FNIC_IO_THROTTLE_COUNT_MIN,
+			    c->io_throttle_count));
+
+	c->link_down_timeout =
+		min_t(u32, VNIC_FNIC_LINK_DOWN_TIMEOUT_MAX,
+		      c->link_down_timeout);
+
+	c->port_down_timeout =
+		min_t(u32, VNIC_FNIC_PORT_DOWN_TIMEOUT_MAX,
+		      c->port_down_timeout);
+
+	c->port_down_io_retries =
+		min_t(u32, VNIC_FNIC_PORT_DOWN_IO_RETRIES_MAX,
+		      c->port_down_io_retries);
+
+	c->luns_per_tgt =
+		min_t(u32, VNIC_FNIC_LUNS_PER_TARGET_MAX,
+		      max_t(u32, VNIC_FNIC_LUNS_PER_TARGET_MIN,
+			    c->luns_per_tgt));
+
+	c->intr_timer = min_t(u16, VNIC_INTR_TIMER_MAX, c->intr_timer);
+	c->intr_timer_type = c->intr_timer_type;
+
+	shost_printk(KERN_INFO, fnic->lport->host,
+		     "vNIC MAC addr %02x:%02x:%02x:%02x:%02x:%02x "
+		     "wq/wq_copy/rq %d/%d/%d\n",
+		     fnic->mac_addr[0], fnic->mac_addr[1], fnic->mac_addr[2],
+		     fnic->mac_addr[3], fnic->mac_addr[4], fnic->mac_addr[5],
+		     c->wq_enet_desc_count, c->wq_copy_desc_count,
+		     c->rq_desc_count);
+	shost_printk(KERN_INFO, fnic->lport->host,
+		     "vNIC node wwn %llx port wwn %llx\n",
+		     c->node_wwn, c->port_wwn);
+	shost_printk(KERN_INFO, fnic->lport->host,
+		     "vNIC ed_tov %d ra_tov %d\n",
+		     c->ed_tov, c->ra_tov);
+	shost_printk(KERN_INFO, fnic->lport->host,
+		     "vNIC mtu %d intr timer %d\n",
+		     c->maxdatafieldsize, c->intr_timer);
+	shost_printk(KERN_INFO, fnic->lport->host,
+		     "vNIC flags 0x%x luns per tgt %d\n",
+		     c->flags, c->luns_per_tgt);
+	shost_printk(KERN_INFO, fnic->lport->host,
+		     "vNIC flogi_retries %d flogi timeout %d\n",
+		     c->flogi_retries, c->flogi_timeout);
+	shost_printk(KERN_INFO, fnic->lport->host,
+		     "vNIC plogi retries %d plogi timeout %d\n",
+		     c->plogi_retries, c->plogi_timeout);
+	shost_printk(KERN_INFO, fnic->lport->host,
+		     "vNIC io throttle count %d link dn timeout %d\n",
+		     c->io_throttle_count, c->link_down_timeout);
+	shost_printk(KERN_INFO, fnic->lport->host,
+		     "vNIC port dn io retries %d port dn timeout %d\n",
+		     c->port_down_io_retries, c->port_down_timeout);
+
+	return 0;
+}
+
+int fnic_set_nic_config(struct fnic *fnic, u8 rss_default_cpu,
+			u8 rss_hash_type,
+			u8 rss_hash_bits, u8 rss_base_cpu, u8 rss_enable,
+			u8 tso_ipid_split_en, u8 ig_vlan_strip_en)
+{
+	u64 a0, a1;
+	u32 nic_cfg;
+	int wait = 1000;
+
+	vnic_set_nic_cfg(&nic_cfg, rss_default_cpu,
+		rss_hash_type, rss_hash_bits, rss_base_cpu,
+		rss_enable, tso_ipid_split_en, ig_vlan_strip_en);
+
+	a0 = nic_cfg;
+	a1 = 0;
+
+	return vnic_dev_cmd(fnic->vdev, CMD_NIC_CFG, &a0, &a1, wait);
+}
+
+void fnic_get_res_counts(struct fnic *fnic)
+{
+	fnic->wq_count = vnic_dev_get_res_count(fnic->vdev, RES_TYPE_WQ);
+	fnic->raw_wq_count = fnic->wq_count - 1;
+	fnic->wq_copy_count = fnic->wq_count - fnic->raw_wq_count;
+	fnic->rq_count = vnic_dev_get_res_count(fnic->vdev, RES_TYPE_RQ);
+	fnic->cq_count = vnic_dev_get_res_count(fnic->vdev, RES_TYPE_CQ);
+	fnic->intr_count = vnic_dev_get_res_count(fnic->vdev,
+		RES_TYPE_INTR_CTRL);
+}
+
+void fnic_free_vnic_resources(struct fnic *fnic)
+{
+	unsigned int i;
+
+	for (i = 0; i < fnic->raw_wq_count; i++)
+		vnic_wq_free(&fnic->wq[i]);
+
+	for (i = 0; i < fnic->wq_copy_count; i++)
+		vnic_wq_copy_free(&fnic->wq_copy[i]);
+
+	for (i = 0; i < fnic->rq_count; i++)
+		vnic_rq_free(&fnic->rq[i]);
+
+	for (i = 0; i < fnic->cq_count; i++)
+		vnic_cq_free(&fnic->cq[i]);
+
+	for (i = 0; i < fnic->intr_count; i++)
+		vnic_intr_free(&fnic->intr[i]);
+}
+
+int fnic_alloc_vnic_resources(struct fnic *fnic)
+{
+	enum vnic_dev_intr_mode intr_mode;
+	unsigned int mask_on_assertion;
+	unsigned int interrupt_offset;
+	unsigned int error_interrupt_enable;
+	unsigned int error_interrupt_offset;
+	unsigned int i, cq_index;
+	unsigned int wq_copy_cq_desc_count;
+	int err;
+
+	intr_mode = vnic_dev_get_intr_mode(fnic->vdev);
+
+	shost_printk(KERN_INFO, fnic->lport->host, "vNIC interrupt mode: %s\n",
+		     intr_mode == VNIC_DEV_INTR_MODE_INTX ? "legacy PCI INTx" :
+		     intr_mode == VNIC_DEV_INTR_MODE_MSI ? "MSI" :
+		     intr_mode == VNIC_DEV_INTR_MODE_MSIX ?
+		     "MSI-X" : "unknown");
+
+	shost_printk(KERN_INFO, fnic->lport->host, "vNIC resources avail: "
+		     "wq %d cp_wq %d raw_wq %d rq %d cq %d intr %d\n",
+		     fnic->wq_count, fnic->wq_copy_count, fnic->raw_wq_count,
+		     fnic->rq_count, fnic->cq_count, fnic->intr_count);
+
+	/* Allocate Raw WQ used for FCS frames */
+	for (i = 0; i < fnic->raw_wq_count; i++) {
+		err = vnic_wq_alloc(fnic->vdev, &fnic->wq[i], i,
+			fnic->config.wq_enet_desc_count,
+			sizeof(struct wq_enet_desc));
+		if (err)
+			goto err_out_cleanup;
+	}
+
+	/* Allocate Copy WQs used for SCSI IOs */
+	for (i = 0; i < fnic->wq_copy_count; i++) {
+		err = vnic_wq_copy_alloc(fnic->vdev, &fnic->wq_copy[i],
+			(fnic->raw_wq_count + i),
+			fnic->config.wq_copy_desc_count,
+			sizeof(struct fcpio_host_req));
+		if (err)
+			goto err_out_cleanup;
+	}
+
+	/* RQ for receiving FCS frames */
+	for (i = 0; i < fnic->rq_count; i++) {
+		err = vnic_rq_alloc(fnic->vdev, &fnic->rq[i], i,
+			fnic->config.rq_desc_count,
+			sizeof(struct rq_enet_desc));
+		if (err)
+			goto err_out_cleanup;
+	}
+
+	/* CQ for each RQ */
+	for (i = 0; i < fnic->rq_count; i++) {
+		cq_index = i;
+		err = vnic_cq_alloc(fnic->vdev,
+			&fnic->cq[cq_index], cq_index,
+			fnic->config.rq_desc_count,
+			sizeof(struct cq_enet_rq_desc));
+		if (err)
+			goto err_out_cleanup;
+	}
+
+	/* CQ for each WQ */
+	for (i = 0; i < fnic->raw_wq_count; i++) {
+		cq_index = fnic->rq_count + i;
+		err = vnic_cq_alloc(fnic->vdev, &fnic->cq[cq_index], cq_index,
+			fnic->config.wq_enet_desc_count,
+			sizeof(struct cq_enet_wq_desc));
+		if (err)
+			goto err_out_cleanup;
+	}
+
+	/* CQ for each COPY WQ */
+	wq_copy_cq_desc_count = (fnic->config.wq_copy_desc_count * 3);
+	for (i = 0; i < fnic->wq_copy_count; i++) {
+		cq_index = fnic->raw_wq_count + fnic->rq_count + i;
+		err = vnic_cq_alloc(fnic->vdev, &fnic->cq[cq_index],
+			cq_index,
+			wq_copy_cq_desc_count,
+			sizeof(struct fcpio_fw_req));
+		if (err)
+			goto err_out_cleanup;
+	}
+
+	for (i = 0; i < fnic->intr_count; i++) {
+		err = vnic_intr_alloc(fnic->vdev, &fnic->intr[i], i);
+		if (err)
+			goto err_out_cleanup;
+	}
+
+	fnic->legacy_pba = vnic_dev_get_res(fnic->vdev,
+				RES_TYPE_INTR_PBA_LEGACY, 0);
+
+	if (!fnic->legacy_pba && intr_mode == VNIC_DEV_INTR_MODE_INTX) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Failed to hook legacy pba resource\n");
+		err = -ENODEV;
+		goto err_out_cleanup;
+	}
+
+	/*
+	 * Init RQ/WQ resources.
+	 *
+	 * RQ[0 to n-1] point to CQ[0 to n-1]
+	 * WQ[0 to m-1] point to CQ[n to n+m-1]
+	 * WQ_COPY[0 to k-1] points to CQ[n+m to n+m+k-1]
+	 *
+	 * Note for copy wq we always initialize with cq_index = 0
+	 *
+	 * Error interrupt is not enabled for MSI.
+	 */
+
+	switch (intr_mode) {
+	case VNIC_DEV_INTR_MODE_INTX:
+	case VNIC_DEV_INTR_MODE_MSIX:
+		error_interrupt_enable = 1;
+		error_interrupt_offset = fnic->err_intr_offset;
+		break;
+	default:
+		error_interrupt_enable = 0;
+		error_interrupt_offset = 0;
+		break;
+	}
+
+	for (i = 0; i < fnic->rq_count; i++) {
+		cq_index = i;
+		vnic_rq_init(&fnic->rq[i],
+			     cq_index,
+			     error_interrupt_enable,
+			     error_interrupt_offset);
+	}
+
+	for (i = 0; i < fnic->raw_wq_count; i++) {
+		cq_index = i + fnic->rq_count;
+		vnic_wq_init(&fnic->wq[i],
+			     cq_index,
+			     error_interrupt_enable,
+			     error_interrupt_offset);
+	}
+
+	for (i = 0; i < fnic->wq_copy_count; i++) {
+		vnic_wq_copy_init(&fnic->wq_copy[i],
+				  0 /* cq_index 0 - always */,
+				  error_interrupt_enable,
+				  error_interrupt_offset);
+	}
+
+	for (i = 0; i < fnic->cq_count; i++) {
+
+		switch (intr_mode) {
+		case VNIC_DEV_INTR_MODE_MSIX:
+			interrupt_offset = i;
+			break;
+		default:
+			interrupt_offset = 0;
+			break;
+		}
+
+		vnic_cq_init(&fnic->cq[i],
+			0 /* flow_control_enable */,
+			1 /* color_enable */,
+			0 /* cq_head */,
+			0 /* cq_tail */,
+			1 /* cq_tail_color */,
+			1 /* interrupt_enable */,
+			1 /* cq_entry_enable */,
+			0 /* cq_message_enable */,
+			interrupt_offset,
+			0 /* cq_message_addr */);
+	}
+
+	/*
+	 * Init INTR resources
+	 *
+	 * mask_on_assertion is not used for INTx due to the level-
+	 * triggered nature of INTx
+	 */
+
+	switch (intr_mode) {
+	case VNIC_DEV_INTR_MODE_MSI:
+	case VNIC_DEV_INTR_MODE_MSIX:
+		mask_on_assertion = 1;
+		break;
+	default:
+		mask_on_assertion = 0;
+		break;
+	}
+
+	for (i = 0; i < fnic->intr_count; i++) {
+		vnic_intr_init(&fnic->intr[i],
+			fnic->config.intr_timer,
+			fnic->config.intr_timer_type,
+			mask_on_assertion);
+	}
+
+	/* init the stats memory by making the first call here */
+	err = vnic_dev_stats_dump(fnic->vdev, &fnic->stats);
+	if (err) {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "vnic_dev_stats_dump failed - x%x\n", err);
+		goto err_out_cleanup;
+	}
+
+	/* Clear LIF stats */
+	vnic_dev_stats_clear(fnic->vdev);
+
+	return 0;
+
+err_out_cleanup:
+	fnic_free_vnic_resources(fnic);
+
+	return err;
+}

drivers/scsi/fnic/fnic_res.h

@@ -0,0 +1,197 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _FNIC_RES_H_
+#define _FNIC_RES_H_
+
+#include "wq_enet_desc.h"
+#include "rq_enet_desc.h"
+#include "vnic_wq.h"
+#include "vnic_rq.h"
+#include "fnic_io.h"
+#include "fcpio.h"
+#include "vnic_wq_copy.h"
+#include "vnic_cq_copy.h"
+
+static inline void fnic_queue_wq_desc(struct vnic_wq *wq,
+				      void *os_buf, dma_addr_t dma_addr,
+				      unsigned int len, unsigned int fc_eof,
+				      int vlan_tag_insert,
+				      unsigned int vlan_tag,
+				      int cq_entry, int sop, int eop)
+{
+	struct wq_enet_desc *desc = vnic_wq_next_desc(wq);
+
+	wq_enet_desc_enc(desc,
+			 (u64)dma_addr | VNIC_PADDR_TARGET,
+			 (u16)len,
+			 0, /* mss_or_csum_offset */
+			 (u16)fc_eof,
+			 0, /* offload_mode */
+			 (u8)eop, (u8)cq_entry,
+			 1, /* fcoe_encap */
+			 (u8)vlan_tag_insert,
+			 (u16)vlan_tag,
+			 0 /* loopback */);
+
+	vnic_wq_post(wq, os_buf, dma_addr, len, sop, eop);
+}
+
+static inline void fnic_queue_wq_copy_desc_icmnd_16(struct vnic_wq_copy *wq,
+						    u32 req_id,
+						    u32 lunmap_id, u8 spl_flags,
+						    u32 sgl_cnt, u32 sense_len,
+						    u64 sgl_addr, u64 sns_addr,
+						    u8 crn, u8 pri_ta,
+						    u8 flags, u8 *scsi_cdb,
+						    u32 data_len, u8 *lun,
+						    u32 d_id, u16 mss,
+						    u32 ratov, u32 edtov)
+{
+	struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq);
+
+	desc->hdr.type = FCPIO_ICMND_16; /* enum fcpio_type */
+	desc->hdr.status = 0;            /* header status entry */
+	desc->hdr._resvd = 0;            /* reserved */
+	desc->hdr.tag.u.req_id = req_id; /* id for this request */
+
+	desc->u.icmnd_16.lunmap_id = lunmap_id; /* index into lunmap table */
+	desc->u.icmnd_16.special_req_flags = spl_flags; /* exch req flags */
+	desc->u.icmnd_16._resvd0[0] = 0;        /* reserved */
+	desc->u.icmnd_16._resvd0[1] = 0;        /* reserved */
+	desc->u.icmnd_16._resvd0[2] = 0;        /* reserved */
+	desc->u.icmnd_16.sgl_cnt = sgl_cnt;     /* scatter-gather list count */
+	desc->u.icmnd_16.sense_len = sense_len; /* sense buffer length */
+	desc->u.icmnd_16.sgl_addr = sgl_addr;   /* scatter-gather list addr */
+	desc->u.icmnd_16.sense_addr = sns_addr; /* sense buffer address */
+	desc->u.icmnd_16.crn = crn;             /* SCSI Command Reference No.*/
+	desc->u.icmnd_16.pri_ta = pri_ta; 	/* SCSI Pri & Task attribute */
+	desc->u.icmnd_16._resvd1 = 0;           /* reserved: should be 0 */
+	desc->u.icmnd_16.flags = flags;         /* command flags */
+	memcpy(desc->u.icmnd_16.scsi_cdb, scsi_cdb, CDB_16);    /* SCSI CDB */
+	desc->u.icmnd_16.data_len = data_len;   /* length of data expected */
+	memcpy(desc->u.icmnd_16.lun, lun, LUN_ADDRESS);  /* LUN address */
+	desc->u.icmnd_16._resvd2 = 0;          	/* reserved */
+	hton24(desc->u.icmnd_16.d_id, d_id);	/* FC vNIC only: Target D_ID */
+	desc->u.icmnd_16.mss = mss;            	/* FC vNIC only: max burst */
+	desc->u.icmnd_16.r_a_tov = ratov; /*FC vNIC only: Res. Alloc Timeout */
+	desc->u.icmnd_16.e_d_tov = edtov; /*FC vNIC only: Err Detect Timeout */
+
+	vnic_wq_copy_post(wq);
+}
+
+static inline void fnic_queue_wq_copy_desc_itmf(struct vnic_wq_copy *wq,
+						u32 req_id, u32 lunmap_id,
+						u32 tm_req, u32 tm_id, u8 *lun,
+						u32 d_id, u32 r_a_tov,
+						u32 e_d_tov)
+{
+	struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq);
+
+	desc->hdr.type = FCPIO_ITMF;     /* enum fcpio_type */
+	desc->hdr.status = 0;            /* header status entry */
+	desc->hdr._resvd = 0;            /* reserved */
+	desc->hdr.tag.u.req_id = req_id; /* id for this request */
+
+	desc->u.itmf.lunmap_id = lunmap_id; /* index into lunmap table */
+	desc->u.itmf.tm_req = tm_req;       /* SCSI Task Management request */
+	desc->u.itmf.t_tag = tm_id;         /* tag of fcpio to be aborted */
+	desc->u.itmf._resvd = 0;
+	memcpy(desc->u.itmf.lun, lun, LUN_ADDRESS);  /* LUN address */
+	desc->u.itmf._resvd1 = 0;
+	hton24(desc->u.itmf.d_id, d_id);    /* FC vNIC only: Target D_ID */
+	desc->u.itmf.r_a_tov = r_a_tov;     /* FC vNIC only: R_A_TOV in msec */
+	desc->u.itmf.e_d_tov = e_d_tov;     /* FC vNIC only: E_D_TOV in msec */
+
+	vnic_wq_copy_post(wq);
+}
+
+static inline void fnic_queue_wq_copy_desc_flogi_reg(struct vnic_wq_copy *wq,
+						     u32 req_id, u8 format,
+						     u32 s_id, u8 *gw_mac)
+{
+	struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq);
+
+	desc->hdr.type = FCPIO_FLOGI_REG;     /* enum fcpio_type */
+	desc->hdr.status = 0;                 /* header status entry */
+	desc->hdr._resvd = 0;                 /* reserved */
+	desc->hdr.tag.u.req_id = req_id;      /* id for this request */
+
+	desc->u.flogi_reg.format = format;
+	hton24(desc->u.flogi_reg.s_id, s_id);
+	memcpy(desc->u.flogi_reg.gateway_mac, gw_mac, ETH_ALEN);
+
+	vnic_wq_copy_post(wq);
+}
+
+static inline void fnic_queue_wq_copy_desc_fw_reset(struct vnic_wq_copy *wq,
+						    u32 req_id)
+{
+	struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq);
+
+	desc->hdr.type = FCPIO_RESET;     /* enum fcpio_type */
+	desc->hdr.status = 0;             /* header status entry */
+	desc->hdr._resvd = 0;             /* reserved */
+	desc->hdr.tag.u.req_id = req_id;  /* id for this request */
+
+	vnic_wq_copy_post(wq);
+}
+
+static inline void fnic_queue_wq_copy_desc_lunmap(struct vnic_wq_copy *wq,
+						  u32 req_id, u64 lunmap_addr,
+						  u32 lunmap_len)
+{
+	struct fcpio_host_req *desc = vnic_wq_copy_next_desc(wq);
+
+	desc->hdr.type = FCPIO_LUNMAP_REQ;	/* enum fcpio_type */
+	desc->hdr.status = 0;			/* header status entry */
+	desc->hdr._resvd = 0;			/* reserved */
+	desc->hdr.tag.u.req_id = req_id;	/* id for this request */
+
+	desc->u.lunmap_req.addr = lunmap_addr;	/* address of the buffer */
+	desc->u.lunmap_req.len = lunmap_len;	/* len of the buffer */
+
+	vnic_wq_copy_post(wq);
+}
+
+static inline void fnic_queue_rq_desc(struct vnic_rq *rq,
+				      void *os_buf, dma_addr_t dma_addr,
+				      u16 len)
+{
+	struct rq_enet_desc *desc = vnic_rq_next_desc(rq);
+
+	rq_enet_desc_enc(desc,
+		(u64)dma_addr | VNIC_PADDR_TARGET,
+		RQ_ENET_TYPE_ONLY_SOP,
+		(u16)len);
+
+	vnic_rq_post(rq, os_buf, 0, dma_addr, len);
+}
+
+
+struct fnic;
+
+int fnic_get_vnic_config(struct fnic *);
+int fnic_alloc_vnic_resources(struct fnic *);
+void fnic_free_vnic_resources(struct fnic *);
+void fnic_get_res_counts(struct fnic *);
+int fnic_set_nic_config(struct fnic *fnic, u8 rss_default_cpu,
+			u8 rss_hash_type, u8 rss_hash_bits, u8 rss_base_cpu,
+			u8 rss_enable, u8 tso_ipid_split_en,
+			u8 ig_vlan_strip_en);
+
+#endif /* _FNIC_RES_H_ */

drivers/scsi/fnic/fnic_scsi.c

@@ -0,0 +1,1850 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/mempool.h>
+#include <linux/errno.h>
+#include <linux/init.h>
+#include <linux/workqueue.h>
+#include <linux/pci.h>
+#include <linux/scatterlist.h>
+#include <linux/skbuff.h>
+#include <linux/spinlock.h>
+#include <linux/if_ether.h>
+#include <linux/if_vlan.h>
+#include <linux/delay.h>
+#include <scsi/scsi.h>
+#include <scsi/scsi_host.h>
+#include <scsi/scsi_device.h>
+#include <scsi/scsi_cmnd.h>
+#include <scsi/scsi_tcq.h>
+#include <scsi/fc/fc_els.h>
+#include <scsi/fc/fc_fcoe.h>
+#include <scsi/libfc.h>
+#include <scsi/fc_frame.h>
+#include "fnic_io.h"
+#include "fnic.h"
+
+const char *fnic_state_str[] = {
+	[FNIC_IN_FC_MODE] =           "FNIC_IN_FC_MODE",
+	[FNIC_IN_FC_TRANS_ETH_MODE] = "FNIC_IN_FC_TRANS_ETH_MODE",
+	[FNIC_IN_ETH_MODE] =          "FNIC_IN_ETH_MODE",
+	[FNIC_IN_ETH_TRANS_FC_MODE] = "FNIC_IN_ETH_TRANS_FC_MODE",
+};
+
+static const char *fnic_ioreq_state_str[] = {
+	[FNIC_IOREQ_CMD_PENDING] = "FNIC_IOREQ_CMD_PENDING",
+	[FNIC_IOREQ_ABTS_PENDING] = "FNIC_IOREQ_ABTS_PENDING",
+	[FNIC_IOREQ_ABTS_COMPLETE] = "FNIC_IOREQ_ABTS_COMPLETE",
+	[FNIC_IOREQ_CMD_COMPLETE] = "FNIC_IOREQ_CMD_COMPLETE",
+};
+
+static const char *fcpio_status_str[] =  {
+	[FCPIO_SUCCESS] = "FCPIO_SUCCESS", /*0x0*/
+	[FCPIO_INVALID_HEADER] = "FCPIO_INVALID_HEADER",
+	[FCPIO_OUT_OF_RESOURCE] = "FCPIO_OUT_OF_RESOURCE",
+	[FCPIO_INVALID_PARAM] = "FCPIO_INVALID_PARAM]",
+	[FCPIO_REQ_NOT_SUPPORTED] = "FCPIO_REQ_NOT_SUPPORTED",
+	[FCPIO_IO_NOT_FOUND] = "FCPIO_IO_NOT_FOUND",
+	[FCPIO_ABORTED] = "FCPIO_ABORTED", /*0x41*/
+	[FCPIO_TIMEOUT] = "FCPIO_TIMEOUT",
+	[FCPIO_SGL_INVALID] = "FCPIO_SGL_INVALID",
+	[FCPIO_MSS_INVALID] = "FCPIO_MSS_INVALID",
+	[FCPIO_DATA_CNT_MISMATCH] = "FCPIO_DATA_CNT_MISMATCH",
+	[FCPIO_FW_ERR] = "FCPIO_FW_ERR",
+	[FCPIO_ITMF_REJECTED] = "FCPIO_ITMF_REJECTED",
+	[FCPIO_ITMF_FAILED] = "FCPIO_ITMF_FAILED",
+	[FCPIO_ITMF_INCORRECT_LUN] = "FCPIO_ITMF_INCORRECT_LUN",
+	[FCPIO_CMND_REJECTED] = "FCPIO_CMND_REJECTED",
+	[FCPIO_NO_PATH_AVAIL] = "FCPIO_NO_PATH_AVAIL",
+	[FCPIO_PATH_FAILED] = "FCPIO_PATH_FAILED",
+	[FCPIO_LUNMAP_CHNG_PEND] = "FCPIO_LUNHMAP_CHNG_PEND",
+};
+
+const char *fnic_state_to_str(unsigned int state)
+{
+	if (state >= ARRAY_SIZE(fnic_state_str) || !fnic_state_str[state])
+		return "unknown";
+
+	return fnic_state_str[state];
+}
+
+static const char *fnic_ioreq_state_to_str(unsigned int state)
+{
+	if (state >= ARRAY_SIZE(fnic_ioreq_state_str) ||
+	    !fnic_ioreq_state_str[state])
+		return "unknown";
+
+	return fnic_ioreq_state_str[state];
+}
+
+static const char *fnic_fcpio_status_to_str(unsigned int status)
+{
+	if (status >= ARRAY_SIZE(fcpio_status_str) || !fcpio_status_str[status])
+		return "unknown";
+
+	return fcpio_status_str[status];
+}
+
+static void fnic_cleanup_io(struct fnic *fnic, int exclude_id);
+
+static inline spinlock_t *fnic_io_lock_hash(struct fnic *fnic,
+					    struct scsi_cmnd *sc)
+{
+	u32 hash = sc->request->tag & (FNIC_IO_LOCKS - 1);
+
+	return &fnic->io_req_lock[hash];
+}
+
+/*
+ * Unmap the data buffer and sense buffer for an io_req,
+ * also unmap and free the device-private scatter/gather list.
+ */
+static void fnic_release_ioreq_buf(struct fnic *fnic,
+				   struct fnic_io_req *io_req,
+				   struct scsi_cmnd *sc)
+{
+	if (io_req->sgl_list_pa)
+		pci_unmap_single(fnic->pdev, io_req->sgl_list_pa,
+				 sizeof(io_req->sgl_list[0]) * io_req->sgl_cnt,
+				 PCI_DMA_TODEVICE);
+	scsi_dma_unmap(sc);
+
+	if (io_req->sgl_cnt)
+		mempool_free(io_req->sgl_list_alloc,
+			     fnic->io_sgl_pool[io_req->sgl_type]);
+	if (io_req->sense_buf_pa)
+		pci_unmap_single(fnic->pdev, io_req->sense_buf_pa,
+				 SCSI_SENSE_BUFFERSIZE, PCI_DMA_FROMDEVICE);
+}
+
+/* Free up Copy Wq descriptors. Called with copy_wq lock held */
+static int free_wq_copy_descs(struct fnic *fnic, struct vnic_wq_copy *wq)
+{
+	/* if no Ack received from firmware, then nothing to clean */
+	if (!fnic->fw_ack_recd[0])
+		return 1;
+
+	/*
+	 * Update desc_available count based on number of freed descriptors
+	 * Account for wraparound
+	 */
+	if (wq->to_clean_index <= fnic->fw_ack_index[0])
+		wq->ring.desc_avail += (fnic->fw_ack_index[0]
+					- wq->to_clean_index + 1);
+	else
+		wq->ring.desc_avail += (wq->ring.desc_count
+					- wq->to_clean_index
+					+ fnic->fw_ack_index[0] + 1);
+
+	/*
+	 * just bump clean index to ack_index+1 accounting for wraparound
+	 * this will essentially free up all descriptors between
+	 * to_clean_index and fw_ack_index, both inclusive
+	 */
+	wq->to_clean_index =
+		(fnic->fw_ack_index[0] + 1) % wq->ring.desc_count;
+
+	/* we have processed the acks received so far */
+	fnic->fw_ack_recd[0] = 0;
+	return 0;
+}
+
+
+/*
+ * fnic_fw_reset_handler
+ * Routine to send reset msg to fw
+ */
+int fnic_fw_reset_handler(struct fnic *fnic)
+{
+	struct vnic_wq_copy *wq = &fnic->wq_copy[0];
+	int ret = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&fnic->wq_copy_lock[0], flags);
+
+	if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0])
+		free_wq_copy_descs(fnic, wq);
+
+	if (!vnic_wq_copy_desc_avail(wq))
+		ret = -EAGAIN;
+	else
+		fnic_queue_wq_copy_desc_fw_reset(wq, SCSI_NO_TAG);
+
+	spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags);
+
+	if (!ret)
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "Issued fw reset\n");
+	else
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "Failed to issue fw reset\n");
+	return ret;
+}
+
+
+/*
+ * fnic_flogi_reg_handler
+ * Routine to send flogi register msg to fw
+ */
+int fnic_flogi_reg_handler(struct fnic *fnic)
+{
+	struct vnic_wq_copy *wq = &fnic->wq_copy[0];
+	u8 gw_mac[ETH_ALEN];
+	int ret = 0;
+	unsigned long flags;
+
+	spin_lock_irqsave(&fnic->wq_copy_lock[0], flags);
+
+	if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0])
+		free_wq_copy_descs(fnic, wq);
+
+	if (!vnic_wq_copy_desc_avail(wq)) {
+		ret = -EAGAIN;
+		goto flogi_reg_ioreq_end;
+	}
+
+	if (fnic->fcoui_mode)
+		memset(gw_mac, 0xff, ETH_ALEN);
+	else
+		memcpy(gw_mac, fnic->dest_addr, ETH_ALEN);
+
+	fnic_queue_wq_copy_desc_flogi_reg(wq, SCSI_NO_TAG,
+					  FCPIO_FLOGI_REG_GW_DEST,
+					  fnic->s_id,
+					  gw_mac);
+
+flogi_reg_ioreq_end:
+	spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags);
+
+	if (!ret)
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "flog reg issued\n");
+
+	return ret;
+}
+
+/*
+ * fnic_queue_wq_copy_desc
+ * Routine to enqueue a wq copy desc
+ */
+static inline int fnic_queue_wq_copy_desc(struct fnic *fnic,
+					  struct vnic_wq_copy *wq,
+					  struct fnic_io_req *io_req,
+					  struct scsi_cmnd *sc,
+					  u32 sg_count)
+{
+	struct scatterlist *sg;
+	struct fc_rport *rport = starget_to_rport(scsi_target(sc->device));
+	struct fc_rport_libfc_priv *rp = rport->dd_data;
+	struct host_sg_desc *desc;
+	u8 pri_tag = 0;
+	unsigned int i;
+	unsigned long intr_flags;
+	int flags;
+	u8 exch_flags;
+	struct scsi_lun fc_lun;
+	char msg[2];
+
+	if (sg_count) {
+		BUG_ON(sg_count < 0);
+		BUG_ON(sg_count > FNIC_MAX_SG_DESC_CNT);
+
+		/* For each SGE, create a device desc entry */
+		desc = io_req->sgl_list;
+		for_each_sg(scsi_sglist(sc), sg, sg_count, i) {
+			desc->addr = cpu_to_le64(sg_dma_address(sg));
+			desc->len = cpu_to_le32(sg_dma_len(sg));
+			desc->_resvd = 0;
+			desc++;
+		}
+
+		io_req->sgl_list_pa = pci_map_single
+			(fnic->pdev,
+			 io_req->sgl_list,
+			 sizeof(io_req->sgl_list[0]) * sg_count,
+			 PCI_DMA_TODEVICE);
+	}
+
+	io_req->sense_buf_pa = pci_map_single(fnic->pdev,
+					      sc->sense_buffer,
+					      SCSI_SENSE_BUFFERSIZE,
+					      PCI_DMA_FROMDEVICE);
+
+	int_to_scsilun(sc->device->lun, &fc_lun);
+
+	pri_tag = FCPIO_ICMND_PTA_SIMPLE;
+	msg[0] = MSG_SIMPLE_TAG;
+	scsi_populate_tag_msg(sc, msg);
+	if (msg[0] == MSG_ORDERED_TAG)
+		pri_tag = FCPIO_ICMND_PTA_ORDERED;
+
+	/* Enqueue the descriptor in the Copy WQ */
+	spin_lock_irqsave(&fnic->wq_copy_lock[0], intr_flags);
+
+	if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0])
+		free_wq_copy_descs(fnic, wq);
+
+	if (unlikely(!vnic_wq_copy_desc_avail(wq))) {
+		spin_unlock_irqrestore(&fnic->wq_copy_lock[0], intr_flags);
+		return SCSI_MLQUEUE_HOST_BUSY;
+	}
+
+	flags = 0;
+	if (sc->sc_data_direction == DMA_FROM_DEVICE)
+		flags = FCPIO_ICMND_RDDATA;
+	else if (sc->sc_data_direction == DMA_TO_DEVICE)
+		flags = FCPIO_ICMND_WRDATA;
+
+	exch_flags = 0;
+	if ((fnic->config.flags & VFCF_FCP_SEQ_LVL_ERR) &&
+	    (rp->flags & FC_RP_FLAGS_RETRY))
+		exch_flags |= FCPIO_ICMND_SRFLAG_RETRY;
+
+	fnic_queue_wq_copy_desc_icmnd_16(wq, sc->request->tag,
+					 0, exch_flags, io_req->sgl_cnt,
+					 SCSI_SENSE_BUFFERSIZE,
+					 io_req->sgl_list_pa,
+					 io_req->sense_buf_pa,
+					 0, /* scsi cmd ref, always 0 */
+					 pri_tag, /* scsi pri and tag */
+					 flags,	/* command flags */
+					 sc->cmnd, scsi_bufflen(sc),
+					 fc_lun.scsi_lun, io_req->port_id,
+					 rport->maxframe_size, rp->r_a_tov,
+					 rp->e_d_tov);
+
+	spin_unlock_irqrestore(&fnic->wq_copy_lock[0], intr_flags);
+	return 0;
+}
+
+/*
+ * fnic_queuecommand
+ * Routine to send a scsi cdb
+ * Called with host_lock held and interrupts disabled.
+ */
+int fnic_queuecommand(struct scsi_cmnd *sc, void (*done)(struct scsi_cmnd *))
+{
+	struct fc_lport *lp;
+	struct fc_rport *rport;
+	struct fnic_io_req *io_req;
+	struct fnic *fnic;
+	struct vnic_wq_copy *wq;
+	int ret;
+	u32 sg_count;
+	unsigned long flags;
+	unsigned long ptr;
+
+	rport = starget_to_rport(scsi_target(sc->device));
+	ret = fc_remote_port_chkready(rport);
+	if (ret) {
+		sc->result = ret;
+		done(sc);
+		return 0;
+	}
+
+	lp = shost_priv(sc->device->host);
+	if (lp->state != LPORT_ST_READY || !(lp->link_up))
+		return SCSI_MLQUEUE_HOST_BUSY;
+
+	/*
+	 * Release host lock, use driver resource specific locks from here.
+	 * Don't re-enable interrupts in case they were disabled prior to the
+	 * caller disabling them.
+	 */
+	spin_unlock(lp->host->host_lock);
+
+	/* Get a new io_req for this SCSI IO */
+	fnic = lport_priv(lp);
+
+	io_req = mempool_alloc(fnic->io_req_pool, GFP_ATOMIC);
+	if (!io_req) {
+		ret = SCSI_MLQUEUE_HOST_BUSY;
+		goto out;
+	}
+	memset(io_req, 0, sizeof(*io_req));
+
+	/* Map the data buffer */
+	sg_count = scsi_dma_map(sc);
+	if (sg_count < 0) {
+		mempool_free(io_req, fnic->io_req_pool);
+		goto out;
+	}
+
+	/* Determine the type of scatter/gather list we need */
+	io_req->sgl_cnt = sg_count;
+	io_req->sgl_type = FNIC_SGL_CACHE_DFLT;
+	if (sg_count > FNIC_DFLT_SG_DESC_CNT)
+		io_req->sgl_type = FNIC_SGL_CACHE_MAX;
+
+	if (sg_count) {
+		io_req->sgl_list =
+			mempool_alloc(fnic->io_sgl_pool[io_req->sgl_type],
+				      GFP_ATOMIC | GFP_DMA);
+		if (!io_req->sgl_list) {
+			ret = SCSI_MLQUEUE_HOST_BUSY;
+			scsi_dma_unmap(sc);
+			mempool_free(io_req, fnic->io_req_pool);
+			goto out;
+		}
+
+		/* Cache sgl list allocated address before alignment */
+		io_req->sgl_list_alloc = io_req->sgl_list;
+		ptr = (unsigned long) io_req->sgl_list;
+		if (ptr % FNIC_SG_DESC_ALIGN) {
+			io_req->sgl_list = (struct host_sg_desc *)
+				(((unsigned long) ptr
+				  + FNIC_SG_DESC_ALIGN - 1)
+				 & ~(FNIC_SG_DESC_ALIGN - 1));
+		}
+	}
+
+	/* initialize rest of io_req */
+	io_req->port_id = rport->port_id;
+	CMD_STATE(sc) = FNIC_IOREQ_CMD_PENDING;
+	CMD_SP(sc) = (char *)io_req;
+	sc->scsi_done = done;
+
+	/* create copy wq desc and enqueue it */
+	wq = &fnic->wq_copy[0];
+	ret = fnic_queue_wq_copy_desc(fnic, wq, io_req, sc, sg_count);
+	if (ret) {
+		/*
+		 * In case another thread cancelled the request,
+		 * refetch the pointer under the lock.
+		 */
+		spinlock_t *io_lock = fnic_io_lock_hash(fnic, sc);
+
+		spin_lock_irqsave(io_lock, flags);
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+		CMD_SP(sc) = NULL;
+		CMD_STATE(sc) = FNIC_IOREQ_CMD_COMPLETE;
+		spin_unlock_irqrestore(io_lock, flags);
+		if (io_req) {
+			fnic_release_ioreq_buf(fnic, io_req, sc);
+			mempool_free(io_req, fnic->io_req_pool);
+		}
+	}
+out:
+	/* acquire host lock before returning to SCSI */
+	spin_lock(lp->host->host_lock);
+	return ret;
+}
+
+/*
+ * fnic_fcpio_fw_reset_cmpl_handler
+ * Routine to handle fw reset completion
+ */
+static int fnic_fcpio_fw_reset_cmpl_handler(struct fnic *fnic,
+					    struct fcpio_fw_req *desc)
+{
+	u8 type;
+	u8 hdr_status;
+	struct fcpio_tag tag;
+	int ret = 0;
+	struct fc_frame *flogi;
+	unsigned long flags;
+
+	fcpio_header_dec(&desc->hdr, &type, &hdr_status, &tag);
+
+	/* Clean up all outstanding io requests */
+	fnic_cleanup_io(fnic, SCSI_NO_TAG);
+
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+
+	flogi = fnic->flogi;
+	fnic->flogi = NULL;
+
+	/* fnic should be in FC_TRANS_ETH_MODE */
+	if (fnic->state == FNIC_IN_FC_TRANS_ETH_MODE) {
+		/* Check status of reset completion */
+		if (!hdr_status) {
+			FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+				      "reset cmpl success\n");
+			/* Ready to send flogi out */
+			fnic->state = FNIC_IN_ETH_MODE;
+		} else {
+			FNIC_SCSI_DBG(KERN_DEBUG,
+				      fnic->lport->host,
+				      "fnic fw_reset : failed %s\n",
+				      fnic_fcpio_status_to_str(hdr_status));
+
+			/*
+			 * Unable to change to eth mode, cannot send out flogi
+			 * Change state to fc mode, so that subsequent Flogi
+			 * requests from libFC will cause more attempts to
+			 * reset the firmware. Free the cached flogi
+			 */
+			fnic->state = FNIC_IN_FC_MODE;
+			ret = -1;
+		}
+	} else {
+		FNIC_SCSI_DBG(KERN_DEBUG,
+			      fnic->lport->host,
+			      "Unexpected state %s while processing"
+			      " reset cmpl\n", fnic_state_to_str(fnic->state));
+		ret = -1;
+	}
+
+	/* Thread removing device blocks till firmware reset is complete */
+	if (fnic->remove_wait)
+		complete(fnic->remove_wait);
+
+	/*
+	 * If fnic is being removed, or fw reset failed
+	 * free the flogi frame. Else, send it out
+	 */
+	if (fnic->remove_wait || ret) {
+		fnic->flogi_oxid = FC_XID_UNKNOWN;
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		if (flogi)
+			dev_kfree_skb_irq(fp_skb(flogi));
+		goto reset_cmpl_handler_end;
+	}
+
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	if (flogi)
+		ret = fnic_send_frame(fnic, flogi);
+
+ reset_cmpl_handler_end:
+	return ret;
+}
+
+/*
+ * fnic_fcpio_flogi_reg_cmpl_handler
+ * Routine to handle flogi register completion
+ */
+static int fnic_fcpio_flogi_reg_cmpl_handler(struct fnic *fnic,
+					     struct fcpio_fw_req *desc)
+{
+	u8 type;
+	u8 hdr_status;
+	struct fcpio_tag tag;
+	int ret = 0;
+	struct fc_frame *flogi_resp = NULL;
+	unsigned long flags;
+	struct sk_buff *skb;
+
+	fcpio_header_dec(&desc->hdr, &type, &hdr_status, &tag);
+
+	/* Update fnic state based on status of flogi reg completion */
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+
+	flogi_resp = fnic->flogi_resp;
+	fnic->flogi_resp = NULL;
+
+	if (fnic->state == FNIC_IN_ETH_TRANS_FC_MODE) {
+
+		/* Check flogi registration completion status */
+		if (!hdr_status) {
+			FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+				      "flog reg succeeded\n");
+			fnic->state = FNIC_IN_FC_MODE;
+		} else {
+			FNIC_SCSI_DBG(KERN_DEBUG,
+				      fnic->lport->host,
+				      "fnic flogi reg :failed %s\n",
+				      fnic_fcpio_status_to_str(hdr_status));
+			fnic->state = FNIC_IN_ETH_MODE;
+			ret = -1;
+		}
+	} else {
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "Unexpected fnic state %s while"
+			      " processing flogi reg completion\n",
+			      fnic_state_to_str(fnic->state));
+		ret = -1;
+	}
+
+	/* Successful flogi reg cmpl, pass frame to LibFC */
+	if (!ret && flogi_resp) {
+		if (fnic->stop_rx_link_events) {
+			spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+			goto reg_cmpl_handler_end;
+		}
+		skb = (struct sk_buff *)flogi_resp;
+		/* Use fr_flags to indicate whether flogi resp or not */
+		fr_flags(flogi_resp) = 1;
+		fr_dev(flogi_resp) = fnic->lport;
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+		skb_queue_tail(&fnic->frame_queue, skb);
+		queue_work(fnic_event_queue, &fnic->frame_work);
+
+	} else {
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		if (flogi_resp)
+			dev_kfree_skb_irq(fp_skb(flogi_resp));
+	}
+
+reg_cmpl_handler_end:
+	return ret;
+}
+
+static inline int is_ack_index_in_range(struct vnic_wq_copy *wq,
+					u16 request_out)
+{
+	if (wq->to_clean_index <= wq->to_use_index) {
+		/* out of range, stale request_out index */
+		if (request_out < wq->to_clean_index ||
+		    request_out >= wq->to_use_index)
+			return 0;
+	} else {
+		/* out of range, stale request_out index */
+		if (request_out < wq->to_clean_index &&
+		    request_out >= wq->to_use_index)
+			return 0;
+	}
+	/* request_out index is in range */
+	return 1;
+}
+
+
+/*
+ * Mark that ack received and store the Ack index. If there are multiple
+ * acks received before Tx thread cleans it up, the latest value will be
+ * used which is correct behavior. This state should be in the copy Wq
+ * instead of in the fnic
+ */
+static inline void fnic_fcpio_ack_handler(struct fnic *fnic,
+					  unsigned int cq_index,
+					  struct fcpio_fw_req *desc)
+{
+	struct vnic_wq_copy *wq;
+	u16 request_out = desc->u.ack.request_out;
+	unsigned long flags;
+
+	/* mark the ack state */
+	wq = &fnic->wq_copy[cq_index - fnic->raw_wq_count - fnic->rq_count];
+	spin_lock_irqsave(&fnic->wq_copy_lock[0], flags);
+
+	if (is_ack_index_in_range(wq, request_out)) {
+		fnic->fw_ack_index[0] = request_out;
+		fnic->fw_ack_recd[0] = 1;
+	}
+	spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags);
+}
+
+/*
+ * fnic_fcpio_icmnd_cmpl_handler
+ * Routine to handle icmnd completions
+ */
+static void fnic_fcpio_icmnd_cmpl_handler(struct fnic *fnic,
+					 struct fcpio_fw_req *desc)
+{
+	u8 type;
+	u8 hdr_status;
+	struct fcpio_tag tag;
+	u32 id;
+	u64 xfer_len = 0;
+	struct fcpio_icmnd_cmpl *icmnd_cmpl;
+	struct fnic_io_req *io_req;
+	struct scsi_cmnd *sc;
+	unsigned long flags;
+	spinlock_t *io_lock;
+
+	/* Decode the cmpl description to get the io_req id */
+	fcpio_header_dec(&desc->hdr, &type, &hdr_status, &tag);
+	fcpio_tag_id_dec(&tag, &id);
+
+	if (id >= FNIC_MAX_IO_REQ)
+		return;
+
+	sc = scsi_host_find_tag(fnic->lport->host, id);
+	WARN_ON_ONCE(!sc);
+	if (!sc)
+		return;
+
+	io_lock = fnic_io_lock_hash(fnic, sc);
+	spin_lock_irqsave(io_lock, flags);
+	io_req = (struct fnic_io_req *)CMD_SP(sc);
+	WARN_ON_ONCE(!io_req);
+	if (!io_req) {
+		spin_unlock_irqrestore(io_lock, flags);
+		return;
+	}
+
+	/* firmware completed the io */
+	io_req->io_completed = 1;
+
+	/*
+	 *  if SCSI-ML has already issued abort on this command,
+	 * ignore completion of the IO. The abts path will clean it up
+	 */
+	if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) {
+		spin_unlock_irqrestore(io_lock, flags);
+		return;
+	}
+
+	/* Mark the IO as complete */
+	CMD_STATE(sc) = FNIC_IOREQ_CMD_COMPLETE;
+
+	icmnd_cmpl = &desc->u.icmnd_cmpl;
+
+	switch (hdr_status) {
+	case FCPIO_SUCCESS:
+		sc->result = (DID_OK << 16) | icmnd_cmpl->scsi_status;
+		xfer_len = scsi_bufflen(sc);
+		scsi_set_resid(sc, icmnd_cmpl->residual);
+
+		if (icmnd_cmpl->flags & FCPIO_ICMND_CMPL_RESID_UNDER)
+			xfer_len -= icmnd_cmpl->residual;
+
+		/*
+		 * If queue_full, then try to reduce queue depth for all
+		 * LUNS on the target. Todo: this should be accompanied
+		 * by a periodic queue_depth rampup based on successful
+		 * IO completion.
+		 */
+		if (icmnd_cmpl->scsi_status == QUEUE_FULL) {
+			struct scsi_device *t_sdev;
+			int qd = 0;
+
+			shost_for_each_device(t_sdev, sc->device->host) {
+				if (t_sdev->id != sc->device->id)
+					continue;
+
+				if (t_sdev->queue_depth > 1) {
+					qd = scsi_track_queue_full
+						(t_sdev,
+						 t_sdev->queue_depth - 1);
+					if (qd == -1)
+						qd = t_sdev->host->cmd_per_lun;
+					shost_printk(KERN_INFO,
+						     fnic->lport->host,
+						     "scsi[%d:%d:%d:%d"
+						     "] queue full detected,"
+						     "new depth = %d\n",
+						     t_sdev->host->host_no,
+						     t_sdev->channel,
+						     t_sdev->id, t_sdev->lun,
+						     t_sdev->queue_depth);
+				}
+			}
+		}
+		break;
+
+	case FCPIO_TIMEOUT:          /* request was timed out */
+		sc->result = (DID_TIME_OUT << 16) | icmnd_cmpl->scsi_status;
+		break;
+
+	case FCPIO_ABORTED:          /* request was aborted */
+		sc->result = (DID_ERROR << 16) | icmnd_cmpl->scsi_status;
+		break;
+
+	case FCPIO_DATA_CNT_MISMATCH: /* recv/sent more/less data than exp. */
+		scsi_set_resid(sc, icmnd_cmpl->residual);
+		sc->result = (DID_ERROR << 16) | icmnd_cmpl->scsi_status;
+		break;
+
+	case FCPIO_OUT_OF_RESOURCE:  /* out of resources to complete request */
+		sc->result = (DID_REQUEUE << 16) | icmnd_cmpl->scsi_status;
+		break;
+	case FCPIO_INVALID_HEADER:   /* header contains invalid data */
+	case FCPIO_INVALID_PARAM:    /* some parameter in request invalid */
+	case FCPIO_REQ_NOT_SUPPORTED:/* request type is not supported */
+	case FCPIO_IO_NOT_FOUND:     /* requested I/O was not found */
+	case FCPIO_SGL_INVALID:      /* request was aborted due to sgl error */
+	case FCPIO_MSS_INVALID:      /* request was aborted due to mss error */
+	case FCPIO_FW_ERR:           /* request was terminated due fw error */
+	default:
+		shost_printk(KERN_ERR, fnic->lport->host, "hdr status = %s\n",
+			     fnic_fcpio_status_to_str(hdr_status));
+		sc->result = (DID_ERROR << 16) | icmnd_cmpl->scsi_status;
+		break;
+	}
+
+	/* Break link with the SCSI command */
+	CMD_SP(sc) = NULL;
+
+	spin_unlock_irqrestore(io_lock, flags);
+
+	fnic_release_ioreq_buf(fnic, io_req, sc);
+
+	mempool_free(io_req, fnic->io_req_pool);
+
+	if (sc->sc_data_direction == DMA_FROM_DEVICE) {
+		fnic->lport->host_stats.fcp_input_requests++;
+		fnic->fcp_input_bytes += xfer_len;
+	} else if (sc->sc_data_direction == DMA_TO_DEVICE) {
+		fnic->lport->host_stats.fcp_output_requests++;
+		fnic->fcp_output_bytes += xfer_len;
+	} else
+		fnic->lport->host_stats.fcp_control_requests++;
+
+	/* Call SCSI completion function to complete the IO */
+	if (sc->scsi_done)
+		sc->scsi_done(sc);
+
+}
+
+/* fnic_fcpio_itmf_cmpl_handler
+ * Routine to handle itmf completions
+ */
+static void fnic_fcpio_itmf_cmpl_handler(struct fnic *fnic,
+					struct fcpio_fw_req *desc)
+{
+	u8 type;
+	u8 hdr_status;
+	struct fcpio_tag tag;
+	u32 id;
+	struct scsi_cmnd *sc;
+	struct fnic_io_req *io_req;
+	unsigned long flags;
+	spinlock_t *io_lock;
+
+	fcpio_header_dec(&desc->hdr, &type, &hdr_status, &tag);
+	fcpio_tag_id_dec(&tag, &id);
+
+	if ((id & FNIC_TAG_MASK) >= FNIC_MAX_IO_REQ)
+		return;
+
+	sc = scsi_host_find_tag(fnic->lport->host, id & FNIC_TAG_MASK);
+	WARN_ON_ONCE(!sc);
+	if (!sc)
+		return;
+
+	io_lock = fnic_io_lock_hash(fnic, sc);
+	spin_lock_irqsave(io_lock, flags);
+	io_req = (struct fnic_io_req *)CMD_SP(sc);
+	WARN_ON_ONCE(!io_req);
+	if (!io_req) {
+		spin_unlock_irqrestore(io_lock, flags);
+		return;
+	}
+
+	if (id & FNIC_TAG_ABORT) {
+		/* Completion of abort cmd */
+		if (CMD_STATE(sc) != FNIC_IOREQ_ABTS_PENDING) {
+			/* This is a late completion. Ignore it */
+			spin_unlock_irqrestore(io_lock, flags);
+			return;
+		}
+		CMD_STATE(sc) = FNIC_IOREQ_ABTS_COMPLETE;
+		CMD_ABTS_STATUS(sc) = hdr_status;
+
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "abts cmpl recd. id %d status %s\n",
+			      (int)(id & FNIC_TAG_MASK),
+			      fnic_fcpio_status_to_str(hdr_status));
+
+		/*
+		 * If scsi_eh thread is blocked waiting for abts to complete,
+		 * signal completion to it. IO will be cleaned in the thread
+		 * else clean it in this context
+		 */
+		if (io_req->abts_done) {
+			complete(io_req->abts_done);
+			spin_unlock_irqrestore(io_lock, flags);
+		} else {
+			FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+				      "abts cmpl, completing IO\n");
+			CMD_SP(sc) = NULL;
+			sc->result = (DID_ERROR << 16);
+
+			spin_unlock_irqrestore(io_lock, flags);
+
+			fnic_release_ioreq_buf(fnic, io_req, sc);
+			mempool_free(io_req, fnic->io_req_pool);
+			if (sc->scsi_done)
+				sc->scsi_done(sc);
+		}
+
+	} else if (id & FNIC_TAG_DEV_RST) {
+		/* Completion of device reset */
+		CMD_LR_STATUS(sc) = hdr_status;
+		CMD_STATE(sc) = FNIC_IOREQ_CMD_COMPLETE;
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "dev reset cmpl recd. id %d status %s\n",
+			      (int)(id & FNIC_TAG_MASK),
+			      fnic_fcpio_status_to_str(hdr_status));
+		if (io_req->dr_done)
+			complete(io_req->dr_done);
+		spin_unlock_irqrestore(io_lock, flags);
+
+	} else {
+		shost_printk(KERN_ERR, fnic->lport->host,
+			     "Unexpected itmf io state %s tag %x\n",
+			     fnic_ioreq_state_to_str(CMD_STATE(sc)), id);
+		spin_unlock_irqrestore(io_lock, flags);
+	}
+
+}
+
+/*
+ * fnic_fcpio_cmpl_handler
+ * Routine to service the cq for wq_copy
+ */
+static int fnic_fcpio_cmpl_handler(struct vnic_dev *vdev,
+				   unsigned int cq_index,
+				   struct fcpio_fw_req *desc)
+{
+	struct fnic *fnic = vnic_dev_priv(vdev);
+	int ret = 0;
+
+	switch (desc->hdr.type) {
+	case FCPIO_ACK: /* fw copied copy wq desc to its queue */
+		fnic_fcpio_ack_handler(fnic, cq_index, desc);
+		break;
+
+	case FCPIO_ICMND_CMPL: /* fw completed a command */
+		fnic_fcpio_icmnd_cmpl_handler(fnic, desc);
+		break;
+
+	case FCPIO_ITMF_CMPL: /* fw completed itmf (abort cmd, lun reset)*/
+		fnic_fcpio_itmf_cmpl_handler(fnic, desc);
+		break;
+
+	case FCPIO_FLOGI_REG_CMPL: /* fw completed flogi_reg */
+		ret = fnic_fcpio_flogi_reg_cmpl_handler(fnic, desc);
+		break;
+
+	case FCPIO_RESET_CMPL: /* fw completed reset */
+		ret = fnic_fcpio_fw_reset_cmpl_handler(fnic, desc);
+		break;
+
+	default:
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "firmware completion type %d\n",
+			      desc->hdr.type);
+		break;
+	}
+
+	return ret;
+}
+
+/*
+ * fnic_wq_copy_cmpl_handler
+ * Routine to process wq copy
+ */
+int fnic_wq_copy_cmpl_handler(struct fnic *fnic, int copy_work_to_do)
+{
+	unsigned int wq_work_done = 0;
+	unsigned int i, cq_index;
+	unsigned int cur_work_done;
+
+	for (i = 0; i < fnic->wq_copy_count; i++) {
+		cq_index = i + fnic->raw_wq_count + fnic->rq_count;
+		cur_work_done = vnic_cq_copy_service(&fnic->cq[cq_index],
+						     fnic_fcpio_cmpl_handler,
+						     copy_work_to_do);
+		wq_work_done += cur_work_done;
+	}
+	return wq_work_done;
+}
+
+static void fnic_cleanup_io(struct fnic *fnic, int exclude_id)
+{
+	unsigned int i;
+	struct fnic_io_req *io_req;
+	unsigned long flags = 0;
+	struct scsi_cmnd *sc;
+	spinlock_t *io_lock;
+
+	for (i = 0; i < FNIC_MAX_IO_REQ; i++) {
+		if (i == exclude_id)
+			continue;
+
+		sc = scsi_host_find_tag(fnic->lport->host, i);
+		if (!sc)
+			continue;
+
+		io_lock = fnic_io_lock_hash(fnic, sc);
+		spin_lock_irqsave(io_lock, flags);
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+		if (!io_req) {
+			spin_unlock_irqrestore(io_lock, flags);
+			goto cleanup_scsi_cmd;
+		}
+
+		CMD_SP(sc) = NULL;
+
+		spin_unlock_irqrestore(io_lock, flags);
+
+		/*
+		 * If there is a scsi_cmnd associated with this io_req, then
+		 * free the corresponding state
+		 */
+		fnic_release_ioreq_buf(fnic, io_req, sc);
+		mempool_free(io_req, fnic->io_req_pool);
+
+cleanup_scsi_cmd:
+		sc->result = DID_TRANSPORT_DISRUPTED << 16;
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, "fnic_cleanup_io:"
+			      " DID_TRANSPORT_DISRUPTED\n");
+
+		/* Complete the command to SCSI */
+		if (sc->scsi_done)
+			sc->scsi_done(sc);
+	}
+}
+
+void fnic_wq_copy_cleanup_handler(struct vnic_wq_copy *wq,
+				  struct fcpio_host_req *desc)
+{
+	u32 id;
+	struct fnic *fnic = vnic_dev_priv(wq->vdev);
+	struct fnic_io_req *io_req;
+	struct scsi_cmnd *sc;
+	unsigned long flags;
+	spinlock_t *io_lock;
+
+	/* get the tag reference */
+	fcpio_tag_id_dec(&desc->hdr.tag, &id);
+	id &= FNIC_TAG_MASK;
+
+	if (id >= FNIC_MAX_IO_REQ)
+		return;
+
+	sc = scsi_host_find_tag(fnic->lport->host, id);
+	if (!sc)
+		return;
+
+	io_lock = fnic_io_lock_hash(fnic, sc);
+	spin_lock_irqsave(io_lock, flags);
+
+	/* Get the IO context which this desc refers to */
+	io_req = (struct fnic_io_req *)CMD_SP(sc);
+
+	/* fnic interrupts are turned off by now */
+
+	if (!io_req) {
+		spin_unlock_irqrestore(io_lock, flags);
+		goto wq_copy_cleanup_scsi_cmd;
+	}
+
+	CMD_SP(sc) = NULL;
+
+	spin_unlock_irqrestore(io_lock, flags);
+
+	fnic_release_ioreq_buf(fnic, io_req, sc);
+	mempool_free(io_req, fnic->io_req_pool);
+
+wq_copy_cleanup_scsi_cmd:
+	sc->result = DID_NO_CONNECT << 16;
+	FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, "wq_copy_cleanup_handler:"
+		      " DID_NO_CONNECT\n");
+
+	if (sc->scsi_done)
+		sc->scsi_done(sc);
+}
+
+static inline int fnic_queue_abort_io_req(struct fnic *fnic, int tag,
+					  u32 task_req, u8 *fc_lun,
+					  struct fnic_io_req *io_req)
+{
+	struct vnic_wq_copy *wq = &fnic->wq_copy[0];
+	unsigned long flags;
+
+	spin_lock_irqsave(&fnic->wq_copy_lock[0], flags);
+
+	if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0])
+		free_wq_copy_descs(fnic, wq);
+
+	if (!vnic_wq_copy_desc_avail(wq)) {
+		spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags);
+		return 1;
+	}
+	fnic_queue_wq_copy_desc_itmf(wq, tag | FNIC_TAG_ABORT,
+				     0, task_req, tag, fc_lun, io_req->port_id,
+				     fnic->config.ra_tov, fnic->config.ed_tov);
+
+	spin_unlock_irqrestore(&fnic->wq_copy_lock[0], flags);
+	return 0;
+}
+
+void fnic_rport_exch_reset(struct fnic *fnic, u32 port_id)
+{
+	int tag;
+	struct fnic_io_req *io_req;
+	spinlock_t *io_lock;
+	unsigned long flags;
+	struct scsi_cmnd *sc;
+	struct scsi_lun fc_lun;
+	enum fnic_ioreq_state old_ioreq_state;
+
+	FNIC_SCSI_DBG(KERN_DEBUG,
+		      fnic->lport->host,
+		      "fnic_rport_reset_exch called portid 0x%06x\n",
+		      port_id);
+
+	if (fnic->in_remove)
+		return;
+
+	for (tag = 0; tag < FNIC_MAX_IO_REQ; tag++) {
+		sc = scsi_host_find_tag(fnic->lport->host, tag);
+		if (!sc)
+			continue;
+
+		io_lock = fnic_io_lock_hash(fnic, sc);
+		spin_lock_irqsave(io_lock, flags);
+
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+
+		if (!io_req || io_req->port_id != port_id) {
+			spin_unlock_irqrestore(io_lock, flags);
+			continue;
+		}
+
+		/*
+		 * Found IO that is still pending with firmware and
+		 * belongs to rport that went away
+		 */
+		if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) {
+			spin_unlock_irqrestore(io_lock, flags);
+			continue;
+		}
+		old_ioreq_state = CMD_STATE(sc);
+		CMD_STATE(sc) = FNIC_IOREQ_ABTS_PENDING;
+		CMD_ABTS_STATUS(sc) = FCPIO_INVALID_CODE;
+
+		BUG_ON(io_req->abts_done);
+
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "fnic_rport_reset_exch: Issuing abts\n");
+
+		spin_unlock_irqrestore(io_lock, flags);
+
+		/* Now queue the abort command to firmware */
+		int_to_scsilun(sc->device->lun, &fc_lun);
+
+		if (fnic_queue_abort_io_req(fnic, tag,
+					    FCPIO_ITMF_ABT_TASK_TERM,
+					    fc_lun.scsi_lun, io_req)) {
+			/*
+			 * Revert the cmd state back to old state, if
+			 * it hasnt changed in between. This cmd will get
+			 * aborted later by scsi_eh, or cleaned up during
+			 * lun reset
+			 */
+			io_lock = fnic_io_lock_hash(fnic, sc);
+
+			spin_lock_irqsave(io_lock, flags);
+			if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING)
+				CMD_STATE(sc) = old_ioreq_state;
+			spin_unlock_irqrestore(io_lock, flags);
+		}
+	}
+
+}
+
+void fnic_terminate_rport_io(struct fc_rport *rport)
+{
+	int tag;
+	struct fnic_io_req *io_req;
+	spinlock_t *io_lock;
+	unsigned long flags;
+	struct scsi_cmnd *sc;
+	struct scsi_lun fc_lun;
+	struct fc_rport_libfc_priv *rdata = rport->dd_data;
+	struct fc_lport *lport = rdata->local_port;
+	struct fnic *fnic = lport_priv(lport);
+	struct fc_rport *cmd_rport;
+	enum fnic_ioreq_state old_ioreq_state;
+
+	FNIC_SCSI_DBG(KERN_DEBUG,
+		      fnic->lport->host, "fnic_terminate_rport_io called"
+		      " wwpn 0x%llx, wwnn0x%llx, portid 0x%06x\n",
+		      rport->port_name, rport->node_name,
+		      rport->port_id);
+
+	if (fnic->in_remove)
+		return;
+
+	for (tag = 0; tag < FNIC_MAX_IO_REQ; tag++) {
+		sc = scsi_host_find_tag(fnic->lport->host, tag);
+		if (!sc)
+			continue;
+
+		cmd_rport = starget_to_rport(scsi_target(sc->device));
+		if (rport != cmd_rport)
+			continue;
+
+		io_lock = fnic_io_lock_hash(fnic, sc);
+		spin_lock_irqsave(io_lock, flags);
+
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+
+		if (!io_req || rport != cmd_rport) {
+			spin_unlock_irqrestore(io_lock, flags);
+			continue;
+		}
+
+		/*
+		 * Found IO that is still pending with firmware and
+		 * belongs to rport that went away
+		 */
+		if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) {
+			spin_unlock_irqrestore(io_lock, flags);
+			continue;
+		}
+		old_ioreq_state = CMD_STATE(sc);
+		CMD_STATE(sc) = FNIC_IOREQ_ABTS_PENDING;
+		CMD_ABTS_STATUS(sc) = FCPIO_INVALID_CODE;
+
+		BUG_ON(io_req->abts_done);
+
+		FNIC_SCSI_DBG(KERN_DEBUG,
+			      fnic->lport->host,
+			      "fnic_terminate_rport_io: Issuing abts\n");
+
+		spin_unlock_irqrestore(io_lock, flags);
+
+		/* Now queue the abort command to firmware */
+		int_to_scsilun(sc->device->lun, &fc_lun);
+
+		if (fnic_queue_abort_io_req(fnic, tag,
+					    FCPIO_ITMF_ABT_TASK_TERM,
+					    fc_lun.scsi_lun, io_req)) {
+			/*
+			 * Revert the cmd state back to old state, if
+			 * it hasnt changed in between. This cmd will get
+			 * aborted later by scsi_eh, or cleaned up during
+			 * lun reset
+			 */
+			io_lock = fnic_io_lock_hash(fnic, sc);
+
+			spin_lock_irqsave(io_lock, flags);
+			if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING)
+				CMD_STATE(sc) = old_ioreq_state;
+			spin_unlock_irqrestore(io_lock, flags);
+		}
+	}
+
+}
+
+static void fnic_block_error_handler(struct scsi_cmnd *sc)
+{
+	struct Scsi_Host *shost = sc->device->host;
+	struct fc_rport *rport = starget_to_rport(scsi_target(sc->device));
+	unsigned long flags;
+
+	spin_lock_irqsave(shost->host_lock, flags);
+	while (rport->port_state == FC_PORTSTATE_BLOCKED) {
+		spin_unlock_irqrestore(shost->host_lock, flags);
+		msleep(1000);
+		spin_lock_irqsave(shost->host_lock, flags);
+	}
+	spin_unlock_irqrestore(shost->host_lock, flags);
+
+}
+
+/*
+ * This function is exported to SCSI for sending abort cmnds.
+ * A SCSI IO is represented by a io_req in the driver.
+ * The ioreq is linked to the SCSI Cmd, thus a link with the ULP's IO.
+ */
+int fnic_abort_cmd(struct scsi_cmnd *sc)
+{
+	struct fc_lport *lp;
+	struct fnic *fnic;
+	struct fnic_io_req *io_req;
+	struct fc_rport *rport;
+	spinlock_t *io_lock;
+	unsigned long flags;
+	int ret = SUCCESS;
+	u32 task_req;
+	struct scsi_lun fc_lun;
+	DECLARE_COMPLETION_ONSTACK(tm_done);
+
+	/* Wait for rport to unblock */
+	fnic_block_error_handler(sc);
+
+	/* Get local-port, check ready and link up */
+	lp = shost_priv(sc->device->host);
+
+	fnic = lport_priv(lp);
+	FNIC_SCSI_DBG(KERN_DEBUG,
+		      fnic->lport->host,
+		      "Abort Cmd called FCID 0x%x, LUN 0x%x TAG %d\n",
+		      (starget_to_rport(scsi_target(sc->device)))->port_id,
+		      sc->device->lun, sc->request->tag);
+
+	if (lp->state != LPORT_ST_READY || !(lp->link_up)) {
+		ret = FAILED;
+		goto fnic_abort_cmd_end;
+	}
+
+	/*
+	 * Avoid a race between SCSI issuing the abort and the device
+	 * completing the command.
+	 *
+	 * If the command is already completed by the fw cmpl code,
+	 * we just return SUCCESS from here. This means that the abort
+	 * succeeded. In the SCSI ML, since the timeout for command has
+	 * happened, the completion wont actually complete the command
+	 * and it will be considered as an aborted command
+	 *
+	 * The CMD_SP will not be cleared except while holding io_req_lock.
+	 */
+	io_lock = fnic_io_lock_hash(fnic, sc);
+	spin_lock_irqsave(io_lock, flags);
+	io_req = (struct fnic_io_req *)CMD_SP(sc);
+	if (!io_req) {
+		spin_unlock_irqrestore(io_lock, flags);
+		goto fnic_abort_cmd_end;
+	}
+
+	io_req->abts_done = &tm_done;
+
+	if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) {
+		spin_unlock_irqrestore(io_lock, flags);
+		goto wait_pending;
+	}
+	/*
+	 * Command is still pending, need to abort it
+	 * If the firmware completes the command after this point,
+	 * the completion wont be done till mid-layer, since abort
+	 * has already started.
+	 */
+	CMD_STATE(sc) = FNIC_IOREQ_ABTS_PENDING;
+	CMD_ABTS_STATUS(sc) = FCPIO_INVALID_CODE;
+
+	spin_unlock_irqrestore(io_lock, flags);
+
+	/*
+	 * Check readiness of the remote port. If the path to remote
+	 * port is up, then send abts to the remote port to terminate
+	 * the IO. Else, just locally terminate the IO in the firmware
+	 */
+	rport = starget_to_rport(scsi_target(sc->device));
+	if (fc_remote_port_chkready(rport) == 0)
+		task_req = FCPIO_ITMF_ABT_TASK;
+	else
+		task_req = FCPIO_ITMF_ABT_TASK_TERM;
+
+	/* Now queue the abort command to firmware */
+	int_to_scsilun(sc->device->lun, &fc_lun);
+
+	if (fnic_queue_abort_io_req(fnic, sc->request->tag, task_req,
+				    fc_lun.scsi_lun, io_req)) {
+		spin_lock_irqsave(io_lock, flags);
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+		if (io_req)
+			io_req->abts_done = NULL;
+		spin_unlock_irqrestore(io_lock, flags);
+		ret = FAILED;
+		goto fnic_abort_cmd_end;
+	}
+
+	/*
+	 * We queued an abort IO, wait for its completion.
+	 * Once the firmware completes the abort command, it will
+	 * wake up this thread.
+	 */
+ wait_pending:
+	wait_for_completion_timeout(&tm_done,
+				    msecs_to_jiffies
+				    (2 * fnic->config.ra_tov +
+				     fnic->config.ed_tov));
+
+	/* Check the abort status */
+	spin_lock_irqsave(io_lock, flags);
+
+	io_req = (struct fnic_io_req *)CMD_SP(sc);
+	if (!io_req) {
+		spin_unlock_irqrestore(io_lock, flags);
+		ret = FAILED;
+		goto fnic_abort_cmd_end;
+	}
+	io_req->abts_done = NULL;
+
+	/* fw did not complete abort, timed out */
+	if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) {
+		spin_unlock_irqrestore(io_lock, flags);
+		ret = FAILED;
+		goto fnic_abort_cmd_end;
+	}
+
+	/*
+	 * firmware completed the abort, check the status,
+	 * free the io_req irrespective of failure or success
+	 */
+	if (CMD_ABTS_STATUS(sc) != FCPIO_SUCCESS)
+		ret = FAILED;
+
+	CMD_SP(sc) = NULL;
+
+	spin_unlock_irqrestore(io_lock, flags);
+
+	fnic_release_ioreq_buf(fnic, io_req, sc);
+	mempool_free(io_req, fnic->io_req_pool);
+
+fnic_abort_cmd_end:
+	FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+		      "Returning from abort cmd %s\n",
+		      (ret == SUCCESS) ?
+		      "SUCCESS" : "FAILED");
+	return ret;
+}
+
+static inline int fnic_queue_dr_io_req(struct fnic *fnic,
+				       struct scsi_cmnd *sc,
+				       struct fnic_io_req *io_req)
+{
+	struct vnic_wq_copy *wq = &fnic->wq_copy[0];
+	struct scsi_lun fc_lun;
+	int ret = 0;
+	unsigned long intr_flags;
+
+	spin_lock_irqsave(&fnic->wq_copy_lock[0], intr_flags);
+
+	if (vnic_wq_copy_desc_avail(wq) <= fnic->wq_copy_desc_low[0])
+		free_wq_copy_descs(fnic, wq);
+
+	if (!vnic_wq_copy_desc_avail(wq)) {
+		ret = -EAGAIN;
+		goto lr_io_req_end;
+	}
+
+	/* fill in the lun info */
+	int_to_scsilun(sc->device->lun, &fc_lun);
+
+	fnic_queue_wq_copy_desc_itmf(wq, sc->request->tag | FNIC_TAG_DEV_RST,
+				     0, FCPIO_ITMF_LUN_RESET, SCSI_NO_TAG,
+				     fc_lun.scsi_lun, io_req->port_id,
+				     fnic->config.ra_tov, fnic->config.ed_tov);
+
+lr_io_req_end:
+	spin_unlock_irqrestore(&fnic->wq_copy_lock[0], intr_flags);
+
+	return ret;
+}
+
+/*
+ * Clean up any pending aborts on the lun
+ * For each outstanding IO on this lun, whose abort is not completed by fw,
+ * issue a local abort. Wait for abort to complete. Return 0 if all commands
+ * successfully aborted, 1 otherwise
+ */
+static int fnic_clean_pending_aborts(struct fnic *fnic,
+				     struct scsi_cmnd *lr_sc)
+{
+	int tag;
+	struct fnic_io_req *io_req;
+	spinlock_t *io_lock;
+	unsigned long flags;
+	int ret = 0;
+	struct scsi_cmnd *sc;
+	struct fc_rport *rport;
+	struct scsi_lun fc_lun;
+	struct scsi_device *lun_dev = lr_sc->device;
+	DECLARE_COMPLETION_ONSTACK(tm_done);
+
+	for (tag = 0; tag < FNIC_MAX_IO_REQ; tag++) {
+		sc = scsi_host_find_tag(fnic->lport->host, tag);
+		/*
+		 * ignore this lun reset cmd or cmds that do not belong to
+		 * this lun
+		 */
+		if (!sc || sc == lr_sc || sc->device != lun_dev)
+			continue;
+
+		io_lock = fnic_io_lock_hash(fnic, sc);
+		spin_lock_irqsave(io_lock, flags);
+
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+
+		if (!io_req || sc->device != lun_dev) {
+			spin_unlock_irqrestore(io_lock, flags);
+			continue;
+		}
+
+		/*
+		 * Found IO that is still pending with firmware and
+		 * belongs to the LUN that we are resetting
+		 */
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "Found IO in %s on lun\n",
+			      fnic_ioreq_state_to_str(CMD_STATE(sc)));
+
+		BUG_ON(CMD_STATE(sc) != FNIC_IOREQ_ABTS_PENDING);
+
+		CMD_ABTS_STATUS(sc) = FCPIO_INVALID_CODE;
+		io_req->abts_done = &tm_done;
+		spin_unlock_irqrestore(io_lock, flags);
+
+		/* Now queue the abort command to firmware */
+		int_to_scsilun(sc->device->lun, &fc_lun);
+		rport = starget_to_rport(scsi_target(sc->device));
+
+		if (fnic_queue_abort_io_req(fnic, tag,
+					    FCPIO_ITMF_ABT_TASK_TERM,
+					    fc_lun.scsi_lun, io_req)) {
+			spin_lock_irqsave(io_lock, flags);
+			io_req = (struct fnic_io_req *)CMD_SP(sc);
+			if (io_req)
+				io_req->abts_done = NULL;
+			spin_unlock_irqrestore(io_lock, flags);
+			ret = 1;
+			goto clean_pending_aborts_end;
+		}
+
+		wait_for_completion_timeout(&tm_done,
+					    msecs_to_jiffies
+					    (fnic->config.ed_tov));
+
+		/* Recheck cmd state to check if it is now aborted */
+		spin_lock_irqsave(io_lock, flags);
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+		if (!io_req) {
+			spin_unlock_irqrestore(io_lock, flags);
+			ret = 1;
+			goto clean_pending_aborts_end;
+		}
+
+		io_req->abts_done = NULL;
+
+		/* if abort is still pending with fw, fail */
+		if (CMD_STATE(sc) == FNIC_IOREQ_ABTS_PENDING) {
+			spin_unlock_irqrestore(io_lock, flags);
+			ret = 1;
+			goto clean_pending_aborts_end;
+		}
+		CMD_SP(sc) = NULL;
+		spin_unlock_irqrestore(io_lock, flags);
+
+		fnic_release_ioreq_buf(fnic, io_req, sc);
+		mempool_free(io_req, fnic->io_req_pool);
+	}
+
+clean_pending_aborts_end:
+	return ret;
+}
+
+/*
+ * SCSI Eh thread issues a Lun Reset when one or more commands on a LUN
+ * fail to get aborted. It calls driver's eh_device_reset with a SCSI command
+ * on the LUN.
+ */
+int fnic_device_reset(struct scsi_cmnd *sc)
+{
+	struct fc_lport *lp;
+	struct fnic *fnic;
+	struct fnic_io_req *io_req;
+	struct fc_rport *rport;
+	int status;
+	int ret = FAILED;
+	spinlock_t *io_lock;
+	unsigned long flags;
+	DECLARE_COMPLETION_ONSTACK(tm_done);
+
+	/* Wait for rport to unblock */
+	fnic_block_error_handler(sc);
+
+	/* Get local-port, check ready and link up */
+	lp = shost_priv(sc->device->host);
+
+	fnic = lport_priv(lp);
+	FNIC_SCSI_DBG(KERN_DEBUG,
+		      fnic->lport->host,
+		      "Device reset called FCID 0x%x, LUN 0x%x\n",
+		      (starget_to_rport(scsi_target(sc->device)))->port_id,
+		      sc->device->lun);
+
+
+	if (lp->state != LPORT_ST_READY || !(lp->link_up))
+		goto fnic_device_reset_end;
+
+	/* Check if remote port up */
+	rport = starget_to_rport(scsi_target(sc->device));
+	if (fc_remote_port_chkready(rport))
+		goto fnic_device_reset_end;
+
+	io_lock = fnic_io_lock_hash(fnic, sc);
+	spin_lock_irqsave(io_lock, flags);
+	io_req = (struct fnic_io_req *)CMD_SP(sc);
+
+	/*
+	 * If there is a io_req attached to this command, then use it,
+	 * else allocate a new one.
+	 */
+	if (!io_req) {
+		io_req = mempool_alloc(fnic->io_req_pool, GFP_ATOMIC);
+		if (!io_req) {
+			spin_unlock_irqrestore(io_lock, flags);
+			goto fnic_device_reset_end;
+		}
+		memset(io_req, 0, sizeof(*io_req));
+		io_req->port_id = rport->port_id;
+		CMD_SP(sc) = (char *)io_req;
+	}
+	io_req->dr_done = &tm_done;
+	CMD_STATE(sc) = FNIC_IOREQ_CMD_PENDING;
+	CMD_LR_STATUS(sc) = FCPIO_INVALID_CODE;
+	spin_unlock_irqrestore(io_lock, flags);
+
+	FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host, "TAG %d\n",
+		      sc->request->tag);
+
+	/*
+	 * issue the device reset, if enqueue failed, clean up the ioreq
+	 * and break assoc with scsi cmd
+	 */
+	if (fnic_queue_dr_io_req(fnic, sc, io_req)) {
+		spin_lock_irqsave(io_lock, flags);
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+		if (io_req)
+			io_req->dr_done = NULL;
+		goto fnic_device_reset_clean;
+	}
+
+	/*
+	 * Wait on the local completion for LUN reset.  The io_req may be
+	 * freed while we wait since we hold no lock.
+	 */
+	wait_for_completion_timeout(&tm_done,
+				    msecs_to_jiffies(FNIC_LUN_RESET_TIMEOUT));
+
+	spin_lock_irqsave(io_lock, flags);
+	io_req = (struct fnic_io_req *)CMD_SP(sc);
+	if (!io_req) {
+		spin_unlock_irqrestore(io_lock, flags);
+		goto fnic_device_reset_end;
+	}
+	io_req->dr_done = NULL;
+
+	status = CMD_LR_STATUS(sc);
+	spin_unlock_irqrestore(io_lock, flags);
+
+	/*
+	 * If lun reset not completed, bail out with failed. io_req
+	 * gets cleaned up during higher levels of EH
+	 */
+	if (status == FCPIO_INVALID_CODE) {
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "Device reset timed out\n");
+		goto fnic_device_reset_end;
+	}
+
+	/* Completed, but not successful, clean up the io_req, return fail */
+	if (status != FCPIO_SUCCESS) {
+		spin_lock_irqsave(io_lock, flags);
+		FNIC_SCSI_DBG(KERN_DEBUG,
+			      fnic->lport->host,
+			      "Device reset completed - failed\n");
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+		goto fnic_device_reset_clean;
+	}
+
+	/*
+	 * Clean up any aborts on this lun that have still not
+	 * completed. If any of these fail, then LUN reset fails.
+	 * clean_pending_aborts cleans all cmds on this lun except
+	 * the lun reset cmd. If all cmds get cleaned, the lun reset
+	 * succeeds
+	 */
+	if (fnic_clean_pending_aborts(fnic, sc)) {
+		spin_lock_irqsave(io_lock, flags);
+		io_req = (struct fnic_io_req *)CMD_SP(sc);
+		FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+			      "Device reset failed"
+			      " since could not abort all IOs\n");
+		goto fnic_device_reset_clean;
+	}
+
+	/* Clean lun reset command */
+	spin_lock_irqsave(io_lock, flags);
+	io_req = (struct fnic_io_req *)CMD_SP(sc);
+	if (io_req)
+		/* Completed, and successful */
+		ret = SUCCESS;
+
+fnic_device_reset_clean:
+	if (io_req)
+		CMD_SP(sc) = NULL;
+
+	spin_unlock_irqrestore(io_lock, flags);
+
+	if (io_req) {
+		fnic_release_ioreq_buf(fnic, io_req, sc);
+		mempool_free(io_req, fnic->io_req_pool);
+	}
+
+fnic_device_reset_end:
+	FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+		      "Returning from device reset %s\n",
+		      (ret == SUCCESS) ?
+		      "SUCCESS" : "FAILED");
+	return ret;
+}
+
+/* Clean up all IOs, clean up libFC local port */
+int fnic_reset(struct Scsi_Host *shost)
+{
+	struct fc_lport *lp;
+	struct fnic *fnic;
+	int ret = SUCCESS;
+
+	lp = shost_priv(shost);
+	fnic = lport_priv(lp);
+
+	FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+		      "fnic_reset called\n");
+
+	/*
+	 * Reset local port, this will clean up libFC exchanges,
+	 * reset remote port sessions, and if link is up, begin flogi
+	 */
+	if (lp->tt.lport_reset(lp))
+		ret = FAILED;
+
+	FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+		      "Returning from fnic reset %s\n",
+		      (ret == SUCCESS) ?
+		      "SUCCESS" : "FAILED");
+
+	return ret;
+}
+
+/*
+ * SCSI Error handling calls driver's eh_host_reset if all prior
+ * error handling levels return FAILED. If host reset completes
+ * successfully, and if link is up, then Fabric login begins.
+ *
+ * Host Reset is the highest level of error recovery. If this fails, then
+ * host is offlined by SCSI.
+ *
+ */
+int fnic_host_reset(struct scsi_cmnd *sc)
+{
+	int ret;
+	unsigned long wait_host_tmo;
+	struct Scsi_Host *shost = sc->device->host;
+	struct fc_lport *lp = shost_priv(shost);
+
+	/*
+	 * If fnic_reset is successful, wait for fabric login to complete
+	 * scsi-ml tries to send a TUR to every device if host reset is
+	 * successful, so before returning to scsi, fabric should be up
+	 */
+	ret = fnic_reset(shost);
+	if (ret == SUCCESS) {
+		wait_host_tmo = jiffies + FNIC_HOST_RESET_SETTLE_TIME * HZ;
+		ret = FAILED;
+		while (time_before(jiffies, wait_host_tmo)) {
+			if ((lp->state == LPORT_ST_READY) &&
+			    (lp->link_up)) {
+				ret = SUCCESS;
+				break;
+			}
+			ssleep(1);
+		}
+	}
+
+	return ret;
+}
+
+/*
+ * This fxn is called from libFC when host is removed
+ */
+void fnic_scsi_abort_io(struct fc_lport *lp)
+{
+	int err = 0;
+	unsigned long flags;
+	enum fnic_state old_state;
+	struct fnic *fnic = lport_priv(lp);
+	DECLARE_COMPLETION_ONSTACK(remove_wait);
+
+	/* Issue firmware reset for fnic, wait for reset to complete */
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+	fnic->remove_wait = &remove_wait;
+	old_state = fnic->state;
+	fnic->state = FNIC_IN_FC_TRANS_ETH_MODE;
+	vnic_dev_del_addr(fnic->vdev, fnic->data_src_addr);
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	err = fnic_fw_reset_handler(fnic);
+	if (err) {
+		spin_lock_irqsave(&fnic->fnic_lock, flags);
+		if (fnic->state == FNIC_IN_FC_TRANS_ETH_MODE)
+			fnic->state = old_state;
+		fnic->remove_wait = NULL;
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+		return;
+	}
+
+	/* Wait for firmware reset to complete */
+	wait_for_completion_timeout(&remove_wait,
+				    msecs_to_jiffies(FNIC_RMDEVICE_TIMEOUT));
+
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+	fnic->remove_wait = NULL;
+	FNIC_SCSI_DBG(KERN_DEBUG, fnic->lport->host,
+		      "fnic_scsi_abort_io %s\n",
+		      (fnic->state == FNIC_IN_ETH_MODE) ?
+		      "SUCCESS" : "FAILED");
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+}
+
+/*
+ * This fxn called from libFC to clean up driver IO state on link down
+ */
+void fnic_scsi_cleanup(struct fc_lport *lp)
+{
+	unsigned long flags;
+	enum fnic_state old_state;
+	struct fnic *fnic = lport_priv(lp);
+
+	/* issue fw reset */
+	spin_lock_irqsave(&fnic->fnic_lock, flags);
+	old_state = fnic->state;
+	fnic->state = FNIC_IN_FC_TRANS_ETH_MODE;
+	vnic_dev_del_addr(fnic->vdev, fnic->data_src_addr);
+	spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+
+	if (fnic_fw_reset_handler(fnic)) {
+		spin_lock_irqsave(&fnic->fnic_lock, flags);
+		if (fnic->state == FNIC_IN_FC_TRANS_ETH_MODE)
+			fnic->state = old_state;
+		spin_unlock_irqrestore(&fnic->fnic_lock, flags);
+	}
+
+}
+
+void fnic_empty_scsi_cleanup(struct fc_lport *lp)
+{
+}
+
+void fnic_exch_mgr_reset(struct fc_lport *lp, u32 sid, u32 did)
+{
+	struct fnic *fnic = lport_priv(lp);
+
+	/* Non-zero sid, nothing to do */
+	if (sid)
+		goto call_fc_exch_mgr_reset;
+
+	if (did) {
+		fnic_rport_exch_reset(fnic, did);
+		goto call_fc_exch_mgr_reset;
+	}
+
+	/*
+	 * sid = 0, did = 0
+	 * link down or device being removed
+	 */
+	if (!fnic->in_remove)
+		fnic_scsi_cleanup(lp);
+	else
+		fnic_scsi_abort_io(lp);
+
+	/* call libFC exch mgr reset to reset its exchanges */
+call_fc_exch_mgr_reset:
+	fc_exch_mgr_reset(lp, sid, did);
+
+}

drivers/scsi/fnic/rq_enet_desc.h

@@ -0,0 +1,58 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _RQ_ENET_DESC_H_
+#define _RQ_ENET_DESC_H_
+
+/* Ethernet receive queue descriptor: 16B */
+struct rq_enet_desc {
+	__le64 address;
+	__le16 length_type;
+	u8 reserved[6];
+};
+
+enum rq_enet_type_types {
+	RQ_ENET_TYPE_ONLY_SOP = 0,
+	RQ_ENET_TYPE_NOT_SOP = 1,
+	RQ_ENET_TYPE_RESV2 = 2,
+	RQ_ENET_TYPE_RESV3 = 3,
+};
+
+#define RQ_ENET_ADDR_BITS		64
+#define RQ_ENET_LEN_BITS		14
+#define RQ_ENET_LEN_MASK		((1 << RQ_ENET_LEN_BITS) - 1)
+#define RQ_ENET_TYPE_BITS		2
+#define RQ_ENET_TYPE_MASK		((1 << RQ_ENET_TYPE_BITS) - 1)
+
+static inline void rq_enet_desc_enc(struct rq_enet_desc *desc,
+	u64 address, u8 type, u16 length)
+{
+	desc->address = cpu_to_le64(address);
+	desc->length_type = cpu_to_le16((length & RQ_ENET_LEN_MASK) |
+		((type & RQ_ENET_TYPE_MASK) << RQ_ENET_LEN_BITS));
+}
+
+static inline void rq_enet_desc_dec(struct rq_enet_desc *desc,
+	u64 *address, u8 *type, u16 *length)
+{
+	*address = le64_to_cpu(desc->address);
+	*length = le16_to_cpu(desc->length_type) & RQ_ENET_LEN_MASK;
+	*type = (u8)((le16_to_cpu(desc->length_type) >> RQ_ENET_LEN_BITS) &
+		RQ_ENET_TYPE_MASK);
+}
+
+#endif /* _RQ_ENET_DESC_H_ */

drivers/scsi/fnic/vnic_cq.c

@@ -0,0 +1,85 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include "vnic_dev.h"
+#include "vnic_cq.h"
+
+void vnic_cq_free(struct vnic_cq *cq)
+{
+	vnic_dev_free_desc_ring(cq->vdev, &cq->ring);
+
+	cq->ctrl = NULL;
+}
+
+int vnic_cq_alloc(struct vnic_dev *vdev, struct vnic_cq *cq, unsigned int index,
+	unsigned int desc_count, unsigned int desc_size)
+{
+	int err;
+
+	cq->index = index;
+	cq->vdev = vdev;
+
+	cq->ctrl = vnic_dev_get_res(vdev, RES_TYPE_CQ, index);
+	if (!cq->ctrl) {
+		printk(KERN_ERR "Failed to hook CQ[%d] resource\n", index);
+		return -EINVAL;
+	}
+
+	err = vnic_dev_alloc_desc_ring(vdev, &cq->ring, desc_count, desc_size);
+	if (err)
+		return err;
+
+	return 0;
+}
+
+void vnic_cq_init(struct vnic_cq *cq, unsigned int flow_control_enable,
+	unsigned int color_enable, unsigned int cq_head, unsigned int cq_tail,
+	unsigned int cq_tail_color, unsigned int interrupt_enable,
+	unsigned int cq_entry_enable, unsigned int cq_message_enable,
+	unsigned int interrupt_offset, u64 cq_message_addr)
+{
+	u64 paddr;
+
+	paddr = (u64)cq->ring.base_addr | VNIC_PADDR_TARGET;
+	writeq(paddr, &cq->ctrl->ring_base);
+	iowrite32(cq->ring.desc_count, &cq->ctrl->ring_size);
+	iowrite32(flow_control_enable, &cq->ctrl->flow_control_enable);
+	iowrite32(color_enable, &cq->ctrl->color_enable);
+	iowrite32(cq_head, &cq->ctrl->cq_head);
+	iowrite32(cq_tail, &cq->ctrl->cq_tail);
+	iowrite32(cq_tail_color, &cq->ctrl->cq_tail_color);
+	iowrite32(interrupt_enable, &cq->ctrl->interrupt_enable);
+	iowrite32(cq_entry_enable, &cq->ctrl->cq_entry_enable);
+	iowrite32(cq_message_enable, &cq->ctrl->cq_message_enable);
+	iowrite32(interrupt_offset, &cq->ctrl->interrupt_offset);
+	writeq(cq_message_addr, &cq->ctrl->cq_message_addr);
+}
+
+void vnic_cq_clean(struct vnic_cq *cq)
+{
+	cq->to_clean = 0;
+	cq->last_color = 0;
+
+	iowrite32(0, &cq->ctrl->cq_head);
+	iowrite32(0, &cq->ctrl->cq_tail);
+	iowrite32(1, &cq->ctrl->cq_tail_color);
+
+	vnic_dev_clear_desc_ring(&cq->ring);
+}

drivers/scsi/fnic/vnic_cq.h

@@ -0,0 +1,121 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _VNIC_CQ_H_
+#define _VNIC_CQ_H_
+
+#include "cq_desc.h"
+#include "vnic_dev.h"
+
+/*
+ * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth
+ * Driver) when both are built with CONFIG options =y
+ */
+#define vnic_cq_service fnic_cq_service
+#define vnic_cq_free fnic_cq_free
+#define vnic_cq_alloc fnic_cq_alloc
+#define vnic_cq_init fnic_cq_init
+#define vnic_cq_clean fnic_cq_clean
+
+/* Completion queue control */
+struct vnic_cq_ctrl {
+	u64 ring_base;			/* 0x00 */
+	u32 ring_size;			/* 0x08 */
+	u32 pad0;
+	u32 flow_control_enable;	/* 0x10 */
+	u32 pad1;
+	u32 color_enable;		/* 0x18 */
+	u32 pad2;
+	u32 cq_head;			/* 0x20 */
+	u32 pad3;
+	u32 cq_tail;			/* 0x28 */
+	u32 pad4;
+	u32 cq_tail_color;		/* 0x30 */
+	u32 pad5;
+	u32 interrupt_enable;		/* 0x38 */
+	u32 pad6;
+	u32 cq_entry_enable;		/* 0x40 */
+	u32 pad7;
+	u32 cq_message_enable;		/* 0x48 */
+	u32 pad8;
+	u32 interrupt_offset;		/* 0x50 */
+	u32 pad9;
+	u64 cq_message_addr;		/* 0x58 */
+	u32 pad10;
+};
+
+struct vnic_cq {
+	unsigned int index;
+	struct vnic_dev *vdev;
+	struct vnic_cq_ctrl __iomem *ctrl;	/* memory-mapped */
+	struct vnic_dev_ring ring;
+	unsigned int to_clean;
+	unsigned int last_color;
+};
+
+static inline unsigned int vnic_cq_service(struct vnic_cq *cq,
+	unsigned int work_to_do,
+	int (*q_service)(struct vnic_dev *vdev, struct cq_desc *cq_desc,
+	u8 type, u16 q_number, u16 completed_index, void *opaque),
+	void *opaque)
+{
+	struct cq_desc *cq_desc;
+	unsigned int work_done = 0;
+	u16 q_number, completed_index;
+	u8 type, color;
+
+	cq_desc = (struct cq_desc *)((u8 *)cq->ring.descs +
+		cq->ring.desc_size * cq->to_clean);
+	cq_desc_dec(cq_desc, &type, &color,
+		&q_number, &completed_index);
+
+	while (color != cq->last_color) {
+
+		if ((*q_service)(cq->vdev, cq_desc, type,
+			q_number, completed_index, opaque))
+			break;
+
+		cq->to_clean++;
+		if (cq->to_clean == cq->ring.desc_count) {
+			cq->to_clean = 0;
+			cq->last_color = cq->last_color ? 0 : 1;
+		}
+
+		cq_desc = (struct cq_desc *)((u8 *)cq->ring.descs +
+			cq->ring.desc_size * cq->to_clean);
+		cq_desc_dec(cq_desc, &type, &color,
+			&q_number, &completed_index);
+
+		work_done++;
+		if (work_done >= work_to_do)
+			break;
+	}
+
+	return work_done;
+}
+
+void vnic_cq_free(struct vnic_cq *cq);
+int vnic_cq_alloc(struct vnic_dev *vdev, struct vnic_cq *cq, unsigned int index,
+	unsigned int desc_count, unsigned int desc_size);
+void vnic_cq_init(struct vnic_cq *cq, unsigned int flow_control_enable,
+	unsigned int color_enable, unsigned int cq_head, unsigned int cq_tail,
+	unsigned int cq_tail_color, unsigned int interrupt_enable,
+	unsigned int cq_entry_enable, unsigned int message_enable,
+	unsigned int interrupt_offset, u64 message_addr);
+void vnic_cq_clean(struct vnic_cq *cq);
+
+#endif /* _VNIC_CQ_H_ */

drivers/scsi/fnic/vnic_cq_copy.h

@@ -0,0 +1,62 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _VNIC_CQ_COPY_H_
+#define _VNIC_CQ_COPY_H_
+
+#include "fcpio.h"
+
+static inline unsigned int vnic_cq_copy_service(
+	struct vnic_cq *cq,
+	int (*q_service)(struct vnic_dev *vdev,
+			 unsigned int index,
+			 struct fcpio_fw_req *desc),
+	unsigned int work_to_do)
+
+{
+	struct fcpio_fw_req *desc;
+	unsigned int work_done = 0;
+	u8 color;
+
+	desc = (struct fcpio_fw_req *)((u8 *)cq->ring.descs +
+		cq->ring.desc_size * cq->to_clean);
+	fcpio_color_dec(desc, &color);
+
+	while (color != cq->last_color) {
+
+		if ((*q_service)(cq->vdev, cq->index, desc))
+			break;
+
+		cq->to_clean++;
+		if (cq->to_clean == cq->ring.desc_count) {
+			cq->to_clean = 0;
+			cq->last_color = cq->last_color ? 0 : 1;
+		}
+
+		desc = (struct fcpio_fw_req *)((u8 *)cq->ring.descs +
+			cq->ring.desc_size * cq->to_clean);
+		fcpio_color_dec(desc, &color);
+
+		work_done++;
+		if (work_done >= work_to_do)
+			break;
+	}
+
+	return work_done;
+}
+
+#endif /* _VNIC_CQ_COPY_H_ */

drivers/scsi/fnic/vnic_dev.c

@@ -0,0 +1,690 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include <linux/if_ether.h>
+#include "vnic_resource.h"
+#include "vnic_devcmd.h"
+#include "vnic_dev.h"
+#include "vnic_stats.h"
+
+struct vnic_res {
+	void __iomem *vaddr;
+	unsigned int count;
+};
+
+struct vnic_dev {
+	void *priv;
+	struct pci_dev *pdev;
+	struct vnic_res res[RES_TYPE_MAX];
+	enum vnic_dev_intr_mode intr_mode;
+	struct vnic_devcmd __iomem *devcmd;
+	struct vnic_devcmd_notify *notify;
+	struct vnic_devcmd_notify notify_copy;
+	dma_addr_t notify_pa;
+	u32 *linkstatus;
+	dma_addr_t linkstatus_pa;
+	struct vnic_stats *stats;
+	dma_addr_t stats_pa;
+	struct vnic_devcmd_fw_info *fw_info;
+	dma_addr_t fw_info_pa;
+};
+
+#define VNIC_MAX_RES_HDR_SIZE \
+	(sizeof(struct vnic_resource_header) + \
+	sizeof(struct vnic_resource) * RES_TYPE_MAX)
+#define VNIC_RES_STRIDE	128
+
+void *vnic_dev_priv(struct vnic_dev *vdev)
+{
+	return vdev->priv;
+}
+
+static int vnic_dev_discover_res(struct vnic_dev *vdev,
+	struct vnic_dev_bar *bar)
+{
+	struct vnic_resource_header __iomem *rh;
+	struct vnic_resource __iomem *r;
+	u8 type;
+
+	if (bar->len < VNIC_MAX_RES_HDR_SIZE) {
+		printk(KERN_ERR "vNIC BAR0 res hdr length error\n");
+		return -EINVAL;
+	}
+
+	rh = bar->vaddr;
+	if (!rh) {
+		printk(KERN_ERR "vNIC BAR0 res hdr not mem-mapped\n");
+		return -EINVAL;
+	}
+
+	if (ioread32(&rh->magic) != VNIC_RES_MAGIC ||
+	    ioread32(&rh->version) != VNIC_RES_VERSION) {
+		printk(KERN_ERR "vNIC BAR0 res magic/version error "
+			"exp (%lx/%lx) curr (%x/%x)\n",
+			VNIC_RES_MAGIC, VNIC_RES_VERSION,
+			ioread32(&rh->magic), ioread32(&rh->version));
+		return -EINVAL;
+	}
+
+	r = (struct vnic_resource __iomem *)(rh + 1);
+
+	while ((type = ioread8(&r->type)) != RES_TYPE_EOL) {
+
+		u8 bar_num = ioread8(&r->bar);
+		u32 bar_offset = ioread32(&r->bar_offset);
+		u32 count = ioread32(&r->count);
+		u32 len;
+
+		r++;
+
+		if (bar_num != 0)  /* only mapping in BAR0 resources */
+			continue;
+
+		switch (type) {
+		case RES_TYPE_WQ:
+		case RES_TYPE_RQ:
+		case RES_TYPE_CQ:
+		case RES_TYPE_INTR_CTRL:
+			/* each count is stride bytes long */
+			len = count * VNIC_RES_STRIDE;
+			if (len + bar_offset > bar->len) {
+				printk(KERN_ERR "vNIC BAR0 resource %d "
+					"out-of-bounds, offset 0x%x + "
+					"size 0x%x > bar len 0x%lx\n",
+					type, bar_offset,
+					len,
+					bar->len);
+				return -EINVAL;
+			}
+			break;
+		case RES_TYPE_INTR_PBA_LEGACY:
+		case RES_TYPE_DEVCMD:
+			len = count;
+			break;
+		default:
+			continue;
+		}
+
+		vdev->res[type].count = count;
+		vdev->res[type].vaddr = (char __iomem *)bar->vaddr + bar_offset;
+	}
+
+	return 0;
+}
+
+unsigned int vnic_dev_get_res_count(struct vnic_dev *vdev,
+	enum vnic_res_type type)
+{
+	return vdev->res[type].count;
+}
+
+void __iomem *vnic_dev_get_res(struct vnic_dev *vdev, enum vnic_res_type type,
+	unsigned int index)
+{
+	if (!vdev->res[type].vaddr)
+		return NULL;
+
+	switch (type) {
+	case RES_TYPE_WQ:
+	case RES_TYPE_RQ:
+	case RES_TYPE_CQ:
+	case RES_TYPE_INTR_CTRL:
+		return (char __iomem *)vdev->res[type].vaddr +
+					index * VNIC_RES_STRIDE;
+	default:
+		return (char __iomem *)vdev->res[type].vaddr;
+	}
+}
+
+unsigned int vnic_dev_desc_ring_size(struct vnic_dev_ring *ring,
+				     unsigned int desc_count,
+				     unsigned int desc_size)
+{
+	/* The base address of the desc rings must be 512 byte aligned.
+	 * Descriptor count is aligned to groups of 32 descriptors.  A
+	 * count of 0 means the maximum 4096 descriptors.  Descriptor
+	 * size is aligned to 16 bytes.
+	 */
+
+	unsigned int count_align = 32;
+	unsigned int desc_align = 16;
+
+	ring->base_align = 512;
+
+	if (desc_count == 0)
+		desc_count = 4096;
+
+	ring->desc_count = ALIGN(desc_count, count_align);
+
+	ring->desc_size = ALIGN(desc_size, desc_align);
+
+	ring->size = ring->desc_count * ring->desc_size;
+	ring->size_unaligned = ring->size + ring->base_align;
+
+	return ring->size_unaligned;
+}
+
+void vnic_dev_clear_desc_ring(struct vnic_dev_ring *ring)
+{
+	memset(ring->descs, 0, ring->size);
+}
+
+int vnic_dev_alloc_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring,
+	unsigned int desc_count, unsigned int desc_size)
+{
+	vnic_dev_desc_ring_size(ring, desc_count, desc_size);
+
+	ring->descs_unaligned = pci_alloc_consistent(vdev->pdev,
+		ring->size_unaligned,
+		&ring->base_addr_unaligned);
+
+	if (!ring->descs_unaligned) {
+		printk(KERN_ERR
+		  "Failed to allocate ring (size=%d), aborting\n",
+			(int)ring->size);
+		return -ENOMEM;
+	}
+
+	ring->base_addr = ALIGN(ring->base_addr_unaligned,
+		ring->base_align);
+	ring->descs = (u8 *)ring->descs_unaligned +
+		(ring->base_addr - ring->base_addr_unaligned);
+
+	vnic_dev_clear_desc_ring(ring);
+
+	ring->desc_avail = ring->desc_count - 1;
+
+	return 0;
+}
+
+void vnic_dev_free_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring)
+{
+	if (ring->descs) {
+		pci_free_consistent(vdev->pdev,
+			ring->size_unaligned,
+			ring->descs_unaligned,
+			ring->base_addr_unaligned);
+		ring->descs = NULL;
+	}
+}
+
+int vnic_dev_cmd(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
+	u64 *a0, u64 *a1, int wait)
+{
+	struct vnic_devcmd __iomem *devcmd = vdev->devcmd;
+	int delay;
+	u32 status;
+	int dev_cmd_err[] = {
+		/* convert from fw's version of error.h to host's version */
+		0,	/* ERR_SUCCESS */
+		EINVAL,	/* ERR_EINVAL */
+		EFAULT,	/* ERR_EFAULT */
+		EPERM,	/* ERR_EPERM */
+		EBUSY,  /* ERR_EBUSY */
+	};
+	int err;
+
+	status = ioread32(&devcmd->status);
+	if (status & STAT_BUSY) {
+		printk(KERN_ERR "Busy devcmd %d\n", _CMD_N(cmd));
+		return -EBUSY;
+	}
+
+	if (_CMD_DIR(cmd) & _CMD_DIR_WRITE) {
+		writeq(*a0, &devcmd->args[0]);
+		writeq(*a1, &devcmd->args[1]);
+		wmb();
+	}
+
+	iowrite32(cmd, &devcmd->cmd);
+
+	if ((_CMD_FLAGS(cmd) & _CMD_FLAGS_NOWAIT))
+			return 0;
+
+	for (delay = 0; delay < wait; delay++) {
+
+		udelay(100);
+
+		status = ioread32(&devcmd->status);
+		if (!(status & STAT_BUSY)) {
+
+			if (status & STAT_ERROR) {
+				err = dev_cmd_err[(int)readq(&devcmd->args[0])];
+				printk(KERN_ERR "Error %d devcmd %d\n",
+					err, _CMD_N(cmd));
+				return -err;
+			}
+
+			if (_CMD_DIR(cmd) & _CMD_DIR_READ) {
+				rmb();
+				*a0 = readq(&devcmd->args[0]);
+				*a1 = readq(&devcmd->args[1]);
+			}
+
+			return 0;
+		}
+	}
+
+	printk(KERN_ERR "Timedout devcmd %d\n", _CMD_N(cmd));
+	return -ETIMEDOUT;
+}
+
+int vnic_dev_fw_info(struct vnic_dev *vdev,
+	struct vnic_devcmd_fw_info **fw_info)
+{
+	u64 a0, a1 = 0;
+	int wait = 1000;
+	int err = 0;
+
+	if (!vdev->fw_info) {
+		vdev->fw_info = pci_alloc_consistent(vdev->pdev,
+			sizeof(struct vnic_devcmd_fw_info),
+			&vdev->fw_info_pa);
+		if (!vdev->fw_info)
+			return -ENOMEM;
+
+		a0 = vdev->fw_info_pa;
+
+		/* only get fw_info once and cache it */
+		err = vnic_dev_cmd(vdev, CMD_MCPU_FW_INFO, &a0, &a1, wait);
+	}
+
+	*fw_info = vdev->fw_info;
+
+	return err;
+}
+
+int vnic_dev_spec(struct vnic_dev *vdev, unsigned int offset, unsigned int size,
+	void *value)
+{
+	u64 a0, a1;
+	int wait = 1000;
+	int err;
+
+	a0 = offset;
+	a1 = size;
+
+	err = vnic_dev_cmd(vdev, CMD_DEV_SPEC, &a0, &a1, wait);
+
+	switch (size) {
+	case 1:
+		*(u8 *)value = (u8)a0;
+		break;
+	case 2:
+		*(u16 *)value = (u16)a0;
+		break;
+	case 4:
+		*(u32 *)value = (u32)a0;
+		break;
+	case 8:
+		*(u64 *)value = a0;
+		break;
+	default:
+		BUG();
+		break;
+	}
+
+	return err;
+}
+
+int vnic_dev_stats_clear(struct vnic_dev *vdev)
+{
+	u64 a0 = 0, a1 = 0;
+	int wait = 1000;
+	return vnic_dev_cmd(vdev, CMD_STATS_CLEAR, &a0, &a1, wait);
+}
+
+int vnic_dev_stats_dump(struct vnic_dev *vdev, struct vnic_stats **stats)
+{
+	u64 a0, a1;
+	int wait = 1000;
+
+	if (!vdev->stats) {
+		vdev->stats = pci_alloc_consistent(vdev->pdev,
+			sizeof(struct vnic_stats), &vdev->stats_pa);
+		if (!vdev->stats)
+			return -ENOMEM;
+	}
+
+	*stats = vdev->stats;
+	a0 = vdev->stats_pa;
+	a1 = sizeof(struct vnic_stats);
+
+	return vnic_dev_cmd(vdev, CMD_STATS_DUMP, &a0, &a1, wait);
+}
+
+int vnic_dev_close(struct vnic_dev *vdev)
+{
+	u64 a0 = 0, a1 = 0;
+	int wait = 1000;
+	return vnic_dev_cmd(vdev, CMD_CLOSE, &a0, &a1, wait);
+}
+
+int vnic_dev_enable(struct vnic_dev *vdev)
+{
+	u64 a0 = 0, a1 = 0;
+	int wait = 1000;
+	return vnic_dev_cmd(vdev, CMD_ENABLE, &a0, &a1, wait);
+}
+
+int vnic_dev_disable(struct vnic_dev *vdev)
+{
+	u64 a0 = 0, a1 = 0;
+	int wait = 1000;
+	return vnic_dev_cmd(vdev, CMD_DISABLE, &a0, &a1, wait);
+}
+
+int vnic_dev_open(struct vnic_dev *vdev, int arg)
+{
+	u64 a0 = (u32)arg, a1 = 0;
+	int wait = 1000;
+	return vnic_dev_cmd(vdev, CMD_OPEN, &a0, &a1, wait);
+}
+
+int vnic_dev_open_done(struct vnic_dev *vdev, int *done)
+{
+	u64 a0 = 0, a1 = 0;
+	int wait = 1000;
+	int err;
+
+	*done = 0;
+
+	err = vnic_dev_cmd(vdev, CMD_OPEN_STATUS, &a0, &a1, wait);
+	if (err)
+		return err;
+
+	*done = (a0 == 0);
+
+	return 0;
+}
+
+int vnic_dev_soft_reset(struct vnic_dev *vdev, int arg)
+{
+	u64 a0 = (u32)arg, a1 = 0;
+	int wait = 1000;
+	return vnic_dev_cmd(vdev, CMD_SOFT_RESET, &a0, &a1, wait);
+}
+
+int vnic_dev_soft_reset_done(struct vnic_dev *vdev, int *done)
+{
+	u64 a0 = 0, a1 = 0;
+	int wait = 1000;
+	int err;
+
+	*done = 0;
+
+	err = vnic_dev_cmd(vdev, CMD_SOFT_RESET_STATUS, &a0, &a1, wait);
+	if (err)
+		return err;
+
+	*done = (a0 == 0);
+
+	return 0;
+}
+
+int vnic_dev_hang_notify(struct vnic_dev *vdev)
+{
+	u64 a0, a1;
+	int wait = 1000;
+	return vnic_dev_cmd(vdev, CMD_HANG_NOTIFY, &a0, &a1, wait);
+}
+
+int vnic_dev_mac_addr(struct vnic_dev *vdev, u8 *mac_addr)
+{
+	u64 a0, a1;
+	int wait = 1000;
+	int err, i;
+
+	for (i = 0; i < ETH_ALEN; i++)
+		mac_addr[i] = 0;
+
+	err = vnic_dev_cmd(vdev, CMD_MAC_ADDR, &a0, &a1, wait);
+	if (err)
+		return err;
+
+	for (i = 0; i < ETH_ALEN; i++)
+		mac_addr[i] = ((u8 *)&a0)[i];
+
+	return 0;
+}
+
+void vnic_dev_packet_filter(struct vnic_dev *vdev, int directed, int multicast,
+	int broadcast, int promisc, int allmulti)
+{
+	u64 a0, a1 = 0;
+	int wait = 1000;
+	int err;
+
+	a0 = (directed ? CMD_PFILTER_DIRECTED : 0) |
+	     (multicast ? CMD_PFILTER_MULTICAST : 0) |
+	     (broadcast ? CMD_PFILTER_BROADCAST : 0) |
+	     (promisc ? CMD_PFILTER_PROMISCUOUS : 0) |
+	     (allmulti ? CMD_PFILTER_ALL_MULTICAST : 0);
+
+	err = vnic_dev_cmd(vdev, CMD_PACKET_FILTER, &a0, &a1, wait);
+	if (err)
+		printk(KERN_ERR "Can't set packet filter\n");
+}
+
+void vnic_dev_add_addr(struct vnic_dev *vdev, u8 *addr)
+{
+	u64 a0 = 0, a1 = 0;
+	int wait = 1000;
+	int err;
+	int i;
+
+	for (i = 0; i < ETH_ALEN; i++)
+		((u8 *)&a0)[i] = addr[i];
+
+	err = vnic_dev_cmd(vdev, CMD_ADDR_ADD, &a0, &a1, wait);
+	if (err)
+		printk(KERN_ERR
+			"Can't add addr [%02x:%02x:%02x:%02x:%02x:%02x], %d\n",
+			addr[0], addr[1], addr[2], addr[3], addr[4], addr[5],
+			err);
+}
+
+void vnic_dev_del_addr(struct vnic_dev *vdev, u8 *addr)
+{
+	u64 a0 = 0, a1 = 0;
+	int wait = 1000;
+	int err;
+	int i;
+
+	for (i = 0; i < ETH_ALEN; i++)
+		((u8 *)&a0)[i] = addr[i];
+
+	err = vnic_dev_cmd(vdev, CMD_ADDR_DEL, &a0, &a1, wait);
+	if (err)
+		printk(KERN_ERR
+			"Can't del addr [%02x:%02x:%02x:%02x:%02x:%02x], %d\n",
+			addr[0], addr[1], addr[2], addr[3], addr[4], addr[5],
+			err);
+}
+
+int vnic_dev_notify_set(struct vnic_dev *vdev, u16 intr)
+{
+	u64 a0, a1;
+	int wait = 1000;
+
+	if (!vdev->notify) {
+		vdev->notify = pci_alloc_consistent(vdev->pdev,
+			sizeof(struct vnic_devcmd_notify),
+			&vdev->notify_pa);
+		if (!vdev->notify)
+			return -ENOMEM;
+	}
+
+	a0 = vdev->notify_pa;
+	a1 = ((u64)intr << 32) & 0x0000ffff00000000ULL;
+	a1 += sizeof(struct vnic_devcmd_notify);
+
+	return vnic_dev_cmd(vdev, CMD_NOTIFY, &a0, &a1, wait);
+}
+
+void vnic_dev_notify_unset(struct vnic_dev *vdev)
+{
+	u64 a0, a1;
+	int wait = 1000;
+
+	a0 = 0;  /* paddr = 0 to unset notify buffer */
+	a1 = 0x0000ffff00000000ULL; /* intr num = -1 to unreg for intr */
+	a1 += sizeof(struct vnic_devcmd_notify);
+
+	vnic_dev_cmd(vdev, CMD_NOTIFY, &a0, &a1, wait);
+}
+
+static int vnic_dev_notify_ready(struct vnic_dev *vdev)
+{
+	u32 *words;
+	unsigned int nwords = sizeof(struct vnic_devcmd_notify) / 4;
+	unsigned int i;
+	u32 csum;
+
+	if (!vdev->notify)
+		return 0;
+
+	do {
+		csum = 0;
+		memcpy(&vdev->notify_copy, vdev->notify,
+			sizeof(struct vnic_devcmd_notify));
+		words = (u32 *)&vdev->notify_copy;
+		for (i = 1; i < nwords; i++)
+			csum += words[i];
+	} while (csum != words[0]);
+
+	return 1;
+}
+
+int vnic_dev_init(struct vnic_dev *vdev, int arg)
+{
+	u64 a0 = (u32)arg, a1 = 0;
+	int wait = 1000;
+	return vnic_dev_cmd(vdev, CMD_INIT, &a0, &a1, wait);
+}
+
+int vnic_dev_link_status(struct vnic_dev *vdev)
+{
+	if (vdev->linkstatus)
+		return *vdev->linkstatus;
+
+	if (!vnic_dev_notify_ready(vdev))
+		return 0;
+
+	return vdev->notify_copy.link_state;
+}
+
+u32 vnic_dev_port_speed(struct vnic_dev *vdev)
+{
+	if (!vnic_dev_notify_ready(vdev))
+		return 0;
+
+	return vdev->notify_copy.port_speed;
+}
+
+u32 vnic_dev_msg_lvl(struct vnic_dev *vdev)
+{
+	if (!vnic_dev_notify_ready(vdev))
+		return 0;
+
+	return vdev->notify_copy.msglvl;
+}
+
+u32 vnic_dev_mtu(struct vnic_dev *vdev)
+{
+	if (!vnic_dev_notify_ready(vdev))
+		return 0;
+
+	return vdev->notify_copy.mtu;
+}
+
+u32 vnic_dev_link_down_cnt(struct vnic_dev *vdev)
+{
+	if (!vnic_dev_notify_ready(vdev))
+		return 0;
+
+	return vdev->notify_copy.link_down_cnt;
+}
+
+void vnic_dev_set_intr_mode(struct vnic_dev *vdev,
+	enum vnic_dev_intr_mode intr_mode)
+{
+	vdev->intr_mode = intr_mode;
+}
+
+enum vnic_dev_intr_mode vnic_dev_get_intr_mode(
+	struct vnic_dev *vdev)
+{
+	return vdev->intr_mode;
+}
+
+void vnic_dev_unregister(struct vnic_dev *vdev)
+{
+	if (vdev) {
+		if (vdev->notify)
+			pci_free_consistent(vdev->pdev,
+				sizeof(struct vnic_devcmd_notify),
+				vdev->notify,
+				vdev->notify_pa);
+		if (vdev->linkstatus)
+			pci_free_consistent(vdev->pdev,
+				sizeof(u32),
+				vdev->linkstatus,
+				vdev->linkstatus_pa);
+		if (vdev->stats)
+			pci_free_consistent(vdev->pdev,
+				sizeof(struct vnic_dev),
+				vdev->stats, vdev->stats_pa);
+		if (vdev->fw_info)
+			pci_free_consistent(vdev->pdev,
+				sizeof(struct vnic_devcmd_fw_info),
+				vdev->fw_info, vdev->fw_info_pa);
+		kfree(vdev);
+	}
+}
+
+struct vnic_dev *vnic_dev_register(struct vnic_dev *vdev,
+	void *priv, struct pci_dev *pdev, struct vnic_dev_bar *bar)
+{
+	if (!vdev) {
+		vdev = kzalloc(sizeof(struct vnic_dev), GFP_KERNEL);
+		if (!vdev)
+			return NULL;
+	}
+
+	vdev->priv = priv;
+	vdev->pdev = pdev;
+
+	if (vnic_dev_discover_res(vdev, bar))
+		goto err_out;
+
+	vdev->devcmd = vnic_dev_get_res(vdev, RES_TYPE_DEVCMD, 0);
+	if (!vdev->devcmd)
+		goto err_out;
+
+	return vdev;
+
+err_out:
+	vnic_dev_unregister(vdev);
+	return NULL;
+}

drivers/scsi/fnic/vnic_dev.h

@@ -0,0 +1,161 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _VNIC_DEV_H_
+#define _VNIC_DEV_H_
+
+#include "vnic_resource.h"
+#include "vnic_devcmd.h"
+
+/*
+ * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth
+ * Driver) when both are built with CONFIG options =y
+ */
+#define vnic_dev_priv fnic_dev_priv
+#define vnic_dev_get_res_count fnic_dev_get_res_count
+#define vnic_dev_get_res fnic_dev_get_res
+#define vnic_dev_desc_ring_size fnic_dev_desc_ring_siz
+#define vnic_dev_clear_desc_ring fnic_dev_clear_desc_ring
+#define vnic_dev_alloc_desc_ring fnic_dev_alloc_desc_ring
+#define vnic_dev_free_desc_ring fnic_dev_free_desc_ring
+#define vnic_dev_cmd fnic_dev_cmd
+#define vnic_dev_fw_info fnic_dev_fw_info
+#define vnic_dev_spec fnic_dev_spec
+#define vnic_dev_stats_clear fnic_dev_stats_clear
+#define vnic_dev_stats_dump fnic_dev_stats_dump
+#define vnic_dev_hang_notify fnic_dev_hang_notify
+#define vnic_dev_packet_filter fnic_dev_packet_filter
+#define vnic_dev_add_addr fnic_dev_add_addr
+#define vnic_dev_del_addr fnic_dev_del_addr
+#define vnic_dev_mac_addr fnic_dev_mac_addr
+#define vnic_dev_notify_set fnic_dev_notify_set
+#define vnic_dev_notify_unset fnic_dev_notify_unset
+#define vnic_dev_link_status fnic_dev_link_status
+#define vnic_dev_port_speed fnic_dev_port_speed
+#define vnic_dev_msg_lvl fnic_dev_msg_lvl
+#define vnic_dev_mtu fnic_dev_mtu
+#define vnic_dev_link_down_cnt fnic_dev_link_down_cnt
+#define vnic_dev_close fnic_dev_close
+#define vnic_dev_enable fnic_dev_enable
+#define vnic_dev_disable fnic_dev_disable
+#define vnic_dev_open fnic_dev_open
+#define vnic_dev_open_done fnic_dev_open_done
+#define vnic_dev_init fnic_dev_init
+#define vnic_dev_soft_reset fnic_dev_soft_reset
+#define vnic_dev_soft_reset_done fnic_dev_soft_reset_done
+#define vnic_dev_set_intr_mode fnic_dev_set_intr_mode
+#define vnic_dev_get_intr_mode fnic_dev_get_intr_mode
+#define vnic_dev_unregister fnic_dev_unregister
+#define vnic_dev_register fnic_dev_register
+
+#ifndef VNIC_PADDR_TARGET
+#define VNIC_PADDR_TARGET	0x0000000000000000ULL
+#endif
+
+#ifndef readq
+static inline u64 readq(void __iomem *reg)
+{
+	return ((u64)readl(reg + 0x4UL) << 32) | (u64)readl(reg);
+}
+
+static inline void writeq(u64 val, void __iomem *reg)
+{
+	writel(val & 0xffffffff, reg);
+	writel(val >> 32, reg + 0x4UL);
+}
+#endif
+
+enum vnic_dev_intr_mode {
+	VNIC_DEV_INTR_MODE_UNKNOWN,
+	VNIC_DEV_INTR_MODE_INTX,
+	VNIC_DEV_INTR_MODE_MSI,
+	VNIC_DEV_INTR_MODE_MSIX,
+};
+
+struct vnic_dev_bar {
+	void __iomem *vaddr;
+	dma_addr_t bus_addr;
+	unsigned long len;
+};
+
+struct vnic_dev_ring {
+	void *descs;
+	size_t size;
+	dma_addr_t base_addr;
+	size_t base_align;
+	void *descs_unaligned;
+	size_t size_unaligned;
+	dma_addr_t base_addr_unaligned;
+	unsigned int desc_size;
+	unsigned int desc_count;
+	unsigned int desc_avail;
+};
+
+struct vnic_dev;
+struct vnic_stats;
+
+void *vnic_dev_priv(struct vnic_dev *vdev);
+unsigned int vnic_dev_get_res_count(struct vnic_dev *vdev,
+				    enum vnic_res_type type);
+void __iomem *vnic_dev_get_res(struct vnic_dev *vdev, enum vnic_res_type type,
+			       unsigned int index);
+unsigned int vnic_dev_desc_ring_size(struct vnic_dev_ring *ring,
+				     unsigned int desc_count,
+				     unsigned int desc_size);
+void vnic_dev_clear_desc_ring(struct vnic_dev_ring *ring);
+int vnic_dev_alloc_desc_ring(struct vnic_dev *vdev, struct vnic_dev_ring *ring,
+			     unsigned int desc_count, unsigned int desc_size);
+void vnic_dev_free_desc_ring(struct vnic_dev *vdev,
+			     struct vnic_dev_ring *ring);
+int vnic_dev_cmd(struct vnic_dev *vdev, enum vnic_devcmd_cmd cmd,
+		 u64 *a0, u64 *a1, int wait);
+int vnic_dev_fw_info(struct vnic_dev *vdev,
+		     struct vnic_devcmd_fw_info **fw_info);
+int vnic_dev_spec(struct vnic_dev *vdev, unsigned int offset,
+		  unsigned int size, void *value);
+int vnic_dev_stats_clear(struct vnic_dev *vdev);
+int vnic_dev_stats_dump(struct vnic_dev *vdev, struct vnic_stats **stats);
+int vnic_dev_hang_notify(struct vnic_dev *vdev);
+void vnic_dev_packet_filter(struct vnic_dev *vdev, int directed, int multicast,
+			    int broadcast, int promisc, int allmulti);
+void vnic_dev_add_addr(struct vnic_dev *vdev, u8 *addr);
+void vnic_dev_del_addr(struct vnic_dev *vdev, u8 *addr);
+int vnic_dev_mac_addr(struct vnic_dev *vdev, u8 *mac_addr);
+int vnic_dev_notify_set(struct vnic_dev *vdev, u16 intr);
+void vnic_dev_notify_unset(struct vnic_dev *vdev);
+int vnic_dev_link_status(struct vnic_dev *vdev);
+u32 vnic_dev_port_speed(struct vnic_dev *vdev);
+u32 vnic_dev_msg_lvl(struct vnic_dev *vdev);
+u32 vnic_dev_mtu(struct vnic_dev *vdev);
+u32 vnic_dev_link_down_cnt(struct vnic_dev *vdev);
+int vnic_dev_close(struct vnic_dev *vdev);
+int vnic_dev_enable(struct vnic_dev *vdev);
+int vnic_dev_disable(struct vnic_dev *vdev);
+int vnic_dev_open(struct vnic_dev *vdev, int arg);
+int vnic_dev_open_done(struct vnic_dev *vdev, int *done);
+int vnic_dev_init(struct vnic_dev *vdev, int arg);
+int vnic_dev_soft_reset(struct vnic_dev *vdev, int arg);
+int vnic_dev_soft_reset_done(struct vnic_dev *vdev, int *done);
+void vnic_dev_set_intr_mode(struct vnic_dev *vdev,
+			    enum vnic_dev_intr_mode intr_mode);
+enum vnic_dev_intr_mode vnic_dev_get_intr_mode(struct vnic_dev *vdev);
+void vnic_dev_unregister(struct vnic_dev *vdev);
+struct vnic_dev *vnic_dev_register(struct vnic_dev *vdev,
+				   void *priv, struct pci_dev *pdev,
+				   struct vnic_dev_bar *bar);
+
+#endif /* _VNIC_DEV_H_ */

drivers/scsi/fnic/vnic_devcmd.h

@@ -0,0 +1,281 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _VNIC_DEVCMD_H_
+#define _VNIC_DEVCMD_H_
+
+#define _CMD_NBITS      14
+#define _CMD_VTYPEBITS	10
+#define _CMD_FLAGSBITS  6
+#define _CMD_DIRBITS	2
+
+#define _CMD_NMASK      ((1 << _CMD_NBITS)-1)
+#define _CMD_VTYPEMASK  ((1 << _CMD_VTYPEBITS)-1)
+#define _CMD_FLAGSMASK  ((1 << _CMD_FLAGSBITS)-1)
+#define _CMD_DIRMASK    ((1 << _CMD_DIRBITS)-1)
+
+#define _CMD_NSHIFT     0
+#define _CMD_VTYPESHIFT (_CMD_NSHIFT+_CMD_NBITS)
+#define _CMD_FLAGSSHIFT (_CMD_VTYPESHIFT+_CMD_VTYPEBITS)
+#define _CMD_DIRSHIFT   (_CMD_FLAGSSHIFT+_CMD_FLAGSBITS)
+
+/*
+ * Direction bits (from host perspective).
+ */
+#define _CMD_DIR_NONE   0U
+#define _CMD_DIR_WRITE  1U
+#define _CMD_DIR_READ   2U
+#define _CMD_DIR_RW     (_CMD_DIR_WRITE | _CMD_DIR_READ)
+
+/*
+ * Flag bits.
+ */
+#define _CMD_FLAGS_NONE 0U
+#define _CMD_FLAGS_NOWAIT 1U
+
+/*
+ * vNIC type bits.
+ */
+#define _CMD_VTYPE_NONE  0U
+#define _CMD_VTYPE_ENET  1U
+#define _CMD_VTYPE_FC    2U
+#define _CMD_VTYPE_SCSI  4U
+#define _CMD_VTYPE_ALL   (_CMD_VTYPE_ENET | _CMD_VTYPE_FC | _CMD_VTYPE_SCSI)
+
+/*
+ * Used to create cmds..
+*/
+#define _CMDCF(dir, flags, vtype, nr)  \
+	(((dir)   << _CMD_DIRSHIFT) | \
+	((flags) << _CMD_FLAGSSHIFT) | \
+	((vtype) << _CMD_VTYPESHIFT) | \
+	((nr)    << _CMD_NSHIFT))
+#define _CMDC(dir, vtype, nr)    _CMDCF(dir, 0, vtype, nr)
+#define _CMDCNW(dir, vtype, nr)  _CMDCF(dir, _CMD_FLAGS_NOWAIT, vtype, nr)
+
+/*
+ * Used to decode cmds..
+*/
+#define _CMD_DIR(cmd)            (((cmd) >> _CMD_DIRSHIFT) & _CMD_DIRMASK)
+#define _CMD_FLAGS(cmd)          (((cmd) >> _CMD_FLAGSSHIFT) & _CMD_FLAGSMASK)
+#define _CMD_VTYPE(cmd)          (((cmd) >> _CMD_VTYPESHIFT) & _CMD_VTYPEMASK)
+#define _CMD_N(cmd)              (((cmd) >> _CMD_NSHIFT) & _CMD_NMASK)
+
+enum vnic_devcmd_cmd {
+	CMD_NONE                = _CMDC(_CMD_DIR_NONE, _CMD_VTYPE_NONE, 0),
+
+	/* mcpu fw info in mem: (u64)a0=paddr to struct vnic_devcmd_fw_info */
+	CMD_MCPU_FW_INFO        = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 1),
+
+	/* dev-specific block member:
+	 *    in: (u16)a0=offset,(u8)a1=size
+	 *    out: a0=value */
+	CMD_DEV_SPEC            = _CMDC(_CMD_DIR_RW, _CMD_VTYPE_ALL, 2),
+
+	/* stats clear */
+	CMD_STATS_CLEAR         = _CMDCNW(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 3),
+
+	/* stats dump in mem: (u64)a0=paddr to stats area,
+	 *                    (u16)a1=sizeof stats area */
+	CMD_STATS_DUMP          = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 4),
+
+	/* set Rx packet filter: (u32)a0=filters (see CMD_PFILTER_*) */
+	CMD_PACKET_FILTER	= _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 7),
+
+	/* hang detection notification */
+	CMD_HANG_NOTIFY         = _CMDC(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 8),
+
+	/* MAC address in (u48)a0 */
+	CMD_MAC_ADDR            = _CMDC(_CMD_DIR_READ,
+					_CMD_VTYPE_ENET | _CMD_VTYPE_FC, 9),
+
+	/* disable/enable promisc mode: (u8)a0=0/1 */
+/***** XXX DEPRECATED *****/
+	CMD_PROMISC_MODE        = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 10),
+
+	/* disable/enable all-multi mode: (u8)a0=0/1 */
+/***** XXX DEPRECATED *****/
+	CMD_ALLMULTI_MODE       = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 11),
+
+	/* add addr from (u48)a0 */
+	CMD_ADDR_ADD            = _CMDCNW(_CMD_DIR_WRITE,
+					_CMD_VTYPE_ENET | _CMD_VTYPE_FC, 12),
+
+	/* del addr from (u48)a0 */
+	CMD_ADDR_DEL            = _CMDCNW(_CMD_DIR_WRITE,
+					_CMD_VTYPE_ENET | _CMD_VTYPE_FC, 13),
+
+	/* add VLAN id in (u16)a0 */
+	CMD_VLAN_ADD            = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 14),
+
+	/* del VLAN id in (u16)a0 */
+	CMD_VLAN_DEL            = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 15),
+
+	/* nic_cfg in (u32)a0 */
+	CMD_NIC_CFG             = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 16),
+
+	/* union vnic_rss_key in mem: (u64)a0=paddr, (u16)a1=len */
+	CMD_RSS_KEY             = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 17),
+
+	/* union vnic_rss_cpu in mem: (u64)a0=paddr, (u16)a1=len */
+	CMD_RSS_CPU             = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 18),
+
+	/* initiate softreset */
+	CMD_SOFT_RESET          = _CMDCNW(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 19),
+
+	/* softreset status:
+	 *    out: a0=0 reset complete, a0=1 reset in progress */
+	CMD_SOFT_RESET_STATUS   = _CMDC(_CMD_DIR_READ, _CMD_VTYPE_ALL, 20),
+
+	/* set struct vnic_devcmd_notify buffer in mem:
+	 * in:
+	 *   (u64)a0=paddr to notify (set paddr=0 to unset)
+	 *   (u32)a1 & 0x00000000ffffffff=sizeof(struct vnic_devcmd_notify)
+	 *   (u16)a1 & 0x0000ffff00000000=intr num (-1 for no intr)
+	 * out:
+	 *   (u32)a1 = effective size
+	 */
+	CMD_NOTIFY              = _CMDC(_CMD_DIR_RW, _CMD_VTYPE_ALL, 21),
+
+	/* UNDI API: (u64)a0=paddr to s_PXENV_UNDI_ struct,
+	 *           (u8)a1=PXENV_UNDI_xxx */
+	CMD_UNDI                = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 22),
+
+	/* initiate open sequence (u32)a0=flags (see CMD_OPENF_*) */
+	CMD_OPEN		= _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 23),
+
+	/* open status:
+	 *    out: a0=0 open complete, a0=1 open in progress */
+	CMD_OPEN_STATUS		= _CMDC(_CMD_DIR_READ, _CMD_VTYPE_ALL, 24),
+
+	/* close vnic */
+	CMD_CLOSE		= _CMDC(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 25),
+
+	/* initialize virtual link: (u32)a0=flags (see CMD_INITF_*) */
+	CMD_INIT		= _CMDCNW(_CMD_DIR_READ, _CMD_VTYPE_ALL, 26),
+
+	/* variant of CMD_INIT, with provisioning info
+	 *     (u64)a0=paddr of vnic_devcmd_provinfo
+	 *     (u32)a1=sizeof provision info */
+	CMD_INIT_PROV_INFO	= _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 27),
+
+	/* enable virtual link */
+	CMD_ENABLE		= _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 28),
+
+	/* disable virtual link */
+	CMD_DISABLE		= _CMDC(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 29),
+
+	/* stats dump all vnics on uplink in mem: (u64)a0=paddr (u32)a1=uif */
+	CMD_STATS_DUMP_ALL	= _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_ALL, 30),
+
+	/* init status:
+	 *    out: a0=0 init complete, a0=1 init in progress
+	 *         if a0=0, a1=errno */
+	CMD_INIT_STATUS		= _CMDC(_CMD_DIR_READ, _CMD_VTYPE_ALL, 31),
+
+	/* INT13 API: (u64)a0=paddr to vnic_int13_params struct
+	 *            (u8)a1=INT13_CMD_xxx */
+	CMD_INT13               = _CMDC(_CMD_DIR_WRITE, _CMD_VTYPE_FC, 32),
+
+	/* logical uplink enable/disable: (u64)a0: 0/1=disable/enable */
+	CMD_LOGICAL_UPLINK      = _CMDCNW(_CMD_DIR_WRITE, _CMD_VTYPE_ENET, 33),
+
+	/* undo initialize of virtual link */
+	CMD_DEINIT		= _CMDCNW(_CMD_DIR_NONE, _CMD_VTYPE_ALL, 34),
+};
+
+/* flags for CMD_OPEN */
+#define CMD_OPENF_OPROM		0x1	/* open coming from option rom */
+
+/* flags for CMD_INIT */
+#define CMD_INITF_DEFAULT_MAC	0x1	/* init with default mac addr */
+
+/* flags for CMD_PACKET_FILTER */
+#define CMD_PFILTER_DIRECTED		0x01
+#define CMD_PFILTER_MULTICAST		0x02
+#define CMD_PFILTER_BROADCAST		0x04
+#define CMD_PFILTER_PROMISCUOUS		0x08
+#define CMD_PFILTER_ALL_MULTICAST	0x10
+
+enum vnic_devcmd_status {
+	STAT_NONE = 0,
+	STAT_BUSY = 1 << 0,	/* cmd in progress */
+	STAT_ERROR = 1 << 1,	/* last cmd caused error (code in a0) */
+};
+
+enum vnic_devcmd_error {
+	ERR_SUCCESS = 0,
+	ERR_EINVAL = 1,
+	ERR_EFAULT = 2,
+	ERR_EPERM = 3,
+	ERR_EBUSY = 4,
+	ERR_ECMDUNKNOWN = 5,
+	ERR_EBADSTATE = 6,
+	ERR_ENOMEM = 7,
+	ERR_ETIMEDOUT = 8,
+	ERR_ELINKDOWN = 9,
+};
+
+struct vnic_devcmd_fw_info {
+	char fw_version[32];
+	char fw_build[32];
+	char hw_version[32];
+	char hw_serial_number[32];
+};
+
+struct vnic_devcmd_notify {
+	u32 csum;		/* checksum over following words */
+
+	u32 link_state;		/* link up == 1 */
+	u32 port_speed;		/* effective port speed (rate limit) */
+	u32 mtu;		/* MTU */
+	u32 msglvl;		/* requested driver msg lvl */
+	u32 uif;		/* uplink interface */
+	u32 status;		/* status bits (see VNIC_STF_*) */
+	u32 error;		/* error code (see ERR_*) for first ERR */
+	u32 link_down_cnt;	/* running count of link down transitions */
+};
+#define VNIC_STF_FATAL_ERR	0x0001	/* fatal fw error */
+
+struct vnic_devcmd_provinfo {
+	u8 oui[3];
+	u8 type;
+	u8 data[0];
+};
+
+/*
+ * Writing cmd register causes STAT_BUSY to get set in status register.
+ * When cmd completes, STAT_BUSY will be cleared.
+ *
+ * If cmd completed successfully STAT_ERROR will be clear
+ * and args registers contain cmd-specific results.
+ *
+ * If cmd error, STAT_ERROR will be set and args[0] contains error code.
+ *
+ * status register is read-only.  While STAT_BUSY is set,
+ * all other register contents are read-only.
+ */
+
+/* Make sizeof(vnic_devcmd) a power-of-2 for I/O BAR. */
+#define VNIC_DEVCMD_NARGS 15
+struct vnic_devcmd {
+	u32 status;			/* RO */
+	u32 cmd;			/* RW */
+	u64 args[VNIC_DEVCMD_NARGS];	/* RW cmd args (little-endian) */
+};
+
+#endif /* _VNIC_DEVCMD_H_ */

drivers/scsi/fnic/vnic_intr.c

@@ -0,0 +1,60 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/kernel.h>
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include "vnic_dev.h"
+#include "vnic_intr.h"
+
+void vnic_intr_free(struct vnic_intr *intr)
+{
+	intr->ctrl = NULL;
+}
+
+int vnic_intr_alloc(struct vnic_dev *vdev, struct vnic_intr *intr,
+	unsigned int index)
+{
+	intr->index = index;
+	intr->vdev = vdev;
+
+	intr->ctrl = vnic_dev_get_res(vdev, RES_TYPE_INTR_CTRL, index);
+	if (!intr->ctrl) {
+		printk(KERN_ERR "Failed to hook INTR[%d].ctrl resource\n",
+			index);
+		return -EINVAL;
+	}
+
+	return 0;
+}
+
+void vnic_intr_init(struct vnic_intr *intr, unsigned int coalescing_timer,
+	unsigned int coalescing_type, unsigned int mask_on_assertion)
+{
+	iowrite32(coalescing_timer, &intr->ctrl->coalescing_timer);
+	iowrite32(coalescing_type, &intr->ctrl->coalescing_type);
+	iowrite32(mask_on_assertion, &intr->ctrl->mask_on_assertion);
+	iowrite32(0, &intr->ctrl->int_credits);
+}
+
+void vnic_intr_clean(struct vnic_intr *intr)
+{
+	iowrite32(0, &intr->ctrl->int_credits);
+}

drivers/scsi/fnic/vnic_intr.h

@@ -0,0 +1,118 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _VNIC_INTR_H_
+#define _VNIC_INTR_H_
+
+#include <linux/pci.h>
+#include "vnic_dev.h"
+
+/*
+ * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth
+ * Driver) when both are built with CONFIG options =y
+ */
+#define vnic_intr_unmask fnic_intr_unmask
+#define vnic_intr_mask fnic_intr_mask
+#define vnic_intr_return_credits fnic_intr_return_credits
+#define vnic_intr_credits fnic_intr_credits
+#define vnic_intr_return_all_credits fnic_intr_return_all_credits
+#define vnic_intr_legacy_pba fnic_intr_legacy_pba
+#define vnic_intr_free fnic_intr_free
+#define vnic_intr_alloc fnic_intr_alloc
+#define vnic_intr_init fnic_intr_init
+#define vnic_intr_clean fnic_intr_clean
+
+#define VNIC_INTR_TIMER_MAX		0xffff
+
+#define VNIC_INTR_TIMER_TYPE_ABS	0
+#define VNIC_INTR_TIMER_TYPE_QUIET	1
+
+/* Interrupt control */
+struct vnic_intr_ctrl {
+	u32 coalescing_timer;		/* 0x00 */
+	u32 pad0;
+	u32 coalescing_value;		/* 0x08 */
+	u32 pad1;
+	u32 coalescing_type;		/* 0x10 */
+	u32 pad2;
+	u32 mask_on_assertion;		/* 0x18 */
+	u32 pad3;
+	u32 mask;			/* 0x20 */
+	u32 pad4;
+	u32 int_credits;		/* 0x28 */
+	u32 pad5;
+	u32 int_credit_return;		/* 0x30 */
+	u32 pad6;
+};
+
+struct vnic_intr {
+	unsigned int index;
+	struct vnic_dev *vdev;
+	struct vnic_intr_ctrl __iomem *ctrl;	/* memory-mapped */
+};
+
+static inline void vnic_intr_unmask(struct vnic_intr *intr)
+{
+	iowrite32(0, &intr->ctrl->mask);
+}
+
+static inline void vnic_intr_mask(struct vnic_intr *intr)
+{
+	iowrite32(1, &intr->ctrl->mask);
+}
+
+static inline void vnic_intr_return_credits(struct vnic_intr *intr,
+	unsigned int credits, int unmask, int reset_timer)
+{
+#define VNIC_INTR_UNMASK_SHIFT		16
+#define VNIC_INTR_RESET_TIMER_SHIFT	17
+
+	u32 int_credit_return = (credits & 0xffff) |
+		(unmask ? (1 << VNIC_INTR_UNMASK_SHIFT) : 0) |
+		(reset_timer ? (1 << VNIC_INTR_RESET_TIMER_SHIFT) : 0);
+
+	iowrite32(int_credit_return, &intr->ctrl->int_credit_return);
+}
+
+static inline unsigned int vnic_intr_credits(struct vnic_intr *intr)
+{
+	return ioread32(&intr->ctrl->int_credits);
+}
+
+static inline void vnic_intr_return_all_credits(struct vnic_intr *intr)
+{
+	unsigned int credits = vnic_intr_credits(intr);
+	int unmask = 1;
+	int reset_timer = 1;
+
+	vnic_intr_return_credits(intr, credits, unmask, reset_timer);
+}
+
+static inline u32 vnic_intr_legacy_pba(u32 __iomem *legacy_pba)
+{
+	/* read PBA without clearing */
+	return ioread32(legacy_pba);
+}
+
+void vnic_intr_free(struct vnic_intr *intr);
+int vnic_intr_alloc(struct vnic_dev *vdev, struct vnic_intr *intr,
+	unsigned int index);
+void vnic_intr_init(struct vnic_intr *intr, unsigned int coalescing_timer,
+	unsigned int coalescing_type, unsigned int mask_on_assertion);
+void vnic_intr_clean(struct vnic_intr *intr);
+
+#endif /* _VNIC_INTR_H_ */

drivers/scsi/fnic/vnic_nic.h

@@ -0,0 +1,69 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _VNIC_NIC_H_
+#define _VNIC_NIC_H_
+
+/*
+ * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth
+ * Driver) when both are built with CONFIG options =y
+ */
+#define vnic_set_nic_cfg fnic_set_nic_cfg
+
+#define NIC_CFG_RSS_DEFAULT_CPU_MASK_FIELD	0xffUL
+#define NIC_CFG_RSS_DEFAULT_CPU_SHIFT		0
+#define NIC_CFG_RSS_HASH_TYPE			(0xffUL << 8)
+#define NIC_CFG_RSS_HASH_TYPE_MASK_FIELD	0xffUL
+#define NIC_CFG_RSS_HASH_TYPE_SHIFT		8
+#define NIC_CFG_RSS_HASH_BITS			(7UL << 16)
+#define NIC_CFG_RSS_HASH_BITS_MASK_FIELD	7UL
+#define NIC_CFG_RSS_HASH_BITS_SHIFT		16
+#define NIC_CFG_RSS_BASE_CPU			(7UL << 19)
+#define NIC_CFG_RSS_BASE_CPU_MASK_FIELD		7UL
+#define NIC_CFG_RSS_BASE_CPU_SHIFT		19
+#define NIC_CFG_RSS_ENABLE			(1UL << 22)
+#define NIC_CFG_RSS_ENABLE_MASK_FIELD		1UL
+#define NIC_CFG_RSS_ENABLE_SHIFT		22
+#define NIC_CFG_TSO_IPID_SPLIT_EN		(1UL << 23)
+#define NIC_CFG_TSO_IPID_SPLIT_EN_MASK_FIELD	1UL
+#define NIC_CFG_TSO_IPID_SPLIT_EN_SHIFT		23
+#define NIC_CFG_IG_VLAN_STRIP_EN		(1UL << 24)
+#define NIC_CFG_IG_VLAN_STRIP_EN_MASK_FIELD	1UL
+#define NIC_CFG_IG_VLAN_STRIP_EN_SHIFT		24
+
+static inline void vnic_set_nic_cfg(u32 *nic_cfg,
+	u8 rss_default_cpu, u8 rss_hash_type,
+	u8 rss_hash_bits, u8 rss_base_cpu,
+	u8 rss_enable, u8 tso_ipid_split_en,
+	u8 ig_vlan_strip_en)
+{
+	*nic_cfg = (rss_default_cpu & NIC_CFG_RSS_DEFAULT_CPU_MASK_FIELD) |
+		((rss_hash_type & NIC_CFG_RSS_HASH_TYPE_MASK_FIELD)
+			<< NIC_CFG_RSS_HASH_TYPE_SHIFT) |
+		((rss_hash_bits & NIC_CFG_RSS_HASH_BITS_MASK_FIELD)
+			<< NIC_CFG_RSS_HASH_BITS_SHIFT) |
+		((rss_base_cpu & NIC_CFG_RSS_BASE_CPU_MASK_FIELD)
+			<< NIC_CFG_RSS_BASE_CPU_SHIFT) |
+		((rss_enable & NIC_CFG_RSS_ENABLE_MASK_FIELD)
+			<< NIC_CFG_RSS_ENABLE_SHIFT) |
+		((tso_ipid_split_en & NIC_CFG_TSO_IPID_SPLIT_EN_MASK_FIELD)
+			<< NIC_CFG_TSO_IPID_SPLIT_EN_SHIFT) |
+		((ig_vlan_strip_en & NIC_CFG_IG_VLAN_STRIP_EN_MASK_FIELD)
+			<< NIC_CFG_IG_VLAN_STRIP_EN_SHIFT);
+}
+
+#endif /* _VNIC_NIC_H_ */

drivers/scsi/fnic/vnic_resource.h

@@ -0,0 +1,61 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _VNIC_RESOURCE_H_
+#define _VNIC_RESOURCE_H_
+
+#define VNIC_RES_MAGIC		0x766E6963L	/* 'vnic' */
+#define VNIC_RES_VERSION	0x00000000L
+
+/* vNIC resource types */
+enum vnic_res_type {
+	RES_TYPE_EOL,			/* End-of-list */
+	RES_TYPE_WQ,			/* Work queues */
+	RES_TYPE_RQ,			/* Receive queues */
+	RES_TYPE_CQ,			/* Completion queues */
+	RES_TYPE_RSVD1,
+	RES_TYPE_NIC_CFG,		/* Enet NIC config registers */
+	RES_TYPE_RSVD2,
+	RES_TYPE_RSVD3,
+	RES_TYPE_RSVD4,
+	RES_TYPE_RSVD5,
+	RES_TYPE_INTR_CTRL,		/* Interrupt ctrl table */
+	RES_TYPE_INTR_TABLE,		/* MSI/MSI-X Interrupt table */
+	RES_TYPE_INTR_PBA,		/* MSI/MSI-X PBA table */
+	RES_TYPE_INTR_PBA_LEGACY,	/* Legacy intr status */
+	RES_TYPE_RSVD6,
+	RES_TYPE_RSVD7,
+	RES_TYPE_DEVCMD,		/* Device command region */
+	RES_TYPE_PASS_THRU_PAGE,	/* Pass-thru page */
+
+	RES_TYPE_MAX,			/* Count of resource types */
+};
+
+struct vnic_resource_header {
+	u32 magic;
+	u32 version;
+};
+
+struct vnic_resource {
+	u8 type;
+	u8 bar;
+	u8 pad[2];
+	u32 bar_offset;
+	u32 count;
+};
+
+#endif /* _VNIC_RESOURCE_H_ */

drivers/scsi/fnic/vnic_rq.c

@@ -0,0 +1,196 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+
+#include <linux/errno.h>
+#include <linux/types.h>
+#include <linux/pci.h>
+#include <linux/delay.h>
+#include "vnic_dev.h"
+#include "vnic_rq.h"
+
+static int vnic_rq_alloc_bufs(struct vnic_rq *rq)
+{
+	struct vnic_rq_buf *buf;
+	struct vnic_dev *vdev;
+	unsigned int i, j, count = rq->ring.desc_count;
+	unsigned int blks = VNIC_RQ_BUF_BLKS_NEEDED(count);
+
+	vdev = rq->vdev;
+
+	for (i = 0; i < blks; i++) {
+		rq->bufs[i] = kzalloc(VNIC_RQ_BUF_BLK_SZ, GFP_ATOMIC);
+		if (!rq->bufs[i]) {
+			printk(KERN_ERR "Failed to alloc rq_bufs\n");
+			return -ENOMEM;
+		}
+	}
+
+	for (i = 0; i < blks; i++) {
+		buf = rq->bufs[i];
+		for (j = 0; j < VNIC_RQ_BUF_BLK_ENTRIES; j++) {
+			buf->index = i * VNIC_RQ_BUF_BLK_ENTRIES + j;
+			buf->desc = (u8 *)rq->ring.descs +
+				rq->ring.desc_size * buf->index;
+			if (buf->index + 1 == count) {
+				buf->next = rq->bufs[0];
+				break;
+			} else if (j + 1 == VNIC_RQ_BUF_BLK_ENTRIES) {
+				buf->next = rq->bufs[i + 1];
+			} else {
+				buf->next = buf + 1;
+				buf++;
+			}
+		}
+	}
+
+	rq->to_use = rq->to_clean = rq->bufs[0];
+	rq->buf_index = 0;
+
+	return 0;
+}
+
+void vnic_rq_free(struct vnic_rq *rq)
+{
+	struct vnic_dev *vdev;
+	unsigned int i;
+
+	vdev = rq->vdev;
+
+	vnic_dev_free_desc_ring(vdev, &rq->ring);
+
+	for (i = 0; i < VNIC_RQ_BUF_BLKS_MAX; i++) {
+		kfree(rq->bufs[i]);
+		rq->bufs[i] = NULL;
+	}
+
+	rq->ctrl = NULL;
+}
+
+int vnic_rq_alloc(struct vnic_dev *vdev, struct vnic_rq *rq, unsigned int index,
+	unsigned int desc_count, unsigned int desc_size)
+{
+	int err;
+
+	rq->index = index;
+	rq->vdev = vdev;
+
+	rq->ctrl = vnic_dev_get_res(vdev, RES_TYPE_RQ, index);
+	if (!rq->ctrl) {
+		printk(KERN_ERR "Failed to hook RQ[%d] resource\n", index);
+		return -EINVAL;
+	}
+
+	vnic_rq_disable(rq);
+
+	err = vnic_dev_alloc_desc_ring(vdev, &rq->ring, desc_count, desc_size);
+	if (err)
+		return err;
+
+	err = vnic_rq_alloc_bufs(rq);
+	if (err) {
+		vnic_rq_free(rq);
+		return err;
+	}
+
+	return 0;
+}
+
+void vnic_rq_init(struct vnic_rq *rq, unsigned int cq_index,
+	unsigned int error_interrupt_enable,
+	unsigned int error_interrupt_offset)
+{
+	u64 paddr;
+	u32 fetch_index;
+
+	paddr = (u64)rq->ring.base_addr | VNIC_PADDR_TARGET;
+	writeq(paddr, &rq->ctrl->ring_base);
+	iowrite32(rq->ring.desc_count, &rq->ctrl->ring_size);
+	iowrite32(cq_index, &rq->ctrl->cq_index);
+	iowrite32(error_interrupt_enable, &rq->ctrl->error_interrupt_enable);
+	iowrite32(error_interrupt_offset, &rq->ctrl->error_interrupt_offset);
+	iowrite32(0, &rq->ctrl->dropped_packet_count);
+	iowrite32(0, &rq->ctrl->error_status);
+
+	/* Use current fetch_index as the ring starting point */
+	fetch_index = ioread32(&rq->ctrl->fetch_index);
+	rq->to_use = rq->to_clean =
+		&rq->bufs[fetch_index / VNIC_RQ_BUF_BLK_ENTRIES]
+			[fetch_index % VNIC_RQ_BUF_BLK_ENTRIES];
+	iowrite32(fetch_index, &rq->ctrl->posted_index);
+
+	rq->buf_index = 0;
+}
+
+unsigned int vnic_rq_error_status(struct vnic_rq *rq)
+{
+	return ioread32(&rq->ctrl->error_status);
+}
+
+void vnic_rq_enable(struct vnic_rq *rq)
+{
+	iowrite32(1, &rq->ctrl->enable);
+}
+
+int vnic_rq_disable(struct vnic_rq *rq)
+{
+	unsigned int wait;
+
+	iowrite32(0, &rq->ctrl->enable);
+
+	/* Wait for HW to ACK disable request */
+	for (wait = 0; wait < 100; wait++) {
+		if (!(ioread32(&rq->ctrl->running)))
+			return 0;
+		udelay(1);
+	}
+
+	printk(KERN_ERR "Failed to disable RQ[%d]\n", rq->index);
+
+	return -ETIMEDOUT;
+}
+
+void vnic_rq_clean(struct vnic_rq *rq,
+	void (*buf_clean)(struct vnic_rq *rq, struct vnic_rq_buf *buf))
+{
+	struct vnic_rq_buf *buf;
+	u32 fetch_index;
+
+	BUG_ON(ioread32(&rq->ctrl->enable));
+
+	buf = rq->to_clean;
+
+	while (vnic_rq_desc_used(rq) > 0) {
+
+		(*buf_clean)(rq, buf);
+
+		buf = rq->to_clean = buf->next;
+		rq->ring.desc_avail++;
+	}
+
+	/* Use current fetch_index as the ring starting point */
+	fetch_index = ioread32(&rq->ctrl->fetch_index);
+	rq->to_use = rq->to_clean =
+		&rq->bufs[fetch_index / VNIC_RQ_BUF_BLK_ENTRIES]
+			[fetch_index % VNIC_RQ_BUF_BLK_ENTRIES];
+	iowrite32(fetch_index, &rq->ctrl->posted_index);
+
+	rq->buf_index = 0;
+
+	vnic_dev_clear_desc_ring(&rq->ring);
+}
+

drivers/scsi/fnic/vnic_rq.h

@@ -0,0 +1,235 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _VNIC_RQ_H_
+#define _VNIC_RQ_H_
+
+#include <linux/pci.h>
+#include "vnic_dev.h"
+#include "vnic_cq.h"
+
+/*
+ * These defines avoid symbol clash between fnic and enic (Cisco 10G Eth
+ * Driver) when both are built with CONFIG options =y
+ */
+#define vnic_rq_desc_avail fnic_rq_desc_avail
+#define vnic_rq_desc_used fnic_rq_desc_used
+#define vnic_rq_next_desc fnic_rq_next_desc
+#define vnic_rq_next_index fnic_rq_next_index
+#define vnic_rq_next_buf_index fnic_rq_next_buf_index
+#define vnic_rq_post fnic_rq_post
+#define vnic_rq_posting_soon fnic_rq_posting_soon
+#define vnic_rq_return_descs fnic_rq_return_descs
+#define vnic_rq_service fnic_rq_service
+#define vnic_rq_fill fnic_rq_fill
+#define vnic_rq_free fnic_rq_free
+#define vnic_rq_alloc fnic_rq_alloc
+#define vnic_rq_init fnic_rq_init
+#define vnic_rq_error_status fnic_rq_error_status
+#define vnic_rq_enable fnic_rq_enable
+#define vnic_rq_disable fnic_rq_disable
+#define vnic_rq_clean fnic_rq_clean
+
+/* Receive queue control */
+struct vnic_rq_ctrl {
+	u64 ring_base;			/* 0x00 */
+	u32 ring_size;			/* 0x08 */
+	u32 pad0;
+	u32 posted_index;		/* 0x10 */
+	u32 pad1;
+	u32 cq_index;			/* 0x18 */
+	u32 pad2;
+	u32 enable;			/* 0x20 */
+	u32 pad3;
+	u32 running;			/* 0x28 */
+	u32 pad4;
+	u32 fetch_index;		/* 0x30 */
+	u32 pad5;
+	u32 error_interrupt_enable;	/* 0x38 */
+	u32 pad6;
+	u32 error_interrupt_offset;	/* 0x40 */
+	u32 pad7;
+	u32 error_status;		/* 0x48 */
+	u32 pad8;
+	u32 dropped_packet_count;	/* 0x50 */
+	u32 pad9;
+	u32 dropped_packet_count_rc;	/* 0x58 */
+	u32 pad10;
+};
+
+/* Break the vnic_rq_buf allocations into blocks of 64 entries */
+#define VNIC_RQ_BUF_BLK_ENTRIES 64
+#define VNIC_RQ_BUF_BLK_SZ \
+	(VNIC_RQ_BUF_BLK_ENTRIES * sizeof(struct vnic_rq_buf))
+#define VNIC_RQ_BUF_BLKS_NEEDED(entries) \
+	DIV_ROUND_UP(entries, VNIC_RQ_BUF_BLK_ENTRIES)
+#define VNIC_RQ_BUF_BLKS_MAX VNIC_RQ_BUF_BLKS_NEEDED(4096)
+
+struct vnic_rq_buf {
+	struct vnic_rq_buf *next;
+	dma_addr_t dma_addr;
+	void *os_buf;
+	unsigned int os_buf_index;
+	unsigned int len;
+	unsigned int index;
+	void *desc;
+};
+
+struct vnic_rq {
+	unsigned int index;
+	struct vnic_dev *vdev;
+	struct vnic_rq_ctrl __iomem *ctrl;	/* memory-mapped */
+	struct vnic_dev_ring ring;
+	struct vnic_rq_buf *bufs[VNIC_RQ_BUF_BLKS_MAX];
+	struct vnic_rq_buf *to_use;
+	struct vnic_rq_buf *to_clean;
+	void *os_buf_head;
+	unsigned int buf_index;
+	unsigned int pkts_outstanding;
+};
+
+static inline unsigned int vnic_rq_desc_avail(struct vnic_rq *rq)
+{
+	/* how many does SW own? */
+	return rq->ring.desc_avail;
+}
+
+static inline unsigned int vnic_rq_desc_used(struct vnic_rq *rq)
+{
+	/* how many does HW own? */
+	return rq->ring.desc_count - rq->ring.desc_avail - 1;
+}
+
+static inline void *vnic_rq_next_desc(struct vnic_rq *rq)
+{
+	return rq->to_use->desc;
+}
+
+static inline unsigned int vnic_rq_next_index(struct vnic_rq *rq)
+{
+	return rq->to_use->index;
+}
+
+static inline unsigned int vnic_rq_next_buf_index(struct vnic_rq *rq)
+{
+	return rq->buf_index++;
+}
+
+static inline void vnic_rq_post(struct vnic_rq *rq,
+	void *os_buf, unsigned int os_buf_index,
+	dma_addr_t dma_addr, unsigned int len)
+{
+	struct vnic_rq_buf *buf = rq->to_use;
+
+	buf->os_buf = os_buf;
+	buf->os_buf_index = os_buf_index;
+	buf->dma_addr = dma_addr;
+	buf->len = len;
+
+	buf = buf->next;
+	rq->to_use = buf;
+	rq->ring.desc_avail--;
+
+	/* Move the posted_index every nth descriptor
+	 */
+
+#ifndef VNIC_RQ_RETURN_RATE
+#define VNIC_RQ_RETURN_RATE		0xf	/* keep 2^n - 1 */
+#endif
+
+	if ((buf->index & VNIC_RQ_RETURN_RATE) == 0) {
+		/* Adding write memory barrier prevents compiler and/or CPU
+		 * reordering, thus avoiding descriptor posting before
+		 * descriptor is initialized. Otherwise, hardware can read
+		 * stale descriptor fields.
+		 */
+		wmb();
+		iowrite32(buf->index, &rq->ctrl->posted_index);
+	}
+}
+
+static inline int vnic_rq_posting_soon(struct vnic_rq *rq)
+{
+	return (rq->to_use->index & VNIC_RQ_RETURN_RATE) == 0;
+}
+
+static inline void vnic_rq_return_descs(struct vnic_rq *rq, unsigned int count)
+{
+	rq->ring.desc_avail += count;
+}
+
+enum desc_return_options {
+	VNIC_RQ_RETURN_DESC,
+	VNIC_RQ_DEFER_RETURN_DESC,
+};
+
+static inline void vnic_rq_service(struct vnic_rq *rq,
+	struct cq_desc *cq_desc, u16 completed_index,
+	int desc_return, void (*buf_service)(struct vnic_rq *rq,
+	struct cq_desc *cq_desc, struct vnic_rq_buf *buf,
+	int skipped, void *opaque), void *opaque)
+{
+	struct vnic_rq_buf *buf;
+	int skipped;
+
+	buf = rq->to_clean;
+	while (1) {
+
+		skipped = (buf->index != completed_index);
+
+		(*buf_service)(rq, cq_desc, buf, skipped, opaque);
+
+		if (desc_return == VNIC_RQ_RETURN_DESC)
+			rq->ring.desc_avail++;
+
+		rq->to_clean = buf->next;
+
+		if (!skipped)
+			break;
+
+		buf = rq->to_clean;
+	}
+}
+
+static inline int vnic_rq_fill(struct vnic_rq *rq,
+	int (*buf_fill)(struct vnic_rq *rq))
+{
+	int err;
+
+	while (vnic_rq_desc_avail(rq) > 1) {
+
+		err = (*buf_fill)(rq);
+		if (err)
+			return err;
+	}
+
+	return 0;
+}
+
+void vnic_rq_free(struct vnic_rq *rq);
+int vnic_rq_alloc(struct vnic_dev *vdev, struct vnic_rq *rq, unsigned int index,
+	unsigned int desc_count, unsigned int desc_size);
+void vnic_rq_init(struct vnic_rq *rq, unsigned int cq_index,
+	unsigned int error_interrupt_enable,
+	unsigned int error_interrupt_offset);
+unsigned int vnic_rq_error_status(struct vnic_rq *rq);
+void vnic_rq_enable(struct vnic_rq *rq);
+int vnic_rq_disable(struct vnic_rq *rq);
+void vnic_rq_clean(struct vnic_rq *rq,
+	void (*buf_clean)(struct vnic_rq *rq, struct vnic_rq_buf *buf));
+
+#endif /* _VNIC_RQ_H_ */

drivers/scsi/fnic/vnic_scsi.h

@@ -0,0 +1,99 @@
+/*
+ * Copyright 2008 Cisco Systems, Inc.  All rights reserved.
+ * Copyright 2007 Nuova Systems, Inc.  All rights reserved.
+ *
+ * This program is free software; you may redistribute it and/or modify
+ * it under the terms of the GNU General Public License as published by
+ * the Free Software Foundation; version 2 of the License.
+ *
+ * THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND,
+ * EXPRESS OR IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF
+ * MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE AND
+ * NONINFRINGEMENT. IN NO EVENT SHALL THE AUTHORS OR COPYRIGHT HOLDERS
+ * BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER LIABILITY, WHETHER IN AN
+ * ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, OUT OF OR IN
+ * CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
+ * SOFTWARE.
+ */
+#ifndef _VNIC_SCSI_H_
+#define _VNIC_SCSI_H_
+
+#define VNIC_FNIC_WQ_COPY_COUNT_MIN         1
+#define VNIC_FNIC_WQ_COPY_COUNT_MAX         1
+
+#define VNIC_FNIC_WQ_DESCS_MIN              64
+#define VNIC_FNIC_WQ_DESCS_MAX              128
+
+#define VNIC_FNIC_WQ_COPY_DESCS_MIN         64
+#define VNIC_FNIC_WQ_COPY_DESCS_MAX         512
+
+#define VNIC_FNIC_RQ_DESCS_MIN              64
+#define VNIC_FNIC_RQ_DESCS_MAX              128
+
+#define VNIC_FNIC_EDTOV_MIN                 1000
+#define VNIC_FNIC_EDTOV_MAX                 255000
+#define VNIC_FNIC_EDTOV_DEF                 2000
+
+#define VNIC_FNIC_RATOV_MIN                 1000
+#define VNIC_FNIC_RATOV_MAX                 255000
+
+#define VNIC_FNIC_MAXDATAFIELDSIZE_MIN      256
+#define VNIC_FNIC_MAXDATAFIELDSIZE_MAX      2112
+
+#define VNIC_FNIC_FLOGI_RETRIES_MIN         0
+#define VNIC_FNIC_FLOGI_RETRIES_MAX         0xffffffff
+#define VNIC_FNIC_FLOGI_RETRIES_DEF         0xffffffff
+
+#define VNIC_FNIC_FLOGI_TIMEOUT_MIN         1000
+#define VNIC_FNIC_FLOGI_TIMEOUT_MAX         255000
+
+#define VNIC_FNIC_PLOGI_RETRIES_MIN         0
+#define VNIC_FNIC_PLOGI_RETRIES_MAX         255
+#define VNIC_FNIC_PLOGI_RETRIES_DEF         8
+
+#define VNIC_FNIC_PLOGI_TIMEOUT_MIN         1000
+#define VNIC_FNIC_PLOGI_TIMEOUT_MAX         255000
+
+#define VNIC_FNIC_IO_THROTTLE_COUNT_MIN     256
+#define VNIC_FNIC_IO_THROTTLE_COUNT_MAX     4096
+
+#define VNIC_FNIC_LINK_DOWN_TIMEOUT_MIN     0
+#define VNIC_FNIC_LINK_DOWN_TIMEOUT_MAX     240000
+
+#define VNIC_FNIC_PORT_DOWN_TIMEOUT_MIN     0
+#define VNIC_FNIC_PORT_DOWN_TIMEOUT_MAX     240000
+
+#define VNIC_FNIC_PORT_DOWN_IO_RETRIES_MIN  0
+#define VNIC_FNIC_PORT_DOWN_IO_RETRIES_MAX  255
+
+#define VNIC_FNIC_LUNS_PER_TARGET_MIN       1
+#define VNIC_FNIC_LUNS_PER_TARGET_MAX       1024
+
+/* Device-specific region: scsi configuration */
+struct vnic_fc_config {
+	u64 node_wwn;
+	u64 port_wwn;
+	u32 flags;
+	u32 wq_enet_desc_count;
+	u32 wq_copy_desc_count;
+	u32 rq_desc_count;
+	u32 flogi_retries;
+	u32 flogi_timeout;
+	u32 plogi_retries;
+	u32 plogi_timeout;
+	u32 io_throttle_count;
+	u32 link_down_timeout;
+	u32 port_down_timeout;
+	u32 port_down_io_retries;
+	u32 luns_per_tgt;
+	u16 maxdatafieldsize;
+	u16 ed_tov;
+	u16 ra_tov;
+	u16 intr_timer;
+	u8 intr_timer_type;
+};
+
+#define VFCF_FCP_SEQ_LVL_ERR	0x1	/* Enable FCP-2 Error Recovery */
+#define VFCF_PERBI		0x2	/* persistent binding info available */
+
+#endif /* _VNIC_SCSI_H_ */