Merge branch 'next' of git://git.monstr.eu/linux-2.6-microblaze

* 'next' of git://git.monstr.eu/linux-2.6-microblaze: (42 commits)
  microblaze: Fix build with make 3.82
  fbdev/xilinxfb: Microblaze driver support
  microblaze: Support C optimized lib functions for little-endian
  microblaze: Separate library optimized functions
  microblaze: Support timer on AXI lite
  microblaze: Add support for little-endian Microblaze
  microblaze: KGDB little endian support
  microblaze: Add PVR for endians plus detection
  net: emaclite: Add support for little-endian platforms
  microblaze: trivial: Add comment for AXI pvr
  microblaze: pci-common cleanup
  microblaze: Support early console on uart16550
  microblaze: Do not compile early console support for uartlite if is disabled
  microblaze: Setup early console dynamically
  microblaze: Rename all uartlite early printk functions
  microblaze: remove early printk uarlite console dependency from header
  microblaze: Remove additional compatible properties
  microblaze: Remove hardcoded asm instraction for PVR loading
  microblaze: Use static const char * const where possible
  microblaze: Define VMALLOC_START/END
  ...

arch/microblaze/Kconfig

@@ -121,6 +121,23 @@ config CMDLINE_FORCE
 	  Set this to have arguments from the default kernel command string
 	  override those passed by the boot loader.
 
+config SECCOMP
+	bool "Enable seccomp to safely compute untrusted bytecode"
+	depends on PROC_FS
+	default y
+	help
+	  This kernel feature is useful for number crunching applications
+	  that may need to compute untrusted bytecode during their
+	  execution. By using pipes or other transports made available to
+	  the process as file descriptors supporting the read/write
+	  syscalls, it's possible to isolate those applications in
+	  their own address space using seccomp. Once seccomp is
+	  enabled via /proc/<pid>/seccomp, it cannot be disabled
+	  and the task is only allowed to execute a few safe syscalls
+	  defined by each seccomp mode.
+
+	  If unsure, say Y. Only embedded should say N here.
+
 endmenu
 
 menu "Advanced setup"

arch/microblaze/Kconfig.debug

@@ -10,7 +10,7 @@ source "lib/Kconfig.debug"
 
 config EARLY_PRINTK
 	bool "Early printk function for kernel"
-	depends on SERIAL_UARTLITE_CONSOLE
+	depends on SERIAL_UARTLITE_CONSOLE || SERIAL_8250_CONSOLE
 	default n
 	help
 	  This option turns on/off early printk messages to console.

arch/microblaze/Makefile

@@ -42,11 +42,8 @@ KBUILD_CFLAGS += -ffixed-r31 $(CPUFLAGS-1) $(CPUFLAGS-2)
 LDFLAGS		:=
 LDFLAGS_vmlinux	:=
 
-LIBGCC := $(shell $(CC) $(KBUILD_CFLAGS) -print-libgcc-file-name)
-
 head-y := arch/microblaze/kernel/head.o
 libs-y += arch/microblaze/lib/
-libs-y += $(LIBGCC)
 core-y += arch/microblaze/kernel/
 core-y += arch/microblaze/mm/
 core-y += arch/microblaze/platform/
@@ -72,12 +69,16 @@ export MMU DTB
 
 all: linux.bin
 
-BOOT_TARGETS = linux.bin linux.bin.gz simpleImage.%
+# With make 3.82 we cannot mix normal and wildcard targets
+BOOT_TARGETS1 = linux.bin linux.bin.gz
+BOOT_TARGETS2 = simpleImage.%
 
 archclean:
 	$(Q)$(MAKE) $(clean)=$(boot)
 
-$(BOOT_TARGETS): vmlinux
+$(BOOT_TARGETS1): vmlinux
+	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
+$(BOOT_TARGETS2): vmlinux
 	$(Q)$(MAKE) $(build)=$(boot) $(boot)/$@
 
 define archhelp

arch/microblaze/include/asm/byteorder.h

@@ -1,6 +1,10 @@
 #ifndef _ASM_MICROBLAZE_BYTEORDER_H
 #define _ASM_MICROBLAZE_BYTEORDER_H
 
+#ifdef __MICROBLAZEEL__
+#include <linux/byteorder/little_endian.h>
+#else
 #include <linux/byteorder/big_endian.h>
+#endif
 
 #endif /* _ASM_MICROBLAZE_BYTEORDER_H */

arch/microblaze/include/asm/checksum.h

@@ -24,8 +24,13 @@ csum_tcpudp_nofold(__be32 saddr, __be32 daddr, unsigned short len,
 		"addc %0, %0, %3\n\t"
 		"addc %0, %0, r0\n\t"
 		: "+&d" (sum)
-		: "d" (saddr), "d" (daddr), "d" (len + proto));
-
+		: "d" (saddr), "d" (daddr),
+#ifdef __MICROBLAZEEL__
+	"d" ((len + proto) << 8)
+#else
+	"d" (len + proto)
+#endif
+);
 	return sum;
 }
 

arch/microblaze/include/asm/cpuinfo.h

@@ -38,6 +38,7 @@ struct cpuinfo {
 	u32 use_exc;
 	u32 ver_code;
 	u32 mmu;
+	u32 endian;
 
 	/* CPU caches */
 	u32 use_icache;
@@ -76,7 +77,6 @@ struct cpuinfo {
 	u32 num_rd_brk;
 	u32 num_wr_brk;
 	u32 cpu_clock_freq; /* store real freq of cpu */
-	u32 freq_div_hz; /* store freq/HZ */
 
 	/* FPGA family */
 	u32 fpga_family_code;
@@ -97,7 +97,8 @@ void set_cpuinfo_pvr_full(struct cpuinfo *ci, struct device_node *cpu);
 static inline unsigned int fcpu(struct device_node *cpu, char *n)
 {
 	int *val;
-	return (val = (int *) of_get_property(cpu, n, NULL)) ? *val : 0;
+	return (val = (int *) of_get_property(cpu, n, NULL)) ?
+							be32_to_cpup(val) : 0;
 }
 
 #endif /* _ASM_MICROBLAZE_CPUINFO_H */

arch/microblaze/include/asm/elf.h

@@ -71,7 +71,7 @@ typedef elf_fpreg_t elf_fpregset_t[ELF_NFPREG];
 
 #define ELF_ET_DYN_BASE         (0x08000000)
 
-#ifdef __LITTLE_ENDIAN__
+#ifdef __MICROBLAZEEL__
 #define ELF_DATA	ELFDATA2LSB
 #else
 #define ELF_DATA	ELFDATA2MSB

arch/microblaze/include/asm/gpio.h

@@ -38,12 +38,9 @@ static inline int gpio_cansleep(unsigned int gpio)
 	return __gpio_cansleep(gpio);
 }
 
-/*
- * Not implemented, yet.
- */
 static inline int gpio_to_irq(unsigned int gpio)
 {
-	return -ENOSYS;
+	return __gpio_to_irq(gpio);
 }
 
 static inline int irq_to_gpio(unsigned int irq)

arch/microblaze/include/asm/io.h

@@ -243,6 +243,8 @@ static inline void __iomem *__ioremap(phys_addr_t address, unsigned long size,
 #define out_8(a, v) __raw_writeb((v), (a))
 #define in_8(a) __raw_readb(a)
 
+#define mmiowb()
+
 #define ioport_map(port, nr)	((void __iomem *)(port))
 #define ioport_unmap(addr)
 

arch/microblaze/include/asm/page.h

@@ -205,9 +205,6 @@ extern int page_is_ram(unsigned long pfn);
 #define TOPHYS(addr)  __virt_to_phys(addr)
 
 #ifdef CONFIG_MMU
-#ifdef CONFIG_CONTIGUOUS_PAGE_ALLOC
-#define WANT_PAGE_VIRTUAL 1 /* page alloc 2 relies on this */
-#endif
 
 #define VM_DATA_DEFAULT_FLAGS	(VM_READ | VM_WRITE | VM_EXEC | \
 				 VM_MAYREAD | VM_MAYWRITE | VM_MAYEXEC)

arch/microblaze/include/asm/pci.h

@@ -165,5 +165,7 @@ extern void __init xilinx_pci_init(void);
 static inline void __init xilinx_pci_init(void) { return; }
 #endif
 
+#include <asm-generic/pci-dma-compat.h>
+
 #endif	/* __KERNEL__ */
 #endif /* __ASM_MICROBLAZE_PCI_H */

arch/microblaze/include/asm/pgalloc.h

@@ -165,7 +165,8 @@ extern inline void pte_free(struct mm_struct *mm, struct page *ptepage)
 
 #define __pte_free_tlb(tlb, pte, addr)	pte_free((tlb)->mm, (pte))
 
-#define pmd_populate(mm, pmd, pte)	(pmd_val(*(pmd)) = page_address(pte))
+#define pmd_populate(mm, pmd, pte) \
+			(pmd_val(*(pmd)) = (unsigned long)page_address(pte))
 
 #define pmd_populate_kernel(mm, pmd, pte) \
 		(pmd_val(*(pmd)) = (unsigned long) (pte))

arch/microblaze/include/asm/pgtable.h

@@ -57,6 +57,13 @@ static inline int pte_file(pte_t pte) { return 0; }
 
 #define pgprot_noncached_wc(prot)	prot
 
+/*
+ * All 32bit addresses are effectively valid for vmalloc...
+ * Sort of meaningless for non-VM targets.
+ */
+#define	VMALLOC_START	0
+#define	VMALLOC_END	0xffffffff
+
 #else /* CONFIG_MMU */
 
 #include <asm-generic/4level-fixup.h>

arch/microblaze/include/asm/prom.h

@@ -27,6 +27,7 @@
 
 /* Other Prototypes */
 extern int early_uartlite_console(void);
+extern int early_uart16550_console(void);
 
 #ifdef CONFIG_PCI
 /*

arch/microblaze/include/asm/pvr.h

@@ -30,7 +30,9 @@ struct pvr_s {
 #define PVR0_USE_EXC_MASK		0x04000000
 #define PVR0_USE_ICACHE_MASK		0x02000000
 #define PVR0_USE_DCACHE_MASK		0x01000000
-#define PVR0_USE_MMU			0x00800000	/* new */
+#define PVR0_USE_MMU			0x00800000
+#define PVR0_USE_BTC			0x00400000
+#define PVR0_ENDI			0x00200000
 #define PVR0_VERSION_MASK		0x0000FF00
 #define PVR0_USER1_MASK			0x000000FF
 
@@ -38,9 +40,9 @@ struct pvr_s {
 #define PVR1_USER2_MASK			0xFFFFFFFF
 
 /* Configuration PVR masks */
-#define PVR2_D_OPB_MASK			0x80000000
+#define PVR2_D_OPB_MASK			0x80000000 /* or AXI */
 #define PVR2_D_LMB_MASK			0x40000000
-#define PVR2_I_OPB_MASK			0x20000000
+#define PVR2_I_OPB_MASK			0x20000000 /* or AXI */
 #define PVR2_I_LMB_MASK			0x10000000
 #define PVR2_INTERRUPT_IS_EDGE_MASK	0x08000000
 #define PVR2_EDGE_IS_POSITIVE_MASK	0x04000000
@@ -63,8 +65,8 @@ struct pvr_s {
 #define PVR2_OPCODE_0x0_ILL_MASK	0x00000040
 #define PVR2_UNALIGNED_EXC_MASK		0x00000020
 #define PVR2_ILL_OPCODE_EXC_MASK	0x00000010
-#define PVR2_IOPB_BUS_EXC_MASK		0x00000008
-#define PVR2_DOPB_BUS_EXC_MASK		0x00000004
+#define PVR2_IOPB_BUS_EXC_MASK		0x00000008 /* or AXI */
+#define PVR2_DOPB_BUS_EXC_MASK		0x00000004 /* or AXI */
 #define PVR2_DIV_ZERO_EXC_MASK		0x00000002
 #define PVR2_FPU_EXC_MASK		0x00000001
 
@@ -208,6 +210,8 @@ struct pvr_s {
 #define PVR_MMU_TLB_ACCESS(pvr)	(pvr.pvr[11] & PVR11_MMU_TLB_ACCESS)
 #define PVR_MMU_ZONES(pvr)	(pvr.pvr[11] & PVR11_MMU_ZONES)
 
+/* endian */
+#define PVR_ENDIAN(pvr)	(pvr.pvr[0] & PVR0_ENDI)
 
 int cpu_has_pvr(void);
 void get_pvr(struct pvr_s *pvr);

arch/microblaze/include/asm/seccomp.h

@@ -0,0 +1,16 @@
+#ifndef _ASM_MICROBLAZE_SECCOMP_H
+#define _ASM_MICROBLAZE_SECCOMP_H
+
+#include <linux/unistd.h>
+
+#define __NR_seccomp_read		__NR_read
+#define __NR_seccomp_write		__NR_write
+#define __NR_seccomp_exit		__NR_exit
+#define __NR_seccomp_sigreturn		__NR_sigreturn
+
+#define __NR_seccomp_read_32		__NR_read
+#define __NR_seccomp_write_32		__NR_write
+#define __NR_seccomp_exit_32		__NR_exit
+#define __NR_seccomp_sigreturn_32	__NR_sigreturn
+
+#endif	/* _ASM_MICROBLAZE_SECCOMP_H */

arch/microblaze/include/asm/setup.h

@@ -25,6 +25,12 @@ void early_printk(const char *fmt, ...);
 int setup_early_printk(char *opt);
 void disable_early_printk(void);
 
+#if defined(CONFIG_EARLY_PRINTK)
+#define eprintk early_printk
+#else
+#define eprintk printk
+#endif
+
 void heartbeat(void);
 void setup_heartbeat(void);
 

arch/microblaze/include/asm/thread_info.h

@@ -127,23 +127,19 @@ static inline struct thread_info *current_thread_info(void)
 #define TIF_SECCOMP		10      /* secure computing */
 #define TIF_FREEZE		14	/* Freezing for suspend */
 
-/* FIXME change in entry.S */
-#define TIF_KERNEL_TRACE	8	/* kernel trace active */
-
 /* true if poll_idle() is polling TIF_NEED_RESCHED */
 #define TIF_POLLING_NRFLAG	16
 
-#define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
-#define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
-#define _TIF_SIGPENDING		(1<<TIF_SIGPENDING)
-#define _TIF_NEED_RESCHED	(1<<TIF_NEED_RESCHED)
-#define _TIF_SINGLESTEP		(1<<TIF_SINGLESTEP)
-#define _TIF_IRET		(1<<TIF_IRET)
-#define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
-#define _TIF_FREEZE		(1<<TIF_FREEZE)
+#define _TIF_SYSCALL_TRACE	(1 << TIF_SYSCALL_TRACE)
+#define _TIF_NOTIFY_RESUME	(1 << TIF_NOTIFY_RESUME)
+#define _TIF_SIGPENDING		(1 << TIF_SIGPENDING)
+#define _TIF_NEED_RESCHED	(1 << TIF_NEED_RESCHED)
+#define _TIF_SINGLESTEP		(1 << TIF_SINGLESTEP)
+#define _TIF_IRET		(1 << TIF_IRET)
+#define _TIF_POLLING_NRFLAG	(1 << TIF_POLLING_NRFLAG)
+#define _TIF_FREEZE		(1 << TIF_FREEZE)
 #define _TIF_SYSCALL_AUDIT	(1 << TIF_SYSCALL_AUDIT)
 #define _TIF_SECCOMP		(1 << TIF_SECCOMP)
-#define _TIF_KERNEL_TRACE	(1 << TIF_KERNEL_TRACE)
 
 /* work to do in syscall trace */
 #define _TIF_WORK_SYSCALL_MASK  (_TIF_SYSCALL_TRACE | _TIF_SINGLESTEP | \

arch/microblaze/include/asm/unaligned.h

@@ -12,12 +12,18 @@
 
 # ifdef __KERNEL__
 
-# include <linux/unaligned/be_struct.h>
+# include <linux/unaligned/be_byteshift.h>
 # include <linux/unaligned/le_byteshift.h>
 # include <linux/unaligned/generic.h>
 
-# define get_unaligned	__get_unaligned_be
-# define put_unaligned	__put_unaligned_be
+
+#  ifdef __MICROBLAZEEL__
+#   define get_unaligned	__get_unaligned_le
+#   define put_unaligned	__put_unaligned_le
+#  else
+#   define get_unaligned	__get_unaligned_be
+#   define put_unaligned	__put_unaligned_be
+#  endif
 
 # endif	/* __KERNEL__ */
 #endif /* _ASM_MICROBLAZE_UNALIGNED_H */

arch/microblaze/include/asm/unistd.h

@@ -383,8 +383,11 @@
 #define __NR_rt_tgsigqueueinfo	365 /* new */
 #define __NR_perf_event_open	366 /* new */
 #define __NR_recvmmsg		367 /* new */
+#define __NR_fanotify_init	368
+#define __NR_fanotify_mark	369
+#define __NR_prlimit64		370
 
-#define __NR_syscalls		368
+#define __NR_syscalls		371
 
 #ifdef __KERNEL__
 #ifndef __ASSEMBLY__

arch/microblaze/kernel/cpu/cpuinfo-pvr-full.c

@@ -72,6 +72,7 @@ void set_cpuinfo_pvr_full(struct cpuinfo *ci, struct device_node *cpu)
 	CI(pvr_user2, USER2);
 
 	CI(mmu, USE_MMU);
+	CI(endian, ENDIAN);
 
 	CI(use_icache, USE_ICACHE);
 	CI(icache_tagbits, ICACHE_ADDR_TAG_BITS);

arch/microblaze/kernel/cpu/cpuinfo-static.c

@@ -119,6 +119,7 @@ void __init set_cpuinfo_static(struct cpuinfo *ci, struct device_node *cpu)
 	ci->pvr_user2 = fcpu(cpu, "xlnx,pvr-user2");
 
 	ci->mmu = fcpu(cpu, "xlnx,use-mmu");
+	ci->endian = fcpu(cpu, "xlnx,endianness");
 
 	ci->ver_code = 0;
 	ci->fpga_family_code = 0;

arch/microblaze/kernel/cpu/cpuinfo.c

@@ -30,6 +30,8 @@ const struct cpu_ver_key cpu_ver_lookup[] = {
 	{"7.20.c", 0x0e},
 	{"7.20.d", 0x0f},
 	{"7.30.a", 0x10},
+	{"7.30.b", 0x11},
+	{"8.00.a", 0x12},
 	{NULL, 0},
 };
 

arch/microblaze/kernel/cpu/mb.c

@@ -51,11 +51,12 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	count = seq_printf(m,
 			"CPU-Family:	MicroBlaze\n"
 			"FPGA-Arch:	%s\n"
-			"CPU-Ver:	%s\n"
+			"CPU-Ver:	%s, %s endian\n"
 			"CPU-MHz:	%d.%02d\n"
 			"BogoMips:	%lu.%02lu\n",
 			fpga_family,
 			cpu_ver,
+			cpuinfo.endian ? "little" : "big",
 			cpuinfo.cpu_clock_freq /
 			1000000,
 			cpuinfo.cpu_clock_freq %

arch/microblaze/kernel/cpu/pvr.c

@@ -27,7 +27,7 @@
 	register unsigned tmp __asm__("r3");			\
 	tmp = 0x0;	/* Prevent warning about unused */	\
 	__asm__ __volatile__ (					\
-			".byte 0x94,0x60,0xa0, " #pvrid "\n\t"	\
+			"mfs	%0, rpvr" #pvrid ";"	\
 			: "=r" (tmp) : : "memory"); 		\
 	val = tmp;						\
 }

arch/microblaze/kernel/early_printk.c

@@ -24,7 +24,8 @@
 static u32 early_console_initialized;
 static u32 base_addr;
 
-static void early_printk_putc(char c)
+#ifdef CONFIG_SERIAL_UARTLITE_CONSOLE
+static void early_printk_uartlite_putc(char c)
 {
 	/*
 	 * Limit how many times we'll spin waiting for TX FIFO status.
@@ -45,25 +46,70 @@ static void early_printk_putc(char c)
 		out_be32(base_addr + 4, c & 0xff);
 }
 
-static void early_printk_write(struct console *unused,
+static void early_printk_uartlite_write(struct console *unused,
 					const char *s, unsigned n)
 {
 	while (*s && n-- > 0) {
-		early_printk_putc(*s);
+		early_printk_uartlite_putc(*s);
 		if (*s == '\n')
-			early_printk_putc('\r');
+			early_printk_uartlite_putc('\r');
 		s++;
 	}
 }
 
-static struct console early_serial_console = {
+static struct console early_serial_uartlite_console = {
 	.name = "earlyser",
-	.write = early_printk_write,
+	.write = early_printk_uartlite_write,
 	.flags = CON_PRINTBUFFER,
 	.index = -1,
 };
+#endif /* CONFIG_SERIAL_UARTLITE_CONSOLE */
 
-static struct console *early_console = &early_serial_console;
+#ifdef CONFIG_SERIAL_8250_CONSOLE
+static void early_printk_uart16550_putc(char c)
+{
+	/*
+	 * Limit how many times we'll spin waiting for TX FIFO status.
+	 * This will prevent lockups if the base address is incorrectly
+	 * set, or any other issue on the UARTLITE.
+	 * This limit is pretty arbitrary, unless we are at about 10 baud
+	 * we'll never timeout on a working UART.
+	 */
+
+	#define UART_LSR_TEMT	0x40 /* Transmitter empty */
+	#define UART_LSR_THRE	0x20 /* Transmit-hold-register empty */
+	#define BOTH_EMPTY (UART_LSR_TEMT | UART_LSR_THRE)
+
+	unsigned retries = 10000;
+
+	while (--retries &&
+		!((in_be32(base_addr + 0x14) & BOTH_EMPTY) == BOTH_EMPTY))
+		;
+
+	if (retries)
+		out_be32(base_addr, c & 0xff);
+}
+
+static void early_printk_uart16550_write(struct console *unused,
+					const char *s, unsigned n)
+{
+	while (*s && n-- > 0) {
+		early_printk_uart16550_putc(*s);
+		if (*s == '\n')
+			early_printk_uart16550_putc('\r');
+		s++;
+	}
+}
+
+static struct console early_serial_uart16550_console = {
+	.name = "earlyser",
+	.write = early_printk_uart16550_write,
+	.flags = CON_PRINTBUFFER,
+	.index = -1,
+};
+#endif /* CONFIG_SERIAL_8250_CONSOLE */
+
+static struct console *early_console;
 
 void early_printk(const char *fmt, ...)
 {
@@ -84,20 +130,43 @@ int __init setup_early_printk(char *opt)
 	if (early_console_initialized)
 		return 1;
 
+#ifdef CONFIG_SERIAL_UARTLITE_CONSOLE
 	base_addr = early_uartlite_console();
 	if (base_addr) {
 		early_console_initialized = 1;
 #ifdef CONFIG_MMU
 		early_console_reg_tlb_alloc(base_addr);
 #endif
+		early_console = &early_serial_uartlite_console;
 		early_printk("early_printk_console is enabled at 0x%08x\n",
 							base_addr);
 
 		/* register_console(early_console); */
 
 		return 0;
-	} else
-		return 1;
+	}
+#endif /* CONFIG_SERIAL_UARTLITE_CONSOLE */
+
+#ifdef CONFIG_SERIAL_8250_CONSOLE
+	base_addr = early_uart16550_console();
+	base_addr &= ~3; /* clear register offset */
+	if (base_addr) {
+		early_console_initialized = 1;
+#ifdef CONFIG_MMU
+		early_console_reg_tlb_alloc(base_addr);
+#endif
+		early_console = &early_serial_uart16550_console;
+
+		early_printk("early_printk_console is enabled at 0x%08x\n",
+							base_addr);
+
+		/* register_console(early_console); */
+
+		return 0;
+	}
+#endif /* CONFIG_SERIAL_8250_CONSOLE */
+
+	return 1;
 }
 
 void __init disable_early_printk(void)

arch/microblaze/kernel/entry.S

@@ -186,6 +186,8 @@
 	swi	r13, r1, PTO+PT_R13;	/* Save SDA2 */			\
 	swi	r14, r1, PTO+PT_PC;	/* PC, before IRQ/trap */	\
 	swi	r15, r1, PTO+PT_R15;	/* Save LP */			\
+	swi	r16, r1, PTO+PT_R16;					\
+	swi	r17, r1, PTO+PT_R17;					\
 	swi	r18, r1, PTO+PT_R18;	/* Save asm scratch reg */	\
 	swi	r19, r1, PTO+PT_R19;					\
 	swi	r20, r1, PTO+PT_R20;					\
@@ -220,6 +222,8 @@
 	lwi	r13, r1, PTO+PT_R13;	/* restore SDA2 */		\
 	lwi	r14, r1, PTO+PT_PC;	/* RESTORE_LINK PC, before IRQ/trap */\
 	lwi	r15, r1, PTO+PT_R15;	/* restore LP */		\
+	lwi	r16, r1, PTO+PT_R16;					\
+	lwi	r17, r1, PTO+PT_R17;					\
 	lwi	r18, r1, PTO+PT_R18;	/* restore asm scratch reg */	\
 	lwi	r19, r1, PTO+PT_R19;					\
 	lwi	r20, r1, PTO+PT_R20;					\
@@ -295,6 +299,8 @@ C_ENTRY(_user_exception):
 	/* addik	r1, r1, -STATE_SAVE_SIZE; */
 	addik r1, r1, THREAD_SIZE + CONFIG_KERNEL_BASE_ADDR - CONFIG_KERNEL_START - STATE_SAVE_SIZE;
 	SAVE_REGS
+	swi	r0, r1, PTO + PT_R3
+	swi	r0, r1, PTO + PT_R4
 
 	lwi	r11, r0, TOPHYS(PER_CPU(ENTRY_SP));
 	swi	r11, r1, PTO+PT_R1;		/* Store user SP.  */
@@ -458,14 +464,8 @@ C_ENTRY(sys_execve):
 	addik	r8, r1, PTO;		/* add user context as 4th arg */
 
 C_ENTRY(sys_rt_sigreturn_wrapper):
-	swi	r3, r1, PTO+PT_R3; /* restore saved r3, r4 registers */
-	swi	r4, r1, PTO+PT_R4;
-	brlid	r15, sys_rt_sigreturn	/* Do real work */
+	brid	sys_rt_sigreturn	/* Do real work */
 	addik	r5, r1, PTO;		/* add user context as 1st arg */
-	lwi	r3, r1, PTO+PT_R3; /* restore saved r3, r4 registers */
-	lwi	r4, r1, PTO+PT_R4;
-	bri ret_from_trap /* fall through will not work here due to align */
-	nop;
 
 /*
  * HW EXCEPTION rutine start
@@ -765,9 +765,7 @@ C_ENTRY(_debug_exception):
 	/* save all regs to pt_reg structure */
 	swi	r0, r1, PTO+PT_R0;	/* R0 must be saved too */
 	swi	r14, r1, PTO+PT_R14	/* rewrite saved R14 value */
-	swi	r16, r1, PTO+PT_R16
 	swi	r16, r1, PTO+PT_PC; /* PC and r16 are the same */
-	swi	r17, r1, PTO+PT_R17
 	/* save special purpose registers to pt_regs */
 	mfs	r11, rear;
 	swi	r11, r1, PTO+PT_EAR;
@@ -801,8 +799,6 @@ C_ENTRY(_debug_exception):
 
 	addik	r1, r1, -STATE_SAVE_SIZE; /* Make room on the stack.  */
 	SAVE_REGS;
-	swi	r17, r1, PTO+PT_R17;
-	swi	r16, r1, PTO+PT_R16;
 	swi	r16, r1, PTO+PT_PC;	/* Save LP */
 	swi	r0, r1, PTO + PT_MODE; /* Was in user-mode.  */
 	lwi	r11, r0, TOPHYS(PER_CPU(ENTRY_SP));
@@ -848,8 +844,6 @@ dbtrap_call: /* Return point for kernel/user entry + 8 because of rtsd r15, 8 */
 	tophys(r1,r1);
 	/* MS: Restore all regs */
 	RESTORE_REGS
-	lwi	r17, r1, PTO+PT_R17;
-	lwi	r16, r1, PTO+PT_R16;
 	addik	r1, r1, STATE_SAVE_SIZE	 /* Clean up stack space */
 	lwi	r1, r1, PT_R1 - PT_SIZE; /* Restore user stack pointer */
 DBTRAP_return_user: /* MS: Make global symbol for debugging */
@@ -863,7 +857,6 @@ DBTRAP_return_user: /* MS: Make global symbol for debugging */
 	RESTORE_REGS
 	lwi	r14, r1, PTO+PT_R14;
 	lwi	r16, r1, PTO+PT_PC;
-	lwi	r17, r1, PTO+PT_R17;
 	addik	r1, r1, STATE_SAVE_SIZE; /* MS: Clean up stack space */
 	tovirt(r1,r1);
 DBTRAP_return_kernel: /* MS: Make global symbol for debugging */

arch/microblaze/kernel/exceptions.c

@@ -72,7 +72,6 @@ asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
 							int fsr, int addr)
 {
 #ifdef CONFIG_MMU
-	int code;
 	addr = regs->pc;
 #endif
 
@@ -86,8 +85,7 @@ asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
 	switch (type & 0x1F) {
 	case MICROBLAZE_ILL_OPCODE_EXCEPTION:
 		if (user_mode(regs)) {
-			pr_debug(KERN_WARNING "Illegal opcode exception " \
-							"in user mode.\n");
+			pr_debug("Illegal opcode exception in user mode\n");
 			_exception(SIGILL, regs, ILL_ILLOPC, addr);
 			return;
 		}
@@ -97,8 +95,7 @@ asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
 		break;
 	case MICROBLAZE_IBUS_EXCEPTION:
 		if (user_mode(regs)) {
-			pr_debug(KERN_WARNING "Instruction bus error " \
-						"exception in user mode.\n");
+			pr_debug("Instruction bus error exception in user mode\n");
 			_exception(SIGBUS, regs, BUS_ADRERR, addr);
 			return;
 		}
@@ -108,8 +105,7 @@ asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
 		break;
 	case MICROBLAZE_DBUS_EXCEPTION:
 		if (user_mode(regs)) {
-			pr_debug(KERN_WARNING "Data bus error exception " \
-							"in user mode.\n");
+			pr_debug("Data bus error exception in user mode\n");
 			_exception(SIGBUS, regs, BUS_ADRERR, addr);
 			return;
 		}
@@ -119,8 +115,7 @@ asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
 		break;
 	case MICROBLAZE_DIV_ZERO_EXCEPTION:
 		if (user_mode(regs)) {
-			pr_debug(KERN_WARNING "Divide by zero exception " \
-							"in user mode\n");
+			pr_debug("Divide by zero exception in user mode\n");
 			_exception(SIGILL, regs, FPE_INTDIV, addr);
 			return;
 		}
@@ -129,7 +124,7 @@ asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
 		die("Divide by zero exception", regs, SIGBUS);
 		break;
 	case MICROBLAZE_FPU_EXCEPTION:
-		pr_debug(KERN_WARNING "FPU exception\n");
+		pr_debug("FPU exception\n");
 		/* IEEE FP exception */
 		/* I removed fsr variable and use code var for storing fsr */
 		if (fsr & FSR_IO)
@@ -147,14 +142,8 @@ asmlinkage void full_exception(struct pt_regs *regs, unsigned int type,
 
 #ifdef CONFIG_MMU
 	case MICROBLAZE_PRIVILEGED_EXCEPTION:
-		pr_debug(KERN_WARNING "Privileged exception\n");
-		/* "brk r0,r0" - used as debug breakpoint - old toolchain */
-		if (get_user(code, (unsigned long *)regs->pc) == 0
-			&& code == 0x980c0000) {
-			_exception(SIGTRAP, regs, TRAP_BRKPT, addr);
-		} else {
-			_exception(SIGILL, regs, ILL_PRVOPC, addr);
-		}
+		pr_debug("Privileged exception\n");
+		_exception(SIGILL, regs, ILL_PRVOPC, addr);
 		break;
 #endif
 	default:

arch/microblaze/kernel/heartbeat.c

@@ -47,11 +47,10 @@ void setup_heartbeat(void)
 	struct device_node *gpio = NULL;
 	int *prop;
 	int j;
-	char *gpio_list[] = {
-				"xlnx,xps-gpio-1.00.a",
-				"xlnx,opb-gpio-1.00.a",
-				NULL
-			};
+	const char * const gpio_list[] = {
+		"xlnx,xps-gpio-1.00.a",
+		NULL
+	};
 
 	for (j = 0; gpio_list[j] != NULL; j++) {
 		gpio = of_find_compatible_node(NULL, NULL, gpio_list[j]);
@@ -60,7 +59,7 @@ void setup_heartbeat(void)
 	}
 
 	if (gpio) {
-		base_addr = *(int *) of_get_property(gpio, "reg", NULL);
+		base_addr = be32_to_cpup(of_get_property(gpio, "reg", NULL));
 		base_addr = (unsigned long) ioremap(base_addr, PAGE_SIZE);
 		printk(KERN_NOTICE "Heartbeat GPIO at 0x%x\n", base_addr);
 

arch/microblaze/kernel/intc.c

@@ -126,11 +126,8 @@ void __init init_IRQ(void)
 				0
 			};
 #endif
-	static char *intc_list[] = {
+	const char * const intc_list[] = {
 				"xlnx,xps-intc-1.00.a",
-				"xlnx,opb-intc-1.00.c",
-				"xlnx,opb-intc-1.00.b",
-				"xlnx,opb-intc-1.00.a",
 				NULL
 			};
 
@@ -141,12 +138,15 @@ void __init init_IRQ(void)
 	}
 	BUG_ON(!intc);
 
-	intc_baseaddr = *(int *) of_get_property(intc, "reg", NULL);
+	intc_baseaddr = be32_to_cpup(of_get_property(intc,
+								"reg", NULL));
 	intc_baseaddr = (unsigned long) ioremap(intc_baseaddr, PAGE_SIZE);
-	nr_irq = *(int *) of_get_property(intc, "xlnx,num-intr-inputs", NULL);
+	nr_irq = be32_to_cpup(of_get_property(intc,
+						"xlnx,num-intr-inputs", NULL));
 
 	intr_type =
-		*(int *) of_get_property(intc, "xlnx,kind-of-intr", NULL);
+		be32_to_cpup(of_get_property(intc,
+						"xlnx,kind-of-intr", NULL));
 	if (intr_type >= (1 << (nr_irq + 1)))
 		printk(KERN_INFO " ERROR: Mismatch in kind-of-intr param\n");
 

arch/microblaze/kernel/kgdb.c

@@ -80,7 +80,7 @@ void gdb_regs_to_pt_regs(unsigned long *gdb_regs, struct pt_regs *regs)
 void microblaze_kgdb_break(struct pt_regs *regs)
 {
 	if (kgdb_handle_exception(1, SIGTRAP, 0, regs) != 0)
-		return 0;
+		return;
 
 	/* Jump over the first arch_kgdb_breakpoint which is barrier to
 	 * get kgdb work. The same solution is used for powerpc */
@@ -114,7 +114,6 @@ int kgdb_arch_handle_exception(int vector, int signo, int err_code,
 {
 	char *ptr;
 	unsigned long address;
-	int cpu = smp_processor_id();
 
 	switch (remcom_in_buffer[0]) {
 	case 'c':
@@ -143,5 +142,9 @@ void kgdb_arch_exit(void)
  * Global data
  */
 struct kgdb_arch arch_kgdb_ops = {
+#ifdef __MICROBLAZEEL__
+	.gdb_bpt_instr = {0x18, 0x00, 0x0c, 0xba}, /* brki r16, 0x18 */
+#else
 	.gdb_bpt_instr = {0xba, 0x0c, 0x00, 0x18}, /* brki r16, 0x18 */
+#endif
 };

arch/microblaze/kernel/microblaze_ksyms.c

@@ -15,37 +15,13 @@
 #include <linux/syscalls.h>
 
 #include <asm/checksum.h>
+#include <asm/cacheflush.h>
 #include <linux/io.h>
 #include <asm/page.h>
 #include <asm/system.h>
 #include <linux/ftrace.h>
 #include <linux/uaccess.h>
 
-/*
- * libgcc functions - functions that are used internally by the
- * compiler... (prototypes are not correct though, but that
- * doesn't really matter since they're not versioned).
- */
-extern void __ashldi3(void);
-EXPORT_SYMBOL(__ashldi3);
-extern void __ashrdi3(void);
-EXPORT_SYMBOL(__ashrdi3);
-extern void __divsi3(void);
-EXPORT_SYMBOL(__divsi3);
-extern void __lshrdi3(void);
-EXPORT_SYMBOL(__lshrdi3);
-extern void __modsi3(void);
-EXPORT_SYMBOL(__modsi3);
-extern void __mulsi3(void);
-EXPORT_SYMBOL(__mulsi3);
-extern void __muldi3(void);
-EXPORT_SYMBOL(__muldi3);
-extern void __ucmpdi2(void);
-EXPORT_SYMBOL(__ucmpdi2);
-extern void __udivsi3(void);
-EXPORT_SYMBOL(__udivsi3);
-extern void __umodsi3(void);
-EXPORT_SYMBOL(__umodsi3);
 extern char *_ebss;
 EXPORT_SYMBOL_GPL(_ebss);
 #ifdef CONFIG_FUNCTION_TRACER
@@ -63,3 +39,9 @@ EXPORT_SYMBOL(__strncpy_user);
 EXPORT_SYMBOL(memcpy);
 EXPORT_SYMBOL(memmove);
 #endif
+
+#ifdef CONFIG_MMU
+EXPORT_SYMBOL(empty_zero_page);
+#endif
+
+EXPORT_SYMBOL(mbc);

arch/microblaze/kernel/prom.c

@@ -72,11 +72,12 @@ static int __init early_init_dt_scan_serial(unsigned long node,
 /* find compatible node with uartlite */
 	p = of_get_flat_dt_prop(node, "compatible", &l);
 	if ((strncmp(p, "xlnx,xps-uartlite", 17) != 0) &&
-			(strncmp(p, "xlnx,opb-uartlite", 17) != 0))
+			(strncmp(p, "xlnx,opb-uartlite", 17) != 0) &&
+			(strncmp(p, "xlnx,axi-uartlite", 17) != 0))
 		return 0;
 
 	addr = of_get_flat_dt_prop(node, "reg", &l);
-	return *addr; /* return address */
+	return be32_to_cpup(addr); /* return address */
 }
 
 /* this function is looking for early uartlite console - Microblaze specific */
@@ -84,6 +85,40 @@ int __init early_uartlite_console(void)
 {
 	return of_scan_flat_dt(early_init_dt_scan_serial, NULL);
 }
+
+/* MS this is Microblaze specifig function */
+static int __init early_init_dt_scan_serial_full(unsigned long node,
+				const char *uname, int depth, void *data)
+{
+	unsigned long l;
+	char *p;
+	unsigned int addr;
+
+	pr_debug("search \"chosen\", depth: %d, uname: %s\n", depth, uname);
+
+/* find all serial nodes */
+	if (strncmp(uname, "serial", 6) != 0)
+		return 0;
+
+	early_init_dt_check_for_initrd(node);
+
+/* find compatible node with uartlite */
+	p = of_get_flat_dt_prop(node, "compatible", &l);
+
+	if ((strncmp(p, "xlnx,xps-uart16550", 18) != 0) &&
+		(strncmp(p, "xlnx,axi-uart16550", 18) != 0))
+		return 0;
+
+	addr = *(u32 *)of_get_flat_dt_prop(node, "reg", &l);
+	addr += *(u32 *)of_get_flat_dt_prop(node, "reg-offset", &l);
+	return be32_to_cpu(addr); /* return address */
+}
+
+/* this function is looking for early uartlite console - Microblaze specific */
+int __init early_uart16550_console(void)
+{
+	return of_scan_flat_dt(early_init_dt_scan_serial_full, NULL);
+}
 #endif
 
 void __init early_init_devtree(void *params)

arch/microblaze/kernel/setup.c

@@ -92,12 +92,6 @@ inline unsigned get_romfs_len(unsigned *addr)
 }
 #endif	/* CONFIG_MTD_UCLINUX_EBSS */
 
-#if defined(CONFIG_EARLY_PRINTK) && defined(CONFIG_SERIAL_UARTLITE_CONSOLE)
-#define eprintk early_printk
-#else
-#define eprintk printk
-#endif
-
 void __init machine_early_init(const char *cmdline, unsigned int ram,
 		unsigned int fdt, unsigned int msr)
 {

arch/microblaze/kernel/syscall_table.S

@@ -372,3 +372,6 @@ ENTRY(sys_call_table)
 	.long sys_rt_tgsigqueueinfo	/* 365 */
 	.long sys_perf_event_open
 	.long sys_recvmmsg
+	.long sys_fanotify_init
+	.long sys_fanotify_mark
+	.long sys_prlimit64	/* 370 */

arch/microblaze/kernel/timer.c

@@ -38,6 +38,9 @@ static unsigned int timer_baseaddr;
 #define TIMER_BASE	timer_baseaddr
 #endif
 
+unsigned int freq_div_hz;
+unsigned int timer_clock_freq;
+
 #define TCSR0	(0x00)
 #define TLR0	(0x04)
 #define TCR0	(0x08)
@@ -115,7 +118,7 @@ static void microblaze_timer_set_mode(enum clock_event_mode mode,
 	switch (mode) {
 	case CLOCK_EVT_MODE_PERIODIC:
 		printk(KERN_INFO "%s: periodic\n", __func__);
-		microblaze_timer0_start_periodic(cpuinfo.freq_div_hz);
+		microblaze_timer0_start_periodic(freq_div_hz);
 		break;
 	case CLOCK_EVT_MODE_ONESHOT:
 		printk(KERN_INFO "%s: oneshot\n", __func__);
@@ -168,7 +171,7 @@ static struct irqaction timer_irqaction = {
 static __init void microblaze_clockevent_init(void)
 {
 	clockevent_microblaze_timer.mult =
-		div_sc(cpuinfo.cpu_clock_freq, NSEC_PER_SEC,
+		div_sc(timer_clock_freq, NSEC_PER_SEC,
 				clockevent_microblaze_timer.shift);
 	clockevent_microblaze_timer.max_delta_ns =
 		clockevent_delta2ns((u32)~0, &clockevent_microblaze_timer);
@@ -201,7 +204,7 @@ static struct cyclecounter microblaze_cc = {
 
 int __init init_microblaze_timecounter(void)
 {
-	microblaze_cc.mult = div_sc(cpuinfo.cpu_clock_freq, NSEC_PER_SEC,
+	microblaze_cc.mult = div_sc(timer_clock_freq, NSEC_PER_SEC,
 				microblaze_cc.shift);
 
 	timecounter_init(&microblaze_tc, &microblaze_cc, sched_clock());
@@ -221,7 +224,7 @@ static struct clocksource clocksource_microblaze = {
 static int __init microblaze_clocksource_init(void)
 {
 	clocksource_microblaze.mult =
-			clocksource_hz2mult(cpuinfo.cpu_clock_freq,
+			clocksource_hz2mult(timer_clock_freq,
 						clocksource_microblaze.shift);
 	if (clocksource_register(&clocksource_microblaze))
 		panic("failed to register clocksource");
@@ -247,6 +250,7 @@ void __init time_init(void)
 	u32 irq, i = 0;
 	u32 timer_num = 1;
 	struct device_node *timer = NULL;
+	const void *prop;
 #ifdef CONFIG_SELFMOD_TIMER
 	unsigned int timer_baseaddr = 0;
 	int arr_func[] = {
@@ -258,12 +262,10 @@ void __init time_init(void)
 				0
 			};
 #endif
-	char *timer_list[] = {
-				"xlnx,xps-timer-1.00.a",
-				"xlnx,opb-timer-1.00.b",
-				"xlnx,opb-timer-1.00.a",
-				NULL
-			};
+	const char * const timer_list[] = {
+		"xlnx,xps-timer-1.00.a",
+		NULL
+	};
 
 	for (i = 0; timer_list[i] != NULL; i++) {
 		timer = of_find_compatible_node(NULL, NULL, timer_list[i]);
@@ -272,13 +274,13 @@ void __init time_init(void)
 	}
 	BUG_ON(!timer);
 
-	timer_baseaddr = *(int *) of_get_property(timer, "reg", NULL);
+	timer_baseaddr = be32_to_cpup(of_get_property(timer, "reg", NULL));
 	timer_baseaddr = (unsigned long) ioremap(timer_baseaddr, PAGE_SIZE);
-	irq = *(int *) of_get_property(timer, "interrupts", NULL);
-	timer_num =
-		*(int *) of_get_property(timer, "xlnx,one-timer-only", NULL);
+	irq = be32_to_cpup(of_get_property(timer, "interrupts", NULL));
+	timer_num = be32_to_cpup(of_get_property(timer,
+						"xlnx,one-timer-only", NULL));
 	if (timer_num) {
-		printk(KERN_EMERG "Please enable two timers in HW\n");
+		eprintk(KERN_EMERG "Please enable two timers in HW\n");
 		BUG();
 	}
 
@@ -288,7 +290,14 @@ void __init time_init(void)
 	printk(KERN_INFO "%s #0 at 0x%08x, irq=%d\n",
 		timer_list[i], timer_baseaddr, irq);
 
-	cpuinfo.freq_div_hz = cpuinfo.cpu_clock_freq / HZ;
+	/* If there is clock-frequency property than use it */
+	prop = of_get_property(timer, "clock-frequency", NULL);
+	if (prop)
+		timer_clock_freq = be32_to_cpup(prop);
+	else
+		timer_clock_freq = cpuinfo.cpu_clock_freq;
+
+	freq_div_hz = timer_clock_freq / HZ;
 
 	setup_irq(irq, &timer_irqaction);
 #ifdef CONFIG_HEART_BEAT

arch/microblaze/kernel/vmlinux.lds.S

@@ -8,7 +8,6 @@
  * for more details.
  */
 
-OUTPUT_FORMAT("elf32-microblaze", "elf32-microblaze", "elf32-microblaze")
 OUTPUT_ARCH(microblaze)
 ENTRY(microblaze_start)
 
@@ -16,7 +15,11 @@ ENTRY(microblaze_start)
 #include <asm-generic/vmlinux.lds.h>
 #include <asm/thread_info.h>
 
+#ifdef __MICROBLAZEEL__
+jiffies = jiffies_64;
+#else
 jiffies = jiffies_64 + 4;
+#endif
 
 SECTIONS {
 	. = CONFIG_KERNEL_START;

arch/microblaze/lib/Makefile

@@ -11,3 +11,13 @@ lib-y += memcpy.o memmove.o
 endif
 
 lib-y += uaccess_old.o
+
+lib-y += ashldi3.o
+lib-y += ashrdi3.o
+lib-y += divsi3.o
+lib-y += lshrdi3.o
+lib-y += modsi3.o
+lib-y += muldi3.o
+lib-y += mulsi3.o
+lib-y += udivsi3.o
+lib-y += umodsi3.o

arch/microblaze/lib/ashldi3.c

@@ -0,0 +1,29 @@
+#include <linux/module.h>
+
+#include "libgcc.h"
+
+long long __ashldi3(long long u, word_type b)
+{
+	DWunion uu, w;
+	word_type bm;
+
+	if (b == 0)
+		return u;
+
+	uu.ll = u;
+	bm = 32 - b;
+
+	if (bm <= 0) {
+		w.s.low = 0;
+		w.s.high = (unsigned int) uu.s.low << -bm;
+	} else {
+		const unsigned int carries = (unsigned int) uu.s.low >> bm;
+
+		w.s.low = (unsigned int) uu.s.low << b;
+		w.s.high = ((unsigned int) uu.s.high << b) | carries;
+	}
+
+	return w.ll;
+}
+
+EXPORT_SYMBOL(__ashldi3);

arch/microblaze/lib/ashrdi3.c

@@ -0,0 +1,31 @@
+#include <linux/module.h>
+
+#include "libgcc.h"
+
+long long __ashrdi3(long long u, word_type b)
+{
+	DWunion uu, w;
+	word_type bm;
+
+	if (b == 0)
+		return u;
+
+	uu.ll = u;
+	bm = 32 - b;
+
+	if (bm <= 0) {
+		/* w.s.high = 1..1 or 0..0 */
+		w.s.high =
+		    uu.s.high >> 31;
+		w.s.low = uu.s.high >> -bm;
+	} else {
+		const unsigned int carries = (unsigned int) uu.s.high << bm;
+
+		w.s.high = uu.s.high >> b;
+		w.s.low = ((unsigned int) uu.s.low >> b) | carries;
+	}
+
+	return w.ll;
+}
+
+EXPORT_SYMBOL(__ashrdi3);

arch/microblaze/lib/divsi3.S

@@ -0,0 +1,73 @@
+#include <linux/linkage.h>
+
+/*
+* Divide operation for 32 bit integers.
+*	Input :	Dividend in Reg r5
+*		Divisor in Reg r6
+*	Output: Result in Reg r3
+*/
+	.text
+	.globl	__divsi3
+	.type __divsi3, @function
+	.ent __divsi3
+__divsi3:
+	.frame	r1, 0, r15
+
+	addik	r1, r1, -16
+	swi	r28, r1, 0
+	swi	r29, r1, 4
+	swi	r30, r1, 8
+	swi	r31, r1, 12
+
+	beqi	r6, div_by_zero /* div_by_zero - division error */
+	beqi	r5, result_is_zero /* result is zero */
+	bgeid	r5, r5_pos
+	xor	r28, r5, r6 /* get the sign of the result */
+	rsubi	r5, r5, 0 /* make r5 positive */
+r5_pos:
+	bgei	r6, r6_pos
+	rsubi	r6, r6, 0 /* make r6 positive */
+r6_pos:
+	addik	r30, r0, 0 /* clear mod */
+	addik	r3, r0, 0 /* clear div */
+	addik	r29, r0, 32 /* initialize the loop count */
+
+	/* first part try to find the first '1' in the r5 */
+div0:
+	blti	r5, div2 /* this traps r5 == 0x80000000 */
+div1:
+	add	r5, r5, r5 /* left shift logical r5 */
+	bgtid	r5, div1
+	addik	r29, r29, -1
+div2:
+	/* left shift logical r5 get the '1' into the carry */
+	add	r5, r5, r5
+	addc	r30, r30, r30 /* move that bit into the mod register */
+	rsub	r31, r6, r30 /* try to subtract (r30 a r6) */
+	blti	r31, mod_too_small
+	/* move the r31 to mod since the result was positive */
+	or	r30, r0, r31
+	addik	r3, r3, 1
+mod_too_small:
+	addik	r29, r29, -1
+	beqi	r29, loop_end
+	add	r3, r3, r3 /* shift in the '1' into div */
+	bri	div2 /* div2 */
+loop_end:
+	bgei	r28, return_here
+	brid	return_here
+	rsubi	r3, r3, 0 /* negate the result */
+div_by_zero:
+result_is_zero:
+	or	r3, r0, r0 /* set result to 0 */
+return_here:
+/* restore values of csrs and that of r3 and the divisor and the dividend */
+	lwi	r28, r1, 0
+	lwi	r29, r1, 4
+	lwi	r30, r1, 8
+	lwi	r31, r1, 12
+	rtsd	r15, 8
+	addik	r1, r1, 16
+
+.size __divsi3, . - __divsi3
+.end __divsi3

arch/microblaze/lib/libgcc.h

@@ -0,0 +1,25 @@
+#ifndef __ASM_LIBGCC_H
+#define __ASM_LIBGCC_H
+
+#include <asm/byteorder.h>
+
+typedef int word_type __attribute__ ((mode (__word__)));
+
+#ifdef __BIG_ENDIAN
+struct DWstruct {
+	int high, low;
+};
+#elif defined(__LITTLE_ENDIAN)
+struct DWstruct {
+	int low, high;
+};
+#else
+#error I feel sick.
+#endif
+
+typedef union {
+	struct DWstruct s;
+	long long ll;
+} DWunion;
+
+#endif /* __ASM_LIBGCC_H */

arch/microblaze/lib/lshrdi3.c

@@ -0,0 +1,29 @@
+#include <linux/module.h>
+
+#include "libgcc.h"
+
+long long __lshrdi3(long long u, word_type b)
+{
+	DWunion uu, w;
+	word_type bm;
+
+	if (b == 0)
+		return u;
+
+	uu.ll = u;
+	bm = 32 - b;
+
+	if (bm <= 0) {
+		w.s.high = 0;
+		w.s.low = (unsigned int) uu.s.high >> -bm;
+	} else {
+		const unsigned int carries = (unsigned int) uu.s.high << bm;
+
+		w.s.high = (unsigned int) uu.s.high >> b;
+		w.s.low = ((unsigned int) uu.s.low >> b) | carries;
+	}
+
+	return w.ll;
+}
+
+EXPORT_SYMBOL(__lshrdi3);

arch/microblaze/lib/memcpy.c

@@ -33,17 +33,24 @@
 #include <asm/system.h>
 
 #ifdef __HAVE_ARCH_MEMCPY
+#ifndef CONFIG_OPT_LIB_FUNCTION
 void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
 {
 	const char *src = v_src;
 	char *dst = v_dst;
-#ifndef CONFIG_OPT_LIB_FUNCTION
+
 	/* Simple, byte oriented memcpy. */
 	while (c--)
 		*dst++ = *src++;
 
 	return v_dst;
-#else
+}
+#else /* CONFIG_OPT_LIB_FUNCTION */
+void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
+{
+	const char *src = v_src;
+	char *dst = v_dst;
+
 	/* The following code tries to optimize the copy by using unsigned
 	 * alignment. This will work fine if both source and destination are
 	 * aligned on the same boundary. However, if they are aligned on
@@ -86,7 +93,7 @@ void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
 		case 0x1:	/* Unaligned - Off by 1 */
 			/* Word align the source */
 			i_src = (const void *) ((unsigned)src & ~3);
-
+#ifndef __MICROBLAZEEL__
 			/* Load the holding buffer */
 			buf_hold = *i_src++ << 8;
 
@@ -95,7 +102,16 @@ void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
 				*i_dst++ = buf_hold | value >> 24;
 				buf_hold = value << 8;
 			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*i_src++ & 0xFFFFFF00) >>8;
 
+			for (; c >= 4; c -= 4) {
+				value = *i_src++;
+				*i_dst++ = buf_hold | ((value & 0xFF) << 24);
+				buf_hold = (value & 0xFFFFFF00) >>8;
+			}
+#endif
 			/* Realign the source */
 			src = (const void *)i_src;
 			src -= 3;
@@ -103,7 +119,7 @@ void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
 		case 0x2:	/* Unaligned - Off by 2 */
 			/* Word align the source */
 			i_src = (const void *) ((unsigned)src & ~3);
-
+#ifndef __MICROBLAZEEL__
 			/* Load the holding buffer */
 			buf_hold = *i_src++ << 16;
 
@@ -112,7 +128,16 @@ void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
 				*i_dst++ = buf_hold | value >> 16;
 				buf_hold = value << 16;
 			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*i_src++ & 0xFFFF0000 )>>16;
 
+			for (; c >= 4; c -= 4) {
+				value = *i_src++;
+				*i_dst++ = buf_hold | ((value & 0xFFFF)<<16);
+				buf_hold = (value & 0xFFFF0000) >>16;
+			}
+#endif
 			/* Realign the source */
 			src = (const void *)i_src;
 			src -= 2;
@@ -120,7 +145,7 @@ void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
 		case 0x3:	/* Unaligned - Off by 3 */
 			/* Word align the source */
 			i_src = (const void *) ((unsigned)src & ~3);
-
+#ifndef __MICROBLAZEEL__
 			/* Load the holding buffer */
 			buf_hold = *i_src++ << 24;
 
@@ -129,7 +154,16 @@ void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
 				*i_dst++ = buf_hold | value >> 8;
 				buf_hold = value << 24;
 			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*i_src++ & 0xFF000000) >> 24;
 
+			for (; c >= 4; c -= 4) {
+				value = *i_src++;
+				*i_dst++ = buf_hold | ((value & 0xFFFFFF) << 8);
+				buf_hold = (value & 0xFF000000) >> 24;
+			}
+#endif
 			/* Realign the source */
 			src = (const void *)i_src;
 			src -= 1;
@@ -150,7 +184,7 @@ void *memcpy(void *v_dst, const void *v_src, __kernel_size_t c)
 	}
 
 	return v_dst;
-#endif
 }
+#endif /* CONFIG_OPT_LIB_FUNCTION */
 EXPORT_SYMBOL(memcpy);
 #endif /* __HAVE_ARCH_MEMCPY */

arch/microblaze/lib/memmove.c

@@ -31,16 +31,12 @@
 #include <linux/string.h>
 
 #ifdef __HAVE_ARCH_MEMMOVE
+#ifndef CONFIG_OPT_LIB_FUNCTION
 void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 {
 	const char *src = v_src;
 	char *dst = v_dst;
 
-#ifdef CONFIG_OPT_LIB_FUNCTION
-	const uint32_t *i_src;
-	uint32_t *i_dst;
-#endif
-
 	if (!c)
 		return v_dst;
 
@@ -48,7 +44,6 @@ void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 	if (v_dst <= v_src)
 		return memcpy(v_dst, v_src, c);
 
-#ifndef CONFIG_OPT_LIB_FUNCTION
 	/* copy backwards, from end to beginning */
 	src += c;
 	dst += c;
@@ -58,7 +53,22 @@ void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 		*--dst = *--src;
 
 	return v_dst;
-#else
+}
+#else /* CONFIG_OPT_LIB_FUNCTION */
+void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
+{
+	const char *src = v_src;
+	char *dst = v_dst;
+	const uint32_t *i_src;
+	uint32_t *i_dst;
+
+	if (!c)
+		return v_dst;
+
+	/* Use memcpy when source is higher than dest */
+	if (v_dst <= v_src)
+		return memcpy(v_dst, v_src, c);
+
 	/* The following code tries to optimize the copy by using unsigned
 	 * alignment. This will work fine if both source and destination are
 	 * aligned on the same boundary. However, if they are aligned on
@@ -104,7 +114,7 @@ void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 		case 0x1:	/* Unaligned - Off by 1 */
 			/* Word align the source */
 			i_src = (const void *) (((unsigned)src + 4) & ~3);
-
+#ifndef __MICROBLAZEEL__
 			/* Load the holding buffer */
 			buf_hold = *--i_src >> 24;
 
@@ -113,7 +123,16 @@ void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 				*--i_dst = buf_hold << 8 | value;
 				buf_hold = value >> 24;
 			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*--i_src & 0xFF) << 24;
 
+			for (; c >= 4; c -= 4) {
+				value = *--i_src;
+				*--i_dst = buf_hold | ((value & 0xFFFFFF00)>>8);
+				buf_hold = (value  & 0xFF) << 24;
+			}
+#endif
 			/* Realign the source */
 			src = (const void *)i_src;
 			src += 1;
@@ -121,7 +140,7 @@ void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 		case 0x2:	/* Unaligned - Off by 2 */
 			/* Word align the source */
 			i_src = (const void *) (((unsigned)src + 4) & ~3);
-
+#ifndef __MICROBLAZEEL__
 			/* Load the holding buffer */
 			buf_hold = *--i_src >> 16;
 
@@ -130,7 +149,16 @@ void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 				*--i_dst = buf_hold << 16 | value;
 				buf_hold = value >> 16;
 			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*--i_src & 0xFFFF) << 16;
 
+			for (; c >= 4; c -= 4) {
+				value = *--i_src;
+				*--i_dst = buf_hold | ((value & 0xFFFF0000)>>16);
+				buf_hold = (value & 0xFFFF) << 16;
+			}
+#endif
 			/* Realign the source */
 			src = (const void *)i_src;
 			src += 2;
@@ -138,7 +166,7 @@ void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 		case 0x3:	/* Unaligned - Off by 3 */
 			/* Word align the source */
 			i_src = (const void *) (((unsigned)src + 4) & ~3);
-
+#ifndef __MICROBLAZEEL__
 			/* Load the holding buffer */
 			buf_hold = *--i_src >> 8;
 
@@ -147,7 +175,16 @@ void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 				*--i_dst = buf_hold << 24 | value;
 				buf_hold = value >> 8;
 			}
+#else
+			/* Load the holding buffer */
+			buf_hold = (*--i_src & 0xFFFFFF) << 8;
 
+			for (; c >= 4; c -= 4) {
+				value = *--i_src;
+				*--i_dst = buf_hold | ((value & 0xFF000000)>> 24);
+				buf_hold = (value & 0xFFFFFF) << 8;;
+			}
+#endif
 			/* Realign the source */
 			src = (const void *)i_src;
 			src += 3;
@@ -169,7 +206,7 @@ void *memmove(void *v_dst, const void *v_src, __kernel_size_t c)
 		*--dst = *--src;
 	}
 	return v_dst;
-#endif
 }
+#endif /* CONFIG_OPT_LIB_FUNCTION */
 EXPORT_SYMBOL(memmove);
 #endif /* __HAVE_ARCH_MEMMOVE */

arch/microblaze/lib/memset.c

@@ -31,17 +31,30 @@
 #include <linux/string.h>
 
 #ifdef __HAVE_ARCH_MEMSET
+#ifndef CONFIG_OPT_LIB_FUNCTION
+void *memset(void *v_src, int c, __kernel_size_t n)
+{
+	char *src = v_src;
+
+	/* Truncate c to 8 bits */
+	c = (c & 0xFF);
+
+	/* Simple, byte oriented memset or the rest of count. */
+	while (n--)
+		*src++ = c;
+
+	return v_src;
+}
+#else /* CONFIG_OPT_LIB_FUNCTION */
 void *memset(void *v_src, int c, __kernel_size_t n)
 {
 	char *src = v_src;
-#ifdef CONFIG_OPT_LIB_FUNCTION
 	uint32_t *i_src;
 	uint32_t w32 = 0;
-#endif
+
 	/* Truncate c to 8 bits */
 	c = (c & 0xFF);
 
-#ifdef CONFIG_OPT_LIB_FUNCTION
 	if (unlikely(c)) {
 		/* Make a repeating word out of it */
 		w32 = c;
@@ -72,12 +85,13 @@ void *memset(void *v_src, int c, __kernel_size_t n)
 
 		src  = (void *)i_src;
 	}
-#endif
+
 	/* Simple, byte oriented memset or the rest of count. */
 	while (n--)
 		*src++ = c;
 
 	return v_src;
 }
+#endif /* CONFIG_OPT_LIB_FUNCTION */
 EXPORT_SYMBOL(memset);
 #endif /* __HAVE_ARCH_MEMSET */

arch/microblaze/lib/modsi3.S

@@ -0,0 +1,73 @@
+#include <linux/linkage.h>
+
+/*
+* modulo operation for 32 bit integers.
+*	Input :	op1 in Reg r5
+*		op2 in Reg r6
+*	Output: op1 mod op2 in Reg r3
+*/
+
+	.text
+	.globl	__modsi3
+	.type __modsi3,  @function
+	.ent __modsi3
+
+__modsi3:
+	.frame	r1, 0, r15
+
+	addik	r1, r1, -16
+	swi	r28, r1, 0
+	swi	r29, r1, 4
+	swi	r30, r1, 8
+	swi	r31, r1, 12
+
+	beqi	r6, div_by_zero /* div_by_zero division error */
+	beqi	r5, result_is_zero /* result is zero */
+	bgeid	r5, r5_pos
+	/* get the sign of the result [ depends only on the first arg] */
+	add	r28, r5, r0
+	rsubi	r5, r5, 0	 /* make r5 positive */
+r5_pos:
+	bgei	r6, r6_pos
+	rsubi	r6, r6, 0	 /* make r6 positive */
+r6_pos:
+	addik	r3, r0, 0 /* clear mod */
+	addik	r30, r0, 0 /* clear div */
+	addik	r29, r0, 32 /* initialize the loop count */
+/* first part try to find the first '1' in the r5 */
+div1:
+	add	r5, r5, r5 /* left shift logical r5 */
+	bgeid	r5, div1
+	addik	r29, r29, -1
+div2:
+	/* left shift logical r5 get the '1' into the carry */
+	add	r5, r5, r5
+	addc	r3, r3, r3 /* move that bit into the mod register */
+	rsub	r31, r6, r3 /* try to subtract (r30 a r6) */
+	blti	r31, mod_too_small
+	/* move the r31 to mod since the result was positive */
+	or	r3, r0, r31
+	addik	r30, r30, 1
+mod_too_small:
+	addik	r29, r29, -1
+	beqi	r29, loop_end
+	add	r30, r30, r30 /* shift in the '1' into div */
+	bri	div2 /* div2 */
+loop_end:
+	bgei	r28, return_here
+	brid	return_here
+	rsubi	r3, r3, 0 /* negate the result */
+div_by_zero:
+result_is_zero:
+	or	r3, r0, r0 /* set result to 0 [both mod as well as div are 0] */
+return_here:
+/* restore values of csrs and that of r3 and the divisor and the dividend */
+	lwi	r28, r1, 0
+	lwi	r29, r1, 4
+	lwi	r30, r1, 8
+	lwi	r31, r1, 12
+	rtsd	r15, 8
+	addik	r1, r1, 16
+
+.size __modsi3,  . - __modsi3
+.end __modsi3

arch/microblaze/lib/muldi3.S

@@ -0,0 +1,121 @@
+#include <linux/linkage.h>
+
+/*
+ * Multiply operation for 64 bit integers, for devices with hard multiply
+ *	Input :	Operand1[H] in Reg r5
+ *		Operand1[L] in Reg r6
+ *		Operand2[H] in Reg r7
+ *		Operand2[L] in Reg r8
+ *	Output: Result[H] in Reg r3
+ *		Result[L] in Reg r4
+ *
+ * Explaination:
+ *
+ *	Both the input numbers are divided into 16 bit number as follows
+ *		op1 = A B C D
+ *		op2 = E F G H
+ *	result = D * H
+ *		 + (C * H + D * G) << 16
+ *		 + (B * H + C * G + D * F) << 32
+ *		 + (A * H + B * G + C * F + D * E) << 48
+ *
+ *	Only 64 bits of the output are considered
+ */
+
+	.text
+	.globl	__muldi3
+	.type __muldi3, @function
+	.ent __muldi3
+
+__muldi3:
+	addi	r1, r1, -40
+
+/* Save the input operands on the caller's stack */
+	swi	r5, r1, 44
+	swi	r6, r1, 48
+	swi	r7, r1, 52
+	swi	r8, r1, 56
+
+/* Store all the callee saved registers */
+	sw	r20, r1, r0
+	swi	r21, r1, 4
+	swi	r22, r1, 8
+	swi	r23, r1, 12
+	swi	r24, r1, 16
+	swi	r25, r1, 20
+	swi	r26, r1, 24
+	swi	r27, r1, 28
+
+/* Load all the 16 bit values for A thru H */
+	lhui	r20, r1, 44 /* A */
+	lhui	r21, r1, 46 /* B */
+	lhui	r22, r1, 48 /* C */
+	lhui	r23, r1, 50 /* D */
+	lhui	r24, r1, 52 /* E */
+	lhui	r25, r1, 54 /* F */
+	lhui	r26, r1, 56 /* G */
+	lhui	r27, r1, 58 /* H */
+
+/* D * H ==> LSB of the result on stack ==> Store1 */
+	mul	r9, r23, r27
+	swi	r9, r1, 36 /* Pos2 and Pos3 */
+
+/* Hi (Store1) + C * H + D * G ==> Store2 ==> Pos1 and Pos2 */
+/* Store the carry generated in position 2 for Pos 3 */
+	lhui	r11, r1, 36 /* Pos2 */
+	mul	r9, r22, r27 /* C * H */
+	mul	r10, r23, r26 /* D * G */
+	add	r9, r9, r10
+	addc	r12, r0, r0
+	add	r9, r9, r11
+	addc	r12, r12, r0 /* Store the Carry */
+	shi	r9, r1, 36 /* Store Pos2 */
+	swi	r9, r1, 32
+	lhui	r11, r1, 32
+	shi	r11, r1, 34 /* Store Pos1 */
+
+/* Hi (Store2) + B * H + C * G + D * F ==> Store3 ==> Pos0 and Pos1 */
+	mul	r9, r21, r27 /* B * H */
+	mul	r10, r22, r26 /* C * G */
+	mul	r7, r23, r25 /* D * F */
+	add	r9, r9, r11
+	add	r9, r9, r10
+	add	r9, r9, r7
+	swi	r9, r1, 32 /* Pos0 and Pos1 */
+
+/* Hi (Store3) + A * H + B * G + C * F + D * E ==> Store3 ==> Pos0 */
+	lhui	r11, r1, 32 /* Pos0 */
+	mul	r9, r20, r27 /* A * H */
+	mul	r10, r21, r26 /* B * G */
+	mul	r7, r22, r25 /* C * F */
+	mul	r8, r23, r24 /* D * E */
+	add	r9, r9, r11
+	add 	r9, r9, r10
+	add	r9, r9, r7
+	add	r9, r9, r8
+	sext16	r9, r9 /* Sign extend the MSB */
+	shi	r9, r1, 32
+
+/* Move results to r3 and r4 */
+	lhui	r3, r1, 32
+	add	r3, r3, r12
+	shi	r3, r1, 32
+	lwi	r3, r1, 32 /* Hi Part */
+	lwi	r4, r1, 36 /* Lo Part */
+
+/* Restore Callee saved registers */
+	lw	r20, r1, r0
+	lwi	r21, r1, 4
+	lwi	r22, r1, 8
+	lwi	r23, r1, 12
+	lwi	r24, r1, 16
+	lwi	r25, r1, 20
+	lwi	r26, r1, 24
+	lwi	r27, r1, 28
+
+/* Restore Frame and return */
+	rtsd	r15, 8
+	addi	r1, r1, 40
+
+.size __muldi3, . - __muldi3
+.end __muldi3

arch/microblaze/lib/mulsi3.S

@@ -0,0 +1,46 @@
+#include <linux/linkage.h>
+
+/*
+ * Multiply operation for 32 bit integers.
+ *	Input :	Operand1 in Reg r5
+ *		Operand2 in Reg r6
+ *	Output: Result [op1 * op2] in Reg r3
+ */
+	.text
+	.globl	__mulsi3
+	.type __mulsi3,  @function
+	.ent __mulsi3
+
+__mulsi3:
+	.frame	r1, 0, r15
+	add	r3, r0, r0
+	beqi	r5, result_is_zero /* multiply by zero */
+	beqi	r6, result_is_zero /* multiply by zero */
+	bgeid	r5, r5_pos
+	xor	r4, r5, r6 /* get the sign of the result */
+	rsubi	r5, r5, 0 /* make r5 positive */
+r5_pos:
+	bgei	r6, r6_pos
+	rsubi	r6, r6, 0 /* make r6 positive */
+r6_pos:
+	bri	l1
+l2:
+	add	r5, r5, r5
+l1:
+	srl	r6, r6
+	addc	r7, r0, r0
+	beqi	r7, l2
+	bneid	r6, l2
+	add	r3, r3, r5
+	blti	r4, negateresult
+	rtsd	r15, 8
+	nop
+negateresult:
+	rtsd	r15, 8
+	rsub	r3, r3, r0
+result_is_zero:
+	rtsd	r15, 8
+	addi	r3, r0, 0
+
+.size __mulsi3,  . - __mulsi3
+.end __mulsi3

arch/microblaze/lib/udivsi3.S

@@ -0,0 +1,84 @@
+#include <linux/linkage.h>
+
+/*
+* Unsigned divide operation.
+*	Input :	Divisor in Reg r5
+*		Dividend in Reg r6
+*	Output: Result in Reg r3
+*/
+
+	.text
+	.globl	__udivsi3
+	.type __udivsi3, @function
+	.ent __udivsi3
+
+__udivsi3:
+
+	.frame	r1, 0, r15
+
+	addik	r1, r1, -12
+	swi	r29, r1, 0
+	swi	r30, r1, 4
+	swi	r31, r1, 8
+
+	beqi	r6, div_by_zero /* div_by_zero /* division error */
+	beqid	r5, result_is_zero /* result is zero */
+	addik	r30, r0, 0 /* clear mod */
+	addik	r29, r0, 32 /* initialize the loop count */
+
+/* check if r6 and r5 are equal - if yes, return 1 */
+	rsub	r18, r5, r6
+	beqid	r18, return_here
+	addik	r3, r0, 1
+
+/* check if (uns)r6 is greater than (uns)r5. in that case, just return 0 */
+	xor	r18, r5, r6
+	bgeid	r18, 16
+	add	r3, r0, r0 /* we would anyways clear r3 */
+	blti	r6, return_here /* r6[bit 31 = 1] hence is greater */
+	bri	checkr6
+	rsub	r18, r6, r5 /* microblazecmp */
+	blti	r18, return_here
+
+/* if r6 [bit 31] is set, then return result as 1 */
+checkr6:
+	bgti	r6, div0
+	brid	return_here
+	addik	r3, r0, 1
+
+/* first part try to find the first '1' in the r5 */
+div0:
+	blti	r5, div2
+div1:
+	add	r5, r5, r5 /* left shift logical r5 */
+	bgtid	r5, div1
+	addik	r29, r29, -1
+div2:
+/* left shift logical r5 get the '1' into the carry */
+	add	r5, r5, r5
+	addc	r30, r30, r30 /* move that bit into the mod register */
+	rsub	r31, r6, r30 /* try to subtract (r30 a r6) */
+	blti	r31, mod_too_small
+/* move the r31 to mod since the result was positive */
+	or	r30, r0, r31
+	addik	r3, r3, 1
+mod_too_small:
+	addik	r29, r29, -1
+	beqi	r29, loop_end
+	add	r3, r3, r3 /* shift in the '1' into div */
+	bri	div2 /* div2 */
+loop_end:
+	bri	return_here
+div_by_zero:
+result_is_zero:
+	or	r3, r0, r0 /* set result to 0 */
+return_here:
+/* restore values of csrs and that of r3 and the divisor and the dividend */
+	lwi	r29, r1, 0
+	lwi	r30, r1, 4
+	lwi	r31, r1, 8
+	rtsd	r15, 8
+	addik	r1, r1, 12
+
+.size __udivsi3, . - __udivsi3
+.end __udivsi3

arch/microblaze/lib/umodsi3.S

@@ -0,0 +1,86 @@
+#include <linux/linkage.h>
+
+/*
+ * Unsigned modulo operation for 32 bit integers.
+ *	Input :	op1 in Reg r5
+ *		op2 in Reg r6
+ *	Output: op1 mod op2 in Reg r3
+ */
+
+	.text
+	.globl	__umodsi3
+	.type __umodsi3, @function
+	.ent __umodsi3
+
+__umodsi3:
+	.frame	r1, 0, r15
+
+	addik	r1, r1, -12
+	swi	r29, r1, 0
+	swi	r30, r1, 4
+	swi	r31, r1, 8
+
+	beqi	r6, div_by_zero /* div_by_zero - division error */
+	beqid	r5, result_is_zero /* result is zero */
+	addik	r3, r0, 0 /* clear div */
+	addik	r30, r0, 0 /* clear mod */
+	addik	r29, r0, 32 /* initialize the loop count */
+
+/* check if r6 and r5 are equal /* if yes, return 0 */
+	rsub	r18, r5, r6
+	beqi	r18, return_here
+
+/* check if (uns)r6 is greater than (uns)r5. in that case, just return r5 */
+	xor	r18, r5, r6
+	bgeid	r18, 16
+	addik	r3, r5, 0
+	blti	r6, return_here
+	bri	$lcheckr6
+	rsub	r18, r5, r6 /* microblazecmp */
+	bgti	r18, return_here
+
+/* if r6 [bit 31] is set, then return result as r5-r6 */
+$lcheckr6:
+	bgtid	r6, div0
+	addik	r3, r0, 0
+	addik	r18, r0, 0x7fffffff
+	and	r5, r5, r18
+	and	r6, r6, r18
+	brid	return_here
+	rsub	r3, r6, r5
+/* first part: try to find the first '1' in the r5 */
+div0:
+	blti	r5, div2
+div1:
+	add	r5, r5, r5 /* left shift logical r5 */
+	bgeid	r5, div1
+	addik	r29, r29, -1
+div2:
+	/* left shift logical r5 get the '1' into the carry */
+	add	r5, r5, r5
+	addc	r3, r3, r3 /* move that bit into the mod register */
+	rsub	r31, r6, r3 /* try to subtract (r3 a r6) */
+	blti	r31, mod_too_small
+	/* move the r31 to mod since the result was positive */
+	or	r3, r0, r31
+	addik	r30, r30, 1
+mod_too_small:
+	addik	r29, r29, -1
+	beqi	r29, loop_end
+	add	r30, r30, r30 /* shift in the '1' into div */
+	bri	div2 /* div2 */
+loop_end:
+	bri	return_here
+div_by_zero:
+result_is_zero:
+	or	r3, r0, r0 /* set result to 0 */
+return_here:
+/* restore values of csrs and that of r3 and the divisor and the dividend */
+	lwi	r29, r1, 0
+	lwi	r30, r1, 4
+	lwi	r31, r1, 8
+	rtsd	r15, 8
+	addik	r1, r1, 12
+
+.size __umodsi3, . - __umodsi3
+.end __umodsi3

arch/microblaze/pci/pci-common.c

@@ -60,21 +60,6 @@ struct dma_map_ops *get_pci_dma_ops(void)
 }
 EXPORT_SYMBOL(get_pci_dma_ops);
 
-int pci_set_dma_mask(struct pci_dev *dev, u64 mask)
-{
-	return dma_set_mask(&dev->dev, mask);
-}
-
-int pci_set_consistent_dma_mask(struct pci_dev *dev, u64 mask)
-{
-	int rc;
-
-	rc = dma_set_mask(&dev->dev, mask);
-	dev->dev.coherent_dma_mask = dev->dma_mask;
-
-	return rc;
-}
-
 struct pci_controller *pcibios_alloc_controller(struct device_node *dev)
 {
 	struct pci_controller *phb;
@@ -1075,8 +1060,6 @@ void __devinit pcibios_setup_bus_devices(struct pci_bus *bus)
 		 bus->number, bus->self ? pci_name(bus->self) : "PHB");
 
 	list_for_each_entry(dev, &bus->devices, bus_list) {
-		struct dev_archdata *sd = &dev->dev.archdata;
-
 		/* Setup OF node pointer in archdata */
 		dev->dev.of_node = pci_device_to_OF_node(dev);
 
@@ -1086,8 +1069,8 @@ void __devinit pcibios_setup_bus_devices(struct pci_bus *bus)
 		set_dev_node(&dev->dev, pcibus_to_node(dev->bus));
 
 		/* Hook up default DMA ops */
-		sd->dma_ops = pci_dma_ops;
-		sd->dma_data = (void *)PCI_DRAM_OFFSET;
+		set_dma_ops(&dev->dev, pci_dma_ops);
+		dev->dev.archdata.dma_data = (void *)PCI_DRAM_OFFSET;
 
 		/* Read default IRQs and fixup if necessary */
 		pci_read_irq_line(dev);

arch/microblaze/platform/generic/system.dts

@@ -85,6 +85,7 @@
 			xlnx,dynamic-bus-sizing = <0x1>;
 			xlnx,edge-is-positive = <0x1>;
 			xlnx,family = "virtex5";
+			xlnx,endianness = <0x1>;
 			xlnx,fpu-exception = <0x1>;
 			xlnx,fsl-data-size = <0x20>;
 			xlnx,fsl-exception = <0x0>;
@@ -218,6 +219,7 @@
 			#address-cells = <1>;
 			#size-cells = <1>;
 			compatible = "xlnx,compound";
+			ranges ;
 			ethernet@81c00000 {
 				compatible = "xlnx,xps-ll-temac-1.01.b", "xlnx,xps-ll-temac-1.00.a";
 				device_type = "network";
@@ -332,6 +334,7 @@
 			#address-cells = <1>;
 			#size-cells = <1>;
 			compatible = "xlnx,mpmc-4.02.a";
+			ranges ;
 			PIM3: sdma@84600180 {
 				compatible = "xlnx,ll-dma-1.00.a";
 				interrupt-parent = <&xps_intc_0>;

arch/microblaze/platform/platform.c

@@ -17,9 +17,6 @@
 
 static struct of_device_id xilinx_of_bus_ids[] __initdata = {
 	{ .compatible = "simple-bus", },
-	{ .compatible = "xlnx,plb-v46-1.00.a", },
-	{ .compatible = "xlnx,opb-v20-1.10.c", },
-	{ .compatible = "xlnx,opb-v20-1.10.b", },
 	{ .compatible = "xlnx,compound", },
 	{}
 };

drivers/net/xilinx_emaclite.c

@@ -430,8 +430,8 @@ static u16 xemaclite_recv_data(struct net_local *drvdata, u8 *data)
 	}
 
 	/* Get the protocol type of the ethernet frame that arrived */
-	proto_type = ((in_be32(addr + XEL_HEADER_OFFSET +
-			XEL_RXBUFF_OFFSET) >> XEL_HEADER_SHIFT) &
+	proto_type = ((ntohl(in_be32(addr + XEL_HEADER_OFFSET +
+			XEL_RXBUFF_OFFSET)) >> XEL_HEADER_SHIFT) &
 			XEL_RPLR_LENGTH_MASK);
 
 	/* Check if received ethernet frame is a raw ethernet frame
@@ -439,9 +439,9 @@ static u16 xemaclite_recv_data(struct net_local *drvdata, u8 *data)
 	if (proto_type > (ETH_FRAME_LEN + ETH_FCS_LEN)) {
 
 		if (proto_type == ETH_P_IP) {
-			length = ((in_be32(addr +
+			length = ((ntohl(in_be32(addr +
 					XEL_HEADER_IP_LENGTH_OFFSET +
-					XEL_RXBUFF_OFFSET) >>
+					XEL_RXBUFF_OFFSET)) >>
 					XEL_HEADER_SHIFT) &
 					XEL_RPLR_LENGTH_MASK);
 			length += ETH_HLEN + ETH_FCS_LEN;

drivers/video/xilinxfb.c

@@ -32,10 +32,14 @@
 #include <linux/dma-mapping.h>
 #include <linux/of_device.h>
 #include <linux/of_platform.h>
+#include <linux/of_address.h>
 #include <linux/io.h>
 #include <linux/xilinxfb.h>
 #include <linux/slab.h>
+
+#ifdef CONFIG_PPC_DCR
 #include <asm/dcr.h>
+#endif
 
 #define DRIVER_NAME		"xilinxfb"
 
@@ -123,10 +127,10 @@ struct xilinxfb_drvdata {
 						registers */
 	void __iomem	*regs;		/* virt. address of the control
 						registers */
-
+#ifdef CONFIG_PPC_DCR
 	dcr_host_t      dcr_host;
 	unsigned int    dcr_len;
-
+#endif
 	void		*fb_virt;	/* virt. address of the frame buffer */
 	dma_addr_t	fb_phys;	/* phys. address of the frame buffer */
 	int		fb_alloced;	/* Flag, was the fb memory alloced? */
@@ -152,9 +156,10 @@ static void xilinx_fb_out_be32(struct xilinxfb_drvdata *drvdata, u32 offset,
 {
 	if (drvdata->flags & PLB_ACCESS_FLAG)
 		out_be32(drvdata->regs + (offset << 2), val);
+#ifdef CONFIG_PPC_DCR
 	else
 		dcr_write(drvdata->dcr_host, offset, val);
-
+#endif
 }
 
 static int
@@ -383,8 +388,11 @@ static int xilinxfb_release(struct device *dev)
 	if (drvdata->flags & PLB_ACCESS_FLAG) {
 		iounmap(drvdata->regs);
 		release_mem_region(drvdata->regs_phys, 8);
-	} else
+	}
+#ifdef CONFIG_PPC_DCR
+	else
 		dcr_unmap(drvdata->dcr_host, drvdata->dcr_len);
+#endif
 
 	kfree(drvdata);
 	dev_set_drvdata(dev, NULL);
@@ -404,7 +412,7 @@ xilinxfb_of_probe(struct platform_device *op, const struct of_device_id *match)
 	u32 tft_access;
 	struct xilinxfb_platform_data pdata;
 	struct resource res;
-	int size, rc, start;
+	int size, rc;
 	struct xilinxfb_drvdata *drvdata;
 
 	/* Copy with the default pdata (not a ptr reference!) */
@@ -437,7 +445,10 @@ xilinxfb_of_probe(struct platform_device *op, const struct of_device_id *match)
 			dev_err(&op->dev, "invalid address\n");
 			goto err;
 		}
-	} else {
+	}
+#ifdef CONFIG_PPC_DCR
+	else {
+		int start;
 		res.start = 0;
 		start = dcr_resource_start(op->dev.of_node, 0);
 		drvdata->dcr_len = dcr_resource_len(op->dev.of_node, 0);
@@ -447,6 +458,7 @@ xilinxfb_of_probe(struct platform_device *op, const struct of_device_id *match)
 			goto err;
 		}
 	}
+#endif
 
 	prop = of_get_property(op->dev.of_node, "phys-size", &size);
 	if ((prop) && (size >= sizeof(u32)*2)) {