Merge bk://linux-dj.bkbits.net/agpgart

into home.osdl.org:/home/torvalds/v2.5/linux

MAINTAINERS

@@ -879,6 +879,8 @@ W:	http://www.developer.ibm.com/welcome/netfinity/serveraid.html
 S:	Supported 
 
 IDE DRIVER [GENERAL]
+P:	Bartlomiej Zolnierkiewicz
+M:	B.Zolnierkiewicz@elka.pw.edu.pl
 L:	linux-kernel@vger.kernel.org
 L:	linux-ide@vger.kernel.org
 S:	Maintained

arch/sparc/boot/Makefile

@@ -32,3 +32,6 @@ $(obj)/tftpboot.img: $(obj)/piggyback $(obj)/System.map $(obj)/image FORCE
 
 $(obj)/btfix.s: $(obj)/btfixupprep vmlinux FORCE
 	$(call if_changed,btfix)
+
+clean:
+	rm $(obj)/System.map

arch/sparc/defconfig

@@ -138,29 +138,61 @@ CONFIG_BLK_DEV_INITRD=y
 CONFIG_SCSI=y
 
 #
-# SCSI support type (disk, tape, CDrom)
+# SCSI support type (disk, tape, CD-ROM)
 #
 CONFIG_BLK_DEV_SD=y
-CONFIG_SD_EXTRA_DEVS=40
 # CONFIG_CHR_DEV_ST is not set
 # CONFIG_CHR_DEV_OSST is not set
 CONFIG_BLK_DEV_SR=m
 # CONFIG_BLK_DEV_SR_VENDOR is not set
-CONFIG_SR_EXTRA_DEVS=2
 CONFIG_CHR_DEV_SG=m
 
 #
 # Some SCSI devices (e.g. CD jukebox) support multiple LUNs
 #
 # CONFIG_SCSI_MULTI_LUN is not set
+CONFIG_SCSI_REPORT_LUNS=y
 # CONFIG_SCSI_CONSTANTS is not set
 # CONFIG_SCSI_LOGGING is not set
 
 #
 # SCSI low-level drivers
 #
-CONFIG_SCSI_SUNESP=y
+# CONFIG_BLK_DEV_3W_XXXX_RAID is not set
+# CONFIG_SCSI_ACARD is not set
+# CONFIG_SCSI_AACRAID is not set
+# CONFIG_SCSI_AIC7XXX is not set
+# CONFIG_SCSI_AIC7XXX_OLD is not set
+# CONFIG_SCSI_AIC79XX is not set
+# CONFIG_SCSI_DPT_I2O is not set
+# CONFIG_SCSI_ADVANSYS is not set
+# CONFIG_SCSI_AM53C974 is not set
+# CONFIG_SCSI_MEGARAID is not set
+# CONFIG_SCSI_BUSLOGIC is not set
+# CONFIG_SCSI_CPQFCTS is not set
+# CONFIG_SCSI_DMX3191D is not set
+# CONFIG_SCSI_EATA is not set
+# CONFIG_SCSI_EATA_PIO is not set
+# CONFIG_SCSI_FUTURE_DOMAIN is not set
+# CONFIG_SCSI_GDTH is not set
+# CONFIG_SCSI_GENERIC_NCR5380 is not set
+# CONFIG_SCSI_GENERIC_NCR5380_MMIO is not set
+# CONFIG_SCSI_INITIO is not set
+# CONFIG_SCSI_INIA100 is not set
+# CONFIG_SCSI_SYM53C8XX_2 is not set
+# CONFIG_SCSI_NCR53C8XX is not set
+# CONFIG_SCSI_PCI2000 is not set
+# CONFIG_SCSI_PCI2220I is not set
+# CONFIG_SCSI_QLOGIC_ISP is not set
+# CONFIG_SCSI_QLOGIC_FC is not set
+# CONFIG_SCSI_QLOGIC_1280 is not set
 CONFIG_SCSI_QLOGICPTI=m
+# CONFIG_SCSI_DC395x is not set
+# CONFIG_SCSI_DC390T is not set
+# CONFIG_SCSI_U14_34F is not set
+# CONFIG_SCSI_NSP32 is not set
+# CONFIG_SCSI_DEBUG is not set
+CONFIG_SCSI_SUNESP=y
 
 #
 # Fibre Channel support

arch/sparc/kernel/Makefile

@@ -12,8 +12,9 @@ obj-y    := entry.o wof.o wuf.o etrap.o rtrap.o traps.o $(IRQ_OBJS) \
 	    sys_sparc.o sunos_asm.o systbls.o \
 	    time.o windows.o cpu.o devices.o sclow.o \
 	    tadpole.o tick14.o ptrace.o sys_solaris.o \
-	    unaligned.o muldiv.o pcic.o semaphore.o sparc_ksyms.o
+	    unaligned.o muldiv.o semaphore.o sparc_ksyms.o
 
+obj-$(CONFIG_PCI) += pcic.o
 obj-$(CONFIG_SUN4) += sun4setup.o
 obj-$(CONFIG_SMP) += trampoline.o smp.o sun4m_smp.o sun4d_smp.o
 obj-$(CONFIG_SUN_AUXIO) += auxio.o

arch/sparc/kernel/pcic.c

@@ -35,27 +35,6 @@
 #include <asm/timer.h>
 #include <asm/uaccess.h>
 
-#ifndef CONFIG_PCI
-
-asmlinkage int sys_pciconfig_read(unsigned long bus,
-				  unsigned long dfn,
-				  unsigned long off,
-				  unsigned long len,
-				  unsigned char *buf)
-{
-	return -EINVAL;
-}
-
-asmlinkage int sys_pciconfig_write(unsigned long bus,
-				   unsigned long dfn,
-				   unsigned long off,
-				   unsigned long len,
-				   unsigned char *buf)
-{
-	return -EINVAL;
-}
-
-#else
 
 struct pci_fixup pcibios_fixups[] = {
 	{ 0 }
@@ -1044,5 +1023,3 @@ void insl(unsigned long addr, void *dst, unsigned long count) {
 }
 
 subsys_initcall(pcic_init);
-
-#endif

arch/sparc/kernel/process.c

@@ -287,13 +287,22 @@ void show_regs(struct pt_regs *r)
 	       rw->ins[4], rw->ins[5], rw->ins[6], rw->ins[7]);
 }
 
+/*
+ * The show_stack is an external API which we do not use ourselves.
+ * The oops is printed in die_if_kernel.
+ */
 void show_stack(struct task_struct *tsk, unsigned long *_ksp)
 {
 	unsigned long pc, fp;
-	unsigned long task_base = (unsigned long) tsk;
+	unsigned long task_base;
 	struct reg_window *rw;
 	int count = 0;
 
+	if (tsk != NULL)
+		task_base = (unsigned long) tsk->thread_info;
+	else
+		task_base = (unsigned long) current_thread_info();
+
 	fp = (unsigned long) _ksp;
 	do {
 		/* Bogus frame pointer? */
@@ -308,13 +317,6 @@ void show_stack(struct task_struct *tsk, unsigned long *_ksp)
 	printk("\n");
 }
 
-void show_trace_task(struct task_struct *tsk)
-{
-	if (tsk)
-		show_stack(tsk,
-			   (unsigned long *) tsk->thread_info->ksp);
-}
-
 /*
  * Note: sparc64 has a pretty intricated thread_saved_pc, check it out.
  */

arch/sparc/kernel/sys_sparc.c

@@ -123,7 +123,10 @@ asmlinkage int sys_ipc (uint call, int first, int second, int third, void __user
 	if (call <= SEMCTL)
 		switch (call) {
 		case SEMOP:
-			err = sys_semop (first, (struct sembuf __user *)ptr, second);
+			err = sys_semtimedop (first, (struct sembuf __user *)ptr, second, NULL);
+			goto out;
+		case SEMTIMEDOP:
+			err = sys_semtimedop (first, (struct sembuf __user *)ptr, second, (const struct timespec __user *) fifth);
 			goto out;
 		case SEMGET:
 			err = sys_semget (first, second, third);

arch/sparc/mm/srmmu.c

@@ -148,6 +148,9 @@ extern unsigned long fix_kmap_end;
 /* 1 bit <=> 256 bytes of nocache <=> 64 PTEs */
 #define SRMMU_NOCACHE_BITMAP_SHIFT (PAGE_SHIFT - 4)
 
+/* The context table is a nocache user with the biggest alignment needs. */
+#define SRMMU_NOCACHE_ALIGN_MAX (sizeof(ctxd_t)*SRMMU_MAX_CONTEXTS)
+
 void *srmmu_nocache_pool;
 void *srmmu_nocache_bitmap;
 static struct bit_map srmmu_nocache_map;
@@ -320,6 +323,10 @@ static unsigned long __srmmu_get_nocache(int size, int align)
 		printk("Size 0x%x unaligned int nocache request\n", size);
 		size += SRMMU_NOCACHE_BITMAP_SHIFT-1;
 	}
+	if (align > SRMMU_NOCACHE_ALIGN_MAX) {
+		BUG();
+		return 0;
+	}
 
 	offset = bit_map_string_get(&srmmu_nocache_map,
 		       			size >> SRMMU_NOCACHE_BITMAP_SHIFT,
@@ -425,7 +432,8 @@ void srmmu_nocache_init(void)
 
 	bitmap_bits = srmmu_nocache_size >> SRMMU_NOCACHE_BITMAP_SHIFT;
 
-	srmmu_nocache_pool = __alloc_bootmem(srmmu_nocache_size, PAGE_SIZE, 0UL);
+	srmmu_nocache_pool = __alloc_bootmem(srmmu_nocache_size,
+		SRMMU_NOCACHE_ALIGN_MAX, 0UL);
 	memset(srmmu_nocache_pool, 0, srmmu_nocache_size);
 
 	srmmu_nocache_bitmap = __alloc_bootmem(bitmap_bits >> 3, SMP_CACHE_BYTES, 0UL);

arch/sparc64/Kconfig

@@ -924,13 +924,6 @@ config DEBUG_SPINLOCK
 	  best used in conjunction with the NMI watchdog so that spinlock
 	  deadlocks are also debuggable.
 
-config KALLSYMS
-	bool "Load all symbols for debugging/ksymoops"
-	help
-	  Say Y here to let the kernel print out symbolic crash information and
-	  symbolic stack backtraces. This increases the size of the kernel
-	  somewhat, as all symbols have to be loaded into the kernel image.
-
 config DEBUG_SPINLOCK_SLEEP
 	bool "Sleep-inside-spinlock checking"
 	help

arch/sparc64/kernel/entry.S

@@ -1749,8 +1749,8 @@ ret_sys_call:
 	cmp		%o0, -ENOIOCTLCMD
 	sllx		%g2, 32, %g2
 	bgeu,pn		%xcc, 1f
-
 	 andcc		%l0, _TIF_SYSCALL_TRACE, %l6	
+80:
 	andn		%g3, %g2, %g3		/* System call success, clear Carry condition code. */
 	stx		%g3, [%sp + PTREGS_OFF + PT_V9_TSTATE]	
 	bne,pn		%icc, linux_syscall_trace2
@@ -1760,9 +1760,21 @@ ret_sys_call:
 	 stx		%l2, [%sp + PTREGS_OFF + PT_V9_TNPC]
 
 1:
+	/* Really a failure?  Check if force_successful_syscall_return()
+	 * was invoked.
+	 */
+	ldx		[%curptr + TI_FLAGS], %l0		! Load
+	andcc		%l0, _TIF_SYSCALL_SUCCESS, %g0
+	be,pt		%icc, 1f
+	 andcc		%l0, _TIF_SYSCALL_TRACE, %l6	
+	andn		%l0, _TIF_SYSCALL_SUCCESS, %l0
+	ba,pt		%xcc, 80b
+	 stx		%l0, [%curptr + TI_FLAGS]
+
 	/* System call failure, set Carry condition code.
 	 * Also, get abs(errno) to return to the process.
 	 */
+1:
 	sub		%g0, %o0, %o0
 	or		%g3, %g2, %g3
 	stx		%o0, [%sp + PTREGS_OFF + PT_V9_I0]

arch/sparc64/kernel/sys_sparc.c

@@ -46,8 +46,10 @@ asmlinkage unsigned long sys_getpagesize(void)
 
 unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsigned long len, unsigned long pgoff, unsigned long flags)
 {
-	struct vm_area_struct * vmm;
+	struct mm_struct *mm = current->mm;
+	struct vm_area_struct * vma;
 	unsigned long task_size = TASK_SIZE;
+	unsigned long start_addr;
 	int do_color_align;
 
 	if (flags & MAP_FIXED) {
@@ -63,30 +65,54 @@ unsigned long arch_get_unmapped_area(struct file *filp, unsigned long addr, unsi
 		task_size = 0xf0000000UL;
 	if (len > task_size || len > -PAGE_OFFSET)
 		return -ENOMEM;
-	if (!addr)
-		addr = TASK_UNMAPPED_BASE;
 
 	do_color_align = 0;
 	if (filp || (flags & MAP_SHARED))
 		do_color_align = 1;
 
+	if (addr) {
+		if (do_color_align)
+			addr = COLOUR_ALIGN(addr, pgoff);
+		else
+			addr = PAGE_ALIGN(addr);
+
+		vma = find_vma(mm, addr);
+		if (task_size - len >= addr &&
+		    (!vma || addr + len <= vma->vm_start))
+			return addr;
+	}
+
+	start_addr = addr = mm->free_area_cache;
+
+	task_size -= len;
+
+full_search:
 	if (do_color_align)
 		addr = COLOUR_ALIGN(addr, pgoff);
 	else
 		addr = PAGE_ALIGN(addr);
-	task_size -= len;
 
-	for (vmm = find_vma(current->mm, addr); ; vmm = vmm->vm_next) {
-		/* At this point:  (!vmm || addr < vmm->vm_end). */
+	for (vma = find_vma(mm, addr); ; vma = vma->vm_next) {
+		/* At this point:  (!vma || addr < vma->vm_end). */
 		if (addr < PAGE_OFFSET && -PAGE_OFFSET - len < addr) {
 			addr = PAGE_OFFSET;
-			vmm = find_vma(current->mm, PAGE_OFFSET);
+			vma = find_vma(mm, PAGE_OFFSET);
 		}
-		if (task_size < addr)
+		if (task_size < addr) {
+			if (start_addr != TASK_UNMAPPED_BASE) {
+				start_addr = addr = TASK_UNMAPPED_BASE;
+				goto full_search;
+			}
 			return -ENOMEM;
-		if (!vmm || addr + len <= vmm->vm_start)
+		}
+		if (!vma || addr + len <= vma->vm_start) {
+			/*
+			 * Remember the place where we stopped the search:
+			 */
+			mm->free_area_cache = addr + len;
 			return addr;
-		addr = vmm->vm_end;
+		}
+		addr = vma->vm_end;
 		if (do_color_align)
 			addr = COLOUR_ALIGN(addr, pgoff);
 	}
@@ -182,7 +208,10 @@ asmlinkage int sys_ipc (unsigned call, int first, int second, unsigned long thir
 	if (call <= SEMCTL)
 		switch (call) {
 		case SEMOP:
-			err = sys_semop (first, (struct sembuf *)ptr, second);
+			err = sys_semtimedop (first, (struct sembuf *)ptr, second, NULL);
+			goto out;
+		case SEMTIMEDOP:
+			err = sys_semtimedop (first, (struct sembuf *)ptr, second, (const struct timespec *) fifth);
 			goto out;
 		case SEMGET:
 			err = sys_semget (first, second, (int)third);

arch/sparc64/kernel/sys_sparc32.c

@@ -480,7 +480,7 @@ static int do_sys32_semctl(int first, int second, int third, void *uptr)
 
 static int do_sys32_msgsnd (int first, int second, int third, void *uptr)
 {
-	struct msgbuf *p = kmalloc (second + sizeof (struct msgbuf) + 4, GFP_USER);
+	struct msgbuf *p = kmalloc (second + sizeof (struct msgbuf), GFP_USER);
 	struct msgbuf32 *up = (struct msgbuf32 *)uptr;
 	mm_segment_t old_fs;
 	int err;
@@ -522,12 +522,12 @@ static int do_sys32_msgrcv (int first, int second, int msgtyp, int third,
 		msgtyp = ipck.msgtyp;
 	}
 	err = -ENOMEM;
-	p = kmalloc (second + sizeof (struct msgbuf) + 4, GFP_USER);
+	p = kmalloc (second + sizeof (struct msgbuf), GFP_USER);
 	if (!p)
 		goto out;
 	old_fs = get_fs ();
 	set_fs (KERNEL_DS);
-	err = sys_msgrcv (first, p, second + 4, msgtyp, third);
+	err = sys_msgrcv (first, p, second, msgtyp, third);
 	set_fs (old_fs);
 	if (err < 0)
 		goto free_then_out;
@@ -736,6 +736,22 @@ static int do_sys32_shmctl (int first, int second, void *uptr)
 	return err;
 }
 
+static int sys32_semtimedop(int semid, struct sembuf *tsems, int nsems,
+			    const struct compat_timespec *timeout32)
+{
+	struct compat_timespec t32;
+	struct timespec *t64 = compat_alloc_user_space(sizeof(*t64));
+
+	if (copy_from_user(&t32, timeout32, sizeof(t32)))
+		return -EFAULT;
+
+	if (put_user(t32.tv_sec, &t64->tv_sec) ||
+	    put_user(t32.tv_nsec, &t64->tv_nsec))
+		return -EFAULT;
+
+	return sys_semtimedop(semid, tsems, nsems, t64);
+}
+
 asmlinkage int sys32_ipc (u32 call, int first, int second, int third, u32 ptr, u32 fifth)
 {
 	int version, err;
@@ -747,8 +763,10 @@ asmlinkage int sys32_ipc (u32 call, int first, int second, int third, u32 ptr, u
 		switch (call) {
 		case SEMOP:
 			/* struct sembuf is the same on 32 and 64bit :)) */
-			err = sys_semop (first, (struct sembuf *)AA(ptr), second);
+			err = sys_semtimedop (first, (struct sembuf *)AA(ptr), second, NULL);
 			goto out;
+		case SEMTIMEDOP:
+			err = sys32_semtimedop (first, (struct sembuf *)AA(ptr), second, (const struct compat_timespec *) AA(fifth));
 		case SEMGET:
 			err = sys_semget (first, second, third);
 			goto out;

drivers/pcmcia/ti113x.h

@@ -176,6 +176,9 @@ static int ti_override(struct yenta_socket *socket)
 	if (new != reg)
 		exca_writeb(socket, I365_INTCTL, new);
 
+#if 0
+	/* THIS CAUSES HANGS! Disabled for now, do not know why */
+
 	/*
 	 * If ISA interrupts don't work, then fall back to routing card
 	 * interrupts to the PCI interrupt of the socket.
@@ -195,6 +198,7 @@ static int ti_override(struct yenta_socket *socket)
 		config_writel(socket, TI122X_IRQMUX, irqmux);
 		config_writeb(socket, TI113X_DEVICE_CONTROL, devctl);
 	}
+#endif
 
 	socket->socket.ss_entry->init = ti_init;
 	return 0;

include/asm-sparc/ipc.h

@@ -14,6 +14,7 @@ struct ipc_kludge {
 #define SEMOP		 1
 #define SEMGET		 2
 #define SEMCTL		 3
+#define SEMTIMEDOP	 4
 #define MSGSND		11
 #define MSGRCV		12
 #define MSGGET		13

include/asm-sparc/pgtsrmmu.h

@@ -13,6 +13,9 @@
 #include <asm/thread_info.h>	/* TI_UWINMASK for WINDOW_FLUSH */
 #endif
 
+/* Number of contexts is implementation-dependent; 64k is the most we support */
+#define SRMMU_MAX_CONTEXTS	65536
+
 /* PMD_SHIFT determines the size of the area a second-level page table entry can map */
 #define SRMMU_PMD_SHIFT         18
 #define SRMMU_PMD_SIZE          (1UL << SRMMU_PMD_SHIFT)

include/asm-sparc64/ipc.h

@@ -14,6 +14,7 @@ struct ipc_kludge {
 #define SEMOP		 1
 #define SEMGET		 2
 #define SEMCTL		 3
+#define SEMTIMEDOP	 4
 #define MSGSND		11
 #define MSGRCV		12
 #define MSGGET		13

include/asm-sparc64/ptrace.h

@@ -94,6 +94,8 @@ struct sparc_trapf {
 #define STACKFRAME32_SZ	sizeof(struct sparc_stackf32)
 
 #ifdef __KERNEL__
+#define force_successful_syscall_return()	\
+	set_thread_flag(TIF_SYSCALL_SUCCESS)
 #define user_mode(regs) (!((regs)->tstate & TSTATE_PRIV))
 #define instruction_pointer(regs) ((regs)->tpc)
 extern void show_regs(struct pt_regs *);

include/asm-sparc64/thread_info.h

@@ -205,7 +205,12 @@ register struct thread_info *current_thread_info_reg asm("g6");
 #define TIF_BLKCOMMIT		9	/* use ASI_BLK_COMMIT_* in copy_user_page */
 
 #define TIF_POLLING_NRFLAG	10
-#define TIF_ABI_PENDING		11
+#define TIF_SYSCALL_SUCCESS	11
+/* NOTE: Thread flags >= 12 should be ones we have no interest
+ *       in using in assembly, else we can't use the mask as
+ *       an immediate value in instructions such as andcc.
+ */
+#define TIF_ABI_PENDING		12
 
 #define _TIF_SYSCALL_TRACE	(1<<TIF_SYSCALL_TRACE)
 #define _TIF_NOTIFY_RESUME	(1<<TIF_NOTIFY_RESUME)
@@ -219,6 +224,7 @@ register struct thread_info *current_thread_info_reg asm("g6");
 #define _TIF_BLKCOMMIT		(1<<TIF_BLKCOMMIT)
 #define _TIF_POLLING_NRFLAG	(1<<TIF_POLLING_NRFLAG)
 #define _TIF_ABI_PENDING	(1<<TIF_ABI_PENDING)
+#define _TIF_SYSCALL_SUCCESS	(1<<TIF_SYSCALL_SUCCESS)
 
 #define _TIF_USER_WORK_MASK	((0xff << TI_FLAG_WSAVED_SHIFT) | \
 				 (_TIF_NOTIFY_RESUME | _TIF_SIGPENDING | \

include/net/ip6_route.h

@@ -61,7 +61,11 @@ extern struct rt6_info		*rt6_lookup(struct in6_addr *daddr,
 					    struct in6_addr *saddr,
 					    int oif, int flags);
 
-extern struct rt6_info		*ip6_dst_alloc(void);
+extern struct dst_entry *ndisc_dst_alloc(struct net_device *dev,
+					 struct neighbour *neigh,
+					 int (*output)(struct sk_buff *));
+extern int ndisc_dst_gc(int *more);
+extern void fib6_force_start_gc(void);
 
 /*
  *	support functions for ND

kernel/sys.c

@@ -233,6 +233,8 @@ cond_syscall(compat_sys_futex)
 cond_syscall(sys_epoll_create)
 cond_syscall(sys_epoll_ctl)
 cond_syscall(sys_epoll_wait)
+cond_syscall(sys_pciconfig_read)
+cond_syscall(sys_pciconfig_write)
 
 static int set_one_prio(struct task_struct *p, int niceval, int error)
 {

net/compat.c

@@ -405,34 +405,6 @@ static int do_set_attach_filter(int fd, int level, int optname,
 			      sizeof(struct sock_fprog));
 }
 
-static int do_set_icmpv6_filter(int fd, int level, int optname,
-				char *optval, int optlen)
-{
-	struct icmp6_filter kfilter;
-	mm_segment_t old_fs;
-	int ret, i;
-
-	if (optlen < sizeof(kfilter))
-		return -EINVAL;
-	if (copy_from_user(&kfilter, optval, sizeof(kfilter)))
-		return -EFAULT;
-
-	for (i = 0; i < 8; i += 2) {
-		u32 tmp = kfilter.data[i];
-
-		kfilter.data[i] = kfilter.data[i + 1];
-		kfilter.data[i + 1] = tmp;
-	}
-
-	old_fs = get_fs();
-	set_fs(KERNEL_DS);
-	ret = sys_setsockopt(fd, level, optname,
-			     (char *) &kfilter, sizeof(kfilter));
-	set_fs(old_fs);
-
-	return ret;
-}
-
 static int do_set_sock_timeout(int fd, int level, int optname, char *optval, int optlen)
 {
 	struct compat_timeval *up = (struct compat_timeval *) optval;
@@ -465,9 +437,6 @@ asmlinkage long compat_sys_setsockopt(int fd, int level, int optname,
 					    optval, optlen);
 	if (optname == SO_RCVTIMEO || optname == SO_SNDTIMEO)
 		return do_set_sock_timeout(fd, level, optname, optval, optlen);
-	if (level == SOL_ICMPV6 && optname == ICMPV6_FILTER)
-		return do_set_icmpv6_filter(fd, level, optname,
-					    optval, optlen);
 
 	return sys_setsockopt(fd, level, optname, optval, optlen);
 }

net/core/net-sysfs.c

 /*
  * net-sysfs.c - network device class and attributes
  *
- * Copyright (c) 2003 Stephen Hemminber <shemminger@osdl.org>
+ * Copyright (c) 2003 Stephen Hemminger <shemminger@osdl.org>
  * 
  */
 

net/ipv4/Kconfig

@@ -375,4 +375,5 @@ config INET_IPCOMP
 	  If unsure, say Y.
 
 source "net/ipv4/netfilter/Kconfig"
+source "net/ipv4/ipvs/Kconfig"
 

net/ipv4/Makefile

@@ -21,5 +21,6 @@ obj-$(CONFIG_INET_ESP) += esp4.o
 obj-$(CONFIG_INET_IPCOMP) += ipcomp.o
 obj-$(CONFIG_IP_PNP) += ipconfig.o
 obj-$(CONFIG_NETFILTER)	+= netfilter/
+obj-$(CONFIG_IP_VS) += ipvs/
 
 obj-y += xfrm4_policy.o xfrm4_state.o xfrm4_input.o xfrm4_tunnel.o

net/ipv4/ipvs/Makefile

+#
+# Makefile for the IPVS modules on top of IPv4.
+#
+# Note! Dependencies are done automagically by 'make dep', which also
+# removes any old dependencies. DON'T put your own dependencies here
+# unless it's something special (ie not a .c file).
+#
+# Note 2! The CFLAGS definition is now in the main makefile...
+
+
+# IPVS transport protocol load balancing support
+ip_vs_proto-objs-y :=
+ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_TCP) += ip_vs_proto_tcp.o
+ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_UDP) += ip_vs_proto_udp.o
+ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_ESP) += ip_vs_proto_esp.o
+ip_vs_proto-objs-$(CONFIG_IP_VS_PROTO_AH) += ip_vs_proto_ah.o
+
+ip_vs-objs :=	ip_vs_conn.o ip_vs_core.o ip_vs_ctl.o ip_vs_sched.o	   \
+		ip_vs_xmit.o ip_vs_app.o ip_vs_sync.o	   		   \
+		ip_vs_est.o ip_vs_proto.o ip_vs_proto_icmp.o		   \
+		$(ip_vs_proto-objs-y)
+
+
+# IPVS core
+obj-$(CONFIG_IP_VS) += ip_vs.o
+
+# IPVS schedulers
+obj-$(CONFIG_IP_VS_RR) += ip_vs_rr.o
+obj-$(CONFIG_IP_VS_WRR) += ip_vs_wrr.o
+obj-$(CONFIG_IP_VS_LC) += ip_vs_lc.o
+obj-$(CONFIG_IP_VS_WLC) += ip_vs_wlc.o
+obj-$(CONFIG_IP_VS_LBLC) += ip_vs_lblc.o
+obj-$(CONFIG_IP_VS_LBLCR) += ip_vs_lblcr.o
+obj-$(CONFIG_IP_VS_DH) += ip_vs_dh.o
+obj-$(CONFIG_IP_VS_SH) += ip_vs_sh.o
+obj-$(CONFIG_IP_VS_SED) += ip_vs_sed.o
+obj-$(CONFIG_IP_VS_NQ) += ip_vs_nq.o
+
+# IPVS application helpers
+obj-$(CONFIG_IP_VS_FTP) += ip_vs_ftp.o

net/ipv4/ipvs/ip_vs_dh.c

+/*
+ * IPVS:        Destination Hashing scheduling module
+ *
+ * Version:     $Id: ip_vs_dh.c,v 1.5 2002/09/15 08:14:08 wensong Exp $
+ *
+ * Authors:     Wensong Zhang <wensong@gnuchina.org>
+ *
+ *              Inspired by the consistent hashing scheduler patch from
+ *              Thomas Proell <proellt@gmx.de>
+ *
+ *              This program is free software; you can redistribute it and/or
+ *              modify it under the terms of the GNU General Public License
+ *              as published by the Free Software Foundation; either version
+ *              2 of the License, or (at your option) any later version.
+ *
+ * Changes:
+ *
+ */
+
+/*
+ * The dh algorithm is to select server by the hash key of destination IP
+ * address. The pseudo code is as follows:
+ *
+ *       n <- servernode[dest_ip];
+ *       if (n is dead) OR
+ *          (n is overloaded) OR (n.weight <= 0) then
+ *                 return NULL;
+ *
+ *       return n;
+ *
+ * Notes that servernode is a 256-bucket hash table that maps the hash
+ * index derived from packet destination IP address to the current server
+ * array. If the dh scheduler is used in cache cluster, it is good to
+ * combine it with cache_bypass feature. When the statically assigned
+ * server is dead or overloaded, the load balancer can bypass the cache
+ * server and send requests to the original server directly.
+ *
+ */
+
+#include <linux/config.h>
+#include <linux/module.h>
+#include <linux/init.h>
+#include <linux/types.h>
+#include <linux/kernel.h>
+#include <linux/errno.h>
+
+#include <net/ip_vs.h>
+
+
+/*
+ *      IPVS DH bucket
+ */
+struct ip_vs_dh_bucket {
+	struct ip_vs_dest       *dest;          /* real server (cache) */
+};
+
+/*
+ *     for IPVS DH entry hash table
+ */
+#ifndef CONFIG_IP_VS_DH_TAB_BITS
+#define CONFIG_IP_VS_DH_TAB_BITS        8
+#endif
+#define IP_VS_DH_TAB_BITS               CONFIG_IP_VS_DH_TAB_BITS
+#define IP_VS_DH_TAB_SIZE               (1 << IP_VS_DH_TAB_BITS)
+#define IP_VS_DH_TAB_MASK               (IP_VS_DH_TAB_SIZE - 1)
+
+
+/*
+ *	Returns hash value for IPVS DH entry
+ */
+static inline unsigned ip_vs_dh_hashkey(__u32 addr)
+{
+	return (ntohl(addr)*2654435761UL) & IP_VS_DH_TAB_MASK;
+}
+
+
+/*
+ *      Get ip_vs_dest associated with supplied parameters.
+ */
+static inline struct ip_vs_dest *
+ip_vs_dh_get(struct ip_vs_dh_bucket *tbl, __u32 addr)
+{
+	return (tbl[ip_vs_dh_hashkey(addr)]).dest;
+}
+
+
+/*
+ *      Assign all the hash buckets of the specified table with the service.
+ */
+static int
+ip_vs_dh_assign(struct ip_vs_dh_bucket *tbl, struct ip_vs_service *svc)
+{
+	int i;
+	struct ip_vs_dh_bucket *b;
+	struct list_head *p;
+	struct ip_vs_dest *dest;
+
+	b = tbl;
+	p = &svc->destinations;
+	for (i=0; i<IP_VS_DH_TAB_SIZE; i++) {
+		if (list_empty(p)) {
+			b->dest = NULL;
+		} else {
+			if (p == &svc->destinations)
+				p = p->next;
+
+			dest = list_entry(p, struct ip_vs_dest, n_list);
+			atomic_inc(&dest->refcnt);
+			b->dest = dest;
+
+			p = p->next;
+		}
+		b++;
+	}
+	return 0;
+}
+
+
+/*
+ *      Flush all the hash buckets of the specified table.
+ */
+static void ip_vs_dh_flush(struct ip_vs_dh_bucket *tbl)
+{
+	int i;
+	struct ip_vs_dh_bucket *b;
+
+	b = tbl;
+	for (i=0; i<IP_VS_DH_TAB_SIZE; i++) {
+		if (b->dest) {
+			atomic_dec(&b->dest->refcnt);
+			b->dest = NULL;
+		}
+		b++;
+	}
+}
+
+
+static int ip_vs_dh_init_svc(struct ip_vs_service *svc)
+{
+	struct ip_vs_dh_bucket *tbl;
+
+	/* allocate the DH table for this service */
+	tbl = kmalloc(sizeof(struct ip_vs_dh_bucket)*IP_VS_DH_TAB_SIZE,
+		      GFP_ATOMIC);
+	if (tbl == NULL) {
+		IP_VS_ERR("ip_vs_dh_init_svc(): no memory\n");
+		return -ENOMEM;
+	}
+	svc->sched_data = tbl;
+	IP_VS_DBG(6, "DH hash table (memory=%dbytes) allocated for "
+		  "current service\n",
+		  sizeof(struct ip_vs_dh_bucket)*IP_VS_DH_TAB_SIZE);
+
+	/* assign the hash buckets with the updated service */
+	ip_vs_dh_assign(tbl, svc);
+
+	return 0;
+}
+
+
+static int ip_vs_dh_done_svc(struct ip_vs_service *svc)
+{
+	struct ip_vs_dh_bucket *tbl = svc->sched_data;
+
+	/* got to clean up hash buckets here */
+	ip_vs_dh_flush(tbl);
+
+	/* release the table itself */
+	kfree(svc->sched_data);
+	IP_VS_DBG(6, "DH hash table (memory=%dbytes) released\n",
+		  sizeof(struct ip_vs_dh_bucket)*IP_VS_DH_TAB_SIZE);
+
+	return 0;
+}
+
+
+static int ip_vs_dh_update_svc(struct ip_vs_service *svc)
+{
+	struct ip_vs_dh_bucket *tbl = svc->sched_data;
+
+	/* got to clean up hash buckets here */
+	ip_vs_dh_flush(tbl);
+
+	/* assign the hash buckets with the updated service */
+	ip_vs_dh_assign(tbl, svc);
+
+	return 0;
+}
+
+
+/*
+ *      If the dest flags is set with IP_VS_DEST_F_OVERLOAD,
+ *      consider that the server is overloaded here.
+ */
+static inline int is_overloaded(struct ip_vs_dest *dest)
+{
+	return dest->flags & IP_VS_DEST_F_OVERLOAD;
+}
+
+
+/*
+ *      Destination hashing scheduling
+ */
+static struct ip_vs_dest *
+ip_vs_dh_schedule(struct ip_vs_service *svc, struct iphdr *iph)
+{
+	struct ip_vs_dest *dest;
+	struct ip_vs_dh_bucket *tbl;
+
+	IP_VS_DBG(6, "ip_vs_dh_schedule(): Scheduling...\n");
+
+	tbl = (struct ip_vs_dh_bucket *)svc->sched_data;
+	dest = ip_vs_dh_get(tbl, iph->daddr);
+	if (!dest
+	    || !(dest->flags & IP_VS_DEST_F_AVAILABLE)
+	    || atomic_read(&dest->weight) <= 0
+	    || is_overloaded(dest)) {
+		return NULL;
+	}
+
+	IP_VS_DBG(6, "DH: destination IP address %u.%u.%u.%u "
+		  "--> server %u.%u.%u.%u:%d\n",
+		  NIPQUAD(iph->daddr),
+		  NIPQUAD(dest->addr),
+		  ntohs(dest->port));
+
+	return dest;
+}
+
+
+/*
+ *      IPVS DH Scheduler structure
+ */
+static struct ip_vs_scheduler ip_vs_dh_scheduler =
+{
+	.name =			"dh",
+	.refcnt =		ATOMIC_INIT(0),
+	.module =		THIS_MODULE,
+	.init_service =		ip_vs_dh_init_svc,
+	.done_service =		ip_vs_dh_done_svc,
+	.update_service =	ip_vs_dh_update_svc,
+	.schedule =		ip_vs_dh_schedule,
+};
+
+
+static int __init ip_vs_dh_init(void)
+{
+	INIT_LIST_HEAD(&ip_vs_dh_scheduler.n_list);
+	return register_ip_vs_scheduler(&ip_vs_dh_scheduler);
+}
+
+
+static void __exit ip_vs_dh_cleanup(void)
+{
+	unregister_ip_vs_scheduler(&ip_vs_dh_scheduler);
+}
+
+
+module_init(ip_vs_dh_init);
+module_exit(ip_vs_dh_cleanup);
+MODULE_LICENSE("GPL");

net/ipv4/ipvs/ip_vs_est.c

+/*
+ * ip_vs_est.c: simple rate estimator for IPVS
+ *
+ * Version:     $Id: ip_vs_est.c,v 1.4 2002/11/30 01:50:35 wensong Exp $
+ *
+ * Authors:     Wensong Zhang <wensong@linuxvirtualserver.org>
+ *
+ *              This program is free software; you can redistribute it and/or
+ *              modify it under the terms of the GNU General Public License
+ *              as published by the Free Software Foundation; either version
+ *              2 of the License, or (at your option) any later version.
+ *
+ * Changes:
+ *
+ */
+#include <linux/kernel.h>
+#include <linux/types.h>
+
+#include <net/ip_vs.h>
+
+/*
+  This code is to estimate rate in a shorter interval (such as 8
+  seconds) for virtual services and real servers. For measure rate in a
+  long interval, it is easy to implement a user level daemon which
+  periodically reads those statistical counters and measure rate.
+
+  Currently, the measurement is activated by slow timer handler. Hope
+  this measurement will not introduce too much load.
+
+  We measure rate during the last 8 seconds every 2 seconds:
+
+    avgrate = avgrate*(1-W) + rate*W
+
+    where W = 2^(-2)
+
+  NOTES.
+
+  * The stored value for average bps is scaled by 2^5, so that maximal
+    rate is ~2.15Gbits/s, average pps and cps are scaled by 2^10.
+
+  * A lot code is taken from net/sched/estimator.c
+ */
+
+
+struct ip_vs_estimator
+{
+	struct ip_vs_estimator	*next;
+	struct ip_vs_stats	*stats;
+
+	u32			last_conns;
+	u32			last_inpkts;
+	u32			last_outpkts;
+	u64			last_inbytes;
+	u64			last_outbytes;
+
+	u32			cps;
+	u32			inpps;
+	u32			outpps;
+	u32			inbps;
+	u32			outbps;
+};
+
+
+static struct ip_vs_estimator *est_list = NULL;
+static rwlock_t est_lock = RW_LOCK_UNLOCKED;
+static struct timer_list est_timer;
+
+static void estimation_timer(unsigned long arg)
+{
+	struct ip_vs_estimator *e;
+	struct ip_vs_stats *s;
+	u32 n_conns;
+	u32 n_inpkts, n_outpkts;
+	u64 n_inbytes, n_outbytes;
+	u32 rate;
+
+	read_lock(&est_lock);
+	for (e = est_list; e; e = e->next) {
+		s = e->stats;
+		n_conns = s->conns;
+		n_inpkts = s->inpkts;
+		n_outpkts = s->outpkts;
+		n_inbytes = s->inbytes;
+		n_outbytes = s->outbytes;
+
+		/* scaled by 2^10, but divided 2 seconds */
+		rate = (n_conns - e->last_conns)<<9;
+		e->last_conns = n_conns;
+		e->cps += ((long)rate - (long)e->cps)>>2;
+		s->cps = (e->cps+0x1FF)>>10;
+
+		rate = (n_inpkts - e->last_inpkts)<<9;
+		e->last_inpkts = n_inpkts;
+		e->inpps += ((long)rate - (long)e->inpps)>>2;
+		s->inpps = (e->inpps+0x1FF)>>10;
+
+		rate = (n_outpkts - e->last_outpkts)<<9;
+		e->last_outpkts = n_outpkts;
+		e->outpps += ((long)rate - (long)e->outpps)>>2;
+		s->outpps = (e->outpps+0x1FF)>>10;
+
+		rate = (n_inbytes - e->last_inbytes)<<4;
+		e->last_inbytes = n_inbytes;
+		e->inbps += ((long)rate - (long)e->inbps)>>2;
+		s->inbps = (e->inbps+0xF)>>5;
+
+		rate = (n_outbytes - e->last_outbytes)<<4;
+		e->last_outbytes = n_outbytes;
+		e->outbps += ((long)rate - (long)e->outbps)>>2;
+		s->outbps = (e->outbps+0xF)>>5;
+	}
+	read_unlock(&est_lock);
+	mod_timer(&est_timer, jiffies + 2*HZ);
+}
+
+int ip_vs_new_estimator(struct ip_vs_stats *stats)
+{
+	struct ip_vs_estimator *est;
+
+	est = kmalloc(sizeof(*est), GFP_KERNEL);
+	if (est == NULL)
+		return -ENOMEM;
+
+	memset(est, 0, sizeof(*est));
+	est->stats = stats;
+	est->last_conns = stats->conns;
+	est->cps = stats->cps<<10;
+
+	est->last_inpkts = stats->inpkts;
+	est->inpps = stats->inpps<<10;
+
+	est->last_outpkts = stats->outpkts;
+	est->outpps = stats->outpps<<10;
+
+	est->last_inbytes = stats->inbytes;
+	est->inbps = stats->inbps<<5;
+
+	est->last_outbytes = stats->outbytes;
+	est->outbps = stats->outbps<<5;
+
+	write_lock_bh(&est_lock);
+	est->next = est_list;
+	if (est->next == NULL) {
+		init_timer(&est_timer);
+		est_timer.expires = jiffies + 2*HZ;
+		est_timer.function = estimation_timer;
+		add_timer(&est_timer);
+	}
+	est_list = est;
+	write_unlock_bh(&est_lock);
+	return 0;
+}
+
+void ip_vs_kill_estimator(struct ip_vs_stats *stats)
+{
+	struct ip_vs_estimator *est, **pest;
+	int killed = 0;
+
+	write_lock_bh(&est_lock);
+	pest = &est_list;
+	while ((est=*pest) != NULL) {
+		if (est->stats != stats) {
+			pest = &est->next;
+			continue;
+		}
+		*pest = est->next;
+		kfree(est);
+		killed++;
+	}
+	if (killed && est_list == NULL)
+		del_timer_sync(&est_timer);
+	write_unlock_bh(&est_lock);
+}
+
+void ip_vs_zero_estimator(struct ip_vs_stats *stats)
+{
+	struct ip_vs_estimator *e;
+
+	write_lock_bh(&est_lock);
+	for (e = est_list; e; e = e->next) {
+		if (e->stats != stats)
+			continue;
+
+		/* set counters zero */
+		e->last_conns = 0;
+		e->last_inpkts = 0;
+		e->last_outpkts = 0;
+		e->last_inbytes = 0;
+		e->last_outbytes = 0;
+		e->cps = 0;
+		e->inpps = 0;
+		e->outpps = 0;
+		e->inbps = 0;
+		e->outbps = 0;
+	}
+	write_unlock_bh(&est_lock);
+}

net/ipv4/netfilter/ip_fw_compat.c

@@ -19,6 +19,12 @@ struct notifier_block;
 
 static struct firewall_ops *fwops;
 
+#ifdef CONFIG_IP_VS
+/* From ip_vs_core.c */
+extern unsigned int
+check_for_ip_vs_out(struct sk_buff **skb_p, int (*okfn)(struct sk_buff *));
+#endif
+
 /* They call these; we do what they want. */
 int register_firewall(int pf, struct firewall_ops *fw)
 {
@@ -134,8 +140,14 @@ fw_in(unsigned int hooknum,
 		return NF_ACCEPT;
 
 	case FW_MASQUERADE:
-		if (hooknum == NF_IP_FORWARD)
+		if (hooknum == NF_IP_FORWARD) {
+#ifdef CONFIG_IP_VS
+			/* check if it is for ip_vs */
+			if (check_for_ip_vs_out(pskb, okfn) == NF_STOLEN)
+				return NF_STOLEN;
+#endif
 			return do_masquerade(pskb, out);
+		}
 		else return NF_ACCEPT;
 
 	case FW_REDIRECT:

net/ipv6/ip6_fib.c

@@ -496,6 +496,12 @@ static __inline__ void fib6_start_gc(struct rt6_info *rt)
 		mod_timer(&ip6_fib_timer, jiffies + ip6_rt_gc_interval);
 }
 
+void fib6_force_start_gc(void)
+{
+	if (ip6_fib_timer.expires == 0)
+		mod_timer(&ip6_fib_timer, jiffies + ip6_rt_gc_interval);
+}
+
 /*
  *	Add routing information to the routing tree.
  *	<destination addr>/<source addr>
@@ -1214,6 +1220,7 @@ void fib6_run_gc(unsigned long dummy)
 
 
 	write_lock_bh(&rt6_lock);
+	ndisc_dst_gc(&gc_args.more);
 	fib6_clean_tree(&ip6_routing_table, fib6_age, 0, NULL);
 	write_unlock_bh(&rt6_lock);
 

net/netlink/af_netlink.c

@@ -430,6 +430,10 @@ int netlink_unicast(struct sock *ssk, struct sk_buff *skb, u32 pid, int nonblock
 		goto no_dst;
 	nlk = nlk_sk(sk);
 
+	/* Don't bother queuing skb if kernel socket has no input function */
+        if (nlk->pid == 0 && !nlk->data_ready)
+        	goto no_dst;
+
 #ifdef NL_EMULATE_DEV
 	if (nlk->handler) {
 		skb_orphan(skb);

net/netsyms.c

@@ -265,6 +265,7 @@ EXPORT_SYMBOL(inet_family_ops);
 EXPORT_SYMBOL(in_aton);
 EXPORT_SYMBOL(ip_mc_inc_group);
 EXPORT_SYMBOL(ip_mc_dec_group);
+EXPORT_SYMBOL(ip_mc_join_group);
 EXPORT_SYMBOL(ip_finish_output);
 EXPORT_SYMBOL(inet_stream_ops);
 EXPORT_SYMBOL(inet_dgram_ops);

net/xfrm/xfrm_policy.c

@@ -855,6 +855,7 @@ xfrm_state_ok(struct xfrm_tmpl *tmpl, struct xfrm_state *x,
 {
 	return	x->id.proto == tmpl->id.proto &&
 		(x->id.spi == tmpl->id.spi || !tmpl->id.spi) &&
+		(x->props.reqid == tmpl->reqid || !tmpl->reqid) &&
 		x->props.mode == tmpl->mode &&
 		(tmpl->aalgos & (1<<x->props.aalgo)) &&
 		!(x->props.mode && xfrm_state_addr_cmp(tmpl, x, family));