Merge bk://kernel.bkbits.net/davem/net-2.5

into home.transmeta.com:/home/torvalds/v2.5/linux

Documentation/networking/e100.txt

 Linux* Base Driver for the Intel(R) PRO/100 Family of Adapters
 ==============================================================
 
-September 16, 2002
+November 19, 2002
 
 
 Contents
@@ -19,7 +19,7 @@ In This Release
 ===============
 
 This file describes the Linux* Base Driver for the Intel(R) PRO/100 Family of
-Adapters, version 2.1.x.  This driver includes support for Itanium(TM)-based 
+Adapters, version 2.2.x.  This driver includes support for Itanium(TM)-based 
 systems.
 
 
@@ -94,8 +94,9 @@ For the latest Intel PRO/100 network driver for Linux, see:
 Command Line Parameters
 =======================
 
-The following optional parameters are used by entering them on the command 
-line with the modprobe or insmod command using this syntax:
+If the driver is built as a module, the  following optional parameters are 
+used by entering them on the command line with the modprobe or insmod command
+using this syntax:
 
      modprobe e100 [<option>=<VAL1>,<VAL2>,...]
 

Documentation/networking/e1000.txt

 Linux* Base Driver for the Intel(R) PRO/1000 Family of Adapters
 ===============================================================
 
-October 12, 2002
+January 8, 2003
 
 
 Contents
@@ -20,27 +20,11 @@ In This Release
 ===============
 
 This file describes the Linux* Base Driver for the Intel(R) PRO/1000 Family
-of Adapters, version 4.4.x.  This driver includes support for 
+of Adapters, version 5.0.x.  This driver includes support for 
 Itanium(TM)-based systems.
 
-This release version includes the following:
-
-   - Support for the ethtool 1.6 interface. A third-party application can use
-     the ethtool interface to get and set driver parameters.
-
-   - Zero copy. This feature provides faster data throughput. Enabled by
-     default in supporting kernels. It is not supported on the Intel(R)
-     PRO/1000 Gigabit Server Adapter.
-
-Features include:
-
-   - Support for the 82545 and 82546-based adapters listed below
-
-   - Wake on LAN* support via ethtool for 82540, 82544, 82545, and 82546-
-     based adapters
-
-   - Adaptive IFS for increased performance at half duplex
 
+Native VLANs are now available with supported kernels.
 
 
 Supported Adapters
@@ -65,6 +49,7 @@ release:
    82544       PRO/1000 T Desktop Adapter           A62947-xxx
 
    82540       PRO/1000 MT Desktop Adapter          A78408-xxx
+   82541                                            C91016-xxx
 
    82545       PRO/1000 MT Server Adapter           A92165-xxx
 
@@ -77,26 +62,32 @@ release:
    82546       PRO/1000 MF Dual Port Server Adapter A91620-xxx
 
 
+
 To verify your Intel adapter is supported, find the board ID number on the 
-adapter. Look for a label that has a barcode and a number in the format of
-123456-001 (six digits hyphen three digits). Match this to the list of
-numbers above.
+adapter. Look for a label that has a barcode and a number in the format  
+A12345-001. Match this to the list of numbers above.
 
 For more information on how to identify your adapter, go to the Adapter & 
 Driver ID Guide at:
 
     http://support.intel.com/support/network/adapter/pro100/21397.htm
 
-For the latest Intel network drivers for Linux, go to:
+For the latest Intel network drivers for Linux, refer to the following
 
-    http://appsr.intel.com/scripts-df/support_intel.asp
+    http://downloadfinder.intel.com/scripts-df/support_intel.asp
 
 
 Command Line Parameters
 =======================
 
 If the driver is built as a module, the  following optional parameters are 
-used by entering them on the command line with the modprobe or insmod command. 
+used by entering them on the command line with the modprobe or insmod command
+using this syntax:
+
+     modprobe e1000 [<option>=<VAL1>,<VAL2>,...]
+
+     insmod e1000 [<option>=<VAL1>,<VAL2>,...] 
+
 For example, with two PRO/1000 PCI adapters, entering:
 
      insmod e1000 TxDescriptors=80,128
@@ -104,6 +95,9 @@ For example, with two PRO/1000 PCI adapters, entering:
 loads the e1000 driver with 80 TX resources for the first adapter and 128 TX 
 resources for the second adapter.
 
+The default value for each parameter is generally the recommended setting,
+unless otherwise noted.
+
 For more information about the AutoNeg, Duplex, and Speed parameters, see the
 "Speed and Duplex Configuration" section in this document.
 
@@ -129,6 +123,20 @@ Default: Read flow control settings from the EEPROM
     This parameter controls the automatic generation(Tx) and response(Rx) to 
     Ethernet PAUSE frames.
 
+InterruptThrottleRate
+Valid Range: 100-100000 (0=off, 1=dynamic)
+Default Value: 1
+    This value represents the maximum number of interrupts per second the 
+    controller generates. InterruptThrottleRate is another setting used in 
+    interrupt moderation. Dynamic mode uses a heuristic algorithm to adjust 
+    InterruptThrottleRate based on the current traffic load.
+
+    NOTE: InterruptThrottleRate takes precedence over the TxAbsIntDelay and 
+          RxAbsIntDelay parameters. In other words, minimizing the receive 
+          and/or transmit absolute delays does not force the controller to 
+          generate more interrupts than what the Interrupt Throttle Rate 
+          allows.
+
 RxDescriptors
 Valid Range: 80-256 for 82542 and 82543-based adapters
              80-4096 for 82540, 82544, 82545, and 82546-based adapters
@@ -176,7 +184,7 @@ Default Value: 0 (auto-negotiate at all supported speeds)
     Speed forces the line speed to the specified value in megabits per second
     (Mbps). If this parameter is not specified or is set to 0 and the link 
     partner is set to auto-negotiate, the board will auto-detect the correct 
-    speed. Duplex must also be set when Speed is set to either 10 or 100.
+    speed. Duplex should also be set when Speed is set to either 10 or 100.
 
 TxDescriptors
 Valid Range: 80-256 for 82542 and 82543-based adapters
@@ -205,7 +213,7 @@ Default Value: 64
     along with TxIntDelay, may improve traffic throughput in specific 
     network conditions.
 
-XsumRX (not available on the PRO/1000 Gigabit Server Adapter)
+XsumRX (not available on the 82542-based adapter)
 Valid Range: 0-1
 Default Value: 1
     A value of '1' indicates that the driver should enable IP checksum
@@ -230,8 +238,8 @@ For copper-based boards, the keywords interact as follows:
   If Speed = 1000, limited auto-negotiation is enabled and only 1000 Mbps is
   advertised (The 1000BaseT spec requires auto-negotiation.)
 
-  If Speed = 10 or 100, then both Speed and Duplex must be set. Auto-
-  negotiation is disabled, and the AutoNeg parameter is ignored. Partner MUST
+  If Speed = 10 or 100, then both Speed and Duplex should be set. Auto-
+  negotiation is disabled, and the AutoNeg parameter is ignored. Partner SHOULD
   also be forced.
 
 The AutoNeg parameter is used when more control is required over the auto-
@@ -263,6 +271,12 @@ Additional Configurations
 
         ifconfig ethx mtu 9000 up
 
+  The maximum MTU setting for Jumbo Frames is 16110. This value coincides 
+  with the maximum Jumbo Frames size of 16128.
+
+  NOTE: Jumbo Frames are supported at 1000 Mbps only. Using Jumbo Frames at 
+  10 or 100 Mbps may result in poor performance or loss of link.
+
 
 Known Issues
 ============
@@ -279,7 +293,7 @@ Known Issues
 Support
 =======
 
-For general information and support, go to the Intel support website at:
+For general information, go to the Intel support website at:
 
     http://support.intel.com
 

MAINTAINERS

@@ -923,10 +923,11 @@ M:	scott.feldman@intel.com
 S:	Supported
 
 INTEL PRO/1000 GIGABIT ETHERNET SUPPORT
-P:	Chris Leech
-M:	christopher.leech@intel.com
+P:	Jeb Cramer
+M:	cramerj@intel.com
 P:	Scott Feldman
 M:	scott.feldman@intel.com
+W:	http://sourceforge.net/projects/e1000/
 S:	Supported
 
 INTERMEZZO FILE SYSTEM

arch/i386/kernel/entry.S

@@ -73,10 +73,10 @@ NT_MASK		= 0x00004000
 VM_MASK		= 0x00020000
 
 /*
- * ESP0 is at offset 4. 0x100 is the size of the TSS, and
+ * ESP0 is at offset 4. 0x200 is the size of the TSS, and
  * also thus the top-of-stack pointer offset of SYSENTER_ESP
  */
-TSS_ESP0_OFFSET = (4 - 0x100)
+TSS_ESP0_OFFSET = (4 - 0x200)
 
 #ifdef CONFIG_PREEMPT
 #define preempt_stop		cli
@@ -479,23 +479,32 @@ device_not_available_emulate:
  * by hand onto the new stack - while updating the return eip past
  * the instruction that would have done it for sysenter.
  */
-#define CHECK_SYSENTER_EIP			\
-	cmpl $sysenter_entry,(%esp);		\
-	jne 1f;					\
-	movl TSS_ESP0_OFFSET+12(%esp),%esp;	\
+#define FIX_STACK(offset, ok, label)		\
+	cmpw $__KERNEL_CS,4(%esp);		\
+	jne ok;					\
+label:						\
+	movl TSS_ESP0_OFFSET+offset(%esp),%esp;	\
 	pushfl;					\
 	pushl $__KERNEL_CS;			\
-	pushl $sysenter_past_esp;		\
-1:
+	pushl $sysenter_past_esp
 
 ENTRY(debug)
-	CHECK_SYSENTER_EIP
+	cmpl $sysenter_entry,(%esp)
+	jne debug_stack_correct
+	FIX_STACK(12, debug_stack_correct, debug_esp_fix_insn)
+debug_stack_correct:
 	pushl $0
 	pushl $do_debug
 	jmp error_code
 
 ENTRY(nmi)
-	CHECK_SYSENTER_EIP
+	cmpl $sysenter_entry,(%esp)
+	je nmi_stack_fixup
+	cmpl $debug - 1,(%esp)
+	jle nmi_stack_correct
+	cmpl $debug_esp_fix_insn,(%esp)
+	jle nmi_debug_stack_fixup
+nmi_stack_correct:
 	pushl %eax
 	SAVE_ALL
 	movl %esp, %edx
@@ -505,6 +514,13 @@ ENTRY(nmi)
 	addl $8, %esp
 	RESTORE_ALL
 
+nmi_stack_fixup:
+	FIX_STACK(12,nmi_stack_correct, 1)
+	jmp nmi_stack_correct
+nmi_debug_stack_fixup:
+	FIX_STACK(24,nmi_stack_correct, 1)
+	jmp nmi_stack_correct
+
 ENTRY(int3)
 	pushl $0
 	pushl $do_int3

arch/sparc/kernel/traps.c

@@ -89,6 +89,7 @@ void instruction_dump (unsigned long *pc)
 
 void die_if_kernel(char *str, struct pt_regs *regs)
 {
+	static int die_counter;
 	int count = 0;
 
 	/* Amuse the user. */
@@ -98,7 +99,7 @@ void die_if_kernel(char *str, struct pt_regs *regs)
 "              /_| \\__/ |_\\\n"
 "                 \\__U_/\n");
 
-	printk("%s(%d): %s\n", current->comm, current->pid, str);
+	printk("%s(%d): %s [#%d]\n", current->comm, current->pid, str, ++die_counter);
 	show_regs(regs);
 
 	__SAVE; __SAVE; __SAVE; __SAVE;

arch/sparc64/kernel/head.S

@@ -671,7 +671,9 @@ set_worklist:
 #endif
 
 	/* Kill PROM timer */
-	wr	%g0, 0, %tick_cmpr
+	sethi	%hi(0x80000000), %g1
+	sllx	%g1, 32, %g1
+	wr	%g1, 0, %tick_cmpr
 
 	BRANCH_IF_ANY_CHEETAH(g1,g5,1f)
 

arch/sparc64/kernel/irq.c

@@ -719,12 +719,8 @@ void handler_irq(int irq, struct pt_regs *regs)
 	 */
 	{
 		unsigned long clr_mask = 1 << irq;
-		unsigned long tick_mask;
+		unsigned long tick_mask = tick_ops->softint_mask;
 
-		if (SPARC64_USE_STICK)
-			tick_mask = (1UL << 16);
-		else
-			tick_mask = (1UL << 0);
 		if ((irq == 14) && (get_softint() & tick_mask)) {
 			irq = 0;
 			clr_mask = tick_mask;
@@ -946,113 +942,6 @@ int probe_irq_off(unsigned long mask)
 	return 0;
 }
 
-/* This is gets the master TICK_INT timer going. */
-void sparc64_init_timers(void (*cfunc)(int, void *, struct pt_regs *),
-			 unsigned long *clock)
-{
-	unsigned long pstate;
-	extern unsigned long timer_tick_offset;
-	int node, err;
-#ifdef CONFIG_SMP
-	extern void smp_tick_init(void);
-#endif
-
-	if (!SPARC64_USE_STICK) {
-		node = linux_cpus[0].prom_node;
-		*clock = prom_getint(node, "clock-frequency");
-	} else {
-		node = prom_root_node;
-		*clock = prom_getint(node, "stick-frequency");
-	}
-	timer_tick_offset = *clock / HZ;
-#ifdef CONFIG_SMP
-	smp_tick_init();
-#endif
-
-	/* Register IRQ handler. */
-	err = request_irq(build_irq(0, 0, 0UL, 0UL), cfunc, SA_STATIC_ALLOC,
-			  "timer", NULL);
-
-	if (err) {
-		prom_printf("Serious problem, cannot register TICK_INT\n");
-		prom_halt();
-	}
-
-	/* Guarentee that the following sequences execute
-	 * uninterrupted.
-	 */
-	__asm__ __volatile__("rdpr	%%pstate, %0\n\t"
-			     "wrpr	%0, %1, %%pstate"
-			     : "=r" (pstate)
-			     : "i" (PSTATE_IE));
-
-	/* Set things up so user can access tick register for profiling
-	 * purposes.  Also workaround BB_ERRATA_1 by doing a dummy
-	 * read back of %tick after writing it.
-	 */
-	__asm__ __volatile__(
-	"	sethi	%%hi(0x80000000), %%g1\n"
-	"	ba,pt	%%xcc, 1f\n"
-	"	 sllx	%%g1, 32, %%g1\n"
-	"	.align	64\n"
-	"1:	rd	%%tick, %%g2\n"
-	"	add	%%g2, 6, %%g2\n"
-	"	andn	%%g2, %%g1, %%g2\n"
-	"	wrpr	%%g2, 0, %%tick\n"
-	"	rdpr	%%tick, %%g0"
-	: /* no outputs */
-	: /* no inputs */
-	: "g1", "g2");
-
-	/* Workaround for Spitfire Errata (#54 I think??), I discovered
-	 * this via Sun BugID 4008234, mentioned in Solaris-2.5.1 patch
-	 * number 103640.
-	 *
-	 * On Blackbird writes to %tick_cmpr can fail, the
-	 * workaround seems to be to execute the wr instruction
-	 * at the start of an I-cache line, and perform a dummy
-	 * read back from %tick_cmpr right after writing to it. -DaveM
-	 */
-	if (!SPARC64_USE_STICK) {
-	__asm__ __volatile__(
-	"	rd	%%tick, %%g1\n"
-	"	ba,pt	%%xcc, 1f\n"
-	"	 add	%%g1, %0, %%g1\n"
-	"	.align	64\n"
-	"1:	wr	%%g1, 0x0, %%tick_cmpr\n"
-	"	rd	%%tick_cmpr, %%g0"
-	: /* no outputs */
-	: "r" (timer_tick_offset)
-	: "g1");
-	} else {
-	/* Let the user get at STICK too. */
-	__asm__ __volatile__(
-	"	sethi	%%hi(0x80000000), %%g1\n"
-	"	sllx	%%g1, 32, %%g1\n"
-	"	rd	%%asr24, %%g2\n"
-	"	andn	%%g2, %%g1, %%g2\n"
-	"	wr	%%g2, 0, %%asr24"
-	: /* no outputs */
-	: /* no inputs */
-	: "g1", "g2");
-
-	__asm__ __volatile__(
-	"	rd	%%asr24, %%g1\n"
-	"	add	%%g1, %0, %%g1\n"
-	"	wr	%%g1, 0x0, %%asr25"
-	: /* no outputs */
-	: "r" (timer_tick_offset)
-	: "g1");
-	}
-
-	/* Restore PSTATE_IE. */
-	__asm__ __volatile__("wrpr	%0, 0x0, %%pstate"
-			     : /* no outputs */
-			     : "r" (pstate));
-
-	local_irq_enable();
-}
-
 #ifdef CONFIG_SMP
 static int retarget_one_irq(struct irqaction *p, int goal_cpu)
 {

arch/sparc64/kernel/sparc64_ksyms.c

@@ -128,20 +128,13 @@ EXPORT_SYMBOL(__write_unlock);
 #endif
 
 /* Hard IRQ locking */
-#ifdef CONFIG_SMP
 EXPORT_SYMBOL(synchronize_irq);
-#endif
 
 #if defined(CONFIG_MCOUNT)
 extern void mcount(void);
 EXPORT_SYMBOL_NOVERS(mcount);
 #endif
 
-/* Uniprocessor clock frequency */
-#ifndef CONFIG_SMP
-EXPORT_SYMBOL(up_clock_tick);
-#endif
-
 /* Per-CPU information table */
 EXPORT_SYMBOL(cpu_data);
 
@@ -162,10 +155,13 @@ EXPORT_SYMBOL(_do_write_lock);
 EXPORT_SYMBOL(_do_write_unlock);
 #endif
 
-#ifdef CONFIG_SMP
 EXPORT_SYMBOL(smp_call_function);
-#endif
+#endif /* CONFIG_SMP */
 
+/* Uniprocessor clock frequency */
+#ifndef CONFIG_SMP
+extern unsigned long up_clock_tick;
+EXPORT_SYMBOL(up_clock_tick);
 #endif
 
 /* semaphores */

arch/sparc64/kernel/trampoline.S

@@ -85,7 +85,10 @@ spitfire_startup:
 
 startup_continue:
 	wrpr		%g0, 15, %pil
-	wr		%g0, 0, %tick_cmpr
+
+	sethi		%hi(0x80000000), %g2
+	sllx		%g2, 32, %g2
+	wr		%g2, 0, %tick_cmpr
 
 	/* Call OBP by hand to lock KERNBASE into i/d tlbs. */
 	mov		%o0, %l0

arch/sparc64/kernel/traps.c

@@ -33,6 +33,7 @@
 #include <asm/chafsr.h>
 #include <asm/psrcompat.h>
 #include <asm/processor.h>
+#include <asm/timer.h>
 #ifdef CONFIG_KMOD
 #include <linux/kmod.h>
 #endif
@@ -588,7 +589,7 @@ unsigned long __init cheetah_tune_scheduling(void)
 	flush_linesize = ecache_flush_linesize;
 	flush_size = ecache_flush_size >> 1;
 
-	__asm__ __volatile__("rd %%tick, %0" : "=r" (tick1));
+	tick1 = tick_ops->get_tick();
 
 	__asm__ __volatile__("1: subcc	%0, %4, %0\n\t"
 			     "   bne,pt	%%xcc, 1b\n\t"
@@ -597,7 +598,7 @@ unsigned long __init cheetah_tune_scheduling(void)
 			     : "0" (flush_size), "r" (flush_base),
 			       "i" (ASI_PHYS_USE_EC), "r" (flush_linesize));
 
-	__asm__ __volatile__("rd %%tick, %0" : "=r" (tick2));
+	tick2 = tick_ops->get_tick();
 
 	raw = (tick2 - tick1);
 
@@ -1598,6 +1599,7 @@ void show_trace_task(struct task_struct *tsk)
 
 void die_if_kernel(char *str, struct pt_regs *regs)
 {
+	static int die_counter;
 	extern void __show_regs(struct pt_regs * regs);
 	extern void smp_report_regs(void);
 	int count = 0;
@@ -1610,7 +1612,7 @@ void die_if_kernel(char *str, struct pt_regs *regs)
 "              /_| \\__/ |_\\\n"
 "                 \\__U_/\n");
 
-	printk("%s(%d): %s\n", current->comm, current->pid, str);
+	printk("%s(%d): %s [#%d]\n", current->comm, current->pid, str, ++die_counter);
 	__asm__ __volatile__("flushw");
 	__show_regs(regs);
 	if (regs->tstate & TSTATE_PRIV) {

arch/sparc64/mm/ultra.S

@@ -560,8 +560,8 @@ xcall_flush_tlb_kernel_range:
 	/* This runs in a very controlled environment, so we do
 	 * not need to worry about BH races etc.
 	 */
-	.globl		xcall_sync_stick
-xcall_sync_stick:
+	.globl		xcall_sync_tick
+xcall_sync_tick:
 	rdpr		%pstate, %g2
 	wrpr		%g2, PSTATE_IG | PSTATE_AG, %pstate
 	rdpr		%pil, %g2
@@ -569,7 +569,7 @@ xcall_sync_stick:
 	sethi		%hi(109f), %g7
 	b,pt		%xcc, etrap_irq
 109:	 or		%g7, %lo(109b), %g7
-	call		smp_synchronize_stick_client
+	call		smp_synchronize_tick_client
 	 nop
 	clr		%l6
 	b		rtrap_xcall

drivers/net/3c509.c

@@ -56,6 +56,10 @@
 		v1.19b 08Nov2002 Marc Zyngier <maz@wild-wind.fr.eu.org>
 		    - Introduce driver model for EISA cards.
 */
+/*
+  FIXES for PC-9800:
+  Shu Iwanaga: 3c569B(PC-9801 C-bus) support
+*/
 
 #define DRV_NAME	"3c509"
 #define DRV_VERSION	"1.19b"
@@ -257,7 +261,7 @@ static struct mca_driver el3_mca_driver = {
 };
 #endif /* CONFIG_MCA */
 
-#ifdef __ISAPNP__
+#if defined(__ISAPNP__) && !defined(CONFIG_X86_PC9800)
 static struct isapnp_device_id el3_isapnp_adapters[] __initdata = {
 	{	ISAPNP_ANY_ID, ISAPNP_ANY_ID,
 		ISAPNP_VENDOR('T', 'C', 'M'), ISAPNP_FUNCTION(0x5090),
@@ -350,7 +354,7 @@ static void el3_common_remove (struct net_device *dev)
 		if (lp->pmdev)
 			pm_unregister(lp->pmdev);
 #endif
-#ifdef __ISAPNP__
+#if defined(__ISAPNP__) && !defined(CONFIG_X86_PC9800)
 		if (lp->type == EL3_PNP)
 			pnp_device_detach(to_pnp_dev(lp->dev));
 #endif
@@ -368,12 +372,12 @@ static int __init el3_probe(int card_idx)
 	int ioaddr, irq, if_port;
 	u16 phys_addr[3];
 	static int current_tag;
-#ifdef __ISAPNP__
+#if defined(__ISAPNP__) && !defined(CONFIG_X86_PC9800)
 	static int pnp_cards;
 	struct pnp_dev *idev = NULL;
 #endif /* __ISAPNP__ */
 
-#ifdef __ISAPNP__
+#if defined(__ISAPNP__) && !defined(CONFIG_X86_PC9800)
 	if (nopnp == 1)
 		goto no_pnp;
 
@@ -421,6 +425,9 @@ static int __init el3_probe(int card_idx)
 no_pnp:
 #endif /* __ISAPNP__ */
 
+#ifdef CONFIG_X86_PC9800
+	id_port = 0x71d0;
+#else
 	/* Select an open I/O location at 0x1*0 to do contention select. */
 	for ( ; id_port < 0x200; id_port += 0x10) {
 		if (check_region(id_port, 1))
@@ -435,6 +442,7 @@ static int __init el3_probe(int card_idx)
 		printk(" WARNING: No I/O port available for 3c509 activation.\n");
 		return -ENODEV;
 	}
+#endif /* CONFIG_X86_PC9800 */
 	/* Next check for all ISA bus boards by sending the ID sequence to the
 	   ID_PORT.  We find cards past the first by setting the 'current_tag'
 	   on cards as they are found.  Cards with their tag set will not
@@ -465,7 +473,7 @@ static int __init el3_probe(int card_idx)
 		phys_addr[i] = htons(id_read_eeprom(i));
 	}
 
-#ifdef __ISAPNP__
+#if defined(__ISAPNP__) && !defined(CONFIG_X86_PC9800)
 	if (nopnp == 0) {
 		/* The ISA PnP 3c509 cards respond to the ID sequence.
 		   This check is needed in order not to register them twice. */
@@ -490,9 +498,19 @@ static int __init el3_probe(int card_idx)
 	{
 		unsigned int iobase = id_read_eeprom(8);
 		if_port = iobase >> 14;
+#ifdef CONFIG_X86_PC9800
+		ioaddr = 0x40d0 + ((iobase & 0x1f) << 8);
+#else
 		ioaddr = 0x200 + ((iobase & 0x1f) << 4);
+#endif
 	}
 	irq = id_read_eeprom(9) >> 12;
+#ifdef CONFIG_X86_PC9800
+	if (irq == 7)
+		irq = 6;
+	else if (irq == 15)
+		irq = 13;
+#endif
 
 	if (!(dev = init_etherdev(NULL, sizeof(struct el3_private))))
 			return -ENOMEM;
@@ -522,7 +540,11 @@ static int __init el3_probe(int card_idx)
 	outb(0xd0 + ++current_tag, id_port);
 
 	/* Activate the adaptor at the EEPROM location. */
+#ifdef CONFIG_X86_PC9800
+	outb((ioaddr >> 8) | 0xe0, id_port);
+#else
 	outb((ioaddr >> 4) | 0xe0, id_port);
+#endif
 
 	EL3WINDOW(0);
 	if (inw(ioaddr) != 0x6d50) {
@@ -534,7 +556,7 @@ static int __init el3_probe(int card_idx)
 	/* Free the interrupt so that some other card can use it. */
 	outw(0x0f00, ioaddr + WN0_IRQ);
 
-#ifdef __ISAPNP__	
+#if defined(__ISAPNP__) && !defined(CONFIG_X86_PC9800)
  found:							/* PNP jumps here... */
 #endif /* __ISAPNP__ */
 
@@ -543,7 +565,7 @@ static int __init el3_probe(int card_idx)
 	dev->irq = irq;
 	dev->if_port = if_port;
 	lp = dev->priv;
-#ifdef __ISAPNP__
+#if defined(__ISAPNP__) && !defined(CONFIG_X86_PC9800)
 	lp->dev = &idev->dev;
 #endif
 
@@ -1388,6 +1410,12 @@ el3_up(struct net_device *dev)
 	outw(0x0001, ioaddr + 4);
 
 	/* Set the IRQ line. */
+#ifdef CONFIG_X86_PC9800
+	if (dev->irq == 6)
+		dev->irq = 7;
+	else if (dev->irq == 13)
+		dev->irq = 15;
+#endif
 	outw((dev->irq << 12) | 0x0f00, ioaddr + WN0_IRQ);
 
 	/* Set the station address in window 2 each time opened. */
@@ -1550,7 +1578,7 @@ MODULE_PARM_DESC(debug, "debug level (0-6)");
 MODULE_PARM_DESC(irq, "IRQ number(s) (assigned)");
 MODULE_PARM_DESC(xcvr,"transceiver(s) (0=internal, 1=external)");
 MODULE_PARM_DESC(max_interrupt_work, "maximum events handled per interrupt");
-#ifdef __ISAPNP__
+#if defined(__ISAPNP__) && !defined(CONFIG_X86_PC9800)
 MODULE_PARM(nopnp, "i");
 MODULE_PARM_DESC(nopnp, "disable ISA PnP support (0-1)");
 MODULE_DEVICE_TABLE(isapnp, el3_isapnp_adapters);

drivers/net/8390.h

@@ -123,7 +123,8 @@ struct ei_device {
 #define inb_p(port)   in_8(port)
 #define outb_p(val,port)  out_8(port,val)
 
-#elif defined(CONFIG_ARM_ETHERH) || defined(CONFIG_ARM_ETHERH_MODULE)
+#elif defined(CONFIG_ARM_ETHERH) || defined(CONFIG_ARM_ETHERH_MODULE) || \
+      defined(CONFIG_NET_CBUS)
 #define EI_SHIFT(x)	(ei_local->reg_offset[x])
 #else
 #define EI_SHIFT(x)	(x)

drivers/net/Kconfig

@@ -663,7 +663,7 @@ config EL16
 	  as <file:Documentation/networking/net-modules.txt>.
 
 config EL3
-	tristate "3c509/3c529 (MCA)/3c579 \"EtherLink III\" support"
+	tristate "3c509/3c529 (MCA)/3c569B (98)/3c579 \"EtherLink III\" support"
 	depends on NET_VENDOR_3COM && (ISA || EISA || MCA)
 	---help---
 	  If you have a network (Ethernet) card belonging to the 3Com
@@ -932,7 +932,7 @@ config NI65
 source "drivers/net/tulip/Kconfig"
 
 config AT1700
-	tristate "AT1700/1720 support (EXPERIMENTAL)"
+	tristate "AT1700/1720/RE1000Plus(C-Bus) support (EXPERIMENTAL)"
 	depends on NET_ETHERNET && (ISA || MCA) && EXPERIMENTAL
 	---help---
 	  If you have a network (Ethernet) card of this type, say Y and read
@@ -978,7 +978,7 @@ config HP100
 
 config NET_ISA
 	bool "Other ISA cards"
-	depends on NET_ETHERNET && ISA
+	depends on NET_ETHERNET && ISA && !X86_PC9800
 	---help---
 	  If your network (Ethernet) card hasn't been mentioned yet and its
 	  bus system (that's the way the cards talks to the other components
@@ -1176,6 +1176,55 @@ config SK_G16
 	  the Ethernet-HOWTO, available from
 	  <http://www.linuxdoc.org/docs.html#howto>.
 
+config NET_CBUS
+	bool "NEC PC-9800 C-bus cards"
+	depends on NET_ETHERNET && ISA && X86_PC9800
+	---help---
+	  If your network (Ethernet) card hasn't been mentioned yet and its
+	  bus system (that's the way the cards talks to the other components
+	  of your computer) is NEC PC-9800 C-Bus, say Y.
+
+config NE2K_CBUS
+	tristate "Most NE2000-based Ethernet support"
+	depends on NET_CBUS
+
+config NE2K_CBUS_EGY98
+	bool "Melco EGY-98 support"
+	depends on NE2K_CBUS
+
+config NE2K_CBUS_LGY98
+	bool "Melco LGY-98 support"
+	depends on NE2K_CBUS
+
+config NE2K_CBUS_ICM
+	bool "ICM IF-27xxET support"
+	depends on NE2K_CBUS
+
+config NE2K_CBUS_IOLA98
+	bool "I-O DATA LA-98 support"
+	depends on NE2K_CBUS
+
+config NE2K_CBUS_CNET98EL
+	bool "Contec C-NET(98)E/L support"
+	depends on NE2K_CBUS
+
+config NE2K_CBUS_CNET98EL_IO_BASE
+	hex "C-NET(98)E/L I/O base address (0xaaed or 0x55ed)"
+	depends on NE2K_CBUS_CNET98EL
+	default "0xaaed"
+
+config NE2K_CBUS_ATLA98
+	bool "Allied Telesis LA-98 Support"
+	depends on NE2K_CBUS
+
+config NE2K_CBUS_BDN
+	bool "ELECOM Laneed LD-BDN[123]A Support"
+	depends on NE2K_CBUS
+
+config NE2K_CBUS_NEC108
+	bool "NEC PC-9801-108 Support"
+	depends on NE2K_CBUS
+
 config SKMC
 	tristate "SKnet MCA support"
 	depends on NET_ETHERNET && MCA
@@ -1904,6 +1953,7 @@ config E1000
 	  82544       PRO/1000 XF Server Adapter           A50484-xxx
 	  82544       PRO/1000 T Desktop Adapter           A62947-xxx
 	  82540       PRO/1000 MT Desktop Adapter          A78408-xxx
+	  82541       PRO/1000 MT Desktop Adapter          C91016-xxx
 	  82545       PRO/1000 MT Server Adapter           A92165-xxx
 	  82546       PRO/1000 MT Dual Port Server Adapter A92111-xxx
 	  82545       PRO/1000 MF Server Adapter           A91622-xxx

drivers/net/Makefile

@@ -81,6 +81,7 @@ obj-$(CONFIG_ARM_ETHERH) += 8390.o
 obj-$(CONFIG_WD80x3) += wd.o 8390.o
 obj-$(CONFIG_EL2) += 3c503.o 8390.o
 obj-$(CONFIG_NE2000) += ne.o 8390.o
+obj-$(CONFIG_NE2K_CBUS) += ne2k_cbus.o 8390.o
 obj-$(CONFIG_NE2_MCA) += ne2.o 8390.o
 obj-$(CONFIG_HPLAN) += hp.o 8390.o
 obj-$(CONFIG_HPLAN_PLUS) += hp-plus.o 8390.o

drivers/net/Makefile.lib

@@ -19,6 +19,7 @@ obj-$(CONFIG_MACE)		+= crc32.o
 obj-$(CONFIG_MACMACE)		+= crc32.o
 obj-$(CONFIG_MIPS_AU1000_ENET)	+= crc32.o
 obj-$(CONFIG_NATSEMI)		+= crc32.o	
+obj-$(CONFIG_NE2K_CBUS)		+= crc32.o
 obj-$(CONFIG_PCMCIA_FMVJ18X)	+= crc32.o
 obj-$(CONFIG_PCMCIA_SMC91C92)	+= crc32.o
 obj-$(CONFIG_PCMCIA_XIRTULIP)	+= crc32.o

drivers/net/Space.c

@@ -233,7 +233,7 @@ static struct devprobe isa_probes[] __initdata = {
 #ifdef CONFIG_E2100		/* Cabletron E21xx series. */
 	{e2100_probe, 0},
 #endif
-#ifdef CONFIG_NE2000		/* ISA (use ne2k-pci for PCI cards) */
+#if defined(CONFIG_NE2000) || defined(CONFIG_NE2K_CBUS)	/* ISA & PC-9800 CBUS (use ne2k-pci for PCI cards) */
 	{ne_probe, 0},
 #endif
 #ifdef CONFIG_LANCE		/* ISA/VLB (use pcnet32 for PCI cards) */

drivers/net/at1700.c

@@ -34,6 +34,10 @@
 	only is it difficult to detect, it also moves around in I/O space in
 	response to inb()s from other device probes!
 */
+/*
+	99/03/03  Allied Telesis RE1000 Plus support by T.Hagawa
+	99/12/30	port to 2.3.35 by K.Takai
+*/
 
 #include <linux/config.h>
 #include <linux/errno.h>
@@ -76,10 +80,17 @@ static int fmv18x_probe_list[] __initdata = {
  *	ISA
  */
 
+#ifndef CONFIG_X86_PC9800
 static int at1700_probe_list[] __initdata = {
 	0x260, 0x280, 0x2a0, 0x240, 0x340, 0x320, 0x380, 0x300, 0
 };
 
+#else /* CONFIG_X86_PC9800 */
+static int at1700_probe_list[] __initdata = {
+	0x1d6, 0x1d8, 0x1da, 0x1d4, 0xd4, 0xd2, 0xd8, 0xd0, 0
+};
+
+#endif /* CONFIG_X86_PC9800 */
 /*
  *	MCA
  */
@@ -122,6 +133,7 @@ struct net_local {
 
 
 /* Offsets from the base address. */
+#ifndef CONFIG_X86_PC9800
 #define STATUS			0
 #define TX_STATUS		0
 #define RX_STATUS		1
@@ -136,6 +148,7 @@ struct net_local {
 #define TX_START		10
 #define COL16CNTL		11		/* Controll Reg for 16 collisions */
 #define MODE13			13
+#define RX_CTRL			14
 /* Configuration registers only on the '865A/B chips. */
 #define EEPROM_Ctrl 	16
 #define EEPROM_Data 	17
@@ -144,8 +157,39 @@ struct net_local {
 #define IOCONFIG		18		/* Either read the jumper, or move the I/O. */
 #define IOCONFIG1		19
 #define	SAPROM			20		/* The station address PROM, if no EEPROM. */
+#define MODE24			24
 #define RESET			31		/* Write to reset some parts of the chip. */
 #define AT1700_IO_EXTENT	32
+#define PORT_OFFSET(o) (o)
+#else /* CONFIG_X86_PC9800 */
+#define STATUS			(0x0000)
+#define TX_STATUS		(0x0000)
+#define RX_STATUS		(0x0001)
+#define TX_INTR			(0x0200)/* Bit-mapped interrupt enable registers. */
+#define RX_INTR			(0x0201)
+#define TX_MODE			(0x0400)
+#define RX_MODE			(0x0401)
+#define CONFIG_0		(0x0600)/* Misc. configuration settings. */
+#define CONFIG_1		(0x0601)
+/* Run-time register bank 2 definitions. */
+#define DATAPORT		(0x0800)/* Word-wide DMA or programmed-I/O dataport. */
+#define TX_START		(0x0a00)
+#define COL16CNTL		(0x0a01)/* Controll Reg for 16 collisions */
+#define MODE13			(0x0c01)
+#define RX_CTRL			(0x0e00)
+/* Configuration registers only on the '865A/B chips. */
+#define EEPROM_Ctrl 	(0x1000)
+#define EEPROM_Data 	(0x1200)
+#define CARDSTATUS	16			/* FMV-18x Card Status */
+#define CARDSTATUS1	17			/* FMV-18x Card Status */
+#define IOCONFIG		(0x1400)/* Either read the jumper, or move the I/O. */
+#define IOCONFIG1		(0x1600)
+#define	SAPROM			20		/* The station address PROM, if no EEPROM. */
+#define	MODE24			(0x1800)/* The station address PROM, if no EEPROM. */
+#define RESET			(0x1e01)/* Write to reset some parts of the chip. */
+#define PORT_OFFSET(o) ({ int _o_ = (o); (_o_ & ~1) * 0x100 + (_o_ & 1); })
+#endif /* CONFIG_X86_PC9800 */
+
 
 #define TX_TIMEOUT		10
 
@@ -225,8 +269,20 @@ static int __init at1700_probe1(struct net_device *dev, int ioaddr)
 	int slot, ret = -ENODEV;
 	struct net_local *lp;
 	
+#ifndef CONFIG_X86_PC9800
 	if (!request_region(ioaddr, AT1700_IO_EXTENT, dev->name))
 		return -EBUSY;
+#else
+	for (i = 0; i < 0x2000; i += 0x0200) {
+		if (!request_region(ioaddr + i, 2, dev->name)) {
+			while (i > 0) {
+				i -= 0x0200;
+				release_region(ioaddr + i, 2);
+			}
+			return -EBUSY;
+		}
+	}
+#endif
 
 		/* Resetting the chip doesn't reset the ISA interface, so don't bother.
 	   That means we have to be careful with the register values we probe for.
@@ -317,10 +373,17 @@ static int __init at1700_probe1(struct net_device *dev, int ioaddr)
 		/* Reset the internal state machines. */
 	outb(0, ioaddr + RESET);
 
-	if (is_at1700)
+	if (is_at1700) {
+#ifndef CONFIG_X86_PC9800
 		irq = at1700_irqmap[(read_eeprom(ioaddr, 12)&0x04)
 						   | (read_eeprom(ioaddr, 0)>>14)];
-	else {
+#else
+		{
+			char re1000plus_irqmap[4] = {3, 5, 6, 12};
+			irq = re1000plus_irqmap[inb(ioaddr + IOCONFIG1) >> 6];
+		}
+#endif
+	} else {
 		/* Check PnP mode for FMV-183/184/183A/184A. */
 		/* This PnP routine is very poor. IO and IRQ should be known. */
 		if (inb(ioaddr + CARDSTATUS1) & 0x20) {
@@ -392,18 +455,22 @@ static int __init at1700_probe1(struct net_device *dev, int ioaddr)
 	/* Set the station address in bank zero. */
 	outb(0x00, ioaddr + CONFIG_1);
 	for (i = 0; i < 6; i++)
-		outb(dev->dev_addr[i], ioaddr + 8 + i);
+		outb(dev->dev_addr[i], ioaddr + PORT_OFFSET(8 + i));
 
 	/* Switch to bank 1 and set the multicast table to accept none. */
 	outb(0x04, ioaddr + CONFIG_1);
 	for (i = 0; i < 8; i++)
-		outb(0x00, ioaddr + 8 + i);
+		outb(0x00, ioaddr + PORT_OFFSET(8 + i));
 
 
 	/* Switch to bank 2 */
 	/* Lock our I/O address, and set manual processing mode for 16 collisions. */
 	outb(0x08, ioaddr + CONFIG_1);
+#ifndef CONFIG_X86_PC9800
 	outb(dev->if_port, ioaddr + MODE13);
+#else
+	outb(0, ioaddr + MODE13);
+#endif
 	outb(0x00, ioaddr + COL16CNTL);
 
 	if (net_debug)
@@ -447,7 +514,12 @@ static int __init at1700_probe1(struct net_device *dev, int ioaddr)
 	kfree(dev->priv);
 	dev->priv = NULL;
 err_out:
+#ifndef CONFIG_X86_PC9800
 	release_region(ioaddr, AT1700_IO_EXTENT);
+#else
+	for (i = 0; i < 0x2000; i += 0x0200)
+		release_region(ioaddr + i, 2);
+#endif
 	return ret;
 }
 
@@ -459,7 +531,11 @@ static int __init at1700_probe1(struct net_device *dev, int ioaddr)
 #define EE_DATA_READ	0x80	/* EEPROM chip data out, in reg. 17. */
 
 /* Delay between EEPROM clock transitions. */
+#ifndef CONFIG_X86_PC9800
 #define eeprom_delay()	do { } while (0)
+#else
+#define eeprom_delay()	__asm__ ("out%B0 %%al,%0" :: "N"(0x5f))
+#endif
 
 /* The EEPROM commands include the alway-set leading bit. */
 #define EE_WRITE_CMD	(5 << 6)
@@ -542,12 +618,12 @@ static void net_tx_timeout (struct net_device *dev)
 		inw (ioaddr + STATUS), inb (ioaddr + TX_STATUS) & 0x80
 		? "IRQ conflict" : "network cable problem");
 	printk ("%s: timeout registers: %04x %04x %04x %04x %04x %04x %04x %04x.\n",
-	 dev->name, inw (ioaddr + 0), inw (ioaddr + 2), inw (ioaddr + 4),
-		inw (ioaddr + 6), inw (ioaddr + 8), inw (ioaddr + 10),
-		inw (ioaddr + 12), inw (ioaddr + 14));
+	 dev->name, inw(ioaddr + TX_STATUS), inw(ioaddr + TX_INTR), inw(ioaddr + TX_MODE),
+		inw(ioaddr + CONFIG_0), inw(ioaddr + DATAPORT), inw(ioaddr + TX_START),
+		inw(ioaddr + MODE13 - 1), inw(ioaddr + RX_CTRL));
 	lp->stats.tx_errors++;
 	/* ToDo: We should try to restart the adaptor... */
-	outw (0xffff, ioaddr + 24);
+	outw(0xffff, ioaddr + MODE24);
 	outw (0xffff, ioaddr + TX_STATUS);
 	outb (0x5a, ioaddr + CONFIG_0);
 	outb (0xe8, ioaddr + CONFIG_1);
@@ -704,7 +780,7 @@ net_rx(struct net_device *dev)
 				   dev->name, inb(ioaddr + RX_MODE), status);
 #ifndef final_version
 		if (status == 0) {
-			outb(0x05, ioaddr + 14);
+			outb(0x05, ioaddr + RX_CTRL);
 			break;
 		}
 #endif
@@ -724,7 +800,7 @@ net_rx(struct net_device *dev)
 					   dev->name, pkt_len);
 				/* Prime the FIFO and then flush the packet. */
 				inw(ioaddr + DATAPORT); inw(ioaddr + DATAPORT);
-				outb(0x05, ioaddr + 14);
+				outb(0x05, ioaddr + RX_CTRL);
 				lp->stats.rx_errors++;
 				break;
 			}
@@ -734,7 +810,7 @@ net_rx(struct net_device *dev)
 					   dev->name, pkt_len);
 				/* Prime the FIFO and then flush the packet. */
 				inw(ioaddr + DATAPORT); inw(ioaddr + DATAPORT);
-				outb(0x05, ioaddr + 14);
+				outb(0x05, ioaddr + RX_CTRL);
 				lp->stats.rx_dropped++;
 				break;
 			}
@@ -761,7 +837,7 @@ net_rx(struct net_device *dev)
 			if ((inb(ioaddr + RX_MODE) & 0x40) == 0x40)
 				break;
 			inw(ioaddr + DATAPORT);				/* dummy status read */
-			outb(0x05, ioaddr + 14);
+			outb(0x05, ioaddr + RX_CTRL);
 		}
 
 		if (net_debug > 5)
@@ -844,24 +920,28 @@ set_rx_mode(struct net_device *dev)
 		outb(0x02, ioaddr + RX_MODE);	/* Use normal mode. */
 	}
 
-	save_flags(flags);
-	cli();
+	spin_lock_irqsave (&lp->lock, flags);
 	if (memcmp(mc_filter, lp->mc_filter, sizeof(mc_filter))) {
 		int saved_bank = inw(ioaddr + CONFIG_0);
 		/* Switch to bank 1 and set the multicast table. */
 		outw((saved_bank & ~0x0C00) | 0x0480, ioaddr + CONFIG_0);
 		for (i = 0; i < 8; i++)
-			outb(mc_filter[i], ioaddr + 8 + i);
+			outb(mc_filter[i], ioaddr + PORT_OFFSET(8 + i));
 		memcpy(lp->mc_filter, mc_filter, sizeof(mc_filter));
 		outw(saved_bank, ioaddr + CONFIG_0);
 	}
-	restore_flags(flags);
+	spin_unlock_irqrestore (&lp->lock, flags);
 	return;
 }
 
 #ifdef MODULE
 static struct net_device dev_at1700;
+#ifndef CONFIG_X86_PC9800
 static int io = 0x260;
+#else
+static int io = 0xd0;
+#endif
+
 static int irq;
 
 MODULE_PARM(io, "i");
@@ -901,7 +981,15 @@ cleanup_module(void)
 
 	/* If we don't do this, we can't re-insmod it later. */
 	free_irq(dev_at1700.irq, NULL);
+#ifndef CONFIG_X86_PC9800
 	release_region(dev_at1700.base_addr, AT1700_IO_EXTENT);
+#else
+	{
+		int i;
+		for (i = 0; i < 0x2000; i += 0x200)
+			release_region(dev_at1700.base_addr + i, 2);
+	}
+#endif
 }
 #endif /* MODULE */
 MODULE_LICENSE("GPL");

drivers/net/e100/e100.h

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 
@@ -57,6 +57,8 @@
 #include <linux/if.h>
 #include <asm/uaccess.h>
 #include <linux/ip.h>
+#include <linux/if_vlan.h>
+#include <linux/mii.h>
 
 #define E100_REGS_LEN 1
 /*
@@ -301,6 +303,9 @@ struct driver_stats {
 
 /* EEPROM bit definitions */
 /*- EEPROM control register bits */
+#define EEPROM_FLAG_ASF  0x8000
+#define EEPROM_FLAG_GCL  0x4000
+
 #define EN_TRNF          0x10	/* Enable turnoff */
 #define EEDO             0x08	/* EEPROM data out */
 #define EEDI             0x04	/* EEPROM data in (set for writing data) */
@@ -319,6 +324,8 @@ struct driver_stats {
 #define EEPROM_COMPATIBILITY_WORD       3
 #define EEPROM_PWA_NO                   8
 #define EEPROM_ID_WORD			0x0A
+#define EEPROM_CONFIG_ASF		0x0D
+#define EEPROM_SMBUS_ADDR		0x90
 
 #define EEPROM_SUM                      0xbaba
 
@@ -358,7 +365,7 @@ struct driver_stats {
 #define CB_STATUS_MASK          BIT_12_15	/* CB Status Mask (4-bits) */
 #define CB_STATUS_COMPLETE      BIT_15	/* CB Complete Bit */
 #define CB_STATUS_OK            BIT_13	/* CB OK Bit */
-#define CB_STATUS_UNDERRUN      BIT_12	/* CB A Bit */
+#define CB_STATUS_VLAN          BIT_12 /* CB Valn detected Bit */
 #define CB_STATUS_FAIL          BIT_11	/* CB Fail (F) Bit */
 
 /*misc command bits */
@@ -851,6 +858,7 @@ struct ethtool_lpbk_data{
 };
 
 struct e100_private {
+	struct vlan_group *vlgrp;
 	u32 flags;		/* board management flags */
 	u32 tx_per_underrun;	/* number of good tx frames per underrun */
 	unsigned int tx_count;	/* count of tx frames, so we can request an interrupt */
@@ -886,7 +894,6 @@ struct e100_private {
 	struct driver_stats drv_stats;
 
 	u8 rev_id;		/* adapter PCI revision ID */
-	unsigned long device_type;	/* device type from e100_vendor.h */
 
 	unsigned int phy_addr;	/* address of PHY component */
 	unsigned int PhyId;	/* ID of PHY component */
@@ -923,8 +930,6 @@ struct e100_private {
 
 	struct cfg_params params;	/* adapter's command line parameters */
 
-	char *id_string;
-
 	u32 speed_duplex_caps;	/* adapter's speed/duplex capabilities */
 
 	/* WOL params for ethtool */

drivers/net/e100/e100_config.c

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 
@@ -521,6 +521,25 @@ e100_config_wol(struct e100_private *bdp)
 	spin_unlock_bh(&(bdp->config_lock));
 }
 
+void
+e100_config_vlan_drop(struct e100_private *bdp, unsigned char enable)
+{
+	spin_lock_bh(&(bdp->config_lock));
+	if (enable) {
+		if (!(bdp->config[22] & CB_CFIG_VLAN_DROP_ENABLE)) {
+			bdp->config[22] |= CB_CFIG_VLAN_DROP_ENABLE;
+			E100_CONFIG(bdp, 22);
+		}
+
+	} else {
+		if ((bdp->config[22] & CB_CFIG_VLAN_DROP_ENABLE)) {
+			bdp->config[22] &= ~CB_CFIG_VLAN_DROP_ENABLE;
+			E100_CONFIG(bdp, 22);
+		}
+	}
+	spin_unlock_bh(&(bdp->config_lock));
+}
+
 /**
  * e100_config_loopback_mode
  * @bdp: atapter's private data struct

drivers/net/e100/e100_config.h

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 
@@ -163,5 +163,5 @@ extern void e100_config_force_dplx(struct e100_private *bdp);
 extern u8 e100_config_loopback_mode(struct e100_private *bdp, u8 mode);
 extern u8 e100_config_dynamic_tbd(struct e100_private *bdp, u8 enable);
 extern u8 e100_config_tcb_ext_enable(struct e100_private *bdp, u8 enable);
-
+extern void e100_config_vlan_drop(struct e100_private *bdp, unsigned char enable);
 #endif /* _E100_CONFIG_INC_ */

drivers/net/e100/e100_eeprom.c

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 

drivers/net/e100/e100_phy.c

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 
@@ -636,7 +636,6 @@ e100_force_speed_duplex(struct e100_private *bdp)
 	control &= ~BMCR_ANENABLE;
 	control &= ~BMCR_LOOPBACK;
 
-	/* Check e100.c values */
 	switch (bdp->params.e100_speed_duplex) {
 	case E100_SPEED_10_HALF:
 		control &= ~BMCR_SPEED100;
@@ -682,6 +681,41 @@ e100_force_speed_duplex(struct e100_private *bdp)
 	} while (true);
 }
 
+void
+e100_force_speed_duplex_to_phy(struct e100_private *bdp)
+{
+	u16 control;
+
+	e100_mdi_read(bdp, MII_BMCR, bdp->phy_addr, &control);
+	control &= ~BMCR_ANENABLE;
+	control &= ~BMCR_LOOPBACK;
+
+	switch (bdp->params.e100_speed_duplex) {
+	case E100_SPEED_10_HALF:
+		control &= ~BMCR_SPEED100;
+		control &= ~BMCR_FULLDPLX;
+		break;
+
+	case E100_SPEED_10_FULL:
+		control &= ~BMCR_SPEED100;
+		control |= BMCR_FULLDPLX;
+		break;
+
+	case E100_SPEED_100_HALF:
+		control |= BMCR_SPEED100;
+		control &= ~BMCR_FULLDPLX;
+		break;
+
+	case E100_SPEED_100_FULL:
+		control |= BMCR_SPEED100;
+		control |= BMCR_FULLDPLX;
+		break;
+	}
+
+	/* Send speed/duplex command to PHY layer. */
+	e100_mdi_write(bdp, MII_BMCR, bdp->phy_addr, control);
+}
+
 /* 
  * Procedure: e100_set_fc
  *

drivers/net/e100/e100_phy.h

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 
@@ -30,8 +30,6 @@
 
 #include "e100.h"
 
-#include <linux/mii.h>
-
 /*
  * Auto-polarity enable/disable
  * e100_autopolarity = 0 => disable auto-polarity

drivers/net/e100/e100_test.c

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 
@@ -242,6 +242,9 @@ e100_diag_config_loopback(struct e100_private* bdp,
 		 *dynamic_tbd = e100_config_dynamic_tbd(bdp,*dynamic_tbd);
 
 	if (set_loopback) {
+		/* ICH PHY loopback is broken */
+		if (bdp->flags & IS_ICH && loopback_mode == PHY_LOOPBACK)
+			loopback_mode = MAC_LOOPBACK;
 		/* Configure loopback on MAC */
 		e100_config_loopback_mode(bdp,loopback_mode);
 	} else {

drivers/net/e100/e100_ucode.h

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 

drivers/net/e1000/e1000.h

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 
@@ -65,6 +65,7 @@
 #include <linux/reboot.h>
 #include <net/checksum.h>
 #include <linux/workqueue.h>
+#include <linux/mii.h>
 #include <linux/ethtool.h>
 #include <linux/if_vlan.h>
 
@@ -95,6 +96,15 @@ struct e1000_adapter;
 #define E1000_RXBUFFER_8192  8192
 #define E1000_RXBUFFER_16384 16384
 
+/* SmartSpeed delimiters */
+#define E1000_SMARTSPEED_DOWNSHIFT 3
+#define E1000_SMARTSPEED_MAX       15
+
+/* Packet Buffer allocations */
+#define E1000_TX_FIFO_SIZE_SHIFT 0xA
+#define E1000_TX_HEAD_ADDR_SHIFT 7
+#define E1000_PBA_TX_MASK 0xFFFF0000
+
 /* Flow Control High-Watermark: 43464 bytes */
 #define E1000_FC_HIGH_THRESH 0xA9C8
 
@@ -107,10 +117,7 @@ struct e1000_adapter;
 /* How many Tx Descriptors do we need to call netif_wake_queue ? */
 #define E1000_TX_QUEUE_WAKE	16
 /* How many Rx Buffers do we bundle into one write to the hardware ? */
-#define E1000_RX_BUFFER_WRITE	16
-
-#define E1000_JUMBO_PBA      0x00000028
-#define E1000_DEFAULT_PBA    0x00000030
+#define E1000_RX_BUFFER_WRITE	16	/* Must be power of 2 */
 
 #define AUTO_ALL_MODES       0
 #define E1000_EEPROM_APME    4
@@ -145,7 +152,8 @@ struct e1000_desc_ring {
 };
 
 #define E1000_DESC_UNUSED(R) \
-((((R)->next_to_clean + (R)->count) - ((R)->next_to_use + 1)) % ((R)->count))
+	((((R)->next_to_clean > (R)->next_to_use) ? 0 : (R)->count) + \
+	(R)->next_to_clean - (R)->next_to_use - 1)
 
 #define E1000_GET_DESC(R, i, type)	(&(((struct type *)((R).desc))[i]))
 #define E1000_RX_DESC(R, i)		E1000_GET_DESC(R, i, e1000_rx_desc)
@@ -155,6 +163,7 @@ struct e1000_desc_ring {
 /* board specific private data structure */
 
 struct e1000_adapter {
+	struct timer_list tx_fifo_stall_timer;
 	struct timer_list watchdog_timer;
 	struct timer_list phy_info_timer;
 	struct vlan_group *vlgrp;
@@ -163,6 +172,7 @@ struct e1000_adapter {
 	uint32_t rx_buffer_len;
 	uint32_t part_num;
 	uint32_t wol;
+	uint32_t smartspeed;
 	uint16_t link_speed;
 	uint16_t link_duplex;
 	spinlock_t stats_lock;
@@ -177,7 +187,11 @@ struct e1000_adapter {
 	uint32_t txd_cmd;
 	uint32_t tx_int_delay;
 	uint32_t tx_abs_int_delay;
-	int max_data_per_txd;
+	uint32_t gotcl;
+	uint32_t tx_fifo_head;
+	uint32_t tx_head_addr;
+	uint32_t tx_fifo_size;
+	atomic_t tx_fifo_stall;
 
 	/* RX */
 	struct e1000_desc_ring rx_ring;
@@ -186,6 +200,10 @@ struct e1000_adapter {
 	uint32_t rx_int_delay;
 	uint32_t rx_abs_int_delay;
 	boolean_t rx_csum;
+	uint32_t gorcl;
+
+	/* Interrupt Throttle Rate */
+	uint32_t itr;
 
 	/* OS defined structs */
 	struct net_device *netdev;

drivers/net/e1000/e1000_ethtool.c

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 
@@ -38,6 +38,7 @@ extern char e1000_driver_version[];
 extern int e1000_up(struct e1000_adapter *adapter);
 extern void e1000_down(struct e1000_adapter *adapter);
 extern void e1000_reset(struct e1000_adapter *adapter);
+extern int e1000_set_spd_dplx(struct e1000_adapter *adapter, uint16_t spddplx);
 
 static char e1000_gstrings_stats[][ETH_GSTRING_LEN] = {
 	"rx_packets", "tx_packets", "rx_bytes", "tx_bytes", "rx_errors",
@@ -129,30 +130,9 @@ e1000_ethtool_sset(struct e1000_adapter *adapter, struct ethtool_cmd *ecmd)
 		hw->autoneg = 1;
 		hw->autoneg_advertised = 0x002F;
 		ecmd->advertising = 0x002F;
-	} else {
-		hw->autoneg = 0;
-		switch(ecmd->speed + ecmd->duplex) {
-		case SPEED_10 + DUPLEX_HALF:
-			hw->forced_speed_duplex = e1000_10_half;
-			break;
-		case SPEED_10 + DUPLEX_FULL:
-			hw->forced_speed_duplex = e1000_10_full;
-			break;
-		case SPEED_100 + DUPLEX_HALF:
-			hw->forced_speed_duplex = e1000_100_half;
-			break;
-		case SPEED_100 + DUPLEX_FULL:
-			hw->forced_speed_duplex = e1000_100_full;
-			break;
-		case SPEED_1000 + DUPLEX_FULL:
-			hw->autoneg = 1;
-			hw->autoneg_advertised = ADVERTISE_1000_FULL;
-			break;
-		case SPEED_1000 + DUPLEX_HALF: /* not supported */
-		default:
+	} else
+		if(e1000_set_spd_dplx(adapter, ecmd->speed + ecmd->duplex))
 			return -EINVAL;
-		}
-	}
 
 	/* reset the link */
 
@@ -165,16 +145,6 @@ e1000_ethtool_sset(struct e1000_adapter *adapter, struct ethtool_cmd *ecmd)
 	return 0;
 }
 
-static inline int
-e1000_eeprom_size(struct e1000_hw *hw)
-{
-	if((hw->mac_type > e1000_82544) &&
-	   (E1000_READ_REG(hw, EECD) & E1000_EECD_SIZE))
-		return 512;
-	else
-		return 128;
-}
-
 static void
 e1000_ethtool_gdrvinfo(struct e1000_adapter *adapter,
                        struct ethtool_drvinfo *drvinfo)
@@ -186,7 +156,7 @@ e1000_ethtool_gdrvinfo(struct e1000_adapter *adapter,
 	drvinfo->n_stats = E1000_STATS_LEN;
 #define E1000_REGS_LEN 32
 	drvinfo->regdump_len  = E1000_REGS_LEN * sizeof(uint32_t);
-	drvinfo->eedump_len  = e1000_eeprom_size(&adapter->hw);
+	drvinfo->eedump_len = adapter->hw.eeprom.word_size * 2;
 }
 
 static void
@@ -220,9 +190,8 @@ e1000_ethtool_geeprom(struct e1000_adapter *adapter,
                       struct ethtool_eeprom *eeprom, uint16_t *eeprom_buff)
 {
 	struct e1000_hw *hw = &adapter->hw;
-	int max_len, first_word, last_word;
+	int first_word, last_word;
 	int ret_val = 0;
-	int i;
 
 	if(eeprom->len == 0) {
 		ret_val = -EINVAL;
@@ -231,22 +200,28 @@ e1000_ethtool_geeprom(struct e1000_adapter *adapter,
 
 	eeprom->magic = hw->vendor_id | (hw->device_id << 16);
 
-	max_len = e1000_eeprom_size(hw);
-
 	if(eeprom->offset > eeprom->offset + eeprom->len) {
 		ret_val = -EINVAL;
 		goto geeprom_error;
 	}
 
-	if((eeprom->offset + eeprom->len) > max_len)
-		eeprom->len = (max_len - eeprom->offset);
+	if((eeprom->offset + eeprom->len) > (hw->eeprom.word_size * 2))
+		eeprom->len = ((hw->eeprom.word_size * 2) - eeprom->offset);
 
 	first_word = eeprom->offset >> 1;
 	last_word = (eeprom->offset + eeprom->len - 1) >> 1;
 
-	for(i = 0; i <= (last_word - first_word); i++)
-		e1000_read_eeprom(hw, first_word + i, &eeprom_buff[i]);
-
+	if(hw->eeprom.type == e1000_eeprom_spi)
+		ret_val = e1000_read_eeprom(hw, first_word,
+					    last_word - first_word + 1,
+					    eeprom_buff);
+	else {
+		uint16_t i;
+		for (i = 0; i < last_word - first_word + 1; i++)
+			if((ret_val = e1000_read_eeprom(hw, first_word + i, 1,
+						       &eeprom_buff[i])))
+				break;
+	}
 geeprom_error:
 	return ret_val;
 }
@@ -257,9 +232,8 @@ e1000_ethtool_seeprom(struct e1000_adapter *adapter,
 {
 	struct e1000_hw *hw = &adapter->hw;
 	uint16_t *eeprom_buff;
-	int max_len, first_word, last_word;
 	void *ptr;
-	int i;
+	int max_len, first_word, last_word, ret_val = 0;
 
 	if(eeprom->len == 0)
 		return -EOPNOTSUPP;
@@ -267,7 +241,7 @@ e1000_ethtool_seeprom(struct e1000_adapter *adapter,
 	if(eeprom->magic != (hw->vendor_id | (hw->device_id << 16)))
 		return -EFAULT;
 
-	max_len = e1000_eeprom_size(hw);
+	max_len = hw->eeprom.word_size * 2;
 
 	if((eeprom->offset + eeprom->len) > max_len)
 		eeprom->len = (max_len - eeprom->offset);
@@ -283,30 +257,31 @@ e1000_ethtool_seeprom(struct e1000_adapter *adapter,
 	if(eeprom->offset & 1) {
 		/* need read/modify/write of first changed EEPROM word */
 		/* only the second byte of the word is being modified */
-		e1000_read_eeprom(hw, first_word, &eeprom_buff[0]);
+		ret_val = e1000_read_eeprom(hw, first_word, 1,
+					    &eeprom_buff[0]);
 		ptr++;
 	}
-	if((eeprom->offset + eeprom->len) & 1) {
+	if(((eeprom->offset + eeprom->len) & 1) && (ret_val == 0)) {
 		/* need read/modify/write of last changed EEPROM word */
 		/* only the first byte of the word is being modified */
-		e1000_read_eeprom(hw, last_word,
+		ret_val = e1000_read_eeprom(hw, last_word, 1,
 		                  &eeprom_buff[last_word - first_word]);
 	}
-	if(copy_from_user(ptr, user_data, eeprom->len)) {
-		kfree(eeprom_buff);
-		return -EFAULT;
+	if((ret_val != 0) || copy_from_user(ptr, user_data, eeprom->len)) {
+		ret_val = -EFAULT;
+		goto seeprom_error;
 	}
 
-	for(i = 0; i <= (last_word - first_word); i++)
-		e1000_write_eeprom(hw, first_word + i, eeprom_buff[i]);
+	ret_val = e1000_write_eeprom(hw, first_word,
+				     last_word - first_word + 1, eeprom_buff);
 
 	/* Update the checksum over the first part of the EEPROM if needed */
-	if(first_word <= EEPROM_CHECKSUM_REG)
+	if((ret_val == 0) && first_word <= EEPROM_CHECKSUM_REG)
 		e1000_update_eeprom_checksum(hw);
 
+seeprom_error:
 	kfree(eeprom_buff);
-
-	return 0;
+	return ret_val;
 }
 
 static void
@@ -333,8 +308,8 @@ e1000_ethtool_gwol(struct e1000_adapter *adapter, struct ethtool_wolinfo *wol)
 		/* Fall Through */
 
 	default:
-		wol->supported = WAKE_UCAST | WAKE_MCAST
-			         | WAKE_BCAST | WAKE_MAGIC;
+		wol->supported = WAKE_UCAST | WAKE_MCAST |
+				 WAKE_BCAST | WAKE_MAGIC;
 
 		wol->wolopts = 0;
 		if(adapter->wol & E1000_WUFC_EX)
@@ -368,7 +343,7 @@ e1000_ethtool_swol(struct e1000_adapter *adapter, struct ethtool_wolinfo *wol)
 		/* Fall Through */
 
 	default:
-		if(wol->wolopts & (WAKE_ARP | WAKE_MAGICSECURE | WAKE_PHY))
+		if(wol->wolopts & (WAKE_PHY | WAKE_ARP | WAKE_MAGICSECURE))
 			return -EOPNOTSUPP;
 
 		adapter->wol = 0;
@@ -542,13 +517,12 @@ e1000_ethtool_ioctl(struct net_device *netdev, struct ifreq *ifr)
 	}
 	case ETHTOOL_GEEPROM: {
 		struct ethtool_eeprom eeprom = {ETHTOOL_GEEPROM};
+		struct e1000_hw *hw = &adapter->hw;
 		uint16_t *eeprom_buff;
 		void *ptr;
-		int max_len, err = 0;
-
-		max_len = e1000_eeprom_size(&adapter->hw);
+		int err = 0;
 
-		eeprom_buff = kmalloc(max_len, GFP_KERNEL);
+		eeprom_buff = kmalloc(hw->eeprom.word_size * 2, GFP_KERNEL);
 
 		if(eeprom_buff == NULL)
 			return -ENOMEM;

drivers/net/e1000/e1000_osdep.h

 /*******************************************************************************
 
   
-  Copyright(c) 1999 - 2002 Intel Corporation. All rights reserved.
+  Copyright(c) 1999 - 2003 Intel Corporation. All rights reserved.
   
   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 
@@ -77,24 +77,22 @@ typedef enum {
 
 
 #define E1000_WRITE_REG(a, reg, value) ( \
-    ((a)->mac_type >= e1000_82543) ? \
-        (writel((value), ((a)->hw_addr + E1000_##reg))) : \
-        (writel((value), ((a)->hw_addr + E1000_82542_##reg))))
+    writel((value), ((a)->hw_addr + \
+        (((a)->mac_type >= e1000_82543) ? E1000_##reg : E1000_82542_##reg))))
 
 #define E1000_READ_REG(a, reg) ( \
-    ((a)->mac_type >= e1000_82543) ? \
-        readl((a)->hw_addr + E1000_##reg) : \
-        readl((a)->hw_addr + E1000_82542_##reg))
+    readl((a)->hw_addr + \
+        (((a)->mac_type >= e1000_82543) ? E1000_##reg : E1000_82542_##reg)))
 
 #define E1000_WRITE_REG_ARRAY(a, reg, offset, value) ( \
-    ((a)->mac_type >= e1000_82543) ? \
-        writel((value), ((a)->hw_addr + E1000_##reg + ((offset) << 2))) : \
-        writel((value), ((a)->hw_addr + E1000_82542_##reg + ((offset) << 2))))
+    writel((value), ((a)->hw_addr + \
+        (((a)->mac_type >= e1000_82543) ? E1000_##reg : E1000_82542_##reg) + \
+        ((offset) << 2))))
 
 #define E1000_READ_REG_ARRAY(a, reg, offset) ( \
-    ((a)->mac_type >= e1000_82543) ? \
-        readl((a)->hw_addr + E1000_##reg + ((offset) << 2)) : \
-        readl((a)->hw_addr + E1000_82542_##reg + ((offset) << 2)))
+    readl((a)->hw_addr + \
+        (((a)->mac_type >= e1000_82543) ? E1000_##reg : E1000_82542_##reg) + \
+        ((offset) << 2)))
 
 #define E1000_WRITE_FLUSH(a) E1000_READ_REG(a, STATUS)