Re: FYI: kgdb 2.1.0 interdiff

From: Amit S. Kale
Date: Fri Jan 23 2004 - 04:11:23 EST

On Friday 23 Jan 2004 1:48 pm, Pavel Machek wrote:
> Hi!
>
> [I hope these are usefull to someone? They do not contain ppc and
> serial].

I guess so. Now onwards I'll send diff of patch trees. They should be a few
lines only between consecutive kgdb versions and will be easier to read.

> entry.S change is interesting... and it changes do_IRQ() calling
> convention. Perhaps that change should be pushed to Andrew? It is
> "interesting" to have different do_IRQ in case of kgdb...

This change of do_IRQ calling convention isn't a good fix. It's just a quick
work around the gdb's problem of not being able to parse common_interrupt
frame correctly.

George Anzinger is working on some cfi directives to tell gdb about the format
of common_interrupt stack frame. Once that's available I'll remove this
work-around.


>
> Pavel
>
> @@ -67,8 +67,8 @@
>
> Index: linux/arch/i386/kernel/entry.S
> ===================================================================
> ---- linux.orig/arch/i386/kernel/entry.S 2004-01-20 14:28:44.000000000
> +0100 -+++ linux/arch/i386/kernel/entry.S 2004-01-20 14:29:19.000000000
> +0100 +--- linux.orig/arch/i386/kernel/entry.S 2004-01-23
> 09:04:23.000000000 +0100 ++++ linux/arch/i386/kernel/entry.S 2004-01-23
> 09:05:16.000000000 +0100 @@ -226,7 +226,7 @@
> jz restore_all
> movl $PREEMPT_ACTIVE,TI_PRE_COUNT(%ebp)
> @@ -87,7 +87,25 @@
> cli # make sure we don't miss an interrupt
> # setting need_resched or sigpending
> # between sampling and the iret
> -@@ -606,6 +606,31 @@
> +@@ -399,7 +399,17 @@
> + ALIGN
> + common_interrupt:
> + SAVE_ALL
> ++ movl %esp, %eax
> ++/* Create a fake function call followed by a fake function prologue to
> fool ++ * gdb into believing that this is a normal function call. */
> ++ pushl EIP(%eax)
> ++
> ++common_interrupt_1:
> ++ pushl %ebp
> ++ movl %esp, %ebp
> ++ pushl %eax
> + call do_IRQ
> ++ addl $12, %esp
> + jmp ret_from_intr
> +
> + #define BUILD_INTERRUPT(name, nr) \
> +@@ -606,6 +616,31 @@
> pushl $do_spurious_interrupt_bug
> jmp error_code
>
> @@ -581,10 +599,41 @@
> + .remove_break = i386_remove_hw_break,
> + .correct_hw_break = i386_correct_hw_break,
> +};
> +Index: linux/arch/i386/kernel/irq.c
> +===================================================================
> +--- linux.orig/arch/i386/kernel/irq.c 2004-01-09 20:26:08.000000000 +0100
> ++++ linux/arch/i386/kernel/irq.c 2004-01-23 09:05:16.000000000 +0100
> +@@ -410,7 +410,7 @@
> + * SMP cross-CPU interrupts have their own specific
> + * handlers).
> + */
> +-asmlinkage unsigned int do_IRQ(struct pt_regs regs)
> ++asmlinkage unsigned int do_IRQ(struct pt_regs *regs)
> + {
> + /*
> + * We ack quickly, we don't want the irq controller
> +@@ -422,7 +422,7 @@
> + * 0 return value means that this irq is already being
> + * handled by some other CPU. (or is disabled)
> + */
> +- int irq = regs.orig_eax & 0xff; /* high bits used in ret_from_ code */
> ++ int irq = regs->orig_eax & 0xff; /* high bits used in ret_from_ code */
> + irq_desc_t *desc = irq_desc + irq;
> + struct irqaction * action;
> + unsigned int status;
> +@@ -488,7 +488,7 @@
> + irqreturn_t action_ret;
> +
> + spin_unlock(&desc->lock);
> +- action_ret = handle_IRQ_event(irq, &regs, action);
> ++ action_ret = handle_IRQ_event(irq, regs, action);
> + spin_lock(&desc->lock);
> + if (!noirqdebug)
> + note_interrupt(irq, desc, action_ret);
> Index: linux/arch/i386/kernel/nmi.c
> ===================================================================
> ---- linux.orig/arch/i386/kernel/nmi.c 2004-01-20 14:28:44.000000000 +0100
> -+++ linux/arch/i386/kernel/nmi.c 2004-01-20 14:29:19.000000000 +0100
> +--- linux.orig/arch/i386/kernel/nmi.c 2004-01-23 09:04:23.000000000 +0100
> ++++ linux/arch/i386/kernel/nmi.c 2004-01-23 09:05:16.000000000 +0100
> @@ -25,6 +25,7 @@
> #include <linux/module.h>
> #include <linux/nmi.h>
> @@ -2031,9 +2080,9 @@
> obj-$(CONFIG_VORTEX) += 3c59x.o
> Index: linux/drivers/net/kgdb_eth.c
> ===================================================================
> ---- linux.orig/drivers/net/kgdb_eth.c 2004-01-20 14:29:19.000000000 +0100
> -+++ linux/drivers/net/kgdb_eth.c 2004-01-20 14:29:19.000000000 +0100
> -@@ -0,0 +1,704 @@
> +--- linux.orig/drivers/net/kgdb_eth.c 2004-01-23 09:05:16.000000000 +0100
> ++++ linux/drivers/net/kgdb_eth.c 2004-01-23 09:05:16.000000000 +0100
> +@@ -0,0 +1,702 @@
> +/*
> + * Network interface GDB stub
> + *
> @@ -2107,7 +2156,7 @@
> +struct net_device *kgdb_netdevice = NULL;
> +
> +/* Indicates dept of recursion for xmitlock hold */
> -+static int xlockholdcount = 0;
> ++static int xlock_hold_count = 0;
> +
> +/* kgdb ethernet ring buffers. Increase the space if you get panics in
> + * kgdbeth_alloc_skb.
> @@ -2122,7 +2171,7 @@
> +static struct sk_buff *send_skb;
> +static int bufnum;
> +
> -+static char kgdb_netdevname[16];
> ++static char kgdb_netdev_name[16];
> +
> +#ifdef CONFIG_KGDB_CONSOLE
> +#error kgdb over ethernet is not yet ready for console messages.
> @@ -2181,7 +2230,7 @@
> +
> +/* Holds xmitlock of the ethernet device
> + * Recursive calls allowed */
> -+static void kgdbeth_holdxlock(void)
> ++static void kgdbeth_lock(void)
> +{
> + if (spin_is_locked(&kgdb_netdevice->xmit_lock)) {
> + if (kgdb_netdevice->xmit_lock_owner == smp_processor_id()) {
> @@ -2192,14 +2241,14 @@
> + kgdb_netdevice->xmit_lock_owner = smp_processor_id();
> +
> +gotit:
> -+ xlockholdcount++;
> ++ xlock_hold_count++;
> +}
> +
> +/* releases xmitlock of the ethernet device
> + * Recursive calls allowed */
> -+static void kgdbeth_relxlock(void)
> ++static void kgdbeth_unlock(void)
> +{
> -+ if (--xlockholdcount) {
> ++ if (--xlock_hold_count) {
> + kgdb_netdevice->xmit_lock_owner = -1;
> + spin_unlock(&kgdb_netdevice->xmit_lock);
> + }
> @@ -2245,7 +2294,7 @@
> + if (!(in_dev->ifa_list)) {
> + panic("No interface address set for interface!\n");
> + }
> -+ kgdbeth_holdxlock();
> ++ kgdbeth_lock();
> +
> + udp_len = len + sizeof(struct udphdr);
> + ip_len = eth_len = udp_len + sizeof(struct iphdr);
> @@ -2284,10 +2333,8 @@
> + memcpy(eth->h_dest, kgdb_remotemac, kgdb_netdevice->addr_len);
> +
> + kgdb_netdevice->hard_start_xmit(skb, kgdb_netdevice);
> -+ if (atomic_read(&skb->users) != 1) {
> -+ BUG();
> -+ }
> -+ kgdbeth_relxlock();
> ++ BUG_ON(atomic_read(&skb->users) != 1);
> ++ kgdbeth_unlock();
> +}
> +
> +static void kgdbeth_flush(void)
> @@ -2317,10 +2364,10 @@
> +kgdb_eth_reply_arp(void)
> +{
> + if (send_skb) {
> -+ kgdbeth_holdxlock();
> ++ kgdbeth_lock();
> + kgdb_netdevice->hard_start_xmit(send_skb, kgdb_netdevice);
> + send_skb = NULL;
> -+ kgdbeth_relxlock();
> ++ kgdbeth_unlock();
> + }
> +}
> +
> @@ -2391,7 +2438,7 @@
> + }
> + /* reply to the ARP request */
> +
> -+ kgdbeth_holdxlock();
> ++ kgdbeth_lock();
> + if (send_skb) {
> + /* Get rid of any previous replies to ARP request. We hope
> + * that regular reply to ARP by network layers would have gone
> @@ -2401,7 +2448,7 @@
> + send_skb = kgdbeth_alloc_skb(sizeof(struct arphdr) +
> + 2 * (kgdb_netdevice->addr_len + 4) +
> + LL_RESERVED_SPACE(kgdb_netdevice));
> -+ kgdbeth_relxlock();
> ++ kgdbeth_unlock();
> +
> + skb_reserve(send_skb, LL_RESERVED_SPACE(kgdb_netdevice));
> + send_skb->nh.raw = send_skb->data;
> @@ -2531,18 +2578,18 @@
> + struct irqaction *ia_ptr;
> + int i;
> +
> -+ sprintf(kgdb_netdevname, "eth%d", kgdb_eth);
> ++ sprintf(kgdb_netdev_name, "eth%d", kgdb_eth);
> +
> + for (kgdb_netdevice = dev_base;
> + kgdb_netdevice != NULL;
> + kgdb_netdevice = kgdb_netdevice->next) {
> -+ if (strncmp(kgdb_netdevice->name, kgdb_netdevname, IFNAMSIZ) == 0) {
> ++ if (strncmp(kgdb_netdevice->name, kgdb_netdev_name, IFNAMSIZ) == 0) {
> + break;
> + }
> + }
> + if (!kgdb_netdevice) {
> + printk("kgdbeth: Unable to find interface %s\n",
> -+ kgdb_netdevname);
> ++ kgdb_netdev_name);
> + return -ENODEV;
> + }
> + if (!(kgdb_netdevice->flags & IFF_UP)) {
> @@ -2550,7 +2597,7 @@
> + }
> + ia_ptr = irq_desc[kgdb_netdevice->irq].action;
> + while (ia_ptr) {
> -+ if (!strncmp(kgdb_netdevname, ia_ptr->name, IFNAMSIZ)) {
> ++ if (!strncmp(kgdb_netdev_name, ia_ptr->name, IFNAMSIZ)) {
> + kgdbeth_irqhandler = ia_ptr->handler;
> + break;
> + }
> @@ -2558,7 +2605,7 @@
> + }
> + if (!kgdbeth_irqhandler) {
> + printk("kgdbeth: Interface %s doesn't have an interrupt"
> -+ " handler cannot use it\n", kgdb_netdevname);
> ++ " handler cannot use it\n", kgdb_netdev_name);
> + return -EINVAL;
> + }
> + for (i = 0; i < NUM_SENDBUF; i++) {
> @@ -2603,7 +2650,7 @@
> + * responded yet.
> + */
> +static void kgdbeth_begin_session(void) {
> -+ kgdbeth_holdxlock();
> ++ kgdbeth_lock();
> + disable_irq(kgdb_netdevice->irq);
> + kgdb_eth_reply_arp();
> +}
> @@ -2611,7 +2658,7 @@
> +static void kgdbeth_end_session(void)
> +{
> + enable_irq(kgdb_netdevice->irq);
> -+ kgdbeth_relxlock();
> ++ kgdbeth_unlock();
> +}
> +
> +struct kgdb_serial kgdbeth_serial = {
> @@ -2661,7 +2708,7 @@
> +
> +int kgdbeth_event(struct notifier_block * self, unsigned long val, void *
> data) +{
> -+ if (strcmp(((struct net_device *)data)->name, kgdb_netdevname)) {
> ++ if (strcmp(((struct net_device *)data)->name, kgdb_netdev_name)) {
> + goto out;
> + }
> + if (val!= NETDEV_UP)
> @@ -2740,8 +2787,8 @@
> +__setup("kgdbeth=", kgdbeth_opt);
> Index: linux/include/asm-i386/kgdb.h
> ===================================================================
> ---- linux.orig/include/asm-i386/kgdb.h 2004-01-20 14:29:19.000000000 +0100
> -+++ linux/include/asm-i386/kgdb.h 2004-01-20 14:29:19.000000000 +0100
> +--- linux.orig/include/asm-i386/kgdb.h 2004-01-23 09:05:16.000000000 +0100
> ++++ linux/include/asm-i386/kgdb.h 2004-01-23 09:05:17.000000000 +0100
> @@ -0,0 +1,49 @@
> +#ifndef _ASM_KGDB_H_
> +#define _ASM_KGDB_H_

--
Amit Kale
EmSysSoft (http://www.emsyssoft.com)
KGDB: Linux Kernel Source Level Debugger (http://kgdb.sourceforge.net)

-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/