* Problem in yenta.c, 2nd edition
@ 2001-11-02 8:39 Jan Marek
2001-11-02 9:26 ` Andrew Morton
0 siblings, 1 reply; 2+ messages in thread
From: Jan Marek @ 2001-11-02 8:39 UTC (permalink / raw)
To: linux-kernel
Hallo l-k,
the first: I'm very sorry for old post about PCMCIA: function
yenta_config_init() is called the first time from yenta_open()
and not from yenta_init(), as I think... It was my error...
I explored yenta.c through printk() function and I found the last
point, where kernel freeze: This point is here:
--- function yenta_open_bh() from drivers/pcmcia/yenta.c
/*
* 'Bottom half' for the yenta_open routine. Allocate the interrupt line
* and register the socket with the upper layers.
*/
static void yenta_open_bh(void * data)
{
pci_socket_t * socket = (pci_socket_t *) data;
/* It's OK to overwrite this now */
socket->tq_task.routine = yenta_bh;
printk("yenta_open_bh: socket->tq_task.routine\n");
if (!socket->cb_irq || request_irq(socket->cb_irq, yenta_interrupt, (SA_SHIRQ | SA_INTERRUPT), socket->dev->name, socket)) {
/* No IRQ or request_irq failed. Poll */
printk("yenta_open_bh: in the if block\n");
socket->cb_irq = 0; /* But zero is a valid IRQ number. */
socket->poll_timer.function = yenta_interrupt_wrapper;
socket->poll_timer.data = (unsigned long)socket;
socket->poll_timer.expires = jiffies + HZ;
printk("yenta_open_bh: before add_timer\n");
add_timer(&socket->poll_timer);
printk("yenta_open_bh: add_timer\n");
}
printk("yenta_open_bh: after if(!socket->cb_irq...\n");
/* Figure out what the dang thing can do for the PCMCIA layer... */
yenta_get_socket_capabilities(socket, isa_interrupts);
printk("yenta_open_bh: after yenta_get_socket_capabilities\n");
printk("Socket status: %08x\n", cb_readl(socket, CB_SOCKET_STATE));
/* Register it with the pcmcia layer.. */
cardbus_register(socket);
printk("yenta_open_bh: cardbus_register()\n");
MOD_DEC_USE_COUNT;
}
--- end of function
Last message from kernel is:
yenta_open_bh: socket->tq_task.routine
Then I mean, that problem is in the calling of request_irq(). As
you see, I tried call this function with flags (SA_SHIRQ |
SA_INTERRUPT), but this is not the right way too...
I wrote, that my Compaq has IRQ 11 shared between PCMCIA
controller and VGA card...
Please, can you cc any follow-ups to me? Thank you very much...
Sincerely
Jan Marek
--
Ing. Jan Marek
University of South Bohemia
Academic Computer Centre
Phone: +420-38-7772080
^ permalink raw reply [flat|nested] 2+ messages in thread
* Re: Problem in yenta.c, 2nd edition
2001-11-02 8:39 Problem in yenta.c, 2nd edition Jan Marek
@ 2001-11-02 9:26 ` Andrew Morton
0 siblings, 0 replies; 2+ messages in thread
From: Andrew Morton @ 2001-11-02 9:26 UTC (permalink / raw)
To: Jan Marek; +Cc: linux-kernel
Jan Marek wrote:
>
> Hallo l-k,
>
> the first: I'm very sorry for old post about PCMCIA: function
> yenta_config_init() is called the first time from yenta_open()
> and not from yenta_init(), as I think... It was my error...
>
> I explored yenta.c through printk() function and I found the last
> point, where kernel freeze: This point is here:
>
Presumably the controller is permanently requesting an
interrupt. So as soon as you enable the IRQ, you lock
up, taking infinite interrupts.
It's always seemed a bit fishy how we enable the IRQ before
initialising the socket controller.
Try moving the block of code which requests the IRQ down
so that it comes after the call to cardbus_register().
> --- function yenta_open_bh() from drivers/pcmcia/yenta.c
> /*
> * 'Bottom half' for the yenta_open routine. Allocate the interrupt line
> * and register the socket with the upper layers.
> */
> static void yenta_open_bh(void * data)
> {
> pci_socket_t * socket = (pci_socket_t *) data;
>
> /* It's OK to overwrite this now */
> socket->tq_task.routine = yenta_bh;
> printk("yenta_open_bh: socket->tq_task.routine\n");
>
>From here
> if (!socket->cb_irq || request_irq(socket->cb_irq, yenta_interrupt, (SA_SHIRQ | SA_INTERRUPT), socket->dev->name, socket)) {
> /* No IRQ or request_irq failed. Poll */
> printk("yenta_open_bh: in the if block\n");
> socket->cb_irq = 0; /* But zero is a valid IRQ number. */
> socket->poll_timer.function = yenta_interrupt_wrapper;
> socket->poll_timer.data = (unsigned long)socket;
> socket->poll_timer.expires = jiffies + HZ;
> printk("yenta_open_bh: before add_timer\n");
> add_timer(&socket->poll_timer);
> printk("yenta_open_bh: add_timer\n");
> }
to here
> printk("yenta_open_bh: after if(!socket->cb_irq...\n");
> /* Figure out what the dang thing can do for the PCMCIA layer... */
> yenta_get_socket_capabilities(socket, isa_interrupts);
> printk("yenta_open_bh: after yenta_get_socket_capabilities\n");
> printk("Socket status: %08x\n", cb_readl(socket, CB_SOCKET_STATE));
>
> /* Register it with the pcmcia layer.. */
> cardbus_register(socket);
> printk("yenta_open_bh: cardbus_register()\n");
move it down to here.
> MOD_DEC_USE_COUNT;
> }
-
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2001-11-02 9:31 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2001-11-02 8:39 Problem in yenta.c, 2nd edition Jan Marek
2001-11-02 9:26 ` Andrew Morton
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox