From: Jeff Dike Date: Thu, 29 Mar 2007 08:20:30 +0000 (-0700) Subject: [PATCH] uml: irq locking fixes X-Git-Tag: v2.6.21-rc6~63 X-Git-Url: http://pilppa.com/gitweb/?a=commitdiff_plain;h=3076212f8d3d9fe4d7b11c6b560c0f424edfcb4e;p=linux-2.6-omap-h63xx.git [PATCH] uml: irq locking fixes As the comment immediately preceding this points out, this list is changed in irq context, so it needs to be protected with spin_lock_irqsave in process context when it is processed. Sometimes, gcc should just compile the comments and forget the code. The IRQ side of this was better, in the sense that it blocked and unblocked interrupts, but it still should have saved and restored them. Signed-off-by: Jeff Dike Cc: Paolo 'Blaisorblade' Giarrusso Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- diff --git a/arch/um/drivers/chan_kern.c b/arch/um/drivers/chan_kern.c index 7b8baf146ac..9fdfad64953 100644 --- a/arch/um/drivers/chan_kern.c +++ b/arch/um/drivers/chan_kern.c @@ -236,11 +236,11 @@ void free_irqs(void) struct chan *chan; LIST_HEAD(list); struct list_head *ele; + unsigned long flags; - spin_lock_irq(&irqs_to_free_lock); + spin_lock_irqsave(&irqs_to_free_lock, flags); list_splice_init(&irqs_to_free, &list); - INIT_LIST_HEAD(&irqs_to_free); - spin_unlock_irq(&irqs_to_free_lock); + spin_unlock_irqrestore(&irqs_to_free_lock, flags); list_for_each(ele, &list){ chan = list_entry(ele, struct chan, free_list); @@ -255,13 +255,15 @@ void free_irqs(void) static void close_one_chan(struct chan *chan, int delay_free_irq) { + unsigned long flags; + if(!chan->opened) return; if(delay_free_irq){ - spin_lock_irq(&irqs_to_free_lock); + spin_lock_irqsave(&irqs_to_free_lock, flags); list_add(&chan->free_list, &irqs_to_free); - spin_unlock_irq(&irqs_to_free_lock); + spin_unlock_irqrestore(&irqs_to_free_lock, flags); } else { if(chan->input)