[slime-devel] Re: Halp, slime isn't working any more (for me)

[T Taneli Vahakangas]

On Mon, Sep 03, 2007 at 12:57:48PM +0200, Helmut Eller wrote:
> * T Taneli Vahakangas [2007-09-03 00:17+0200] writes:
> 
> > Currently I don't have time to investigate further, so I'm just
> > dumping this on the list, hoping that somebody else will take a
> > look. The kernel patch makes slight changes to how /proc files
> > are accessed, why would that have any effect on slime or sbcl?
> 
> Slime calls (unnecessarily) MACHINE-VERSION, which seems to open
> /proc/cpuinfo in SBCL.

Thanks, Helmut, for the info. In the end, I contacted Alexey
Dobriyan who introduced the bug ;) and now fixed it with the
kernel patch that follows. It applies cleanly at least against
2.6.23-rc5.

	Taneli

--- a/fs/proc/inode.c
+++ b/fs/proc/inode.c
@@ -11,6 +11,7 @@
 #include <linux/string.h>
 #include <linux/stat.h>
 #include <linux/completion.h>
+#include <linux/poll.h>
 #include <linux/file.h>
 #include <linux/limits.h>
 #include <linux/init.h>
@@ -232,7 +233,7 @@ static ssize_t proc_reg_write(struct file *file, const char __user *buf, size_t
 static unsigned int proc_reg_poll(struct file *file, struct poll_table_struct *pts)
 {
 	struct proc_dir_entry *pde = PDE(file->f_path.dentry->d_inode);
-	unsigned int rv = 0;
+	unsigned int rv = POLLIN | POLLOUT | POLLRDNORM | POLLWRNORM;
 	unsigned int (*poll)(struct file *, struct poll_table_struct *);
 
 	spin_lock(&pde->pde_unload_lock);