Re: [PATCH 0/5] seqlock: introduce SEQLOCK_READ_SECTION()

[Oleg Nesterov]

On 10/05, Al Viro wrote:
>
> On Sun, Oct 05, 2025 at 04:49:29PM +0200, Oleg Nesterov wrote:
>
> > 	static char *__dentry_path(const struct dentry *d, struct prepend_buffer *p)
> > 	{
> > 		const struct dentry *dentry;
> > 		struct prepend_buffer b;
> >
> > 		rcu_read_lock();
> > 		__SEQLOCK_READ_SECTION(&rename_lock, lockless, seq, NULL) {
> > 			dentry = d;
> > 			b = *p;
> > 			while (!IS_ROOT(dentry)) {
> > 				const struct dentry *parent = dentry->d_parent;
> >
> > 				prefetch(parent);
> > 				if (!prepend_name(&b, &dentry->d_name))
> > 					break;
> > 				dentry = parent;
> > 			}
> > 			if (lockless)
> > 				rcu_read_unlock();
>
> <cringe>

Weird or buggy?

> prepend_path() would be interesting to look at...

Yeah, I have already looked at it.

Not sure yet what the "good" cleanup could do. Will think about it more.

Oleg.