linux-kernel - Re: [PATCH v3 2/3] augmented rbtree: add new RB_DECLARE_CALLBACKS

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Wed, 31 Jul 2019 16:20:11 +0200
From:   Uladzislau Rezki <urezki@...il.com>
To:     Michel Lespinasse <walken@...gle.com>
Cc:     Uladzislau Rezki <urezki@...il.com>,
        Andrew Morton <akpm@...ux-foundation.org>,
        Davidlohr Bueso <dave@...olabs.net>,
        Peter Zijlstra <peterz@...radead.org>,
        David Howells <dhowells@...hat.com>,
        LKML <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH v3 2/3] augmented rbtree: add new
 RB_DECLARE_CALLBACKS_MAX macro

Hello, Michel.

> 
> Hmmm, I had not thought about that. Agree that this can be useful -
> there is already similar test code in rbtree_test.c and also
> vma_compute_subtree_gap() in mmap.c, ...
> 
> With patch 3/3 of this series, the RBCOMPUTE function (typically
> generated through the RB_DECLARE_CALLBACKS_MAX macro) will return a
> bool indicating if the node's augmented value was already correctly
> set. Maybe this can be used for test code, through in the false case,
> the node's augmented value is already overwritten with the correct
> value. Not sure if that is a problem though - the files I mentioned
> above have test code that will dump the values if there is a mismatch,
> but really I think in every realistic case just noting that there was
> one would be just as helpful as being able to dump the old (incorrect)
> value....
> 
> What do you think - is the RBCOMPUTE(node, true) function sufficient
> for such debugging ?
>
I think so, at least i do not see any issues with that. If it returns
"false" then it will indicate that the node was not correctly augmented.

Also, i see in many places across your patches there is below code:

<snip>
	RBSTRUCT *child;						      \
	RBTYPE max = RBCOMPUTE(node);					      \
	if (node->RBFIELD.rb_left) {					      \
		child = rb_entry(node->RBFIELD.rb_left, RBSTRUCT, RBFIELD);   \
		if (child->RBAUGMENTED > max)				      \
			max = child->RBAUGMENTED;			      \
	}								      \
	if (node->RBFIELD.rb_right) {					      \
		child = rb_entry(node->RBFIELD.rb_right, RBSTRUCT, RBFIELD);  \
		if (child->RBAUGMENTED > max)				      \
			max = child->RBAUGMENTED;			      \
	}								      \
	if (exit && node->RBAUGMENTED == max)				      \
		return true;						      \
	node->RBAUGMENTED = max;					      \
	return false;
<snip>

i think it can be simplified by using max3 macro. For example:

<snip>
get_subtree_max(struct rb_node *node)
{
	struct something *foo;

	va = rb_entry_safe(node, struct something, rb_node);
	return foo ? foo->subtree_max : 0;
}

compute_subtree_max_size(struct vmap_area *va)
{
	return max3(va_size(va),
		get_subtree_max_size(va->rb_node.rb_left),
		get_subtree_max_size(va->rb_node.rb_right));
}
<snip>

What do you think about that?

Thank you.

--
Vlad Rezki