Re: [Xen-devel] [PATCH V2 3/6] [RFC] xen/common: Introduce _xrealloc function

From: Oleksandr <olekstysh@gmail.com>
To: Jan Beulich <JBeulich@suse.com>
Cc: "sstabellini@kernel.org" <sstabellini@kernel.org>,
	Wei Liu <wl@xen.org>,
	Konrad Rzeszutek Wilk <konrad.wilk@oracle.com>,
	George Dunlap <George.Dunlap@eu.citrix.com>,
	Andrew Cooper <andrew.cooper3@citrix.com>,
	Ian Jackson <ian.jackson@eu.citrix.com>, Tim Deegan <tim@xen.org>,
	Oleksandr Tyshchenko <oleksandr_tyshchenko@epam.com>,
	"julien.grall@arm.com" <julien.grall@arm.com>,
	"xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>
Subject: Re: [Xen-devel] [PATCH V2 3/6] [RFC] xen/common: Introduce _xrealloc function
Date: Tue, 6 Aug 2019 21:50:12 +0300	[thread overview]
Message-ID: <760df686-ef36-e400-4ef3-c51334694b68@gmail.com> (raw)
In-Reply-To: <921ff67c-5a61-9f43-965e-e1b2075c2280@suse.com>

On 05.08.19 13:02, Jan Beulich wrote:

Hi, Jan

> While I can see why having a re-allocation function may be handy,
> explicit / direct use of _xmalloc() and _xzalloc() are discouraged,
> in favor of the more type-safe underscore-less variants.

took into account

> I can't
> see though how a type-safe "realloc" could look like, except for
> arrays. If resizing arrays is all you're after, I'd like to
> recommend to go that route rather then the suggested one here. If
> resizing arbitrary objects is the goal, then what you suggest may
> be the only route, but I'd still be not overly happy to see such
> added.

My main goal is to get "ported" from Linux "iommu_fwspec" support 
(xrealloc user) in [1].

I tried to retain code as much as possible while porting. So, this patch 
adds almost the same thing what the ported code expects.

But, I would be OK to consider modifying a code in a way to resize an 
array as well as any other variants if present.

>
> Furthermore you don't even use internals of the allocator: It is
> common practice to avoid re-allocation if the requested size fits
> within the already allocated block. That's not the least helpful
> because in such a case you can't possibly suffer any -ENOMEM
> condition.

agree, took into account as well.

>
> And finally - please note _xmalloc()'s and _xfree()'s use /
> special casing of ZERO_BLOCK_PTR: You absolutely would need to
> mirror this here.

got it, will use for zero-size allocation

[1] 
https://lists.xenproject.org/archives/html/xen-devel/2019-08/msg00257.html

Thank you.

-- 
Regards,

Oleksandr Tyshchenko

_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xenproject.org
https://lists.xenproject.org/mailman/listinfo/xen-devel