Received-SPF: pass (google.com: domain of linux-kernel-owner@vger.kernel.org designates 2620:137:e000::1:20 as permitted sender) client-ip=2620:137:e000::1:20;
Message-ID: <52da839c-f23f-3341-961d-a253003bfbb3@bytedance.com>
Date:   Tue, 16 May 2023 08:39:29 +0800
MIME-Version: 1.0
User-Agent: Mozilla/5.0 (Macintosh; Intel Mac OS X 10.15; rv:102.0)
 Gecko/20100101 Thunderbird/102.10.1
Subject: Re: [PATCH 02/10] maple_tree: Drop mtree_alloc_{range,rrange}() and
 related functions.
To:     "Liam R. Howlett" <Liam.Howlett@Oracle.com>
References: <20230515131757.60035-1-zhangpeng.00@bytedance.com>
 <20230515131757.60035-3-zhangpeng.00@bytedance.com>
 <ZGJrketUDnGkWw8b@casper.infradead.org>
 <20230515173516.5tmytjvnzvdogfm7@revolver>
Cc:     Matthew Wilcox <willy@infradead.org>,
        Peng Zhang <zhangpeng.00@bytedance.com>,
        akpm@linux-foundation.org, maple-tree@lists.infradead.org,
        linux-kernel@vger.kernel.org, linux-mm@kvack.org
From:   Peng Zhang <zhangpeng.00@bytedance.com>
In-Reply-To: <20230515173516.5tmytjvnzvdogfm7@revolver>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit
Precedence: bulk


在 2023/5/16 01:35, Liam R. Howlett 写道:
> * Matthew Wilcox <willy@infradead.org> [230515 13:27]:
>> On Mon, May 15, 2023 at 09:17:49PM +0800, Peng Zhang wrote:
>>> Drop mtree_alloc_{range,rrange}(), mas_{rev_}alloc() and mas_fill_gap().
>>> mtree_alloc_{range,rrange}() currently have no users and can be easily
>>> implemented with mas_empty_area{_rev}(). mas_{rev_}alloc() and
>>> mas_fill_gap() are just their internal functions, drop them together.
>>
>> No, I think this is the wrong way to go.
>>
>> Most users should not be using the mas_* API.  These are the advanced
>> APIs.  Most users will want to use mtree_alloc_range().  Just like most
>> users of the XArray use the xa_insert() API rather than open-coding the
>> xa_state and calling the xas_* APIs.
>>
>> Please read Documentation/core-api/xarray.rst and maple_tree.rst for more
>> details on Normal vs Advanced API.  The real problem is that we have so
>> few users today, so you haven't seen that most people don't want to use
>> the advanced APIs.
> 
> 
> Peng, Apologies on the confusion.  Please do as Matthew said.  If you
> have a way to unify the functionality to use the same internal
> functions, then I think that would be a welcome change.
> 
I will implement new mtree_alloc_{range,rrange}() using other internal 
functions.