Received-SPF: pass (google.com: domain of linux-kernel+bounces-147969-linux.lists.archive=gmail.com@vger.kernel.org designates 147.75.48.161 as permitted sender) client-ip=147.75.48.161;
Message-ID: <63d9540f-61bc-4eba-819c-c05d2e486bdb@linux.dev>
Date: Wed, 17 Apr 2024 13:51:21 +0800
Precedence: bulk
MIME-Version: 1.0
Subject: Re: [PATCH] bcachefs: Align the display format of
 `btrees/inodes/keys`
To: Kent Overstreet <kent.overstreet@linux.dev>
Cc: Hongbo Li <lihongbo22@huawei.com>, linux-bcachefs@vger.kernel.org,
 linux-kernel@vger.kernel.org, Youling Tang <tangyouling@kylinos.cn>
References: <20240417015448.84294-1-youling.tang@linux.dev>
 <toosz4pt2la6wf2575vwe6efebcbzsy735cq7exmlqhzrgicrf@s7g3rrbvge2l>
 <66ecea56-e4d3-4241-a1dc-378d70555321@linux.dev>
 <fpvnushjgr5txyduvpihevvghv64sdso4p6t3nhhddcjimqh2k@ehzpswsmnxin>
 <d17de3df-a3f2-4317-92cc-1fd02641328b@huawei.com>
 <29d5ea19-21b5-4076-9acc-8286a050c33f@huawei.com>
Content-Language: en-US, en-AU
From: Youling Tang <youling.tang@linux.dev>
In-Reply-To: <29d5ea19-21b5-4076-9acc-8286a050c33f@huawei.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 8bit

Hi, Kent & Hongbo
On 17/04/2024 11:21, Hongbo Li wrote:
>
>
> On 2024/4/17 11:16, Hongbo Li wrote:
>>
>>
>> On 2024/4/17 10:59, Kent Overstreet wrote:
>>> On Wed, Apr 17, 2024 at 10:50:10AM +0800, Youling Tang wrote:
>>>> Hi, Kent
>>>> On 17/04/2024 10:20, Kent Overstreet wrote:
>>>>> On Wed, Apr 17, 2024 at 09:54:48AM +0800, Youling Tang wrote:
>>>>>> From: Youling Tang <tangyouling@kylinos.cn>
>>>>>>
>>>>>> Before patch:
>>>>>> ```
>>>>>>    #cat btrees/inodes/keys
>>>>>>    u64s 17 type inode_v3 0:4096:U32_MAX len 0 ver 0: mode=40755
>>>>>>      flags= (16300000)
>>>>>> ```
>>>>>>
>>>>>> After patch:
>>>>>> ```
>>>>>>    #cat btrees/inodes/keys
>>>>>>    u64s 17 type inode_v3 0:4096:U32_MAX len 0 ver 0:
>>>>>>      mode=40755
>>>>>>      flags= (16300000)
>> The flags also with the space after "=". Is it reseonable?
> Sorry, I misspell. I mean whether it is reasonable.
>>>>> This would print a newline for keys that don't have a value...
>>>> The original intention was to make the display of the printed 
>>>> content in
>>>> '__bch2_inode_unpacked_to_text ()' consistent, without considering 
>>>> other
>>>> callbacks.
>>>>
>>>> Or just modify it in the following way?
>>>
>>> Yeah, that's better
>>>
>>> Do it off my master branch though, there's some printbuf 
>>> imprevements in
>>> there.
>>>
>>> https://evilpiepirate.org/git/bcachefs.git
I will make the following changes based on the master branch,

--- a/fs/bcachefs/inode.c

+++ b/fs/bcachefs/inode.c
@@ -534,12 +534,13 @@ int bch2_inode_v3_invalid(struct bch_fs *c, struct 
bkey_s_c k,
  static void __bch2_inode_unpacked_to_text(struct printbuf *out,
                                           struct bch_inode_unpacked *inode)
  {
+       prt_printf(out, "\n");
         printbuf_indent_add(out, 2);
         prt_printf(out, "mode=%o\n", inode->bi_mode);

         prt_str(out, "flags=");
         prt_bitflags(out, bch2_inode_flag_strs, inode->bi_flags & ((1U 
<< 20) - 1));
-       prt_printf(out, " (%x)\n", inode->bi_flags);
+       prt_printf(out, "(%x)\n", inode->bi_flags);
>>>
>>>> --- a/fs/bcachefs/inode.c
>>>> +++ b/fs/bcachefs/inode.c
>>>> @@ -534,6 +534,8 @@ int bch2_inode_v3_invalid(struct bch_fs *c, struct
>>>> bkey_s_c k,
>>>>   static void __bch2_inode_unpacked_to_text(struct printbuf *out,
>>>> struct bch_inode_unpacked *inode)
>>>>   {
>>>> +       prt_newline(out);
>>>> +
>>>>          printbuf_indent_add(out, 2);
>>>>          prt_printf(out, "mode=%o", inode->bi_mode);
>>>>          prt_newline(out);
>>>>
>>>>
>>>> Thanks,
>>>> Youling.
>>>
>>