linux-kernel - Re: [PATCH v4 3/3] fs/file.c: add fast path in find_next

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

Message-ID: <398cd906-c31a-465b-9400-d8d81a3cf049@intel.com>
Date: Tue, 16 Jul 2024 20:37:04 +0800
From: "Ma, Yu" <yu.ma@...el.com>
To: Jan Kara <jack@...e.cz>
Cc: viro@...iv.linux.org.uk, brauner@...nel.org, mjguzik@...il.com,
 edumazet@...gle.com, linux-fsdevel@...r.kernel.org,
 linux-kernel@...r.kernel.org, pan.deng@...el.com, tianyou.li@...el.com,
 tim.c.chen@...el.com, tim.c.chen@...ux.intel.com, yu.ma@...el.com
Subject: Re: [PATCH v4 3/3] fs/file.c: add fast path in find_next_fd()


On 7/16/2024 7:19 PM, Jan Kara wrote:
> On Fri 12-07-24 22:39:17, Yu Ma wrote:
>> Skip 2-levels searching via find_next_zero_bit() when there is free slot in the
>> word contains next_fd, as:
>> (1) next_fd indicates the lower bound for the first free fd.
>> (2) There is fast path inside of find_next_zero_bit() when size<=64 to speed up
>> searching.
>> (3) After fdt is expanded (the bitmap size doubled for each time of expansion),
>> it would never be shrunk. The search size increases but there are few open fds
>> available here.
>>
>> This fast path is proposed by Mateusz Guzik <mjguzik@...il.com>, and agreed by
>> Jan Kara <jack@...e.cz>, which is more generic and scalable than previous
>> versions. And on top of patch 1 and 2, it improves pts/blogbench-1.1.0 read by
>> 8% and write by 4% on Intel ICX 160 cores configuration with v6.10-rc7.
>>
>> Reviewed-by: Tim Chen <tim.c.chen@...ux.intel.com>
>> Signed-off-by: Yu Ma <yu.ma@...el.com>
> Looks good. Just some code style nits below.

Copy that, thanks Honza, I'll revise and send out updated version soon.


>
>> diff --git a/fs/file.c b/fs/file.c
>> index 1be2a5bcc7c4..a3ce6ba30c8c 100644
>> --- a/fs/file.c
>> +++ b/fs/file.c
>> @@ -488,9 +488,20 @@ struct files_struct init_files = {
>>   
>>   static unsigned int find_next_fd(struct fdtable *fdt, unsigned int start)
>>   {
>> +	unsigned int bitbit = start / BITS_PER_LONG;
>> +	unsigned int bit;
>> +
>> +	/*
>> +	 * Try to avoid looking at the second level bitmap
>> +	 */
>> +	bit = find_next_zero_bit(&fdt->open_fds[bitbit], BITS_PER_LONG,
>> +				 start & (BITS_PER_LONG -1));
> 							^^ Either
> (BITS_PER_LONG-1) or (BITS_PER_LONG - 1) please. Your combination looks
> particularly weird :)
>
>> +	if (bit < BITS_PER_LONG) {
>> +		return bit + bitbit * BITS_PER_LONG;
>> +	}
> No need for braces around the above block.
>
>>   	unsigned int maxfd = fdt->max_fds; /* always multiple of BITS_PER_LONG */
>>   	unsigned int maxbit = maxfd / BITS_PER_LONG;
> We keep declarations at the beginning of the block. Usually it keeps the
> code more readable and the compiler should be clever enough to perform the
> loads & arithmetics only when needed.
>
> After fixing these style nits feel free to add:
>
> Reviewed-by: Jan Kara <jack@...e.cz>
>
> 								Honza

Yes, I'll polish this part of code accordingly, thanks for all the 
comments here :)