All of lore.kernel.org
 help / color / mirror / Atom feed
From: Sasha Levin <levinsasha928@gmail.com>
To: Josh Triplett <josh@joshtriplett.org>
Cc: Tejun Heo <tj@kernel.org>,
	torvalds@linux-foundation.org, akpm@linux-foundation.org,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org,
	paul.gortmaker@windriver.com
Subject: Re: [RFC 1/4] hashtable: introduce a small and naive hashtable
Date: Thu, 02 Aug 2012 13:23:39 +0200	[thread overview]
Message-ID: <501A633B.3010509@gmail.com> (raw)
In-Reply-To: <20120802103244.GA23318@leaf>

On 08/02/2012 12:32 PM, Josh Triplett wrote:
> On Thu, Aug 02, 2012 at 12:00:33PM +0200, Sasha Levin wrote:
>> On 08/02/2012 12:45 AM, Tejun Heo wrote:
>>> On Thu, Aug 02, 2012 at 12:41:56AM +0200, Sasha Levin wrote:
>>>> How would your DEFINE_HASHTABLE look like if we got for the simple
>>>> 'struct hash_table' approach?
>>>
>>> I think defining a different enclosing anonymous struct which the
>>> requested number of array entries and then aliasing the actual
>>> hash_table to that symbol should work.  It's rather horrible and I'm
>>> not sure it's worth the trouble.
>>
>> I agree that this is probably not worth the trouble.
>>
>> At the moment I see two alternatives:
>>
>> 1. Dynamically allocate the hash buckets.
>>
>> 2. Use the first bucket to store size. Something like the follows:
>>
>> 	#define HASH_TABLE(name, bits)	\
>>         	struct hlist_head name[1 << bits + 1];
>>
>> 	#define HASH_TABLE_INIT (bits) ({name[0].next = bits});
>>
>> And then have hash_{add,get} just skip the first bucket.
>>
>>
>> While it's not a pretty hack, I don't see a nice way to avoid having to dynamically allocate buckets for all cases.
> 
> What about using a C99 flexible array member?  Kernel style prohibits
> variable-length arrays, but I don't think the same rationale applies to
> flexible array members.
> 
> struct hash_table {
>     size_t count;
>     struct hlist_head buckets[];
> };
> 
> #define DEFINE_HASH_TABLE(name, length) struct hash_table name = { .count = length, .buckets = { [0 ... (length - 1)] = HLIST_HEAD_INIT } }

The limitation of this approach is that the struct hash_table variable must be 'static', which is a bit limiting - see for example the use of hashtable in 'struct user_namespace'.

> 
> - Josh Triplett
> 

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

WARNING: multiple messages have this Message-ID (diff)
From: Sasha Levin <levinsasha928@gmail.com>
To: Josh Triplett <josh@joshtriplett.org>
Cc: Tejun Heo <tj@kernel.org>,
	torvalds@linux-foundation.org, akpm@linux-foundation.org,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org,
	paul.gortmaker@windriver.com
Subject: Re: [RFC 1/4] hashtable: introduce a small and naive hashtable
Date: Thu, 02 Aug 2012 13:23:39 +0200	[thread overview]
Message-ID: <501A633B.3010509@gmail.com> (raw)
In-Reply-To: <20120802103244.GA23318@leaf>

On 08/02/2012 12:32 PM, Josh Triplett wrote:
> On Thu, Aug 02, 2012 at 12:00:33PM +0200, Sasha Levin wrote:
>> On 08/02/2012 12:45 AM, Tejun Heo wrote:
>>> On Thu, Aug 02, 2012 at 12:41:56AM +0200, Sasha Levin wrote:
>>>> How would your DEFINE_HASHTABLE look like if we got for the simple
>>>> 'struct hash_table' approach?
>>>
>>> I think defining a different enclosing anonymous struct which the
>>> requested number of array entries and then aliasing the actual
>>> hash_table to that symbol should work.  It's rather horrible and I'm
>>> not sure it's worth the trouble.
>>
>> I agree that this is probably not worth the trouble.
>>
>> At the moment I see two alternatives:
>>
>> 1. Dynamically allocate the hash buckets.
>>
>> 2. Use the first bucket to store size. Something like the follows:
>>
>> 	#define HASH_TABLE(name, bits)	\
>>         	struct hlist_head name[1 << bits + 1];
>>
>> 	#define HASH_TABLE_INIT (bits) ({name[0].next = bits});
>>
>> And then have hash_{add,get} just skip the first bucket.
>>
>>
>> While it's not a pretty hack, I don't see a nice way to avoid having to dynamically allocate buckets for all cases.
> 
> What about using a C99 flexible array member?  Kernel style prohibits
> variable-length arrays, but I don't think the same rationale applies to
> flexible array members.
> 
> struct hash_table {
>     size_t count;
>     struct hlist_head buckets[];
> };
> 
> #define DEFINE_HASH_TABLE(name, length) struct hash_table name = { .count = length, .buckets = { [0 ... (length - 1)] = HLIST_HEAD_INIT } }

The limitation of this approach is that the struct hash_table variable must be 'static', which is a bit limiting - see for example the use of hashtable in 'struct user_namespace'.

> 
> - Josh Triplett
> 


  reply	other threads:[~2012-08-02 11:23 UTC|newest]

Thread overview: 75+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2012-07-31 18:05 [RFC 0/4] generic hashtable implementation Sasha Levin
2012-07-31 18:05 ` Sasha Levin
2012-07-31 18:05 ` [RFC 1/4] hashtable: introduce a small and naive hashtable Sasha Levin
2012-07-31 18:05   ` Sasha Levin
2012-07-31 18:23   ` Tejun Heo
2012-07-31 18:23     ` Tejun Heo
2012-07-31 20:31     ` Sasha Levin
2012-07-31 20:31       ` Sasha Levin
2012-08-01 18:19     ` Sasha Levin
2012-08-01 18:19       ` Sasha Levin
2012-08-01 18:21       ` Tejun Heo
2012-08-01 18:21         ` Tejun Heo
2012-08-01 18:24         ` Sasha Levin
2012-08-01 18:24           ` Sasha Levin
2012-08-01 18:27           ` Tejun Heo
2012-08-01 18:27             ` Tejun Heo
2012-08-01 19:06             ` Sasha Levin
2012-08-01 19:06               ` Sasha Levin
2012-08-01 20:24               ` Tejun Heo
2012-08-01 20:24                 ` Tejun Heo
2012-08-01 22:41                 ` Sasha Levin
2012-08-01 22:41                   ` Sasha Levin
2012-08-01 22:45                   ` Tejun Heo
2012-08-01 22:45                     ` Tejun Heo
2012-08-02 10:00                     ` Sasha Levin
2012-08-02 10:00                       ` Sasha Levin
2012-08-02 10:32                       ` Josh Triplett
2012-08-02 10:32                         ` Josh Triplett
2012-08-02 11:23                         ` Sasha Levin [this message]
2012-08-02 11:23                           ` Sasha Levin
2012-08-02 13:04                           ` Sasha Levin
2012-08-02 13:04                             ` Sasha Levin
2012-08-02 16:15                             ` Josh Triplett
2012-08-02 16:15                               ` Josh Triplett
2012-08-02 16:48                               ` Sasha Levin
2012-08-02 16:48                                 ` Sasha Levin
2012-08-02 17:44                                 ` Josh Triplett
2012-08-02 17:44                                   ` Josh Triplett
2012-08-02 17:54                                   ` Sasha Levin
2012-08-02 17:54                                     ` Sasha Levin
2012-08-02 20:41                                     ` Josh Triplett
2012-08-02 20:41                                       ` Josh Triplett
2012-08-02 21:47                                       ` Sasha Levin
2012-08-02 21:47                                         ` Sasha Levin
2012-08-03 17:59                                         ` Josh Triplett
2012-08-03 17:59                                           ` Josh Triplett
2012-08-02 16:03                           ` Eric W. Biederman
2012-08-02 16:03                             ` Eric W. Biederman
2012-08-02 16:34                             ` Sasha Levin
2012-08-02 16:34                               ` Sasha Levin
2012-08-02 16:40                               ` Eric W. Biederman
2012-08-02 16:40                                 ` Eric W. Biederman
2012-08-02 17:32                                 ` Linus Torvalds
2012-08-02 17:32                                   ` Linus Torvalds
2012-08-02 17:48                                   ` Eric Dumazet
2012-08-02 17:48                                     ` Eric Dumazet
2012-08-02 17:59                                   ` Josh Triplett
2012-08-02 17:59                                     ` Josh Triplett
2012-08-02 18:08                                     ` Linus Torvalds
2012-08-02 18:08                                       ` Linus Torvalds
2012-08-02 20:25                                       ` Josh Triplett
2012-08-02 20:25                                         ` Josh Triplett
2012-08-02 20:32                                         ` Linus Torvalds
2012-08-02 21:21                                           ` Josh Triplett
2012-08-02 21:21                                             ` Josh Triplett
2012-08-02 21:50                                             ` Linus Torvalds
2012-08-02 21:50                                               ` Linus Torvalds
2012-08-02  9:35             ` Josh Triplett
2012-08-02  9:35               ` Josh Triplett
2012-07-31 18:05 ` [RFC 2/4] user_ns: use new hashtable implementation Sasha Levin
2012-07-31 18:05   ` Sasha Levin
2012-07-31 18:05 ` [RFC 3/4] mm,ksm: " Sasha Levin
2012-07-31 18:05   ` Sasha Levin
2012-07-31 18:05 ` [RFC 4/4] workqueue: " Sasha Levin
2012-07-31 18:05   ` Sasha Levin

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=501A633B.3010509@gmail.com \
    --to=levinsasha928@gmail.com \
    --cc=akpm@linux-foundation.org \
    --cc=josh@joshtriplett.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=paul.gortmaker@windriver.com \
    --cc=tj@kernel.org \
    --cc=torvalds@linux-foundation.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body.
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.