DPDK patches and discussions
 help / color / mirror / Atom feed
From: "Medvedkin, Vladimir" <vladimir.medvedkin@intel.com>
To: "Ananyev, Konstantin" <konstantin.ananyev@intel.com>,
	"dev@dpdk.org" <dev@dpdk.org>
Cc: "Chilikin, Andrey" <andrey.chilikin@intel.com>,
	"Kinsella, Ray" <ray.kinsella@intel.com>,
	"Wang, Yipeng1" <yipeng1.wang@intel.com>,
	"Gobriel, Sameh" <sameh.gobriel@intel.com>,
	"Richardson, Bruce" <bruce.richardson@intel.com>
Subject: Re: [dpdk-dev] [PATCH v2 2/3] hash: add predictable RSS implementation
Date: Tue, 13 Apr 2021 15:28:08 +0300
Message-ID: <8cb12932-aba3-66f2-4560-f7baffe950ea@intel.com> (raw)
In-Reply-To: <DM6PR11MB4491CE573779470E934BE7C49A709@DM6PR11MB4491.namprd11.prod.outlook.com>

Hi Konstantin,

On 12/04/2021 12:47, Ananyev, Konstantin wrote:
>>
>> <snip>
>>
>>>> +#defineRETA_SZ_MIN2U
>>>> +#defineRETA_SZ_MAX16U
>>>
>>> Should these RETA_SZ defines be in public header?
>>> So user can know what are allowed values?
>>>
>>
>> I don't think this is necessary, because the user chooses it not
>> arbitrary, but depending on the NIC.
> 
> Sure thing, but it would be goo for the user to know what are the SW
> Limitations on it without digging through .c.
> 

I see, I'll move it to .h

>>
>>>> +#define RETA_SZ_IN_RANGE(reta_sz)((reta_sz >= RETA_SZ_MIN) && \
>>
>> <snip>
>>
>>>> +uint32_t i;
>>>
>>> Empty line is  missing.
>>>
>>
>> Thanks
>>
>>>> +if ((name == NULL) || (key_len == 0) || !RETA_SZ_IN_RANGE(reta_sz)) {
>>>> +rte_errno = EINVAL;
>>>> +return NULL;
>>>> +}
>>
>> <snip>
>>
>>>> +static inline void
>>>> +set_bit(uint8_t *ptr, uint32_t bit, uint32_t pos)
>>>> +{
>>>> +uint32_t byte_idx = pos >> 3;
>>>
>>> Just as a nit to be consistent with the line below:
>>> pos / CHAR_BIT;
>>>
>>
>> Fixed
>>
>>>> +uint32_t bit_idx = (CHAR_BIT - 1) - (pos & (CHAR_BIT - 1));
>>>> +uint8_t tmp;
>>
>> <snip>
>>
>>>> +ent = rte_zmalloc(NULL, sizeof(struct rte_thash_subtuple_helper) +
>>>> +sizeof(uint32_t) * (1 << ctx->reta_sz_log), 0);
>>>
>>> Helper can be used by data-path code (via rte_thash_get_compliment()) right?
>>> Then might be better to align it at cache-line.
>>>
>>
>> Agree, I'll fix it
>>
>>>> +if (ent == NULL)
>>>> +return -ENOMEM;
>>
>> <snip>
>>
>>>>    uint32_t
>>>> -rte_thash_get_compliment(struct rte_thash_subtuple_helper *h __rte_unused,
>>>> -uint32_t hash __rte_unused, uint32_t desired_hash __rte_unused)
>>>> +rte_thash_get_compliment(struct rte_thash_subtuple_helper *h,
>>>> +uint32_t hash, uint32_t desired_hash)
>>>>    {
>>>> -return 0;
>>>> +return h->compl_table[(hash ^ desired_hash) & h->lsb_msk];
>>>>    }
>>>
>>> Would it make sense to add another-one for multi values:
>>> rte_thash_get_compliment(uint32_t hash, const uint32_t desired_hashes[], uint32_t adj_hash[], uint32_t num);
>>> So user can get adjustment values for multiple queues at once?
>>>
>>
>> At the moment I can't find scenarios why do we need to have a bulk
>> version for this function
> 
> My thought was about case when number of configured
> HW queues is less than reta_size.
> Let say reta_size==4, but user configured only 3 queues and reta={0,1,2,0}.
> In that case for queue 0, both 0 and 3 values would suit.
> 

Ah, yes, I remember out discussion about this. I'll think about this and 
add next releases.
I should be kind of

int
rte_thash_get_compliment_reta(struct rte_thash_subtuple_helper *helper, 
uint32_t hash, const uint16_t queue_id, struct rte_eth_rss_reta_entry64 
*reta, int reta_sz);

where reta is completed by rte_eth_dev_rss_reta_query()


-- 
Regards,
Vladimir

  reply	other threads:[~2021-04-13 12:28 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-16 18:24 [dpdk-dev] [PATCH v1 0/3] Predictable RSS feature Vladimir Medvedkin
2021-03-16 18:24 ` [dpdk-dev] [PATCH v1 1/3] hash: add predictable RSS API Vladimir Medvedkin
2021-03-16 18:24 ` [dpdk-dev] [PATCH v1 2/3] hash: add predictable RSS implementation Vladimir Medvedkin
2021-03-16 18:24 ` [dpdk-dev] [PATCH v1 3/3] test/hash: add additional thash tests Vladimir Medvedkin
2021-04-06 19:50 ` [dpdk-dev] [PATCH v2 0/3] Predictable RSS feature Vladimir Medvedkin
2021-04-08 15:56   ` Stephen Hemminger
2021-04-11 18:51     ` Medvedkin, Vladimir
2021-04-10  0:32   ` Wang, Yipeng1
2021-04-11 18:51     ` Medvedkin, Vladimir
2021-04-11 19:11   ` [dpdk-dev] [PATCH v3 " Vladimir Medvedkin
2021-04-13 13:19     ` [dpdk-dev] [PATCH v4 " Vladimir Medvedkin
2021-04-14 18:04       ` Wang, Yipeng1
2021-04-15  8:29         ` David Marchand
2021-04-19 15:59       ` [dpdk-dev] [PATCH v5 0/5] " Vladimir Medvedkin
2021-04-20 21:31         ` Thomas Monjalon
2021-04-19 15:59       ` [dpdk-dev] [PATCH v5 1/5] hash: add predictable RSS API Vladimir Medvedkin
2021-04-19 15:59       ` [dpdk-dev] [PATCH v5 2/5] hash: add predictable RSS implementation Vladimir Medvedkin
2021-04-19 15:59       ` [dpdk-dev] [PATCH v5 3/5] test/hash: add additional thash tests Vladimir Medvedkin
2021-04-29  9:13         ` David Marchand
2021-04-29  9:17           ` Medvedkin, Vladimir
2021-04-29 18:45             ` Stanislaw Kardach
2021-05-04 14:06               ` Medvedkin, Vladimir
2021-04-19 15:59       ` [dpdk-dev] [PATCH v5 4/5] doc: add thash documentation Vladimir Medvedkin
2021-04-20 11:25         ` Mcnamara, John
2021-04-19 15:59       ` [dpdk-dev] [PATCH v5 5/5] maintainers: claim maintainership of the hash library Vladimir Medvedkin
2021-04-13 13:19     ` [dpdk-dev] [PATCH v4 1/3] hash: add predictable RSS API Vladimir Medvedkin
2021-04-14 17:06       ` Wang, Yipeng1
2021-04-13 13:19     ` [dpdk-dev] [PATCH v4 2/3] hash: add predictable RSS implementation Vladimir Medvedkin
2021-04-14 17:51       ` Wang, Yipeng1
2021-04-13 13:19     ` [dpdk-dev] [PATCH v4 3/3] test/hash: add additional thash tests Vladimir Medvedkin
2021-04-14 17:56       ` Wang, Yipeng1
2021-04-11 19:11   ` [dpdk-dev] [PATCH v3 1/3] hash: add predictable RSS API Vladimir Medvedkin
2021-04-11 19:11   ` [dpdk-dev] [PATCH v3 2/3] hash: add predictable RSS implementation Vladimir Medvedkin
2021-04-11 19:11   ` [dpdk-dev] [PATCH v3 3/3] test/hash: add additional thash tests Vladimir Medvedkin
2021-04-06 19:50 ` [dpdk-dev] [PATCH v2 1/3] hash: add predictable RSS API Vladimir Medvedkin
2021-04-10  0:05   ` Wang, Yipeng1
2021-04-11 18:52     ` Medvedkin, Vladimir
2021-04-06 19:50 ` [dpdk-dev] [PATCH v2 2/3] hash: add predictable RSS implementation Vladimir Medvedkin
2021-04-07 12:53   ` Ananyev, Konstantin
2021-04-11 18:51     ` Medvedkin, Vladimir
2021-04-12  9:47       ` Ananyev, Konstantin
2021-04-13 12:28         ` Medvedkin, Vladimir [this message]
2021-04-10  0:10   ` Wang, Yipeng1
2021-04-11 18:52     ` Medvedkin, Vladimir
2021-04-06 19:50 ` [dpdk-dev] [PATCH v2 3/3] test/hash: add additional thash tests Vladimir Medvedkin

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=8cb12932-aba3-66f2-4560-f7baffe950ea@intel.com \
    --to=vladimir.medvedkin@intel.com \
    --cc=andrey.chilikin@intel.com \
    --cc=bruce.richardson@intel.com \
    --cc=dev@dpdk.org \
    --cc=konstantin.ananyev@intel.com \
    --cc=ray.kinsella@intel.com \
    --cc=sameh.gobriel@intel.com \
    --cc=yipeng1.wang@intel.com \
    /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

DPDK patches and discussions

This inbox may be cloned and mirrored by anyone:

	git clone --mirror https://inbox.dpdk.org/dev/0 dev/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 dev dev/ https://inbox.dpdk.org/dev \
		dev@dpdk.org
	public-inbox-index dev

Example config snippet for mirrors.
Newsgroup available over NNTP:
	nntp://inbox.dpdk.org/inbox.dpdk.dev


AGPL code for this site: git clone https://public-inbox.org/public-inbox.git