All of lore.kernel.org
 help / color / mirror / Atom feed
From: Kalle Valo <kvalo@kernel.org>
To: Baochen Qiang <quic_bqiang@quicinc.com>
Cc: <ath12k@lists.infradead.org>,  <linux-wireless@vger.kernel.org>
Subject: Re: [PATCH 1/8] wifi: ath12k: Add MLO station state change handling
Date: Wed, 20 Nov 2024 21:32:50 +0200	[thread overview]
Message-ID: <87bjy9zonx.fsf@kernel.org> (raw)
In-Reply-To: <97f3f465-6ebd-4ca1-b672-4c8c7f42220d@quicinc.com> (Baochen Qiang's message of "Wed, 13 Nov 2024 10:55:22 +0800")

Baochen Qiang <quic_bqiang@quicinc.com> writes:

>>>> +static void ath12k_mac_free_unassign_link_sta(struct ath12k_hw *ah,
>>>> +					      struct ath12k_sta *ahsta,
>>>> +					      u8 link_id)
>>>> +{
>>>> +	struct ath12k_link_sta *arsta;
>>>> +
>>>> +	lockdep_assert_wiphy(ah->hw->wiphy);
>>>> +
>>>> +	if (WARN_ON(link_id >= IEEE80211_MLD_MAX_NUM_LINKS))
>>>> +		return;
>>>> +
>>>> +	arsta = wiphy_dereference(ah->hw->wiphy, ahsta->link[link_id]);
>>>> +
>>>> +	if (WARN_ON(!arsta))
>>>> +		return;
>>>> +
>>>> +	ath12k_mac_unassign_link_sta(ah, ahsta, link_id);
>>>> +
>>>> +	arsta->link_id = ATH12K_INVALID_LINK_ID;
>>>> +	arsta->ahsta = NULL;
>>>> +	arsta->arvif = NULL;
>>>
>>> if arsta is not deflink and would be freed, can we avoid these
>>> cleanup?
>> 
>> I think that's something we can cleanup later if needed. Sure, it's
>> extra assignments but it's not really doing any harm.
> exactly, but ideally we should avoid unnecessary effort if possible.
>
>> 
>>>> +	if (arsta != &ahsta->deflink)
>>>> +		kfree(arsta);
>>>
>>> I know the actual free happens here, but why split them?
>> 
>> You mean why have a separate function ath12k_mac_unassign_link_sta() and
>> instead just have all code the in
>> ath12k_mac_free_unassign_link_sta()?
>
> yes. such that we can have synchronize_rcu() and kfree() located together.

Ok, I think I now get what you mean. Does this look better:

static void ath12k_mac_free_unassign_link_sta(struct ath12k_hw *ah,
					      struct ath12k_sta *ahsta,
					      u8 link_id)
{
	struct ath12k_link_sta *arsta;

	lockdep_assert_wiphy(ah->hw->wiphy);

	if (WARN_ON(link_id >= IEEE80211_MLD_MAX_NUM_LINKS))
		return;

	arsta = wiphy_dereference(ah->hw->wiphy, ahsta->link[link_id]);
	if (WARN_ON(!arsta))
		return;

	ahsta->links_map &= ~BIT(link_id);
	rcu_assign_pointer(ahsta->link[link_id], NULL);

	if (arsta == &ahsta->deflink) {
		arsta->link_id = ATH12K_INVALID_LINK_ID;
		arsta->ahsta = NULL;
		arsta->arvif = NULL;
		return;
	}

	synchronize_rcu();
	kfree(arsta);
}

BTW your lines are really long, please check your settings:

https://patchwork.kernel.org/project/linux-wireless/patch/20241106142617.660901-2-kvalo@kernel.org/

-- 
https://patchwork.kernel.org/project/linux-wireless/list/

https://wireless.wiki.kernel.org/en/developers/documentation/submittingpatches


  reply	other threads:[~2024-11-20 19:32 UTC|newest]

Thread overview: 32+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2024-11-06 14:26 [PATCH 0/8] [0/8] wifi: ath12k: MLO support part 3 Kalle Valo
2024-11-06 14:26 ` [PATCH 1/8] wifi: ath12k: Add MLO station state change handling Kalle Valo
2024-11-06 16:33   ` Jeff Johnson
2024-11-07  2:26   ` Baochen Qiang
2024-11-12 16:03     ` Kalle Valo
2024-11-13  2:33       ` Baochen Qiang
2024-11-20 15:57         ` Kalle Valo
2024-11-07  2:45   ` Baochen Qiang
2024-11-12 17:10     ` Kalle Valo
2024-11-13  2:55       ` Baochen Qiang
2024-11-20 19:32         ` Kalle Valo [this message]
2024-11-21  3:35           ` Baochen Qiang
2024-11-21 14:51             ` Kalle Valo
2024-11-06 14:26 ` [PATCH 2/8] wifi: ath12k: support change_sta_links() mac80211 op Kalle Valo
2024-11-06 16:33   ` Jeff Johnson
2024-11-07  7:14   ` Baochen Qiang
2024-11-12 16:55     ` Kalle Valo
2024-11-12 17:11       ` Kalle Valo
2024-11-06 14:26 ` [PATCH 3/8] wifi: ath12k: add primary link for data path operations Kalle Valo
2024-11-06 16:33   ` Jeff Johnson
2024-11-06 14:26 ` [PATCH 4/8] wifi: ath12k: use arsta instead of sta Kalle Valo
2024-11-06 16:33   ` Jeff Johnson
2024-11-06 14:26 ` [PATCH 5/8] wifi: ath12k: add reo queue lookup table for ML peers Kalle Valo
2024-11-06 16:33   ` Jeff Johnson
2024-11-06 14:26 ` [PATCH 6/8] wifi: ath12k: modify chanctx iterators for MLO Kalle Valo
2024-11-06 16:33   ` Jeff Johnson
2024-11-06 14:26 ` [PATCH 7/8] wifi: ath12k: Use mac80211 vif's link_conf instead of bss_conf Kalle Valo
2024-11-06 16:33   ` Jeff Johnson
2024-11-06 14:26 ` [PATCH 8/8] wifi: ath12k: Use mac80211 sta's link_sta instead of deflink Kalle Valo
2024-11-06 16:33   ` Jeff Johnson
  -- strict thread matches above, loose matches on Subject: below --
2024-11-21 15:57 [PATCH 0/8] [0/8] wifi: ath12k: MLO support part 3 Kalle Valo
2024-11-21 15:57 ` [PATCH 1/8] wifi: ath12k: Add MLO station state change handling Kalle Valo
2024-11-25 17:03   ` Kalle Valo

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=87bjy9zonx.fsf@kernel.org \
    --to=kvalo@kernel.org \
    --cc=ath12k@lists.infradead.org \
    --cc=linux-wireless@vger.kernel.org \
    --cc=quic_bqiang@quicinc.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
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.