From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-wm0-x229.google.com (mail-wm0-x229.google.com [IPv6:2a00:1450:400c:c09::229]) (using TLSv1.2 with cipher ECDHE-RSA-AES128-GCM-SHA256 (128/128 bits)) (No client certificate requested) by lists.bufferbloat.net (Postfix) with ESMTPS id E46863B2A0 for ; Thu, 24 Mar 2016 03:49:12 -0400 (EDT) Received: by mail-wm0-x229.google.com with SMTP id p65so53798501wmp.0 for ; Thu, 24 Mar 2016 00:49:12 -0700 (PDT) DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=tieto.com; s=google; h=mime-version:in-reply-to:references:date:message-id:subject:from:to :cc:content-transfer-encoding; bh=+uPNyLSsv0RfYSNd1U+MqY6Zqp1xeh63a8B9TugXho0=; b=hPOQ/OixbF9EizOCiKqWS8nsG4bdH/Wa8CW4SH3M66ZFeRdQaHaQjmALwkkDpWuKMm +k+iFjUyo50uFy4etpypVPF4ZRGk2GWKtUSxWztg/Q0ZLvauQwriDXibAKGZe9uUz+Wc ij0DXcGby8oPbjQ2RylWwAB3AZFxdpUF985Vo= X-Google-DKIM-Signature: v=1; a=rsa-sha256; c=relaxed/relaxed; d=1e100.net; s=20130820; h=x-gm-message-state:mime-version:in-reply-to:references:date :message-id:subject:from:to:cc:content-transfer-encoding; bh=+uPNyLSsv0RfYSNd1U+MqY6Zqp1xeh63a8B9TugXho0=; b=L6DFD/B8ADCxnxfpFj4e6UOanzpx+PigQAA9iSZO3NSPfqfJpNWHREsOMFJiNWYsIx rXKBJfiUB54UrsLlxPLw1VDaviTVjpEcKggp0sFmY2wqU8qWHd4JRpEAqyL5mnQeTSIs qkbKfy1ugQbVxlBr7kSB/resfYZp6cmV9KnIba3Iwvdi2/+8Mej7h9KHcphP4qtjWTtJ glEEKoYi9eoN/e27xIdPjgpK4wV+mbcDzd/GG9LWLkrhI4GFKOPhUfeGAchxSsB0EbIo 6HfJpPnnUOYLKM5GEikx7D8g4ujbd4QdSYpccLNaVqvUM5MvxljFxceywZgTWj2Vg0Bd z0zg== X-Gm-Message-State: AD7BkJKsAUYHIcIoeJkNHvONVSRSF1z1BXGOzrwUQubcG7TPLu1IyRfnH1fYuW2yAZYbNWoWViMXo1sSYAC/JRydYVs03i10xstPWzQ7wEXZC6FDaK0CL0d2LrTQxiJD1Nxt5NcPI+WXyHYpdjdewb7Pyq3BaV0rAJgwQA== MIME-Version: 1.0 X-Received: by 10.194.81.103 with SMTP id z7mr7960848wjx.25.1458805752088; Thu, 24 Mar 2016 00:49:12 -0700 (PDT) Received: by 10.194.21.73 with HTTP; Thu, 24 Mar 2016 00:49:12 -0700 (PDT) In-Reply-To: <20160324071923.GB23410@atheros-ThinkPad-T61> References: <1458123478-1795-1-git-send-email-michal.kazior@tieto.com> <1458123478-1795-3-git-send-email-michal.kazior@tieto.com> <20160324071923.GB23410@atheros-ThinkPad-T61> Date: Thu, 24 Mar 2016 08:49:12 +0100 Message-ID: From: Michal Kazior To: Mohammed Shafi Shajakhan Cc: linux-wireless , Felix Fietkau , Emmanuel Grumbach , Network Development , Dave Taht , "ath10k@lists.infradead.org" , codel@lists.bufferbloat.net, make-wifi-fast@lists.bufferbloat.net, Johannes Berg , Tim Shepard Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: quoted-printable X-DomainID: tieto.com X-Mailman-Approved-At: Fri, 25 Mar 2016 11:52:32 -0400 Subject: Re: [Make-wifi-fast] [RFCv2 2/3] ath10k: report per-station tx/rate rates to mac80211 X-BeenThere: make-wifi-fast@lists.bufferbloat.net X-Mailman-Version: 2.1.20 Precedence: list List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Thu, 24 Mar 2016 07:49:13 -0000 On 24 March 2016 at 08:19, Mohammed Shafi Shajakhan wrote: > Hi Michal, > > On Wed, Mar 16, 2016 at 11:17:57AM +0100, Michal Kazior wrote: >> The rate control is offloaded by firmware so it's >> challanging to provide expected throughput value >> for given station. >> >> This approach is naive as it reports last tx rate >> used for given station as provided by firmware >> stat event. >> >> This should be sufficient for airtime estimation >> used for fq-codel-in-mac80211 tx scheduling >> purposes now. >> >> This patch uses a very hacky way to get the stats. >> This is sufficient for proof-of-concept but must >> be cleaned up properly eventually. >> >> Signed-off-by: Michal Kazior >> --- >> drivers/net/wireless/ath/ath10k/core.h | 5 +++ >> drivers/net/wireless/ath/ath10k/debug.c | 61 ++++++++++++++++++++++++++= +++---- >> drivers/net/wireless/ath/ath10k/mac.c | 26 ++++++++------ >> drivers/net/wireless/ath/ath10k/wmi.h | 2 +- >> 4 files changed, 76 insertions(+), 18 deletions(-) >> >> diff --git a/drivers/net/wireless/ath/ath10k/core.h b/drivers/net/wirele= ss/ath/ath10k/core.h >> index 23ba03fb7a5f..3f76669d44cf 100644 >> --- a/drivers/net/wireless/ath/ath10k/core.h >> +++ b/drivers/net/wireless/ath/ath10k/core.h >> @@ -331,6 +331,9 @@ struct ath10k_sta { >> /* protected by conf_mutex */ >> bool aggr_mode; >> u64 rx_duration; >> + >> + u32 tx_rate_kbps; >> + u32 rx_rate_kbps; >> #endif >> }; >> >> @@ -372,6 +375,8 @@ struct ath10k_vif { >> s8 def_wep_key_idx; >> >> u16 tx_seq_no; >> + u32 tx_rate_kbps; >> + u32 rx_rate_kbps; >> >> union { >> struct { >> diff --git a/drivers/net/wireless/ath/ath10k/debug.c b/drivers/net/wirel= ess/ath/ath10k/debug.c >> index 076d29b53ddf..cc7ebf04ae00 100644 >> --- a/drivers/net/wireless/ath/ath10k/debug.c >> +++ b/drivers/net/wireless/ath/ath10k/debug.c >> @@ -316,6 +316,58 @@ static void ath10k_debug_fw_stats_reset(struct ath1= 0k *ar) >> spin_unlock_bh(&ar->data_lock); >> } >> >> +static void ath10k_mac_update_txrx_rate_iter(void *data, >> + u8 *mac, >> + struct ieee80211_vif *vif) >> +{ >> + struct ath10k_fw_stats_peer *peer =3D data; >> + struct ath10k_vif *arvif; >> + >> + if (memcmp(vif->addr, peer->peer_macaddr, ETH_ALEN)) >> + return; >> + >> + arvif =3D (void *)vif->drv_priv; >> + arvif->tx_rate_kbps =3D peer->peer_tx_rate; >> + arvif->rx_rate_kbps =3D peer->peer_rx_rate; >> +} >> + >> +static void ath10k_mac_update_txrx_rate(struct ath10k *ar, >> + struct ath10k_fw_stats *stats) >> +{ >> + struct ieee80211_hw *hw =3D ar->hw; >> + struct ath10k_fw_stats_peer *peer; >> + struct ath10k_sta *arsta; >> + struct ieee80211_sta *sta; >> + const u8 *localaddr =3D NULL; >> + >> + rcu_read_lock(); >> + >> + list_for_each_entry(peer, &stats->peers, list) { >> + /* This doesn't account for multiple STA connected on diff= erent >> + * vifs. Unfortunately there's no way to derive that from = the available >> + * information. >> + */ >> + sta =3D ieee80211_find_sta_by_ifaddr(hw, >> + peer->peer_macaddr, >> + localaddr); >> + if (!sta) { >> + /* This tries to update multicast rates */ >> + ieee80211_iterate_active_interfaces_atomic( >> + hw, >> + IEEE80211_IFACE_ITER_NORMAL, >> + ath10k_mac_update_txrx_rate_iter, >> + peer); >> + continue; >> + } >> + >> + arsta =3D (void *)sta->drv_priv; >> + arsta->tx_rate_kbps =3D peer->peer_tx_rate; >> + arsta->rx_rate_kbps =3D peer->peer_rx_rate; >> + } >> + >> + rcu_read_unlock(); >> +} >> + >> void ath10k_debug_fw_stats_process(struct ath10k *ar, struct sk_buff *s= kb) >> { >> struct ath10k_fw_stats stats =3D {}; >> @@ -335,6 +387,8 @@ void ath10k_debug_fw_stats_process(struct ath10k *ar= , struct sk_buff *skb) >> goto free; >> } >> >> + ath10k_mac_update_txrx_rate(ar, &stats); >> + >> /* Stat data may exceed htc-wmi buffer limit. In such case firmwar= e >> * splits the stats data and delivers it in a ping-pong fashion of >> * request cmd-update event. >> @@ -351,13 +405,6 @@ void ath10k_debug_fw_stats_process(struct ath10k *a= r, struct sk_buff *skb) >> if (peer_stats_svc) >> ath10k_sta_update_rx_duration(ar, &stats.peers); >> >> - if (ar->debug.fw_stats_done) { >> - if (!peer_stats_svc) >> - ath10k_warn(ar, "received unsolicited stats update= event\n"); >> - >> - goto free; >> - } >> - > > [shafi] As you had suggested previously, should we completely clean up th= is ping > - pong response approach for f/w stats, (or) this should be retained to s= upport > backward compatibility and also for supporting ping - pong response when = user > cats for fw-stats (via debugfs) (i did see in the commit message this nee= ds to > be cleaned up) I think it makes sense to remove the ping-pong logic and rely on periodic updates alone, including fw_stats and ethstats handling. >> - if (test_bit(WMI_SERVICE_PEER_STATS, ar->wmi.svc_map)) { >> - param =3D ar->wmi.pdev_param->peer_stats_update_period; >> - ret =3D ath10k_wmi_pdev_set_param(ar, param, >> - PEER_DEFAULT_STATS_UPDATE_= PERIOD); >> - if (ret) { >> - ath10k_warn(ar, >> - "failed to set peer stats period : %d\= n", >> - ret); >> - goto err_core_stop; >> - } >> + param =3D ar->wmi.pdev_param->peer_stats_update_period; >> + ret =3D ath10k_wmi_pdev_set_param(ar, param, >> + PEER_DEFAULT_STATS_UPDATE_PERIOD); >> + if (ret) { >> + ath10k_warn(ar, >> + "failed to set peer stats period : %d\n", >> + ret); >> + goto err_core_stop; >> } > > [shafi] If i am correct this change requires 'PEER_STATS' to be enabled b= y > default. No, it does not. Periodic stats have been available since forever. >> diff --git a/drivers/net/wireless/ath/ath10k/wmi.h b/drivers/net/wireles= s/ath/ath10k/wmi.h >> index 4d3cbc44fcd2..2877a3a27b95 100644 >> --- a/drivers/net/wireless/ath/ath10k/wmi.h >> +++ b/drivers/net/wireless/ath/ath10k/wmi.h >> @@ -3296,7 +3296,7 @@ struct wmi_csa_event { >> /* the definition of different PDEV parameters */ >> #define PDEV_DEFAULT_STATS_UPDATE_PERIOD 500 >> #define VDEV_DEFAULT_STATS_UPDATE_PERIOD 500 >> -#define PEER_DEFAULT_STATS_UPDATE_PERIOD 500 >> +#define PEER_DEFAULT_STATS_UPDATE_PERIOD 100 > > [shafi] Is this for more granularity since 500ms is not sufficient ?, I u= nderstand > the firmware has default stats_update_period as 500ms and i hope it suppo= rts > 100ms as well, Also if we are going to support period stats update we may= need to > accumulate the information in driver (like this change and rx_duration The patch is used for rough rate estimation which is used to keep Tx queues filled only with 1-2 txops worth of data. Signal conditions can change vastly so I figured I need the peer stat update events come more often. I didn't really verify if they come every 100ms. The patch already served it's job as a proof-of-concept for smarter tx queuing. > I will try to take this change, rebase it to TOT and see how it goes. There's really no benefit it taking this patch as a basis for periodic stat handling. Majority of this patch is just handling peer_tx_rate for rate estimation purposes. Feel free to knock yourself out with ripping out the ping-pong stat stuff though :) Micha=C5=82