linux-wireless.vger.kernel.org archive mirror
 help / color / mirror / Atom feed
From: Maya Erez <merez@codeaurora.org>
To: Kalle Valo <kvalo@codeaurora.org>
Cc: Ahmad Masri <amasri@codeaurora.org>,
	linux-wireless@vger.kernel.org, wil6210@qti.qualcomm.com,
	Maya Erez <merez@codeaurora.org>
Subject: [PATCH 03/11] wil6210: increase the frequency of status ring hw tail update
Date: Sun, 16 Jun 2019 10:25:59 +0300	[thread overview]
Message-ID: <1560669967-23706-4-git-send-email-merez@codeaurora.org> (raw)
In-Reply-To: <1560669967-23706-1-git-send-email-merez@codeaurora.org>

From: Ahmad Masri <amasri@codeaurora.org>

The driver updates Tx status ring HW tail only after it finishes
processing the whole status ring, while the HW is still transmitting
from other transmit rings. This can cause back-pressure on HW if
no status entries are available.

Update HW tail of Tx status ring without waiting for the end of the
processing to help feeding back the HW with status entries and to allow
additional packet transmission.

Signed-off-by: Ahmad Masri <amasri@codeaurora.org>
Signed-off-by: Maya Erez <merez@codeaurora.org>
---
 drivers/net/wireless/ath/wil6210/txrx_edma.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)

diff --git a/drivers/net/wireless/ath/wil6210/txrx_edma.c b/drivers/net/wireless/ath/wil6210/txrx_edma.c
index 6140db5..dc040cd 100644
--- a/drivers/net/wireless/ath/wil6210/txrx_edma.c
+++ b/drivers/net/wireless/ath/wil6210/txrx_edma.c
@@ -26,6 +26,10 @@
 #include "txrx.h"
 #include "trace.h"
 
+/* Max number of entries (packets to complete) to update the hwtail of tx
+ * status ring. Should be power of 2
+ */
+#define WIL_EDMA_TX_SRING_UPDATE_HW_TAIL 128
 #define WIL_EDMA_MAX_DATA_OFFSET (2)
 /* RX buffer size must be aligned to 4 bytes */
 #define WIL_EDMA_RX_BUF_LEN_DEFAULT (2048)
@@ -1155,7 +1159,7 @@ int wil_tx_sring_handler(struct wil6210_priv *wil,
 	struct wil_net_stats *stats;
 	struct wil_tx_enhanced_desc *_d;
 	unsigned int ring_id;
-	unsigned int num_descs;
+	unsigned int num_descs, num_statuses = 0;
 	int i;
 	u8 dr_bit; /* Descriptor Ready bit */
 	struct wil_ring_tx_status msg;
@@ -1276,6 +1280,11 @@ int wil_tx_sring_handler(struct wil6210_priv *wil,
 		}
 
 again:
+		num_statuses++;
+		if (num_statuses % WIL_EDMA_TX_SRING_UPDATE_HW_TAIL == 0)
+			/* update HW tail to allow HW to push new statuses */
+			wil_w(wil, sring->hwtail, sring->swhead);
+
 		wil_sring_advance_swhead(sring);
 
 		wil_get_next_tx_status_msg(sring, &msg);
@@ -1286,8 +1295,9 @@ int wil_tx_sring_handler(struct wil6210_priv *wil,
 	if (desc_cnt)
 		wil_update_net_queues(wil, vif, NULL, false);
 
-	/* Update the HW tail ptr (RD ptr) */
-	wil_w(wil, sring->hwtail, (sring->swhead - 1) % sring->size);
+	if (num_statuses % WIL_EDMA_TX_SRING_UPDATE_HW_TAIL != 0)
+		/* Update the HW tail ptr (RD ptr) */
+		wil_w(wil, sring->hwtail, (sring->swhead - 1) % sring->size);
 
 	return desc_cnt;
 }
-- 
1.9.1


  parent reply	other threads:[~2019-06-16  7:33 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-06-16  7:25 [PATCH 00/11] wil6210 patches Maya Erez
2019-06-16  7:25 ` [PATCH 01/11] wil6210: do not reset FW in STA to P2P client interface switch Maya Erez
2019-06-27 17:51   ` Kalle Valo
2019-06-16  7:25 ` [PATCH 02/11] wil6210: enlarge Tx status ring size Maya Erez
2019-06-16  7:25 ` Maya Erez [this message]
2019-06-16  7:26 ` [PATCH 04/11] wil6210: Add support for setting RBUFCAP configuration Maya Erez
2019-06-16  7:26 ` [PATCH 05/11] wil6210: fix printout in wil_read_pmccfg Maya Erez
2019-06-16  7:26 ` [PATCH 06/11] wil6210: clear FW and ucode log address Maya Erez
2019-06-16  7:26 ` [PATCH 07/11] wil6210: update cid boundary check of wil_find_cid/_by_idx() Maya Erez
2019-06-16  7:26 ` [PATCH 08/11] wil6210: publish max_msdu_size to FW on BCAST ring Maya Erez
2019-06-16  7:26 ` [PATCH 09/11] wil6210: add support for reading multiple RFs temperature via debugfs Maya Erez
2019-06-16  7:26 ` [PATCH 10/11] wil6210: set WIL_WMI_CALL_GENERAL_TO_MS as wmi_call timeout Maya Erez
2019-06-16  7:26 ` [PATCH 11/11] wil6210: drop old event after " Maya Erez

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=1560669967-23706-4-git-send-email-merez@codeaurora.org \
    --to=merez@codeaurora.org \
    --cc=amasri@codeaurora.org \
    --cc=kvalo@codeaurora.org \
    --cc=linux-wireless@vger.kernel.org \
    --cc=wil6210@qti.qualcomm.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).