From: Jérôme Pouiller <jerome.pouiller@xxxxxxxxxx> [ Upstream commit 6598b12d6635e8e3060863b84c04e472546ee126 ] Before to start the scan request, the firmware signals (with a null frame) to the AP it won't be able to receive data. This frame can be long to send: up to 512TU. The current calculus of the scan timeout does not take into account this delay. Signed-off-by: Jérôme Pouiller <jerome.pouiller@xxxxxxxxxx> Link: https://lore.kernel.org/r/20200515083325.378539-5-Jerome.Pouiller@xxxxxxxxxx Signed-off-by: Greg Kroah-Hartman <gregkh@xxxxxxxxxxxxxxxxxxx> Signed-off-by: Sasha Levin <sashal@xxxxxxxxxx> --- drivers/staging/wfx/hif_tx.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/staging/wfx/hif_tx.c b/drivers/staging/wfx/hif_tx.c index 77bca43aca428..20b3045d76674 100644 --- a/drivers/staging/wfx/hif_tx.c +++ b/drivers/staging/wfx/hif_tx.c @@ -268,7 +268,7 @@ int hif_scan(struct wfx_vif *wvif, struct cfg80211_scan_request *req, tmo_chan_bg = le32_to_cpu(body->max_channel_time) * USEC_PER_TU; tmo_chan_fg = 512 * USEC_PER_TU + body->probe_delay; tmo_chan_fg *= body->num_of_probe_requests; - tmo = chan_num * max(tmo_chan_bg, tmo_chan_fg); + tmo = chan_num * max(tmo_chan_bg, tmo_chan_fg) + 512 * USEC_PER_TU; wfx_fill_header(hif, wvif->id, HIF_REQ_ID_START_SCAN, buf_len); ret = wfx_cmd_send(wvif->wdev, hif, NULL, 0, false); -- 2.25.1