Return-path: Received: from mail-wi0-f179.google.com ([209.85.212.179]:41506 "EHLO mail-wi0-f179.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751808AbaHSMak convert rfc822-to-8bit (ORCPT ); Tue, 19 Aug 2014 08:30:40 -0400 Received: by mail-wi0-f179.google.com with SMTP id f8so5200752wiw.0 for ; Tue, 19 Aug 2014 05:30:39 -0700 (PDT) MIME-Version: 1.0 In-Reply-To: <87a977o450.fsf@kamboji.qca.qualcomm.com> References: <1407402260-29854-1-git-send-email-michal.kazior@tieto.com> <1407402260-29854-5-git-send-email-michal.kazior@tieto.com> <87a977o450.fsf@kamboji.qca.qualcomm.com> Date: Tue, 19 Aug 2014 14:30:39 +0200 Message-ID: (sfid-20140819_143044_007243_F177B7F7) Subject: Re: [PATCH 4/5] ath10k: split ce irq/handler setup From: Michal Kazior To: Kalle Valo Cc: "ath10k@lists.infradead.org" , linux-wireless Content-Type: text/plain; charset=UTF-8 Sender: linux-wireless-owner@vger.kernel.org List-ID: On 14 August 2014 10:40, Kalle Valo wrote: > Michal Kazior writes: > >> It doesn't make much sense to overwrite send_cb >> and recv_cb callbacks over and over again whenever >> transport starts. Just make sure to unmask copy >> engine interrupts when starting. >> >> Signed-off-by: Michal Kazior > > The patch looks, just a followup question for the future: > >> @@ -1674,7 +1639,9 @@ static int ath10k_pci_ce_init(struct ath10k *ar) >> pipe_info->hif_ce_state = ar; >> attr = &host_ce_config_wlan[pipe_num]; >> >> - ret = ath10k_ce_init_pipe(ar, pipe_num, attr); >> + ret = ath10k_ce_init_pipe(ar, pipe_num, attr, >> + ath10k_pci_ce_send_done, >> + ath10k_pci_ce_recv_data); > > As we call ath10k_ce_init_pipe() only once and seem to have only one set > of functions, why even bother bother with function pointers? What if we > just call the functions directly? Yeah, we can remove this abstraction later. I don't there's anything depending on it anymore. Michał