Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S934864AbcKQKBE (ORCPT ); Thu, 17 Nov 2016 05:01:04 -0500 Received: from atlantic540.startdedicated.de ([188.138.9.77]:49563 "EHLO atlantic540.startdedicated.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S933535AbcKQKBA (ORCPT ); Thu, 17 Nov 2016 05:01:00 -0500 From: Daniel Wagner To: linux-kernel@vger.kernel.org Cc: Ming Lei , "Luis R . Rodriguez" , Greg Kroah-Hartman , "Rafael J . Wysocki" , Daniel Vetter , Takashi Iwai , Bjorn Andersson , Arend van Spriel , Yves-Alexis Perez , Daniel Wagner Subject: [PATCH v7 0/4] firmware: refactor loading status Date: Thu, 17 Nov 2016 11:00:47 +0100 Message-Id: <1479376851-2307-1-git-send-email-wagi@monom.org> X-Mailer: git-send-email 2.7.4 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2415 Lines: 71 From: Daniel Wagner Hi, after the lengthy discussion online and in person I decided to remove all things which might change behavoir. So this is refactore clean up version. I think it worth to consider it because this version gets rid of the fw_lock dependency for protecting the loading status. It is based on Yves-Alexis' patch "firmware: fix usermode helper fallback loading". According this patch I fixed the type cast error in __fw_state_wait_common(). This function returns a long now. The last patch could be merged into second last. I decided to split it to make the review easier. cheers, daniel changes since v6: - removed all controversial patches which change behavoir - fixed casting error in return type in __fw_state_wait_common() - rebased in on v4.9-rc5 + Yves-Alexis' patch. changes since v5: - add enum name (enum fw_status) - renamed function prefix and structure fw_umh to fw_state (avoid to clash with enum fw_status) - add back patch 1 from v3 - move unused functions for direct firmeware loading into UMH section changes since v4: - replaced "firmware: Move umh locking code into fw_load_from_user_helper()" with "firmware: document user mode helper lock usage" - changed prefix fw_status_ to fw_umh_ - fixed a couple of bux pointed out by Ming - changed type of fw_umh::status to u8 and updated commit message to point out that all states are exclusive changes since v3: - added 'firmware: Move umh locking code into fw_load_from_user_helper()' - dropped loading_tiemout and firmware_loading_time() for !CONFG_FW_LOADER_USER_HELPER - rebased on Luis patches changes since v2: - more splitting out - first patch factors out all the bit ops into fw_status - second patch gets rid of the bit ops - third get rid of fw_lock by using swait changes since v1: - moved swait change into its own patch - added ifdef section for FW_LOADER_USER_HELPER_FALLBACK - updated commit message highlighting the mutex usage drop a bit Daniel Wagner (4): firmware: refactor loading status firmware: drop bit ops in favor of simple state machine firmware: do not use fw_lock for fw_state protection firmware: move fw_state_is_done() into UHM section drivers/base/firmware_class.c | 156 ++++++++++++++++++++++++++---------------- 1 file changed, 96 insertions(+), 60 deletions(-) -- 2.7.4