Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S261170AbVBFL04 (ORCPT ); Sun, 6 Feb 2005 06:26:56 -0500 Received: (majordomo@vger.kernel.org) by vger.kernel.org id S261179AbVBFL04 (ORCPT ); Sun, 6 Feb 2005 06:26:56 -0500 Received: from [211.58.254.17] ([211.58.254.17]:57534 "EHLO hemosu.com") by vger.kernel.org with ESMTP id S261183AbVBFL0c (ORCPT ); Sun, 6 Feb 2005 06:26:32 -0500 Date: Sun, 6 Feb 2005 20:26:28 +0900 From: Tejun Heo To: bzolnier@gmail.com, linux-kernel@vger.kernel.org, linux-ide@vger.kernel.org Subject: [PATCH 2.6.11-rc2] ide: driver updates, series 3 Message-ID: <20050206112628.GA31274@htj.dyndns.org> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.6+20040907i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2556 Lines: 75 Hello, Bartlomiej. Here are four patches which implement taskfile handling functions and merge flagged_taskfile() into do_rw_taskfile(). I didn't implement separate writing function for flagged taskfile, rationale being... 1. A single function which properly handles both unflagged and flagged cases has simpler semantics and, accordingly, less likely to cause bugs. If there were two separate functions ide_write_taskfile() and ide_write_flagged_taskfile(), when somebody tries to add custom in/out flags to a piece of code which uses unflagged taskfile and ide_write_taskfile(), it would be easy to set in/out flags but forget to call ide_write_flagged_taskfile() instead. 2. There virtually is no overhead. In #03, I moved taskfile reading part of ide_end_drive_cmd() into a separate function named ide_read_taskfile(), so that taskfile handling codes can live side-by-side in ide-taskfile.c. Also, I didn't use any of HWIF, IDE_*_REG macros, as they should go away. Please tell me what you think about the new referencing style. [ Start of patch descriptions ] 01_ide_write_taskfile.patch : ide_write_taskfile() implemented. ide_write_taskfile(), which writes the content of a ide_task_t into the IDE taskfile registers, is implemented, and do_rw_taskfile() and flagged_taskfile() are converted to use ide_write_taskfile(). Behavior changes are - flagged taskfile now honors HOB feature register. - No do_rw_taskfile() HIHI check on select register. Except for the DEV bit, all bits are honored. 02_ide_do_rw_disk_use_write_taskfile.patch : __ide_do_rw_disk() rewritten ide_write_taskfile(). __ide_do_rw_disk() rewritten using ide_write_taskfile(). 03_ide_read_taskfile.patch : ide_read_taskfile() implemented. Status reading part of ide_end_drive_cmd() is moved into a separate function ide_read_taskfile(). 04_ide_merge_rw_and_flagged_taskfile.patch : merge flagged_taskfile() into do_rw_taskfile() Merged flagged_taskfile() into do_rw_taskfile(). During the merge, the following change took place. - Uses taskfile->data_phase to determine if dma trasfer is requested. (previously, do_rw_taskfile() directly switched on taskfile->command for all dma commands) [ End of patch descriptions ] Thanks a lot. :-) -- tejun - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/