From mboxrd@z Thu Jan 1 00:00:00 1970 From: Bartlomiej Zolnierkiewicz Subject: [PATCH 2/2] pmac: remove needless pmac_ide_destroy_dmatable() wrapper Date: Tue, 2 Sep 2008 20:31:22 +0200 Message-ID: <200809022031.22922.bzolnier@gmail.com> Mime-Version: 1.0 Content-Type: text/plain; charset="iso-8859-1" Content-Transfer-Encoding: 7bit Return-path: Received: from gv-out-0910.google.com ([216.239.58.190]:10259 "EHLO gv-out-0910.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751409AbYIBSdz (ORCPT ); Tue, 2 Sep 2008 14:33:55 -0400 Received: by gv-out-0910.google.com with SMTP id e6so67510gvc.37 for ; Tue, 02 Sep 2008 11:33:53 -0700 (PDT) Content-Disposition: inline Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: linux-ide@vger.kernel.org Cc: linux-kernel@vger.kernel.org, Benjamin Herrenschmidt hwif->sg_nents is always != 0 when this function is called and there is also no need to explicitely zero hwif->sg_nents. Cc: Benjamin Herrenschmidt Signed-off-by: Bartlomiej Zolnierkiewicz --- drivers/ide/ppc/pmac.c | 16 +++------------- 1 file changed, 3 insertions(+), 13 deletions(-) Index: b/drivers/ide/ppc/pmac.c =================================================================== --- a/drivers/ide/ppc/pmac.c +++ b/drivers/ide/ppc/pmac.c @@ -1523,18 +1523,6 @@ use_pio_instead: return 0; /* revert to PIO for this request */ } -/* Teardown mappings after DMA has completed. */ -static void -pmac_ide_destroy_dmatable (ide_drive_t *drive) -{ - ide_hwif_t *hwif = drive->hwif; - - if (hwif->sg_nents) { - ide_destroy_dmatable(drive); - hwif->sg_nents = 0; - } -} - /* * Prepare a DMA transfer. We build the DMA table, adjust the timings for * a read on KeyLargo ATA/66 and mark us as waiting for DMA completion @@ -1606,7 +1594,9 @@ pmac_ide_dma_end (ide_drive_t *drive) drive->waiting_for_dma = 0; dstat = readl(&dma->status); writel(((RUN|WAKE|DEAD) << 16), &dma->control); - pmac_ide_destroy_dmatable(drive); + + ide_destroy_dmatable(drive); + /* verify good dma status. we don't check for ACTIVE beeing 0. We should... * in theory, but with ATAPI decices doing buffer underruns, that would * cause us to disable DMA, which isn't what we want