From mboxrd@z Thu Jan 1 00:00:00 1970 From: Jeff Garzik Subject: Re: [git patches] libata build fix Date: Fri, 09 Oct 2009 17:53:13 -0400 Message-ID: <4ACFB0C9.605@garzik.org> References: <20091009043248.GA12304@havoc.gtf.org> <20091009083903.GA20687@elte.hu> Mime-Version: 1.0 Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: QUOTED-PRINTABLE Return-path: Received: from [207.36.208.214] ([207.36.208.214]:42783 "EHLO mail.dvmed.net" rhost-flags-FAIL-FAIL-OK-OK) by vger.kernel.org with ESMTP id S934514AbZJIVyE (ORCPT ); Fri, 9 Oct 2009 17:54:04 -0400 In-Reply-To: <20091009083903.GA20687@elte.hu> Sender: linux-ide-owner@vger.kernel.org List-Id: linux-ide@vger.kernel.org To: Ingo Molnar Cc: Andrew Morton , Linus Torvalds , linux-ide@vger.kernel.org, LKML On 10/09/2009 04:39 AM, Ingo Molnar wrote: > From: Ingo Molnar > Date: Fri, 9 Oct 2009 10:13:56 +0200 > Subject: [PATCH] ahci: Fix "filter FPDMA non-zero offset enable for A= spire 3810T" on !CONFIG_ATA_ACPI > MIME-Version: 1.0 > Content-Type: text/plain; charset=3Dutf-8 > Content-Transfer-Encoding: 8bit > > Commit f80ae7e ("ahci: filter FPDMA non-zero offset enable for Aspire > 3810T") introduced a new build breakage: > > drivers/ata/ahci.c: In function =E2=80=98ahci_gtf_filter_workaroun= d=E2=80=99: > drivers/ata/ahci.c:2927: error: =E2=80=98struct ata_device=E2=80=99= has no member named =E2=80=98gtf_filter=E2=80=99 > make[2]: *** [drivers/ata/ahci.o] Error 1 > make[1]: *** [drivers/ata] Error 2 > make[1]: *** Waiting for unfinished jobs.... > > As the acpi taskfile command filter support is a property of > libata-acpi - so the ata_device::gtf_filter field only exists > under CONFIG_ATA_ACPI. > > Add the trivial #ifdef to turn ahci_gtf_filter_workaround() > into a NOP on !CONFIG_ATA_ACPI. > > Cc: Marcus Meissner > Cc: Jeff Garzik > Signed-off-by: Ingo Molnar > --- > drivers/ata/ahci.c | 2 ++ > 1 files changed, 2 insertions(+), 0 deletions(-) > > diff --git a/drivers/ata/ahci.c b/drivers/ata/ahci.c > index 4edca6e..fc8d527 100644 > --- a/drivers/ata/ahci.c > +++ b/drivers/ata/ahci.c > @@ -2886,6 +2886,7 @@ static bool ahci_broken_online(struct pci_dev *= pdev) > > static void ahci_gtf_filter_workaround(struct ata_host *host) > { > +#ifdef CONFIG_ATA_ACPI > static const struct dmi_system_id sysids[] =3D { > /* > * Aspire 3810T issues a bunch of SATA enable commands > @@ -2926,6 +2927,7 @@ static void ahci_gtf_filter_workaround(struct a= ta_host *host) > ata_for_each_dev(dev, link, ALL) > dev->gtf_filter |=3D filter; > } > +#endif > } Either way is fine with me. I tend to prefer conditional compilation t= o=20 be outside a function, but it's not a big deal. Linus can pull my=20 submit, or add Acked-by: Jeff Garzik to this Ingo patch. Same end result; whichever makes people happier, and fixes builds the=20 most rapidly... Jeff