From mboxrd@z Thu Jan 1 00:00:00 1970 From: Takashi Iwai Subject: Re: SB16 build error. Date: Thu, 30 Jun 2011 15:14:23 +0200 Message-ID: References: <20110630091754.GA12119@linux-mips.org> <20110630105254.GA25732@linux-mips.org> <20110630123212.GA6690@linux-mips.org> <20110630124333.GA9727@linux-mips.org> Mime-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: In-Reply-To: <20110630124333.GA9727@linux-mips.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: alsa-devel-bounces@alsa-project.org Errors-To: alsa-devel-bounces@alsa-project.org To: Ralf Baechle Cc: linux-arch@vger.kernel.org, linux-mips@linux-mips.org, alsa-devel@alsa-project.org, florian@linux-mips.org, "David S. Miller" , Benjamin Herrenschmidt , linux-kernel@vger.kernel.org, Ivan Kokshaysky , sparclinux@vger.kernel.org, Paul Mackerras , Matt Turner , Florian Fainelli , Richard Henderson List-Id: linux-arch.vger.kernel.org At Thu, 30 Jun 2011 13:43:33 +0100, Ralf Baechle wrote: > > On Thu, Jun 30, 2011 at 02:38:20PM +0200, Takashi Iwai wrote: > > > > In userland an empty definition will be used for _IOC_TYPECHECK so there > > > won't be an error. So userland already is already using the existing > > > value for SNDRV_SB_CSP_IOCTL_LOAD_CODE ... > > > > Right. It has an invalid direction (3), but apps won't care such > > details anyway. > > > > > With a crude hack like > > > > > > #define SNDRV_SB_CSP_IOCTL_LOAD_CODE \ > > > _IOC(_IOC_WRITE,'H', 0x11, sizeof(struct snd_sb_csp_microcode)) > > > > > > error checking can be bypassed and all will be fine as long as the > > > resulting value doesn't result in in a a duplicate case value - which it > > > doesn't, at least not in my testing. > > > > > > Should work but isn't nice. > > > > Indeed. But which is uglier is hard to answer :) > > > > If you are fine with the hacked ioctl number above, I can put it > > with some comments. This won't break anything, at least. > > Go ahead then and yes, this really deserves a comment. OK, here is the patch. Takashi === From: Takashi Iwai Subject: [PATCH] ALSA: sb16 - Fix build errors on MIPS and others with 13bit ioctl size One of ioctl definition in sound/sb16_csp.h contains the data size over 8kB, and this causes build errors on architectures like MIPS, which define _IOC_SIZEBITS=13. For avoiding this build errors but keeping the compatibility, manually exapnd with _IOC() instead of using _IOW() for the problematic ioctl. Signed-off-by: Takashi Iwai --- include/sound/sb16_csp.h | 9 ++++++++- 1 files changed, 8 insertions(+), 1 deletions(-) diff --git a/include/sound/sb16_csp.h b/include/sound/sb16_csp.h index 736eac7..af1b49e 100644 --- a/include/sound/sb16_csp.h +++ b/include/sound/sb16_csp.h @@ -99,7 +99,14 @@ struct snd_sb_csp_info { /* get CSP information */ #define SNDRV_SB_CSP_IOCTL_INFO _IOR('H', 0x10, struct snd_sb_csp_info) /* load microcode to CSP */ -#define SNDRV_SB_CSP_IOCTL_LOAD_CODE _IOW('H', 0x11, struct snd_sb_csp_microcode) +/* NOTE: struct snd_sb_csp_microcode overflows the max size (13 bits) + * defined for some architectures like MIPS, and it leads to build errors. + * (x86 and co have 14-bit size, thus it's valid, though.) + * As a workaround for skipping the size-limit check, here we don't use the + * normal _IOW() macro but _IOC() with the manual argument. + */ +#define SNDRV_SB_CSP_IOCTL_LOAD_CODE \ + _IOC(_IOC_WRITE, 'H', 0x11, sizeof(struct snd_sb_csp_microcode)) /* unload microcode from CSP */ #define SNDRV_SB_CSP_IOCTL_UNLOAD_CODE _IO('H', 0x12) /* start CSP */ -- 1.7.6 From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cantor2.suse.de ([195.135.220.15]:45259 "EHLO mx2.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752038Ab1F3NOY (ORCPT ); Thu, 30 Jun 2011 09:14:24 -0400 Date: Thu, 30 Jun 2011 15:14:23 +0200 Message-ID: From: Takashi Iwai Subject: Re: SB16 build error. In-Reply-To: <20110630124333.GA9727@linux-mips.org> References: <20110630091754.GA12119@linux-mips.org> <20110630105254.GA25732@linux-mips.org> <20110630123212.GA6690@linux-mips.org> <20110630124333.GA9727@linux-mips.org> MIME-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Sender: linux-arch-owner@vger.kernel.org List-ID: To: Ralf Baechle Cc: Jaroslav Kysela , alsa-devel@alsa-project.org, linux-kernel@vger.kernel.org, linux-mips@linux-mips.org, florian@linux-mips.org, Florian Fainelli , linux-arch@vger.kernel.org, Richard Henderson , Ivan Kokshaysky , Matt Turner , Benjamin Herrenschmidt , Paul Mackerras , "David S. Miller" , sparclinux@vger.kernel.org Message-ID: <20110630131423.LtYgfbKgYu4CGc8rJNj-kylgU_mEXUJyrjlzaJWmu78@z> At Thu, 30 Jun 2011 13:43:33 +0100, Ralf Baechle wrote: > > On Thu, Jun 30, 2011 at 02:38:20PM +0200, Takashi Iwai wrote: > > > > In userland an empty definition will be used for _IOC_TYPECHECK so there > > > won't be an error. So userland already is already using the existing > > > value for SNDRV_SB_CSP_IOCTL_LOAD_CODE ... > > > > Right. It has an invalid direction (3), but apps won't care such > > details anyway. > > > > > With a crude hack like > > > > > > #define SNDRV_SB_CSP_IOCTL_LOAD_CODE \ > > > _IOC(_IOC_WRITE,'H', 0x11, sizeof(struct snd_sb_csp_microcode)) > > > > > > error checking can be bypassed and all will be fine as long as the > > > resulting value doesn't result in in a a duplicate case value - which it > > > doesn't, at least not in my testing. > > > > > > Should work but isn't nice. > > > > Indeed. But which is uglier is hard to answer :) > > > > If you are fine with the hacked ioctl number above, I can put it > > with some comments. This won't break anything, at least. > > Go ahead then and yes, this really deserves a comment. OK, here is the patch. Takashi === From: Takashi Iwai Subject: [PATCH] ALSA: sb16 - Fix build errors on MIPS and others with 13bit ioctl size One of ioctl definition in sound/sb16_csp.h contains the data size over 8kB, and this causes build errors on architectures like MIPS, which define _IOC_SIZEBITS=13. For avoiding this build errors but keeping the compatibility, manually exapnd with _IOC() instead of using _IOW() for the problematic ioctl. Signed-off-by: Takashi Iwai --- include/sound/sb16_csp.h | 9 ++++++++- 1 files changed, 8 insertions(+), 1 deletions(-) diff --git a/include/sound/sb16_csp.h b/include/sound/sb16_csp.h index 736eac7..af1b49e 100644 --- a/include/sound/sb16_csp.h +++ b/include/sound/sb16_csp.h @@ -99,7 +99,14 @@ struct snd_sb_csp_info { /* get CSP information */ #define SNDRV_SB_CSP_IOCTL_INFO _IOR('H', 0x10, struct snd_sb_csp_info) /* load microcode to CSP */ -#define SNDRV_SB_CSP_IOCTL_LOAD_CODE _IOW('H', 0x11, struct snd_sb_csp_microcode) +/* NOTE: struct snd_sb_csp_microcode overflows the max size (13 bits) + * defined for some architectures like MIPS, and it leads to build errors. + * (x86 and co have 14-bit size, thus it's valid, though.) + * As a workaround for skipping the size-limit check, here we don't use the + * normal _IOW() macro but _IOC() with the manual argument. + */ +#define SNDRV_SB_CSP_IOCTL_LOAD_CODE \ + _IOC(_IOC_WRITE, 'H', 0x11, sizeof(struct snd_sb_csp_microcode)) /* unload microcode from CSP */ #define SNDRV_SB_CSP_IOCTL_UNLOAD_CODE _IO('H', 0x12) /* start CSP */ -- 1.7.6