From mboxrd@z Thu Jan 1 00:00:00 1970 From: Takashi Iwai Subject: Re: ALSA: line6: Fix volume calculation for big-endian Date: Thu, 05 Mar 2015 13:01:39 +0100 Message-ID: References: <20150305102149.GA10971@mwanda> Mime-Version: 1.0 (generated by SEMI 1.14.6 - "Maruoka") Content-Type: text/plain; charset=US-ASCII Return-path: In-Reply-To: <20150305102149.GA10971@mwanda> Sender: kernel-janitors-owner@vger.kernel.org To: Dan Carpenter Cc: alsa-devel@alsa-project.org, kernel-janitors@vger.kernel.org List-Id: alsa-devel@alsa-project.org At Thu, 5 Mar 2015 13:21:49 +0300, Dan Carpenter wrote: > > Hello Takashi Iwai, > > The patch 0416980d0a2b: "ALSA: line6: Fix volume calculation for > big-endian" from Jan 28, 2015, leads to the following static checker > warning: > > sound/usb/line6/playback.c:42 change_volume() > warn: always clamps to s16min > sound/usb/line6/playback.c:57 change_volume() > warn: always clamps to (-8388608) > sound/usb/line6/playback.c:129 add_monitor_signal() > warn: always clamps to s16min > > sound/usb/line6/playback.c > 25 static void change_volume(struct urb *urb_out, int volume[], > 26 int bytes_per_frame) > 27 { > 28 int chn = 0; > 29 > 30 if (volume[0] == 256 && volume[1] == 256) > 31 return; /* maximum volume - no change */ > 32 > 33 if (bytes_per_frame == 4) { > 34 __le16 *p, *buf_end; > 35 > 36 p = (__le16 *)urb_out->transfer_buffer; > 37 buf_end = p + urb_out->transfer_buffer_length / sizeof(*p); > 38 > 39 for (; p < buf_end; ++p) { > 40 short pv = le16_to_cpu(*p); > 41 int val = (pv * volume[chn & 1]) >> 8; > 42 pv = clamp(val, 0x7fff, -0x8000); > ^^^^^^^^^^^^^^^^ > You didn't really add this, but you might know what was intended here. > It is a complete mystery to me. :) No mystery, just left/right swapped. It's a common mistake, and you'll feel the same when you go from Paris to London and look at a street. I'll fix them (the code, not the street). thanks, Takashi From mboxrd@z Thu Jan 1 00:00:00 1970 From: Takashi Iwai Date: Thu, 05 Mar 2015 12:01:39 +0000 Subject: Re: ALSA: line6: Fix volume calculation for big-endian Message-Id: List-Id: References: <20150305102149.GA10971@mwanda> In-Reply-To: <20150305102149.GA10971@mwanda> MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: Dan Carpenter Cc: alsa-devel@alsa-project.org, kernel-janitors@vger.kernel.org At Thu, 5 Mar 2015 13:21:49 +0300, Dan Carpenter wrote: > > Hello Takashi Iwai, > > The patch 0416980d0a2b: "ALSA: line6: Fix volume calculation for > big-endian" from Jan 28, 2015, leads to the following static checker > warning: > > sound/usb/line6/playback.c:42 change_volume() > warn: always clamps to s16min > sound/usb/line6/playback.c:57 change_volume() > warn: always clamps to (-8388608) > sound/usb/line6/playback.c:129 add_monitor_signal() > warn: always clamps to s16min > > sound/usb/line6/playback.c > 25 static void change_volume(struct urb *urb_out, int volume[], > 26 int bytes_per_frame) > 27 { > 28 int chn = 0; > 29 > 30 if (volume[0] = 256 && volume[1] = 256) > 31 return; /* maximum volume - no change */ > 32 > 33 if (bytes_per_frame = 4) { > 34 __le16 *p, *buf_end; > 35 > 36 p = (__le16 *)urb_out->transfer_buffer; > 37 buf_end = p + urb_out->transfer_buffer_length / sizeof(*p); > 38 > 39 for (; p < buf_end; ++p) { > 40 short pv = le16_to_cpu(*p); > 41 int val = (pv * volume[chn & 1]) >> 8; > 42 pv = clamp(val, 0x7fff, -0x8000); > ^^^^^^^^^^^^^^^^ > You didn't really add this, but you might know what was intended here. > It is a complete mystery to me. :) No mystery, just left/right swapped. It's a common mistake, and you'll feel the same when you go from Paris to London and look at a street. I'll fix them (the code, not the street). thanks, Takashi