March 2007 - Alsa-devel - mailman.alsa-project.org

Re: [alsa-devel] [Alsa-devel] [i8x0] docking station audio-out ceased to work
by Takashi Iwai 02 Apr '07

02 Apr '07

At Wed, 21 Mar 2007 23:57:32 +0100, Andreas Happe wrote: > > On Mon, 2007-03-19 at 18:53 +0100, Takashi Iwai wrote: > > Interesting. Usually AC97-SPSA specifies the AC97 slot for SPDIF > > output. The full (= 3) means the slot 10/11, 0 means slot 3/4. > > IIUC, AC97-SPSA must be 3 for docking station output working? > > If i remember correctly - but I was on holiday before the hardware > stopped to work (it was either set to 0 or 3, but I think that it was > 3). > > > If yes, try to set spdif_aclink=1 with the latest driver (HG > > version)... > > do you mean alsa-driver or alsa-kernel? I'm somehow reluctant to replace > my distribution's userspace utilities with self-compiled stuff (while i > would happily build a custom kernel). You need both trees to compile the driver. > BTW: a short link to a tutorial howto build hg kernels (or against what > kernel release). I checked it out, copied over the sound/ directory and > rebuild my 2.6.21-rc4 kernel. Copying to Linus kernel may work in most cases but sometimes not since alsa-kernel repo is updated for mm tree. The building HG driver is easy. Clone (both) repos, and run hgcompile script instead of configure. See alsa-driver/INSTALL. Takashi

2 10

[alsa-devel] how does hw:0, 1 eventually get mapped to /dev/snd/controlC0 and company? is this on the wiki somewhere?
by John Utz 02 Apr '07

02 Apr '07

how does this work? an explicit explanation would be great, but i'd even settle for a vague handwavy one at this point.

3 9

[alsa-devel] usleep() and nanosleep() timings seem inaccurate using ALSA
by Carlo Florendo 31 Mar '07

31 Mar '07

Good Day! After studying the intricacies of MIDI, I ended up writing an implementation of the MIDI protocol and file format. I then studied the ALSA sequencer API to be able to control a synthesizer keyboard and play MIDI files. I've used aplaymidi, aconnect, arecordmidi, and all those great ALSA utilities. They're very good! However, I wanted to have a simple ncurses based, command line MIDI sequencer meant for small Linux distributions such as DSL or Trustix so I began to write a command line sequencer using ALSA. I've encountered one problem about using usleep() and nanosleep() especially in 2.4 kernels. The core of my problem in my application is that the MIDI files slow down their tempo when pitch changes are made on a particular channel/track, while a NOTE-ON event is being played. For simplicity's sake, I'll give timing examples in terms of seconds and not bars nor measures. For example, if I've got a 2 second NOTE-ON playing a guitar note, all timings are ok. If the NOTE-ON is interspersed with CONTROLLER messages such as a pitch bend, the part which is supposed to sound in approximately 2 seconds, sounds much longer than that. The effect of this is that a ritardando is heard everytime short-duration CONTROLLER messages are sent. The problem disappears when I use the sequencer queues. However, the problem manifests itself when using direct output (i.e. not using queues, using snd_seq_event_output and snd_seq_drain_output one after the other). The reason why I would not want to use the queue is that I'd want the UI of my program (such as metronome ticks and counts) to somehow synch with the sound that is being heard. I've tried removing all UI elements, thinking that the UI controls might have caused the delay, but still the ritardando effect persists. The structure of my code is like this: parse_midi_file(); merge_all_tracks_to_one_single_track(); normalize_the_delta_times() for all events; do usleep(delta_time_for_each_event); play_the_event_using_alsa_direct_output(); done The ritardando effect is a bit lesser using 2.6 kernels. If I use queues, the problem disappears completely in 2.4 and 2.6 kernels. I don't hear the effect with simple MIDI files, such as with piano pieces. By the way, I intend to use my app with a 2.4 kernel. Does anybody have a solution to the delay problem? Please point out to me if I'm doing anything wrong since I've just been working with alsa-seq for a few months. Thank you very much. Best Regards, Carlo -- Carlo Florendo Softare Engineer/Network Co-Administrator Astra Philippines Inc. UP-Ayala Technopark, Diliman 1101, Quezon City Philippines http://www.astra.ph -- The Astra Group of Companies 5-3-11 Sekido, Tama City Tokyo 206-0011, Japan http://www.astra.co.jp

2 4

[alsa-devel] is statically linking libasound supported? because if i do it, i cant open the device!
by John Utz 31 Mar '07

31 Mar '07

i have a test app that works successfully with hw:0,1 on gnto if i link it dynamically: dmx-gnto jutz # ./atestminhw1 Playback device is hw:0,1 Stream parameters are 48000Hz, S16_LE, 6 channels Rate set to 48000Hz (requested 48000Hz) Buffer size range from 6 to 5461 Period size range from 3 to 2730 Using max buffer size 5460 Periods = 4 was set period_size = 1365 was set buffer_size = 5460 Channel: 0 Freq: 220.000000 but it fails to find hw:0,1 if i link it statically, is that expected behavior? dmx-gnto jutz # ./atestminhw1 Playback device is hw:0,1 Stream parameters are 48000Hz, S16_LE, 6 channels ALSA lib pcm.c:2109:(snd_pcm_open_conf) symbol _snd_pcm_hw_open is not defined inside (null) Failed: No such device or address this line succeeds with a dynamic linking and fails with if(0 > (iErr = snd_pcm_open(&pPcm, gs_pcNameDev, SND_PCM_STREAM_PLAYBACK, 0))) goto fin; here is my static link arguments: gcc -g -static -o atestminhw1 atestmin.c -lasound -lm -lpthread -ldl here is my dynamic link arguments: gcc -g -o atestminhw1 atestmin.c -lasound and, for completeness, here is my lame test app, derived from other test apps that come with the source: #include <stdio.h> #include <stdlib.h> #include <string.h> #include <sched.h> #include <errno.h> #include <getopt.h> #include <inttypes.h> #include <ctype.h> #include <byteswap.h> #define ALSA_PCM_NEW_HW_PARAMS_API #define ALSA_PCM_NEW_SW_PARAMS_API #include <alsa/asoundlib.h> #include <sys/time.h> #include <math.h> #define MAX_CHANNELS 16 static char *gs_pcNameDev = "hw:0,1"; /* playback device */ //static char *gs_pcNameDev = "plughw"; /* playback device */ //static char *gs_pcNameDev = "default"; /* playback device */ static snd_pcm_format_t gs_SndPcmFmt = SND_PCM_FORMAT_S16; /* sample format */ static unsigned int gs_uRate = 48000; /* stream rate */ static unsigned int gs_uCntChan = 6; /* count of channels */ static unsigned int gs_uTimBuf = 0; /* ring buffer length in us */ static unsigned int gs_uTimPer = 0; /* period time in us */ static unsigned int gs_uCntPer = 4; /* number of periods */ static double gs_dFreq = 440; /* sinusoidal wave frequency in Hz */ static snd_pcm_uframes_t gs_uSizBuf; static snd_pcm_uframes_t gs_uSizPer; static int set_hwparams(snd_pcm_t *handle, snd_pcm_hw_params_t *params, snd_pcm_access_t access) { unsigned int uRate; int iErr; snd_pcm_uframes_t period_size_min; snd_pcm_uframes_t period_size_max; snd_pcm_uframes_t buffer_size_min; snd_pcm_uframes_t buffer_size_max; /* choose all parameters */ if (0 > (iErr = snd_pcm_hw_params_any(handle, params))) { printf("Broken configuration for playback: no configurations available: %s\n", snd_strerror(iErr)); return iErr; } /* set the interleaved read/write format */ if (0 > (iErr = snd_pcm_hw_params_set_access(handle, params, access))) { printf("Access type not available for playback: %s\n", snd_strerror(iErr)); return iErr; } /* set the sample format */ if (0 > (iErr = snd_pcm_hw_params_set_format(handle, params, gs_SndPcmFmt))) { printf("Sample format not available for playback: %s\n", snd_strerror(iErr)); return iErr; } /* set the count of channels */ if (0 > (iErr = snd_pcm_hw_params_set_channels(handle, params, gs_uCntChan))) { printf("Channels count (%i) not available for playbacks: %s\n", gs_uCntChan, snd_strerror(iErr)); return iErr; } /* set the stream rate */ uRate = gs_uRate; if (0 > (iErr = snd_pcm_hw_params_set_rate(handle, params, gs_uRate, 0))) { printf("Rate %iHz not available for playback: %s\n", gs_uRate, snd_strerror(iErr)); return iErr; } if (uRate != gs_uRate) { printf("Rate doesn't match (requested %iHz, get %iHz, err %d)\n", gs_uRate, uRate, iErr); return -EINVAL; } printf("Rate set to %iHz (requested %iHz)\n", uRate, gs_uRate); /* set the buffer time */ iErr = snd_pcm_hw_params_get_buffer_size_min(params, &buffer_size_min); iErr = snd_pcm_hw_params_get_buffer_size_max(params, &buffer_size_max); iErr = snd_pcm_hw_params_get_period_size_min(params, &period_size_min, NULL); iErr = snd_pcm_hw_params_get_period_size_max(params, &period_size_max, NULL); printf("Buffer size range from %lu to %lu\n",buffer_size_min, buffer_size_max); printf("Period size range from %lu to %lu\n",period_size_min, period_size_max); if (gs_uTimPer > 0) { printf("Requested period time %u us\n", gs_uTimPer); if (0 > (iErr = snd_pcm_hw_params_set_period_time_near(handle, params, &gs_uTimPer, NULL))) { printf("Unable to set period time %u us for playback: %s\n", gs_uTimPer, snd_strerror(iErr)); return iErr; } } if (gs_uTimBuf > 0) { printf("Requested buffer time %u us\n", gs_uTimBuf); if (0 > (iErr = snd_pcm_hw_params_set_buffer_time_near(handle, params, &gs_uTimBuf, NULL))) { printf("Unable to set buffer time %u us for playback: %s\n", gs_uTimBuf, snd_strerror(iErr)); return iErr; } } if (! gs_uTimBuf && ! gs_uTimPer) { gs_uSizBuf = buffer_size_max; if (! gs_uTimPer) gs_uSizBuf = (gs_uSizBuf / gs_uCntPer) * gs_uCntPer; printf("Using max buffer size %lu\n", gs_uSizBuf); if (0 > (iErr = snd_pcm_hw_params_set_buffer_size_near(handle, params, &gs_uSizBuf))) { printf("Unable to set buffer size %lu for playback: %s\n", gs_uSizBuf, snd_strerror(iErr)); return iErr; } } if (! gs_uTimBuf || ! gs_uTimPer) { printf("Periods = %u\n", gs_uCntPer); if (0 > (iErr = snd_pcm_hw_params_set_periods_near(handle, params, &gs_uCntPer, NULL))) { printf("Unable to set nperiods %u for playback: %s\n", gs_uCntPer, snd_strerror(iErr)); return iErr; } } snd_pcm_hw_params_get_buffer_size(params, &gs_uSizBuf); snd_pcm_hw_params_get_period_size(params, &gs_uSizPer, NULL); printf("was set period_size = %lu\n",gs_uSizPer); printf("was set buffer_size = %lu\n",gs_uSizBuf); if (2*gs_uSizPer > gs_uSizBuf) { printf("buffer to small, could not use\n"); return iErr; } /* write the parameters to device */ iErr = snd_pcm_hw_params(handle, params); if (iErr < 0) { printf("Unable to set hw params for playback: %s\n", snd_strerror(iErr)); return iErr; } return 0; } static int set_swparams(snd_pcm_t *handle, snd_pcm_sw_params_t *swparams) { int iErr; /* get the current swparams */ if (0 > (iErr = snd_pcm_sw_params_current(handle, swparams))) { printf("Unable to determine current swparams for playback: %s\n", snd_strerror(iErr)); return iErr; } /* start the transfer when a buffer is full */ if (0 > (iErr = snd_pcm_sw_params_set_start_threshold(handle, swparams, gs_uSizBuf))) { printf("Unable to set start threshold mode for playback: %s\n", snd_strerror(iErr)); return iErr; } /* allow the transfer when at least period_size frames can be processed */ if (0 > (iErr = snd_pcm_sw_params_set_avail_min(handle, swparams, gs_uSizPer))) { printf("Unable to set avail min for playback: %s\n", snd_strerror(iErr)); return iErr; } /* align all transfers to 1 sample */ if (0 > (iErr = snd_pcm_sw_params_set_xfer_align(handle, swparams, 1))) { printf("Unable to set transfer align for playback: %s\n", snd_strerror(iErr)); return iErr; } /* write the parameters to the playback device */ if (0 > (iErr = snd_pcm_sw_params(handle, swparams))) { printf("Unable to set sw params for playback: %s\n", snd_strerror(iErr)); return iErr; } return 0; } static void generate_sine(uint8_t *pu8Frames, int iChannel, int iCntFrames, double *pdPhase, float fFreq) { double dPhase = *pdPhase; double dPhaseMax = 1.0 / fFreq; double dSizStep = 1.0 / (double)gs_uRate; double dRes; int iIdxChan; int32_t i32Res; int16_t *pi16Samp = (int16_t*) pu8Frames; while (iCntFrames-- > 0) { for(iIdxChan=0;iIdxChan<gs_uCntChan;iIdxChan++) if (iIdxChan==iChannel) { dRes = (sin((dPhase * 2 * M_PI) / dPhaseMax - M_PI)) * 0x03fffffff; /* Don't use MAX volume */ i32Res = dRes; *pi16Samp++ = i32Res >>16; } else *pi16Samp++ = 0; dPhase += dSizStep; if (dPhase >= dPhaseMax) dPhase -= dPhaseMax; } *pdPhase = dPhase; } /* * Underrun and suspend recovery */ static int xrun_recovery(snd_pcm_t *pPcm, int iErr) { if (iErr == -EPIPE) { /* under-run */ iErr = snd_pcm_prepare(pPcm); if (iErr < 0) printf("Can't recovery from underrun, prepare failed: %s\n", snd_strerror(iErr)); return 0; } else if (iErr == -ESTRPIPE) { while ((iErr = snd_pcm_resume(pPcm)) == -EAGAIN) sleep(1); /* wait until the suspend flag is released */ if (iErr < 0) { iErr = snd_pcm_prepare(pPcm); if (iErr < 0) printf("Can't recovery from suspend, prepare failed: %s\n", snd_strerror(iErr)); } return 0; } return iErr; } static int write_loop(snd_pcm_t *pPcm, int iChan, int iNumPer, uint8_t *pu8Frames) { double phase = 0; int iErr, iIdxPer, iCntFrames; uint8_t *pu8FramesLocal; float fFreq = gs_dFreq*(iChan+1)/2; // provide an audible chan change cue printf("Channel: %d Freq: %f\n", iChan, fFreq); for(iIdxPer = 0; iIdxPer < iNumPer; iIdxPer++) { iCntFrames = gs_uSizPer; pu8FramesLocal = pu8Frames; generate_sine(pu8FramesLocal, iChan, gs_uSizPer, &phase, fFreq); while (0 < iCntFrames) { if (-EAGAIN == (iErr = snd_pcm_writei(pPcm, pu8FramesLocal, iCntFrames))) continue; if (0 > iErr) { printf("Write error: %d,%s\n", iErr, snd_strerror(iErr)); if (0 > xrun_recovery(pPcm, iErr)) { printf("xrun_recovery failed: %d,%s\n", iErr, snd_strerror(iErr)); return -1; } break; // skip one period } pu8FramesLocal += snd_pcm_frames_to_bytes(pPcm, iErr); iCntFrames -= iErr; } } if (gs_uSizBuf > iIdxPer * gs_uSizPer) { snd_pcm_drain(pPcm); snd_pcm_prepare(pPcm ); } return 0; } int main(int argc, char *argv[]) { int iErr; int iIdxChan; uint8_t *pu8Frames = NULL; snd_pcm_t *pPcm = NULL; snd_pcm_hw_params_t *hwparams; snd_pcm_sw_params_t *swparams; snd_pcm_hw_params_alloca(&hwparams); snd_pcm_sw_params_alloca(&swparams); printf("Playback device is %s\n", gs_pcNameDev); printf("Stream parameters are %iHz, %s, %i channels\n", gs_uRate, snd_pcm_format_name(gs_SndPcmFmt), gs_uCntChan); if(0 > (iErr = snd_pcm_open(&pPcm, gs_pcNameDev, SND_PCM_STREAM_PLAYBACK, 0))) goto fin; if(0 > (iErr = set_hwparams(pPcm, hwparams, SND_PCM_ACCESS_RW_INTERLEAVED))) goto fin; if(0 > (iErr = set_swparams(pPcm, swparams))) goto fin; if(!(pu8Frames = malloc(snd_pcm_frames_to_bytes(pPcm, gs_uSizPer)))) goto fin; while(1) for(iIdxChan = 0; iIdxChan < gs_uCntChan; iIdxChan++) if (0> (iErr = write_loop(pPcm, iIdxChan, (gs_uRate*3)/gs_uSizPer, pu8Frames))) goto fin; iErr = EXIT_SUCCESS; fin: if(iErr) printf("Failed: %s\n",snd_strerror(iErr)); if(pu8Frames) free(pu8Frames); if(pPcm) snd_pcm_close(pPcm); exit(iErr); }

2 6

[alsa-devel] [2/4] 2.6.21-rc5: known regressions (v2)
by Adrian Bunk 30 Mar '07

30 Mar '07

This email lists some known regressions in Linus' tree compared to 2.6.20. If you find your name in the Cc header, you are either submitter of one of the bugs, maintainer of an affectected subsystem or driver, a patch of you caused a breakage or I'm considering you in any other way possibly involved with one or more of these issues. Due to the huge amount of recipients, please trim the Cc when answering. Subject : snd_hda_intel doesn't work with ASUS M2V mainboard References : http://bugzilla.kernel.org/show_bug.cgi?id=8273 Submitter : Hans-Georg Rist <hg.rist(a)web.de> Status : unknown Subject : snd_intel8x0: divide error: 0000 References : http://lkml.org/lkml/2007/3/5/252 Submitter : Michal Piotrowski <michal.k.k.piotrowski(a)gmail.com> Status : unknown Subject : hal daemon crashes after pulling a USB serial device References : http://www.opensubscriber.com/message/linux-usb-devel@lists.sourceforge.net… Submitter : Andi Kleen <ak(a)suse.de> Handled-By : Oliver Neukum <oneukum(a)suse.de> Status : problem is being debugged Subject : USB: iPod doesn't work (CONFIG_USB_SUSPEND) References : http://lkml.org/lkml/2007/3/21/320 Submitter : Tino Keitel <tino.keitel(a)gmx.de> Caused-By : Marcelo Tosatti <marcelo(a)kvack.org> commit 1d619f128ba911cd3e6d6ad3475f146eb92f5c27 Handled-By : Oliver Neukum <oneukum(a)suse.de> Status : problem is being debuggged Subject : USB: Oops when changing DVB-T adapter References : http://lkml.org/lkml/2007/3/9/212 Submitter : CIJOML <cijoml(a)volny.cz> Status : unknown Subject : forcedeth: sporadic under-load crashes References : http://lkml.org/lkml/2007/3/26/63 Submitter : Ingo Molnar <mingo(a)elte.hu> Handled-By : Ingo Molnar <mingo(a)elte.hu> Ayaz Abdulla <aabdulla(a)nvidia.com> Status : problem is being debugged

1 0

[alsa-devel] 14rc3 fails to compile on vanilla 2.6.20.3
by Andrea Gozzi 30 Mar '07

30 Mar '07

Hi. I recently compiled a new kernel, 2.6.20.3, but when I tried to build alsa 14rc3 it failed. I also had troubles with the ati fglrx driver (missing declarations in headers - solved) so it could maybe be related to the new 20.x kernels. I attached logs for both the "configure" and "make". Andrea

2 3

[alsa-devel] Questions on Bluetooth Audio & Alsa Relationship
by Allyn, Mark A 30 Mar '07

30 Mar '07

Hello: Where is the correct place to discuss issues pertaining to snd-bt-sco and its relationship with Alsa? I am having trouble getting snd-bt-sco to work with Alsa 1.0.14rc3. I am getting a kernel oops for a NULL pointer somewhere in snd_bt_sco_playback_open. I looked on the blutooth-alsa sourceforge site, it seems that snd-bt-sco is no longer maintained and that they are going toward a solution using something called plugz. Plugz, however requires a kernel patch that works for kernels 2.6.19 and above, but not with 2.6.14, which is the kernel in the distro that I am using. I have attempted to ask some questions on the bluez developer email list on sourceforge, but I only got one response saying that the snd-bt-sco solution is more mature than the plugz development effort. I then asked about the kernel oops and have not got any response. What is the state of bluetooth audio development? Is it slacking off? Perhaps is the a plugz email list that I don't know about? Thank you for your help. Truly, Mark Allyn

2 1

[alsa-devel] Custom ioctl
by intellij 30 Mar '07

30 Mar '07

hi, how can I send custom request to my driver? Looking into the sources I suspect that sending ioctl() on opened hardware pcm will not reach the playback_ops.ioctl in my driver. thank you very much.

2 1

[alsa-devel] [PATCH] snd-usb-caiaq: Make playback work
by Karsten Wiese 29 Mar '07

29 Mar '07

Hi Takashi, some typo fixes. best regards, Karsten -----------------------------------------------------------------------> From: Karsten Wiese <fzu(a)wemgehoertderstaat.de> Date: Thu, 29 Mar 2007 16:39:47 +0200 Subject: [PATCH] snd-usb-caiaq: Make playback work Signed-off-by: Karsten Wiese <fzu(a)wemgehoertderstaat.de> --- sound/usb/caiaq/caiaq-audio.c | 15 ++++++++------- 1 files changed, 8 insertions(+), 7 deletions(-) diff --git a/sound/usb/caiaq/caiaq-audio.c b/sound/usb/caiaq/caiaq-audio.c index e80c8db..0414d76 100644 --- a/sound/usb/caiaq/caiaq-audio.c +++ b/sound/usb/caiaq/caiaq-audio.c @@ -388,8 +388,8 @@ static void read_in_urb_mode2(struct snd_usb_caiaqdev *dev, struct snd_pcm_runtime *rt = sub->runtime; char *audio_buf = rt->dma_area; int sz = frames_to_bytes(rt, rt->buffer_size); - audio_buf[dev->audio_in_buf_pos[stream]++] - = usb_buf[i]; + audio_buf[dev->audio_in_buf_pos[stream]++] = + usb_buf[i]; dev->period_in_count[stream]++; if (dev->audio_in_buf_pos[stream] == sz) dev->audio_in_buf_pos[stream] = 0; @@ -436,19 +436,21 @@ static void fill_out_urb(struct snd_usb_caiaqdev *dev, spin_lock(&dev->spinlock); for (i = 0; i < iso->length;) { - for (stream = 0; stream < dev->n_streams; stream++) { + for (stream = 0; stream < dev->n_streams; stream++, i++) { sub = dev->sub_playback[stream]; if (sub) { struct snd_pcm_runtime *rt = sub->runtime; char *audio_buf = rt->dma_area; int sz = frames_to_bytes(rt, rt->buffer_size); - usb_buf[i++] - = audio_buf[dev->audio_out_buf_pos[stream]++]; + usb_buf[i] = + audio_buf[dev->audio_out_buf_pos[stream]]; + dev->period_out_count[stream]++; dev->audio_out_buf_pos[stream]++; if (dev->audio_out_buf_pos[stream] == sz) dev->audio_out_buf_pos[stream] = 0; } else - usb_buf[i++] = 0; + usb_buf[i] = 0; + } /* fill in the check bytes */ if (dev->spec.data_alignment == 2 && @@ -456,7 +458,6 @@ static void fill_out_urb(struct snd_usb_caiaqdev *dev, (dev->n_streams * CHANNELS_PER_STREAM)) for (stream = 0; stream < dev->n_streams; stream++, i++) usb_buf[i] = MAKE_CHECKBYTE(dev, stream, i); - } } spin_unlock(&dev->spinlock);

2 1

[alsa-devel] Alsa OSS - who is maintaining it?
by Stefan de Konink 29 Mar '07

29 Mar '07

-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA512 Hello, I'm looking for the maintainers of the Alsa OSS layer. It seems it is getting more seriously broken each month. The last serious updates were almost 7 months ago by tiwai and perex. Dev must have noticed the 'skips' on the end of an OSS stream. The complete broken big endian emulation and so on. What is currently happening in this area? Yours Sincerely, Stefan de Konink -----BEGIN PGP SIGNATURE----- Version: GnuPG v2.0.3 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org iD8DBQFGByXUYH1+F2Rqwn0RCmU3AJ9O25z5wcCpXLDUnx61idhB7MGcYACcD0fh Fl95KN+x77vuC006U0mexJ8= =guiG -----END PGP SIGNATURE-----

4 8