August 2019 - Alsa-devel - mailman.alsa-project.org

[alsa-devel] [PATCH] ASoC: SOF: Intel: hda: fix MSI handling
by Pierre-Louis Bossart 07 Aug '19

07 Aug '19

The addition of a kernel module parameter to optionally disable MSI had the side effect of permanently disabling it. The return value of pci_alloc_irq_vectors() is the number of allocated vectors or a negative number on error, so testing with the ! operator is not quite right. It was one optimization too far. Restore previous behavior to use MSI by default, unless the user selects not to do so or the allocation of irq_vectors fails. Fixes: 672ff5e3596ee ('ASoC: SOF: Intel: hda: add a parameter to disable MSI') Signed-off-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> --- sound/soc/sof/intel/hda.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/sof/intel/hda.c b/sound/soc/sof/intel/hda.c index 724a494702fa..c7de5065b961 100644 --- a/sound/soc/sof/intel/hda.c +++ b/sound/soc/sof/intel/hda.c @@ -543,7 +543,7 @@ int hda_dsp_probe(struct snd_sof_dev *sdev) * if it fails, use legacy interrupt mode * TODO: support msi multiple vectors */ - if (hda_use_msi && !pci_alloc_irq_vectors(pci, 1, 1, PCI_IRQ_MSI)) { + if (hda_use_msi && pci_alloc_irq_vectors(pci, 1, 1, PCI_IRQ_MSI) > 0) { dev_info(sdev->dev, "use msi interrupt mode\n"); hdev->irq = pci_irq_vector(pci, 0); /* ipc irq number is the same of hda irq */ -- 2.20.1

2 1

[alsa-devel] [PATCH] ASoC: rt1011: Add a flag for the R0 calibration test
by shumingf＠realtek.com 07 Aug '19

07 Aug '19

From: Shuming Fan <shumingf(a)realtek.com> The factory test needs to know whether the calibration completed. This flag helps to confirm the calibration completed or not. Signed-off-by: Shuming Fan <shumingf(a)realtek.com> --- sound/soc/codecs/rt1011.c | 14 +++++++++++++- sound/soc/codecs/rt1011.h | 2 +- 2 files changed, 14 insertions(+), 2 deletions(-) diff --git a/sound/soc/codecs/rt1011.c b/sound/soc/codecs/rt1011.c index b25b4c1127ca..638abcaf52b3 100644 --- a/sound/soc/codecs/rt1011.c +++ b/sound/soc/codecs/rt1011.c @@ -1225,7 +1225,10 @@ static int rt1011_bq_drc_info(struct snd_kcontrol *kcontrol, static int rt1011_r0_cali_get(struct snd_kcontrol *kcontrol, struct snd_ctl_elem_value *ucontrol) { - ucontrol->value.integer.value[0] = 0; + struct snd_soc_component *component = snd_kcontrol_chip(kcontrol); + struct rt1011_priv *rt1011 = snd_soc_component_get_drvdata(component); + + ucontrol->value.integer.value[0] = rt1011->cali_done; return 0; } @@ -1239,6 +1242,7 @@ static int rt1011_r0_cali_put(struct snd_kcontrol *kcontrol, if (!component->card->instantiated) return 0; + rt1011->cali_done = 0; if (snd_soc_component_get_bias_level(component) == SND_SOC_BIAS_OFF && ucontrol->value.integer.value[0]) rt1011_calibrate(rt1011, 1); @@ -2135,6 +2139,7 @@ static int rt1011_calibrate(struct rt1011_priv *rt1011, unsigned char cali_flag) r0_factor = ((format / r0[0] * 100) / 128) - (r0_integer * 100); rt1011->r0_reg = r0[0]; + rt1011->cali_done = 1; dev_info(dev, "r0 resistance about %d.%02d ohm, reg=0x%X\n", r0_integer, r0_factor, r0[0]); } @@ -2185,6 +2190,13 @@ static void rt1011_calibration_work(struct work_struct *work) rt1011_calibrate(rt1011, 1); + /* + * This flag should reset after booting. + * The factory test will do calibration again and use this flag to check + * whether the calibration completed + */ + rt1011->cali_done = 0; + /* initial */ rt1011_reg_init(component); } diff --git a/sound/soc/codecs/rt1011.h b/sound/soc/codecs/rt1011.h index 5d431d7f1d16..2d65983f3d0f 100644 --- a/sound/soc/codecs/rt1011.h +++ b/sound/soc/codecs/rt1011.h @@ -666,7 +666,7 @@ struct rt1011_priv { int pll_out; int bq_drc_set; - unsigned int r0_reg; + unsigned int r0_reg, cali_done; int recv_spk_mode; }; -- 2.22.0

2 1

[alsa-devel] [PATCH RESEND] ASoC: rt1011: Add R0 temperature and TDM1 ADC2DAT Swap control
by shumingf＠realtek.com 07 Aug '19

07 Aug '19

From: Shuming Fan <shumingf(a)realtek.com> - The user level application could set the R0 temperature after booting system. The degree Celsius of R0 temperature store in the non-volatile space when doing R0 calibration. - TDM1 ADC2DAT Swap controls use to control TDM slot2/3 data Signed-off-by: Shuming Fan <shumingf(a)realtek.com> --- sound/soc/codecs/rt1011.c | 9 ++++++++- sound/soc/codecs/rt1011.h | 1 + 2 files changed, 9 insertions(+), 1 deletion(-) diff --git a/sound/soc/codecs/rt1011.c b/sound/soc/codecs/rt1011.c index 0a6ff13d76e1..b25b4c1127ca 100644 --- a/sound/soc/codecs/rt1011.c +++ b/sound/soc/codecs/rt1011.c @@ -1029,6 +1029,8 @@ static const char * const rt1011_tdm_adc_swap_select[] = { static SOC_ENUM_SINGLE_DECL(rt1011_tdm_adc1_1_enum, RT1011_TDM1_SET_3, 6, rt1011_tdm_adc_swap_select); +static SOC_ENUM_SINGLE_DECL(rt1011_tdm_adc2_1_enum, RT1011_TDM1_SET_3, 4, + rt1011_tdm_adc_swap_select); static void rt1011_reset(struct regmap *regmap) { @@ -1333,7 +1335,8 @@ static const struct snd_kcontrol_new rt1011_snd_controls[] = { /* TDM1 Data Out Selection */ SOC_ENUM("TDM1 DOUT Source", rt1011_tdm1_adc1_dat_enum), SOC_ENUM("TDM1 DOUT Location", rt1011_tdm1_adc1_loc_enum), - SOC_ENUM("TDM1 ADCDAT Swap Select", rt1011_tdm_adc1_1_enum), + SOC_ENUM("TDM1 ADC1DAT Swap Select", rt1011_tdm_adc1_1_enum), + SOC_ENUM("TDM1 ADC2DAT Swap Select", rt1011_tdm_adc2_1_enum), /* Data Out Mode */ SOC_ENUM("I2S ADC DOUT Mode", rt1011_adc_dout_mode_enum), @@ -1355,6 +1358,10 @@ static const struct snd_kcontrol_new rt1011_snd_controls[] = { SOC_SINGLE_EXT("R0 Calibration", SND_SOC_NOPM, 0, 1, 0, rt1011_r0_cali_get, rt1011_r0_cali_put), RT1011_R0_LOAD("R0 Load Mode"), + + /* R0 temperature */ + SOC_SINGLE("R0 Temperature", RT1011_STP_INITIAL_RESISTANCE_TEMP, + 2, 255, 0), }; static int rt1011_is_sys_clk_from_pll(struct snd_soc_dapm_widget *source, diff --git a/sound/soc/codecs/rt1011.h b/sound/soc/codecs/rt1011.h index 98a38800c4df..5d431d7f1d16 100644 --- a/sound/soc/codecs/rt1011.h +++ b/sound/soc/codecs/rt1011.h @@ -227,6 +227,7 @@ #define RT1011_STP_CALIB_RS_TEMP 0x152a #define RT1011_INIT_RECIPROCAL_REG_24_16 0x1538 #define RT1011_INIT_RECIPROCAL_REG_15_0 0x1539 +#define RT1011_STP_INITIAL_RESISTANCE_TEMP 0x153c #define RT1011_STP_ALPHA_RECIPROCAL_MSB 0x153e #define RT1011_SPK_RESISTANCE_1 0x1544 #define RT1011_SPK_RESISTANCE_2 0x1546 -- 2.22.0

2 1

[alsa-devel] [PATCH v3 0/5] Add support for new SAI IP version
by Daniel Baluta 07 Aug '19

07 Aug '19

So far SAI IPs integrated with imx6 only supported one data line. Starting with imx7 and imx8 SAI integration support up to 8 data lines. First patch introduce register definition to support this. New SAI IP version introduces two new registers (Version and Parmeter registers) which are placed at the beginning of register address space. For this reason we need to fix the register's address. Support for this is introduced in patch 3. Changes since v2: - removed patches regarding data line mask because I need to find a better way to describe to model data lines. Perhaps,we only need to specify how many datalines a specific SAI instance supports and then let SAI driver to activate datalines based on the number of channels. Will open the discussion on this on a separate thread. - fixed devicetree documentation as per Nicolin comments and will send a separate patch to convert it to yaml. Changes since v1: - removed patches from Lucas as they were already accepted - addressed comments from Lucas and Nicolin regarding device tree property naming - removed comment saying that "datalines" must be always consecutively enabled (this is not true, checked with IP owner) - added new patch to document newly introduced compatbile strings - removed patch introducing combined mode as I will still need some time to figure out how to properly allow users to set it. Daniel Baluta (5): ASoC: fsl_sai: Add registers definition for multiple datalines ASoC: fsl_sai: Update Tx/Rx channel enable mask ASoC: fsl_sai: Add support for SAI new version ASoC: fsl_sai: Add support for imx7ulp/imx8mq ASoC: dt-bindings: Introduce compatible strings for 7ULP and 8MQ .../devicetree/bindings/sound/fsl-sai.txt | 3 +- sound/soc/fsl/fsl_sai.c | 320 ++++++++++++------ sound/soc/fsl/fsl_sai.h | 78 +++-- 3 files changed, 273 insertions(+), 128 deletions(-) -- 2.17.1

3 14

[alsa-devel] [PATCH v2 0/2] ASoC: Intel: IPC framework updates
by Cezary Rojewski 07 Aug '19

07 Aug '19

Existing IPC framework omits crucial part of the entire communication: reply header. Some IPCs cannot function at all without the access to said header. Update the sst-ipc with new sst_ipc_message structure to represent both request and reply allowing reply-processing handlers to save received responses. Despite the range of changes required for model to be updated, no functional changes are made for core hanswell, baytrail and skylake message handlers. Reply-processing handlers now save received response header yet no usage is added by default. To allow for future changes, righful kings of IPC kingdom need to be put back on the throne. This update addresses one of them: LARGE_CONFIG_GET. Changes since v1: - Combined first four patches into single one - Rebased onto v5.4 Cezary Rojewski (2): ASoC: Intel: Update request-reply IPC model ASoC: Intel: Skylake: large_config_get overhaul sound/soc/intel/baytrail/sst-baytrail-ipc.c | 65 ++++---- sound/soc/intel/common/sst-ipc.c | 69 ++++---- sound/soc/intel/common/sst-ipc.h | 27 ++-- sound/soc/intel/haswell/sst-haswell-ipc.c | 164 +++++++++++--------- sound/soc/intel/skylake/cnl-sst.c | 6 +- sound/soc/intel/skylake/skl-messages.c | 3 +- sound/soc/intel/skylake/skl-sst-ipc.c | 152 ++++++++++-------- sound/soc/intel/skylake/skl-sst-ipc.h | 3 +- 8 files changed, 260 insertions(+), 229 deletions(-) -- 2.17.1

2 4

[alsa-devel] [PATCH v3] ALSA: hiface: fix multiple memory leak bugs
by Wenwen Wang 07 Aug '19

07 Aug '19

In hiface_pcm_init(), 'rt' is firstly allocated through kzalloc(). Later on, hiface_pcm_init_urb() is invoked to initialize 'rt->out_urbs[i]'. In hiface_pcm_init_urb(), 'rt->out_urbs[i].buffer' is allocated through kzalloc(). However, if hiface_pcm_init_urb() fails, both 'rt' and 'rt->out_urbs[i].buffer' are not deallocated, leading to memory leak bugs. Also, 'rt->out_urbs[i].buffer' is not deallocated if snd_pcm_new() fails. To fix the above issues, free 'rt' and 'rt->out_urbs[i].buffer'. Signed-off-by: Wenwen Wang <wenwen(a)cs.uga.edu> --- sound/usb/hiface/pcm.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-) diff --git a/sound/usb/hiface/pcm.c b/sound/usb/hiface/pcm.c index 14fc1e1..c406497 100644 --- a/sound/usb/hiface/pcm.c +++ b/sound/usb/hiface/pcm.c @@ -600,14 +600,13 @@ int hiface_pcm_init(struct hiface_chip *chip, u8 extra_freq) ret = hiface_pcm_init_urb(&rt->out_urbs[i], chip, OUT_EP, hiface_pcm_out_urb_handler); if (ret < 0) - return ret; + goto error; } ret = snd_pcm_new(chip->card, "USB-SPDIF Audio", 0, 1, 0, &pcm); if (ret < 0) { - kfree(rt); dev_err(&chip->dev->dev, "Cannot create pcm instance\n"); - return ret; + goto error; } pcm->private_data = rt; @@ -620,4 +619,10 @@ int hiface_pcm_init(struct hiface_chip *chip, u8 extra_freq) chip->pcm = rt; return 0; + +error: + for (i = 0; i < PCM_N_URBS; i++) + kfree(rt->out_urbs[i].buffer); + kfree(rt); + return ret; } -- 2.7.4

2 1

[alsa-devel] [PATCH v2] ALSA: pcm: fix multiple memory leak bugs
by Wenwen Wang 07 Aug '19

07 Aug '19

In hiface_pcm_init(), 'rt' is firstly allocated through kzalloc(). Later on, hiface_pcm_init_urb() is invoked to initialize 'rt->out_urbs[i]'. In hiface_pcm_init_urb(), 'rt->out_urbs[i].buffer' is allocated through kzalloc(). However, if hiface_pcm_init_urb() fails, both 'rt' and 'rt->out_urbs[i].buffer' are not deallocated, leading to memory leak bugs. Also, 'rt->out_urbs[i].buffer' is not deallocated if snd_pcm_new() fails. To fix the above issues, free 'rt' and 'rt->out_urbs[i].buffer'. Signed-off-by: Wenwen Wang <wenwen(a)cs.uga.edu> --- sound/usb/hiface/pcm.c | 8 +++++++- 1 file changed, 7 insertions(+), 1 deletion(-) diff --git a/sound/usb/hiface/pcm.c b/sound/usb/hiface/pcm.c index 14fc1e1..9b132aa 100644 --- a/sound/usb/hiface/pcm.c +++ b/sound/usb/hiface/pcm.c @@ -599,12 +599,18 @@ int hiface_pcm_init(struct hiface_chip *chip, u8 extra_freq) for (i = 0; i < PCM_N_URBS; i++) { ret = hiface_pcm_init_urb(&rt->out_urbs[i], chip, OUT_EP, hiface_pcm_out_urb_handler); - if (ret < 0) + if (ret < 0) { + for (; i >= 0; i--) + kfree(rt->out_urbs[i].buffer); + kfree(rt); return ret; + } } ret = snd_pcm_new(chip->card, "USB-SPDIF Audio", 0, 1, 0, &pcm); if (ret < 0) { + for (i = 0; i < PCM_N_URBS; i++) + kfree(rt->out_urbs[i].buffer); kfree(rt); dev_err(&chip->dev->dev, "Cannot create pcm instance\n"); return ret; -- 2.7.4

2 2

[alsa-devel] [PATCH] ALSA: hda - Workaround for crackled sound on AMD controller (1022:1457)
by Takashi Iwai 07 Aug '19

07 Aug '19

A long-time problem on the recent AMD chip (1022:1457) with Realtek codecs (X370, X470, B450, etc) was the crackled or distorted sound for capture streams, as well as occasional playback hiccups. After lengthy debugging sessions, the workaround we've found is like below: - Set up the proper driver caps for this controller, similar as the other AMD controller. - Correct the DMA position reporting with the fixed FIFO size, which is similar like as workaround used for VIA chip set. - Even after the position correction, PA still shows mysterious stalls of playback streams when a capture is triggered in timer-scheduled mode. Since we have no clear way to eliminate the stall, pass the BATCH PCM flag for PA to suppress the tsched mode as a temporary workaround. This patch implements the workarounds. For the driver caps, it defines a new preset, AXZ_DCAPS_PRESET_AMD_SB. It enables the FIFO- corrected position reporting (corresponding to the new position_fix=6) and enforces the SNDRV_PCM_INFO_BATCH flag. Note that the current implementation is merely a workaround. Hopefully we'll find a better alternative in future, especially about removing the BATCH flag hack again. BugLink: https://bugzilla.kernel.org/show_bug.cgi?id=195303 Cc: <stable(a)vger.kernel.org> Signed-off-by: Takashi Iwai <tiwai(a)suse.de> --- sound/pci/hda/hda_controller.c | 7 +++++ sound/pci/hda/hda_controller.h | 2 +- sound/pci/hda/hda_intel.c | 63 +++++++++++++++++++++++++++++++++++++++++- 3 files changed, 70 insertions(+), 2 deletions(-) diff --git a/sound/pci/hda/hda_controller.c b/sound/pci/hda/hda_controller.c index 2fbdde239936..48d863736b3c 100644 --- a/sound/pci/hda/hda_controller.c +++ b/sound/pci/hda/hda_controller.c @@ -613,6 +613,13 @@ static int azx_pcm_open(struct snd_pcm_substream *substream) 20, 178000000); + /* by some reason, the playback stream stalls on PulseAudio with + * tsched=1 when a capture stream triggers. Until we figure out the + * real cause, disable tsched mode by telling the PCM info flag. + */ + if (chip->driver_caps & AZX_DCAPS_AMD_WORKAROUND) + runtime->hw.info |= SNDRV_PCM_INFO_BATCH; + if (chip->align_buffer_size) /* constrain buffer sizes to be multiple of 128 bytes. This is more efficient in terms of memory diff --git a/sound/pci/hda/hda_controller.h b/sound/pci/hda/hda_controller.h index baa15374fbcb..f2a6df5e6bcb 100644 --- a/sound/pci/hda/hda_controller.h +++ b/sound/pci/hda/hda_controller.h @@ -31,7 +31,7 @@ /* 14 unused */ #define AZX_DCAPS_CTX_WORKAROUND (1 << 15) /* X-Fi workaround */ #define AZX_DCAPS_POSFIX_LPIB (1 << 16) /* Use LPIB as default */ -/* 17 unused */ +#define AZX_DCAPS_AMD_WORKAROUND (1 << 17) /* AMD-specific workaround */ #define AZX_DCAPS_NO_64BIT (1 << 18) /* No 64bit address */ #define AZX_DCAPS_SYNC_WRITE (1 << 19) /* sync each cmd write */ #define AZX_DCAPS_OLD_SSYNC (1 << 20) /* Old SSYNC reg for ICH */ diff --git a/sound/pci/hda/hda_intel.c b/sound/pci/hda/hda_intel.c index 1e14d7270adf..a6d8c0d77b84 100644 --- a/sound/pci/hda/hda_intel.c +++ b/sound/pci/hda/hda_intel.c @@ -64,6 +64,7 @@ enum { POS_FIX_VIACOMBO, POS_FIX_COMBO, POS_FIX_SKL, + POS_FIX_FIFO, }; /* Defines for ATI HD Audio support in SB450 south bridge */ @@ -135,7 +136,7 @@ module_param_array(model, charp, NULL, 0444); MODULE_PARM_DESC(model, "Use the given board model."); module_param_array(position_fix, int, NULL, 0444); MODULE_PARM_DESC(position_fix, "DMA pointer read method." - "(-1 = system default, 0 = auto, 1 = LPIB, 2 = POSBUF, 3 = VIACOMBO, 4 = COMBO, 5 = SKL+)."); + "(-1 = system default, 0 = auto, 1 = LPIB, 2 = POSBUF, 3 = VIACOMBO, 4 = COMBO, 5 = SKL+, 6 = FIFO)."); module_param_array(bdl_pos_adj, int, NULL, 0644); MODULE_PARM_DESC(bdl_pos_adj, "BDL position adjustment offset."); module_param_array(probe_mask, int, NULL, 0444); @@ -332,6 +333,11 @@ enum { #define AZX_DCAPS_PRESET_ATI_HDMI_NS \ (AZX_DCAPS_PRESET_ATI_HDMI | AZX_DCAPS_SNOOP_OFF) +/* quirks for AMD SB */ +#define AZX_DCAPS_PRESET_AMD_SB \ + (AZX_DCAPS_NO_TCSEL | AZX_DCAPS_SYNC_WRITE | AZX_DCAPS_AMD_WORKAROUND |\ + AZX_DCAPS_SNOOP_TYPE(ATI) | AZX_DCAPS_PM_RUNTIME) + /* quirks for Nvidia */ #define AZX_DCAPS_PRESET_NVIDIA \ (AZX_DCAPS_NO_MSI | AZX_DCAPS_CORBRP_SELF_CLEAR |\ @@ -841,6 +847,49 @@ static unsigned int azx_via_get_position(struct azx *chip, return bound_pos + mod_dma_pos; } +#define AMD_FIFO_SIZE 32 + +/* get the current DMA position with FIFO size correction */ +static unsigned int azx_get_pos_fifo(struct azx *chip, struct azx_dev *azx_dev) +{ + struct snd_pcm_substream *substream = azx_dev->core.substream; + struct snd_pcm_runtime *runtime = substream->runtime; + unsigned int pos, delay; + + pos = snd_hdac_stream_get_pos_lpib(azx_stream(azx_dev)); + if (!runtime) + return pos; + + runtime->delay = AMD_FIFO_SIZE; + delay = frames_to_bytes(runtime, AMD_FIFO_SIZE); + if (azx_dev->insufficient) { + if (pos < delay) { + delay = pos; + runtime->delay = bytes_to_frames(runtime, pos); + } else { + azx_dev->insufficient = 0; + } + } + + /* correct the DMA position for capture stream */ + if (substream->stream == SNDRV_PCM_STREAM_CAPTURE) { + if (pos < delay) + pos += azx_dev->core.bufsize; + pos -= delay; + } + + return pos; +} + +static int azx_get_delay_from_fifo(struct azx *chip, struct azx_dev *azx_dev, + unsigned int pos) +{ + struct snd_pcm_substream *substream = azx_dev->core.substream; + + /* just read back the calculated value in the above */ + return substream->runtime->delay; +} + static unsigned int azx_skl_get_dpib_pos(struct azx *chip, struct azx_dev *azx_dev) { @@ -1417,6 +1466,7 @@ static int check_position_fix(struct azx *chip, int fix) case POS_FIX_VIACOMBO: case POS_FIX_COMBO: case POS_FIX_SKL: + case POS_FIX_FIFO: return fix; } @@ -1433,6 +1483,10 @@ static int check_position_fix(struct azx *chip, int fix) dev_dbg(chip->card->dev, "Using VIACOMBO position fix\n"); return POS_FIX_VIACOMBO; } + if (chip->driver_caps & AZX_DCAPS_AMD_WORKAROUND) { + dev_dbg(chip->card->dev, "Using FIFO position fix\n"); + return POS_FIX_FIFO; + } if (chip->driver_caps & AZX_DCAPS_POSFIX_LPIB) { dev_dbg(chip->card->dev, "Using LPIB position fix\n"); return POS_FIX_LPIB; @@ -1453,6 +1507,7 @@ static void assign_position_fix(struct azx *chip, int fix) [POS_FIX_VIACOMBO] = azx_via_get_position, [POS_FIX_COMBO] = azx_get_pos_lpib, [POS_FIX_SKL] = azx_get_pos_skl, + [POS_FIX_FIFO] = azx_get_pos_fifo, }; chip->get_position[0] = chip->get_position[1] = callbacks[fix]; @@ -1467,6 +1522,9 @@ static void assign_position_fix(struct azx *chip, int fix) azx_get_delay_from_lpib; } + if (fix == POS_FIX_FIFO) + chip->get_delay[0] = chip->get_delay[1] = + azx_get_delay_from_fifo; } /* @@ -2447,6 +2505,9 @@ static const struct pci_device_id azx_ids[] = { /* AMD Hudson */ { PCI_DEVICE(0x1022, 0x780d), .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_ATI_SB }, + /* AMD, X370 & co */ + { PCI_DEVICE(0x1022, 0x1457), + .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_AMD_SB }, /* AMD Stoney */ { PCI_DEVICE(0x1022, 0x157a), .driver_data = AZX_DRIVER_GENERIC | AZX_DCAPS_PRESET_ATI_SB | -- 2.16.4

1 0

[alsa-devel] [PATCH] ALSA: pcm: fix a memory leak bug
by Wenwen Wang 07 Aug '19

07 Aug '19

In hiface_pcm_init(), 'rt' is firstly allocated through kzalloc(). Later on, hiface_pcm_init_urb() is invoked to initialize 'rt->out_urbs[i]'. However, if the initialization fails, 'rt' is not deallocated, leading to a memory leak bug. To fix the above issue, free 'rt' before returning the error. Signed-off-by: Wenwen Wang <wenwen(a)cs.uga.edu> --- sound/usb/hiface/pcm.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) diff --git a/sound/usb/hiface/pcm.c b/sound/usb/hiface/pcm.c index 14fc1e1..5dbcd0d 100644 --- a/sound/usb/hiface/pcm.c +++ b/sound/usb/hiface/pcm.c @@ -599,8 +599,10 @@ int hiface_pcm_init(struct hiface_chip *chip, u8 extra_freq) for (i = 0; i < PCM_N_URBS; i++) { ret = hiface_pcm_init_urb(&rt->out_urbs[i], chip, OUT_EP, hiface_pcm_out_urb_handler); - if (ret < 0) + if (ret < 0) { + kfree(rt); return ret; + } } ret = snd_pcm_new(chip->card, "USB-SPDIF Audio", 0, 1, 0, &pcm); -- 2.7.4

2 2

[alsa-devel] [PATCH] ALSA: usb-midi: fix a memory leak bug
by Wenwen Wang 07 Aug '19

07 Aug '19

In __snd_usbmidi_create(), a MIDI streaming interface structure is allocated through kzalloc() and the pointer is saved to 'umidi'. Later on, the endpoint structures are created by invoking snd_usbmidi_create_endpoints_midiman() or snd_usbmidi_create_endpoints(), depending on the type of the audio quirk type. However, if the creation fails, the allocated 'umidi' is not deallocated, leading to a memory leak bug. To fix the above issue, free 'umidi' before returning the error. Signed-off-by: Wenwen Wang <wenwen(a)cs.uga.edu> --- sound/usb/midi.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/usb/midi.c b/sound/usb/midi.c index b737f0e..22db37f 100644 --- a/sound/usb/midi.c +++ b/sound/usb/midi.c @@ -2476,7 +2476,7 @@ int __snd_usbmidi_create(struct snd_card *card, else err = snd_usbmidi_create_endpoints(umidi, endpoints); if (err < 0) - goto exit; + goto free_midi; usb_autopm_get_interface_no_resume(umidi->iface); -- 2.7.4

2 2