December 2022 - Sound-open-firmware - mailman.alsa-project.org

[Sound-open-firmware] [PATCH v1 0/3] Add support of MediaTek mt8188 to SOF
by Tinghan Shen 10 Jan '23

10 Jan '23

Add support of MediaTek mt8188 SoC DSP to SOF. The sof driver patches in this series are taken from thesofproject/linux/tree/topic/sof-dev-rebase. Tinghan Shen (3): dt-bindings: dsp: mediatek: Add mt8188 dsp compatible ASoC: SOF: mediatek: Support mt8188 platform ASoC: SOF: mediatek: Provide debugfs_add_region_item ops for core .../bindings/dsp/mediatek,mt8186-dsp.yaml | 4 +++- sound/soc/sof/mediatek/mt8186/mt8186.c | 20 +++++++++++++++++++ sound/soc/sof/mediatek/mt8186/mt8186.h | 3 ++- sound/soc/sof/mediatek/mt8195/mt8195.c | 1 + 4 files changed, 26 insertions(+), 2 deletions(-) -- 2.18.0

5 9

[Sound-open-firmware] [PATCH AUTOSEL 6.1 1/7] ASoC: SOF: Revert: "core: unregister clients and machine drivers in .shutdown"
by Sasha Levin 09 Jan '23

09 Jan '23

From: Kai Vehmanen <kai.vehmanen(a)linux.intel.com> [ Upstream commit 44fda61d2bcfb74a942df93959e083a4e8eff75f ] The unregister machine drivers call is not safe to do when kexec is used. Kexec-lite gets blocked with following backtrace: [ 84.943749] Freezing user space processes ... (elapsed 0.111 seconds) done. [ 246.784446] INFO: task kexec-lite:5123 blocked for more than 122 seconds. [ 246.819035] Call Trace: [ 246.821782] <TASK> [ 246.824186] __schedule+0x5f9/0x1263 [ 246.828231] schedule+0x87/0xc5 [ 246.831779] snd_card_disconnect_sync+0xb5/0x127 ... [ 246.889249] snd_sof_device_shutdown+0xb4/0x150 [ 246.899317] pci_device_shutdown+0x37/0x61 [ 246.903990] device_shutdown+0x14c/0x1d6 [ 246.908391] kernel_kexec+0x45/0xb9 This reverts commit 83bfc7e793b555291785136c3ae86abcdc046887. Reported-by: Ricardo Ribalda <ribalda(a)chromium.org> Cc: Ricardo Ribalda <ribalda(a)chromium.org> Signed-off-by: Kai Vehmanen <kai.vehmanen(a)linux.intel.com> Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> Reviewed-by: Péter Ujfalusi <peter.ujfalusi(a)linux.intel.com> Reviewed-by: Ranjani Sridharan <ranjani.sridharan(a)linux.intel.com> Link: https://lore.kernel.org/r/20221209114529.3909192-3-kai.vehmanen@linux.intel… Signed-off-by: Mark Brown <broonie(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- sound/soc/sof/core.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/sound/soc/sof/core.c b/sound/soc/sof/core.c index 3e6141d03770..625977a29d8a 100644 --- a/sound/soc/sof/core.c +++ b/sound/soc/sof/core.c @@ -475,19 +475,10 @@ EXPORT_SYMBOL(snd_sof_device_remove); int snd_sof_device_shutdown(struct device *dev) { struct snd_sof_dev *sdev = dev_get_drvdata(dev); - struct snd_sof_pdata *pdata = sdev->pdata; if (IS_ENABLED(CONFIG_SND_SOC_SOF_PROBE_WORK_QUEUE)) cancel_work_sync(&sdev->probe_work); - /* - * make sure clients and machine driver(s) are unregistered to force - * all userspace devices to be closed prior to the DSP shutdown sequence - */ - sof_unregister_clients(sdev); - - snd_sof_machine_unregister(sdev, pdata); - if (sdev->fw_state == SOF_FW_BOOT_COMPLETE) return snd_sof_shutdown(sdev); -- 2.35.1

2 3

[Sound-open-firmware] [PATCH AUTOSEL 6.0 1/7] ASoC: SOF: Revert: "core: unregister clients and machine drivers in .shutdown"
by Sasha Levin 31 Dec '22

31 Dec '22

From: Kai Vehmanen <kai.vehmanen(a)linux.intel.com> [ Upstream commit 44fda61d2bcfb74a942df93959e083a4e8eff75f ] The unregister machine drivers call is not safe to do when kexec is used. Kexec-lite gets blocked with following backtrace: [ 84.943749] Freezing user space processes ... (elapsed 0.111 seconds) done. [ 246.784446] INFO: task kexec-lite:5123 blocked for more than 122 seconds. [ 246.819035] Call Trace: [ 246.821782] <TASK> [ 246.824186] __schedule+0x5f9/0x1263 [ 246.828231] schedule+0x87/0xc5 [ 246.831779] snd_card_disconnect_sync+0xb5/0x127 ... [ 246.889249] snd_sof_device_shutdown+0xb4/0x150 [ 246.899317] pci_device_shutdown+0x37/0x61 [ 246.903990] device_shutdown+0x14c/0x1d6 [ 246.908391] kernel_kexec+0x45/0xb9 This reverts commit 83bfc7e793b555291785136c3ae86abcdc046887. Reported-by: Ricardo Ribalda <ribalda(a)chromium.org> Cc: Ricardo Ribalda <ribalda(a)chromium.org> Signed-off-by: Kai Vehmanen <kai.vehmanen(a)linux.intel.com> Reviewed-by: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> Reviewed-by: Péter Ujfalusi <peter.ujfalusi(a)linux.intel.com> Reviewed-by: Ranjani Sridharan <ranjani.sridharan(a)linux.intel.com> Link: https://lore.kernel.org/r/20221209114529.3909192-3-kai.vehmanen@linux.intel… Signed-off-by: Mark Brown <broonie(a)kernel.org> Signed-off-by: Sasha Levin <sashal(a)kernel.org> --- sound/soc/sof/core.c | 9 --------- 1 file changed, 9 deletions(-) diff --git a/sound/soc/sof/core.c b/sound/soc/sof/core.c index c99b5e6c026c..694a2d94a222 100644 --- a/sound/soc/sof/core.c +++ b/sound/soc/sof/core.c @@ -472,19 +472,10 @@ EXPORT_SYMBOL(snd_sof_device_remove); int snd_sof_device_shutdown(struct device *dev) { struct snd_sof_dev *sdev = dev_get_drvdata(dev); - struct snd_sof_pdata *pdata = sdev->pdata; if (IS_ENABLED(CONFIG_SND_SOC_SOF_PROBE_WORK_QUEUE)) cancel_work_sync(&sdev->probe_work); - /* - * make sure clients and machine driver(s) are unregistered to force - * all userspace devices to be closed prior to the DSP shutdown sequence - */ - sof_unregister_clients(sdev); - - snd_sof_machine_unregister(sdev, pdata); - if (sdev->fw_state == SOF_FW_BOOT_COMPLETE) return snd_sof_shutdown(sdev); -- 2.35.1

1 1

[Sound-open-firmware] [PATCH] ASoC: SOF: mediatek: mt8195: remove a redundant comparison of sram
by YC Hung 27 Dec '22

27 Dec '22

DSP SRAM is not used for audio shared buffer between host and DSP so TOTAL_SIZE_SHARED_SRAM_FROM_TAIL is zero. Remove the definition and redundant comparison to fix coverity "unsigned compared against 0". Signed-off-by: YC Hung <yc.hung(a)mediatek.com> --- sound/soc/sof/mediatek/mt8195/mt8195.c | 5 ----- sound/soc/sof/mediatek/mt8195/mt8195.h | 2 -- 2 files changed, 7 deletions(-) diff --git a/sound/soc/sof/mediatek/mt8195/mt8195.c b/sound/soc/sof/mediatek/mt8195/mt8195.c index 3c81e84fcecf..7c831e18483c 100644 --- a/sound/soc/sof/mediatek/mt8195/mt8195.c +++ b/sound/soc/sof/mediatek/mt8195/mt8195.c @@ -215,11 +215,6 @@ static int platform_parse_resource(struct platform_device *pdev, void *data) adsp->pa_sram = (phys_addr_t)mmio->start; adsp->sramsize = resource_size(mmio); - if (adsp->sramsize < TOTAL_SIZE_SHARED_SRAM_FROM_TAIL) { - dev_err(dev, "adsp SRAM(%#x) is not enough for share\n", - adsp->sramsize); - return -EINVAL; - } dev_dbg(dev, "sram pbase=%pa,%#x\n", &adsp->pa_sram, adsp->sramsize); diff --git a/sound/soc/sof/mediatek/mt8195/mt8195.h b/sound/soc/sof/mediatek/mt8195/mt8195.h index 7ffd523f936c..b4229170049f 100644 --- a/sound/soc/sof/mediatek/mt8195/mt8195.h +++ b/sound/soc/sof/mediatek/mt8195/mt8195.h @@ -139,8 +139,6 @@ struct snd_sof_dev; #define DSP_MBOX1_BAR 6 #define DSP_MBOX2_BAR 7 -#define TOTAL_SIZE_SHARED_SRAM_FROM_TAIL 0x0 - #define SIZE_SHARED_DRAM_DL 0x40000 /*Shared buffer for Downlink*/ #define SIZE_SHARED_DRAM_UL 0x40000 /*Shared buffer for Uplink*/ -- 2.18.0

3 2

[Sound-open-firmware] [PATCH] ASoC: SOF: amd: Enable cache for AMD Rembrandt platform
by V sujith kumar Reddy 27 Dec '22

27 Dec '22

Enable DSP cache for ACP memory Signed-off-by: V sujith kumar Reddy <Vsujithkumar.Reddy(a)amd.com> --- sound/soc/sof/amd/acp-dsp-offset.h | 4 ++++ sound/soc/sof/amd/acp-loader.c | 7 +++++++ sound/soc/sof/amd/acp.h | 2 ++ 3 files changed, 13 insertions(+) diff --git a/sound/soc/sof/amd/acp-dsp-offset.h b/sound/soc/sof/amd/acp-dsp-offset.h index de5726251dc6..920155dee819 100644 --- a/sound/soc/sof/amd/acp-dsp-offset.h +++ b/sound/soc/sof/amd/acp-dsp-offset.h @@ -85,4 +85,8 @@ #define ACP_SCRATCH_REG_0 0x10000 #define ACP6X_DSP_FUSION_RUNSTALL 0x0644 + +/* Cache window registers */ +#define ACP_DSP0_CACHE_OFFSET0 0x0420 +#define ACP_DSP0_CACHE_SIZE0 0x0424 #endif diff --git a/sound/soc/sof/amd/acp-loader.c b/sound/soc/sof/amd/acp-loader.c index 090c8b18c83c..a4bce5a3ae48 100644 --- a/sound/soc/sof/amd/acp-loader.c +++ b/sound/soc/sof/amd/acp-loader.c @@ -151,6 +151,7 @@ static void configure_pte_for_fw_loading(int type, int num_pages, struct acp_dev int acp_dsp_pre_fw_run(struct snd_sof_dev *sdev) { struct pci_dev *pci = to_pci_dev(sdev->dev); + const struct sof_amd_acp_desc *desc = get_chip_info(sdev->pdata); struct acp_dev_data *adata; unsigned int src_addr, size_fw; u32 page_count, dma_size; @@ -183,6 +184,12 @@ int acp_dsp_pre_fw_run(struct snd_sof_dev *sdev) if (ret < 0) dev_err(sdev->dev, "acp dma transfer status: %d\n", ret); + if (desc->rev > 3) { + /* Cache Window enable */ + snd_sof_dsp_write(sdev, ACP_DSP_BAR, ACP_DSP0_CACHE_OFFSET0, desc->sram_pte_offset); + snd_sof_dsp_write(sdev, ACP_DSP_BAR, ACP_DSP0_CACHE_SIZE0, SRAM1_SIZE | BIT(31)); + } + /* Free memory once DMA is complete */ dma_size = (PAGE_ALIGN(sdev->basefw.fw->size) >> PAGE_SHIFT) * ACP_PAGE_SIZE; dma_free_coherent(&pci->dev, dma_size, adata->bin_buf, adata->sha_dma_addr); diff --git a/sound/soc/sof/amd/acp.h b/sound/soc/sof/amd/acp.h index 09e16ef8afa0..4314094a97fd 100644 --- a/sound/soc/sof/amd/acp.h +++ b/sound/soc/sof/amd/acp.h @@ -72,6 +72,8 @@ #define EXCEPT_MAX_HDR_SIZE 0x400 #define AMD_STACK_DUMP_SIZE 32 +#define SRAM1_SIZE 0x13A000 + enum clock_source { ACP_CLOCK_96M = 0, ACP_CLOCK_48M, -- 2.25.1

2 1

[Sound-open-firmware] [PATCH 00/15] ASoC: Drop empty platform remove functions
by Uwe Kleine-König 27 Dec '22

27 Dec '22

From: Uwe Kleine-König <uwe(a)kleine-koenig.org> Hello, this patch series removes all platform remove functions that only return zero below sound/soc. There is no reason to have these, as the only caller is platform core code doing: if (drv->remove) { int ret = drv->remove(dev); ... } (in platform_remove()) and so having no remove function is both equivalent and simpler. Best regards Uwe Uwe Kleine-König (15): ASoC: pxa: Drop empty platform remove function ASoC: amd: Drop empty platform remove function ASoC: atmel-classd: Drop empty platform remove function ASoC: atmel-pdmic: Drop empty platform remove function ASoC: 88pm860x: Drop empty platform remove function ASoC: ac97: Drop empty platform remove function ASoC: adau7002: Drop empty platform remove function ASoC: bt-sco: Drop empty platform remove function ASoC: cq93vc: Drop empty platform remove function ASoC: mc13783: Drop empty platform remove function ASoC: wl1273: Drop empty platform remove function ASoC: pxa: e740_wm9705: Drop empty platform remove function ASoC: pxa: e750_wm9705: Drop empty platform remove function ASoC: pxa: e800_wm9712: Drop empty platform remove function ASoC: SOF: nocodec: Drop empty platform remove function sound/soc/amd/raven/acp3x-i2s.c | 8 -------- sound/soc/atmel/atmel-classd.c | 6 ------ sound/soc/atmel/atmel-pdmic.c | 6 ------ sound/soc/codecs/88pm860x-codec.c | 6 ------ sound/soc/codecs/ac97.c | 6 ------ sound/soc/codecs/adau7002.c | 6 ------ sound/soc/codecs/bt-sco.c | 6 ------ sound/soc/codecs/cq93vc.c | 6 ------ sound/soc/codecs/mc13783.c | 6 ------ sound/soc/codecs/wl1273.c | 6 ------ sound/soc/pxa/e740_wm9705.c | 6 ------ sound/soc/pxa/e750_wm9705.c | 6 ------ sound/soc/pxa/e800_wm9712.c | 6 ------ sound/soc/pxa/spitz.c | 6 ------ sound/soc/sof/nocodec.c | 6 ------ 15 files changed, 92 deletions(-) base-commit: 3a28c2c89f4b412b648761430720d40a8dc326ef -- 2.38.1

3 3

[Sound-open-firmware] [PATCH] ASoC: SOF: mediatek: initialize panic_info to zero
by YC Hung 14 Dec '22

14 Dec '22

Coverity spotted that panic_info is not initialized to zero in mtk_adsp_dump. Using uninitialized value panic_info.linenum when calling snd_sof_get_status. Fix this coverity by initializing panic_info struct as zero. Signed-off-by: YC Hung <yc.hung(a)mediatek.com> --- sound/soc/sof/mediatek/mtk-adsp-common.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/sound/soc/sof/mediatek/mtk-adsp-common.c b/sound/soc/sof/mediatek/mtk-adsp-common.c index 1e0769c668a7..de8dbe27cd0d 100644 --- a/sound/soc/sof/mediatek/mtk-adsp-common.c +++ b/sound/soc/sof/mediatek/mtk-adsp-common.c @@ -60,7 +60,7 @@ void mtk_adsp_dump(struct snd_sof_dev *sdev, u32 flags) { char *level = (flags & SOF_DBG_DUMP_OPTIONAL) ? KERN_DEBUG : KERN_ERR; struct sof_ipc_dsp_oops_xtensa xoops; - struct sof_ipc_panic_info panic_info; + struct sof_ipc_panic_info panic_info = {}; u32 stack[MTK_ADSP_STACK_DUMP_SIZE]; u32 status; -- 2.18.0

3 2

Re: [Sound-open-firmware] [PATCH v8 3/3] ASoC: SOF: Fix deadlock when shutdown a frozen userspace
by Kai Vehmanen 09 Dec '22

09 Dec '22

Hi, On Thu, 1 Dec 2022, Ricardo Ribalda wrote: > On Thu, 1 Dec 2022 at 14:22, 'Oliver Neukum' via Chromeos Kdump <chromeos-kdump(a)google.com> wrote: > > > > On 01.12.22 14:03, Ricardo Ribalda wrote: > > > This patchset does not modify this behaviour. It simply fixes the > > > stall for kexec(). > > > > > > The patch that introduced the stall: > > > 83bfc7e793b5 ("ASoC: SOF: core: unregister clients and machine drivers > > > in .shutdown") > > > > That patch is problematic. I would go as far as saying that > > it needs to be reverted. > > It fixes a real issue. We have not had any complaints until we tried > to kexec in the platform. > I wont recommend reverting it until we have an alternative implementation. > > kexec is far less common than suspend/reboot. I've posted an alternative to ALSA list that reverts the problematic patch and fixes the problem (the patch was originally addressing) in a different way: https://mailman.alsa-project.org/pipermail/alsa-devel/2022-December/209776.… No changes outside sound/soc/ are needed with this approach. Br, Kai

1 0

[Sound-open-firmware] [PATCH 2/2] ASoC : SOF: amd: Add support for IPC and DSP dumps
by V sujith kumar Reddy 05 Dec '22

05 Dec '22

Add support for IPC and DSP dumps for AMD platforms. Signed-off-by: V sujith kumar Reddy <Vsujithkumar.Reddy(a)amd.com> --- sound/soc/sof/amd/Kconfig | 1 + sound/soc/sof/amd/acp-common.c | 108 +++++++++++++++++++++++++++++++++ sound/soc/sof/amd/acp-ipc.c | 13 ++++ sound/soc/sof/amd/acp.h | 6 ++ 4 files changed, 128 insertions(+) diff --git a/sound/soc/sof/amd/Kconfig b/sound/soc/sof/amd/Kconfig index a305ea6efea9..1cb92d6030e3 100644 --- a/sound/soc/sof/amd/Kconfig +++ b/sound/soc/sof/amd/Kconfig @@ -20,6 +20,7 @@ config SND_SOC_SOF_AMD_COMMON select SND_SOC_SOF_IPC3 select SND_SOC_SOF_PCI_DEV select SND_AMD_ACP_CONFIG + select SND_SOC_SOF_XTENSA select SND_SOC_ACPI if ACPI help This option is not user-selectable but automatically handled by diff --git a/sound/soc/sof/amd/acp-common.c b/sound/soc/sof/amd/acp-common.c index 150e042e4039..bd6c1b198736 100644 --- a/sound/soc/sof/amd/acp-common.c +++ b/sound/soc/sof/amd/acp-common.c @@ -16,6 +16,7 @@ #include "../sof-audio.h" #include "acp.h" #include "acp-dsp-offset.h" +#include <sound/sof/xtensa.h> int acp_dai_probe(struct snd_soc_dai *dai) { @@ -33,6 +34,107 @@ int acp_dai_probe(struct snd_soc_dai *dai) } EXPORT_SYMBOL_NS(acp_dai_probe, SND_SOC_SOF_AMD_COMMON); +/** + * amd_sof_ipc_dump() - This function is called when IPC tx times out. + * @sdev: SOF device. + */ +void amd_sof_ipc_dump(struct snd_sof_dev *sdev) +{ + const struct sof_amd_acp_desc *desc = get_chip_info(sdev->pdata); + u32 base = desc->dsp_intr_base; + u32 dsp_msg_write = sdev->debug_box.offset + + offsetof(struct scratch_ipc_conf, sof_dsp_msg_write); + u32 dsp_ack_write = sdev->debug_box.offset + + offsetof(struct scratch_ipc_conf, sof_dsp_ack_write); + u32 host_msg_write = sdev->debug_box.offset + + offsetof(struct scratch_ipc_conf, sof_host_msg_write); + u32 host_ack_write = sdev->debug_box.offset + + offsetof(struct scratch_ipc_conf, sof_host_ack_write); + u32 dsp_msg, dsp_ack, host_msg, host_ack, irq_stat; + + dsp_msg = snd_sof_dsp_read(sdev, ACP_DSP_BAR, ACP_SCRATCH_REG_0 + dsp_msg_write); + dsp_ack = snd_sof_dsp_read(sdev, ACP_DSP_BAR, ACP_SCRATCH_REG_0 + dsp_ack_write); + host_msg = snd_sof_dsp_read(sdev, ACP_DSP_BAR, ACP_SCRATCH_REG_0 + host_msg_write); + host_ack = snd_sof_dsp_read(sdev, ACP_DSP_BAR, ACP_SCRATCH_REG_0 + host_ack_write); + irq_stat = snd_sof_dsp_read(sdev, ACP_DSP_BAR, base + DSP_SW_INTR_STAT_OFFSET); + + dev_err(sdev->dev, + "dsp_msg = %#x dsp_ack = %#x host_msg = %#x host_ack = %#x irq_stat = %#x\n", + dsp_msg, dsp_ack, host_msg, host_ack, irq_stat); +} + +/** + * amd_get_registers() - This function is called in case of DSP oops + * in order to gather information about the registers, filename and + * linenumber and stack. + * @sdev: SOF device. + * @xoops: Stores information about registers. + * @panic_info: Stores information about filename and line number. + * @stack: Stores the stack dump. + * @stack_words: Size of the stack dump. + */ +static void amd_get_registers(struct snd_sof_dev *sdev, + struct sof_ipc_dsp_oops_xtensa *xoops, + struct sof_ipc_panic_info *panic_info, + u32 *stack, size_t stack_words) +{ + u32 offset = sdev->dsp_oops_offset; + + /* first read registers */ + acp_mailbox_read(sdev, offset, xoops, sizeof(*xoops)); + + /* then get panic info */ + if (xoops->arch_hdr.totalsize > EXCEPT_MAX_HDR_SIZE) { + dev_err(sdev->dev, "invalid header size 0x%x. FW oops is bogus\n", + xoops->arch_hdr.totalsize); + return; + } + + offset += xoops->arch_hdr.totalsize; + acp_mailbox_read(sdev, offset, panic_info, sizeof(*panic_info)); + + /* then get the stack */ + offset += sizeof(*panic_info); + acp_mailbox_read(sdev, offset, stack, stack_words * sizeof(u32)); +} + +/** + * amd_sof_dump() - This function is called when a panic message is + * received from the firmware. + * @sdev: SOF device. + * @flags: parameter not used but required by ops prototype + */ +void amd_sof_dump(struct snd_sof_dev *sdev, u32 flags) +{ + struct sof_ipc_dsp_oops_xtensa xoops; + struct sof_ipc_panic_info panic_info; + u32 stack[AMD_STACK_DUMP_SIZE]; + u32 status; + + /* Get information about the panic status from the debug box area. + * Compute the trace point based on the status. + */ + if (sdev->dsp_oops_offset > sdev->debug_box.offset) { + acp_mailbox_read(sdev, sdev->debug_box.offset, &status, sizeof(u32)); + } else { + /* Read DSP Panic status from dsp_box. + * As window information for exception box offset and size is not available + * before FW_READY + */ + acp_mailbox_read(sdev, sdev->dsp_box.offset, &status, sizeof(u32)); + sdev->dsp_oops_offset = sdev->dsp_box.offset + sizeof(status); + } + + /* Get information about the registers, the filename and line + * number and the stack. + */ + amd_get_registers(sdev, &xoops, &panic_info, stack, AMD_STACK_DUMP_SIZE); + + /* Print the information to the console */ + sof_print_oops_and_stack(sdev, KERN_ERR, status, status, &xoops, + &panic_info, stack, AMD_STACK_DUMP_SIZE); +} + struct snd_soc_acpi_mach *amd_sof_machine_select(struct snd_sof_dev *sdev) { struct snd_sof_pdata *sof_pdata = sdev->pdata; @@ -104,9 +206,15 @@ struct snd_sof_dsp_ops sof_acp_common_ops = { /* PM */ .suspend = amd_sof_acp_suspend, .resume = amd_sof_acp_resume, + + .ipc_dump = amd_sof_ipc_dump, + .dbg_dump = amd_sof_dump, + .debugfs_add_region_item = snd_sof_debugfs_add_region_item_iomem, + .dsp_arch_ops = &sof_xtensa_arch_ops, }; EXPORT_SYMBOL_NS(sof_acp_common_ops, SND_SOC_SOF_AMD_COMMON); MODULE_IMPORT_NS(SND_SOC_SOF_AMD_COMMON); +MODULE_IMPORT_NS(SND_SOC_SOF_XTENSA); MODULE_DESCRIPTION("ACP SOF COMMON Driver"); MODULE_LICENSE("Dual BSD/GPL"); diff --git a/sound/soc/sof/amd/acp-ipc.c b/sound/soc/sof/amd/acp-ipc.c index dd6e53c63407..5a02753c4610 100644 --- a/sound/soc/sof/amd/acp-ipc.c +++ b/sound/soc/sof/amd/acp-ipc.c @@ -154,8 +154,15 @@ irqreturn_t acp_sof_ipc_irq_thread(int irq, void *context) offsetof(struct scratch_ipc_conf, sof_dsp_ack_write); bool ipc_irq = false; int dsp_msg, dsp_ack; + unsigned int status; if (sdev->first_boot && sdev->fw_state != SOF_FW_BOOT_COMPLETE) { + acp_mailbox_read(sdev, sdev->dsp_box.offset, &status, sizeof(status)); + if ((status & SOF_IPC_PANIC_MAGIC_MASK) == SOF_IPC_PANIC_MAGIC) { + snd_sof_dsp_panic(sdev, sdev->dsp_box.offset + sizeof(status), + true); + return IRQ_HANDLED; + } snd_sof_ipc_msgs_rx(sdev); acp_dsp_ipc_host_done(sdev); return IRQ_HANDLED; @@ -180,6 +187,12 @@ irqreturn_t acp_sof_ipc_irq_thread(int irq, void *context) ipc_irq = true; } + acp_mailbox_read(sdev, sdev->debug_box.offset, &status, sizeof(u32)); + if ((status & SOF_IPC_PANIC_MAGIC_MASK) == SOF_IPC_PANIC_MAGIC) { + snd_sof_dsp_panic(sdev, sdev->dsp_oops_offset, true); + return IRQ_HANDLED; + } + if (!ipc_irq) dev_dbg_ratelimited(sdev->dev, "nothing to do in IPC IRQ thread\n"); diff --git a/sound/soc/sof/amd/acp.h b/sound/soc/sof/amd/acp.h index 76ad963faec1..09e16ef8afa0 100644 --- a/sound/soc/sof/amd/acp.h +++ b/sound/soc/sof/amd/acp.h @@ -69,6 +69,9 @@ #define BOX_SIZE_512 0x200 #define BOX_SIZE_1024 0x400 +#define EXCEPT_MAX_HDR_SIZE 0x400 +#define AMD_STACK_DUMP_SIZE 32 + enum clock_source { ACP_CLOCK_96M = 0, ACP_CLOCK_48M, @@ -254,6 +257,9 @@ int acp_sof_trace_release(struct snd_sof_dev *sdev); int amd_sof_acp_suspend(struct snd_sof_dev *sdev, u32 target_state); int amd_sof_acp_resume(struct snd_sof_dev *sdev); +void amd_sof_ipc_dump(struct snd_sof_dev *sdev); +void amd_sof_dump(struct snd_sof_dev *sdev, u32 flags); + static inline const struct sof_amd_acp_desc *get_chip_info(struct snd_sof_pdata *pdata) { const struct sof_dev_desc *desc = pdata->desc; -- 2.25.1

1 0

[Sound-open-firmware] [PATCH 1/2] ASoC: SOF: amd: Use poll function instead to read ACP_SHA_DSP_FW_QUALIFIER
by V sujith kumar Reddy 05 Dec '22

05 Dec '22

From: Ajye Huang <ajye_huang(a)compal.corp-partner.google.com> The Skyrim project and Whiterun met error when DSP loading during device boot. Ex, error in kernel log, ERR kernel: [ 16.124537] snd_sof_amd_rembrandt 0000:04:00.5: PSP validation failed. Use the snd_sof_dsp_read_poll_timeout function to successfully read the FW_QUALIFIER register Signed-off-by: Ajye Huang <ajye_huang(a)compal.corp-partner.google.com> Signed-off-by: V sujith kumar Reddy <Vsujithkumar.Reddy(a)amd.com> --- sound/soc/sof/amd/acp.c | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-) diff --git a/sound/soc/sof/amd/acp.c b/sound/soc/sof/amd/acp.c index 47115a77c92c..6bd2888fbb66 100644 --- a/sound/soc/sof/amd/acp.c +++ b/sound/soc/sof/amd/acp.c @@ -255,10 +255,12 @@ int configure_and_run_sha_dma(struct acp_dev_data *adata, void *image_addr, if (ret) return ret; - fw_qualifier = snd_sof_dsp_read(sdev, ACP_DSP_BAR, ACP_SHA_DSP_FW_QUALIFIER); - if (!(fw_qualifier & DSP_FW_RUN_ENABLE)) { + ret = snd_sof_dsp_read_poll_timeout(sdev, ACP_DSP_BAR, ACP_SHA_DSP_FW_QUALIFIER, + fw_qualifier, fw_qualifier & DSP_FW_RUN_ENABLE, + ACP_REG_POLL_INTERVAL, ACP_DMA_COMPLETE_TIMEOUT_US); + if (ret < 0) { dev_err(sdev->dev, "PSP validation failed\n"); - return -EINVAL; + return ret; } return 0; -- 2.25.1

1 0

Re: [Sound-open-firmware] [PATCH v8 2/3] freezer: refactor pm_freezing into a function.
by Ricardo Ribalda 05 Dec '22

05 Dec '22

Hi Rafael On Fri, 2 Dec 2022 at 18:48, Rafael J. Wysocki <rafael(a)kernel.org> wrote: > > On Thu, Dec 1, 2022 at 12:08 PM Ricardo Ribalda <ribalda(a)chromium.org> wrote: > > > > Add a way to let the drivers know if the processes are frozen. > > > > This is needed by drivers that are waiting for processes to end on their > > shutdown path. > > > > Convert pm_freezing into a function and export it, so it can be used by > > drivers that are either built-in or modules. > > > > Cc: stable(a)vger.kernel.org > > Fixes: 83bfc7e793b5 ("ASoC: SOF: core: unregister clients and machine drivers in .shutdown") > > Signed-off-by: Ricardo Ribalda <ribalda(a)chromium.org> > > Why can't you export the original pm_freezing variable and why is this > fixing anything? Because then any module will be able to modify the content of the variable. The Fixes: is because the last patch on the set is doing a real fix. If you only cherry-pick the last patch on a stable branch, the build will fail. (Also, the zero-day builder complains) Anyway, I think we can hold this patch for a bit. The snd people are discussing if this the way to handle it, or if we should handle .shutdown in a different way. Thanks! > > > --- > > include/linux/freezer.h | 3 ++- > > kernel/freezer.c | 3 +-- > > kernel/power/process.c | 24 ++++++++++++++++++++---- > > 3 files changed, 23 insertions(+), 7 deletions(-) > > > > diff --git a/include/linux/freezer.h b/include/linux/freezer.h > > index b303472255be..3413c869d68b 100644 > > --- a/include/linux/freezer.h > > +++ b/include/linux/freezer.h > > @@ -13,7 +13,7 @@ > > #ifdef CONFIG_FREEZER > > DECLARE_STATIC_KEY_FALSE(freezer_active); > > > > -extern bool pm_freezing; /* PM freezing in effect */ > > +bool pm_freezing(void); > > extern bool pm_nosig_freezing; /* PM nosig freezing in effect */ > > > > /* > > @@ -80,6 +80,7 @@ static inline int freeze_processes(void) { return -ENOSYS; } > > static inline int freeze_kernel_threads(void) { return -ENOSYS; } > > static inline void thaw_processes(void) {} > > static inline void thaw_kernel_threads(void) {} > > +static inline bool pm_freezing(void) { return false; } > > > > static inline bool try_to_freeze(void) { return false; } > > > > diff --git a/kernel/freezer.c b/kernel/freezer.c > > index 4fad0e6fca64..2d3530ebdb7e 100644 > > --- a/kernel/freezer.c > > +++ b/kernel/freezer.c > > @@ -20,7 +20,6 @@ EXPORT_SYMBOL(freezer_active); > > * indicate whether PM freezing is in effect, protected by > > * system_transition_mutex > > */ > > -bool pm_freezing; > > bool pm_nosig_freezing; > > > > /* protects freezing and frozen transitions */ > > @@ -46,7 +45,7 @@ bool freezing_slow_path(struct task_struct *p) > > if (pm_nosig_freezing || cgroup_freezing(p)) > > return true; > > > > - if (pm_freezing && !(p->flags & PF_KTHREAD)) > > + if (pm_freezing() && !(p->flags & PF_KTHREAD)) > > return true; > > > > return false; > > diff --git a/kernel/power/process.c b/kernel/power/process.c > > index ddd9988327fe..8a4d0e2c8c20 100644 > > --- a/kernel/power/process.c > > +++ b/kernel/power/process.c > > @@ -108,6 +108,22 @@ static int try_to_freeze_tasks(bool user_only) > > return todo ? -EBUSY : 0; > > } > > > > +/* > > + * Indicate whether PM freezing is in effect, protected by > > + * system_transition_mutex. > > + */ > > +static bool pm_freezing_internal; > > + > > +/** > > + * pm_freezing - indicate whether PM freezing is in effect. > > + * > > + */ > > +bool pm_freezing(void) > > +{ > > + return pm_freezing_internal; > > +} > > +EXPORT_SYMBOL(pm_freezing); > > Use EXPORT_SYMBOL_GPL() instead, please. > > > + > > /** > > * freeze_processes - Signal user space processes to enter the refrigerator. > > * The current thread will not be frozen. The same process that calls > > @@ -126,12 +142,12 @@ int freeze_processes(void) > > /* Make sure this task doesn't get frozen */ > > current->flags |= PF_SUSPEND_TASK; > > > > - if (!pm_freezing) > > + if (!pm_freezing()) > > static_branch_inc(&freezer_active); > > > > pm_wakeup_clear(0); > > pr_info("Freezing user space processes ... "); > > - pm_freezing = true; > > + pm_freezing_internal = true; > > error = try_to_freeze_tasks(true); > > if (!error) { > > __usermodehelper_set_disable_depth(UMH_DISABLED); > > @@ -187,9 +203,9 @@ void thaw_processes(void) > > struct task_struct *curr = current; > > > > trace_suspend_resume(TPS("thaw_processes"), 0, true); > > - if (pm_freezing) > > + if (pm_freezing()) > > static_branch_dec(&freezer_active); > > - pm_freezing = false; > > + pm_freezing_internal = false; > > pm_nosig_freezing = false; > > > > oom_killer_enable(); > > > > -- > > -- > You received this message because you are subscribed to the Google Groups "Chromeos Kdump" group. > To unsubscribe from this group and stop receiving emails from it, send an email to chromeos-kdump+unsubscribe(a)google.com. > To view this discussion on the web, visit https://groups.google.com/a/google.com/d/msgid/chromeos-kdump/CAJZ5v0jbKSTQ…. -- Ricardo Ribalda

1 0

[Sound-open-firmware] [PATCH v7 0/2] ASoC: SOF: Fix deadlock when shutdown a frozen userspace
by Ricardo Ribalda 01 Dec '22

01 Dec '22

Since: 83bfc7e793b5 ("ASoC: SOF: core: unregister clients and machine drivers in .shutdown") we wait for all the workloads to be completed during shutdown. This was done to avoid a stall once the device is started again. Unfortunately this has the side effect of stalling kexec(), if the userspace is frozen. Let's handle that case. To: Pierre-Louis Bossart <pierre-louis.bossart(a)linux.intel.com> To: Liam Girdwood <lgirdwood(a)gmail.com> To: Peter Ujfalusi <peter.ujfalusi(a)linux.intel.com> To: Bard Liao <yung-chuan.liao(a)linux.intel.com> To: Ranjani Sridharan <ranjani.sridharan(a)linux.intel.com> To: Kai Vehmanen <kai.vehmanen(a)linux.intel.com> To: Daniel Baluta <daniel.baluta(a)nxp.com> To: Mark Brown <broonie(a)kernel.org> To: Jaroslav Kysela <perex(a)perex.cz> To: Takashi Iwai <tiwai(a)suse.com> To: Eric Biederman <ebiederm(a)xmission.com> To: Chromeos Kdump <chromeos-kdump(a)google.com> To: Steven Rostedt <rostedt(a)goodmis.org> Cc: stable(a)vger.kernel.org Cc: sound-open-firmware(a)alsa-project.org Cc: alsa-devel(a)alsa-project.org Cc: linux-kernel(a)vger.kernel.org Cc: kexec(a)lists.infradead.org Signed-off-by: Ricardo Ribalda <ribalda(a)chromium.org> --- Changes in v7: - Fix commit message (Thanks Pierre-Louis). - Link to v6: https://lore.kernel.org/r/20221127-snd-freeze-v6-0-3e90553f64a5@chromium.org Changes in v6: - Check if we are in kexec with the userspace frozen. - Link to v5: https://lore.kernel.org/r/20221127-snd-freeze-v5-0-4ededeb08ba0@chromium.org Changes in v5: - Edit subject prefix. - Link to v4: https://lore.kernel.org/r/20221127-snd-freeze-v4-0-51ca64b7f2ab@chromium.org Changes in v4: - Do not call snd_sof_machine_unregister from shutdown. - Link to v3: https://lore.kernel.org/r/20221127-snd-freeze-v3-0-a2eda731ca14@chromium.org Changes in v3: - Wrap pm_freezing in a function. - Link to v2: https://lore.kernel.org/r/20221127-snd-freeze-v2-0-d8a425ea9663@chromium.org Changes in v2: - Only use pm_freezing if CONFIG_FREEZER . - Link to v1: https://lore.kernel.org/r/20221127-snd-freeze-v1-0-57461a366ec2@chromium.org --- Ricardo Ribalda (2): kexec: Introduce kexec_with_frozen_processes ASoC: SOF: Fix deadlock when shutdown a frozen userspace include/linux/kexec.h | 3 +++ kernel/kexec_core.c | 5 +++++ sound/soc/sof/core.c | 4 +++- 3 files changed, 11 insertions(+), 1 deletion(-) --- base-commit: 4312098baf37ee17a8350725e6e0d0e8590252d4 change-id: 20221127-snd-freeze-1ee143228326 Best regards, -- Ricardo Ribalda <ribalda(a)chromium.org>

3 5