[BUG] error 7 unsupported request to SET_PIPELINE_STATE IPC, could not alloc size = 1536

[kv2019i]

Describe the bug IPC error while running "~/sof-test/test-case/check-xrun-injection.sh -c 10 -m playback" as part of larger test plan (daily sof-test plan).

To Reproduce Run test plan including check-xrun-injection.sh.

Reproduction Rate ?

Expected behavior No IPC errors.

Impact Loss of audio.

Environment

1. Branch name and commit hash of the 2 repositories: sof (firmware/topology) and linux (kernel driver).
   • Kernel: 184f97930d2d
   • SOF: https://github.com/thesofproject/sof/commit/dff733232673cdbedf8553a1e7a199b76d9e856d
2. Name of the topology file
   • Topology: sof-mtl-nocodec.tplg
3. Name of the platform(s) on which the bug is observed.
   • Platform: Intel mtl

Screenshots or console output Seen on Intel daily test plan #38973

 [ 3945.314285] <inf> ipc: ipc_cmd: rx	: 0x13000004|0x1
[ 3945.314380] <err> buffer: buffer_alloc: buffer_alloc(): could not alloc size = 1536 bytes of type = 32
[ 3945.314386] <err> dai_comp: dai_set_dma_buffer: comp:1 0x10004 dai_set_dma_buffer(): failed to alloc dma buffer
[ 3945.314391] <err> dai_comp: dai_common_params: comp:1 0x10004 dai_zephyr_params(): alloc dma buffer failed.
[ 3945.314403] <err> module_adapter: module_prepare: comp:1 0x10004 module_prepare() error -12: module specific prepare failed, comp_id 65540
[ 3945.314410] <err> module_adapter: module_adapter_prepare: comp:1 0x10004 module_adapter_prepare() error fffffff4: module prepare failed
[ 3945.314416] <err> pipe: pipeline_prepare: pipe:1 0x0 pipeline_prepare(): ret = -12, dev->comp.id = 3
[ 3945.314421] <err> ipc: ipc4_pcm_params: ipc: pipe 1 comp 0 prepare failed -12
[ 3945.314460] <inf> component: comp_set_state: comp:1 0x10004 comp_set_state(), state already set to 1
[ 3945.314468] <err> ipc: ipc_cmd: ipc4: FW_GEN_MSG failed with err 7 

[fredoh9]

LNL NOCODEC has same/similar errors.

check-signal-stop-start TPLG=/lib/firmware/intel/development/sof-lnl-nocodec.tplg MODEL=LNLM_RVP_NOCODEC SOF_TEST_INTERVAL=5 ~/sof-test/test-case/check-signal-stop-start.sh -m capture -c 50

[ 3832.775173] <err> buffer: buffer_alloc: buffer_alloc(): could not alloc size = 1536 bytes of type = 32
[ 3832.775176] <err> dai_comp: dai_set_dma_buffer: comp:0 0x4 dai_set_dma_buffer(): failed to alloc dma buffer
[ 3832.775188] <err> dai_comp: dai_common_params: comp:0 0x4 dai_zephyr_params(): alloc dma buffer failed.
[ 3832.775193] <err> module_adapter: module_prepare: comp:0 0x4 module_prepare() error -12: module specific prepare failed, comp_id 4
[ 3832.775198] <err> module_adapter: module_adapter_prepare: comp:0 0x4 module_adapter_prepare() error fffffff4: module prepare failed
[ 3832.775201] <err> pipe: pipeline_prepare: pipe:0 0x0 pipeline_prepare(): ret = -12, dev->comp.id = 4
[ 3832.775206] <err> ipc: ipc4_pcm_params: ipc: pipe 0 comp 0 prepare failed -12
[ 3832.775216] <inf> component: comp_set_state: comp:0 0x4 comp_set_state(), state already set to 1
[ 3832.775220] <err> ipc: ipc_cmd: ipc4: FW_GEN_MSG failed with err 7
[ 3832.775301] <inf> host_comp: host_get_copy_bytes_normal: comp:1 0x10004 no bytes to copy, available samples: 0, free_samples: 384

and multiple-pipeline-playback-50 test TPLG=/lib/firmware/intel/development/sof-lnl-nocodec.tplg MODEL=LNLM_RVP_NOCODEC SOF_TEST_INTERVAL=5 ~/sof-test/test-case/c.sh -f p -l 50

[ 3856.342458] <inf> ipc: buffer_new: buffer new size 0x300 id 0.0 flags 0x0
[ 3856.342463] <err> buffer: buffer_alloc: buffer_alloc(): could not alloc structure
[ 3856.342468] <err> ipc: ipc_comp_connect: failed to allocate buffer to bind 4 to 6
[ 3856.342471] <err> ipc: ipc_cmd: ipc4: MODULE_MSG failed with err 3

Intel internal daily test link: planresultdetail/39105

[plbossart]

This is the last bug preventing us from adding LNL to the CI pool

see Intel daily test #/result/planresultdetail/39452?model=LNLM_RVP_NOCODEC&testcase=check-signal-stop-start-capture-50

Not sure why it's marked as P2 @kv2019i

[kv2019i]

Might be related (and even a fix): https://github.com/thesofproject/sof/pull/9006

[plbossart]

Still happening today in daily tests, see #/result/planresultdetail/39481?model=MTLP_RVP_NOCODEC&testcase=check-xrun-injection-playback-10

[marc-hb]

Again today in result/planresultdetail/39547?model=LNLM_RVP_NOCODEC&testcase=volume-basic-test-50

It behaves like a memory leak?

EDIT: like #9005?

[marc-hb]

Not seen in today's daily tests.