FreeCalypso > hg > fc-tourmaline
view src/cs/services/audio/audio_sr_reco.c @ 220:0ed36de51973
ABB semaphore protection overhaul
The ABB semaphone protection logic that came with TCS211 from TI
was broken in several ways:
* Some semaphore-protected functions were called from Application_Initialize()
context. NU_Obtain_Semaphore() called with NU_SUSPEND fails with
NU_INVALID_SUSPEND in this context, but the return value wasn't checked,
and NU_Release_Semaphore() would be called unconditionally at the end.
The latter call would increment the semaphore count past 1, making the
semaphore no longer binary and thus no longer effective for resource
protection. The fix is to check the return value from NU_Obtain_Semaphore()
and skip the NU_Release_Semaphore() call if the semaphore wasn't properly
obtained.
* Some SPI hardware manipulation was being done before entering the semaphore-
protected critical section. The fix is to reorder the code: first obtain
the semaphore, then do everything else.
* In the corner case of L1/DSP recovery, l1_abb_power_on() would call some
non-semaphore-protected ABB & SPI init functions. The fix is to skip those
calls in the case of recovery.
* A few additional corner cases existed, all of which are fixed by making
ABB semaphore protection 100% consistent for all ABB functions and code paths.
There is still one remaining problem of priority inversion: suppose a low-
priority task calls an ABB function, and some medium-priority task just happens
to preempt right in the middle of that semaphore-protected ABB operation. Then
the high-priority SPI task is locked out for a non-deterministic time until
that medium-priority task finishes its work and goes back to sleep. This
priority inversion problem remains outstanding for now.
| author | Mychaela Falconia <falcon@freecalypso.org> |
|---|---|
| date | Mon, 26 Apr 2021 20:55:25 +0000 |
| parents | 4e78acac3d88 |
| children |
line wrap: on
line source
/****************************************************************************/ /* */ /* File Name: audio_sr_reco.c */ /* */ /* Purpose: This file contains all the functions used to manage the */ /* recognition of a word for the speech recognition module. */ /* */ /* Version 0.1 */ /* */ /* Date Modification */ /* ------------------------------------ */ /* 09 Oct. 2001 Create */ /* */ /* Author */ /* Francois Mazard - Stephanie Gerthoux */ /* */ /* (C) Copyright 2001 by Texas Instruments Incorporated, All Rights Reserved*/ /****************************************************************************/ #include "rv/rv_defined_swe.h" #ifdef RVM_AUDIO_MAIN_SWE #ifndef _WINDOWS #include "config/swconfig.cfg" #include "config/sys.cfg" #include "config/chipset.cfg" #endif /* include the usefull L1 header */ #include "l1_confg.h" #if (SPEECH_RECO) #include "rv/rv_general.h" #include "rvm/rvm_gen.h" #include "audio/audio_features_i.h" #include "audio/audio_ffs_i.h" #include "audio/audio_api.h" #include "audio/audio_structs_i.h" #include "audio/audio_var_i.h" #include "audio/audio_messages_i.h" #include "rvf/rvf_target.h" #include "audio/audio_const_i.h" #include "audio/audio_error_hdlr_i.h" #include "audio/audio_macro_i.h" /* include the usefull L1 header */ #define BOOL_FLAG #define CHAR_FLAG #include "l1_types.h" #include "l1audio_cust.h" #include "l1audio_msgty.h" #include "l1audio_signa.h" #include "l1audio_cust.h" #include <string.h> /********************************************************************************/ /* */ /* Function Name: audio_sr_reco_convert_parameter */ /* */ /* Purpose: Convert the speech reco parameters from the entity to */ /* the l1 parameters */ /* */ /* Input Parameters: */ /* entity speech reco message */ /* */ /* Output Parameters: */ /* layer 1 speech reco message */ /* */ /* Note: */ /* None. */ /* */ /* Revision History: */ /* None. */ /* */ /********************************************************************************/ void audio_sr_reco_convert_parameter( T_AUDIO_SR_RECO_START *entity_parameter, T_MMI_SR_RECO_REQ *l1_parameter) { l1_parameter->database_id = 0; l1_parameter->vocabulary_size = entity_parameter->vocabulary_size; } /********************************************************************************/ /* */ /* Function Name: audio_sr_reco_send_status */ /* */ /* Purpose: This function sends the speech reco status to the */ /* entity. */ /* */ /* Input Parameters: */ /* status, */ /* return path */ /* */ /* Output Parameters: */ /* None. */ /* */ /* Note: */ /* None. */ /* */ /* Revision History: */ /* None. */ /* */ /********************************************************************************/ void audio_sr_reco_send_status (T_AUDIO_RET status, UINT8 word_index, T_RV_RETURN return_path) { void *p_send_message = NULL; T_RVF_MB_STATUS mb_status = RVF_RED; #ifndef _WINDOWS UINT8 i; char *p_database; char *p_word_name; UINT8 model_name_lengh; #endif while (mb_status == RVF_RED) { /* allocate the message buffer */ mb_status = rvf_get_buf (p_audio_gbl_var->mb_external, sizeof (T_AUDIO_SR_RECO_STATUS), (T_RVF_BUFFER **) (&p_send_message)); /* If insufficient resources, then report a memory error and abort. */ /* and wait until more ressource is given */ if (mb_status == RVF_RED) { audio_sr_error_trace(AUDIO_ENTITY_NO_MEMORY); rvf_delay(RVF_MS_TO_TICKS(1000)); } } /*fill the header of the message */ ((T_AUDIO_SR_RECO_STATUS *)(p_send_message))->os_hdr.msg_id = AUDIO_SR_RECO_STATUS_MSG; if (status > 0) { status = (INT8)((-2) - status); } if (status == SC_NO_ERROR) /* no error occured dring the recognition */ { #ifndef _WINDOWS /* search the name of the recognized word */ p_database = p_audio_gbl_var->speech_reco.sr_reco.p_database; for(i=0; i< word_index; i++) { p_database += AUDIO_PATH_NAME_MAX_SIZE; } /* the _sr need to be suppress */ model_name_lengh = (strlen(p_database) - 3); /* copy this path in the message */ strncpy(((T_AUDIO_SR_RECO_STATUS *)(p_send_message))->word_recognized, p_database, model_name_lengh); p_word_name = (char *)(((T_AUDIO_SR_RECO_STATUS *)(p_send_message))->word_recognized); p_word_name += model_name_lengh; *p_word_name = 0; #endif } /* fill the status parameters */ ((T_AUDIO_SR_RECO_STATUS *)(p_send_message))->status = status; if (return_path.callback_func == NULL) { /* send the message to the entity */ rvf_send_msg (return_path.addr_id, p_send_message); } else { /* call the callback function */ (*return_path.callback_func)((void *)(p_send_message)); rvf_free_buf((T_RVF_BUFFER *)p_send_message); } } /********************************************************************************/ /* */ /* Function Name: audio_sr_reco_manager */ /* */ /* Purpose: This function is called to manage a speech reco */ /* process. */ /* */ /* Input Parameters: */ /* Audio Speech reco reco Parameters, */ /* */ /* Output Parameters: */ /* None. */ /* */ /* Note: */ /* None. */ /* */ /* Revision History: */ /* None. */ /* */ /********************************************************************************/ void audio_sr_reco_manager (T_RV_HDR *p_message) { /* Declare local variables. */ void *p_send_message; /**************** audio_sr_reco_manager function begins *********************/ switch(p_audio_gbl_var->speech_reco.sr_reco.state) { case AUDIO_SR_RECO_IDLE: { switch(p_message->msg_id) { case AUDIO_SR_RECO_START_REQ: { /* save the task id of the entity */ p_audio_gbl_var->speech_reco.sr_reco.task_id = ((T_AUDIO_SR_RECO_START*)(p_message))->os_hdr.src_addr_id; /* save the return path */ p_audio_gbl_var->speech_reco.sr_reco.return_path.callback_func = ((T_AUDIO_SR_RECO_START*)(p_message))->return_path.callback_func; p_audio_gbl_var->speech_reco.sr_reco.return_path.addr_id = ((T_AUDIO_SR_RECO_START*)(p_message))->return_path.addr_id; /* save the message parameters */ p_audio_gbl_var->speech_reco.sr_reco.vocabulary_size = ((T_AUDIO_SR_RECO_START*)(p_message))->vocabulary_size; p_audio_gbl_var->speech_reco.sr_reco.p_database = ((T_AUDIO_SR_RECO_START*)(p_message))->p_database; /* allocate the buffer for the message to the L1 */ p_send_message = audio_allocate_l1_message(sizeof(T_MMI_SR_RECO_REQ)); if ( p_send_message != NULL) { /* Convert the entity parameters to the audio L1 parameters */ audio_sr_reco_convert_parameter((T_AUDIO_SR_RECO_START *)p_message, (T_MMI_SR_RECO_REQ *)p_send_message); /* send the start command to the audio L1 */ audio_send_l1_message(MMI_SR_RECO_START_REQ, p_send_message); } /* change to the state AUDIO_SR_RECO_WAIT_START_CONFIRMATION */ p_audio_gbl_var->speech_reco.sr_reco.state = AUDIO_SR_RECO_WAIT_START_CON; break; } case AUDIO_SR_RECO_STOP_REQ: { /* event error - send an error message*/ audio_sr_reco_send_status (AUDIO_ERROR, 0, ((T_AUDIO_SR_RECO_STOP *)(p_message))->return_path); audio_sr_error_trace(AUDIO_ERROR_STOP_EVENT); break; } } break; } /* case AUDIO_SR_RECO_IDLE */ case AUDIO_SR_RECO_WAIT_START_CON: { switch(p_message->msg_id) { case MMI_SR_RECO_START_CON: { /* change to the state AUDIO_SR_RECO_WAIT_STOP_COMMAND */ p_audio_gbl_var->speech_reco.sr_reco.state = AUDIO_SR_RECO_WAIT_STOP_COMMAND; break; } case AUDIO_SR_RECO_STOP_REQ: /*.Before stopping this task, control that */ /* stop task id caller = SR reco task id */ { if ( p_audio_gbl_var->speech_reco.sr_reco.task_id == ((T_AUDIO_SR_RECO_STOP*)(p_message))->os_hdr.src_addr_id) { /* save the return path */ p_audio_gbl_var->speech_reco.sr_reco.return_path.callback_func = ((T_AUDIO_SR_RECO_STOP*)(p_message))->return_path.callback_func; p_audio_gbl_var->speech_reco.sr_reco.return_path.addr_id = ((T_AUDIO_SR_RECO_STOP*)(p_message))->return_path.addr_id; /* change to the state AUDIO_SR_RECO_WAIT_STOP_CONFIRMATION */ p_audio_gbl_var->speech_reco.sr_reco.state = AUDIO_SR_RECO_WAIT_START_CON_TO_STOP; } else { /* A stop request from an other task is sent during a start connection */ /* event error - send an error message */ audio_sr_reco_send_status (AUDIO_ERROR, 0, ((T_AUDIO_SR_RECO_STOP *)(p_message))->return_path); audio_sr_error_trace(AUDIO_ERROR_STOP_EVENT); } break; } case AUDIO_SR_RECO_START_REQ: { /* deallocate the vocabulary data buffer */ rvf_free_buf( (T_RVF_BUFFER *)(((T_AUDIO_SR_RECO_START *)(p_message))->p_database)); /* event error - send an error message*/ audio_sr_reco_send_status (AUDIO_ERROR, 0, ((T_AUDIO_SR_RECO_START *)(p_message))->return_path); audio_sr_error_trace(AUDIO_ERROR_START_EVENT); break; } } break; } /* case AUDIO_SR_RECO_WAIT_START_CON */ case AUDIO_SR_RECO_WAIT_START_CON_TO_STOP: { switch(p_message->msg_id) { case MMI_SR_RECO_START_CON: { /* send the stop command to the audio L1 */ /* allocate the buffer for the message to the L1 */ p_send_message = audio_allocate_l1_message(0); if ( p_send_message != NULL) { /* send the start command to the audio L1 */ audio_send_l1_message(MMI_SR_RECO_STOP_REQ, p_send_message); } /* change to the state AUDIO_SR_RECO_WAIT_STOP_CON */ p_audio_gbl_var->speech_reco.sr_reco.state = AUDIO_SR_RECO_WAIT_STOP_CON; break; } case AUDIO_SR_RECO_START_REQ: { /* deallocate the vocabulary data buffer */ rvf_free_buf( (T_RVF_BUFFER *)(((T_AUDIO_SR_RECO_START *)(p_message))->p_database)); /* event error - send an error message*/ audio_sr_reco_send_status (AUDIO_ERROR, 0, ((T_AUDIO_SR_RECO_START *)(p_message))->return_path); audio_sr_error_trace(AUDIO_ERROR_START_EVENT); break; } case AUDIO_SR_RECO_STOP_REQ: { /* event error - send an error message*/ audio_sr_reco_send_status (AUDIO_ERROR, 0, ((T_AUDIO_SR_RECO_STOP *)(p_message))->return_path); audio_sr_error_trace(AUDIO_ERROR_STOP_EVENT); break; } } break; } case AUDIO_SR_RECO_WAIT_STOP_COMMAND: { switch(p_message->msg_id) { case MMI_SR_RECO_STOP_CON: { /* deallocate the vocabulary data buffer */ rvf_free_buf( (T_RVF_BUFFER *)(p_audio_gbl_var->speech_reco.sr_reco.p_database)); audio_sr_reco_send_status (((T_MMI_SR_RECO_STOP_CON *)p_message)->error_id, (UINT8)((T_MMI_SR_RECO_STOP_CON *)p_message)->best_word_index, p_audio_gbl_var->speech_reco.sr_reco.return_path); /* change to the state AUDIO_SR_RECO_IDLE */ p_audio_gbl_var->speech_reco.sr_reco.state = AUDIO_SR_RECO_IDLE; break; } case AUDIO_SR_RECO_STOP_REQ: { /*.Before stopping this task, control that */ /* stop task id caller = SR reco task id */ if ( p_audio_gbl_var->speech_reco.sr_reco.task_id == ((T_AUDIO_SR_RECO_STOP*)(p_message))->os_hdr.src_addr_id) { /* save the return path */ p_audio_gbl_var->speech_reco.sr_reco.return_path.callback_func = ((T_AUDIO_SR_RECO_STOP*)(p_message))->return_path.callback_func; p_audio_gbl_var->speech_reco.sr_reco.return_path.addr_id = ((T_AUDIO_SR_RECO_STOP*)(p_message))->return_path.addr_id; /* send the stop command to the audio L1 */ /* allocate the buffer for the message to the L1 */ p_send_message = audio_allocate_l1_message(0); if ( p_send_message != NULL) { /* send the start command to the audio L1 */ audio_send_l1_message(MMI_SR_RECO_STOP_REQ, p_send_message); } /* change to the state AUDIO_SR_RECO_WAIT_STOP_CONFIRMATION */ p_audio_gbl_var->speech_reco.sr_reco.state = AUDIO_SR_RECO_WAIT_STOP_CON; } else { /* A stop request from an other task is sent during a start connection */ /* event error - send an error message */ audio_sr_reco_send_status (AUDIO_ERROR, 0, ((T_AUDIO_SR_RECO_STOP *)(p_message))->return_path); audio_sr_error_trace(AUDIO_ERROR_STOP_EVENT); } break; } case AUDIO_SR_RECO_START_REQ: { /* deallocate the vocabulary data buffer */ rvf_free_buf( (T_RVF_BUFFER *)(((T_AUDIO_SR_RECO_START *)(p_message))->p_database)); /* event error - send an error message*/ audio_sr_reco_send_status (AUDIO_ERROR, 0, ((T_AUDIO_SR_RECO_START *)(p_message))->return_path); audio_sr_error_trace(AUDIO_ERROR_START_EVENT); break; } } break; } /* case AUDIO_SR_RECO_WAIT_STOP_COMMAND */ case AUDIO_SR_RECO_WAIT_STOP_CON: { switch(p_message->msg_id) { case MMI_SR_RECO_STOP_CON: { /* deallocate the vocabulary data buffer */ rvf_free_buf( (T_RVF_BUFFER *)(p_audio_gbl_var->speech_reco.sr_reco.p_database)); audio_sr_reco_send_status (((T_MMI_SR_RECO_STOP_CON *)p_message)->error_id, (UINT8)((T_MMI_SR_RECO_STOP_CON *)p_message)->best_word_index, p_audio_gbl_var->speech_reco.sr_reco.return_path); /* change to the state AUDIO_SR_RECO_IDLE */ p_audio_gbl_var->speech_reco.sr_reco.state = AUDIO_SR_RECO_IDLE; break; } case AUDIO_SR_RECO_STOP_REQ: { /* event error - send an error message*/ audio_sr_reco_send_status (AUDIO_ERROR, 0, ((T_AUDIO_SR_RECO_STOP *)(p_message))->return_path); audio_sr_error_trace(AUDIO_ERROR_STOP_EVENT); break; } case AUDIO_SR_RECO_START_REQ: { /* deallocate the vocabulary data buffer */ rvf_free_buf( (T_RVF_BUFFER *)(((T_AUDIO_SR_RECO_START *)(p_message))->p_database)); /* event error - send an error message*/ audio_sr_reco_send_status (AUDIO_ERROR, 0, ((T_AUDIO_SR_RECO_START *)(p_message))->return_path); audio_sr_error_trace(AUDIO_ERROR_START_EVENT); break; } } break; } /* case AUDIO_SR_RECO_WAIT_STOP_CON */ } /* switch(p_audio_gbl_var->speech_reco.state) */ } /*********************** End of audio_sr_reco_manager function **********************/ #endif /* #if (SPEECH_RECO) */ #endif /* #ifdef RVM_AUDIO_MAIN_SWE */