libsdio: use libsdio in WHD HAL #78
Conversation
JIRA: RTOS-212
| unsigned int dir : 1; /* 31 */ | ||
| } cmd53; | ||
| uint32_t uint; | ||
| } sdio_cmd_arg_t; |
There was a problem hiding this comment.
This type is already defined by whd, so we don't have to redefine it. It's named 'sdio_cmd_argument_t' and is placed whd_bus_sdio_protocol.h.
|
|
||
|
|
||
| /* NOTE: obj is ignored - state is kept in sdio_common */ | ||
| /* NOTE: obj is ignored, state instance is held by the SDIO API*/ |
There was a problem hiding this comment.
Since obj is used, you could add a line (void)obj; so that the compiler does not complain about an unused variable. Also a nitpick:
| /* NOTE: obj is ignored, state instance is held by the SDIO API*/ | |
| /* NOTE: obj is ignored, state instance is held by the SDIO API */ |
| cyhal_gpio_init(0, 0, 0, 0); /* all args are ignored */ | ||
| usleep(10000); /* 10 ms for regulator discharge */ | ||
| cyhal_gpio_write(0, 1); /* only second arg is used */ | ||
| usleep(250000); /* 250 ms for WLAN power up */ |
There was a problem hiding this comment.
A newline before the return statement would be cool ;)
| } | ||
|
|
||
|
|
||
| /* NOTE: obj is ignored - state is kept in sdio_common */ | ||
| /* NOTE: obj is ignored, state instance is held by the SDIO API*/ |
There was a problem hiding this comment.
Nitpick:
| /* NOTE: obj is ignored, state instance is held by the SDIO API*/ | |
| /* NOTE: obj is ignored, state instance is held by the SDIO API */ |
| } | ||
|
|
||
|
|
||
| /* NOTE: obj is ignored - state is kept in sdio_common */ | ||
| /* NOTE: obj is ignored, state instance is held by the SDIO API*/ | ||
| cy_rslt_t cyhal_sdio_configure(cyhal_sdio_t *obj, const cyhal_sdio_cfg_t *config) |
There was a problem hiding this comment.
Well, basically if you don't use this obj variable you should add a (void)obj statement everywhere.
| *response = val; | ||
|
|
||
| return CY_RSLT_SUCCESS; | ||
| /* block size is 0 for some reason, hardcode instead */ |
There was a problem hiding this comment.
I don't understand this comment. Does whd always set blocksz to 0? If yes, then how do you know it should be 64?
| /* WHD only uses send cmd api call for | ||
| * transfers after init commands which are | ||
| * issued by sdio_init function, hence | ||
| * just indicate success and exit */ |
There was a problem hiding this comment.
I'm afraid we can't do it this way - assuming a specifiic behaviour of the upper layer, in this case the whd driver. It might change in the future, we need to keep it generic still. Unfortunately I see no other option here than exporting sdio_sendCmd, which is now static, and maybe check if the cmd is CYHAL_SDIO_CMD_IO_RW_DIRECT and use transferDirect function in this case and sendCmd otherwise.
| } | ||
|
|
||
| size_t len; | ||
| sdio_cmd_arg_t arg; |
There was a problem hiding this comment.
It could be done in an even smarter way:
| sdio_cmd_arg_t arg; | |
| whd_bus_sdio_cmd53_argument_t arg = (whd_bus_sdio_cmd53_argument_t)argument; | |
| val = *(sdio_common.base + int_status); | ||
| if (val & (1 << 8)) | ||
| break; | ||
| cyhal_sdio_irq_handler_t conv; |
There was a problem hiding this comment.
At least it should be static. But I think the most portable solution would be to add a structure for keeping the sdio context with only this value here and make use of the (cyhal_sdio_t *obj) pointer, which gets passed to basically every function.
JIRA: RTOS-212
Description
Motivation and Context
Types of changes
How Has This Been Tested?
Checklist:
Special treatment