consider this situation: no more fresh block, meantime an item with such a big size come in, and currently no block inuse has sufficient space to save this item, the code will run into the modified line, and KV_MGR_WORKSPACE will be set to KV_BLK_INVALID. this will cause no matter what size of an item wanna come in again, kv_blk_search_suitable just return KV_BLK_INVALID because "if (KV_MGR_WORKSPACE == KV_BLK_INVALID)" is true.
1. a true wear-leveling kv fs for norflash, especially optimize for some onchip norflash with "write once on one single write unit" like stm32l4, a true "no earse before write" flash algorithm.
2. an "as less as possible" gc strategy, do best to save norflash's life.
3. full "power down protection" support
4. see "examples" of kv, project in "TencentOS_tiny_EVB_MX_Plus", with onchip flash and qspiflash sample.
1. add TOS_CFG_LIBC_PRINTF_EN
1: enable printf to implement tos_kprintf
when you call tos_kprintf(actually currently it's a MACRO rather than a function) , the libc will be invloved into the image, most of the time it means a larger image size.
0: disable printf, means tos_kprintf is defined as an empty MACRO
most of the time, means you will not get a log output from the uart(currently we donnot have a console yet) but a smaller image size.
2. fix a bug of tos_mail_q and tos_priority_mail_q
if we are post-ing a mail and there is[are] task[s] just pend-ing for the mail queue, the mail delivered to the pend-er will be a pointer, and probably it's a pointer to the local variable on the stack. Actually we should do a memcpy rather than a pointer assignment. for more details, see the patch.
