On Sun, Feb 5, 2017 at 11:36 AM, Vinod Koul <vinod.koul@xxxxxxxxx> wrote:
> On Thu, Feb 02, 2017 at 10:17:15AM +0530, Anup Patel wrote:
>> +config BCM_SBA_RAID
>> + tristate "Broadcom SBA RAID engine support"
>> + depends on (ARM64 && MAILBOX && RAID6_PQ) || COMPILE_TEST
>> + select DMA_ENGINE
>> + select DMA_ENGINE_RAID
>> + select ASYNC_TX_ENABLE_CHANNEL_SWITCH
>> + default ARCH_BCM_IPROC
>
> whats with the funny alignement?
Sure, I will use tabs here.
>
>> +/* SBA command related defines */
>> +#define SBA_TYPE_SHIFT 48
>> +#define SBA_TYPE_MASK 0x3
>> +#define SBA_TYPE_A 0x0
>> +#define SBA_TYPE_B 0x2
>> +#define SBA_TYPE_C 0x3
>> +#define SBA_USER_DEF_SHIFT 32
>> +#define SBA_USER_DEF_MASK 0xffff
>> +#define SBA_R_MDATA_SHIFT 24
>> +#define SBA_R_MDATA_MASK 0xff
>> +#define SBA_C_MDATA_MS_SHIFT 18
>> +#define SBA_C_MDATA_MS_MASK 0x3
>> +#define SBA_INT_SHIFT 17
>> +#define SBA_INT_MASK 0x1
>> +#define SBA_RESP_SHIFT 16
>> +#define SBA_RESP_MASK 0x1
>> +#define SBA_C_MDATA_SHIFT 8
>> +#define SBA_C_MDATA_MASK 0xff
>> +#define SBA_CMD_SHIFT 0
>> +#define SBA_CMD_MASK 0xf
>> +#define SBA_CMD_ZERO_ALL_BUFFERS 0x8
>> +#define SBA_CMD_LOAD_BUFFER 0x9
>> +#define SBA_CMD_XOR 0xa
>> +#define SBA_CMD_GALOIS_XOR 0xb
>> +#define SBA_CMD_ZERO_BUFFER 0x4
>> +#define SBA_CMD_WRITE_BUFFER 0xc
>
> Try using BIT and GENMAST for hardware descriptions
Sure, will do.
>
>> +
>> +/* SBA C_MDATA helper macros */
>> +#define SBA_C_MDATA_LOAD_VAL(__bnum0) ((__bnum0) & 0x3)
>> +#define SBA_C_MDATA_WRITE_VAL(__bnum0) ((__bnum0) & 0x3)
>> +#define SBA_C_MDATA_XOR_VAL(__bnum1, __bnum0) \
>> + ({ u32 __v = ((__bnum0) & 0x3); \
>> + __v |= ((__bnum1) & 0x3) << 2; \
>> + __v; \
>> + })
>> +#define SBA_C_MDATA_PQ_VAL(__dnum, __bnum1, __bnum0) \
>> + ({ u32 __v = ((__bnum0) & 0x3); \
>> + __v |= ((__bnum1) & 0x3) << 2; \
>> + __v |= ((__dnum) & 0x1f) << 5; \
>> + __v; \
>> + })
>
> ah why are we usig complex macros, why can't these be simple functions..
"static inline functions" seemed too complicated here because most of
these macros are two lines of c-code.
Do you still insist on using "static inline functions"?
>
>> +#define SBA_C_MDATA_LS(__c_mdata_val) ((__c_mdata_val) & 0xff)
>> +#define SBA_C_MDATA_MS(__c_mdata_val) (((__c_mdata_val) >> 8) & 0x3)
>> +
>> +/* Driver helper macros */
>> +#define to_sba_request(tx) \
>> + container_of(tx, struct sba_request, tx)
>> +#define to_sba_device(dchan) \
>> + container_of(dchan, struct sba_device, dma_chan)
>> +
>> +enum sba_request_state {
>> + SBA_REQUEST_STATE_FREE = 1,
>> + SBA_REQUEST_STATE_ALLOCED = 2,
>> + SBA_REQUEST_STATE_PENDING = 3,
>> + SBA_REQUEST_STATE_ACTIVE = 4,
>> + SBA_REQUEST_STATE_COMPLETED = 5,
>> + SBA_REQUEST_STATE_ABORTED = 6,
>
> whats up with a very funny indentation setting, we use 8 chars.
>
> Please re-read the Documentation/process/coding-style.rst
I have double checked this enum. The indentation is fine
and as-per coding style. Am I missing anything else?
>
>> +static int sba_alloc_chan_resources(struct dma_chan *dchan)
>> +{
>> + /*
>> + * We only have one channel so we have pre-alloced
>> + * channel resources. Over here we just return number
>> + * of free request.
>> + */
>> + return sba_free_request_count(to_sba_device(dchan));
>> +}
>
> essentially you are not doing much, so you can skip it. Its an optional
> call.
Sure, will do.
>
>> +static void sba_free_chan_resources(struct dma_chan *dchan)
>> +{
>> + /*
>> + * Channel resources are pre-alloced so we just free-up
>> + * whatever we can so that we can re-use pre-alloced
>> + * channel resources next time.
>> + */
>> + sba_cleanup_inflight_requests(to_sba_device(dchan));
>
> well this one checks for pending requests as well, which shouldn't be there
> when freeing a channel, something seems not quite right here..
>
>> +static int sba_send_mbox_request(struct sba_device *sba,
>> + struct sba_request *req)
>> +{
>> + int mchans_idx, ret = 0;
>> +
>> + /* Select mailbox channel in round-robin fashion */
>> + mchans_idx = atomic_inc_return(&sba->mchans_current);
>> + mchans_idx = mchans_idx % sba->mchans_count;
>> +
>> + /* Send batch message for the request */
>> + req->bmsg.batch.msgs_queued = 0;
>> + ret = mbox_send_message(sba->mchans[mchans_idx], &req->bmsg);
>> + if (ret < 0) {
>> + dev_info(sba->dev, "channel %d message %d (total %d)",
>> + mchans_idx, req->bmsg.batch.msgs_queued,
>> + req->bmsg.batch.msgs_count);
>
> dev_err?
Sure, will use dev_err.
>
>> + dev_err(sba->dev, "send message failed with error %d", ret);
>> + return ret;
>> + }
>> + ret = req->bmsg.error;
>> + if (ret < 0) {
>> + dev_info(sba->dev,
>> + "mbox channel %d message %d (total %d)",
>> + mchans_idx, req->bmsg.batch.msgs_queued,
>> + req->bmsg.batch.msgs_count);
>
> same here
OK.
>
>> +static dma_cookie_t sba_tx_submit(struct dma_async_tx_descriptor *tx)
>> +{
>> + unsigned long flags;
>> + dma_cookie_t cookie;
>> + struct sba_request *req;
>> + struct sba_device *sba;
>> +
>> + if (unlikely(!tx))
>> + return -EINVAL;
>> +
>> + sba = to_sba_device(tx->chan);
>> + req = to_sba_request(tx);
>> +
>> + /* Assign cookie and mark request pending */
>> + spin_lock_irqsave(&sba->reqs_lock, flags);
>> + cookie = dma_cookie_assign(tx);
>> + _sba_pending_request(sba, req);
>> + spin_unlock_irqrestore(&sba->reqs_lock, flags);
>> +
>> + /* Try to submit pending request */
>> + sba_issue_pending(&sba->dma_chan);
>
> Nope, thats wrong, caller needs to call .issue_pending for that
This was giving minor performance improvement but I will
remove this since its against API usage.
>
>> +static enum dma_status sba_tx_status(struct dma_chan *dchan,
>> + dma_cookie_t cookie,
>> + struct dma_tx_state *txstate)
>> +{
>> + int mchan_idx;
>> + enum dma_status ret;
>> + struct sba_device *sba = to_sba_device(dchan);
>> +
>> + ret = dma_cookie_status(dchan, cookie, txstate);
>> + if (ret == DMA_COMPLETE)
>> + return ret;
>> +
>> + for (mchan_idx = 0; mchan_idx < sba->mchans_count; mchan_idx++)
>> + mbox_client_peek_data(sba->mchans[mchan_idx]);
>
> what is this achieving?
The mbox_client_peek_data() is a hint to mailbox controller driver
to check for available messages.
This gives good performance improvement when some DMA client
code is polling using tx_status() callback.
>
>> +static struct dma_async_tx_descriptor *
>> +sba_prep_dma_memcpy(struct dma_chan *dchan, dma_addr_t dst, dma_addr_t src,
>> + size_t len, unsigned long flags)
>> +{
>> + size_t msg_len;
>> + dma_addr_t msg_offset = 0;
>> + unsigned int msgs_count = 0, cmds_count, cmds_idx = 0;
>> + struct sba_device *sba = to_sba_device(dchan);
>> + struct sba_request *req = NULL;
>> +
>> + /* Sanity checks */
>> + if (unlikely(len > sba->req_size))
>> + return NULL;
>
> why is that an error, you can create multiple txn of max length
Sure, I will extend driver to create multiple txn when
"len > req->size"
>
>> +static int sba_async_register(struct sba_device *sba)
>> +{
>> + int ret;
>> + struct dma_device *dma_dev = &sba->dma_dev;
>> +
>> + /* Initialize DMA channel cookie */
>> + sba->dma_chan.device = dma_dev;
>> + dma_cookie_init(&sba->dma_chan);
>> +
>> + /* Initialize DMA device capability mask */
>> + dma_cap_zero(dma_dev->cap_mask);
>> + dma_cap_set(DMA_MEMCPY, dma_dev->cap_mask);
>> + dma_cap_set(DMA_XOR, dma_dev->cap_mask);
>> + dma_cap_set(DMA_PQ, dma_dev->cap_mask);
>> +
>> + /*
>> + * Set mailbox channel device as the base device of
>> + * our dma_device because the actual memory accesses
>> + * will be done by mailbox controller
>> + */
>> + dma_dev->dev = sba->mbox_dev;
>> +
>> + /* Set base prep routines */
>> + dma_dev->device_alloc_chan_resources = sba_alloc_chan_resources;
>> + dma_dev->device_free_chan_resources = sba_free_chan_resources;
>> + dma_dev->device_issue_pending = sba_issue_pending;
>> + dma_dev->device_tx_status = sba_tx_status;
>
> Please add terminate callback support, also add the capabilities, we need to
> advertise that and use in clients
OK, I will add terminate callback.
>
> Also you can simplify bunch of code by using virt-chan support for managing
> channels and descriptors
OK, I will surely explore virt-chan.
Regards,
Anup
