On Mon, Feb 23, 2009 at 11:47:19AM +0530, arun c wrote:
...
> Yes I don't need any lock to protect my data
> here, as host is write only and target is read only.
>
> Here is what I did.
>
> typedef command_pkt {
> u32 valid;
> ......
> ......
> ......
> u32 data[];
> }
>
> typedef struct my_circ_buf {
> u32 read_index;
> u32 write_index;
> command_pkt_t cmd_pkt[30];
> } my_circ_buf_t;
>
> Host writes commands,
> makes valid field to VALID,
> and increments write_index.
I suspect you need/want at least one wmb() between those three steps
depending on what the remote host is polling.
The "VALID" field sounds redundant to me and I would drop it.
Updating the write_index should be enough clue for the remote side
to know which indexs are valid.
> Target reads commands at read_index,
> if and only if valid field is VALID and
> after processing it makes valid field
> INVALID and increments read_index.
So "VALID" really means "BUSY" or "INFLIGHT".
Don't really need share both VALID bit and head/tail indexes.
Avoiding sharing of the index could conserve a fair amount of CPU cycles
on the host side.
> It seems to be working for me
> I need to investigate further for any
> cache coherency issues, write now
> I am running with dcache off.
This is arch specific. Host side is usually cache coherent.
IIRC, some embedded PPC are not.
> Regards,
> Arun C
--
To unsubscribe from this list: send the line "unsubscribe linux-pci" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at http://vger.kernel.org/majordomo-info.html
