Dear Eric Nelson,
On 03/06/2012 12:49 PM, Marek Vasut wrote:
Dear Eric Nelson,
On 03/06/2012 10:06 AM, Eric Nelson wrote:
On 03/05/2012 01:06 PM, Marek Vasut wrote:
Dear Eric Nelson,
<snip> > > + if (!fec->rbd_base) { > + ret = -ENOMEM; > + goto err2; > + } > + memset(fec->rbd_base, 0, size); > + }
We might want to flush the descriptors to memory after they have been inited?
We're also missing a flush after the call to fec_rbd_init().
I think we need only one flush in the whole allocation sequence. But you're probably right here.
Yeah. There's no point in doing memset() and flush() just to write them again and not flush the proper values.
I'm inclined to move that call to right after the memset and before a newly-added flush and do the same with the call to tbd_init().
You mean into fec_rbd_init() ?
fec_tbd_init(). See below.
diff --git a/drivers/net/fec_mxc.c b/drivers/net/fec_mxc.c index 0db5ca9..d5d0d5e 100644 --- a/drivers/net/fec_mxc.c +++ b/drivers/net/fec_mxc.c @@ -544,6 +544,7 @@ static int fec_init(struct eth_device *dev, bd_t* bd) goto err1; } memset(fec->tbd_base, 0, size);
fec_tbd_init(fec); flush_dcache_range((unsigned)fec->tbd_base, size); }@@ -560,6 +561,13 @@ static int fec_init(struct eth_device *dev, bd_t* bd) goto err2; } memset(fec->rbd_base, 0, size);
/** Initialize RxBD ring*/if (fec_rbd_init(fec, FEC_RBD_NUM, FEC_MAX_PKT_SIZE) < 0) {ret = -ENOMEM;goto err3;} flush_dcache_range((unsigned)fec->rbd_base, (unsigned)fec->rbd_base + size); }@@ -619,16 +627,6 @@ static int fec_init(struct eth_device *dev, bd_t* bd) writel((uint32_t)fec->tbd_base, &fec->eth->etdsr); writel((uint32_t)fec->rbd_base, &fec->eth->erdsr);
/** Initialize RxBD/TxBD rings*/if (fec_rbd_init(fec, FEC_RBD_NUM, FEC_MAX_PKT_SIZE) < 0) {ret = -ENOMEM;goto err3;}fec_tbd_init(fec);
I see, makes sense.
Best regards, Marek Vasut