On Tue, 9 Sep 2025 11:00:50 +0800
Dust Li [off-list ref] wrote:

On 2025-09-09 00:01:49, Halil Pasic wrote:

quoted

Think SMC_WR_BUF_CNT_SEND := SMC_WR_BUF_CNT used in send context and
SMC_WR_BUF_CNT_RECV := 3 * SMC_WR_BUF_CNT used in recv context. Those
get replaced with lgr->pref_send_wr and lgr->max_recv_wr respective.  

Yes it is just in the commit message, I messed up the search and replace
in the commit message. :(

                            ^                       ^
                            better to use the same prefix

I personally prefer max_send_wr/max_recv_wr.

Will go back to that then for v3

quoted

While at it let us also remove a confusing comment that is either not
about the context in which it resides (describing
qp_attr.cap.pref_send_wr and qp_attr.cap.max_recv_wr) or not applicable  

                ^
I haven't found pref_send_wr in qp_attr.cap

Again search and replace. Sorry!

[..]

quoted

+
+	Please be aware that all the buffers need to be allocated as a physically
+	continuous array in which each element is a single buffer and has the size
+	of SMC_WR_BUF_SIZE (48) bytes. If the allocation fails we give up much
+	like before having this control.
+	this control.  

The final 'this control' looks unwanted.

 

You are right

[..]

quoted

@@ -741,50 +742,51 @@ int smc_wr_alloc_lgr_mem(struct smc_link_group *lgr)

int smc_wr_alloc_link_mem(struct smc_link *link)
{
	/* allocate link related memory */
-	link->wr_tx_bufs = kcalloc(SMC_WR_BUF_CNT, SMC_WR_BUF_SIZE, GFP_KERNEL);
+	link->wr_tx_bufs = kcalloc(link->lgr->pref_send_wr,
+				   SMC_WR_BUF_SIZE, GFP_KERNEL);
	if (!link->wr_tx_bufs)
		goto no_mem;
-	link->wr_rx_bufs = kcalloc(SMC_WR_BUF_CNT * 3, link->wr_rx_buflen,
+	link->wr_rx_bufs = kcalloc(link->lgr->pref_recv_wr, SMC_WR_BUF_SIZE,
				   GFP_KERNEL);  

I will have to do some digging, let's assume for now that it is my
mistake. Unfortunately I won't be able to revisit this before next
Wednesday.

Thank you for your review!

Regards,
Halil