Skip to contentSkip to conversation

net: mvpp2: prs: Fix parser range for VID filtering

PATCH In review Maxime Chevallier 2 participants 4 messages Jun 10, 2019 – Jun 11, 2019

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

PATCH 1/2 net: mvpp2: prs: Fix parser range for VID filtering

VID filtering is implemented in the Header Parser, with one range of 11 vids being assigned for each no-loopback port.

Make sure we use the per-port range when looking for existing entries in the Parser.

Since we used a global range instead of a per-port one, this causes VIDs to be removed from the whitelist from all ports of the same PPv2 instance.

Fixes 56beda3db602 ("net: mvpp2: Add hardware offloading for VLAN filtering")
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_prs.c | 17 ++++++++---------
 1 file changed, 8 insertions(+), 9 deletions(-)
drivers/net/ethernet/marvell/mvpp2/mvpp2_prs.c +8 -9
index 392fd895f278..e0da4db3bf56 100644
@@ -1905,8 +1905,7 @@ static int mvpp2_prs_ip6_init(struct mvpp2 *priv)
 }
 
 /* Find tcam entry with matched pair <vid,port> */
-static int mvpp2_prs_vid_range_find(struct mvpp2 *priv, int pmap, u16 vid,
-				    u16 mask)
+static int mvpp2_prs_vid_range_find(struct mvpp2_port *port, u16 vid, u16 mask)
 {
 	unsigned char byte[2], enable[2];
 	struct mvpp2_prs_entry pe;
@@ -1914,13 +1913,13 @@ static int mvpp2_prs_vid_range_find(struct mvpp2 *priv, int pmap, u16 vid,
 	int tid;
 
 	/* Go through the all entries with MVPP2_PRS_LU_VID */
-	for (tid = MVPP2_PE_VID_FILT_RANGE_START;
-	     tid <= MVPP2_PE_VID_FILT_RANGE_END; tid++) {
-		if (!priv->prs_shadow[tid].valid ||
-		    priv->prs_shadow[tid].lu != MVPP2_PRS_LU_VID)
+	for (tid = MVPP2_PRS_VID_PORT_FIRST(port->id);
+	     tid <= MVPP2_PRS_VID_PORT_LAST(port->id); tid++) {
+		if (!port->priv->prs_shadow[tid].valid ||
+		    port->priv->prs_shadow[tid].lu != MVPP2_PRS_LU_VID)
 			continue;
 
-		mvpp2_prs_init_from_hw(priv, &pe, tid);
+		mvpp2_prs_init_from_hw(port->priv, &pe, tid);
 
 		mvpp2_prs_tcam_data_byte_get(&pe, 2, &byte[0], &enable[0]);
 		mvpp2_prs_tcam_data_byte_get(&pe, 3, &byte[1], &enable[1]);
@@ -1950,7 +1949,7 @@ int mvpp2_prs_vid_entry_add(struct mvpp2_port *port, u16 vid)
 	memset(&pe, 0, sizeof(pe));
 
 	/* Scan TCAM and see if entry with this <vid,port> already exist */
-	tid = mvpp2_prs_vid_range_find(priv, (1 << port->id), vid, mask);
+	tid = mvpp2_prs_vid_range_find(port, vid, mask);
 
 	reg_val = mvpp2_read(priv, MVPP2_MH_REG(port->id));
 	if (reg_val & MVPP2_DSA_EXTENDED)
@@ -2008,7 +2007,7 @@ void mvpp2_prs_vid_entry_remove(struct mvpp2_port *port, u16 vid)
 	int tid;
 
 	/* Scan TCAM and see if entry with this <vid,port> already exist */
-	tid = mvpp2_prs_vid_range_find(priv, (1 << port->id), vid, 0xfff);
+	tid = mvpp2_prs_vid_range_find(port, vid, 0xfff);
 
 	/* No such entry */
 	if (tid < 0)
Signature · 1 line
2.20.1
#1 lore raw

PATCH 2/2 net: mvpp2: prs: Use the correct helpers when removing all VID filters

When removing all VID filters, the mvpp2_prs_vid_entry_remove would be called with the TCAM id incorrectly used as a VID, causing the wrong TCAM entries to be invalidated.

Fix this by directly invalidating entries in the VID range.

Fixes 56beda3db602 ("net: mvpp2: Add hardware offloading for VLAN filtering")
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2_prs.c | 6 ++++--
 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/net/ethernet/marvell/mvpp2/mvpp2_prs.c +4 -2
index e0da4db3bf56..ae2240074d8e 100644
@@ -2025,8 +2025,10 @@ void mvpp2_prs_vid_remove_all(struct mvpp2_port *port)
 
 	for (tid = MVPP2_PRS_VID_PORT_FIRST(port->id);
 	     tid <= MVPP2_PRS_VID_PORT_LAST(port->id); tid++) {
-		if (priv->prs_shadow[tid].valid)
-			mvpp2_prs_vid_entry_remove(port, tid);
+		if (priv->prs_shadow[tid].valid) {
+			mvpp2_prs_hw_inv(priv, tid);
+			priv->prs_shadow[tid].valid = false;
+		}
 	}
 }
 
Signature · 1 line
2.20.1
maintainer
#2 lore raw
In reply to Maxime Chevallier

Please start providing proper header postings with each and every patch series, explaining at a high level what the patch series is doing, how it is doing it, and why it is doing it that way.

Thanks.

#3 lore raw

Hello David,

Quoted 3 lines · On Mon, 10 Jun 2019 09:23:00 -0700 (PDT) David Miller <[email protected]>…

Please start providing proper header postings with each and every patch series, explaining at a high level what the patch series is doing, how it is doing it, and why it is doing it that way.

Sure, I'll resend with a proper cover-letter, sorry about that.

Thanks,

Maxime