Skip to contentSkip to conversation

bnx2x: tx_start_bd->vlan_or_ethertype is le16

PATCH Applied Chas Williams 4 participants 11 messages Nov 3, 2015 – Dec 6, 2015 Network drivers

lore
Share & tools
lore
#0 lore raw

PATCH 1/1 bnx2x: tx_start_bd->vlan_or_ethertype is le16

---
 drivers/net/bnx2x/bnx2x.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/bnx2x/bnx2x.c +2 -1
index fed7a06..76444eb 100644
@@ -2169,7 +2169,8 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq, struct rte_mbuf **m_head, int m_p
 				struct ether_hdr *eh
 				    = rte_pktmbuf_mtod(m0, struct ether_hdr *);
 
-				tx_start_bd->vlan_or_ethertype = eh->ether_type;
+				tx_start_bd->vlan_or_ethertype
+				    = rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));
 			}
 		}
 
Signature · 1 line
2.1.0
#1 lore raw

Anyone to review please?

2015-11-03 12:26, Chas Williams:

Quoted 18 lines
---
 drivers/net/bnx2x/bnx2x.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/bnx2x/bnx2x.c +2 -1
index fed7a06..76444eb 100644
@@ -2169,7 +2169,8 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq, struct rte_mbuf **m_head, int m_p
 				struct ether_hdr *eh
 				    = rte_pktmbuf_mtod(m0, struct ether_hdr *);
 
-				tx_start_bd->vlan_or_ethertype = eh->ether_type;
+				tx_start_bd->vlan_or_ethertype
+				    = rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));
 			}
 		}
Acked-by
#2 lore raw
Quoted 18 lines

Anyone to review please?

2015-11-03 12:26, Chas Williams:

Quoted 14 lines
---
 drivers/net/bnx2x/bnx2x.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/bnx2x/bnx2x.c +0 -0
index fed7a06..76444eb 100644
@@ -2169,7 +2169,8 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq,

struct rte_mbuf **m_head, int m_p
struct ether_hdr *eh
= rte_pktmbuf_mtod(m0, struct ether_hdr *);

Quoted 7 lines

drivers/net/bnx2x/bnx2x.c int bnx2x_tx_encap(struct bnx2x_tx_queue *txq, struct rte_mbuf **m_head, int m_p · line 2175

Quoted 5 lines
drivers/net/bnx2x/bnx2x.c +2 -1
-                            tx_start_bd->vlan_or_ethertype = eh->ether_type;
+                            tx_start_bd->vlan_or_ethertype
+                                = rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));
                     }
             }

Minor question - any specific reason to use rte_be_to_cpu_16() on ether_type alone before converting from native order to le16?

Thanks,
Harish

________________________________

This message and any attached documents contain information from the sending company or its parent company(s), subsidiaries, divisions or branch offices that may be confidential. If you are not the intended recipient, you may not read, copy, distribute, or use this information. If you have received this transmission in error, please notify the sender immediately by reply e-mail and then delete this message.

#3 lore raw
Quoted 26 lines · On Tue, 1 Dec 2015 21:53:59 +0000 Harish Patil <[email protected]> wrote:
Quoted 18 lines

Anyone to review please?

2015-11-03 12:26, Chas Williams:

Quoted 14 lines
---
 drivers/net/bnx2x/bnx2x.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/bnx2x/bnx2x.c +0 -0
index fed7a06..76444eb 100644
@@ -2169,7 +2169,8 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq,

struct rte_mbuf **m_head, int m_p
struct ether_hdr *eh
= rte_pktmbuf_mtod(m0, struct ether_hdr *);

Quoted 7 lines

drivers/net/bnx2x/bnx2x.c int bnx2x_tx_encap(struct bnx2x_tx_queue *txq, struct rte_mbuf **m_head, int m_p · line 2175

Quoted 5 lines
drivers/net/bnx2x/bnx2x.c +2 -1
-                            tx_start_bd->vlan_or_ethertype = eh->ether_type;
+                            tx_start_bd->vlan_or_ethertype
+                                = rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));
                     }
             }  
Quoted 5 lines

Minor question - any specific reason to use rte_be_to_cpu_16() on ether_type alone before converting from native order to le16?

ether_type is in network byte order (big endian)
and hardware wants little endian. On x86 the second step is a nop.

#4 lore raw
Quoted 33 lines
Quoted 29 lines · On Tue, 1 Dec 2015 21:53:59 +0000 Harish Patil <[email protected]> wrote:
Quoted 9 lines

Anyone to review please?

2015-11-03 12:26, Chas Williams:

Quoted 5 lines
---
 drivers/net/bnx2x/bnx2x.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
Quoted 8 lines
Quoted 8 lines
drivers/net/bnx2x/bnx2x.c +0 -0
index fed7a06..76444eb 100644
@@ -2169,7 +2169,8 @@ int bnx2x_tx_encap(struct bnx2x_tx_queue *txq,

struct rte_mbuf **m_head, int m_p
struct ether_hdr *eh
= rte_pktmbuf_mtod(m0, struct

ether_hdr *);

Quoted 2 lines
Quoted 2 lines

- tx_start_bd->vlan_or_ethertype =

eh->ether_type;

Quoted 2 lines

drivers/net/bnx2x/bnx2x.c

Quoted 2 lines
drivers/net/bnx2x/bnx2x.c +2 -0
+                            tx_start_bd->vlan_or_ethertype
+                                =

rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));

Quoted 4 lines
Quoted 2 lines
                     }
             }
Quoted 9 lines
Quoted 5 lines

Minor question - any specific reason to use rte_be_to_cpu_16() on ether_type alone before converting from native order to le16?

ether_type is in network byte order (big endian)
and hardware wants little endian. On x86 the second step is a nop.

Thanks. Yes the question was for second step, second step would be a no-op on x86 - thanks for clarifying.

________________________________

This message and any attached documents contain information from the sending company or its parent company(s), subsidiaries, divisions or branch offices that may be confidential. If you are not the intended recipient, you may not read, copy, distribute, or use this information. If you have received this transmission in error, please notify the sender immediately by reply e-mail and then delete this message.

#5 lore raw
In reply to Stephen Hemminger

2015-12-01 14:37, Stephen Hemminger:

Quoted 11 lines
Quoted 9 lines · Harish Patil
Quoted 6 lines

2015-11-03 12:26, Chas Williams:

drivers/net/bnx2x/bnx2x.c int bnx2x_tx_encap(struct bnx2x_tx_queue *txq, struct rte_mbuf **m_head, int m_p · line 2173

Quoted 5 lines
drivers/net/bnx2x/bnx2x.c +2 -1
-                            tx_start_bd->vlan_or_ethertype = eh->ether_type;
+                            tx_start_bd->vlan_or_ethertype
+                                = rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));

Minor question - any specific reason to use rte_be_to_cpu_16() on ether_type alone before converting from native order to le16?

Quoted 2 lines

ether_type is in network byte order (big endian)
and hardware wants little endian. On x86 the second step is a nop.

Doesn't it deserve a macro rte_ntole16()?
It may be in lib/librte_eal/common/include/generic/rte_byteorder.h.

#6 lore raw
Quoted 12 lines · Thomas Monjalon

2015-12-01 14:37, Stephen Hemminger:

Quoted 11 lines
Quoted 9 lines · Harish Patil
Quoted 6 lines

2015-11-03 12:26, Chas Williams:

drivers/net/bnx2x/bnx2x.c int bnx2x_tx_encap(struct bnx2x_tx_queue *txq, struct rte_mbuf **m_head, int m_p · line 2173

Quoted 5 lines
drivers/net/bnx2x/bnx2x.c +2 -1
-                            tx_start_bd->vlan_or_ethertype = eh->ether_type;
+                            tx_start_bd->vlan_or_ethertype
+                                = rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));

Minor question - any specific reason to use rte_be_to_cpu_16() on ether_type alone before converting from native order to le16?

Quoted 5 lines
Quoted 2 lines

ether_type is in network byte order (big endian)
and hardware wants little endian. On x86 the second step is a nop.

Doesn't it deserve a macro rte_ntole16()?
It may be in lib/librte_eal/common/include/generic/rte_byteorder.h.

I looked I didn't see anything. This value, according to the linux driver, wants to be little endian regardless of the host endian.

#7 lore raw

2015-12-01 18:58, Charles Williams:

Quoted 13 lines · Thomas Monjalon
Quoted 12 lines

2015-12-01 14:37, Stephen Hemminger:

Quoted 11 lines
Quoted 9 lines · Harish Patil
Quoted 6 lines

2015-11-03 12:26, Chas Williams:

drivers/net/bnx2x/bnx2x.c int bnx2x_tx_encap(struct bnx2x_tx_queue *txq, struct rte_mbuf **m_head, int m_p · line 2173

Quoted 5 lines
drivers/net/bnx2x/bnx2x.c +2 -1
-                            tx_start_bd->vlan_or_ethertype = eh->ether_type;
+                            tx_start_bd->vlan_or_ethertype
+                                = rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));

Minor question - any specific reason to use rte_be_to_cpu_16() on ether_type alone before converting from native order to le16?

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

ether_type is in network byte order (big endian)
and hardware wants little endian. On x86 the second step is a nop.

Doesn't it deserve a macro rte_ntole16()?
It may be in lib/librte_eal/common/include/generic/rte_byteorder.h.

I looked I didn't see anything. This value, according to the linux driver, wants to be little endian regardless of the host endian.

Yes, that's why I suggest to create some macros to do this kind of conversion.
Example: rte_ntole16 means "network to little endian 16-bit".
Do you think it would be clearer to use?

#8 lore raw
Quoted 20 lines · Thomas Monjalon

2015-12-01 18:58, Charles Williams:

Quoted 19 lines · Thomas Monjalon
Quoted 17 lines

2015-12-01 14:37, Stephen Hemminger:

Quoted 11 lines
Quoted 9 lines · Harish Patil
Quoted 6 lines

2015-11-03 12:26, Chas Williams:

drivers/net/bnx2x/bnx2x.c

Quoted 5 lines
drivers/net/bnx2x/bnx2x.c +2 -1
-                            tx_start_bd->vlan_or_ethertype = eh->ether_type;
+                            tx_start_bd->vlan_or_ethertype
+                                = rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));

Minor question - any specific reason to use rte_be_to_cpu_16() on ether_type alone before converting from native order to le16?

Quoted 2 lines

ether_type is in network byte order (big endian)
and hardware wants little endian. On x86 the second step is a nop.

Doesn't it deserve a macro rte_ntole16()?
It may be in lib/librte_eal/common/include/generic/rte_byteorder.h.

Quoted 6 lines
Quoted 2 lines

I looked I didn't see anything. This value, according to the linux driver, wants to be little endian regardless of the host endian.

Yes, that's why I suggest to create some macros to do this kind of conversion.
Example: rte_ntole16 means "network to little endian 16-bit".
Do you think it would be clearer to use?

This is the only example of this kind of conversion in the source code so it would be a macro for one user. If you create rte_ntole16() you might feel obligated to create the various permutations for which there are no consumers.

#9 lore raw

2015-12-02 05:18, Charles Williams:

Quoted 24 lines · Thomas Monjalon
Quoted 23 lines

2015-12-01 18:58, Charles Williams:

Quoted 21 lines · Thomas Monjalon
Quoted 12 lines

2015-12-01 14:37, Stephen Hemminger:

Quoted 11 lines
Quoted 9 lines · Harish Patil
Quoted 6 lines

2015-11-03 12:26, Chas Williams:

drivers/net/bnx2x/bnx2x.c

Quoted 5 lines
drivers/net/bnx2x/bnx2x.c +2 -1
-                            tx_start_bd->vlan_or_ethertype = eh->ether_type;
+                            tx_start_bd->vlan_or_ethertype
+                                = rte_cpu_to_le_16(rte_be_to_cpu_16(eh->ether_type));

Minor question - any specific reason to use rte_be_to_cpu_16() on ether_type alone before converting from native order to le16?

Quoted 5 lines
Quoted 2 lines

ether_type is in network byte order (big endian)
and hardware wants little endian. On x86 the second step is a nop.

Doesn't it deserve a macro rte_ntole16()?
It may be in lib/librte_eal/common/include/generic/rte_byteorder.h.

I looked I didn't see anything. This value, according to the linux driver, wants to be little endian regardless of the host endian.

Quoted 8 lines
Quoted 3 lines

Yes, that's why I suggest to create some macros to do this kind of conversion.
Example: rte_ntole16 means "network to little endian 16-bit".
Do you think it would be clearer to use?

This is the only example of this kind of conversion in the source code so it would be a macro for one user. If you create rte_ntole16() you might feel obligated to create the various permutations for which there are no consumers.

Yes, that's why I was not sure of the interest.