* [PATCH 00/15] net/ena: driver release 2.10.0
@ 2024-07-02 14:46 shaibran
2024-07-02 14:46 ` [PATCH 01/15] net/ena/base: add descriptor dump capability shaibran
` (15 more replies)
0 siblings, 16 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
Hi all,
This release contains:
1. HAL upgrade to latest divided into separate patches.
2. three bug fixes.
3. restructuring of the Rx checksum code for readability
4. restructuring of the device uninit flow which is also needed
for adding hot un/plug support later.
5. modification of logs on the datapath that leads to improved performance.
Will appreciate your feedbacks!
All the best and thanks in advance,
Shai
Shai Brandes (15):
net/ena/base: add descriptor dump capability
net/ena/base: remove unused param
net/ena/base: remove redundant assert checks
net/ena/base: update memory barrier comment
net/ena/base: add method to check used entries
net/ena/base: add an additional reset reason
net/ena/base: update copyrights comments
net/ena/base: add macro for bitfield access
net/ena: logger change to improve performance
net/ena: rework device uninit
net/ena: fix bad checksum handling
net/ena: fix invalid return value check
net/ena: fix wrong handling of checksum
net/ena: rework Rx checksum inspection
net/ena: upgrade driver version to 2.10.0
doc/guides/rel_notes/release_24_07.rst | 11 +
drivers/net/ena/base/ena_com.c | 167 +++++++------
drivers/net/ena/base/ena_com.h | 14 +-
.../net/ena/base/ena_defs/ena_admin_defs.h | 5 +-
.../net/ena/base/ena_defs/ena_common_defs.h | 4 +-
.../net/ena/base/ena_defs/ena_eth_io_defs.h | 5 +-
drivers/net/ena/base/ena_defs/ena_includes.h | 4 +-
drivers/net/ena/base/ena_defs/ena_regs_defs.h | 5 +-
drivers/net/ena/base/ena_eth_com.c | 231 +++++++++++-------
drivers/net/ena/base/ena_eth_com.h | 33 ++-
drivers/net/ena/base/ena_plat.h | 4 +-
drivers/net/ena/base/ena_plat_dpdk.h | 24 +-
drivers/net/ena/ena_ethdev.c | 134 +++++-----
13 files changed, 374 insertions(+), 267 deletions(-)
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 01/15] net/ena/base: add descriptor dump capability
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 02/15] net/ena/base: remove unused param shaibran
` (14 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
This patch adds the capability to print rx/tx descriptors.
This patch introduces a new function ena_com_tx_cdesc_idx_to_ptr
which is the equivalent of ena_com_rx_cdesc_idx_to_ptr but for tx cdesc.
Finally, this patch moves the io_cq header incrementation in
ena_com_cdesc_rx_pkt_get() function to be after the possible if fail
cases since it makes more sense to point into the next descriptor only
if the last one is valid.
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
drivers/net/ena/base/ena_eth_com.c | 56 +++++++++++++++++++++++++++---
drivers/net/ena/base/ena_eth_com.h | 8 +++++
2 files changed, 60 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ena/base/ena_eth_com.c b/drivers/net/ena/base/ena_eth_com.c
index 0de736fbe0..9ba0beb868 100644
--- a/drivers/net/ena/base/ena_eth_com.c
+++ b/drivers/net/ena/base/ena_eth_com.c
@@ -5,8 +5,7 @@
#include "ena_eth_com.h"
-static struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc(
- struct ena_com_io_cq *io_cq)
+struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc(struct ena_com_io_cq *io_cq)
{
struct ena_eth_io_rx_cdesc_base *cdesc;
u16 expected_phase, head_masked;
@@ -32,6 +31,55 @@ static struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc(
return cdesc;
}
+void ena_com_dump_single_rx_cdesc(struct ena_com_io_cq *io_cq,
+ struct ena_eth_io_rx_cdesc_base *desc)
+{
+ if (desc) {
+ uint32_t *desc_arr = (uint32_t *)desc;
+
+ ena_trc_err(ena_com_io_cq_to_ena_dev(io_cq),
+ "RX descriptor value[0x%08x 0x%08x 0x%08x 0x%08x] phase[%u] first[%u] last[%u] MBZ7[%u] MZB17[%u]\n",
+ desc_arr[0], desc_arr[1], desc_arr[2], desc_arr[3],
+ ENA_FIELD_GET(desc->status, (uint32_t)ENA_ETH_IO_RX_DESC_PHASE_MASK,
+ 0),
+ ENA_FIELD_GET(desc->status, (uint32_t)ENA_ETH_IO_RX_DESC_FIRST_MASK,
+ ENA_ETH_IO_RX_DESC_FIRST_SHIFT),
+ ENA_FIELD_GET(desc->status, (uint32_t)ENA_ETH_IO_RX_DESC_LAST_MASK,
+ ENA_ETH_IO_RX_DESC_LAST_SHIFT),
+ ENA_FIELD_GET(desc->status,
+ (uint32_t)ENA_ETH_IO_RX_CDESC_BASE_MBZ7_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_MBZ7_SHIFT),
+ ENA_FIELD_GET(desc->status,
+ (uint32_t)ENA_ETH_IO_RX_CDESC_BASE_MBZ17_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_MBZ17_SHIFT));
+ }
+}
+
+void ena_com_dump_single_tx_cdesc(struct ena_com_io_cq *io_cq,
+ struct ena_eth_io_tx_cdesc *desc)
+{
+ if (desc) {
+ uint32_t *desc_arr = (uint32_t *)desc;
+
+ ena_trc_err(ena_com_io_cq_to_ena_dev(io_cq),
+ "TX descriptor value[0x%08x 0x%08x] phase[%u] MBZ6[%u]\n",
+ desc_arr[0], desc_arr[1],
+ ENA_FIELD_GET(desc->flags, (uint32_t)ENA_ETH_IO_TX_CDESC_PHASE_MASK,
+ 0),
+ ENA_FIELD_GET(desc->flags, (uint32_t)ENA_ETH_IO_TX_CDESC_MBZ6_MASK,
+ ENA_ETH_IO_TX_CDESC_MBZ6_SHIFT));
+ }
+}
+
+struct ena_eth_io_tx_cdesc *ena_com_tx_cdesc_idx_to_ptr(struct ena_com_io_cq *io_cq, u16 idx)
+{
+ idx &= (io_cq->q_depth - 1);
+
+ return (struct ena_eth_io_tx_cdesc *)
+ ((uintptr_t)io_cq->cdesc_addr.virt_addr +
+ idx * io_cq->cdesc_entry_size_in_bytes);
+}
+
static void *get_sq_desc_regular_queue(struct ena_com_io_sq *io_sq)
{
u16 tail_masked;
@@ -228,7 +276,7 @@ static int ena_com_sq_update_tail(struct ena_com_io_sq *io_sq)
return ena_com_sq_update_reqular_queue_tail(io_sq);
}
-static struct ena_eth_io_rx_cdesc_base *
+struct ena_eth_io_rx_cdesc_base *
ena_com_rx_cdesc_idx_to_ptr(struct ena_com_io_cq *io_cq, u16 idx)
{
idx &= (io_cq->q_depth - 1);
@@ -254,7 +302,6 @@ static int ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq,
break;
status = READ_ONCE32(cdesc->status);
- ena_com_cq_inc_head(io_cq);
if (unlikely((status & ENA_ETH_IO_RX_CDESC_BASE_FIRST_MASK) >>
ENA_ETH_IO_RX_CDESC_BASE_FIRST_SHIFT && count != 0)) {
ena_trc_err(dev,
@@ -272,6 +319,7 @@ static int ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq,
return ENA_COM_FAULT;
}
+ ena_com_cq_inc_head(io_cq);
count++;
last = (status & ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK) >>
ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT;
diff --git a/drivers/net/ena/base/ena_eth_com.h b/drivers/net/ena/base/ena_eth_com.h
index 2fac10e678..4e3d0fb6fd 100644
--- a/drivers/net/ena/base/ena_eth_com.h
+++ b/drivers/net/ena/base/ena_eth_com.h
@@ -16,6 +16,14 @@ extern "C" {
#define ENA_LLQ_HEADER (128UL - ENA_LLQ_ENTRY_DESC_CHUNK_SIZE)
#define ENA_LLQ_LARGE_HEADER (256UL - ENA_LLQ_ENTRY_DESC_CHUNK_SIZE)
+void ena_com_dump_single_rx_cdesc(struct ena_com_io_cq *io_cq,
+ struct ena_eth_io_rx_cdesc_base *desc);
+void ena_com_dump_single_tx_cdesc(struct ena_com_io_cq *io_cq,
+ struct ena_eth_io_tx_cdesc *desc);
+struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc(struct ena_com_io_cq *io_cq);
+struct ena_eth_io_rx_cdesc_base *ena_com_rx_cdesc_idx_to_ptr(struct ena_com_io_cq *io_cq, u16 idx);
+struct ena_eth_io_tx_cdesc *ena_com_tx_cdesc_idx_to_ptr(struct ena_com_io_cq *io_cq, u16 idx);
+
struct ena_com_tx_ctx {
struct ena_com_tx_meta ena_meta;
struct ena_com_buf *ena_bufs;
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 02/15] net/ena/base: remove unused param
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
2024-07-02 14:46 ` [PATCH 01/15] net/ena/base: add descriptor dump capability shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 03/15] net/ena/base: remove redundant assert checks shaibran
` (13 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
Remove an unused dev_node parameter when allocating DMA memory.
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
drivers/net/ena/base/ena_com.c | 9 ++-------
drivers/net/ena/base/ena_plat_dpdk.h | 11 +++++------
2 files changed, 7 insertions(+), 13 deletions(-)
diff --git a/drivers/net/ena/base/ena_com.c b/drivers/net/ena/base/ena_com.c
index 24756e5e76..9e1fa40c0c 100644
--- a/drivers/net/ena/base/ena_com.c
+++ b/drivers/net/ena/base/ena_com.c
@@ -328,7 +328,6 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev,
struct ena_com_io_sq *io_sq)
{
size_t size;
- int dev_node = 0;
memset(&io_sq->desc_addr, 0x0, sizeof(io_sq->desc_addr));
@@ -347,8 +346,7 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev,
io_sq->desc_addr.virt_addr,
io_sq->desc_addr.phys_addr,
io_sq->desc_addr.mem_handle,
- ctx->numa_node,
- dev_node);
+ ctx->numa_node);
if (!io_sq->desc_addr.virt_addr) {
ENA_MEM_ALLOC_COHERENT(ena_dev->dmadev,
size,
@@ -377,8 +375,7 @@ static int ena_com_init_io_sq(struct ena_com_dev *ena_dev,
ENA_MEM_ALLOC_NODE(ena_dev->dmadev,
size,
io_sq->bounce_buf_ctrl.base_buffer,
- ctx->numa_node,
- dev_node);
+ ctx->numa_node);
if (!io_sq->bounce_buf_ctrl.base_buffer)
io_sq->bounce_buf_ctrl.base_buffer = ENA_MEM_ALLOC(ena_dev->dmadev, size);
@@ -417,7 +414,6 @@ static int ena_com_init_io_cq(struct ena_com_dev *ena_dev,
struct ena_com_io_cq *io_cq)
{
size_t size;
- int prev_node = 0;
memset(&io_cq->cdesc_addr, 0x0, sizeof(io_cq->cdesc_addr));
@@ -436,7 +432,6 @@ static int ena_com_init_io_cq(struct ena_com_dev *ena_dev,
io_cq->cdesc_addr.phys_addr,
io_cq->cdesc_addr.mem_handle,
ctx->numa_node,
- prev_node,
ENA_CDESC_RING_SIZE_ALIGNMENT);
if (!io_cq->cdesc_addr.virt_addr) {
ENA_MEM_ALLOC_COHERENT_ALIGNED(ena_dev->dmadev,
diff --git a/drivers/net/ena/base/ena_plat_dpdk.h b/drivers/net/ena/base/ena_plat_dpdk.h
index dffe60705d..eaa509bc2a 100644
--- a/drivers/net/ena/base/ena_plat_dpdk.h
+++ b/drivers/net/ena/base/ena_plat_dpdk.h
@@ -240,23 +240,22 @@ ena_mem_alloc_coherent(struct rte_eth_dev_data *data, size_t size,
rte_memzone_free(mem_handle); })
#define ENA_MEM_ALLOC_COHERENT_NODE_ALIGNED( \
- dmadev, size, virt, phys, mem_handle, node, dev_node, alignment) \
+ dmadev, size, virt, phys, mem_handle, node, alignment) \
do { \
void *virt_addr; \
dma_addr_t phys_addr; \
- ENA_TOUCH(dev_node); \
(mem_handle) = ena_mem_alloc_coherent((dmadev), (size), \
(node), (alignment), &virt_addr, &phys_addr); \
(virt) = virt_addr; \
(phys) = phys_addr; \
} while (0)
#define ENA_MEM_ALLOC_COHERENT_NODE( \
- dmadev, size, virt, phys, mem_handle, node, dev_node) \
+ dmadev, size, virt, phys, mem_handle, node) \
ENA_MEM_ALLOC_COHERENT_NODE_ALIGNED(dmadev, size, virt, phys, \
- mem_handle, node, dev_node, RTE_CACHE_LINE_SIZE)
-#define ENA_MEM_ALLOC_NODE(dmadev, size, virt, node, dev_node) \
+ mem_handle, node, RTE_CACHE_LINE_SIZE)
+#define ENA_MEM_ALLOC_NODE(dmadev, size, virt, node) \
do { \
- ENA_TOUCH(dmadev); ENA_TOUCH(dev_node); \
+ ENA_TOUCH(dmadev); \
virt = rte_zmalloc_socket(NULL, size, 0, node); \
} while (0)
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 03/15] net/ena/base: remove redundant assert checks
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
2024-07-02 14:46 ` [PATCH 01/15] net/ena/base: add descriptor dump capability shaibran
2024-07-02 14:46 ` [PATCH 02/15] net/ena/base: remove unused param shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 04/15] net/ena/base: update memory barrier comment shaibran
` (12 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
Remove ENA_WARN checks from ena_com_wait_and_process_admin_cq_polling
since once the execution flow reaches the check, it must be
ENA_CMD_COMPLETED because it can't be either of the other options:
1. ENA_CMD_ABORTED - in such case it will perform "goto err" in the "if"
block above, thus skipping the ENA_WARN check.
2. ENA_CMD_SUBMITTED - in such case it will timeout inside the while(1)
loop above and perform "goto err", thus skipping the ENA_WARN check.
Remove ENA_WARN check from ena_com_wait_and_process_admin_cq_interrupts
since once the execution flow reaches the check, it must be
ENA_CMD_COMPLETED because it can't be either of the other options:
1. ENA_CMD_ABORTED - same as above, i will perform "goto err" in the
"if" block above, thus skipping the ENA_WARN check.
2. ENA_CMD_SUBMITTED - in such case it will perform "goto err" in the
nested if block above, since "admin_queue->polling" is false (because
of the interrupt mode execution of admin commands)
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
drivers/net/ena/base/ena_com.c | 8 --------
1 file changed, 8 deletions(-)
diff --git a/drivers/net/ena/base/ena_com.c b/drivers/net/ena/base/ena_com.c
index 9e1fa40c0c..f9dd086484 100644
--- a/drivers/net/ena/base/ena_com.c
+++ b/drivers/net/ena/base/ena_com.c
@@ -598,10 +598,6 @@ static int ena_com_wait_and_process_admin_cq_polling(struct ena_comp_ctx *comp_c
goto err;
}
- ENA_WARN(comp_ctx->status != ENA_CMD_COMPLETED,
- admin_queue->ena_dev, "Invalid comp status %d\n",
- comp_ctx->status);
-
ret = ena_com_comp_status_to_errno(admin_queue, comp_ctx->comp_status);
err:
comp_ctxt_release(admin_queue, comp_ctx);
@@ -828,10 +824,6 @@ static int ena_com_wait_and_process_admin_cq_interrupts(struct ena_comp_ctx *com
goto err;
}
- ENA_WARN(comp_ctx->status != ENA_CMD_COMPLETED,
- admin_queue->ena_dev, "Invalid comp status %d\n",
- comp_ctx->status);
-
ret = ena_com_comp_status_to_errno(admin_queue, comp_ctx->comp_status);
err:
comp_ctxt_release(admin_queue, comp_ctx);
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 04/15] net/ena/base: update memory barrier comment
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (2 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 03/15] net/ena/base: remove redundant assert checks shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 05/15] net/ena/base: add method to check used entries shaibran
` (11 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
Update the comment above the phase bit descriptor
read in AENQ processing.
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
drivers/net/ena/base/ena_com.c | 16 ++++++++++++----
1 file changed, 12 insertions(+), 4 deletions(-)
diff --git a/drivers/net/ena/base/ena_com.c b/drivers/net/ena/base/ena_com.c
index f9dd086484..ad4f3f9431 100644
--- a/drivers/net/ena/base/ena_com.c
+++ b/drivers/net/ena/base/ena_com.c
@@ -2409,8 +2409,12 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *ena_dev, void *data)
/* Go over all the events */
while ((READ_ONCE8(aenq_common->flags) &
ENA_ADMIN_AENQ_COMMON_DESC_PHASE_MASK) == phase) {
- /* Make sure the device finished writing the rest of the descriptor
- * before reading it.
+ /* When the phase bit of the AENQ descriptor aligns with the driver's phase bit,
+ * it signifies the readiness of the entire AENQ descriptor.
+ * The driver should proceed to read the descriptor's data only after confirming
+ * and synchronizing the phase bit.
+ * This memory fence guarantees the correct sequence of accesses to the
+ * descriptor's memory.
*/
dma_rmb();
@@ -2468,8 +2472,12 @@ bool ena_com_aenq_has_keep_alive(struct ena_com_dev *ena_dev)
/* Go over all the events */
while ((READ_ONCE8(aenq_common->flags) &
ENA_ADMIN_AENQ_COMMON_DESC_PHASE_MASK) == phase) {
- /* Make sure the device finished writing the rest of the descriptor
- * before reading it.
+ /* When the phase bit of the AENQ descriptor aligns with the driver's phase bit,
+ * it signifies the readiness of the entire AENQ descriptor.
+ * The driver should proceed to read the descriptor's data only after confirming
+ * and synchronizing the phase bit.
+ * This memory fence guarantees the correct sequence of accesses to the
+ * descriptor's memory.
*/
dma_rmb();
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 05/15] net/ena/base: add method to check used entries
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (3 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 04/15] net/ena/base: update memory barrier comment shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 06/15] net/ena/base: add an additional reset reason shaibran
` (10 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
Provide a method to check the number of used entries in the send queue
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
drivers/net/ena/base/ena_eth_com.h | 13 ++++++-------
1 file changed, 6 insertions(+), 7 deletions(-)
diff --git a/drivers/net/ena/base/ena_eth_com.h b/drivers/net/ena/base/ena_eth_com.h
index 4e3d0fb6fd..877668612c 100644
--- a/drivers/net/ena/base/ena_eth_com.h
+++ b/drivers/net/ena/base/ena_eth_com.h
@@ -82,15 +82,14 @@ static inline void ena_com_unmask_intr(struct ena_com_io_cq *io_cq,
ENA_REG_WRITE32(io_cq->bus, intr_reg->intr_control, io_cq->unmask_reg);
}
-static inline int ena_com_free_q_entries(struct ena_com_io_sq *io_sq)
+static inline u16 ena_com_used_q_entries(struct ena_com_io_sq *io_sq)
{
- u16 tail, next_to_comp, cnt;
-
- next_to_comp = io_sq->next_to_comp;
- tail = io_sq->tail;
- cnt = tail - next_to_comp;
+ return io_sq->tail - io_sq->next_to_comp;
+}
- return io_sq->q_depth - 1 - cnt;
+static inline int ena_com_free_q_entries(struct ena_com_io_sq *io_sq)
+{
+ return io_sq->q_depth - 1 - ena_com_used_q_entries(io_sq);
}
/* Check if the submission queue has enough space to hold required_buffers */
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 06/15] net/ena/base: add an additional reset reason
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (4 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 05/15] net/ena/base: add method to check used entries shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 07/15] net/ena/base: update copyrights comments shaibran
` (9 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
This commit adds the support for a new reset reason
for `MISS_FIRST_INTERRUPT` in order to distinguish
between resets where no interrupts have been received
and sporadic missed interrupts.
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
drivers/net/ena/base/ena_defs/ena_regs_defs.h | 1 +
1 file changed, 1 insertion(+)
diff --git a/drivers/net/ena/base/ena_defs/ena_regs_defs.h b/drivers/net/ena/base/ena_defs/ena_regs_defs.h
index dd9b629f10..e12a578fac 100644
--- a/drivers/net/ena/base/ena_defs/ena_regs_defs.h
+++ b/drivers/net/ena/base/ena_defs/ena_regs_defs.h
@@ -26,6 +26,7 @@ enum ena_regs_reset_reason_types {
ENA_REGS_RESET_TX_DESCRIPTOR_MALFORMED = 17,
ENA_REGS_RESET_MISSING_ADMIN_INTERRUPT = 18,
ENA_REGS_RESET_DEVICE_REQUEST = 19,
+ ENA_REGS_RESET_MISS_FIRST_INTERRUPT = 20,
ENA_REGS_RESET_LAST,
};
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 07/15] net/ena/base: update copyrights comments
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (5 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 06/15] net/ena/base: add an additional reset reason shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-07 18:57 ` Ferruh Yigit
2024-07-08 11:45 ` Ferruh Yigit
2024-07-02 14:46 ` [PATCH 08/15] net/ena/base: add macro for bitfield access shaibran
` (8 subsequent siblings)
15 siblings, 2 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
copyright dates are not mandatory to be maintained,
therefore the range of years was removed.
In addition, the copyrights lines were separated
into two comments.
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
drivers/net/ena/base/ena_com.c | 4 ++--
drivers/net/ena/base/ena_com.h | 4 ++--
drivers/net/ena/base/ena_defs/ena_admin_defs.h | 5 +++--
drivers/net/ena/base/ena_defs/ena_common_defs.h | 4 ++--
drivers/net/ena/base/ena_defs/ena_eth_io_defs.h | 5 +++--
drivers/net/ena/base/ena_defs/ena_includes.h | 4 ++--
drivers/net/ena/base/ena_defs/ena_regs_defs.h | 4 ++--
drivers/net/ena/base/ena_eth_com.c | 4 ++--
drivers/net/ena/base/ena_eth_com.h | 4 ++--
drivers/net/ena/base/ena_plat.h | 4 ++--
drivers/net/ena/base/ena_plat_dpdk.h | 4 ++--
11 files changed, 24 insertions(+), 22 deletions(-)
diff --git a/drivers/net/ena/base/ena_com.c b/drivers/net/ena/base/ena_com.c
index ad4f3f9431..5f46e692b3 100644
--- a/drivers/net/ena/base/ena_com.c
+++ b/drivers/net/ena/base/ena_com.c
@@ -1,5 +1,5 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
diff --git a/drivers/net/ena/base/ena_com.h b/drivers/net/ena/base/ena_com.h
index 737747f64b..5d38b69c0f 100644
--- a/drivers/net/ena/base/ena_com.h
+++ b/drivers/net/ena/base/ena_com.h
@@ -1,5 +1,5 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
diff --git a/drivers/net/ena/base/ena_defs/ena_admin_defs.h b/drivers/net/ena/base/ena_defs/ena_admin_defs.h
index cff6451c96..8a1bb0bb76 100644
--- a/drivers/net/ena/base/ena_defs/ena_admin_defs.h
+++ b/drivers/net/ena/base/ena_defs/ena_admin_defs.h
@@ -1,7 +1,8 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
+
#ifndef _ENA_ADMIN_H_
#define _ENA_ADMIN_H_
diff --git a/drivers/net/ena/base/ena_defs/ena_common_defs.h b/drivers/net/ena/base/ena_defs/ena_common_defs.h
index d1ee40de32..ed5359cb99 100644
--- a/drivers/net/ena/base/ena_defs/ena_common_defs.h
+++ b/drivers/net/ena/base/ena_defs/ena_common_defs.h
@@ -1,5 +1,5 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
diff --git a/drivers/net/ena/base/ena_defs/ena_eth_io_defs.h b/drivers/net/ena/base/ena_defs/ena_eth_io_defs.h
index f811dd261e..c93cd85632 100644
--- a/drivers/net/ena/base/ena_defs/ena_eth_io_defs.h
+++ b/drivers/net/ena/base/ena_defs/ena_eth_io_defs.h
@@ -1,7 +1,8 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
+
#ifndef _ENA_ETH_IO_H_
#define _ENA_ETH_IO_H_
diff --git a/drivers/net/ena/base/ena_defs/ena_includes.h b/drivers/net/ena/base/ena_defs/ena_includes.h
index 20dba04d52..4bcb1e3a4a 100644
--- a/drivers/net/ena/base/ena_defs/ena_includes.h
+++ b/drivers/net/ena/base/ena_defs/ena_includes.h
@@ -1,5 +1,5 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2019 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
diff --git a/drivers/net/ena/base/ena_defs/ena_regs_defs.h b/drivers/net/ena/base/ena_defs/ena_regs_defs.h
index e12a578fac..823dccd841 100644
--- a/drivers/net/ena/base/ena_defs/ena_regs_defs.h
+++ b/drivers/net/ena/base/ena_defs/ena_regs_defs.h
@@ -1,5 +1,5 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
#ifndef _ENA_REGS_H_
diff --git a/drivers/net/ena/base/ena_eth_com.c b/drivers/net/ena/base/ena_eth_com.c
index 9ba0beb868..e26678827c 100644
--- a/drivers/net/ena/base/ena_eth_com.c
+++ b/drivers/net/ena/base/ena_eth_com.c
@@ -1,5 +1,5 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
diff --git a/drivers/net/ena/base/ena_eth_com.h b/drivers/net/ena/base/ena_eth_com.h
index 877668612c..f91cf67c09 100644
--- a/drivers/net/ena/base/ena_eth_com.h
+++ b/drivers/net/ena/base/ena_eth_com.h
@@ -1,5 +1,5 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
diff --git a/drivers/net/ena/base/ena_plat.h b/drivers/net/ena/base/ena_plat.h
index abd2d8b6c4..52104f3029 100644
--- a/drivers/net/ena/base/ena_plat.h
+++ b/drivers/net/ena/base/ena_plat.h
@@ -1,5 +1,5 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2019 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
diff --git a/drivers/net/ena/base/ena_plat_dpdk.h b/drivers/net/ena/base/ena_plat_dpdk.h
index eaa509bc2a..87f7083ce9 100644
--- a/drivers/net/ena/base/ena_plat_dpdk.h
+++ b/drivers/net/ena/base/ena_plat_dpdk.h
@@ -1,5 +1,5 @@
-/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
+/* SPDX-License-Identifier: BSD-3-Clause */
+/* Copyright (c) Amazon.com, Inc. or its affiliates.
* All rights reserved.
*/
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 08/15] net/ena/base: add macro for bitfield access
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (6 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 07/15] net/ena/base: update copyrights comments shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 09/15] net/ena: logger change to improve performance shaibran
` (7 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
Add ENA_FIELD_GET and ENA_FIELD_PREP macro to make access to
fields more readable.
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
drivers/net/ena/base/ena_com.c | 111 ++++++++++++--------
drivers/net/ena/base/ena_com.h | 10 +-
drivers/net/ena/base/ena_eth_com.c | 146 +++++++++++++++------------
drivers/net/ena/base/ena_eth_com.h | 8 +-
drivers/net/ena/base/ena_plat_dpdk.h | 3 +
5 files changed, 165 insertions(+), 113 deletions(-)
diff --git a/drivers/net/ena/base/ena_com.c b/drivers/net/ena/base/ena_com.c
index 5f46e692b3..e5f1a31c9e 100644
--- a/drivers/net/ena/base/ena_com.c
+++ b/drivers/net/ena/base/ena_com.c
@@ -162,10 +162,13 @@ static int ena_com_admin_init_aenq(struct ena_com_dev *ena_dev,
ENA_REG_WRITE32(ena_dev->bus, addr_high, ena_dev->reg_bar + ENA_REGS_AENQ_BASE_HI_OFF);
aenq_caps = 0;
- aenq_caps |= ena_dev->aenq.q_depth & ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK;
- aenq_caps |= (sizeof(struct ena_admin_aenq_entry) <<
- ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_SHIFT) &
- ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_MASK;
+ aenq_caps |= ENA_FIELD_PREP(ena_dev->aenq.q_depth,
+ ENA_REGS_AENQ_CAPS_AENQ_DEPTH_MASK,
+ ENA_ZERO_SHIFT);
+
+ aenq_caps |= ENA_FIELD_PREP(sizeof(struct ena_admin_aenq_entry),
+ ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_MASK,
+ ENA_REGS_AENQ_CAPS_AENQ_ENTRY_SIZE_SHIFT);
ENA_REG_WRITE32(ena_dev->bus, aenq_caps, ena_dev->reg_bar + ENA_REGS_AENQ_CAPS_OFF);
if (unlikely(!aenq_handlers)) {
@@ -856,8 +859,9 @@ static u32 ena_com_reg_bar_read32(struct ena_com_dev *ena_dev, u16 offset)
mmio_read->seq_num++;
read_resp->req_id = mmio_read->seq_num + 0xDEAD;
- mmio_read_reg = (offset << ENA_REGS_MMIO_REG_READ_REG_OFF_SHIFT) &
- ENA_REGS_MMIO_REG_READ_REG_OFF_MASK;
+ mmio_read_reg = ENA_FIELD_PREP(offset,
+ ENA_REGS_MMIO_REG_READ_REG_OFF_MASK,
+ ENA_REGS_MMIO_REG_READ_REG_OFF_SHIFT);
mmio_read_reg |= mmio_read->seq_num &
ENA_REGS_MMIO_REG_READ_REQ_ID_MASK;
@@ -927,9 +931,10 @@ static int ena_com_destroy_io_sq(struct ena_com_dev *ena_dev,
else
direction = ENA_ADMIN_SQ_DIRECTION_RX;
- destroy_cmd.sq.sq_identity |= (direction <<
- ENA_ADMIN_SQ_SQ_DIRECTION_SHIFT) &
- ENA_ADMIN_SQ_SQ_DIRECTION_MASK;
+ destroy_cmd.sq.sq_identity |=
+ ENA_FIELD_PREP(direction,
+ ENA_ADMIN_SQ_SQ_DIRECTION_MASK,
+ ENA_ADMIN_SQ_SQ_DIRECTION_SHIFT);
destroy_cmd.sq.sq_idx = io_sq->idx;
destroy_cmd.aq_common_descriptor.opcode = ENA_ADMIN_DESTROY_SQ;
@@ -1267,16 +1272,18 @@ static int ena_com_create_io_sq(struct ena_com_dev *ena_dev,
else
direction = ENA_ADMIN_SQ_DIRECTION_RX;
- create_cmd.sq_identity |= (direction <<
- ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_SHIFT) &
- ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_MASK;
+ create_cmd.sq_identity |=
+ ENA_FIELD_PREP(direction,
+ ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_MASK,
+ ENA_ADMIN_AQ_CREATE_SQ_CMD_SQ_DIRECTION_SHIFT);
create_cmd.sq_caps_2 |= io_sq->mem_queue_type &
ENA_ADMIN_AQ_CREATE_SQ_CMD_PLACEMENT_POLICY_MASK;
- create_cmd.sq_caps_2 |= (ENA_ADMIN_COMPLETION_POLICY_DESC <<
- ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_SHIFT) &
- ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_MASK;
+ create_cmd.sq_caps_2 |=
+ ENA_FIELD_PREP(ENA_ADMIN_COMPLETION_POLICY_DESC,
+ ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_MASK,
+ ENA_ADMIN_AQ_CREATE_SQ_CMD_COMPLETION_POLICY_SHIFT);
create_cmd.sq_caps_3 |=
ENA_ADMIN_AQ_CREATE_SQ_CMD_IS_PHYSICALLY_CONTIGUOUS_MASK;
@@ -1616,8 +1623,9 @@ int ena_com_get_dma_width(struct ena_com_dev *ena_dev)
return ENA_COM_TIMER_EXPIRED;
}
- width = (caps & ENA_REGS_CAPS_DMA_ADDR_WIDTH_MASK) >>
- ENA_REGS_CAPS_DMA_ADDR_WIDTH_SHIFT;
+ width = ENA_FIELD_GET(caps,
+ ENA_REGS_CAPS_DMA_ADDR_WIDTH_MASK,
+ ENA_REGS_CAPS_DMA_ADDR_WIDTH_SHIFT);
ena_trc_dbg(ena_dev, "ENA dma width: %d\n", width);
@@ -1651,18 +1659,26 @@ int ena_com_validate_version(struct ena_com_dev *ena_dev)
}
ena_trc_info(ena_dev, "ENA device version: %d.%d\n",
- (ver & ENA_REGS_VERSION_MAJOR_VERSION_MASK) >>
- ENA_REGS_VERSION_MAJOR_VERSION_SHIFT,
- ver & ENA_REGS_VERSION_MINOR_VERSION_MASK);
+ ENA_FIELD_GET(ver,
+ ENA_REGS_VERSION_MAJOR_VERSION_MASK,
+ ENA_REGS_VERSION_MAJOR_VERSION_SHIFT),
+ ENA_FIELD_GET(ver,
+ ENA_REGS_VERSION_MINOR_VERSION_MASK,
+ ENA_ZERO_SHIFT));
ena_trc_info(ena_dev, "ENA controller version: %d.%d.%d implementation version %d\n",
- (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK)
- >> ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT,
- (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK)
- >> ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_SHIFT,
- (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_SUBMINOR_VERSION_MASK),
- (ctrl_ver & ENA_REGS_CONTROLLER_VERSION_IMPL_ID_MASK) >>
- ENA_REGS_CONTROLLER_VERSION_IMPL_ID_SHIFT);
+ ENA_FIELD_GET(ctrl_ver,
+ ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK,
+ ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_SHIFT),
+ ENA_FIELD_GET(ctrl_ver,
+ ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_MASK,
+ ENA_REGS_CONTROLLER_VERSION_MINOR_VERSION_SHIFT),
+ ENA_FIELD_GET(ctrl_ver,
+ ENA_REGS_CONTROLLER_VERSION_SUBMINOR_VERSION_MASK,
+ ENA_ZERO_SHIFT),
+ ENA_FIELD_GET(ctrl_ver,
+ ENA_REGS_CONTROLLER_VERSION_IMPL_ID_MASK,
+ ENA_REGS_CONTROLLER_VERSION_IMPL_ID_SHIFT));
ctrl_ver_masked =
(ctrl_ver & ENA_REGS_CONTROLLER_VERSION_MAJOR_VERSION_MASK) |
@@ -2117,16 +2133,20 @@ int ena_com_admin_init(struct ena_com_dev *ena_dev,
ENA_REG_WRITE32(ena_dev->bus, addr_high, ena_dev->reg_bar + ENA_REGS_ACQ_BASE_HI_OFF);
aq_caps = 0;
- aq_caps |= admin_queue->q_depth & ENA_REGS_AQ_CAPS_AQ_DEPTH_MASK;
- aq_caps |= (sizeof(struct ena_admin_aq_entry) <<
- ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_SHIFT) &
- ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_MASK;
+ aq_caps |= ENA_FIELD_PREP(admin_queue->q_depth,
+ ENA_REGS_AQ_CAPS_AQ_DEPTH_MASK,
+ ENA_ZERO_SHIFT);
+ aq_caps |= ENA_FIELD_PREP(sizeof(struct ena_admin_aq_entry),
+ ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_MASK,
+ ENA_REGS_AQ_CAPS_AQ_ENTRY_SIZE_SHIFT);
acq_caps = 0;
- acq_caps |= admin_queue->q_depth & ENA_REGS_ACQ_CAPS_ACQ_DEPTH_MASK;
- acq_caps |= (sizeof(struct ena_admin_acq_entry) <<
- ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_SHIFT) &
- ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_MASK;
+ acq_caps |= ENA_FIELD_PREP(admin_queue->q_depth,
+ ENA_REGS_ACQ_CAPS_ACQ_DEPTH_MASK,
+ ENA_ZERO_SHIFT);
+ acq_caps |= ENA_FIELD_PREP(sizeof(struct ena_admin_acq_entry),
+ ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_MASK,
+ ENA_REGS_ACQ_CAPS_ACQ_ENTRY_SIZE_SHIFT);
ENA_REG_WRITE32(ena_dev->bus, aq_caps, ena_dev->reg_bar + ENA_REGS_AQ_CAPS_OFF);
ENA_REG_WRITE32(ena_dev->bus, acq_caps, ena_dev->reg_bar + ENA_REGS_ACQ_CAPS_OFF);
@@ -2521,8 +2541,9 @@ int ena_com_dev_reset(struct ena_com_dev *ena_dev,
return ENA_COM_INVAL;
}
- timeout = (cap & ENA_REGS_CAPS_RESET_TIMEOUT_MASK) >>
- ENA_REGS_CAPS_RESET_TIMEOUT_SHIFT;
+ timeout = ENA_FIELD_GET(cap,
+ ENA_REGS_CAPS_RESET_TIMEOUT_MASK,
+ ENA_REGS_CAPS_RESET_TIMEOUT_SHIFT);
if (timeout == 0) {
ena_trc_err(ena_dev, "Invalid timeout value\n");
return ENA_COM_INVAL;
@@ -2534,10 +2555,12 @@ int ena_com_dev_reset(struct ena_com_dev *ena_dev,
/* For backward compatibility, device will interpret
* bits 24-27 as MSB, bits 28-31 as LSB
*/
- reset_reason_lsb = ENA_FIELD_GET(reset_reason, ENA_RESET_REASON_LSB_MASK,
+ reset_reason_lsb = ENA_FIELD_GET(reset_reason,
+ ENA_RESET_REASON_LSB_MASK,
ENA_RESET_REASON_LSB_OFFSET);
- reset_reason_msb = ENA_FIELD_GET(reset_reason, ENA_RESET_REASON_MSB_MASK,
+ reset_reason_msb = ENA_FIELD_GET(reset_reason,
+ ENA_RESET_REASON_MSB_MASK,
ENA_RESET_REASON_MSB_OFFSET);
reset_val |= reset_reason_lsb << ENA_REGS_DEV_CTL_RESET_REASON_SHIFT;
@@ -2549,8 +2572,9 @@ int ena_com_dev_reset(struct ena_com_dev *ena_dev,
* extended reset reason fallback to generic
*/
reset_val = ENA_REGS_DEV_CTL_DEV_RESET_MASK;
- reset_val |= (ENA_REGS_RESET_GENERIC << ENA_REGS_DEV_CTL_RESET_REASON_SHIFT) &
- ENA_REGS_DEV_CTL_RESET_REASON_MASK;
+ reset_val |= ENA_FIELD_PREP(ENA_REGS_RESET_GENERIC,
+ ENA_REGS_DEV_CTL_RESET_REASON_MASK,
+ ENA_REGS_DEV_CTL_RESET_REASON_SHIFT);
}
ENA_REG_WRITE32(ena_dev->bus, reset_val, ena_dev->reg_bar + ENA_REGS_DEV_CTL_OFF);
@@ -2572,8 +2596,9 @@ int ena_com_dev_reset(struct ena_com_dev *ena_dev,
return rc;
}
- timeout = (cap & ENA_REGS_CAPS_ADMIN_CMD_TO_MASK) >>
- ENA_REGS_CAPS_ADMIN_CMD_TO_SHIFT;
+ timeout = ENA_FIELD_GET(cap,
+ ENA_REGS_CAPS_ADMIN_CMD_TO_MASK,
+ ENA_REGS_CAPS_ADMIN_CMD_TO_SHIFT);
if (timeout)
/* the resolution of timeout reg is 100ms */
ena_dev->admin_queue.completion_timeout = timeout * 100000;
diff --git a/drivers/net/ena/base/ena_com.h b/drivers/net/ena/base/ena_com.h
index 5d38b69c0f..fbb0ea39ec 100644
--- a/drivers/net/ena/base/ena_com.h
+++ b/drivers/net/ena/base/ena_com.h
@@ -1201,15 +1201,17 @@ static inline void ena_com_update_intr_reg(struct ena_eth_io_intr_reg *intr_reg,
ENA_ETH_IO_INTR_REG_RX_INTR_DELAY_MASK;
intr_reg->intr_control |=
- (tx_delay_interval << ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_SHIFT)
- & ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_MASK;
+ ENA_FIELD_PREP(tx_delay_interval,
+ ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_MASK,
+ ENA_ETH_IO_INTR_REG_TX_INTR_DELAY_SHIFT);
if (unmask)
intr_reg->intr_control |= ENA_ETH_IO_INTR_REG_INTR_UNMASK_MASK;
intr_reg->intr_control |=
- (((u32)no_moderation_update) << ENA_ETH_IO_INTR_REG_NO_MODERATION_UPDATE_SHIFT) &
- ENA_ETH_IO_INTR_REG_NO_MODERATION_UPDATE_MASK;
+ ENA_FIELD_PREP(((u32)no_moderation_update),
+ ENA_ETH_IO_INTR_REG_NO_MODERATION_UPDATE_MASK,
+ ENA_ETH_IO_INTR_REG_NO_MODERATION_UPDATE_SHIFT);
}
static inline u8 *ena_com_get_next_bounce_buffer(struct ena_com_io_bounce_buffer_control *bounce_buf_ctrl)
diff --git a/drivers/net/ena/base/ena_eth_com.c b/drivers/net/ena/base/ena_eth_com.c
index e26678827c..29cc331b1b 100644
--- a/drivers/net/ena/base/ena_eth_com.c
+++ b/drivers/net/ena/base/ena_eth_com.c
@@ -17,8 +17,9 @@ struct ena_eth_io_rx_cdesc_base *ena_com_get_next_rx_cdesc(struct ena_com_io_cq
cdesc = (struct ena_eth_io_rx_cdesc_base *)(io_cq->cdesc_addr.virt_addr
+ (head_masked * io_cq->cdesc_entry_size_in_bytes));
- desc_phase = (READ_ONCE32(cdesc->status) & ENA_ETH_IO_RX_CDESC_BASE_PHASE_MASK) >>
- ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT;
+ desc_phase = ENA_FIELD_GET(READ_ONCE32(cdesc->status),
+ ENA_ETH_IO_RX_CDESC_BASE_PHASE_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_PHASE_SHIFT);
if (desc_phase != expected_phase)
return NULL;
@@ -302,8 +303,10 @@ static int ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq,
break;
status = READ_ONCE32(cdesc->status);
- if (unlikely((status & ENA_ETH_IO_RX_CDESC_BASE_FIRST_MASK) >>
- ENA_ETH_IO_RX_CDESC_BASE_FIRST_SHIFT && count != 0)) {
+ if (unlikely(ENA_FIELD_GET(status,
+ ENA_ETH_IO_RX_CDESC_BASE_FIRST_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_FIRST_SHIFT) &&
+ count != 0)) {
ena_trc_err(dev,
"First bit is on in descriptor #%u on q_id: %u, req_id: %u\n",
count, io_cq->qid, cdesc->req_id);
@@ -321,8 +324,9 @@ static int ena_com_cdesc_rx_pkt_get(struct ena_com_io_cq *io_cq,
ena_com_cq_inc_head(io_cq);
count++;
- last = (status & ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK) >>
- ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT;
+ last = ENA_FIELD_GET(status,
+ ENA_ETH_IO_RX_CDESC_BASE_LAST_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_LAST_SHIFT);
} while (!last);
if (last) {
@@ -361,32 +365,37 @@ static int ena_com_create_meta(struct ena_com_io_sq *io_sq,
meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_EXT_VALID_MASK;
/* bits 0-9 of the mss */
- meta_desc->word2 |= ((u32)ena_meta->mss <<
- ENA_ETH_IO_TX_META_DESC_MSS_LO_SHIFT) &
- ENA_ETH_IO_TX_META_DESC_MSS_LO_MASK;
+ meta_desc->word2 |=
+ ENA_FIELD_PREP((u32)ena_meta->mss,
+ ENA_ETH_IO_TX_META_DESC_MSS_LO_MASK,
+ ENA_ETH_IO_TX_META_DESC_MSS_LO_SHIFT);
/* bits 10-13 of the mss */
- meta_desc->len_ctrl |= ((ena_meta->mss >> 10) <<
- ENA_ETH_IO_TX_META_DESC_MSS_HI_SHIFT) &
- ENA_ETH_IO_TX_META_DESC_MSS_HI_MASK;
+ meta_desc->len_ctrl |=
+ ENA_FIELD_PREP((ena_meta->mss >> 10),
+ ENA_ETH_IO_TX_META_DESC_MSS_HI_MASK,
+ ENA_ETH_IO_TX_META_DESC_MSS_HI_SHIFT);
/* Extended meta desc */
meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_ETH_META_TYPE_MASK;
- meta_desc->len_ctrl |= ((u32)io_sq->phase <<
- ENA_ETH_IO_TX_META_DESC_PHASE_SHIFT) &
- ENA_ETH_IO_TX_META_DESC_PHASE_MASK;
+ meta_desc->len_ctrl |=
+ ENA_FIELD_PREP((u32)io_sq->phase,
+ ENA_ETH_IO_TX_META_DESC_PHASE_MASK,
+ ENA_ETH_IO_TX_META_DESC_PHASE_SHIFT);
meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_FIRST_MASK;
meta_desc->len_ctrl |= ENA_ETH_IO_TX_META_DESC_META_STORE_MASK;
meta_desc->word2 |= ena_meta->l3_hdr_len &
ENA_ETH_IO_TX_META_DESC_L3_HDR_LEN_MASK;
- meta_desc->word2 |= (ena_meta->l3_hdr_offset <<
- ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_SHIFT) &
- ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_MASK;
+ meta_desc->word2 |=
+ ENA_FIELD_PREP(ena_meta->l3_hdr_offset,
+ ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_MASK,
+ ENA_ETH_IO_TX_META_DESC_L3_HDR_OFF_SHIFT);
- meta_desc->word2 |= ((u32)ena_meta->l4_hdr_len <<
- ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_SHIFT) &
- ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_MASK;
+ meta_desc->word2 |=
+ ENA_FIELD_PREP((u32)ena_meta->l4_hdr_len,
+ ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_MASK,
+ ENA_ETH_IO_TX_META_DESC_L4_HDR_LEN_IN_WORDS_SHIFT);
return ena_com_sq_update_tail(io_sq);
}
@@ -424,21 +433,26 @@ static void ena_com_rx_set_flags(struct ena_com_io_cq *io_cq,
ena_rx_ctx->l3_proto = cdesc->status &
ENA_ETH_IO_RX_CDESC_BASE_L3_PROTO_IDX_MASK;
ena_rx_ctx->l4_proto =
- (cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_MASK) >>
- ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_SHIFT;
+ ENA_FIELD_GET(cdesc->status,
+ ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_L4_PROTO_IDX_SHIFT);
ena_rx_ctx->l3_csum_err =
- !!((cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_MASK) >>
- ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_SHIFT);
+ !!(ENA_FIELD_GET(cdesc->status,
+ ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_L3_CSUM_ERR_SHIFT));
ena_rx_ctx->l4_csum_err =
- !!((cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_MASK) >>
- ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_SHIFT);
+ !!(ENA_FIELD_GET(cdesc->status,
+ ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_ERR_SHIFT));
ena_rx_ctx->l4_csum_checked =
- !!((cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_CHECKED_MASK) >>
- ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_CHECKED_SHIFT);
+ !!(ENA_FIELD_GET(cdesc->status,
+ ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_CHECKED_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_L4_CSUM_CHECKED_SHIFT));
ena_rx_ctx->hash = cdesc->hash;
ena_rx_ctx->frag =
- (cdesc->status & ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_MASK) >>
- ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_SHIFT;
+ ENA_FIELD_GET(cdesc->status,
+ ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_MASK,
+ ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_SHIFT);
ena_trc_dbg(ena_com_io_cq_to_ena_dev(io_cq),
"l3_proto %d l4_proto %d l3_csum_err %d l4_csum_err %d hash %u frag %d cdesc_status %x\n",
@@ -523,46 +537,48 @@ int ena_com_prepare_tx(struct ena_com_io_sq *io_sq,
if (!have_meta)
desc->len_ctrl |= ENA_ETH_IO_TX_DESC_FIRST_MASK;
- desc->buff_addr_hi_hdr_sz |= ((u32)header_len <<
- ENA_ETH_IO_TX_DESC_HEADER_LENGTH_SHIFT) &
- ENA_ETH_IO_TX_DESC_HEADER_LENGTH_MASK;
- desc->len_ctrl |= ((u32)io_sq->phase << ENA_ETH_IO_TX_DESC_PHASE_SHIFT) &
- ENA_ETH_IO_TX_DESC_PHASE_MASK;
+ desc->buff_addr_hi_hdr_sz |= ENA_FIELD_PREP((u32)header_len,
+ ENA_ETH_IO_TX_DESC_HEADER_LENGTH_MASK,
+ ENA_ETH_IO_TX_DESC_HEADER_LENGTH_SHIFT);
+
+ desc->len_ctrl |= ENA_FIELD_PREP((u32)io_sq->phase,
+ ENA_ETH_IO_TX_DESC_PHASE_MASK,
+ ENA_ETH_IO_TX_DESC_PHASE_SHIFT);
desc->len_ctrl |= ENA_ETH_IO_TX_DESC_COMP_REQ_MASK;
/* Bits 0-9 */
- desc->meta_ctrl |= ((u32)ena_tx_ctx->req_id <<
- ENA_ETH_IO_TX_DESC_REQ_ID_LO_SHIFT) &
- ENA_ETH_IO_TX_DESC_REQ_ID_LO_MASK;
+ desc->meta_ctrl |= ENA_FIELD_PREP((u32)ena_tx_ctx->req_id,
+ ENA_ETH_IO_TX_DESC_REQ_ID_LO_MASK,
+ ENA_ETH_IO_TX_DESC_REQ_ID_LO_SHIFT);
- desc->meta_ctrl |= (ena_tx_ctx->df <<
- ENA_ETH_IO_TX_DESC_DF_SHIFT) &
- ENA_ETH_IO_TX_DESC_DF_MASK;
+ desc->meta_ctrl |= ENA_FIELD_PREP(ena_tx_ctx->df,
+ ENA_ETH_IO_TX_DESC_DF_MASK,
+ ENA_ETH_IO_TX_DESC_DF_SHIFT);
/* Bits 10-15 */
- desc->len_ctrl |= ((ena_tx_ctx->req_id >> 10) <<
- ENA_ETH_IO_TX_DESC_REQ_ID_HI_SHIFT) &
- ENA_ETH_IO_TX_DESC_REQ_ID_HI_MASK;
+ desc->len_ctrl |= ENA_FIELD_PREP((ena_tx_ctx->req_id >> 10),
+ ENA_ETH_IO_TX_DESC_REQ_ID_HI_MASK,
+ ENA_ETH_IO_TX_DESC_REQ_ID_HI_SHIFT);
if (ena_tx_ctx->meta_valid) {
- desc->meta_ctrl |= (ena_tx_ctx->tso_enable <<
- ENA_ETH_IO_TX_DESC_TSO_EN_SHIFT) &
- ENA_ETH_IO_TX_DESC_TSO_EN_MASK;
+ desc->meta_ctrl |= ENA_FIELD_PREP(ena_tx_ctx->tso_enable,
+ ENA_ETH_IO_TX_DESC_TSO_EN_MASK,
+ ENA_ETH_IO_TX_DESC_TSO_EN_SHIFT);
desc->meta_ctrl |= ena_tx_ctx->l3_proto &
ENA_ETH_IO_TX_DESC_L3_PROTO_IDX_MASK;
- desc->meta_ctrl |= (ena_tx_ctx->l4_proto <<
- ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_SHIFT) &
- ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_MASK;
- desc->meta_ctrl |= (ena_tx_ctx->l3_csum_enable <<
- ENA_ETH_IO_TX_DESC_L3_CSUM_EN_SHIFT) &
- ENA_ETH_IO_TX_DESC_L3_CSUM_EN_MASK;
- desc->meta_ctrl |= (ena_tx_ctx->l4_csum_enable <<
- ENA_ETH_IO_TX_DESC_L4_CSUM_EN_SHIFT) &
- ENA_ETH_IO_TX_DESC_L4_CSUM_EN_MASK;
- desc->meta_ctrl |= (ena_tx_ctx->l4_csum_partial <<
- ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_SHIFT) &
- ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_MASK;
+ desc->meta_ctrl |= ENA_FIELD_PREP(ena_tx_ctx->l4_proto,
+ ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_MASK,
+ ENA_ETH_IO_TX_DESC_L4_PROTO_IDX_SHIFT);
+ desc->meta_ctrl |= ENA_FIELD_PREP(ena_tx_ctx->l3_csum_enable,
+ ENA_ETH_IO_TX_DESC_L3_CSUM_EN_MASK,
+ ENA_ETH_IO_TX_DESC_L3_CSUM_EN_SHIFT);
+ desc->meta_ctrl |= ENA_FIELD_PREP(ena_tx_ctx->l4_csum_enable,
+ ENA_ETH_IO_TX_DESC_L4_CSUM_EN_MASK,
+ ENA_ETH_IO_TX_DESC_L4_CSUM_EN_SHIFT);
+ desc->meta_ctrl |= ENA_FIELD_PREP(ena_tx_ctx->l4_csum_partial,
+ ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_MASK,
+ ENA_ETH_IO_TX_DESC_L4_CSUM_PARTIAL_SHIFT);
}
for (i = 0; i < num_bufs; i++) {
@@ -581,9 +597,9 @@ int ena_com_prepare_tx(struct ena_com_io_sq *io_sq,
memset(desc, 0x0, sizeof(struct ena_eth_io_tx_desc));
- desc->len_ctrl |= ((u32)io_sq->phase <<
- ENA_ETH_IO_TX_DESC_PHASE_SHIFT) &
- ENA_ETH_IO_TX_DESC_PHASE_MASK;
+ desc->len_ctrl |= ENA_FIELD_PREP((u32)io_sq->phase,
+ ENA_ETH_IO_TX_DESC_PHASE_MASK,
+ ENA_ETH_IO_TX_DESC_PHASE_SHIFT);
}
desc->len_ctrl |= ena_bufs->len &
@@ -702,7 +718,9 @@ int ena_com_add_single_rx_desc(struct ena_com_io_sq *io_sq,
desc->ctrl = ENA_ETH_IO_RX_DESC_FIRST_MASK |
ENA_ETH_IO_RX_DESC_LAST_MASK |
ENA_ETH_IO_RX_DESC_COMP_REQ_MASK |
- (io_sq->phase & ENA_ETH_IO_RX_DESC_PHASE_MASK);
+ ENA_FIELD_GET(io_sq->phase,
+ ENA_ETH_IO_RX_DESC_PHASE_MASK,
+ ENA_ZERO_SHIFT);
desc->req_id = req_id;
diff --git a/drivers/net/ena/base/ena_eth_com.h b/drivers/net/ena/base/ena_eth_com.h
index f91cf67c09..8a12ed5fba 100644
--- a/drivers/net/ena/base/ena_eth_com.h
+++ b/drivers/net/ena/base/ena_eth_com.h
@@ -188,7 +188,9 @@ static inline void ena_com_update_numa_node(struct ena_com_io_cq *io_cq,
if (!io_cq->numa_node_cfg_reg)
return;
- numa_cfg.numa_cfg = (numa_node & ENA_ETH_IO_NUMA_NODE_CFG_REG_NUMA_MASK)
+ numa_cfg.numa_cfg = (ENA_FIELD_GET(numa_node,
+ ENA_ETH_IO_NUMA_NODE_CFG_REG_NUMA_MASK,
+ ENA_ZERO_SHIFT))
| ENA_ETH_IO_NUMA_NODE_CFG_REG_ENABLED_MASK;
ENA_REG_WRITE32(io_cq->bus, numa_cfg.numa_cfg, io_cq->numa_node_cfg_reg);
@@ -230,7 +232,9 @@ static inline int ena_com_tx_comp_req_id_get(struct ena_com_io_cq *io_cq,
* expected, it mean that the device still didn't update
* this completion.
*/
- cdesc_phase = flags & ENA_ETH_IO_TX_CDESC_PHASE_MASK;
+ cdesc_phase = ENA_FIELD_GET(flags,
+ ENA_ETH_IO_TX_CDESC_PHASE_MASK,
+ ENA_ZERO_SHIFT);
if (cdesc_phase != expected_phase)
return ENA_COM_TRY_AGAIN;
diff --git a/drivers/net/ena/base/ena_plat_dpdk.h b/drivers/net/ena/base/ena_plat_dpdk.h
index 87f7083ce9..03f29a2f1d 100644
--- a/drivers/net/ena/base/ena_plat_dpdk.h
+++ b/drivers/net/ena/base/ena_plat_dpdk.h
@@ -325,6 +325,9 @@ void ena_rss_key_fill(void *key, size_t size);
#define ENA_BITS_PER_U64(bitmap) (ena_bits_per_u64(bitmap))
#define ENA_FIELD_GET(value, mask, offset) (((value) & (mask)) >> (offset))
+#define ENA_FIELD_PREP(value, mask, offset) (((value) << (offset)) & (mask))
+
+#define ENA_ZERO_SHIFT 0
static __rte_always_inline int ena_bits_per_u64(uint64_t bitmap)
{
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 09/15] net/ena: logger change to improve performance
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (7 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 08/15] net/ena/base: add macro for bitfield access shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 10/15] net/ena: rework device uninit shaibran
` (6 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
Current implementation of ena_trc_dbg on every TX packet has a major
performance impact on DPDK TX flow.
Profiling revealed that these calls, which trigger rte_log usage,
consume a significant amount of CPU resources.
Change details:
1. Several warning prints that incorrectly used ena_trc_dbg will now be
compiled out. They have been changed to ena_trc_warn to avoid
compiler warnings, such as empty if/else body or unused parameter.
2. Removed variables which is used only inside prints and thus
may be unreferenced.
3. calls for ena_trc_dbg will be enabled only if
RTE_ETHDEV_DEBUG_TX or RTE_ETHDEV_DEBUG_RX defined
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
doc/guides/rel_notes/release_24_07.rst | 4 ++++
drivers/net/ena/base/ena_com.c | 19 +++++--------------
drivers/net/ena/base/ena_eth_com.c | 25 ++++++++++++-------------
drivers/net/ena/base/ena_plat_dpdk.h | 6 +++++-
drivers/net/ena/ena_ethdev.c | 2 +-
5 files changed, 27 insertions(+), 29 deletions(-)
diff --git a/doc/guides/rel_notes/release_24_07.rst b/doc/guides/rel_notes/release_24_07.rst
index e68a53d757..1e46a4b7c7 100644
--- a/doc/guides/rel_notes/release_24_07.rst
+++ b/doc/guides/rel_notes/release_24_07.rst
@@ -73,6 +73,10 @@ New Features
``bpf_obj_get()`` for an xskmap pinned (by the AF_XDP DP) inside the
container.
+* **Updated Amazon ena (Elastic Network Adapter) net driver.**
+
+ * Reworked the driver logger usage in order to improve Tx performance.
+
* **Update Tap PMD driver.**
* Updated to support up to 8 queues when used by secondary process.
diff --git a/drivers/net/ena/base/ena_com.c b/drivers/net/ena/base/ena_com.c
index e5f1a31c9e..24bad19848 100644
--- a/drivers/net/ena/base/ena_com.c
+++ b/drivers/net/ena/base/ena_com.c
@@ -1392,11 +1392,7 @@ int ena_com_execute_admin_command(struct ena_com_admin_queue *admin_queue,
comp, comp_size);
if (IS_ERR(comp_ctx)) {
ret = PTR_ERR(comp_ctx);
- if (ret == ENA_COM_NO_DEVICE)
- ena_trc_dbg(admin_queue->ena_dev,
- "Failed to submit command [%d]\n",
- ret);
- else
+ if (ret != ENA_COM_NO_DEVICE)
ena_trc_err(admin_queue->ena_dev,
"Failed to submit command [%d]\n",
ret);
@@ -1408,10 +1404,8 @@ int ena_com_execute_admin_command(struct ena_com_admin_queue *admin_queue,
if (unlikely(ret)) {
if (admin_queue->running_state)
ena_trc_err(admin_queue->ena_dev,
- "Failed to process command. ret = %d\n", ret);
- else
- ena_trc_dbg(admin_queue->ena_dev,
- "Failed to process command. ret = %d\n", ret);
+ "Failed to process command [%d]\n",
+ ret);
}
return ret;
}
@@ -2416,7 +2410,6 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *ena_dev, void *data)
struct ena_admin_aenq_entry *aenq_e;
struct ena_admin_aenq_common_desc *aenq_common;
struct ena_com_aenq *aenq = &ena_dev->aenq;
- u64 timestamp;
ena_aenq_handler handler_cb;
u16 masked_head, processed = 0;
u8 phase;
@@ -2438,13 +2431,11 @@ void ena_com_aenq_intr_handler(struct ena_com_dev *ena_dev, void *data)
*/
dma_rmb();
- timestamp = (u64)aenq_common->timestamp_low |
- ((u64)aenq_common->timestamp_high << 32);
-
ena_trc_dbg(ena_dev, "AENQ! Group[%x] Syndrome[%x] timestamp: [%" ENA_PRIu64 "s]\n",
aenq_common->group,
aenq_common->syndrome,
- timestamp);
+ ((u64)aenq_common->timestamp_low |
+ ((u64)aenq_common->timestamp_high << 32)));
/* Handle specific event*/
handler_cb = ena_com_get_specific_aenq_cb(ena_dev,
diff --git a/drivers/net/ena/base/ena_eth_com.c b/drivers/net/ena/base/ena_eth_com.c
index 29cc331b1b..90dd85c7ff 100644
--- a/drivers/net/ena/base/ena_eth_com.c
+++ b/drivers/net/ena/base/ena_eth_com.c
@@ -426,8 +426,7 @@ static int ena_com_create_and_store_tx_meta_desc(struct ena_com_io_sq *io_sq,
return ENA_COM_OK;
}
-static void ena_com_rx_set_flags(struct ena_com_io_cq *io_cq,
- struct ena_com_rx_ctx *ena_rx_ctx,
+static void ena_com_rx_set_flags(struct ena_com_rx_ctx *ena_rx_ctx,
struct ena_eth_io_rx_cdesc_base *cdesc)
{
ena_rx_ctx->l3_proto = cdesc->status &
@@ -453,16 +452,6 @@ static void ena_com_rx_set_flags(struct ena_com_io_cq *io_cq,
ENA_FIELD_GET(cdesc->status,
ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_MASK,
ENA_ETH_IO_RX_CDESC_BASE_IPV4_FRAG_SHIFT);
-
- ena_trc_dbg(ena_com_io_cq_to_ena_dev(io_cq),
- "l3_proto %d l4_proto %d l3_csum_err %d l4_csum_err %d hash %u frag %d cdesc_status %x\n",
- ena_rx_ctx->l3_proto,
- ena_rx_ctx->l4_proto,
- ena_rx_ctx->l3_csum_err,
- ena_rx_ctx->l4_csum_err,
- ena_rx_ctx->hash,
- ena_rx_ctx->frag,
- cdesc->status);
}
/*****************************************************************************/
@@ -689,7 +678,17 @@ int ena_com_rx_pkt(struct ena_com_io_cq *io_cq,
io_sq->qid, io_sq->next_to_comp);
/* Get rx flags from the last pkt */
- ena_com_rx_set_flags(io_cq, ena_rx_ctx, cdesc);
+ ena_com_rx_set_flags(ena_rx_ctx, cdesc);
+
+ ena_trc_dbg(ena_com_io_cq_to_ena_dev(io_cq),
+ "l3_proto %d l4_proto %d l3_csum_err %d l4_csum_err %d hash %d frag %d cdesc_status %x\n",
+ ena_rx_ctx->l3_proto,
+ ena_rx_ctx->l4_proto,
+ ena_rx_ctx->l3_csum_err,
+ ena_rx_ctx->l4_csum_err,
+ ena_rx_ctx->hash,
+ ena_rx_ctx->frag,
+ cdesc->status);
ena_rx_ctx->descs = nb_hw_desc;
diff --git a/drivers/net/ena/base/ena_plat_dpdk.h b/drivers/net/ena/base/ena_plat_dpdk.h
index 03f29a2f1d..21b96113c7 100644
--- a/drivers/net/ena/base/ena_plat_dpdk.h
+++ b/drivers/net/ena/base/ena_plat_dpdk.h
@@ -122,7 +122,11 @@ extern int ena_logtype_com;
"[ENA_COM: %s]" fmt, __func__, ##arg) \
)
-#define ena_trc_dbg(dev, format, arg...) ena_trc_log(dev, DEBUG, format, ##arg)
+#if (defined RTE_ETHDEV_DEBUG_TX) || (defined RTE_ETHDEV_DEBUG_RX)
+#define ena_trc_dbg(dev, format, ...) ena_trc_log(dev, DEBUG, format, ##__VA_ARGS__)
+#else
+#define ena_trc_dbg(dev, format, ...)
+#endif
#define ena_trc_info(dev, format, arg...) ena_trc_log(dev, INFO, format, ##arg)
#define ena_trc_warn(dev, format, arg...) ena_trc_log(dev, WARNING, format, ##arg)
#define ena_trc_err(dev, format, arg...) ena_trc_log(dev, ERR, format, ##arg)
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index 66fc287faf..56dbe3b9cd 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -3124,7 +3124,7 @@ static int ena_xmit_mbuf(struct ena_ring *tx_ring, struct rte_mbuf *mbuf)
*/
if (!ena_com_sq_have_enough_space(tx_ring->ena_com_io_sq,
mbuf->nb_segs + 2)) {
- PMD_DRV_LOG(DEBUG, "Not enough space in the tx queue\n");
+ PMD_TX_LOG(DEBUG, "Not enough space in the tx queue\n");
return ENA_COM_NO_MEM;
}
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 10/15] net/ena: rework device uninit
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (8 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 09/15] net/ena: logger change to improve performance shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 11/15] net/ena: fix bad checksum handling shaibran
` (5 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
Rework device uninitialization flow to ensure complete resource
cleanup, and lay the groundwork for hot-unplug support. With this
change, `ena_destroy_device()` is removed, its functionality now
incorporated into `ena_close()`.
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
doc/guides/rel_notes/release_24_07.rst | 2 +
drivers/net/ena/ena_ethdev.c | 55 ++++++++++++--------------
2 files changed, 27 insertions(+), 30 deletions(-)
diff --git a/doc/guides/rel_notes/release_24_07.rst b/doc/guides/rel_notes/release_24_07.rst
index 1e46a4b7c7..a59fb2a21f 100644
--- a/doc/guides/rel_notes/release_24_07.rst
+++ b/doc/guides/rel_notes/release_24_07.rst
@@ -76,6 +76,8 @@ New Features
* **Updated Amazon ena (Elastic Network Adapter) net driver.**
* Reworked the driver logger usage in order to improve Tx performance.
+ * Reworked the device uninitialization flow to ensure complete resource
+ cleanup and lay the groundwork for hot-unplug support.
* **Update Tap PMD driver.**
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index 56dbe3b9cd..4e7171e629 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -280,8 +280,8 @@ static int ena_infos_get(struct rte_eth_dev *dev,
static void ena_control_path_handler(void *cb_arg);
static void ena_control_path_poll_handler(void *cb_arg);
static void ena_timer_wd_callback(struct rte_timer *timer, void *arg);
-static void ena_destroy_device(struct rte_eth_dev *eth_dev);
static int eth_ena_dev_init(struct rte_eth_dev *eth_dev);
+static int eth_ena_dev_uninit(struct rte_eth_dev *eth_dev);
static int ena_xstats_get_names(struct rte_eth_dev *dev,
struct rte_eth_xstat_name *xstats_names,
unsigned int n);
@@ -880,12 +880,16 @@ static int ena_close(struct rte_eth_dev *dev)
struct rte_pci_device *pci_dev = RTE_ETH_DEV_TO_PCI(dev);
struct rte_intr_handle *intr_handle = pci_dev->intr_handle;
struct ena_adapter *adapter = dev->data->dev_private;
+ struct ena_com_dev *ena_dev = &adapter->ena_dev;
int ret = 0;
int rc;
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
return 0;
+ if (adapter->state == ENA_ADAPTER_STATE_CLOSED)
+ return 0;
+
if (adapter->state == ENA_ADAPTER_STATE_RUNNING)
ret = ena_stop(dev);
adapter->state = ENA_ADAPTER_STATE_CLOSED;
@@ -905,6 +909,19 @@ static int ena_close(struct rte_eth_dev *dev)
rte_free(adapter->drv_stats);
adapter->drv_stats = NULL;
+ ena_com_set_admin_running_state(ena_dev, false);
+
+ ena_com_rss_destroy(ena_dev);
+
+ ena_com_delete_debug_area(ena_dev);
+ ena_com_delete_host_info(ena_dev);
+
+ ena_com_abort_admin_commands(ena_dev);
+ ena_com_wait_for_abort_completion(ena_dev);
+ ena_com_admin_destroy(ena_dev);
+ ena_com_mmio_reg_read_request_destroy(ena_dev);
+ ena_com_delete_customer_metrics_buffer(ena_dev);
+
/*
* MAC is not allocated dynamically. Setting NULL should prevent from
* release of the resource in the rte_eth_dev_release_port().
@@ -925,7 +942,12 @@ ena_dev_reset(struct rte_eth_dev *dev)
return -EPERM;
}
- ena_destroy_device(dev);
+ rc = eth_ena_dev_uninit(dev);
+ if (rc) {
+ PMD_INIT_LOG(CRIT, "Failed to un-initialize device\n");
+ return rc;
+ }
+
rc = eth_ena_dev_init(dev);
if (rc)
PMD_INIT_LOG(CRIT, "Cannot initialize device\n");
@@ -2434,39 +2456,12 @@ static int eth_ena_dev_init(struct rte_eth_dev *eth_dev)
return rc;
}
-static void ena_destroy_device(struct rte_eth_dev *eth_dev)
-{
- struct ena_adapter *adapter = eth_dev->data->dev_private;
- struct ena_com_dev *ena_dev = &adapter->ena_dev;
-
- if (adapter->state == ENA_ADAPTER_STATE_FREE)
- return;
-
- ena_com_set_admin_running_state(ena_dev, false);
-
- if (adapter->state != ENA_ADAPTER_STATE_CLOSED)
- ena_close(eth_dev);
-
- ena_com_rss_destroy(ena_dev);
-
- ena_com_delete_debug_area(ena_dev);
- ena_com_delete_host_info(ena_dev);
-
- ena_com_abort_admin_commands(ena_dev);
- ena_com_wait_for_abort_completion(ena_dev);
- ena_com_admin_destroy(ena_dev);
- ena_com_mmio_reg_read_request_destroy(ena_dev);
- ena_com_delete_customer_metrics_buffer(ena_dev);
-
- adapter->state = ENA_ADAPTER_STATE_FREE;
-}
-
static int eth_ena_dev_uninit(struct rte_eth_dev *eth_dev)
{
if (rte_eal_process_type() != RTE_PROC_PRIMARY)
return 0;
- ena_destroy_device(eth_dev);
+ ena_close(eth_dev);
return 0;
}
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 11/15] net/ena: fix bad checksum handling
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (9 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 10/15] net/ena: rework device uninit shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 12/15] net/ena: fix invalid return value check shaibran
` (4 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes, stable
From: Shai Brandes <shaibran@amazon.com>
Removed a workaround for a false L4 bad Rx csum
indication from the device. The workaround was to set it
as unknown so the application would check it instead.
The issue was fixed in the device, thus the driver bad csum
handling should be fixed in the PMD.
Fixes: b2d2f1cf89a6 ("net/ena: fix checksum flag for L4")
Cc: stable@dpdk.org
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
doc/guides/rel_notes/release_24_07.rst | 1 +
drivers/net/ena/ena_ethdev.c | 8 +-------
2 files changed, 2 insertions(+), 7 deletions(-)
diff --git a/doc/guides/rel_notes/release_24_07.rst b/doc/guides/rel_notes/release_24_07.rst
index a59fb2a21f..f000dec54b 100644
--- a/doc/guides/rel_notes/release_24_07.rst
+++ b/doc/guides/rel_notes/release_24_07.rst
@@ -78,6 +78,7 @@ New Features
* Reworked the driver logger usage in order to improve Tx performance.
* Reworked the device uninitialization flow to ensure complete resource
cleanup and lay the groundwork for hot-unplug support.
+ * Removed an obsolete workaround for a false L4 bad Rx checksum indication.
* **Update Tap PMD driver.**
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index 4e7171e629..b43b913903 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -674,13 +674,7 @@ static inline void ena_rx_mbuf_prepare(struct ena_ring *rx_ring,
} else {
if (unlikely(ena_rx_ctx->l4_csum_err)) {
++rx_stats->l4_csum_bad;
- /*
- * For the L4 Rx checksum offload the HW may indicate
- * bad checksum although it's valid. Because of that,
- * we're setting the UNKNOWN flag to let the app
- * re-verify the checksum.
- */
- ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_UNKNOWN;
+ ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_BAD;
} else {
++rx_stats->l4_csum_good;
ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_GOOD;
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 12/15] net/ena: fix invalid return value check
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (10 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 11/15] net/ena: fix bad checksum handling shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-07 18:57 ` Ferruh Yigit
2024-07-02 14:46 ` [PATCH 13/15] net/ena: fix wrong handling of checksum shaibran
` (3 subsequent siblings)
15 siblings, 1 reply; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes, stable
From: Shai Brandes <shaibran@amazon.com>
Removed the sign inversion for when checking if
ena_com_set_host_attributes returns ENA_COM_UNSUPPORTED.
ENA_COM_UNSUPPORTED is defined as -EOPNOTSUPP, so the extra sign
inversion is wrong.
Fixes: 3adcba9a8987 ("net/ena: update HAL to the newer version")
Cc: stable@dpdk.org
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
doc/guides/rel_notes/release_24_07.rst | 1 +
drivers/net/ena/ena_ethdev.c | 4 ++--
2 files changed, 3 insertions(+), 2 deletions(-)
diff --git a/doc/guides/rel_notes/release_24_07.rst b/doc/guides/rel_notes/release_24_07.rst
index f000dec54b..24bb91ad46 100644
--- a/doc/guides/rel_notes/release_24_07.rst
+++ b/doc/guides/rel_notes/release_24_07.rst
@@ -79,6 +79,7 @@ New Features
* Reworked the device uninitialization flow to ensure complete resource
cleanup and lay the groundwork for hot-unplug support.
* Removed an obsolete workaround for a false L4 bad Rx checksum indication.
+ * Fixed an invalid return value check.
* **Update Tap PMD driver.**
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index b43b913903..67a1d86f9a 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -812,7 +812,7 @@ static void ena_config_host_info(struct ena_com_dev *ena_dev)
rc = ena_com_set_host_attributes(ena_dev);
if (rc) {
- if (rc == -ENA_COM_UNSUPPORTED)
+ if (rc == ENA_COM_UNSUPPORTED)
PMD_DRV_LOG(WARNING, "Cannot set host attributes\n");
else
PMD_DRV_LOG(ERR, "Cannot set host attributes\n");
@@ -856,7 +856,7 @@ static void ena_config_debug_area(struct ena_adapter *adapter)
rc = ena_com_set_host_attributes(&adapter->ena_dev);
if (rc) {
- if (rc == -ENA_COM_UNSUPPORTED)
+ if (rc == ENA_COM_UNSUPPORTED)
PMD_DRV_LOG(WARNING, "Cannot set host attributes\n");
else
PMD_DRV_LOG(ERR, "Cannot set host attributes\n");
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 13/15] net/ena: fix wrong handling of checksum
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (11 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 12/15] net/ena: fix invalid return value check shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 14/15] net/ena: rework Rx checksum inspection shaibran
` (2 subsequent siblings)
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes, stable
From: Shai Brandes <shaibran@amazon.com>
This change fixes an issue where a non tcp/udp packet can be indicated
to have an invalid csum. If the device erroneously tries to verify the
csum on a non tcp/udp packet it will result in false indication that
there is a csum error. This change make the driver ignore the
indication for csum error on such packets.
Fixes: 84daba9962b5 ("net/ena: add extra Rx checksum related xstats")
Cc: stable@dpdk.org
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
doc/guides/rel_notes/release_24_07.rst | 1 +
drivers/net/ena/ena_ethdev.c | 3 ++-
2 files changed, 3 insertions(+), 1 deletion(-)
diff --git a/doc/guides/rel_notes/release_24_07.rst b/doc/guides/rel_notes/release_24_07.rst
index 24bb91ad46..ec960d93cc 100644
--- a/doc/guides/rel_notes/release_24_07.rst
+++ b/doc/guides/rel_notes/release_24_07.rst
@@ -80,6 +80,7 @@ New Features
cleanup and lay the groundwork for hot-unplug support.
* Removed an obsolete workaround for a false L4 bad Rx checksum indication.
* Fixed an invalid return value check.
+ * Fixed Rx chcecksum inspection to check only TCP/UDP packets.
* **Update Tap PMD driver.**
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index 67a1d86f9a..a18c94df28 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -669,7 +669,8 @@ static inline void ena_rx_mbuf_prepare(struct ena_ring *rx_ring,
packet_type |= RTE_PTYPE_L3_IPV6;
}
- if (!ena_rx_ctx->l4_csum_checked || ena_rx_ctx->frag) {
+ if (!ena_rx_ctx->l4_csum_checked || ena_rx_ctx->frag ||
+ !(packet_type & (RTE_PTYPE_L4_TCP | RTE_PTYPE_L4_UDP))) {
ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_UNKNOWN;
} else {
if (unlikely(ena_rx_ctx->l4_csum_err)) {
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 14/15] net/ena: rework Rx checksum inspection
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (12 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 13/15] net/ena: fix wrong handling of checksum shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-02 14:46 ` [PATCH 15/15] net/ena: upgrade driver version to 2.10.0 shaibran
2024-07-07 19:11 ` [PATCH 00/15] net/ena: driver release 2.10.0 Ferruh Yigit
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
This restructure is a simplification of the
Rx checksum inspection logic in ena_rx_mbuf_prepare.
Its purpose is to improve readability and maintainability
by consolidating conditions.
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
doc/guides/rel_notes/release_24_07.rst | 2 +
drivers/net/ena/ena_ethdev.c | 66 +++++++++++++++-----------
2 files changed, 39 insertions(+), 29 deletions(-)
diff --git a/doc/guides/rel_notes/release_24_07.rst b/doc/guides/rel_notes/release_24_07.rst
index ec960d93cc..d2253999fa 100644
--- a/doc/guides/rel_notes/release_24_07.rst
+++ b/doc/guides/rel_notes/release_24_07.rst
@@ -81,6 +81,8 @@ New Features
* Removed an obsolete workaround for a false L4 bad Rx checksum indication.
* Fixed an invalid return value check.
* Fixed Rx chcecksum inspection to check only TCP/UDP packets.
+ * Reworked the Rx checksum inspection routine to improve
+ readability and maintainability.
* **Update Tap PMD driver.**
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index a18c94df28..feb229c5ec 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -53,8 +53,6 @@
*/
#define ENA_CLEANUP_BUF_THRESH 256
-#define ENA_PTYPE_HAS_HASH (RTE_PTYPE_L4_TCP | RTE_PTYPE_L4_UDP)
-
struct ena_stats {
char name[ETH_GSTRING_LEN];
int stat_offset;
@@ -645,19 +643,14 @@ static inline void ena_trigger_reset(struct ena_adapter *adapter,
static inline void ena_rx_mbuf_prepare(struct ena_ring *rx_ring,
struct rte_mbuf *mbuf,
- struct ena_com_rx_ctx *ena_rx_ctx,
- bool fill_hash)
+ struct ena_com_rx_ctx *ena_rx_ctx)
{
struct ena_stats_rx *rx_stats = &rx_ring->rx_stats;
uint64_t ol_flags = 0;
uint32_t packet_type = 0;
- if (ena_rx_ctx->l4_proto == ENA_ETH_IO_L4_PROTO_TCP)
- packet_type |= RTE_PTYPE_L4_TCP;
- else if (ena_rx_ctx->l4_proto == ENA_ETH_IO_L4_PROTO_UDP)
- packet_type |= RTE_PTYPE_L4_UDP;
-
- if (ena_rx_ctx->l3_proto == ENA_ETH_IO_L3_PROTO_IPV4) {
+ switch (ena_rx_ctx->l3_proto) {
+ case ENA_ETH_IO_L3_PROTO_IPV4:
packet_type |= RTE_PTYPE_L3_IPV4;
if (unlikely(ena_rx_ctx->l3_csum_err)) {
++rx_stats->l3_csum_bad;
@@ -665,27 +658,45 @@ static inline void ena_rx_mbuf_prepare(struct ena_ring *rx_ring,
} else {
ol_flags |= RTE_MBUF_F_RX_IP_CKSUM_GOOD;
}
- } else if (ena_rx_ctx->l3_proto == ENA_ETH_IO_L3_PROTO_IPV6) {
+ break;
+ case ENA_ETH_IO_L3_PROTO_IPV6:
packet_type |= RTE_PTYPE_L3_IPV6;
+ break;
+ default:
+ break;
}
- if (!ena_rx_ctx->l4_csum_checked || ena_rx_ctx->frag ||
- !(packet_type & (RTE_PTYPE_L4_TCP | RTE_PTYPE_L4_UDP))) {
- ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_UNKNOWN;
- } else {
- if (unlikely(ena_rx_ctx->l4_csum_err)) {
- ++rx_stats->l4_csum_bad;
- ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_BAD;
+ switch (ena_rx_ctx->l4_proto) {
+ case ENA_ETH_IO_L4_PROTO_TCP:
+ packet_type |= RTE_PTYPE_L4_TCP;
+ break;
+ case ENA_ETH_IO_L4_PROTO_UDP:
+ packet_type |= RTE_PTYPE_L4_UDP;
+ break;
+ default:
+ break;
+ }
+
+ /* L4 csum is relevant only for TCP/UDP packets */
+ if ((packet_type & (RTE_PTYPE_L4_TCP | RTE_PTYPE_L4_UDP)) && !ena_rx_ctx->frag) {
+ if (ena_rx_ctx->l4_csum_checked) {
+ if (likely(!ena_rx_ctx->l4_csum_err)) {
+ ++rx_stats->l4_csum_good;
+ ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_GOOD;
+ } else {
+ ++rx_stats->l4_csum_bad;
+ ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_BAD;
+ }
} else {
- ++rx_stats->l4_csum_good;
- ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_GOOD;
+ ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_UNKNOWN;
}
- }
- if (fill_hash &&
- likely((packet_type & ENA_PTYPE_HAS_HASH) && !ena_rx_ctx->frag)) {
- ol_flags |= RTE_MBUF_F_RX_RSS_HASH;
- mbuf->hash.rss = ena_rx_ctx->hash;
+ if (rx_ring->offloads & RTE_ETH_RX_OFFLOAD_RSS_HASH) {
+ ol_flags |= RTE_MBUF_F_RX_RSS_HASH;
+ mbuf->hash.rss = ena_rx_ctx->hash;
+ }
+ } else {
+ ol_flags |= RTE_MBUF_F_RX_L4_CKSUM_UNKNOWN;
}
mbuf->ol_flags = ol_flags;
@@ -2765,7 +2776,6 @@ static uint16_t eth_ena_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
uint16_t completed;
struct ena_com_rx_ctx ena_rx_ctx;
int i, rc = 0;
- bool fill_hash;
#ifdef RTE_ETHDEV_DEBUG_RX
/* Check adapter state */
@@ -2776,8 +2786,6 @@ static uint16_t eth_ena_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
}
#endif
- fill_hash = rx_ring->offloads & RTE_ETH_RX_OFFLOAD_RSS_HASH;
-
descs_in_use = rx_ring->ring_size -
ena_com_free_q_entries(rx_ring->ena_com_io_sq) - 1;
nb_pkts = RTE_MIN(descs_in_use, nb_pkts);
@@ -2823,7 +2831,7 @@ static uint16_t eth_ena_recv_pkts(void *rx_queue, struct rte_mbuf **rx_pkts,
}
/* fill mbuf attributes if any */
- ena_rx_mbuf_prepare(rx_ring, mbuf, &ena_rx_ctx, fill_hash);
+ ena_rx_mbuf_prepare(rx_ring, mbuf, &ena_rx_ctx);
if (unlikely(mbuf->ol_flags &
(RTE_MBUF_F_RX_IP_CKSUM_BAD | RTE_MBUF_F_RX_L4_CKSUM_BAD)))
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* [PATCH 15/15] net/ena: upgrade driver version to 2.10.0
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (13 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 14/15] net/ena: rework Rx checksum inspection shaibran
@ 2024-07-02 14:46 ` shaibran
2024-07-07 19:11 ` [PATCH 00/15] net/ena: driver release 2.10.0 Ferruh Yigit
15 siblings, 0 replies; 24+ messages in thread
From: shaibran @ 2024-07-02 14:46 UTC (permalink / raw)
To: ferruh.yigit; +Cc: dev, Shai Brandes
From: Shai Brandes <shaibran@amazon.com>
upgrade driver version to 2.10.0.
Signed-off-by: Shai Brandes <shaibran@amazon.com>
---
drivers/net/ena/ena_ethdev.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/drivers/net/ena/ena_ethdev.c b/drivers/net/ena/ena_ethdev.c
index feb229c5ec..e0c239e88f 100644
--- a/drivers/net/ena/ena_ethdev.c
+++ b/drivers/net/ena/ena_ethdev.c
@@ -22,7 +22,7 @@
#include <ena_eth_io_defs.h>
#define DRV_MODULE_VER_MAJOR 2
-#define DRV_MODULE_VER_MINOR 9
+#define DRV_MODULE_VER_MINOR 10
#define DRV_MODULE_VER_SUBMINOR 0
#define __MERGE_64B_H_L(h, l) (((uint64_t)h << 32) | l)
--
2.17.1
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 07/15] net/ena/base: update copyrights comments
2024-07-02 14:46 ` [PATCH 07/15] net/ena/base: update copyrights comments shaibran
@ 2024-07-07 18:57 ` Ferruh Yigit
2024-07-08 4:08 ` Hemant Agrawal
2024-07-08 11:45 ` Ferruh Yigit
1 sibling, 1 reply; 24+ messages in thread
From: Ferruh Yigit @ 2024-07-07 18:57 UTC (permalink / raw)
To: shaibran; +Cc: dev, Hemant Agrawal
On 7/2/2024 3:46 PM, shaibran@amazon.com wrote:
> From: Shai Brandes <shaibran@amazon.com>
>
> copyright dates are not mandatory to be maintained,
> therefore the range of years was removed.
> In addition, the copyrights lines were separated
> into two comments.
>
> Signed-off-by: Shai Brandes <shaibran@amazon.com>
> ---
> drivers/net/ena/base/ena_com.c | 4 ++--
> drivers/net/ena/base/ena_com.h | 4 ++--
> drivers/net/ena/base/ena_defs/ena_admin_defs.h | 5 +++--
> drivers/net/ena/base/ena_defs/ena_common_defs.h | 4 ++--
> drivers/net/ena/base/ena_defs/ena_eth_io_defs.h | 5 +++--
> drivers/net/ena/base/ena_defs/ena_includes.h | 4 ++--
> drivers/net/ena/base/ena_defs/ena_regs_defs.h | 4 ++--
> drivers/net/ena/base/ena_eth_com.c | 4 ++--
> drivers/net/ena/base/ena_eth_com.h | 4 ++--
> drivers/net/ena/base/ena_plat.h | 4 ++--
> drivers/net/ena/base/ena_plat_dpdk.h | 4 ++--
> 11 files changed, 24 insertions(+), 22 deletions(-)
>
> diff --git a/drivers/net/ena/base/ena_com.c b/drivers/net/ena/base/ena_com.c
> index ad4f3f9431..5f46e692b3 100644
> --- a/drivers/net/ena/base/ena_com.c
> +++ b/drivers/net/ena/base/ena_com.c
> @@ -1,5 +1,5 @@
> -/* SPDX-License-Identifier: BSD-3-Clause
> - * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
> +/* SPDX-License-Identifier: BSD-3-Clause */
> +/* Copyright (c) Amazon.com, Inc. or its affiliates.
> * All rights reserved.
> */
>
Hi Shai,
I don't know about implication of dropping the copyright year
altogether, we need to ask to an expert.
What I know is, a copyright expires after some year, so to hold the
copyright, there needs to be a starting point. And people use ranges to
extend this copyright year.
Instead of removing the year altogether, it make sense to me to just
keep the initial year and don't add range later as code updated, this
also will be more consistent with rest of the code.
Let me drop this patch from the set, and learn more from the experts. It
can be included later, I don't think this is something urgent.
Another aspect is, the history of a source file can be acquired from the
pubic git repository, so when it is first merged, when updated, all
these information is available.
So this feels like no need to record these days in the source code files
but I don't know if these git repository metadata information is legally
valid and accepted.
And I am not quite sure if it has really any value to worry about
copyright for a publicly available device driver code anyway, but this
is not my domain.
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 12/15] net/ena: fix invalid return value check
2024-07-02 14:46 ` [PATCH 12/15] net/ena: fix invalid return value check shaibran
@ 2024-07-07 18:57 ` Ferruh Yigit
0 siblings, 0 replies; 24+ messages in thread
From: Ferruh Yigit @ 2024-07-07 18:57 UTC (permalink / raw)
To: shaibran; +Cc: dev, stable
On 7/2/2024 3:46 PM, shaibran@amazon.com wrote:
> From: Shai Brandes <shaibran@amazon.com>
>
> Removed the sign inversion for when checking if
> ena_com_set_host_attributes returns ENA_COM_UNSUPPORTED.
> ENA_COM_UNSUPPORTED is defined as -EOPNOTSUPP, so the extra sign
> inversion is wrong.
>
> Fixes: 3adcba9a8987 ("net/ena: update HAL to the newer version")
> Cc: stable@dpdk.org
>
> Signed-off-by: Shai Brandes <shaibran@amazon.com>
> ---
> doc/guides/rel_notes/release_24_07.rst | 1 +
> drivers/net/ena/ena_ethdev.c | 4 ++--
> 2 files changed, 3 insertions(+), 2 deletions(-)
>
> diff --git a/doc/guides/rel_notes/release_24_07.rst b/doc/guides/rel_notes/release_24_07.rst
> index f000dec54b..24bb91ad46 100644
> --- a/doc/guides/rel_notes/release_24_07.rst
> +++ b/doc/guides/rel_notes/release_24_07.rst
> @@ -79,6 +79,7 @@ New Features
> * Reworked the device uninitialization flow to ensure complete resource
> cleanup and lay the groundwork for hot-unplug support.
> * Removed an obsolete workaround for a false L4 bad Rx checksum indication.
> + * Fixed an invalid return value check.
>
Dropped this release notes update while merging, as it looks not
significant enough for the release notes.
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 00/15] net/ena: driver release 2.10.0
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
` (14 preceding siblings ...)
2024-07-02 14:46 ` [PATCH 15/15] net/ena: upgrade driver version to 2.10.0 shaibran
@ 2024-07-07 19:11 ` Ferruh Yigit
15 siblings, 0 replies; 24+ messages in thread
From: Ferruh Yigit @ 2024-07-07 19:11 UTC (permalink / raw)
To: shaibran; +Cc: dev
On 7/2/2024 3:46 PM, shaibran@amazon.com wrote:
> From: Shai Brandes <shaibran@amazon.com>
>
> Hi all,
> This release contains:
> 1. HAL upgrade to latest divided into separate patches.
> 2. three bug fixes.
> 3. restructuring of the Rx checksum code for readability
> 4. restructuring of the device uninit flow which is also needed
> for adding hot un/plug support later.
> 5. modification of logs on the datapath that leads to improved performance.
>
> Will appreciate your feedbacks!
> All the best and thanks in advance,
> Shai
>
> Shai Brandes (15):
> net/ena/base: add descriptor dump capability
> net/ena/base: remove unused param
> net/ena/base: remove redundant assert checks
> net/ena/base: update memory barrier comment
> net/ena/base: add method to check used entries
> net/ena/base: add an additional reset reason
> net/ena/base: update copyrights comments
> net/ena/base: add macro for bitfield access
> net/ena: logger change to improve performance
> net/ena: rework device uninit
> net/ena: fix bad checksum handling
> net/ena: fix invalid return value check
> net/ena: fix wrong handling of checksum
> net/ena: rework Rx checksum inspection
> net/ena: upgrade driver version to 2.10.0
>
Except from 7/15 (copyright one),
Series applied to dpdk-next-net/main, thanks.
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 07/15] net/ena/base: update copyrights comments
2024-07-07 18:57 ` Ferruh Yigit
@ 2024-07-08 4:08 ` Hemant Agrawal
2024-07-08 7:02 ` Brandes, Shai
0 siblings, 1 reply; 24+ messages in thread
From: Hemant Agrawal @ 2024-07-08 4:08 UTC (permalink / raw)
To: Ferruh Yigit, shaibran; +Cc: dev, Hemant Agrawal
On 08-07-2024 00:27, Ferruh Yigit wrote:
> On 7/2/2024 3:46 PM, shaibran@amazon.com wrote:
>> From: Shai Brandes <shaibran@amazon.com>
>>
>> copyright dates are not mandatory to be maintained,
>> therefore the range of years was removed.
>> In addition, the copyrights lines were separated
>> into two comments.
>>
>> Signed-off-by: Shai Brandes <shaibran@amazon.com>
>> ---
>> drivers/net/ena/base/ena_com.c | 4 ++--
>> drivers/net/ena/base/ena_com.h | 4 ++--
>> drivers/net/ena/base/ena_defs/ena_admin_defs.h | 5 +++--
>> drivers/net/ena/base/ena_defs/ena_common_defs.h | 4 ++--
>> drivers/net/ena/base/ena_defs/ena_eth_io_defs.h | 5 +++--
>> drivers/net/ena/base/ena_defs/ena_includes.h | 4 ++--
>> drivers/net/ena/base/ena_defs/ena_regs_defs.h | 4 ++--
>> drivers/net/ena/base/ena_eth_com.c | 4 ++--
>> drivers/net/ena/base/ena_eth_com.h | 4 ++--
>> drivers/net/ena/base/ena_plat.h | 4 ++--
>> drivers/net/ena/base/ena_plat_dpdk.h | 4 ++--
>> 11 files changed, 24 insertions(+), 22 deletions(-)
>>
>> diff --git a/drivers/net/ena/base/ena_com.c b/drivers/net/ena/base/ena_com.c
>> index ad4f3f9431..5f46e692b3 100644
>> --- a/drivers/net/ena/base/ena_com.c
>> +++ b/drivers/net/ena/base/ena_com.c
>> @@ -1,5 +1,5 @@
>> -/* SPDX-License-Identifier: BSD-3-Clause
>> - * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
>> +/* SPDX-License-Identifier: BSD-3-Clause */
>> +/* Copyright (c) Amazon.com, Inc. or its affiliates.
>> * All rights reserved.
>> */
>>
> Hi Shai,
>
> I don't know about implication of dropping the copyright year
> altogether, we need to ask to an expert.
>
> What I know is, a copyright expires after some year, so to hold the
> copyright, there needs to be a starting point. And people use ranges to
> extend this copyright year.
>
>
> Instead of removing the year altogether, it make sense to me to just
> keep the initial year and don't add range later as code updated, this
> also will be more consistent with rest of the code.
> Let me drop this patch from the set, and learn more from the experts. It
> can be included later, I don't think this is something urgent.
>
>
> Another aspect is, the history of a source file can be acquired from the
> pubic git repository, so when it is first merged, when updated, all
> these information is available.
> So this feels like no need to record these days in the source code files
> but I don't know if these git repository metadata information is legally
> valid and accepted.
>
>
> And I am not quite sure if it has really any value to worry about
> copyright for a publicly available device driver code anyway, but this
> is not my domain.
Hi Shai,
Will you please just update the copyright year instead of removing
the year completely? e.g. 2015-2020, 2024 or 2015-2024
-Hemant
^ permalink raw reply [flat|nested] 24+ messages in thread
* RE: [PATCH 07/15] net/ena/base: update copyrights comments
2024-07-08 4:08 ` Hemant Agrawal
@ 2024-07-08 7:02 ` Brandes, Shai
2024-07-08 8:38 ` Ferruh Yigit
0 siblings, 1 reply; 24+ messages in thread
From: Brandes, Shai @ 2024-07-08 7:02 UTC (permalink / raw)
To: Hemant Agrawal, Ferruh Yigit; +Cc: dev, Hemant Agrawal
> -----Original Message-----
> From: Hemant Agrawal <hemant.agrawal@oss.nxp.com>
> Sent: Monday, July 8, 2024 7:09 AM
> To: Ferruh Yigit <ferruh.yigit@amd.com>; Brandes, Shai
> <shaibran@amazon.com>
> Cc: dev@dpdk.org; Hemant Agrawal <hemant.agrawal@nxp.com>
> Subject: RE: [EXTERNAL] [PATCH 07/15] net/ena/base: update copyrights
> comments
>
> CAUTION: This email originated from outside of the organization. Do not click
> links or open attachments unless you can confirm the sender and know the
> content is safe.
>
>
>
> On 08-07-2024 00:27, Ferruh Yigit wrote:
> > On 7/2/2024 3:46 PM, shaibran@amazon.com wrote:
> >> From: Shai Brandes <shaibran@amazon.com>
> >>
> >> copyright dates are not mandatory to be maintained, therefore the
> >> range of years was removed.
> >> In addition, the copyrights lines were separated into two comments.
> >>
> >> Signed-off-by: Shai Brandes <shaibran@amazon.com>
> >> ---
> >> drivers/net/ena/base/ena_com.c | 4 ++--
> >> drivers/net/ena/base/ena_com.h | 4 ++--
> >> drivers/net/ena/base/ena_defs/ena_admin_defs.h | 5 +++--
> >> drivers/net/ena/base/ena_defs/ena_common_defs.h | 4 ++--
> >> drivers/net/ena/base/ena_defs/ena_eth_io_defs.h | 5 +++--
> >> drivers/net/ena/base/ena_defs/ena_includes.h | 4 ++--
> >> drivers/net/ena/base/ena_defs/ena_regs_defs.h | 4 ++--
> >> drivers/net/ena/base/ena_eth_com.c | 4 ++--
> >> drivers/net/ena/base/ena_eth_com.h | 4 ++--
> >> drivers/net/ena/base/ena_plat.h | 4 ++--
> >> drivers/net/ena/base/ena_plat_dpdk.h | 4 ++--
> >> 11 files changed, 24 insertions(+), 22 deletions(-)
> >>
> >> diff --git a/drivers/net/ena/base/ena_com.c
> >> b/drivers/net/ena/base/ena_com.c index ad4f3f9431..5f46e692b3 100644
> >> --- a/drivers/net/ena/base/ena_com.c
> >> +++ b/drivers/net/ena/base/ena_com.c
> >> @@ -1,5 +1,5 @@
> >> -/* SPDX-License-Identifier: BSD-3-Clause
> >> - * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
> >> +/* SPDX-License-Identifier: BSD-3-Clause */
> >> +/* Copyright (c) Amazon.com, Inc. or its affiliates.
> >> * All rights reserved.
> >> */
> >>
> > Hi Shai,
> >
> > I don't know about implication of dropping the copyright year
> > altogether, we need to ask to an expert.
> >
> > What I know is, a copyright expires after some year, so to hold the
> > copyright, there needs to be a starting point. And people use ranges
> > to extend this copyright year.
> >
> >
> > Instead of removing the year altogether, it make sense to me to just
> > keep the initial year and don't add range later as code updated, this
> > also will be more consistent with rest of the code.
> > Let me drop this patch from the set, and learn more from the experts.
> > It can be included later, I don't think this is something urgent.
> >
> >
> > Another aspect is, the history of a source file can be acquired from
> > the pubic git repository, so when it is first merged, when updated,
> > all these information is available.
> > So this feels like no need to record these days in the source code
> > files but I don't know if these git repository metadata information is legally
> > valid and accepted.
> >
> >
> > And I am not quite sure if it has really any value to worry about
> > copyright for a publicly available device driver code anyway, but this
> > is not my domain.
>
> Hi Shai,
>
> Will you please just update the copyright year instead of removing the
> year completely? e.g. 2015-2020, 2024 or 2015-2024
>
> -Hemant
[Brandes, Shai]
Hi Ferruh and Hemant, aws removed the copyrights' years range completely for all our open source drivers based on our legal team and the open source article that states that there is no need to provide years or ranges of years for copyright notices in an OSS project.
https://www.linuxfoundation.org/blog/blog/copyright-notices-in-open-source-software-projects
leaving this patch out or using a different years range will create code difference between the dpdk driver and the HAL code it consumes, which is common to all ENA drivers (which is less maintainable.)
Thanks,
Shai
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 07/15] net/ena/base: update copyrights comments
2024-07-08 7:02 ` Brandes, Shai
@ 2024-07-08 8:38 ` Ferruh Yigit
2024-07-08 8:48 ` Brandes, Shai
0 siblings, 1 reply; 24+ messages in thread
From: Ferruh Yigit @ 2024-07-08 8:38 UTC (permalink / raw)
To: Brandes, Shai, Hemant Agrawal; +Cc: dev, Hemant Agrawal, techboard
On 7/8/2024 8:02 AM, Brandes, Shai wrote:
>
>
>> -----Original Message-----
>> From: Hemant Agrawal <hemant.agrawal@oss.nxp.com>
>> Sent: Monday, July 8, 2024 7:09 AM
>> To: Ferruh Yigit <ferruh.yigit@amd.com>; Brandes, Shai
>> <shaibran@amazon.com>
>> Cc: dev@dpdk.org; Hemant Agrawal <hemant.agrawal@nxp.com>
>> Subject: RE: [EXTERNAL] [PATCH 07/15] net/ena/base: update copyrights
>> comments
>>
>> CAUTION: This email originated from outside of the organization. Do not click
>> links or open attachments unless you can confirm the sender and know the
>> content is safe.
>>
>>
>>
>> On 08-07-2024 00:27, Ferruh Yigit wrote:
>>> On 7/2/2024 3:46 PM, shaibran@amazon.com wrote:
>>>> From: Shai Brandes <shaibran@amazon.com>
>>>>
>>>> copyright dates are not mandatory to be maintained, therefore the
>>>> range of years was removed.
>>>> In addition, the copyrights lines were separated into two comments.
>>>>
>>>> Signed-off-by: Shai Brandes <shaibran@amazon.com>
>>>> ---
>>>> drivers/net/ena/base/ena_com.c | 4 ++--
>>>> drivers/net/ena/base/ena_com.h | 4 ++--
>>>> drivers/net/ena/base/ena_defs/ena_admin_defs.h | 5 +++--
>>>> drivers/net/ena/base/ena_defs/ena_common_defs.h | 4 ++--
>>>> drivers/net/ena/base/ena_defs/ena_eth_io_defs.h | 5 +++--
>>>> drivers/net/ena/base/ena_defs/ena_includes.h | 4 ++--
>>>> drivers/net/ena/base/ena_defs/ena_regs_defs.h | 4 ++--
>>>> drivers/net/ena/base/ena_eth_com.c | 4 ++--
>>>> drivers/net/ena/base/ena_eth_com.h | 4 ++--
>>>> drivers/net/ena/base/ena_plat.h | 4 ++--
>>>> drivers/net/ena/base/ena_plat_dpdk.h | 4 ++--
>>>> 11 files changed, 24 insertions(+), 22 deletions(-)
>>>>
>>>> diff --git a/drivers/net/ena/base/ena_com.c
>>>> b/drivers/net/ena/base/ena_com.c index ad4f3f9431..5f46e692b3 100644
>>>> --- a/drivers/net/ena/base/ena_com.c
>>>> +++ b/drivers/net/ena/base/ena_com.c
>>>> @@ -1,5 +1,5 @@
>>>> -/* SPDX-License-Identifier: BSD-3-Clause
>>>> - * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
>>>> +/* SPDX-License-Identifier: BSD-3-Clause */
>>>> +/* Copyright (c) Amazon.com, Inc. or its affiliates.
>>>> * All rights reserved.
>>>> */
>>>>
>>> Hi Shai,
>>>
>>> I don't know about implication of dropping the copyright year
>>> altogether, we need to ask to an expert.
>>>
>>> What I know is, a copyright expires after some year, so to hold the
>>> copyright, there needs to be a starting point. And people use ranges
>>> to extend this copyright year.
>>>
>>>
>>> Instead of removing the year altogether, it make sense to me to just
>>> keep the initial year and don't add range later as code updated, this
>>> also will be more consistent with rest of the code.
>>> Let me drop this patch from the set, and learn more from the experts.
>>> It can be included later, I don't think this is something urgent.
>>>
>>>
>>> Another aspect is, the history of a source file can be acquired from
>>> the pubic git repository, so when it is first merged, when updated,
>>> all these information is available.
>>> So this feels like no need to record these days in the source code
>>> files but I don't know if these git repository metadata information is legally
>>> valid and accepted.
>>>
>>>
>>> And I am not quite sure if it has really any value to worry about
>>> copyright for a publicly available device driver code anyway, but this
>>> is not my domain.
>>
>> Hi Shai,
>>
>> Will you please just update the copyright year instead of removing the
>> year completely? e.g. 2015-2020, 2024 or 2015-2024
>>
>> -Hemant
> [Brandes, Shai]
> Hi Ferruh and Hemant, aws removed the copyrights' years range completely for all our open source drivers based on our legal team and the open source article that states that there is no need to provide years or ranges of years for copyright notices in an OSS project.
> https://www.linuxfoundation.org/blog/blog/copyright-notices-in-open-source-software-projects
> leaving this patch out or using a different years range will create code difference between the dpdk driver and the HAL code it consumes, which is common to all ENA drivers (which is less maintainable.)
>
Hi Shai,
I made the comment to make sure you are protected, but if this is the
request coming from your legal team I think you already know what you
are doing and I am OK to continue with the patch.
By the way, thanks for the blog, it can be reference for future discussions.
^ permalink raw reply [flat|nested] 24+ messages in thread
* RE: [PATCH 07/15] net/ena/base: update copyrights comments
2024-07-08 8:38 ` Ferruh Yigit
@ 2024-07-08 8:48 ` Brandes, Shai
0 siblings, 0 replies; 24+ messages in thread
From: Brandes, Shai @ 2024-07-08 8:48 UTC (permalink / raw)
To: Ferruh Yigit, Hemant Agrawal; +Cc: dev, Hemant Agrawal, techboard
> -----Original Message-----
> From: Ferruh Yigit <ferruh.yigit@amd.com>
> Sent: Monday, July 8, 2024 11:38 AM
> To: Brandes, Shai <shaibran@amazon.com>; Hemant Agrawal
> <hemant.agrawal@oss.nxp.com>
> Cc: dev@dpdk.org; Hemant Agrawal <hemant.agrawal@nxp.com>;
> techboard@dpdk.org
> Subject: RE: [EXTERNAL] [PATCH 07/15] net/ena/base: update copyrights
> comments
>
> CAUTION: This email originated from outside of the organization. Do not click
> links or open attachments unless you can confirm the sender and know the
> content is safe.
>
>
>
> On 7/8/2024 8:02 AM, Brandes, Shai wrote:
> >
> >
> >> -----Original Message-----
> >> From: Hemant Agrawal <hemant.agrawal@oss.nxp.com>
> >> Sent: Monday, July 8, 2024 7:09 AM
> >> To: Ferruh Yigit <ferruh.yigit@amd.com>; Brandes, Shai
> >> <shaibran@amazon.com>
> >> Cc: dev@dpdk.org; Hemant Agrawal <hemant.agrawal@nxp.com>
> >> Subject: RE: [EXTERNAL] [PATCH 07/15] net/ena/base: update copyrights
> >> comments
> >>
> >> CAUTION: This email originated from outside of the organization. Do
> >> not click links or open attachments unless you can confirm the sender
> >> and know the content is safe.
> >>
> >>
> >>
> >> On 08-07-2024 00:27, Ferruh Yigit wrote:
> >>> On 7/2/2024 3:46 PM, shaibran@amazon.com wrote:
> >>>> From: Shai Brandes <shaibran@amazon.com>
> >>>>
> >>>> copyright dates are not mandatory to be maintained, therefore the
> >>>> range of years was removed.
> >>>> In addition, the copyrights lines were separated into two comments.
> >>>>
> >>>> Signed-off-by: Shai Brandes <shaibran@amazon.com>
> >>>> ---
> >>>> drivers/net/ena/base/ena_com.c | 4 ++--
> >>>> drivers/net/ena/base/ena_com.h | 4 ++--
> >>>> drivers/net/ena/base/ena_defs/ena_admin_defs.h | 5 +++--
> >>>> drivers/net/ena/base/ena_defs/ena_common_defs.h | 4 ++--
> >>>> drivers/net/ena/base/ena_defs/ena_eth_io_defs.h | 5 +++--
> >>>> drivers/net/ena/base/ena_defs/ena_includes.h | 4 ++--
> >>>> drivers/net/ena/base/ena_defs/ena_regs_defs.h | 4 ++--
> >>>> drivers/net/ena/base/ena_eth_com.c | 4 ++--
> >>>> drivers/net/ena/base/ena_eth_com.h | 4 ++--
> >>>> drivers/net/ena/base/ena_plat.h | 4 ++--
> >>>> drivers/net/ena/base/ena_plat_dpdk.h | 4 ++--
> >>>> 11 files changed, 24 insertions(+), 22 deletions(-)
> >>>>
> >>>> diff --git a/drivers/net/ena/base/ena_com.c
> >>>> b/drivers/net/ena/base/ena_com.c index ad4f3f9431..5f46e692b3
> >>>> 100644
> >>>> --- a/drivers/net/ena/base/ena_com.c
> >>>> +++ b/drivers/net/ena/base/ena_com.c
> >>>> @@ -1,5 +1,5 @@
> >>>> -/* SPDX-License-Identifier: BSD-3-Clause
> >>>> - * Copyright (c) 2015-2020 Amazon.com, Inc. or its affiliates.
> >>>> +/* SPDX-License-Identifier: BSD-3-Clause */
> >>>> +/* Copyright (c) Amazon.com, Inc. or its affiliates.
> >>>> * All rights reserved.
> >>>> */
> >>>>
> >>> Hi Shai,
> >>>
> >>> I don't know about implication of dropping the copyright year
> >>> altogether, we need to ask to an expert.
> >>>
> >>> What I know is, a copyright expires after some year, so to hold the
> >>> copyright, there needs to be a starting point. And people use ranges
> >>> to extend this copyright year.
> >>>
> >>>
> >>> Instead of removing the year altogether, it make sense to me to just
> >>> keep the initial year and don't add range later as code updated,
> >>> this also will be more consistent with rest of the code.
> >>> Let me drop this patch from the set, and learn more from the experts.
> >>> It can be included later, I don't think this is something urgent.
> >>>
> >>>
> >>> Another aspect is, the history of a source file can be acquired from
> >>> the pubic git repository, so when it is first merged, when updated,
> >>> all these information is available.
> >>> So this feels like no need to record these days in the source code
> >>> files but I don't know if these git repository metadata information is
> legally
> >>> valid and accepted.
> >>>
> >>>
> >>> And I am not quite sure if it has really any value to worry about
> >>> copyright for a publicly available device driver code anyway, but
> >>> this is not my domain.
> >>
> >> Hi Shai,
> >>
> >> Will you please just update the copyright year instead of
> >> removing the year completely? e.g. 2015-2020, 2024 or 2015-2024
> >>
> >> -Hemant
> > [Brandes, Shai]
> > Hi Ferruh and Hemant, aws removed the copyrights' years range
> completely for all our open source drivers based on our legal team and the
> open source article that states that there is no need to provide years or
> ranges of years for copyright notices in an OSS project.
> > https://www.linuxfoundation.org/blog/blog/copyright-notices-in-open-so
> > urce-software-projects leaving this patch out or using a different
> > years range will create code difference between the dpdk driver and
> > the HAL code it consumes, which is common to all ENA drivers (which is
> > less maintainable.)
> >
>
> Hi Shai,
>
> I made the comment to make sure you are protected, but if this is the
> request coming from your legal team I think you already know what you are
> doing and I am OK to continue with the patch.
>
> By the way, thanks for the blog, it can be reference for future discussions.
>
>
[Brandes, Shai] sure thing. And thank you! I appreciate the time you invested for reviewing the patchset.
^ permalink raw reply [flat|nested] 24+ messages in thread
* Re: [PATCH 07/15] net/ena/base: update copyrights comments
2024-07-02 14:46 ` [PATCH 07/15] net/ena/base: update copyrights comments shaibran
2024-07-07 18:57 ` Ferruh Yigit
@ 2024-07-08 11:45 ` Ferruh Yigit
1 sibling, 0 replies; 24+ messages in thread
From: Ferruh Yigit @ 2024-07-08 11:45 UTC (permalink / raw)
To: shaibran; +Cc: dev
On 7/2/2024 3:46 PM, shaibran@amazon.com wrote:
> From: Shai Brandes <shaibran@amazon.com>
>
> copyright dates are not mandatory to be maintained,
> therefore the range of years was removed.
> In addition, the copyrights lines were separated
> into two comments.
>
> Signed-off-by: Shai Brandes <shaibran@amazon.com>
>
Applied to dpdk-next-net/main, thanks.
^ permalink raw reply [flat|nested] 24+ messages in thread
end of thread, other threads:[~2024-07-08 11:45 UTC | newest]
Thread overview: 24+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2024-07-02 14:46 [PATCH 00/15] net/ena: driver release 2.10.0 shaibran
2024-07-02 14:46 ` [PATCH 01/15] net/ena/base: add descriptor dump capability shaibran
2024-07-02 14:46 ` [PATCH 02/15] net/ena/base: remove unused param shaibran
2024-07-02 14:46 ` [PATCH 03/15] net/ena/base: remove redundant assert checks shaibran
2024-07-02 14:46 ` [PATCH 04/15] net/ena/base: update memory barrier comment shaibran
2024-07-02 14:46 ` [PATCH 05/15] net/ena/base: add method to check used entries shaibran
2024-07-02 14:46 ` [PATCH 06/15] net/ena/base: add an additional reset reason shaibran
2024-07-02 14:46 ` [PATCH 07/15] net/ena/base: update copyrights comments shaibran
2024-07-07 18:57 ` Ferruh Yigit
2024-07-08 4:08 ` Hemant Agrawal
2024-07-08 7:02 ` Brandes, Shai
2024-07-08 8:38 ` Ferruh Yigit
2024-07-08 8:48 ` Brandes, Shai
2024-07-08 11:45 ` Ferruh Yigit
2024-07-02 14:46 ` [PATCH 08/15] net/ena/base: add macro for bitfield access shaibran
2024-07-02 14:46 ` [PATCH 09/15] net/ena: logger change to improve performance shaibran
2024-07-02 14:46 ` [PATCH 10/15] net/ena: rework device uninit shaibran
2024-07-02 14:46 ` [PATCH 11/15] net/ena: fix bad checksum handling shaibran
2024-07-02 14:46 ` [PATCH 12/15] net/ena: fix invalid return value check shaibran
2024-07-07 18:57 ` Ferruh Yigit
2024-07-02 14:46 ` [PATCH 13/15] net/ena: fix wrong handling of checksum shaibran
2024-07-02 14:46 ` [PATCH 14/15] net/ena: rework Rx checksum inspection shaibran
2024-07-02 14:46 ` [PATCH 15/15] net/ena: upgrade driver version to 2.10.0 shaibran
2024-07-07 19:11 ` [PATCH 00/15] net/ena: driver release 2.10.0 Ferruh Yigit
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).