[dpdk-dev] [PATCH] vhost: add event areas to vq_is_ready()
Consider a virtqueue ready when, apart from the the descriptor area,
both event supression areas have been mapped.
Fixes: 2d1541e2b6b3 ("vhost: add vring address setup for packed queues")
Cc: y...@fridaylinux.org
Cc: sta...@dpdk.org
Signed-off-by: Adrian Moreno
---
lib/librte_vhost/vhost_user.c | 3 ++-
1 file changed, 2 insertions(+), 1 deletion(-)
diff --git a/lib/librte_vhost/vhost_user.c b/lib/librte_vhost/vhost_user.c
index 69b84a882..c9cc4d648 100644
--- a/lib/librte_vhost/vhost_user.c
+++ b/lib/librte_vhost/vhost_user.c
@@ -1298,7 +1298,8 @@ vq_is_ready(struct virtio_net *dev, struct
vhost_virtqueue *vq)
return false;
if (vq_is_packed(dev))
- rings_ok = !!vq->desc_packed;
+ rings_ok = vq->desc_packed && vq->driver_event &&
+ vq->device_event;
else
rings_ok = vq->desc && vq->avail && vq->used;
--
2.21.1
[dpdk-dev] [PATCH] doc: add rte_ring_xxx_elem introduction to rel notes
Added introduction of rte_ring_xxx_elem APIs to release notes. Signed-off-by: Honnappa Nagarahalli --- doc/guides/rel_notes/release_20_02.rst | 4 1 file changed, 4 insertions(+) diff --git a/doc/guides/rel_notes/release_20_02.rst b/doc/guides/rel_notes/release_20_02.rst index 50e2c1484..eecc938ed 100644 --- a/doc/guides/rel_notes/release_20_02.rst +++ b/doc/guides/rel_notes/release_20_02.rst @@ -61,6 +61,10 @@ New Features A new API has been added to wait for a memory location to be updated with a 16-bit, 32-bit, 64-bit value. +* **Added rte_ring_xxx_elem APIs.** + + New APIs have been added to support rings with custom element size. + * **Updated rte_flow api to support L2TPv3 over IP flows.** Added support for new flow item to handle L2TPv3 over IP rte_flow patterns. -- 2.17.1
Re: [dpdk-dev] [dpdk-stable] [PATCH 2/2] app/testpmd: fix invalid port detaching
Hi Ferruh
From: Ferruh Yigit
> On 1/23/2020 7:25 PM, Matan Azrad wrote:
> > Hi
> >
> > From: Ferruh Yigit
> >> On 1/23/2020 3:29 PM, Matan Azrad wrote:
> >>>
> >>> Hi
> >>>
> >>> From: Ferruh Yigit
> On 1/23/2020 2:05 PM, Matan Azrad wrote:
> > Hi
> >
> > From: Yigit, Ferruh
> >> On 11/12/2019 8:47 AM, Matan Azrad wrote:
> >>> The port was not validated before detaching.
> >>>
> >>> Ignore port detach operation when the port is not valid.
> >>>
> >>> Fixes: f8e5baa2662d ("app/testpmd: check not detaching device
> >>> twice")
> >>> Cc: tho...@monjalon.net
> >>> Cc: sta...@dpdk.org
> >>>
> >>> Signed-off-by: Matan Azrad
> >>> ---
> >>> app/test-pmd/testpmd.c | 3 +++
> >>> 1 file changed, 3 insertions(+)
> >>>
> >>> diff --git a/app/test-pmd/testpmd.c b/app/test-pmd/testpmd.c
> >>> index 346..370eefe 100644
> >>> --- a/app/test-pmd/testpmd.c
> >>> +++ b/app/test-pmd/testpmd.c
> >>> @@ -2545,6 +2545,9 @@ struct extmem_param {
> >>>
> >>> printf("Removing a device...\n");
> >>>
> >>> + if (port_id_is_invalid(port_id, ENABLED_WARN))
> >>> + return;
> >>> +
> >>> dev = rte_eth_devices[port_id].device;
> >>> if (dev == NULL) {
> >>> printf("Device already removed\n");
> >>>
> >>
> >> The patch is already in 19.11 [1] but it is breaking the testpmd
> >> hotplug support.
> >> Before 'detach_port_device()' called, the port has been stopped
> >> and closed [2], which will make port fail from 'port_id_is_invalid()'
> >> check and the device removal path never fully called.
> >> The implication is, since device not detached, vfio request
> >> interrupt keeps triggered continuously and re-starts the detach
> >> path, but because of the half cleaned device it fails and app
> >> gets stuck with a
> continuous log [3].
> >>
> >> I wonder if the actual hotplug has been tested with this patch,
> >> the commit log is not clear about the motivation and implication
> >> of the patch, I am not clear why this check is added but I am
> >> sending a patch soon to remove it back.
> >
> > The motivation of this patch was to prevent double detach on same
> > port,
> so the user cannot call detach of invalid port.
>
> What is the definition of the 'invalid port', if you mean device
> already detached case, in the second call of the function "if (dev
> == NULL)" check should prevent it going forward.
> >>>
> >>> No, ethdev doesn't zero the device pointer when it release a port.
> >>
> >> As far as I can see it does, please see below.
> >
> > The code below is problematic because:
> >
> > 1. It is very bad that the application changing ethdev structure directly.
>
> Where the application is changing the ethdev structure?
See it in the function we talk on:
rte_eth_devices[sibling].device = NULL;
The application shouldn't do it - it should be done only by ethdev lib or by
the PMDs.
Are you agree here?
> Application calls the 'rte_dev_remove()' API, which does the job.
Agree, This function is freeing(rte_free) the rte_device (actually makes the
rte_eth_devices[sibling].device pointer dangled)
and releases its related resources what makes the device detached.
> > 2. The below code run over valid port only, not on invalid port(UNUSED
> state).
> >
> > So, the device pointer will still be valid if the port is invalid.
> >
> > All of this shows that this function try to detach only a valid port
> > (probably
> mainly because it is called by Testpmd detach command).
> >
> >>> So even if the port is in unused state already - means invalid, the
> >>> device
> >> pointer still may be valid and point to the last port that used the same
> >> id.
> >>
> >> If the port is closed, it is unused state, and ethdev layer resources
> >> freed but as you said device related structures are still there,
> >> device pointer is still valid and it is still in probed device list
> >> etc.. We need to able to detach the device even after it is unused state.
> >
> > Yes, but detach is for device, not for port.
> > The device pointer must be taken only when the port is in valid state.
> > Why?
> > Because if the port is in UNUSED state it is free to be allocated again by
> ethdev layer for other device, then, the device pointer may point to other
> device.
> >
Do you agree on the above statement I wrote?
> >> "stop -> close -> detach" is a normal order, we shouldn't prevent it,
> >> but your check does prevent it.
> >
> > Yes, this is good order, but the pointer of the device should be taken
> before close.
> > My patch prevent accessing invalid structure.
>
> The ethdev close() dev_ops, frees ethdev related resources, the rte_device
> is still valid in that struct.
That’s exactly my concern.
I think you wrong here, the rte_device may be invalid in that struct,
e
[dpdk-dev] [PATCH] app/pdump: fix build with clang
Clang checks indentation and found incorrect indentation in pdump.
app/pdump/main.c:598:3: error: misleading indentation; statement is not part of
the previous 'if' [-Werror,-Wmisleading-indentation]
Cc: reshma.pat...@intel.com
Cc: sta...@dpdk.org
Fixes: caa7028276b8 ("app/pdump: add tool for packet capturing")
Signed-off-by: Stephen Hemminger
---
app/pdump/main.c | 2 +-
1 file changed, 1 insertion(+), 1 deletion(-)
diff --git a/app/pdump/main.c b/app/pdump/main.c
index 903d02f48233..d05a0236670a 100644
--- a/app/pdump/main.c
+++ b/app/pdump/main.c
@@ -595,7 +595,7 @@ configure_vdev(uint16_t port_id)
if (ret != 0)
rte_exit(EXIT_FAILURE, "dev config failed\n");
-for (q = 0; q < txRings; q++) {
+ for (q = 0; q < txRings; q++) {
ret = rte_eth_tx_queue_setup(port_id, q, TX_DESC_PER_QUEUE,
rte_eth_dev_socket_id(port_id), NULL);
if (ret < 0)
--
2.20.1
[dpdk-dev] [PATCH 0/5] l3fwd and lpm cleanups
While working on l3fwd, saw some minor things that should be cleaned up. Stephen Hemminger (5): lpm: make ipv6 address immutable examples/l3fwd: use RTE_DIM examples/l3fwd: make lookup struct static examples/l3fwd: make route array constant examples/l3fwd: improve readability for destination lookup examples/l3fwd/l3fwd_lpm.c | 42 -- lib/librte_lpm/rte_lpm6.c | 8 lib/librte_lpm/rte_lpm6.h | 6 +++--- 3 files changed, 29 insertions(+), 27 deletions(-) -- 2.20.1
[dpdk-dev] [PATCH 3/5] examples/l3fwd: make lookup struct static
The lookup structure is only used in the lpm code and does
not have to be global.
Signed-off-by: Stephen Hemminger
---
examples/l3fwd/l3fwd_lpm.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/examples/l3fwd/l3fwd_lpm.c b/examples/l3fwd/l3fwd_lpm.c
index b6802d63ba1d..486f19963fc4 100644
--- a/examples/l3fwd/l3fwd_lpm.c
+++ b/examples/l3fwd/l3fwd_lpm.c
@@ -73,8 +73,8 @@ static struct ipv6_l3fwd_lpm_route
ipv6_l3fwd_lpm_route_array[] = {
#define IPV6_L3FWD_LPM_MAX_RULES 1024
#define IPV6_L3FWD_LPM_NUMBER_TBL8S (1 << 16)
-struct rte_lpm *ipv4_l3fwd_lpm_lookup_struct[NB_SOCKETS];
-struct rte_lpm6 *ipv6_l3fwd_lpm_lookup_struct[NB_SOCKETS];
+static struct rte_lpm *ipv4_l3fwd_lpm_lookup_struct[NB_SOCKETS];
+static struct rte_lpm6 *ipv6_l3fwd_lpm_lookup_struct[NB_SOCKETS];
static inline uint16_t
lpm_get_ipv4_dst_port(void *ipv4_hdr, uint16_t portid, void *lookup_struct)
--
2.20.1
[dpdk-dev] [PATCH 1/5] lpm: make ipv6 address immutable
Both the table setup and lookup do no modify their arguments.
Therefore the parameter should be constant.
This is not actually an API breakage since programs can be
recompiled without change. This is not an ABI breakage because
old programs will still run.
Signed-off-by: Stephen Hemminger
---
lib/librte_lpm/rte_lpm6.c | 8
lib/librte_lpm/rte_lpm6.h | 6 +++---
2 files changed, 7 insertions(+), 7 deletions(-)
diff --git a/lib/librte_lpm/rte_lpm6.c b/lib/librte_lpm/rte_lpm6.c
index c46e557e2389..5ea818abd6e9 100644
--- a/lib/librte_lpm/rte_lpm6.c
+++ b/lib/librte_lpm/rte_lpm6.c
@@ -854,8 +854,8 @@ simulate_add(struct rte_lpm6 *lpm, const uint8_t
*masked_ip, uint8_t depth)
* Add a route
*/
int
-rte_lpm6_add(struct rte_lpm6 *lpm, uint8_t *ip, uint8_t depth,
- uint32_t next_hop)
+rte_lpm6_add(struct rte_lpm6 *lpm, const uint8_t *ip, uint8_t depth,
+uint32_t next_hop)
{
struct rte_lpm6_tbl_entry *tbl;
struct rte_lpm6_tbl_entry *tbl_next = NULL;
@@ -913,7 +913,7 @@ rte_lpm6_add(struct rte_lpm6 *lpm, uint8_t *ip, uint8_t
depth,
*/
static inline int
lookup_step(const struct rte_lpm6 *lpm, const struct rte_lpm6_tbl_entry *tbl,
- const struct rte_lpm6_tbl_entry **tbl_next, uint8_t *ip,
+ const struct rte_lpm6_tbl_entry **tbl_next, const uint8_t *ip,
uint8_t first_byte, uint32_t *next_hop)
{
uint32_t tbl8_index, tbl_entry;
@@ -943,7 +943,7 @@ lookup_step(const struct rte_lpm6 *lpm, const struct
rte_lpm6_tbl_entry *tbl,
* Looks up an IP
*/
int
-rte_lpm6_lookup(const struct rte_lpm6 *lpm, uint8_t *ip,
+rte_lpm6_lookup(const struct rte_lpm6 *lpm, const uint8_t *ip,
uint32_t *next_hop)
{
const struct rte_lpm6_tbl_entry *tbl;
diff --git a/lib/librte_lpm/rte_lpm6.h b/lib/librte_lpm/rte_lpm6.h
index 37dfb20249a8..042991f8cdb7 100644
--- a/lib/librte_lpm/rte_lpm6.h
+++ b/lib/librte_lpm/rte_lpm6.h
@@ -94,8 +94,8 @@ rte_lpm6_free(struct rte_lpm6 *lpm);
* 0 on success, negative value otherwise
*/
int
-rte_lpm6_add(struct rte_lpm6 *lpm, uint8_t *ip, uint8_t depth,
- uint32_t next_hop);
+rte_lpm6_add(struct rte_lpm6 *lpm, const uint8_t *ip, uint8_t depth,
+uint32_t next_hop);
/**
* Check if a rule is present in the LPM table,
@@ -171,7 +171,7 @@ rte_lpm6_delete_all(struct rte_lpm6 *lpm);
* -EINVAL for incorrect arguments, -ENOENT on lookup miss, 0 on lookup hit
*/
int
-rte_lpm6_lookup(const struct rte_lpm6 *lpm, uint8_t *ip, uint32_t *next_hop);
+rte_lpm6_lookup(const struct rte_lpm6 *lpm, const uint8_t *ip, uint32_t
*next_hop);
/**
* Lookup multiple IP addresses in an LPM table.
--
2.20.1
[dpdk-dev] [PATCH 5/5] examples/l3fwd: improve readability for destination lookup
The functions to lookup ipv4 and ipv6 were both using opaque
pointers (void *) when they should use a typed pointer instead.
The ip headers are not modified during lookup.
Get rid of unnecessary cast on the return from the function.
Replace complex trigraph expression with simple if to improve
readability.
Signed-off-by: Stephen Hemminger
---
examples/l3fwd/l3fwd_lpm.c | 28
1 file changed, 16 insertions(+), 12 deletions(-)
diff --git a/examples/l3fwd/l3fwd_lpm.c b/examples/l3fwd/l3fwd_lpm.c
index 30f6385419ee..bd6e7fe03d77 100644
--- a/examples/l3fwd/l3fwd_lpm.c
+++ b/examples/l3fwd/l3fwd_lpm.c
@@ -77,27 +77,31 @@ static struct rte_lpm
*ipv4_l3fwd_lpm_lookup_struct[NB_SOCKETS];
static struct rte_lpm6 *ipv6_l3fwd_lpm_lookup_struct[NB_SOCKETS];
static inline uint16_t
-lpm_get_ipv4_dst_port(void *ipv4_hdr, uint16_t portid, void *lookup_struct)
+lpm_get_ipv4_dst_port(const struct rte_ipv4_hdr *ipv4_hdr,
+ uint16_t portid,
+ struct rte_lpm *ipv4_l3fwd_lookup_struct)
{
+ uint32_t dst_ip = rte_be_to_cpu_32(ipv4_hdr->dst_addr);
uint32_t next_hop;
- struct rte_lpm *ipv4_l3fwd_lookup_struct =
- (struct rte_lpm *)lookup_struct;
- return (uint16_t) ((rte_lpm_lookup(ipv4_l3fwd_lookup_struct,
- rte_be_to_cpu_32(((struct rte_ipv4_hdr *)ipv4_hdr)->dst_addr),
- &next_hop) == 0) ? next_hop : portid);
+ if (rte_lpm_lookup(ipv4_l3fwd_lookup_struct, dst_ip, &next_hop) == 0)
+ return next_hop;
+ else
+ return portid;
}
static inline uint16_t
-lpm_get_ipv6_dst_port(void *ipv6_hdr, uint16_t portid, void *lookup_struct)
+lpm_get_ipv6_dst_port(const struct rte_ipv6_hdr *ipv6_hdr,
+ uint16_t portid,
+ struct rte_lpm6 *ipv6_l3fwd_lookup_struct)
{
+ const uint8_t *dst_ip = ipv6_hdr->dst_addr;
uint32_t next_hop;
- struct rte_lpm6 *ipv6_l3fwd_lookup_struct =
- (struct rte_lpm6 *)lookup_struct;
- return (uint16_t) ((rte_lpm6_lookup(ipv6_l3fwd_lookup_struct,
- ((struct rte_ipv6_hdr *)ipv6_hdr)->dst_addr,
- &next_hop) == 0) ? next_hop : portid);
+ if (rte_lpm6_lookup(ipv6_l3fwd_lookup_struct, dst_ip, &next_hop) == 0)
+ return next_hop;
+ else
+ return portid;
}
static __rte_always_inline uint16_t
--
2.20.1
[dpdk-dev] [PATCH 4/5] examples/l3fwd: make route array constant
The initial route setup array is unmodified by the lpm code
and can be made constant. This depends on earlier patch to
fix the rte_lpm6 to use const.
Signed-off-by: Stephen Hemminger
---
examples/l3fwd/l3fwd_lpm.c | 4 ++--
1 file changed, 2 insertions(+), 2 deletions(-)
diff --git a/examples/l3fwd/l3fwd_lpm.c b/examples/l3fwd/l3fwd_lpm.c
index 486f19963fc4..30f6385419ee 100644
--- a/examples/l3fwd/l3fwd_lpm.c
+++ b/examples/l3fwd/l3fwd_lpm.c
@@ -42,7 +42,7 @@ struct ipv6_l3fwd_lpm_route {
};
/* 198.18.0.0/16 are set aside for RFC2544 benchmarking (RFC5735). */
-static struct ipv4_l3fwd_lpm_route ipv4_l3fwd_lpm_route_array[] = {
+static const struct ipv4_l3fwd_lpm_route ipv4_l3fwd_lpm_route_array[] = {
{RTE_IPV4(198, 18, 0, 0), 24, 0},
{RTE_IPV4(198, 18, 1, 0), 24, 1},
{RTE_IPV4(198, 18, 2, 0), 24, 2},
@@ -54,7 +54,7 @@ static struct ipv4_l3fwd_lpm_route
ipv4_l3fwd_lpm_route_array[] = {
};
/* 2001:0200::/48 is IANA reserved range for IPv6 benchmarking (RFC5180) */
-static struct ipv6_l3fwd_lpm_route ipv6_l3fwd_lpm_route_array[] = {
+static const struct ipv6_l3fwd_lpm_route ipv6_l3fwd_lpm_route_array[] = {
{{32, 1, 2, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0, 0}, 48, 0},
{{32, 1, 2, 0, 0, 0, 0, 0, 0, 1, 0, 0, 0, 0, 0, 0}, 48, 1},
{{32, 1, 2, 0, 0, 0, 0, 0, 0, 2, 0, 0, 0, 0, 0, 0}, 48, 2},
--
2.20.1
[dpdk-dev] [PATCH 2/5] examples/l3fwd: use RTE_DIM
Use the standard RTE_DIM macro to compute array size.
Signed-off-by: Stephen Hemminger
---
examples/l3fwd/l3fwd_lpm.c | 6 ++
1 file changed, 2 insertions(+), 4 deletions(-)
diff --git a/examples/l3fwd/l3fwd_lpm.c b/examples/l3fwd/l3fwd_lpm.c
index 349de2703cd1..b6802d63ba1d 100644
--- a/examples/l3fwd/l3fwd_lpm.c
+++ b/examples/l3fwd/l3fwd_lpm.c
@@ -65,10 +65,8 @@ static struct ipv6_l3fwd_lpm_route
ipv6_l3fwd_lpm_route_array[] = {
{{32, 1, 2, 0, 0, 0, 0, 0, 0, 7, 0, 0, 0, 0, 0, 0}, 48, 7},
};
-#define IPV4_L3FWD_LPM_NUM_ROUTES \
- (sizeof(ipv4_l3fwd_lpm_route_array) /
sizeof(ipv4_l3fwd_lpm_route_array[0]))
-#define IPV6_L3FWD_LPM_NUM_ROUTES \
- (sizeof(ipv6_l3fwd_lpm_route_array) /
sizeof(ipv6_l3fwd_lpm_route_array[0]))
+#define IPV4_L3FWD_LPM_NUM_ROUTES RTE_DIM(ipv4_l3fwd_lpm_route_array)
+#define IPV6_L3FWD_LPM_NUM_ROUTES RTE_DIM(ipv6_l3fwd_lpm_route_array)
#define IPV4_L3FWD_LPM_MAX_RULES 1024
#define IPV4_L3FWD_LPM_NUMBER_TBL8S (1 << 8)
--
2.20.1
