Re: [PATCH] eventdev: fix symbol export for port maintenance
On Wed, Oct 11, 2023 at 8:47 AM Mattias Rönnblom wrote:
>
> On 2023-10-10 16:00, David Marchand wrote:
> > Trying to call rte_event_maintain out of the eventdev library triggers
> > a link failure, as the tracepoint symbol associated to this inline
> > helper was not exported.
> >
> > Fixes: 54f17843a887 ("eventdev: add port maintenance API")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: David Marchand
> > ---
> > Caught by the CI when testing the dispatcher library.
> > See for example:
> > https://github.com/ovsrobot/dpdk/actions/runs/6460514355/job/17538348529#step:19:5506
> >
> > ---
> > lib/eventdev/version.map | 1 +
> > 1 file changed, 1 insertion(+)
> >
> > diff --git a/lib/eventdev/version.map b/lib/eventdev/version.map
> > index b03c10d99f..249eb115b1 100644
> > --- a/lib/eventdev/version.map
> > +++ b/lib/eventdev/version.map
> > @@ -5,6 +5,7 @@ DPDK_24 {
> > __rte_eventdev_trace_deq_burst;
> > __rte_eventdev_trace_enq_burst;
> > __rte_eventdev_trace_eth_tx_adapter_enqueue;
> > + __rte_eventdev_trace_maintain;
> > __rte_eventdev_trace_timer_arm_burst;
> > __rte_eventdev_trace_timer_arm_tmo_tick_burst;
> > __rte_eventdev_trace_timer_cancel_burst;
>
> I can't say I know why it's needed, but the change seems consistent with
> other Eventdev trace points.
The trace point framework in DPDK relies on a per trace point global
variable (whose name is __):
#define __RTE_TRACE_POINT(_mode, _tp, _args, ...) \
extern rte_trace_point_t __##_tp; \
static __rte_always_inline void \
_tp _args \
{ \
__rte_trace_point_emit_header_##_mode(&__##_tp); \
__VA_ARGS__ \
}
When tracepoints are called from within a shared library code, and
because all symbols of a group of objects are visible, the tracepoint
symbols are resolved by the linker.
But when this tracepoint is called via an inline helper from some code
out of the shared library, this symbol must be exported in the shared
library map or it won't be visible to "external" users.
--
David Marchand
Re: [PATCH] eventdev: fix symbol export for port maintenance
Hello Jerin,
On Wed, Oct 11, 2023 at 8:51 AM Jerin Jacob wrote:
>
> On Tue, Oct 10, 2023 at 7:30 PM David Marchand
> wrote:
> >
> > Trying to call rte_event_maintain out of the eventdev library triggers
> > a link failure, as the tracepoint symbol associated to this inline
> > helper was not exported.
> >
> > Fixes: 54f17843a887 ("eventdev: add port maintenance API")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: David Marchand
>
> Acked-by: Jerin Jacob
>
> David, If it is stopping dispatcher library integration, please take
> this patch through main tree, if not, I will add through event tree
> for rc2.
I was going to suggest merging directly in main :-).
I will delegate it to me in patchwork.
Thanks.
--
David Marchand
[PATCH] crypto/cnxk: fix RISC-V compilation
Fixing RISC-V compilation failure by adding check
for NULL pointer.
Fixes: 905537accdd1 ("crypto/cnxk: support raw APIs")
Signed-off-by: Tejasree Kondoj
---
drivers/crypto/cnxk/cnxk_se.h | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/crypto/cnxk/cnxk_se.h b/drivers/crypto/cnxk/cnxk_se.h
index 5d138163f0..c2a807fa94 100644
--- a/drivers/crypto/cnxk/cnxk_se.h
+++ b/drivers/crypto/cnxk/cnxk_se.h
@@ -1834,6 +1834,10 @@ cpt_kasumi_enc_prep(uint32_t req_flags, uint64_t d_offs,
uint64_t d_lens,
cpt_inst_w4.s.param2 = auth_data_len;
inst->w4.u64 = cpt_inst_w4.u64;
+
+ if (unlikely(iv_s == NULL))
+ return -1;
+
if (is_sg_ver2)
sg2_inst_prep(params, inst, offset_ctrl, iv_s, iv_len, 0, 0,
inputlen, outputlen, 0,
req_flags, 0, 0);
@@ -1890,6 +1894,10 @@ cpt_kasumi_dec_prep(uint64_t d_offs, uint64_t d_lens,
struct roc_se_fc_params *p
}
inst->w4.u64 = cpt_inst_w4.u64;
+
+ if (unlikely(params->iv_buf == NULL))
+ return -1;
+
if (is_sg_ver2)
sg2_inst_prep(params, inst, offset_ctrl, params->iv_buf,
iv_len, 0, 0, inputlen,
outputlen, 0, 0, 0, 1);
--
2.25.1
Re: [PATCH] eventdev: fix symbol export for port maintenance
Jerin,
On Wed, Oct 11, 2023 at 9:03 AM David Marchand
wrote:
>
> On Wed, Oct 11, 2023 at 8:47 AM Mattias Rönnblom
> wrote:
> >
> > On 2023-10-10 16:00, David Marchand wrote:
> > > Trying to call rte_event_maintain out of the eventdev library triggers
> > > a link failure, as the tracepoint symbol associated to this inline
> > > helper was not exported.
> > >
> > > Fixes: 54f17843a887 ("eventdev: add port maintenance API")
> > > Cc: sta...@dpdk.org
> > >
> > > Signed-off-by: David Marchand
> > > ---
> > > Caught by the CI when testing the dispatcher library.
> > > See for example:
> > > https://github.com/ovsrobot/dpdk/actions/runs/6460514355/job/17538348529#step:19:5506
> > >
> > > ---
> > > lib/eventdev/version.map | 1 +
> > > 1 file changed, 1 insertion(+)
> > >
> > > diff --git a/lib/eventdev/version.map b/lib/eventdev/version.map
> > > index b03c10d99f..249eb115b1 100644
> > > --- a/lib/eventdev/version.map
> > > +++ b/lib/eventdev/version.map
> > > @@ -5,6 +5,7 @@ DPDK_24 {
> > > __rte_eventdev_trace_deq_burst;
> > > __rte_eventdev_trace_enq_burst;
> > > __rte_eventdev_trace_eth_tx_adapter_enqueue;
> > > + __rte_eventdev_trace_maintain;
> > > __rte_eventdev_trace_timer_arm_burst;
> > > __rte_eventdev_trace_timer_arm_tmo_tick_burst;
> > > __rte_eventdev_trace_timer_cancel_burst;
> >
> > I can't say I know why it's needed, but the change seems consistent with
> > other Eventdev trace points.
>
> The trace point framework in DPDK relies on a per trace point global
> variable (whose name is __):
>
> #define __RTE_TRACE_POINT(_mode, _tp, _args, ...) \
> extern rte_trace_point_t __##_tp; \
> static __rte_always_inline void \
> _tp _args \
> { \
> __rte_trace_point_emit_header_##_mode(&__##_tp); \
> __VA_ARGS__ \
> }
>
> When tracepoints are called from within a shared library code, and
> because all symbols of a group of objects are visible, the tracepoint
> symbols are resolved by the linker.
> But when this tracepoint is called via an inline helper from some code
> out of the shared library, this symbol must be exported in the shared
> library map or it won't be visible to "external" users.
Could we describe / mention this in the trace point library doc?
Or maybe I read too quickly and there is already something but it was
not obvious to me.
--
David Marchand
[PATCH v7 0/3] Add dispatcher library
The purpose of the dispatcher library is to decouple different parts of an eventdev-based application (e.g., processing pipeline stages), sharing the same underlying event device. The dispatcher replaces the conditional logic (often, a switch statement) that typically follows an event device dequeue operation, where events are dispatched to different parts of the application based on event meta data, such as the queue id or scheduling type. The concept is similar to a UNIX file descriptor event loop library. Instead of tying callback functions to fds as for example libevent does, the dispatcher relies on application-supplied matching callback functions to decide where to deliver events. A dispatcher is configured to dequeue events from a specific event device, and ties into the service core framework, to do its (and the application's) work. The dispatcher provides a convenient way for an eventdev-based application to use service cores for application-level processing, and thus for sharing those cores with other DPDK services. Although the dispatcher adds some overhead, experience suggests that the net effect on the application (both synthetic benchmarks and more real-world applications) may well be positive. This is primarily due to clustering (see programming guide) reducing cache misses. Benchmarking indicates that the overhead is ~10 cc/event (on a large core), with a handful of often-used handlers. The dispatcher does not support run-time reconfiguration. The use of the dispatcher library is optional, and an eventdev-based application may still opt to access the event device using direct eventdev API calls, or by some other means. Mattias Rönnblom (3): lib: introduce dispatcher library test: add dispatcher test suite doc: add dispatcher programming guide MAINTAINERS |6 + app/test/meson.build |1 + app/test/test_dispatcher.c | 1074 ++ doc/api/doxy-api-index.md|1 + doc/api/doxy-api.conf.in |1 + doc/guides/prog_guide/dispatcher_lib.rst | 433 + doc/guides/prog_guide/index.rst |1 + doc/guides/rel_notes/release_23_11.rst |5 + lib/dispatcher/meson.build | 13 + lib/dispatcher/rte_dispatcher.c | 691 ++ lib/dispatcher/rte_dispatcher.h | 466 ++ lib/dispatcher/version.map | 20 + lib/meson.build |2 + 13 files changed, 2714 insertions(+) create mode 100644 app/test/test_dispatcher.c create mode 100644 doc/guides/prog_guide/dispatcher_lib.rst create mode 100644 lib/dispatcher/meson.build create mode 100644 lib/dispatcher/rte_dispatcher.c create mode 100644 lib/dispatcher/rte_dispatcher.h create mode 100644 lib/dispatcher/version.map -- 2.34.1
[PATCH v7 1/3] lib: introduce dispatcher library
The purpose of the dispatcher library is to help reduce coupling in an Eventdev-based DPDK application. In addition, the dispatcher also provides a convenient and flexible way for the application to use service cores for application-level processing. Signed-off-by: Mattias Rönnblom Tested-by: Peter Nilsson Reviewed-by: Heng Wang -- PATCH v6: o Use single tab as indentation for continuation lines in multiple-line function prototypes. (David Marchand) o Add dispatcher library release note. (David Marchand) o Various indentation and spelling improvements. (David Marchand) o Add direct , and includes, instead of relying on . (David Marchand) o Avoid Doxygen post annotations for struct fields. (David Marchand) PATCH v5: o Move from using an integer id to a pointer to reference a dispatcher instance, to simplify the API. o Fix bug where dispatcher stats retrieval function erroneously depended on the user-supplied stats buffer being all-zero. PATCH v4: o Fix bugs in handler and finalizer unregistration. (Naga Harish) o Return -EINVAL in cases where NULL pointers were provided in calls requiring non-NULL pointers. (Naga Harish) o Add experimental warning for the whole API. (Jerin Jacob) PATCH v3: o To underline its optional character and since it does not provide hardware abstraction, the event dispatcher is now a separate library. o Change name from rte_event_dispatcher -> rte_dispatcher, to make it shorter and to avoid the rte_event_* namespace. PATCH v2: o Add dequeue batch count statistic. o Add statistics reset function to API. o Clarify MT safety guarantees (or lack thereof) in the API documentation. o Change loop variable type in evd_lcore_get_handler_by_id() to uint16_t, to be consistent with similar loops elsewhere in the dispatcher. o Fix variable names in finalizer unregister function. PATCH: o Change prefix from RED to EVD, to avoid confusion with random early detection. RFC v4: o Move handlers to per-lcore data structures. o Introduce mechanism which rearranges handlers so that often-used handlers tend to be tried first. o Terminate dispatch loop in case all events are delivered. o To avoid the dispatcher's service function hogging the CPU, process only one batch per call. o Have service function return -EAGAIN if no work is performed. o Events delivered in the process function is no longer marked 'const', since modifying them may be useful for the application and cause no difficulties for the dispatcher. o Various minor API documentation improvements. RFC v3: o Add stats_get() function to the version.map file. --- MAINTAINERS| 4 + doc/api/doxy-api-index.md | 1 + doc/api/doxy-api.conf.in | 1 + doc/guides/rel_notes/release_23_11.rst | 5 + lib/dispatcher/meson.build | 13 + lib/dispatcher/rte_dispatcher.c| 691 + lib/dispatcher/rte_dispatcher.h| 466 + lib/dispatcher/version.map | 20 + lib/meson.build| 2 + 9 files changed, 1203 insertions(+) create mode 100644 lib/dispatcher/meson.build create mode 100644 lib/dispatcher/rte_dispatcher.c create mode 100644 lib/dispatcher/rte_dispatcher.h create mode 100644 lib/dispatcher/version.map diff --git a/MAINTAINERS b/MAINTAINERS index 00f5a5f9e6..a4372701c4 100644 --- a/MAINTAINERS +++ b/MAINTAINERS @@ -1733,6 +1733,10 @@ M: Nithin Dabilpuram M: Pavan Nikhilesh F: lib/node/ +Dispatcher - EXPERIMENTAL +M: Mattias Rönnblom +F: lib/dispatcher/ + Test Applications - diff --git a/doc/api/doxy-api-index.md b/doc/api/doxy-api-index.md index fdeda13932..7d0cad9fed 100644 --- a/doc/api/doxy-api-index.md +++ b/doc/api/doxy-api-index.md @@ -155,6 +155,7 @@ The public API headers are grouped by topics: - **classification** [reorder](@ref rte_reorder.h), + [dispatcher](@ref rte_dispatcher.h), [distributor](@ref rte_distributor.h), [EFD](@ref rte_efd.h), [ACL](@ref rte_acl.h), diff --git a/doc/api/doxy-api.conf.in b/doc/api/doxy-api.conf.in index df801d32f9..93709e1d2c 100644 --- a/doc/api/doxy-api.conf.in +++ b/doc/api/doxy-api.conf.in @@ -34,6 +34,7 @@ INPUT = @TOPDIR@/doc/api/doxy-api-index.md \ @TOPDIR@/lib/cmdline \ @TOPDIR@/lib/compressdev \ @TOPDIR@/lib/cryptodev \ + @TOPDIR@/lib/dispatcher \ @TOPDIR@/lib/distributor \ @TOPDIR@/lib/dmadev \ @TOPDIR@/lib/efd \ diff --git a/doc/guides/rel_notes/release_23_11.rst b/doc/guides/rel_notes/release_23_11.rst index 9319c86cd8..b5c5073018 100644 --- a/doc/guides/rel_notes/release_23_11.rst +++ b/doc/guides/rel_notes/release_23_11.rst @@ -80,6 +80,11 @@ New Features device is different from the Tx Ethernet device with re
[PATCH v7 3/3] doc: add dispatcher programming guide
Provide programming guide for the dispatcher library.
Signed-off-by: Mattias Rönnblom
--
PATCH v7:
o Mark pseudo code blocks as being type "none", to avoid Sphinx failures
on non-Ubuntu systems. (David Marchand)
o "Necessarily" necessarily needs to be spelled just so. (David Marchand)
PATCH v6:
o Eliminate unneeded white space in code blocks. (David Marchand)
PATCH v5:
o Update guide to match API changes related to dispatcher ids.
PATCH v3:
o Adapt guide to the dispatcher API name changes.
PATCH:
o Improve grammar and spelling.
RFC v4:
o Extend event matching section of the programming guide.
o Improve grammar and spelling.
---
MAINTAINERS | 1 +
doc/guides/prog_guide/dispatcher_lib.rst | 433 +++
doc/guides/prog_guide/index.rst | 1 +
3 files changed, 435 insertions(+)
create mode 100644 doc/guides/prog_guide/dispatcher_lib.rst
diff --git a/MAINTAINERS b/MAINTAINERS
index 262401d43d..748c15cfe9 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1737,6 +1737,7 @@ Dispatcher - EXPERIMENTAL
M: Mattias Rönnblom
F: lib/dispatcher/
F: app/test/test_dispatcher.c
+F: doc/guides/prog_guide/dispatcher_lib.rst
Test Applications
diff --git a/doc/guides/prog_guide/dispatcher_lib.rst
b/doc/guides/prog_guide/dispatcher_lib.rst
new file mode 100644
index 00..6de1ea78b0
--- /dev/null
+++ b/doc/guides/prog_guide/dispatcher_lib.rst
@@ -0,0 +1,433 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+Copyright(c) 2023 Ericsson AB.
+
+Dispatcher
+==
+
+Overview
+
+
+The purpose of the dispatcher is to help reduce coupling in an
+:doc:`Eventdev `-based DPDK application.
+
+In particular, the dispatcher addresses a scenario where an
+application's modules share the same event device and event device
+ports, and performs work on the same lcore threads.
+
+The dispatcher replaces the conditional logic that follows an event
+device dequeue operation, where events are dispatched to different
+parts of the application, typically based on fields in the
+``rte_event``, such as the ``queue_id``, ``sub_event_type``, or
+``sched_type``.
+
+Below is an excerpt from a fictitious application consisting of two
+modules; A and B. In this example, event-to-module routing is based
+purely on queue id, where module A expects all events to a certain
+queue id, and module B two other queue ids. [#Mapping]_
+
+.. code-block:: c
+
+for (;;) {
+struct rte_event events[MAX_BURST];
+unsigned int n;
+
+n = rte_event_dequeue_burst(dev_id, port_id, events,
+ MAX_BURST, 0);
+
+for (i = 0; i < n; i++) {
+const struct rte_event *event = &events[i];
+
+switch (event->queue_id) {
+case MODULE_A_QUEUE_ID:
+module_a_process(event);
+break;
+case MODULE_B_STAGE_0_QUEUE_ID:
+module_b_process_stage_0(event);
+break;
+case MODULE_B_STAGE_1_QUEUE_ID:
+module_b_process_stage_1(event);
+break;
+}
+}
+}
+
+The issue this example attempts to illustrate is that the centralized
+conditional logic has knowledge of things that should be private to
+the modules. In other words, this pattern leads to a violation of
+module encapsulation.
+
+The shared conditional logic contains explicit knowledge about what
+events should go where. In case, for example, the
+``module_a_process()`` is broken into two processing stages — a
+module-internal affair — the shared conditional code must be updated
+to reflect this change.
+
+The centralized event routing code becomes an issue in larger
+applications, where modules are developed by different organizations.
+This pattern also makes module reuse across different application more
+difficult. The part of the conditional logic relevant for a particular
+application may need to be duplicated across many module
+instantiations (e.g., applications and test setups).
+
+The dispatcher separates the mechanism (routing events to their
+receiver) from the policy (which events should go where).
+
+The basic operation of the dispatcher is as follows:
+
+* Dequeue a batch of events from the event device.
+* For each event determine which handler should receive the event, using
+ a set of application-provided, per-handler event matching callback
+ functions.
+* Provide events matching a particular handler, to that handler, using
+ its process callback.
+
+If the above application would have made use of the dispatcher, the
+code relevant for its module A may have looked something like this:
+
+.. code-block:: c
+
+static bool
+module_a_match(const struct rte_event *event, void *cb_data)
+{
+ return event->queue_id == MODULE_A_QUEUE_
[PATCH v7 2/3] test: add dispatcher test suite
Add unit tests for the dispatcher.
--
PATCH v7:
o Skip (not fail) tests in case too few lcores are available or if
the DSW event device is not available. (David Marchand)
o Properly clean up resources in the above-mentioned scenarios.
PATCH v6:
o Register test as "fast". (David Marchand)
o Use single tab as indentation for continuation lines in multiple-line
function prototypes. (David Marchand)
o Add Signed-off-by line. (David Marchand)
o Use DPDK atomics wrapper API instead of C11 atomics.
PATCH v5:
o Update test suite to use pointer and not integer id when calling
dispatcher functions.
PATCH v3:
o Adapt the test suite to dispatcher API name changes.
PATCH v2:
o Test finalize callback functionality.
o Test handler and finalizer count upper limits.
o Add statistics reset test.
o Make sure dispatcher supply the proper event dev id and port id back
to the application.
PATCH:
o Extend test to cover often-used handler optimization feature.
RFC v4:
o Adapt to non-const events in process function prototype.
Signed-off-by: Mattias Rönnblom
---
MAINTAINERS|1 +
app/test/meson.build |1 +
app/test/test_dispatcher.c | 1074
3 files changed, 1076 insertions(+)
create mode 100644 app/test/test_dispatcher.c
diff --git a/MAINTAINERS b/MAINTAINERS
index a4372701c4..262401d43d 100644
--- a/MAINTAINERS
+++ b/MAINTAINERS
@@ -1736,6 +1736,7 @@ F: lib/node/
Dispatcher - EXPERIMENTAL
M: Mattias Rönnblom
F: lib/dispatcher/
+F: app/test/test_dispatcher.c
Test Applications
diff --git a/app/test/meson.build b/app/test/meson.build
index bf9fc90612..ace10327f8 100644
--- a/app/test/meson.build
+++ b/app/test/meson.build
@@ -59,6 +59,7 @@ source_file_deps = {
'test_cycles.c': [],
'test_debug.c': [],
'test_devargs.c': ['kvargs'],
+'test_dispatcher.c': ['dispatcher'],
'test_distributor.c': ['distributor'],
'test_distributor_perf.c': ['distributor'],
'test_dmadev.c': ['dmadev', 'bus_vdev'],
diff --git a/app/test/test_dispatcher.c b/app/test/test_dispatcher.c
new file mode 100644
index 00..f9dc097f32
--- /dev/null
+++ b/app/test/test_dispatcher.c
@@ -0,0 +1,1074 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(c) 2023 Ericsson AB
+ */
+
+#include
+#include
+#include
+#include
+#include
+#include
+
+#include "test.h"
+
+#define NUM_WORKERS 3
+#define NUM_PORTS (NUM_WORKERS + 1)
+#define WORKER_PORT_ID(worker_idx) (worker_idx)
+#define DRIVER_PORT_ID (NUM_PORTS - 1)
+
+#define NUM_SERVICE_CORES NUM_WORKERS
+#define MIN_LCORES (NUM_SERVICE_CORES + 1)
+
+/* Eventdev */
+#define NUM_QUEUES 8
+#define LAST_QUEUE_ID (NUM_QUEUES - 1)
+#define MAX_EVENTS 4096
+#define NEW_EVENT_THRESHOLD (MAX_EVENTS / 2)
+#define DEQUEUE_BURST_SIZE 32
+#define ENQUEUE_BURST_SIZE 32
+
+#define NUM_EVENTS 1000
+#define NUM_FLOWS 16
+
+#define DSW_VDEV "event_dsw0"
+
+struct app_queue {
+ uint8_t queue_id;
+ uint64_t sn[NUM_FLOWS];
+ int dispatcher_reg_id;
+};
+
+struct cb_count {
+ uint8_t expected_event_dev_id;
+ uint8_t expected_event_port_id[RTE_MAX_LCORE];
+ RTE_ATOMIC(int) count;
+};
+
+struct test_app {
+ uint8_t event_dev_id;
+ struct rte_dispatcher *dispatcher;
+ uint32_t dispatcher_service_id;
+
+ unsigned int service_lcores[NUM_SERVICE_CORES];
+
+ int never_match_reg_id;
+ uint64_t never_match_count;
+ struct cb_count never_process_count;
+
+ struct app_queue queues[NUM_QUEUES];
+
+ int finalize_reg_id;
+ struct cb_count finalize_count;
+
+ bool running;
+
+ RTE_ATOMIC(int) completed_events;
+ RTE_ATOMIC(int) errors;
+};
+
+static struct test_app *
+test_app_create(void)
+{
+ int i;
+ struct test_app *app;
+
+ app = calloc(1, sizeof(struct test_app));
+
+ if (app == NULL)
+ return NULL;
+
+ for (i = 0; i < NUM_QUEUES; i++)
+ app->queues[i].queue_id = i;
+
+ return app;
+}
+
+static void
+test_app_free(struct test_app *app)
+{
+ free(app);
+}
+
+static int
+test_app_create_vdev(struct test_app *app)
+{
+ int rc;
+
+ rc = rte_vdev_init(DSW_VDEV, NULL);
+ if (rc < 0)
+ return TEST_SKIPPED;
+
+ rc = rte_event_dev_get_dev_id(DSW_VDEV);
+
+ app->event_dev_id = (uint8_t)rc;
+
+ return TEST_SUCCESS;
+}
+
+static int
+test_app_destroy_vdev(struct test_app *app)
+{
+ int rc;
+
+ rc = rte_event_dev_close(app->event_dev_id);
+ TEST_ASSERT_SUCCESS(rc, "Error while closing event device");
+
+ rc = rte_vdev_uninit(DSW_VDEV);
+ TEST_ASSERT_SUCCESS(rc, "Error while uninitializing virtual device");
+
+ return TEST_SUCCESS;
+}
+
+static int
+test_app_setup_event_dev(struct test_app *app)
+{
+ int rc;
+ int i;
+
+ rc = test_app_create_vdev(app);
+ if (rc != TEST_SUCCESS)
+ return rc;
+
+ stru
RE: [EXT] Re: [PATCH v2 13/15] crypto/cnxk: add support for raw APIs
Hi Thomas, Please find the fix at https://patchwork.dpdk.org/project/dpdk/patch/20231011071825.3416866-1-ktejas...@marvell.com/ Thanks Tejasree > -Original Message- > From: Thomas Monjalon > Sent: Wednesday, October 11, 2023 1:07 AM > To: Akhil Goyal ; Anoob Joseph > ; Vidya Sagar Velumuri ; > Tejasree Kondoj > Cc: dev@dpdk.org; Aakash Sasidharan ; > Gowrishankar Muthukrishnan ; > dev@dpdk.org > Subject: [EXT] Re: [PATCH v2 13/15] crypto/cnxk: add support for raw APIs > > External Email > > -- > 26/09/2023 07:58, Tejasree Kondoj: > > From: Anoob Joseph > > > > Add crypto RAW API support in cnxk PMD Enable the flag to allow > > execution of raw test suite. > > > > Signed-off-by: Vidya Sagar Velumuri > > Signed-off-by: Anoob Joseph > > --- > > doc/guides/cryptodevs/features/cn10k.ini | 1 + > > drivers/crypto/cnxk/cn10k_cryptodev_ops.c | 459 > ++ > > drivers/crypto/cnxk/cnxk_cryptodev.c | 20 +- > > drivers/crypto/cnxk/cnxk_cryptodev_ops.h | 1 + > > drivers/crypto/cnxk/cnxk_se.h | 293 ++ > > 5 files changed, 761 insertions(+), 13 deletions(-) > > This patch does not compile on RISC-V: > drivers/crypto/cnxk/cnxk_se.h:499:25: error: argument 2 null where non-null > expected >
Re: [PATCH v6 2/3] test: add dispatcher test suite
On Wed, Oct 11, 2023 at 8:28 AM Mattias Rönnblom wrote:
>
> On 2023-10-10 13:56, David Marchand wrote:
> > On Mon, Oct 9, 2023 at 8:22 PM Mattias Rönnblom
> > wrote:
> >> +static int
> >> +test_dispatcher(void)
> >> +{
> >> + return unit_test_suite_runner(&test_suite);
> >> +}
> >> +
> >> +REGISTER_FAST_TEST(dispatcher_autotest, false, true, test_dispatcher);
> >
> > Since this test expects some lcores, wdyt of adding:
> >
> > @@ -1044,6 +1044,12 @@ static struct unit_test_suite test_suite = {
> > static int
> > test_dispatcher(void)
> > {
> > + if (rte_lcore_count() < NUM_SERVICE_CORES + 1) {
> > + printf("Not enough cores for dispatcher_autotest,
> > expecting at least %u\n",
> > + NUM_SERVICE_CORES + 1);
> > + return TEST_SKIPPED;
> > + }
> > +
> > return unit_test_suite_runner(&test_suite);
> > }
> >
> > This should avoid the failures we get with some CI env.
> > (additionnally, I tested this on my laptop and the test runs fine)
> >
> >
>
> Indeed, this is a much better way than to fail the test case.
>
> I'm thinking this is best done in test_setup(), since it's related to
> the setup. In case other test cases are added that required a different
> setup, there may be no minimum lcore requirement.
This is what I had tried as a first attempt but as I hit some crashes
in the teardown step, I went with the easiest fix.
>
> You will get multiple (four, for the moment) print-outs though, in case
> you run with fewer than 4 lcores.
>
> I'll also make sure I skip (and not fail) the tests in case the DSW
> event device is not included in the build.
>
Yep, it is better like this.
Thanks for v7, I'll have a look today.
--
David Marchand
Re: [PATCH] doc: remove confusing command to send patch
On Tue, Oct 10, 2023 at 6:26 PM Thomas Monjalon wrote: > > In the contributor guide, it was said that no need to Cc maintainers > for new additions, probably for new directories not having a maintainer. > There is no harm, and it is a good habit, to always Cc maintainers. > > Remove this case as it can mislead to not Cc maintainers when needed. > > Signed-off-by: Thomas Monjalon I agree Cc: maintainers should be the default / recommended way of sending patches. Just to convince myself, adding some meson skeleton for a "plop" library, adding an entry in the release notes and hooking in lib/meson.build: $ git show --stat doc/guides/rel_notes/release_23_11.rst | 4 lib/meson.build| 1 + lib/plop/meson.build | 2 ++ $ ./devtools/get-maintainer.sh 0001-new-awesome-library.patch In this case, it translates to an empty To: list if you follow the example command line: git send-email --to-cmd ./devtools/get-maintainer.sh --cc dev@dpdk.org 000*.patch We could add a default list of recipients if no maintainer is found by the script. And the next question is who should be in that list.. -- David Marchand
Re: [PATCH] common/qat: enable gen4 c devices
Hello Ciara, On Mon, Aug 21, 2023 at 11:37 AM Power, Ciara wrote: > > > > > > +-+-+-+-+--+---+---++--- > > -+--+++ > > > | Yes | No | No | 4 | 401xxx | IDZ/ N/A | qat_401xxx| > > > 4xxx | > > 4942 | 2| 4943 | 16 | > > > > > > +-+-+-+-+--+---+---+-- > > > --++--+++ > > > + | Yes | Yes | Yes | 4 | 402xxx | linux/6.4+| qat_402xxx| > > > 4xxx | > > 4944 | 2| 4945 | 16 | > > > + > > > +-+-+-+-+--+---+---++-- > > --+--+++ > > > + | Yes | No | No | 4 | 402xxx | IDZ/ N/A | qat_402xxx| > > > 4xxx | > > 4944 | 2| 4945 | 16 | > > > + > > > + +-+-+-+-+--+---+---+ > > > + ++--+++ > > > > Is there such a kernel module named qat_402xxx upstream? > > I can only find qat_4xxx. > > > Good catch, you're right, there is no kernel module 402xxx. > These devices fall under the original 4xxx driver. > Will update here, and send a fix for the 401xxx entry later. I noticed this patch for 402xxx pulled in the main branch. Don't forget to send the fix on 401xxx entry please. -- David Marchand
[PATCH] doc: sort build and EAL features in the release notes
When adding build and EAL features in 23.11, the format and sorting order was unusual. This change is making these features similar as others. Signed-off-by: Thomas Monjalon --- doc/guides/rel_notes/release_23_11.rst | 68 ++ 1 file changed, 37 insertions(+), 31 deletions(-) diff --git a/doc/guides/rel_notes/release_23_11.rst b/doc/guides/rel_notes/release_23_11.rst index b0a957d7b4..671be09003 100644 --- a/doc/guides/rel_notes/release_23_11.rst +++ b/doc/guides/rel_notes/release_23_11.rst @@ -20,23 +20,6 @@ DPDK Release 23.11 ninja -C build doc xdg-open build/doc/guides/html/rel_notes/release_23_11.html -* Build Requirements: From DPDK 23.11 onwards, - building DPDK will require a C compiler which supports the C11 standard, - including support for C11 standard atomics. - - More specifically, the requirements will be: - - * Support for flag "-std=c11" (or similar) - * __STDC_NO_ATOMICS__ is *not defined* when using c11 flag - - Please note: - - * C11, including standard atomics, is supported from GCC version 5 onwards, -and is the default language version in that release -(Ref: https://gcc.gnu.org/gcc-5/changes.html) - * C11 is the default compilation mode in Clang from version 3.6, -which also added support for standard atomics -(Ref: https://releases.llvm.org/3.6.0/tools/clang/docs/ReleaseNotes.html) New Features @@ -72,6 +55,43 @@ New Features Also, make sure to start the actual text at the margin. === +* **Build requirements increased for C11.** + + From DPDK 23.11 onwards, + building DPDK will require a C compiler which supports the C11 standard, + including support for C11 standard atomics. + + More specifically, the requirements will be: + + * Support for flag "-std=c11" (or similar) + * __STDC_NO_ATOMICS__ is *not defined* when using c11 flag + + Please note: + + * C11, including standard atomics, is supported from GCC version 5 onwards, +and is the default language version in that release +(Ref: https://gcc.gnu.org/gcc-5/changes.html) + * C11 is the default compilation mode in Clang from version 3.6, +which also added support for standard atomics +(Ref: https://releases.llvm.org/3.6.0/tools/clang/docs/ReleaseNotes.html) + +* **Added new build options.** + + * Enabling deprecated libraries is now done using +the new ``enable_deprecated_libraries`` build option. + * Optional libraries can now be selected with the new ``enable_libs`` +build option similarly to the existing ``enable_drivers`` build option. + +* **Introduced a new API for atomic operations.** + + This new API serves as a wrapper for transitioning + to standard atomic operations as described in the C11 standard. + This API implementation points at the compiler intrinsics by default. + The implementation using C11 standard atomic operations is enabled + via the ``enable_stdatomic`` build option. + +* **Added support for power intrinsics with AMD processors.** + * **Added mbuf recycling support.** Added ``rte_eth_recycle_rx_queue_info_get`` and ``rte_eth_recycle_mbufs`` @@ -153,20 +173,6 @@ New Features * Added SM2 algorithm support in asymmetric crypto operations. -* build: Enabling deprecated libraries is now done using the new - ``enable_deprecated_libraries`` build option. - -* build: Optional libraries can now be selected with the new ``enable_libs`` - build option similarly to the existing ``enable_drivers`` build option. - -* eal: Introduced a new API for atomic operations. This new API serves as a - wrapper for transitioning to standard atomic operations as described in the - C11 standard. This API implementation points at the compiler intrinsics by - default. The implementation using C11 standard atomic operations is enabled - via the ``enable_stdatomic`` build option. - -* eal: Added support for power intrinsics with AMD processors. - Removed Items - -- 2.42.0
Re: [PATCH] doc: remove confusing command to send patch
11/10/2023 09:30, David Marchand: > On Tue, Oct 10, 2023 at 6:26 PM Thomas Monjalon wrote: > > > > In the contributor guide, it was said that no need to Cc maintainers > > for new additions, probably for new directories not having a maintainer. > > There is no harm, and it is a good habit, to always Cc maintainers. > > > > Remove this case as it can mislead to not Cc maintainers when needed. > > > > Signed-off-by: Thomas Monjalon > > I agree Cc: maintainers should be the default / recommended way of > sending patches. > > Just to convince myself, adding some meson skeleton for a "plop" > library, adding an entry in the release notes and hooking in > lib/meson.build: > $ git show --stat > doc/guides/rel_notes/release_23_11.rst | 4 > lib/meson.build| 1 + > lib/plop/meson.build | 2 ++ > > $ ./devtools/get-maintainer.sh 0001-new-awesome-library.patch > > In this case, it translates to an empty To: list if you follow the > example command line: >git send-email --to-cmd ./devtools/get-maintainer.sh --cc > dev@dpdk.org 000*.patch > > We could add a default list of recipients if no maintainer is found by > the script. > And the next question is who should be in that list.. Or we can send to dev@dpdk.org, Cc maintainers. This is what I do: git send-email --to dev@dpdk.org --cc-cmd devtools/get-maintainer.sh
Re: [PATCH] crypto/cnxk: fix RISC-V compilation
11/10/2023 09:18, Tejasree Kondoj:
> Fixing RISC-V compilation failure by adding check
> for NULL pointer.
>
> Fixes: 905537accdd1 ("crypto/cnxk: support raw APIs")
>
> Signed-off-by: Tejasree Kondoj
Squashed in "crypto/cnxk: support raw APIs", thank you.
[Bug 1296] net/mlx5 driver causing indirect pool leakage when testing IPv6 fragmentation
https://bugs.dpdk.org/show_bug.cgi?id=1296 Bug ID: 1296 Summary: net/mlx5 driver causing indirect pool leakage when testing IPv6 fragmentation Product: DPDK Version: 22.11 Hardware: x86 OS: Linux Status: UNCONFIRMED Severity: normal Priority: Normal Component: ethdev Assignee: dev@dpdk.org Reporter: java@ericsson.com Target Milestone: --- When testing IPv6 fragmentation uisng net/mlx5 driver we noticed the indirect buffer is not returning to the pool and will eventually cause indirect pool to be exausted. when using intel nic with iavf driver with same application code the issue is not seen. There is also no issue for pcap driver. when review code https://github.com/DPDK/dpdk/blob/main/drivers/net/mlx5/mlx5_tx.h function mlx5_tx_free_mbuf(). Seems in rte_pktmbuf_prefree_seg(*pkts) the pkts->next will be set to NULL for indirect buffer which is linked to the direct buffer. Test is done by explicitly putting the indirect buffer back to mem pool at end of function mlx5_tx_free_mbuf(). -- You are receiving this mail because: You are the assignee for the bug.
[PATCH] net/cpfl: reset devargs during the first probe
From: Beilei Xing
Reset devargs during the first probe. Otherwise, probe again will
be affected.
Fixes: a607312291b3 ("net/cpfl: support probe again")
Signed-off-by: Beilei Xing
---
drivers/net/cpfl/cpfl_ethdev.c | 6 +++---
1 file changed, 3 insertions(+), 3 deletions(-)
diff --git a/drivers/net/cpfl/cpfl_ethdev.c b/drivers/net/cpfl/cpfl_ethdev.c
index 762fbddfe6..890a027a1d 100644
--- a/drivers/net/cpfl/cpfl_ethdev.c
+++ b/drivers/net/cpfl/cpfl_ethdev.c
@@ -1611,11 +1611,12 @@ cpfl_parse_devargs(struct rte_pci_device *pci_dev,
struct cpfl_adapter_ext *adap
struct rte_kvargs *kvlist;
int ret;
- cpfl_args->req_vport_nb = 0;
-
if (devargs == NULL)
return 0;
+ if (first)
+ memset(cpfl_args, 0, sizeof(struct cpfl_devargs));
+
kvlist = rte_kvargs_parse(devargs->args,
first ? cpfl_valid_args_first : cpfl_valid_args_again);
if (kvlist == NULL) {
@@ -1628,7 +1629,6 @@ cpfl_parse_devargs(struct rte_pci_device *pci_dev, struct
cpfl_adapter_ext *adap
return -EINVAL;
}
- cpfl_args->repr_args_num = 0;
ret = rte_kvargs_process(kvlist, CPFL_REPRESENTOR, &parse_repr,
cpfl_args);
if (ret != 0)
--
2.34.1
Re: [PATCH] doc: remove confusing command to send patch
On Wed, Oct 11, 2023 at 10:03:07AM +0200, Thomas Monjalon wrote: > 11/10/2023 09:30, David Marchand: > > On Tue, Oct 10, 2023 at 6:26 PM Thomas Monjalon wrote: > > > > > > In the contributor guide, it was said that no need to Cc maintainers > > > for new additions, probably for new directories not having a maintainer. > > > There is no harm, and it is a good habit, to always Cc maintainers. > > > > > > Remove this case as it can mislead to not Cc maintainers when needed. > > > > > > Signed-off-by: Thomas Monjalon > > > > I agree Cc: maintainers should be the default / recommended way of > > sending patches. > > > > Just to convince myself, adding some meson skeleton for a "plop" > > library, adding an entry in the release notes and hooking in > > lib/meson.build: > > $ git show --stat > > doc/guides/rel_notes/release_23_11.rst | 4 > > lib/meson.build| 1 + > > lib/plop/meson.build | 2 ++ > > > > $ ./devtools/get-maintainer.sh 0001-new-awesome-library.patch > > > > In this case, it translates to an empty To: list if you follow the > > example command line: > >git send-email --to-cmd ./devtools/get-maintainer.sh --cc > > dev@dpdk.org 000*.patch > > > > We could add a default list of recipients if no maintainer is found by > > the script. > > And the next question is who should be in that list.. > > Or we can send to dev@dpdk.org, Cc maintainers. > This is what I do: > git send-email --to dev@dpdk.org --cc-cmd devtools/get-maintainer.sh > +1 for this, mainly on the basis of it being what I do too! :-)
Re: [PATCH] ethdev: add calculate hash function
On 10/11/2023 3:11 AM, fengchengwen wrote: > Hi, > > On 2023/10/10 19:05, Ferruh Yigit wrote: >> On 9/26/2023 12:37 PM, Ori Kam wrote: >>> rte_flow supports insert by index table[1]. >>> >>> Using the above table, the application can create rules >>> that are based on hash. >>> For example application can create the following logic in order >>> to create load balancing: >>> 1. Create insert by index table with 2 rules, that hashes based on dmac >>> 2. Insert to index 0 a rule that sends the traffic to port A. >>> 3. Insert to index 1 a rule that sends the traffic to port B. >>> >>> Let's also assume that before this table, there is a 5 tuple >>> match table that jumps to the above table. >>> >>> So each packet that matches one of the 5 tuple rules is RSSed >>> to port A or B, based on dmac hash. >>> >>> The issue arises when there is a miss on the 5 tuple table, >>> which resulted due to the packet being the first packet of this flow, or >>> fragmented packet or any other reason. >>> In this case, the application must calculate what would be the >>> hash calculated by the HW so it can send the packet to the correct >>> port. >>> >>> This new API allows applications to calculate the hash value of a given >>> packet for a given table. >>> >>> [1] - >>> http://patches.dpdk.org/project/dpdk/patch/20230208030624.78465-2-akozy...@nvidia.com/ >>> >>> Signed-off-by: Ori Kam >>> --- >>> app/test-pmd/cmdline_flow.c | 86 +++- >>> app/test-pmd/config.c| 54 ++ >>> app/test-pmd/testpmd.h | 2 + >>> lib/ethdev/rte_flow.c| 21 + >>> lib/ethdev/rte_flow.h| 32 ++ >>> lib/ethdev/rte_flow_driver.h | 5 +++ >>> lib/ethdev/version.map | 1 + >>> 7 files changed, 200 insertions(+), 1 deletion(-) >>> >> >> This is a new rte_flow API but unfortunately there isn't any >> review/comment, at least it is experimental API. If there is no >> objection/discussion in next few days, I will merge the feature. >> >> Probably it will be another rte flow feature that only NVIDIA knows and >> uses. While mentioned from using, is the driver update for the feature > > The hns3 driver support subset of rte_flow, we found the rte_flow feature is > very flexible. > And its implementation varies according to vendors. > > Can the rte_flow be standardized ? > Hi Chengwen, Yes rte_flow is already implemented by many vendors, each uses some subset of it. It is flexible and useful, no concern about it. My point was, most of the new rte_flow features are coming from single vendor and most of them are not fully reviewed by the wider community. As some of the features merged without much review from wider community, not everyone aware of them, and features are not fully benefited from, although that is somewhat related to HW support as Jerin pointed before. As hns3 is a user of the rte_flow already, it would be great to get more feedback and review from hns3 maintainers, that boosts the confidence to the new proposed features/APIs. Thanks, ferruh >> planned for this release? >> >> >> Meanwhile, can you please update the documentation, `rte_flow.rst` and >> `testpmd_funcs.rst`? >> Also can you please rebase on top of latest next-net, this patch >> conflicts with merged group set miss action feature. >> >> . >>
[PATCH v2 0/3] rewrite fastpath routines
This series adds new fastpath routines for cn10k & cn9k endpoint devices and supports 32B Tx desciptor format which improves the performance. v2 change: - Fixed CI Shijith Thotton (1): net/octeon_ep: support 32B IQ descriptor size Vamsi Attunuru (2): net/octeon_ep: clean up receive routine net/octeon_ep: add new fastpath routines drivers/net/octeon_ep/cnxk_ep_rx.c| 309 ++ drivers/net/octeon_ep/cnxk_ep_tx.c| 209 + drivers/net/octeon_ep/cnxk_ep_vf.c| 12 +- drivers/net/octeon_ep/cnxk_ep_vf.h| 13 ++ drivers/net/octeon_ep/meson.build | 12 + drivers/net/octeon_ep/otx2_ep_vf.c| 11 +- drivers/net/octeon_ep/otx_ep_common.h | 127 ++- drivers/net/octeon_ep/otx_ep_ethdev.c | 69 +- drivers/net/octeon_ep/otx_ep_rxtx.c | 263 +++--- drivers/net/octeon_ep/otx_ep_rxtx.h | 38 +++- drivers/net/octeon_ep/otx_ep_vf.c | 8 + 11 files changed, 814 insertions(+), 257 deletions(-) create mode 100644 drivers/net/octeon_ep/cnxk_ep_rx.c create mode 100644 drivers/net/octeon_ep/cnxk_ep_tx.c -- 2.25.1
[PATCH v2 1/3] net/octeon_ep: support 32B IQ descriptor size
From: Shijith Thotton
Update input queue setup to consider descriptor size in driver conf.
The default instruction size for otx2 and cnxk devices has been updated
to 32 bytes.
Signed-off-by: Shijith Thotton
---
drivers/net/octeon_ep/cnxk_ep_vf.c| 10 +-
drivers/net/octeon_ep/meson.build | 10 ++
drivers/net/octeon_ep/otx2_ep_vf.c| 10 +-
drivers/net/octeon_ep/otx_ep_common.h | 4
drivers/net/octeon_ep/otx_ep_rxtx.c | 8 +++-
drivers/net/octeon_ep/otx_ep_vf.c | 8
6 files changed, 43 insertions(+), 7 deletions(-)
diff --git a/drivers/net/octeon_ep/cnxk_ep_vf.c
b/drivers/net/octeon_ep/cnxk_ep_vf.c
index 92c2d2ca5c..7b3669fe0c 100644
--- a/drivers/net/octeon_ep/cnxk_ep_vf.c
+++ b/drivers/net/octeon_ep/cnxk_ep_vf.c
@@ -106,6 +106,14 @@ cnxk_ep_vf_setup_iq_regs(struct otx_ep_device *otx_ep,
uint32_t iq_no)
return -EIO;
}
+ /* Configure input queue instruction size. */
+ if (otx_ep->conf->iq.instr_type == OTX_EP_32BYTE_INSTR)
+ reg_val &= ~(CNXK_EP_R_IN_CTL_IS_64B);
+ else
+ reg_val |= CNXK_EP_R_IN_CTL_IS_64B;
+ oct_ep_write64(reg_val, otx_ep->hw_addr + CNXK_EP_R_IN_CONTROL(iq_no));
+ iq->desc_size = otx_ep->conf->iq.instr_type;
+
/* Write the start of the input queue's ring and its size */
oct_ep_write64(iq->base_addr_dma, otx_ep->hw_addr +
CNXK_EP_R_IN_INSTR_BADDR(iq_no));
oct_ep_write64(iq->nb_desc, otx_ep->hw_addr +
CNXK_EP_R_IN_INSTR_RSIZE(iq_no));
@@ -354,7 +362,7 @@ static const struct otx_ep_config default_cnxk_ep_conf = {
/* IQ attributes */
.iq= {
.max_iqs = OTX_EP_CFG_IO_QUEUES,
- .instr_type= OTX_EP_64BYTE_INSTR,
+ .instr_type= OTX_EP_32BYTE_INSTR,
.pending_list_size = (OTX_EP_MAX_IQ_DESCRIPTORS *
OTX_EP_CFG_IO_QUEUES),
},
diff --git a/drivers/net/octeon_ep/meson.build
b/drivers/net/octeon_ep/meson.build
index e698bf9792..4538c0396e 100644
--- a/drivers/net/octeon_ep/meson.build
+++ b/drivers/net/octeon_ep/meson.build
@@ -10,3 +10,13 @@ sources = files(
'cnxk_ep_vf.c',
'otx_ep_mbox.c',
)
+
+if (toolchain == 'gcc' and cc.version().version_compare('>=11.0.0'))
+error_cflags += ['-Wno-array-bounds']
+endif
+
+foreach flag: error_cflags
+if cc.has_argument(flag)
+c_args += flag
+endif
+endforeach
diff --git a/drivers/net/octeon_ep/otx2_ep_vf.c
b/drivers/net/octeon_ep/otx2_ep_vf.c
index ced3a415a5..f72b8d25d7 100644
--- a/drivers/net/octeon_ep/otx2_ep_vf.c
+++ b/drivers/net/octeon_ep/otx2_ep_vf.c
@@ -256,6 +256,14 @@ otx2_vf_setup_iq_regs(struct otx_ep_device *otx_ep,
uint32_t iq_no)
return -EIO;
}
+ /* Configure input queue instruction size. */
+ if (otx_ep->conf->iq.instr_type == OTX_EP_32BYTE_INSTR)
+ reg_val &= ~(SDP_VF_R_IN_CTL_IS_64B);
+ else
+ reg_val |= SDP_VF_R_IN_CTL_IS_64B;
+ oct_ep_write64(reg_val, otx_ep->hw_addr + SDP_VF_R_IN_CONTROL(iq_no));
+ iq->desc_size = otx_ep->conf->iq.instr_type;
+
/* Write the start of the input queue's ring and its size */
oct_ep_write64(iq->base_addr_dma, otx_ep->hw_addr +
SDP_VF_R_IN_INSTR_BADDR(iq_no));
oct_ep_write64(iq->nb_desc, otx_ep->hw_addr +
SDP_VF_R_IN_INSTR_RSIZE(iq_no));
@@ -500,7 +508,7 @@ static const struct otx_ep_config default_otx2_ep_conf = {
/* IQ attributes */
.iq= {
.max_iqs = OTX_EP_CFG_IO_QUEUES,
- .instr_type= OTX_EP_64BYTE_INSTR,
+ .instr_type= OTX_EP_32BYTE_INSTR,
.pending_list_size = (OTX_EP_MAX_IQ_DESCRIPTORS *
OTX_EP_CFG_IO_QUEUES),
},
diff --git a/drivers/net/octeon_ep/otx_ep_common.h
b/drivers/net/octeon_ep/otx_ep_common.h
index c150cbe619..90e059cad0 100644
--- a/drivers/net/octeon_ep/otx_ep_common.h
+++ b/drivers/net/octeon_ep/otx_ep_common.h
@@ -11,6 +11,7 @@
#define OTX_EP_MAX_RINGS_PER_VF(8)
#define OTX_EP_CFG_IO_QUEUESOTX_EP_MAX_RINGS_PER_VF
+#define OTX_EP_32BYTE_INSTR (32)
#define OTX_EP_64BYTE_INSTR (64)
/*
* Backpressure for SDP is configured on Octeon, and the minimum queue sizes
@@ -215,6 +216,9 @@ struct otx_ep_instr_queue {
/* Number of descriptors in this ring. */
uint32_t nb_desc;
+ /* Size of the descriptor. */
+ uint8_t desc_size;
+
/* Input ring index, where the driver should write the next packet */
uint32_t host_write_index;
diff --git a/drivers/net/octeon_ep/otx_ep_rxtx.c
b/drivers/net/octeon_ep/otx_ep_rxtx.c
index b37fc8109f..5b759d759b 100644
--- a/drivers/net/octeon_ep/otx_ep_rxtx.c
+++ b/drivers/net/octeon_ep/otx_ep_rxtx.c
@@ -484,7 +484,7 @@ otx_ep_ri
[PATCH v2 2/3] net/octeon_ep: clean up receive routine
Patch improves Rx routine and pkt count update routines,
packet count update routines need to drain inflight ISM
memory updates while decrementing the packet count register.
Signed-off-by: Vamsi Attunuru
---
drivers/net/octeon_ep/otx_ep_rxtx.c | 162
1 file changed, 68 insertions(+), 94 deletions(-)
diff --git a/drivers/net/octeon_ep/otx_ep_rxtx.c
b/drivers/net/octeon_ep/otx_ep_rxtx.c
index 5b759d759b..ea7c9a5d62 100644
--- a/drivers/net/octeon_ep/otx_ep_rxtx.c
+++ b/drivers/net/octeon_ep/otx_ep_rxtx.c
@@ -442,7 +442,14 @@ otx_vf_update_read_index(struct otx_ep_instr_queue *iq)
* when count above halfway to saturation.
*/
rte_write32(val, iq->inst_cnt_reg);
- *iq->inst_cnt_ism = 0;
+ rte_mb();
+
+ rte_write64(OTX2_SDP_REQUEST_ISM, iq->inst_cnt_reg);
+ while (__atomic_load_n(iq->inst_cnt_ism, __ATOMIC_RELAXED) >=
val) {
+ rte_write64(OTX2_SDP_REQUEST_ISM, iq->inst_cnt_reg);
+ rte_mb();
+ }
+
iq->inst_cnt_ism_prev = 0;
}
rte_write64(OTX2_SDP_REQUEST_ISM, iq->inst_cnt_reg);
@@ -565,9 +572,7 @@ prepare_xmit_gather_list(struct otx_ep_instr_queue *iq,
struct rte_mbuf *m, uint
finfo = &iq->req_list[iq->host_write_index].finfo;
*dptr = rte_mem_virt2iova(finfo->g.sg);
- ih->s.tlen = pkt_len + ih->s.fsz;
- ih->s.gsz = frags;
- ih->s.gather = 1;
+ ih->u64 |= ((1ULL << 62) | ((uint64_t)frags << 48) | (pkt_len +
ih->s.fsz));
while (frags--) {
finfo->g.sg[(j >> 2)].ptr[(j & mask)] = rte_mbuf_data_iova(m);
@@ -750,36 +755,26 @@ otx2_ep_xmit_pkts(void *tx_queue, struct rte_mbuf **pkts,
uint16_t nb_pkts)
static uint32_t
otx_ep_droq_refill(struct otx_ep_droq *droq)
{
- struct otx_ep_droq_desc *desc_ring;
+ struct otx_ep_droq_desc *desc_ring = droq->desc_ring;
struct otx_ep_droq_info *info;
struct rte_mbuf *buf = NULL;
uint32_t desc_refilled = 0;
- desc_ring = droq->desc_ring;
-
while (droq->refill_count && (desc_refilled < droq->nb_desc)) {
- /* If a valid buffer exists (happens if there is no dispatch),
-* reuse the buffer, else allocate.
-*/
- if (droq->recv_buf_list[droq->refill_idx] != NULL)
- break;
-
buf = rte_pktmbuf_alloc(droq->mpool);
/* If a buffer could not be allocated, no point in
* continuing
*/
- if (buf == NULL) {
+ if (unlikely(!buf)) {
droq->stats.rx_alloc_failure++;
break;
}
info = rte_pktmbuf_mtod(buf, struct otx_ep_droq_info *);
- memset(info, 0, sizeof(*info));
+ info->length = 0;
droq->recv_buf_list[droq->refill_idx] = buf;
desc_ring[droq->refill_idx].buffer_ptr =
rte_mbuf_data_iova_default(buf);
-
-
droq->refill_idx = otx_ep_incr_index(droq->refill_idx, 1,
droq->nb_desc);
@@ -791,21 +786,18 @@ otx_ep_droq_refill(struct otx_ep_droq *droq)
}
static struct rte_mbuf *
-otx_ep_droq_read_packet(struct otx_ep_device *otx_ep,
- struct otx_ep_droq *droq, int next_fetch)
+otx_ep_droq_read_packet(struct otx_ep_device *otx_ep, struct otx_ep_droq
*droq, int next_fetch)
{
volatile struct otx_ep_droq_info *info;
- struct rte_mbuf *droq_pkt2 = NULL;
- struct rte_mbuf *droq_pkt = NULL;
- struct rte_net_hdr_lens hdr_lens;
- struct otx_ep_droq_info *info2;
+ struct rte_mbuf *mbuf_next = NULL;
+ struct rte_mbuf *mbuf = NULL;
uint64_t total_pkt_len;
uint32_t pkt_len = 0;
int next_idx;
- droq_pkt = droq->recv_buf_list[droq->read_idx];
- droq_pkt2 = droq->recv_buf_list[droq->read_idx];
- info = rte_pktmbuf_mtod(droq_pkt, struct otx_ep_droq_info *);
+ mbuf = droq->recv_buf_list[droq->read_idx];
+ info = rte_pktmbuf_mtod(mbuf, struct otx_ep_droq_info *);
+
/* make sure info is available */
rte_rmb();
if (unlikely(!info->length)) {
@@ -826,32 +818,25 @@ otx_ep_droq_read_packet(struct otx_ep_device *otx_ep,
assert(0);
}
}
+
if (next_fetch) {
next_idx = otx_ep_incr_index(droq->read_idx, 1, droq->nb_desc);
- droq_pkt2 = droq->recv_buf_list[next_idx];
- info2 = rte_pktmbuf_mtod(droq_pkt2, struct otx_ep_droq_info *);
- rte_prefetch_non_temporal((const void *)info2);
+ mbuf_next = droq->recv_buf_list[next_idx];
+ rte_prefetch0(rte_pktmbuf_mtod(mbuf_next, void *));
}
- info->
[PATCH v2 3/3] net/octeon_ep: add new fastpath routines
Adds new fastpath routines for cn10k & cn9k endpoint
devices and assigns the fastpath routines based on
the offload flags.
Patch also adds misc changes to improve performance
and code-readability.
Signed-off-by: Vamsi Attunuru
---
drivers/net/octeon_ep/cnxk_ep_rx.c| 309 ++
drivers/net/octeon_ep/cnxk_ep_tx.c| 209 +
drivers/net/octeon_ep/cnxk_ep_vf.c| 2 +
drivers/net/octeon_ep/cnxk_ep_vf.h| 13 ++
drivers/net/octeon_ep/meson.build | 2 +
drivers/net/octeon_ep/otx2_ep_vf.c| 1 +
drivers/net/octeon_ep/otx_ep_common.h | 125 ++-
drivers/net/octeon_ep/otx_ep_ethdev.c | 69 +-
drivers/net/octeon_ep/otx_ep_rxtx.c | 93 +---
drivers/net/octeon_ep/otx_ep_rxtx.h | 38 +++-
10 files changed, 704 insertions(+), 157 deletions(-)
diff --git a/drivers/net/octeon_ep/cnxk_ep_rx.c
b/drivers/net/octeon_ep/cnxk_ep_rx.c
new file mode 100644
index 00..74f0011283
--- /dev/null
+++ b/drivers/net/octeon_ep/cnxk_ep_rx.c
@@ -0,0 +1,309 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2023 Marvell.
+ */
+
+#include "otx_ep_common.h"
+#include "otx2_ep_vf.h"
+#include "otx_ep_rxtx.h"
+
+static inline int
+cnxk_ep_rx_refill_mbuf(struct otx_ep_droq *droq, uint32_t count)
+{
+ struct otx_ep_droq_desc *desc_ring = droq->desc_ring;
+ struct rte_mbuf **recv_buf_list = droq->recv_buf_list;
+ uint32_t refill_idx = droq->refill_idx;
+ struct rte_mbuf *buf;
+ uint32_t i;
+ int rc;
+
+ rc = rte_pktmbuf_alloc_bulk(droq->mpool, &recv_buf_list[refill_idx],
count);
+ if (unlikely(rc)) {
+ droq->stats.rx_alloc_failure++;
+ return rc;
+ }
+
+ for (i = 0; i < count; i++) {
+ buf = recv_buf_list[refill_idx];
+ desc_ring[refill_idx].buffer_ptr =
rte_mbuf_data_iova_default(buf);
+ refill_idx++;
+ }
+
+ droq->refill_idx = otx_ep_incr_index(droq->refill_idx, count,
droq->nb_desc);
+ droq->refill_count -= count;
+
+ return 0;
+}
+
+static inline void
+cnxk_ep_rx_refill(struct otx_ep_droq *droq)
+{
+ uint32_t desc_refilled = 0, count;
+ uint32_t nb_desc = droq->nb_desc;
+ uint32_t refill_idx = droq->refill_idx;
+ int rc;
+
+ if (unlikely(droq->read_idx == refill_idx))
+ return;
+
+ if (refill_idx < droq->read_idx) {
+ count = droq->read_idx - refill_idx;
+ rc = cnxk_ep_rx_refill_mbuf(droq, count);
+ if (unlikely(rc)) {
+ droq->stats.rx_alloc_failure++;
+ return;
+ }
+ desc_refilled = count;
+ } else {
+ count = nb_desc - refill_idx;
+ rc = cnxk_ep_rx_refill_mbuf(droq, count);
+ if (unlikely(rc)) {
+ droq->stats.rx_alloc_failure++;
+ return;
+ }
+
+ desc_refilled = count;
+ count = droq->read_idx;
+ rc = cnxk_ep_rx_refill_mbuf(droq, count);
+ if (unlikely(rc)) {
+ droq->stats.rx_alloc_failure++;
+ return;
+ }
+ desc_refilled += count;
+ }
+
+ /* Flush the droq descriptor data to memory to be sure
+* that when we update the credits the data in memory is
+* accurate.
+*/
+ rte_io_wmb();
+ rte_write32(desc_refilled, droq->pkts_credit_reg);
+}
+
+static inline uint32_t
+cnxk_ep_check_rx_pkts(struct otx_ep_droq *droq)
+{
+ uint32_t new_pkts;
+ uint32_t val;
+
+ /* Batch subtractions from the HW counter to reduce PCIe traffic
+* This adds an extra local variable, but almost halves the
+* number of PCIe writes.
+*/
+ val = __atomic_load_n(droq->pkts_sent_ism, __ATOMIC_RELAXED);
+ new_pkts = val - droq->pkts_sent_ism_prev;
+ droq->pkts_sent_ism_prev = val;
+
+ if (val > (uint32_t)(1 << 31)) {
+ /* Only subtract the packet count in the HW counter
+* when count above halfway to saturation.
+*/
+ rte_write64((uint64_t)val, droq->pkts_sent_reg);
+ rte_mb();
+
+ rte_write64(OTX2_SDP_REQUEST_ISM, droq->pkts_sent_reg);
+ while (__atomic_load_n(droq->pkts_sent_ism, __ATOMIC_RELAXED)
>= val) {
+ rte_write64(OTX2_SDP_REQUEST_ISM, droq->pkts_sent_reg);
+ rte_mb();
+ }
+
+ droq->pkts_sent_ism_prev = 0;
+ }
+ rte_write64(OTX2_SDP_REQUEST_ISM, droq->pkts_sent_reg);
+ droq->pkts_pending += new_pkts;
+
+ return new_pkts;
+}
+
+static inline int16_t __rte_hot
+cnxk_ep_rx_pkts_to_process(struct otx_ep_droq *droq, uint16_t nb_pkts)
+{
+ if (droq->pkts_pending < nb_pkts)
+ cnxk_ep_check_rx_pkts(d
Re: [PATCH] doc: remove confusing command to send patch
On 10/11/2023 9:30 AM, Bruce Richardson wrote: > On Wed, Oct 11, 2023 at 10:03:07AM +0200, Thomas Monjalon wrote: >> 11/10/2023 09:30, David Marchand: >>> On Tue, Oct 10, 2023 at 6:26 PM Thomas Monjalon wrote: In the contributor guide, it was said that no need to Cc maintainers for new additions, probably for new directories not having a maintainer. There is no harm, and it is a good habit, to always Cc maintainers. Remove this case as it can mislead to not Cc maintainers when needed. Signed-off-by: Thomas Monjalon >>> >>> I agree Cc: maintainers should be the default / recommended way of >>> sending patches. >>> >>> Just to convince myself, adding some meson skeleton for a "plop" >>> library, adding an entry in the release notes and hooking in >>> lib/meson.build: >>> $ git show --stat >>> doc/guides/rel_notes/release_23_11.rst | 4 >>> lib/meson.build| 1 + >>> lib/plop/meson.build | 2 ++ >>> >>> $ ./devtools/get-maintainer.sh 0001-new-awesome-library.patch >>> >>> In this case, it translates to an empty To: list if you follow the >>> example command line: >>>git send-email --to-cmd ./devtools/get-maintainer.sh --cc >>> dev@dpdk.org 000*.patch >>> >>> We could add a default list of recipients if no maintainer is found by >>> the script. >>> And the next question is who should be in that list.. >> >> Or we can send to dev@dpdk.org, Cc maintainers. >> This is what I do: >> git send-email --to dev@dpdk.org --cc-cmd devtools/get-maintainer.sh >> > +1 for this, mainly on the basis of it being what I do too! :-) > I am for "--to-cmd=./devtools/get-maintainer.sh --cc dev@dpdk.org" To highlight response is expected from the maintainers, and community is informed. Also people may have filters to give higher priority to emails they are in 'to' list, high priority is what we want from maintainers :)
RE: [PATCH v2] ethdev: add TCP/IP modify field IDs
Hi, > -Original Message- > From: Suanming Mou > Sent: Wednesday, October 11, 2023 8:11 AM > To: Ferruh Yigit ; Ori Kam ; Aman > Singh ; Yuying Zhang ; > NBU-Contact-Thomas Monjalon (EXTERNAL) ; Andrew > Rybchenko > Cc: dev@dpdk.org > Subject: RE: [PATCH v2] ethdev: add TCP/IP modify field IDs > > Hi, > > > -Original Message- > > From: Ferruh Yigit > > Sent: Tuesday, October 10, 2023 6:24 PM > > To: Suanming Mou ; Ori Kam ; > > Aman Singh ; Yuying Zhang > > ; NBU-Contact-Thomas Monjalon (EXTERNAL) > > ; Andrew Rybchenko > > > > Cc: dev@dpdk.org > > Subject: Re: [PATCH v2] ethdev: add TCP/IP modify field IDs > > > > On 9/8/2023 4:49 AM, Suanming Mou wrote: > > > Currently, get TCP/IP header or data length information from traffic > > > is missing in the modify field IDs. This commit adds the missing TCP > > > data_offset, IPv4 IHL/total_len, IPv6 payload_len to modify filed IDs. > > > This allows users be able to manager more TCP/IP fields. > > > > > > Signed-off-by: Suanming Mou > > > --- > > > > > > v2: fix typo tcp_date_off -> tcp_data_off > > > > > > --- > > > app/test-pmd/cmdline_flow.c | 1 + > > > lib/ethdev/rte_flow.h | 4 > > > 2 files changed, 5 insertions(+) > > > > > > > Hi Suanming, > > > > Patch looks good. But, testpmd modify flow action support seems not > > documented at all, can you please first add it [1], later update that > > document with this patch? > > > > Also can you please check if `rte_flow.rst` also needs to be updated or not? > > Sure, will check and update. `rte_flow.rst` has the modify_field action description in `Action: ``MODIFY_FIELD``` section. Will update only the ` testpmd_funcs.rst`. > Thanks > > > > > > [1]: `doc/guides/testpmd_app_ug/testpmd_funcs.rst`, `Flow rules > > management` section
Re: [PATCH 0/3] enable PPC in test-meson-builds on ubuntu
31/08/2023 14:10, Bruce Richardson: > To help developers catch errors as soon as possible, we want to ensure > that as wide a variety of builds are done by test-meson-builds as > possible. Unfortunately, for those using Ubuntu, the shipped version > of GCC compiler for PowerPC on that system is not correctly detected by > the script. We fix this by detecting if Ubuntu is in use, and pointing > to the correct cross-file in that case. > > On enabling those PPC builds on my system, a couple of other build > issues we encountered. These are fixed by the set before enabling the > build in the script. > > Bruce Richardson (3): > vhost: fix build for powerpc > build: fix failures due to incompatible IPSec lib > devtools: enable testing ppc builds on ubuntu Applied, thanks.
RE: [PATCH v1] app/testpmd: refine encap content
Hi Zhang,
> -Original Message-
> From: Zhang, Yuying
> Sent: Tuesday, August 22, 2023 4:13 AM
>
> From: Yuying Zhang
>
> Refine vxlan encap content of all protocol headers.
>
> Fixes: 1960be7d32f8 ("app/testpmd: add VXLAN encap/decap")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Yuying Zhang
> ---
> app/test-pmd/cmdline_flow.c | 12 ++--
> 1 file changed, 10 insertions(+), 2 deletions(-)
>
> diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
> index 95c0a19beb..a488864e2b 100644
> --- a/app/test-pmd/cmdline_flow.c
> +++ b/app/test-pmd/cmdline_flow.c
> @@ -8523,7 +8523,7 @@ parse_setup_vxlan_encap_data(struct
> action_vxlan_encap_data *action_vxlan_encap_
> .type = RTE_FLOW_ITEM_TYPE_END,
> },
> },
> - .item_eth.hdr.ether_type = 0,
> + .item_eth.hdr.ether_type =
> rte_cpu_to_be_16(RTE_ETHER_TYPE_IPV4),
> .item_vlan = {
> .hdr.vlan_tci = vxlan_encap_conf.vlan_tci,
> .hdr.eth_proto = 0,
> @@ -8531,24 +8531,32 @@ parse_setup_vxlan_encap_data(struct
> action_vxlan_encap_data *action_vxlan_encap_
> .item_ipv4.hdr = {
> .src_addr = vxlan_encap_conf.ipv4_src,
> .dst_addr = vxlan_encap_conf.ipv4_dst,
> + .version_ihl = RTE_IPV4_VHL_DEF,
> + .next_proto_id = IPPROTO_UDP,
> + .time_to_live = IPDEFTTL,
You can use the following command, right?
set vxlan-tos-ttl ip-version ipv4|ipv6 vni udp-src udp-dst
ip-tos ip-ttl ip-src ip-dst
eth-src eth-dst
> + .hdr_checksum = rte_cpu_to_be_16(1),
This checksum should be calculated by HW so why set it?
> },
> .item_udp.hdr = {
> .src_port = vxlan_encap_conf.udp_src,
> .dst_port = vxlan_encap_conf.udp_dst,
> + .dgram_cksum = RTE_BE16(0x01),
This should be calculated by HW or set to 0
> },
> - .item_vxlan.hdr.flags = 0,
> + .item_vxlan.hdr.flags = 0x08,
> };
> memcpy(action_vxlan_encap_data-
> >item_eth.hdr.dst_addr.addr_bytes,
> vxlan_encap_conf.eth_dst, RTE_ETHER_ADDR_LEN);
> memcpy(action_vxlan_encap_data->item_eth.hdr.src_addr.addr_bytes,
> vxlan_encap_conf.eth_src, RTE_ETHER_ADDR_LEN);
> if (!vxlan_encap_conf.select_ipv4) {
> + action_vxlan_encap_data->item_eth.type =
> RTE_BE16(RTE_ETHER_TYPE_IPV6);
> memcpy(&action_vxlan_encap_data->item_ipv6.hdr.src_addr,
> &vxlan_encap_conf.ipv6_src,
> sizeof(vxlan_encap_conf.ipv6_src));
> memcpy(&action_vxlan_encap_data->item_ipv6.hdr.dst_addr,
> &vxlan_encap_conf.ipv6_dst,
> sizeof(vxlan_encap_conf.ipv6_dst));
> + action_vxlan_encap_data->item_ipv6.hdr.proto =
> IPPROTO_UDP;
> + action_vxlan_encap_data->item_ipv6.hdr.hop_limits =
> IPDEFTTL;
> action_vxlan_encap_data->items[2] = (struct rte_flow_item){
> .type = RTE_FLOW_ITEM_TYPE_IPV6,
> .spec = &action_vxlan_encap_data->item_ipv6,
> --
> 2.25.1
Best,
Ori
[PATCH v5 00/40] support setting and querying RSS algorithms
This patchset is to support setting and querying RSS algorithms. -- v5: 1. rewrite some comments. 2. check RSS algorithm for drivers supporting RSS. 3. change field "func" of rss_conf to "algorithm". 4. fix commit log for [PATCH v4 4/7]. 5. add Acked-by Reshma Pattan. 6. add symmetric_toeplitz_sort for showing. 7. change "hf" to "hash function" for showing. v4: 1. recomment some definitions related to RSS. 2. allocate static memory for rss_key instead of dynamic. 3. use array of strings to get the name of rss algorithm. 4. add display of rss algorithm with testpmd. v3: 1. fix commit log for PATCH [1/5]. 2. make RSS ABI changes description to start the actual text at the margin. 3. move defnition of enum rte_eth_hash_function to rte_ethdev.h. 4. fix some comment codes. v2: 1. return error if "func" is invalid. 2. modify the comments of the "func" field. 3. modify commit log of patch [3/5]. 4. use malloc instead of rte_malloc. 5. adjust display format of RSS info. 6. remove the string display of rss_hf. Huisong Li (1): net/hns3: support setting and querying RSS hash function Jie Hai (39): ethdev: overwrite some comment related to RSS ethdev: support setting and querying RSS algorithm net/atlantic: check RSS hash algorithms net/axgbe: check RSS hash algorithms net/bnx2x: check RSS hash algorithms net/bnxt: check RSS hash algorithms net/bonding: check RSS hash algorithms net/cnxk: check RSS hash algorithms net/cpfl: check RSS hash algorithms net/cxgbe: check RSS hash algorithms net/dpaa: check RSS hash algorithms net/dpaa2: check RSS hash algorithms net/ena: check RSS hash algorithms net/enic: check RSS hash algorithms net/fm10k: check RSS hash algorithms net/hinic: check RSS hash algorithms net/i40e: check RSS hash algorithms net/iavf: check RSS hash algorithms net/ice: check RSS hash algorithms net/idpf: check RSS hash algorithms net/igc: check RSS hash algorithms net/ionic: check RSS hash algorithms net/ixgbe: check RSS hash algorithms net/mana: check RSS hash algorithms net/mlx5: check RSS hash algorithms net/mvpp2: check RSS hash algorithms net/netvsc: check RSS hash algorithms net/ngbe: : check RSS hash algorithms net/nfp: check RSS hash algorithms net/null: check RSS hash algorithms net/qede: check RSS hash algorithms net/sfc: check RSS hash algorithms net/tap: check RSS hash algorithms net/thunderx: check RSS hash algorithms net/txgbe: check RSS hash algorithms app/proc-info: fix never show RSS info app/proc-info: adjust the display format of RSS info app/proc-info: support querying RSS hash algorithm app/testpmd: add RSS hash algorithms display app/proc-info/main.c | 32 ++- app/test-pmd/cmdline.c | 29 ++--- app/test-pmd/config.c | 38 - app/test-pmd/testpmd.h | 2 +- doc/guides/rel_notes/release_23_11.rst | 2 + drivers/net/atlantic/atl_ethdev.c | 2 + drivers/net/axgbe/axgbe_ethdev.c | 9 + drivers/net/bnx2x/bnx2x_ethdev.c | 4 ++ drivers/net/bnxt/bnxt_ethdev.c | 6 +++ drivers/net/bonding/rte_eth_bond_pmd.c | 6 +++ drivers/net/cnxk/cnxk_ethdev.c | 5 +++ drivers/net/cnxk/cnxk_ethdev_ops.c | 3 ++ drivers/net/cpfl/cpfl_ethdev.c | 6 +++ drivers/net/cxgbe/cxgbe_ethdev.c | 9 - drivers/net/dpaa/dpaa_ethdev.c | 7 drivers/net/dpaa2/dpaa2_ethdev.c | 7 drivers/net/ena/ena_rss.c | 3 ++ drivers/net/enic/enic_ethdev.c | 1 + drivers/net/enic/enic_main.c | 3 ++ drivers/net/fm10k/fm10k_ethdev.c | 9 - drivers/net/hinic/hinic_pmd_ethdev.c | 3 ++ drivers/net/hinic/hinic_pmd_rx.c | 3 ++ drivers/net/hns3/hns3_rss.c| 47 - drivers/net/i40e/i40e_ethdev.c | 7 drivers/net/iavf/iavf_ethdev.c | 6 +++ drivers/net/ice/ice_dcf.c | 3 ++ drivers/net/ice/ice_dcf_ethdev.c | 3 ++ drivers/net/ice/ice_ethdev.c | 7 drivers/net/idpf/idpf_ethdev.c | 6 +++ drivers/net/igc/igc_ethdev.c | 4 ++ drivers/net/igc/igc_txrx.c | 5 +++ drivers/net/ionic/ionic_ethdev.c | 6 +++ drivers/net/ixgbe/ixgbe_ethdev.c | 12 +- drivers/net/ixgbe/ixgbe_rxtx.c | 4 ++ drivers/net/mana/mana.c| 11 - drivers/net/mlx5/mlx5_ethdev.c | 4 ++ drivers/net/mlx5/mlx5_rss.c| 3 +- drivers/net/mvpp2/mrvl_ethdev.c| 3 ++ drivers/net/netvsc/hn_ethdev.c | 6 +++ drivers/net/nfp/nfp_common.c | 9 - drivers/net/ngbe/ngbe_ethdev.c | 6 ++- drivers/net/ngbe/ngbe_rxtx.c | 3 ++ drivers/net/null/rte_eth_null.c| 8 drivers/net/qede/qede_ethdev.c | 9 - drivers/net/sfc/sfc_ethdev.c | 3 ++ drivers/net/sfc/sfc_rx.c | 3 ++ drivers/ne
[PATCH v5 01/40] ethdev: overwrite some comment related to RSS
1. overwrite the comments of fields of 'rte_eth_rss_conf'.
2. Add comments for RTE_ETH_HASH_FUNCTION_DEFAULT.
Signed-off-by: Jie Hai
---
lib/ethdev/rte_ethdev.h | 29 ++---
lib/ethdev/rte_flow.h | 3 +++
2 files changed, 17 insertions(+), 15 deletions(-)
diff --git a/lib/ethdev/rte_ethdev.h b/lib/ethdev/rte_ethdev.h
index 8542257721c9..b9e4e21189d2 100644
--- a/lib/ethdev/rte_ethdev.h
+++ b/lib/ethdev/rte_ethdev.h
@@ -448,24 +448,23 @@ struct rte_vlan_filter_conf {
/**
* A structure used to configure the Receive Side Scaling (RSS) feature
* of an Ethernet port.
- * If not NULL, the *rss_key* pointer of the *rss_conf* structure points
- * to an array holding the RSS key to use for hashing specific header
- * fields of received packets. The length of this array should be indicated
- * by *rss_key_len* below. Otherwise, a default random hash key is used by
- * the device driver.
- *
- * The *rss_key_len* field of the *rss_conf* structure indicates the length
- * in bytes of the array pointed by *rss_key*. To be compatible, this length
- * will be checked in i40e only. Others assume 40 bytes to be used as before.
- *
- * The *rss_hf* field of the *rss_conf* structure indicates the different
- * types of IPv4/IPv6 packets to which the RSS hashing must be applied.
- * Supplying an *rss_hf* equal to zero disables the RSS feature.
*/
struct rte_eth_rss_conf {
- uint8_t *rss_key;/**< If not NULL, 40-byte hash key. */
+ /**
+* If used to query, the'rss_key_len' indicates the size of rss key of
+* the hardware. And only when rss_key_len is not zero, the 'rss_key'
+* is valid.
+* If used to configure, rss_key_len indicates the length of the
+* 'rss_key' if 'rss_key' is not empty.
+*/
+ uint8_t *rss_key;
uint8_t rss_key_len; /**< hash key length in bytes. */
- uint64_t rss_hf; /**< Hash functions to apply - see below. */
+ /**
+* Indicating which type of packets and which part of the packets
+* to apply for RSS hash, (see RTE_ETH_RSS_*).
+* Setting *rss_hf* to zero disables the RSS feature.
+*/
+ uint64_t rss_hf;
};
/*
diff --git a/lib/ethdev/rte_flow.h b/lib/ethdev/rte_flow.h
index b385741fba6d..5d9e3c68af7b 100644
--- a/lib/ethdev/rte_flow.h
+++ b/lib/ethdev/rte_flow.h
@@ -3227,6 +3227,9 @@ struct rte_flow_query_count {
* Hash function types.
*/
enum rte_eth_hash_function {
+ /**
+* DEFAULT means driver decides which hash algorithm to pick.
+*/
RTE_ETH_HASH_FUNCTION_DEFAULT = 0,
RTE_ETH_HASH_FUNCTION_TOEPLITZ, /**< Toeplitz */
RTE_ETH_HASH_FUNCTION_SIMPLE_XOR, /**< Simple XOR */
--
2.30.0
[PATCH v5 03/40] net/atlantic: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/atlantic/atl_ethdev.c | 2 ++
1 file changed, 2 insertions(+)
diff --git a/drivers/net/atlantic/atl_ethdev.c
b/drivers/net/atlantic/atl_ethdev.c
index 3a028f429002..c6e772d47263 100644
--- a/drivers/net/atlantic/atl_ethdev.c
+++ b/drivers/net/atlantic/atl_ethdev.c
@@ -1857,6 +1857,8 @@ atl_rss_hash_update(struct rte_eth_dev *dev,
};
cfg->is_rss = !!rss_conf->rss_hf;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
if (rss_conf->rss_key) {
memcpy(cfg->aq_rss.hash_secret_key, rss_conf->rss_key,
rss_conf->rss_key_len);
--
2.30.0
[PATCH v5 04/40] net/axgbe: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/axgbe/axgbe_ethdev.c | 9 +
1 file changed, 9 insertions(+)
diff --git a/drivers/net/axgbe/axgbe_ethdev.c b/drivers/net/axgbe/axgbe_ethdev.c
index d08ea4893c37..f67faff2d636 100644
--- a/drivers/net/axgbe/axgbe_ethdev.c
+++ b/drivers/net/axgbe/axgbe_ethdev.c
@@ -339,6 +339,12 @@ static int
axgbe_dev_configure(struct rte_eth_dev *dev)
{
struct axgbe_port *pdata = dev->data->dev_private;
+
+ if (dev->data->dev_conf.rxmode.mq_mode == RTE_ETH_MQ_RX_RSS &&
+ dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* Checksum offload to hardware */
pdata->rx_csum_enable = dev->data->dev_conf.rxmode.offloads &
RTE_ETH_RX_OFFLOAD_CHECKSUM;
@@ -582,6 +588,9 @@ axgbe_dev_rss_hash_update(struct rte_eth_dev *dev,
return -EINVAL;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_key != NULL &&
rss_conf->rss_key_len == AXGBE_RSS_HASH_KEY_SIZE) {
rte_memcpy(pdata->rss_key, rss_conf->rss_key,
--
2.30.0
[PATCH v5 02/40] ethdev: support setting and querying RSS algorithm
Currently, rte_eth_rss_conf supports configuring and querying
RSS hash functions, rss key and it's length, but not RSS hash
algorithm.
The structure ``rte_eth_rss_conf`` is extended by adding a new
field "algorithm". This represents the RSS algorithms to apply.
The following API will be affected:
- rte_eth_dev_configure
- rte_eth_dev_rss_hash_update
- rte_eth_dev_rss_hash_conf_get
If the value of "algorithm" used for configuration is a gibberish
value, report the error and return. Do the same for
rte_eth_dev_rss_hash_update and rte_eth_dev_configure.
To check whether the drivers report valid "algorithm", it is set
to default value before querying.
Signed-off-by: Jie Hai
Signed-off-by: Dongdong Liu
---
doc/guides/rel_notes/release_23_11.rst | 2 ++
lib/ethdev/rte_ethdev.c| 17
lib/ethdev/rte_ethdev.h| 27 +
lib/ethdev/rte_flow.c | 1 -
lib/ethdev/rte_flow.h | 28 ++
5 files changed, 48 insertions(+), 27 deletions(-)
diff --git a/doc/guides/rel_notes/release_23_11.rst
b/doc/guides/rel_notes/release_23_11.rst
index e13d57728071..92a445ab2ed3 100644
--- a/doc/guides/rel_notes/release_23_11.rst
+++ b/doc/guides/rel_notes/release_23_11.rst
@@ -197,6 +197,8 @@ ABI Changes
fields, to move ``rxq`` and ``txq`` fields, to change the size of
``reserved1`` and ``reserved2`` fields.
+* ethdev: Added "algorithm" field to ``rte_eth_rss_conf`` structure for RSS
+ hash algorithm.
Known Issues
diff --git a/lib/ethdev/rte_ethdev.c b/lib/ethdev/rte_ethdev.c
index 18a4b950b184..2eda1b8072e5 100644
--- a/lib/ethdev/rte_ethdev.c
+++ b/lib/ethdev/rte_ethdev.c
@@ -1464,6 +1464,14 @@ rte_eth_dev_configure(uint16_t port_id, uint16_t
nb_rx_q, uint16_t nb_tx_q,
goto rollback;
}
+ if (dev_conf->rx_adv_conf.rss_conf.algorithm >=
RTE_ETH_HASH_FUNCTION_MAX) {
+ RTE_ETHDEV_LOG(ERR,
+ "Ethdev port_id=%u invalid RSS algorithm:
0x%"PRIx64"\n",
+ port_id, dev_conf->rx_adv_conf.rss_conf.algorithm);
+ ret = -EINVAL;
+ goto rollback;
+ }
+
/* Check if Rx RSS distribution is disabled but RSS hash is enabled. */
if (((dev_conf->rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) == 0) &&
(dev_conf->rxmode.offloads & RTE_ETH_RX_OFFLOAD_RSS_HASH)) {
@@ -4673,6 +4681,13 @@ rte_eth_dev_rss_hash_update(uint16_t port_id,
return -ENOTSUP;
}
+ if (rss_conf->algorithm >= RTE_ETH_HASH_FUNCTION_MAX) {
+ RTE_ETHDEV_LOG(ERR,
+ "Ethdev port_id=%u invalid RSS algorithm:
0x%"PRIx64"\n",
+ port_id, rss_conf->algorithm);
+ return -EINVAL;
+ }
+
if (*dev->dev_ops->rss_hash_update == NULL)
return -ENOTSUP;
ret = eth_err(port_id, (*dev->dev_ops->rss_hash_update)(dev,
@@ -4700,6 +4715,8 @@ rte_eth_dev_rss_hash_conf_get(uint16_t port_id,
return -EINVAL;
}
+ rss_conf->algorithm = RTE_ETH_HASH_FUNCTION_DEFAULT;
+
if (*dev->dev_ops->rss_hash_conf_get == NULL)
return -ENOTSUP;
ret = eth_err(port_id, (*dev->dev_ops->rss_hash_conf_get)(dev,
diff --git a/lib/ethdev/rte_ethdev.h b/lib/ethdev/rte_ethdev.h
index b9e4e21189d2..42c4250bd509 100644
--- a/lib/ethdev/rte_ethdev.h
+++ b/lib/ethdev/rte_ethdev.h
@@ -445,6 +445,32 @@ struct rte_vlan_filter_conf {
uint64_t ids[64];
};
+/**
+ * Hash function types.
+ */
+enum rte_eth_hash_function {
+ /**
+* DEFAULT means driver decides which hash algorithm to pick.
+*/
+ RTE_ETH_HASH_FUNCTION_DEFAULT = 0,
+ RTE_ETH_HASH_FUNCTION_TOEPLITZ, /**< Toeplitz */
+ RTE_ETH_HASH_FUNCTION_SIMPLE_XOR, /**< Simple XOR */
+ /**
+* Symmetric Toeplitz: src, dst will be replaced by
+* xor(src, dst). For the case with src/dst only,
+* src or dst address will xor with zero pair.
+*/
+ RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ,
+ /**
+* Symmetric Toeplitz: L3 and L4 fields are sorted prior to
+* the hash function.
+* If src_ip > dst_ip, swap src_ip and dst_ip.
+* If src_port > dst_port, swap src_port and dst_port.
+*/
+ RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ_SORT,
+ RTE_ETH_HASH_FUNCTION_MAX,
+};
+
/**
* A structure used to configure the Receive Side Scaling (RSS) feature
* of an Ethernet port.
@@ -465,6 +491,7 @@ struct rte_eth_rss_conf {
* Setting *rss_hf* to zero disables the RSS feature.
*/
uint64_t rss_hf;
+ enum rte_eth_hash_function algorithm; /**< Hash algorithm. */
};
/*
diff --git a/lib/ethdev/rte_flow.c b/lib/ethdev/rte_flow.c
index ba8bf27090fb..deedce08fb0a 100644
--- a/lib/ethdev/rte_flow.c
+++ b/lib/ethdev/rte_flow.c
[PATCH v5 05/40] net/bnx2x: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/bnx2x/bnx2x_ethdev.c | 4
1 file changed, 4 insertions(+)
diff --git a/drivers/net/bnx2x/bnx2x_ethdev.c b/drivers/net/bnx2x/bnx2x_ethdev.c
index 4448cf2de2d7..078d6db75d1b 100644
--- a/drivers/net/bnx2x/bnx2x_ethdev.c
+++ b/drivers/net/bnx2x/bnx2x_ethdev.c
@@ -196,6 +196,10 @@ bnx2x_dev_configure(struct rte_eth_dev *dev)
PMD_DRV_LOG(DEBUG, sc, "num_queues=%d, mtu=%d",
sc->num_queues, sc->mtu);
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* allocate ilt */
if (bnx2x_alloc_ilt_mem(sc) != 0) {
PMD_DRV_LOG(ERR, sc, "bnx2x_alloc_ilt_mem was failed");
--
2.30.0
[PATCH v5 06/40] net/bnxt: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/bnxt/bnxt_ethdev.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/bnxt/bnxt_ethdev.c b/drivers/net/bnxt/bnxt_ethdev.c
index ee1552452a11..92ce05b26dc7 100644
--- a/drivers/net/bnxt/bnxt_ethdev.c
+++ b/drivers/net/bnxt/bnxt_ethdev.c
@@ -1166,6 +1166,9 @@ static int bnxt_dev_configure_op(struct rte_eth_dev
*eth_dev)
rx_offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH;
eth_dev->data->dev_conf.rxmode.offloads = rx_offloads;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* application provides the hash key to program */
if (rss_conf->rss_key != NULL) {
if (rss_conf->rss_key_len != HW_HASH_KEY_SIZE)
@@ -2168,6 +2171,9 @@ static int bnxt_rss_hash_update_op(struct rte_eth_dev
*eth_dev,
return -EINVAL;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* Update the default RSS VNIC(s) */
vnic = bnxt_get_default_vnic(bp);
vnic->hash_type = bnxt_rte_to_hwrm_hash_types(rss_conf->rss_hf);
--
2.30.0
[PATCH v5 07/40] net/bonding: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/bonding/rte_eth_bond_pmd.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/bonding/rte_eth_bond_pmd.c
b/drivers/net/bonding/rte_eth_bond_pmd.c
index 122b1187fd9c..75ad0270af03 100644
--- a/drivers/net/bonding/rte_eth_bond_pmd.c
+++ b/drivers/net/bonding/rte_eth_bond_pmd.c
@@ -3187,6 +3187,9 @@ bond_ethdev_rss_hash_update(struct rte_eth_dev *dev,
if (bond_rss_conf.rss_hf != 0)
dev->data->dev_conf.rx_adv_conf.rss_conf.rss_hf =
bond_rss_conf.rss_hf;
+ if (bond_rss_conf.algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (bond_rss_conf.rss_key) {
if (bond_rss_conf.rss_key_len < internals->rss_key_len)
return -EINVAL;
@@ -3915,6 +3918,9 @@ bond_ethdev_configure(struct rte_eth_dev *dev)
struct rte_eth_rss_conf *rss_conf =
&dev->data->dev_conf.rx_adv_conf.rss_conf;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (internals->rss_key_len == 0) {
internals->rss_key_len = sizeof(default_rss_key);
}
--
2.30.0
[PATCH v5 08/40] net/cnxk: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/cnxk/cnxk_ethdev.c | 5 +
drivers/net/cnxk/cnxk_ethdev_ops.c | 3 +++
2 files changed, 8 insertions(+)
diff --git a/drivers/net/cnxk/cnxk_ethdev.c b/drivers/net/cnxk/cnxk_ethdev.c
index 01b707b6c4ac..dc150de745df 100644
--- a/drivers/net/cnxk/cnxk_ethdev.c
+++ b/drivers/net/cnxk/cnxk_ethdev.c
@@ -976,6 +976,10 @@ nix_rss_default_setup(struct cnxk_eth_dev *dev)
if (rss_hash_level)
rss_hash_level -= 1;
+ if (eth_dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
flowkey_cfg = cnxk_rss_ethdev_to_nix(dev, rss_hf, rss_hash_level);
return roc_nix_rss_default_setup(&dev->nix, flowkey_cfg);
}
@@ -1373,6 +1377,7 @@ cnxk_nix_configure(struct rte_eth_dev *eth_dev)
}
/* Configure RSS */
+
rc = nix_rss_default_setup(dev);
if (rc) {
plt_err("Failed to configure rss rc=%d", rc);
diff --git a/drivers/net/cnxk/cnxk_ethdev_ops.c
b/drivers/net/cnxk/cnxk_ethdev_ops.c
index 3ade8eed3626..b6cba99cbb7f 100644
--- a/drivers/net/cnxk/cnxk_ethdev_ops.c
+++ b/drivers/net/cnxk/cnxk_ethdev_ops.c
@@ -1054,6 +1054,9 @@ cnxk_nix_rss_hash_update(struct rte_eth_dev *eth_dev,
int rc = -EINVAL;
uint8_t alg_idx;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ goto fail;
+
if (rss_conf->rss_key && rss_conf->rss_key_len != ROC_NIX_RSS_KEY_LEN) {
plt_err("Hash key size mismatch %d vs %d",
rss_conf->rss_key_len, ROC_NIX_RSS_KEY_LEN);
--
2.30.0
[PATCH v5 10/40] net/cxgbe: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/cxgbe/cxgbe_ethdev.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/net/cxgbe/cxgbe_ethdev.c b/drivers/net/cxgbe/cxgbe_ethdev.c
index 45bbeaef0ceb..8de57bbfe661 100644
--- a/drivers/net/cxgbe/cxgbe_ethdev.c
+++ b/drivers/net/cxgbe/cxgbe_ethdev.c
@@ -440,9 +440,13 @@ int cxgbe_dev_configure(struct rte_eth_dev *eth_dev)
CXGBE_FUNC_TRACE();
- if (eth_dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (eth_dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
eth_dev->data->dev_conf.rxmode.offloads |=
RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ if (eth_dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+ }
if (!(adapter->flags & FW_QUEUE_BOUND)) {
err = cxgbe_setup_sge_fwevtq(adapter);
@@ -1165,6 +1169,9 @@ static int cxgbe_dev_rss_hash_update(struct rte_eth_dev
*dev,
struct adapter *adapter = pi->adapter;
int err;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
err = cxgbe_write_rss_conf(pi, rss_conf->rss_hf);
if (err)
return err;
--
2.30.0
[PATCH v5 09/40] net/cpfl: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/cpfl/cpfl_ethdev.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/cpfl/cpfl_ethdev.c b/drivers/net/cpfl/cpfl_ethdev.c
index c4ca9343c3e0..6acb6ce9fd22 100644
--- a/drivers/net/cpfl/cpfl_ethdev.c
+++ b/drivers/net/cpfl/cpfl_ethdev.c
@@ -450,6 +450,9 @@ cpfl_init_rss(struct idpf_vport *vport)
rss_conf = &dev_data->dev_conf.rx_adv_conf.rss_conf;
nb_q = dev_data->nb_rx_queues;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_key == NULL) {
for (i = 0; i < vport->rss_key_size; i++)
vport->rss_key[i] = (uint8_t)rte_rand();
@@ -568,6 +571,9 @@ cpfl_rss_hash_update(struct rte_eth_dev *dev,
return -ENOTSUP;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (!rss_conf->rss_key || rss_conf->rss_key_len == 0) {
PMD_DRV_LOG(DEBUG, "No key to be configured");
goto skip_rss_key;
--
2.30.0
[PATCH v5 11/40] net/dpaa: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/dpaa/dpaa_ethdev.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/net/dpaa/dpaa_ethdev.c b/drivers/net/dpaa/dpaa_ethdev.c
index a6c86113d125..c4d5de9ba21d 100644
--- a/drivers/net/dpaa/dpaa_ethdev.c
+++ b/drivers/net/dpaa/dpaa_ethdev.c
@@ -258,6 +258,10 @@ dpaa_eth_dev_configure(struct rte_eth_dev *dev)
}
if (!(default_q || fmc_q)) {
+ if (eth_conf->rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (dpaa_fm_config(dev,
eth_conf->rx_adv_conf.rss_conf.rss_hf)) {
dpaa_write_fm_config_to_file();
@@ -1446,6 +1450,9 @@ dpaa_dev_rss_hash_update(struct rte_eth_dev *dev,
PMD_INIT_FUNC_TRACE();
if (!(default_q || fmc_q)) {
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (dpaa_fm_config(dev, rss_conf->rss_hf)) {
DPAA_PMD_ERR("FM port configuration: Failed\n");
return -1;
--
2.30.0
[PATCH v5 12/40] net/dpaa2: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/dpaa2/dpaa2_ethdev.c | 7 +++
1 file changed, 7 insertions(+)
diff --git a/drivers/net/dpaa2/dpaa2_ethdev.c b/drivers/net/dpaa2/dpaa2_ethdev.c
index 679f33ae1a08..355c94b6112f 100644
--- a/drivers/net/dpaa2/dpaa2_ethdev.c
+++ b/drivers/net/dpaa2/dpaa2_ethdev.c
@@ -583,6 +583,10 @@ dpaa2_eth_dev_configure(struct rte_eth_dev *dev)
}
if (eth_conf->rxmode.mq_mode == RTE_ETH_MQ_RX_RSS) {
+ if (eth_conf->rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
for (tc_index = 0; tc_index < priv->num_rx_tc; tc_index++) {
ret = dpaa2_setup_flow_dist(dev,
eth_conf->rx_adv_conf.rss_conf.rss_hf,
@@ -2194,6 +2198,9 @@ dpaa2_dev_rss_hash_update(struct rte_eth_dev *dev,
PMD_INIT_FUNC_TRACE();
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_hf) {
for (tc_index = 0; tc_index < priv->num_rx_tc; tc_index++) {
ret = dpaa2_setup_flow_dist(dev, rss_conf->rss_hf,
--
2.30.0
[PATCH v5 13/40] net/ena: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/ena/ena_rss.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/ena/ena_rss.c b/drivers/net/ena/ena_rss.c
index d0ba9d5c0a14..06aff9f3bd49 100644
--- a/drivers/net/ena/ena_rss.c
+++ b/drivers/net/ena/ena_rss.c
@@ -398,6 +398,9 @@ static int ena_rss_hash_set(struct ena_com_dev *ena_dev,
uint8_t *rss_key;
int rc;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_key != NULL) {
/* Reorder the RSS key bytes for the hardware requirements. */
ena_reorder_rss_hash_key(hw_rss_key, rss_conf->rss_key,
--
2.30.0
[PATCH v5 14/40] net/enic: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/enic/enic_ethdev.c | 1 +
drivers/net/enic/enic_main.c | 3 +++
2 files changed, 4 insertions(+)
diff --git a/drivers/net/enic/enic_ethdev.c b/drivers/net/enic/enic_ethdev.c
index cdf091559196..164f423a85c8 100644
--- a/drivers/net/enic/enic_ethdev.c
+++ b/drivers/net/enic/enic_ethdev.c
@@ -834,6 +834,7 @@ static int enicpmd_dev_rss_hash_conf_get(struct rte_eth_dev
*dev,
ENICPMD_FUNC_TRACE();
if (rss_conf == NULL)
return -EINVAL;
+
if (rss_conf->rss_key != NULL &&
rss_conf->rss_key_len < ENIC_RSS_HASH_KEY_SIZE) {
dev_err(enic, "rss_hash_conf_get: wrong rss_key_len. given=%u"
diff --git a/drivers/net/enic/enic_main.c b/drivers/net/enic/enic_main.c
index 19a99a82c501..2eafe7637b3a 100644
--- a/drivers/net/enic/enic_main.c
+++ b/drivers/net/enic/enic_main.c
@@ -1428,6 +1428,9 @@ int enic_set_rss_conf(struct enic *enic, struct
rte_eth_rss_conf *rss_conf)
}
}
+ if (rss_enable && rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
ret = enic_set_niccfg(enic, ENIC_RSS_DEFAULT_CPU, rss_hash_type,
ENIC_RSS_HASH_BITS, ENIC_RSS_BASE_CPU,
rss_enable);
--
2.30.0
[PATCH v5 15/40] net/fm10k: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/fm10k/fm10k_ethdev.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/net/fm10k/fm10k_ethdev.c b/drivers/net/fm10k/fm10k_ethdev.c
index 4d3c4c10cfa4..d8b27e84d836 100644
--- a/drivers/net/fm10k/fm10k_ethdev.c
+++ b/drivers/net/fm10k/fm10k_ethdev.c
@@ -452,8 +452,12 @@ fm10k_dev_configure(struct rte_eth_dev *dev)
PMD_INIT_FUNC_TRACE();
- if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
dev->data->dev_conf.rxmode.offloads |=
RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+ }
/* multiple queue mode checking */
ret = fm10k_check_mq_mode(dev);
@@ -2195,6 +2199,9 @@ fm10k_rss_hash_update(struct rte_eth_dev *dev,
FM10K_RSSRK_ENTRIES_PER_REG))
return -EINVAL;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (hf == 0)
return -EINVAL;
--
2.30.0
[PATCH v5 16/40] net/hinic: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/hinic/hinic_pmd_ethdev.c | 3 +++
drivers/net/hinic/hinic_pmd_rx.c | 3 +++
2 files changed, 6 insertions(+)
diff --git a/drivers/net/hinic/hinic_pmd_ethdev.c
b/drivers/net/hinic/hinic_pmd_ethdev.c
index 7aa5e7d8e929..e818dc939073 100644
--- a/drivers/net/hinic/hinic_pmd_ethdev.c
+++ b/drivers/net/hinic/hinic_pmd_ethdev.c
@@ -1937,6 +1937,9 @@ static int hinic_rss_hash_update(struct rte_eth_dev *dev,
return HINIC_OK;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return HINIC_ERROR;
+
if (rss_conf->rss_key_len > HINIC_RSS_KEY_SIZE) {
PMD_DRV_LOG(ERR, "Invalid rss key, rss_key_len: %d",
rss_conf->rss_key_len);
diff --git a/drivers/net/hinic/hinic_pmd_rx.c b/drivers/net/hinic/hinic_pmd_rx.c
index 7adb6e365993..7736b61c0b0f 100644
--- a/drivers/net/hinic/hinic_pmd_rx.c
+++ b/drivers/net/hinic/hinic_pmd_rx.c
@@ -670,6 +670,9 @@ int hinic_rx_configure(struct rte_eth_dev *dev)
goto rss_config_err;
}
+ if (rss_conf.algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ goto rss_config_err;
+
err = hinic_rss_init(nic_dev, NULL, &rss_conf);
if (err) {
PMD_DRV_LOG(ERR, "Init rss failed");
--
2.30.0
[PATCH v5 17/40] net/i40e: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it in case of ignoring unsupported values. Signed-off-by: Jie Hai --- drivers/net/i40e/i40e_ethdev.c | 7 +++ 1 file changed, 7 insertions(+) diff --git a/drivers/net/i40e/i40e_ethdev.c b/drivers/net/i40e/i40e_ethdev.c index 50ba9aac9498..fcd6d4eb0495 100644 --- a/drivers/net/i40e/i40e_ethdev.c +++ b/drivers/net/i40e/i40e_ethdev.c @@ -7749,6 +7749,9 @@ i40e_dev_rss_hash_update(struct rte_eth_dev *dev, if (rss_hf == 0) /* Disable RSS */ return -EINVAL; + if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT) + return -EINVAL; + return i40e_hw_rss_hash_set(pf, rss_conf); } @@ -8986,6 +8989,10 @@ i40e_pf_config_rss(struct i40e_pf *pf) !(mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)) return 0; + if (pf->dev_data->dev_conf.rx_adv_conf.rss_conf.algorithm != + RTE_ETH_HASH_FUNCTION_DEFAULT) + return -EINVAL; + hw = I40E_PF_TO_HW(pf); hena = i40e_config_hena(pf->adapter, rss_hf); i40e_write_rx_ctl(hw, I40E_PFQF_HENA(0), (uint32_t)hena); -- 2.30.0
[PATCH v5 18/40] net/iavf: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/iavf/iavf_ethdev.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/iavf/iavf_ethdev.c b/drivers/net/iavf/iavf_ethdev.c
index f2fc5a56216d..9d6950d462a6 100644
--- a/drivers/net/iavf/iavf_ethdev.c
+++ b/drivers/net/iavf/iavf_ethdev.c
@@ -523,6 +523,9 @@ iavf_init_rss(struct iavf_adapter *adapter)
return -ENOTSUP;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* configure RSS key */
if (!rss_conf->rss_key) {
/* Calculate the default hash key */
@@ -1588,6 +1591,9 @@ iavf_dev_rss_hash_update(struct rte_eth_dev *dev,
if (!(vf->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RSS_PF))
return -ENOTSUP;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* Set hash key. */
ret = iavf_set_rss_key(adapter, rss_conf->rss_key,
rss_conf->rss_key_len);
--
2.30.0
[PATCH v5 19/40] net/ice: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/ice/ice_dcf.c| 3 +++
drivers/net/ice/ice_dcf_ethdev.c | 3 +++
drivers/net/ice/ice_ethdev.c | 7 +++
3 files changed, 13 insertions(+)
diff --git a/drivers/net/ice/ice_dcf.c b/drivers/net/ice/ice_dcf.c
index 7f8f5163acef..bff39cf9d145 100644
--- a/drivers/net/ice/ice_dcf.c
+++ b/drivers/net/ice/ice_dcf.c
@@ -1120,6 +1120,9 @@ ice_dcf_init_rss(struct ice_dcf_hw *hw)
return ice_dcf_configure_rss_lut(hw);
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* In IAVF, RSS enablement is set by PF driver. It is not supported
* to set based on rss_conf->rss_hf.
*/
diff --git a/drivers/net/ice/ice_dcf_ethdev.c b/drivers/net/ice/ice_dcf_ethdev.c
index 30ad18d8fc20..f0db68c5dee5 100644
--- a/drivers/net/ice/ice_dcf_ethdev.c
+++ b/drivers/net/ice/ice_dcf_ethdev.c
@@ -1405,6 +1405,9 @@ ice_dcf_dev_rss_hash_update(struct rte_eth_dev *dev,
if (!(hw->vf_res->vf_cap_flags & VIRTCHNL_VF_OFFLOAD_RSS_PF))
return -ENOTSUP;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* HENA setting, it is enabled by default, no change */
if (!rss_conf->rss_key || rss_conf->rss_key_len == 0) {
PMD_DRV_LOG(DEBUG, "No key to be configured");
diff --git a/drivers/net/ice/ice_ethdev.c b/drivers/net/ice/ice_ethdev.c
index 4bad39c2c1c9..0cba6f7d7b6a 100644
--- a/drivers/net/ice/ice_ethdev.c
+++ b/drivers/net/ice/ice_ethdev.c
@@ -3383,6 +3383,10 @@ static int ice_init_rss(struct ice_pf *pf)
return -ENOMEM;
}
}
+
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* configure RSS key */
if (!rss_conf->rss_key)
ice_get_default_rss_key(vsi->rss_key, vsi->rss_key_size);
@@ -5046,6 +5050,9 @@ ice_rss_hash_update(struct rte_eth_dev *dev,
struct ice_pf *pf = ICE_DEV_PRIVATE_TO_PF(dev->data->dev_private);
struct ice_vsi *vsi = pf->main_vsi;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
/* set hash key */
status = ice_set_rss_key(vsi, rss_conf->rss_key, rss_conf->rss_key_len);
if (status)
--
2.30.0
[PATCH v5 20/40] net/idpf: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/idpf/idpf_ethdev.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/idpf/idpf_ethdev.c b/drivers/net/idpf/idpf_ethdev.c
index 3af7cf0bb7e0..013db04ac8fc 100644
--- a/drivers/net/idpf/idpf_ethdev.c
+++ b/drivers/net/idpf/idpf_ethdev.c
@@ -426,6 +426,9 @@ idpf_init_rss(struct idpf_vport *vport)
rss_conf = &dev_data->dev_conf.rx_adv_conf.rss_conf;
nb_q = dev_data->nb_rx_queues;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_key == NULL) {
for (i = 0; i < vport->rss_key_size; i++)
vport->rss_key[i] = (uint8_t)rte_rand();
@@ -541,6 +544,9 @@ idpf_rss_hash_update(struct rte_eth_dev *dev,
return -ENOTSUP;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (!rss_conf->rss_key || rss_conf->rss_key_len == 0) {
PMD_DRV_LOG(DEBUG, "No key to be configured");
goto skip_rss_key;
--
2.30.0
[PATCH v5 21/40] net/igc: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/igc/igc_ethdev.c | 4
drivers/net/igc/igc_txrx.c | 5 +
2 files changed, 9 insertions(+)
diff --git a/drivers/net/igc/igc_ethdev.c b/drivers/net/igc/igc_ethdev.c
index 58c4f8092772..11c0f5ff231b 100644
--- a/drivers/net/igc/igc_ethdev.c
+++ b/drivers/net/igc/igc_ethdev.c
@@ -2442,6 +2442,10 @@ eth_igc_rss_hash_update(struct rte_eth_dev *dev,
struct rte_eth_rss_conf *rss_conf)
{
struct igc_hw *hw = IGC_DEV_PRIVATE_HW(dev);
+
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
igc_hw_rss_hash_set(hw, rss_conf);
return 0;
}
diff --git a/drivers/net/igc/igc_txrx.c b/drivers/net/igc/igc_txrx.c
index 5c60e3e99709..5e62e00d2ad9 100644
--- a/drivers/net/igc/igc_txrx.c
+++ b/drivers/net/igc/igc_txrx.c
@@ -818,6 +818,7 @@ igc_rss_configure(struct rte_eth_dev *dev)
rss_conf = dev->data->dev_conf.rx_adv_conf.rss_conf;
if (rss_conf.rss_key == NULL)
rss_conf.rss_key = default_rss_key;
+
igc_hw_rss_hash_set(hw, &rss_conf);
}
@@ -958,6 +959,10 @@ igc_dev_mq_rx_configure(struct rte_eth_dev *dev)
return -EINVAL;
}
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
switch (dev->data->dev_conf.rxmode.mq_mode) {
case RTE_ETH_MQ_RX_RSS:
igc_rss_configure(dev);
--
2.30.0
[PATCH v5 22/40] net/ionic: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/ionic/ionic_ethdev.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/ionic/ionic_ethdev.c b/drivers/net/ionic/ionic_ethdev.c
index 340fd0cd5923..e2e4d23c069b 100644
--- a/drivers/net/ionic/ionic_ethdev.c
+++ b/drivers/net/ionic/ionic_ethdev.c
@@ -642,6 +642,9 @@ ionic_dev_rss_hash_update(struct rte_eth_dev *eth_dev,
if (rss_conf->rss_key)
key = rss_conf->rss_key;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if ((rss_conf->rss_hf & IONIC_ETH_RSS_OFFLOAD_ALL) == 0) {
/*
* Can't disable rss through hash flags,
@@ -826,6 +829,9 @@ ionic_dev_configure(struct rte_eth_dev *eth_dev)
IONIC_PRINT_CALL();
+ if (eth_dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
ionic_lif_configure(lif);
return 0;
--
2.30.0
[PATCH v5 23/40] net/ixgbe: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/ixgbe/ixgbe_ethdev.c | 12 ++--
drivers/net/ixgbe/ixgbe_rxtx.c | 4
2 files changed, 14 insertions(+), 2 deletions(-)
diff --git a/drivers/net/ixgbe/ixgbe_ethdev.c b/drivers/net/ixgbe/ixgbe_ethdev.c
index d6cf00317e77..b92cd746a061 100644
--- a/drivers/net/ixgbe/ixgbe_ethdev.c
+++ b/drivers/net/ixgbe/ixgbe_ethdev.c
@@ -2434,8 +2434,12 @@ ixgbe_dev_configure(struct rte_eth_dev *dev)
PMD_INIT_FUNC_TRACE();
- if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
dev->data->dev_conf.rxmode.offloads |=
RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+ }
/* multiple queue mode checking */
ret = ixgbe_check_mq_mode(dev);
@@ -5326,8 +5330,12 @@ ixgbevf_dev_configure(struct rte_eth_dev *dev)
PMD_INIT_LOG(DEBUG, "Configured Virtual Function port id: %d",
dev->data->port_id);
- if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
dev->data->dev_conf.rxmode.offloads |=
RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+ }
/*
* VF has no ability to enable/disable HW CRC
diff --git a/drivers/net/ixgbe/ixgbe_rxtx.c b/drivers/net/ixgbe/ixgbe_rxtx.c
index 90b0a7004f50..150f8065ab60 100644
--- a/drivers/net/ixgbe/ixgbe_rxtx.c
+++ b/drivers/net/ixgbe/ixgbe_rxtx.c
@@ -3631,6 +3631,10 @@ ixgbe_dev_rss_hash_update(struct rte_eth_dev *dev,
/* RSS enabled */
if (rss_hf == 0) /* Disable RSS */
return -(EINVAL);
+
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
ixgbe_hw_rss_hash_set(hw, rss_conf);
return 0;
}
--
2.30.0
[PATCH v5 24/40] net/mana: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/mana/mana.c | 11 +--
1 file changed, 9 insertions(+), 2 deletions(-)
diff --git a/drivers/net/mana/mana.c b/drivers/net/mana/mana.c
index 41c3cf259584..e3055bb545d9 100644
--- a/drivers/net/mana/mana.c
+++ b/drivers/net/mana/mana.c
@@ -80,8 +80,12 @@ mana_dev_configure(struct rte_eth_dev *dev)
struct mana_priv *priv = dev->data->dev_private;
struct rte_eth_conf *dev_conf = &dev->data->dev_conf;
- if (dev_conf->rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
- dev_conf->rxmode.offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ if (dev_conf->rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
+ dev->data->dev_conf.rxmode.offloads |=
RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+ }
if (dev->data->nb_rx_queues != dev->data->nb_tx_queues) {
DRV_LOG(ERR, "Only support equal number of rx/tx queues");
@@ -413,6 +417,9 @@ mana_rss_hash_update(struct rte_eth_dev *dev,
return -ENODEV;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_hf & ~MANA_ETH_RSS_SUPPORT) {
DRV_LOG(ERR, "Port %u invalid RSS HF 0x%" PRIx64,
dev->data->port_id, rss_conf->rss_hf);
--
2.30.0
[PATCH v5 26/40] net/mvpp2: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/mvpp2/mrvl_ethdev.c | 3 +++
1 file changed, 3 insertions(+)
diff --git a/drivers/net/mvpp2/mrvl_ethdev.c b/drivers/net/mvpp2/mrvl_ethdev.c
index 359a5d1df7ab..817153af2ef1 100644
--- a/drivers/net/mvpp2/mrvl_ethdev.c
+++ b/drivers/net/mvpp2/mrvl_ethdev.c
@@ -440,6 +440,9 @@ mrvl_configure_rss(struct mrvl_priv *priv, struct
rte_eth_rss_conf *rss_conf)
if (rss_conf->rss_key)
MRVL_LOG(WARNING, "Changing hash key is not supported");
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_hf == 0) {
priv->ppio_params.inqs_params.hash_type = PP2_PPIO_HASH_T_NONE;
} else if (rss_conf->rss_hf & RTE_ETH_RSS_IPV4) {
--
2.30.0
[PATCH v5 25/40] net/mlx5: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/mlx5/mlx5_ethdev.c | 4
drivers/net/mlx5/mlx5_rss.c| 3 ++-
2 files changed, 6 insertions(+), 1 deletion(-)
diff --git a/drivers/net/mlx5/mlx5_ethdev.c b/drivers/net/mlx5/mlx5_ethdev.c
index 4a85415ff38d..5437fa531738 100644
--- a/drivers/net/mlx5/mlx5_ethdev.c
+++ b/drivers/net/mlx5/mlx5_ethdev.c
@@ -74,6 +74,10 @@ mlx5_dev_configure(struct rte_eth_dev *dev)
!!dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key;
int ret = 0;
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (use_app_rss_key &&
(dev->data->dev_conf.rx_adv_conf.rss_conf.rss_key_len !=
MLX5_RSS_HASH_KEY_LEN)) {
diff --git a/drivers/net/mlx5/mlx5_rss.c b/drivers/net/mlx5/mlx5_rss.c
index e2b478b75ce7..25c410e6010a 100644
--- a/drivers/net/mlx5/mlx5_rss.c
+++ b/drivers/net/mlx5/mlx5_rss.c
@@ -37,7 +37,8 @@ mlx5_rss_hash_update(struct rte_eth_dev *dev,
unsigned int i;
unsigned int idx;
- if (rss_conf->rss_hf & MLX5_RSS_HF_MASK) {
+ if (rss_conf->rss_hf & MLX5_RSS_HF_MASK ||
+ rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT) {
rte_errno = EINVAL;
return -rte_errno;
}
--
2.30.0
[PATCH v5 27/40] net/netvsc: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/netvsc/hn_ethdev.c | 6 ++
1 file changed, 6 insertions(+)
diff --git a/drivers/net/netvsc/hn_ethdev.c b/drivers/net/netvsc/hn_ethdev.c
index d0bbc0a4c0c0..9e4205816a6f 100644
--- a/drivers/net/netvsc/hn_ethdev.c
+++ b/drivers/net/netvsc/hn_ethdev.c
@@ -395,6 +395,9 @@ static int hn_rss_hash_update(struct rte_eth_dev *dev,
return err;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
hn_rss_hash_init(hv, rss_conf);
if (rss_conf->rss_hf != 0) {
@@ -756,6 +759,9 @@ static int hn_dev_configure(struct rte_eth_dev *dev)
for (i = 0; i < NDIS_HASH_INDCNT; i++)
hv->rss_ind[i] = i % dev->data->nb_rx_queues;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
hn_rss_hash_init(hv, rss_conf);
subchan = hv->num_queues - 1;
--
2.30.0
[PATCH v5 28/40] net/ngbe: : check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/ngbe/ngbe_ethdev.c | 6 +-
drivers/net/ngbe/ngbe_rxtx.c | 3 +++
2 files changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/net/ngbe/ngbe_ethdev.c b/drivers/net/ngbe/ngbe_ethdev.c
index 478da014b2f8..bb4b8afb0a80 100644
--- a/drivers/net/ngbe/ngbe_ethdev.c
+++ b/drivers/net/ngbe/ngbe_ethdev.c
@@ -921,8 +921,12 @@ ngbe_dev_configure(struct rte_eth_dev *dev)
PMD_INIT_FUNC_TRACE();
- if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
dev->data->dev_conf.rxmode.offloads |=
RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ }
/* set flag to update link status after init */
intr->flags |= NGBE_FLAG_NEED_LINK_UPDATE;
diff --git a/drivers/net/ngbe/ngbe_rxtx.c b/drivers/net/ngbe/ngbe_rxtx.c
index f31906cc2fe3..0b22d567fcb6 100644
--- a/drivers/net/ngbe/ngbe_rxtx.c
+++ b/drivers/net/ngbe/ngbe_rxtx.c
@@ -2500,6 +2500,9 @@ ngbe_dev_rss_hash_update(struct rte_eth_dev *dev,
return -ENOTSUP;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
hash_key = rss_conf->rss_key;
if (hash_key) {
/* Fill in RSS hash key */
--
2.30.0
[PATCH v5 30/40] net/null: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/null/rte_eth_null.c | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/net/null/rte_eth_null.c b/drivers/net/null/rte_eth_null.c
index 31081af79752..8427d7484178 100644
--- a/drivers/net/null/rte_eth_null.c
+++ b/drivers/net/null/rte_eth_null.c
@@ -186,6 +186,11 @@ eth_null_copy_tx(void *q, struct rte_mbuf **bufs, uint16_t
nb_bufs)
static int
eth_dev_configure(struct rte_eth_dev *dev __rte_unused)
{
+ if ((dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) &&
+dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
return 0;
}
@@ -444,6 +449,9 @@ eth_rss_hash_update(struct rte_eth_dev *dev, struct
rte_eth_rss_conf *rss_conf)
rte_spinlock_lock(&internal->rss_lock);
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if ((rss_conf->rss_hf & internal->flow_type_rss_offloads) != 0)
dev->data->dev_conf.rx_adv_conf.rss_conf.rss_hf =
rss_conf->rss_hf &
internal->flow_type_rss_offloads;
--
2.30.0
[PATCH v5 31/40] net/qede: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/qede/qede_ethdev.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/net/qede/qede_ethdev.c b/drivers/net/qede/qede_ethdev.c
index 22cd4706467a..7db69e8f35e0 100644
--- a/drivers/net/qede/qede_ethdev.c
+++ b/drivers/net/qede/qede_ethdev.c
@@ -1272,8 +1272,12 @@ static int qede_dev_configure(struct rte_eth_dev
*eth_dev)
PMD_INIT_FUNC_TRACE(edev);
- if (rxmode->mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (rxmode->mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
+ if (eth_dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
rxmode->offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ }
/* We need to have min 1 RX queue.There is no min check in
* rte_eth_dev_configure(), so we are checking it here.
@@ -2119,6 +2123,9 @@ int qede_rss_hash_update(struct rte_eth_dev *eth_dev,
DP_INFO(edev, "RSS hf = 0x%lx len = %u key = %p\n",
(unsigned long)hf, len, key);
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (hf != 0) {
/* Enabling RSS */
DP_INFO(edev, "Enabling rss\n");
--
2.30.0
[PATCH v5 29/40] net/nfp: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/nfp/nfp_common.c | 9 -
1 file changed, 8 insertions(+), 1 deletion(-)
diff --git a/drivers/net/nfp/nfp_common.c b/drivers/net/nfp/nfp_common.c
index 5683afc40a31..780316c764df 100644
--- a/drivers/net/nfp/nfp_common.c
+++ b/drivers/net/nfp/nfp_common.c
@@ -390,8 +390,12 @@ nfp_net_configure(struct rte_eth_dev *dev)
rxmode = &dev_conf->rxmode;
txmode = &dev_conf->txmode;
- if (rxmode->mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (rxmode->mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
rxmode->offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ if (dev_conf->rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+ }
/* Checking TX mode */
if (txmode->mq_mode) {
@@ -1805,6 +1809,9 @@ nfp_net_rss_hash_update(struct rte_eth_dev *dev,
return 0; /* Nothing to do */
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_key_len > NFP_NET_CFG_RSS_KEY_SZ) {
PMD_DRV_LOG(ERR, "hash key too long");
return -EINVAL;
--
2.30.0
[PATCH v5 32/40] net/sfc: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/sfc/sfc_ethdev.c | 3 +++
drivers/net/sfc/sfc_rx.c | 3 +++
2 files changed, 6 insertions(+)
diff --git a/drivers/net/sfc/sfc_ethdev.c b/drivers/net/sfc/sfc_ethdev.c
index 1efe64a36a7f..19d7761e2664 100644
--- a/drivers/net/sfc/sfc_ethdev.c
+++ b/drivers/net/sfc/sfc_ethdev.c
@@ -1728,6 +1728,9 @@ sfc_dev_rss_hash_update(struct rte_eth_dev *dev,
goto fail_scale_mode_set;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_key != NULL) {
if (sa->state == SFC_ETHDEV_STARTED) {
for (key_i = 0; key_i < n_contexts; key_i++) {
diff --git a/drivers/net/sfc/sfc_rx.c b/drivers/net/sfc/sfc_rx.c
index 1dde2c111001..71c151a07162 100644
--- a/drivers/net/sfc/sfc_rx.c
+++ b/drivers/net/sfc/sfc_rx.c
@@ -1519,6 +1519,9 @@ sfc_rx_process_adv_conf_rss(struct sfc_adapter *sa,
return rc;
}
+ if (conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (conf->rss_key != NULL) {
if (conf->rss_key_len != sizeof(rss->key)) {
sfc_err(sa, "RSS key size is wrong (should be %zu)",
--
2.30.0
[PATCH v5 33/40] net/tap: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/tap/rte_eth_tap.c | 8
1 file changed, 8 insertions(+)
diff --git a/drivers/net/tap/rte_eth_tap.c b/drivers/net/tap/rte_eth_tap.c
index b25a52655fa2..5e4813637f0b 100644
--- a/drivers/net/tap/rte_eth_tap.c
+++ b/drivers/net/tap/rte_eth_tap.c
@@ -1038,6 +1038,10 @@ tap_dev_configure(struct rte_eth_dev *dev)
TAP_LOG(INFO, "%s: %s: RX configured queues number: %u",
dev->device->name, pmd->name, dev->data->nb_rx_queues);
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
return 0;
}
@@ -1894,6 +1898,10 @@ tap_rss_hash_update(struct rte_eth_dev *dev,
rte_errno = EINVAL;
return -rte_errno;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT) {
+ rte_errno = EINVAL;
+ return -rte_errno;
+ }
if (rss_conf->rss_key && rss_conf->rss_key_len) {
/*
* Currently TAP RSS key is hard coded
--
2.30.0
[PATCH v5 34/40] net/thunderx: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/thunderx/nicvf_ethdev.c | 10 +-
1 file changed, 9 insertions(+), 1 deletion(-)
diff --git a/drivers/net/thunderx/nicvf_ethdev.c
b/drivers/net/thunderx/nicvf_ethdev.c
index ab1e714d9767..2fc54c521c88 100644
--- a/drivers/net/thunderx/nicvf_ethdev.c
+++ b/drivers/net/thunderx/nicvf_ethdev.c
@@ -621,6 +621,9 @@ nicvf_dev_rss_hash_update(struct rte_eth_dev *dev,
struct nicvf *nic = nicvf_pmd_priv(dev);
uint64_t nic_rss;
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
if (rss_conf->rss_key &&
rss_conf->rss_key_len != RSS_HASH_KEY_BYTE_SIZE) {
PMD_DRV_LOG(ERR, "Hash key size mismatch %u",
@@ -1984,8 +1987,13 @@ nicvf_dev_configure(struct rte_eth_dev *dev)
PMD_INIT_FUNC_TRACE();
- if (rxmode->mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (rxmode->mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
+ if (conf->rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
rxmode->offloads |= RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ }
if (!rte_eal_has_hugepages()) {
PMD_INIT_LOG(INFO, "Huge page is not configured");
--
2.30.0
[PATCH v5 35/40] net/txgbe: check RSS hash algorithms
A new field 'algorithm' has been added to rss_conf, check it
in case of ignoring unsupported values.
Signed-off-by: Jie Hai
---
drivers/net/txgbe/txgbe_ethdev.c| 7 ++-
drivers/net/txgbe/txgbe_ethdev_vf.c | 7 ++-
drivers/net/txgbe/txgbe_rxtx.c | 3 +++
3 files changed, 15 insertions(+), 2 deletions(-)
diff --git a/drivers/net/txgbe/txgbe_ethdev.c b/drivers/net/txgbe/txgbe_ethdev.c
index 6bc231a13063..b0e6ea6d171b 100644
--- a/drivers/net/txgbe/txgbe_ethdev.c
+++ b/drivers/net/txgbe/txgbe_ethdev.c
@@ -1525,8 +1525,13 @@ txgbe_dev_configure(struct rte_eth_dev *dev)
PMD_INIT_FUNC_TRACE();
- if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
dev->data->dev_conf.rxmode.offloads |=
RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ }
/* multiple queue mode checking */
ret = txgbe_check_mq_mode(dev);
diff --git a/drivers/net/txgbe/txgbe_ethdev_vf.c
b/drivers/net/txgbe/txgbe_ethdev_vf.c
index f1341fbf7e22..b0bc8441ac12 100644
--- a/drivers/net/txgbe/txgbe_ethdev_vf.c
+++ b/drivers/net/txgbe/txgbe_ethdev_vf.c
@@ -579,8 +579,13 @@ txgbevf_dev_configure(struct rte_eth_dev *dev)
PMD_INIT_LOG(DEBUG, "Configured Virtual Function port id: %d",
dev->data->port_id);
- if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG)
+ if (dev->data->dev_conf.rxmode.mq_mode & RTE_ETH_MQ_RX_RSS_FLAG) {
+ if (dev->data->dev_conf.rx_adv_conf.rss_conf.algorithm !=
+ RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
dev->data->dev_conf.rxmode.offloads |=
RTE_ETH_RX_OFFLOAD_RSS_HASH;
+ }
/*
* VF has no ability to enable/disable HW CRC
diff --git a/drivers/net/txgbe/txgbe_rxtx.c b/drivers/net/txgbe/txgbe_rxtx.c
index f7cd2333abee..3ffc2f4a57a3 100644
--- a/drivers/net/txgbe/txgbe_rxtx.c
+++ b/drivers/net/txgbe/txgbe_rxtx.c
@@ -2894,6 +2894,9 @@ txgbe_dev_rss_hash_update(struct rte_eth_dev *dev,
return -ENOTSUP;
}
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ return -EINVAL;
+
hash_key = rss_conf->rss_key;
if (hash_key) {
/* Fill in RSS hash key */
--
2.30.0
[PATCH v5 36/40] net/hns3: support setting and querying RSS hash function
From: Huisong Li
Support setting and querying RSS hash function by ethdev ops.
Signed-off-by: Huisong Li
Signed-off-by: Dongdong Liu
---
drivers/net/hns3/hns3_rss.c | 47 +
1 file changed, 27 insertions(+), 20 deletions(-)
diff --git a/drivers/net/hns3/hns3_rss.c b/drivers/net/hns3/hns3_rss.c
index 6126512bd780..010a759f23d9 100644
--- a/drivers/net/hns3/hns3_rss.c
+++ b/drivers/net/hns3/hns3_rss.c
@@ -646,14 +646,14 @@ hns3_dev_rss_hash_update(struct rte_eth_dev *dev,
if (ret)
goto set_tuple_fail;
- if (key) {
- ret = hns3_rss_set_algo_key(hw, hw->rss_info.hash_algo,
- key, hw->rss_key_size);
- if (ret)
- goto set_algo_key_fail;
- /* Update the shadow RSS key with user specified */
+ ret = hns3_update_rss_algo_key(hw, rss_conf->algorithm, key, key_len);
+ if (ret != 0)
+ goto set_algo_key_fail;
+
+ if (rss_conf->algorithm != RTE_ETH_HASH_FUNCTION_DEFAULT)
+ hw->rss_info.hash_algo =
hns3_hash_func_map[rss_conf->algorithm];
+ if (key != NULL)
memcpy(hw->rss_info.key, key, hw->rss_key_size);
- }
hw->rss_info.rss_hf = rss_hf;
rte_spinlock_unlock(&hw->lock);
@@ -769,7 +769,13 @@ int
hns3_dev_rss_hash_conf_get(struct rte_eth_dev *dev,
struct rte_eth_rss_conf *rss_conf)
{
+ const uint8_t hash_func_map[] = {
+ [HNS3_RSS_HASH_ALGO_TOEPLITZ] = RTE_ETH_HASH_FUNCTION_TOEPLITZ,
+ [HNS3_RSS_HASH_ALGO_SIMPLE] = RTE_ETH_HASH_FUNCTION_SIMPLE_XOR,
+ [HNS3_RSS_HASH_ALGO_SYMMETRIC_TOEP] =
RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ,
+ };
struct hns3_adapter *hns = dev->data->dev_private;
+ uint8_t rss_key[HNS3_RSS_KEY_SIZE_MAX] = {0};
struct hns3_hw *hw = &hns->hw;
uint8_t hash_algo;
int ret;
@@ -777,26 +783,27 @@ hns3_dev_rss_hash_conf_get(struct rte_eth_dev *dev,
rte_spinlock_lock(&hw->lock);
ret = hns3_rss_hash_get_rss_hf(hw, &rss_conf->rss_hf);
if (ret != 0) {
+ rte_spinlock_unlock(&hw->lock);
hns3_err(hw, "obtain hash tuples failed, ret = %d", ret);
- goto out;
+ return ret;
+ }
+
+ ret = hns3_rss_get_algo_key(hw, &hash_algo, rss_key, hw->rss_key_size);
+ if (ret != 0) {
+ rte_spinlock_unlock(&hw->lock);
+ hns3_err(hw, "obtain hash algo and key failed, ret = %d", ret);
+ return ret;
}
+ rte_spinlock_unlock(&hw->lock);
- /* Get the RSS Key required by the user */
+ /* Get the RSS Key if user required. */
if (rss_conf->rss_key && rss_conf->rss_key_len >= hw->rss_key_size) {
- ret = hns3_rss_get_algo_key(hw, &hash_algo, rss_conf->rss_key,
- hw->rss_key_size);
- if (ret != 0) {
- hns3_err(hw, "obtain hash algo and key failed, ret =
%d",
-ret);
- goto out;
- }
+ memcpy(rss_conf->rss_key, rss_key, hw->rss_key_size);
rss_conf->rss_key_len = hw->rss_key_size;
}
+ rss_conf->algorithm = hash_func_map[hash_algo];
-out:
- rte_spinlock_unlock(&hw->lock);
-
- return ret;
+ return 0;
}
/*
--
2.30.0
[PATCH v5 37/40] app/proc-info: fix never show RSS info
Command show-port should show RSS info (rss_key, len and rss_hf),
However, the information is shown only when rss_conf.rss_key is not
NULL. Since no memory is allocated for rss_conf.rss_key, rss_key
will always be NULL and the rss_info will never show. This patch
fixes it.
Fixes: 8a37f37fc243 ("app/procinfo: add --show-port")
Cc: sta...@dpdk.org
Signed-off-by: Jie Hai
Signed-off-by: Dongdong Liu
Acked-by: Reshma Pattan
---
app/proc-info/main.c | 21 -
1 file changed, 12 insertions(+), 9 deletions(-)
diff --git a/app/proc-info/main.c b/app/proc-info/main.c
index af4c1d8bcbd4..4509b3c16e36 100644
--- a/app/proc-info/main.c
+++ b/app/proc-info/main.c
@@ -152,6 +152,8 @@ struct desc_param {
static struct desc_param rx_desc_param;
static struct desc_param tx_desc_param;
+#define RSS_HASH_KEY_SIZE 64
+
/* display usage */
static void
proc_info_usage(const char *prgname)
@@ -1013,6 +1015,7 @@ show_port(void)
struct rte_eth_fc_conf fc_conf;
struct rte_ether_addr mac;
struct rte_eth_dev_owner owner;
+ uint8_t rss_key[RSS_HASH_KEY_SIZE];
/* Skip if port is not in mask */
if ((enabled_port_mask & (1ul << i)) == 0)
@@ -1171,17 +1174,17 @@ show_port(void)
printf("\n");
}
+ rss_conf.rss_key = rss_key;
+ rss_conf.rss_key_len = dev_info.hash_key_size;
ret = rte_eth_dev_rss_hash_conf_get(i, &rss_conf);
if (ret == 0) {
- if (rss_conf.rss_key) {
- printf(" - RSS\n");
- printf("\t -- RSS len %u key (hex):",
- rss_conf.rss_key_len);
- for (k = 0; k < rss_conf.rss_key_len; k++)
- printf(" %x", rss_conf.rss_key[k]);
- printf("\t -- hf 0x%"PRIx64"\n",
- rss_conf.rss_hf);
- }
+ printf(" - RSS\n");
+ printf("\t -- RSS len %u key (hex):",
+ rss_conf.rss_key_len);
+ for (k = 0; k < rss_conf.rss_key_len; k++)
+ printf(" %x", rss_conf.rss_key[k]);
+ printf("\t -- hf 0x%"PRIx64"\n",
+ rss_conf.rss_hf);
}
#ifdef RTE_LIB_SECURITY
--
2.30.0
[PATCH v5 38/40] app/proc-info: adjust the display format of RSS info
This patch splits the length and value of RSS key into two parts,
removes spaces between RSS keys, and adds line breaks between RSS
key and RSS hf.
Before the adjustment, RSS info is shown as:
- RSS
-- RSS len 40 key (hex): 6d 5a 56 da 25 5b e c2 41 67 \
25 3d 43 a3 8f b0 d0 ca 2b cb ae 7b 30 b4 77 cb 2d \
a3 80 30 f2 c 6a 42 b7 3b be ac 1 fa -- hf 0x0
and after:
- RSS info
-- key len : 40
-- key (hex) : 6d5a56da255b0ec24167253d43a38fb0d0c \
a2bcbae7b30b477cb2da38030f20c6a42b73bbeac01fa
-- hash function : 0x0
Fixes: 8a37f37fc243 ("app/procinfo: add --show-port")
Cc: sta...@dpdk.org
Signed-off-by: Jie Hai
Signed-off-by: Dongdong Liu
Acked-by: Reshma Pattan
---
app/proc-info/main.c | 9 +
1 file changed, 5 insertions(+), 4 deletions(-)
diff --git a/app/proc-info/main.c b/app/proc-info/main.c
index 4509b3c16e36..e98352118db1 100644
--- a/app/proc-info/main.c
+++ b/app/proc-info/main.c
@@ -1178,12 +1178,13 @@ show_port(void)
rss_conf.rss_key_len = dev_info.hash_key_size;
ret = rte_eth_dev_rss_hash_conf_get(i, &rss_conf);
if (ret == 0) {
- printf(" - RSS\n");
- printf("\t -- RSS len %u key (hex):",
+ printf(" - RSS info\n");
+ printf("\t -- key len : %u\n",
rss_conf.rss_key_len);
+ printf("\t -- key (hex) : ");
for (k = 0; k < rss_conf.rss_key_len; k++)
- printf(" %x", rss_conf.rss_key[k]);
- printf("\t -- hf 0x%"PRIx64"\n",
+ printf("%02x", rss_conf.rss_key[k]);
+ printf("\n\t -- hash function : 0x%"PRIx64"\n",
rss_conf.rss_hf);
}
--
2.30.0
[PATCH v5 39/40] app/proc-info: support querying RSS hash algorithm
Display RSS hash algorithm with command show-port as below.
- RSS info
-- hash algorithm : toeplitz
Signed-off-by: Jie Hai
Signed-off-by: Dongdong Liu
Acked-by: Reshma Pattan
---
app/proc-info/main.c | 10 ++
1 file changed, 10 insertions(+)
diff --git a/app/proc-info/main.c b/app/proc-info/main.c
index e98352118db1..43f264848a6f 100644
--- a/app/proc-info/main.c
+++ b/app/proc-info/main.c
@@ -153,6 +153,14 @@ static struct desc_param rx_desc_param;
static struct desc_param tx_desc_param;
#define RSS_HASH_KEY_SIZE 64
+static const char * const rss_hash_algos[] = {
+ [RTE_ETH_HASH_FUNCTION_SIMPLE_XOR] = "simple_xor",
+ [RTE_ETH_HASH_FUNCTION_TOEPLITZ] = "toeplitz",
+ [RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ] = "symmetric_toeplitz",
+ [RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ_SORT] =
"symmetric_toeplitz_sort",
+ [RTE_ETH_HASH_FUNCTION_DEFAULT] = "default",
+ [RTE_ETH_HASH_FUNCTION_MAX] = "unknown"
+};
/* display usage */
static void
@@ -1186,6 +1194,8 @@ show_port(void)
printf("%02x", rss_conf.rss_key[k]);
printf("\n\t -- hash function : 0x%"PRIx64"\n",
rss_conf.rss_hf);
+ printf("\t -- hash algorithm : %s\n",
+ rss_hash_algos[rss_conf.algorithm]);
}
#ifdef RTE_LIB_SECURITY
--
2.30.0
[PATCH v5 40/40] app/testpmd: add RSS hash algorithms display
Add the command "show port X rss-hash algorithm" to display
the RSS hash algorithms of port X. An example is shown:
testpmd> show port 0 rss-hash algorithm
RSS algorithms:
toeplitz
Signed-off-by: Jie Hai
---
app/test-pmd/cmdline.c | 29 -
app/test-pmd/config.c | 38 +++---
app/test-pmd/testpmd.h | 2 +-
3 files changed, 44 insertions(+), 25 deletions(-)
diff --git a/app/test-pmd/cmdline.c b/app/test-pmd/cmdline.c
index 679ca47b9401..d0eafd7f1254 100644
--- a/app/test-pmd/cmdline.c
+++ b/app/test-pmd/cmdline.c
@@ -174,8 +174,8 @@ static void cmd_help_long_parsed(void *parsed_result,
" by masks on port X. size is used to indicate the"
" hardware supported reta size\n\n"
- "show port (port_id) rss-hash [key]\n"
- "Display the RSS hash functions and RSS hash key of
port\n\n"
+ "show port (port_id) rss-hash [key | algorithm]\n"
+ "Display the RSS hash functions, RSS hash key and
RSS hash algorithms of port\n\n"
"clear port (info|stats|xstats|fdir) (port_id|all)\n"
"Clear information for port_id, or all.\n\n"
@@ -3026,15 +3026,17 @@ struct cmd_showport_rss_hash {
cmdline_fixed_string_t rss_hash;
cmdline_fixed_string_t rss_type;
cmdline_fixed_string_t key; /* optional argument */
+ cmdline_fixed_string_t algorithm; /* optional argument */
};
static void cmd_showport_rss_hash_parsed(void *parsed_result,
__rte_unused struct cmdline *cl,
- void *show_rss_key)
+ __rte_unused void *data)
{
struct cmd_showport_rss_hash *res = parsed_result;
- port_rss_hash_conf_show(res->port_id, show_rss_key != NULL);
+ port_rss_hash_conf_show(res->port_id,
+ !strcmp(res->key, "key"), !strcmp(res->algorithm, "algorithm"));
}
static cmdline_parse_token_string_t cmd_showport_rss_hash_show =
@@ -3049,6 +3051,8 @@ static cmdline_parse_token_string_t
cmd_showport_rss_hash_rss_hash =
"rss-hash");
static cmdline_parse_token_string_t cmd_showport_rss_hash_rss_key =
TOKEN_STRING_INITIALIZER(struct cmd_showport_rss_hash, key, "key");
+static cmdline_parse_token_string_t cmd_showport_rss_hash_rss_algo =
+ TOKEN_STRING_INITIALIZER(struct cmd_showport_rss_hash, algorithm,
"algorithm");
static cmdline_parse_inst_t cmd_showport_rss_hash = {
.f = cmd_showport_rss_hash_parsed,
@@ -3065,7 +3069,7 @@ static cmdline_parse_inst_t cmd_showport_rss_hash = {
static cmdline_parse_inst_t cmd_showport_rss_hash_key = {
.f = cmd_showport_rss_hash_parsed,
- .data = (void *)1,
+ .data = NULL,
.help_str = "show port rss-hash key",
.tokens = {
(void *)&cmd_showport_rss_hash_show,
@@ -3077,6 +3081,20 @@ static cmdline_parse_inst_t cmd_showport_rss_hash_key = {
},
};
+static cmdline_parse_inst_t cmd_showport_rss_hash_algo = {
+ .f = cmd_showport_rss_hash_parsed,
+ .data = NULL,
+ .help_str = "show port rss-hash algorithm",
+ .tokens = {
+ (void *)&cmd_showport_rss_hash_show,
+ (void *)&cmd_showport_rss_hash_port,
+ (void *)&cmd_showport_rss_hash_port_id,
+ (void *)&cmd_showport_rss_hash_rss_hash,
+ (void *)&cmd_showport_rss_hash_rss_algo,
+ NULL,
+ },
+};
+
/* *** Configure DCB *** */
struct cmd_config_dcb {
cmdline_fixed_string_t port;
@@ -12953,6 +12971,7 @@ static cmdline_parse_ctx_t builtin_ctx[] = {
(cmdline_parse_inst_t *)&cmd_tunnel_udp_config,
(cmdline_parse_inst_t *)&cmd_showport_rss_hash,
(cmdline_parse_inst_t *)&cmd_showport_rss_hash_key,
+ (cmdline_parse_inst_t *)&cmd_showport_rss_hash_algo,
(cmdline_parse_inst_t *)&cmd_config_rss_hash_key,
(cmdline_parse_inst_t *)&cmd_cleanup_txq_mbufs,
(cmdline_parse_inst_t *)&cmd_dump,
diff --git a/app/test-pmd/config.c b/app/test-pmd/config.c
index 7034fa125bc0..deb9414191b7 100644
--- a/app/test-pmd/config.c
+++ b/app/test-pmd/config.c
@@ -1485,6 +1485,15 @@ rss_types_display(uint64_t rss_types, uint16_t
char_num_per_line)
printf("\n");
}
+static const char * const rss_hash_algos[] = {
+ [RTE_ETH_HASH_FUNCTION_SIMPLE_XOR] = "simple_xor",
+ [RTE_ETH_HASH_FUNCTION_TOEPLITZ] = "toeplitz",
+ [RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ] = "symmetric_toeplitz",
+ [RTE_ETH_HASH_FUNCTION_SYMMETRIC_TOEPLITZ_SORT] =
"symmetric_toeplitz_sort",
+ [RTE_ETH_HASH_FUNCTION_DEFAULT] = "default",
+ [RTE_ETH_HASH_FUNCTION_MAX] = "Unknown function"
+};
+
static void
rss_config_display(struct rte_flow_action_rss *rss_conf)
{
@@ -1503,24 +1512,7 @@ rss_config_d
RE: [PATCH v6 3/3] power: amd power monitor support
[AMD Official Use Only - General]
Hi David,
> -Original Message-
> From: David Marchand
> Sent: Tuesday, October 10, 2023 2:30 PM
> To: Tummala, Sivaprasad
> Cc: david.h...@intel.com; konstantin.v.anan...@yandex.ru;
> roret...@linux.microsoft.com; anatoly.bura...@intel.com; tho...@monjalon.net;
> Yigit, Ferruh ; dev@dpdk.org
> Subject: Re: [PATCH v6 3/3] power: amd power monitor support
>
> Caution: This message originated from an External Source. Use proper caution
> when opening attachments, clicking links, or responding.
>
>
> Hello Siva,
>
> On Mon, Oct 9, 2023 at 4:06 PM Sivaprasad Tummala
> wrote:
> >
> > mwaitx allows EPYC processors to enter a implementation dependent
> > power/performance optimized state (C1 state) for a specific period or
> > until a store to the monitored address range.
> >
> > Signed-off-by: Sivaprasad Tummala
> > Acked-by: Anatoly Burakov
> > ---
>
> Please put some changelog to make life easier for reviewers (and me).
>
> I diffed with the previous version to check what had been changed and I see
> this:
>
> static void amd_mwaitx(const uint64_t timeout) ...
> - "c"(2), /* enable timer */
> - "b"(timeout));
> + "c"(0)); /* no time-out */
> /
> I will take this series as is, but please confirm why this change was needed.
>
Thanks for applying the patch set. The change was needed to fix compilation
issue for 32-bit DPDK.
>
> > lib/eal/x86/rte_power_intrinsics.c | 108
> > ++---
> > 1 file changed, 84 insertions(+), 24 deletions(-)
> >
> > diff --git a/lib/eal/x86/rte_power_intrinsics.c
> > b/lib/eal/x86/rte_power_intrinsics.c
> > index 664cde01e9..0d2953f570 100644
> > --- a/lib/eal/x86/rte_power_intrinsics.c
> > +++ b/lib/eal/x86/rte_power_intrinsics.c
> > @@ -17,6 +17,78 @@ static struct power_wait_status {
> > volatile void *monitor_addr; /**< NULL if not currently
> > sleeping */ } __rte_cache_aligned wait_status[RTE_MAX_LCORE];
> >
> > +/**
> > + * This functions uses UMONITOR/UMWAIT instructions and will enter C0.2
> > state.
>
> Fixed while applying, function*
>
> > + * For more information about usage of these instructions, please
> > +refer to
> > + * Intel(R) 64 and IA-32 Architectures Software Developer's Manual.
> > + */
> > +static void intel_umonitor(volatile void *addr) { #if
> > +defined(RTE_TOOLCHAIN_MSVC) || defined(__WAITPKG__)
> > + /* cast away "volatile" when using the intrinsic */
> > + _umonitor((void *)(uintptr_t)addr); #else
> > + /*
> > +* we're using raw byte codes for compiler versions which
> > +* don't support this instruction natively.
> > +*/
> > + asm volatile(".byte 0xf3, 0x0f, 0xae, 0xf7;"
> > + :
> > + : "D"(addr));
> > +#endif
> > +}
> > +
> > +static void intel_umwait(const uint64_t timeout) {
> > + const uint32_t tsc_l = (uint32_t)timeout;
> > + const uint32_t tsc_h = (uint32_t)(timeout >> 32); #if
> > +defined(RTE_TOOLCHAIN_MSVC) || defined(__WAITPKG__)
> > + _umwait(tsc_l, tsc_h);
> > +#else
> > + asm volatile(".byte 0xf2, 0x0f, 0xae, 0xf7;"
> > + : /* ignore rflags */
> > + : "D"(0), /* enter C0.2 */
> > + "a"(tsc_l), "d"(tsc_h)); #endif }
> > +
> > +/**
> > + * This functions uses MONITORX/MWAITX instructions and will enter C1
> > state.
> > + * For more information about usage of these instructions, please
> > +refer to
> > + * AMD64 Architecture Programmer’s Manual.
> > + */
> > +static void amd_monitorx(volatile void *addr) { #if
> > +defined(__MWAITX__)
>
> This part probably breaks build with MSVC.
>
> I could not determine whether MSVC supports this intrinsic.
> I'll rely on Tyler to fix it later.
>
>
> Series applied.
>
> --
> David Marchand
Re: [PATCH v4] dmadev: add tracepoints
Hi Thomas,
Sorry for the late reply.
On 2023/8/14 22:16, Thomas Monjalon wrote:
> jeudi 3 août 2023, fengchengwen:
>> Hi Thomas,
>>
>> On 2023/7/31 20:48, Thomas Monjalon wrote:
>>> 10/07/2023 09:50, fengchengwen:
Hi Thomas,
On 2023/7/10 14:49, Thomas Monjalon wrote:
> 09/07/2023 05:23, fengchengwen:
>> Hi Thomas,
>>
>> On 2023/7/7 18:40, Thomas Monjalon wrote:
>>> 26/05/2023 10:42, Chengwen Feng:
Add tracepoints at important APIs for tracing support.
Signed-off-by: Chengwen Feng
Acked-by: Morten Brørup
---
v4: Fix asan smoke fail.
v3: Address Morten's comment:
Move stats_get and vchan_status and to trace_fp.h.
v2: Address Morten's comment:
Make stats_get as fast-path trace-points.
Place fast-path trace-point functions behind in version.map.
>>>
>>> There are more things to fix.
>>> First you must export rte_dmadev_trace_fp.h as it is included by
>>> rte_dmadev.h.
>>
>> It was already included by rte_dmadev.h:
>> diff --git a/lib/dmadev/rte_dmadev.h b/lib/dmadev/rte_dmadev.h
>> index e61d71959e..e792b90ef8 100644
>> --- a/lib/dmadev/rte_dmadev.h
>> +++ b/lib/dmadev/rte_dmadev.h
>> @@ -796,6 +796,7 @@ struct rte_dma_sge {
>> };
>>
>> #include "rte_dmadev_core.h"
>> +#include "rte_dmadev_trace_fp.h"
>>
>>
>>> Note: you could have caught this if testing the example app for DMA.
>>> Second, you must avoid structs and enum in this header file,
>>
>> Let me explain the #if #endif logic:
>>
>> For the function:
>> uint16_t
>> rte_dma_completed(int16_t dev_id, uint16_t vchan, const uint16_t nb_cpls,
>>uint16_t *last_idx, bool *has_error)
>>
>> The common trace implementation:
>> RTE_TRACE_POINT_FP(
>> rte_dma_trace_completed,
>> RTE_TRACE_POINT_ARGS(int16_t dev_id, uint16_t vchan,
>> const uint16_t nb_cpls, uint16_t *last_idx,
>> bool *has_error, uint16_t ret),
>> rte_trace_point_emit_i16(dev_id);
>> rte_trace_point_emit_u16(vchan);
>> rte_trace_point_emit_u16(nb_cpls);
>> rte_trace_point_emit_ptr(idx_val);
>> rte_trace_point_emit_ptr(has_error);
>> rte_trace_point_emit_u16(ret);
>> )
>>
>> But it has a problem: for pointer parameter (e.g. last_idx and
>> has_error), only record
>> the pointer value (i.e. address value).
>>
>> I think the pointer value has no mean (in particular, many of there
>> pointers are stack
>> variables), the value of the pointer point to is meaningful.
>>
>> So I add the pointer reference like below (as V3 did):
>> RTE_TRACE_POINT_FP(
>> rte_dma_trace_completed,
>> RTE_TRACE_POINT_ARGS(int16_t dev_id, uint16_t vchan,
>> const uint16_t nb_cpls, uint16_t *last_idx,
>> bool *has_error, uint16_t ret),
>> int has_error_val = *has_error;// pointer reference
>> int last_idx_val = *last_idx; // pointer reference
>> rte_trace_point_emit_i16(dev_id);
>> rte_trace_point_emit_u16(vchan);
>> rte_trace_point_emit_u16(nb_cpls);
>> rte_trace_point_emit_int(last_idx_val);// record the value of
>> pointer
>> rte_trace_point_emit_int(has_error_val); // record the value of
>> pointer
>> rte_trace_point_emit_u16(ret);
>> )
>>
>> Unfortunately, the above lead to asan failed. because in:
>> RTE_TRACE_POINT_REGISTER(rte_dma_trace_completed,
>> lib.dmadev.completed)
>> it will invoke rte_dma_trace_completed() with the parameter is undefined.
>>
>>
>> To solve this problem, consider the rte_dmadev_trace_points.c will
>> include rte_trace_point_register.h,
>> and the rte_trace_point_register.h will defined macro:
>> _RTE_TRACE_POINT_REGISTER_H_.
>>
>> so we update trace points as (as V4 did):
>> RTE_TRACE_POINT_FP(
>> rte_dma_trace_completed,
>> RTE_TRACE_POINT_ARGS(int16_t dev_id, uint16_t vchan,
>> const uint16_t nb_cpls, uint16_t *last_idx,
>> bool *has_error, uint16_t ret),
>> #ifdef _RTE_TRACE_POINT_REGISTER_H_
>> uint16_t __last_idx = 0;
>> bool __has_error = false;
>> last_idx = &__last_idx; // make sure the pointer has
>> meaningful value.
>> has_error = &__has_error;// so that the next pointer
>> reference will work well.
>> #endif /* _RTE_TRACE_POINT_REGISTER_H_ */
>> int has_error_val = *has_error;
>> int last_idx_val = *last_idx;
>> rte_trace_point_emit_i16(dev_id);
>> rte_trace_point_emit_u16(vchan);
>> rte_trace_point_emit_u16(nb_cpls);
>> rte_trace_point_emit_int(last_idx_val);
>> rte_trace_point_emit_int(ha
Re: [PATCH] doc: remove confusing command to send patch
11/10/2023 10:41, Ferruh Yigit: > On 10/11/2023 9:30 AM, Bruce Richardson wrote: > > On Wed, Oct 11, 2023 at 10:03:07AM +0200, Thomas Monjalon wrote: > >> 11/10/2023 09:30, David Marchand: > >>> On Tue, Oct 10, 2023 at 6:26 PM Thomas Monjalon > >>> wrote: > > In the contributor guide, it was said that no need to Cc maintainers > for new additions, probably for new directories not having a maintainer. > There is no harm, and it is a good habit, to always Cc maintainers. > > Remove this case as it can mislead to not Cc maintainers when needed. > > Signed-off-by: Thomas Monjalon > >>> > >>> I agree Cc: maintainers should be the default / recommended way of > >>> sending patches. > >>> > >>> Just to convince myself, adding some meson skeleton for a "plop" > >>> library, adding an entry in the release notes and hooking in > >>> lib/meson.build: > >>> $ git show --stat > >>> doc/guides/rel_notes/release_23_11.rst | 4 > >>> lib/meson.build| 1 + > >>> lib/plop/meson.build | 2 ++ > >>> > >>> $ ./devtools/get-maintainer.sh 0001-new-awesome-library.patch > >>> > >>> In this case, it translates to an empty To: list if you follow the > >>> example command line: > >>>git send-email --to-cmd ./devtools/get-maintainer.sh --cc > >>> dev@dpdk.org 000*.patch > >>> > >>> We could add a default list of recipients if no maintainer is found by > >>> the script. > >>> And the next question is who should be in that list.. > >> > >> Or we can send to dev@dpdk.org, Cc maintainers. > >> This is what I do: > >> git send-email --to dev@dpdk.org --cc-cmd devtools/get-maintainer.sh > >> > > +1 for this, mainly on the basis of it being what I do too! :-) > > > > I am for "--to-cmd=./devtools/get-maintainer.sh --cc dev@dpdk.org" > > To highlight response is expected from the maintainers, and community is > informed. > > Also people may have filters to give higher priority to emails they are > in 'to' list, high priority is what we want from maintainers :) They should give high priority when they are Cc as well. The problem is that we may have patches with empty "To", especially for cover letters and new libs.
Re: [PATCH v4] app/testpmd: enable cli for programmable action
On 10/11/2023 3:24 AM, Zhang, Qi Z wrote:
>
>> -Original Message-
>> From: Ferruh Yigit
>> Sent: Tuesday, October 10, 2023 6:49 PM
>> To: Zhang, Qi Z ; Singh, Aman Deep
>> ; Zhang, Yuying
>> Cc: dev@dpdk.org; Dumitrescu, Cristian ;
>> or...@nvidia.com
>> Subject: Re: [PATCH v4] app/testpmd: enable cli for programmable action
>>
>> On 10/7/2023 11:47 AM, Qi Zhang wrote:
>>> Parsing command line for rte_flow_action_prog.
>>>
>>> Syntax:
>>>
>>> "prog name [arguments \
>>> ... end]"
>>>
>>
>> Can you please put full rte flow command in the commit log? Like what is the
>> 'pattern' for above command?
>
> The pattern part should be independent of the action part,
>
> though for our P4 device, we will prefer use rte_flow_flex_item, something
> like:
>
> flow create 0 pattern flex item is xxx pattern is xxx / flex item is xxx
> pattern is / actions prog name ..
>
> but it does not limit PMD to support flow like below
>
I think agreement was to use flex pattern, and my understand is "struct
rte_flow_item_flex" will be used to present the table_id.
Without not using flex, how driver will detect which table to update?
> flow create 0 pattern eth / ipv4 src is 1.1.1.1 / actions prog name ..
>
> So I think it may not be necessary to highlight the pattern format here.
>
Complete samples helps a lot to user, can you please include the full
rte flow command, you can have flex pattern sample and if you want add
more samples with other patterns but we need to clarify it first.
>>
>>
>>> Use parse_string0 to parse name string.
>>> Use parse_hex to parse hex string.
>>> Use struct action_prog_data to store parsed result.
>>>
>>> Example:
>>>
>>> Action with 2 arguments:
>>>
>>> "prog name action0 arguments field0 03FF field1 55AA end"
>>>
>>> Action without argument:
>>>
>>> "prog name action1"
>>>
>>> Signed-off-by: Qi Zhang
>>>
>>
>> Is there an existing driver implementation, checking it helps to understand
>> feature implementation?
>
> This work is still ongoing, currently we target to upstream on DPDK 24.03
>
If you won't have driver yet, do you have a way to test these commands?
Or is this implementation just theoretical at this stage?
>>
>>
>>> ---
>>>
>>> v4:
>>> - be more generous on the max size of name and value.
>>>
>>> v3:
>>> - refine struct action_prog_data
>>> - enlarge the max size
>>>
>>> v2:
>>> - fix title
>>> - minor coding style refine.
>>>
>>> app/test-pmd/cmdline_flow.c | 232
>>>
>>> 1 file changed, 232 insertions(+)
>>>
>>
>> Hi Qi,
>>
>> Can you please update documentation too,
>> `doc/guides/testpmd_app_ug/testpmd_funcs.rst`, `Flow rules management`
>> section.
>
> Sure.
>
>>
>>
>>> diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
>>> index 21828c144c..ae5556e704 100644
>>> --- a/app/test-pmd/cmdline_flow.c
>>> +++ b/app/test-pmd/cmdline_flow.c
>>> @@ -719,6 +719,13 @@ enum index {
>>> ACTION_IPV6_EXT_PUSH,
>>> ACTION_IPV6_EXT_PUSH_INDEX,
>>> ACTION_IPV6_EXT_PUSH_INDEX_VALUE,
>>> + ACTION_PROG,
>>> + ACTION_PROG_NAME,
>>> + ACTION_PROG_NAME_STRING,
>>> + ACTION_PROG_ARGUMENTS,
>>> + ACTION_PROG_ARG_NAME,
>>> + ACTION_PROG_ARG_VALUE,
>>> + ACTION_PROG_ARG_END,
>>> };
>>>
>>> /** Maximum size for pattern in struct rte_flow_item_raw. */ @@
>>> -749,6 +756,23 @@ struct action_rss_data {
>>> uint16_t queue[ACTION_RSS_QUEUE_NUM]; };
>>>
>>> +#define ACTION_PROG_NAME_SIZE_MAX 256 #define
>> ACTION_PROG_ARG_NUM_MAX
>>> +16 #define ACTION_PROG_ARG_VALUE_SIZE_MAX 64
>>> +
>>> +/** Storage for struct rte_flow_action_prog including external data.
>>> +*/ struct action_prog_data {
>>> + struct rte_flow_action_prog conf;
>>> + struct {
>>> + char name[ACTION_PROG_NAME_SIZE_MAX];
>>> + struct rte_flow_action_prog_argument
>> args[ACTION_PROG_ARG_NUM_MAX];
>>> + struct {
>>> + char names[ACTION_PROG_NAME_SIZE_MAX];
>>> + uint8_t
>> value[ACTION_PROG_ARG_VALUE_SIZE_MAX];
>>> + } arg_data[ACTION_PROG_ARG_NUM_MAX];
>>> + } data;
>>> +};
>>> +
>>> /** Maximum data size in struct rte_flow_action_raw_encap. */
>>> #define ACTION_RAW_ENCAP_MAX_DATA 512 #define
>> RAW_ENCAP_CONFS_MAX_NUM
>>> 8 @@ -2169,6 +2193,7 @@ static const enum index next_action[] = {
>>> ACTION_QUOTA_QU,
>>> ACTION_IPV6_EXT_REMOVE,
>>> ACTION_IPV6_EXT_PUSH,
>>> + ACTION_PROG,
>>> ZERO,
>>> };
>>>
>>> @@ -2510,6 +2535,13 @@ static const enum index
>> action_represented_port[] = {
>>> ZERO,
>>> };
>>>
>>> +static const enum index action_prog[] = {
>>> + ACTION_PROG_NAME,
>>> + ACTION_PROG_ARGUMENTS,
>>> + ACTION_NEXT,
>>> + ZERO,
>>> +};
>>> +
>>> static int parse_set_raw_encap_decap(struct context *, const struct token
>>> *,
>>> const char *, unsigned int,
>>> void *, unsigned int);
>>> @@ -2786,6 +2818,18 @@ static int
>>> par
Re: [PATCH] doc: remove confusing command to send patch
On 10/11/2023 11:20 AM, Thomas Monjalon wrote: > 11/10/2023 10:41, Ferruh Yigit: >> On 10/11/2023 9:30 AM, Bruce Richardson wrote: >>> On Wed, Oct 11, 2023 at 10:03:07AM +0200, Thomas Monjalon wrote: 11/10/2023 09:30, David Marchand: > On Tue, Oct 10, 2023 at 6:26 PM Thomas Monjalon > wrote: >> >> In the contributor guide, it was said that no need to Cc maintainers >> for new additions, probably for new directories not having a maintainer. >> There is no harm, and it is a good habit, to always Cc maintainers. >> >> Remove this case as it can mislead to not Cc maintainers when needed. >> >> Signed-off-by: Thomas Monjalon > > I agree Cc: maintainers should be the default / recommended way of > sending patches. > > Just to convince myself, adding some meson skeleton for a "plop" > library, adding an entry in the release notes and hooking in > lib/meson.build: > $ git show --stat > doc/guides/rel_notes/release_23_11.rst | 4 > lib/meson.build| 1 + > lib/plop/meson.build | 2 ++ > > $ ./devtools/get-maintainer.sh 0001-new-awesome-library.patch > > In this case, it translates to an empty To: list if you follow the > example command line: >git send-email --to-cmd ./devtools/get-maintainer.sh --cc > dev@dpdk.org 000*.patch > > We could add a default list of recipients if no maintainer is found by > the script. > And the next question is who should be in that list.. Or we can send to dev@dpdk.org, Cc maintainers. This is what I do: git send-email --to dev@dpdk.org --cc-cmd devtools/get-maintainer.sh >>> +1 for this, mainly on the basis of it being what I do too! :-) >>> >> >> I am for "--to-cmd=./devtools/get-maintainer.sh --cc dev@dpdk.org" >> >> To highlight response is expected from the maintainers, and community is >> informed. >> >> Also people may have filters to give higher priority to emails they are >> in 'to' list, high priority is what we want from maintainers :) > > They should give high priority when they are Cc as well. > > The problem is that we may have patches with empty "To", > especially for cover letters and new libs. > There are indeed, for those cases I am putting 'dev' to "To:".
Re: [PATCH v3] net/netvsc: add support for mtu_set
On 10/10/2023 8:08 PM, Long Li wrote: >> Subject: [PATCH v3] net/netvsc: add support for mtu_set >> >> Add support for changing the netvsc MTU. The MTU can only be set at nvs >> initialization, therefore to change the MTU the underlying vmbus >> channel(s) are torn down and the vmbus device unmapped and remapped. The >> existing rx and tx queue(s) are reconnected to the new vmbus channel(s). >> >> Signed-off-by: Sam Andrew >> Acked-by: Stephen Hemminger > > Acked-by: Long Li > Applied to dpdk-next-net/main, thanks.
Re: [PATCH v2 1/3] net/ark: support for single function with multiple port
On 10/10/2023 9:42 PM, Ed Czeck wrote:
> Support the creation of multiple ports from one ark device via
> the use of ark pmd extension. I.e., one device with q queue can
> seen a p ports each with q/p queues.
>
> Add unique dev_private data for each port to manage queue assignment.
>
> This patch repairs a latent issue uncovered during testing.
> Fixes: 6799275eeea6 ("net/ark: support virtual functions")
> Cc: sta...@dpdk.org
> Backporting is not requested.
>
> Signed-off-by: Ed Czeck
>
Series applied to dpdk-next-net/main, thanks.
[PATCH v3 1/2] doc: add modify_field action description
This commit adds the missing modify_field action description to
`testpmd_funcs.rst`.
Signed-off-by: Suanming Mou
---
v3: add modify_field description.
---
doc/guides/testpmd_app_ug/testpmd_funcs.rst | 21 +
1 file changed, 21 insertions(+)
diff --git a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
index 67968ecb7f..04c502798a 100644
--- a/doc/guides/testpmd_app_ug/testpmd_funcs.rst
+++ b/doc/guides/testpmd_app_ug/testpmd_funcs.rst
@@ -4108,6 +4108,27 @@ This section lists supported actions and their
attributes, if any.
- ``mtr_init_color {value}``: initial color value (green/yellow/red)
- ``mtr_state {unsigned}``: meter state (disabled/enabled)
+- ``modify_field``: Modify packet field
+
+ - ``op``: modify operation (set/add/sub)
+ - ``dst_type``: the destination field to be modified, the supported fields as
+``enum rte_flow_field_id`` listed.
+ - ``dst_level``: destination field level.
+ - ``dst_tag_index``: destination field tag array.
+ - ``dst_type_id``: destination field type ID.
+ - ``dst_class``: destination field class ID.
+ - ``dst_offset``: destination field bit offset.
+ - ``src_type``: the modify source field, the supported fields as
+``enum rte_flow_field_id`` listed.
+ - ``src_level``: source field level.
+ - ``src_tag_index``: source field tag array.
+ - ``src_type_id``: source field type ID.
+ - ``src_class``: source field class ID.
+ - ``src_offset``: source field bit offset.
+ - ``src_value``: source immediate value.
+ - ``src_ptr``: pointer to source immediate value.
+ - ``width``: number of bits to copy.
+
Destroying flow rules
~
--
2.34.1
[PATCH v3 2/2] ethdev: add TCP/IP modify field IDs
Currently, get TCP/IP header or data length information from traffic
is missing in the modify field IDs. This commit adds the missing
TCP data_offset, IPv4 IHL/total_len, IPv6 payload_len to modify filed
IDs. This allows users be able to manager more TCP/IP fields.
Signed-off-by: Suanming Mou
---
v3: add modify_field description.
v2: fix typo tcp_date_off -> tcp_data_off
---
app/test-pmd/cmdline_flow.c | 1 +
lib/ethdev/rte_flow.h | 4
2 files changed, 5 insertions(+)
diff --git a/app/test-pmd/cmdline_flow.c b/app/test-pmd/cmdline_flow.c
index 10b9b25a3c..cf757d4c96 100644
--- a/app/test-pmd/cmdline_flow.c
+++ b/app/test-pmd/cmdline_flow.c
@@ -950,6 +950,7 @@ static const char *const modify_field_ids[] = {
"flex_item",
"hash_result",
"geneve_opt_type", "geneve_opt_class", "geneve_opt_data", "mpls",
+ "tcp_data_off", "ipv4_ihl", "ipv4_total_len", "ipv6_payload_len",
NULL
};
diff --git a/lib/ethdev/rte_flow.h b/lib/ethdev/rte_flow.h
index b79bb9968b..3b0485fcde 100644
--- a/lib/ethdev/rte_flow.h
+++ b/lib/ethdev/rte_flow.h
@@ -3921,6 +3921,10 @@ enum rte_flow_field_id {
RTE_FLOW_FIELD_GENEVE_OPT_CLASS,/**< GENEVE option class. */
RTE_FLOW_FIELD_GENEVE_OPT_DATA, /**< GENEVE option data. */
RTE_FLOW_FIELD_MPLS,/**< MPLS header. */
+ RTE_FLOW_FIELD_TCP_DATA_OFFSET, /**< TCP data offset. */
+ RTE_FLOW_FIELD_IPV4_IHL,/**< IPv4 IHL. */
+ RTE_FLOW_FIELD_IPV4_TOTAL_LEN, /**< IPv4 total length. */
+ RTE_FLOW_FIELD_IPV6_PAYLOAD_LEN /**< IPv6 payload length. */
};
/**
--
2.34.1
Re: [PATCH] eventdev: fix symbol export for port maintenance
On Wed, Oct 11, 2023 at 8:51 AM Jerin Jacob wrote:
> On Tue, Oct 10, 2023 at 7:30 PM David Marchand
> wrote:
> >
> > Trying to call rte_event_maintain out of the eventdev library triggers
> > a link failure, as the tracepoint symbol associated to this inline
> > helper was not exported.
> >
> > Fixes: 54f17843a887 ("eventdev: add port maintenance API")
> > Cc: sta...@dpdk.org
> >
> > Signed-off-by: David Marchand
> Acked-by: Jerin Jacob
Applied thanks.
--
David Marchand
Re: [PATCH] doc: sort build and EAL features in the release notes
On Wed, Oct 11, 2023 at 9:54 AM Thomas Monjalon wrote: > > When adding build and EAL features in 23.11, > the format and sorting order was unusual. Indeed, my bad. > This change is making these features similar as others. > > Signed-off-by: Thomas Monjalon Acked-by: David Marchand Applied, thanks. -- David Marchand
RE: [PATCH v2 25/29] regexdev: remove experimental tag
Hi Stephen,
I know that Nvidia, is going to remove the support for regex in the future.
The only other HW that implements this is Marvel.
I don't know about other manufacturers plan to support it. I suggest keeping it
as is for now, and maybe
we will need to remove this lib.
Anyone from Marvel can comment on this lib support?
Thanks,
Ori
> -Original Message-
> From: Stephen Hemminger
> Sent: Wednesday, August 9, 2023 3:10 AM
> To: dev@dpdk.org
> Cc: Stephen Hemminger ; Ori Kam
>
> Subject: [PATCH v2 25/29] regexdev: remove experimental tag
>
> This library was added in 22.11.
> Time to make it not experimental.
>
> Signed-off-by: Stephen Hemminger
> ---
> lib/regexdev/rte_regexdev.h | 92 -
> lib/regexdev/version.map| 2 +-
> 2 files changed, 1 insertion(+), 93 deletions(-)
>
> diff --git a/lib/regexdev/rte_regexdev.h b/lib/regexdev/rte_regexdev.h
> index 25476f1f739d..e8e194e29540 100644
> --- a/lib/regexdev/rte_regexdev.h
> +++ b/lib/regexdev/rte_regexdev.h
> @@ -226,9 +226,6 @@ extern int rte_regexdev_logtype;
> } while (0)
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Check if dev_id is ready.
> *
> * @param dev_id
> @@ -238,27 +235,19 @@ extern int rte_regexdev_logtype;
> * - 0 if device state is not in ready state.
> * - 1 if device state is ready state.
> */
> -__rte_experimental
> int rte_regexdev_is_valid_dev(uint16_t dev_id);
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Get the total number of RegEx devices that have been successfully
> * initialised.
> *
> * @return
> * The total number of usable RegEx devices.
> */
> -__rte_experimental
> uint8_t
> rte_regexdev_count(void);
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Get the device identifier for the named RegEx device.
> *
> * @param name
> @@ -268,7 +257,6 @@ rte_regexdev_count(void);
> * Returns RegEx device identifier on success.
> * - <0: Failure to find named RegEx device.
> */
> -__rte_experimental
> int
> rte_regexdev_get_dev_id(const char *name);
>
> @@ -628,9 +616,6 @@ struct rte_regexdev_info {
> };
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Retrieve the contextual information of a RegEx device.
> *
> * @param dev_id
> @@ -644,7 +629,6 @@ struct rte_regexdev_info {
> * - 0: Success, driver updates the contextual information of the RegEx
> device
> * - <0: Error code returned by the driver info get function.
> */
> -__rte_experimental
> int
> rte_regexdev_info_get(uint8_t dev_id, struct rte_regexdev_info *dev_info);
>
> @@ -723,9 +707,6 @@ struct rte_regexdev_config {
> };
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Configure a RegEx device.
> *
> * This function must be invoked first before any other function in the
> @@ -743,7 +724,6 @@ struct rte_regexdev_config {
> * @return
> * - 0: Success, device configured. Otherwise negative errno is returned.
> */
> -__rte_experimental
> int
> rte_regexdev_configure(uint8_t dev_id, const struct rte_regexdev_config
> *cfg);
>
> @@ -782,9 +762,6 @@ struct rte_regexdev_qp_conf {
> };
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Allocate and set up a RegEx queue pair for a RegEx device.
> *
> * @param dev_id
> @@ -799,15 +776,11 @@ struct rte_regexdev_qp_conf {
> * @return
> * 0 on success. Otherwise negative errno is returned.
> */
> -__rte_experimental
> int
> rte_regexdev_queue_pair_setup(uint8_t dev_id, uint16_t queue_pair_id,
> const struct rte_regexdev_qp_conf *qp_conf);
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Start a RegEx device.
> *
> * The device start step is the last one and consists of setting the RegEx
> @@ -822,14 +795,10 @@ rte_regexdev_queue_pair_setup(uint8_t dev_id,
> uint16_t queue_pair_id,
> * @return
> * 0 on success. Otherwise negative errno is returned.
> */
> -__rte_experimental
> int
> rte_regexdev_start(uint8_t dev_id);
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Stop a RegEx device.
> *
> * Stop a RegEx device. The device can be restarted with a call to
> @@ -845,14 +814,10 @@ rte_regexdev_start(uint8_t dev_id);
> * @return
> * 0 on success. Otherwise negative errno is returned.
> */
> -__rte_experimental
> int
> rte_regexdev_stop(uint8_t dev_id);
>
> /**
> - * @warning
> - * @b EXPERIMENTAL: this API may change without prior notice.
> - *
> * Close a RegEx device. The device cannot be restarted!
> *
> * @param dev_id
> @@ -861,7 +826,6 @@ rte_regexdev_stop(uint8_t dev_id);
> * @return
>
[RFC v2 0/2] add pointer compression API
This patchset is proposing adding a new EAL header with utility functions that allow compression of arrays of pointers. When passing caches full of pointers between threads, memory containing the pointers is copied multiple times which is especially costly between cores. A compression method will allow us to shrink the memory size copied. The compression takes advantage of the fact that pointers are usually located in a limited memory region (like a mempool). We can compress them by converting them to offsets from a base memory address. Offsets can be stored in fewer bytes (dictated by the memory region size and alignment of the pointer). For example: an 8 byte aligned pointer which is part of a 32GB memory pool can be stored in 4 bytes. The API is very generic and does not assume mempool pointers, any pointer can be passed in. Compression is based on few and fast operations and especially with vector instructions leveraged creates minimal overhead. The API accepts and returns arrays because the overhead means it only is worth it when done in bulk. Test is added that shows potential performance gain from compression. In this test an array of pointers is passed through a ring between two cores. It shows the gain which is dependent on the bulk operation size. In this synthetic test run on ampere altra a substantial (up to 25%) performance gain is seen if done in bulk size larger than 32. At 32 it breaks even and lower sizes create a small (less than 5%) slowdown due to overhead. In a more realistic mock application running the l3 forwarding dpdk example that works in pipeline mode this translated into a ~5% throughput increase on an ampere altra. v2: * addressed review comments (style, explanations and typos) * lowered bulk iterations closer to original numbers to keep runtime short * fixed pointer size warning on 32-bit arch Paul Szczepanek (2): eal: add pointer compression functions test: add pointer compress tests to ring perf test .mailmap | 1 + app/test/test_ring.h | 59 +- app/test/test_ring_perf.c | 324 ++--- lib/eal/include/meson.build| 1 + lib/eal/include/rte_ptr_compress.h | 160 ++ 5 files changed, 421 insertions(+), 124 deletions(-) create mode 100644 lib/eal/include/rte_ptr_compress.h -- 2.25.1
[RFC v2 1/2] eal: add pointer compression functions
Add a new utility header for compressing pointers. The provided
functions can store pointers in 32-bit offsets.
The compression takes advantage of the fact that pointers are
usually located in a limited memory region (like a mempool).
We can compress them by converting them to offsets from a base
memory address. Offsets can be stored in fewer bytes (dictated
by the memory region size and alignment of the pointer).
For example: an 8 byte aligned pointer which is part of a 32GB
memory pool can be stored in 4 bytes.
Suggested-by: Honnappa Nagarahalli
Signed-off-by: Paul Szczepanek
Signed-off-by: Kamalakshitha Aligeri
Reviewed-by: Honnappa Nagarahalli
---
.mailmap | 1 +
lib/eal/include/meson.build| 1 +
lib/eal/include/rte_ptr_compress.h | 160 +
3 files changed, 162 insertions(+)
create mode 100644 lib/eal/include/rte_ptr_compress.h
diff --git a/.mailmap b/.mailmap
index 864d33ee46..3f0c9d32f5 100644
--- a/.mailmap
+++ b/.mailmap
@@ -1058,6 +1058,7 @@ Paul Greenwalt
Paulis Gributs
Paul Luse
Paul M Stillwell Jr
+Paul Szczepanek
Pavan Kumar Linga
Pavan Nikhilesh
Pavel Belous
diff --git a/lib/eal/include/meson.build b/lib/eal/include/meson.build
index a0463efac7..17d8373648 100644
--- a/lib/eal/include/meson.build
+++ b/lib/eal/include/meson.build
@@ -36,6 +36,7 @@ headers += files(
'rte_pci_dev_features.h',
'rte_per_lcore.h',
'rte_pflock.h',
+ 'rte_ptr_compress.h',
'rte_random.h',
'rte_reciprocal.h',
'rte_seqcount.h',
diff --git a/lib/eal/include/rte_ptr_compress.h
b/lib/eal/include/rte_ptr_compress.h
new file mode 100644
index 00..73bde22973
--- /dev/null
+++ b/lib/eal/include/rte_ptr_compress.h
@@ -0,0 +1,160 @@
+/* SPDX-License-Identifier: BSD-shift-Clause
+ * Copyright(c) 2023 Arm Limited
+ */
+
+#ifndef RTE_PTR_COMPRESS_H
+#define RTE_PTR_COMPRESS_H
+
+/**
+ * @file
+ * Pointer compression and decompression.
+ */
+
+#include
+#include
+
+#include
+#include
+#include
+#include
+
+#ifdef __cplusplus
+extern "C" {
+#endif
+
+/**
+ * Compress pointers into 32-bit offsets from base pointer.
+ *
+ * @note It is programmer's responsibility to ensure the resulting offsets fit
+ * into 32 bits. Alignment of the structures pointed to by the pointers allows
+ * us to drop bits from the offsets. This is controlled by the bit_shift
+ * parameter. This means that if structures are aligned by 8 bytes they must be
+ * within 32GB of the base pointer. If there is no such alignment guarantee
they
+ * must be within 4GB.
+ *
+ * @param ptr_base
+ * A pointer used to calculate offsets of pointers in src_table.
+ * @param src_table
+ * A pointer to an array of pointers.
+ * @param dest_table
+ * A pointer to an array of compressed pointers returned by this function.
+ * @param n
+ * The number of objects to compress, must be strictly positive.
+ * @param bit_shift
+ * Byte alignment of memory pointed to by the pointers allows for
+ * bits to be dropped from the offset and hence widen the memory region that
+ * can be covered. This controls how many bits are right shifted.
+ **/
+static __rte_always_inline void
+rte_ptr_compress_32(void *ptr_base, void **src_table,
+ uint32_t *dest_table, unsigned int n, unsigned int bit_shift)
+{
+ unsigned int i = 0;
+#if defined RTE_HAS_SVE_ACLE
+ svuint64_t v_src_table;
+ svuint64_t v_dest_table;
+ svbool_t pg = svwhilelt_b64(i, n);
+ do {
+ v_src_table = svld1_u64(pg, (uint64_t *)src_table + i);
+ v_dest_table = svsub_x(pg, v_src_table, (uint64_t)ptr_base);
+ v_dest_table = svlsr_x(pg, v_dest_table, bit_shift);
+ svst1w(pg, &dest_table[i], v_dest_table);
+ i += svcntd();
+ pg = svwhilelt_b64(i, n);
+ } while (svptest_any(svptrue_b64(), pg));
+#elif defined __ARM_NEON
+ uint64_t ptr_diff;
+ uint64x2_t v_src_table;
+ uint64x2_t v_dest_table;
+ /* right shift is done by left shifting by negative int */
+ int64x2_t v_shift = vdupq_n_s64(-bit_shift);
+ uint64x2_t v_ptr_base = vdupq_n_u64((uint64_t)ptr_base);
+ for (; i < (n & ~0x1); i += 2) {
+ v_src_table = vld1q_u64((const uint64_t *)src_table + i);
+ v_dest_table = vsubq_u64(v_src_table, v_ptr_base);
+ v_dest_table = vshlq_u64(v_dest_table, v_shift);
+ vst1_u32(dest_table + i, vqmovn_u64(v_dest_table));
+ }
+ /* process leftover single item in case of odd number of n */
+ if (unlikely(n & 0x1)) {
+ ptr_diff = RTE_PTR_DIFF(src_table[i], ptr_base);
+ dest_table[i] = (uint32_t) (ptr_diff >> bit_shift);
+ }
+#else
+ uintptr_t ptr_diff;
+ for (; i < n; i++) {
+ ptr_diff = RTE_PTR_DIFF(src_table[i], ptr_base);
+ /* save extra bits that are redundant du
[RFC v2 2/2] test: add pointer compress tests to ring perf test
Add a test that runs a zero copy burst enqueue and dequeue on a ring
of raw pointers and compressed pointers at different burst sizes to
showcase performance benefits of newly added pointer compression APIs.
Refactored threading code to pass more parameters to threads to
reuse existing code. Added more bulk sizes to showcase their effects
on compression. Adjusted loop iteration numbers to take into account
bulk sizes to keep runtime constant (instead of number of operations).
Adjusted old printfs to match new ones which have aligned numbers.
Signed-off-by: Paul Szczepanek
Reviewed-by: Honnappa Nagarahalli
---
app/test/test_ring.h | 59 ++-
app/test/test_ring_perf.c | 324 +++---
2 files changed, 259 insertions(+), 124 deletions(-)
diff --git a/app/test/test_ring.h b/app/test/test_ring.h
index 45c263f3ff..e8b7525c23 100644
--- a/app/test/test_ring.h
+++ b/app/test/test_ring.h
@@ -1,10 +1,12 @@
/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright(c) 2019 Arm Limited
+ * Copyright(c) 2019-2023 Arm Limited
*/
#include
#include
#include
+#include
+#include
/* API type to call
* rte_ring__enqueue_
@@ -25,6 +27,9 @@
#define TEST_RING_ELEM_BULK 16
#define TEST_RING_ELEM_BURST 32
+#define TEST_RING_ELEM_BURST_ZC 64
+#define TEST_RING_ELEM_BURST_ZC_COMPRESS_PTR_32 128
+
#define TEST_RING_IGNORE_API_TYPE ~0U
/* This function is placed here as it is required for both
@@ -101,6 +106,9 @@ static inline unsigned int
test_ring_enqueue(struct rte_ring *r, void **obj, int esize, unsigned int n,
unsigned int api_type)
{
+ unsigned int ret;
+ struct rte_ring_zc_data zcd = {0};
+
/* Legacy queue APIs? */
if (esize == -1)
switch (api_type) {
@@ -152,6 +160,29 @@ test_ring_enqueue(struct rte_ring *r, void **obj, int
esize, unsigned int n,
case (TEST_RING_THREAD_MPMC | TEST_RING_ELEM_BURST):
return rte_ring_mp_enqueue_burst_elem(r, obj, esize, n,
NULL);
+ case (TEST_RING_ELEM_BURST_ZC):
+ ret = rte_ring_enqueue_zc_burst_elem_start(
+ r, esize, n, &zcd, NULL);
+ if (unlikely(ret == 0))
+ return 0;
+ rte_memcpy(zcd.ptr1, (char *)obj, zcd.n1 * esize);
+ if (unlikely(zcd.ptr2 != NULL))
+ rte_memcpy(zcd.ptr2,
+ (char *)obj + zcd.n1 * esize,
+ (ret - zcd.n1) * esize);
+ rte_ring_enqueue_zc_finish(r, ret);
+ return ret;
+ case (TEST_RING_ELEM_BURST_ZC_COMPRESS_PTR_32):
+ ret = rte_ring_enqueue_zc_burst_elem_start(
+ r, sizeof(uint32_t), n, &zcd, NULL);
+ if (unlikely(ret == 0))
+ return 0;
+ rte_ptr_compress_32(0, obj, zcd.ptr1, zcd.n1, 3);
+ if (unlikely(zcd.ptr2 != NULL))
+ rte_ptr_compress_32(0, obj + zcd.n1,
+ zcd.ptr2, ret - zcd.n1, 3);
+ rte_ring_enqueue_zc_finish(r, ret);
+ return ret;
default:
printf("Invalid API type\n");
return 0;
@@ -162,6 +193,9 @@ static inline unsigned int
test_ring_dequeue(struct rte_ring *r, void **obj, int esize, unsigned int n,
unsigned int api_type)
{
+ unsigned int ret;
+ struct rte_ring_zc_data zcd = {0};
+
/* Legacy queue APIs? */
if (esize == -1)
switch (api_type) {
@@ -213,6 +247,29 @@ test_ring_dequeue(struct rte_ring *r, void **obj, int
esize, unsigned int n,
case (TEST_RING_THREAD_MPMC | TEST_RING_ELEM_BURST):
return rte_ring_mc_dequeue_burst_elem(r, obj, esize,
n, NULL);
+ case (TEST_RING_ELEM_BURST_ZC):
+ ret = rte_ring_dequeue_zc_burst_elem_start(
+ r, esize, n, &zcd, NULL);
+ if (unlikely(ret == 0))
+ return 0;
+ rte_memcpy((char *)obj, zcd.ptr1, zcd.n1 * esize);
+ if (unlikely(zcd.ptr2 != NULL))
+ rte_memcpy((char *)obj + zcd.n1 * esize,
+ zcd.ptr2,
+ (ret - zcd.n1) * esize);
+ rte_ring_dequeue_zc_finish(r, ret);
+ return ret;
+ case (TEST_RING_ELEM_
Re: [PATCH v4 0/3] Release ethdev shared memory on port cleanup
27/09/2023 13:45, David Marchand: > This series was triggered after investigating why the > eal_flags_file_prefix_autotest unit test was failing in the case of > statically built binaries [1]). > > For now, I went with a simple (naive) approach and put all accesses to the > shared data under a single lock: ethdev maintainers, it is your turn to > shine and give me reasons why we should keep the locks the way they > were ;-). > And let's see what the CI reports... Applied, thanks.
[PATCH v3 0/3] rewrite fastpath routines
This series adds new fastpath routines for cn10k & cn9k endpoint devices and supports 32B Tx desciptor format which improves the performance. v2 & v3 changes: - Fixed CI Shijith Thotton (1): net/octeon_ep: support 32B IQ descriptor size Vamsi Attunuru (2): net/octeon_ep: clean up receive routine net/octeon_ep: add new fastpath routines drivers/net/octeon_ep/cnxk_ep_rx.c| 309 ++ drivers/net/octeon_ep/cnxk_ep_tx.c| 209 + drivers/net/octeon_ep/cnxk_ep_vf.c| 12 +- drivers/net/octeon_ep/cnxk_ep_vf.h| 13 ++ drivers/net/octeon_ep/meson.build | 2 + drivers/net/octeon_ep/otx2_ep_vf.c| 11 +- drivers/net/octeon_ep/otx_ep_common.h | 127 ++- drivers/net/octeon_ep/otx_ep_ethdev.c | 69 +- drivers/net/octeon_ep/otx_ep_rxtx.c | 255 +++-- drivers/net/octeon_ep/otx_ep_rxtx.h | 38 +++- drivers/net/octeon_ep/otx_ep_vf.c | 8 + 11 files changed, 801 insertions(+), 252 deletions(-) create mode 100644 drivers/net/octeon_ep/cnxk_ep_rx.c create mode 100644 drivers/net/octeon_ep/cnxk_ep_tx.c -- 2.25.1
[PATCH v3 1/3] net/octeon_ep: support 32B IQ descriptor size
From: Shijith Thotton
Update input queue setup to consider descriptor size in driver conf.
The default instruction size for otx2 and cnxk devices has been updated
to 32 bytes.
Signed-off-by: Shijith Thotton
---
drivers/net/octeon_ep/cnxk_ep_vf.c| 10 +-
drivers/net/octeon_ep/otx2_ep_vf.c| 10 +-
drivers/net/octeon_ep/otx_ep_common.h | 4
drivers/net/octeon_ep/otx_ep_vf.c | 8
4 files changed, 30 insertions(+), 2 deletions(-)
diff --git a/drivers/net/octeon_ep/cnxk_ep_vf.c
b/drivers/net/octeon_ep/cnxk_ep_vf.c
index 92c2d2ca5c..7b3669fe0c 100644
--- a/drivers/net/octeon_ep/cnxk_ep_vf.c
+++ b/drivers/net/octeon_ep/cnxk_ep_vf.c
@@ -106,6 +106,14 @@ cnxk_ep_vf_setup_iq_regs(struct otx_ep_device *otx_ep,
uint32_t iq_no)
return -EIO;
}
+ /* Configure input queue instruction size. */
+ if (otx_ep->conf->iq.instr_type == OTX_EP_32BYTE_INSTR)
+ reg_val &= ~(CNXK_EP_R_IN_CTL_IS_64B);
+ else
+ reg_val |= CNXK_EP_R_IN_CTL_IS_64B;
+ oct_ep_write64(reg_val, otx_ep->hw_addr + CNXK_EP_R_IN_CONTROL(iq_no));
+ iq->desc_size = otx_ep->conf->iq.instr_type;
+
/* Write the start of the input queue's ring and its size */
oct_ep_write64(iq->base_addr_dma, otx_ep->hw_addr +
CNXK_EP_R_IN_INSTR_BADDR(iq_no));
oct_ep_write64(iq->nb_desc, otx_ep->hw_addr +
CNXK_EP_R_IN_INSTR_RSIZE(iq_no));
@@ -354,7 +362,7 @@ static const struct otx_ep_config default_cnxk_ep_conf = {
/* IQ attributes */
.iq= {
.max_iqs = OTX_EP_CFG_IO_QUEUES,
- .instr_type= OTX_EP_64BYTE_INSTR,
+ .instr_type= OTX_EP_32BYTE_INSTR,
.pending_list_size = (OTX_EP_MAX_IQ_DESCRIPTORS *
OTX_EP_CFG_IO_QUEUES),
},
diff --git a/drivers/net/octeon_ep/otx2_ep_vf.c
b/drivers/net/octeon_ep/otx2_ep_vf.c
index ced3a415a5..f72b8d25d7 100644
--- a/drivers/net/octeon_ep/otx2_ep_vf.c
+++ b/drivers/net/octeon_ep/otx2_ep_vf.c
@@ -256,6 +256,14 @@ otx2_vf_setup_iq_regs(struct otx_ep_device *otx_ep,
uint32_t iq_no)
return -EIO;
}
+ /* Configure input queue instruction size. */
+ if (otx_ep->conf->iq.instr_type == OTX_EP_32BYTE_INSTR)
+ reg_val &= ~(SDP_VF_R_IN_CTL_IS_64B);
+ else
+ reg_val |= SDP_VF_R_IN_CTL_IS_64B;
+ oct_ep_write64(reg_val, otx_ep->hw_addr + SDP_VF_R_IN_CONTROL(iq_no));
+ iq->desc_size = otx_ep->conf->iq.instr_type;
+
/* Write the start of the input queue's ring and its size */
oct_ep_write64(iq->base_addr_dma, otx_ep->hw_addr +
SDP_VF_R_IN_INSTR_BADDR(iq_no));
oct_ep_write64(iq->nb_desc, otx_ep->hw_addr +
SDP_VF_R_IN_INSTR_RSIZE(iq_no));
@@ -500,7 +508,7 @@ static const struct otx_ep_config default_otx2_ep_conf = {
/* IQ attributes */
.iq= {
.max_iqs = OTX_EP_CFG_IO_QUEUES,
- .instr_type= OTX_EP_64BYTE_INSTR,
+ .instr_type= OTX_EP_32BYTE_INSTR,
.pending_list_size = (OTX_EP_MAX_IQ_DESCRIPTORS *
OTX_EP_CFG_IO_QUEUES),
},
diff --git a/drivers/net/octeon_ep/otx_ep_common.h
b/drivers/net/octeon_ep/otx_ep_common.h
index c150cbe619..90e059cad0 100644
--- a/drivers/net/octeon_ep/otx_ep_common.h
+++ b/drivers/net/octeon_ep/otx_ep_common.h
@@ -11,6 +11,7 @@
#define OTX_EP_MAX_RINGS_PER_VF(8)
#define OTX_EP_CFG_IO_QUEUESOTX_EP_MAX_RINGS_PER_VF
+#define OTX_EP_32BYTE_INSTR (32)
#define OTX_EP_64BYTE_INSTR (64)
/*
* Backpressure for SDP is configured on Octeon, and the minimum queue sizes
@@ -215,6 +216,9 @@ struct otx_ep_instr_queue {
/* Number of descriptors in this ring. */
uint32_t nb_desc;
+ /* Size of the descriptor. */
+ uint8_t desc_size;
+
/* Input ring index, where the driver should write the next packet */
uint32_t host_write_index;
diff --git a/drivers/net/octeon_ep/otx_ep_vf.c
b/drivers/net/octeon_ep/otx_ep_vf.c
index 4f3538146b..236b7a874c 100644
--- a/drivers/net/octeon_ep/otx_ep_vf.c
+++ b/drivers/net/octeon_ep/otx_ep_vf.c
@@ -120,6 +120,14 @@ otx_ep_setup_iq_regs(struct otx_ep_device *otx_ep,
uint32_t iq_no)
return -EIO;
}
+ /* Configure input queue instruction size. */
+ if (iq->desc_size == OTX_EP_32BYTE_INSTR)
+ reg_val &= ~(OTX_EP_R_IN_CTL_IS_64B);
+ else
+ reg_val |= OTX_EP_R_IN_CTL_IS_64B;
+ oct_ep_write64(reg_val, otx_ep->hw_addr + OTX_EP_R_IN_CONTROL(iq_no));
+ iq->desc_size = otx_ep->conf->iq.instr_type;
+
/* Write the start of the input queue's ring and its size */
otx_ep_write64(iq->base_addr_dma, otx_ep->hw_addr,
OTX_EP_R_IN_INSTR_
[PATCH v3 2/3] net/octeon_ep: clean up receive routine
Patch improves Rx routine and pkt count update routines,
packet count update routines need to drain inflight ISM
memory updates while decrementing the packet count register.
Signed-off-by: Vamsi Attunuru
---
drivers/net/octeon_ep/otx_ep_rxtx.c | 162
1 file changed, 68 insertions(+), 94 deletions(-)
diff --git a/drivers/net/octeon_ep/otx_ep_rxtx.c
b/drivers/net/octeon_ep/otx_ep_rxtx.c
index b37fc8109f..4c509a419f 100644
--- a/drivers/net/octeon_ep/otx_ep_rxtx.c
+++ b/drivers/net/octeon_ep/otx_ep_rxtx.c
@@ -442,7 +442,14 @@ otx_vf_update_read_index(struct otx_ep_instr_queue *iq)
* when count above halfway to saturation.
*/
rte_write32(val, iq->inst_cnt_reg);
- *iq->inst_cnt_ism = 0;
+ rte_mb();
+
+ rte_write64(OTX2_SDP_REQUEST_ISM, iq->inst_cnt_reg);
+ while (__atomic_load_n(iq->inst_cnt_ism, __ATOMIC_RELAXED) >=
val) {
+ rte_write64(OTX2_SDP_REQUEST_ISM, iq->inst_cnt_reg);
+ rte_mb();
+ }
+
iq->inst_cnt_ism_prev = 0;
}
rte_write64(OTX2_SDP_REQUEST_ISM, iq->inst_cnt_reg);
@@ -567,9 +574,7 @@ prepare_xmit_gather_list(struct otx_ep_instr_queue *iq,
struct rte_mbuf *m, uint
finfo = &iq->req_list[iq->host_write_index].finfo;
*dptr = rte_mem_virt2iova(finfo->g.sg);
- ih->s.tlen = pkt_len + ih->s.fsz;
- ih->s.gsz = frags;
- ih->s.gather = 1;
+ ih->u64 |= ((1ULL << 62) | ((uint64_t)frags << 48) | (pkt_len +
ih->s.fsz));
while (frags--) {
finfo->g.sg[(j >> 2)].ptr[(j & mask)] = rte_mbuf_data_iova(m);
@@ -752,36 +757,26 @@ otx2_ep_xmit_pkts(void *tx_queue, struct rte_mbuf **pkts,
uint16_t nb_pkts)
static uint32_t
otx_ep_droq_refill(struct otx_ep_droq *droq)
{
- struct otx_ep_droq_desc *desc_ring;
+ struct otx_ep_droq_desc *desc_ring = droq->desc_ring;
struct otx_ep_droq_info *info;
struct rte_mbuf *buf = NULL;
uint32_t desc_refilled = 0;
- desc_ring = droq->desc_ring;
-
while (droq->refill_count && (desc_refilled < droq->nb_desc)) {
- /* If a valid buffer exists (happens if there is no dispatch),
-* reuse the buffer, else allocate.
-*/
- if (droq->recv_buf_list[droq->refill_idx] != NULL)
- break;
-
buf = rte_pktmbuf_alloc(droq->mpool);
/* If a buffer could not be allocated, no point in
* continuing
*/
- if (buf == NULL) {
+ if (unlikely(!buf)) {
droq->stats.rx_alloc_failure++;
break;
}
info = rte_pktmbuf_mtod(buf, struct otx_ep_droq_info *);
- memset(info, 0, sizeof(*info));
+ info->length = 0;
droq->recv_buf_list[droq->refill_idx] = buf;
desc_ring[droq->refill_idx].buffer_ptr =
rte_mbuf_data_iova_default(buf);
-
-
droq->refill_idx = otx_ep_incr_index(droq->refill_idx, 1,
droq->nb_desc);
@@ -793,21 +788,18 @@ otx_ep_droq_refill(struct otx_ep_droq *droq)
}
static struct rte_mbuf *
-otx_ep_droq_read_packet(struct otx_ep_device *otx_ep,
- struct otx_ep_droq *droq, int next_fetch)
+otx_ep_droq_read_packet(struct otx_ep_device *otx_ep, struct otx_ep_droq
*droq, int next_fetch)
{
volatile struct otx_ep_droq_info *info;
- struct rte_mbuf *droq_pkt2 = NULL;
- struct rte_mbuf *droq_pkt = NULL;
- struct rte_net_hdr_lens hdr_lens;
- struct otx_ep_droq_info *info2;
+ struct rte_mbuf *mbuf_next = NULL;
+ struct rte_mbuf *mbuf = NULL;
uint64_t total_pkt_len;
uint32_t pkt_len = 0;
int next_idx;
- droq_pkt = droq->recv_buf_list[droq->read_idx];
- droq_pkt2 = droq->recv_buf_list[droq->read_idx];
- info = rte_pktmbuf_mtod(droq_pkt, struct otx_ep_droq_info *);
+ mbuf = droq->recv_buf_list[droq->read_idx];
+ info = rte_pktmbuf_mtod(mbuf, struct otx_ep_droq_info *);
+
/* make sure info is available */
rte_rmb();
if (unlikely(!info->length)) {
@@ -828,32 +820,25 @@ otx_ep_droq_read_packet(struct otx_ep_device *otx_ep,
assert(0);
}
}
+
if (next_fetch) {
next_idx = otx_ep_incr_index(droq->read_idx, 1, droq->nb_desc);
- droq_pkt2 = droq->recv_buf_list[next_idx];
- info2 = rte_pktmbuf_mtod(droq_pkt2, struct otx_ep_droq_info *);
- rte_prefetch_non_temporal((const void *)info2);
+ mbuf_next = droq->recv_buf_list[next_idx];
+ rte_prefetch0(rte_pktmbuf_mtod(mbuf_next, void *));
}
- info->
[PATCH v3 3/3] net/octeon_ep: add new fastpath routines
Adds new fastpath routines for cn10k & cn9k endpoint
devices and assigns the fastpath routines based on
the offload flags.
Patch also adds misc changes to improve performance
and code-readability.
Signed-off-by: Vamsi Attunuru
---
drivers/net/octeon_ep/cnxk_ep_rx.c| 309 ++
drivers/net/octeon_ep/cnxk_ep_tx.c| 209 +
drivers/net/octeon_ep/cnxk_ep_vf.c| 2 +
drivers/net/octeon_ep/cnxk_ep_vf.h| 13 ++
drivers/net/octeon_ep/meson.build | 2 +
drivers/net/octeon_ep/otx2_ep_vf.c| 1 +
drivers/net/octeon_ep/otx_ep_common.h | 125 ++-
drivers/net/octeon_ep/otx_ep_ethdev.c | 69 +-
drivers/net/octeon_ep/otx_ep_rxtx.c | 93 +---
drivers/net/octeon_ep/otx_ep_rxtx.h | 38 +++-
10 files changed, 704 insertions(+), 157 deletions(-)
diff --git a/drivers/net/octeon_ep/cnxk_ep_rx.c
b/drivers/net/octeon_ep/cnxk_ep_rx.c
new file mode 100644
index 00..74f0011283
--- /dev/null
+++ b/drivers/net/octeon_ep/cnxk_ep_rx.c
@@ -0,0 +1,309 @@
+/* SPDX-License-Identifier: BSD-3-Clause
+ * Copyright(C) 2023 Marvell.
+ */
+
+#include "otx_ep_common.h"
+#include "otx2_ep_vf.h"
+#include "otx_ep_rxtx.h"
+
+static inline int
+cnxk_ep_rx_refill_mbuf(struct otx_ep_droq *droq, uint32_t count)
+{
+ struct otx_ep_droq_desc *desc_ring = droq->desc_ring;
+ struct rte_mbuf **recv_buf_list = droq->recv_buf_list;
+ uint32_t refill_idx = droq->refill_idx;
+ struct rte_mbuf *buf;
+ uint32_t i;
+ int rc;
+
+ rc = rte_pktmbuf_alloc_bulk(droq->mpool, &recv_buf_list[refill_idx],
count);
+ if (unlikely(rc)) {
+ droq->stats.rx_alloc_failure++;
+ return rc;
+ }
+
+ for (i = 0; i < count; i++) {
+ buf = recv_buf_list[refill_idx];
+ desc_ring[refill_idx].buffer_ptr =
rte_mbuf_data_iova_default(buf);
+ refill_idx++;
+ }
+
+ droq->refill_idx = otx_ep_incr_index(droq->refill_idx, count,
droq->nb_desc);
+ droq->refill_count -= count;
+
+ return 0;
+}
+
+static inline void
+cnxk_ep_rx_refill(struct otx_ep_droq *droq)
+{
+ uint32_t desc_refilled = 0, count;
+ uint32_t nb_desc = droq->nb_desc;
+ uint32_t refill_idx = droq->refill_idx;
+ int rc;
+
+ if (unlikely(droq->read_idx == refill_idx))
+ return;
+
+ if (refill_idx < droq->read_idx) {
+ count = droq->read_idx - refill_idx;
+ rc = cnxk_ep_rx_refill_mbuf(droq, count);
+ if (unlikely(rc)) {
+ droq->stats.rx_alloc_failure++;
+ return;
+ }
+ desc_refilled = count;
+ } else {
+ count = nb_desc - refill_idx;
+ rc = cnxk_ep_rx_refill_mbuf(droq, count);
+ if (unlikely(rc)) {
+ droq->stats.rx_alloc_failure++;
+ return;
+ }
+
+ desc_refilled = count;
+ count = droq->read_idx;
+ rc = cnxk_ep_rx_refill_mbuf(droq, count);
+ if (unlikely(rc)) {
+ droq->stats.rx_alloc_failure++;
+ return;
+ }
+ desc_refilled += count;
+ }
+
+ /* Flush the droq descriptor data to memory to be sure
+* that when we update the credits the data in memory is
+* accurate.
+*/
+ rte_io_wmb();
+ rte_write32(desc_refilled, droq->pkts_credit_reg);
+}
+
+static inline uint32_t
+cnxk_ep_check_rx_pkts(struct otx_ep_droq *droq)
+{
+ uint32_t new_pkts;
+ uint32_t val;
+
+ /* Batch subtractions from the HW counter to reduce PCIe traffic
+* This adds an extra local variable, but almost halves the
+* number of PCIe writes.
+*/
+ val = __atomic_load_n(droq->pkts_sent_ism, __ATOMIC_RELAXED);
+ new_pkts = val - droq->pkts_sent_ism_prev;
+ droq->pkts_sent_ism_prev = val;
+
+ if (val > (uint32_t)(1 << 31)) {
+ /* Only subtract the packet count in the HW counter
+* when count above halfway to saturation.
+*/
+ rte_write64((uint64_t)val, droq->pkts_sent_reg);
+ rte_mb();
+
+ rte_write64(OTX2_SDP_REQUEST_ISM, droq->pkts_sent_reg);
+ while (__atomic_load_n(droq->pkts_sent_ism, __ATOMIC_RELAXED)
>= val) {
+ rte_write64(OTX2_SDP_REQUEST_ISM, droq->pkts_sent_reg);
+ rte_mb();
+ }
+
+ droq->pkts_sent_ism_prev = 0;
+ }
+ rte_write64(OTX2_SDP_REQUEST_ISM, droq->pkts_sent_reg);
+ droq->pkts_pending += new_pkts;
+
+ return new_pkts;
+}
+
+static inline int16_t __rte_hot
+cnxk_ep_rx_pkts_to_process(struct otx_ep_droq *droq, uint16_t nb_pkts)
+{
+ if (droq->pkts_pending < nb_pkts)
+ cnxk_ep_check_rx_pkts(d
RE: [PATCH v4] app/testpmd: enable cli for programmable action
> -Original Message-
> From: Ferruh Yigit
> Sent: Wednesday, October 11, 2023 6:21 PM
> To: Zhang, Qi Z ; Singh, Aman Deep
> ; Zhang, Yuying
> Cc: dev@dpdk.org; Dumitrescu, Cristian ;
> or...@nvidia.com
> Subject: Re: [PATCH v4] app/testpmd: enable cli for programmable action
>
> On 10/11/2023 3:24 AM, Zhang, Qi Z wrote:
> >
> >> -Original Message-
> >> From: Ferruh Yigit
> >> Sent: Tuesday, October 10, 2023 6:49 PM
> >> To: Zhang, Qi Z ; Singh, Aman Deep
> >> ; Zhang, Yuying
> >> Cc: dev@dpdk.org; Dumitrescu, Cristian
> >> ; or...@nvidia.com
> >> Subject: Re: [PATCH v4] app/testpmd: enable cli for programmable
> >> action
> >>
> >> On 10/7/2023 11:47 AM, Qi Zhang wrote:
> >>> Parsing command line for rte_flow_action_prog.
> >>>
> >>> Syntax:
> >>>
> >>> "prog name [arguments \
> >>> ... end]"
> >>>
> >>
> >> Can you please put full rte flow command in the commit log? Like what
> >> is the 'pattern' for above command?
> >
> > The pattern part should be independent of the action part,
> >
> > though for our P4 device, we will prefer use rte_flow_flex_item, something
> like:
> >
> > flow create 0 pattern flex item is xxx pattern is xxx / flex item is xxx
> > pattern
> is / actions prog name ..
> >
> > but it does not limit PMD to support flow like below
> >
>
> I think agreement was to use flex pattern, and my understand is "struct
> rte_flow_item_flex" will be used to present the table_id.
>
> Without not using flex, how driver will detect which table to update?
>
>
> > flow create 0 pattern eth / ipv4 src is 1.1.1.1 / actions prog name ..
> >
> > So I think it may not be necessary to highlight the pattern format here.
> >
>
> Complete samples helps a lot to user, can you please include the full rte flow
> command, you can have flex pattern sample and if you want add more
> samples with other patterns but we need to clarify it first.
Agree, I have added full sample on v6 in testpmd document as below:
...
A rule use Programmable Action to perform a customized tunnel header encap
for specific IP packets
testpmd> flow create 0 ingress pattern eth / ipv4 src is 1.2.3.4 / end
actions prog name cust_tun_encap arguments tunn_id 55AA meta0 2E meta1 9000 end
/ end"
...
The reason I did not include a sample with a flex item is that, in the context
of our P4 device, there is no requirement to utilize the
'rte_flow_item_flex_handle' within 'rte_flow_item_flex.'
However, the current testpmd does not offer support for this configuration.
Therefore, it may be necessary to introduce a new syntax, such as ... pattern
flex pattern is / flex pattern is xxx / end ..., which would also map to
'rte_flow_item_flex'.
>
>
> >>
> >>
> >>> Use parse_string0 to parse name string.
> >>> Use parse_hex to parse hex string.
> >>> Use struct action_prog_data to store parsed result.
> >>>
> >>> Example:
> >>>
> >>> Action with 2 arguments:
> >>>
> >>> "prog name action0 arguments field0 03FF field1 55AA end"
> >>>
> >>> Action without argument:
> >>>
> >>> "prog name action1"
> >>>
> >>> Signed-off-by: Qi Zhang
> >>>
> >>
> >> Is there an existing driver implementation, checking it helps to
> >> understand feature implementation?
> >
> > This work is still ongoing, currently we target to upstream on DPDK
> > 24.03
> >
>
> If you won't have driver yet, do you have a way to test these commands?
> Or is this implementation just theoretical at this stage?
Yes, internally, we are very close to have an implementation that will leverage
this new API,
The JSON file loaded by CPFL PMD contains the mapping rule that direct PMD how
to handling
an action_prog, currently we didn't see any gap of the new API.
>
>
> >>
> >>
> >>> ---
> >>>
> >>> v4:
> >>> - be more generous on the max size of name and value.
> >>>
> >>> v3:
> >>> - refine struct action_prog_data
> >>> - enlarge the max size
> >>>
> >>> v2:
> >>> - fix title
> >>> - minor coding style refine.
> >>>
> >>> app/test-pmd/cmdline_flow.c | 232
> >>>
> >>> 1 file changed, 232 insertions(+)
> >>>
> >>
> >> Hi Qi,
> >>
> >> Can you please update documentation too,
> >> `doc/guides/testpmd_app_ug/testpmd_funcs.rst`, `Flow rules
> >> management` section.
> >
> > Sure.
> >
> >>
> >>
> >>> diff --git a/app/test-pmd/cmdline_flow.c
> >>> b/app/test-pmd/cmdline_flow.c index 21828c144c..ae5556e704 100644
> >>> --- a/app/test-pmd/cmdline_flow.c
> >>> +++ b/app/test-pmd/cmdline_flow.c
> >>> @@ -719,6 +719,13 @@ enum index {
> >>> ACTION_IPV6_EXT_PUSH,
> >>> ACTION_IPV6_EXT_PUSH_INDEX,
> >>> ACTION_IPV6_EXT_PUSH_INDEX_VALUE,
> >>> + ACTION_PROG,
> >>> + ACTION_PROG_NAME,
> >>> + ACTION_PROG_NAME_STRING,
> >>> + ACTION_PROG_ARGUMENTS,
> >>> + ACTION_PROG_ARG_NAME,
> >>> + ACTION_PROG_ARG_VALUE,
> >>> + ACTION_PROG_ARG_END,
> >>> };
> >>>
> >>> /** Maximum size for pattern in struct rte_flow_item_raw. */ @@
> >>> -749,6 +756,23 @@ struct action_rss_data {
> >>> uint16_t queue[ACTION
[PATCH v3 0/5] document and simplify use of cmdline
The DPDK commandline library is widely used by apps and examples within DPDK, but it is not documented in our programmers guide and it requires a lot of boilerplate code definitions in order to used. We can improve this situation by creating a simple python script to automatically generate the boilerplate from a list of commands. This patchset contains a new documentation chapter on cmdline library, going through step-by-step how to add commands and create the necessary token lists and parse contexts. Following that initial doc patch, the set then contains a boilerplate-generating script, as well as a set of three patches showing its use, by converting three examples to use the script instead of having the hard-coded boilerplate. Once the script is used, adding a new command becomes as simple as adding the desired command to the .list file, and then writing the required function which will be called for that command. No other boilerplate coding is necessary. Script obviously does not cover the full range of capabilities of the commandline lib, but does cover the most used parts. The code-saving to each of the examples by auto-generating the boilerplate is significant, and probably more examples with commandlines can be converted over in future. The "cmdline" example itself, is not converted over, as it should probably remain as a simple example of direct library use without the script. V3: * Added lots of documentation * Added support for help text for each command * Cleaned up script a little so it passes pycodestyle and most flake8 checks, when line-length is set to max 100. * Removed RFC tag, as I consider this patchset stable enough for consideration in a release. V2-RFC: * Add support for IP addresses in commands * Move to buildtools directory and make installable * Convert 3 examples to use script, and eliminate their boilerplate Bruce Richardson (5): doc/prog_guide: new chapter on cmdline library buildtools: script to generate cmdline boilerplate examples/simple_mp: auto-generate cmdline boilerplate examples/hotplug_mp: auto-generate cmdline boilerplate examples/bond: auto-generate cmdline boilerplate app/test/commands.c | 2 + buildtools/dpdk-cmdline-gen.py| 167 +++ buildtools/meson.build| 7 + doc/guides/prog_guide/cmdline.rst | 466 ++ doc/guides/prog_guide/index.rst | 1 + examples/bond/Makefile| 12 +- examples/bond/commands.list | 6 + examples/bond/main.c | 161 +- examples/bond/main.h | 10 - examples/bond/meson.build | 8 + examples/multi_process/hotplug_mp/Makefile| 12 +- examples/multi_process/hotplug_mp/commands.c | 147 +- examples/multi_process/hotplug_mp/commands.h | 10 - .../multi_process/hotplug_mp/commands.list| 5 + examples/multi_process/hotplug_mp/meson.build | 9 + examples/multi_process/simple_mp/Makefile | 12 +- examples/multi_process/simple_mp/meson.build | 9 + .../multi_process/simple_mp/mp_commands.c | 106 +--- .../multi_process/simple_mp/mp_commands.h | 14 - .../multi_process/simple_mp/mp_commands.list | 3 + 20 files changed, 745 insertions(+), 422 deletions(-) create mode 100755 buildtools/dpdk-cmdline-gen.py create mode 100644 doc/guides/prog_guide/cmdline.rst create mode 100644 examples/bond/commands.list delete mode 100644 examples/bond/main.h delete mode 100644 examples/multi_process/hotplug_mp/commands.h create mode 100644 examples/multi_process/hotplug_mp/commands.list delete mode 100644 examples/multi_process/simple_mp/mp_commands.h create mode 100644 examples/multi_process/simple_mp/mp_commands.list -- 2.39.2
[PATCH v3 1/5] doc/prog_guide: new chapter on cmdline library
The cmdline library was not documented in our programmers guide, so add
a new chapter on it. This chapter covers step-by-step how to use the
library, rather than focusing on the library internals. This complements
the existing cmdline example app document, providing more details on the
process of using the library.
Signed-off-by: Bruce Richardson
---
app/test/commands.c | 2 +
doc/guides/prog_guide/cmdline.rst | 337 ++
doc/guides/prog_guide/index.rst | 1 +
3 files changed, 340 insertions(+)
create mode 100644 doc/guides/prog_guide/cmdline.rst
diff --git a/app/test/commands.c b/app/test/commands.c
index 31259e5c21..497d8e9952 100644
--- a/app/test/commands.c
+++ b/app/test/commands.c
@@ -108,6 +108,7 @@ dump_struct_sizes(void)
#undef DUMP_SIZE
}
+/* Add the dump_* tests cases 8< */
static void cmd_dump_parsed(void *parsed_result,
__rte_unused struct cmdline *cl,
__rte_unused void *data)
@@ -155,6 +156,7 @@ cmdline_parse_inst_t cmd_dump = {
NULL,
},
};
+/* >8 End of add the dump_* tests cases */
//
diff --git a/doc/guides/prog_guide/cmdline.rst
b/doc/guides/prog_guide/cmdline.rst
new file mode 100644
index 00..40f49a30cc
--- /dev/null
+++ b/doc/guides/prog_guide/cmdline.rst
@@ -0,0 +1,337 @@
+.. SPDX-License-Identifier: BSD-3-Clause
+Copyright(c) 2023 Intel Corporation.
+
+Command-line Library
+
+
+Since its earliest versions, DPDK has included a command-line library -
+primarily for internal use by, for example, ``dpdk-testpmd`` and the
``dpdk-test`` binaries,
+but the library is also exported on install and can be used by any end
application.
+This chapter covers the basics of the command-line library and how to use it
in an application.
+
+Library Features
+
+
+The DPDK command-line library supports the following features:
+
+* Tab-completion available for interactive terminal sessions
+
+* Ability to read and process commands taken from an input file, e.g. startup
script
+
+* Parameterized commands able to take multiple parameters with different
datatypes:
+
+ * Strings
+ * Signed/unsigned 16/32/64-bit integers
+ * IP Addresses
+ * Ethernet Addresses
+
+* Ability to multiplex multiple commands to a single callback function
+
+Adding Command-line to an Application
+-
+
+Adding a command-line instance to an application involves a number of coding
steps.
+
+1. Define the result structure for the command, specifying the command
parameters
+
+2. Provide an initializer for each field in the result
+
+3. Define the callback function for the command
+
+4. Provide a parse result structure instance for the command, linking the
callback to the command
+
+5. Add the parse result structure to a command-line context
+
+6. Within your main application code, create a new command-line instance
passing in the context.
+
+The next few subsections will cover each of these steps in more detail,
+working through an example to add two commands to a command-line instance.
+Those two commands will be:
+
+1. ``quit`` - as the name suggests, to close the application
+
+2. ``show port stats `` - to display on-screen the statistics for a given
ethernet port
+
+.. note::
+
+ For further examples of use of the command-line, see
+ :doc:`cmdline example application <../sample_app_ug/cmd_line>`
+
+Defining Command Result Structure
+~
+
+The first structure to be defined is the structure which will be created on
successful parse of a command.
+This structure contains one member field for each token, or word, in the
command.
+The simplest case is for a one-word command, like ``quit``.
+For this, we only need to define a structure with a single string parameter to
contain that word.
+
+.. code-block:: c
+
+ struct cmd_quit_result {
+ cmdline_fixed_string_t quit;
+ };
+
+For readability, the name of the struct member should match that of the token
in the command.
+
+For our second command, we need a structure with four member fields in it,
+as there are four words/tokens in our command.
+The first three are strings, and the final one is a 16-bit numeric value.
+The resulting struct looks like:
+
+.. code-block:: c
+
+ struct cmd_show_port_stats_result {
+ cmdline_fixed_string_t show;
+ cmdline_fixed_string_t port;
+ cmdline_fixed_string_t stats;
+ uint16_t n;
+ };
+
+As before, we choose names to match the tokens in the command.
+Since our numeric parameter is a 16-bit value, we use ``uint16_t`` type for it.
+Any of the standard sized integer types can be used as parameters, depending
on the desired result.
+
+Beyond the standard integer types,
+the library also allows variable parameters to be of a number of other types,
+as called out in the feature list above.
+
+* For variable string parameters,
[PATCH v3 2/5] buildtools: script to generate cmdline boilerplate
Provide a "dpdk-cmdline-gen.py" script for application developers to
quickly generate the boilerplate code necessary for using the cmdline
library.
Example of use:
The script takes an input file with a list of commands the user wants in
the app, where the parameter variables are tagged with the type.
For example:
$ cat commands.list
list
add x y
echo message
add socket path
quit
When run through the script as "./dpdk-cmdline-gen.py commands.list",
the output will be the contents of a header file with all the
boilerplate necessary for a commandline instance with those commands.
If the flag --stubs is passed, an output header filename must also be
passed, in which case both a header file with the definitions and a C
file with function stubs in it is written to disk. The separation is so
that the header file can be rewritten at any future point to add more
commands, while the C file can be kept as-is and extended by the user
with any additional functions needed.
Signed-off-by: Bruce Richardson
---
buildtools/dpdk-cmdline-gen.py| 167 ++
buildtools/meson.build| 7 ++
doc/guides/prog_guide/cmdline.rst | 131 ++-
3 files changed, 304 insertions(+), 1 deletion(-)
create mode 100755 buildtools/dpdk-cmdline-gen.py
diff --git a/buildtools/dpdk-cmdline-gen.py b/buildtools/dpdk-cmdline-gen.py
new file mode 100755
index 00..3b41fb0493
--- /dev/null
+++ b/buildtools/dpdk-cmdline-gen.py
@@ -0,0 +1,167 @@
+#!/usr/bin/env python3
+# SPDX-License-Identifier: BSD-3-Clause
+# Copyright(c) 2023 Intel Corporation
+#
+"""Script to automatically generate boilerplate for using DPDK cmdline
library."""
+
+import argparse
+import sys
+
+PARSE_FN_PARAMS = 'void *parsed_result, struct cmdline *cl, void *data'
+PARSE_FN_BODY = """
+/* TODO: command action */
+RTE_SET_USED(parsed_result);
+RTE_SET_USED(cl);
+RTE_SET_USED(data);
+"""
+
+
+def process_command(tokens, cfile, comment):
+"""Generate the structures and definitions for a single command."""
+name = []
+
+if tokens[0].startswith('<'):
+print('Error: each command must start with at least one literal
string', file=sys.stderr)
+sys.exit(1)
+for t in tokens:
+if t.startswith('<'):
+break
+name.append(t)
+name = '_'.join(name)
+
+result_struct = []
+initializers = []
+token_list = []
+for t in tokens:
+if t.startswith('<'):
+t_type, t_name = t[1:].split('>')
+t_val = 'NULL'
+else:
+t_type = 'STRING'
+t_name = t
+t_val = f'"{t}"'
+
+if t_type == 'STRING':
+result_struct.append(f'\tcmdline_fixed_string_t {t_name};')
+initializers.append(
+f'static cmdline_parse_token_string_t
cmd_{name}_{t_name}_tok =\n' +
+f'\tTOKEN_STRING_INITIALIZER(struct cmd_{name}_result,
{t_name}, {t_val});')
+elif t_type in ['UINT8', 'UINT16', 'UINT32', 'UINT64', 'INT8',
'INT16', 'INT32', 'INT64']:
+result_struct.append(f'\t{t_type.lower()}_t {t_name};')
+initializers.append(
+f'static cmdline_parse_token_num_t cmd_{name}_{t_name}_tok
=\n' +
+f'\tTOKEN_NUM_INITIALIZER(struct cmd_{name}_result,
{t_name}, RTE_{t_type});')
+elif t_type in ['IP', 'IP_ADDR', 'IPADDR']:
+result_struct.append(f'\tcmdline_ipaddr_t {t_name};')
+initializers.append(
+f'cmdline_parse_token_ipaddr_t cmd_{name}_{t_name}_tok
=\n' +
+f'\tTOKEN_IPV4_INITIALIZER(struct cmd_{name}_result,
{t_name});')
+else:
+print(f'Error: unknown token-type {t}', file=sys.stderr)
+sys.exit(1)
+token_list.append(f'cmd_{name}_{t_name}_tok')
+
+print(f'/* Auto-generated handling for command "{" ".join(tokens)}" */')
+# output function prototype
+func_sig = f'void\ncmd_{name}_parsed({PARSE_FN_PARAMS})'
+print(f'extern {func_sig};\n')
+# output function template if C file being written
+if (cfile):
+print(f'{func_sig}\n{{{PARSE_FN_BODY}}}\n', file=cfile)
+# output result data structure
+print(
+f'struct cmd_{name}_result {{\n' +
+'\n'.join(result_struct) +
+'\n};\n')
+# output the initializer tokens
+print('\n'.join(initializers) + '\n')
+# output the instance structure
+print(
+f'static cmdline_parse_inst_t cmd_{name} = {{\n' +
+f'\t.f = cmd_{name}_parsed,\n' +
+'\t.data = NULL,\n' +
+f'\t.help_str = "{comment}",\n' +
+'\t.tokens = {')
+for t in token_list:
+print(f'\t\t(void *)&{t},')
+print('\t\tNULL\n' + '\t}\n' + '};\n')
+
+# return the instance structure name
+return f'cmd_{name}'
+
+
+def process_commands(infile, hfile, cfile, c
[PATCH v3 3/5] examples/simple_mp: auto-generate cmdline boilerplate
Use the dpdk-cmdline-gen script to autogenerate all the boilerplate
structs and defines for the commandline part of the app.
Signed-off-by: Bruce Richardson
---
examples/multi_process/simple_mp/Makefile | 12 +-
examples/multi_process/simple_mp/meson.build | 9 ++
.../multi_process/simple_mp/mp_commands.c | 106 ++
.../multi_process/simple_mp/mp_commands.h | 14 ---
.../multi_process/simple_mp/mp_commands.list | 3 +
5 files changed, 30 insertions(+), 114 deletions(-)
delete mode 100644 examples/multi_process/simple_mp/mp_commands.h
create mode 100644 examples/multi_process/simple_mp/mp_commands.list
diff --git a/examples/multi_process/simple_mp/Makefile
b/examples/multi_process/simple_mp/Makefile
index 1d0a260e64..890b6b7e62 100644
--- a/examples/multi_process/simple_mp/Makefile
+++ b/examples/multi_process/simple_mp/Makefile
@@ -6,6 +6,7 @@ APP = simple_mp
# all source are stored in SRCS-y
SRCS-y := main.c mp_commands.c
+SRC-DEPS := build/mp_commands.h
PKGCONF ?= pkg-config
@@ -22,10 +23,13 @@ static: build/$(APP)-static
ln -sf $(APP)-static build/$(APP)
PC_FILE := $(shell $(PKGCONF) --path libdpdk 2>/dev/null)
-CFLAGS += -O3 $(shell $(PKGCONF) --cflags libdpdk)
+CFLAGS += -O3 $(shell $(PKGCONF) --cflags libdpdk) -I build/
LDFLAGS_SHARED = $(shell $(PKGCONF) --libs libdpdk)
LDFLAGS_STATIC = $(shell $(PKGCONF) --static --libs libdpdk)
+build/mp_commands.h: mp_commands.list Makefile
+ dpdk-cmdline-gen.py -o $@ --context-name=simple_mp_ctx $<
+
ifeq ($(MAKECMDGOALS),static)
# check for broken pkg-config
ifeq ($(shell echo $(LDFLAGS_STATIC) | grep
'whole-archive.*l:lib.*no-whole-archive'),)
@@ -36,10 +40,10 @@ endif
CFLAGS += -DALLOW_EXPERIMENTAL_API
-build/$(APP)-shared: $(SRCS-y) Makefile $(PC_FILE) | build
+build/$(APP)-shared: $(SRCS-y) Makefile $(PC_FILE) | build $(SRC-DEPS)
$(CC) $(CFLAGS) $(SRCS-y) -o $@ $(LDFLAGS) $(LDFLAGS_SHARED)
-build/$(APP)-static: $(SRCS-y) Makefile $(PC_FILE) | build
+build/$(APP)-static: $(SRCS-y) Makefile $(PC_FILE) | build $(SRC-DEPS)
$(CC) $(CFLAGS) $(SRCS-y) -o $@ $(LDFLAGS) $(LDFLAGS_STATIC)
build:
@@ -47,5 +51,5 @@ build:
.PHONY: clean
clean:
- rm -f build/$(APP) build/$(APP)-static build/$(APP)-shared
+ rm -f build/$(APP) build/$(APP)-static build/$(APP)-shared
build/mp_commands.h
test -d build && rmdir -p build || true
diff --git a/examples/multi_process/simple_mp/meson.build
b/examples/multi_process/simple_mp/meson.build
index 359af4384d..e99b7a3f6f 100644
--- a/examples/multi_process/simple_mp/meson.build
+++ b/examples/multi_process/simple_mp/meson.build
@@ -7,7 +7,16 @@
# DPDK instance, use 'make'
allow_experimental_apis = true
+
+cmd_h = custom_target('commands_hdr',
+ output: 'mp_commands.h',
+ input: files('mp_commands.list'),
+ capture: true,
+ command: [cmdline_gen_cmd, '--context-name=simple_mp_ctx', '@INPUT@']
+)
+
sources = files(
'mp_commands.c',
'main.c',
)
+sources += cmd_h
diff --git a/examples/multi_process/simple_mp/mp_commands.c
b/examples/multi_process/simple_mp/mp_commands.c
index a5f91b00be..df9fa94208 100644
--- a/examples/multi_process/simple_mp/mp_commands.c
+++ b/examples/multi_process/simple_mp/mp_commands.c
@@ -1,44 +1,18 @@
/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright(c) 2010-2014 Intel Corporation
+ * Copyright(c) 2010-2023 Intel Corporation
*/
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-
-#include
-#include
-#include
-#include
-#include
-#include
-#include
-#include
#include
-#include
#include
#include
-#include
-#include
-#include
-#include
-#include
#include "mp_commands.h"
-/**/
-
-struct cmd_send_result {
- cmdline_fixed_string_t action;
- cmdline_fixed_string_t message;
-};
+extern struct rte_ring *send_ring, *recv_ring;
+extern struct rte_mempool *message_pool;
+extern volatile int quit;
-static void cmd_send_parsed(void *parsed_result,
+void
+cmd_send_parsed(void *parsed_result,
__rte_unused struct cmdline *cl,
__rte_unused void *data)
{
@@ -54,29 +28,8 @@ static void cmd_send_parsed(void *parsed_result,
}
}
-cmdline_parse_token_string_t cmd_send_action =
- TOKEN_STRING_INITIALIZER(struct cmd_send_result, action, "send");
-cmdline_parse_token_string_t cmd_send_message =
- TOKEN_STRING_INITIALIZER(struct cmd_send_result, message, NULL);
-
-cmdline_parse_inst_t cmd_send = {
- .f = cmd_send_parsed, /* function to call */
- .data = NULL, /* 2nd arg of func */
- .help_str = "send a string to another process",
- .tokens = {/* token list, NULL terminated */
- (void *)&cmd_send_action,
- (void *)&cmd_send_message,
- NULL,
- },
-};
-
-/*
[PATCH v3 4/5] examples/hotplug_mp: auto-generate cmdline boilerplate
Use the dpdk-cmdline-gen script to autogenerate all the boilerplate
structs and defines for the commandline part of the app.
Signed-off-by: Bruce Richardson
---
examples/multi_process/hotplug_mp/Makefile| 12 +-
examples/multi_process/hotplug_mp/commands.c | 147 ++
examples/multi_process/hotplug_mp/commands.h | 10 --
.../multi_process/hotplug_mp/commands.list| 5 +
examples/multi_process/hotplug_mp/meson.build | 9 ++
5 files changed, 35 insertions(+), 148 deletions(-)
delete mode 100644 examples/multi_process/hotplug_mp/commands.h
create mode 100644 examples/multi_process/hotplug_mp/commands.list
diff --git a/examples/multi_process/hotplug_mp/Makefile
b/examples/multi_process/hotplug_mp/Makefile
index 6b20d6e49a..81ee85cd6b 100644
--- a/examples/multi_process/hotplug_mp/Makefile
+++ b/examples/multi_process/hotplug_mp/Makefile
@@ -6,6 +6,7 @@ APP = hotplug_mp
# all source are stored in SRCS-y
SRCS-y := main.c commands.c
+SRC-DEPS := build/commands.h
PKGCONF ?= pkg-config
@@ -22,10 +23,13 @@ static: build/$(APP)-static
ln -sf $(APP)-static build/$(APP)
PC_FILE := $(shell $(PKGCONF) --path libdpdk 2>/dev/null)
-CFLAGS += -O3 $(shell $(PKGCONF) --cflags libdpdk)
+CFLAGS += -O3 $(shell $(PKGCONF) --cflags libdpdk) -I build/
LDFLAGS_SHARED = $(shell $(PKGCONF) --libs libdpdk)
LDFLAGS_STATIC = $(shell $(PKGCONF) --static --libs libdpdk)
+build/commands.h: commands.list Makefile
+ dpdk-cmdline-gen.py -o $@ --context-name=main_ctx $<
+
ifeq ($(MAKECMDGOALS),static)
# check for broken pkg-config
ifeq ($(shell echo $(LDFLAGS_STATIC) | grep
'whole-archive.*l:lib.*no-whole-archive'),)
@@ -36,10 +40,10 @@ endif
CFLAGS += -DALLOW_EXPERIMENTAL_API
-build/$(APP)-shared: $(SRCS-y) Makefile $(PC_FILE) | build
+build/$(APP)-shared: $(SRCS-y) Makefile $(PC_FILE) | build $(SRC-DEPS)
$(CC) $(CFLAGS) $(SRCS-y) -o $@ $(LDFLAGS) $(LDFLAGS_SHARED)
-build/$(APP)-static: $(SRCS-y) Makefile $(PC_FILE) | build
+build/$(APP)-static: $(SRCS-y) Makefile $(PC_FILE) | build $(SRC-DEPS)
$(CC) $(CFLAGS) $(SRCS-y) -o $@ $(LDFLAGS) $(LDFLAGS_STATIC)
build:
@@ -47,5 +51,5 @@ build:
.PHONY: clean
clean:
- rm -f build/$(APP) build/$(APP)-static build/$(APP)-shared
+ rm -f build/$(APP) build/$(APP)-static build/$(APP)-shared
build/commands.h
test -d build && rmdir -p build || true
diff --git a/examples/multi_process/hotplug_mp/commands.c
b/examples/multi_process/hotplug_mp/commands.c
index 88f44e00a0..900eb9f774 100644
--- a/examples/multi_process/hotplug_mp/commands.c
+++ b/examples/multi_process/hotplug_mp/commands.c
@@ -1,24 +1,12 @@
/* SPDX-License-Identifier: BSD-3-Clause
- * Copyright(c) 2018 Intel Corporation.
+ * Copyright(c) 2018-2023 Intel Corporation.
*/
-#include
-#include
-#include
-#include
-#include
-#include
-
#include
#include
+#include "commands.h"
-/**/
-
-struct cmd_help_result {
- cmdline_fixed_string_t help;
-};
-
-static void cmd_help_parsed(__rte_unused void *parsed_result,
+void cmd_help_parsed(__rte_unused void *parsed_result,
struct cmdline *cl,
__rte_unused void *data)
{
@@ -29,52 +17,16 @@ static void cmd_help_parsed(__rte_unused void
*parsed_result,
"- list\n\n");
}
-cmdline_parse_token_string_t cmd_help_help =
- TOKEN_STRING_INITIALIZER(struct cmd_help_result, help, "help");
-
-cmdline_parse_inst_t cmd_help = {
- .f = cmd_help_parsed, /* function to call */
- .data = NULL, /* 2nd arg of func */
- .help_str = "show help",
- .tokens = {/* token list, NULL terminated */
- (void *)&cmd_help_help,
- NULL,
- },
-};
-
-/**/
-
-struct cmd_quit_result {
- cmdline_fixed_string_t quit;
-};
-
-static void cmd_quit_parsed(__rte_unused void *parsed_result,
+void
+cmd_quit_parsed(__rte_unused void *parsed_result,
struct cmdline *cl,
__rte_unused void *data)
{
cmdline_quit(cl);
}
-cmdline_parse_token_string_t cmd_quit_quit =
- TOKEN_STRING_INITIALIZER(struct cmd_quit_result, quit, "quit");
-
-cmdline_parse_inst_t cmd_quit = {
- .f = cmd_quit_parsed, /* function to call */
- .data = NULL, /* 2nd arg of func */
- .help_str = "quit",
- .tokens = {/* token list, NULL terminated */
- (void *)&cmd_quit_quit,
- NULL,
- },
-};
-
-/**/
-
-struct cmd_list_result {
- cmdline_fixed_string_t list;
-};
-
-static void cmd_list_parsed(__rte_unused void *parsed_result,
+void
+cmd_list_parsed(__rte_unused void *parsed_result,
struct cmdline *cl,
__rte_unused void *data
[PATCH v3 5/5] examples/bond: auto-generate cmdline boilerplate
Use the dpdk-cmdline-gen script to autogenerate all the boilerplate
structs and defines for the commandline part of the app.
Signed-off-by: Bruce Richardson
---
Note: the original help text on some of the commands in this example
were not useful "this command do not handle any arguments". Therefore,
when converting over to the command script, the per-command help
info has been updated with reference to the code rather than a literal
transfer of the existing help text, as was done with the previous 2
example apps.
---
examples/bond/Makefile | 12 ++-
examples/bond/commands.list | 6 ++
examples/bond/main.c| 161
examples/bond/main.h| 10 ---
examples/bond/meson.build | 8 ++
5 files changed, 37 insertions(+), 160 deletions(-)
create mode 100644 examples/bond/commands.list
delete mode 100644 examples/bond/main.h
diff --git a/examples/bond/Makefile b/examples/bond/Makefile
index ad711a5bee..d87c7a32ba 100644
--- a/examples/bond/Makefile
+++ b/examples/bond/Makefile
@@ -6,6 +6,7 @@ APP = bond_app
# all source are stored in SRCS-y
SRCS-y := main.c
+SRC-DEPS := build/commands.h
PKGCONF ?= pkg-config
@@ -24,10 +25,13 @@ static: build/$(APP)-static
LDFLAGS += -lrte_net_bond
PC_FILE := $(shell $(PKGCONF) --path libdpdk 2>/dev/null)
-CFLAGS += -O3 $(shell $(PKGCONF) --cflags libdpdk)
+CFLAGS += -O3 $(shell $(PKGCONF) --cflags libdpdk) -I build/
LDFLAGS_SHARED = $(shell $(PKGCONF) --libs libdpdk)
LDFLAGS_STATIC = $(shell $(PKGCONF) --static --libs libdpdk)
+build/commands.h: commands.list Makefile
+ dpdk-cmdline-gen.py -o $@ --context-name=main_ctx $<
+
ifeq ($(MAKECMDGOALS),static)
# check for broken pkg-config
ifeq ($(shell echo $(LDFLAGS_STATIC) | grep
'whole-archive.*l:lib.*no-whole-archive'),)
@@ -38,10 +42,10 @@ endif
CFLAGS += -DALLOW_EXPERIMENTAL_API
-build/$(APP)-shared: $(SRCS-y) Makefile $(PC_FILE) | build
+build/$(APP)-shared: $(SRCS-y) Makefile $(PC_FILE) | build $(SRC-DEPS)
$(CC) $(CFLAGS) $(SRCS-y) -o $@ $(LDFLAGS) $(LDFLAGS_SHARED)
-build/$(APP)-static: $(SRCS-y) Makefile $(PC_FILE) | build
+build/$(APP)-static: $(SRCS-y) Makefile $(PC_FILE) | build $(SRC-DEPS)
$(CC) $(CFLAGS) $(SRCS-y) -o $@ $(LDFLAGS) $(LDFLAGS_STATIC)
build:
@@ -49,5 +53,5 @@ build:
.PHONY: clean
clean:
- rm -f build/$(APP) build/$(APP)-static build/$(APP)-shared
+ rm -f build/$(APP) build/$(APP)-static build/$(APP)-shared build/*.h
test -d build && rmdir -p build || true
diff --git a/examples/bond/commands.list b/examples/bond/commands.list
new file mode 100644
index 00..a10bf75708
--- /dev/null
+++ b/examples/bond/commands.list
@@ -0,0 +1,6 @@
+send ip # sends one ARPrequest through bonding for IP
+start # starts listening if not started at startup
+stop # stops listening
+show # shows some bond info, e.g. active members
+help # show help
+quit # close application
diff --git a/examples/bond/main.c b/examples/bond/main.c
index 90f422ec11..8528abf675 100644
--- a/examples/bond/main.c
+++ b/examples/bond/main.c
@@ -45,16 +45,8 @@
#include
#include
-#include
-#include
-#include
-#include
-#include
-#include
#include
-#include
-
-#include "main.h"
+#include "commands.h"
#define RTE_LOGTYPE_DCB RTE_LOGTYPE_USER1
@@ -462,11 +454,7 @@ static int lcore_main(__rte_unused void *arg1)
return 0;
}
-struct cmd_obj_send_result {
- cmdline_fixed_string_t action;
- cmdline_ipaddr_t ip;
-};
-static inline void get_string(struct cmd_obj_send_result *res, char *buf,
uint8_t size)
+static inline void get_string(struct cmd_send_result *res, char *buf, uint8_t
size)
{
snprintf(buf, size, NIPQUAD_FMT,
((unsigned)((unsigned char *)&(res->ip.addr.ipv4))[0]),
@@ -475,12 +463,11 @@ static inline void get_string(struct cmd_obj_send_result
*res, char *buf, uint8_
((unsigned)((unsigned char *)&(res->ip.addr.ipv4))[3])
);
}
-static void cmd_obj_send_parsed(void *parsed_result,
- __rte_unused struct cmdline *cl,
- __rte_unused void *data)
+void
+cmd_send_parsed(void *parsed_result, __rte_unused struct cmdline *cl,
__rte_unused void *data)
{
- struct cmd_obj_send_result *res = parsed_result;
+ struct cmd_send_result *res = parsed_result;
char ip_str[INET6_ADDRSTRLEN];
struct rte_ether_addr bond_mac_addr;
@@ -544,29 +531,8 @@ static void cmd_obj_send_parsed(void *parsed_result,
cmdline_printf(cl, "\n");
}
-cmdline_parse_token_string_t cmd_obj_action_send =
- TOKEN_STRING_INITIALIZER(struct cmd_obj_send_result, action, "send");
-cmdline_parse_token_ipaddr_t cmd_obj_ip =
- TOKEN_IPV4_INITIALIZER(struct cmd_obj_send_result, ip);
-
-cmdline_parse_inst_t cmd_obj_send = {
- .f = cmd_obj_send_parsed, /* function to call */
- .data = NULL, /*
RE: [RFC 1/2] eal: add pointer compression functions
> -Original Message-
> From: Thomas Monjalon
> Sent: Monday, October 9, 2023 10:54 AM
> To: Paul Szczepanek
> Cc: dev@dpdk.org; Honnappa Nagarahalli ;
> Kamalakshitha Aligeri
> Subject: Re: [RFC 1/2] eal: add pointer compression functions
>
> 27/09/2023 17:08, Paul Szczepanek:
> > Add a new utility header for compressing pointers. Pointers are
> > compressed by taking advantage of their locality. Instead of storing
> > the full address only an offset from a known base is stored.
>
> You probably need to insert some explanations from the cover letter.
>
> > The provided functions can store pointers in 32bit offsets.
> >
> > Suggested-by: Honnappa Nagarahalli
> > Signed-off-by: Paul Szczepanek
> > Signed-off-by: Kamalakshitha Aligeri
> > Reviewed-by: Honnappa Nagarahalli
> [...]
> > --- a/lib/eal/include/meson.build
> > +++ b/lib/eal/include/meson.build
> > @@ -35,6 +35,7 @@ headers += files(
> > 'rte_pci_dev_feature_defs.h',
> > 'rte_pci_dev_features.h',
> > 'rte_per_lcore.h',
> > + 'rte_ptr_compress.h',
> > 'rte_pflock.h',
> > 'rte_random.h',
> > 'rte_reciprocal.h',
>
> Did you try to sort alphabetically? failed :)
>
> > +#ifndef _RTE_PTR_COMPRESS_H_
> > +#define _RTE_PTR_COMPRESS_H_
>
> No need extra underscores.
>
> > +
> > +/**
> > + * @file
> > + * RTE pointer compression and decompression.
>
> RTE has no mean here I think.
>
> > + */
> > +
> > +#include
> > +#include
> > +
> > +#include
> > +#include
> > +#include
> > +#include
> > +
> > +#ifdef __cplusplus
> > +extern "C" {
> > +#endif
> > +
> > +/**
> > + * Compress pointers into 32 bit offsets from base pointer.
>
> I think it should be "32-bit".
>
> > + *
> > + * @note Offsets from the base pointer must fit within 32bits.
> > + Alignment allows
> > + * us to drop bits from the offsets - this means that for pointers
> > + aligned by
> > + * 8 bytes they must be within 32GB of the base pointer. Unaligned
> > + pointers
> > + * must be within 4GB.
>
> Not clear what is "alignment".
> > + *
> > + * @param ptr_base
> > + * A pointer used to calculate offsets of pointers in src_table.
> > + * @param src_table
> > + * A pointer to an array of pointers.
> > + * @param dest_table
> > + * A pointer to an array of compressed pointers returned by this
> > function.
> > + * @param n
> > + * The number of objects to compress, must be strictly positive.
> > + * @param bit_shift
> > + * Byte alignment of memory pointed to by the pointers allows for
> > + * bits to be dropped from the offset and hence widen the memory region
> that
> > + * can be covered. This controls how many bits are right shifted.
> > + **/
> > +static __rte_always_inline void
> > +rte_ptr_compress_32(void *ptr_base, void **src_table,
> > + uint32_t *dest_table, unsigned int n, unsigned int bit_shift) {
> > + unsigned int i = 0;
> > +#if defined RTE_HAS_SVE_ACLE
> > + svuint64_t v_src_table;
> > + svuint64_t v_dest_table;
> > + svbool_t pg = svwhilelt_b64(i, n);
> > + do {
> > + v_src_table = svld1_u64(pg, (uint64_t *)src_table + i);
> > + v_dest_table = svsub_x(pg, v_src_table, (uint64_t)ptr_base);
> > + v_dest_table = svlsr_x(pg, v_dest_table, bit_shift);
> > + svst1w(pg, &dest_table[i], v_dest_table);
> > + i += svcntd();
> > + pg = svwhilelt_b64(i, n);
> > + } while (svptest_any(svptrue_b64(), pg)); #elif defined __ARM_NEON
> > + uint64_t ptr_diff;
> > + uint64x2_t v_src_table;
> > + uint64x2_t v_dest_table;
> > + /* right shift is done by left shifting by negative int */
> > + int64x2_t v_shift = vdupq_n_s64(-bit_shift);
> > + uint64x2_t v_ptr_base = vdupq_n_u64((uint64_t)ptr_base);
> > + for (; i < (n & ~0x1); i += 2) {
> > + v_src_table = vld1q_u64((const uint64_t *)src_table + i);
> > + v_dest_table = vsubq_u64(v_src_table, v_ptr_base);
> > + v_dest_table = vshlq_u64(v_dest_table, v_shift);
> > + vst1_u32(dest_table + i, vqmovn_u64(v_dest_table));
> > + }
> > + /* process leftover single item in case of odd number of n */
> > + if (unlikely(n & 0x1)) {
> > + ptr_diff = RTE_PTR_DIFF(src_table[i], ptr_base);
> > + dest_table[i] = (uint32_t) (ptr_diff >> bit_shift);
> > + }
> > +#else
> > + uint64_t ptr_diff;
> > + for (; i < n; i++) {
> > + ptr_diff = RTE_PTR_DIFF(src_table[i], ptr_base);
> > + /* save extra bits that are redundant due to alignment */
> > + ptr_diff = ptr_diff >> bit_shift;
> > + /* make sure no truncation will happen when casting */
> > + RTE_ASSERT(ptr_diff <= UINT32_MAX);
> > + dest_table[i] = (uint32_t) ptr_diff;
> > + }
> > +#endif
> > +}
>
> I see it is providing some per-CPU optimizations, so it is in favor of having
> it in
> DPDK.
> Other than that, it looks very generic, so it is questionable to have in DPDK.
We had it done for
Re: [PATCH v3 4/4] net/tap: use rte_ether_unformat_address
On 10/3/2023 9:29 PM, Stephen Hemminger wrote: > From: David Christensen > > Building DPDK with gcc 12 on a ppc64le system generates a > stringop-overflow warning. Replace the local MAC address > validation function parse_user_mac() with a call to > rte_ether_unformat_addr() instead. > > Bugzilla ID: 1197 > For record, this should be 1195, caught by David: Bugzilla ID: 1195
Re: [PATCH v4 0/3] Spec changes to support multi I/O models
02/10/2023 11:58, Srikanth Yalavarthi: > This series implements changes to mldev spec to extend support > for ML models with multiple inputs and outputs. Changes include > introduction of I/O layout to support packed and split buffers > for model input and output. Extended the rte_ml_model_info > structure to support multiple inputs and outputs. > > Updated rte_ml_op and quantize / dequantize APIs to support an > array of input and output ML buffer segments. > > Support for batches option is dropped from test application. Applied, thanks.
Re: [PATCH v5 1/3] lib: introduce dispatcher library
On Mon, Oct 9, 2023 at 6:50 PM Mattias Rönnblom wrote:
[snip]
> +static int
> +evd_set_service_runstate(struct rte_dispatcher *dispatcher, int state)
> +{
> + int rc;
> +
> + rc = rte_service_component_runstate_set(dispatcher->service_id,
> + state);
> +
> + if (rc != 0) {
> + RTE_EDEV_LOG_ERR("Unexpected error %d occurred while
> setting "
> +"service component run state to %d\n",
> rc,
> +state);
> + RTE_ASSERT(0);
> >>>
> >>> Why not propagating the error to callers?
> >>>
> >>>
> >>
> >> The root cause would be a programming error, hence an assertion is more
> >> appropriate way to deal with the situation.
> >
> > Without building RTE_ENABLE_ASSERT (disabled by default), the code
> > later in this function will still be executed.
> >
>
> If RTE_ASSERT() is not the way to assure a consistent internal library
> state, what is? RTE_VERIFY()?
The usual way in DPDK is to use RTE_VERIFY or rte_panic with the error message.
There is also libc assert().
RTE_ASSERT is more of a debug macro since it is under a build option.
But by making the library "panic" on some assertion, I have followup comments:
- what is the point of returning an int for rte_dispatcher_start() /
rte_dispatcher_stop()?
- rte_dispatcher_start() and rte_dispatcher_stop() (doxygen)
documentation needs updating, as they can't return anything but 0.
--
David Marchand
Re: [PATCH] bus/pci: fix device ID print
13/09/2023 10:23, Qiming Yang:
> This patch fixes the issue where device id first 0 does not print.
>
> Fixes: e4f27af0f448 ("bus/pci: reduce boot-up logs to absolute minimum")
> Cc: sta...@dpdk.org
>
> Signed-off-by: Qiming Yang
Applied, thanks.
