Skip to contentSkip to conversation

build: update DPDK to use C11 standard

PATCH Superseded Bruce Richardson 8 participants 45 messages Jul 31, 2023 – Aug 11, 2023

Subsystems (2)
lore
Share & tools
lore
#0 lore raw

PATCH 1/1 build: update DPDK to use C11 standard

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

---

When moving the information about the new requirement to the release notes, a change like this doesn't seem to fit into any existing section. Given its global scope and importance, I've therefore just put it on top of the file, rather than in any section.

---
 doc/guides/linux_gsg/sys_reqs.rst      |  3 ++-
 doc/guides/rel_notes/deprecation.rst   | 18 ------------------
 doc/guides/rel_notes/release_23_11.rst | 17 +++++++++++++++++
 meson.build                            |  1 +
 4 files changed, 20 insertions(+), 19 deletions(-)
doc/guides/linux_gsg/sys_reqs.rst +2 -1
index dfeaf4e1c5..13be715933 100644
@@ -27,7 +27,8 @@ Compilation of the DPDK
     The setup commands and installed packages needed on various systems may be different.
     For details on Linux distributions and the versions tested, please consult the DPDK Release Notes.
 
-*   General development tools including a supported C compiler such as gcc (version 4.9+) or clang (version 3.4+),
+*   General development tools including a C compiler supporting the C11 standard,
+    including standard atomics, for example: GCC (version 5.0+) or Clang (version 3.6+),
     and ``pkg-config`` or ``pkgconf`` to be used when building end-user binaries against DPDK.
 
     * For RHEL/Fedora systems these can be installed using ``dnf groupinstall "Development Tools"``
doc/guides/rel_notes/deprecation.rst +0 -18
index 494b401cda..cc939d3c67 100644
@@ -17,24 +17,6 @@ Other API and ABI deprecation notices are to be posted below.
 Deprecation Notices
 -------------------
 
-* C Compiler: 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)
-
 * build: Enabling deprecated libraries (``flow_classify``, ``kni``)
   won't be possible anymore through the use of the ``disable_libs`` build option.
   A new build option for deprecated libraries will be introduced instead.
doc/guides/rel_notes/release_23_11.rst +17 -0
index 6b4dd21fd0..c8b9ed456c 100644
@@ -20,6 +20,23 @@ 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
 ------------
meson.build +1 -0
index 39cb73846d..70b54f0c98 100644
@@ -9,6 +9,7 @@ project('DPDK', 'c',
         license: 'BSD',
         default_options: [
             'buildtype=release',
+            'c_std=c11',
             'default_library=static',
             'warning_level=2',
         ],
Signature · 1 line
2.39.2
#2 lore raw

PATCH v2 1/1 build: update DPDK to use C11 standard

In reply to Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

---
V2:
* Resubmit now that 23.11-rc0 patch applied
* Add _POSIX_C_SOURCE macro to eal_common_errno.c to get POSIX definition of strerror_r() with c11 standard.

---
 doc/guides/linux_gsg/sys_reqs.rst      |  3 ++-
 doc/guides/rel_notes/deprecation.rst   | 18 ------------------
 doc/guides/rel_notes/release_23_11.rst | 17 +++++++++++++++++
 lib/eal/common/eal_common_errno.c      |  1 +
 meson.build                            |  1 +
 5 files changed, 21 insertions(+), 19 deletions(-)
doc/guides/linux_gsg/sys_reqs.rst +2 -1
index dfeaf4e1c5..13be715933 100644
@@ -27,7 +27,8 @@ Compilation of the DPDK
     The setup commands and installed packages needed on various systems may be different.
     For details on Linux distributions and the versions tested, please consult the DPDK Release Notes.
 
-*   General development tools including a supported C compiler such as gcc (version 4.9+) or clang (version 3.4+),
+*   General development tools including a C compiler supporting the C11 standard,
+    including standard atomics, for example: GCC (version 5.0+) or Clang (version 3.6+),
     and ``pkg-config`` or ``pkgconf`` to be used when building end-user binaries against DPDK.
 
     * For RHEL/Fedora systems these can be installed using ``dnf groupinstall "Development Tools"``
doc/guides/rel_notes/deprecation.rst +0 -18
index 494b401cda..cc939d3c67 100644
@@ -17,24 +17,6 @@ Other API and ABI deprecation notices are to be posted below.
 Deprecation Notices
 -------------------
 
-* C Compiler: 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)
-
 * build: Enabling deprecated libraries (``flow_classify``, ``kni``)
   won't be possible anymore through the use of the ``disable_libs`` build option.
   A new build option for deprecated libraries will be introduced instead.
doc/guides/rel_notes/release_23_11.rst +17 -0
index 6b4dd21fd0..c8b9ed456c 100644
@@ -20,6 +20,23 @@ 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
 ------------
lib/eal/common/eal_common_errno.c +1 -0
index ef8f782abb..b30e2f0ad4 100644
@@ -4,6 +4,7 @@
 
 /* Use XSI-compliant portable version of strerror_r() */
 #undef _GNU_SOURCE
+#define _POSIX_C_SOURCE 200809L
 
 #include <stdio.h>
 #include <string.h>
meson.build +1 -0
index 39cb73846d..70b54f0c98 100644
@@ -9,6 +9,7 @@ project('DPDK', 'c',
         license: 'BSD',
         default_options: [
             'buildtype=release',
+            'c_std=c11',
             'default_library=static',
             'warning_level=2',
         ],
Signature · 1 line
2.39.2
#4 lore raw
In reply to Bruce Richardson
Quoted 108 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

---
V2:
* Resubmit now that 23.11-rc0 patch applied
* Add _POSIX_C_SOURCE macro to eal_common_errno.c to get POSIX definition of strerror_r() with c11 standard.

---
 doc/guides/linux_gsg/sys_reqs.rst      |  3 ++-
 doc/guides/rel_notes/deprecation.rst   | 18 ------------------
 doc/guides/rel_notes/release_23_11.rst | 17 +++++++++++++++++
 lib/eal/common/eal_common_errno.c      |  1 +
 meson.build                            |  1 +
 5 files changed, 21 insertions(+), 19 deletions(-)
doc/guides/linux_gsg/sys_reqs.rst +2 -1
index dfeaf4e1c5..13be715933 100644
@@ -27,7 +27,8 @@ Compilation of the DPDK
     The setup commands and installed packages needed on various systems may be different.
     For details on Linux distributions and the versions tested, please consult the DPDK Release Notes.
 
-*   General development tools including a supported C compiler such as gcc (version 4.9+) or clang (version 3.4+),
+*   General development tools including a C compiler supporting the C11 standard,
+    including standard atomics, for example: GCC (version 5.0+) or Clang (version 3.6+),
     and ``pkg-config`` or ``pkgconf`` to be used when building end-user binaries against DPDK.
 
     * For RHEL/Fedora systems these can be installed using ``dnf groupinstall "Development Tools"``
doc/guides/rel_notes/deprecation.rst +0 -18
index 494b401cda..cc939d3c67 100644
@@ -17,24 +17,6 @@ Other API and ABI deprecation notices are to be posted below.
 Deprecation Notices
 -------------------
 
-* C Compiler: 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)
-
 * build: Enabling deprecated libraries (``flow_classify``, ``kni``)
   won't be possible anymore through the use of the ``disable_libs`` build option.
   A new build option for deprecated libraries will be introduced instead.
doc/guides/rel_notes/release_23_11.rst +17 -0
index 6b4dd21fd0..c8b9ed456c 100644
@@ -20,6 +20,23 @@ 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
 ------------
lib/eal/common/eal_common_errno.c +1 -0
index ef8f782abb..b30e2f0ad4 100644
@@ -4,6 +4,7 @@
 
 /* Use XSI-compliant portable version of strerror_r() */
 #undef _GNU_SOURCE
+#define _POSIX_C_SOURCE 200809L
 
 #include <stdio.h>
 #include <string.h>
meson.build +0 -0
index 39cb73846d..70b54f0c98 100644

meson.build project('DPDK', 'c', · line 15

Quoted 7 lines
meson.build +1 -0
@@ -9,6 +9,7 @@ project('DPDK', 'c',
         license: 'BSD',
         default_options: [
             'buildtype=release',
+            'c_std=c11',
             'default_library=static',
             'warning_level=2',
         ],

oh I acked v2 (and you can maintain that ack) but one additional removal of forced -std=gnu99 is maybe necessary?

drivers/net/failsafe/meson.build

    probably should remove cflags += '-std=gnu99'

if we remove it we inherit the -std=c11 from meson project configuration and define a _POSIX_C_SOURCE narrowly where necessary (if it is needed).

#5 lore raw

PATCH v3 1/1 build: update DPDK to use C11 standard

In reply to Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

---
V3:
* remove (now unneeded) use of -std=gnu99 in failsafe net driver.

V2:
* Resubmit now that 23.11-rc0 patch applied
* Add _POSIX_C_SOURCE macro to eal_common_errno.c to get POSIX definition of strerror_r() with c11 standard.

---
 doc/guides/linux_gsg/sys_reqs.rst      |  3 ++-
 doc/guides/rel_notes/deprecation.rst   | 18 ------------------
 doc/guides/rel_notes/release_23_11.rst | 17 +++++++++++++++++
 drivers/net/failsafe/meson.build       |  1 -
 lib/eal/common/eal_common_errno.c      |  1 +
 meson.build                            |  1 +
 6 files changed, 21 insertions(+), 20 deletions(-)
doc/guides/linux_gsg/sys_reqs.rst +2 -1
index dfeaf4e1c5..13be715933 100644
@@ -27,7 +27,8 @@ Compilation of the DPDK
     The setup commands and installed packages needed on various systems may be different.
     For details on Linux distributions and the versions tested, please consult the DPDK Release Notes.
 
-*   General development tools including a supported C compiler such as gcc (version 4.9+) or clang (version 3.4+),
+*   General development tools including a C compiler supporting the C11 standard,
+    including standard atomics, for example: GCC (version 5.0+) or Clang (version 3.6+),
     and ``pkg-config`` or ``pkgconf`` to be used when building end-user binaries against DPDK.
 
     * For RHEL/Fedora systems these can be installed using ``dnf groupinstall "Development Tools"``
doc/guides/rel_notes/deprecation.rst +0 -18
index 494b401cda..cc939d3c67 100644
@@ -17,24 +17,6 @@ Other API and ABI deprecation notices are to be posted below.
 Deprecation Notices
 -------------------
 
-* C Compiler: 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)
-
 * build: Enabling deprecated libraries (``flow_classify``, ``kni``)
   won't be possible anymore through the use of the ``disable_libs`` build option.
   A new build option for deprecated libraries will be introduced instead.
doc/guides/rel_notes/release_23_11.rst +17 -0
index 6b4dd21fd0..c8b9ed456c 100644
@@ -20,6 +20,23 @@ 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
 ------------
drivers/net/failsafe/meson.build +0 -1
index 6013e13722..c1d361083b 100644
@@ -7,7 +7,6 @@ if is_windows
     subdir_done()
 endif
 
-cflags += '-std=gnu99'
 cflags += '-D_DEFAULT_SOURCE'
 cflags += '-D_XOPEN_SOURCE=700'
 cflags += '-pedantic'
lib/eal/common/eal_common_errno.c +1 -0
index ef8f782abb..b30e2f0ad4 100644
@@ -4,6 +4,7 @@
 
 /* Use XSI-compliant portable version of strerror_r() */
 #undef _GNU_SOURCE
+#define _POSIX_C_SOURCE 200809L
 
 #include <stdio.h>
 #include <string.h>
meson.build +1 -0
index 39cb73846d..70b54f0c98 100644
@@ -9,6 +9,7 @@ project('DPDK', 'c',
         license: 'BSD',
         default_options: [
             'buildtype=release',
+            'c_std=c11',
             'default_library=static',
             'warning_level=2',
         ],
Signature · 1 line
2.39.2
Acked-by on v3 1/1
#6 lore raw
Quoted 11 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

---
V3:
* remove (now unneeded) use of -std=gnu99 in failsafe net driver.

Quoted 5 lines

V2:
* Resubmit now that 23.11-rc0 patch applied
* Add _POSIX_C_SOURCE macro to eal_common_errno.c to get POSIX definition of strerror_r() with c11 standard.
---

if there is no objection from others i would like to suggest expedited merge of this series to make it easy to take a dependency on it for atomics changes.

thanks!

#7 lore raw

Hi Bruce,

I see some failures for this series for our Ubuntu 20.04 containers. And, our DTS testbeds which are on ubuntu 20.04 are skipping running testsuites because they can't compile DPDK. So, that's why it has some missing results for a couple of the Intel NICs. For context, I'll paste below where the compile job terminates in one of our containerized compile test runs. The GCC in use here is version 9.4, so it meets the requirements as described in your patch as far as I can tell. I'll check it out more tomorrow to see whether it's an infra failure, like some missing dependencies. Please let me know if we expect to have no issues with 20.04 or if this is anticipated.

Thanks!

[1638/2730] Generating symbol file 'drivers/a715181@@rte_net_ixgbe@sha /librte_net_ixgbe.so.24.0.symbols'.
[1639/2730] Compiling C object 'drivers/a715181@@tmp_rte_net_mlx4@sta /net_mlx4_mlx4.c.o'.
FAILED: drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o ccache cc -Idrivers/a715181@@tmp_rte_net_mlx4@sta -Idrivers -I../drivers -Idrivers/net/mlx4 -I../drivers/net/mlx4 -Ilib/ethdev -I../lib/ethdev -I. -I../ -Iconfig -I../config -Ilib/eal/include -I../lib/eal/include -Ilib/eal/linux/include -I../lib/eal/linux/include -Ilib/eal/x86/include -I../lib/eal/x86/include -Ilib/eal/common -I../lib/eal/common -Ilib/eal -I../lib/eal -Ilib/kvargs -I../lib/kvargs -Ilib/telemetry/../metrics -I../lib/telemetry/../metrics -Ilib/telemetry -I../lib/telemetry -Ilib/net -I../lib/net -Ilib/mbuf -I../lib/mbuf -Ilib/mempool -I../lib/mempool -Ilib/ring -I../lib/ring -Ilib/meter -I../lib/meter -Idrivers/bus/pci -I../drivers/bus/pci -I../drivers/bus/pci/linux -Ilib/pci -I../lib/pci -Idrivers/bus/vdev -I../drivers/bus/vdev -I/usr/include/libnl3 -fdiagnostics-color=always -pipe -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wextra -Werror -std=c11 -O3 -include rte_config.h -Wcast-qual -Wdeprecated -Wformat -Wformat-nonliteral -Wformat-security -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wpointer-arith -Wsign-compare -Wstrict-prototypes -Wundef -Wwrite-strings -Wno-address-of-packed-member -Wno-packed-not-aligned -Wno-missing-field-initializers -D_GNU_SOURCE -fPIC -march=native -DALLOW_EXPERIMENTAL_API -DALLOW_INTERNAL_API -Wno-format-truncation -std=c11 -Wno-strict-prototypes -D_BSD_SOURCE -D_DEFAULT_SOURCE -D_XOPEN_SOURCE=600 -UPEDANTIC -DRTE_LOG_DEFAULT_LOGTYPE=pmd.net.mlx4 -MD -MQ 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -MF 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o.d' -o 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -c ../drivers/net/mlx4/mlx4.c
In file included from ../drivers/net/mlx4/mlx4_rxtx.h:27,

                 from ../drivers/net/mlx4/mlx4.c:49:
../drivers/net/mlx4/mlx4_prm.h:111:8: error: redefinition of 'struct
mlx4_wqe_lso_seg'
  111 | struct mlx4_wqe_lso_seg {
      |        ^~~~~~~~~~~~~~~~
In file included from ../drivers/net/mlx4/mlx4_glue.h:16,
                 from ../drivers/net/mlx4/mlx4.c:46:
/usr/include/infiniband/mlx4dv.h:410:8: note: originally defined here
  410 | struct mlx4_wqe_lso_seg {
      |        ^~~~~~~~~~~~~~~~

ninja: build stopped: subcommand failed.

#8 lore raw
Quoted 4 lines · Patrick Robb

Hi Bruce,
I see some failures for this series for our Ubuntu 20.04 containers. And, our DTS testbeds which are on ubuntu 20.04 are skipping running testsuites because they can't compile DPDK. So, that's why it has some

Quoted 8 lines

missing results for a couple of the Intel NICs. For context, I'll paste below where the compile job terminates in one of our containerized compile test runs. The GCC in use here is version 9.4, so it meets the requirements as described in your patch as far as I can tell. I'll check it out more tomorrow to see whether it's an infra failure, like some missing dependencies. Please let me know if we expect to have no issues with 20.04 or if this is anticipated.
Thanks!

While not anticipated, this is why we have CI so these range of distros can all be checked. There are a lot of failure reports from Ubuntu 20.04 so I think this is a genuine issue. I will fire up a 20.04 VM myself and see what I can find, and submit a new patch revision.

/Bruce

#9 lore raw
In reply to Patrick Robb
Quoted 44 lines · Patrick Robb

Hi Bruce,
I see some failures for this series for our Ubuntu 20.04 containers. And, our DTS testbeds which are on ubuntu 20.04 are skipping running testsuites because they can't compile DPDK. So, that's why it has some missing results for a couple of the Intel NICs. For context, I'll paste below where the compile job terminates in one of our containerized compile test runs. The GCC in use here is version 9.4, so it meets the requirements as described in your patch as far as I can tell. I'll check it out more tomorrow to see whether it's an infra failure, like some missing dependencies. Please let me know if we expect to have no issues with 20.04 or if this is anticipated.
Thanks!
[1638/2730] Generating symbol file 'drivers/a715181@@rte_net_ixgbe@sha/librte_net_ixgbe.so.24.0.symbols'. [1639/2730] Compiling C object 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o'.
FAILED: drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o ccache cc -Idrivers/a715181@@tmp_rte_net_mlx4@sta -Idrivers -I../drivers -Idrivers/net/mlx4 -I../drivers/net/mlx4 -Ilib/ethdev -I../lib/ethdev -I. -I../ -Iconfig -I../config -Ilib/eal/include -I../lib/eal/include -Ilib/eal/linux/include -I../lib/eal/linux/include -Ilib/eal/x86/include -I../lib/eal/x86/include -Ilib/eal/common -I../lib/eal/common -Ilib/eal -I../lib/eal -Ilib/kvargs -I../lib/kvargs -Ilib/telemetry/../metrics -I../lib/telemetry/../metrics -Ilib/telemetry -I../lib/telemetry -Ilib/net -I../lib/net -Ilib/mbuf -I../lib/mbuf -Ilib/mempool -I../lib/mempool -Ilib/ring -I../lib/ring -Ilib/meter -I../lib/meter -Idrivers/bus/pci -I../drivers/bus/pci -I../drivers/bus/pci/linux -Ilib/pci -I../lib/pci -Idrivers/bus/vdev -I../drivers/bus/vdev -I/usr/include/libnl3 -fdiagnostics-color=always -pipe -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wextra -Werror -std=c11 -O3 -include rte_config.h -Wcast-qual -Wdeprecated -Wformat -Wformat-nonliteral -Wformat-security -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wpointer-arith -Wsign-compare -Wstrict-prototypes -Wundef -Wwrite-strings -Wno-address-of-packed-member -Wno-packed-not-aligned -Wno-missing-field-initializers -D_GNU_SOURCE -fPIC -march=native -DALLOW_EXPERIMENTAL_API -DALLOW_INTERNAL_API -Wno-format-truncation -std=c11 -Wno-strict-prototypes -D_BSD_SOURCE -D_DEFAULT_SOURCE -D_XOPEN_SOURCE=600 -UPEDANTIC -DRTE_LOG_DEFAULT_LOGTYPE=pmd.net.mlx4 -MD -MQ 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -MF 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o.d' -o 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -c ../drivers/net/mlx4/mlx4.c
In file included from ../drivers/net/mlx4/mlx4_rxtx.h:27,

Quoted 11 lines
                    from ../drivers/net/mlx4/mlx4.c:49:
   ../drivers/net/mlx4/mlx4_prm.h:111:8: error: redefinition of 'struct
   mlx4_wqe_lso_seg'
     111 | struct mlx4_wqe_lso_seg {
         |        ^~~~~~~~~~~~~~~~
   In file included from ../drivers/net/mlx4/mlx4_glue.h:16,
                    from ../drivers/net/mlx4/mlx4.c:46:
   /usr/include/infiniband/mlx4dv.h:410:8: note: originally defined here
     410 | struct mlx4_wqe_lso_seg {
         |        ^~~~~~~~~~~~~~~~
   ninja: build stopped: subcommand failed.

Hi again,

I've attempted to reproduce this on my Ubuntu 20.04 VM and failed, everything seems to build ok.

Looking through the logs, though, there does appear to be a difference in the configurations in the two cases. I suspect my Ubuntu has an updated verbs package compared to the image you are using. In the log of the failed build, I see:

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO Configuring mlx4_autoconf.h using configuration

While building in my VM, I have:

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: YES (cached) Configuring mlx4_autoconf.h using configuration

So my verbs mlx4 header has got a different set of definitions to those in the CI machine. My Ubuntu reports as 20.04.6 with libibverbs-dev package version "28.0-1ubuntu1"

Can the CI image be updated to latest 20.04 packages?

/Bruce

#10 lore raw
Quoted 76 lines · Bruce Richardson
Quoted 44 lines · Patrick Robb

Hi Bruce,
I see some failures for this series for our Ubuntu 20.04 containers. And, our DTS testbeds which are on ubuntu 20.04 are skipping running testsuites because they can't compile DPDK. So, that's why it has some missing results for a couple of the Intel NICs. For context, I'll paste below where the compile job terminates in one of our containerized compile test runs. The GCC in use here is version 9.4, so it meets the requirements as described in your patch as far as I can tell. I'll check it out more tomorrow to see whether it's an infra failure, like some missing dependencies. Please let me know if we expect to have no issues with 20.04 or if this is anticipated.
Thanks!
[1638/2730] Generating symbol file 'drivers/a715181@@rte_net_ixgbe@sha/librte_net_ixgbe.so.24.0.symbols'. [1639/2730] Compiling C object 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o'.
FAILED: drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o ccache cc -Idrivers/a715181@@tmp_rte_net_mlx4@sta -Idrivers -I../drivers -Idrivers/net/mlx4 -I../drivers/net/mlx4 -Ilib/ethdev -I../lib/ethdev -I. -I../ -Iconfig -I../config -Ilib/eal/include -I../lib/eal/include -Ilib/eal/linux/include -I../lib/eal/linux/include -Ilib/eal/x86/include -I../lib/eal/x86/include -Ilib/eal/common -I../lib/eal/common -Ilib/eal -I../lib/eal -Ilib/kvargs -I../lib/kvargs -Ilib/telemetry/../metrics -I../lib/telemetry/../metrics -Ilib/telemetry -I../lib/telemetry -Ilib/net -I../lib/net -Ilib/mbuf -I../lib/mbuf -Ilib/mempool -I../lib/mempool -Ilib/ring -I../lib/ring -Ilib/meter -I../lib/meter -Idrivers/bus/pci -I../drivers/bus/pci -I../drivers/bus/pci/linux -Ilib/pci -I../lib/pci -Idrivers/bus/vdev -I../drivers/bus/vdev -I/usr/include/libnl3 -fdiagnostics-color=always -pipe -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wextra -Werror -std=c11 -O3 -include rte_config.h -Wcast-qual -Wdeprecated -Wformat -Wformat-nonliteral -Wformat-security -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wpointer-arith -Wsign-compare -Wstrict-prototypes -Wundef -Wwrite-strings -Wno-address-of-packed-member -Wno-packed-not-aligned -Wno-missing-field-initializers -D_GNU_SOURCE -fPIC -march=native -DALLOW_EXPERIMENTAL_API -DALLOW_INTERNAL_API -Wno-format-truncation -std=c11 -Wno-strict-prototypes -D_BSD_SOURCE -D_DEFAULT_SOURCE -D_XOPEN_SOURCE=600 -UPEDANTIC -DRTE_LOG_DEFAULT_LOGTYPE=pmd.net.mlx4 -MD -MQ 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -MF 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o.d' -o 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -c ../drivers/net/mlx4/mlx4.c
In file included from ../drivers/net/mlx4/mlx4_rxtx.h:27,

Quoted 11 lines
                    from ../drivers/net/mlx4/mlx4.c:49:
   ../drivers/net/mlx4/mlx4_prm.h:111:8: error: redefinition of 'struct
   mlx4_wqe_lso_seg'
     111 | struct mlx4_wqe_lso_seg {
         |        ^~~~~~~~~~~~~~~~
   In file included from ../drivers/net/mlx4/mlx4_glue.h:16,
                    from ../drivers/net/mlx4/mlx4.c:46:
   /usr/include/infiniband/mlx4dv.h:410:8: note: originally defined here
     410 | struct mlx4_wqe_lso_seg {
         |        ^~~~~~~~~~~~~~~~
   ninja: build stopped: subcommand failed.

Hi again,

I've attempted to reproduce this on my Ubuntu 20.04 VM and failed, everything seems to build ok.

Looking through the logs, though, there does appear to be a difference in the configurations in the two cases. I suspect my Ubuntu has an updated verbs package compared to the image you are using. In the log of the failed build, I see:

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO Configuring mlx4_autoconf.h using configuration

While building in my VM, I have:

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: YES (cached) Configuring mlx4_autoconf.h using configuration

Quoted 8 lines

So my verbs mlx4 header has got a different set of definitions to those in the CI machine. My Ubuntu reports as 20.04.6 with libibverbs-dev package version "28.0-1ubuntu1"

Can the CI image be updated to latest 20.04 packages?

/Bruce

I can reproduce the issue seen at UNH, with a 20.04.6 container and the same libibverbs as you:
ii libibverbs-dev:amd64 28.0-1ubuntu1
amd64 Development files for the libibverbs library

So I suspect something is different in container images..

Pasting the (hopefully) relevant meson logs:

Running compile:

Working directory:  /root/dpdk/build/meson-private/tmp0ovvvd9g
Command line:  ccache cc -I/usr/include/libnl3

/root/dpdk/build/meson-private/tmp0ovvvd9g/testfile.c -o /root/dpdk/build/meson-private/tmp0ovvvd9g/output.obj -pipe -c -D_FILE_OFFSET_BITS=64 -O0 -std=c11

Code:

 #include <infiniband/mlx4dv.h>
        void bar(void) {
            struct mlx4_wqe_lso_seg foo;
            foo.mss_hdr_size;

        };

Compiler stdout:

Compiler stderr:
In file included from /root/dpdk/build/meson-private/tmp0ovvvd9g/testfile.c:1:
/usr/include/infiniband/mlx4dv.h:176:2: error: unknown type name 'off_t'

  176 |  off_t   uar_mmap_offset;
      |  ^~~~~

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO

Signature · 1 line
David Marchand
#11 lore raw
Quoted 6 lines · David Marchand
Quoted 4 lines · Bruce Richardson
Quoted 2 lines · Patrick Robb

Hi Bruce,
I see some failures for this series for our Ubuntu 20.04 containers.

<snip>

Quoted 53 lines
Quoted 19 lines

Hi again,

I've attempted to reproduce this on my Ubuntu 20.04 VM and failed, everything seems to build ok.

Looking through the logs, though, there does appear to be a difference in the configurations in the two cases. I suspect my Ubuntu has an updated verbs package compared to the image you are using. In the log of the failed build, I see:

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO Configuring mlx4_autoconf.h using configuration

While building in my VM, I have:

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: YES (cached) Configuring mlx4_autoconf.h using configuration

Quoted 8 lines

So my verbs mlx4 header has got a different set of definitions to those in the CI machine. My Ubuntu reports as 20.04.6 with libibverbs-dev package version "28.0-1ubuntu1"

Can the CI image be updated to latest 20.04 packages?

/Bruce

I can reproduce the issue seen at UNH, with a 20.04.6 container and the same libibverbs as you:
ii libibverbs-dev:amd64 28.0-1ubuntu1
amd64 Development files for the libibverbs library

So I suspect something is different in container images..

Pasting the (hopefully) relevant meson logs:

Running compile:

Working directory:  /root/dpdk/build/meson-private/tmp0ovvvd9g
Command line:  ccache cc -I/usr/include/libnl3

/root/dpdk/build/meson-private/tmp0ovvvd9g/testfile.c -o /root/dpdk/build/meson-private/tmp0ovvvd9g/output.obj -pipe -c -D_FILE_OFFSET_BITS=64 -O0 -std=c11

Code:

 #include <infiniband/mlx4dv.h>
        void bar(void) {
            struct mlx4_wqe_lso_seg foo;
            foo.mss_hdr_size;

        };

Compiler stdout:

Quoted 9 lines

Compiler stderr:
In file included from /root/dpdk/build/meson-private/tmp0ovvvd9g/testfile.c:1:
/usr/include/infiniband/mlx4dv.h:176:2: error: unknown type name 'off_t'

  176 |  off_t   uar_mmap_offset;
      |  ^~~~~

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO

Thanks. I'll dig some more.

#12 lore raw
Quoted 67 lines · Bruce Richardson
Quoted 6 lines · David Marchand
Quoted 4 lines · Bruce Richardson
Quoted 2 lines · Patrick Robb

Hi Bruce,
I see some failures for this series for our Ubuntu 20.04 containers.

<snip>

Quoted 59 lines
Quoted 19 lines

Hi again,

I've attempted to reproduce this on my Ubuntu 20.04 VM and failed, everything seems to build ok.

Looking through the logs, though, there does appear to be a difference in the configurations in the two cases. I suspect my Ubuntu has an updated verbs package compared to the image you are using. In the log of the failed build, I see:

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO Configuring mlx4_autoconf.h using configuration

While building in my VM, I have:

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: YES (cached) Configuring mlx4_autoconf.h using configuration

Quoted 8 lines

So my verbs mlx4 header has got a different set of definitions to those in the CI machine. My Ubuntu reports as 20.04.6 with libibverbs-dev package version "28.0-1ubuntu1"

Can the CI image be updated to latest 20.04 packages?

/Bruce

I can reproduce the issue seen at UNH, with a 20.04.6 container and the same libibverbs as you:
ii libibverbs-dev:amd64 28.0-1ubuntu1
amd64 Development files for the libibverbs library

So I suspect something is different in container images..

Pasting the (hopefully) relevant meson logs:

Running compile:

Working directory:  /root/dpdk/build/meson-private/tmp0ovvvd9g
Command line:  ccache cc -I/usr/include/libnl3

/root/dpdk/build/meson-private/tmp0ovvvd9g/testfile.c -o /root/dpdk/build/meson-private/tmp0ovvvd9g/output.obj -pipe -c -D_FILE_OFFSET_BITS=64 -O0 -std=c11

Code:

 #include <infiniband/mlx4dv.h>
        void bar(void) {
            struct mlx4_wqe_lso_seg foo;
            foo.mss_hdr_size;

        };

Compiler stdout:

Compiler stderr:
In file included from /root/dpdk/build/meson-private/tmp0ovvvd9g/testfile.c:1:
/usr/include/infiniband/mlx4dv.h:176:2: error: unknown type name 'off_t'

  176 |  off_t   uar_mmap_offset;
      |  ^~~~~
Quoted 4 lines
Quoted 3 lines

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO

Thanks. I'll dig some more.

I think the meson version may be the culprit here. In my meson log I don't see the -std=c11 flag appended to the test compilation command.

Let me downgrade my meson version and see if I can reproduce.

/Bruce

#13 lore raw
Quoted 86 lines

-----Original Message-----
From: Bruce Richardson <[email protected]>
Sent: Tuesday, August 1, 2023 1:51 PM
To: David Marchand <[email protected]>
Cc: Patrick Robb <[email protected]>; Tyler Retzlaff <[email protected]>; [email protected]; Morten Brørup <[email protected]>; NBU-Contact-Thomas Monjalon (EXTERNAL) <[email protected]>
Subject: Re: [PATCH v3] build: update DPDK to use C11 standard

Quoted 9 lines · Bruce Richardson
Quoted 6 lines · David Marchand
Quoted 4 lines · Bruce Richardson
Quoted 2 lines · Patrick Robb

Hi Bruce,
I see some failures for this series for our Ubuntu 20.04 containers.

<snip>

Quoted 1 line
Quoted 1 line
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

Hi again,

I've attempted to reproduce this on my Ubuntu 20.04 VM and failed, everything seems to build ok.

Looking through the logs, though, there does appear to be a difference in the configurations in the two cases. I suspect my Ubuntu has an updated verbs package compared to the image you are using. In the log of the

failed

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

build, I see:

Checking whether type "struct mlx4_wqe_lso_seg" has member

"mss_hdr_size" with dependencies libmlx4, libibverbs: NO

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

Configuring mlx4_autoconf.h using configuration

While building in my VM, I have:

Checking whether type "struct mlx4_wqe_lso_seg" has member

"mss_hdr_size" with dependencies libmlx4, libibverbs: YES (cached)

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

Configuring mlx4_autoconf.h using configuration

So my verbs mlx4 header has got a different set of definitions to those in the CI machine. My Ubuntu reports as 20.04.6 with libibverbs-dev

package

Quoted 29 lines
Quoted 29 lines
Quoted 6 lines

version "28.0-1ubuntu1"

Can the CI image be updated to latest 20.04 packages?

/Bruce

I can reproduce the issue seen at UNH, with a 20.04.6 container and the same libibverbs as you:
ii libibverbs-dev:amd64 28.0-1ubuntu1
amd64 Development files for the libibverbs library

So I suspect something is different in container images..

Pasting the (hopefully) relevant meson logs:

Running compile:

Working directory:  /root/dpdk/build/meson-private/tmp0ovvvd9g
Command line:  ccache cc -I/usr/include/libnl3

/root/dpdk/build/meson-private/tmp0ovvvd9g/testfile.c -o /root/dpdk/build/meson-private/tmp0ovvvd9g/output.obj -pipe -c -D_FILE_OFFSET_BITS=64 -O0 -std=c11

Code:

 #include <infiniband/mlx4dv.h>
        void bar(void) {
            struct mlx4_wqe_lso_seg foo;
            foo.mss_hdr_size;
Quoted 5 lines
Quoted 5 lines
        };

Compiler stdout:

Compiler stderr:
In file included from /root/dpdk/build/meson-

private/tmp0ovvvd9g/testfile.c:1:

Quoted 7 lines
Quoted 7 lines

/usr/include/infiniband/mlx4dv.h:176:2: error: unknown type name 'off_t'

  176 |  off_t   uar_mmap_offset;
      |  ^~~~~

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO

Quoted 8 lines
Quoted 1 line

Thanks. I'll dig some more.

I think the meson version may be the culprit here. In my meson log I don't see the -std=c11 flag appended to the test compilation command.

Let me downgrade my meson version and see if I can reproduce.

/Bruce

Hello,

I see two other build failures.

On Ubuntu 20.04 with clang 10 and rdma-core v47.0 (built from source), I see errors similar to these:

drivers/common/mlx5/linux/mlx5_glue.h:58:6: error: redefinition of 'mlx5_ib_uapi_flow_action_packet_reformat_type' [..]
drivers/common/mlx5/linux/mlx5_glue.h:59:6: error: redefinition of 'mlx5_ib_uapi_flow_table_type' [..]
drivers/common/mlx5/linux/mlx5_glue.h:121:2: error: redefinition of enumerator 'MLX5DV_DR_ACTION_DEST_REFORMAT' [..]

On Ubuntu 20.04 while cross compiling for ppc64le with powerpc64le-linux-gnu-gcc 9.4, I see:

[..]
lib/acl/acl_run_altivec.h:44:16: error: two or more data types in declaration specifiers
[..]
lib/acl/acl_run_altivec.h:44:2: error: use of boolean types in AltiVec types is invalid
[..]
error: incompatible types when assigning to type '__vector unsigned char' {aka '__vector(16) unsigned char'} from type '__vector __bool int' {aka '__vector(4) __bool int'} [..]
lib/acl/acl_run_altivec.h:66:4: error: invalid parameter combination for AltiVec intrinsic '__builtin_vec_sel'
[..]

Regards,
Ali

#14 lore raw
Quoted 99 lines · Ali Alnubani
Quoted 86 lines

-----Original Message-----
From: Bruce Richardson <[email protected]>
Sent: Tuesday, August 1, 2023 1:51 PM
To: David Marchand <[email protected]>
Cc: Patrick Robb <[email protected]>; Tyler Retzlaff <[email protected]>; [email protected]; Morten Brørup <[email protected]>; NBU-Contact-Thomas Monjalon (EXTERNAL) <[email protected]>
Subject: Re: [PATCH v3] build: update DPDK to use C11 standard

Quoted 9 lines · Bruce Richardson
Quoted 6 lines · David Marchand
Quoted 4 lines · Bruce Richardson
Quoted 2 lines · Patrick Robb

Hi Bruce,
I see some failures for this series for our Ubuntu 20.04 containers.

<snip>

Quoted 1 line
Quoted 1 line
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines

Hi again,

I've attempted to reproduce this on my Ubuntu 20.04 VM and failed, everything seems to build ok.

Looking through the logs, though, there does appear to be a difference in the configurations in the two cases. I suspect my Ubuntu has an updated verbs package compared to the image you are using. In the log of the

failed

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

build, I see:

Checking whether type "struct mlx4_wqe_lso_seg" has member

"mss_hdr_size" with dependencies libmlx4, libibverbs: NO

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines

Configuring mlx4_autoconf.h using configuration

While building in my VM, I have:

Checking whether type "struct mlx4_wqe_lso_seg" has member

"mss_hdr_size" with dependencies libmlx4, libibverbs: YES (cached)

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

Configuring mlx4_autoconf.h using configuration

So my verbs mlx4 header has got a different set of definitions to those in the CI machine. My Ubuntu reports as 20.04.6 with libibverbs-dev

package

Quoted 29 lines
Quoted 29 lines
Quoted 6 lines

version "28.0-1ubuntu1"

Can the CI image be updated to latest 20.04 packages?

/Bruce

I can reproduce the issue seen at UNH, with a 20.04.6 container and the same libibverbs as you:
ii libibverbs-dev:amd64 28.0-1ubuntu1
amd64 Development files for the libibverbs library

So I suspect something is different in container images..

Pasting the (hopefully) relevant meson logs:

Running compile:

Working directory:  /root/dpdk/build/meson-private/tmp0ovvvd9g
Command line:  ccache cc -I/usr/include/libnl3

/root/dpdk/build/meson-private/tmp0ovvvd9g/testfile.c -o /root/dpdk/build/meson-private/tmp0ovvvd9g/output.obj -pipe -c -D_FILE_OFFSET_BITS=64 -O0 -std=c11

Code:

 #include <infiniband/mlx4dv.h>
        void bar(void) {
            struct mlx4_wqe_lso_seg foo;
            foo.mss_hdr_size;
Quoted 5 lines
Quoted 5 lines
        };

Compiler stdout:

Compiler stderr:
In file included from /root/dpdk/build/meson-

private/tmp0ovvvd9g/testfile.c:1:

Quoted 7 lines
Quoted 7 lines

/usr/include/infiniband/mlx4dv.h:176:2: error: unknown type name 'off_t'

  176 |  off_t   uar_mmap_offset;
      |  ^~~~~

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO

Quoted 8 lines
Quoted 1 line

Thanks. I'll dig some more.

I think the meson version may be the culprit here. In my meson log I don't see the -std=c11 flag appended to the test compilation command.

Let me downgrade my meson version and see if I can reproduce.

/Bruce

Hello,

I see two other build failures.

Quoted 9 lines

On Ubuntu 20.04 with clang 10 and rdma-core v47.0 (built from source), I see errors similar to these:

drivers/common/mlx5/linux/mlx5_glue.h:58:6: error: redefinition of 'mlx5_ib_uapi_flow_action_packet_reformat_type' [..]
drivers/common/mlx5/linux/mlx5_glue.h:59:6: error: redefinition of 'mlx5_ib_uapi_flow_table_type' [..]
drivers/common/mlx5/linux/mlx5_glue.h:121:2: error: redefinition of enumerator 'MLX5DV_DR_ACTION_DEST_REFORMAT' [..]

I believe I have a fix for these. The checks for the various structure members need the driver "cflags" passed as args to the check, as the std=c11 means that some things like off_t type are no longer defined (off_t is posix)

Quoted 12 lines

On Ubuntu 20.04 while cross compiling for ppc64le with powerpc64le-linux-gnu-gcc 9.4, I see:

[..]
lib/acl/acl_run_altivec.h:44:16: error: two or more data types in declaration specifiers
[..]
lib/acl/acl_run_altivec.h:44:2: error: use of boolean types in AltiVec types is invalid
[..]
error: incompatible types when assigning to type '__vector unsigned char' {aka '__vector(16) unsigned char'} from type '__vector __bool int' {aka '__vector(4) __bool int'} [..]
lib/acl/acl_run_altivec.h:66:4: error: invalid parameter combination for AltiVec intrinsic '__builtin_vec_sel'
[..]

This I still need to investigate.

In the meantime, I'll do a new version to see if the Ubuntu 20.04 issues with the mlx drivers goes away with my fix.

/Bruce

#15 lore raw
In reply to Ali Alnubani

On Tue, Aug 01, 2023 at 12:42:36PM +0000, Ali Alnubani wrote:
<snip>

Quoted 13 lines

On Ubuntu 20.04 while cross compiling for ppc64le with powerpc64le-linux-gnu-gcc 9.4, I see:

[..]
lib/acl/acl_run_altivec.h:44:16: error: two or more data types in declaration specifiers
[..]
lib/acl/acl_run_altivec.h:44:2: error: use of boolean types in AltiVec types is invalid
[..]
error: incompatible types when assigning to type '__vector unsigned char' {aka '__vector(16) unsigned char'} from type '__vector __bool int' {aka '__vector(4) __bool int'} [..]
lib/acl/acl_run_altivec.h:66:4: error: invalid parameter combination for AltiVec intrinsic '__builtin_vec_sel'
[..]

Hi Ali,

where can I get the full log for this error? Is it recorded in patchwork under a particular test set? I see compile runs for loongarch and aarch(64), but nothing listed for PPC.

/Bruce

#16 lore raw
Quoted 28 lines

-----Original Message-----
From: Bruce Richardson <[email protected]>
Sent: Tuesday, August 1, 2023 4:22 PM
To: Ali Alnubani <[email protected]>
Cc: David Marchand <[email protected]>; Patrick Robb <[email protected]>; Tyler Retzlaff <[email protected]>; [email protected]; Morten Brørup <[email protected]>; NBU- Contact-Thomas Monjalon (EXTERNAL) <[email protected]>
Subject: Re: [PATCH v3] build: update DPDK to use C11 standard

On Tue, Aug 01, 2023 at 12:42:36PM +0000, Ali Alnubani wrote:
<snip>

Quoted 2 lines

On Ubuntu 20.04 while cross compiling for ppc64le with powerpc64le-linux-

gnu-gcc 9.4, I see:

Quoted 3 lines

[..]
lib/acl/acl_run_altivec.h:44:16: error: two or more data types in declaration

specifiers

Quoted 2 lines

[..]
lib/acl/acl_run_altivec.h:44:2: error: use of boolean types in AltiVec types is

invalid

Quoted 2 lines

[..]
error: incompatible types when assigning to type '__vector unsigned char'

{aka '__vector(16) unsigned char'} from type '__vector __bool int' {aka '__vector(4) __bool int'}

Quoted 2 lines

[..]
lib/acl/acl_run_altivec.h:66:4: error: invalid parameter combination for

Quoted 8 lines

AltiVec intrinsic '__builtin_vec_sel'

Quoted 2 lines

[..]

Hi Ali,

where can I get the full log for this error? Is it recorded in patchwork under a particular test set? I see compile runs for loongarch and aarch(64), but nothing listed for PPC.

Hello Bruce,

I run this build internally. Attaching build log with more info about the environment.

Note: I'm using an out-of-tree cross file because the in-tree one doesn't set binaries.pkgconfig.

$ meson --werror --buildtype=debugoptimized --cross-file /tmp/ppc64le-power8-linux-gcc -Dexamples=all -Ddisable_drivers=*/cnxk,*/octeontx -Dc_args='-I/opt/cudart' /tmp/build-ppc64-gcc
The Meson build system
Version: 1.2.0
Source dir: /root/dpdk
Build dir: /tmp/build-ppc64-gcc
Build type: cross build
Program cat found: YES (/usr/bin/cat)
Project name: DPDK
Project version: 23.11.0-rc0
C compiler for the host machine: powerpc64le-linux-gnu-gcc (gcc 9.4.0 "powerpc64le-linux-gnu-gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0")
C linker for the host machine: powerpc64le-linux-gnu-gcc ld.bfd 2.34
C compiler for the build machine: ccache cc (gcc 9.4.0 "cc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0")
C linker for the build machine: cc ld.bfd 2.34
Build machine cpu family: x86_64
Build machine cpu: x86_64
Host machine cpu family: ppc64
Host machine cpu: power8
Target machine cpu family: ppc64
Target machine cpu: power8
Message: ## Building in Developer Mode ##
Program pkg-config found: YES (/usr/bin/pkg-config)
Program check-symbols.sh found: YES (/root/dpdk/buildtools/check-symbols.sh)
Program options-ibverbs-static.sh found: YES (/root/dpdk/buildtools/options-ibverbs-static.sh)
Program objdump found: YES (/usr/bin/objdump)
Program python3 found: YES (/usr/bin/python3)
Program cat found: YES (/usr/bin/cat)
Checking for size of "void *" : 8 
Checking for size of "void *" : 8 
Library m found: YES
Library numa found: NO
Library fdt found: NO
Library execinfo found: NO
Has header "execinfo.h" : YES 
Found pkg-config: /usr/bin/powerpc64le-linux-gnu-pkg-config (0.29.1)
Run-time dependency libarchive found: NO (tried pkgconfig)
Run-time dependency libbsd found: NO (tried pkgconfig)
Run-time dependency jansson found: NO (tried pkgconfig)
Run-time dependency openssl found: NO (tried pkgconfig)
Run-time dependency libpcap found: NO (tried pkgconfig)
Library pcap found: NO
Compiler for C supports arguments -Wcast-qual: YES 
Compiler for C supports arguments -Wdeprecated: YES 
Compiler for C supports arguments -Wformat: YES 
Compiler for C supports arguments -Wformat-nonliteral: YES 
Compiler for C supports arguments -Wformat-security: YES 
Compiler for C supports arguments -Wmissing-declarations: YES 
Compiler for C supports arguments -Wmissing-prototypes: YES 
Compiler for C supports arguments -Wnested-externs: YES 
Compiler for C supports arguments -Wold-style-definition: YES 
Compiler for C supports arguments -Wpointer-arith: YES 
Compiler for C supports arguments -Wsign-compare: YES 
Compiler for C supports arguments -Wstrict-prototypes: YES 
Compiler for C supports arguments -Wundef: YES 
Compiler for C supports arguments -Wwrite-strings: YES 
Compiler for C supports arguments -Wno-address-of-packed-member: YES 
Compiler for C supports arguments -Wno-packed-not-aligned: YES 
Compiler for C supports arguments -Wno-missing-field-initializers: YES 
Compiler for C supports arguments -mcpu=power9: YES 
Compiler for C supports arguments -Wno-format-truncation: YES 
Message: lib/kvargs: Defining dependency "kvargs"
Message: lib/telemetry: Defining dependency "telemetry"
Checking for function "getentropy" : NO 
Message: lib/eal: Defining dependency "eal"
Message: lib/ring: Defining dependency "ring"
Message: lib/rcu: Defining dependency "rcu"
Message: lib/mempool: Defining dependency "mempool"
Message: lib/mbuf: Defining dependency "mbuf"
Message: lib/net: Defining dependency "net"
Message: lib/meter: Defining dependency "meter"
Message: lib/ethdev: Defining dependency "ethdev"
Message: lib/pci: Defining dependency "pci"
Message: lib/cmdline: Defining dependency "cmdline"
Message: lib/metrics: Defining dependency "metrics"
Message: lib/hash: Defining dependency "hash"
Message: lib/timer: Defining dependency "timer"
Message: lib/acl: Defining dependency "acl"
Message: lib/bbdev: Defining dependency "bbdev"
Message: lib/bitratestats: Defining dependency "bitratestats"
Run-time dependency libelf found: NO (tried pkgconfig)
lib/bpf/meson.build:36: WARNING: libelf is missing, rte_bpf_elf_load API will be disabled
lib/bpf/meson.build:43: WARNING: libpcap is missing, rte_bpf_convert API will be disabled
Message: lib/bpf: Defining dependency "bpf"
Message: lib/cfgfile: Defining dependency "cfgfile"
Message: lib/compressdev: Defining dependency "compressdev"
Message: lib/cryptodev: Defining dependency "cryptodev"
Message: lib/distributor: Defining dependency "distributor"
Message: lib/efd: Defining dependency "efd"
Message: lib/eventdev: Defining dependency "eventdev"
Message: lib/gpudev: Defining dependency "gpudev"
Message: lib/gro: Defining dependency "gro"
Message: lib/gso: Defining dependency "gso"
Message: lib/ip_frag: Defining dependency "ip_frag"
Message: lib/jobstats: Defining dependency "jobstats"
Message: lib/latencystats: Defining dependency "latencystats"
Message: lib/lpm: Defining dependency "lpm"
Message: lib/member: Defining dependency "member"
Message: lib/pcapng: Defining dependency "pcapng"
Compiler for C supports arguments -Wno-cast-qual: YES 
Message: lib/power: Defining dependency "power"
Message: lib/rawdev: Defining dependency "rawdev"
Message: lib/regexdev: Defining dependency "regexdev"
Message: lib/mldev: Defining dependency "mldev"
Message: lib/dmadev: Defining dependency "dmadev"
Message: lib/rib: Defining dependency "rib"
Message: lib/reorder: Defining dependency "reorder"
Message: lib/sched: Defining dependency "sched"
Message: lib/security: Defining dependency "security"
Message: lib/stack: Defining dependency "stack"
Has header "linux/userfaultfd.h" : YES 
Has header "linux/vduse.h" : NO 
Message: lib/vhost: Defining dependency "vhost"
Message: lib/ipsec: Defining dependency "ipsec"
Message: lib/pdcp: Defining dependency "pdcp"
Message: lib/fib: Defining dependency "fib"
Message: lib/port: Defining dependency "port"
Message: lib/pdump: Defining dependency "pdump"
Message: lib/table: Defining dependency "table"
Message: lib/pipeline: Defining dependency "pipeline"
Message: lib/graph: Defining dependency "graph"
Message: lib/node: Defining dependency "node"
Compiler for C supports arguments -Wno-format-truncation: YES (cached)
Message: drivers/common/cpt: Defining dependency "common_cpt"
Compiler for C supports arguments -Wno-cast-qual: YES (cached)
Compiler for C supports arguments -Wno-pointer-arith: YES 
Message: drivers/common/dpaax: Defining dependency "common_dpaax"
Compiler for C supports arguments -Wno-pointer-to-int-cast: YES 
Message: drivers/common/iavf: Defining dependency "common_iavf"
Message: drivers/common/idpf: Defining dependency "common_idpf"
Run-time dependency libmusdk found: NO (tried pkgconfig)
Message: drivers/bus/auxiliary: Defining dependency "bus_auxiliary"
Message: drivers/bus/cdx: Defining dependency "bus_cdx"
Compiler for C supports arguments -Wno-cast-qual: YES (cached)
Compiler for C supports arguments -Wno-pointer-arith: YES (cached)
Message: drivers/bus/dpaa: Defining dependency "bus_dpaa"
Message: drivers/bus/fslmc: Defining dependency "bus_fslmc"
Message: drivers/bus/ifpga: Defining dependency "bus_ifpga"
Message: drivers/bus/pci: Defining dependency "bus_pci"
Message: drivers/bus/platform: Defining dependency "bus_platform"
Message: drivers/bus/vdev: Defining dependency "bus_vdev"
Message: drivers/bus/vmbus: Defining dependency "bus_vmbus"
Compiler for C supports arguments -std=c11: YES 
Compiler for C supports arguments -Wno-strict-prototypes: YES 
Compiler for C supports arguments -D_BSD_SOURCE: YES 
Compiler for C supports arguments -D_DEFAULT_SOURCE: YES 
Compiler for C supports arguments -D_XOPEN_SOURCE=600: YES 
Run-time dependency libmlx5 found: YES 1.24.48.0
Run-time dependency libibverbs found: YES 1.14.48.0
Library mtcr_ul found: NO
Header "infiniband/verbs.h" has symbol "IBV_FLOW_SPEC_ESP" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/verbs.h" has symbol "IBV_RX_HASH_IPSEC_SPI" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/verbs.h" has symbol "IBV_ACCESS_RELAXED_ORDERING " with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_CQE_RES_FORMAT_CSUM_STRIDX" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_CONTEXT_MASK_TUNNEL_OFFLOADS" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_CONTEXT_FLAGS_MPW_ALLOWED" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_CONTEXT_FLAGS_CQE_128B_COMP" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_CQ_INIT_ATTR_FLAGS_CQE_PAD" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_create_flow_action_packet_reformat" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/verbs.h" has symbol "IBV_FLOW_SPEC_MPLS" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/verbs.h" has symbol "IBV_WQ_FLAGS_PCI_WRITE_END_PADDING" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/verbs.h" has symbol "IBV_WQ_FLAG_RX_END_PADDING" with dependencies libmlx5, libibverbs: NO 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_query_devx_port" with dependencies libmlx5, libibverbs: NO 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_query_port" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_action_create_dest_ib_port" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_devx_obj_create" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_FLOW_ACTION_COUNTERS_DEVX" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_FLOW_ACTION_DEFAULT_MISS" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_devx_obj_query_async" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_devx_qp_query" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_pp_alloc" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_action_create_dest_devx_tir" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_devx_get_event" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_action_create_flow_meter" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5_MMAP_GET_NC_PAGES_CMD" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_DR_DOMAIN_TYPE_NIC_RX" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_DR_DOMAIN_TYPE_FDB" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_action_create_push_vlan" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_alloc_var" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5_OPCODE_ENHANCED_MPSW" with dependencies libmlx5, libibverbs: NO 
Header "infiniband/mlx5dv.h" has symbol "MLX5_OPCODE_SEND_EN" with dependencies libmlx5, libibverbs: NO 
Header "infiniband/mlx5dv.h" has symbol "MLX5_OPCODE_WAIT" with dependencies libmlx5, libibverbs: NO 
Header "infiniband/mlx5dv.h" has symbol "MLX5_OPCODE_ACCESS_ASO" with dependencies libmlx5, libibverbs: NO 
Header "linux/ethtool.h" has symbol "SUPPORTED_40000baseKR4_Full" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "SUPPORTED_40000baseCR4_Full" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "SUPPORTED_40000baseSR4_Full" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "SUPPORTED_40000baseLR4_Full" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "SUPPORTED_56000baseKR4_Full" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "SUPPORTED_56000baseCR4_Full" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "SUPPORTED_56000baseSR4_Full" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "SUPPORTED_56000baseLR4_Full" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "ETHTOOL_LINK_MODE_25000baseCR_Full_BIT" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "ETHTOOL_LINK_MODE_50000baseCR2_Full_BIT" with dependencies libmlx5, libibverbs: YES 
Header "linux/ethtool.h" has symbol "ETHTOOL_LINK_MODE_100000baseKR4_Full_BIT" with dependencies libmlx5, libibverbs: YES 
Header "linux/if_link.h" has symbol "IFLA_NUM_VF" with dependencies libmlx5, libibverbs: YES 
Header "linux/if_link.h" has symbol "IFLA_EXT_MASK" with dependencies libmlx5, libibverbs: YES 
Header "linux/if_link.h" has symbol "IFLA_PHYS_SWITCH_ID" with dependencies libmlx5, libibverbs: YES 
Header "linux/if_link.h" has symbol "IFLA_PHYS_PORT_NAME" with dependencies libmlx5, libibverbs: YES 
Header "rdma/rdma_netlink.h" has symbol "RDMA_NL_NLDEV" with dependencies libmlx5, libibverbs: YES 
Header "rdma/rdma_netlink.h" has symbol "RDMA_NLDEV_CMD_GET" with dependencies libmlx5, libibverbs: YES 
Header "rdma/rdma_netlink.h" has symbol "RDMA_NLDEV_CMD_PORT_GET" with dependencies libmlx5, libibverbs: YES 
Header "rdma/rdma_netlink.h" has symbol "RDMA_NLDEV_ATTR_DEV_INDEX" with dependencies libmlx5, libibverbs: YES 
Header "rdma/rdma_netlink.h" has symbol "RDMA_NLDEV_ATTR_DEV_NAME" with dependencies libmlx5, libibverbs: YES 
Header "rdma/rdma_netlink.h" has symbol "RDMA_NLDEV_ATTR_PORT_INDEX" with dependencies libmlx5, libibverbs: YES 
Header "rdma/rdma_netlink.h" has symbol "RDMA_NLDEV_ATTR_PORT_STATE" with dependencies libmlx5, libibverbs: YES 
Header "rdma/rdma_netlink.h" has symbol "RDMA_NLDEV_ATTR_NDEV_INDEX" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dump_dr_domain" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_action_create_flow_sampler" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_domain_set_reclaim_device_memory" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_action_create_dest_array" with dependencies libmlx5, libibverbs: YES 
Header "linux/devlink.h" has symbol "DEVLINK_GENL_NAME" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_action_create_aso" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/verbs.h" has symbol "INFINIBAND_VERBS_H" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5_WQE_UMR_CTRL_FLAG_INLINE" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dump_dr_rule" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "MLX5DV_DR_ACTION_FLAGS_ASO_CT_DIRECTION_INITIATOR" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_domain_allow_duplicate_rules" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/verbs.h" has symbol "ibv_reg_mr_iova" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/verbs.h" has symbol "ibv_import_device" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_dr_action_create_dest_root_table" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/mlx5dv.h" has symbol "mlx5dv_create_steering_anchor" with dependencies libmlx5, libibverbs: YES 
Header "infiniband/verbs.h" has symbol "ibv_is_fork_initialized" with dependencies libmlx5, libibverbs: YES 
Checking whether type "struct mlx5dv_sw_parsing_caps" has member "sw_parsing_offloads" with dependencies libmlx5, libibverbs: YES 
Checking whether type "struct ibv_counter_set_init_attr" has member "counter_set_id" with dependencies libmlx5, libibverbs: NO 
Checking whether type "struct ibv_counters_init_attr" has member "comp_mask" with dependencies libmlx5, libibverbs: YES 
Checking whether type "struct mlx5dv_devx_uar" has member "mmap_off" with dependencies libmlx5, libibverbs: YES 
Checking whether type "struct mlx5dv_flow_matcher_attr" has member "ft_type" with dependencies libmlx5, libibverbs: YES 
Configuring mlx5_autoconf.h using configuration
Message: drivers/common/mlx5: Defining dependency "common_mlx5"
Run-time dependency libcrypto found: NO (tried pkgconfig)
Library IPSec_MB found: NO
Message: drivers/common/qat: Defining dependency "common_qat"
Compiler for C supports arguments -Wdisabled-optimization: YES 
Compiler for C supports arguments -Waggregate-return: YES 
Compiler for C supports arguments -Wbad-function-cast: YES 
Compiler for C supports arguments -Wno-sign-compare: YES 
Compiler for C supports arguments -Wno-unused-parameter: YES 
Compiler for C supports arguments -Wno-unused-variable: YES 
Compiler for C supports arguments -Wno-empty-body: YES 
Compiler for C supports arguments -Wno-unused-but-set-variable: YES 
Message: drivers/mempool/bucket: Defining dependency "mempool_bucket"
Message: drivers/mempool/dpaa: Defining dependency "mempool_dpaa"
Message: drivers/mempool/dpaa2: Defining dependency "mempool_dpaa2"
Message: drivers/mempool/ring: Defining dependency "mempool_ring"
Message: drivers/mempool/stack: Defining dependency "mempool_stack"
Compiler for C supports arguments -Wno-pointer-arith: YES (cached)
Message: drivers/dma/dpaa: Defining dependency "dma_dpaa"
Compiler for C supports arguments -Wno-pointer-arith: YES (cached)
Message: drivers/dma/dpaa2: Defining dependency "dma_dpaa2"
Message: drivers/dma/skeleton: Defining dependency "dma_skeleton"
Message: drivers/net/af_packet: Defining dependency "net_af_packet"
Run-time dependency libxdp found: NO (tried pkgconfig)
Run-time dependency libbpf found: NO (tried pkgconfig)
Library bpf found: NO
Has header "linux/if_xdp.h" : YES 
Message: drivers/net/ark: Defining dependency "net_ark"
Message: drivers/net/atlantic: Defining dependency "net_atlantic"
Message: drivers/net/avp: Defining dependency "net_avp"
Message: drivers/net/axgbe: Defining dependency "net_axgbe"
Run-time dependency zlib found: NO (tried pkgconfig)
Compiler for C supports arguments -DSUPPORT_CFA_HW_ALL=1: YES 
Message: drivers/net/bnxt: Defining dependency "net_bnxt"
Message: drivers/net/bonding: Defining dependency "net_bond"
Message: drivers/net/cpfl: Defining dependency "net_cpfl"
Message: drivers/net/cxgbe: Defining dependency "net_cxgbe"
Compiler for C supports arguments -Wno-pointer-arith: YES (cached)
Message: drivers/net/dpaa: Defining dependency "net_dpaa"
Message: drivers/net/dpaa2: Defining dependency "net_dpaa2"
Compiler for C supports arguments -Wno-uninitialized: YES 
Compiler for C supports arguments -Wno-unused-parameter: YES (cached)
Compiler for C supports arguments -Wno-unused-variable: YES (cached)
Compiler for C supports arguments -Wno-misleading-indentation: YES 
Compiler for C supports arguments -Wno-implicit-fallthrough: YES 
Message: drivers/net/e1000: Defining dependency "net_e1000"
Message: drivers/net/ena: Defining dependency "net_ena"
Message: drivers/net/enetc: Defining dependency "net_enetc"
Message: drivers/net/enetfec: Defining dependency "net_enetfec"
Fetching value of define "__AVX2__" :  
Compiler for C supports arguments -mavx2: NO 
Message: drivers/net/enic: Defining dependency "net_enic"
Message: drivers/net/failsafe: Defining dependency "net_failsafe"
Compiler for C supports arguments -Wno-unused-parameter: YES (cached)
Compiler for C supports arguments -Wno-unused-value: YES 
Compiler for C supports arguments -Wno-strict-aliasing: YES 
Compiler for C supports arguments -Wno-format-extra-args: YES 
Compiler for C supports arguments -Wno-unused-variable: YES (cached)
Compiler for C supports arguments -Wno-implicit-fallthrough: YES (cached)
Message: drivers/net/fm10k: Defining dependency "net_fm10k"
Message: drivers/net/gve: Defining dependency "net_gve"
Message: drivers/net/hinic: Defining dependency "net_hinic"
Compiler for C supports arguments -Wno-sign-compare: YES (cached)
Compiler for C supports arguments -Wno-unused-value: YES (cached)
Compiler for C supports arguments -Wno-format: YES 
Compiler for C supports arguments -Wno-format-security: YES 
Compiler for C supports arguments -Wno-format-nonliteral: YES 
Compiler for C supports arguments -Wno-strict-aliasing: YES (cached)
Compiler for C supports arguments -Wno-unused-but-set-variable: YES (cached)
Compiler for C supports arguments -Wno-unused-parameter: YES (cached)
Message: drivers/net/i40e: Defining dependency "net_i40e"
Message: drivers/net/iavf: Defining dependency "net_iavf"
Compiler for C supports arguments -Wno-unused-value: YES (cached)
Compiler for C supports arguments -Wno-unused-but-set-variable: YES (cached)
Compiler for C supports arguments -Wno-unused-variable: YES (cached)
Compiler for C supports arguments -Wno-unused-parameter: YES (cached)
Message: drivers/net/ice: Defining dependency "net_ice"
Message: drivers/net/idpf: Defining dependency "net_idpf"
Message: drivers/net/igc: Defining dependency "net_igc"
Message: drivers/net/ionic: Defining dependency "net_ionic"
Compiler for C supports arguments -Wno-unused-value: YES (cached)
Compiler for C supports arguments -Wno-unused-but-set-variable: YES (cached)
Compiler for C supports arguments -Wno-unused-parameter: YES (cached)
Message: drivers/net/ixgbe: Defining dependency "net_ixgbe"
Message: Disabling kni [drivers/net/kni]: missing internal dependency "kni"
Message: drivers/net/memif: Defining dependency "net_memif"
Run-time dependency libmlx4 found: YES 1.0.48.0
Dependency libibverbs found: YES 1.14.48.0 (cached)
Compiler for C supports arguments -std=c11: YES (cached)
Compiler for C supports arguments -Wno-strict-prototypes: YES (cached)
Compiler for C supports arguments -D_BSD_SOURCE: YES (cached)
Compiler for C supports arguments -D_DEFAULT_SOURCE: YES (cached)
Compiler for C supports arguments -D_XOPEN_SOURCE=600: YES (cached)
Header "infiniband/mlx4dv.h" has symbol "MLX4DV_SET_CTX_ATTR_BUF_ALLOCATORS" with dependencies libmlx4, libibverbs: YES 
Header "infiniband/mlx4dv.h" has symbol "MLX4DV_QP_MASK_UAR_MMAP_OFFSET" with dependencies libmlx4, libibverbs: YES 
Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: YES 
Configuring mlx4_autoconf.h using configuration
Message: drivers/net/mlx4: Defining dependency "net_mlx4"
Compiler for C supports arguments -std=c11: YES (cached)
Compiler for C supports arguments -Wno-strict-prototypes: YES (cached)
Compiler for C supports arguments -D_BSD_SOURCE: YES (cached)
Compiler for C supports arguments -D_DEFAULT_SOURCE: YES (cached)
Compiler for C supports arguments -D_XOPEN_SOURCE=600: YES (cached)
Message: drivers/net/mlx5: Defining dependency "net_mlx5"
Run-time dependency libmusdk found: NO (tried pkgconfig)
Run-time dependency libmusdk found: NO (tried pkgconfig)
Message: drivers/net/netvsc: Defining dependency "net_netvsc"
Run-time dependency netcope-common found: NO (tried pkgconfig)
Message: drivers/net/nfp: Defining dependency "net_nfp"
Message: drivers/net/ngbe: Defining dependency "net_ngbe"
Message: drivers/net/null: Defining dependency "net_null"
Message: drivers/net/octeon_ep: Defining dependency "net_octeon_ep"
Compiler for C supports arguments -Wno-pointer-arith: YES (cached)
Message: drivers/net/pfe: Defining dependency "net_pfe"
Compiler for C supports arguments -Wno-unused-parameter: YES (cached)
Compiler for C supports arguments -Wno-sign-compare: YES (cached)
Compiler for C supports arguments -Wno-missing-prototypes: YES 
Compiler for C supports arguments -Wno-cast-qual: YES (cached)
Compiler for C supports arguments -Wno-unused-function: YES 
Compiler for C supports arguments -Wno-unused-variable: YES (cached)
Compiler for C supports arguments -Wno-strict-aliasing: YES (cached)
Compiler for C supports arguments -Wno-missing-prototypes: YES (cached)
Compiler for C supports arguments -Wno-unused-value: YES (cached)
Compiler for C supports arguments -Wno-format-nonliteral: YES (cached)
Compiler for C supports arguments -Wno-shift-negative-value: YES 
Compiler for C supports arguments -Wno-unused-but-set-variable: YES (cached)
Compiler for C supports arguments -Wno-missing-declarations: YES 
Compiler for C supports arguments -Wno-maybe-uninitialized: YES 
Compiler for C supports arguments -Wno-strict-prototypes: YES (cached)
Compiler for C supports arguments -Wno-shift-negative-value: YES (cached)
Compiler for C supports arguments -Wno-implicit-fallthrough: YES (cached)
Compiler for C supports arguments -Wno-format-extra-args: YES (cached)
Compiler for C supports arguments -Wno-visibility: NO 
Compiler for C supports arguments -Wno-empty-body: YES (cached)
Compiler for C supports arguments -Wno-invalid-source-encoding: NO 
Compiler for C supports arguments -Wno-sometimes-uninitialized: NO 
Compiler for C supports arguments -Wno-pointer-bool-conversion: NO 
Compiler for C supports arguments -Wno-format-nonliteral: YES (cached)
Message: drivers/net/qede: Defining dependency "net_qede"
Message: drivers/net/ring: Defining dependency "net_ring"
Message: drivers/net/softnic: Defining dependency "net_softnic"
Header "linux/pkt_cls.h" has symbol "TCA_FLOWER_UNSPEC" : YES 
Header "linux/pkt_cls.h" has symbol "TCA_FLOWER_KEY_VLAN_PRIO" : YES 
Header "linux/pkt_cls.h" has symbol "TCA_BPF_UNSPEC" : YES 
Header "linux/pkt_cls.h" has symbol "TCA_BPF_FD" : YES 
Header "linux/tc_act/tc_bpf.h" has symbol "TCA_ACT_BPF_UNSPEC" : YES 
Header "linux/tc_act/tc_bpf.h" has symbol "TCA_ACT_BPF_FD" : YES 
Configuring tap_autoconf.h using configuration
Message: drivers/net/tap: Defining dependency "net_tap"
Compiler for C supports arguments -fno-prefetch-loop-arrays: YES 
Compiler for C supports arguments -Wno-maybe-uninitialized: YES (cached)
Message: drivers/net/thunderx: Defining dependency "net_thunderx"
Message: drivers/net/txgbe: Defining dependency "net_txgbe"
Compiler for C supports arguments -D_BSD_SOURCE: YES (cached)
Compiler for C supports arguments -D_DEFAULT_SOURCE: YES (cached)
Compiler for C supports arguments -D_XOPEN_SOURCE=600: YES (cached)
Message: drivers/net/vdev_netvsc: Defining dependency "net_vdev_netvsc"
Message: drivers/net/vhost: Defining dependency "net_vhost"
Message: drivers/net/virtio: Defining dependency "net_virtio"
Compiler for C supports arguments -Wno-unused-parameter: YES (cached)
Compiler for C supports arguments -Wno-unused-value: YES (cached)
Compiler for C supports arguments -Wno-strict-aliasing: YES (cached)
Compiler for C supports arguments -Wno-format-extra-args: YES (cached)
Message: drivers/net/vmxnet3: Defining dependency "net_vmxnet3"
Message: Disabling cnxk_bphy [drivers/raw/cnxk_bphy]: missing internal dependency "common_cnxk"
Message: Disabling cnxk_gpio [drivers/raw/cnxk_gpio]: missing internal dependency "common_cnxk"
Message: drivers/raw/dpaa2_cmdif: Defining dependency "raw_dpaa2_cmdif"
Message: drivers/raw/ntb: Defining dependency "raw_ntb"
Message: drivers/raw/skeleton: Defining dependency "raw_skeleton"
Run-time dependency libaarch64crypto found: NO (tried pkgconfig)
Message: drivers/crypto/bcmfs: Defining dependency "crypto_bcmfs"
Message: drivers/crypto/caam_jr: Defining dependency "crypto_caam_jr"
Run-time dependency libcrypto found: NO (tried pkgconfig)
Message: drivers/crypto/dpaa_sec: Defining dependency "crypto_dpaa_sec"
Message: drivers/crypto/dpaa2_sec: Defining dependency "crypto_dpaa2_sec"
Library IPSec_MB found: NO
Compiler for C supports arguments -std=c11: YES (cached)
Compiler for C supports arguments -Wno-strict-prototypes: YES (cached)
Compiler for C supports arguments -D_BSD_SOURCE: YES (cached)
Compiler for C supports arguments -D_DEFAULT_SOURCE: YES (cached)
Compiler for C supports arguments -D_XOPEN_SOURCE=600: YES (cached)
Message: drivers/crypto/mlx5: Defining dependency "crypto_mlx5"
Run-time dependency libmusdk found: NO (tried pkgconfig)
Message: drivers/crypto/nitrox: Defining dependency "crypto_nitrox"
Message: drivers/crypto/null: Defining dependency "crypto_null"
Run-time dependency libcrypto found: NO (tried pkgconfig)
Message: drivers/crypto/scheduler: Defining dependency "crypto_scheduler"
Run-time dependency libwd_crypto found: NO (tried pkgconfig)
Run-time dependency libwd found: NO (tried pkgconfig)
Message: drivers/crypto/virtio: Defining dependency "crypto_virtio"
Run-time dependency libisal found: NO (tried pkgconfig)
Compiler for C supports arguments -std=c11: YES (cached)
Compiler for C supports arguments -Wno-strict-prototypes: YES (cached)
Compiler for C supports arguments -D_BSD_SOURCE: YES (cached)
Compiler for C supports arguments -D_DEFAULT_SOURCE: YES (cached)
Compiler for C supports arguments -D_XOPEN_SOURCE=600: YES (cached)
Message: drivers/compress/mlx5: Defining dependency "compress_mlx5"
Run-time dependency zlib found: NO (tried pkgconfig)
Compiler for C supports arguments -std=c11: YES (cached)
Compiler for C supports arguments -Wno-strict-prototypes: YES (cached)
Compiler for C supports arguments -D_BSD_SOURCE: YES (cached)
Compiler for C supports arguments -D_DEFAULT_SOURCE: YES (cached)
Compiler for C supports arguments -D_XOPEN_SOURCE=600: YES (cached)
Message: drivers/regex/mlx5: Defining dependency "regex_mlx5"
Message: Disabling cn9k [drivers/regex/cn9k]: missing internal dependency "common_cnxk"
Message: drivers/vdpa/ifc: Defining dependency "vdpa_ifc"
Compiler for C supports arguments -std=c11: YES (cached)
Compiler for C supports arguments -Wno-strict-prototypes: YES (cached)
Compiler for C supports arguments -D_BSD_SOURCE: YES (cached)
Compiler for C supports arguments -D_DEFAULT_SOURCE: YES (cached)
Compiler for C supports arguments -D_XOPEN_SOURCE=600: YES (cached)
Message: drivers/vdpa/mlx5: Defining dependency "vdpa_mlx5"
Message: drivers/event/dpaa: Defining dependency "event_dpaa"
Message: drivers/event/dpaa2: Defining dependency "event_dpaa2"
Compiler for C supports arguments -Wno-format-nonliteral: YES (cached)
Message: drivers/event/dsw: Defining dependency "event_dsw"
Message: drivers/event/opdl: Defining dependency "event_opdl"
Message: drivers/event/skeleton: Defining dependency "event_skeleton"
Message: drivers/event/sw: Defining dependency "event_sw"
Message: drivers/baseband/acc: Defining dependency "baseband_acc"
Message: drivers/baseband/fpga_5gnr_fec: Defining dependency "baseband_fpga_5gnr_fec"
Message: drivers/baseband/fpga_lte_fec: Defining dependency "baseband_fpga_lte_fec"
Message: drivers/baseband/la12xx: Defining dependency "baseband_la12xx"
Message: drivers/baseband/null: Defining dependency "baseband_null"
Found CMake: NO
Run-time dependency flexran_sdk_turbo found: NO (tried pkgconfig and cmake)
Run-time dependency flexran_sdk_ldpc_decoder_5gnr found: NO (tried pkgconfig and cmake)
Message: drivers/baseband/turbo_sw: Defining dependency "baseband_turbo_sw"
Has header "cuda.h" : YES 
Has header "cudaTypedefs.h" : YES 
Has header "gdrapi.h" : NO 
Message: drivers/gpu/cuda: Defining dependency "gpu_cuda"
Compiler for C supports arguments -Wno-format-truncation: YES (cached)
Run-time dependency zlib found: NO (tried pkgconfig)
Message: hugepage availability: false
Program test_telemetry.sh found: YES (/root/dpdk/app/test/test_telemetry.sh)
Program doxygen found: YES (/usr/bin/doxygen)
Configuring doxy-api.conf using configuration
Program sphinx-build found: YES (/usr/bin/sphinx-build)
Compiler for C supports arguments -Wno-format-truncation: YES (cached)
Message: Missing dependency "flow_classify" for example "flow_classify"
Message: Skipping example "flow_classify"
Has header "sys/epoll.h" : YES 
Library pqos found: NO
Message: Skipping example "l2fwd-cat"
Library rt found: YES
Has header "sys/epoll.h" : YES (cached)
Has header "linux/virtio_blk.h" : YES 
Library virt found: NO
Run-time dependency jansson found: NO (tried pkgconfig)
Message: Skipping example "vm_power_manager"
Library virt found: NO
Message: Skipping example "guest_cli"
Configuring rte_build_config.h using configuration
Message: 
=================
Applications Enabled
=================

apps:
pdump, proc-info, test-acl, test-bbdev, test-cmdline, test-compress-perf, test-crypto-perf, test-dma-perf, test-eventdev, test-fib, test-flow-perf, test-gpudev, test-mldev, test-pipeline, test-pmd, test-regex, test-sad, test-security-perf,

Message:
=================
Libraries Enabled
=================

libs:
kvargs, telemetry, eal, ring, rcu, mempool, mbuf, net,
meter, ethdev, pci, cmdline, metrics, hash, timer, acl,
bbdev, bitratestats, bpf, cfgfile, compressdev, cryptodev, distributor, efd, eventdev, gpudev, gro, gso, ip_frag, jobstats, latencystats, lpm, member, pcapng, power, rawdev, regexdev, mldev, dmadev, rib, reorder, sched, security, stack, vhost, ipsec, pdcp, fib, port, pdump, table, pipeline, graph, node,

Message:
===============
Drivers Enabled
===============

common:
cpt, dpaax, iavf, idpf, mlx5, qat,
bus:
auxiliary, cdx, dpaa, fslmc, ifpga, pci, platform, vdev, vmbus,
mempool:
bucket, dpaa, dpaa2, ring, stack,
dma:
dpaa, dpaa2, skeleton,
net:
af_packet, ark, atlantic, avp, axgbe, bnxt, bond, cpfl, cxgbe, dpaa, dpaa2, e1000, ena, enetc, enetfec, enic, failsafe, fm10k, gve, hinic, i40e, iavf, ice, idpf, igc, ionic, ixgbe, memif, mlx4, mlx5, netvsc, nfp, ngbe, null, octeon_ep, pfe, qede, ring, softnic, tap, thunderx, txgbe, vdev_netvsc, vhost, virtio, vmxnet3,
raw:
dpaa2_cmdif, ntb, skeleton,
crypto:
bcmfs, caam_jr, dpaa_sec, dpaa2_sec, mlx5, nitrox, null, scheduler, virtio,
compress:

	mlx5, 
regex:
	mlx5, 

ml:

vdpa:

	ifc, mlx5, 

event:
dpaa, dpaa2, dsw, opdl, skeleton, sw,
baseband:
acc, fpga_5gnr_fec, fpga_lte_fec, la12xx, null, turbo_sw,
gpu:
cuda,

Message:
=================
Content Skipped
=================

apps:
dumpcap: missing dependency, "libpcap"

libs:
kni: explicitly disabled via build config (deprecated lib)
flow_classify: explicitly disabled via build config (deprecated lib)

drivers:
common/mvep: missing dependency, "libmusdk"
common/octeontx: explicitly disabled via build config
common/cnxk: explicitly disabled via build config
crypto/qat: missing dependency, libipsecmb or libcrypto
common/sfc_efx: only supported on x86_64 and aarch64
mempool/cnxk: explicitly disabled via build config
mempool/octeontx: explicitly disabled via build config
dma/cnxk: explicitly disabled via build config
dma/hisilicon: only supported on x86_64 and aarch64
dma/idxd: only supported on x86
dma/ioat: only supported on x86
net/af_xdp: missing dependency, "libxdp >=1.2.2" and "libbpf"
net/bnx2x: missing dependency, "zlib"
net/cnxk: explicitly disabled via build config
net/hns3: only supported on x86_64 and aarch64
net/ipn3ke: missing dependency, "libfdt"
net/kni: missing internal dependency, "kni" (deprecated lib)
net/mana: only supported on x86_64 Linux
net/mvneta: missing dependency, "libmusdk"
net/mvpp2: missing dependency, "libmusdk"
net/nfb: missing dependency, "libnfb"
net/octeontx: explicitly disabled via build config
net/pcap: missing dependency, "libpcap"
net/sfc: only supported on x86_64 and aarch64
raw/cnxk_bphy: missing internal dependency, "common_cnxk"
raw/cnxk_gpio: missing internal dependency, "common_cnxk"
raw/ifpga: missing dependency, "libfdt"
crypto/armv8: missing dependency, "libAArch64crypto"
crypto/ccp: missing dependency, "libcrypto"
crypto/cnxk: explicitly disabled via build config
crypto/ipsec_mb: missing dependency, "libIPSec_MB"
crypto/mvsam: missing dependency, "libmusdk"
crypto/octeontx: explicitly disabled via build config
crypto/openssl: missing dependency, "libcrypto"
crypto/uadk: missing dependency, "libwd"
compress/isal: missing dependency, "libisal"
compress/octeontx: explicitly disabled via build config
compress/zlib: missing dependency, "zlib"
regex/cn9k: missing internal dependency, "common_cnxk"
ml/cnxk: explicitly disabled via build config
vdpa/sfc: only supported on x86_64 and aarch64
event/cnxk: explicitly disabled via build config
event/dlb2: only supported on x86_64 Linux
event/octeontx: explicitly disabled via build config

Build targets in project: 671
NOTICE: Future-deprecated features used:
* 0.58.0: {'meson.get_cross_property'}

DPDK 23.11.0-rc0

User defined options

    Cross files    : /tmp/ppc64le-power8-linux-gcc
    buildtype      : debugoptimized
    werror         : True
    c_args         : -I/opt/cudart

disable_drivers: */cnxk,*/octeontx
examples : all

Found ninja-1.10.0 at /usr/bin/ninja
WARNING: Running the setup command as meson [options] instead of meson setup [options] is ambiguous and deprecated.

$ ninja -C /tmp/build-ppc64-gcc
ninja: Entering directory `/tmp/build-ppc64-gcc'
[1/2330] Compiling C object lib/librte_eal.a.p/eal_ppc_rte_hypervisor.c.o
[2/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_hypervisor.c.o
[3/2330] Compiling C object lib/librte_eal.a.p/eal_common_rte_version.c.o
[4/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_errno.c.o
[5/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_class.c.o
[6/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_cpuflags.c.o
[7/2330] Compiling C object lib/librte_eal.a.p/eal_common_rte_reciprocal.c.o
[8/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_hexdump.c.o
[9/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_log.c.o
[10/2330] Compiling C object lib/librte_eal.a.p/eal_unix_eal_unix_timer.c.o
[11/2330] Compiling C object lib/librte_eal.a.p/eal_ppc_rte_power_intrinsics.c.o
[12/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_debug.c.o
[13/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_uuid.c.o
[14/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_cpuflags.c.o
[15/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_string_fns.c.o
[16/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_timer.c.o
[17/2330] Compiling C object lib/librte_kvargs.a.p/kvargs_rte_kvargs.c.o
[18/2330] Linking static target lib/librte_kvargs.a
[19/2330] Compiling C object lib/librte_eal.a.p/eal_ppc_rte_cycles.c.o
[20/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_config.c.o
[21/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_timer.c.o
[22/2330] Compiling C object lib/librte_eal.a.p/eal_unix_eal_unix_thread.c.o
[23/2330] Compiling C object lib/librte_telemetry.a.p/telemetry_telemetry_data.c.o
[24/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_tailqs.c.o
[25/2330] Compiling C object lib/librte_eal.a.p/eal_unix_eal_debug.c.o
[26/2330] Compiling C object lib/librte_eal.a.p/eal_unix_eal_unix_memory.c.o
[27/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_launch.c.o
[28/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_bus.c.o
[29/2330] Compiling C object lib/librte_eal.a.p/eal_common_rte_keepalive.c.o
[30/2330] Compiling C object lib/librte_eal.a.p/eal_unix_eal_firmware.c.o
[31/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_thread.c.o
[32/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_vfio_mp_sync.c.o
[33/2330] Compiling C object lib/librte_eal.a.p/eal_common_rte_random.c.o
[34/2330] Compiling C object lib/librte_eal.a.p/eal_unix_eal_file.c.o
[35/2330] Compiling C object lib/librte_telemetry.a.p/telemetry_telemetry_legacy.c.o
[36/2330] Compiling C object lib/librte_eal.a.p/eal_ppc_rte_cpuflags.c.o
[37/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_mcfg.c.o
[38/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_lcore.c.o
[39/2330] Compiling C object lib/librte_eal.a.p/eal_unix_eal_filesystem.c.o
[40/2330] Compiling C object lib/librte_eal.a.p/eal_unix_rte_thread.c.o
[41/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_trace_ctf.c.o
[42/2330] Compiling C object lib/librte_ring.a.p/ring_rte_ring.c.o
[43/2330] Compiling C object lib/librte_eal.a.p/eal_common_malloc_elem.c.o
[44/2330] Compiling C object lib/librte_eal.a.p/eal_common_hotplug_mp.c.o
[45/2330] Linking static target lib/librte_ring.a
[46/2330] Compiling C object lib/librte_mempool.a.p/mempool_rte_mempool_ops_default.c.o
[47/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_memalloc.c.o
[48/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_interrupts.c.o
[49/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_log.c.o
[50/2330] Compiling C object lib/librte_net.a.p/net_rte_net_crc.c.o
[51/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_dynmem.c.o
[52/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_alarm.c.o
[53/2330] Compiling C object lib/librte_mbuf.a.p/mbuf_rte_mbuf_pool_ops.c.o
[54/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_trace_points.c.o
[55/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_memzone.c.o
[56/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_devargs.c.o
[57/2330] Compiling C object lib/librte_mbuf.a.p/mbuf_rte_mbuf_ptype.c.o
[58/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_trace.c.o
[59/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_thread.c.o
[60/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_trace_utils.c.o
[61/2330] Compiling C object lib/librte_mempool.a.p/mempool_rte_mempool_ops.c.o
[62/2330] Generating lib/kvargs.sym_chk with a custom command (wrapped by meson to capture output)
[63/2330] Compiling C object lib/librte_eal.a.p/eal_common_malloc_mp.c.o
[64/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_dev.c.o
[65/2330] Compiling C object lib/librte_net.a.p/net_rte_ether.c.o
[66/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_dev.c.o
[67/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_vt100.c.o
[68/2330] Compiling C object lib/librte_mempool.a.p/mempool_mempool_trace_points.c.o
[69/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_hugepage_info.c.o
[70/2330] Compiling C object lib/librte_meter.a.p/meter_rte_meter.c.o
[71/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_lcore.c.o
[72/2330] Linking target lib/librte_kvargs.so.24.0
[73/2330] Linking static target lib/librte_meter.a
[74/2330] Compiling C object lib/librte_power.a.p/power_power_common.c.o
[75/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_parse_portlist.c.o
[76/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_parse_string.c.o
[77/2330] Compiling C object lib/librte_pci.a.p/pci_rte_pci.c.o
[78/2330] Linking static target lib/librte_pci.a
[79/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline.c.o
[80/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_parse_num.c.o
[81/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_socket.c.o
[82/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_cirbuf.c.o
[83/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_os_unix.c.o
[84/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_parse_etheraddr.c.o
[85/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_ethdev_profile.c.o
[86/2330] Compiling C object lib/librte_eal.a.p/eal_common_rte_malloc.c.o
[87/2330] Generating symbol file lib/librte_kvargs.so.24.0.p/librte_kvargs.so.24.0.symbols
[88/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_parse.c.o
[89/2330] Compiling C object lib/librte_mbuf.a.p/mbuf_rte_mbuf_dyn.c.o
[90/2330] Compiling C object lib/librte_net.a.p/net_rte_net.c.o
[91/2330] Compiling C object lib/librte_net.a.p/net_rte_arp.c.o
[92/2330] Compiling C object lib/librte_eal.a.p/eal_common_malloc_heap.c.o
[93/2330] Linking static target lib/librte_net.a
[94/2330] Compiling C object lib/librte_acl.a.p/acl_acl_run_altivec.c.o
FAILED: lib/librte_acl.a.p/acl_acl_run_altivec.c.o 
powerpc64le-linux-gnu-gcc -Ilib/librte_acl.a.p -Ilib -I../../root/dpdk/lib -Ilib/acl -I../../root/dpdk/lib/acl -I. -I../../root/dpdk -Iconfig -I../../root/dpdk/config -Ilib/eal/include -I../../root/dpdk/lib/eal/include -Ilib/eal/linux/include -I../../root/dpdk/lib/eal/linux/include -Ilib/eal/ppc/include -I../../root/dpdk/lib/eal/ppc/include -Ilib/eal/common -I../../root/dpdk/lib/eal/common -Ilib/eal -I../../root/dpdk/lib/eal -Ilib/kvargs -I../../root/dpdk/lib/kvargs -Ilib/metrics -I../../root/dpdk/lib/metrics -Ilib/telemetry -I../../root/dpdk/lib/telemetry -I/opt/cudart -fdiagnostics-color=always -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wextra -Werror -std=c11 -O2 -g -include rte_config.h -Wcast-qual -Wdeprecated -Wformat -Wformat-nonliteral -Wformat-security -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wpointer-arith -Wsign-compare -Wstrict-prototypes -Wundef -Wwrite-strings -Wno-address-of-packed-member -Wno-packed-not-aligned -Wno-missing-field-initializers -D_GNU_SOURCE -fPIC -mcpu=power8 -mtune=power8 -DALLOW_EXPERIMENTAL_API -DALLOW_INTERNAL_API -Wno-format-truncation -DRTE_LOG_DEFAULT_LOGTYPE=lib.acl -MD -MQ lib/librte_acl.a.p/acl_acl_run_altivec.c.o -MF lib/librte_acl.a.p/acl_acl_run_altivec.c.o.d -o lib/librte_acl.a.p/acl_acl_run_altivec.c.o -c ../../root/dpdk/lib/acl/acl_run_altivec.c
In file included from ../../root/dpdk/lib/acl/acl_run_altivec.c:6:
../../root/dpdk/lib/acl/acl_run_altivec.h: In function 'resolve_priority_altivec':
../../root/dpdk/lib/acl/acl_run_altivec.h:44:16: error: two or more data types in declaration specifiers
   44 |  __vector bool int selector;
      |                ^~~
../../root/dpdk/lib/acl/acl_run_altivec.h:44:2: error: use of boolean types in AltiVec types is invalid
   44 |  __vector bool int selector;
      |  ^~~~~~~~
../../root/dpdk/lib/acl/acl_run_altivec.h:65:4: note: use '-flax-vector-conversions' to permit conversions between vectors with differing element types or numbers of subparts
   65 |    selector = vec_cmpgt(priority1, priority);
      |    ^~~~~~~~
In file included from ../../root/dpdk/lib/eal/ppc/include/rte_altivec.h:10,
                 from ../../root/dpdk/lib/eal/ppc/include/rte_vect.h:9,
                 from ../../root/dpdk/lib/acl/rte_acl_osdep.h:37,
                 from ../../root/dpdk/lib/acl/rte_acl.h:14,
                 from ../../root/dpdk/lib/acl/acl_run.h:8,
                 from ../../root/dpdk/lib/acl/acl_run_altivec.h:6,
                 from ../../root/dpdk/lib/acl/acl_run_altivec.c:6:
../../root/dpdk/lib/acl/acl_run_altivec.h:65:15: error: incompatible types when assigning to type '__vector unsigned char' {aka '__vector(16) unsigned char'} from type '__vector __bool int' {aka '__vector(4) __bool int'}
   65 |    selector = vec_cmpgt(priority1, priority);
      |               ^~~~~~~~~
In file included from ../../root/dpdk/lib/acl/acl_run_altivec.c:6:
../../root/dpdk/lib/acl/acl_run_altivec.h:66:4: error: invalid parameter combination for AltiVec intrinsic '__builtin_vec_sel'
   66 |    results = vec_sel(results, results1, selector);
      |    ^~~~~~~
../../root/dpdk/lib/acl/acl_run_altivec.h:68:5: error: invalid parameter combination for AltiVec intrinsic '__builtin_vec_sel'
   68 |     selector);
      |     ^~~~~~~~
../../root/dpdk/lib/acl/acl_run_altivec.h: In function 'transition4':
../../root/dpdk/lib/acl/acl_run_altivec.h:113:16: error: two or more data types in declaration specifiers
  113 |  __vector bool int dfa_msk;
      |                ^~~
../../root/dpdk/lib/acl/acl_run_altivec.h:113:2: error: use of boolean types in AltiVec types is invalid
  113 |  __vector bool int dfa_msk;
      |  ^~~~~~~~
In file included from ../../root/dpdk/lib/eal/ppc/include/rte_altivec.h:10,
                 from ../../root/dpdk/lib/eal/ppc/include/rte_vect.h:9,
                 from ../../root/dpdk/lib/acl/rte_acl_osdep.h:37,
                 from ../../root/dpdk/lib/acl/rte_acl.h:14,
                 from ../../root/dpdk/lib/acl/acl_run.h:8,
                 from ../../root/dpdk/lib/acl/acl_run_altivec.h:6,
                 from ../../root/dpdk/lib/acl/acl_run_altivec.c:6:
../../root/dpdk/lib/acl/acl_run_altivec.h:137:12: error: incompatible types when assigning to type '__vector unsigned char' {aka '__vector(16) unsigned char'} from type '__vector __bool int' {aka '__vector(4) __bool int'}
  137 |  dfa_msk = vec_cmpeq(node_type, t);
      |            ^~~~~~~~~
In file included from ../../root/dpdk/lib/acl/acl_run_altivec.c:6:
../../root/dpdk/lib/acl/acl_run_altivec.h:167:2: error: invalid parameter combination for AltiVec intrinsic '__builtin_vec_sel'
  167 |  t = vec_sel(quad_ofs, dfa_ofs, dfa_msk);
      |  ^
[95/2330] Generating lib/ring.sym_chk with a custom command (wrapped by meson to capture output)
[96/2330] Compiling C object lib/librte_acl.a.p/acl_tb_mem.c.o
[97/2330] Compiling C object lib/librte_hash.a.p/hash_rte_fbk_hash.c.o
[98/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_fbarray.c.o
[99/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_parse_ipaddr.c.o
[100/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_rte_class_eth.c.o
[101/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_sff_telemetry.c.o
[102/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_sff_8472.c.o
[103/2330] Generating lib/pci.sym_chk with a custom command (wrapped by meson to capture output)
[104/2330] Compiling C object lib/librte_metrics.a.p/metrics_rte_metrics.c.o
[105/2330] Compiling C object lib/librte_eal.a.p/eal_common_rte_service.c.o
[106/2330] Generating lib/meter.sym_chk with a custom command (wrapped by meson to capture output)
[107/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_memory.c.o
[108/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_ethdev_private.c.o
[109/2330] Compiling C object lib/librte_bpf.a.p/bpf_bpf_load.c.o
[110/2330] Compiling C object lib/librte_metrics.a.p/metrics_rte_metrics_telemetry.c.o
[111/2330] Compiling C object lib/librte_bpf.a.p/bpf_bpf_stub.c.o
[112/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal.c.o
[113/2330] Compiling C object lib/librte_bpf.a.p/bpf_bpf.c.o
[114/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_proc.c.o
[115/2330] Compiling C object lib/librte_bpf.a.p/bpf_bpf_dump.c.o
[116/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_interrupts.c.o
[117/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_ethdev_driver.c.o
[118/2330] Compiling C object lib/librte_distributor.a.p/distributor_rte_distributor_match_generic.c.o
[119/2330] Compiling C object lib/librte_acl.a.p/acl_rte_acl.c.o
[120/2330] Compiling C object lib/librte_compressdev.a.p/compressdev_rte_compressdev_pmd.c.o
[121/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_memory.c.o
[122/2330] Generating lib/net.sym_chk with a custom command (wrapped by meson to capture output)
[123/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_rte_ethdev_cman.c.o
[124/2330] Compiling C object lib/librte_eventdev.a.p/eventdev_eventdev_private.c.o
[125/2330] Compiling C object lib/librte_bitratestats.a.p/bitratestats_rte_bitrate.c.o
[126/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_sff_8079.c.o
[127/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_memalloc.c.o
[128/2330] Compiling C object lib/librte_cryptodev.a.p/cryptodev_cryptodev_pmd.c.o
[129/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_sff_common.c.o
[130/2330] Compiling C object lib/librte_mempool.a.p/mempool_rte_mempool.c.o
[131/2330] Compiling C object lib/librte_cmdline.a.p/cmdline_cmdline_rdline.c.o
[132/2330] Compiling C object lib/librte_acl.a.p/acl_acl_gen.c.o
[133/2330] Compiling C object lib/librte_compressdev.a.p/compressdev_rte_comp.c.o
[134/2330] Compiling C object lib/librte_cfgfile.a.p/cfgfile_rte_cfgfile.c.o
[135/2330] Compiling C object lib/librte_acl.a.p/acl_acl_run_scalar.c.o
[136/2330] Compiling C object lib/librte_eal.a.p/eal_linux_eal_vfio.c.o
[137/2330] Compiling C object lib/librte_timer.a.p/timer_rte_timer.c.o
[138/2330] Compiling C object lib/librte_eventdev.a.p/eventdev_rte_event_ring.c.o
[139/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_sff_8636.c.o
[140/2330] Compiling C object lib/librte_compressdev.a.p/compressdev_rte_compressdev.c.o
[141/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_rte_ethdev_telemetry.c.o
[142/2330] Compiling C object lib/librte_distributor.a.p/distributor_rte_distributor_single.c.o
[143/2330] Compiling C object lib/librte_rcu.a.p/rcu_rte_rcu_qsbr.c.o
[144/2330] Compiling C object lib/librte_eal.a.p/eal_common_eal_common_options.c.o
[145/2330] Compiling C object lib/librte_hash.a.p/hash_rte_thash.c.o
[146/2330] Compiling C object lib/librte_bpf.a.p/bpf_bpf_exec.c.o
[147/2330] Compiling C object lib/librte_distributor.a.p/distributor_rte_distributor.c.o
[148/2330] Compiling C object lib/librte_bbdev.a.p/bbdev_rte_bbdev.c.o
[149/2330] Compiling C object lib/librte_acl.a.p/acl_acl_bld.c.o
[150/2330] Compiling C object lib/librte_cryptodev.a.p/cryptodev_cryptodev_trace_points.c.o
[151/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_rte_mtr.c.o
[152/2330] Compiling C object lib/librte_telemetry.a.p/telemetry_telemetry.c.o
[153/2330] Compiling C object lib/librte_bpf.a.p/bpf_bpf_validate.c.o
[154/2330] Compiling C object lib/librte_bpf.a.p/bpf_bpf_pkt.c.o
[155/2330] Compiling C object lib/librte_eventdev.a.p/eventdev_eventdev_trace_points.c.o
[156/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_rte_tm.c.o
[157/2330] Compiling C object lib/librte_mbuf.a.p/mbuf_rte_mbuf.c.o
[158/2330] Compiling C object lib/librte_efd.a.p/efd_rte_efd.c.o
[159/2330] Compiling C object lib/librte_eventdev.a.p/eventdev_rte_event_timer_adapter.c.o
[160/2330] Compiling C object lib/librte_eventdev.a.p/eventdev_rte_event_crypto_adapter.c.o
[161/2330] Compiling C object lib/librte_eventdev.a.p/eventdev_rte_event_eth_tx_adapter.c.o
[162/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_rte_flow.c.o
[163/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_ethdev_trace_points.c.o
[164/2330] Compiling C object lib/librte_hash.a.p/hash_rte_cuckoo_hash.c.o
[165/2330] Compiling C object lib/librte_cryptodev.a.p/cryptodev_rte_cryptodev.c.o
[166/2330] Compiling C object lib/librte_eventdev.a.p/eventdev_rte_event_eth_rx_adapter.c.o
[167/2330] Compiling C object lib/librte_ethdev.a.p/ethdev_rte_ethdev.c.o
ninja: build stopped: subcommand failed.

Cross-file (/tmp/ppc64le-power8-linux-gcc):
"""
[binaries]
c = 'powerpc64le-linux-gnu-gcc'
cpp = 'powerpc64le-linux-gnu-cpp'
ar = 'powerpc64le-linux-gnu-gcc-ar'
strip = 'powerpc64le-linux-gnu-strip'
pkgconfig = 'powerpc64le-linux-gnu-pkg-config'

[host_machine]
system = 'linux'
cpu_family = 'ppc64'
cpu = 'power8'
endian = 'little'
"""

OS: Ubuntu 20.04.6 LTS
Compiler: powerpc64le-linux-gnu-gcc (Ubuntu 9.4.0-1ubuntu1~20.04.1) 9.4.0
rdma-core: v47.0

Environment:
CC=gcc PKG_CONFIG_PATH=:/opt/rdma-core/build-ppc64/lib/pkgconfig

#17 lore raw

+David Christensen

Quoted 42 lines · Ali Alnubani
Quoted 28 lines

-----Original Message-----
From: Bruce Richardson <[email protected]>
Sent: Tuesday, August 1, 2023 4:22 PM
To: Ali Alnubani <[email protected]>
Cc: David Marchand <[email protected]>; Patrick Robb <[email protected]>; Tyler Retzlaff <[email protected]>; [email protected]; Morten Brørup <[email protected]>; NBU- Contact-Thomas Monjalon (EXTERNAL) <[email protected]>
Subject: Re: [PATCH v3] build: update DPDK to use C11 standard

On Tue, Aug 01, 2023 at 12:42:36PM +0000, Ali Alnubani wrote:
<snip>

Quoted 2 lines

On Ubuntu 20.04 while cross compiling for ppc64le with powerpc64le-linux-

gnu-gcc 9.4, I see:

Quoted 3 lines

[..]
lib/acl/acl_run_altivec.h:44:16: error: two or more data types in declaration

specifiers

Quoted 2 lines

[..]
lib/acl/acl_run_altivec.h:44:2: error: use of boolean types in AltiVec types is

invalid

Quoted 2 lines

[..]
error: incompatible types when assigning to type '__vector unsigned char'

{aka '__vector(16) unsigned char'} from type '__vector __bool int' {aka '__vector(4) __bool int'}

Quoted 2 lines

[..]
lib/acl/acl_run_altivec.h:66:4: error: invalid parameter combination for

Quoted 8 lines

AltiVec intrinsic '__builtin_vec_sel'

Quoted 2 lines

[..]

Hi Ali,

where can I get the full log for this error? Is it recorded in patchwork under a particular test set? I see compile runs for loongarch and aarch(64), but nothing listed for PPC.

Hello Bruce,

I run this build internally. Attaching build log with more info about the environment.

Note: I'm using an out-of-tree cross file because the in-tree one doesn't set binaries.pkgconfig.

<snip>

Quoted 8 lines

In file included from ../../root/dpdk/lib/acl/acl_run_altivec.c:6:
../../root/dpdk/lib/acl/acl_run_altivec.h: In function 'resolve_priority_altivec':
../../root/dpdk/lib/acl/acl_run_altivec.h:44:16: error: two or more data types in declaration specifiers
44 | __vector bool int selector;
| ^~~
../../root/dpdk/lib/acl/acl_run_altivec.h:44:2: error: use of boolean types in AltiVec types is invalid

   44 |  __vector bool int selector;
      |  ^~~~~~~~

I don't know anything about altivec, unfortunately, and the use of "bool int" together looks very strange. Adding PPC maintainer to thread.

David, can you perhaps help out with these issues?

Thanks,
/Bruce

#18 lore raw
Quoted 57 lines · Bruce Richardson

+David Christensen

Quoted 42 lines · Ali Alnubani
Quoted 28 lines

-----Original Message-----
From: Bruce Richardson <[email protected]>
Sent: Tuesday, August 1, 2023 4:22 PM
To: Ali Alnubani <[email protected]>
Cc: David Marchand <[email protected]>; Patrick Robb <[email protected]>; Tyler Retzlaff <[email protected]>; [email protected]; Morten Brørup <[email protected]>; NBU- Contact-Thomas Monjalon (EXTERNAL) <[email protected]>
Subject: Re: [PATCH v3] build: update DPDK to use C11 standard

On Tue, Aug 01, 2023 at 12:42:36PM +0000, Ali Alnubani wrote:
<snip>

Quoted 2 lines

On Ubuntu 20.04 while cross compiling for ppc64le with powerpc64le-linux-

gnu-gcc 9.4, I see:

Quoted 3 lines

[..]
lib/acl/acl_run_altivec.h:44:16: error: two or more data types in declaration

specifiers

Quoted 2 lines

[..]
lib/acl/acl_run_altivec.h:44:2: error: use of boolean types in AltiVec types is

invalid

Quoted 2 lines

[..]
error: incompatible types when assigning to type '__vector unsigned char'

{aka '__vector(16) unsigned char'} from type '__vector __bool int' {aka '__vector(4) __bool int'}

Quoted 2 lines

[..]
lib/acl/acl_run_altivec.h:66:4: error: invalid parameter combination for

Quoted 8 lines

AltiVec intrinsic '__builtin_vec_sel'

Quoted 2 lines

[..]

Hi Ali,

where can I get the full log for this error? Is it recorded in patchwork under a particular test set? I see compile runs for loongarch and aarch(64), but nothing listed for PPC.

Hello Bruce,

I run this build internally. Attaching build log with more info about the environment.

Note: I'm using an out-of-tree cross file because the in-tree one doesn't set binaries.pkgconfig.

<snip>

Quoted 8 lines

In file included from ../../root/dpdk/lib/acl/acl_run_altivec.c:6:
../../root/dpdk/lib/acl/acl_run_altivec.h: In function 'resolve_priority_altivec':
../../root/dpdk/lib/acl/acl_run_altivec.h:44:16: error: two or more data types in declaration specifiers
44 | __vector bool int selector;
| ^~~
../../root/dpdk/lib/acl/acl_run_altivec.h:44:2: error: use of boolean types in AltiVec types is invalid

   44 |  __vector bool int selector;
      |  ^~~~~~~~
Quoted 5 lines

I don't know anything about altivec, unfortunately, and the use of "bool int" together looks very strange. Adding PPC maintainer to thread.

David, can you perhaps help out with these issues?

Doing some google searching throws up this patch which looks relevant:
https://patches.dpdk.org/project/dpdk/patch/[email protected]/

#19 lore raw
In reply to Bruce Richardson

01/08/2023 12:19, Bruce Richardson:

Quoted 72 lines · Patrick Robb
Quoted 44 lines

Hi Bruce,
I see some failures for this series for our Ubuntu 20.04 containers. And, our DTS testbeds which are on ubuntu 20.04 are skipping running testsuites because they can't compile DPDK. So, that's why it has some missing results for a couple of the Intel NICs. For context, I'll paste below where the compile job terminates in one of our containerized compile test runs. The GCC in use here is version 9.4, so it meets the requirements as described in your patch as far as I can tell. I'll check it out more tomorrow to see whether it's an infra failure, like some missing dependencies. Please let me know if we expect to have no issues with 20.04 or if this is anticipated.
Thanks!
[1638/2730] Generating symbol file 'drivers/a715181@@rte_net_ixgbe@sha/librte_net_ixgbe.so.24.0.symbols'. [1639/2730] Compiling C object 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o'.
FAILED: drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o ccache cc -Idrivers/a715181@@tmp_rte_net_mlx4@sta -Idrivers -I../drivers -Idrivers/net/mlx4 -I../drivers/net/mlx4 -Ilib/ethdev -I../lib/ethdev -I. -I../ -Iconfig -I../config -Ilib/eal/include -I../lib/eal/include -Ilib/eal/linux/include -I../lib/eal/linux/include -Ilib/eal/x86/include -I../lib/eal/x86/include -Ilib/eal/common -I../lib/eal/common -Ilib/eal -I../lib/eal -Ilib/kvargs -I../lib/kvargs -Ilib/telemetry/../metrics -I../lib/telemetry/../metrics -Ilib/telemetry -I../lib/telemetry -Ilib/net -I../lib/net -Ilib/mbuf -I../lib/mbuf -Ilib/mempool -I../lib/mempool -Ilib/ring -I../lib/ring -Ilib/meter -I../lib/meter -Idrivers/bus/pci -I../drivers/bus/pci -I../drivers/bus/pci/linux -Ilib/pci -I../lib/pci -Idrivers/bus/vdev -I../drivers/bus/vdev -I/usr/include/libnl3 -fdiagnostics-color=always -pipe -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wextra -Werror -std=c11 -O3 -include rte_config.h -Wcast-qual -Wdeprecated -Wformat -Wformat-nonliteral -Wformat-security -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wpointer-arith -Wsign-compare -Wstrict-prototypes -Wundef -Wwrite-strings -Wno-address-of-packed-member -Wno-packed-not-aligned -Wno-missing-field-initializers -D_GNU_SOURCE -fPIC -march=native -DALLOW_EXPERIMENTAL_API -DALLOW_INTERNAL_API -Wno-format-truncation -std=c11 -Wno-strict-prototypes -D_BSD_SOURCE -D_DEFAULT_SOURCE -D_XOPEN_SOURCE=600 -UPEDANTIC -DRTE_LOG_DEFAULT_LOGTYPE=pmd.net.mlx4 -MD -MQ 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -MF 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o.d' -o 'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -c ../drivers/net/mlx4/mlx4.c
In file included from ../drivers/net/mlx4/mlx4_rxtx.h:27,

Quoted 11 lines
                    from ../drivers/net/mlx4/mlx4.c:49:
   ../drivers/net/mlx4/mlx4_prm.h:111:8: error: redefinition of 'struct
   mlx4_wqe_lso_seg'
     111 | struct mlx4_wqe_lso_seg {
         |        ^~~~~~~~~~~~~~~~
   In file included from ../drivers/net/mlx4/mlx4_glue.h:16,
                    from ../drivers/net/mlx4/mlx4.c:46:
   /usr/include/infiniband/mlx4dv.h:410:8: note: originally defined here
     410 | struct mlx4_wqe_lso_seg {
         |        ^~~~~~~~~~~~~~~~
   ninja: build stopped: subcommand failed.

Hi again,

I've attempted to reproduce this on my Ubuntu 20.04 VM and failed, everything seems to build ok.

Looking through the logs, though, there does appear to be a difference in the configurations in the two cases. I suspect my Ubuntu has an updated verbs package compared to the image you are using. In the log of the failed build, I see:

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: NO Configuring mlx4_autoconf.h using configuration

While building in my VM, I have:

Quoted 8 lines

Checking whether type "struct mlx4_wqe_lso_seg" has member "mss_hdr_size" with dependencies libmlx4, libibverbs: YES (cached) Configuring mlx4_autoconf.h using configuration

So my verbs mlx4 header has got a different set of definitions to those in the CI machine. My Ubuntu reports as 20.04.6 with libibverbs-dev package version "28.0-1ubuntu1"

Can the CI image be updated to latest 20.04 packages?

I'm surprised there is such issue.
The autoconf mechanism should manage any old package.

Ali, do you have an idea what happens?

#20 lore raw
Quoted 97 lines

-----Original Message-----
From: Thomas Monjalon <[email protected]>
Sent: Tuesday, August 1, 2023 1:37 PM
To: Patrick Robb <[email protected]>; Bruce Richardson <[email protected]>; Ali Alnubani <[email protected]>
Cc: Tyler Retzlaff <[email protected]>; [email protected]; Morten Brørup <[email protected]>; [email protected]; Raslan Darawsheh <[email protected]>
Subject: Re: [PATCH v3] build: update DPDK to use C11 standard

01/08/2023 12:19, Bruce Richardson:

Quoted 7 lines · Patrick Robb
Quoted 6 lines

Hi Bruce,
I see some failures for this series for our Ubuntu 20.04 containers. And, our DTS testbeds which are on ubuntu 20.04 are skipping running testsuites because they can't compile DPDK. So, that's why it has some missing results for a couple of the Intel NICs. For context, I'll paste below where the compile job terminates in one of our containerized

Quoted 8 lines
Quoted 8 lines

compile test runs. The GCC in use here is version 9.4, so it meets the requirements as described in your patch as far as I can tell. I'll check it out more tomorrow to see whether it's an infra failure, like some missing dependencies. Please let me know if we expect to have no issues with 20.04 or if this is anticipated.
Thanks!
[1638/2730] Generating symbol file

'drivers/a715181@@rte_net_ixgbe@sha/librte_net_ixgbe.so.24.0.symbols'.

Quoted 13 lines
Quoted 13 lines
   [1639/2730] Compiling C object
   'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o'.
   FAILED: drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o
   ccache cc -Idrivers/a715181@@tmp_rte_net_mlx4@sta -Idrivers
   -I../drivers -Idrivers/net/mlx4 -I../drivers/net/mlx4 -Ilib/ethdev
   -I../lib/ethdev -I. -I../ -Iconfig -I../config -Ilib/eal/include
   -I../lib/eal/include -Ilib/eal/linux/include -I../lib/eal/linux/include
   -Ilib/eal/x86/include -I../lib/eal/x86/include -Ilib/eal/common
   -I../lib/eal/common -Ilib/eal -I../lib/eal -Ilib/kvargs -I../lib/kvargs
   -Ilib/telemetry/../metrics -I../lib/telemetry/../metrics
   -Ilib/telemetry -I../lib/telemetry -Ilib/net -I../lib/net -Ilib/mbuf
   -I../lib/mbuf -Ilib/mempool -I../lib/mempool -Ilib/ring -I../lib/ring
   -Ilib/meter -I../lib/meter -Idrivers/bus/pci -I../drivers/bus/pci
Quoted 8 lines
Quoted 8 lines

-I../drivers/bus/pci/linux -Ilib/pci -I../lib/pci -Idrivers/bus/vdev -I../drivers/bus/vdev -I/usr/include/libnl3 -fdiagnostics-color=always -pipe -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wextra -Werror -std=c11 -O3 -include rte_config.h -Wcast-qual -Wdeprecated -Wformat -Wformat-nonliteral -Wformat-security -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wpointer-arith -Wsign-compare -Wstrict-prototypes -Wundef -Wwrite-strings -Wno-address-of-packed-member -Wno-packed-not-

aligned

Quoted 2 lines
Quoted 2 lines

-Wno-missing-field-initializers -D_GNU_SOURCE -fPIC -march=native -DALLOW_EXPERIMENTAL_API -DALLOW_INTERNAL_API -Wno-format-

truncation

Quoted 2 lines
Quoted 2 lines

-std=c11 -Wno-strict-prototypes -D_BSD_SOURCE -D_DEFAULT_SOURCE -D_XOPEN_SOURCE=600 -UPEDANTIC -

DRTE_LOG_DEFAULT_LOGTYPE=pmd.net.mlx4

Quoted 1 line
Quoted 1 line
   -MD -MQ

'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -MF

Quoted 21 lines
Quoted 7 lines
   'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o.d' -o
   'drivers/a715181@@tmp_rte_net_mlx4@sta/net_mlx4_mlx4.c.o' -c
   ../drivers/net/mlx4/mlx4.c
   In file included from ../drivers/net/mlx4/mlx4_rxtx.h:27,
                    from ../drivers/net/mlx4/mlx4.c:49:
   ../drivers/net/mlx4/mlx4_prm.h:111:8: error: redefinition of 'struct
   mlx4_wqe_lso_seg'
Quoted 8 lines
     111 | struct mlx4_wqe_lso_seg {
         |        ^~~~~~~~~~~~~~~~
   In file included from ../drivers/net/mlx4/mlx4_glue.h:16,
                    from ../drivers/net/mlx4/mlx4.c:46:
   /usr/include/infiniband/mlx4dv.h:410:8: note: originally defined here
     410 | struct mlx4_wqe_lso_seg {
         |        ^~~~~~~~~~~~~~~~
   ninja: build stopped: subcommand failed.

Hi again,

I've attempted to reproduce this on my Ubuntu 20.04 VM and failed, everything seems to build ok.

Quoted 6 lines

Looking through the logs, though, there does appear to be a difference in the configurations in the two cases. I suspect my Ubuntu has an updated verbs package compared to the image you are using. In the log of the failed build, I see:

Checking whether type "struct mlx4_wqe_lso_seg" has member

"mss_hdr_size" with dependencies libmlx4, libibverbs: NO

Quoted 5 lines

Configuring mlx4_autoconf.h using configuration

While building in my VM, I have:

Checking whether type "struct mlx4_wqe_lso_seg" has member

"mss_hdr_size" with dependencies libmlx4, libibverbs: YES (cached)

Quoted 6 lines

Configuring mlx4_autoconf.h using configuration

So my verbs mlx4 header has got a different set of definitions to those in the CI machine. My Ubuntu reports as 20.04.6 with libibverbs-dev package version "28.0-1ubuntu1"

Quoted 7 lines
Quoted 1 line

Can the CI image be updated to latest 20.04 packages?

I'm surprised there is such issue.
The autoconf mechanism should manage any old package.

Ali, do you have an idea what happens?

28.0-1ubuntu1 is the latest for focal (20.04LTS), and I can reproduce with it as well.
https://packages.ubuntu.com/focal/libibverbs-dev

It's resolved for me with v4.

#21 lore raw

PATCH v4 1/1 build: update DPDK to use C11 standard

In reply to Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

---
V4:
* pass cflags to the structure and definition checks in mlx* drivers to ensure posix definitions - as well as C-standard ones - are available.

V3:
* remove (now unneeded) use of -std=gnu99 in failsafe net driver.

V2:
* Resubmit now that 23.11-rc0 patch applied
* Add _POSIX_C_SOURCE macro to eal_common_errno.c to get POSIX definition of strerror_r() with c11 standard.

---
 doc/guides/linux_gsg/sys_reqs.rst      |  3 ++-
 doc/guides/rel_notes/deprecation.rst   | 18 ------------------
 doc/guides/rel_notes/release_23_11.rst | 17 +++++++++++++++++
 drivers/common/mlx5/linux/meson.build  |  5 +++--
 drivers/net/failsafe/meson.build       |  1 -
 drivers/net/mlx4/meson.build           |  4 ++--
 lib/eal/common/eal_common_errno.c      |  1 +
 meson.build                            |  1 +
 8 files changed, 26 insertions(+), 24 deletions(-)
doc/guides/linux_gsg/sys_reqs.rst +2 -1
index dfeaf4e1c5..13be715933 100644
@@ -27,7 +27,8 @@ Compilation of the DPDK
     The setup commands and installed packages needed on various systems may be different.
     For details on Linux distributions and the versions tested, please consult the DPDK Release Notes.
 
-*   General development tools including a supported C compiler such as gcc (version 4.9+) or clang (version 3.4+),
+*   General development tools including a C compiler supporting the C11 standard,
+    including standard atomics, for example: GCC (version 5.0+) or Clang (version 3.6+),
     and ``pkg-config`` or ``pkgconf`` to be used when building end-user binaries against DPDK.
 
     * For RHEL/Fedora systems these can be installed using ``dnf groupinstall "Development Tools"``
doc/guides/rel_notes/deprecation.rst +0 -18
index 494b401cda..cc939d3c67 100644
@@ -17,24 +17,6 @@ Other API and ABI deprecation notices are to be posted below.
 Deprecation Notices
 -------------------
 
-* C Compiler: 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)
-
 * build: Enabling deprecated libraries (``flow_classify``, ``kni``)
   won't be possible anymore through the use of the ``disable_libs`` build option.
   A new build option for deprecated libraries will be introduced instead.
doc/guides/rel_notes/release_23_11.rst +17 -0
index 6b4dd21fd0..c8b9ed456c 100644
@@ -20,6 +20,23 @@ 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
 ------------
drivers/common/mlx5/linux/meson.build +3 -2
index 15edc13041..b3a64547c5 100644
@@ -231,11 +231,12 @@ if  libmtcr_ul_found
 endif
 
 foreach arg:has_sym_args
-    mlx5_config.set(arg[0], cc.has_header_symbol(arg[1], arg[2], dependencies: libs))
+    mlx5_config.set(arg[0], cc.has_header_symbol(arg[1], arg[2], dependencies: libs, args: cflags))
 endforeach
 foreach arg:has_member_args
     file_prefix = '#include <' + arg[1] + '>'
-    mlx5_config.set(arg[0], cc.has_member(arg[2], arg[3], prefix : file_prefix, dependencies: libs))
+    mlx5_config.set(arg[0],
+            cc.has_member(arg[2], arg[3], prefix : file_prefix, dependencies: libs, args: cflags))
 endforeach
 
 # Build Glue Library
drivers/net/failsafe/meson.build +0 -1
index 6013e13722..c1d361083b 100644
@@ -7,7 +7,6 @@ if is_windows
     subdir_done()
 endif
 
-cflags += '-std=gnu99'
 cflags += '-D_DEFAULT_SOURCE'
 cflags += '-D_XOPEN_SOURCE=700'
 cflags += '-pedantic'
drivers/net/mlx4/meson.build +2 -2
index a038c1ec1b..3c5ee24186 100644
@@ -103,12 +103,12 @@ has_sym_args = [
 config = configuration_data()
 foreach arg:has_sym_args
     config.set(arg[0], cc.has_header_symbol(arg[1], arg[2],
-        dependencies: libs))
+        dependencies: libs, args: cflags))
 endforeach
 foreach arg:has_member_args
     file_prefix = '#include <' + arg[1] + '>'
     config.set(arg[0], cc.has_member(arg[2], arg[3],
-        prefix: file_prefix, dependencies: libs))
+        prefix: file_prefix, dependencies: libs, args: cflags))
 endforeach
 configure_file(output : 'mlx4_autoconf.h', configuration : config)
 
lib/eal/common/eal_common_errno.c +1 -0
index ef8f782abb..b30e2f0ad4 100644
@@ -4,6 +4,7 @@
 
 /* Use XSI-compliant portable version of strerror_r() */
 #undef _GNU_SOURCE
+#define _POSIX_C_SOURCE 200809L
 
 #include <stdio.h>
 #include <string.h>
meson.build +1 -0
index 39cb73846d..70b54f0c98 100644
@@ -9,6 +9,7 @@ project('DPDK', 'c',
         license: 'BSD',
         default_options: [
             'buildtype=release',
+            'c_std=c11',
             'default_library=static',
             'warning_level=2',
         ],
Signature · 1 line
2.39.2
#22 lore raw
Quoted 10 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 8 lines

---
V4:
* pass cflags to the structure and definition checks in mlx* drivers to ensure posix definitions - as well as C-standard ones - are available.

With this v4, mlx4 builds fine in my Ubuntu 20.04.6 container. However, I think the mlx4dv.h includes are probably faulty: as this header is using off_t, it should include sys/types.h in the first place.
https://github.com/linux-rdma/rdma-core/blob/master/providers/mlx4/mlx4dv.h#L36

This had been fixed in the mlx5 header in some rdma-core change in the
past: https://github.com/linux-rdma/rdma-core/commit/d2389b34ccc5

Signature · 1 line
David Marchand
#23 lore raw
Quoted 21 lines · David Marchand
Quoted 10 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 8 lines

---
V4:
* pass cflags to the structure and definition checks in mlx* drivers to ensure posix definitions - as well as C-standard ones - are available.

Quoted 9 lines

With this v4, mlx4 builds fine in my Ubuntu 20.04.6 container. However, I think the mlx4dv.h includes are probably faulty: as this header is using off_t, it should include sys/types.h in the first place.
https://github.com/linux-rdma/rdma-core/blob/master/providers/mlx4/mlx4dv.h#L36

This had been fixed in the mlx5 header in some rdma-core change in the
past: https://github.com/linux-rdma/rdma-core/commit/d2389b34ccc5

Even if that were fixed, I still think the correct behaviour in our build here is to test the structures using the same flags as will be used to build the final lib.

/Bruce

#24 lore raw
Quoted 29 lines · Bruce Richardson
Quoted 27 lines · David Marchand
Quoted 10 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 8 lines

---
V4:
* pass cflags to the structure and definition checks in mlx* drivers to ensure posix definitions - as well as C-standard ones - are available.

With this v4, mlx4 builds fine in my Ubuntu 20.04.6 container. However, I think the mlx4dv.h includes are probably faulty: as this header is using off_t, it should include sys/types.h in the first place.
https://github.com/linux-rdma/rdma-core/blob/master/providers/mlx4/mlx4dv.h#L36

Quoted 6 lines
Quoted 3 lines

This had been fixed in the mlx5 header in some rdma-core change in the
past: https://github.com/linux-rdma/rdma-core/commit/d2389b34ccc5

Even if that were fixed, I still think the correct behaviour in our build here is to test the structures using the same flags as will be used to build the final lib.

Yes, compiling for testing and using the structures must be aligned.

Signature · 1 line
David Marchand
#25 lore raw
In reply to Bruce Richardson
Quoted 18 lines

-----Original Message-----
From: Bruce Richardson <[email protected]>
Sent: Tuesday, August 1, 2023 4:16 PM
To: [email protected]
Cc: Bruce Richardson <[email protected]>; Morten Brørup <[email protected]>; Tyler Retzlaff <[email protected]>
Subject: [PATCH v4] build: update DPDK to use C11 standard

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 4 lines

---

By the way, I also see this build failure in RHEL 7:

[827/1011] Compiling C object app/dpdk-testpmd.p/test-pmd_cmdline_flow.c.o
FAILED: app/dpdk-testpmd.p/test-pmd_cmdline_flow.c.o
ccache cc -Iapp/dpdk-testpmd.p -Iapp -I../app -Iapp/test-pmd -I../app/test-pmd -Ilib/ethdev -I../lib/ethdev -I. -I.. -Iconfig -I../config -Ilib/eal/include -I../lib/eal/include -Ilib/eal/linux/include -I../lib/eal/linux/include -Ilib/eal/x86/include -I../lib/eal/x86/include -Ilib/eal/common -I../lib/eal/common -Ilib/eal -I../lib/eal -Ilib/kvargs -I../lib/kvargs -Ilib/metrics -I../lib/metrics -Ilib/telemetry -I../lib/telemetry -Ilib/net -I../lib/net -Ilib/mbuf -I../lib/mbuf -Ilib/mempool -I../lib/mempool -Ilib/ring -I../lib/ring -Ilib/meter -I../lib/meter -Ilib/cmdline -I../lib/cmdline -Ilib/bitratestats -I../lib/bitratestats -Ilib/bpf -I../lib/bpf -Ilib/gro -I../lib/gro -Ilib/gso -I../lib/gso -Ilib/latencystats -I../lib/latencystats -Ilib/pdump -I../lib/pdump -Ilib/pcapng -I../lib/pcapng -Idrivers/net/mlx5 -I../drivers/net/mlx5 -Idrivers/net/mlx5/linux -I../drivers/net/mlx5/linux -Idrivers/net/mlx5/hws -I../drivers/net/mlx5/hws -Idrivers/bus/pci -I../drivers/bus/pci -I../drivers/bus/pci/linux -Ilib/pci -I../lib/pci -Idrivers/bus/vdev -I../drivers/bus/vdev -Ilib/hash -I../lib/hash -Ilib/rcu -I../lib/rcu -Idrivers/common/mlx5 -I../drivers/common/mlx5 -Idrivers/common/mlx5/linux -I../drivers/common/mlx5/linux -Idrivers/bus/auxiliary -I../drivers/bus/auxiliary -pipe -D_FILE_OFFSET_BITS=64 -Wall -Winvalid-pch -Wextra -Werror -std=c11 -O3 -include rte_config.h -Wcast-qual -Wdeprecated -Wformat -Wformat-nonliteral -Wformat-security -Wmissing-declarations -Wmissing-prototypes -Wnested-externs -Wold-style-definition -Wpointer-arith -Wsign-compare -Wstrict-prototypes -Wundef -Wwrite-strings -Wno-missing-field-initializers -D_GNU_SOURCE -march=native -DALLOW_EXPERIMENTAL_API -Wno-deprecated-declarations -MD -MQ app/dpdk-testpmd.p/test-pmd_cmdline_flow.c.o -MF app/dpdk-testpmd.p/test-pmd_cmdline_flow.c.o.d -o app/dpdk-testpmd.p/test-pmd_cmdline_flow.c.o -c ../app/test-pmd/cmdline_flow.c ../app/test-pmd/cmdline_flow.c: In function ‘parse_vc_spec’:
../app/test-pmd/cmdline_flow.c:1035:37: error: array initialized from non-constant array expression

 #define NEXT_ENTRY(...) (const enum index []){ __VA_ARGS__, ZERO, }
                                     ^
../app/test-pmd/cmdline_flow.c:8049:38: note: in expansion of macro ‘NEXT_ENTRY’
   static const enum index prefix[] = NEXT_ENTRY(COMMON_PREFIX);
                                      ^
../app/test-pmd/cmdline_flow.c: In function ‘parse_vc_action_rss_type’:
../app/test-pmd/cmdline_flow.c:1035:37: error: array initialized from non-constant array expression
 #define NEXT_ENTRY(...) (const enum index []){ __VA_ARGS__, ZERO, }
                                     ^
../app/test-pmd/cmdline_flow.c:8391:35: note: in expansion of macro ‘NEXT_ENTRY’
  static const enum index next[] = NEXT_ENTRY(ACTION_RSS_TYPE);
                                   ^
../app/test-pmd/cmdline_flow.c: In function ‘parse_vc_action_rss_queue’:
../app/test-pmd/cmdline_flow.c:1035:37: error: array initialized from non-constant array expression
 #define NEXT_ENTRY(...) (const enum index []){ __VA_ARGS__, ZERO, }
                                     ^
../app/test-pmd/cmdline_flow.c:8435:35: note: in expansion of macro ‘NEXT_ENTRY’
  static const enum index next[] = NEXT_ENTRY(ACTION_RSS_QUEUE);
                                   ^

[834/1011] Compiling C object app/dpdk-testpmd.p/test-pmd_config.c.o
ninja: build stopped: subcommand failed.

Meson: 1.2.0
Gcc: 4.8.5

#26 lore raw
Quoted 18 lines · Ali Alnubani
Quoted 18 lines

-----Original Message-----
From: Bruce Richardson <[email protected]>
Sent: Tuesday, August 1, 2023 4:16 PM
To: [email protected]
Cc: Bruce Richardson <[email protected]>; Morten Brørup <[email protected]>; Tyler Retzlaff <[email protected]>
Subject: [PATCH v4] build: update DPDK to use C11 standard

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 7 lines
Quoted 4 lines

---

By the way, I also see this build failure in RHEL 7:

It's my understanding that we no longer support the default compilers in RHEL 7, since gcc 4.8.5 doesn't support the necessary c11 standard atomics.

/Bruce

#27 lore raw
Quoted 24 lines · Bruce Richardson
Quoted 23 lines · Ali Alnubani
Quoted 22 lines

-----Original Message-----
From: Bruce Richardson <[email protected]>
Sent: Tuesday, August 1, 2023 4:16 PM
To: [email protected]
Cc: Bruce Richardson <[email protected]>; Morten Brørup <[email protected]>; Tyler Retzlaff <[email protected]>
Subject: [PATCH v4] build: update DPDK to use C11 standard

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

---

Quoted 4 lines
Quoted 2 lines

By the way, I also see this build failure in RHEL 7:

It's my understanding that we no longer support the default compilers in RHEL 7, since gcc 4.8.5 doesn't support the necessary c11 standard atomics.

yes, support is being dropped for RHEL 7

http://mails.dpdk.org/archives/dev/2023-February/263516.html

it seems like now that we are in 23.11 merge window the CI pipelines for the unsupported targets can be decomissioned?

Quoted 2 lines

/Bruce

#29 lore raw
Quoted 4 lines · Patrick Robb
Quoted 3 lines · Tyler Retzlaff

yes, support is being dropped for RHEL 7

in the main branch.

Quoted 7 lines
Quoted 5 lines

http://mails.dpdk.org/archives/dev/2023-February/263516.html

it seems like now that we are in 23.11 merge window the CI pipelines for the unsupported targets can be decomissioned?

The Community Lab will discontinue CI testing on RHEL7 at start of day tomorrow

What about the LTS releases testing?

Signature · 1 line
David Marchand
#30 lore raw
Quoted 7 lines · David Marchand

What about the LTS releases testing?

--
David Marchand

Okay, we will disable rhel7 testing for dpdk main and next branches, but leave testing on for the LTS releases.

#31 lore raw
Quoted 5 lines · Patrick Robb
Quoted 2 lines · David Marchand

What about the LTS releases testing?

Okay, we will disable rhel7 testing for dpdk main and next branches, but leave testing on for the LTS releases.

Thank you Patrick.

Signature · 1 line
David Marchand
#32 lore raw

PATCH v5 1/1 build: update DPDK to use C11 standard

In reply to Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

---
V5:
* Fix build issues with bool type in altivec code, due to bool type being in C11. Use __bool for altivec-specific version instead.

V4:
* pass cflags to the structure and definition checks in mlx* drivers to ensure posix definitions - as well as C-standard ones - are available.

V3:
* remove (now unneeded) use of -std=gnu99 in failsafe net driver.

V2:
* Resubmit now that 23.11-rc0 patch applied
* Add _POSIX_C_SOURCE macro to eal_common_errno.c to get POSIX definition of strerror_r() with c11 standard.

---
 doc/guides/linux_gsg/sys_reqs.rst      |  3 ++-
 doc/guides/rel_notes/deprecation.rst   | 18 ------------------
 doc/guides/rel_notes/release_23_11.rst | 17 +++++++++++++++++
 drivers/common/mlx5/linux/meson.build  |  5 +++--
 drivers/net/failsafe/meson.build       |  1 -
 drivers/net/mlx4/meson.build           |  4 ++--
 lib/acl/acl_run_altivec.h              |  4 ++--
 lib/eal/common/eal_common_errno.c      |  1 +
 meson.build                            |  1 +
 9 files changed, 28 insertions(+), 26 deletions(-)
doc/guides/linux_gsg/sys_reqs.rst +2 -1
index dfeaf4e1c5..13be715933 100644
@@ -27,7 +27,8 @@ Compilation of the DPDK
     The setup commands and installed packages needed on various systems may be different.
     For details on Linux distributions and the versions tested, please consult the DPDK Release Notes.
 
-*   General development tools including a supported C compiler such as gcc (version 4.9+) or clang (version 3.4+),
+*   General development tools including a C compiler supporting the C11 standard,
+    including standard atomics, for example: GCC (version 5.0+) or Clang (version 3.6+),
     and ``pkg-config`` or ``pkgconf`` to be used when building end-user binaries against DPDK.
 
     * For RHEL/Fedora systems these can be installed using ``dnf groupinstall "Development Tools"``
doc/guides/rel_notes/deprecation.rst +0 -18
index 494b401cda..cc939d3c67 100644
@@ -17,24 +17,6 @@ Other API and ABI deprecation notices are to be posted below.
 Deprecation Notices
 -------------------
 
-* C Compiler: 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)
-
 * build: Enabling deprecated libraries (``flow_classify``, ``kni``)
   won't be possible anymore through the use of the ``disable_libs`` build option.
   A new build option for deprecated libraries will be introduced instead.
doc/guides/rel_notes/release_23_11.rst +17 -0
index 6b4dd21fd0..c8b9ed456c 100644
@@ -20,6 +20,23 @@ 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
 ------------
drivers/common/mlx5/linux/meson.build +3 -2
index 15edc13041..b3a64547c5 100644
@@ -231,11 +231,12 @@ if  libmtcr_ul_found
 endif
 
 foreach arg:has_sym_args
-    mlx5_config.set(arg[0], cc.has_header_symbol(arg[1], arg[2], dependencies: libs))
+    mlx5_config.set(arg[0], cc.has_header_symbol(arg[1], arg[2], dependencies: libs, args: cflags))
 endforeach
 foreach arg:has_member_args
     file_prefix = '#include <' + arg[1] + '>'
-    mlx5_config.set(arg[0], cc.has_member(arg[2], arg[3], prefix : file_prefix, dependencies: libs))
+    mlx5_config.set(arg[0],
+            cc.has_member(arg[2], arg[3], prefix : file_prefix, dependencies: libs, args: cflags))
 endforeach
 
 # Build Glue Library
drivers/net/failsafe/meson.build +0 -1
index 6013e13722..c1d361083b 100644
@@ -7,7 +7,6 @@ if is_windows
     subdir_done()
 endif
 
-cflags += '-std=gnu99'
 cflags += '-D_DEFAULT_SOURCE'
 cflags += '-D_XOPEN_SOURCE=700'
 cflags += '-pedantic'
drivers/net/mlx4/meson.build +2 -2
index a038c1ec1b..3c5ee24186 100644
@@ -103,12 +103,12 @@ has_sym_args = [
 config = configuration_data()
 foreach arg:has_sym_args
     config.set(arg[0], cc.has_header_symbol(arg[1], arg[2],
-        dependencies: libs))
+        dependencies: libs, args: cflags))
 endforeach
 foreach arg:has_member_args
     file_prefix = '#include <' + arg[1] + '>'
     config.set(arg[0], cc.has_member(arg[2], arg[3],
-        prefix: file_prefix, dependencies: libs))
+        prefix: file_prefix, dependencies: libs, args: cflags))
 endforeach
 configure_file(output : 'mlx4_autoconf.h', configuration : config)
 
lib/acl/acl_run_altivec.h +2 -2
index 4556e1503b..3c30466d2d 100644
@@ -41,7 +41,7 @@ resolve_priority_altivec(uint64_t transition, int n,
 {
 	uint32_t x;
 	xmm_t results, priority, results1, priority1;
-	__vector bool int selector;
+	__vector __bool int selector;
 	xmm_t *saved_results, *saved_priority;
 
 	for (x = 0; x < categories; x += RTE_ACL_RESULTS_MULTIPLIER) {
@@ -110,7 +110,7 @@ transition4(xmm_t next_input, const uint64_t *trans,
 	xmm_t in, node_type, r, t;
 	xmm_t dfa_ofs, quad_ofs;
 	xmm_t *index_mask, *tp;
-	__vector bool int dfa_msk;
+	__vector __bool int dfa_msk;
 	__vector signed char zeroes = {};
 	union {
 		uint64_t d64[2];
lib/eal/common/eal_common_errno.c +1 -0
index ef8f782abb..b30e2f0ad4 100644
@@ -4,6 +4,7 @@
 
 /* Use XSI-compliant portable version of strerror_r() */
 #undef _GNU_SOURCE
+#define _POSIX_C_SOURCE 200809L
 
 #include <stdio.h>
 #include <string.h>
meson.build +1 -0
index 39cb73846d..70b54f0c98 100644
@@ -9,6 +9,7 @@ project('DPDK', 'c',
         license: 'BSD',
         default_options: [
             'buildtype=release',
+            'c_std=c11',
             'default_library=static',
             'warning_level=2',
         ],
Signature · 1 line
2.39.2
#33 lore raw
Quoted 9 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 8 lines

---
V5:
* Fix build issues with bool type in altivec code, due to bool type being in C11. Use __bool for altivec-specific version instead.

Just a note on the altivec builds: from my testing on latest ubuntu with "powerpc64le-linux-gnu-gcc (Ubuntu 12.3.0-1ubuntu1~23.04)", there are quite a few warnings/errors building the current DPDK mainline. This patch doesn't fix any of those errors, but neither does it add any new ones, as far as I can see! However, we may need some more PPC fixes and updated CI coverage to catch these issues sooner in future.

/Bruce

Tested-by on v5 1/1
#34 lore raw
In reply to Bruce Richardson
Quoted 34 lines

-----Original Message-----
From: Bruce Richardson <[email protected]>
Sent: Wednesday, August 2, 2023 3:32 PM
To: [email protected]
Cc: Bruce Richardson <[email protected]>; Morten Brørup <[email protected]>; Tyler Retzlaff <[email protected]>
Subject: [PATCH v5] build: update DPDK to use C11 standard

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

---
V5:
* Fix build issues with bool type in altivec code, due to bool type being in C11. Use __bool for altivec-specific version instead.

V4:
* pass cflags to the structure and definition checks in mlx* drivers to ensure posix definitions - as well as C-standard ones - are available.

V3:
* remove (now unneeded) use of -std=gnu99 in failsafe net driver.

Quoted 6 lines

V2:
* Resubmit now that 23.11-rc0 patch applied
* Add _POSIX_C_SOURCE macro to eal_common_errno.c to get POSIX definition of strerror_r() with c11 standard.

---

Thanks,
Ali

Tested-by Ali Alnubani on v5 1/1 Applied
#35 lore raw
In reply to Bruce Richardson
Quoted 10 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 3 lines

The CI results look good.

Applied, thanks!

Signature · 1 line
David Marchand
#36 lore raw

03/08/2023 15:36, David Marchand:

Quoted 12 lines
Quoted 10 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 8 lines
Quoted 3 lines

The CI results look good.

Applied, thanks!

The compiler support is updated, that's fine.
Should we go further and document some major Linux distributions? One concern is to make clear RHEL 7 is not supported anymore. Should it be a release note?

#37 lore raw
Quoted 18 lines · On Thu, 10 Aug 2023 15:34:43 +0200 Thomas Monjalon <[email protected]> wrote:

03/08/2023 15:36, David Marchand:

Quoted 17 lines
Quoted 10 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 3 lines
Quoted 10 lines
Quoted 3 lines

The CI results look good.

Applied, thanks!

The compiler support is updated, that's fine.
Should we go further and document some major Linux distributions? One concern is to make clear RHEL 7 is not supported anymore. Should it be a release note?

Should be addressed in linux/sys_reqs.rst as well as deprecation notice. Also, is it possible to add automated check in build for compiler version?

#38 lore raw
Quoted 23 lines · Stephen Hemminger
Quoted 20 lines · On Thu, 10 Aug 2023 15:34:43 +0200 Thomas Monjalon <[email protected]> wrote:

03/08/2023 15:36, David Marchand:

Quoted 19 lines
Quoted 10 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 3 lines

The CI results look good.

Quoted 7 lines
Quoted 7 lines
Quoted 1 line

Applied, thanks!

The compiler support is updated, that's fine.
Should we go further and document some major Linux distributions? One concern is to make clear RHEL 7 is not supported anymore. Should it be a release note?

Well, DPDK currently is still building fine on Centos 7 for me, so let's hold off on claiming anything until it's definitely broken.

Quoted 4 lines
Quoted 1 line

Should be addressed in linux/sys_reqs.rst as well as deprecation notice. Also, is it possible to add automated check in build for compiler version?

I'd be a little careful about what we claim, and I think current docs are accurate vs our original plans. What we didn't plan to support was the GCC and Clang compiler versions in RHEL 7, but if one installs an updated GCC, for example, the build should be fine on RHEL 7.

Now, though, we are having to re-evaluate our use of stdatomics, which means we may not actually break RHEL 7 compatibility after all. We'll have to "watch this space" as the saying goes!

Overall, I think the approach of build-time checks is the best, but not for specific versions, but instead for capabilities. If/when we add support for stdatomics to DPDK builds on Linux/BSD, at that point we put in the initial compiler checks a suitable check for them being present and output a suitable error if not found.

/Bruce

#39 lore raw

10/08/2023 18:35, Bruce Richardson:

Quoted 49 lines · Stephen Hemminger
Quoted 23 lines
Quoted 20 lines · On Thu, 10 Aug 2023 15:34:43 +0200 Thomas Monjalon <[email protected]> wrote:

03/08/2023 15:36, David Marchand:

Quoted 19 lines
Quoted 10 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 3 lines

The CI results look good.

Quoted 7 lines
Quoted 7 lines
Quoted 1 line

Applied, thanks!

The compiler support is updated, that's fine.
Should we go further and document some major Linux distributions? One concern is to make clear RHEL 7 is not supported anymore. Should it be a release note?

Well, DPDK currently is still building fine on Centos 7 for me, so let's hold off on claiming anything until it's definitely broken.

Quoted 4 lines
Quoted 1 line

Should be addressed in linux/sys_reqs.rst as well as deprecation notice. Also, is it possible to add automated check in build for compiler version?

I'd be a little careful about what we claim, and I think current docs are accurate vs our original plans. What we didn't plan to support was the GCC and Clang compiler versions in RHEL 7, but if one installs an updated GCC, for example, the build should be fine on RHEL 7.

Now, though, we are having to re-evaluate our use of stdatomics, which means we may not actually break RHEL 7 compatibility after all. We'll have to "watch this space" as the saying goes!

Quoted 5 lines

Overall, I think the approach of build-time checks is the best, but not for specific versions, but instead for capabilities. If/when we add support for stdatomics to DPDK builds on Linux/BSD, at that point we put in the initial compiler checks a suitable check for them being present and output a suitable error if not found.

OK looks good

#40 lore raw
Quoted 50 lines · On Thu, 10 Aug 2023 18:49:09 +0200 Thomas Monjalon <[email protected]> wrote:

10/08/2023 18:35, Bruce Richardson:

Quoted 49 lines · Stephen Hemminger
Quoted 23 lines
Quoted 20 lines · On Thu, 10 Aug 2023 15:34:43 +0200 Thomas Monjalon <[email protected]> wrote:

03/08/2023 15:36, David Marchand:

Quoted 19 lines
Quoted 10 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11 supporting compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are not in the C standard are unavailable by default. We fix this by ensuring the correct defines or cflags are passed to the components that need them.

Quoted 3 lines

The CI results look good.

Quoted 7 lines
Quoted 7 lines
Quoted 1 line

Applied, thanks!

The compiler support is updated, that's fine.
Should we go further and document some major Linux distributions? One concern is to make clear RHEL 7 is not supported anymore. Should it be a release note?

Well, DPDK currently is still building fine on Centos 7 for me, so let's hold off on claiming anything until it's definitely broken.

Quoted 4 lines
Quoted 1 line

Should be addressed in linux/sys_reqs.rst as well as deprecation notice. Also, is it possible to add automated check in build for compiler version?

I'd be a little careful about what we claim, and I think current docs are accurate vs our original plans. What we didn't plan to support was the GCC and Clang compiler versions in RHEL 7, but if one installs an updated GCC, for example, the build should be fine on RHEL 7.

Now, though, we are having to re-evaluate our use of stdatomics, which means we may not actually break RHEL 7 compatibility after all. We'll have to "watch this space" as the saying goes!

Quoted 7 lines
Quoted 5 lines

Overall, I think the approach of build-time checks is the best, but not for specific versions, but instead for capabilities. If/when we add support for stdatomics to DPDK builds on Linux/BSD, at that point we put in the initial compiler checks a suitable check for them being present and output a suitable error if not found.

OK looks good

Note: RHEL 7 official end of maintenance support is not until June 2024.

#41 lore raw
Quoted 68 lines

From: Stephen Hemminger [mailto:[email protected]]
Sent: Thursday, 10 August 2023 19.03

Quoted 10 lines · On Thu, 10 Aug 2023 18:49:09 +0200 Thomas Monjalon <[email protected]> wrote:

10/08/2023 18:35, Bruce Richardson:

Quoted 9 lines · Stephen Hemminger
Quoted 8 lines
Quoted 5 lines · On Thu, 10 Aug 2023 15:34:43 +0200 Thomas Monjalon <[email protected]> wrote:

03/08/2023 15:36, David Marchand:

Quoted 4 lines
Quoted 2 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11

supporting

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are

not in

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

the C standard are unavailable by default. We fix this by

ensuring

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

the correct defines or cflags are passed to the components

that

Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
Quoted 10 lines
Quoted 5 lines

need them.

The CI results look good.

Applied, thanks!

Quoted 7 lines
Quoted 7 lines
Quoted 5 lines
Quoted 5 lines

The compiler support is updated, that's fine.
Should we go further and document some major Linux distributions? One concern is to make clear RHEL 7 is not supported anymore. Should it be a release note?

Well, DPDK currently is still building fine on Centos 7 for me, so

let's

Quoted 5 lines
Quoted 5 lines

hold off on claiming anything until it's definitely broken.

Quoted 3 lines
Quoted 1 line

Should be addressed in linux/sys_reqs.rst as well as deprecation

notice.

Quoted 1 line
Quoted 1 line
Quoted 1 line

Also, is it possible to add automated check in build for compiler

version?

Quoted 2 lines
Quoted 2 lines

I'd be a little careful about what we claim, and I think current docs

are

Quoted 1 line
Quoted 1 line

accurate vs our original plans. What we didn't plan to support was

the GCC

Quoted 1 line
Quoted 1 line

and Clang compiler versions in RHEL 7, but if one installs an updated

GCC,

Quoted 3 lines
Quoted 3 lines

for example, the build should be fine on RHEL 7.

Now, though, we are having to re-evaluate our use of stdatomics,

which

Quoted 1 line
Quoted 1 line

means we may not actually break RHEL 7 compatibility after all. We'll

have

Quoted 2 lines
Quoted 2 lines

to "watch this space" as the saying goes!

Quoted 9 lines
Quoted 1 line
Quoted 1 line

Overall, I think the approach of build-time checks is the best, but

not

Quoted 1 line
Quoted 1 line

for specific versions, but instead for capabilities. If/when we add

support

Quoted 1 line
Quoted 1 line

for stdatomics to DPDK builds on Linux/BSD, at that point we put in

the

Quoted 1 line
Quoted 1 line

initial compiler checks a suitable check for them being present and

output

Quoted 1 line
Quoted 1 line

a suitable error if not found.

Exactly. Capabilities checks is the right way to go when cross compiling.

Quoted 5 lines
Quoted 2 lines

OK looks good

Note: RHEL 7 official end of maintenance support is not until June 2024.

It was agreed to abandon RHEL 7, mainly driven by the need for C11 stdatomic.h, which is not supported by the GCC C compiler included with RHEL 7. So it pains me to admit that Stephen has a valid point here, after it turned out that the GCC g++ is not C11 compatible.

Regardless, I think that DPDK 23.11 support for RHEL 7 should be limited to "might work on RHEL 7", rather than guaranteed support for RHEL 7 (which would require DPDK CI to resume testing on RHEL 7).

IIRC, there was also the argument that DPDK 23.11 LTS support ends after June 2024.

Here's another argument for abandoning RHEL 7: RHEL 7 uses Linux Kernel 3.10. Although DPDK requires Linux Kernel >= 4.14, we promise backwards compatibility for RHEL/CentOS 7. Do we really still want to do that? (Note: RHEL 8 uses Linux Kernel 4.18.)

While we're discussing the Linux Kernel version required... Is it documented anywhere why a specific Linux Kernel version is required by DPDK? Or more specifically: Is it documented anywhere which DPDK features require which specific Linux Kernel versions?

#42 lore raw
Quoted 81 lines · Morten Brørup
Quoted 68 lines

From: Stephen Hemminger [mailto:[email protected]]
Sent: Thursday, 10 August 2023 19.03

Quoted 10 lines · On Thu, 10 Aug 2023 18:49:09 +0200 Thomas Monjalon <[email protected]> wrote:

10/08/2023 18:35, Bruce Richardson:

Quoted 9 lines · Stephen Hemminger
Quoted 8 lines
Quoted 5 lines · On Thu, 10 Aug 2023 15:34:43 +0200 Thomas Monjalon <[email protected]> wrote:

03/08/2023 15:36, David Marchand:

Quoted 4 lines
Quoted 2 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11

supporting

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are

not in

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

the C standard are unavailable by default. We fix this by

ensuring

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

the correct defines or cflags are passed to the components

that

Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
Quoted 10 lines
Quoted 5 lines

need them.

The CI results look good.

Applied, thanks!

Quoted 7 lines
Quoted 7 lines
Quoted 5 lines
Quoted 5 lines

The compiler support is updated, that's fine.
Should we go further and document some major Linux distributions? One concern is to make clear RHEL 7 is not supported anymore. Should it be a release note?

Well, DPDK currently is still building fine on Centos 7 for me, so

let's

Quoted 5 lines
Quoted 5 lines

hold off on claiming anything until it's definitely broken.

Quoted 3 lines
Quoted 1 line

Should be addressed in linux/sys_reqs.rst as well as deprecation

notice.

Quoted 1 line
Quoted 1 line
Quoted 1 line

Also, is it possible to add automated check in build for compiler

version?

Quoted 2 lines
Quoted 2 lines

I'd be a little careful about what we claim, and I think current docs

are

Quoted 1 line
Quoted 1 line

accurate vs our original plans. What we didn't plan to support was

the GCC

Quoted 1 line
Quoted 1 line

and Clang compiler versions in RHEL 7, but if one installs an updated

GCC,

Quoted 3 lines
Quoted 3 lines

for example, the build should be fine on RHEL 7.

Now, though, we are having to re-evaluate our use of stdatomics,

which

Quoted 1 line
Quoted 1 line

means we may not actually break RHEL 7 compatibility after all. We'll

have

Quoted 2 lines
Quoted 2 lines

to "watch this space" as the saying goes!

Quoted 9 lines
Quoted 1 line
Quoted 1 line

Overall, I think the approach of build-time checks is the best, but

not

Quoted 1 line
Quoted 1 line

for specific versions, but instead for capabilities. If/when we add

support

Quoted 1 line
Quoted 1 line

for stdatomics to DPDK builds on Linux/BSD, at that point we put in

the

Quoted 1 line
Quoted 1 line

initial compiler checks a suitable check for them being present and

output

Quoted 1 line
Quoted 1 line

a suitable error if not found.

Exactly. Capabilities checks is the right way to go when cross compiling.

Quoted 1 line
Quoted 1 line
Quoted 6 lines
Quoted 4 lines
Quoted 1 line

OK looks good

Note: RHEL 7 official end of maintenance support is not until June 2024.

It was agreed to abandon RHEL 7, mainly driven by the need for C11 stdatomic.h, which is not supported by the GCC C compiler included with RHEL 7. So it pains me to admit that Stephen has a valid point here, after it turned out that the GCC g++ is not C11 compatible.

we would substantially reduce porting delta to retain C11, there are a number of other things that help with portability from C11 that we can utilized that i hadn't brought up before since it had been resolved to adopt it.

it would be really unfortunate to say we aren't going to require C11 since that would cause me to have to bring a lot more conditional compile into the tree.

just fyi

Quoted 9 lines

Regardless, I think that DPDK 23.11 support for RHEL 7 should be limited to "might work on RHEL 7", rather than guaranteed support for RHEL 7 (which would require DPDK CI to resume testing on RHEL 7).

IIRC, there was also the argument that DPDK 23.11 LTS support ends after June 2024.

Here's another argument for abandoning RHEL 7: RHEL 7 uses Linux Kernel 3.10. Although DPDK requires Linux Kernel >= 4.14, we promise backwards compatibility for RHEL/CentOS 7. Do we really still want to do that? (Note: RHEL 8 uses Linux Kernel 4.18.)

While we're discussing the Linux Kernel version required... Is it documented anywhere why a specific Linux Kernel version is required by DPDK? Or more specifically: Is it documented anywhere which DPDK features require which specific Linux Kernel versions?

#43 lore raw
Quoted 94 lines · Tyler Retzlaff
Quoted 81 lines · Morten Brørup
Quoted 68 lines

From: Stephen Hemminger [mailto:[email protected]]
Sent: Thursday, 10 August 2023 19.03

Quoted 10 lines · On Thu, 10 Aug 2023 18:49:09 +0200 Thomas Monjalon <[email protected]> wrote:

10/08/2023 18:35, Bruce Richardson:

Quoted 9 lines · Stephen Hemminger
Quoted 8 lines
Quoted 5 lines · On Thu, 10 Aug 2023 15:34:43 +0200 Thomas Monjalon <[email protected]> wrote:

03/08/2023 15:36, David Marchand:

Quoted 4 lines
Quoted 2 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11

supporting

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the standard with GNU extensions, means that some posix definitions which are

not in

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

the C standard are unavailable by default. We fix this by

ensuring

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

the correct defines or cflags are passed to the components

that

Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
Quoted 11 lines
Quoted 10 lines
Quoted 5 lines

need them.

The CI results look good.

Applied, thanks!

Quoted 7 lines
Quoted 7 lines
Quoted 5 lines
Quoted 5 lines

The compiler support is updated, that's fine.
Should we go further and document some major Linux distributions? One concern is to make clear RHEL 7 is not supported anymore. Should it be a release note?

Well, DPDK currently is still building fine on Centos 7 for me, so

let's

Quoted 5 lines
Quoted 5 lines

hold off on claiming anything until it's definitely broken.

Quoted 3 lines
Quoted 1 line

Should be addressed in linux/sys_reqs.rst as well as deprecation

notice.

Quoted 1 line
Quoted 1 line
Quoted 1 line

Also, is it possible to add automated check in build for compiler

version?

Quoted 2 lines
Quoted 2 lines

I'd be a little careful about what we claim, and I think current docs

are

Quoted 1 line
Quoted 1 line

accurate vs our original plans. What we didn't plan to support was

the GCC

Quoted 1 line
Quoted 1 line

and Clang compiler versions in RHEL 7, but if one installs an updated

GCC,

Quoted 3 lines
Quoted 3 lines

for example, the build should be fine on RHEL 7.

Now, though, we are having to re-evaluate our use of stdatomics,

which

Quoted 1 line
Quoted 1 line

means we may not actually break RHEL 7 compatibility after all. We'll

have

Quoted 2 lines
Quoted 2 lines

to "watch this space" as the saying goes!

Quoted 9 lines
Quoted 1 line
Quoted 1 line

Overall, I think the approach of build-time checks is the best, but

not

Quoted 1 line
Quoted 1 line

for specific versions, but instead for capabilities. If/when we add

support

Quoted 1 line
Quoted 1 line

for stdatomics to DPDK builds on Linux/BSD, at that point we put in

the

Quoted 1 line
Quoted 1 line

initial compiler checks a suitable check for them being present and

output

Quoted 1 line
Quoted 1 line

a suitable error if not found.

Exactly. Capabilities checks is the right way to go when cross compiling.

Quoted 1 line
Quoted 1 line
Quoted 6 lines
Quoted 4 lines
Quoted 1 line

OK looks good

Note: RHEL 7 official end of maintenance support is not until June 2024.

It was agreed to abandon RHEL 7, mainly driven by the need for C11 stdatomic.h, which is not supported by the GCC C compiler included with RHEL 7. So it pains me to admit that Stephen has a valid point here, after it turned out that the GCC g++ is not C11 compatible.

we would substantially reduce porting delta to retain C11, there are a number of other things that help with portability from C11 that we can utilized that i hadn't brought up before since it had been resolved to adopt it.

Quoted 6 lines

it would be really unfortunate to say we aren't going to require C11 since that would cause me to have to bring a lot more conditional compile into the tree.

just fyi

As far as I know we are requiring C11, and the meson.build file on main tree currently has set the minimum for that. We just haven't got code in there yet that uses the C11 atomics, which is the bit that is missing from the RHEL 7 compiler.

I agree that we should not actually actively support RHEL 7 - I just wouldn't call attention to the fact that it's no longer supported if it still happens to work. Once we actually break it, I'm all for documenting that it's not supported. [NOTE: I both cases, I'm not saying that we call it out explicitly as supported either - I'm just talking about a release note entry saying the opposite!]

/Bruce

#44 lore raw
Quoted 136 lines

From: Bruce Richardson [mailto:[email protected]]
Sent: Friday, 11 August 2023 10.53

Quoted 9 lines · Tyler Retzlaff
Quoted 8 lines · Morten Brørup
Quoted 8 lines

From: Stephen Hemminger [mailto:[email protected]]
Sent: Thursday, 10 August 2023 19.03

Quoted 2 lines · On Thu, 10 Aug 2023 18:49:09 +0200 Thomas Monjalon <[email protected]> wrote:

10/08/2023 18:35, Bruce Richardson:

Quoted 1 line

On Thu, Aug 10, 2023 at 07:48:39AM -0700, Stephen Hemminger

Quoted 4 lines · wrote:
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 1 line · On Thu, 10 Aug 2023 15:34:43 +0200 Thomas Monjalon <[email protected]> wrote:

03/08/2023 15:36, David Marchand:

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 2 lines · Bruce Richardson

As previously announced, DPDK 23.11 will require a C11

supporting

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

compiler and will use the C11 standard in all builds.

Forcing use of the C standard, rather than the

standard with

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

GNU extensions, means that some posix definitions

which are

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

not in

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

the C standard are unavailable by default. We fix this

by

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

ensuring

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

the correct defines or cflags are passed to the

components

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

that

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

need them.

Signed-off-by Bruce Richardson

<[email protected]>

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Quoted 2 lines
Acked-by Tyler Retzlaff

<[email protected]>

Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 8 lines
Quoted 5 lines

The CI results look good.

Applied, thanks!

The compiler support is updated, that's fine. Should we go further and document some major Linux

distributions?

Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line
Quoted 1 line

One concern is to make clear RHEL 7 is not supported

anymore.

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 4 lines
Quoted 2 lines
Quoted 2 lines

Should it be a release note?

Well, DPDK currently is still building fine on Centos 7 for

me, so

Quoted 6 lines
Quoted 6 lines
Quoted 6 lines

let's

Quoted 5 lines
Quoted 5 lines

hold off on claiming anything until it's definitely broken.

Quoted 3 lines
Quoted 1 line

Should be addressed in linux/sys_reqs.rst as well as

deprecation

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

notice.

Quoted 1 line
Quoted 1 line
Quoted 1 line

Also, is it possible to add automated check in build for

compiler

Quoted 3 lines
Quoted 3 lines
Quoted 3 lines

version?

Quoted 2 lines
Quoted 2 lines

I'd be a little careful about what we claim, and I think

current docs

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

are

Quoted 1 line
Quoted 1 line

accurate vs our original plans. What we didn't plan to support

was

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

the GCC

Quoted 1 line
Quoted 1 line

and Clang compiler versions in RHEL 7, but if one installs an

updated

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

GCC,

Quoted 3 lines
Quoted 3 lines

for example, the build should be fine on RHEL 7.

Now, though, we are having to re-evaluate our use of

stdatomics,

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

which

Quoted 1 line
Quoted 1 line

means we may not actually break RHEL 7 compatibility after

all. We'll

Quoted 4 lines
Quoted 4 lines
Quoted 4 lines

have

Quoted 3 lines
Quoted 3 lines

to "watch this space" as the saying goes!

Overall, I think the approach of build-time checks is the

best, but

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

not

Quoted 1 line
Quoted 1 line

for specific versions, but instead for capabilities. If/when

we add

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

support

Quoted 1 line
Quoted 1 line

for stdatomics to DPDK builds on Linux/BSD, at that point we

put in

Quoted 2 lines
Quoted 2 lines
Quoted 2 lines

the

Quoted 1 line
Quoted 1 line

initial compiler checks a suitable check for them being

present and

Quoted 4 lines
Quoted 4 lines
Quoted 2 lines

output

Quoted 1 line
Quoted 1 line

a suitable error if not found.

Exactly. Capabilities checks is the right way to go when cross

compiling.

Quoted 5 lines
Quoted 5 lines
Quoted 5 lines
Quoted 3 lines

OK looks good

Note: RHEL 7 official end of maintenance support is not until June

2024.

Quoted 3 lines
Quoted 3 lines
Quoted 1 line

It was agreed to abandon RHEL 7, mainly driven by the need for C11

stdatomic.h, which is not supported by the GCC C compiler included with RHEL 7. So it pains me to admit that Stephen has a valid point here, after it turned out that the GCC g++ is not C11 compatible.

Quoted 6 lines

we would substantially reduce porting delta to retain C11, there are a number of other things that help with portability from C11 that we can utilized that i hadn't brought up before since it had been resolved to adopt it.

Quoted 6 lines

it would be really unfortunate to say we aren't going to require C11 since that would cause me to have to bring a lot more conditional compile into the tree.

just fyi

As far as I know we are requiring C11, and the meson.build file on main tree currently has set the minimum for that. We just haven't got code in there yet that uses the C11 atomics, which is the bit that is missing from
the RHEL 7 compiler.

Quoted 8 lines

I agree that we should not actually actively support RHEL 7 - I just wouldn't call attention to the fact that it's no longer supported if it still happens to work. Once we actually break it, I'm all for documenting
that it's not supported. [NOTE: I both cases, I'm not saying that we call
it out explicitly as supported either - I'm just talking about a release note entry saying the opposite!]

If I was using RHEL 7, and DPDK degraded the RHEL 7 support level (e.g. not testing it in CI anymore) with a release, I would certainly want this to be highlighted in the release notes! "Not known to be broken, might work" and "passed validation" are two very different things in a production environment. :-)

Anyway, Bruce's point is valid; it's not binary, so let's not scare RHEL 7 users more than we have to.

Quoted 2 lines

/Bruce