Improve core EAL musl compatibility

PATCH Superseded Anatoly Burakov 4 participants 36 messages Aug 29, 2018 – Oct 22, 2018

Subsystems (4)

Network drivers Library code Arm/Amlogic Meson SoC support Crypto API

lore

Share & tools

lore

Anatoly Burakov

author

2018-08-29 11:56 UTC #0 lore raw

cover Improve core EAL musl compatibility

This patchset fixes numerous issues with musl compatibility in the core EAL libraries. It does not fix anything beyond core EAL (so, PCI driver is still broken, so are a few other drivers), but it's a good start.

Tested on container with Alpine Linux. Alpine dependencies:

build-base bsd-compat-headers libexecinfo-dev linux-headers numactl-dev

For numactl-dev, testing repository needs to be enabled:

echo "http://dl-cdn.alpinelinux.org/alpine/edge/testing" >> /etc/apk/repositories

If successful (using a very broad definition of "success"), the build should fail somewhere in PCI bus driver in UIO.

Anatoly Burakov (7):
linuxapp: build with _GNU_SOURCE defined by default pci/vfio: improve musl compatibility
fbarray: improve musl compatibility eal/hugepage_info: improve musl compatibility
mem: improve musl compatibility
string_fns: improve musl compatibility
eal: improve musl compatibility

 app/meson.build                                 |  9 ++++++++-
 drivers/bus/pci/linux/Makefile                  |  2 --
 drivers/bus/pci/linux/pci_vfio.c                |  8 ++++----
 drivers/meson.build                             |  6 ++++++
 drivers/net/softnic/conn.c                      |  1 -
 examples/meson.build                            |  6 ++++++
 lib/librte_eal/common/eal_common_fbarray.c      |  1 +
 lib/librte_eal/common/eal_common_memory.c       |  1 +
 lib/librte_eal/common/include/rte_string_fns.h  |  1 +
 lib/librte_eal/linuxapp/eal/Makefile            | 16 ----------------
 lib/librte_eal/linuxapp/eal/eal.c               |  5 +++--
 lib/librte_eal/linuxapp/eal/eal_hugepage_info.c |  1 +
 lib/librte_eal/linuxapp/eal/eal_memory.c        |  1 +
 lib/librte_eal/linuxapp/eal/eal_thread.c        |  5 +++--
 lib/meson.build                                 |  6 ++++++
 mk/exec-env/linuxapp/rte.vars.mk                |  2 ++
 test/test/meson.build                           |  5 +++++
 17 files changed, 48 insertions(+), 28 deletions(-)

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-08-29 11:56 UTC #1 lore raw

PATCH 1/7 linuxapp: build with _GNU_SOURCE defined by default

We use _GNU_SOURCE all over the place, but often times we miss defining it, resulting in broken builds on musl. Rather than fixing every library's and driver's and application's makefile, fix it by simply defining _GNU_SOURCE by default for all Linuxapp builds.

Signed-off-by Anatoly Burakov <[email protected]>

---
 app/meson.build                      |  9 ++++++++-
 drivers/bus/pci/linux/Makefile       |  2 --
 drivers/meson.build                  |  6 ++++++
 drivers/net/softnic/conn.c           |  1 -
 examples/meson.build                 |  6 ++++++
 lib/librte_eal/linuxapp/eal/Makefile | 16 ----------------
 lib/meson.build                      |  6 ++++++
 mk/exec-env/linuxapp/rte.vars.mk     |  2 ++
 test/test/meson.build                |  5 +++++
 9 files changed, 33 insertions(+), 20 deletions(-)

app/meson.build +8 -1

index 99e0b93ec..c9a52a22b 100644
@@ -11,13 +11,20 @@ apps = ['pdump',
 # for BSD only
 lib_execinfo = cc.find_library('execinfo', required: false)
 
+default_cflags = machine_args
+
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 foreach app:apps
 	build = true
 	name = app
 	allow_experimental_apis = false
 	sources = []
 	includes = []
-	cflags = machine_args
+	cflags = default_cflags
 	objs = [] # other object files to link against, used e.g. for
 	          # instruction-set optimized versions of code

drivers/bus/pci/linux/Makefile +0 -2

index 96ea1d540..90404468b 100644
@@ -4,5 +4,3 @@
 SRCS += pci.c
 SRCS += pci_uio.c
 SRCS += pci_vfio.c
-
-CFLAGS += -D_GNU_SOURCE

drivers/meson.build +6 -0

index f94e2fe67..7ff97ef4e 100644
@@ -15,6 +15,12 @@ default_cflags = machine_args
 if cc.has_argument('-Wno-format-truncation')
 	default_cflags += '-Wno-format-truncation'
 endif
+
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 foreach class:driver_classes
 	drivers = []
 	std_deps = []

drivers/net/softnic/conn.c +0 -1

index 990cf40fc..8b6658088 100644
@@ -8,7 +8,6 @@
 #include <unistd.h>
 #include <sys/types.h>
 
-#define __USE_GNU
 #include <sys/socket.h>
 
 #include <sys/epoll.h>

examples/meson.build +6 -0

index 4ee7a1114..70c22eb62 100644
@@ -22,6 +22,12 @@ default_cflags = machine_args
 if cc.has_argument('-Wno-format-truncation')
 	default_cflags += '-Wno-format-truncation'
 endif
+
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 foreach example: examples
 	name = example
 	build = true

lib/librte_eal/linuxapp/eal/Makefile +0 ~~-16~~

index fd92c75c2..bfee453bc 100644
@@ -85,22 +85,6 @@ SRCS-y += rte_cycles.c
 
 CFLAGS_eal_common_cpuflags.o := $(CPUFLAGS_LIST)
 
-CFLAGS_eal.o := -D_GNU_SOURCE
-CFLAGS_eal_interrupts.o := -D_GNU_SOURCE
-CFLAGS_eal_vfio_mp_sync.o := -D_GNU_SOURCE
-CFLAGS_eal_timer.o := -D_GNU_SOURCE
-CFLAGS_eal_lcore.o := -D_GNU_SOURCE
-CFLAGS_eal_memalloc.o := -D_GNU_SOURCE
-CFLAGS_eal_thread.o := -D_GNU_SOURCE
-CFLAGS_eal_log.o := -D_GNU_SOURCE
-CFLAGS_eal_common_log.o := -D_GNU_SOURCE
-CFLAGS_eal_hugepage_info.o := -D_GNU_SOURCE
-CFLAGS_eal_common_whitelist.o := -D_GNU_SOURCE
-CFLAGS_eal_common_options.o := -D_GNU_SOURCE
-CFLAGS_eal_common_thread.o := -D_GNU_SOURCE
-CFLAGS_eal_common_lcore.o := -D_GNU_SOURCE
-CFLAGS_rte_cycles.o := -D_GNU_SOURCE
-
 # workaround for a gcc bug with noreturn attribute
 # http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603
 ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y)

lib/meson.build +6 -0

index eb91f100b..4c1577571 100644
@@ -30,6 +30,12 @@ default_cflags = machine_args
 if cc.has_argument('-Wno-format-truncation')
 	default_cflags += '-Wno-format-truncation'
 endif
+
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 foreach l:libraries
 	build = true
 	name = l

mk/exec-env/linuxapp/rte.vars.mk +2 -0

index 3129edc8c..91b778fcc 100644
@@ -17,6 +17,8 @@ else
 EXECENV_CFLAGS  = -pthread
 endif
 
+EXECENV_CFLAGS += -D_GNU_SOURCE
+
 EXECENV_LDLIBS  =
 EXECENV_ASFLAGS =

test/test/meson.build +5 -0

index b1dd6eca2..c81fca439 100644
@@ -242,6 +242,11 @@ if cc.has_argument('-Wno-format-truncation')
     cflags += '-Wno-format-truncation'
 endif
 
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 test_dep_objs = []
 compress_test_dep = dependency('zlib', required: false)
 if compress_test_dep.found()

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-08-29 11:57 UTC #2 lore raw

PATCH 2/7 pci/vfio: improve musl compatibility

In reply to Anatoly Burakov

Musl already has PAGE_SIZE defined, and our define clashed with it. Rename our define to SYS_PAGE_SIZE.

Bugzilla ID: 36

Signed-off-by Anatoly Burakov <[email protected]>

---
 drivers/bus/pci/linux/pci_vfio.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

drivers/bus/pci/linux/pci_vfio.c +4 -4

index 686386d6a..88bcfb88b 100644
@@ -35,8 +35,8 @@
 
 #ifdef VFIO_PRESENT
 
-#define PAGE_SIZE   (sysconf(_SC_PAGESIZE))
-#define PAGE_MASK   (~(PAGE_SIZE - 1))
+#define SYS_PAGE_SIZE   (sysconf(_SC_PAGESIZE))
+#define SYS_PAGE_MASK   (~(SYS_PAGE_SIZE - 1))
 
 static struct rte_tailq_elem rte_vfio_tailq = {
 	.name = "VFIO_RESOURCE_LIST",

@@ -344,8 +344,8 @@ pci_vfio_mmap_bar(int vfio_dev_fd, struct mapped_pci_resource *vfio_res,
 		 */
 		uint32_t table_start = msix_table->offset;
 		uint32_t table_end = table_start + msix_table->size;
-		table_end = (table_end + ~PAGE_MASK) & PAGE_MASK;
-		table_start &= PAGE_MASK;
+		table_end = (table_end + ~SYS_PAGE_MASK) & SYS_PAGE_MASK;
+		table_start &= SYS_PAGE_MASK;
 
 		if (table_start == 0 && table_end >= bar->size) {
 			/* Cannot map this BAR */

Signature · 1 line

2.17.1

Bruce Richardson

2018-08-29 12:35 UTC #3 lore raw

Re: [PATCH 2/7] pci/vfio: improve musl compatibility

Quoted 7 lines · Anatoly Burakov

Musl already has PAGE_SIZE defined, and our define clashed with it. Rename our define to SYS_PAGE_SIZE.
Bugzilla ID: 36
Signed-off-by Anatoly Burakov <[email protected]>
---

Would it not be easier to just do?

#ifndef PAGE_SIZE
#define PAGE_SIZE ...
#endif

Quoted 31 lines

 drivers/bus/pci/linux/pci_vfio.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)

drivers/bus/pci/linux/pci_vfio.c +4 -4

index 686386d6a..88bcfb88b 100644
@@ -35,8 +35,8 @@
 
 #ifdef VFIO_PRESENT
 
-#define PAGE_SIZE   (sysconf(_SC_PAGESIZE))
-#define PAGE_MASK   (~(PAGE_SIZE - 1))
+#define SYS_PAGE_SIZE   (sysconf(_SC_PAGESIZE))
+#define SYS_PAGE_MASK   (~(SYS_PAGE_SIZE - 1))
 
 static struct rte_tailq_elem rte_vfio_tailq = {
 	.name = "VFIO_RESOURCE_LIST",

@@ -344,8 +344,8 @@ pci_vfio_mmap_bar(int vfio_dev_fd, struct mapped_pci_resource *vfio_res,
 		 */
 		uint32_t table_start = msix_table->offset;
 		uint32_t table_end = table_start + msix_table->size;
-		table_end = (table_end + ~PAGE_MASK) & PAGE_MASK;
-		table_start &= PAGE_MASK;
+		table_end = (table_end + ~SYS_PAGE_MASK) & SYS_PAGE_MASK;
+		table_start &= SYS_PAGE_MASK;
 
 		if (table_start == 0 && table_end >= bar->size) {
 			/* Cannot map this BAR */

Signature · 1 line

2.17.1

Burakov, Anatoly

author

2018-08-29 14:09 UTC #4 lore raw

Re: [PATCH 2/7] pci/vfio: improve musl compatibility

Quoted 6 lines · Bruce Richardson

Quoted 5 lines · Anatoly Burakov
Musl already has PAGE_SIZE defined, and our define clashed with it. Rename our define to SYS_PAGE_SIZE.
Bugzilla ID: 36

Quoted 8 lines

Quoted 2 lines
Signed-off-by Anatoly Burakov <[email protected]>
---
Would it not be easier to just do?
#ifndef PAGE_SIZE
#define PAGE_SIZE ...
#endif

Sure, that can work.

Signature · 2 lines

Thanks,
Anatoly

Anatoly Burakov

author

2018-08-29 11:56 UTC #5 lore raw

PATCH 3/7 fbarray: improve musl compatibility

In reply to Anatoly Burakov

When built against musl, fcntl.h doesn't silently get included. Fix by including it explicitly.

Bugzilla ID: 34

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/common/eal_common_fbarray.c | 1 +
 1 file changed, 1 insertion(+)

lib/librte_eal/common/eal_common_fbarray.c +1 -0

index 43caf3ced..6f0169a5c 100644
@@ -2,6 +2,7 @@
  * Copyright(c) 2017-2018 Intel Corporation
  */
 
+#include <fcntl.h>
 #include <inttypes.h>
 #include <limits.h>
 #include <sys/mman.h>

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-08-29 11:56 UTC #6 lore raw

PATCH 4/7 eal/hugepage_info: improve musl compatibility

In reply to Anatoly Burakov

When built against musl, fcntl.h doesn't silently get included. Fix by including it explicitly.

Bugzilla ID: 33

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/linuxapp/eal/eal_hugepage_info.c | 1 +
 1 file changed, 1 insertion(+)

lib/librte_eal/linuxapp/eal/eal_hugepage_info.c +1 -0

index 3a7d4b222..0eab1cf71 100644
@@ -6,6 +6,7 @@
 #include <sys/types.h>
 #include <sys/file.h>
 #include <dirent.h>
+#include <fcntl.h>
 #include <stdint.h>
 #include <stdlib.h>
 #include <stdio.h>

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-08-29 11:56 UTC #7 lore raw

PATCH 5/7 mem: improve musl compatibility

In reply to Anatoly Burakov

When built against musl, fcntl.h doesn't silently get included. Fix by including it explicitly.

Bugzilla ID: 31

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/common/eal_common_memory.c | 1 +
 lib/librte_eal/linuxapp/eal/eal_memory.c  | 1 +
 2 files changed, 2 insertions(+)

lib/librte_eal/common/eal_common_memory.c +1 -0

index fbfb1b055..b061f7528 100644
@@ -2,6 +2,7 @@
  * Copyright(c) 2010-2014 Intel Corporation
  */
 
+#include <fcntl.h>
 #include <errno.h>
 #include <stdio.h>
 #include <stdint.h>

lib/librte_eal/linuxapp/eal/eal_memory.c +1 -0

index dbf19499e..a62665257 100644
@@ -5,6 +5,7 @@
 
 #define _FILE_OFFSET_BITS 64
 #include <errno.h>
+#include <fcntl.h>
 #include <stdarg.h>
 #include <stdbool.h>
 #include <stdlib.h>

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-08-29 11:57 UTC #8 lore raw

PATCH 6/7 string_fns: improve musl compatibility

In reply to Anatoly Burakov

Musl wraps various string functions such as strlcpy in order to harden them. However, the fortify wrappers are included without including the actual string functions being wrapped, which throws missing definition compile errors. Fix by including string.h in string functions header.

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/common/include/rte_string_fns.h | 1 +
 1 file changed, 1 insertion(+)

lib/librte_eal/common/include/rte_string_fns.h +1 -0

index 97597a148..ffeed2cd2 100644
@@ -16,6 +16,7 @@ extern "C" {
 #endif
 
 #include <stdio.h>
+#include <string.h>
 
 /**
  * Takes string "string" parameter and splits it at character "delim"

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-08-29 11:57 UTC #9 lore raw

PATCH 7/7 eal: improve musl compatibility

In reply to Anatoly Burakov

Musl complains about pthread id being of wrong size. Fix it by casting to 64-bit and printing 64-bit hex unconditionally.

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/linuxapp/eal/eal.c        | 5 +++--
 lib/librte_eal/linuxapp/eal/eal_thread.c | 5 +++--
 2 files changed, 6 insertions(+), 4 deletions(-)

lib/librte_eal/linuxapp/eal/eal.c +3 -2

index e59ac6577..abd61d346 100644
@@ -3,6 +3,7 @@
  * Copyright(c) 2012-2014 6WIND S.A.
  */
 
+#include <inttypes.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <stdint.h>

@@ -979,8 +980,8 @@ rte_eal_init(int argc, char **argv)
 
 	ret = eal_thread_dump_affinity(cpuset, sizeof(cpuset));
 
-	RTE_LOG(DEBUG, EAL, "Master lcore %u is ready (tid=%x;cpuset=[%s%s])\n",
-		rte_config.master_lcore, (int)thread_id, cpuset,
+	RTE_LOG(DEBUG, EAL, "Master lcore %u is ready (tid=%" PRIx64" ;cpuset=[%s%s])\n",
+		rte_config.master_lcore, (uint64_t)thread_id, cpuset,
 		ret == 0 ? "" : "...");
 
 	RTE_LCORE_FOREACH_SLAVE(i) {

lib/librte_eal/linuxapp/eal/eal_thread.c +3 -2

index b496fc711..c818375d9 100644
@@ -3,6 +3,7 @@
  */
 
 #include <errno.h>
+#include <inttypes.h>
 #include <stdio.h>
 #include <stdlib.h>
 #include <stdint.h>

@@ -121,8 +122,8 @@ eal_thread_loop(__attribute__((unused)) void *arg)
 
 	ret = eal_thread_dump_affinity(cpuset, sizeof(cpuset));
 
-	RTE_LOG(DEBUG, EAL, "lcore %u is ready (tid=%x;cpuset=[%s%s])\n",
-		lcore_id, (int)thread_id, cpuset, ret == 0 ? "" : "...");
+	RTE_LOG(DEBUG, EAL, "lcore %u is ready (tid=%" PRIx64 ";cpuset=[%s%s])\n",
+		lcore_id, (uint64_t)thread_id, cpuset, ret == 0 ? "" : "...");
 
 	/* read on our pipe to get commands */
 	while (1) {

Signature · 1 line

2.17.1

Bruce Richardson

2018-08-29 12:39 UTC #10 lore raw

Re: [PATCH 7/7] eal: improve musl compatibility

Quoted 5 lines · Anatoly Burakov

Musl complains about pthread id being of wrong size. Fix it by casting to 64-bit and printing 64-bit hex unconditionally.
Signed-off-by Anatoly Burakov <[email protected]>
---

Given that on linux pthread_t is a pointer type, will this not give other warnings of casting from pointer to integer of a different type when compiling 32-bit? For safety I suggest casting to long or uintptr_t instead, to ensure we always get an int of the right size.

/Bruce

Bruce Richardson

2018-08-29 12:40 UTC #11 lore raw

Re: [PATCH 7/7] eal: improve musl compatibility

Quoted 8 lines · Bruce Richardson

Quoted 5 lines · Anatoly Burakov
Musl complains about pthread id being of wrong size. Fix it by casting to 64-bit and printing 64-bit hex unconditionally.
Signed-off-by Anatoly Burakov <[email protected]>
---
Given that on linux pthread_t is a pointer type, will this not give other warnings of casting from pointer to integer of a different type when

s/type/size/

Quoted 4 lines

compiling 32-bit? For safety I suggest casting to long or uintptr_t instead, to ensure we always get an int of the right size.
/Bruce

Burakov, Anatoly

author

2018-08-29 14:09 UTC #12 lore raw

Re: [PATCH 7/7] eal: improve musl compatibility

In reply to Bruce Richardson

Quoted 4 lines · Bruce Richardson

Quoted 3 lines · Anatoly Burakov
Musl complains about pthread id being of wrong size. Fix it by casting to 64-bit and printing 64-bit hex unconditionally.

Quoted 9 lines

Quoted 2 lines
Signed-off-by Anatoly Burakov <[email protected]>
---
Given that on linux pthread_t is a pointer type, will this not give other warnings of casting from pointer to integer of a different type when compiling 32-bit? For safety I suggest casting to long or uintptr_t instead, to ensure we always get an int of the right size.
/Bruce

Sure, will fix.

Signature · 2 lines

Thanks,
Anatoly

Stephen Hemminger

2018-08-29 16:47 UTC #13 lore raw

Re: [PATCH 7/7] eal: improve musl compatibility

Quoted 12 lines · Anatoly

Quoted 11 lines · Bruce Richardson
Quoted 5 lines · Anatoly Burakov
Musl complains about pthread id being of wrong size. Fix it by casting to 64-bit and printing 64-bit hex unconditionally.
Signed-off-by Anatoly Burakov <[email protected]>
---
Given that on linux pthread_t is a pointer type, will this not give other warnings of casting from pointer to integer of a different type when compiling 32-bit? For safety I suggest casting to long or uintptr_t instead, to ensure we always get an int of the right size.

Quoted 8 lines

Quoted 2 lines
/Bruce
Sure, will fix.
Signature · 2 lines
Thanks,
Anatoly

Maybe use gettid() to get thread id which is actually way more useful than the pointer value. Of course, glibc doesn't want to provide a syscall wrapper for this.

Burakov, Anatoly

author

2018-08-30 09:13 UTC #14 lore raw

Re: [PATCH 7/7] eal: improve musl compatibility

Quoted 20 lines · Stephen Hemminger

Quoted 17 lines · Anatoly
Quoted 4 lines · Bruce Richardson
Quoted 3 lines · Anatoly Burakov
Musl complains about pthread id being of wrong size. Fix it by casting to 64-bit and printing 64-bit hex unconditionally.
Quoted 9 lines
Quoted 2 lines
Signed-off-by Anatoly Burakov <[email protected]>
---
Given that on linux pthread_t is a pointer type, will this not give other warnings of casting from pointer to integer of a different type when compiling 32-bit? For safety I suggest casting to long or uintptr_t instead, to ensure we always get an int of the right size.
/Bruce
Sure, will fix.

Quoted 9 lines

Quoted 3 lines
Signature · 2 lines
Thanks,
Anatoly
Maybe use gettid() to get thread id which is actually way more useful than the pointer value. Of course, glibc doesn't want to provide a syscall wrapper for this.

We have rte_gettid() call.

Signature · 2 lines

Thanks,
Anatoly

Stephen Hemminger

2018-08-29 16:43 UTC #15 lore raw

Re: [PATCH 7/7] eal: improve musl compatibility

In reply to Anatoly Burakov

Quoted 4 lines · On Wed, 29 Aug 2018 12:56:21 +0100 Anatoly Burakov <[email protected]>…

Musl complains about pthread id being of wrong size. Fix it by casting to 64-bit and printing 64-bit hex unconditionally.
Signed-off-by Anatoly Burakov <[email protected]>

What is pthread_t on musl? On Linux it is unsigned long.

Anatoly Burakov

author

2018-09-20 15:28 UTC #16 lore raw

v2 cover Improve core EAL musl compatibility

In reply to Anatoly Burakov

Tested on container with Alpine Linux. Alpine dependencies:

build-base bsd-compat-headers libexecinfo-dev linux-headers numactl-dev

For numactl-dev, testing repository needs to be enabled:

echo "http://dl-cdn.alpinelinux.org/alpine/edge/testing" >> /etc/apk/repositories

If successful (using a very broad definition of "success"), the build should fail somewhere in PCI bus driver in UIO.

v2 -> v1:
- Fixed patch 2 to use existing define if available
- Fixed patch 7 to use proper format specifier and cast pthread ID to unsigned pointer type

 app/meson.build                                 |  9 ++++++++-
 drivers/bus/pci/linux/Makefile                  |  2 --
 drivers/bus/pci/linux/pci_vfio.c                |  2 ++
 drivers/meson.build                             |  6 ++++++
 drivers/net/softnic/conn.c                      |  1 -
 examples/meson.build                            |  6 ++++++
 lib/librte_eal/common/eal_common_fbarray.c      |  1 +
 lib/librte_eal/common/eal_common_memory.c       |  1 +
 lib/librte_eal/common/include/rte_string_fns.h  |  1 +
 lib/librte_eal/linuxapp/eal/Makefile            | 16 ----------------
 lib/librte_eal/linuxapp/eal/eal.c               |  4 ++--
 lib/librte_eal/linuxapp/eal/eal_hugepage_info.c |  1 +
 lib/librte_eal/linuxapp/eal/eal_memory.c        |  1 +
 lib/librte_eal/linuxapp/eal/eal_thread.c        |  4 ++--
 lib/meson.build                                 |  6 ++++++
 mk/exec-env/linuxapp/rte.vars.mk                |  2 ++
 test/test/meson.build                           |  5 +++++
 17 files changed, 44 insertions(+), 24 deletions(-)

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-09-20 15:27 UTC #17 lore raw

PATCH v2 1/7 linuxapp: build with _GNU_SOURCE defined by default

Signed-off-by Anatoly Burakov <[email protected]>

---
 app/meson.build                      |  9 ++++++++-
 drivers/bus/pci/linux/Makefile       |  2 --
 drivers/meson.build                  |  6 ++++++
 drivers/net/softnic/conn.c           |  1 -
 examples/meson.build                 |  6 ++++++
 lib/librte_eal/linuxapp/eal/Makefile | 16 ----------------
 lib/meson.build                      |  6 ++++++
 mk/exec-env/linuxapp/rte.vars.mk     |  2 ++
 test/test/meson.build                |  5 +++++
 9 files changed, 33 insertions(+), 20 deletions(-)

app/meson.build +8 -1

index 99e0b93ec..c9a52a22b 100644
@@ -11,13 +11,20 @@ apps = ['pdump',
 # for BSD only
 lib_execinfo = cc.find_library('execinfo', required: false)
 
+default_cflags = machine_args
+
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 foreach app:apps
 	build = true
 	name = app
 	allow_experimental_apis = false
 	sources = []
 	includes = []
-	cflags = machine_args
+	cflags = default_cflags
 	objs = [] # other object files to link against, used e.g. for
 	          # instruction-set optimized versions of code

drivers/bus/pci/linux/Makefile +0 -2

index 96ea1d540..90404468b 100644
@@ -4,5 +4,3 @@
 SRCS += pci.c
 SRCS += pci_uio.c
 SRCS += pci_vfio.c
-
-CFLAGS += -D_GNU_SOURCE

drivers/meson.build +6 -0

index 47b4215a3..74fec716d 100644
@@ -16,6 +16,12 @@ default_cflags = machine_args
 if cc.has_argument('-Wno-format-truncation')
 	default_cflags += '-Wno-format-truncation'
 endif
+
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 foreach class:driver_classes
 	drivers = []
 	std_deps = []

drivers/net/softnic/conn.c +0 -1

index 990cf40fc..8b6658088 100644
@@ -8,7 +8,6 @@
 #include <unistd.h>
 #include <sys/types.h>
 
-#define __USE_GNU
 #include <sys/socket.h>
 
 #include <sys/epoll.h>

examples/meson.build +6 -0

index 4ee7a1114..70c22eb62 100644
@@ -22,6 +22,12 @@ default_cflags = machine_args
 if cc.has_argument('-Wno-format-truncation')
 	default_cflags += '-Wno-format-truncation'
 endif
+
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 foreach example: examples
 	name = example
 	build = true

lib/librte_eal/linuxapp/eal/Makefile +0 ~~-16~~

index fd92c75c2..bfee453bc 100644
@@ -85,22 +85,6 @@ SRCS-y += rte_cycles.c
 
 CFLAGS_eal_common_cpuflags.o := $(CPUFLAGS_LIST)
 
-CFLAGS_eal.o := -D_GNU_SOURCE
-CFLAGS_eal_interrupts.o := -D_GNU_SOURCE
-CFLAGS_eal_vfio_mp_sync.o := -D_GNU_SOURCE
-CFLAGS_eal_timer.o := -D_GNU_SOURCE
-CFLAGS_eal_lcore.o := -D_GNU_SOURCE
-CFLAGS_eal_memalloc.o := -D_GNU_SOURCE
-CFLAGS_eal_thread.o := -D_GNU_SOURCE
-CFLAGS_eal_log.o := -D_GNU_SOURCE
-CFLAGS_eal_common_log.o := -D_GNU_SOURCE
-CFLAGS_eal_hugepage_info.o := -D_GNU_SOURCE
-CFLAGS_eal_common_whitelist.o := -D_GNU_SOURCE
-CFLAGS_eal_common_options.o := -D_GNU_SOURCE
-CFLAGS_eal_common_thread.o := -D_GNU_SOURCE
-CFLAGS_eal_common_lcore.o := -D_GNU_SOURCE
-CFLAGS_rte_cycles.o := -D_GNU_SOURCE
-
 # workaround for a gcc bug with noreturn attribute
 # http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603
 ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y)

lib/meson.build +6 -0

index 3acc67e6e..2c7ea436a 100644
@@ -32,6 +32,12 @@ if cc.has_argument('-Wno-format-truncation')
 endif
 
 enabled_libs = [] # used to print summary at the end
+
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 foreach l:libraries
 	build = true
 	name = l

mk/exec-env/linuxapp/rte.vars.mk +2 -0

index 3129edc8c..91b778fcc 100644
@@ -17,6 +17,8 @@ else
 EXECENV_CFLAGS  = -pthread
 endif
 
+EXECENV_CFLAGS += -D_GNU_SOURCE
+
 EXECENV_LDLIBS  =
 EXECENV_ASFLAGS =

test/test/meson.build +5 -0

index b1dd6eca2..c81fca439 100644
@@ -242,6 +242,11 @@ if cc.has_argument('-Wno-format-truncation')
     cflags += '-Wno-format-truncation'
 endif
 
+# on Linux, specify -D_GNU_SOURCE unconditionally
+if host_machine.system() == 'linux'
+	default_cflags += '-D_GNU_SOURCE'
+endif
+
 test_dep_objs = []
 compress_test_dep = dependency('zlib', required: false)
 if compress_test_dep.found()

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-09-20 15:28 UTC #18 lore raw

PATCH v2 2/7 pci/vfio: improve musl compatibility

In reply to Anatoly Burakov

Musl already has PAGE_SIZE defined, and our define clashed with it. Rename our define to SYS_PAGE_SIZE.

Bugzilla ID: 36

Signed-off-by Anatoly Burakov <[email protected]>

---
 drivers/bus/pci/linux/pci_vfio.c | 2 ++
 1 file changed, 2 insertions(+)

drivers/bus/pci/linux/pci_vfio.c +2 -0

index 686386d6a..85f6c9803 100644
@@ -35,7 +35,9 @@
 
 #ifdef VFIO_PRESENT
 
+#ifndef PAGE_SIZE
 #define PAGE_SIZE   (sysconf(_SC_PAGESIZE))
+#endif
 #define PAGE_MASK   (~(PAGE_SIZE - 1))
 
 static struct rte_tailq_elem rte_vfio_tailq = {

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-09-20 15:27 UTC #19 lore raw

PATCH v2 3/7 fbarray: improve musl compatibility

In reply to Anatoly Burakov

When built against musl, fcntl.h doesn't silently get included. Fix by including it explicitly.

Bugzilla ID: 34

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/common/eal_common_fbarray.c | 1 +
 1 file changed, 1 insertion(+)

lib/librte_eal/common/eal_common_fbarray.c +1 -0

index ba6c4ae39..ea0735cb8 100644
@@ -2,6 +2,7 @@
  * Copyright(c) 2017-2018 Intel Corporation
  */
 
+#include <fcntl.h>
 #include <inttypes.h>
 #include <limits.h>
 #include <sys/mman.h>

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-09-20 15:28 UTC #20 lore raw

PATCH v2 4/7 eal/hugepage_info: improve musl compatibility

In reply to Anatoly Burakov

When built against musl, fcntl.h doesn't silently get included. Fix by including it explicitly.

Bugzilla ID: 33

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/linuxapp/eal/eal_hugepage_info.c | 1 +
 1 file changed, 1 insertion(+)

lib/librte_eal/linuxapp/eal/eal_hugepage_info.c +1 -0

index 3a7d4b222..0eab1cf71 100644
@@ -6,6 +6,7 @@
 #include <sys/types.h>
 #include <sys/file.h>
 #include <dirent.h>
+#include <fcntl.h>
 #include <stdint.h>
 #include <stdlib.h>
 #include <stdio.h>

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-09-20 15:27 UTC #21 lore raw

PATCH v2 5/7 mem: improve musl compatibility

In reply to Anatoly Burakov

When built against musl, fcntl.h doesn't silently get included. Fix by including it explicitly.

Bugzilla ID: 31

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/common/eal_common_memory.c | 1 +
 lib/librte_eal/linuxapp/eal/eal_memory.c  | 1 +
 2 files changed, 2 insertions(+)

lib/librte_eal/common/eal_common_memory.c +1 -0

index 0b69804ff..9b5eacc57 100644
@@ -2,6 +2,7 @@
  * Copyright(c) 2010-2014 Intel Corporation
  */
 
+#include <fcntl.h>
 #include <errno.h>
 #include <stdio.h>
 #include <stdint.h>

lib/librte_eal/linuxapp/eal/eal_memory.c +1 -0

index e3ac24815..256cab526 100644
@@ -5,6 +5,7 @@
 
 #define _FILE_OFFSET_BITS 64
 #include <errno.h>
+#include <fcntl.h>
 #include <stdarg.h>
 #include <stdbool.h>
 #include <stdlib.h>

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-09-20 15:27 UTC #22 lore raw

PATCH v2 6/7 string_fns: improve musl compatibility

In reply to Anatoly Burakov

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/common/include/rte_string_fns.h | 1 +
 1 file changed, 1 insertion(+)

lib/librte_eal/common/include/rte_string_fns.h +1 -0

index ecd141b85..295844ad2 100644
@@ -16,6 +16,7 @@ extern "C" {
 #endif
 
 #include <stdio.h>
+#include <string.h>
 
 /**
  * Takes string "string" parameter and splits it at character "delim"

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-09-20 15:27 UTC #23 lore raw

PATCH v2 7/7 eal: improve musl compatibility

In reply to Anatoly Burakov

Musl complains about pthread id being of wrong size, because on musl, pthread_t is a struct pointer, not an unsinged int. Fix the printing code by casting pthread id to unsigned pointer type and adjusting the format specifier to be of appropriate size.

Signed-off-by Anatoly Burakov <[email protected]>

---
 lib/librte_eal/linuxapp/eal/eal.c        | 4 ++--
 lib/librte_eal/linuxapp/eal/eal_thread.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

lib/librte_eal/linuxapp/eal/eal.c +2 -2

index e59ac6577..1d6a9ac44 100644
@@ -979,8 +979,8 @@ rte_eal_init(int argc, char **argv)
 
 	ret = eal_thread_dump_affinity(cpuset, sizeof(cpuset));
 
-	RTE_LOG(DEBUG, EAL, "Master lcore %u is ready (tid=%x;cpuset=[%s%s])\n",
-		rte_config.master_lcore, (int)thread_id, cpuset,
+	RTE_LOG(DEBUG, EAL, "Master lcore %u is ready (tid=%zx;cpuset=[%s%s])\n",
+		rte_config.master_lcore, (uintptr_t)thread_id, cpuset,
 		ret == 0 ? "" : "...");
 
 	RTE_LCORE_FOREACH_SLAVE(i) {

lib/librte_eal/linuxapp/eal/eal_thread.c +2 -2

index b496fc711..379773b68 100644
@@ -121,8 +121,8 @@ eal_thread_loop(__attribute__((unused)) void *arg)
 
 	ret = eal_thread_dump_affinity(cpuset, sizeof(cpuset));
 
-	RTE_LOG(DEBUG, EAL, "lcore %u is ready (tid=%x;cpuset=[%s%s])\n",
-		lcore_id, (int)thread_id, cpuset, ret == 0 ? "" : "...");
+	RTE_LOG(DEBUG, EAL, "lcore %u is ready (tid=%zx;cpuset=[%s%s])\n",
+		lcore_id, (uintptr_t)thread_id, cpuset, ret == 0 ? "" : "...");
 
 	/* read on our pipe to get commands */
 	while (1) {

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-10-04 10:22 UTC #24 lore raw

v3 cover Improve core EAL musl compatibility

In reply to Anatoly Burakov

Tested on container with Alpine Linux. Alpine dependencies:

build-base bsd-compat-headers libexecinfo-dev linux-headers numactl-dev

For numactl-dev, testing repository needs to be enabled:

echo "http://dl-cdn.alpinelinux.org/alpine/edge/testing" >> /etc/apk/repositories

If successful (using a very broad definition of "success"), the build should fail somewhere in PCI bus driver in UIO.

v3 -> v2:
- Made _GNU_SOURCE unconditional for all DPDK targets
- Fixed usage of __USE_GNU and _GNU_SOURCE accross DPDK

v2 -> v1:
- Fixed patch 2 to use existing define if available
- Fixed patch 7 to use proper format specifier and cast pthread ID to unsigned pointer type

Anatoly Burakov (7):
mk: build with _GNU_SOURCE defined by default
pci/vfio: improve musl compatibility
fbarray: improve musl compatibility
eal/hugepage_info: improve musl compatibility
mem: improve musl compatibility
string_fns: improve musl compatibility
eal: improve musl compatibility

 app/meson.build                               |  7 +++++-
 app/test-pmd/Makefile                         |  2 --
 buildtools/pmdinfogen/pmdinfogen.c            |  1 -
 drivers/bus/dpaa/Makefile                     |  1 -
 drivers/bus/dpaa/meson.build                  |  1 -
 drivers/bus/fslmc/meson.build                 |  1 -
 drivers/bus/pci/linux/Makefile                |  2 --
 drivers/bus/pci/linux/pci_vfio.c              |  2 ++
 drivers/bus/pci/meson.build                   |  1 -
 drivers/crypto/dpaa2_sec/Makefile             |  1 -
 drivers/crypto/dpaa_sec/Makefile              |  1 -
 drivers/mempool/dpaa/Makefile                 |  1 -
 drivers/meson.build                           |  4 ++++
 drivers/net/mlx5/mlx5_ethdev.c                |  2 --
 drivers/net/mlx5/mlx5_socket.c                |  2 --
 drivers/net/softnic/conn.c                    |  1 -
 examples/cmdline/Makefile                     |  1 -
 examples/ethtool/ethtool-app/Makefile         |  2 +-
 examples/ip_pipeline/conn.c                   |  1 -
 examples/l2fwd-cat/Makefile                   |  2 --
 examples/l2fwd-cat/meson.build                |  1 -
 examples/load_balancer/Makefile               |  1 -
 examples/meson.build                          |  4 ++++
 .../performance-thread/l3fwd-thread/main.c    |  2 --
 .../performance-thread/pthread_shim/main.c    |  1 -
 .../pthread_shim/pthread_shim.c               |  1 -
 examples/qos_sched/Makefile                   |  2 --
 examples/tep_termination/Makefile             |  1 -
 examples/vhost/Makefile                       |  1 -
 examples/vhost_crypto/Makefile                |  1 -
 examples/vhost_crypto/meson.build             |  2 +-
 examples/vhost_scsi/Makefile                  |  4 ++--
 examples/vhost_scsi/meson.build               |  2 +-
 lib/librte_cmdline/Makefile                   |  1 -
 lib/librte_cmdline/cmdline.c                  | 24 -------------------
 lib/librte_eal/bsdapp/eal/Makefile            |  5 ----
 lib/librte_eal/common/eal_common_fbarray.c    |  1 +
 lib/librte_eal/common/eal_common_memory.c     |  1 +
 .../common/include/rte_string_fns.h           |  1 +
 lib/librte_eal/linuxapp/eal/Makefile          | 16 -------------
 lib/librte_eal/linuxapp/eal/eal.c             |  4 ++--
 .../linuxapp/eal/eal_hugepage_info.c          |  1 +
 lib/librte_eal/linuxapp/eal/eal_memory.c      |  1 +
 lib/librte_eal/linuxapp/eal/eal_thread.c      |  4 ++--
 lib/librte_eal/meson.build                    |  1 -
 lib/librte_pdump/Makefile                     |  1 -
 lib/librte_sched/Makefile                     |  2 --
 lib/meson.build                               |  4 ++++
 mk/target/generic/rte.vars.mk                 |  3 +++
 test/test/Makefile                            |  2 --
 test/test/meson.build                         |  3 +++
 51 files changed, 40 insertions(+), 94 deletions(-)

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-10-04 10:21 UTC #25 lore raw

PATCH v3 1/7 mk: build with _GNU_SOURCE defined by default

Remove all usages of _GNU_SOURCE in source files and makefiles, and also fixup a couple of instances of using __USE_GNU instead of _GNU_SOURCE.

Signed-off-by Anatoly Burakov <[email protected]>

---
 app/meson.build                               |  7 +++++-
 app/test-pmd/Makefile                         |  2 --
 buildtools/pmdinfogen/pmdinfogen.c            |  1 -
 drivers/bus/dpaa/Makefile                     |  1 -
 drivers/bus/dpaa/meson.build                  |  1 -
 drivers/bus/fslmc/meson.build                 |  1 -
 drivers/bus/pci/linux/Makefile                |  2 --
 drivers/bus/pci/meson.build                   |  1 -
 drivers/crypto/dpaa2_sec/Makefile             |  1 -
 drivers/crypto/dpaa_sec/Makefile              |  1 -
 drivers/mempool/dpaa/Makefile                 |  1 -
 drivers/meson.build                           |  4 ++++
 drivers/net/mlx5/mlx5_ethdev.c                |  2 --
 drivers/net/mlx5/mlx5_socket.c                |  2 --
 drivers/net/softnic/conn.c                    |  1 -
 examples/cmdline/Makefile                     |  1 -
 examples/ethtool/ethtool-app/Makefile         |  2 +-
 examples/ip_pipeline/conn.c                   |  1 -
 examples/l2fwd-cat/Makefile                   |  2 --
 examples/l2fwd-cat/meson.build                |  1 -
 examples/load_balancer/Makefile               |  1 -
 examples/meson.build                          |  4 ++++
 .../performance-thread/l3fwd-thread/main.c    |  2 --
 .../performance-thread/pthread_shim/main.c    |  1 -
 .../pthread_shim/pthread_shim.c               |  1 -
 examples/qos_sched/Makefile                   |  2 --
 examples/tep_termination/Makefile             |  1 -
 examples/vhost/Makefile                       |  1 -
 examples/vhost_crypto/Makefile                |  1 -
 examples/vhost_crypto/meson.build             |  2 +-
 examples/vhost_scsi/Makefile                  |  4 ++--
 examples/vhost_scsi/meson.build               |  2 +-
 lib/librte_cmdline/Makefile                   |  1 -
 lib/librte_cmdline/cmdline.c                  | 24 -------------------
 lib/librte_eal/bsdapp/eal/Makefile            |  5 ----
 lib/librte_eal/linuxapp/eal/Makefile          | 16 -------------
 lib/librte_eal/meson.build                    |  1 -
 lib/librte_pdump/Makefile                     |  1 -
 lib/librte_sched/Makefile                     |  2 --
 lib/meson.build                               |  4 ++++
 mk/target/generic/rte.vars.mk                 |  3 +++
 test/test/Makefile                            |  2 --
 test/test/meson.build                         |  3 +++
 43 files changed, 29 insertions(+), 90 deletions(-)

app/meson.build +6 -1

index 99e0b93ec..e68d949e9 100644
@@ -11,13 +11,18 @@ apps = ['pdump',
 # for BSD only
 lib_execinfo = cc.find_library('execinfo', required: false)
 
+default_cflags = machine_args
+
+# specify -D_GNU_SOURCE unconditionally
+default_cflags += '-D_GNU_SOURCE'
+
 foreach app:apps
 	build = true
 	name = app
 	allow_experimental_apis = false
 	sources = []
 	includes = []
-	cflags = machine_args
+	cflags = default_cflags
 	objs = [] # other object files to link against, used e.g. for
 	          # instruction-set optimized versions of code

app/test-pmd/Makefile +0 -2

index 2b4d604b8..3694a5e31 100644
@@ -70,8 +70,6 @@ endif
 
 endif
 
-CFLAGS_cmdline.o := -D_GNU_SOURCE
-
 include $(RTE_SDK)/mk/rte.app.mk
 
 endif

buildtools/pmdinfogen/pmdinfogen.c +0 -1

index 0f35ca46b..dc0b6d5ff 100644
@@ -8,7 +8,6 @@
  *
  */
 
-#define _GNU_SOURCE
 #include <stdio.h>
 #include <ctype.h>
 #include <string.h>

drivers/bus/dpaa/Makefile +0 -1

index bffaa9d92..9c90e53f8 100644
@@ -14,7 +14,6 @@ CFLAGS := -I$(SRCDIR) $(CFLAGS)
 CFLAGS += -O3 $(WERROR_FLAGS)
 CFLAGS += -Wno-pointer-arith
 CFLAGS += -Wno-cast-qual
-CFLAGS += -D _GNU_SOURCE
 CFLAGS += -I$(RTE_BUS_DPAA)/
 CFLAGS += -I$(RTE_BUS_DPAA)/include
 CFLAGS += -I$(RTE_BUS_DPAA)/base/qbman

drivers/bus/dpaa/meson.build +0 -1

index d10b62c03..065a07dfd 100644
@@ -26,4 +26,3 @@ if cc.has_argument('-Wno-cast-qual')
 endif
 
 includes += include_directories('include', 'base/qbman')
-cflags += ['-D_GNU_SOURCE']

drivers/bus/fslmc/meson.build +0 -1

index 22a56a6fc..6b70882b2 100644
@@ -24,4 +24,3 @@ sources = files('fslmc_bus.c',
 allow_experimental_apis = true
 
 includes += include_directories('mc', 'qbman/include', 'portal')
-cflags += ['-D_GNU_SOURCE']

drivers/bus/pci/linux/Makefile +0 -2

index 96ea1d540..90404468b 100644
@@ -4,5 +4,3 @@
 SRCS += pci.c
 SRCS += pci_uio.c
 SRCS += pci_vfio.c
-
-CFLAGS += -D_GNU_SOURCE

drivers/bus/pci/meson.build +0 -1

index 23d6a5fec..1236d477e 100644
@@ -11,7 +11,6 @@ if host_machine.system() == 'linux'
 			'linux/pci_uio.c',
 			'linux/pci_vfio.c')
 	includes += include_directories('linux')
-	cflags += ['-D_GNU_SOURCE']
 else
 	sources += files('bsd/pci.c')
 	includes += include_directories('bsd')

drivers/crypto/dpaa2_sec/Makefile +0 -1

index da3d8f84f..6197452be 100644
@@ -20,7 +20,6 @@ LIB = librte_pmd_dpaa2_sec.a
 CFLAGS += -DALLOW_EXPERIMENTAL_API
 CFLAGS += -O3
 CFLAGS += $(WERROR_FLAGS)
-CFLAGS += -D _GNU_SOURCE
 
 ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y)
 ifeq ($(shell test $(GCC_VERSION) -gt 70 && echo 1), 1)

drivers/crypto/dpaa_sec/Makefile +0 -1

index 9be447041..afe8639e3 100644
@@ -11,7 +11,6 @@ LIB = librte_pmd_dpaa_sec.a
 
 # build flags
 CFLAGS += -DALLOW_EXPERIMENTAL_API
-CFLAGS += -D _GNU_SOURCE
 CFLAGS += -O3
 CFLAGS += $(WERROR_FLAGS)

drivers/mempool/dpaa/Makefile +0 -1

index da8da1e90..4f22fbc76 100644
@@ -10,7 +10,6 @@ LIB = librte_mempool_dpaa.a
 
 CFLAGS := -I$(SRCDIR) $(CFLAGS)
 CFLAGS += -O3 $(WERROR_FLAGS)
-CFLAGS += -D _GNU_SOURCE
 CFLAGS += -I$(RTE_SDK)/drivers/bus/dpaa
 CFLAGS += -I$(RTE_SDK)/drivers/bus/dpaa/include/
 CFLAGS += -I$(RTE_SDK)/drivers/mempool/dpaa

drivers/meson.build +4 -0

index 47b4215a3..567f4a19c 100644
@@ -16,6 +16,10 @@ default_cflags = machine_args
 if cc.has_argument('-Wno-format-truncation')
 	default_cflags += '-Wno-format-truncation'
 endif
+
+# specify -D_GNU_SOURCE unconditionally
+default_cflags += '-D_GNU_SOURCE'
+
 foreach class:driver_classes
 	drivers = []
 	std_deps = []

drivers/net/mlx5/mlx5_ethdev.c +0 -2

index 34c5b95ee..3ab448a21 100644
@@ -3,8 +3,6 @@
  * Copyright 2015 Mellanox Technologies, Ltd
  */
 
-#define _GNU_SOURCE
-
 #include <stddef.h>
 #include <assert.h>
 #include <inttypes.h>

drivers/net/mlx5/mlx5_socket.c +0 -2

index a3a522911..00106171d 100644
@@ -3,8 +3,6 @@
  * Copyright 2016 Mellanox Technologies, Ltd
  */
 
-#define _GNU_SOURCE
-
 #include <sys/types.h>
 #include <sys/socket.h>
 #include <sys/un.h>

drivers/net/softnic/conn.c +0 -1

index 990cf40fc..8b6658088 100644
@@ -8,7 +8,6 @@
 #include <unistd.h>
 #include <sys/types.h>
 
-#define __USE_GNU
 #include <sys/socket.h>
 
 #include <sys/epoll.h>

examples/cmdline/Makefile +0 -1

index 7893c85b3..a617cce11 100644
@@ -56,7 +56,6 @@ SRCS-y := main.c commands.c parse_obj_list.c
 
 CFLAGS += -O3
 CFLAGS += $(WERROR_FLAGS)
-CFLAGS_parse_obj_list.o := -D_GNU_SOURCE
 
 include $(RTE_SDK)/mk/rte.extapp.mk

examples/ethtool/ethtool-app/Makefile +1 -1

index 4cd9efdd5..9ecfc0b89 100644
@@ -16,7 +16,7 @@ APP = ethtool
 # all source are stored in SRCS-y
 SRCS-y := main.c ethapp.c
 
-CFLAGS += -O3 -D_GNU_SOURCE -pthread -I$(SRCDIR)/../lib
+CFLAGS += -O3 -pthread -I$(SRCDIR)/../lib
 CFLAGS += $(WERROR_FLAGS)
 
 LDLIBS += -L$(subst ethtool-app,lib,$(RTE_OUTPUT))/lib

examples/ip_pipeline/conn.c +0 -1

index 6b08e9e8e..30fca80c1 100644
@@ -8,7 +8,6 @@
 #include <unistd.h>
 #include <sys/types.h>
 
-#define __USE_GNU
 #include <sys/socket.h>
 
 #include <sys/epoll.h>

examples/l2fwd-cat/Makefile +0 -2

index aec770c28..b6eeabde1 100644
@@ -23,7 +23,6 @@ CFLAGS += -O3 $(shell pkg-config --cflags libdpdk)
 LDFLAGS_SHARED = $(shell pkg-config --libs libdpdk)
 LDFLAGS_STATIC = -Wl,-Bstatic $(shell pkg-config --static --libs libdpdk)
 
-CFLAGS += -D_GNU_SOURCE
 LDFLAGS += -lpqos
 
 build/$(APP)-shared: $(SRCS-y) Makefile $(PC_FILE) | build

@@ -66,7 +65,6 @@ endif
 EXTRA_CFLAGS += -O3 -g -Wfatal-errors
 
 CFLAGS += -I$(PQOS_INSTALL_PATH)/../include
-CFLAGS_cat.o := -D_GNU_SOURCE
 
 LDLIBS += -L$(PQOS_INSTALL_PATH)
 LDLIBS += -lpqos

examples/l2fwd-cat/meson.build +0 -1

index 1234e7b55..4e2777a03 100644
@@ -9,7 +9,6 @@
 pqos = cc.find_library('pqos', required: false)
 build = pqos.found()
 ext_deps += pqos
-cflags += '-D_GNU_SOURCE'
 cflags += '-I/usr/local/include' # assume pqos lib installed in /usr/local
 sources = files(
 	'cat.c', 'l2fwd-cat.c'

examples/load_balancer/Makefile +0 -1

index fc8df71e8..197b019d5 100644
@@ -50,7 +50,6 @@ include $(RTE_SDK)/mk/rte.vars.mk
 
 CFLAGS += -O3 -g
 CFLAGS += $(WERROR_FLAGS)
-CFLAGS_config.o := -D_GNU_SOURCE
 
 # workaround for a gcc bug with noreturn attribute
 # http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603

examples/meson.build +4 -0

index 4ee7a1114..af81c762e 100644
@@ -22,6 +22,10 @@ default_cflags = machine_args
 if cc.has_argument('-Wno-format-truncation')
 	default_cflags += '-Wno-format-truncation'
 endif
+
+# specify -D_GNU_SOURCE unconditionally
+default_cflags += '-D_GNU_SOURCE'
+
 foreach example: examples
 	name = example
 	build = true

examples/performance-thread/l3fwd-thread/main.c +0 -2

index 50fd1b00a..4f8747bc3 100644
@@ -2,8 +2,6 @@
  * Copyright(c) 2010-2016 Intel Corporation
  */
 
-#define _GNU_SOURCE
-
 #include <stdio.h>
 #include <stdlib.h>
 #include <stdint.h>

examples/performance-thread/pthread_shim/main.c +0 -1

index 7d0d58116..03ff39436 100644
@@ -2,7 +2,6 @@
  * Copyright(c) 2015 Intel Corporation
  */
 
-#define _GNU_SOURCE
 #include <stdio.h>
 #include <stdlib.h>
 #include <stdint.h>

examples/performance-thread/pthread_shim/pthread_shim.c +0 -1

index 53f12437f..a02de0677 100644
@@ -6,7 +6,6 @@
 #include <stdlib.h>
 #include <sys/types.h>
 #include <errno.h>
-#define __USE_GNU
 #include <sched.h>
 #include <dlfcn.h>

examples/qos_sched/Makefile +0 -2

index a7ecf9788..45b0a9eb6 100644
@@ -57,8 +57,6 @@ else
 
 CFLAGS += -O3
 CFLAGS += $(WERROR_FLAGS)
-CFLAGS_args.o := -D_GNU_SOURCE
-CFLAGS_cfg_file.o := -D_GNU_SOURCE
 
 include $(RTE_SDK)/mk/rte.extapp.mk

examples/tep_termination/Makefile +0 -1

index 8ec1a38ef..4c1564325 100644
@@ -60,7 +60,6 @@ endif
 CFLAGS += -DALLOW_EXPERIMENTAL_API
 CFLAGS += -O3
 CFLAGS += $(WERROR_FLAGS)
-CFLAGS += -D_GNU_SOURCE
 
 include $(RTE_SDK)/mk/rte.extapp.mk
 endif

examples/vhost/Makefile +0 -1

index a2ea97a0c..c6964381b 100644
@@ -61,7 +61,6 @@ else
 CFLAGS += -DALLOW_EXPERIMENTAL_API
 CFLAGS += -O2 -D_FILE_OFFSET_BITS=64
 CFLAGS += $(WERROR_FLAGS)
-CFLAGS += -D_GNU_SOURCE
 
 include $(RTE_SDK)/mk/rte.extapp.mk

examples/vhost_crypto/Makefile +0 -1

index 83d331012..a620abf49 100644
@@ -25,7 +25,6 @@ SRCS-y := main.c
 CFLAGS += -DALLOW_EXPERIMENTAL_API
 CFLAGS += -O2 -D_FILE_OFFSET_BITS=64
 CFLAGS += $(WERROR_FLAGS)
-CFLAGS += -D_GNU_SOURCE
 
 include $(RTE_SDK)/mk/rte.extapp.mk

examples/vhost_crypto/meson.build +1 -1

index 0f4876f06..daf19fb87 100644
@@ -8,7 +8,7 @@
 
 allow_experimental_apis = true
 deps += ['vhost', 'cryptodev']
-cflags += ['-D_GNU_SOURCE','-D_FILE_OFFSET_BITS=64']
+cflags += ['-D_FILE_OFFSET_BITS=64']
 sources = files(
 	'main.c'
 )

examples/vhost_scsi/Makefile +2 -2

index fa0cf727e..523aee0bf 100644
@@ -18,7 +18,7 @@ shared: build/$(APP)-shared
 static: build/$(APP)-static
 	ln -sf $(APP)-static build/$(APP)
 
-CFLAGS += -D_GNU_SOURCE -D_FILE_OFFSET_BITS=64
+CFLAGS += -D_FILE_OFFSET_BITS=64
 LDFLAGS += -pthread
 
 PC_FILE := $(shell pkg-config --path libdpdk)

@@ -57,7 +57,7 @@ please change the definition of the RTE_TARGET environment variable)
 all:
 else
 
-CFLAGS += -D_GNU_SOURCE -D_FILE_OFFSET_BITS=64
+CFLAGS += -D_FILE_OFFSET_BITS=64
 CFLAGS += -O2
 CFLAGS += $(WERROR_FLAGS)

examples/vhost_scsi/meson.build +1 -1

index 5f92370f5..2303bcaed 100644
@@ -10,7 +10,7 @@ if host_machine.system() != 'linux'
 	build = false
 endif
 deps += 'vhost'
-cflags += ['-D_GNU_SOURCE','-D_FILE_OFFSET_BITS=64']
+cflags += ['-D_FILE_OFFSET_BITS=64']
 sources = files(
 	'scsi.c', 'vhost_scsi.c'
 )

lib/librte_cmdline/Makefile +0 -1

index ddae1cfde..c64142b8d 100644
@@ -25,7 +25,6 @@ SRCS-$(CONFIG_RTE_LIBRTE_CMDLINE) += cmdline_vt100.c
 SRCS-$(CONFIG_RTE_LIBRTE_CMDLINE) += cmdline_socket.c
 SRCS-$(CONFIG_RTE_LIBRTE_CMDLINE) += cmdline_parse_portlist.c
 
-CFLAGS += -D_GNU_SOURCE
 LDLIBS += -lrte_eal
 
 # install includes

lib/librte_cmdline/cmdline.c +0 ~~-24~~

index 591b78b0f..d9042f048 100644
@@ -126,35 +126,11 @@ cmdline_printf(const struct cmdline *cl, const char *fmt, ...)
 	if (!cl || !fmt)
 		return;
 
-#ifdef _GNU_SOURCE
 	if (cl->s_out < 0)
 		return;
 	va_start(ap, fmt);
 	vdprintf(cl->s_out, fmt, ap);
 	va_end(ap);
-#else
-	int ret;
-	char *buf;
-
-	if (cl->s_out < 0)
-		return;
-
-	buf = malloc(BUFSIZ);
-	if (buf == NULL)
-		return;
-	va_start(ap, fmt);
-	ret = vsnprintf(buf, BUFSIZ, fmt, ap);
-	va_end(ap);
-	if (ret < 0) {
-		free(buf);
-		return;
-	}
-	if (ret >= BUFSIZ)
-		ret = BUFSIZ - 1;
-	ret = write(cl->s_out, buf, ret);
-	(void)ret;
-	free(buf);
-#endif
 }
 
 int

lib/librte_eal/bsdapp/eal/Makefile +0 -5

index d27da3d15..7c8e00a2e 100644
@@ -77,11 +77,6 @@ SRCS-y += rte_cycles.c
 
 CFLAGS_eal_common_cpuflags.o := $(CPUFLAGS_LIST)
 
-CFLAGS_eal.o := -D_GNU_SOURCE
-#CFLAGS_eal_thread.o := -D_GNU_SOURCE
-CFLAGS_eal_log.o := -D_GNU_SOURCE
-CFLAGS_eal_common_log.o := -D_GNU_SOURCE
-
 # workaround for a gcc bug with noreturn attribute
 # http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603
 ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y)

lib/librte_eal/linuxapp/eal/Makefile +0 ~~-16~~

index fd92c75c2..bfee453bc 100644
@@ -85,22 +85,6 @@ SRCS-y += rte_cycles.c
 
 CFLAGS_eal_common_cpuflags.o := $(CPUFLAGS_LIST)
 
-CFLAGS_eal.o := -D_GNU_SOURCE
-CFLAGS_eal_interrupts.o := -D_GNU_SOURCE
-CFLAGS_eal_vfio_mp_sync.o := -D_GNU_SOURCE
-CFLAGS_eal_timer.o := -D_GNU_SOURCE
-CFLAGS_eal_lcore.o := -D_GNU_SOURCE
-CFLAGS_eal_memalloc.o := -D_GNU_SOURCE
-CFLAGS_eal_thread.o := -D_GNU_SOURCE
-CFLAGS_eal_log.o := -D_GNU_SOURCE
-CFLAGS_eal_common_log.o := -D_GNU_SOURCE
-CFLAGS_eal_hugepage_info.o := -D_GNU_SOURCE
-CFLAGS_eal_common_whitelist.o := -D_GNU_SOURCE
-CFLAGS_eal_common_options.o := -D_GNU_SOURCE
-CFLAGS_eal_common_thread.o := -D_GNU_SOURCE
-CFLAGS_eal_common_lcore.o := -D_GNU_SOURCE
-CFLAGS_rte_cycles.o := -D_GNU_SOURCE
-
 # workaround for a gcc bug with noreturn attribute
 # http://gcc.gnu.org/bugzilla/show_bug.cgi?id=12603
 ifeq ($(CONFIG_RTE_TOOLCHAIN_GCC),y)

lib/librte_eal/meson.build +0 -1

index e1fde15d1..f68c6733c 100644
@@ -25,7 +25,6 @@ version = 8  # the version of the EAL API
 allow_experimental_apis = true
 deps += 'compat'
 deps += 'kvargs'
-cflags += '-D_GNU_SOURCE'
 sources = common_sources + env_sources
 objs = common_objs + env_objs
 headers = common_headers + env_headers

lib/librte_pdump/Makefile +0 -1

index ee14dba7a..b241151dc 100644
@@ -8,7 +8,6 @@ LIB = librte_pdump.a
 
 CFLAGS += -DALLOW_EXPERIMENTAL_API
 CFLAGS += $(WERROR_FLAGS) -I$(SRCDIR) -O3
-CFLAGS += -D_GNU_SOURCE
 LDLIBS += -lrte_eal -lrte_mempool -lrte_mbuf -lrte_ethdev
 
 EXPORT_MAP := rte_pdump_version.map

lib/librte_sched/Makefile +0 -2

index 55d9c6989..46c53ed71 100644
@@ -11,8 +11,6 @@ LIB = librte_sched.a
 CFLAGS += -O3
 CFLAGS += $(WERROR_FLAGS)
 
-CFLAGS_rte_red.o := -D_GNU_SOURCE
-
 LDLIBS += -lm
 LDLIBS += -lrt
 LDLIBS += -lrte_eal -lrte_mempool -lrte_mbuf -lrte_net

lib/meson.build +4 -0

index 3acc67e6e..24351cc40 100644
@@ -32,6 +32,10 @@ if cc.has_argument('-Wno-format-truncation')
 endif
 
 enabled_libs = [] # used to print summary at the end
+
+# -D_GNU_SOURCE unconditionally
+default_cflags += '-D_GNU_SOURCE'
+
 foreach l:libraries
 	build = true
 	name = l

mk/target/generic/rte.vars.mk +3 -0

index 98085cd31..dd149acc9 100644
@@ -108,6 +108,9 @@ CFLAGS += -include $(RTE_SDK_BIN)/include/rte_config.h
 LDFLAGS += -L$(RTE_SDK_BIN)/lib
 endif
 
+# always define _GNU_SOURCE
+CFLAGS += -D_GNU_SOURCE
+
 export CFLAGS
 export LDFLAGS

test/test/Makefile +0 -2

index dcea4410d..0b7b56ca2 100644
@@ -209,8 +209,6 @@ CFLAGS += -DALLOW_EXPERIMENTAL_API
 CFLAGS += -O3
 CFLAGS += $(WERROR_FLAGS)
 
-CFLAGS += -D_GNU_SOURCE
-
 LDLIBS += -lm
 ifeq ($(CONFIG_RTE_COMPRESSDEV_TEST),y)
 ifeq ($(CONFIG_RTE_LIBRTE_COMPRESSDEV),y)

test/test/meson.build +3 -0

index bacb5b144..02ce75959 100644
@@ -251,6 +251,9 @@ if cc.has_argument('-Wno-format-truncation')
     cflags += '-Wno-format-truncation'
 endif
 
+# specify -D_GNU_SOURCE unconditionally
+default_cflags += '-D_GNU_SOURCE'
+
 test_dep_objs = []
 compress_test_dep = dependency('zlib', required: false)
 if compress_test_dep.found()

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-10-04 10:20 UTC #26 lore raw

PATCH v3 2/7 pci/vfio: improve musl compatibility

In reply to Anatoly Burakov

Musl already has PAGE_SIZE defined, and our define clashed with it. Rename our define to SYS_PAGE_SIZE.

Bugzilla ID: 36

Signed-off-by Anatoly Burakov <[email protected]>

Acked-by Bruce Richardson <[email protected]>

---
 drivers/bus/pci/linux/pci_vfio.c | 2 ++
 1 file changed, 2 insertions(+)

drivers/bus/pci/linux/pci_vfio.c +2 -0

index 686386d6a..85f6c9803 100644
@@ -35,7 +35,9 @@
 
 #ifdef VFIO_PRESENT
 
+#ifndef PAGE_SIZE
 #define PAGE_SIZE   (sysconf(_SC_PAGESIZE))
+#endif
 #define PAGE_MASK   (~(PAGE_SIZE - 1))
 
 static struct rte_tailq_elem rte_vfio_tailq = {

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-10-04 10:20 UTC #27 lore raw

PATCH v3 3/7 fbarray: improve musl compatibility

In reply to Anatoly Burakov

When built against musl, fcntl.h doesn't silently get included. Fix by including it explicitly.

Bugzilla ID: 34

Signed-off-by Anatoly Burakov <[email protected]>

Acked-by Bruce Richardson <[email protected]>

---
 lib/librte_eal/common/eal_common_fbarray.c | 1 +
 1 file changed, 1 insertion(+)

lib/librte_eal/common/eal_common_fbarray.c +1 -0

index ba6c4ae39..ea0735cb8 100644
@@ -2,6 +2,7 @@
  * Copyright(c) 2017-2018 Intel Corporation
  */
 
+#include <fcntl.h>
 #include <inttypes.h>
 #include <limits.h>
 #include <sys/mman.h>

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-10-04 10:25 UTC #28 lore raw

PATCH v3 4/7 eal/hugepage_info: improve musl compatibility

In reply to Anatoly Burakov

When built against musl, fcntl.h doesn't silently get included. Fix by including it explicitly.

Bugzilla ID: 33

Signed-off-by Anatoly Burakov <[email protected]>

Acked-by Bruce Richardson <[email protected]>

---
 lib/librte_eal/linuxapp/eal/eal_hugepage_info.c | 1 +
 1 file changed, 1 insertion(+)

lib/librte_eal/linuxapp/eal/eal_hugepage_info.c +1 -0

index 3a7d4b222..0eab1cf71 100644
@@ -6,6 +6,7 @@
 #include <sys/types.h>
 #include <sys/file.h>
 #include <dirent.h>
+#include <fcntl.h>
 #include <stdint.h>
 #include <stdlib.h>
 #include <stdio.h>

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-10-04 10:20 UTC #29 lore raw

PATCH v3 5/7 mem: improve musl compatibility

In reply to Anatoly Burakov

When built against musl, fcntl.h doesn't silently get included. Fix by including it explicitly.

Bugzilla ID: 31

Signed-off-by Anatoly Burakov <[email protected]>

Acked-by Bruce Richardson <[email protected]>

---
 lib/librte_eal/common/eal_common_memory.c | 1 +
 lib/librte_eal/linuxapp/eal/eal_memory.c  | 1 +
 2 files changed, 2 insertions(+)

lib/librte_eal/common/eal_common_memory.c +1 -0

index 0b69804ff..9b5eacc57 100644
@@ -2,6 +2,7 @@
  * Copyright(c) 2010-2014 Intel Corporation
  */
 
+#include <fcntl.h>
 #include <errno.h>
 #include <stdio.h>
 #include <stdint.h>

lib/librte_eal/linuxapp/eal/eal_memory.c +1 -0

index e3ac24815..256cab526 100644
@@ -5,6 +5,7 @@
 
 #define _FILE_OFFSET_BITS 64
 #include <errno.h>
+#include <fcntl.h>
 #include <stdarg.h>
 #include <stdbool.h>
 #include <stdlib.h>

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-10-04 10:26 UTC #30 lore raw

PATCH v3 6/7 string_fns: improve musl compatibility

In reply to Anatoly Burakov

Signed-off-by Anatoly Burakov <[email protected]>

Acked-by Bruce Richardson <[email protected]>

---
 lib/librte_eal/common/include/rte_string_fns.h | 1 +
 1 file changed, 1 insertion(+)

lib/librte_eal/common/include/rte_string_fns.h +1 -0

index ecd141b85..295844ad2 100644
@@ -16,6 +16,7 @@ extern "C" {
 #endif
 
 #include <stdio.h>
+#include <string.h>
 
 /**
  * Takes string "string" parameter and splits it at character "delim"

Signature · 1 line

2.17.1

Anatoly Burakov

author

2018-10-04 10:20 UTC #31 lore raw

PATCH v3 7/7 eal: improve musl compatibility

In reply to Anatoly Burakov

Signed-off-by Anatoly Burakov <[email protected]>

Acked-by Bruce Richardson <[email protected]>

---
 lib/librte_eal/linuxapp/eal/eal.c        | 4 ++--
 lib/librte_eal/linuxapp/eal/eal_thread.c | 4 ++--
 2 files changed, 4 insertions(+), 4 deletions(-)

lib/librte_eal/linuxapp/eal/eal.c +2 -2

index e59ac6577..1d6a9ac44 100644
@@ -979,8 +979,8 @@ rte_eal_init(int argc, char **argv)
 
 	ret = eal_thread_dump_affinity(cpuset, sizeof(cpuset));
 
-	RTE_LOG(DEBUG, EAL, "Master lcore %u is ready (tid=%x;cpuset=[%s%s])\n",
-		rte_config.master_lcore, (int)thread_id, cpuset,
+	RTE_LOG(DEBUG, EAL, "Master lcore %u is ready (tid=%zx;cpuset=[%s%s])\n",
+		rte_config.master_lcore, (uintptr_t)thread_id, cpuset,
 		ret == 0 ? "" : "...");
 
 	RTE_LCORE_FOREACH_SLAVE(i) {

lib/librte_eal/linuxapp/eal/eal_thread.c +2 -2

index b496fc711..379773b68 100644
@@ -121,8 +121,8 @@ eal_thread_loop(__attribute__((unused)) void *arg)
 
 	ret = eal_thread_dump_affinity(cpuset, sizeof(cpuset));
 
-	RTE_LOG(DEBUG, EAL, "lcore %u is ready (tid=%x;cpuset=[%s%s])\n",
-		lcore_id, (int)thread_id, cpuset, ret == 0 ? "" : "...");
+	RTE_LOG(DEBUG, EAL, "lcore %u is ready (tid=%zx;cpuset=[%s%s])\n",
+		lcore_id, (uintptr_t)thread_id, cpuset, ret == 0 ? "" : "...");
 
 	/* read on our pipe to get commands */
 	while (1) {

Signature · 1 line

2.17.1

Thomas Monjalon

Applied

2018-10-22 09:33 UTC #32 lore raw

In reply to Anatoly Burakov

04/10/2018 12:20, Anatoly Burakov:

Quoted 25 lines

This patchset fixes numerous issues with musl compatibility in the core EAL libraries. It does not fix anything beyond core EAL (so, PCI driver is still broken, so are a few other drivers), but it's a good start.
Tested on container with Alpine Linux. Alpine dependencies:
build-base bsd-compat-headers libexecinfo-dev linux-headers numactl-dev
For numactl-dev, testing repository needs to be enabled:
echo "http://dl-cdn.alpinelinux.org/alpine/edge/testing" >> /etc/apk/repositories
If successful (using a very broad definition of "success"), the build should fail somewhere in PCI bus driver in UIO.
v3 -> v2:
- Made _GNU_SOURCE unconditional for all DPDK targets
- Fixed usage of __USE_GNU and _GNU_SOURCE accross DPDK
v2 -> v1:
- Fixed patch 2 to use existing define if available
- Fixed patch 7 to use proper format specifier and cast pthread ID to unsigned pointer type

Quoted 8 lines

Anatoly Burakov (7):
mk: build with _GNU_SOURCE defined by default
pci/vfio: improve musl compatibility
fbarray: improve musl compatibility
eal/hugepage_info: improve musl compatibility
mem: improve musl compatibility
string_fns: improve musl compatibility
eal: improve musl compatibility

Applied, thanks

Hope the _GNU_SOURCE source won't bring too much compilation surprises.

Bruce Richardson

Acked-by on the v1 series

2018-09-28 15:27 UTC #33 lore raw

In reply to Anatoly Burakov

Quoted 7 lines · Anatoly Burakov

This patchset fixes numerous issues with musl compatibility in the core EAL libraries. It does not fix anything beyond core EAL (so, PCI driver is still broken, so are a few other drivers), but it's a good start.
Tested on container with Alpine Linux. Alpine dependencies:

Quoted 9 lines

build-base bsd-compat-headers libexecinfo-dev linux-headers numactl-dev
For numactl-dev, testing repository needs to be enabled:
echo "http://dl-cdn.alpinelinux.org/alpine/edge/testing" >> /etc/apk/repositories
If successful (using a very broad definition of "success"), the build should fail somewhere in PCI bus driver in UIO.

Disabling the kernel drivers, I get a build of EAL and the other libraries to compile as static libraries. However, shared library builds - and therefore meson builds fail due to missing backtrace function when linking. There is still work to do here, but this does fix a number of build errors on alpine.

Acked-by Bruce Richardson <[email protected]>

Thomas Monjalon

2018-10-03 22:56 UTC #34 lore raw

28/09/2018 17:25, Bruce Richardson:

Quoted 18 lines · Anatoly Burakov

Quoted 7 lines
This patchset fixes numerous issues with musl compatibility in the core EAL libraries. It does not fix anything beyond core EAL (so, PCI driver is still broken, so are a few other drivers), but it's a good start.
Tested on container with Alpine Linux. Alpine dependencies:
Quoted 9 lines
build-base bsd-compat-headers libexecinfo-dev linux-headers numactl-dev
For numactl-dev, testing repository needs to be enabled:
echo "http://dl-cdn.alpinelinux.org/alpine/edge/testing" >> /etc/apk/repositories
If successful (using a very broad definition of "success"), the build should fail somewhere in PCI bus driver in UIO.

Quoted 7 lines

Disabling the kernel drivers, I get a build of EAL and the other libraries to compile as static libraries. However, shared library builds - and therefore meson builds fail due to missing backtrace function when linking. There is still work to do here, but this does fix a number of build errors on alpine.
Acked-by Bruce Richardson <[email protected]>

I see this error with meson GCC-8.2 static:
examples/ip_pipeline/conn.c:11: error: "__USE_GNU" redefined

Burakov, Anatoly

author

2018-10-04 09:15 UTC #35 lore raw

Quoted 25 lines · Thomas Monjalon

28/09/2018 17:25, Bruce Richardson:
Quoted 24 lines · Anatoly Burakov
Quoted 7 lines
This patchset fixes numerous issues with musl compatibility in the core EAL libraries. It does not fix anything beyond core EAL (so, PCI driver is still broken, so are a few other drivers), but it's a good start.
Tested on container with Alpine Linux. Alpine dependencies:
Quoted 9 lines
build-base bsd-compat-headers libexecinfo-dev linux-headers numactl-dev
For numactl-dev, testing repository needs to be enabled:
echo "http://dl-cdn.alpinelinux.org/alpine/edge/testing" >> /etc/apk/repositories
If successful (using a very broad definition of "success"), the build should fail somewhere in PCI bus driver in UIO.
Disabling the kernel drivers, I get a build of EAL and the other libraries to compile as static libraries. However, shared library builds - and therefore meson builds fail due to missing backtrace function when linking. There is still work to do here, but this does fix a number of build errors on alpine.

Quoted 5 lines

Quoted 1 line
Acked-by Bruce Richardson <[email protected]>
I see this error with meson GCC-8.2 static:
examples/ip_pipeline/conn.c:11: error: "__USE_GNU" redefined

Why are we defining these? __USE_GNU is defined by <features.h> and isn't supposed to be defined/undefined by the user in the first place.

Signature · 2 lines

Thanks,
Anatoly