Skip to contentSkip to conversation

multilib_header_wrapper.h: Use #pragma once

PATCH In review Khem Raj 4 participants 11 messages Feb 5, 2019 – Feb 7, 2019

lore
Share & tools
lore

Khem Raj

author
#0 lore raw

PATCH v3 1/2 multilib_header_wrapper.h: Use #pragma once

Avoid infinite include loops, especially with bits/wordsize.h which is now possible with the synthesized headers since we now also synthesize bits/wordsize.h itelf for some arches e.g. arm/aarch64

In cases where extra preprocessing tools are used such as clang-tidy e.g. and these tools are not passed the knowledge about architecture then case comes where we enter into header include loop for bits/wordsize.h, since this template does explicitly include bits/wordsize.h

To fix this emits the pragma once at beginning of file, this is better solution than include guards, and pragma once is practically supported on all compilers except few e.g. cray c/c++ compiler

---
V2:
- No Changes, just rebased
V3:
- Do it using pragma once instead

 scripts/multilib_header_wrapper.h | 1 +
 1 file changed, 1 insertion(+)
scripts/multilib_header_wrapper.h +1 -0
index 9660225fdd..4824790783 100644
@@ -21,6 +21,7 @@
  * 
  */
 
+#pragma once
 
 #if defined (__bpf__)
 #define __MHWORDSIZE			64
Signature · 1 line
2.20.1
author
#1 lore raw

PATCH 2/2 mpfr: Upgrade to 4.0.2 release

Drop backported patch which is now applied upstream in 4.0.2

---
 ...001-Fix-obsolete-ARC-asm-constraints.patch | 45 ---------
 meta/recipes-support/mpfr/mpfr_4.0.1.bb       | 19 ----
 meta/recipes-support/mpfr/mpfr_4.0.2.bb       | 94 +++++++++++++++++++
 3 files changed, 94 insertions(+), 64 deletions(-)
 delete mode 100644 meta/recipes-support/mpfr/mpfr/0001-Fix-obsolete-ARC-asm-constraints.patch
 delete mode 100644 meta/recipes-support/mpfr/mpfr_4.0.1.bb
 create mode 100644 meta/recipes-support/mpfr/mpfr_4.0.2.bb
meta/recipes-support/mpfr/mpfr/0001-Fix-obsolete-ARC-asm-constraints.patch deleted +0 -45
deleted file mode 100644
index 59d1b0db67..0000000000
@@ -1,45 +0,0 @@
-mpfr-longlong.h: Fix obsolete ARC asm constraints
-
-This patch replaces obsolete ARC "J" asm constraint with
-up-to-date "Cal" constraint.
-
-"J" constraint only existed in pre-upstream GCC port for ARC.
-In current upstream port "Cal" constraint is used which leads
-to compile-time error.
-
-Proposed fix is known to work in Buildroot, Crosstool-NG etc.
-
-[1] https://gcc.gnu.org/git/?p=gcc.git;a=commit;h=09cb6a17e71bd40d2fbfaf82a1502fc210e33c87
-
-Signed-off-by: Claudiu Zissulescu <[email protected]>
-Signed-off-by: Vlad Zakharov <[email protected]>
-Signed-off-by: Alexey Brodkin <[email protected]>
-
-Upstream-Status: Backport [https://gforge.inria.fr/scm/viewvc.php/mpfr?view=revision&revision=13251]
----
-Index: src/mpfr-longlong.h
-===================================================================
---- a/src/mpfr-longlong.h	(revision 10963)
-+++ b/src/mpfr-longlong.h	(working copy)
-@@ -416,17 +416,17 @@
- 	   : "=r" (sh),							\
- 	     "=&r" (sl)							\
- 	   : "r"  ((USItype) (ah)),					\
--	     "rIJ" ((USItype) (bh)),					\
-+	     "rICal" ((USItype) (bh)),					\
- 	     "%r" ((USItype) (al)),					\
--	     "rIJ" ((USItype) (bl)))
-+	     "rICal" ((USItype) (bl)))
- #define sub_ddmmss(sh, sl, ah, al, bh, bl) \
-   __asm__ ("sub.f\t%1, %4, %5\n\tsbc\t%0, %2, %3"			\
- 	   : "=r" (sh),							\
- 	     "=&r" (sl)							\
- 	   : "r" ((USItype) (ah)),					\
--	     "rIJ" ((USItype) (bh)),					\
-+	     "rICal" ((USItype) (bh)),					\
- 	     "r" ((USItype) (al)),					\
--	     "rIJ" ((USItype) (bl)))
-+	     "rICal" ((USItype) (bl)))
- #endif
- 
- #if defined (__arm__) && (defined (__thumb2__) || !defined (__thumb__)) \
meta/recipes-support/mpfr/mpfr_4.0.1.bb deleted +0 -19
deleted file mode 100644
index 2eee76a357..0000000000
@@ -1,19 +0,0 @@
-SUMMARY = "C library for multiple-precision floating-point computations with exact rounding"
-HOMEPAGE = "http://www.mpfr.org/"
-LICENSE = "LGPLv3+"
-SECTION = "devel"
-
-inherit autotools texinfo
-
-LIC_FILES_CHKSUM = "file://COPYING;md5=d32239bcb673463ab874e80d47fae504 \
-		    file://COPYING.LESSER;md5=6a6a8e020838b23406c81b19c1d46df6"
-DEPENDS = "gmp autoconf-archive"
-
-SRC_URI = "http://www.mpfr.org/mpfr-${PV}/mpfr-${PV}.tar.xz \
-           file://0001-Fix-obsolete-ARC-asm-constraints.patch"
-SRC_URI[md5sum] = "b8dd19bd9bb1ec8831a6a582a7308073"
-SRC_URI[sha256sum] = "67874a60826303ee2fb6affc6dc0ddd3e749e9bfcb4c8655e3953d0458a6e16e"
-
-UPSTREAM_CHECK_URI = "http://www.mpfr.org/mpfr-current/"
-
-BBCLASSEXTEND = "native nativesdk"
meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +94 -0

94 lines not on this page · raw message

Signature · 1 line
2.20.1
#2 lore raw

Re: [PATCH 2/2] mpfr: Upgrade to 4.0.2 release

Quoted 98 lines · Khem Raj
meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +1 -0
@@ -0,0 +1,94 @@
+# FIXME: the LIC_FILES_CHKSUM values have been updated by 'devtool

upgrade'.
+# The following is the difference between the old and the new license text.
+# Please update the LICENSE value if needed, and summarize the changes in

meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +12 -0
+# the commit message via 'License-Update:' tag.
+# (example: 'License-Update: copyright years updated.')
+#
+# The changes:
+#
+# --- COPYING
+# +++ COPYING
+# @@ -1,7 +1,7 @@
+#                      GNU GENERAL PUBLIC LICENSE
+#                         Version 3, 29 June 2007
+#  
+# - Copyright (C) 2007 Free Software Foundation, Inc. <

http://fsf.org/>
+# + Copyright (C) 2007 Free Software Foundation, Inc. <
https://fsf.org/>

meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +7 -0
+#   Everyone is permitted to copy and distribute verbatim copies
+#   of this license document, but changing it is not allowed.
+#  
+# @@ -645,7 +645,7 @@
+#      GNU General Public License for more details.
+#  
+#      You should have received a copy of the GNU General Public

License
+# - along with this program. If not, see <
http://www.gnu.org/licenses/>;.
+# + along with this program. If not, see <
https://www.gnu.org/licenses/>;.

meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +2 -0
+#  
+#  Also add information on how to contact you by electronic and

paper mail.

meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +3 -0
+#  
+# @@ -664,11 +664,11 @@
+#    You should also get your employer (if you work as a programmer)

or school,
+# if any, to sign a "copyright disclaimer" for the program, if necessary.
+# For more information on this, and how to apply and follow the GNU GPL, see

meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +4 -0
+# -<http://www.gnu.org/licenses/>;.
+# +<https://www.gnu.org/licenses/>;.
+#  
+#    The GNU General Public License does not permit incorporating

your program
+# into proprietary programs. If your program is a subroutine library, you
+# may consider it more useful to permit linking proprietary applications with
+# the library. If this is what you want to do, use the GNU Lesser General

meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +11 -0
+#  Public License instead of this License.  But first, please read
+# -<http://www.gnu.org/philosophy/why-not-lgpl.html>;.
+# +<https://www.gnu.org/licenses/why-not-lgpl.html>;.
+# --- COPYING.LESSER
+# +++ COPYING.LESSER
+# @@ -1,7 +1,7 @@
+# -		   GNU LESSER GENERAL PUBLIC LICENSE
+# +                   GNU LESSER GENERAL PUBLIC LICENSE
+#                         Version 3, 29 June 2007
+#  
+# - Copyright (C) 2007 Free Software Foundation, Inc. <

http://fsf.org/>
+# + Copyright (C) 2007 Free Software Foundation, Inc. <
https://fsf.org/>

meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +11 -0
+#   Everyone is permitted to copy and distribute verbatim copies
+#   of this license document, but changing it is not allowed.
+#  
+# @@ -10,7 +10,7 @@
+#  the terms and conditions of version 3 of the GNU General Public
+#  License, supplemented by the additional permissions listed below.
+#  
+# -  0. Additional Definitions. 
+# +  0. Additional Definitions.
+#  
+#    As used herein, "this License" refers to version 3 of the GNU

Lesser
+# General Public License, and the "GNU GPL" refers to version 3 of the GNU

meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +2 -0
+# @@ -111,7 +111,7 @@
+#         a copy of the Library already present on the user's

computer
+# system, and (b) will operate properly with a modified
version
+# of the Library that is interface-compatible with the
Linked

meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +4 -0
+# -       Version. 
+# +       Version.
+#  
+#     e) Provide Installation Information, but only if you would

otherwise

meta/recipes-support/mpfr/mpfr_4.0.2.bb line 74

Quoted 3 lines
meta/recipes-support/mpfr/mpfr_4.0.2.bb new file +3 -0
+#     be required to provide such information under section 6 of the
+# 
+#

Could we fix the FIXME please? :)

Cheers,

Richard

#3 lore raw
In reply to Khem Raj
Quoted 11 lines · Khem Raj

Avoid infinite include loops, especially with bits/wordsize.h which is now possible with the synthesized headers since we now also synthesize bits/wordsize.h itelf for some arches e.g. arm/aarch64

In cases where extra preprocessing tools are used such as clang-tidy e.g. and these tools are not passed the knowledge about architecture then case comes where we enter into header include loop for bits/wordsize.h, since this template does explicitly include bits/wordsize.h

Quoted 5 lines

To fix this emits the pragma once at beginning of file, this is better solution than include guards, and pragma once is practically supported on all compilers except few e.g. cray c/c++ compiler

Something about this is breaking a local recipe (for tlsdate fwiw, but I don't think that's relevant):

| checking openssl/ssl.h usability... no
| checking openssl/ssl.h presence... yes
| configure: WARNING: openssl/ssl.h: present but cannot be compiled
| configure: WARNING: openssl/ssl.h:     check for missing prerequisite headers?
| configure: WARNING: openssl/ssl.h: see the Autoconf documentation
| configure: WARNING: openssl/ssl.h:     section "Present But Cannot
Be Compiled"
| configure: WARNING: openssl/ssl.h: proceeding with the compiler's result
| configure: WARNING:     ## ------------------------------------- ##
| configure: WARNING:     ## Report this to jacob at appelbaum.net ##
| configure: WARNING:     ## ------------------------------------- ##
| checking for openssl/ssl.h... no
| configure: error: OpenSSL is not installed, openssl/sslh is missing

Somehow it's not getting the definition for THIRTY_TWO_BIT from openssl/opensslconf.h (from openssl10) when testing for openssl/ssl.h so BN_ULONG ends up undefined... still digging as to what's going on.

Signature · 1 line
Alex Kiernan
author
#4 lore raw
Quoted 36 lines · Alex Kiernan
Quoted 11 lines · Khem Raj

Avoid infinite include loops, especially with bits/wordsize.h which is now possible with the synthesized headers since we now also synthesize bits/wordsize.h itelf for some arches e.g. arm/aarch64

In cases where extra preprocessing tools are used such as clang-tidy e.g. and these tools are not passed the knowledge about architecture then case comes where we enter into header include loop for bits/wordsize.h, since this template does explicitly include bits/wordsize.h

Quoted 5 lines

To fix this emits the pragma once at beginning of file, this is better solution than include guards, and pragma once is practically supported on all compilers except few e.g. cray c/c++ compiler

Something about this is breaking a local recipe (for tlsdate fwiw, but I don't think that's relevant):

| checking openssl/ssl.h usability... no
| checking openssl/ssl.h presence... yes
| configure: WARNING: openssl/ssl.h: present but cannot be compiled
| configure: WARNING: openssl/ssl.h:     check for missing prerequisite headers?
| configure: WARNING: openssl/ssl.h: see the Autoconf documentation
| configure: WARNING: openssl/ssl.h:     section "Present But Cannot
Be Compiled"
| configure: WARNING: openssl/ssl.h: proceeding with the compiler's result
| configure: WARNING:     ## ------------------------------------- ##
| configure: WARNING:     ## Report this to jacob at appelbaum.net ##
| configure: WARNING:     ## ------------------------------------- ##
| checking for openssl/ssl.h... no
| configure: error: OpenSSL is not installed, openssl/sslh is missing
Quoted 4 lines

Somehow it's not getting the definition for THIRTY_TWO_BIT from openssl/opensslconf.h (from openssl10) when testing for openssl/ssl.h so BN_ULONG ends up undefined... still digging as to what's going on.

thanks, is it using openssl10 ?

Quoted 2 lines

--
Alex Kiernan

#5 lore raw
Quoted 38 lines · Khem Raj
Quoted 36 lines · Alex Kiernan
Quoted 11 lines · Khem Raj

Avoid infinite include loops, especially with bits/wordsize.h which is now possible with the synthesized headers since we now also synthesize bits/wordsize.h itelf for some arches e.g. arm/aarch64

In cases where extra preprocessing tools are used such as clang-tidy e.g. and these tools are not passed the knowledge about architecture then case comes where we enter into header include loop for bits/wordsize.h, since this template does explicitly include bits/wordsize.h

Quoted 5 lines

To fix this emits the pragma once at beginning of file, this is better solution than include guards, and pragma once is practically supported on all compilers except few e.g. cray c/c++ compiler

Something about this is breaking a local recipe (for tlsdate fwiw, but I don't think that's relevant):

| checking openssl/ssl.h usability... no
| checking openssl/ssl.h presence... yes
| configure: WARNING: openssl/ssl.h: present but cannot be compiled
| configure: WARNING: openssl/ssl.h:     check for missing prerequisite headers?
| configure: WARNING: openssl/ssl.h: see the Autoconf documentation
| configure: WARNING: openssl/ssl.h:     section "Present But Cannot
Be Compiled"
| configure: WARNING: openssl/ssl.h: proceeding with the compiler's result
| configure: WARNING:     ## ------------------------------------- ##
| configure: WARNING:     ## Report this to jacob at appelbaum.net ##
| configure: WARNING:     ## ------------------------------------- ##
| checking for openssl/ssl.h... no
| configure: error: OpenSSL is not installed, openssl/sslh is missing
Quoted 7 lines
Quoted 4 lines

Somehow it's not getting the definition for THIRTY_TWO_BIT from openssl/opensslconf.h (from openssl10) when testing for openssl/ssl.h so BN_ULONG ends up undefined... still digging as to what's going on.

thanks, is it using openssl10 ?

It is... I'm sure that's relevant, but I'm failing to spot it.

Signature · 1 line
Alex Kiernan
author
#6 lore raw
Quoted 45 lines · Alex Kiernan
Quoted 43 lines · Khem Raj
Quoted 40 lines · Alex Kiernan
Quoted 11 lines · Khem Raj

Avoid infinite include loops, especially with bits/wordsize.h which is now possible with the synthesized headers since we now also synthesize bits/wordsize.h itelf for some arches e.g. arm/aarch64

In cases where extra preprocessing tools are used such as clang-tidy e.g. and these tools are not passed the knowledge about architecture then case comes where we enter into header include loop for bits/wordsize.h, since this template does explicitly include bits/wordsize.h

Quoted 5 lines

To fix this emits the pragma once at beginning of file, this is better solution than include guards, and pragma once is practically supported on all compilers except few e.g. cray c/c++ compiler

Something about this is breaking a local recipe (for tlsdate fwiw, but I don't think that's relevant):

| checking openssl/ssl.h usability... no
| checking openssl/ssl.h presence... yes
| configure: WARNING: openssl/ssl.h: present but cannot be compiled
| configure: WARNING: openssl/ssl.h:     check for missing prerequisite headers?
| configure: WARNING: openssl/ssl.h: see the Autoconf documentation
| configure: WARNING: openssl/ssl.h:     section "Present But Cannot
Be Compiled"
| configure: WARNING: openssl/ssl.h: proceeding with the compiler's result
| configure: WARNING:     ## ------------------------------------- ##
| configure: WARNING:     ## Report this to jacob at appelbaum.net ##
| configure: WARNING:     ## ------------------------------------- ##
| checking for openssl/ssl.h... no
| configure: error: OpenSSL is not installed, openssl/sslh is missing

Somehow it's not getting the definition for THIRTY_TWO_BIT from openssl/opensslconf.h (from openssl10) when testing for openssl/ssl.h so BN_ULONG ends up undefined... still digging as to what's going on.

Quoted 5 lines
Quoted 2 lines

thanks, is it using openssl10 ?

It is... I'm sure that's relevant, but I'm failing to spot it.

so where all do you have opensslconf.h in target sysroot ?

Quoted 2 lines

--
Alex Kiernan

#7 lore raw
Quoted 47 lines · Khem Raj
Quoted 45 lines · Alex Kiernan
Quoted 43 lines · Khem Raj
Quoted 40 lines · Alex Kiernan
Quoted 11 lines · Khem Raj

Avoid infinite include loops, especially with bits/wordsize.h which is now possible with the synthesized headers since we now also synthesize bits/wordsize.h itelf for some arches e.g. arm/aarch64

In cases where extra preprocessing tools are used such as clang-tidy e.g. and these tools are not passed the knowledge about architecture then case comes where we enter into header include loop for bits/wordsize.h, since this template does explicitly include bits/wordsize.h

Quoted 5 lines

To fix this emits the pragma once at beginning of file, this is better solution than include guards, and pragma once is practically supported on all compilers except few e.g. cray c/c++ compiler

Something about this is breaking a local recipe (for tlsdate fwiw, but I don't think that's relevant):

| checking openssl/ssl.h usability... no
| checking openssl/ssl.h presence... yes
| configure: WARNING: openssl/ssl.h: present but cannot be compiled
| configure: WARNING: openssl/ssl.h:     check for missing prerequisite headers?
| configure: WARNING: openssl/ssl.h: see the Autoconf documentation
| configure: WARNING: openssl/ssl.h:     section "Present But Cannot
Be Compiled"
| configure: WARNING: openssl/ssl.h: proceeding with the compiler's result
| configure: WARNING:     ## ------------------------------------- ##
| configure: WARNING:     ## Report this to jacob at appelbaum.net ##
| configure: WARNING:     ## ------------------------------------- ##
| checking for openssl/ssl.h... no
| configure: error: OpenSSL is not installed, openssl/sslh is missing

Somehow it's not getting the definition for THIRTY_TWO_BIT from openssl/opensslconf.h (from openssl10) when testing for openssl/ssl.h so BN_ULONG ends up undefined... still digging as to what's going on.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

thanks, is it using openssl10 ?

It is... I'm sure that's relevant, but I'm failing to spot it.

so where all do you have opensslconf.h in target sysroot ?

$ find tmp -name opensslconf.h -ls
  1589118      8 -rw-r--r--   3 vagrant  vagrant      4534 Feb  6
11:40 tmp/work/x86_64-linux/qemu-helper-native/1.0-r1/recipe-sysroot-native/usr/include/openssl/opensslconf.h
  1451388      4 -rw-r--r--   4 vagrant  vagrant      1971 Feb  6
21:31 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/tlsdate/0.0.13+gitAUTOINC+751f82dc6f-r0/recipe-sysroot/usr/include/openssl/opensslconf.h
  1589118      8 -rw-r--r--   3 vagrant  vagrant      4534 Feb  6
11:40 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/recipe-sysroot-native/usr/include/openssl/opensslconf.h
  1451388      4 -rw-r--r--   4 vagrant  vagrant      1971 Feb  6
21:31 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/sysroot-destdir/usr/include/openssl/opensslconf.h
  1574753      8 -rw-r--r--   1 vagrant  vagrant      7423 Feb  6
21:30 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/openssl-1.0.2q/crypto/opensslconf.h
  1717763      4 -rw-r--r--   1 vagrant  vagrant      3201 Nov 20
13:44 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/openssl-1.0.2q/MacOS/opensslconf.h
  1574836      0 lrwxrwxrwx   1 vagrant  vagrant        26 Feb  6
21:30 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/openssl-1.0.2q/include/openssl/opensslconf.h
-> ../../crypto/opensslconf.h
  1451388      4 -rw-r--r--   4 vagrant  vagrant      1971 Feb  6
21:31 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/image/usr/include/openssl/opensslconf.h
  1589118      8 -rw-r--r--   3 vagrant  vagrant      4534 Feb  6
11:40 tmp/sysroots-components/x86_64/openssl-native/usr/include/openssl/opensslconf.h
  1849804      4 -rw-r--r--   1 vagrant  vagrant      1971 Feb  6
13:28 tmp/sysroots-components/armv7at2hf-neon/openssl/usr/include/openssl/opensslconf.h
  1451388      4 -rw-r--r--   4 vagrant  vagrant      1971 Feb  6
21:31 tmp/sysroots-components/armv7at2hf-neon/openssl10/usr/include/openssl/opensslconf.h
Signature · 1 line
Alex Kiernan
#8 lore raw
Quoted 44 lines · Alex Kiernan
Quoted 42 lines · Khem Raj
Quoted 40 lines · Alex Kiernan
Quoted 38 lines · Khem Raj
Quoted 36 lines · Alex Kiernan
Quoted 11 lines · Khem Raj

Avoid infinite include loops, especially with bits/wordsize.h which is now possible with the synthesized headers since we now also synthesize bits/wordsize.h itelf for some arches e.g. arm/aarch64

In cases where extra preprocessing tools are used such as clang-tidy e.g. and these tools are not passed the knowledge about architecture then case comes where we enter into header include loop for bits/wordsize.h, since this template does explicitly include bits/wordsize.h

Quoted 5 lines

To fix this emits the pragma once at beginning of file, this is better solution than include guards, and pragma once is practically supported on all compilers except few e.g. cray c/c++ compiler

Something about this is breaking a local recipe (for tlsdate fwiw, but I don't think that's relevant):

| checking openssl/ssl.h usability... no
| checking openssl/ssl.h presence... yes
| configure: WARNING: openssl/ssl.h: present but cannot be compiled
| configure: WARNING: openssl/ssl.h:     check for missing prerequisite headers?
| configure: WARNING: openssl/ssl.h: see the Autoconf documentation
| configure: WARNING: openssl/ssl.h:     section "Present But Cannot
Be Compiled"
| configure: WARNING: openssl/ssl.h: proceeding with the compiler's result
| configure: WARNING:     ## ------------------------------------- ##
| configure: WARNING:     ## Report this to jacob at appelbaum.net ##
| configure: WARNING:     ## ------------------------------------- ##
| checking for openssl/ssl.h... no
| configure: error: OpenSSL is not installed, openssl/sslh is missing
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 6 lines
Quoted 3 lines

Somehow it's not getting the definition for THIRTY_TWO_BIT from openssl/opensslconf.h (from openssl10) when testing for openssl/ssl.h so BN_ULONG ends up undefined... still digging as to what's going on.

thanks, is it using openssl10 ?

I'm also seeing build issues with openssl10 which seem to be related to this change:

| i586-oe-linux-gcc -m32 -march=i586
--sysroot=/.../tmp-glibc/work/i586-oe-linux/mosquitto/1.4.8-r0/recipe-sysroot
-O2 -pipe -g -feliminate-unused-debug-types -fdebug-prefix-map=/.../tmp-glibc/work/i586-oe-linux/mosquitto/1.4.8-r0=/usr/src/debug/mosquitto/1.4.8-r0 -fdebug-prefix-map=/.../tmp-glibc/work/i586-oe-linux/mosquitto/1.4.8-r0/recipe-sysroot= -fdebug-prefix-map=/.../tmp-glibc/work/i586-oe-linux/mosquitto/1.4.8-r0/recipe-sysroot-native= -I. -I.. -I../lib -fPIC -DWITH_TLS -DWITH_TLS_PSK -DWITH_THREADING
-DWITH_SOCKS -c read_handle_shared.c -o read_handle_shared.o
| In file included from
/.../tmp-glibc/work/i586-oe-linux/mosquitto/1.4.8-r0/recipe-sysroot/usr/include/openssl/engine.h:75,
| from net_mosq.c:55:
| /.../tmp-glibc/work/i586-oe-linux/mosquitto/1.4.8-r0/recipe-sysroot/usr/include/openssl/bn.h:321:5:
error: unknown type name 'BN_ULONG'

 |      BN_ULONG *d;                /* Pointer to an array of 'BN_BITS2' bit
 |      ^~~~~~~~
 | /.../tmp-glibc/work/i586-oe-linux/mosquitto/1.4.8-r0/recipe-sysroot/usr/include/openssl/bn.h:337:5:
error: unknown type name 'BN_ULONG'
 |      BN_ULONG n0[2];             /* least significant word(s) of Ni; (type
 |      ^~~~~~~~
author
#9 lore raw
In reply to Alex Kiernan

Please try the patch I posted for openssl10 and report back
https://patchwork.openembedded.org/patch/158573/

Quoted 85 lines · Alex Kiernan
Quoted 47 lines · Khem Raj
Quoted 45 lines · Alex Kiernan
Quoted 43 lines · Khem Raj
Quoted 40 lines · Alex Kiernan
Quoted 11 lines · Khem Raj

Avoid infinite include loops, especially with bits/wordsize.h which is now possible with the synthesized headers since we now also synthesize bits/wordsize.h itelf for some arches e.g. arm/aarch64

In cases where extra preprocessing tools are used such as clang-tidy e.g. and these tools are not passed the knowledge about architecture then case comes where we enter into header include loop for bits/wordsize.h, since this template does explicitly include bits/wordsize.h

Quoted 5 lines

To fix this emits the pragma once at beginning of file, this is better solution than include guards, and pragma once is practically supported on all compilers except few e.g. cray c/c++ compiler

Something about this is breaking a local recipe (for tlsdate fwiw, but I don't think that's relevant):

| checking openssl/ssl.h usability... no
| checking openssl/ssl.h presence... yes
| configure: WARNING: openssl/ssl.h: present but cannot be compiled
| configure: WARNING: openssl/ssl.h:     check for missing prerequisite headers?
| configure: WARNING: openssl/ssl.h: see the Autoconf documentation
| configure: WARNING: openssl/ssl.h:     section "Present But Cannot
Be Compiled"
| configure: WARNING: openssl/ssl.h: proceeding with the compiler's result
| configure: WARNING:     ## ------------------------------------- ##
| configure: WARNING:     ## Report this to jacob at appelbaum.net ##
| configure: WARNING:     ## ------------------------------------- ##
| checking for openssl/ssl.h... no
| configure: error: OpenSSL is not installed, openssl/sslh is missing

Somehow it's not getting the definition for THIRTY_TWO_BIT from openssl/opensslconf.h (from openssl10) when testing for openssl/ssl.h so BN_ULONG ends up undefined... still digging as to what's going on.

Quoted 8 lines
Quoted 5 lines
Quoted 2 lines

thanks, is it using openssl10 ?

It is... I'm sure that's relevant, but I'm failing to spot it.

so where all do you have opensslconf.h in target sysroot ?

$ find tmp -name opensslconf.h -ls
  1589118      8 -rw-r--r--   3 vagrant  vagrant      4534 Feb  6
11:40 tmp/work/x86_64-linux/qemu-helper-native/1.0-r1/recipe-sysroot-native/usr/include/openssl/opensslconf.h
  1451388      4 -rw-r--r--   4 vagrant  vagrant      1971 Feb  6
21:31 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/tlsdate/0.0.13+gitAUTOINC+751f82dc6f-r0/recipe-sysroot/usr/include/openssl/opensslconf.h
  1589118      8 -rw-r--r--   3 vagrant  vagrant      4534 Feb  6
11:40 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/recipe-sysroot-native/usr/include/openssl/opensslconf.h
  1451388      4 -rw-r--r--   4 vagrant  vagrant      1971 Feb  6
21:31 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/sysroot-destdir/usr/include/openssl/opensslconf.h
  1574753      8 -rw-r--r--   1 vagrant  vagrant      7423 Feb  6
21:30 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/openssl-1.0.2q/crypto/opensslconf.h
  1717763      4 -rw-r--r--   1 vagrant  vagrant      3201 Nov 20
13:44 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/openssl-1.0.2q/MacOS/opensslconf.h
  1574836      0 lrwxrwxrwx   1 vagrant  vagrant        26 Feb  6
21:30 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/openssl-1.0.2q/include/openssl/opensslconf.h
-> ../../crypto/opensslconf.h
  1451388      4 -rw-r--r--   4 vagrant  vagrant      1971 Feb  6
21:31 tmp/work/armv7at2hf-neon-poky-linux-gnueabi/openssl10/1.0.2q-r0/image/usr/include/openssl/opensslconf.h
  1589118      8 -rw-r--r--   3 vagrant  vagrant      4534 Feb  6
11:40 tmp/sysroots-components/x86_64/openssl-native/usr/include/openssl/opensslconf.h
  1849804      4 -rw-r--r--   1 vagrant  vagrant      1971 Feb  6
13:28 tmp/sysroots-components/armv7at2hf-neon/openssl/usr/include/openssl/opensslconf.h
  1451388      4 -rw-r--r--   4 vagrant  vagrant      1971 Feb  6
21:31 tmp/sysroots-components/armv7at2hf-neon/openssl10/usr/include/openssl/opensslconf.h

--
Alex Kiernan