Building on SmartOS x64, I don't know why 2.8 can be built without problem, but building 3.1.1.1 encountered the following failure:
[root@hswander ~/network-3.1.1.1]# cabal -v3 v2-build
...
/root/network-3.1.1.1/dist-newstyle/build/x86_64-solaris/ghc-8.6.5/network-3.1.1.1/build/network.buildinfo
/opt/local/bin/gcc -m64 -fno-stack-protector -m64 /tmp/8975-4.c -o /tmp/8975-5 '-D__GLASGOW_HASKELL__=806' '-Dsolaris2_BUILD_OS=1' '-Dx86_64_BUILD_ARCH=1' '-Dsolaris2_HOST_OS=1' '-Dx86_64_HOST_ARCH=1' -I/root/network-3.1.1.1/dist-newstyle/build/x86_64-solaris/ghc-8.6.5/network-3.1.1.1/build/autogen -I/root/network-3.1.1.1/dist-newstyle/build/x86_64-solaris/ghc-8.6.5/network-3.1.1.1/build/include -I./include -I. -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/bytestring-0.10.8.2/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/base-4.12.0.0/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/integer-gmp-1.0.2.0/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/include -lnsl -lsocket -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/bytestring-0.10.8.2 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/deepseq-1.4.4.0 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/array-0.5.3.0 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/base-4.12.0.0 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/integer-gmp-1.0.2.0 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/ghc-prim-0.5.3 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/rts
/opt/local/bin/gcc returned ExitFailure 1 with error message:
In file included from /tmp/8975-4.c:1:0:
./include/HsNet.h:96:14: error: unknown type name 'u_int32_t'; did you mean
'uint32_t'?
u_int32_t addr
^~~~~~~~~
uint32_t
/opt/local/bin/gcc -m64 -fno-stack-protector -m64 /tmp/8975-6.c -o /tmp/8975-7 -lnsl -lsocket -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/bytestring-0.10.8.2 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/deepseq-1.4.4.0 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/array-0.5.3.0 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/base-4.12.0.0 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/integer-gmp-1.0.2.0 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/ghc-prim-0.5.3 -L/opt/local/ghc8.6.5/lib/ghc-8.6.5/rts
/opt/local/bin/gcc -m64 -fno-stack-protector -m64 /tmp/8975-8.c -o /tmp/8975-9 -c '-D__GLASGOW_HASKELL__=806' '-Dsolaris2_BUILD_OS=1' '-Dx86_64_BUILD_ARCH=1' '-Dsolaris2_HOST_OS=1' '-Dx86_64_HOST_ARCH=1' -I/root/network-3.1.1.1/dist-newstyle/build/x86_64-solaris/ghc-8.6.5/network-3.1.1.1/build/autogen -I/root/network-3.1.1.1/dist-newstyle/build/x86_64-solaris/ghc-8.6.5/network-3.1.1.1/build/include -I./include -I. -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/bytestring-0.10.8.2/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/base-4.12.0.0/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/integer-gmp-1.0.2.0/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/include
/opt/local/bin/gcc returned ExitFailure 1 with error message:
In file included from /tmp/8975-8.c:1:0:
./include/HsNet.h:96:14: error: unknown type name 'u_int32_t'; did you mean
'uint32_t'?
u_int32_t addr
^~~~~~~~~
uint32_t
/opt/local/bin/gcc -m64 -fno-stack-protector -m64 /tmp/8975-10.c -o /tmp/8975-11 -E '-D__GLASGOW_HASKELL__=806' '-Dsolaris2_BUILD_OS=1' '-Dx86_64_BUILD_ARCH=1' '-Dsolaris2_HOST_OS=1' '-Dx86_64_HOST_ARCH=1' -I/root/network-3.1.1.1/dist-newstyle/build/x86_64-solaris/ghc-8.6.5/network-3.1.1.1/build/autogen -I/root/network-3.1.1.1/dist-newstyle/build/x86_64-solaris/ghc-8.6.5/network-3.1.1.1/build/include -I./include -I. -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/bytestring-0.10.8.2/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/base-4.12.0.0/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/integer-gmp-1.0.2.0/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/include
/opt/local/bin/gcc -m64 -fno-stack-protector -m64 /tmp/8975-12.c -o /tmp/8975-13 -c '-D__GLASGOW_HASKELL__=806' '-Dsolaris2_BUILD_OS=1' '-Dx86_64_BUILD_ARCH=1' '-Dsolaris2_HOST_OS=1' '-Dx86_64_HOST_ARCH=1' -I/root/network-3.1.1.1/dist-newstyle/build/x86_64-solaris/ghc-8.6.5/network-3.1.1.1/build/autogen -I/root/network-3.1.1.1/dist-newstyle/build/x86_64-solaris/ghc-8.6.5/network-3.1.1.1/build/include -I./include -I. -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/bytestring-0.10.8.2/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/base-4.12.0.0/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/integer-gmp-1.0.2.0/include -I/opt/local/ghc8.6.5/lib/ghc-8.6.5/include
/opt/local/bin/gcc returned ExitFailure 1 with error message:
In file included from /tmp/8975-12.c:1:0:
./include/HsNet.h:96:14: error: unknown type name 'u_int32_t'; did you mean
'uint32_t'?
u_int32_t addr
^~~~~~~~~
uint32_t
CallStack (from HasCallStack):
die', called at ./Distribution/Simple/Configure.hs:1879:34 in Cabal-3.0.0.0-LUhNDbiv0rJB8Z6Pf9aFIJ:Distribution.Simple.Configure
checkForeignDeps, called at ./Distribution/Simple.hs:643:7 in Cabal-3.0.0.0-LUhNDbiv0rJB8Z6Pf9aFIJ:Distribution.Simple
cabal: Bad header file: HsNet.h
The header file contains a compile error. You can re-run configure with the
verbosity flag -v3 to see the error messages from the C compiler.
...
I suppose uint32_t is equivalent and more portable according to https://lists.freedesktop.org/archives/release-wranglers/2004-August/000926.html
uint32_t is the form specified by the ISO/ANSI C99 standard, the
current version of the Single Unix Specification (the combined
replacement for the POSIX & XPG series of specs, and LSB 1.3.
while with caveat as said https://lists.freedesktop.org/archives/release-wranglers/2004-August/000932.html
uint32_t might be preferred, but that does not mean it is defined in
<sys/types.h> on all platforms. On Darwin one needs to use <stdint.h>
or <inttypes.h> as suggested in the Solaris 10 documentation. In any
case, <sys/types.h> is definitely not the right cross platform header
file and causes build breakage.
I can draft a PR for review if you folks agree on the change, but the source change is so small (just to remove one character), I guess major effort is to have CI on all supported platforms pass, which I have no idea where to start.
network/include/HsNet.h
Line 96 in 467a59a
Building on SmartOS x64, I don't know why 2.8 can be built without problem, but building 3.1.1.1 encountered the following failure:
I suppose
uint32_tis equivalent and more portable according to https://lists.freedesktop.org/archives/release-wranglers/2004-August/000926.htmlwhile with caveat as said https://lists.freedesktop.org/archives/release-wranglers/2004-August/000932.html
I can draft a PR for review if you folks agree on the change, but the source change is so small (just to remove one character), I guess major effort is to have CI on all supported platforms pass, which I have no idea where to start.