search

mkschreder / orangerpcd

Websocket RPC server for OpenWRT
Other
27 stars 11 forks source link

Build fails on musl with -Werror=type-limits #8

Open mk-f opened 7 years ago

mk-f commented 7 years ago

orange-rpcd-2.16.08-c4ad159efa5fc40f610fa0cb11e25fda19f131ad referenced by 2.16.09:

OpenWrt-libtool: compile:  arm-openwrt-linux-muslgnueabi-gcc -DPACKAGE_NAME=\"orangerpcd\" -DPACKAGE_TARNAME=\"orangerpcd\" -DPACKAGE_VERSION=\"0.1.0\" "-DPACKAGE_STRING=\"orangerpcd 0.1.0\"" -DPACKAGE_BUGREPORT=\"mkschreder.uk@gmail.com\" -DPACKAGE_URL=\"\" -DPACKAGE=\"orangerpcd\" -DVERSION=\"0.1.0\" -DSTDC_HEADERS=1 -DHAVE_SYS_TYPES_H=1 -DHAVE_SYS_STAT_H=1 -DHAVE_STDLIB_H=1 -DHAVE_STRING_H=1 -DHAVE_MEMORY_H=1 -DHAVE_STRINGS_H=1 -DHAVE_INTTYPES_H=1 -DHAVE_STDINT_H=1 -DHAVE_UNISTD_H=1 -DHAVE_DLFCN_H=1 -DLT_OBJDIR=\".libs/\" -DHAVE_LUA_H=1 -DHAVE_UCI_H=1 -I. -I/home/bob/openwrt.pac.juci-v2.16.09/staging_dir/target-arm_arm926ej-s_musl-1.1.14_eabi/usr/include -I/home/bob/openwrt.pac.juci-v2.16.09/staging_dir/target-arm_arm926ej-s_musl-1.1.14_eabi/include -I/home/bob/openwrt.pac.juci-v2.16.09/staging_dir/toolchain-arm_arm926ej-s_gcc-5.3.0_musl-1.1.14_eabi/usr/include -I/home/bob/openwrt.pac.juci-v2.16.09/staging_dir/toolchain-arm_arm926ej-s_gcc-5.3.0_musl-1.1.14_eabi/include/fortify -I/home/bob/openwrt.pac.juci-v2.16.09/staging_dir/toolchain-arm_arm926ej-s_gcc-5.3.0_musl-1.1.14_eabi/include -Wall -fPIC -Wno-format-y2k -W -Wstrict-prototypes -Wmissing-prototypes -Wpointer-arith -Wreturn-type -Wwrite-strings -Wswitch -Wcast-align -Wchar-subscripts -Winline -Wnested-externs -Wredundant-decls -Wmissing-field-initializers -Wextra -Wformat=2 -Wno-format-nonliteral -Wpointer-arith -Wno-missing-braces -Wno-unused-parameter -Wno-unused-variable -Wno-inline -std=gnu99 -fstack-protector-all -Wall -Werror -Os -pipe -march=armv5te -mtune=arm926ej-s -g3 -fno-caller-saves -fno-plt -fhonour-copts -Wno-error=unused-but-set-variable -Wno-error=unused-result -mfloat-abi=soft -iremap /home/bob/openwrt.pac.juci-v2.16.09/build_dir/target-arm_arm926ej-s_musl-1.1.14_eabi/orange-rpcd-2.16.08-c4ad159efa5fc40f610fa0cb11e25fda19f131ad:orange-rpcd-2.16.08-c4ad159efa5fc40f610fa0cb11e25fda19f131ad -Wformat -Werror=format-security -fstack-protector -D_FORTIFY_SOURCE=1 -Wl,-z,now -Wl,-z,relro -MT liborange_la-base64.lo -MD -MP -MF .deps/liborange_la-base64.Tpo -c base64.c  -fPIC -DPIC -o .libs/liborange_la-base64.o
base64.c: In function 'base64_decode_value':
base64.c:135:15: error: comparison is always false due to limited range of data type [-Werror=type-limits]
  if (value_in < 0 || value_in > decoding_size) return -1;
               ^
base64.c: In function 'base64_decode_block':
base64.c:164:22: error: comparison is always false due to limited range of data type [-Werror=type-limits]
    } while (fragment < 0);
                      ^
base64.c:176:22: error: comparison is always false due to limited range of data type [-Werror=type-limits]
    } while (fragment < 0);
                      ^
base64.c:189:22: error: comparison is always false due to limited range of data type [-Werror=type-limits]
    } while (fragment < 0);
                      ^
base64.c:202:22: error: comparison is always false due to limited range of data type [-Werror=type-limits]
    } while (fragment < 0);
                      ^
cc1: all warnings being treated as errors
Makefile:530: recipe for target 'liborange_la-base64.lo' failed
make[5]: *** [liborange_la-base64.lo] Error 1
mkschreder commented 7 years ago

How can value_in < 0 be always false when value_in is a signed char? Surely it can be negative? Am I missing something? Can you try changing type of value_in and fragment to int and see if that gets rid of the error? Actually it can be slightly more "correct" since if value_in is less than -85 that function will wrap.. so I think using an int there is actually better. Can you see if that fixes it for you?

mkschreder commented 7 years ago

I have added -Wtype-limits to makefile and changes these to ints for now.

mk-f commented 7 years ago

char is not signed char, by c-standard "it's up to the compiler" whether if char denotes an unsigned or signed char. GCC seems to default to unsigned char.

mkschreder commented 7 years ago

I see. Can we be sure default int is always signed?

On Thu, Sep 8, 2016 at 10:47 AM, Fabian Müller-Knapp < notifications@github.com> wrote:

char is not signed char, by c-standard "it's up to the compiler" whether if char denotes an unsigned or signed char. GCC seems to default to unsigned char.

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/mkschreder/orangerpcd/issues/8#issuecomment-245533485, or mute the thread https://github.com/notifications/unsubscribe-auth/AERqiRgYK239dY7OfUxZainAWptdcj0Iks5qn8uzgaJpZM4J12C4 .

mk-f commented 7 years ago

I think so. The C89-Draft reads (3.1.2.5):

There are four signed integer types, designated as signed char, short int, int, and long int. (The signed integer and other types may be designated in several additional ways, as described in 3.5.2)

mkschreder commented 7 years ago

I use c99.

On 8 Sep 2016 11:42, "Fabian Müller-Knapp" notifications@github.com wrote:

I think so. The C89-Draft reads (3.1.2.5):

There are four signed integer types, designated as signed char, short int, int, and long int. (The signed integer and other types may be designated in several additional ways, as described in 3.5.2)

— You are receiving this because you commented.

Reply to this email directly, view it on GitHub https://github.com/mkschreder/orangerpcd/issues/8#issuecomment-245546829, or mute the thread https://github.com/notifications/unsubscribe-auth/AERqidC7wW-bXZXEbDHUE0349IR39F7Yks5qn9hpgaJpZM4J12C4 .

mk-f commented 7 years ago

There are five standard signed integer types, designated as signed char, short int, int, long int, and long long int. (These and other types may be designated in several additional ways, as described in 6.7.2.) There may also be implementation-defined extended signed integer types.25) The standard and extended signed integer types are collectively called signed integer types.26)

http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf (6.2.5) (this is the C99-final-draft)

mkschreder commented 7 years ago

Sounds like it answers it. Good.

On 8 Sep 2016 11:51, "Fabian Müller-Knapp" notifications@github.com wrote:

There are five standard signed integer types, designated as signed char, short int, int, long int, and long long int. (These and other types may be designated in several additional ways, as described in 6.7.2.) There may also be implementation-defined extended signed integer types.25) The standard and extended signed integer types are collectively called signed integer types.26)

http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf (6.2.5)

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/mkschreder/orangerpcd/issues/8#issuecomment-245549024, or mute the thread https://github.com/notifications/unsubscribe-auth/AERqiblP6N4KtUR0HvTO5nHOn2yOf0HVks5qn9qXgaJpZM4J12C4 .

mkschreder commented 7 years ago

Feel free to go over my latest concurrency implementation and see if you notice any issues. I currently run 10 threads that wait on rx queue and as messages arrive they load and run scripts and push back responses to the tx queue in the orange_ws_server which is then handled by the websocket thread (memory caching of scripts will need to go. it is not used anymore). Now you can run long tasks in the lua scripts, like spawning other processes etc and responding when ready without it affecting performance of the rest of the server. I should probably make the number of threads configurable though..

Most important right now is to make sure I have not introduced any concurrency issues in my last few commits. So far both helgrind and drd checks pass on the ws_server unit test. Good idea to add more tests and stress test the server..

On 8 Sep 2016 11:51, "Fabian Müller-Knapp" notifications@github.com wrote:

There are five standard signed integer types, designated as signed char, short int, int, long int, and long long int. (These and other types may be designated in several additional ways, as described in 6.7.2.) There may also be implementation-defined extended signed integer types.25) The standard and extended signed integer types are collectively called signed integer types.26)

http://www.open-std.org/jtc1/sc22/wg14/www/docs/n1256.pdf (6.2.5)

— You are receiving this because you commented. Reply to this email directly, view it on GitHub https://github.com/mkschreder/orangerpcd/issues/8#issuecomment-245549024, or mute the thread https://github.com/notifications/unsubscribe-auth/AERqiblP6N4KtUR0HvTO5nHOn2yOf0HVks5qn9qXgaJpZM4J12C4 .