1$NetBSD: README.gcc12,v 1.20 2023/10/22 00:09:20 mrg Exp $ 2 3 4new stuff: 5 libmpx 6 liboffloadmic 7 libvtv 8 libcet 9 libphobos 10 d 11 gfortran 12 libbacktrace: +dwarf5_CFLAGS = $(libbacktrace_TEST_CFLAGS) -gdwarf-5 13 c++tools 14 15todo: 16 - audit all uses of new warning-disable variables and 17 convert some from -Wno- to -Wno-error= 18 19arch/feature list. 20 21tools: does build.sh tools work? 22kernels: does a kernel run? y (yes), b (builds/ready), p (partially), ? (not ready) 23libgcc: does libgcc build? 24native-gcc: does a full mknative-gcc complete? 25sani: has sanitizer been ported to this CPU again for GCC 12? b (build) y (works) 26make release: does build.sh release complete? (before sanitizer - bs) 27runs: does the system boot with a full world? 28atf: does atf run / compare well 29switched: has port switched? y (yes), n (no/ready), ? (not ready) 30 31 32architecture tools kernels libgcc native-gcc/sani make release runs atf switched 33------------ ----- ------- ------ --------------- ------------ ---- --- -------- 34aarch64 y y y y y[17] y y y ? 35aarch64eb y y y y y[17] y y y ? 36alpha y y y y b y y y ? 37earmv4 y b ? y ? ? ? ? ? 38earmv4eb y b ? y ? ? ? ? ? 39earmv5 y y y y b y y y ? 40earmv5eb y y y y b y y y ? 41earmv5hf y b ? y ? ? ? ? ? 42earmv5hfeb y b ? y ? ? ? ? ? 43earmv6 y b ? y ? ? ? ? ? 44earmv6eb y b ? y ? ? ? ? ? 45earmv6hf y y y y b y y y ? 46earmv6hfeb y y y y b y y y ? 47earmv7 y b y y ? ? ? ? ? 48earmv7eb y b ? y ? ? ? ? ? 49earmv7hf y y y y b y y y ? 50earmv7hfeb y y y y b y y y ? 51hppa y y y y ? y ? ? ? 52i386 y y y y n[18] y y n[19] ? 53ia64 y b y y y y y[3] n ? 54m68000 y y y y n[11] n y[14] ? ? 55m68k y y y y b y y[14] y[14] ? 56mipseb y b y y b y ? ? ? 57mipsel y b y y b y ? ? ? 58mips64eb y b y y b y ? ? ? 59mips64el y b y y b y ? ? ? 60powerpc y y y y ? n[4] ? ? ? 61powerpc64 y ? ? y ? ? ? ? ? 62sh3eb y ? y y b y ? ? ? 63sh3el y n[20] y y b y n[22] ? ? 64sparc y y y y n[21] y y y ? 65sparc64 y b y y n[16] y y ? ? 66vax y y y y n[23] n n[24,25] ? ? 67x86_64 y y y y y[13] y y y ? 68riscv32 y y y y b y n[2] ? ? 69riscv64 y y y y b y y y ? 70------------ ----- ------- ------ --------------- ------------ ---- --- 71architecture tools kernels libgcc native-gcc/sani make release runs atf 72 73[2]: riscv32 kernels are not really implemented yet, not GCC 12 issue. 74[3]: GENERIC.SKI boots just as well as GCC 10. 75[4]: rs6000 has a new generator for rs6000-builtins.cc and rs6000-builtins.h rs6000-vecdefines.h 76[7]: fails to build kernels: 77 In file included from sys/arch/powerpc/oea/pmap64.c:4: 78 sys/arch/powerpc/oea/pmap.c:2186:2: error: #error PPC_OEA64 not supported 79[11]: sanitizers don't build yet. 80[13]: on amd64, with very basic test, lsan, ubsan, and asan work, but tsan crashes at start up: 81 ThreadSanitizer: CHECK failed: tsan_rtl.cpp:149 "((reinterpret_cast<uptr>(this) % 64)) == ((0))" (0x10, 0x0) (tid=21499) 82 ThreadSanitizer:DEADLYSIGNAL 83 ==21499==ERROR: ThreadSanitizer: SEGV on unknown address 0x000000000017 (pc 0x7f7ff4e5e14e bp 0x7f7fffffdbd0 sp 0x7f7fffffdb40 T21499) 84 ==21499==The signal is caused by a READ memory access. 85 ==21499==Hint: address points to the zero page. 86 ThreadSanitizer:DEADLYSIGNAL 87 ThreadSanitizer: nested bug in the same thread, aborting. 88[14]: "dd count=1", "env LC_CTYPE=en_US.UTF-8 locale", and "tests/crypto/libcrypto/h_evp_test evpmd_blake.txt" abort. 89 adding -fno-stack-protector to strsuftoll.c, citrus_module.c, and blake2_prov.c, respectively, works around the problem. 90[16]: sanitizers crash early: 91 Program received signal SIGSEGV, Segmentation fault. 92 (gdb) bt 93 #0 0x000000004064265c in __sanitizer::CheckASLR () at /usr/src/external/gpl3/gcc/dist/libsanitizer/sanitizer_common/sanitizer_linux.cpp:2181 94 #1 0xffffffffffffc704 in ?? () 95 Backtrace stopped: previous frame identical to this frame (corrupt stack?) 96[17]: (NOT A REGRESSION) libtsan is newly introduced, which fails with: 97 FATAL: ThreadSanitizer: unexpected memory mapping 0x200100000-0x200101000 98[18]: *san requires 64-bit atomic ops: 99 ld: /usr/lib/libasan.so: undefined reference to `__sync_val_compare_and_swap_8' 100 If lib*san is compiled with -march=i586, all tests in usr.bin/{cc,c++} successfully pass. 101[19]: kernel freezes with garbage parmanently output to console, after lib/libc/sys/t_ptrace_wait:syscall_signal_on_sce fails: 102 syscall_signal_on_sce: [[ 3013.0266693] sorry pid 1847 was killed: orphanes 103 ss 104 ss 105 ss 106 ... 107[20]: kernel needs -O1 for ffs32() and ffs_nodealloccg() for multi-user boot. Even if the whole kernel is built with -O1, it panics during ATF: 108 kernel/t_sysv (92/935): 3 test cases 109 msg: [ 3981.4751382] panic: kernel diagnostic assertion "msghdr->msg_spot >= 0" failed: file "../../../../kern/sysv_msg.c", line 426 110[21]: asan has v7/v8/atomic issues: 111 ld: /usr/lib/libasan.so: undefined reference to `__sync_val_compare_and_swap_8' 112 nm shows some more: 113 U __sync_add_and_fetch_4 114 U __sync_fetch_and_add_4 115 U __sync_lock_test_and_set_4 116 U __sync_val_compare_and_swap_1 117 U __sync_val_compare_and_swap_4 118 liblsan and libubsan are mssing the _4 and _8 versions, too. 119[22]: single-user shell crashes. if userland is built with DBG=-O1, it boots into multi-user mode. 120 however, at least db(3) is broken, by which files are corrupted when edited by vi(1). 121[23]: sanitizers fail to build: 122 sanitizer_flag_parser.h:141:72: error: format '%zx' expects argument of type 'size_t', but argument 4 has type 'unsigned int' [-Werror=format=] 123[24]: (NOT A REGRESSION) c++ exception handling is broken. mechanically applying 124 http://www.execsw.org/netbsd/changeset.cgi?id=20140318T234341Z.c59ac4642e7d4b345f6dc5aa9887e9e1f9aa459a#src/gnu/dist/gcc4/gcc/config/vax/vax.c 125 does not help (with adjustments for gcc 10.5.0, which is broken, too): 126 (gdb) invalid-command 127 [1] Illegal instruction (core dumped) gdb 128[25]: (maybe not related to [24]) native gcc is broken: 129 simh# cc hello.c 130 during RTL pass: final 131 hello.c: In function 'main': 132 hello.c:9:1: internal compiler error: Segmentation fault 133 134 135CPU vs platform test table (for CPUs with multiple ports). this is "make release" or just kernels. 136values: y (yes), k (kernels only), n (failed), r (running), ? (not attempted), nx (yes, no x11) 137 138CPU platform list 139--- -------- ---- 140 acorn32 cats epoc32 evbarm-eb evbarm-el hpcarm iyonix netwinder shark zaurus 141earmv4: k k k k k k k y 142earm: nx nx k k 143earmhf: k k 144earmv6: k k 145earmv6hf: nx nx 146earmv7: k k 147earmv7hf: nx nx 148 149 amiga atari cesfic hp300 luna68k mac68k mvme68k news68k next68k sun3 x68k 150m68k: nx k k k k nx k k k nx k 151 152 evbmips emips ews4800mips mipsco newsmips sgimips 153mipseb: k k k k k y 154 155 evbmips algor arc cobalt hpcmips pmax 156mipsel: k k k k k y 157 158 algor evbmips pmax 159mips64eb: y 160mips64el: k y y 161 162 amigappc bebox evbppc ibmnws macppc mvmeppc ofppc prep rs6000 sandpoint 163powerpc: k k k k kn[4] k k k k k 164 165 evbppc macppc ofppc 166powerpc64: k k n[7] 167 168 dreamcast evbsh3 hpcsh landisk mmeye 169sh3eb: y k 170sh3el: k y k k 171