I have a few modifications I'd like to get reviewed, is it possible for you to look at these commits (from https://github.com/MikaelUrankar/rust-ports/commits/master)
https://github.com/MikaelUrankar/rust-ports/commit/e0bd7fa10394b00e3273a41da4cec33b83128272
https://github.com/MikaelUrankar/rust-ports/commit/ad62db74f035e86fcc3954232f59ab7c9862336f
https://github.com/MikaelUrankar/rust-ports/commit/2b7ac06ba4f33a066cef65db907297ea95d43af4
https://github.com/MikaelUrankar/rust-ports/commit/c496f973b2b14c2949d37c06b0dcf3af37cc84c7

I would also like to add a wasm32 and a nightly options:
https://github.com/MikaelUrankar/rust-ports/commit/c5e727207275b1ba0a74ba5d4b0e0b2de5b77739
https://github.com/MikaelUrankar/rust-ports/commit/7ba1fff787828a9064aac60a786f7bc63c3b2ae7

Our work has collided at some point, I can rebase my work if needed.

Build is ongoing on aarch64

In D23385#513089, @mikael wrote:

I would also like to add a wasm32 and a nightly options:
https://github.com/MikaelUrankar/rust-ports/commit/c5e727207275b1ba0a74ba5d4b0e0b2de5b77739
https://github.com/MikaelUrankar/rust-ports/commit/7ba1fff787828a9064aac60a786f7bc63c3b2ae7

Let's do those in separate reviews after 1.41.0.

It's ok on armv7, ppc64 elfv1/elfv2.
The builds hangs on my aarch64, not sure if it's a problem with rust or the system.
armv6 doesn't build:

  = note: /tmp/rustcdnGbao/librustc_codegen_llvm-9f9b2279994669d5.rlib(rustc_codegen_llvm-9f9b2279994669d5.rustc_codegen_llvm.e2557spx-cgu.11.rcgu.o): In function `_$LT$rustc_target..abi..call..CastTarget$u20$a
s$u20$rustc_codegen_llvm..abi..LlvmType$GT$::llvm_type::h1296210ab461fc57':                                                                                                                                       
          rustc_codegen_llvm.e2557spx-cgu.11:(.text._ZN89_$LT$rustc_target..abi..call..CastTarget$u20$as$u20$rustc_codegen_llvm..abi..LlvmType$GT$9llvm_type17h1296210ab461fc57E+0x54): relocation truncated to fi
t: R_ARM_CALL against symbol `__aeabi_uldivmod' defined in .text.__aeabi_uldivmod section in /tmp/rustcdnGbao/libcompiler_builtins-ee65b414e4115a8f.rlib(compiler_builtins-ee65b414e4115a8f.compiler_builtins.ay8p
39ey-cgu.13.rcgu.o)

In D23385#513496, @mikael wrote:

It's ok on armv7, ppc64 elfv1/elfv2.
The builds hangs on my aarch64, not sure if it's a problem with rust or the system.

When/where does it hang?

armv6 doesn't build:

  = note: /tmp/rustcdnGbao/librustc_codegen_llvm-9f9b2279994669d5.rlib(rustc_codegen_llvm-9f9b2279994669d5.rustc_codegen_llvm.e2557spx-cgu.11.rcgu.o): In function `_$LT$rustc_target..abi..call..CastTarget$u20$a
s$u20$rustc_codegen_llvm..abi..LlvmType$GT$::llvm_type::h1296210ab461fc57':                                                                                                                                       
          rustc_codegen_llvm.e2557spx-cgu.11:(.text._ZN89_$LT$rustc_target..abi..call..CastTarget$u20$as$u20$rustc_codegen_llvm..abi..LlvmType$GT$9llvm_type17h1296210ab461fc57E+0x54): relocation truncated to fi
t: R_ARM_CALL against symbol `__aeabi_uldivmod' defined in .text.__aeabi_uldivmod section in /tmp/rustcdnGbao/libcompiler_builtins-ee65b414e4115a8f.rlib(compiler_builtins-ee65b414e4115a8f.compiler_builtins.ay8p
39ey-cgu.13.rcgu.o)

That's a really weird error? Do you think that maybe forcing the build to use ld.lld (or newer ld.bfd from devel/binutils) would help at all? Cross-compiling armv6-rust-bootstrap-1.41.0 was fine.

--- Makefile.orig
+++ Makefile
@@ -165,7 +165,13 @@
 	@${ECHO_CMD} '[target.${_RUST_TARGET}]' >> ${WRKSRC}/config.toml
 	@${ECHO_CMD} 'cc="${CC}"' >> ${WRKSRC}/config.toml
 	@${ECHO_CMD} 'cxx="${CXX}"' >> ${WRKSRC}/config.toml
+.if ${ARCH} == armv6
+	@${PRINTF} '#!/bin/sh\nexec ${CC} -fuse-ld=lld "$$@"' > ${WRKDIR}/cc-wrapper 	 
+	@${CHMOD} +x ${WRKDIR}/cc-wrapper 	 
+	@${ECHO_CMD} 'linker="${WRKDIR}/cc-wrapper"' >> ${WRKSRC}/config.toml
+.else
 	@${ECHO_CMD} 'linker="${CC}"' >> ${WRKSRC}/config.toml
+.endif
 	@${ECHO_CMD} '[dist]' >> ${WRKSRC}/config.toml
 	@${ECHO_CMD} 'src-tarball=false' >> ${WRKSRC}/config.toml
 	@${REINPLACE_CMD} 's,%CC%,${CC},g' \

In D23385#513500, @tobik wrote:

In D23385#513496, @mikael wrote:

It's ok on armv7, ppc64 elfv1/elfv2.
The builds hangs on my aarch64, not sure if it's a problem with rust or the system.

When/where does it hang?

armv6 doesn't build:

  = note: /tmp/rustcdnGbao/librustc_codegen_llvm-9f9b2279994669d5.rlib(rustc_codegen_llvm-9f9b2279994669d5.rustc_codegen_llvm.e2557spx-cgu.11.rcgu.o): In function `_$LT$rustc_target..abi..call..CastTarget$u20$a
s$u20$rustc_codegen_llvm..abi..LlvmType$GT$::llvm_type::h1296210ab461fc57':                                                                                                                                       
          rustc_codegen_llvm.e2557spx-cgu.11:(.text._ZN89_$LT$rustc_target..abi..call..CastTarget$u20$as$u20$rustc_codegen_llvm..abi..LlvmType$GT$9llvm_type17h1296210ab461fc57E+0x54): relocation truncated to fi
t: R_ARM_CALL against symbol `__aeabi_uldivmod' defined in .text.__aeabi_uldivmod section in /tmp/rustcdnGbao/libcompiler_builtins-ee65b414e4115a8f.rlib(compiler_builtins-ee65b414e4115a8f.compiler_builtins.ay8p
39ey-cgu.13.rcgu.o)

That's a really weird error? Do you think that maybe forcing the build to use ld.lld (or newer ld.bfd from devel/binutils) would help at all? Cross-compiling armv6-rust-bootstrap-1.41.0 was fine.

--- Makefile.orig
+++ Makefile
@@ -165,7 +165,13 @@
 	@${ECHO_CMD} '[target.${_RUST_TARGET}]' >> ${WRKSRC}/config.toml
 	@${ECHO_CMD} 'cc="${CC}"' >> ${WRKSRC}/config.toml
 	@${ECHO_CMD} 'cxx="${CXX}"' >> ${WRKSRC}/config.toml
+.if ${ARCH} == armv6
+	@${PRINTF} '#!/bin/sh\nexec ${CC} -fuse-ld=lld "$$@"' > ${WRKDIR}/cc-wrapper 	 
+	@${CHMOD} +x ${WRKDIR}/cc-wrapper 	 
+	@${ECHO_CMD} 'linker="${WRKDIR}/cc-wrapper"' >> ${WRKSRC}/config.toml
+.else
 	@${ECHO_CMD} 'linker="${CC}"' >> ${WRKSRC}/config.toml
+.endif
 	@${ECHO_CMD} '[dist]' >> ${WRKSRC}/config.toml
 	@${ECHO_CMD} 'src-tarball=false' >> ${WRKSRC}/config.toml
 	@${REINPLACE_CMD} 's,%CC%,${CC},g' \

It's ok on armv6 with binutils:ld or ld.lld.
aarch64 doesn't hang but it's extremely slow, I don't remember it being so slow.

In D23385#513639, @mikael wrote:

aarch64 doesn't hang but it's extremely slow, I don't remember it being so slow.

I think it's unlikely to be caused by the update. To find out maybe reattempt to build Rust 1.40.0 and see how that goes. But maybe it is something simple like running out of free disk space on the builder?

Can I commit it or do you want more time to debug it?