precheck #291 - I'm suspicious the host_alias change may affect Windows.

Wrong suspicion on my part - the Windows parts do use ac_tool_prefix.

Is ad56983 fixing an actual build failure (I'm trying to weigh up the severity, given that 4.08.1 is at RC2 and the original author of the line is on vacation, so I can't check if there's a subtle and uncommented reason for its being wrong!)

Is ad56983 fixing an actual build failure

Yes, when building after configuring with --host x86_64-linux-gnu.

I can reproduce the failure on my machine, using the setting --host x86_64-pc-linux-gnu that seems to be the auto-detected default here:

$ make world.opt
[...]
gcc -c -O2 -fno-strict-aliasing -fwrapv -Wall -Werror -fno-tree-vrp -g -D_FILE_OFFSET_BITS=64 -D_REENTRANT -DCAML_NAME_SPACE  -DOCAML_STDLIB_DIR='"/usr/local/lib/ocaml"'  -DNATIVE_CODE -DTARGET_amd64 -DMODEL_default -DSYS_linux   -o bigarray_n.o bigarray.c
x86_64-pc-linux-gnu-gcc -c -DSYS_linux -I../runtime -DMODEL_default -o amd64.o amd64.S || \
{ echo "If your assembler produced syntax errors, it is probably";\
          echo "unhappy with the preprocessor. Check your assembler, or";\
          echo "try producing amd64.o by hand.";\
          exit 2; }
/bin/sh: x86_64-pc-linux-gnu-gcc: command not found
If your assembler produced syntax errors, it is probably
unhappy with the preprocessor. Check your assembler, or
try producing amd64.o by hand.

This does look like something that we want to include in 4.08.1; I'll discuss this with @Octachron but that is our current consensus.

(If we do merge this and do another release-candidate, we might as well grab #8826 as well to get rid of it.)

@glondu, I'm assuming that you spotted this thanks to some magical and useful Debian testing infrastructure. Do you have other reports coming, or do you know that there are no further issues coming up?

cc @rwmjones

@glondu could you include a Changes entry?

Do you have other reports coming, or do you know that there are no further issues coming up?

Yes, there is at least this patch that I would like to see merged upstream. It fixes FTBFS on kfreebsd-i386.

FYI, you can have a look at our build logs for 4.08.0 (+ some patches). As you can see, there are (at least) 3 failures. I've still not investigated the kfreebsd-amd64 failure, I will probably "fix" powerpc and x32 by disabling the native compiler there.

could you include a Changes entry?

Done in my last push (also rebased on latest 4.08).

@glondu: there is a check-typo warning on your PR, could you fix it?

./Changes:21.81: [long-line] line is over 80 columns

(that's line 21. You can run ./tools/check-typo <base-branch> to run check-typo at home.)

It is, we should have a round of RC for configuration fixes.