addpatch: freehdl #4454
addpatch: freehdl #4454
Conversation
|
Could you leave some comment why these changes is needed to help maintaining? Please also try to upstream them if applicable. Thanks. |
freehdl/riscv64.patch
Outdated
| - rm ieee/math_real.cc \ | ||
| - ieee/numeric_bit.cc \ | ||
| - ieee/numeric_std.cc \ | ||
| - ieee/std_logic_1164.cc \ | ||
| - ieee/std_logic_arith.cc \ | ||
| - ieee/std_logic_signed.cc \ | ||
| - ieee/std_logic_unsigned.cc \ | ||
| - ieee/vital_timing.cc |
There was a problem hiding this comment.
How about this? Is the removal needed on x86_64?
There was a problem hiding this comment.
Not must, I have compiled on x86_64, all passed,
If removed, It will occurs some errors on both x86_64 and riscv64.
There was a problem hiding this comment.
If removed, It will occurs some errors on both x86_64 and riscv64.
What error would occur? Removing some source code deserves an explanation.
There was a problem hiding this comment.
Remove some source code, here is build log: freehdl-0.0.8-13-riscv64-build.log
There was a problem hiding this comment.
Remove some source code, here is build log: freehdl-0.0.8-13-riscv64-build.log
Could you point out which line indicates the related error? I'm really confused. 😕
There was a problem hiding this comment.
Basically, our principle is to be as consistent as we can with the original build script, or give a concrete sound reason why they are different (not just that there will be core dumped).
In the original build script, they were already rm these .cc files, after reading the discussion thread, I also still get confused why you want to remove the rm .cc lines.
There was a problem hiding this comment.
If these .cc files are supposed to be regenerated, and freehdl_v2cc will 'core dumped', then we should figure out why it crashes and try to fix that, instead of leaving an unknown bug there.
There was a problem hiding this comment.
I'm keeping these rm .cc lines now, looking for another way to eliminate core dump or directly report to upstream.
There was a problem hiding this comment.
There was a problem hiding this comment.
Well, I read the codebase and figured this should have no bad effects, so let's try it! Thanks for the explanation.
BTW may you try to register an account on Arch Linux GitLab to report the issue? We don't want to steal your patch. :)
OK, I register an account to report to upstream.
No description provided.