initialization of large int64 attribute fails
9383227c5b
nixbld: consistent netif variables
233998b8f3
nixbld: work around tunnel bring-up race condition
90a6b84c09
nixbld: work around tunnel TCPMSS issues
Change LVDS configuration into groups
Or you can abandon your lazy approach and make it supported in gateware, if it matters. That's an easy thing to do.
8c540d1033
[core] codegen/stmt: Add more casts for boolean types
0cc60a3d33
[core] codegen/expr: Fix missing cast to i1
Fix several missing i1/i8 casts
a59c26aa99
[artiq] Fix RPC of ndarrays from host
02d93b11d1
[meta] Update dependencies
RPC of ndarray from host to device results in garbage data
Fix RPC of ndarrays from host to device
Use postfix options for routing mails through tunnel
call it trump0
, he's the one who started this shit.
Fix several missing i1/i8 casts
How about using i1 for bools instead, and casting them when calling foreign functions?
Then information about what is a bool stays in the LLVM IR.
Fix several missing i1/i8 casts
This is a bit of a shotgun solution. Can't we be more subtle and just have the casts explicitly where they are needed?
Fix several missing i1/i8 casts
Hmm OK, but then I don't remember why we can't just use i1 for the Python bool type. Then we would not need any casts?