Use named constants in the switcher #339
Conversation
nwf
force-pushed
the
202411-switcher-symbolize
branch
from
af59dd8 to
4d21cc6
Compare
| * macro will report an error if the provided value does not equal the constexpr | ||
| * evaluation of `expression`. | ||
| */ | ||
| # define EXPORT_ASSEMBLY_NAME(name, val) \ |
There was a problem hiding this comment.
It took me a minute to work out why we need both of these macros. I guess this one is just the same as EXPORT_ASSEMBLY_EXPRESSION(name, name, val) but without a name collision? That makes me think maybe we should use {} around the definition below?
There was a problem hiding this comment.
We might want the names available in C++, maybe, and the invoking source could, in principle, choose to put things in scope to hide them if desired, but because these are otherwise at top-level, just using {} doesn't do the trick, eliciting "expected unqualified-id" errors from the compiler.
| @@ -75,6 +75,7 @@ namespace priv | |||
| constexpr size_t MCAUSE_LOAD_PAGE_FAULT = 13; | |||
| constexpr size_t MCAUSE_STORE_PAGE_FAULT = 15; | |||
| constexpr size_t MCAUSE_THREAD_EXIT = 24; | |||
| constexpr size_t MCAUSE_THREAD_INTERRUPT = 25; | |||
| constexpr size_t MCAUSE_CHERI = 28; | |||
There was a problem hiding this comment.
This could be a separate PR but these constants are needed in error handlers so we should make them more publicly available in the SDK.
There was a problem hiding this comment.
This is a public heading (and this is why I don't like abbreviations in names of things: priv is short for 'privileged spec' not 'private').
There was a problem hiding this comment.
I guess there's nothing stopping users from include priv/riscv.h it just seems a bit weird. At any rate we should do this in relevant places in the examples/tests.
nwf
force-pushed
the
202411-switcher-symbolize
branch
2 times, most recently
from
1bdb9d3 to
da330fe
Compare
nwf
force-pushed
the
202411-switcher-symbolize
branch
from
da330fe to
8e7188c
Compare
These are not really "trusted stack", and we're about to have some more. While here, tweak a comment as per Robert's suggestion. Co-authored-by: Robert Norton <robert.norton@microsoft.com>
Co-authored-by: Robert Norton <robert.norton@microsoft.com>
Co-authored-by: Robert Norton <robert.norton@microsoft.com>
nwf
force-pushed
the
202411-switcher-symbolize
branch
from
8e7188c to
c661519
Compare
Pull yet more out of #320. Thanks to @rmn30 for the suggestion to make these self-check a la
assembly-helpers.h.