CPUID lets the application know if it is running on a CPU with hybrid
CPU clusters.
This matches big.little fairly easily. Walk the affinity mask and
check if we are running on a big.little system and report it to the
guest.
For x86-64 host just pass through the flag.
Only a partial fix for #1330, still needs preemption disabled to work.
On x86-64 hosts the Linux kernel resides in the top bit of VA which
isn't mapped in to userspace.
This means that userspace will never receive pointers living with that
top bit set unless you're running a 57bit VA host.
This results in userspace pointers never needing to do the sign
extending pointer canonicalization. But additionally some applications
actually don't understand the pointer canonicalization.
This results in bugs like: https://github.com/golang/go/issues/49405
Now if you're running on a 57bit VA host, this will end up behaving like
FEX but it seems like no one in golang land has really messed with 57bit
VA yet.
In AArch64, when configured with a 48bit VA, the userspace gets the full
48bit VA space and on EL mode switch has the full address range change
to the kernel's 48bit VA.
This means that we will /very/ likely allocate pointers in the high
48bit space since Linux currently allocates top-down.
So behave more like x86-64, hide the top 128TB of memory space from the
guest before boot.
Testing: Took the M1Max 15ms to 21ms allocate the top 128TB.
We've run out of the range of 32 bits already and will need to use
another flag in upcoming changes, so we need to expand our flags to be
64-bit.
While we're at it, we can use a dedicated type alias for the instruction
flags to make the interface changeable from one spot in the future.
We can share this between the interpreter and the JIT. Necessary to
support the TSO-correct interpreter path.
With this change the interpreter is TSO-correct for GPRs. Just not FPRs
yet.
There's zero need for these to have an independent size field and it was
just confusing.
For stores it was always set to zero and for loads it was just
duplicated.
In addition this allows introspection of the store op without casting
the op, which can be useful in edge cases
This is a pretty straightforward operation that can be nicely modeled
by the BIC instruction on ARMv8, which is nice since we can get rid of
the need to manually perform the And and Not operations.
This lets us have JITsymbols grouped by library.
Useful for determining where to thunk.
Sadly perf doesn't have an option to deduplicate regions by name, so
some external tooling is necessary to make it look nice.
Previously the offset mismatch error was expecting two arguments, but
only one was provided.
While we're in the area we can convert the logging type over to the
fmt-capable one which can catch these.
While the previous change eliminated much of the codegen caused by
constructing everything individually on the stack, it didn't eliminate a
memcpy of all the elements onto the stack.
This eliminates the memcpys by allowing the compiler to place all the
data into RO and just reference that data.
Keeping this in the commit history to go back to later.
While these are required for static-pie builds to work, with the glibc
bug we can't use those yet.
Disable for now since it is is unnecessary.
If the thunk configuration is enabled then preemptively load the
libraries. Since they will need to be loaded for any thunking library.
This is because we need to ALWAYS share the allocators to the thunks.
Ryan Houdek