iii-i wrote:
I just checked with a few small examples, and while the ABI does not seem to
change, the code generation looks broken for tail recursion:
```
$ cat 1.c
void baz(long double);
void quux(void) { baz((long double)1); }
$ ./bin/clang -O3 -S 1.c
$ cat 1.s
[...]
aghi %r15, -176 # stack frame
larl %r1, .LCPI0_0 # address of (long double)1 in the literal pool
lxeb %f0, 0(%r1) # f0:f2=(long double)1
la %r2, 160(%r15) # r2=&var_160
std %f0, 160(%r15) # var_160=(long double)1
std %f2, 168(%r15)
aghi %r15, 176 # var_160 is above the stack pointer now!
jg baz@PLT # baz(&var_160)
[...]
```
Regarding MSan/DFSan, they need to know whether to copy value's or pointer's
shadow into arguments' TLS. MSan already has checks for that
(`CB.paramHasAttr(ArgNo, Attribute::ByVal)`), and for DFSan I've added them in
my local branch.
https://github.com/llvm/llvm-project/pull/66404
_______________________________________________
cfe-commits mailing list
cfe-commits@lists.llvm.org
https://lists.llvm.org/cgi-bin/mailman/listinfo/cfe-commits
