On Wed, 2021-02-17 at 22:58 -0800, Yonghong Song wrote:
> On 2/15/21 5:12 PM, Ilya Leoshkevich wrote:
> > The logic follows that of BTF_KIND_INT most of the time.
> > Sanitization
> > replaces BTF_KIND_FLOATs with equally-sized BTF_KIND_INTs on older
> > kernels.
> >
> > Signed-off-by: Ilya Leoshkevich <iii@xxxxxxxxxxxxx>
> > ---
> > tools/lib/bpf/btf.c | 44
> > +++++++++++++++++++++++++++++++++
> > tools/lib/bpf/btf.h | 8 ++++++
> > tools/lib/bpf/btf_dump.c | 4 +++
> > tools/lib/bpf/libbpf.c | 29 +++++++++++++++++++++-
> > tools/lib/bpf/libbpf.map | 5 ++++
> > tools/lib/bpf/libbpf_internal.h | 2 ++
> > 6 files changed, 91 insertions(+), 1 deletion(-)
> >
> > diff --git a/tools/lib/bpf/btf.c b/tools/lib/bpf/btf.c
> > index d9c10830d749..07a30e98c3de 100644
> > --- a/tools/lib/bpf/btf.c
> > +++ b/tools/lib/bpf/btf.c
[...]
> > @@ -2373,6 +2377,42 @@ int btf__add_datasec(struct btf *btf, const
> > char *name, __u32 byte_sz)
> > return btf_commit_type(btf, sz);
> > }
> >
> > +/*
> > + * Append new BTF_KIND_FLOAT type with:
> > + * - *name* - non-empty, non-NULL type name;
> > + * - *sz* - size of the type, in bytes;
> > + * Returns:
> > + * - >0, type ID of newly added BTF type;
> > + * - <0, on error.
> > + */
> > +int btf__add_float(struct btf *btf, const char *name, size_t
> > byte_sz)
> > +{
> > + struct btf_type *t;
> > + int sz, name_off;
> > +
> > + /* non-empty name */
> > + if (!name || !name[0])
> > + return -EINVAL;
>
> Do we want to ensure byte_sz to be 2/4/8/16?
> Currently, the int type supports 1/2/4/8/16.
>
> In LLVM, the following are supported float types:
>
> case BuiltinType::Half:
> case BuiltinType::Float:
> case BuiltinType::LongDouble:
> case BuiltinType::Float16:
> case BuiltinType::BFloat16:
> case BuiltinType::Float128:
> case BuiltinType::Double:
There can be 80-bit floats on x86:
#include <stdio.h>
int main() { printf("%zu\n", sizeof(long double)); }
prints 12 when compiled with -m32 (not 10 due to alignment I assume).
I guess this now completely kills the idea with sanitizing FLOATs to
equally-sized INTs...
[...]
