On Thu, May 30, 2019 at 1:23 PM Jiong Wang <jiong.wang@xxxxxxxxxxxxx> wrote: > > There has been quite a few progress around the two steps mentioned in the > answer to the following question: > > Q: BPF 32-bit subregister requirements > > This patch updates the answer to reflect what has been done. > > v2: > - Add missing full stop. (Song Liu) > - Minor tweak on one sentence. (Song Liu) > > v1: > - Integrated rephrase from Quentin and Jakub > > Reviewed-by: Quentin Monnet <quentin.monnet@xxxxxxxxxxxxx> > Reviewed-by: Jakub Kicinski <jakub.kicinski@xxxxxxxxxxxxx> > Signed-off-by: Jiong Wang <jiong.wang@xxxxxxxxxxxxx> Acked-by: Song Liu <songliubraving@xxxxxx> > --- > Documentation/bpf/bpf_design_QA.rst | 30 +++++++++++++++++++++++++----- > 1 file changed, 25 insertions(+), 5 deletions(-) > > diff --git a/Documentation/bpf/bpf_design_QA.rst b/Documentation/bpf/bpf_design_QA.rst > index cb402c5..12a246f 100644 > --- a/Documentation/bpf/bpf_design_QA.rst > +++ b/Documentation/bpf/bpf_design_QA.rst > @@ -172,11 +172,31 @@ registers which makes BPF inefficient virtual machine for 32-bit > CPU architectures and 32-bit HW accelerators. Can true 32-bit registers > be added to BPF in the future? > > -A: NO. The first thing to improve performance on 32-bit archs is to teach > -LLVM to generate code that uses 32-bit subregisters. Then second step > -is to teach verifier to mark operations where zero-ing upper bits > -is unnecessary. Then JITs can take advantage of those markings and > -drastically reduce size of generated code and improve performance. > +A: NO. > + > +But some optimizations on zero-ing the upper 32 bits for BPF registers are > +available, and can be leveraged to improve the performance of JITed BPF > +programs for 32-bit architectures. > + > +Starting with version 7, LLVM is able to generate instructions that operate > +on 32-bit subregisters, provided the option -mattr=+alu32 is passed for > +compiling a program. Furthermore, the verifier can now mark the > +instructions for which zero-ing the upper bits of the destination register > +is required, and insert an explicit zero-extension (zext) instruction > +(a mov32 variant). This means that for architectures without zext hardware > +support, the JIT back-ends do not need to clear the upper bits for > +subregisters written by alu32 instructions or narrow loads. Instead, the > +back-ends simply need to support code generation for that mov32 variant, > +and to overwrite bpf_jit_needs_zext() to make it return "true" (in order to > +enable zext insertion in the verifier). > + > +Note that it is possible for a JIT back-end to have partial hardware > +support for zext. In that case, if verifier zext insertion is enabled, > +it could lead to the insertion of unnecessary zext instructions. Such > +instructions could be removed by creating a simple peephole inside the JIT > +back-end: if one instruction has hardware support for zext and if the next > +instruction is an explicit zext, then the latter can be skipped when doing > +the code generation. > > Q: Does BPF have a stable ABI? > ------------------------------ > -- > 2.7.4 >
