mirror of
https://github.com/RPCS3/llvm-mirror.git
synced 2024-11-26 04:32:44 +01:00
0cfb2248a6
Commit 13f6c81c5d9a ("[BPF] simplify zero extension
with MOV_32_64") tried to use MOV_32_64 instructions
instead of lshift/rshift instructions for zero extension.
This has the benefit to remove the number of instructions
and may help verifier too.
But the same commit also removed the old MOV_32_64
pruning as it deems unsafe as MOV_32_64 does have the
side effect, zeroing out the top 32bit in the register.
This caused the following failure in kernel selftest
test_cls_redirect.o. In linux kernel, we have
struct __sk_buff {
__u32 data;
__u32 data_end;
};
The compiler will generate 32bit load for __sk_buff->data
and __sk_buff->data_end. But kernel verifier will actually
loads an address (64bit address on 64bit kernel) to the
result register. In this particular example, the explicit zext
was not optimized away and destroyed top 32bit
address and the verifier rejected the program :
w2 = *(u32 *)(r1 + 76)
...
r2 = w2 /* MOV_32_64: this will clear top 32bit */
Currently, if the load and the zext are next to each other, the
instruction pattern match can actually capture this to
avoid MOV_32_64, e.g., in BPFInstrInfo.td, we have
def : Pat<(i64 (zextloadi32 ADDRri:$src)),
(SUBREG_TO_REG (i64 0), (LDW32 ADDRri:$src), sub_32)>;
However, if they are not next to each other, LDW32 and
MOV_32_64 are generated, which may cause the above mentioned
problem.
BPF Backend already tried to optimize away pattern
mov_32_64 + lshift + rshift
Commit 13f6c81c5d9a may generate mov_32_64 not followed by shifts.
This patch added optimization for only mov_32_64 too.
Differential Revision: https://reviews.llvm.org/D81048
56 lines
1.4 KiB
LLVM
56 lines
1.4 KiB
LLVM
; RUN: llc < %s -march=bpf -mattr=+alu32 -verify-machineinstrs | FileCheck %s
|
|
;
|
|
; Source:
|
|
; struct __sk_buff {
|
|
; unsigned data;
|
|
; unsigned data_end;
|
|
; };
|
|
;
|
|
; void * test(int flag, struct __sk_buff *skb)
|
|
; {
|
|
; void *p;
|
|
;
|
|
; if (flag) {
|
|
; p = (void *)(long)skb->data;
|
|
; __asm__ __volatile__("": : :"memory");
|
|
; } else {
|
|
; p = (void *)(long)skb->data_end;
|
|
; __asm__ __volatile__("": : :"memory");
|
|
; }
|
|
;
|
|
; return p;
|
|
; }
|
|
; Compilation flag:
|
|
; clang -target bpf -O2 -S -emit-llvm t.c
|
|
|
|
%struct.__sk_buff = type { i32, i32 }
|
|
|
|
define dso_local i8* @test(i32 %flag, %struct.__sk_buff* nocapture readonly %skb) local_unnamed_addr {
|
|
entry:
|
|
%tobool = icmp eq i32 %flag, 0
|
|
br i1 %tobool, label %if.else, label %if.then
|
|
|
|
if.then:
|
|
%data = getelementptr inbounds %struct.__sk_buff, %struct.__sk_buff* %skb, i64 0, i32 0
|
|
%0 = load i32, i32* %data, align 4
|
|
tail call void asm sideeffect "", "~{memory}"()
|
|
br label %if.end
|
|
|
|
if.else:
|
|
%data_end = getelementptr inbounds %struct.__sk_buff, %struct.__sk_buff* %skb, i64 0, i32 1
|
|
%1 = load i32, i32* %data_end, align 4
|
|
tail call void asm sideeffect "", "~{memory}"()
|
|
br label %if.end
|
|
|
|
if.end:
|
|
%p.0.in.in = phi i32 [ %0, %if.then ], [ %1, %if.else ]
|
|
%p.0.in = zext i32 %p.0.in.in to i64
|
|
%p.0 = inttoptr i64 %p.0.in to i8*
|
|
ret i8* %p.0
|
|
}
|
|
|
|
; CHECK: w0 = *(u32 *)(r2 + 0)
|
|
; CHECK: w0 = *(u32 *)(r2 + 4)
|
|
; CHECK-NOT: r[[#]] = w[[#]]
|
|
; CHECK: exit
|