https://llvm.org/bugs/show_bug.cgi?id=26452
Bug ID: 26452
Summary: LLVM emits divide by zero where there was none
Product: libraries
Version: trunk
Hardware: PC
OS: All
Status: NEW
Severity: normal
Priority: P
Component: Backend: WebAssembly
Assignee: unassignedb...@nondot.org
Reporter: j...@google.com
CC: llvm-bugs@lists.llvm.org
Classification: Unclassified
This test:
https://github.com/gcc-mirror/gcc/tree/master/gcc/testsuite/gcc.c-torture/execute/pr60960.c
Uses 4x vectors and scalarizes the operations. It doesn't divide by zero, yet
LLVM generates the following for f1:
f1: # @f1
.param i32, i32, i32, i32, i32
# BB#0: # %entry
i32.const $push0=, 255
i32.and $push3=, $2, $pop0
i32.const $push8=, 0
i32.div_u $2=, $pop3, $pop8
i32.const $push19=, 255
i32.and $push2=, $3, $pop19
i32.const $push18=, 0
i32.div_u $3=, $pop2, $pop18
i32.const $push10=, 3
i32.add $push11=, $0, $pop10
i32.const $push17=, 255
i32.and $push1=, $4, $pop17
i32.const $push16=, 0
i32.div_u $push9=, $pop1, $pop16
i32.store8 $discard=, 0($pop11):p2align=2, $pop9
i32.const $push12=, 2
i32.add $push13=, $0, $pop12
i32.store8 $discard=, 0($pop13):p2align=2, $3
i32.const $push6=, 1
i32.add $push14=, $0, $pop6
i32.store8 $discard=, 0($pop14):p2align=2, $2
i32.const $push4=, 254
i32.and $push5=, $1, $pop4
i32.const $push15=, 1
i32.shr_u $push7=, $pop5, $pop15
i32.store8 $discard=, 0($0):p2align=2, $pop7
return
.endfunc
.Lfunc_end0:
.size f1, .Lfunc_end0-f1
--
You are receiving this mail because:
You are on the CC list for the bug.
_______________________________________________ llvm-bugs mailing list llvm-bugs@lists.llvm.org http://lists.llvm.org/cgi-bin/mailman/listinfo/llvm-bugs
