https://gcc.gnu.org/bugzilla/show_bug.cgi?id=124019
Bug ID: 124019
Summary: Improve subword arithmetic and comparisons
Product: gcc
Version: unknown
Status: UNCONFIRMED
Severity: normal
Priority: P3
Component: rtl-optimization
Assignee: unassigned at gcc dot gnu.org
Reporter: law at gcc dot gnu.org
Target Milestone: ---
Consider this code on rv64gcbv_zicond:
typedef unsigned char uint8_t;
typedef signed char int8_t;
bool f1(int8_t a)
{
int8_t b = a << 4;
return b == -128;
}
Generates something like this:
slli a0,a0,4 # 6 [c=4 l=4] ashldi3
sext.b a0,a0 # 9 [c=4 l=4] *extendqidi2_bitmanip/0
addi a0,a0,128 # 11 [c=4 l=4] *adddi3/1
seqz a0,a0 # 20 [c=4 l=4] *seq_zero_didi
What would be better would be to use andi to mask off all but the low 4 bits
then use addi+seqz to check of the low 4 bits are precisely 0x8.
