+OPENSSL_ia32_cpuid:
+ mov %rbx,%r8
+
+ xor %eax,%eax
+ cpuid
+ mov %eax,%r11d # max value for standard query level
+
+ xor %eax,%eax
+ cmp \$0x756e6547,%ebx # "Genu"
+ setne %al
+ mov %eax,%r9d
+ cmp \$0x49656e69,%edx # "ineI"
+ setne %al
+ or %eax,%r9d
+ cmp \$0x6c65746e,%ecx # "ntel"
+ setne %al
+ or %eax,%r9d # 0 indicates Intel CPU
+ jz .Lintel
+
+ cmp \$0x68747541,%ebx # "Auth"
+ setne %al
+ mov %eax,%r10d
+ cmp \$0x69746E65,%edx # "enti"
+ setne %al
+ or %eax,%r10d
+ cmp \$0x444D4163,%ecx # "cAMD"
+ setne %al
+ or %eax,%r10d # 0 indicates AMD CPU
+ jnz .Lintel
+
+ # AMD specific
+ mov \$0x80000000,%eax
+ cpuid
+ cmp \$0x80000008,%eax
+ jb .Lintel
+
+ mov \$0x80000008,%eax
+ cpuid
+ movzb %cl,%r10 # number of cores - 1
+ inc %r10 # number of cores
+
+ mov \$1,%eax
+ cpuid
+ bt \$28,%edx # test hyper-threading bit
+ jnc .Ldone
+ shr \$16,%ebx # number of logical processors
+ cmp %r10b,%bl
+ ja .Ldone
+ and \$0xefffffff,%edx # ~(1<<28)
+ jmp .Ldone
+
+.Lintel:
+ cmp \$4,%r11d
+ mov \$-1,%r10d
+ jb .Lnocacheinfo
+
+ mov \$4,%eax
+ mov \$0,%ecx # query L1D
+ cpuid
+ mov %eax,%r10d
+ shr \$14,%r10d
+ and \$0xfff,%r10d # number of cores -1 per L1D
+
+.Lnocacheinfo:
+ mov \$1,%eax
+ cpuid
+ cmp \$0,%r9d
+ jne .Lnotintel
+ or \$0x00100000,%edx # use reserved 20th bit to engage RC4_CHAR
+ and \$15,%ah
+ cmp \$15,%ah # examine Family ID
+ je .Lnotintel
+ or \$0x40000000,%edx # use reserved bit to skip unrolled loop
+.Lnotintel:
+ bt \$28,%edx # test hyper-threading bit
+ jnc .Ldone
+ and \$0xefffffff,%edx # ~(1<<28)
+ cmp \$0,%r10d
+ je .Ldone
+
+ or \$0x10000000,%edx # 1<<28
+ shr \$16,%ebx
+ cmp \$1,%bl # see if cache is shared
+ ja .Ldone
+ and \$0xefffffff,%edx # ~(1<<28)
+.Ldone:
+ shl \$32,%rcx
+ mov %edx,%eax
+ mov %r8,%rbx
+ or %rcx,%rax