On 5/18/23 02:35, Juan Quintela wrote:
Richard Henderson <richard.hender...@linaro.org> wrote:
Add cpuinfo.h for i386 and x86_64, and the initialization
for that in util/. Populate that with a slightly altered
copy of the tcg host probing code. Other uses of cpuid.h
will be adjusted one patch at a time.
Signed-off-by: Richard Henderson <richard.hender...@linaro.org>
Reviewed-by: Juan Quintela <quint...@redhat.com>
For what is worth my vote O:-)
+#define CPUINFO_ALWAYS (1u << 0) /* so cpuinfo is nonzero */
+#define CPUINFO_CMOV (1u << 1)
+#define CPUINFO_MOVBE (1u << 2)
+#define CPUINFO_LZCNT (1u << 3)
+#define CPUINFO_POPCNT (1u << 4)
+#define CPUINFO_BMI1 (1u << 5)
+#define CPUINFO_BMI2 (1u << 6)
+#define CPUINFO_SSE2 (1u << 7)
+#define CPUINFO_SSE4 (1u << 8)
+#define CPUINFO_AVX1 (1u << 9)
+#define CPUINFO_AVX2 (1u << 10)
+#define CPUINFO_AVX512F (1u << 11)
+#define CPUINFO_AVX512VL (1u << 12)
+#define CPUINFO_AVX512BW (1u << 13)
+#define CPUINFO_AVX512DQ (1u << 14)
+#define CPUINFO_AVX512VBMI2 (1u << 15)
+#define CPUINFO_ATOMIC_VMOVDQA (1u << 16)
+
+/* Initialized with a constructor. */
+extern unsigned cpuinfo;
On one hand, it is weird having a flags variable that is only 32bit. I
am so user to put 64 bit flags. Future proof, blah, blah, ...
On the other hand, if tcg has survived for so long with only 16 bits, it
is inside posibility that 32bits are more than enough.
Indeed. Nor is this an public abi that needs future-proofing -- in the event we need more
bits, we change it.
+/* Called both as constructor and (possibly) via other constructors. */
+unsigned __attribute__((constructor)) cpuinfo_init(void)
+{
+ unsigned info = cpuinfo;
+
+ if (info) {
+ return info;
+ }
Have to look several times to this, because info "needed to be"" a
static variable, right? O:-)
:-)
r~
