arm64: arm_pmuv3: Correctly extract and check the PMUVer

Currently we're using "sbfx" to extract the PMUVer from ID_AA64DFR0_EL1
and skip the init/reset if no PMU present when the extracted PMUVer is
negative or is zero. However for PMUv3p8 the PMUVer will be 0b1000 and
PMUVer extracted by "sbfx" will always be negative and we'll skip the
init/reset in __init_el2_debug/reset_pmuserenr_el0 unexpectedly.

So this patch use "ubfx" instead of "sbfx" to extract the PMUVer. If
the PMUVer is implementation defined (0b1111) or not implemented(0b0000)
then skip the reset/init. Previously we'll also skip the init/reset
if the PMUVer is higher than the version we known (currently PMUv3p9),
with this patch we'll only skip if the PMU is not implemented or
implementation defined. This keeps consistence with how we probe
the PMU in the driver with pmuv3_implemented().
Signed-off-by: Yicong Yang <yangyicong@hisilicon.com>
Link: https://lore.kernel.org/r/20240411123030.7201-1-yangyicong@huawei.comSigned-off-by: Will Deacon <will@kernel.org>

arch/arm64/include/asm/assembler.h

@@ -480,9 +480,10 @@ alternative_endif
  */
 	.macro	reset_pmuserenr_el0, tmpreg
 	mrs	\tmpreg, id_aa64dfr0_el1
-	sbfx	\tmpreg, \tmpreg, #ID_AA64DFR0_EL1_PMUVer_SHIFT, #4
-	cmp	\tmpreg, #1			// Skip if no PMU present
-	b.lt	9000f
+	ubfx	\tmpreg, \tmpreg, #ID_AA64DFR0_EL1_PMUVer_SHIFT, #4
+	cmp	\tmpreg, #ID_AA64DFR0_EL1_PMUVer_NI
+	ccmp	\tmpreg, #ID_AA64DFR0_EL1_PMUVer_IMP_DEF, #4, ne
+	b.eq	9000f				// Skip if no PMU present or IMP_DEF
 	msr	pmuserenr_el0, xzr		// Disable PMU access from EL0
 9000:
 	.endm

arch/arm64/include/asm/el2_setup.h

@@ -59,13 +59,14 @@
 
 .macro __init_el2_debug
 	mrs	x1, id_aa64dfr0_el1
-	sbfx	x0, x1, #ID_AA64DFR0_EL1_PMUVer_SHIFT, #4
-	cmp	x0, #1
-	b.lt	.Lskip_pmu_\@			// Skip if no PMU present
+	ubfx	x0, x1, #ID_AA64DFR0_EL1_PMUVer_SHIFT, #4
+	cmp	x0, #ID_AA64DFR0_EL1_PMUVer_NI
+	ccmp	x0, #ID_AA64DFR0_EL1_PMUVer_IMP_DEF, #4, ne
+	b.eq	.Lskip_pmu_\@			// Skip if no PMU present or IMP_DEF
 	mrs	x0, pmcr_el0			// Disable debug access traps
 	ubfx	x0, x0, #11, #5			// to EL2 and allow access to
 .Lskip_pmu_\@:
-	csel	x2, xzr, x0, lt			// all PMU counters from EL1
+	csel	x2, xzr, x0, eq			// all PMU counters from EL1
 
 	/* Statistical profiling */
 	ubfx	x0, x1, #ID_AA64DFR0_EL1_PMSVer_SHIFT, #4