Merge tag 'arc-4.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc

Pull ARC updates from Vineet Gupta:

- more changes for HS48 cores: supporting MMUv5, detecting new
micro-arch gizmos

- axs10x platform wiring up reset driver merged in this cycle

- ARC perf driver optimizations

* tag 'arc-4.15-rc1' of git://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc:
ARC: perf: avoid vmalloc backed mmap
ARCv2: perf: optimize given that num counters <= 32
ARCv2: perf: tweak overflow interrupt
ARC: [plat-axs10x] DTS: Add reset controller node to manage ethernet reset
ARCv2: boot log: updates for HS48: dual-issue, ECC, Loop Buffer
ARCv2: Accomodate HS48 MMUv5 by relaxing MMU ver checking
ARC: [plat-axs10x] auto-select AXS101 or AXS103 given the ISA config

arch/arc/Kconfig

···
 	select OF
 	select OF_EARLY_FLATTREE
 	select OF_RESERVED_MEM
-	select PERF_USE_VMALLOC
+	select PERF_USE_VMALLOC if ARC_CACHE_VIPT_ALIASING
 	select HAVE_DEBUG_STACKOVERFLOW
 	select HAVE_GENERIC_DMA_COHERENT
 	select HAVE_KERNEL_GZIP

arch/arc/boot/dts/axs10x_mb.dtsi

···
 		ranges = <0x00000000 0x0 0xe0000000 0x10000000>;
 		interrupt-parent = <&mb_intc>;
 
+		creg_rst: reset-controller@11220 {
+			compatible = "snps,axs10x-reset";
+			#reset-cells = <1>;
+			reg = <0x11220 0x4>;
+		};
+
 		i2sclk: i2sclk@100a0 {
 			compatible = "snps,axs10x-i2s-pll-clock";
 			reg = <0x100a0 0x10>;
···
 			clocks = <&apbclk>;
 			clock-names = "stmmaceth";
 			max-speed = <100>;
+			resets = <&creg_rst 5>;
+			reset-names = "stmmaceth";
 		};
 
 		ehci@0x40000 {

arch/arc/include/asm/arcregs.h

···
 
 /* Build Configuration Registers */
 #define ARC_REG_AUX_DCCM	0x18	/* DCCM Base Addr ARCv2 */
+#define ARC_REG_ERP_CTRL	0x3F	/* ARCv2 Error protection control */
 #define ARC_REG_DCCM_BASE_BUILD	0x61	/* DCCM Base Addr ARCompact */
 #define ARC_REG_CRC_BCR		0x62
 #define ARC_REG_VECBASE_BCR	0x68
 #define ARC_REG_PERIBASE_BCR	0x69
 #define ARC_REG_FP_BCR		0x6B	/* ARCompact: Single-Precision FPU */
 #define ARC_REG_DPFP_BCR	0x6C	/* ARCompact: Dbl Precision FPU */
+#define ARC_REG_ERP_BUILD	0xc7	/* ARCv2 Error protection Build: ECC/Parity */
 #define ARC_REG_FP_V2_BCR	0xc8	/* ARCv2 FPU */
 #define ARC_REG_SLC_BCR		0xce
 #define ARC_REG_DCCM_BUILD	0x74	/* DCCM size (common) */
···
 #define ARC_REG_D_UNCACH_BCR	0x6A
 #define ARC_REG_BPU_BCR		0xc0
 #define ARC_REG_ISA_CFG_BCR	0xc1
+#define ARC_REG_LPB_BUILD	0xE9	/* ARCv2 Loop Buffer Build */
 #define ARC_REG_RTT_BCR		0xF2
 #define ARC_REG_IRQ_BCR		0xF3
+#define ARC_REG_MICRO_ARCH_BCR	0xF9	/* ARCv2 Product revision */
 #define ARC_REG_SMART_BCR	0xFF
 #define ARC_REG_CLUSTER_BCR	0xcf
 #define ARC_REG_AUX_ICCM	0x208	/* ICCM Base Addr (ARCv2) */
+#define ARC_REG_LPB_CTRL	0x488	/* ARCv2 Loop Buffer control */
 
 /* Common for ARCompact and ARCv2 status register */
 #define ARC_REG_STATUS32	0x0A
···
 #endif
 };
 
+/* Error Protection Build: ECC/Parity */
+struct bcr_erp {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int pad3:5, mmu:3, pad2:4, ic:3, dc:3, pad1:6, ver:8;
+#else
+	unsigned int ver:8, pad1:6, dc:3, ic:3, pad2:4, mmu:3, pad3:5;
+#endif
+};
+
+/* Error Protection Control */
+struct ctl_erp {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int pad2:27, mpd:1, pad1:2, dpd:1, dpi:1;
+#else
+	unsigned int dpi:1, dpd:1, pad1:2, mpd:1, pad2:27;
+#endif
+};
+
+struct bcr_lpb {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int pad:16, entries:8, ver:8;
+#else
+	unsigned int ver:8, entries:8, pad:16;
+#endif
+};
+
 struct bcr_generic {
 #ifdef CONFIG_CPU_BIG_ENDIAN
 	unsigned int info:24, ver:8;
···
 	struct cpuinfo_arc_ccm iccm, dccm;
 	struct {
 		unsigned int swap:1, norm:1, minmax:1, barrel:1, crc:1, swape:1, pad1:2,
-			     fpu_sp:1, fpu_dp:1, dual_iss_enb:1, dual_iss_exist:1, pad2:4,
+			     fpu_sp:1, fpu_dp:1, dual:1, dual_enb:1, pad2:4,
 			     debug:1, ap:1, smart:1, rtt:1, pad3:4,
 			     timer0:1, timer1:1, rtc:1, gfrc:1, pad4:4;
 	} extn;

arch/arc/kernel/perf_event.c

···
 	struct hw_perf_event *hwc = &event->hw;
 	int idx = hwc->idx;
 
-	if (__test_and_set_bit(idx, pmu_cpu->used_mask)) {
-		idx = find_first_zero_bit(pmu_cpu->used_mask,
-					  arc_pmu->n_counters);
-		if (idx == arc_pmu->n_counters)
-			return -EAGAIN;
+	idx = ffz(pmu_cpu->used_mask[0]);
+	if (idx == arc_pmu->n_counters)
+		return -EAGAIN;
 
-		__set_bit(idx, pmu_cpu->used_mask);
-		hwc->idx = idx;
-	}
+	__set_bit(idx, pmu_cpu->used_mask);
+	hwc->idx = idx;
 
 	write_aux_reg(ARC_REG_PCT_INDEX, idx);
 
···
 	struct perf_sample_data data;
 	struct arc_pmu_cpu *pmu_cpu = this_cpu_ptr(&arc_pmu_cpu);
 	struct pt_regs *regs;
-	int active_ints;
+	unsigned int active_ints;
 	int idx;
 
 	arc_pmu_disable(&arc_pmu->pmu);
 
 	active_ints = read_aux_reg(ARC_REG_PCT_INT_ACT);
+	if (!active_ints)
+		goto done;
 
 	regs = get_irq_regs();
 
-	for (idx = 0; idx < arc_pmu->n_counters; idx++) {
-		struct perf_event *event = pmu_cpu->act_counter[idx];
+	do {
+		struct perf_event *event;
 		struct hw_perf_event *hwc;
 
-		if (!(active_ints & (1 << idx)))
-			continue;
+		idx = __ffs(active_ints);
 
 		/* Reset interrupt flag by writing of 1 */
 		write_aux_reg(ARC_REG_PCT_INT_ACT, 1 << idx);
···
 		write_aux_reg(ARC_REG_PCT_INT_CTRL,
 			read_aux_reg(ARC_REG_PCT_INT_CTRL) | (1 << idx));
 
+		event = pmu_cpu->act_counter[idx];
 		hwc = &event->hw;
 
 		WARN_ON_ONCE(hwc->idx != idx);
 
 		arc_perf_event_update(event, &event->hw, event->hw.idx);
 		perf_sample_data_init(&data, 0, hwc->last_period);
-		if (!arc_pmu_event_set_period(event))
-			continue;
+		if (arc_pmu_event_set_period(event)) {
+			if (perf_event_overflow(event, &data, regs))
+				arc_pmu_stop(event, 0);
+		}
 
-		if (perf_event_overflow(event, &data, regs))
-			arc_pmu_stop(event, 0);
-	}
+		active_ints &= ~(1U << idx);
+	} while (active_ints);
 
+done:
 	arc_pmu_enable(&arc_pmu->pmu);
 
 	return IRQ_HANDLED;
···
 		pr_err("This core does not have performance counters!\n");
 		return -ENODEV;
 	}
+	BUILD_BUG_ON(ARC_PERF_MAX_COUNTERS > 32);
 	BUG_ON(pct_bcr.c > ARC_PERF_MAX_COUNTERS);
 
 	READ_BCR(ARC_REG_CC_BUILD, cc_bcr);

arch/arc/kernel/setup.c

···
 			unsigned int exec_ctrl;
 
 			READ_BCR(AUX_EXEC_CTRL, exec_ctrl);
-			cpu->extn.dual_iss_exist = 1;
-			cpu->extn.dual_iss_enb = exec_ctrl & 1;
+			cpu->extn.dual_enb = exec_ctrl & 1;
+
+			/* dual issue always present for this core */
+			cpu->extn.dual = 1;
 		}
 	}
 
···
 		       cpu_id, cpu->name, cpu->details,
 		       is_isa_arcompact() ? "ARCompact" : "ARCv2",
 		       IS_AVAIL1(cpu->isa.be, "[Big-Endian]"),
-		       IS_AVAIL3(cpu->extn.dual_iss_exist, cpu->extn.dual_iss_enb, " Dual-Issue"));
+		       IS_AVAIL3(cpu->extn.dual, cpu->extn.dual_enb, " Dual-Issue "));
 
 	n += scnprintf(buf + n, len - n, "Timers\t\t: %s%s%s%s%s%s\nISA Extn\t: ",
 		       IS_AVAIL1(cpu->extn.timer0, "Timer0 "),
···
 
 	if (cpu->bpu.ver)
 		n += scnprintf(buf + n, len - n,
-			      "BPU\t\t: %s%s match, cache:%d, Predict Table:%d\n",
+			      "BPU\t\t: %s%s match, cache:%d, Predict Table:%d",
 			      IS_AVAIL1(cpu->bpu.full, "full"),
 			      IS_AVAIL1(!cpu->bpu.full, "partial"),
 			      cpu->bpu.num_cache, cpu->bpu.num_pred);
 
+	if (is_isa_arcv2()) {
+		struct bcr_lpb lpb;
+
+		READ_BCR(ARC_REG_LPB_BUILD, lpb);
+		if (lpb.ver) {
+			unsigned int ctl;
+			ctl = read_aux_reg(ARC_REG_LPB_CTRL);
+
+			n += scnprintf(buf + n, len - n, " Loop Buffer:%d %s",
+				lpb.entries,
+				IS_DISABLED_RUN(!ctl));
+		}
+	}
+
+	n += scnprintf(buf + n, len - n, "\n");
 	return buf;
 }
 
···
 		n += scnprintf(buf + n, len - n, "Extn [CCM]\t: DCCM @ %x, %d KB / ICCM: @ %x, %d KB\n",
 			       cpu->dccm.base_addr, TO_KB(cpu->dccm.sz),
 			       cpu->iccm.base_addr, TO_KB(cpu->iccm.sz));
+
+	if (is_isa_arcv2()) {
+
+		/* Error Protection: ECC/Parity */
+		struct bcr_erp erp;
+		READ_BCR(ARC_REG_ERP_BUILD, erp);
+
+		if (erp.ver) {
+			struct  ctl_erp ctl;
+			READ_BCR(ARC_REG_ERP_CTRL, ctl);
+
+			/* inverted bits: 0 means enabled */
+			n += scnprintf(buf + n, len - n, "Extn [ECC]\t: %s%s%s%s%s%s\n",
+				IS_AVAIL3(erp.ic,  !ctl.dpi, "IC "),
+				IS_AVAIL3(erp.dc,  !ctl.dpd, "DC "),
+				IS_AVAIL3(erp.mmu, !ctl.mpd, "MMU "));
+		}
+	}
 
 	n += scnprintf(buf + n, len - n, "OS ABI [v%d]\t: %s\n",
 			EF_ARC_OSABI_CURRENT >> 8,

arch/arc/mm/tlb.c

···
 	tmp = read_aux_reg(ARC_REG_MMU_BCR);
 	mmu->ver = (tmp >> 24);
 
-	if (mmu->ver <= 2) {
-		mmu2 = (struct bcr_mmu_1_2 *)&tmp;
-		mmu->pg_sz_k = TO_KB(0x2000);
-		mmu->sets = 1 << mmu2->sets;
-		mmu->ways = 1 << mmu2->ways;
-		mmu->u_dtlb = mmu2->u_dtlb;
-		mmu->u_itlb = mmu2->u_itlb;
-	} else if (mmu->ver == 3) {
-		mmu3 = (struct bcr_mmu_3 *)&tmp;
-		mmu->pg_sz_k = 1 << (mmu3->pg_sz - 1);
-		mmu->sets = 1 << mmu3->sets;
-		mmu->ways = 1 << mmu3->ways;
-		mmu->u_dtlb = mmu3->u_dtlb;
-		mmu->u_itlb = mmu3->u_itlb;
-		mmu->sasid = mmu3->sasid;
+	if (is_isa_arcompact()) {
+		if (mmu->ver <= 2) {
+			mmu2 = (struct bcr_mmu_1_2 *)&tmp;
+			mmu->pg_sz_k = TO_KB(0x2000);
+			mmu->sets = 1 << mmu2->sets;
+			mmu->ways = 1 << mmu2->ways;
+			mmu->u_dtlb = mmu2->u_dtlb;
+			mmu->u_itlb = mmu2->u_itlb;
+		} else {
+			mmu3 = (struct bcr_mmu_3 *)&tmp;
+			mmu->pg_sz_k = 1 << (mmu3->pg_sz - 1);
+			mmu->sets = 1 << mmu3->sets;
+			mmu->ways = 1 << mmu3->ways;
+			mmu->u_dtlb = mmu3->u_dtlb;
+			mmu->u_itlb = mmu3->u_itlb;
+			mmu->sasid = mmu3->sasid;
+		}
 	} else {
 		mmu4 = (struct bcr_mmu_4 *)&tmp;
 		mmu->pg_sz_k = 1 << (mmu4->sz0 - 1);
···
 
 void arc_mmu_init(void)
 {
-	char str[256];
 	struct cpuinfo_arc_mmu *mmu = &cpuinfo_arc700[smp_processor_id()].mmu;
+	char str[256];
+	int compat = 0;
 
 	pr_info("%s", arc_mmu_mumbojumbo(0, str, sizeof(str)));
 
···
 	 */
 	BUILD_BUG_ON(!IS_ALIGNED(STACK_TOP, PMD_SIZE));
 
-	/* For efficiency sake, kernel is compile time built for a MMU ver
-	 * This must match the hardware it is running on.
-	 * Linux built for MMU V2, if run on MMU V1 will break down because V1
-	 *  hardware doesn't understand cmds such as WriteNI, or IVUTLB
-	 * On the other hand, Linux built for V1 if run on MMU V2 will do
-	 *   un-needed workarounds to prevent memcpy thrashing.
-	 * Similarly MMU V3 has new features which won't work on older MMU
+	/*
+	 * Ensure that MMU features assumed by kernel exist in hardware.
+	 * For older ARC700 cpus, it has to be exact match, since the MMU
+	 * revisions were not backwards compatible (MMUv3 TLB layout changed
+	 * so even if kernel for v2 didn't use any new cmds of v3, it would
+	 * still not work.
+	 * For HS cpus, MMUv4 was baseline and v5 is backwards compatible
+	 * (will run older software).
 	 */
-	if (mmu->ver != CONFIG_ARC_MMU_VER) {
+	if (is_isa_arcompact() && mmu->ver == CONFIG_ARC_MMU_VER)
+		compat = 1;
+	else if (is_isa_arcv2() && mmu->ver >= CONFIG_ARC_MMU_VER)
+		compat = 1;
+
+	if (!compat) {
 		panic("MMU ver %d doesn't match kernel built for %d...\n",
 		      mmu->ver, CONFIG_ARC_MMU_VER);
 	}

arch/arc/plat-axs10x/Kconfig

···
 	select MIGHT_HAVE_PCI
 	select GENERIC_IRQ_CHIP
 	select GPIOLIB
+	select AXS101 if ISA_ARCOMPACT
+	select AXS103 if ISA_ARCV2
 	help
 	  Support for the ARC AXS10x Software Development Platforms.
 

arch/arc/plat-axs10x/axs10x.c

···
 
 	axs10x_enable_gpio_intc_wire();
 
-	/*
-	 * Reset ethernet IP core.
-	 * TODO: get rid of this quirk after axs10x reset driver (or simple
-	 * reset driver) will be available in upstream.
-	 */
-	iowrite32((1 << 5), (void __iomem *) CREG_MB_SW_RESET);
-
 	scnprintf(mb, 32, "MainBoard v%d", mb_rev);
 	axs10x_print_board_ver(CREG_MB_VER, mb);
 }