From 4d3696801bad2a037832c15a8d21dfe0c529d9cd Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Sat, 12 Aug 2023 18:23:59 -0700
Subject: ARC: -Wmissing-prototype warning fixes

Anrd reported [1] new compiler warnings due to -Wmissing-protype.
These are for non static functions mostly used in asm code hence not
exported already. Fix this by adding the prototypes.

[1] https://lore.kernel.org/lkml/20230810141947.1236730-1-arnd@kernel.org

Reviewed-by: Arnd Bergmann <arnd@arndb.de>
Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/ctx_sw.c     | 2 +-
 arch/arc/kernel/devtree.c    | 1 +
 arch/arc/kernel/intc-arcv2.c | 2 +-
 arch/arc/kernel/signal.c     | 1 +
 arch/arc/kernel/smp.c        | 7 ++++---
 arch/arc/kernel/stacktrace.c | 1 +
 arch/arc/kernel/traps.c      | 1 +
 7 files changed, 10 insertions(+), 5 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/ctx_sw.c b/arch/arc/kernel/ctx_sw.c
index 1a76f2d6f694..bf16f777a0bc 100644
--- a/arch/arc/kernel/ctx_sw.c
+++ b/arch/arc/kernel/ctx_sw.c
@@ -12,7 +12,7 @@
  */
 
 #include <asm/asm-offsets.h>
-#include <linux/sched.h>
+#include <asm/switch_to.h>
 #include <linux/sched/debug.h>
 
 #define KSP_WORD_OFF 	((TASK_THREAD + THREAD_KSP) / 4)
diff --git a/arch/arc/kernel/devtree.c b/arch/arc/kernel/devtree.c
index 721d465f1580..4c9e61457b2f 100644
--- a/arch/arc/kernel/devtree.c
+++ b/arch/arc/kernel/devtree.c
@@ -12,6 +12,7 @@
 #include <linux/of.h>
 #include <linux/of_fdt.h>
 #include <asm/mach_desc.h>
+#include <asm/serial.h>
 
 #ifdef CONFIG_SERIAL_EARLYCON
 
diff --git a/arch/arc/kernel/intc-arcv2.c b/arch/arc/kernel/intc-arcv2.c
index 5cda19d0aa91..678898757e47 100644
--- a/arch/arc/kernel/intc-arcv2.c
+++ b/arch/arc/kernel/intc-arcv2.c
@@ -108,7 +108,7 @@ static void arcv2_irq_unmask(struct irq_data *data)
 	write_aux_reg(AUX_IRQ_ENABLE, 1);
 }
 
-void arcv2_irq_enable(struct irq_data *data)
+static void arcv2_irq_enable(struct irq_data *data)
 {
 	/* set default priority */
 	write_aux_reg(AUX_IRQ_SELECT, data->hwirq);
diff --git a/arch/arc/kernel/signal.c b/arch/arc/kernel/signal.c
index 3c1590c27fae..0b3bb529d246 100644
--- a/arch/arc/kernel/signal.c
+++ b/arch/arc/kernel/signal.c
@@ -53,6 +53,7 @@
 #include <linux/sched/task_stack.h>
 
 #include <asm/ucontext.h>
+#include <asm/entry.h>
 
 struct rt_sigframe {
 	struct siginfo info;
diff --git a/arch/arc/kernel/smp.c b/arch/arc/kernel/smp.c
index 409cfa4675b4..8d9b188caa27 100644
--- a/arch/arc/kernel/smp.c
+++ b/arch/arc/kernel/smp.c
@@ -23,9 +23,10 @@
 #include <linux/export.h>
 #include <linux/of_fdt.h>
 
-#include <asm/processor.h>
-#include <asm/setup.h>
 #include <asm/mach_desc.h>
+#include <asm/setup.h>
+#include <asm/smp.h>
+#include <asm/processor.h>
 
 #ifndef CONFIG_ARC_HAS_LLSC
 arch_spinlock_t smp_atomic_ops_lock = __ARCH_SPIN_LOCK_UNLOCKED;
@@ -351,7 +352,7 @@ static inline int __do_IPI(unsigned long msg)
  * arch-common ISR to handle for inter-processor interrupts
  * Has hooks for platform specific IPI
  */
-irqreturn_t do_IPI(int irq, void *dev_id)
+static irqreturn_t do_IPI(int irq, void *dev_id)
 {
 	unsigned long pending;
 	unsigned long __maybe_unused copy;
diff --git a/arch/arc/kernel/stacktrace.c b/arch/arc/kernel/stacktrace.c
index 5372dc04e784..ea99c066ef25 100644
--- a/arch/arc/kernel/stacktrace.c
+++ b/arch/arc/kernel/stacktrace.c
@@ -29,6 +29,7 @@
 
 #include <asm/arcregs.h>
 #include <asm/unwind.h>
+#include <asm/stacktrace.h>
 #include <asm/switch_to.h>
 
 /*-------------------------------------------------------------------------
diff --git a/arch/arc/kernel/traps.c b/arch/arc/kernel/traps.c
index 6b83e3f2b41c..2f7eb786695b 100644
--- a/arch/arc/kernel/traps.c
+++ b/arch/arc/kernel/traps.c
@@ -16,6 +16,7 @@
 #include <linux/ptrace.h>
 #include <linux/kprobes.h>
 #include <linux/kgdb.h>
+#include <asm/entry.h>
 #include <asm/setup.h>
 #include <asm/unaligned.h>
 #include <asm/kprobes.h>
-- 
cgit v1.2.3


From 72d861f2d22792eddc91d4617e410ec74db8d814 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Wed, 10 Jun 2020 23:03:22 -0700
Subject: ARC: boot log: eliminate struct cpuinfo_arc #1: mm

This is first step in eliminating struct cpuinfo_arc[NR_CPUS]

Back when we had just ARCompact ISA, the idea was to read/bit-fiddle
the BCRs once and and cache decoded information in a global struct ready
to use.

With ARCv2 it was modified to contained abstract / ISA agnostic
information.

However with ARCv3 there 's too much disparity to abstract in common
structures. So drop the entire decode once and store paradigm. Afterall
there's only 2 users of this machinery anyways:  boot printing and
cat /proc/cpuinfo. None is performance critical to warrant locking away
resident memory per cpu.

This patch is first step in that direction
 - decouples struct cpuinfo_arc_mmu from global struct cpuinfo_arc
 - mmu code still has a trimmed down static version of
   struct cpuinfo_arc_mmu to cache information needed in performance
   critical code such as tlb flush routines
 - folds read_decode_mmu_bcr() into arc_mmu_mumbojumbo()
 - setup_processor() directly calls arc_mmu_init() and not via
   arc_cpu_init()

Tested-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202308151213.qKZPMiyz-lkp@intel.com/
Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/setup.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/setup.c b/arch/arc/kernel/setup.c
index 41f07b3e594e..094461540f8a 100644
--- a/arch/arc/kernel/setup.c
+++ b/arch/arc/kernel/setup.c
@@ -186,7 +186,6 @@ static void read_arc_build_cfg_regs(void)
 	/* Read CCM BCRs for boot reporting even if not enabled in Kconfig */
 	read_decode_ccm_bcr(cpu);
 
-	read_decode_mmu_bcr();
 	read_decode_cache_bcr();
 
 	if (is_isa_arcompact()) {
@@ -256,7 +255,7 @@ static void read_arc_build_cfg_regs(void)
 		cpu->isa.be = IS_ENABLED(CONFIG_CPU_BIG_ENDIAN);
 
 		 /* there's no direct way to distinguish 750 vs. 770 */
-		if (unlikely(cpu->core.family < 0x34 || cpu->mmu.ver < 3))
+		if (unlikely(cpu->core.family < 0x34))
 			cpu->name = "ARC750";
 	} else {
 		cpu->isa = isa;
@@ -463,6 +462,7 @@ void setup_processor(void)
 	arc_init_IRQ();
 
 	pr_info("%s", arc_cpu_mumbojumbo(cpu_id, str, sizeof(str)));
+	pr_info("%s", arc_mmu_mumbojumbo(cpu_id, str, sizeof(str)));
 
 	arc_mmu_init();
 	arc_cache_init();
-- 
cgit v1.2.3


From 17a5ed563aaf0d66bf0438ee2df016f636846a84 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Wed, 10 Jun 2020 23:44:59 -0700
Subject: ARC: boot log: eliminate struct cpuinfo_arc #2: cache

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/setup.c | 3 +--
 1 file changed, 1 insertion(+), 2 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/setup.c b/arch/arc/kernel/setup.c
index 094461540f8a..3ea834941c1f 100644
--- a/arch/arc/kernel/setup.c
+++ b/arch/arc/kernel/setup.c
@@ -186,8 +186,6 @@ static void read_arc_build_cfg_regs(void)
 	/* Read CCM BCRs for boot reporting even if not enabled in Kconfig */
 	read_decode_ccm_bcr(cpu);
 
-	read_decode_cache_bcr();
-
 	if (is_isa_arcompact()) {
 		struct bcr_fp_arcompact sp, dp;
 		struct bcr_bpu_arcompact bpu;
@@ -463,6 +461,7 @@ void setup_processor(void)
 
 	pr_info("%s", arc_cpu_mumbojumbo(cpu_id, str, sizeof(str)));
 	pr_info("%s", arc_mmu_mumbojumbo(cpu_id, str, sizeof(str)));
+	pr_info("%s", arc_cache_mumbojumbo(cpu_id, str, sizeof(str)));
 
 	arc_mmu_init();
 	arc_cache_init();
-- 
cgit v1.2.3


From c5b678b379e7772d553f1b4ec052420d25bf9c7a Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Fri, 12 Jun 2020 15:30:47 -0700
Subject: ARC: boot log: eliminate struct cpuinfo_arc #3: don't export

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/mcip.c | 2 --
 1 file changed, 2 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/mcip.c b/arch/arc/kernel/mcip.c
index f9fdb557c263..55373ca0d28b 100644
--- a/arch/arc/kernel/mcip.c
+++ b/arch/arc/kernel/mcip.c
@@ -165,8 +165,6 @@ static void mcip_probe_n_setup(void)
 		IS_AVAIL1(mp.idu, "IDU "),
 		IS_AVAIL1(mp.dbg, "DEBUG "),
 		IS_AVAIL1(mp.gfrc, "GFRC"));
-
-	cpuinfo_arc700[0].extn.gfrc = mp.gfrc;
 }
 
 struct plat_smp_ops plat_smp_ops = {
-- 
cgit v1.2.3


From fad84e39f116035ae8d550c6020107b8ac113b45 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Thu, 11 Jun 2020 11:08:45 -0700
Subject: ARC: boot log: eliminate struct cpuinfo_arc #4: boot log per ISA

 - boot log now clearly per ISA
 - global struct cpuinfo_arc[] elimiated
 - local struct struct arcinfo kept for passing info
   between functions

Tested-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202308162101.Ve5jBg80-lkp@intel.com
Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/setup.c | 560 ++++++++++++++++++++++--------------------------
 1 file changed, 257 insertions(+), 303 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/setup.c b/arch/arc/kernel/setup.c
index 3ea834941c1f..0aa49308d792 100644
--- a/arch/arc/kernel/setup.c
+++ b/arch/arc/kernel/setup.c
@@ -29,6 +29,7 @@
 #include <asm/mach_desc.h>
 #include <asm/smp.h>
 #include <asm/dsp-impl.h>
+#include <soc/arc/mcip.h>
 
 #define FIX_PTR(x)  __asm__ __volatile__(";" : "+r"(x))
 
@@ -43,19 +44,22 @@ const struct machine_desc *machine_desc;
 
 struct task_struct *_current_task[NR_CPUS];	/* For stack switching */
 
-struct cpuinfo_arc cpuinfo_arc700[NR_CPUS];
+struct cpuinfo_arc {
+	int arcver;
+	unsigned int t0:1, t1:1;
+	struct {
+		unsigned long base;
+		unsigned int sz;
+	} iccm, dccm;
+};
+
+#ifdef CONFIG_ISA_ARCV2
 
-static const struct id_to_str arc_legacy_rel[] = {
+static const struct id_to_str arc_hs_rel[] = {
 	/* ID.ARCVER,	Release */
-#ifdef CONFIG_ISA_ARCOMPACT
-	{ 0x34, 	"R4.10"},
-	{ 0x35, 	"R4.11"},
-#else
 	{ 0x51, 	"R2.0" },
 	{ 0x52, 	"R2.1" },
 	{ 0x53,		"R3.0" },
-#endif
-	{ 0x00,		NULL   }
 };
 
 static const struct id_to_str arc_hs_ver54_rel[] = {
@@ -66,320 +70,294 @@ static const struct id_to_str arc_hs_ver54_rel[] = {
 	{  3,		"R4.00a"},
 	{  0xFF,	NULL   }
 };
+#endif
 
-static void read_decode_ccm_bcr(struct cpuinfo_arc *cpu)
+static int
+arcompact_mumbojumbo(int c, struct cpuinfo_arc *info, char *buf, int len)
 {
-	if (is_isa_arcompact()) {
-		struct bcr_iccm_arcompact iccm;
-		struct bcr_dccm_arcompact dccm;
+	int n = 0;
+#ifdef CONFIG_ISA_ARCOMPACT
+	char *cpu_nm, *isa_nm = "ARCompact";
+	struct bcr_fp_arcompact fpu_sp, fpu_dp;
+	int atomic = 0, be, present;
+	int bpu_full, bpu_cache, bpu_pred;
+	struct bcr_bpu_arcompact bpu;
+	struct bcr_iccm_arcompact iccm;
+	struct bcr_dccm_arcompact dccm;
+	struct bcr_generic isa;
 
-		READ_BCR(ARC_REG_ICCM_BUILD, iccm);
-		if (iccm.ver) {
-			cpu->iccm.sz = 4096 << iccm.sz;	/* 8K to 512K */
-			cpu->iccm.base_addr = iccm.base << 16;
-		}
+	READ_BCR(ARC_REG_ISA_CFG_BCR, isa);
 
-		READ_BCR(ARC_REG_DCCM_BUILD, dccm);
-		if (dccm.ver) {
-			unsigned long base;
-			cpu->dccm.sz = 2048 << dccm.sz;	/* 2K to 256K */
+	if (!isa.ver)	/* ISA BCR absent, use Kconfig info */
+		atomic = IS_ENABLED(CONFIG_ARC_HAS_LLSC);
+	else {
+		/* ARC700_BUILD only has 2 bits of isa info */
+		atomic = isa.info & 1;
+	}
 
-			base = read_aux_reg(ARC_REG_DCCM_BASE_BUILD);
-			cpu->dccm.base_addr = base & ~0xF;
-		}
-	} else {
-		struct bcr_iccm_arcv2 iccm;
-		struct bcr_dccm_arcv2 dccm;
-		unsigned long region;
-
-		READ_BCR(ARC_REG_ICCM_BUILD, iccm);
-		if (iccm.ver) {
-			cpu->iccm.sz = 256 << iccm.sz00;	/* 512B to 16M */
-			if (iccm.sz00 == 0xF && iccm.sz01 > 0)
-				cpu->iccm.sz <<= iccm.sz01;
-
-			region = read_aux_reg(ARC_REG_AUX_ICCM);
-			cpu->iccm.base_addr = region & 0xF0000000;
-		}
+	be = IS_ENABLED(CONFIG_CPU_BIG_ENDIAN);
 
-		READ_BCR(ARC_REG_DCCM_BUILD, dccm);
-		if (dccm.ver) {
-			cpu->dccm.sz = 256 << dccm.sz0;
-			if (dccm.sz0 == 0xF && dccm.sz1 > 0)
-				cpu->dccm.sz <<= dccm.sz1;
+	if (info->arcver < 0x34)
+		cpu_nm = "ARC750";
+	else
+		cpu_nm = "ARC770";
 
-			region = read_aux_reg(ARC_REG_AUX_DCCM);
-			cpu->dccm.base_addr = region & 0xF0000000;
-		}
-	}
-}
+	n += scnprintf(buf + n, len - n, "processor [%d]\t: %s (%s ISA) %s\n",
+		       c, cpu_nm, isa_nm, IS_AVAIL1(be, "[Big-Endian]"));
 
-static void decode_arc_core(struct cpuinfo_arc *cpu)
-{
-	struct bcr_uarch_build_arcv2 uarch;
-	const struct id_to_str *tbl;
-
-	if (cpu->core.family < 0x54) { /* includes arc700 */
+	READ_BCR(ARC_REG_FP_BCR, fpu_sp);
+	READ_BCR(ARC_REG_DPFP_BCR, fpu_dp);
 
-		for (tbl = &arc_legacy_rel[0]; tbl->id != 0; tbl++) {
-			if (cpu->core.family == tbl->id) {
-				cpu->release = tbl->str;
-				break;
-			}
-		}
+	if (fpu_sp.ver | fpu_dp.ver)
+		n += scnprintf(buf + n, len - n, "FPU\t\t: %s%s\n",
+			       IS_AVAIL1(fpu_sp.ver, "SP "),
+			       IS_AVAIL1(fpu_dp.ver, "DP "));
 
-		if (is_isa_arcompact())
-			cpu->name = "ARC700";
-		else if (tbl->str)
-			cpu->name = "HS38";
-		else
-			cpu->name = cpu->release = "Unknown";
+	READ_BCR(ARC_REG_BPU_BCR, bpu);
+	bpu_full = bpu.fam ? 1 : 0;
+	bpu_cache = 256 << (bpu.ent - 1);
+	bpu_pred = 256 << (bpu.ent - 1);
 
-		return;
+	n += scnprintf(buf + n, len - n,
+			"BPU\t\t: %s%s match, cache:%d, Predict Table:%d\n",
+			IS_AVAIL1(bpu_full, "full"),
+			IS_AVAIL1(!bpu_full, "partial"),
+			bpu_cache, bpu_pred);
+
+	READ_BCR(ARC_REG_ICCM_BUILD, iccm);
+	if (iccm.ver) {
+		info->iccm.sz = 4096 << iccm.sz;	/* 8K to 512K */
+		info->iccm.base = iccm.base << 16;
 	}
 
-	/*
-	 * Initial HS cores bumped AUX IDENTITY.ARCVER for each release until
-	 * ARCVER 0x54 which introduced AUX MICRO_ARCH_BUILD and subsequent
-	 * releases only update it.
-	 */
-	READ_BCR(ARC_REG_MICRO_ARCH_BCR, uarch);
-
-	if (uarch.prod == 4) {
-		cpu->name = "HS48";
-		cpu->extn.dual = 1;
+	READ_BCR(ARC_REG_DCCM_BUILD, dccm);
+	if (dccm.ver) {
+		unsigned long base;
+		info->dccm.sz = 2048 << dccm.sz;	/* 2K to 256K */
 
-	} else {
-		cpu->name = "HS38";
+		base = read_aux_reg(ARC_REG_DCCM_BASE_BUILD);
+		info->dccm.base = base & ~0xF;
 	}
 
-	for (tbl = &arc_hs_ver54_rel[0]; tbl->id != 0xFF; tbl++) {
-		if (uarch.maj == tbl->id) {
-			cpu->release = tbl->str;
-			break;
-		}
-	}
+	/* ARCompact ISA specific sanity checks */
+	present = fpu_dp.ver;	/* SP has no arch visible regs */
+	CHK_OPT_STRICT(CONFIG_ARC_FPU_SAVE_RESTORE, present);
+#endif
+	return n;
+
 }
 
-static void read_arc_build_cfg_regs(void)
+static int arcv2_mumbojumbo(int c, struct cpuinfo_arc *info, char *buf, int len)
 {
-	struct bcr_timer timer;
-	struct bcr_generic bcr;
-	struct cpuinfo_arc *cpu = &cpuinfo_arc700[smp_processor_id()];
+	int n = 0;
+#ifdef CONFIG_ISA_ARCV2
+	const char *release, *cpu_nm, *isa_nm = "ARCv2";
+	int dual_issue = 0, dual_enb = 0, mpy_opt, present;
+	int bpu_full, bpu_cache, bpu_pred, bpu_ret_stk;
+	char mpy_nm[16], lpb_nm[32];
 	struct bcr_isa_arcv2 isa;
-	struct bcr_actionpoint ap;
-
-	FIX_PTR(cpu);
-
-	READ_BCR(AUX_IDENTITY, cpu->core);
-	decode_arc_core(cpu);
+	struct bcr_mpy mpy;
+	struct bcr_fp_arcv2 fpu;
+	struct bcr_bpu_arcv2 bpu;
+	struct bcr_lpb lpb;
+	struct bcr_iccm_arcv2 iccm;
+	struct bcr_dccm_arcv2 dccm;
+	struct bcr_erp erp;
 
-	READ_BCR(ARC_REG_TIMERS_BCR, timer);
-	cpu->extn.timer0 = timer.t0;
-	cpu->extn.timer1 = timer.t1;
-	cpu->extn.rtc = timer.rtc;
+	/*
+	 * Initial HS cores bumped AUX IDENTITY.ARCVER for each release until
+	 * ARCVER 0x54 which introduced AUX MICRO_ARCH_BUILD and subsequent
+	 * releases only update it.
+	 */
 
-	cpu->vec_base = read_aux_reg(AUX_INTR_VEC_BASE);
+	cpu_nm = "HS38";
 
-	READ_BCR(ARC_REG_MUL_BCR, cpu->extn_mpy);
+	if (info->arcver > 0x50 && info->arcver <= 0x53) {
+		release = arc_hs_rel[info->arcver - 0x51].str;
+	} else {
+		const struct id_to_str *tbl;
+		struct bcr_uarch_build uarch;
 
-	/* Read CCM BCRs for boot reporting even if not enabled in Kconfig */
-	read_decode_ccm_bcr(cpu);
+		READ_BCR(ARC_REG_MICRO_ARCH_BCR, uarch);
 
-	if (is_isa_arcompact()) {
-		struct bcr_fp_arcompact sp, dp;
-		struct bcr_bpu_arcompact bpu;
-
-		READ_BCR(ARC_REG_FP_BCR, sp);
-		READ_BCR(ARC_REG_DPFP_BCR, dp);
-		cpu->extn.fpu_sp = sp.ver ? 1 : 0;
-		cpu->extn.fpu_dp = dp.ver ? 1 : 0;
-
-		READ_BCR(ARC_REG_BPU_BCR, bpu);
-		cpu->bpu.ver = bpu.ver;
-		cpu->bpu.full = bpu.fam ? 1 : 0;
-		if (bpu.ent) {
-			cpu->bpu.num_cache = 256 << (bpu.ent - 1);
-			cpu->bpu.num_pred = 256 << (bpu.ent - 1);
+		for (tbl = &arc_hs_ver54_rel[0]; tbl->id != 0xFF; tbl++) {
+			if (uarch.maj == tbl->id) {
+				release = tbl->str;
+				break;
+			}
 		}
-	} else {
-		struct bcr_fp_arcv2 spdp;
-		struct bcr_bpu_arcv2 bpu;
-
-		READ_BCR(ARC_REG_FP_V2_BCR, spdp);
-		cpu->extn.fpu_sp = spdp.sp ? 1 : 0;
-		cpu->extn.fpu_dp = spdp.dp ? 1 : 0;
-
-		READ_BCR(ARC_REG_BPU_BCR, bpu);
-		cpu->bpu.ver = bpu.ver;
-		cpu->bpu.full = bpu.ft;
-		cpu->bpu.num_cache = 256 << bpu.bce;
-		cpu->bpu.num_pred = 2048 << bpu.pte;
-		cpu->bpu.ret_stk = 4 << bpu.rse;
-
-		/* if dual issue hardware, is it enabled ? */
-		if (cpu->extn.dual) {
+		if (uarch.prod == 4) {
 			unsigned int exec_ctrl;
 
+			cpu_nm = "HS48";
+			dual_issue = 1;
+			/* if dual issue hardware, is it enabled ? */
 			READ_BCR(AUX_EXEC_CTRL, exec_ctrl);
-			cpu->extn.dual_enb = !(exec_ctrl & 1);
+			dual_enb = !(exec_ctrl & 1);
 		}
 	}
 
-	READ_BCR(ARC_REG_AP_BCR, ap);
-	if (ap.ver) {
-		cpu->extn.ap_num = 2 << ap.num;
-		cpu->extn.ap_full = !ap.min;
-	}
-
-	READ_BCR(ARC_REG_SMART_BCR, bcr);
-	cpu->extn.smart = bcr.ver ? 1 : 0;
-
-	READ_BCR(ARC_REG_RTT_BCR, bcr);
-	cpu->extn.rtt = bcr.ver ? 1 : 0;
-
 	READ_BCR(ARC_REG_ISA_CFG_BCR, isa);
 
-	/* some hacks for lack of feature BCR info in old ARC700 cores */
-	if (is_isa_arcompact()) {
-		if (!isa.ver)	/* ISA BCR absent, use Kconfig info */
-			cpu->isa.atomic = IS_ENABLED(CONFIG_ARC_HAS_LLSC);
-		else {
-			/* ARC700_BUILD only has 2 bits of isa info */
-			struct bcr_generic bcr = *(struct bcr_generic *)&isa;
-			cpu->isa.atomic = bcr.info & 1;
-		}
-
-		cpu->isa.be = IS_ENABLED(CONFIG_CPU_BIG_ENDIAN);
+	n += scnprintf(buf + n, len - n, "processor [%d]\t: %s %s (%s ISA) %s%s%s\n",
+		       c, cpu_nm, release, isa_nm,
+		       IS_AVAIL1(isa.be, "[Big-Endian]"),
+		       IS_AVAIL3(dual_issue, dual_enb, " Dual-Issue "));
+
+	READ_BCR(ARC_REG_MPY_BCR, mpy);
+	mpy_opt = 2;	/* stock MPY/MPYH */
+	if (mpy.dsp)	/* OPT 7-9 */
+		mpy_opt = mpy.dsp + 6;
+
+	scnprintf(mpy_nm, 16, "mpy[opt %d] ", mpy_opt);
+
+	READ_BCR(ARC_REG_FP_V2_BCR, fpu);
+
+	n += scnprintf(buf + n, len - n, "ISA Extn\t: %s%s%s%s%s%s%s%s%s%s%s\n",
+		       IS_AVAIL2(isa.atomic, "atomic ", CONFIG_ARC_HAS_LLSC),
+		       IS_AVAIL2(isa.ldd, "ll64 ", CONFIG_ARC_HAS_LL64),
+		       IS_AVAIL2(isa.unalign, "unalign ", CONFIG_ARC_USE_UNALIGNED_MEM_ACCESS),
+		       IS_AVAIL1(mpy.ver, mpy_nm),
+		       IS_AVAIL1(isa.div_rem, "div_rem "),
+		       IS_AVAIL1((fpu.sp | fpu.dp), "  FPU:"),
+		       IS_AVAIL1(fpu.sp, " sp"),
+		       IS_AVAIL1(fpu.dp, " dp"));
+
+	READ_BCR(ARC_REG_BPU_BCR, bpu);
+	bpu_full = bpu.ft;
+	bpu_cache = 256 << bpu.bce;
+	bpu_pred = 2048 << bpu.pte;
+	bpu_ret_stk = 4 << bpu.rse;
+
+	READ_BCR(ARC_REG_LPB_BUILD, lpb);
+	if (lpb.ver) {
+		unsigned int ctl;
+		ctl = read_aux_reg(ARC_REG_LPB_CTRL);
+
+		scnprintf(lpb_nm, sizeof(lpb_nm), " Loop Buffer:%d %s",
+			  lpb.entries, IS_DISABLED_RUN(!ctl));
+	}
 
-		 /* there's no direct way to distinguish 750 vs. 770 */
-		if (unlikely(cpu->core.family < 0x34))
-			cpu->name = "ARC750";
-	} else {
-		cpu->isa = isa;
+	n += scnprintf(buf + n, len - n,
+			"BPU\t\t: %s%s match, cache:%d, Predict Table:%d Return stk: %d%s\n",
+			IS_AVAIL1(bpu_full, "full"),
+			IS_AVAIL1(!bpu_full, "partial"),
+			bpu_cache, bpu_pred, bpu_ret_stk,
+			lpb_nm);
+
+	READ_BCR(ARC_REG_ICCM_BUILD, iccm);
+	if (iccm.ver) {
+		unsigned long base;
+		info->iccm.sz = 256 << iccm.sz00;	/* 512B to 16M */
+		if (iccm.sz00 == 0xF && iccm.sz01 > 0)
+			info->iccm.sz <<= iccm.sz01;
+		base = read_aux_reg(ARC_REG_AUX_ICCM);
+		info->iccm.base = base & 0xF0000000;
 	}
-}
 
-static char *arc_cpu_mumbojumbo(int cpu_id, char *buf, int len)
-{
-	struct cpuinfo_arc *cpu = &cpuinfo_arc700[cpu_id];
-	struct bcr_identity *core = &cpu->core;
-	char mpy_opt[16];
-	int n = 0;
+	READ_BCR(ARC_REG_DCCM_BUILD, dccm);
+	if (dccm.ver) {
+		unsigned long base;
+		info->dccm.sz = 256 << dccm.sz0;
+		if (dccm.sz0 == 0xF && dccm.sz1 > 0)
+			info->dccm.sz <<= dccm.sz1;
+		base = read_aux_reg(ARC_REG_AUX_DCCM);
+		info->dccm.base = base & 0xF0000000;
+	}
 
-	FIX_PTR(cpu);
+	/* Error Protection: ECC/Parity */
+	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,
-		       "\nIDENTITY\t: ARCVER [%#02x] ARCNUM [%#02x] CHIPID [%#4x]\n",
-		       core->family, core->cpu_id, core->chip_id);
+	/* ARCv2 ISA specific sanity checks */
+	present = fpu.sp | fpu.dp | mpy.dsp;	/* DSP and/or FPU */
+	CHK_OPT_STRICT(CONFIG_ARC_HAS_ACCL_REGS, present);
 
-	n += scnprintf(buf + n, len - n, "processor [%d]\t: %s %s (%s ISA) %s%s%s\n",
-		       cpu_id, cpu->name, cpu->release,
-		       is_isa_arcompact() ? "ARCompact" : "ARCv2",
-		       IS_AVAIL1(cpu->isa.be, "[Big-Endian]"),
-		       IS_AVAIL3(cpu->extn.dual, cpu->extn.dual_enb, " Dual-Issue "));
+	dsp_config_check();
+#endif
+	return n;
+}
 
-	n += scnprintf(buf + n, len - n, "Timers\t\t: %s%s%s%s%s%s\nISA Extn\t: ",
-		       IS_AVAIL1(cpu->extn.timer0, "Timer0 "),
-		       IS_AVAIL1(cpu->extn.timer1, "Timer1 "),
-		       IS_AVAIL2(cpu->extn.rtc, "RTC [UP 64-bit] ", CONFIG_ARC_TIMERS_64BIT),
-		       IS_AVAIL2(cpu->extn.gfrc, "GFRC [SMP 64-bit] ", CONFIG_ARC_TIMERS_64BIT));
+static char *arc_cpu_mumbojumbo(int c, struct cpuinfo_arc *info, char *buf, int len)
+{
+	struct bcr_identity ident;
+	struct bcr_timer timer;
+	struct bcr_generic bcr;
+	struct mcip_bcr mp;
+	struct bcr_actionpoint ap;
+	unsigned long vec_base;
+	int ap_num, ap_full, smart, rtt, n;
 
-	if (cpu->extn_mpy.ver) {
-		if (is_isa_arcompact()) {
-			scnprintf(mpy_opt, 16, "mpy");
-		} else {
+	memset(info, 0, sizeof(struct cpuinfo_arc));
 
-			int opt = 2;	/* stock MPY/MPYH */
+	READ_BCR(AUX_IDENTITY, ident);
+	info->arcver = ident.family;
 
-			if (cpu->extn_mpy.dsp)	/* OPT 7-9 */
-				opt = cpu->extn_mpy.dsp + 6;
+	n = scnprintf(buf, len,
+		       "\nIDENTITY\t: ARCVER [%#02x] ARCNUM [%#02x] CHIPID [%#4x]\n",
+		       ident.family, ident.cpu_id, ident.chip_id);
 
-			scnprintf(mpy_opt, 16, "mpy[opt %d] ", opt);
-		}
+	if (is_isa_arcompact()) {
+		n += arcompact_mumbojumbo(c, info, buf + n, len - n);
+	} else if (is_isa_arcv2()){
+		n += arcv2_mumbojumbo(c, info, buf + n, len - n);
 	}
 
-	n += scnprintf(buf + n, len - n, "%s%s%s%s%s%s%s%s\n",
-		       IS_AVAIL2(cpu->isa.atomic, "atomic ", CONFIG_ARC_HAS_LLSC),
-		       IS_AVAIL2(cpu->isa.ldd, "ll64 ", CONFIG_ARC_HAS_LL64),
-		       IS_AVAIL2(cpu->isa.unalign, "unalign ", CONFIG_ARC_USE_UNALIGNED_MEM_ACCESS),
-		       IS_AVAIL1(cpu->extn_mpy.ver, mpy_opt),
-		       IS_AVAIL1(cpu->isa.div_rem, "div_rem "));
+	n += arc_mmu_mumbojumbo(c, buf + n, len - n);
+	n += arc_cache_mumbojumbo(c, buf + n, len - n);
 
-	if (cpu->bpu.ver) {
-		n += scnprintf(buf + n, len - n,
-			      "BPU\t\t: %s%s match, cache:%d, Predict Table:%d Return stk: %d",
-			      IS_AVAIL1(cpu->bpu.full, "full"),
-			      IS_AVAIL1(!cpu->bpu.full, "partial"),
-			      cpu->bpu.num_cache, cpu->bpu.num_pred, cpu->bpu.ret_stk);
-
-		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");
-	}
+	READ_BCR(ARC_REG_TIMERS_BCR, timer);
+	info->t0 = timer.t0;
+	info->t1 = timer.t1;
 
-	return buf;
-}
+	READ_BCR(ARC_REG_MCIP_BCR, mp);
+	vec_base = read_aux_reg(AUX_INTR_VEC_BASE);
 
-static char *arc_extn_mumbojumbo(int cpu_id, char *buf, int len)
-{
-	int n = 0;
-	struct cpuinfo_arc *cpu = &cpuinfo_arc700[cpu_id];
+	n += scnprintf(buf + n, len - n,
+		       "Timers\t\t: %s%s%s%s%s%s\nVector Table\t: %#lx\n",
+		       IS_AVAIL1(timer.t0, "Timer0 "),
+		       IS_AVAIL1(timer.t1, "Timer1 "),
+		       IS_AVAIL2(timer.rtc, "RTC [UP 64-bit] ", CONFIG_ARC_TIMERS_64BIT),
+		       IS_AVAIL2(mp.gfrc, "GFRC [SMP 64-bit] ", CONFIG_ARC_TIMERS_64BIT),
+		       vec_base);
 
-	FIX_PTR(cpu);
+	READ_BCR(ARC_REG_AP_BCR, ap);
+	if (ap.ver) {
+		ap_num = 2 << ap.num;
+		ap_full = !ap.min;
+	}
 
-	n += scnprintf(buf + n, len - n, "Vector Table\t: %#x\n", cpu->vec_base);
+	READ_BCR(ARC_REG_SMART_BCR, bcr);
+	smart = bcr.ver ? 1 : 0;
 
-	if (cpu->extn.fpu_sp || cpu->extn.fpu_dp)
-		n += scnprintf(buf + n, len - n, "FPU\t\t: %s%s\n",
-			       IS_AVAIL1(cpu->extn.fpu_sp, "SP "),
-			       IS_AVAIL1(cpu->extn.fpu_dp, "DP "));
+	READ_BCR(ARC_REG_RTT_BCR, bcr);
+	rtt = bcr.ver ? 1 : 0;
 
-	if (cpu->extn.ap_num | cpu->extn.smart | cpu->extn.rtt) {
+	if (ap.ver | smart | rtt) {
 		n += scnprintf(buf + n, len - n, "DEBUG\t\t: %s%s",
-			       IS_AVAIL1(cpu->extn.smart, "smaRT "),
-			       IS_AVAIL1(cpu->extn.rtt, "RTT "));
-		if (cpu->extn.ap_num) {
+			       IS_AVAIL1(smart, "smaRT "),
+			       IS_AVAIL1(rtt, "RTT "));
+		if (ap.ver) {
 			n += scnprintf(buf + n, len - n, "ActionPoint %d/%s",
-				       cpu->extn.ap_num,
-				       cpu->extn.ap_full ? "full":"min");
+				       ap_num,
+				       ap_full ? "full":"min");
 		}
 		n += scnprintf(buf + n, len - n, "\n");
 	}
 
-	if (cpu->dccm.sz || cpu->iccm.sz)
-		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 "));
-		}
-	}
+	if (info->dccm.sz || info->iccm.sz)
+		n += scnprintf(buf + n, len - n,
+			       "Extn [CCM]\t: DCCM @ %lx, %d KB / ICCM: @ %lx, %d KB\n",
+			       info->dccm.base, TO_KB(info->dccm.sz),
+			       info->iccm.base, TO_KB(info->iccm.sz));
 
 	return buf;
 }
@@ -398,15 +376,15 @@ void chk_opt_weak(char *opt_name, bool hw_exists, bool opt_ena)
 		panic("Disable %s, hardware NOT present\n", opt_name);
 }
 
-static void arc_chk_core_config(void)
+/*
+ * ISA agnostic sanity checks
+ */
+static void arc_chk_core_config(struct cpuinfo_arc *info)
 {
-	struct cpuinfo_arc *cpu = &cpuinfo_arc700[smp_processor_id()];
-	int present = 0;
-
-	if (!cpu->extn.timer0)
+	if (!info->t0)
 		panic("Timer0 is not present!\n");
 
-	if (!cpu->extn.timer1)
+	if (!info->t1)
 		panic("Timer1 is not present!\n");
 
 #ifdef CONFIG_ARC_HAS_DCCM
@@ -414,35 +392,17 @@ static void arc_chk_core_config(void)
 	 * DCCM can be arbit placed in hardware.
 	 * Make sure it's placement/sz matches what Linux is built with
 	 */
-	if ((unsigned int)__arc_dccm_base != cpu->dccm.base_addr)
+	if ((unsigned int)__arc_dccm_base != info->dccm.base)
 		panic("Linux built with incorrect DCCM Base address\n");
 
-	if (CONFIG_ARC_DCCM_SZ * SZ_1K != cpu->dccm.sz)
+	if (CONFIG_ARC_DCCM_SZ * SZ_1K != info->dccm.sz)
 		panic("Linux built with incorrect DCCM Size\n");
 #endif
 
 #ifdef CONFIG_ARC_HAS_ICCM
-	if (CONFIG_ARC_ICCM_SZ * SZ_1K != cpu->iccm.sz)
+	if (CONFIG_ARC_ICCM_SZ * SZ_1K != info->iccm.sz)
 		panic("Linux built with incorrect ICCM Size\n");
 #endif
-
-	/*
-	 * FP hardware/software config sanity
-	 * -If hardware present, kernel needs to save/restore FPU state
-	 * -If not, it will crash trying to save/restore the non-existant regs
-	 */
-
-	if (is_isa_arcompact()) {
-		/* only DPDP checked since SP has no arch visible regs */
-		present = cpu->extn.fpu_dp;
-		CHK_OPT_STRICT(CONFIG_ARC_FPU_SAVE_RESTORE, present);
-	} else {
-		/* Accumulator Low:High pair (r58:59) present if DSP MPY or FPU */
-		present = cpu->extn_mpy.dsp | cpu->extn.fpu_sp | cpu->extn.fpu_dp;
-		CHK_OPT_STRICT(CONFIG_ARC_HAS_ACCL_REGS, present);
-
-		dsp_config_check();
-	}
 }
 
 /*
@@ -453,23 +413,19 @@ static void arc_chk_core_config(void)
 
 void setup_processor(void)
 {
+	struct cpuinfo_arc info;
+	int c = smp_processor_id();
 	char str[512];
-	int cpu_id = smp_processor_id();
 
-	read_arc_build_cfg_regs();
-	arc_init_IRQ();
+	pr_info("%s", arc_cpu_mumbojumbo(c, &info, str, sizeof(str)));
+	pr_info("%s", arc_platform_smp_cpuinfo());
 
-	pr_info("%s", arc_cpu_mumbojumbo(cpu_id, str, sizeof(str)));
-	pr_info("%s", arc_mmu_mumbojumbo(cpu_id, str, sizeof(str)));
-	pr_info("%s", arc_cache_mumbojumbo(cpu_id, str, sizeof(str)));
+	arc_chk_core_config(&info);
 
+	arc_init_IRQ();
 	arc_mmu_init();
 	arc_cache_init();
 
-	pr_info("%s", arc_extn_mumbojumbo(cpu_id, str, sizeof(str)));
-	pr_info("%s", arc_platform_smp_cpuinfo());
-
-	arc_chk_core_config();
 }
 
 static inline bool uboot_arg_invalid(unsigned long addr)
@@ -616,6 +572,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	char *str;
 	int cpu_id = ptr_to_cpu(v);
 	struct device *cpu_dev = get_cpu_device(cpu_id);
+	struct cpuinfo_arc info;
 	struct clk *cpu_clk;
 	unsigned long freq = 0;
 
@@ -628,7 +585,7 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 	if (!str)
 		goto done;
 
-	seq_printf(m, arc_cpu_mumbojumbo(cpu_id, str, PAGE_SIZE));
+	seq_printf(m, arc_cpu_mumbojumbo(cpu_id, &info, str, PAGE_SIZE));
 
 	cpu_clk = clk_get(cpu_dev, NULL);
 	if (IS_ERR(cpu_clk)) {
@@ -645,9 +602,6 @@ static int show_cpuinfo(struct seq_file *m, void *v)
 		   loops_per_jiffy / (500000 / HZ),
 		   (loops_per_jiffy / (5000 / HZ)) % 100);
 
-	seq_printf(m, arc_mmu_mumbojumbo(cpu_id, str, PAGE_SIZE));
-	seq_printf(m, arc_cache_mumbojumbo(cpu_id, str, PAGE_SIZE));
-	seq_printf(m, arc_extn_mumbojumbo(cpu_id, str, PAGE_SIZE));
 	seq_printf(m, arc_platform_smp_cpuinfo());
 
 	free_page((unsigned long)str);
-- 
cgit v1.2.3


From cfca4b5abe0cc13f9d9f45f760efd8260e31200f Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Tue, 12 May 2020 22:18:08 -0700
Subject: ARC: entry: use gp to cache task pointer (vs. r25)

The motivation is eventual ABI considerations for ARCv3 but even without
it this change us worthwhile as diffstat reduces 100 net lines

r25 is a callee saved register, normally not saved by entry code in
pt_regs. However because of its usage in CONFIG_ARC_CURR_IN_REG it needs
to be. This in turn requires a whole bunch of special casing when we
need to access r25. Then there is distinction between user mode r25 vs.
kernel mode r25 - hence distinct SAVE_CALLEE_SAVED_{USER,KERNEL}

Instead use gp which is a scratch register and thus saved already in entry
code. This cleans things up significantly and much nocer on eyes:

 - SAVE_CALLEE_SAVED_{USER,KERNEL} are now exactly same
 - no special user_r25 slot in pt_reggs

Note that typical global asm registers are callee-saved (r25), but gp is
not callee-saved thus needs additional -ffixed-<reg> toggle

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/asm-offsets.c |  2 --
 arch/arc/kernel/ctx_sw.c      | 13 ++-----------
 arch/arc/kernel/ctx_sw_asm.S  |  2 +-
 arch/arc/kernel/entry.S       |  3 +--
 arch/arc/kernel/process.c     | 11 -----------
 arch/arc/kernel/ptrace.c      |  2 --
 6 files changed, 4 insertions(+), 29 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/asm-offsets.c b/arch/arc/kernel/asm-offsets.c
index 0e884036ab74..37324fd9a72f 100644
--- a/arch/arc/kernel/asm-offsets.c
+++ b/arch/arc/kernel/asm-offsets.c
@@ -63,8 +63,6 @@ int main(void)
 	DEFINE(PT_blink, offsetof(struct pt_regs, blink));
 	DEFINE(PT_lpe, offsetof(struct pt_regs, lp_end));
 	DEFINE(PT_lpc, offsetof(struct pt_regs, lp_count));
-	DEFINE(PT_user_r25, offsetof(struct pt_regs, user_r25));
-
 	DEFINE(SZ_CALLEE_REGS, sizeof(struct callee_regs));
 	DEFINE(SZ_PT_REGS, sizeof(struct pt_regs));
 
diff --git a/arch/arc/kernel/ctx_sw.c b/arch/arc/kernel/ctx_sw.c
index bf16f777a0bc..40d89440b7e4 100644
--- a/arch/arc/kernel/ctx_sw.c
+++ b/arch/arc/kernel/ctx_sw.c
@@ -38,11 +38,7 @@ __switch_to(struct task_struct *prev_task, struct task_struct *next_task)
 		"st.a    r22, [sp, -4]   \n\t"
 		"st.a    r23, [sp, -4]   \n\t"
 		"st.a    r24, [sp, -4]   \n\t"
-#ifndef CONFIG_ARC_CURR_IN_REG
 		"st.a    r25, [sp, -4]   \n\t"
-#else
-		"sub     sp, sp, 4      \n\t"	/* usual r25 placeholder */
-#endif
 
 		/* set ksp of outgoing task in tsk->thread.ksp */
 #if KSP_WORD_OFF <= 255
@@ -58,7 +54,7 @@ __switch_to(struct task_struct *prev_task, struct task_struct *next_task)
 
 		/*
 		 * setup _current_task with incoming tsk.
-		 * optionally, set r25 to that as well
+		 * optionally, set caching reg to that as well
 		 * For SMP extra work to get to &_current_task[cpu]
 		 * (open coded SET_CURR_TASK_ON_CPU)
 		 */
@@ -72,19 +68,14 @@ __switch_to(struct task_struct *prev_task, struct task_struct *next_task)
 		"st   %2,  [r24]		\n\t"
 #endif
 #ifdef CONFIG_ARC_CURR_IN_REG
-		"mov r25, %2   \n\t"
+		"mov gp, %2   \n\t"
 #endif
 
 		/* get ksp of incoming task from tsk->thread.ksp */
 		"ld.as  sp, [%2, %1]   \n\t"
 
 		/* start loading it's CALLEE reg file */
-
-#ifndef CONFIG_ARC_CURR_IN_REG
 		"ld.ab   r25, [sp, 4]   \n\t"
-#else
-		"add    sp, sp, 4       \n\t"
-#endif
 		"ld.ab   r24, [sp, 4]   \n\t"
 		"ld.ab   r23, [sp, 4]   \n\t"
 		"ld.ab   r22, [sp, 4]   \n\t"
diff --git a/arch/arc/kernel/ctx_sw_asm.S b/arch/arc/kernel/ctx_sw_asm.S
index 02c461484761..95cba6265e2b 100644
--- a/arch/arc/kernel/ctx_sw_asm.S
+++ b/arch/arc/kernel/ctx_sw_asm.S
@@ -49,7 +49,7 @@ __switch_to:
 	SET_CURR_TASK_ON_CPU  r1, r3
 
 	/* reload SP with kernel mode stack pointer in task->thread.ksp */
-	ld.as  sp, [r1, (TASK_THREAD + THREAD_KSP)/4]
+	ld.as  sp, [r1, KSP_WORD_OFF]
 
 	/* restore the registers */
 	RESTORE_CALLEE_SAVED_KERNEL
diff --git a/arch/arc/kernel/entry.S b/arch/arc/kernel/entry.S
index 54e91df678dd..cd26e0fa5044 100644
--- a/arch/arc/kernel/entry.S
+++ b/arch/arc/kernel/entry.S
@@ -210,7 +210,6 @@ trap_with_param:
 
 	; Save callee regs in case gdb wants to have a look
 	; SP will grow up by size of CALLEE Reg-File
-	; NOTE: clobbers r12
 	SAVE_CALLEE_SAVED_USER
 
 	; save location of saved Callee Regs @ thread_struct->pc
@@ -318,7 +317,7 @@ resume_user_mode_begin:
 	;      tracer might call PEEKUSR(CALLEE reg)
 	;
 	; NOTE: SP will grow up by size of CALLEE Reg-File
-	SAVE_CALLEE_SAVED_USER		; clobbers r12
+	SAVE_CALLEE_SAVED_USER
 
 	; save location of saved Callee Regs @ thread_struct->callee
 	GET_CURR_TASK_FIELD_PTR   TASK_THREAD, r10
diff --git a/arch/arc/kernel/process.c b/arch/arc/kernel/process.c
index 980b71da2f61..96f591508142 100644
--- a/arch/arc/kernel/process.c
+++ b/arch/arc/kernel/process.c
@@ -162,7 +162,6 @@ asmlinkage void ret_from_fork(void);
  * |      SP        |
  * |    orig_r0     |
  * |    event/ECR   |
- * |    user_r25    |
  * ------------------  <===== END of PAGE
  */
 int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
@@ -243,16 +242,6 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
 	 */
 	c_callee->r25 = task_thread_info(p)->thr_ptr;
 
-#ifdef CONFIG_ARC_CURR_IN_REG
-	/*
-	 * setup usermode thread pointer #2:
-	 * however for this special use of r25 in kernel, __switch_to() sets
-	 * r25 for kernel needs and only in the final return path is usermode
-	 * r25 setup, from pt_regs->user_r25. So set that up as well
-	 */
-	c_regs->user_r25 = c_callee->r25;
-#endif
-
 	return 0;
 }
 
diff --git a/arch/arc/kernel/ptrace.c b/arch/arc/kernel/ptrace.c
index 2abdcd9b09e8..8226df3163fe 100644
--- a/arch/arc/kernel/ptrace.c
+++ b/arch/arc/kernel/ptrace.c
@@ -47,7 +47,6 @@ static const struct pt_regs_offset regoffset_table[] = {
 	REG_OFFSET_NAME(sp),
 	REG_OFFSET_NAME(orig_r0),
 	REG_OFFSET_NAME(event),
-	REG_OFFSET_NAME(user_r25),
 	REG_OFFSET_END,
 };
 
@@ -57,7 +56,6 @@ static const struct pt_regs_offset regoffset_table[] = {
 	REG_OFFSET_NAME(orig_r0),
 	REG_OFFSET_NAME(event),
 	REG_OFFSET_NAME(bta),
-	REG_OFFSET_NAME(user_r25),
 	REG_OFFSET_NAME(r26),
 	REG_OFFSET_NAME(fp),
 	REG_OFFSET_NAME(sp),
-- 
cgit v1.2.3


From b060b7d0c105d495eb9338a3f6c7bc0c09830a2c Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Thu, 14 May 2020 00:16:34 -0700
Subject: ARC: __switch_to: asm with dwarf ops (vs. inline asm)

__switch_to() is final step of context switch, swapping kernel modes
stack (and callee regs) of outgoing task with next task.

It is also the starting point of stack unwinging of a sleeping task and
captures SP, FP, BLINK and the corresponding dwarf info. Back when
dinosaurs still roamed around, ARC gas didn't support CFI pseudo ops and
gcc was responsible for generating dwarf info. Thus it had to be written
in "C" with inline asm to do the hand crafting of stack. The function
prologue (and crucial saving of blink etc) was still gcc generated but
not visible in code. Likewise dwarf info was missing.

Now with modern tools, we can make things more obvious by writing the
code in asm and adding approproate dwarf cfi pseudo ops.

This is mostly non functional change, except for slight chnages to asm

 - ARCompact doesn't support MOV_S fp, sp, so we use MOV

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/Makefile     |   9 +---
 arch/arc/kernel/ctx_sw.c     | 103 -------------------------------------------
 arch/arc/kernel/ctx_sw_asm.S |  72 ++++++++++++++++--------------
 3 files changed, 41 insertions(+), 143 deletions(-)
 delete mode 100644 arch/arc/kernel/ctx_sw.c

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/Makefile b/arch/arc/kernel/Makefile
index 0723d888ac44..95fbf9364c67 100644
--- a/arch/arc/kernel/Makefile
+++ b/arch/arc/kernel/Makefile
@@ -5,6 +5,8 @@
 
 obj-y	:= head.o arcksyms.o setup.o irq.o reset.o ptrace.o process.o devtree.o
 obj-y	+= signal.o traps.o sys.o troubleshoot.o stacktrace.o disasm.o
+obj-y	+= ctx_sw_asm.o
+
 obj-$(CONFIG_ISA_ARCOMPACT)		+= entry-compact.o intc-compact.o
 obj-$(CONFIG_ISA_ARCV2)			+= entry-arcv2.o intc-arcv2.o
 
@@ -24,11 +26,4 @@ ifdef CONFIG_ISA_ARCOMPACT
 CFLAGS_fpu.o   += -mdpfp
 endif
 
-ifdef CONFIG_ARC_DW2_UNWIND
-CFLAGS_ctx_sw.o += -fno-omit-frame-pointer
-obj-y += ctx_sw.o
-else
-obj-y += ctx_sw_asm.o
-endif
-
 extra-y := vmlinux.lds
diff --git a/arch/arc/kernel/ctx_sw.c b/arch/arc/kernel/ctx_sw.c
deleted file mode 100644
index 40d89440b7e4..000000000000
--- a/arch/arc/kernel/ctx_sw.c
+++ /dev/null
@@ -1,103 +0,0 @@
-// SPDX-License-Identifier: GPL-2.0-only
-/*
- * Copyright (C) 2004, 2007-2010, 2011-2012 Synopsys, Inc. (www.synopsys.com)
- *
- * Vineetg: Aug 2009
- *  -"C" version of lowest level context switch asm macro called by schedular
- *   gcc doesn't generate the dward CFI info for hand written asm, hence can't
- *   backtrace out of it (e.g. tasks sleeping in kernel).
- *   So we cheat a bit by writing almost similar code in inline-asm.
- *  -This is a hacky way of doing things, but there is no other simple way.
- *   I don't want/intend to extend unwinding code to understand raw asm
- */
-
-#include <asm/asm-offsets.h>
-#include <asm/switch_to.h>
-#include <linux/sched/debug.h>
-
-#define KSP_WORD_OFF 	((TASK_THREAD + THREAD_KSP) / 4)
-
-struct task_struct *__sched
-__switch_to(struct task_struct *prev_task, struct task_struct *next_task)
-{
-	unsigned int tmp;
-	unsigned int prev = (unsigned int)prev_task;
-	unsigned int next = (unsigned int)next_task;
-
-	__asm__ __volatile__(
-		/* FP/BLINK save generated by gcc (standard function prologue */
-		"st.a    r13, [sp, -4]   \n\t"
-		"st.a    r14, [sp, -4]   \n\t"
-		"st.a    r15, [sp, -4]   \n\t"
-		"st.a    r16, [sp, -4]   \n\t"
-		"st.a    r17, [sp, -4]   \n\t"
-		"st.a    r18, [sp, -4]   \n\t"
-		"st.a    r19, [sp, -4]   \n\t"
-		"st.a    r20, [sp, -4]   \n\t"
-		"st.a    r21, [sp, -4]   \n\t"
-		"st.a    r22, [sp, -4]   \n\t"
-		"st.a    r23, [sp, -4]   \n\t"
-		"st.a    r24, [sp, -4]   \n\t"
-		"st.a    r25, [sp, -4]   \n\t"
-
-		/* set ksp of outgoing task in tsk->thread.ksp */
-#if KSP_WORD_OFF <= 255
-		"st.as   sp, [%3, %1]    \n\t"
-#else
-		/*
-		 * Workaround for NR_CPUS=4k
-		 * %1 is bigger than 255 (S9 offset for st.as)
-		 */
-		"add2    r24, %3, %1     \n\t"
-		"st      sp, [r24]       \n\t"
-#endif
-
-		/*
-		 * setup _current_task with incoming tsk.
-		 * optionally, set caching reg to that as well
-		 * For SMP extra work to get to &_current_task[cpu]
-		 * (open coded SET_CURR_TASK_ON_CPU)
-		 */
-#ifndef CONFIG_SMP
-		"st  %2, [@_current_task]	\n\t"
-#else
-		"lr   r24, [identity]		\n\t"
-		"lsr  r24, r24, 8		\n\t"
-		"bmsk r24, r24, 7		\n\t"
-		"add2 r24, @_current_task, r24	\n\t"
-		"st   %2,  [r24]		\n\t"
-#endif
-#ifdef CONFIG_ARC_CURR_IN_REG
-		"mov gp, %2   \n\t"
-#endif
-
-		/* get ksp of incoming task from tsk->thread.ksp */
-		"ld.as  sp, [%2, %1]   \n\t"
-
-		/* start loading it's CALLEE reg file */
-		"ld.ab   r25, [sp, 4]   \n\t"
-		"ld.ab   r24, [sp, 4]   \n\t"
-		"ld.ab   r23, [sp, 4]   \n\t"
-		"ld.ab   r22, [sp, 4]   \n\t"
-		"ld.ab   r21, [sp, 4]   \n\t"
-		"ld.ab   r20, [sp, 4]   \n\t"
-		"ld.ab   r19, [sp, 4]   \n\t"
-		"ld.ab   r18, [sp, 4]   \n\t"
-		"ld.ab   r17, [sp, 4]   \n\t"
-		"ld.ab   r16, [sp, 4]   \n\t"
-		"ld.ab   r15, [sp, 4]   \n\t"
-		"ld.ab   r14, [sp, 4]   \n\t"
-		"ld.ab   r13, [sp, 4]   \n\t"
-
-		/* last (ret value) = prev : although for ARC it mov r0, r0 */
-		"mov     %0, %3        \n\t"
-
-		/* FP/BLINK restore generated by gcc (standard func epilogue */
-
-		: "=r"(tmp)
-		: "n"(KSP_WORD_OFF), "r"(next), "r"(prev)
-		: "blink"
-	);
-
-	return (struct task_struct *)tmp;
-}
diff --git a/arch/arc/kernel/ctx_sw_asm.S b/arch/arc/kernel/ctx_sw_asm.S
index 95cba6265e2b..59d779004e64 100644
--- a/arch/arc/kernel/ctx_sw_asm.S
+++ b/arch/arc/kernel/ctx_sw_asm.S
@@ -13,48 +13,54 @@
 
 #define KSP_WORD_OFF 	((TASK_THREAD + THREAD_KSP) / 4)
 
-;################### Low Level Context Switch ##########################
+; IN
+;  - r0: prev task (also current)
+;  - r1: next task
+; OUT
+;  - r0: prev task (so r0 not touched)
 
 	.section .sched.text,"ax",@progbits
-	.align 4
-	.global __switch_to
-	.type   __switch_to, @function
-__switch_to:
-	CFI_STARTPROC
-
-	/* Save regs on kernel mode stack of task */
-	st.a    blink, [sp, -4]
-	st.a    fp, [sp, -4]
+ENTRY_CFI(__switch_to)
+
+	/* save kernel stack frame regs of @prev task */
+	push	blink
+	CFI_DEF_CFA_OFFSET 4
+	CFI_OFFSET r31, -4
+
+	push	fp
+	CFI_DEF_CFA_OFFSET 8
+	CFI_OFFSET r27, -8
+
+	mov	fp, sp
+	CFI_DEF_CFA_REGISTER r27
+
+	/* kernel mode callee regs of @prev */
 	SAVE_CALLEE_SAVED_KERNEL
 
-	/* Save the now KSP in task->thread.ksp */
+	/* save final SP to @prev->thread.ksp */
 #if KSP_WORD_OFF  <= 255
 	st.as  sp, [r0, KSP_WORD_OFF]
 #else
 	/* Workaround for NR_CPUS=4k as ST.as can only take s9 offset */
-	add2	r24, r0, KSP_WORD_OFF
-	st	sp, [r24]
+	add2	r10, r0, KSP_WORD_OFF
+	st	sp, [r10]
 #endif
-	/*
-	* Return last task in r0 (return reg)
-	* On ARC, Return reg = First Arg reg = r0.
-	* Since we already have last task in r0,
-	* don't need to do anything special to return it
-	*/
-
-	/*
-	 * switch to new task, contained in r1
-	 * Temp reg r3 is required to get the ptr to store val
-	 */
-	SET_CURR_TASK_ON_CPU  r1, r3
-
-	/* reload SP with kernel mode stack pointer in task->thread.ksp */
-	ld.as  sp, [r1, KSP_WORD_OFF]
-
-	/* restore the registers */
+	/* update @next in _current_task[] and GP register caching it */
+	SET_CURR_TASK_ON_CPU  r1, r10
+
+	/* load SP from @next->thread.ksp */
+	ld.as	sp, [r1, KSP_WORD_OFF]
+
+	/* restore callee regs, stack frame regs of @next */
 	RESTORE_CALLEE_SAVED_KERNEL
-	ld.ab   fp, [sp, 4]
-	ld.ab   blink, [sp, 4]
-	j       [blink]
 
+	pop	fp
+	CFI_RESTORE r27
+	CFI_DEF_CFA r28, 4
+
+	pop	blink
+	CFI_RESTORE r31
+	CFI_DEF_CFA_OFFSET 0
+
+	j      [blink]
 END_CFI(__switch_to)
-- 
cgit v1.2.3


From fd476197c63dd3aae85e286dab7da23b159562b9 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Thu, 14 May 2020 00:51:26 -0700
Subject: ARC: __switch_to: move ksp to thread_info from thread_struct

task's arch specific bits are carried in 2 places
 - embedded thread_struct in task_struct
 - associated thread_info (hoisted in task's stack page) and
   syntactically: (thread_info *)(task_struct->stack)

ksp (dynamic kernel stack top) currently lives in thread_struct but
given its deep location in task struct likely to cache miss when
accessed from  __switch_to(). Moving it to thread_info would be more
efficient given proximity to frequently accessed items such as
preempt_count thus very likely to be in cache, specially in schedular
code.

Note however that currently tsk.thread.ksp takes 1 memory access (off
of tsk pointer) while new code tsk->stack.ksp would take 2, but likely
to be in cache. Moreover if task is current the 2nd reference can be
elided and instead derived from SP as (SP & ~(THREAD_SIZE - 1))

All of this also makes __switch_to() code simpler and we can see the 2
ways of retirving ksp (descrobed above) in new code.

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/asm-offsets.c |  2 +-
 arch/arc/kernel/ctx_sw_asm.S  | 22 ++++++++++------------
 arch/arc/kernel/process.c     |  6 +++---
 3 files changed, 14 insertions(+), 16 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/asm-offsets.c b/arch/arc/kernel/asm-offsets.c
index 37324fd9a72f..5b47d09b925e 100644
--- a/arch/arc/kernel/asm-offsets.c
+++ b/arch/arc/kernel/asm-offsets.c
@@ -20,13 +20,13 @@ int main(void)
 
 	BLANK();
 
-	DEFINE(THREAD_KSP, offsetof(struct thread_struct, ksp));
 	DEFINE(THREAD_CALLEE_REG, offsetof(struct thread_struct, callee_reg));
 	DEFINE(THREAD_FAULT_ADDR,
 	       offsetof(struct thread_struct, fault_address));
 
 	BLANK();
 
+	DEFINE(THREAD_INFO_KSP, offsetof(struct thread_info, ksp));
 	DEFINE(THREAD_INFO_FLAGS, offsetof(struct thread_info, flags));
 	DEFINE(THREAD_INFO_PREEMPT_COUNT,
 	       offsetof(struct thread_info, preempt_count));
diff --git a/arch/arc/kernel/ctx_sw_asm.S b/arch/arc/kernel/ctx_sw_asm.S
index 59d779004e64..48e1f21976ed 100644
--- a/arch/arc/kernel/ctx_sw_asm.S
+++ b/arch/arc/kernel/ctx_sw_asm.S
@@ -11,8 +11,6 @@
 #include <asm/entry.h>       /* For the SAVE_* macros */
 #include <asm/asm-offsets.h>
 
-#define KSP_WORD_OFF 	((TASK_THREAD + THREAD_KSP) / 4)
-
 ; IN
 ;  - r0: prev task (also current)
 ;  - r1: next task
@@ -37,19 +35,19 @@ ENTRY_CFI(__switch_to)
 	/* kernel mode callee regs of @prev */
 	SAVE_CALLEE_SAVED_KERNEL
 
-	/* save final SP to @prev->thread.ksp */
-#if KSP_WORD_OFF  <= 255
-	st.as  sp, [r0, KSP_WORD_OFF]
-#else
-	/* Workaround for NR_CPUS=4k as ST.as can only take s9 offset */
-	add2	r10, r0, KSP_WORD_OFF
-	st	sp, [r10]
-#endif
+	/*
+	 * save final SP to @prev->thread_info.ksp
+	 * @prev is "current" so thread_info derived from SP
+	 */
+	GET_CURR_THR_INFO_FROM_SP  r10
+	st	sp,  [r10, THREAD_INFO_KSP]
+
 	/* update @next in _current_task[] and GP register caching it */
 	SET_CURR_TASK_ON_CPU  r1, r10
 
-	/* load SP from @next->thread.ksp */
-	ld.as	sp, [r1, KSP_WORD_OFF]
+	/* load SP from @next->thread_info.ksp */
+	ld	r10, [r1, TASK_THREAD_INFO]
+	ld	sp,  [r10, THREAD_INFO_KSP]
 
 	/* restore callee regs, stack frame regs of @next */
 	RESTORE_CALLEE_SAVED_KERNEL
diff --git a/arch/arc/kernel/process.c b/arch/arc/kernel/process.c
index 96f591508142..186ceab661eb 100644
--- a/arch/arc/kernel/process.c
+++ b/arch/arc/kernel/process.c
@@ -141,7 +141,7 @@ asmlinkage void ret_from_fork(void);
  * |    unused      |
  * |                |
  * ------------------
- * |     r25        |   <==== top of Stack (thread.ksp)
+ * |     r25        |   <==== top of Stack (thread_info.ksp)
  * ~                ~
  * |    --to--      |   (CALLEE Regs of kernel mode)
  * |     r13        |
@@ -181,14 +181,14 @@ int copy_thread(struct task_struct *p, const struct kernel_clone_args *args)
 	c_callee = ((struct callee_regs *)childksp) - 1;
 
 	/*
-	 * __switch_to() uses thread.ksp to start unwinding stack
+	 * __switch_to() uses thread_info.ksp to start unwinding stack
 	 * For kernel threads we don't need to create callee regs, the
 	 * stack layout nevertheless needs to remain the same.
 	 * Also, since __switch_to anyways unwinds callee regs, we use
 	 * this to populate kernel thread entry-pt/args into callee regs,
 	 * so that ret_from_kernel_thread() becomes simpler.
 	 */
-	p->thread.ksp = (unsigned long)c_callee;	/* THREAD_KSP */
+	task_thread_info(p)->ksp = (unsigned long)c_callee;	/* THREAD_INFO_KSP */
 
 	/* __switch_to expects FP(0), BLINK(return addr) at top */
 	childksp[0] = 0;			/* fp */
-- 
cgit v1.2.3


From c505b0da76a67139e073a5a5c4a1986b1cf168d3 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Tue, 19 May 2020 16:01:52 -0700
Subject: ARC: entry: rework (non-functional)

 - comments update
 - rename syscall_trace_entry
 - use PT_xxx in entry code

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/asm-offsets.c |  1 +
 arch/arc/kernel/entry.S       | 45 +++++++++++++++++++------------------------
 arch/arc/kernel/ptrace.c      |  2 +-
 3 files changed, 22 insertions(+), 26 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/asm-offsets.c b/arch/arc/kernel/asm-offsets.c
index 5b47d09b925e..e46688975868 100644
--- a/arch/arc/kernel/asm-offsets.c
+++ b/arch/arc/kernel/asm-offsets.c
@@ -47,6 +47,7 @@ int main(void)
 
 	DEFINE(PT_status32, offsetof(struct pt_regs, status32));
 	DEFINE(PT_event, offsetof(struct pt_regs, event));
+	DEFINE(PT_bta, offsetof(struct pt_regs, bta));
 	DEFINE(PT_sp, offsetof(struct pt_regs, sp));
 	DEFINE(PT_r0, offsetof(struct pt_regs, r0));
 	DEFINE(PT_r1, offsetof(struct pt_regs, r1));
diff --git a/arch/arc/kernel/entry.S b/arch/arc/kernel/entry.S
index cd26e0fa5044..f291fc8476d7 100644
--- a/arch/arc/kernel/entry.S
+++ b/arch/arc/kernel/entry.S
@@ -160,20 +160,19 @@ END(EV_Extension)
 ; syscall Tracing
 ; ---------------------------------------------
 tracesys:
-	; save EFA in case tracer wants the PC of traced task
-	; using ERET won't work since next-PC has already committed
+	; safekeep EFA (r12) if syscall tracer wanted PC
+	; for traps, ERET is pre-commit so points to next-PC
 	GET_CURR_TASK_FIELD_PTR   TASK_THREAD, r11
 	st  r12, [r11, THREAD_FAULT_ADDR]	; thread.fault_address
 
-	; PRE Sys Call Ptrace hook
-	mov r0, sp			; pt_regs needed
-	bl  @syscall_trace_entry
+	; PRE syscall trace hook
+	mov r0, sp				; pt_regs
+	bl  @syscall_trace_enter
 
 	; Tracing code now returns the syscall num (orig or modif)
 	mov r8, r0
 
 	; Do the Sys Call as we normally would.
-	; Validate the Sys Call number
 	cmp     r8,  NR_syscalls - 1
 	mov.hi  r0, -ENOSYS
 	bhi     tracesys_exit
@@ -190,36 +189,36 @@ tracesys:
 	ld  r6, [sp, PT_r6]
 	ld  r7, [sp, PT_r7]
 	ld.as   r9, [sys_call_table, r8]
-	jl      [r9]        ; Entry into Sys Call Handler
+	jl      [r9]
 
 tracesys_exit:
-	st  r0, [sp, PT_r0]     ; sys call return value in pt_regs
+	st  r0, [sp, PT_r0]
 
-	;POST Sys Call Ptrace Hook
+	; POST syscall trace hook
 	mov r0, sp		; pt_regs needed
 	bl  @syscall_trace_exit
-	b   ret_from_exception ; NOT ret_from_system_call at is saves r0 which
-	; we'd done before calling post hook above
+
+	; don't call ret_from_system_call as it saves r0, already done above
+	b   ret_from_exception
 
 ; ---------------------------------------------
 ; Breakpoint TRAP
 ; ---------------------------------------------
 trap_with_param:
 	mov r0, r12	; EFA in case ptracer/gdb wants stop_pc
-	mov r1, sp
+	mov r1, sp	; pt_regs
 
-	; Save callee regs in case gdb wants to have a look
-	; SP will grow up by size of CALLEE Reg-File
+	; save callee regs in case tracer/gdb wants to peek
 	SAVE_CALLEE_SAVED_USER
 
-	; save location of saved Callee Regs @ thread_struct->pc
+	; safekeep ref to callee regs
 	GET_CURR_TASK_FIELD_PTR   TASK_THREAD, r10
 	st  sp, [r10, THREAD_CALLEE_REG]
 
-	; Call the trap handler
+	; call the non syscall trap handler
 	bl  do_non_swi_trap
 
-	; unwind stack to discard Callee saved Regs
+	; unwind stack to discard callee regs
 	DISCARD_CALLEE_SAVED_USER
 
 	b   ret_from_exception
@@ -237,31 +236,27 @@ ENTRY(EV_Trap)
 
 	FAKE_RET_FROM_EXCPN
 
-	;============ TRAP 1   :breakpoints
-	; Check ECR for trap with arg (PROLOGUE ensures r10 has ECR)
+	;============ TRAP N : breakpoints, kprobes etc
 	bmsk.f 0, r10, 7
 	bnz    trap_with_param
 
-	;============ TRAP  (no param): syscall top level
+	;============ TRAP 0 (no param): syscall
 
-	; If syscall tracing ongoing, invoke pre-post-hooks
+	; syscall tracing ongoing, invoke pre-post-hooks around syscall
 	GET_CURR_THR_INFO_FLAGS   r10
 	and.f 0, r10, _TIF_SYSCALL_WORK
 	bnz   tracesys  ; this never comes back
 
 	;============ Normal syscall case
 
-	; syscall num shd not exceed the total system calls avail
 	cmp     r8,  NR_syscalls - 1
 	mov.hi  r0, -ENOSYS
 	bhi     .Lret_from_system_call
 
-	; Offset into the syscall_table and call handler
 	ld.as   r9,[sys_call_table, r8]
-	jl      [r9]        ; Entry into Sys Call Handler
+	jl      [r9]
 
 .Lret_from_system_call:
-
 	st  r0, [sp, PT_r0]     ; sys call return value in pt_regs
 
 	; fall through to ret_from_exception
diff --git a/arch/arc/kernel/ptrace.c b/arch/arc/kernel/ptrace.c
index 8226df3163fe..14ea7406f5cd 100644
--- a/arch/arc/kernel/ptrace.c
+++ b/arch/arc/kernel/ptrace.c
@@ -339,7 +339,7 @@ long arch_ptrace(struct task_struct *child, long request,
 	return ret;
 }
 
-asmlinkage int syscall_trace_entry(struct pt_regs *regs)
+asmlinkage int syscall_trace_enter(struct pt_regs *regs)
 {
 	if (test_thread_flag(TIF_SYSCALL_TRACE))
 		if (ptrace_report_syscall_entry(regs))
-- 
cgit v1.2.3


From 6b606c8d6625aeda0b526cb687367f72bb98cd30 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Wed, 20 May 2020 00:47:20 -0700
Subject: ARC: entry: ARcompact EV_ProtV to use r10 directly

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/entry-compact.S | 6 ++----
 1 file changed, 2 insertions(+), 4 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/entry-compact.S b/arch/arc/kernel/entry-compact.S
index 5cb0cd7e4eab..77f0090554c5 100644
--- a/arch/arc/kernel/entry-compact.S
+++ b/arch/arc/kernel/entry-compact.S
@@ -254,9 +254,8 @@ END(handle_interrupt_level1)
 
 ENTRY(EV_TLBProtV)
 
-	EXCEPTION_PROLOGUE
+	EXCEPTION_PROLOGUE	; ECR returned in r10
 
-	mov r2, r10	; ECR set into r10 already
 	lr  r0, [efa]	; Faulting Data address (not part of pt_regs saved above)
 
 	; Exception auto-disables further Intr/exceptions.
@@ -273,8 +272,7 @@ ENTRY(EV_TLBProtV)
 	;   -Access Violation	: 00_23_(00|01|02|03)_00
 	;			         x  r  w  r+w
 	;   -Unaligned Access	: 00_23_04_00
-	;
-	bbit1 r2, ECR_C_BIT_PROTV_MISALIG_DATA, 4f
+	bbit1 r10, ECR_C_BIT_PROTV_MISALIG_DATA, 4f
 
 	;========= (6a) Access Violation Processing ========
 	bl  do_page_fault
-- 
cgit v1.2.3


From 0e93ecaeebd5d0f189d1fa50a614ac0816a8c703 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Wed, 20 May 2020 00:52:45 -0700
Subject: ARC: entry: EV_MachineCheck dont re-read ECR

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/entry.S | 5 ++---
 1 file changed, 2 insertions(+), 3 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/entry.S b/arch/arc/kernel/entry.S
index f291fc8476d7..46582fbebcf2 100644
--- a/arch/arc/kernel/entry.S
+++ b/arch/arc/kernel/entry.S
@@ -95,16 +95,15 @@ END(instr_service)
 
 ENTRY(EV_MachineCheck)
 
-	EXCEPTION_PROLOGUE
+	EXCEPTION_PROLOGUE	; ECR returned in r10
 
-	lr  r2, [ecr]
 	lr  r0, [efa]
 	mov r1, sp
 
 	; MC excpetions disable MMU
 	ARC_MMU_REENABLE r3
 
-	lsr  	r3, r2, 8
+	lsr  	r3, r10, 8
 	bmsk 	r3, r3, 7
 	brne    r3, ECR_C_MCHK_DUP_TLB, 1f
 
-- 
cgit v1.2.3


From 13347c10396055c4c6c38a54d10bc6ed5024fbe9 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Wed, 20 May 2020 00:39:09 -0700
Subject: ARC: entry: Add more common chores to EXCEPTION_PROLOGUE

THe high level structure of most ARC exception handlers is
 1. save regfile with EXCEPTION_PROLOGUE
 2. setup r0: EFA (not part of pt_regs)
 3. setup r1: pointer to pt_regs (SP)
 4. drop down to pure kernel mode (from exception)
 5. call the Linux "C" handler

Remove the boiler plate code by moving #2, #3, #4 into #1.

The exceptions to most exceptions are syscall Trap and Machine check
which don't do some of above for various reasons, so call a newly
introduced variant EXCEPTION_PROLOGUE_KEEP_AE (same as original
EXCEPTION_PROLOGUE)

Tested-by: Pavel Kozlov <Pavel.Kozlov@synopsys.com>
Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/entry-arcv2.S   | 15 ---------------
 arch/arc/kernel/entry-compact.S | 13 -------------
 arch/arc/kernel/entry.S         | 19 ++-----------------
 3 files changed, 2 insertions(+), 45 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/entry-arcv2.S b/arch/arc/kernel/entry-arcv2.S
index a7e6a2174187..2e49c81c8086 100644
--- a/arch/arc/kernel/entry-arcv2.S
+++ b/arch/arc/kernel/entry-arcv2.S
@@ -125,11 +125,6 @@ ENTRY(mem_service)
 
 	EXCEPTION_PROLOGUE
 
-	lr  r0, [efa]
-	mov r1, sp
-
-	FAKE_RET_FROM_EXCPN
-
 	bl  do_memory_error
 	b   ret_from_exception
 END(mem_service)
@@ -138,11 +133,6 @@ ENTRY(EV_Misaligned)
 
 	EXCEPTION_PROLOGUE
 
-	lr  r0, [efa]	; Faulting Data address
-	mov r1, sp
-
-	FAKE_RET_FROM_EXCPN
-
 	SAVE_CALLEE_SAVED_USER
 	mov r2, sp              ; callee_regs
 
@@ -163,11 +153,6 @@ ENTRY(EV_TLBProtV)
 
 	EXCEPTION_PROLOGUE
 
-	lr  r0, [efa]	; Faulting Data address
-	mov r1, sp	; pt_regs
-
-	FAKE_RET_FROM_EXCPN
-
 	mov blink, ret_from_exception
 	b   do_page_fault
 
diff --git a/arch/arc/kernel/entry-compact.S b/arch/arc/kernel/entry-compact.S
index 77f0090554c5..774c03cc1d1a 100644
--- a/arch/arc/kernel/entry-compact.S
+++ b/arch/arc/kernel/entry-compact.S
@@ -256,16 +256,6 @@ ENTRY(EV_TLBProtV)
 
 	EXCEPTION_PROLOGUE	; ECR returned in r10
 
-	lr  r0, [efa]	; Faulting Data address (not part of pt_regs saved above)
-
-	; Exception auto-disables further Intr/exceptions.
-	; Re-enable them by pretending to return from exception
-	; (so rest of handler executes in pure K mode)
-
-	FAKE_RET_FROM_EXCPN
-
-	mov   r1, sp	; Handle to pt_regs
-
 	;------ (5) Type of Protection Violation? ----------
 	;
 	; ProtV Hardware Exception is triggered for Access Faults of 2 types
@@ -301,9 +291,6 @@ END(EV_TLBProtV)
 ENTRY(call_do_page_fault)
 
 	EXCEPTION_PROLOGUE
-	lr  r0, [efa]	; Faulting Data address
-	mov   r1, sp
-	FAKE_RET_FROM_EXCPN
 
 	mov blink, ret_from_exception
 	b  do_page_fault
diff --git a/arch/arc/kernel/entry.S b/arch/arc/kernel/entry.S
index 46582fbebcf2..089f6680518f 100644
--- a/arch/arc/kernel/entry.S
+++ b/arch/arc/kernel/entry.S
@@ -80,11 +80,6 @@ ENTRY(instr_service)
 
 	EXCEPTION_PROLOGUE
 
-	lr  r0, [efa]
-	mov r1, sp
-
-	FAKE_RET_FROM_EXCPN
-
 	bl  do_insterror_or_kprobe
 	b   ret_from_exception
 END(instr_service)
@@ -95,7 +90,7 @@ END(instr_service)
 
 ENTRY(EV_MachineCheck)
 
-	EXCEPTION_PROLOGUE	; ECR returned in r10
+	EXCEPTION_PROLOGUE_KEEP_AE	; ECR returned in r10
 
 	lr  r0, [efa]
 	mov r1, sp
@@ -128,11 +123,6 @@ ENTRY(EV_PrivilegeV)
 
 	EXCEPTION_PROLOGUE
 
-	lr  r0, [efa]
-	mov r1, sp
-
-	FAKE_RET_FROM_EXCPN
-
 	bl  do_privilege_fault
 	b   ret_from_exception
 END(EV_PrivilegeV)
@@ -144,11 +134,6 @@ ENTRY(EV_Extension)
 
 	EXCEPTION_PROLOGUE
 
-	lr  r0, [efa]
-	mov r1, sp
-
-	FAKE_RET_FROM_EXCPN
-
 	bl  do_extension_fault
 	b   ret_from_exception
 END(EV_Extension)
@@ -229,7 +214,7 @@ trap_with_param:
 
 ENTRY(EV_Trap)
 
-	EXCEPTION_PROLOGUE
+	EXCEPTION_PROLOGUE_KEEP_AE
 
 	lr  r12, [efa]
 
-- 
cgit v1.2.3


From d4624bf6a6c9d9ff084eb2cba6d3cf6aeda9f974 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Fri, 22 May 2020 16:24:53 -0700
Subject: ARCv2: entry: rearrange pt_regs slightly

Instead of r26,fp,sp,r12,r30 order as fp,r30,r12,r26,sp

 - keeps SP at well known position (right abive hardware autosave)
 - r26,r12 saved specifically for ARCv2 (and not in ARCv3) kept
   closer for easy ifdef'ry later

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/asm-offsets.c | 7 ++++---
 1 file changed, 4 insertions(+), 3 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/asm-offsets.c b/arch/arc/kernel/asm-offsets.c
index e46688975868..478768c88f46 100644
--- a/arch/arc/kernel/asm-offsets.c
+++ b/arch/arc/kernel/asm-offsets.c
@@ -62,11 +62,9 @@ int main(void)
 	DEFINE(PT_r26, offsetof(struct pt_regs, r26));
 	DEFINE(PT_ret, offsetof(struct pt_regs, ret));
 	DEFINE(PT_blink, offsetof(struct pt_regs, blink));
+	OFFSET(PT_fp, pt_regs, fp);
 	DEFINE(PT_lpe, offsetof(struct pt_regs, lp_end));
 	DEFINE(PT_lpc, offsetof(struct pt_regs, lp_count));
-	DEFINE(SZ_CALLEE_REGS, sizeof(struct callee_regs));
-	DEFINE(SZ_PT_REGS, sizeof(struct pt_regs));
-
 #ifdef CONFIG_ISA_ARCV2
 	OFFSET(PT_r12, pt_regs, r12);
 	OFFSET(PT_r30, pt_regs, r30);
@@ -79,5 +77,8 @@ int main(void)
 	OFFSET(PT_DSP_CTRL, pt_regs, DSP_CTRL);
 #endif
 
+	DEFINE(SZ_CALLEE_REGS, sizeof(struct callee_regs));
+	DEFINE(SZ_PT_REGS, sizeof(struct pt_regs));
+
 	return 0;
 }
-- 
cgit v1.2.3


From 58d9ceb7d9f56bd74b8e904e26511d27a4220827 Mon Sep 17 00:00:00 2001
From: Vineet Gupta <vgupta@kernel.org>
Date: Thu, 21 May 2020 17:44:32 -0700
Subject: ARC: pt_regs: create seperate type for ecr

Reduces duplication in each ISA specific pt_regs

Tested-by: kernel test robot <lkp@intel.com>
Closes: https://lore.kernel.org/oe-kbuild-all/202308151342.ROQ9Urvv-lkp@intel.com
Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/kernel/asm-offsets.c  |  2 +-
 arch/arc/kernel/kgdb.c         |  2 +-
 arch/arc/kernel/ptrace.c       |  4 ++--
 arch/arc/kernel/traps.c        |  4 +---
 arch/arc/kernel/troubleshoot.c | 13 +++++++------
 5 files changed, 12 insertions(+), 13 deletions(-)

(limited to 'arch/arc/kernel')

diff --git a/arch/arc/kernel/asm-offsets.c b/arch/arc/kernel/asm-offsets.c
index 478768c88f46..f77deb799175 100644
--- a/arch/arc/kernel/asm-offsets.c
+++ b/arch/arc/kernel/asm-offsets.c
@@ -46,7 +46,7 @@ int main(void)
 	BLANK();
 
 	DEFINE(PT_status32, offsetof(struct pt_regs, status32));
-	DEFINE(PT_event, offsetof(struct pt_regs, event));
+	DEFINE(PT_event, offsetof(struct pt_regs, ecr));
 	DEFINE(PT_bta, offsetof(struct pt_regs, bta));
 	DEFINE(PT_sp, offsetof(struct pt_regs, sp));
 	DEFINE(PT_r0, offsetof(struct pt_regs, r0));
diff --git a/arch/arc/kernel/kgdb.c b/arch/arc/kernel/kgdb.c
index 345a0000554c..4f2b5951454f 100644
--- a/arch/arc/kernel/kgdb.c
+++ b/arch/arc/kernel/kgdb.c
@@ -175,7 +175,7 @@ void kgdb_trap(struct pt_regs *regs)
 	 * with trap_s 4 (compiled) breakpoints, continuation needs to
 	 * start after the breakpoint.
 	 */
-	if (regs->ecr_param == 3)
+	if (regs->ecr.param == 3)
 		instruction_pointer(regs) -= BREAK_INSTR_SIZE;
 
 	kgdb_handle_exception(1, SIGTRAP, 0, regs);
diff --git a/arch/arc/kernel/ptrace.c b/arch/arc/kernel/ptrace.c
index 14ea7406f5cd..e0c233c178b1 100644
--- a/arch/arc/kernel/ptrace.c
+++ b/arch/arc/kernel/ptrace.c
@@ -46,7 +46,7 @@ static const struct pt_regs_offset regoffset_table[] = {
 	REG_OFFSET_NAME(r0),
 	REG_OFFSET_NAME(sp),
 	REG_OFFSET_NAME(orig_r0),
-	REG_OFFSET_NAME(event),
+	REG_OFFSET_NAME(ecr),
 	REG_OFFSET_END,
 };
 
@@ -54,7 +54,7 @@ static const struct pt_regs_offset regoffset_table[] = {
 
 static const struct pt_regs_offset regoffset_table[] = {
 	REG_OFFSET_NAME(orig_r0),
-	REG_OFFSET_NAME(event),
+	REG_OFFSET_NAME(ecr),
 	REG_OFFSET_NAME(bta),
 	REG_OFFSET_NAME(r26),
 	REG_OFFSET_NAME(fp),
diff --git a/arch/arc/kernel/traps.c b/arch/arc/kernel/traps.c
index 2f7eb786695b..9b9570b79362 100644
--- a/arch/arc/kernel/traps.c
+++ b/arch/arc/kernel/traps.c
@@ -110,9 +110,7 @@ void do_machine_check_fault(unsigned long address, struct pt_regs *regs)
  */
 void do_non_swi_trap(unsigned long address, struct pt_regs *regs)
 {
-	unsigned int param = regs->ecr_param;
-
-	switch (param) {
+	switch (regs->ecr.param) {
 	case 1:
 		trap_is_brkpt(address, regs);
 		break;
diff --git a/arch/arc/kernel/troubleshoot.c b/arch/arc/kernel/troubleshoot.c
index 7654c2e42dc0..d5b3ed2c58f5 100644
--- a/arch/arc/kernel/troubleshoot.c
+++ b/arch/arc/kernel/troubleshoot.c
@@ -115,8 +115,8 @@ static void show_ecr_verbose(struct pt_regs *regs)
 	/* For Data fault, this is data address not inst