[PATCH 00/20] ARC updates

[PATCH 00/20] ARC updates

[Vineet Gupta]

Hi,

This is a pile of miscll improvements/updates sitting in one of my local trees.
Given the recent warning fix, we coudl also push them out.
@Alexey, @Shahab: care to give these a spin on hsdk (and test ARC700 build/boot on nSIM if possible).

Thx,
-Vineet

Vineet Gupta (20):
  ARC: uaccess: remove arc specific out-of-line handles for -Os
  ARC: uaccess: use optimized generic __strnlen_user/__strncpy_from_user
  ARC: uaccess: elide unaliged handling if hardware supports
  ARCv2: memset: don't prefetch for len == 0 which happens a alot
  ARC: boot log: eliminate struct cpuinfo_arc #1: mm
  ARC: boot log: eliminate struct cpuinfo_arc #2: cache
  ARC: boot log: eliminate struct cpuinfo_arc #3: don't export
  ARC: boot log: eliminate struct cpuinfo_arc #4: boot log per ISA
  ARC: entry: use gp to cache task pointer (vs. r25)
  ARC: kernel stack: INIT_THREAD need not setup @init_stack in @ksp
  ARC: __switch_to: asm with dwarf ops (vs. inline asm)
  ARC: __switch_to: move ksp to thread_info from thread_struct
  ARC: entry: rework (non-functional)
  ARC: entry: ARcompact EV_ProtV to use r10 directly
  ARC: entry: EV_MachineCheck dont re-read ECR
  ARC: entry: Add more common chores to EXCEPTION_PROLOGUE
  ARC: entry: replace 8 byte OR with 4 byte BSET
  ARC: entry: replace 8 byte ADD.ne with 4 byte ADD2.ne
  ARCv2: entry: rearrange pt_regs slightly
  ARC: pt_regs: create seperate type for ecr

 arch/arc/Kconfig                     |   8 +-
 arch/arc/Makefile                    |   6 +-
 arch/arc/include/asm/arcregs.h       |  99 ++---
 arch/arc/include/asm/current.h       |   2 +-
 arch/arc/include/asm/dwarf.h         |  32 +-
 arch/arc/include/asm/entry-arcv2.h   |  66 ++--
 arch/arc/include/asm/entry-compact.h |  50 ++-
 arch/arc/include/asm/entry.h         | 107 ++---
 arch/arc/include/asm/processor.h     |   7 +-
 arch/arc/include/asm/ptrace.h        |  62 ++-
 arch/arc/include/asm/setup.h         |   6 +-
 arch/arc/include/asm/thread_info.h   |  10 +-
 arch/arc/include/asm/uaccess.h       |  21 +-
 arch/arc/kernel/Makefile             |   9 +-
 arch/arc/kernel/asm-offsets.c        |  14 +-
 arch/arc/kernel/ctx_sw.c             | 112 ------
 arch/arc/kernel/ctx_sw_asm.S         |  76 ++--
 arch/arc/kernel/entry-arcv2.S        |  15 -
 arch/arc/kernel/entry-compact.S      |  16 +-
 arch/arc/kernel/entry.S              |  70 ++--
 arch/arc/kernel/kgdb.c               |   2 +-
 arch/arc/kernel/mcip.c               |   2 -
 arch/arc/kernel/process.c            |  17 +-
 arch/arc/kernel/ptrace.c             |   8 +-
 arch/arc/kernel/setup.c              | 561 ++++++++++++---------------
 arch/arc/kernel/traps.c              |   4 +-
 arch/arc/kernel/troubleshoot.c       |  13 +-
 arch/arc/lib/memset-archs.S          |   3 +-
 arch/arc/mm/cache.c                  | 171 +++-----
 arch/arc/mm/extable.c                |  11 -
 arch/arc/mm/fault.c                  |   6 +-
 arch/arc/mm/tlb.c                    |  94 ++---
 32 files changed, 667 insertions(+), 1013 deletions(-)
 delete mode 100644 arch/arc/kernel/ctx_sw.c

-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 01/20] ARC: uaccess: remove arc specific out-of-line handles for -Os

[Vineet Gupta]

Everything is now out-of-line in lib/usercopy.c

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/uaccess.h | 11 ++---------
 arch/arc/mm/extable.c          | 11 -----------
 2 files changed, 2 insertions(+), 20 deletions(-)

diff --git a/arch/arc/include/asm/uaccess.h b/arch/arc/include/asm/uaccess.h
index 99712471c96a..d2da159bb80a 100644
--- a/arch/arc/include/asm/uaccess.h
+++ b/arch/arc/include/asm/uaccess.h
@@ -584,7 +584,7 @@ raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 	return res;
 }
 
-static inline unsigned long __arc_clear_user(void __user *to, unsigned long n)
+static inline unsigned long __clear_user(void __user *to, unsigned long n)
 {
 	long res = n;
 	unsigned char *d_char = to;
@@ -626,17 +626,10 @@ static inline unsigned long __arc_clear_user(void __user *to, unsigned long n)
 	return res;
 }
 
-#ifndef CONFIG_CC_OPTIMIZE_FOR_SIZE
-
 #define INLINE_COPY_TO_USER
 #define INLINE_COPY_FROM_USER
 
-#define __clear_user(d, n)		__arc_clear_user(d, n)
-#else
-extern unsigned long arc_clear_user_noinline(void __user *to,
-		unsigned long n);
-#define __clear_user(d, n)		arc_clear_user_noinline(d, n)
-#endif
+#define __clear_user			__clear_user
 
 #include <asm-generic/uaccess.h>
 
diff --git a/arch/arc/mm/extable.c b/arch/arc/mm/extable.c
index 4e14c4244ea2..88fa3a4d4906 100644
--- a/arch/arc/mm/extable.c
+++ b/arch/arc/mm/extable.c
@@ -22,14 +22,3 @@ int fixup_exception(struct pt_regs *regs)
 
 	return 0;
 }
-
-#ifdef CONFIG_CC_OPTIMIZE_FOR_SIZE
-
-unsigned long arc_clear_user_noinline(void __user *to,
-		unsigned long n)
-{
-	return __arc_clear_user(to, n);
-}
-EXPORT_SYMBOL(arc_clear_user_noinline);
-
-#endif
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 02/20] ARC: uaccess: use optimized generic __strnlen_user/__strncpy_from_user

[Vineet Gupta]

The existing ARC variants have 2 issues
 - Use ZOL which may not be present in forthcoming architecture
 - Byte loop based vs. generic version which is word loop based

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

diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index 96cf8720bb93..47b4acc7d0c9 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -26,6 +26,8 @@ config ARC
 	select GENERIC_PENDING_IRQ if SMP
 	select GENERIC_SCHED_CLOCK
 	select GENERIC_SMP_IDLE_THREAD
+	select GENERIC_STRNCPY_FROM_USER if MMU
+	select GENERIC_STRNLEN_USER if MMU
 	select HAVE_ARCH_KGDB
 	select HAVE_ARCH_TRACEHOOK
 	select HAVE_ARCH_TRANSPARENT_HUGEPAGE if ARC_MMU_V4
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 03/20] ARC: uaccess: elide unaliged handling if hardware supports

[Vineet Gupta]

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/uaccess.h | 10 ++++++----
 1 file changed, 6 insertions(+), 4 deletions(-)

diff --git a/arch/arc/include/asm/uaccess.h b/arch/arc/include/asm/uaccess.h
index d2da159bb80a..1e8809ea000a 100644
--- a/arch/arc/include/asm/uaccess.h
+++ b/arch/arc/include/asm/uaccess.h
@@ -146,8 +146,9 @@ raw_copy_from_user(void *to, const void __user *from, unsigned long n)
 	if (n == 0)
 		return 0;
 
-	/* unaligned */
-	if (((unsigned long)to & 0x3) || ((unsigned long)from & 0x3)) {
+	/* fallback for unaligned access when hardware doesn't support */
+	if (!IS_ENABLED(CONFIG_ARC_USE_UNALIGNED_MEM_ACCESS) &&
+	     (((unsigned long)to & 0x3) || ((unsigned long)from & 0x3))) {
 
 		unsigned char tmp;
 
@@ -373,8 +374,9 @@ raw_copy_to_user(void __user *to, const void *from, unsigned long n)
 	if (n == 0)
 		return 0;
 
-	/* unaligned */
-	if (((unsigned long)to & 0x3) || ((unsigned long)from & 0x3)) {
+	/* fallback for unaligned access when hardware doesn't support */
+	if (!IS_ENABLED(CONFIG_ARC_USE_UNALIGNED_MEM_ACCESS) &&
+	     (((unsigned long)to & 0x3) || ((unsigned long)from & 0x3))) {
 
 		unsigned char tmp;
 
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 04/20] ARCv2: memset: don't prefetch for len == 0 which happens a alot

[Vineet Gupta]

This avoids potential "bleeding" when size == 0 as cache line would be
dirtied (and possibly fetched from other cores) and due to the same
reaons more optimal too.

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

diff --git a/arch/arc/lib/memset-archs.S b/arch/arc/lib/memset-archs.S
index d2e09fece5bc..d0a5cec4cdca 100644
--- a/arch/arc/lib/memset-archs.S
+++ b/arch/arc/lib/memset-archs.S
@@ -36,12 +36,13 @@
 #endif
 
 ENTRY_CFI(memset)
-	PREFETCHW_INSTR	r0, 0	; Prefetch the first write location
 	mov.f	0, r2
 ;;; if size is zero
 	jz.d	[blink]
 	mov	r3, r0		; don't clobber ret val
 
+	PREFETCHW_INSTR	r0, 0	; Prefetch the first write location
+
 ;;; if length < 8
 	brls.d.nt	r2, 8, .Lsmallchunk
 	mov.f	lp_count,r2
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 05/20] ARC: boot log: eliminate struct cpuinfo_arc #1: mm

[Vineet Gupta]

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()

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/arcregs.h | 27 +++++++---
 arch/arc/include/asm/setup.h   |  1 -
 arch/arc/kernel/setup.c        |  4 +-
 arch/arc/mm/tlb.c              | 92 +++++++++++++---------------------
 4 files changed, 57 insertions(+), 67 deletions(-)

diff --git a/arch/arc/include/asm/arcregs.h b/arch/arc/include/asm/arcregs.h
index 2162023195c5..af00cbe9b850 100644
--- a/arch/arc/include/asm/arcregs.h
+++ b/arch/arc/include/asm/arcregs.h
@@ -185,6 +185,27 @@ struct bcr_uarch_build_arcv2 {
 #endif
 };
 
+struct bcr_mmu_3 {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int ver:8, ways:4, sets:4, res:3, sasid:1, pg_sz:4,
+		     u_itlb:4, u_dtlb:4;
+#else
+	unsigned int u_dtlb:4, u_itlb:4, pg_sz:4, sasid:1, res:3, sets:4,
+		     ways:4, ver:8;
+#endif
+};
+
+struct bcr_mmu_4 {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int ver:8, sasid:1, sz1:4, sz0:4, res:2, pae:1,
+		     n_ways:2, n_entry:2, n_super:2, u_itlb:3, u_dtlb:3;
+#else
+	/*           DTLB      ITLB      JES        JE         JA      */
+	unsigned int u_dtlb:3, u_itlb:3, n_super:2, n_entry:2, n_ways:2,
+		     pae:1, res:2, sz0:4, sz1:4, sasid:1, ver:8;
+#endif
+};
+
 struct bcr_mpy {
 #ifdef CONFIG_CPU_BIG_ENDIAN
 	unsigned int pad:8, x1616:8, dsp:4, cycles:2, type:2, ver:8;
@@ -307,11 +328,6 @@ struct bcr_generic {
  * Generic structures to hold build configuration used at runtime
  */
 
-struct cpuinfo_arc_mmu {
-	unsigned int ver:4, pg_sz_k:8, s_pg_sz_m:8, pad:10, sasid:1, pae:1;
-	unsigned int sets:12, ways:4, u_dtlb:8, u_itlb:8;
-};
-
 struct cpuinfo_arc_cache {
 	unsigned int sz_k:14, line_len:8, assoc:4, alias:1, vipt:1, pad:4;
 };
@@ -326,7 +342,6 @@ struct cpuinfo_arc_ccm {
 
 struct cpuinfo_arc {
 	struct cpuinfo_arc_cache icache, dcache, slc;
-	struct cpuinfo_arc_mmu mmu;
 	struct cpuinfo_arc_bpu bpu;
 	struct bcr_identity core;
 	struct bcr_isa_arcv2 isa;
diff --git a/arch/arc/include/asm/setup.h b/arch/arc/include/asm/setup.h
index 374138832c5a..76443f198778 100644
--- a/arch/arc/include/asm/setup.h
+++ b/arch/arc/include/asm/setup.h
@@ -36,7 +36,6 @@ long __init arc_get_mem_sz(void);
 
 extern void arc_mmu_init(void);
 extern char *arc_mmu_mumbojumbo(int cpu_id, char *buf, int len);
-extern void read_decode_mmu_bcr(void);
 
 extern void arc_cache_init(void);
 extern char *arc_cache_mumbojumbo(int cpu_id, char *buf, int len);
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();
diff --git a/arch/arc/mm/tlb.c b/arch/arc/mm/tlb.c
index 2a3105a682c3..17e32c707367 100644
--- a/arch/arc/mm/tlb.c
+++ b/arch/arc/mm/tlb.c
@@ -18,7 +18,9 @@
 /* A copy of the ASID from the PID reg is kept in asid_cache */
 DEFINE_PER_CPU(unsigned int, asid_cache) = MM_CTXT_FIRST_CYCLE;
 
-static int __read_mostly pae_exists;
+static struct cpuinfo_arc_mmu {
+	unsigned int ver, pg_sz_k, s_pg_sz_m, pae, sets, ways;
+} mmuinfo;
 
 /*
  * Utility Routine to erase a J-TLB entry
@@ -131,7 +133,7 @@ static void tlb_entry_insert(unsigned int pd0, phys_addr_t pd1)
 
 noinline void local_flush_tlb_all(void)
 {
-	struct cpuinfo_arc_mmu *mmu = &cpuinfo_arc700[smp_processor_id()].mmu;
+	struct cpuinfo_arc_mmu *mmu = &mmuinfo;
 	unsigned long flags;
 	unsigned int entry;
 	int num_tlb = mmu->sets * mmu->ways;
@@ -560,89 +562,63 @@ void local_flush_pmd_tlb_range(struct vm_area_struct *vma, unsigned long start,
  * the cpuinfo structure for later use.
  * No Validation is done here, simply read/convert the BCRs
  */
-void read_decode_mmu_bcr(void)
+char *arc_mmu_mumbojumbo(int c, char *buf, int len)
 {
-	struct cpuinfo_arc_mmu *mmu = &cpuinfo_arc700[smp_processor_id()].mmu;
-	unsigned int tmp;
-	struct bcr_mmu_3 {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-	unsigned int ver:8, ways:4, sets:4, res:3, sasid:1, pg_sz:4,
-		     u_itlb:4, u_dtlb:4;
-#else
-	unsigned int u_dtlb:4, u_itlb:4, pg_sz:4, sasid:1, res:3, sets:4,
-		     ways:4, ver:8;
-#endif
-	} *mmu3;
-
-	struct bcr_mmu_4 {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-	unsigned int ver:8, sasid:1, sz1:4, sz0:4, res:2, pae:1,
-		     n_ways:2, n_entry:2, n_super:2, u_itlb:3, u_dtlb:3;
-#else
-	/*           DTLB      ITLB      JES        JE         JA      */
-	unsigned int u_dtlb:3, u_itlb:3, n_super:2, n_entry:2, n_ways:2,
-		     pae:1, res:2, sz0:4, sz1:4, sasid:1, ver:8;
-#endif
-	} *mmu4;
+	struct cpuinfo_arc_mmu *mmu = &mmuinfo;
+	unsigned int bcr, u_dtlb, u_itlb, sasid;
+	struct bcr_mmu_3 *mmu3;
+	struct bcr_mmu_4 *mmu4;
+	char super_pg[64] = "";
+	int n = 0;
 
-	tmp = read_aux_reg(ARC_REG_MMU_BCR);
-	mmu->ver = (tmp >> 24);
+	bcr = read_aux_reg(ARC_REG_MMU_BCR);
+	mmu->ver = (bcr >> 24);
 
 	if (is_isa_arcompact() && mmu->ver == 3) {
-		mmu3 = (struct bcr_mmu_3 *)&tmp;
+		mmu3 = (struct bcr_mmu_3 *)&bcr;
 		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;
+		u_dtlb = mmu3->u_dtlb;
+		u_itlb = mmu3->u_itlb;
+		sasid = mmu3->sasid;
 	} else {
-		mmu4 = (struct bcr_mmu_4 *)&tmp;
+		mmu4 = (struct bcr_mmu_4 *)&bcr;
 		mmu->pg_sz_k = 1 << (mmu4->sz0 - 1);
 		mmu->s_pg_sz_m = 1 << (mmu4->sz1 - 11);
 		mmu->sets = 64 << mmu4->n_entry;
 		mmu->ways = mmu4->n_ways * 2;
-		mmu->u_dtlb = mmu4->u_dtlb * 4;
-		mmu->u_itlb = mmu4->u_itlb * 4;
-		mmu->sasid = mmu4->sasid;
-		pae_exists = mmu->pae = mmu4->pae;
+		u_dtlb = mmu4->u_dtlb * 4;
+		u_itlb = mmu4->u_itlb * 4;
+		sasid = mmu4->sasid;
+		mmu->pae = mmu4->pae;
 	}
-}
 
-char *arc_mmu_mumbojumbo(int cpu_id, char *buf, int len)
-{
-	int n = 0;
-	struct cpuinfo_arc_mmu *p_mmu = &cpuinfo_arc700[cpu_id].mmu;
-	char super_pg[64] = "";
-
-	if (p_mmu->s_pg_sz_m)
-		scnprintf(super_pg, 64, "%dM Super Page %s",
-			  p_mmu->s_pg_sz_m,
-			  IS_USED_CFG(CONFIG_TRANSPARENT_HUGEPAGE));
+	if (mmu->s_pg_sz_m)
+		scnprintf(super_pg, 64, "/%dM%s",
+			  mmu->s_pg_sz_m,
+			  IS_ENABLED(CONFIG_TRANSPARENT_HUGEPAGE) ? " (THP enabled)":"");
 
 	n += scnprintf(buf + n, len - n,
-		      "MMU [v%x]\t: %dk PAGE, %s, swalk %d lvl, JTLB %d (%dx%d), uDTLB %d, uITLB %d%s%s\n",
-		       p_mmu->ver, p_mmu->pg_sz_k, super_pg,  CONFIG_PGTABLE_LEVELS,
-		       p_mmu->sets * p_mmu->ways, p_mmu->sets, p_mmu->ways,
-		       p_mmu->u_dtlb, p_mmu->u_itlb,
-		       IS_AVAIL2(p_mmu->pae, ", PAE40 ", CONFIG_ARC_HAS_PAE40));
+		      "MMU [v%x]\t: %dk%s, swalk %d lvl, JTLB %dx%d, uDTLB %d, uITLB %d%s%s\n",
+		       mmu->ver, mmu->pg_sz_k, super_pg, CONFIG_PGTABLE_LEVELS,
+		       mmu->sets, mmu->ways,
+		       u_dtlb, u_itlb,
+		       IS_AVAIL2(mmu->pae, ", PAE40 ", CONFIG_ARC_HAS_PAE40));
 
 	return buf;
 }
 
 int pae40_exist_but_not_enab(void)
 {
-	return pae_exists && !is_pae40_enabled();
+	return mmuinfo.pae && !is_pae40_enabled();
 }
 
 void arc_mmu_init(void)
 {
-	struct cpuinfo_arc_mmu *mmu = &cpuinfo_arc700[smp_processor_id()].mmu;
-	char str[256];
+	struct cpuinfo_arc_mmu *mmu = &mmuinfo;
 	int compat = 0;
 
-	pr_info("%s", arc_mmu_mumbojumbo(0, str, sizeof(str)));
-
 	/*
 	 * Can't be done in processor.h due to header include dependencies
 	 */
@@ -719,7 +695,7 @@ volatile int dup_pd_silent; /* Be silent abt it or complain (default) */
 void do_tlb_overlap_fault(unsigned long cause, unsigned long address,
 			  struct pt_regs *regs)
 {
-	struct cpuinfo_arc_mmu *mmu = &cpuinfo_arc700[smp_processor_id()].mmu;
+	struct cpuinfo_arc_mmu *mmu = &mmuinfo;
 	unsigned long flags;
 	int set, n_ways = mmu->ways;
 
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 06/20] ARC: boot log: eliminate struct cpuinfo_arc #2: cache

[Vineet Gupta]

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/arcregs.h |  37 ++++++-
 arch/arc/include/asm/setup.h   |   1 -
 arch/arc/kernel/setup.c        |   3 +-
 arch/arc/mm/cache.c            | 171 ++++++++++++---------------------
 4 files changed, 97 insertions(+), 115 deletions(-)

diff --git a/arch/arc/include/asm/arcregs.h b/arch/arc/include/asm/arcregs.h
index af00cbe9b850..cb1ad1bb4ece 100644
--- a/arch/arc/include/asm/arcregs.h
+++ b/arch/arc/include/asm/arcregs.h
@@ -206,6 +206,38 @@ struct bcr_mmu_4 {
 #endif
 };
 
+struct bcr_cache {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int pad:12, line_len:4, sz:4, config:4, ver:8;
+#else
+	unsigned int ver:8, config:4, sz:4, line_len:4, pad:12;
+#endif
+};
+
+struct bcr_slc_cfg {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int pad:24, way:2, lsz:2, sz:4;
+#else
+	unsigned int sz:4, lsz:2, way:2, pad:24;
+#endif
+};
+
+struct bcr_clust_cfg {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int pad:7, c:1, num_entries:8, num_cores:8, ver:8;
+#else
+	unsigned int ver:8, num_cores:8, num_entries:8, c:1, pad:7;
+#endif
+};
+
+struct bcr_volatile {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+	unsigned int start:4, limit:4, pad:22, order:1, disable:1;
+#else
+	unsigned int disable:1, order:1, pad:22, limit:4, start:4;
+#endif
+};
+
 struct bcr_mpy {
 #ifdef CONFIG_CPU_BIG_ENDIAN
 	unsigned int pad:8, x1616:8, dsp:4, cycles:2, type:2, ver:8;
@@ -328,10 +360,6 @@ struct bcr_generic {
  * Generic structures to hold build configuration used at runtime
  */
 
-struct cpuinfo_arc_cache {
-	unsigned int sz_k:14, line_len:8, assoc:4, alias:1, vipt:1, pad:4;
-};
-
 struct cpuinfo_arc_bpu {
 	unsigned int ver, full, num_cache, num_pred, ret_stk;
 };
@@ -341,7 +369,6 @@ struct cpuinfo_arc_ccm {
 };
 
 struct cpuinfo_arc {
-	struct cpuinfo_arc_cache icache, dcache, slc;
 	struct cpuinfo_arc_bpu bpu;
 	struct bcr_identity core;
 	struct bcr_isa_arcv2 isa;
diff --git a/arch/arc/include/asm/setup.h b/arch/arc/include/asm/setup.h
index 76443f198778..4c0bacd0ff5c 100644
--- a/arch/arc/include/asm/setup.h
+++ b/arch/arc/include/asm/setup.h
@@ -39,7 +39,6 @@ extern char *arc_mmu_mumbojumbo(int cpu_id, char *buf, int len);
 
 extern void arc_cache_init(void);
 extern char *arc_cache_mumbojumbo(int cpu_id, char *buf, int len);
-extern void read_decode_cache_bcr(void);
 
 extern void __init handle_uboot_args(void);
 
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();
diff --git a/arch/arc/mm/cache.c b/arch/arc/mm/cache.c
index bdaa4aa40947..7197bb845a40 100644
--- a/arch/arc/mm/cache.c
+++ b/arch/arc/mm/cache.c
@@ -28,6 +28,10 @@ int slc_enable = 1, ioc_enable = 1;
 unsigned long perip_base = ARC_UNCACHED_ADDR_SPACE; /* legacy value for boot */
 unsigned long perip_end = 0xFFFFFFFF; /* legacy value */
 
+static struct cpuinfo_arc_cache {
+	unsigned int sz_k, line_len, colors;
+} ic_info, dc_info, slc_info;
+
 void (*_cache_line_loop_ic_fn)(phys_addr_t paddr, unsigned long vaddr,
 			       unsigned long sz, const int op, const int full_page);
 
@@ -35,78 +39,24 @@ void (*__dma_cache_wback_inv)(phys_addr_t start, unsigned long sz);
 void (*__dma_cache_inv)(phys_addr_t start, unsigned long sz);
 void (*__dma_cache_wback)(phys_addr_t start, unsigned long sz);
 
-char *arc_cache_mumbojumbo(int c, char *buf, int len)
+static char *read_decode_cache_bcr_arcv2(int c, char *buf, int len)
 {
-	int n = 0;
-	struct cpuinfo_arc_cache *p;
-
-#define PR_CACHE(p, cfg, str)						\
-	if (!(p)->line_len)						\
-		n += scnprintf(buf + n, len - n, str"\t\t: N/A\n");	\
-	else								\
-		n += scnprintf(buf + n, len - n,			\
-			str"\t\t: %uK, %dway/set, %uB Line, %s%s%s\n",	\
-			(p)->sz_k, (p)->assoc, (p)->line_len,		\
-			(p)->vipt ? "VIPT" : "PIPT",			\
-			(p)->alias ? " aliasing" : "",			\
-			IS_USED_CFG(cfg));
-
-	PR_CACHE(&cpuinfo_arc700[c].icache, CONFIG_ARC_HAS_ICACHE, "I-Cache");
-	PR_CACHE(&cpuinfo_arc700[c].dcache, CONFIG_ARC_HAS_DCACHE, "D-Cache");
-
-	p = &cpuinfo_arc700[c].slc;
-	if (p->line_len)
-		n += scnprintf(buf + n, len - n,
-			       "SLC\t\t: %uK, %uB Line%s\n",
-			       p->sz_k, p->line_len, IS_USED_RUN(slc_enable));
-
-	n += scnprintf(buf + n, len - n, "Peripherals\t: %#lx%s%s\n",
-		       perip_base,
-		       IS_AVAIL3(ioc_exists, ioc_enable, ", IO-Coherency (per-device) "));
-
-	return buf;
-}
-
-/*
- * Read the Cache Build Confuration Registers, Decode them and save into
- * the cpuinfo structure for later use.
- * No Validation done here, simply read/convert the BCRs
- */
-static void read_decode_cache_bcr_arcv2(int cpu)
-{
-	struct cpuinfo_arc_cache *p_slc = &cpuinfo_arc700[cpu].slc;
+	struct cpuinfo_arc_cache *p_slc = &slc_info;
+	struct bcr_identity ident;
 	struct bcr_generic sbcr;
-
-	struct bcr_slc_cfg {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-		unsigned int pad:24, way:2, lsz:2, sz:4;
-#else
-		unsigned int sz:4, lsz:2, way:2, pad:24;
-#endif
-	} slc_cfg;
-
-	struct bcr_clust_cfg {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-		unsigned int pad:7, c:1, num_entries:8, num_cores:8, ver:8;
-#else
-		unsigned int ver:8, num_cores:8, num_entries:8, c:1, pad:7;
-#endif
-	} cbcr;
-
-	struct bcr_volatile {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-		unsigned int start:4, limit:4, pad:22, order:1, disable:1;
-#else
-		unsigned int disable:1, order:1, pad:22, limit:4, start:4;
-#endif
-	} vol;
-
+	struct bcr_clust_cfg cbcr;
+	struct bcr_volatile vol;
+	int n = 0;
 
 	READ_BCR(ARC_REG_SLC_BCR, sbcr);
 	if (sbcr.ver) {
+		struct bcr_slc_cfg  slc_cfg;
 		READ_BCR(ARC_REG_SLC_CFG, slc_cfg);
 		p_slc->sz_k = 128 << slc_cfg.sz;
 		l2_line_sz = p_slc->line_len = (slc_cfg.lsz == 0) ? 128 : 64;
+		n += scnprintf(buf + n, len - n,
+			       "SLC\t\t: %uK, %uB Line%s\n",
+			       p_slc->sz_k, p_slc->line_len, IS_USED_RUN(slc_enable));
 	}
 
 	READ_BCR(ARC_REG_CLUSTER_BCR, cbcr);
@@ -129,70 +79,83 @@ static void read_decode_cache_bcr_arcv2(int cpu)
 		ioc_enable = 0;
 	}
 
+	READ_BCR(AUX_IDENTITY, ident);
+
 	/* HS 2.0 didn't have AUX_VOL */
-	if (cpuinfo_arc700[cpu].core.family > 0x51) {
+	if (ident.family > 0x51) {
 		READ_BCR(AUX_VOL, vol);
 		perip_base = vol.start << 28;
 		/* HS 3.0 has limit and strict-ordering fields */
-		if (cpuinfo_arc700[cpu].core.family > 0x52)
+		if (ident.family > 0x52)
 			perip_end = (vol.limit << 28) - 1;
 	}
+
+	n += scnprintf(buf + n, len - n, "Peripherals\t: %#lx%s%s\n",
+		       perip_base,
+		       IS_AVAIL3(ioc_exists, ioc_enable, ", IO-Coherency (per-device) "));
+
+	return buf;
 }
 
-void read_decode_cache_bcr(void)
+char *arc_cache_mumbojumbo(int c, char *buf, int len)
 {
-	struct cpuinfo_arc_cache *p_ic, *p_dc;
-	unsigned int cpu = smp_processor_id();
-	struct bcr_cache {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-		unsigned int pad:12, line_len:4, sz:4, config:4, ver:8;
-#else
-		unsigned int ver:8, config:4, sz:4, line_len:4, pad:12;
-#endif
-	} ibcr, dbcr;
+	struct cpuinfo_arc_cache *p_ic = &ic_info, *p_dc = &dc_info;
+	struct bcr_cache ibcr, dbcr;
+	int vipt, assoc;
+	int n = 0;
 
-	p_ic = &cpuinfo_arc700[cpu].icache;
 	READ_BCR(ARC_REG_IC_BCR, ibcr);
-
 	if (!ibcr.ver)
 		goto dc_chk;
 
-	if (ibcr.ver <= 3) {
+	if (is_isa_arcompact() && (ibcr.ver <= 3)) {
 		BUG_ON(ibcr.config != 3);
-		p_ic->assoc = 2;		/* Fixed to 2w set assoc */
-	} else if (ibcr.ver >= 4) {
-		p_ic->assoc = 1 << ibcr.config;	/* 1,2,4,8 */
+		assoc = 2;		/* Fixed to 2w set assoc */
+	} else if (is_isa_arcv2() && (ibcr.ver >= 4)) {
+		assoc = 1 << ibcr.config;	/* 1,2,4,8 */
 	}
 
 	p_ic->line_len = 8 << ibcr.line_len;
 	p_ic->sz_k = 1 << (ibcr.sz - 1);
-	p_ic->vipt = 1;
-	p_ic->alias = p_ic->sz_k/p_ic->assoc/TO_KB(PAGE_SIZE) > 1;
+	p_ic->colors = p_ic->sz_k/assoc/TO_KB(PAGE_SIZE);
+
+	n += scnprintf(buf + n, len - n,
+			"I-Cache\t\t: %uK, %dway/set, %uB Line, VIPT%s%s\n",
+			p_ic->sz_k, assoc, p_ic->line_len,
+			p_ic->colors > 1 ? " aliasing" : "",
+			IS_USED_CFG(CONFIG_ARC_HAS_ICACHE));
 
 dc_chk:
-	p_dc = &cpuinfo_arc700[cpu].dcache;
 	READ_BCR(ARC_REG_DC_BCR, dbcr);
-
 	if (!dbcr.ver)
 		goto slc_chk;
 
-	if (dbcr.ver <= 3) {
+	if (is_isa_arcompact() && (dbcr.ver <= 3)) {
 		BUG_ON(dbcr.config != 2);
-		p_dc->assoc = 4;		/* Fixed to 4w set assoc */
-		p_dc->vipt = 1;
-		p_dc->alias = p_dc->sz_k/p_dc->assoc/TO_KB(PAGE_SIZE) > 1;
-	} else if (dbcr.ver >= 4) {
-		p_dc->assoc = 1 << dbcr.config;	/* 1,2,4,8 */
-		p_dc->vipt = 0;
-		p_dc->alias = 0;		/* PIPT so can't VIPT alias */
+		vipt = 1;
+		assoc = 4;		/* Fixed to 4w set assoc */
+		p_dc->colors = p_dc->sz_k/assoc/TO_KB(PAGE_SIZE);
+	} else if (is_isa_arcv2() && (dbcr.ver >= 4)) {
+		vipt = 0;
+		assoc = 1 << dbcr.config;	/* 1,2,4,8 */
+		p_dc->colors = 1;		/* PIPT so can't VIPT alias */
 	}
 
 	p_dc->line_len = 16 << dbcr.line_len;
 	p_dc->sz_k = 1 << (dbcr.sz - 1);
 
+	n += scnprintf(buf + n, len - n,
+			"D-Cache\t\t: %uK, %dway/set, %uB Line, %s%s%s\n",
+			p_dc->sz_k, assoc, p_dc->line_len,
+			vipt ? "VIPT" : "PIPT",
+			p_dc->colors > 1 ? " aliasing" : "",
+			IS_USED_CFG(CONFIG_ARC_HAS_DCACHE));
+
 slc_chk:
 	if (is_isa_arcv2())
-                read_decode_cache_bcr_arcv2(cpu);
+		read_decode_cache_bcr_arcv2(c, buf + n, len - n);
+
+	return buf;
 }
 
 /*
@@ -1133,10 +1096,8 @@ static noinline void __init arc_ioc_setup(void)
  */
 static noinline void __init arc_cache_init_master(void)
 {
-	unsigned int __maybe_unused cpu = smp_processor_id();
-
 	if (IS_ENABLED(CONFIG_ARC_HAS_ICACHE)) {
-		struct cpuinfo_arc_cache *ic = &cpuinfo_arc700[cpu].icache;
+		struct cpuinfo_arc_cache *ic = &ic_info;
 
 		if (!ic->line_len)
 			panic("cache support enabled but non-existent cache\n");
@@ -1149,14 +1110,14 @@ static noinline void __init arc_cache_init_master(void)
 		 * In MMU v4 (HS38x) the aliasing icache config uses IVIL/PTAG
 		 * pair to provide vaddr/paddr respectively, just as in MMU v3
 		 */
-		if (is_isa_arcv2() && ic->alias)
+		if (is_isa_arcv2() && ic->colors > 1)
 			_cache_line_loop_ic_fn = __cache_line_loop_v3;
 		else
 			_cache_line_loop_ic_fn = __cache_line_loop;
 	}
 
 	if (IS_ENABLED(CONFIG_ARC_HAS_DCACHE)) {
-		struct cpuinfo_arc_cache *dc = &cpuinfo_arc700[cpu].dcache;
+		struct cpuinfo_arc_cache *dc = &dc_info;
 
 		if (!dc->line_len)
 			panic("cache support enabled but non-existent cache\n");
@@ -1168,14 +1129,13 @@ static noinline void __init arc_cache_init_master(void)
 		/* check for D-Cache aliasing on ARCompact: ARCv2 has PIPT */
 		if (is_isa_arcompact()) {
 			int handled = IS_ENABLED(CONFIG_ARC_CACHE_VIPT_ALIASING);
-			int num_colors = dc->sz_k/dc->assoc/TO_KB(PAGE_SIZE);
 
-			if (dc->alias) {
+			if (dc->colors > 1) {
 				if (!handled)
 					panic("Enable CONFIG_ARC_CACHE_VIPT_ALIASING\n");
-				if (CACHE_COLORS_NUM != num_colors)
+				if (CACHE_COLORS_NUM != dc->colors)
 					panic("CACHE_COLORS_NUM not optimized for config\n");
-			} else if (!dc->alias && handled) {
+			} else if (handled && dc->colors == 1) {
 				panic("Disable CONFIG_ARC_CACHE_VIPT_ALIASING\n");
 			}
 		}
@@ -1218,9 +1178,6 @@ static noinline void __init arc_cache_init_master(void)
 void __ref arc_cache_init(void)
 {
 	unsigned int __maybe_unused cpu = smp_processor_id();
-	char str[256];
-
-	pr_info("%s", arc_cache_mumbojumbo(0, str, sizeof(str)));
 
 	if (!cpu)
 		arc_cache_init_master();
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 07/20] ARC: boot log: eliminate struct cpuinfo_arc #3: don't export

[Vineet Gupta]

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/arcregs.h | 2 --
 arch/arc/kernel/mcip.c         | 2 --
 2 files changed, 4 deletions(-)

diff --git a/arch/arc/include/asm/arcregs.h b/arch/arc/include/asm/arcregs.h
index cb1ad1bb4ece..160ee3fab1bd 100644
--- a/arch/arc/include/asm/arcregs.h
+++ b/arch/arc/include/asm/arcregs.h
@@ -384,8 +384,6 @@ struct cpuinfo_arc {
 	struct bcr_mpy extn_mpy;
 };
 
-extern struct cpuinfo_arc cpuinfo_arc700[];
-
 static inline int is_isa_arcv2(void)
 {
 	return IS_ENABLED(CONFIG_ISA_ARCV2);
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 = {
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 08/20] ARC: boot log: eliminate struct cpuinfo_arc #4: boot log per ISA

[Vineet Gupta]

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

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/arcregs.h |  33 +-
 arch/arc/include/asm/setup.h   |   4 +-
 arch/arc/kernel/setup.c        | 560 +++++++++++++++------------------
 arch/arc/mm/cache.c            |  10 +-
 arch/arc/mm/tlb.c              |   4 +-
 5 files changed, 268 insertions(+), 343 deletions(-)

diff --git a/arch/arc/include/asm/arcregs.h b/arch/arc/include/asm/arcregs.h
index 160ee3fab1bd..4b13f60fe7ca 100644
--- a/arch/arc/include/asm/arcregs.h
+++ b/arch/arc/include/asm/arcregs.h
@@ -23,7 +23,7 @@
 #define ARC_REG_ICCM_BUILD	0x78	/* ICCM size (common) */
 #define ARC_REG_XY_MEM_BCR	0x79
 #define ARC_REG_MAC_BCR		0x7a
-#define ARC_REG_MUL_BCR		0x7b
+#define ARC_REG_MPY_BCR		0x7b
 #define ARC_REG_SWAP_BCR	0x7c
 #define ARC_REG_NORM_BCR	0x7d
 #define ARC_REG_MIXMAX_BCR	0x7e
@@ -177,7 +177,7 @@ struct bcr_isa_arcv2 {
 #endif
 };
 
-struct bcr_uarch_build_arcv2 {
+struct bcr_uarch_build {
 #ifdef CONFIG_CPU_BIG_ENDIAN
 	unsigned int pad:8, prod:8, maj:8, min:8;
 #else
@@ -355,35 +355,6 @@ struct bcr_generic {
 #endif
 };
 
-/*
- *******************************************************************
- * Generic structures to hold build configuration used at runtime
- */
-
-struct cpuinfo_arc_bpu {
-	unsigned int ver, full, num_cache, num_pred, ret_stk;
-};
-
-struct cpuinfo_arc_ccm {
-	unsigned int base_addr, sz;
-};
-
-struct cpuinfo_arc {
-	struct cpuinfo_arc_bpu bpu;
-	struct bcr_identity core;
-	struct bcr_isa_arcv2 isa;
-	const char *release, *name;
-	unsigned int vec_base;
-	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:1, dual_enb:1, pad2:4,
-			     ap_num:4, ap_full:1, smart:1, rtt:1, pad3:1,
-			     timer0:1, timer1:1, rtc:1, gfrc:1, pad4:4;
-	} extn;
-	struct bcr_mpy extn_mpy;
-};
-
 static inline int is_isa_arcv2(void)
 {
 	return IS_ENABLED(CONFIG_ISA_ARCV2);
diff --git a/arch/arc/include/asm/setup.h b/arch/arc/include/asm/setup.h
index 4c0bacd0ff5c..1c6db599e1fc 100644
--- a/arch/arc/include/asm/setup.h
+++ b/arch/arc/include/asm/setup.h
@@ -35,10 +35,10 @@ long __init arc_get_mem_sz(void);
 #define IS_AVAIL3(v, v2, s)	IS_AVAIL1(v, s), IS_AVAIL1(v, IS_DISABLED_RUN(v2))
 
 extern void arc_mmu_init(void);
-extern char *arc_mmu_mumbojumbo(int cpu_id, char *buf, int len);
+extern int arc_mmu_mumbojumbo(int cpu_id, char *buf, int len);
 
 extern void arc_cache_init(void);
-extern char *arc_cache_mumbojumbo(int cpu_id, char *buf, int len);
+extern int arc_cache_mumbojumbo(int cpu_id, char *buf, int len);
 
 extern void __init handle_uboot_args(void);
 
diff --git a/arch/arc/kernel/setup.c b/arch/arc/kernel/setup.c
index 3ea834941c1f..8749fcc8cfe3 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(info));
 
-			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);
diff --git a/arch/arc/mm/cache.c b/arch/arc/mm/cache.c
index 7197bb845a40..4510f805b21d 100644
--- a/arch/arc/mm/cache.c
+++ b/arch/arc/mm/cache.c
@@ -39,7 +39,7 @@ void (*__dma_cache_wback_inv)(phys_addr_t start, unsigned long sz);
 void (*__dma_cache_inv)(phys_addr_t start, unsigned long sz);
 void (*__dma_cache_wback)(phys_addr_t start, unsigned long sz);
 
-static char *read_decode_cache_bcr_arcv2(int c, char *buf, int len)
+static int read_decode_cache_bcr_arcv2(int c, char *buf, int len)
 {
 	struct cpuinfo_arc_cache *p_slc = &slc_info;
 	struct bcr_identity ident;
@@ -94,10 +94,10 @@ static char *read_decode_cache_bcr_arcv2(int c, char *buf, int len)
 		       perip_base,
 		       IS_AVAIL3(ioc_exists, ioc_enable, ", IO-Coherency (per-device) "));
 
-	return buf;
+	return n;
 }
 
-char *arc_cache_mumbojumbo(int c, char *buf, int len)
+int arc_cache_mumbojumbo(int c, char *buf, int len)
 {
 	struct cpuinfo_arc_cache *p_ic = &ic_info, *p_dc = &dc_info;
 	struct bcr_cache ibcr, dbcr;
@@ -153,9 +153,9 @@ char *arc_cache_mumbojumbo(int c, char *buf, int len)
 
 slc_chk:
 	if (is_isa_arcv2())
-		read_decode_cache_bcr_arcv2(c, buf + n, len - n);
+		n += read_decode_cache_bcr_arcv2(c, buf + n, len - n);
 
-	return buf;
+	return n;
 }
 
 /*
diff --git a/arch/arc/mm/tlb.c b/arch/arc/mm/tlb.c
index 17e32c707367..19deb8ca9c28 100644
--- a/arch/arc/mm/tlb.c
+++ b/arch/arc/mm/tlb.c
@@ -562,7 +562,7 @@ void local_flush_pmd_tlb_range(struct vm_area_struct *vma, unsigned long start,
  * the cpuinfo structure for later use.
  * No Validation is done here, simply read/convert the BCRs
  */
-char *arc_mmu_mumbojumbo(int c, char *buf, int len)
+int arc_mmu_mumbojumbo(int c, char *buf, int len)
 {
 	struct cpuinfo_arc_mmu *mmu = &mmuinfo;
 	unsigned int bcr, u_dtlb, u_itlb, sasid;
@@ -606,7 +606,7 @@ char *arc_mmu_mumbojumbo(int c, char *buf, int len)
 		       u_dtlb, u_itlb,
 		       IS_AVAIL2(mmu->pae, ", PAE40 ", CONFIG_ARC_HAS_PAE40));
 
-	return buf;
+	return n;
 }
 
 int pae40_exist_but_not_enab(void)
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 09/20] ARC: entry: use gp to cache task pointer (vs. r25)

[Vineet Gupta]

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/Kconfig                     |   6 +-
 arch/arc/Makefile                    |   6 +-
 arch/arc/include/asm/current.h       |   2 +-
 arch/arc/include/asm/entry-arcv2.h   |  17 ++---
 arch/arc/include/asm/entry-compact.h |  35 +++------
 arch/arc/include/asm/entry.h         | 107 ++++++++-------------------
 arch/arc/include/asm/ptrace.h        |   6 +-
 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 -
 13 files changed, 58 insertions(+), 154 deletions(-)

diff --git a/arch/arc/Kconfig b/arch/arc/Kconfig
index 47b4acc7d0c9..c92bacc1ff4c 100644
--- a/arch/arc/Kconfig
+++ b/arch/arc/Kconfig
@@ -492,11 +492,11 @@ config ARC_KVADDR_SIZE
 	  kernel-user gutter)
 
 config ARC_CURR_IN_REG
-	bool "Dedicate Register r25 for current_task pointer"
+	bool "cache current task pointer in gp"
 	default y
 	help
-	  This reserved Register R25 to point to Current Task in
-	  kernel mode. This saves memory access for each such access
+	  This reserves gp register to point to Current Task in
+	  kernel mode eliding memory access for each access
 
 
 config ARC_EMUL_UNALIGNED
diff --git a/arch/arc/Makefile b/arch/arc/Makefile
index 329400a1c355..2390dd042e36 100644
--- a/arch/arc/Makefile
+++ b/arch/arc/Makefile
@@ -28,14 +28,14 @@ cflags-y				+= $(tune-mcpu-def-y)
 endif
 endif
 
-
 ifdef CONFIG_ARC_CURR_IN_REG
 # For a global register definition, make sure it gets passed to every file
 # We had a customer reported bug where some code built in kernel was NOT using
-# any kernel headers, and missing the r25 global register
+# any kernel headers, and missing the global register
 # Can't do unconditionally because of recursive include issues
 # due to <linux/thread_info.h>
 LINUXINCLUDE	+=  -include $(srctree)/arch/arc/include/asm/current.h
+cflags-y	+= -ffixed-gp
 endif
 
 cflags-y				+= -fsection-anchors
@@ -67,7 +67,7 @@ cflags-$(CONFIG_ARC_DW2_UNWIND)		+= -fasynchronous-unwind-tables $(cfi)
 # small data is default for elf32 tool-chain. If not usable, disable it
 # This also allows repurposing GP as scratch reg to gcc reg allocator
 disable_small_data := y
-cflags-$(disable_small_data)		+= -mno-sdata -fcall-used-gp
+cflags-$(disable_small_data)		+= -mno-sdata
 
 cflags-$(CONFIG_CPU_BIG_ENDIAN)		+= -mbig-endian
 ldflags-$(CONFIG_CPU_BIG_ENDIAN)	+= -EB
diff --git a/arch/arc/include/asm/current.h b/arch/arc/include/asm/current.h
index 9b9bdd3e6538..06be89f6f2f0 100644
--- a/arch/arc/include/asm/current.h
+++ b/arch/arc/include/asm/current.h
@@ -13,7 +13,7 @@
 
 #ifdef CONFIG_ARC_CURR_IN_REG
 
-register struct task_struct *curr_arc asm("r25");
+register struct task_struct *curr_arc asm("gp");
 #define current (curr_arc)
 
 #else
diff --git a/arch/arc/include/asm/entry-arcv2.h b/arch/arc/include/asm/entry-arcv2.h
index 0ff4c0610561..858742feab71 100644
--- a/arch/arc/include/asm/entry-arcv2.h
+++ b/arch/arc/include/asm/entry-arcv2.h
@@ -18,7 +18,6 @@
  *              |      orig_r0      |
  *              |      event/ECR    |
  *              |      bta          |
- *              |      user_r25     |
  *              |      gp           |
  *              |      fp           |
  *              |      sp           |
@@ -56,7 +55,7 @@
 	;                 hardware does even if CONFIG_ARC_IRQ_NO_AUTOSAVE
 	;   4. Auto save: (optional) r0-r11, blink, LPE,LPS,LPC, JLI,LDI,EI
 	;
-	; (B) Manually saved some regs: r12,r25,r30, sp,fp,gp, ACCL pair
+	; (B) Manually saved some regs: r12,r30, sp,fp,gp, ACCL pair
 
 #ifdef CONFIG_ARC_IRQ_NO_AUTOSAVE
 	; carve pt_regs on stack (case #3), PC/STAT32 already on stack
@@ -157,17 +156,17 @@
 
 	st	r10, [sp, PT_sp]	; SP (pt_regs->sp)
 
-#ifdef CONFIG_ARC_CURR_IN_REG
-	st	r25, [sp, PT_user_r25]
-	GET_CURR_TASK_ON_CPU	r25
-#endif
-
 #ifdef CONFIG_ARC_HAS_ACCL_REGS
 	ST2	r58, r59, PT_r58
 #endif
 
 	/* clobbers r10, r11 registers pair */
 	DSP_SAVE_REGFILE_IRQ
+
+#ifdef CONFIG_ARC_CURR_IN_REG
+	GET_CURR_TASK_ON_CPU	gp
+#endif
+
 .endm
 
 /*------------------------------------------------------------------------*/
@@ -188,10 +187,6 @@
 	sr	r10, [AUX_USER_SP]
 1:
 
-#ifdef CONFIG_ARC_CURR_IN_REG
-	ld	r25, [sp, PT_user_r25]
-#endif
-
 	/* clobbers r10, r11 registers pair */
 	DSP_RESTORE_REGFILE_IRQ
 
diff --git a/arch/arc/include/asm/entry-compact.h b/arch/arc/include/asm/entry-compact.h
index 67ff06e15cea..e3383e1cb040 100644
--- a/arch/arc/include/asm/entry-compact.h
+++ b/arch/arc/include/asm/entry-compact.h
@@ -151,14 +151,6 @@
 	/* ARC700 doesn't provide auto-stack switching */
 	SWITCH_TO_KERNEL_STK
 
-#ifdef CONFIG_ARC_CURR_IN_REG
-	/* Treat r25 as scratch reg (save on stack) and load with "current" */
-	PUSH    r25
-	GET_CURR_TASK_ON_CPU   r25
-#else
-	sub     sp, sp, 4
-#endif
-
 	st.a	r0, [sp, -8]    /* orig_r0 needed for syscall (skip ECR slot) */
 	sub	sp, sp, 4	/* skip pt_regs->sp, already saved above */
 
@@ -179,6 +171,11 @@
 
 	lr	r10, [ecr]
 	st      r10, [sp, PT_event]    /* EV_Trap expects r10 to have ECR */
+
+#ifdef CONFIG_ARC_CURR_IN_REG
+	/* gp already saved on stack: now load with "current" */
+	GET_CURR_TASK_ON_CPU   gp
+#endif
 .endm
 
 /*--------------------------------------------------------------
@@ -208,11 +205,8 @@
 	POP	gp
 	RESTORE_R12_TO_R0
 
-#ifdef CONFIG_ARC_CURR_IN_REG
-	ld	r25, [sp, 12]
-#endif
 	ld  sp, [sp] /* restore original sp */
-	/* orig_r0, ECR, user_r25 skipped automatically */
+	/* orig_r0, ECR skipped automatically */
 .endm
 
 /* Dummy ECR values for Interrupts */
@@ -229,13 +223,6 @@
 
 	SWITCH_TO_KERNEL_STK
 
-#ifdef CONFIG_ARC_CURR_IN_REG
-	/* Treat r25 as scratch reg (save on stack) and load with "current" */
-	PUSH    r25
-	GET_CURR_TASK_ON_CPU   r25
-#else
-	sub     sp, sp, 4
-#endif
 
 	PUSH	0x003\LVL\()abcd    /* Dummy ECR */
 	sub	sp, sp, 8	    /* skip orig_r0 (not needed)
@@ -255,6 +242,10 @@
 	PUSHAX	lp_start
 	PUSHAX	bta_l\LVL\()
 
+#ifdef CONFIG_ARC_CURR_IN_REG
+	/* gp already saved on stack: now load with "current" */
+	GET_CURR_TASK_ON_CPU   gp
+#endif
 .endm
 
 /*--------------------------------------------------------------
@@ -282,11 +273,7 @@
 	POP	gp
 	RESTORE_R12_TO_R0
 
-#ifdef CONFIG_ARC_CURR_IN_REG
-	ld	r25, [sp, 12]
-#endif
-	ld  sp, [sp] /* restore original sp */
-	/* orig_r0, ECR, user_r25 skipped automatically */
+	ld  sp, [sp] /* restore original sp; orig_r0, ECR skipped implicitly */
 .endm
 
 /* Get thread_info of "current" tsk */
diff --git a/arch/arc/include/asm/entry.h b/arch/arc/include/asm/entry.h
index 2980bc9b7653..49c2e090cb5c 100644
--- a/arch/arc/include/asm/entry.h
+++ b/arch/arc/include/asm/entry.h
@@ -91,7 +91,7 @@
  * Helpers to save/restore callee-saved regs:
  * used by several macros below
  *-------------------------------------------------------------*/
-.macro SAVE_R13_TO_R24
+.macro SAVE_R13_TO_R25
 	PUSH	r13
 	PUSH	r14
 	PUSH	r15
@@ -104,9 +104,11 @@
 	PUSH	r22
 	PUSH	r23
 	PUSH	r24
+	PUSH	r25
 .endm
 
-.macro RESTORE_R24_TO_R13
+.macro RESTORE_R25_TO_R13
+	POP	r25
 	POP	r24
 	POP	r23
 	POP	r22
@@ -121,81 +123,31 @@
 	POP	r13
 .endm
 
-/*--------------------------------------------------------------
- * Collect User Mode callee regs as struct callee_regs - needed by
- * fork/do_signal/unaligned-access-emulation.
- * (By default only scratch regs are saved on entry to kernel)
- *
- * Special handling for r25 if used for caching Task Pointer.
- * It would have been saved in task->thread.user_r25 already, but to keep
- * the interface same it is copied into regular r25 placeholder in
- * struct callee_regs.
- *-------------------------------------------------------------*/
+/*
+ * save user mode callee regs as struct callee_regs
+ *  - needed by fork/do_signal/unaligned-access-emulation.
+ */
 .macro SAVE_CALLEE_SAVED_USER
+	SAVE_R13_TO_R25
+.endm
 
-	mov	r12, sp		; save SP as ref to pt_regs
-	SAVE_R13_TO_R24
-
-#ifdef CONFIG_ARC_CURR_IN_REG
-	; Retrieve orig r25 and save it with rest of callee_regs
-	ld	r12, [r12, PT_user_r25]
-	PUSH	r12
-#else
-	PUSH	r25
-#endif
-
+/*
+ * restore user mode callee regs as struct callee_regs
+ *  - could have been changed by ptrace tracer or unaligned-access fixup
+ */
+.macro RESTORE_CALLEE_SAVED_USER
+	RESTORE_R25_TO_R13
 .endm
 
-/*--------------------------------------------------------------
- * Save kernel Mode callee regs at the time of Contect Switch.
- *
- * Special handling for r25 if used for caching Task Pointer.
- * Kernel simply skips saving it since it will be loaded with
- * incoming task pointer anyways
- *-------------------------------------------------------------*/
+/*
+ * save/restore kernel mode callee regs at the time of context switch
+ */
 .macro SAVE_CALLEE_SAVED_KERNEL
-
-	SAVE_R13_TO_R24
-
-#ifdef CONFIG_ARC_CURR_IN_REG
-	sub     sp, sp, 4
-#else
-	PUSH	r25
-#endif
+	SAVE_R13_TO_R25
 .endm
 
-/*--------------------------------------------------------------
- * Opposite of SAVE_CALLEE_SAVED_KERNEL
- *-------------------------------------------------------------*/
 .macro RESTORE_CALLEE_SAVED_KERNEL
-
-#ifdef CONFIG_ARC_CURR_IN_REG
-	add     sp, sp, 4  /* skip usual r25 placeholder */
-#else
-	POP	r25
-#endif
-	RESTORE_R24_TO_R13
-.endm
-
-/*--------------------------------------------------------------
- * Opposite of SAVE_CALLEE_SAVED_USER
- *
- * ptrace tracer or unaligned-access fixup might have changed a user mode
- * callee reg which is saved back to usual r25 storage location
- *-------------------------------------------------------------*/
-.macro RESTORE_CALLEE_SAVED_USER
-
-#ifdef CONFIG_ARC_CURR_IN_REG
-	POP	r12
-#else
-	POP	r25
-#endif
-	RESTORE_R24_TO_R13
-
-	; SP is back to start of pt_regs
-#ifdef CONFIG_ARC_CURR_IN_REG
-	st	r12, [sp, PT_user_r25]
-#endif
+	RESTORE_R25_TO_R13
 .endm
 
 /*--------------------------------------------------------------
@@ -231,10 +183,10 @@
 
 #ifdef CONFIG_SMP
 
-/*-------------------------------------------------
+/*
  * Retrieve the current running task on this CPU
- * 1. Determine curr CPU id.
- * 2. Use it to index into _current_task[ ]
+ *  - loads it from backing _current_task[] (and can't use the
+ *    caching reg for current task
  */
 .macro  GET_CURR_TASK_ON_CPU   reg
 	GET_CPU_ID  \reg
@@ -256,7 +208,7 @@
 	add2 \tmp, @_current_task, \tmp
 	st   \tsk, [\tmp]
 #ifdef CONFIG_ARC_CURR_IN_REG
-	mov r25, \tsk
+	mov gp, \tsk
 #endif
 
 .endm
@@ -271,21 +223,20 @@
 .macro  SET_CURR_TASK_ON_CPU    tsk, tmp
 	st  \tsk, [@_current_task]
 #ifdef CONFIG_ARC_CURR_IN_REG
-	mov r25, \tsk
+	mov gp, \tsk
 #endif
 .endm
 
 #endif /* SMP / UNI */
 
-/* ------------------------------------------------------------------
+/*
  * Get the ptr to some field of Current Task at @off in task struct
- *  -Uses r25 for Current task ptr if that is enabled
+ *  - Uses current task cached in reg if enabled
  */
-
 #ifdef CONFIG_ARC_CURR_IN_REG
 
 .macro GET_CURR_TASK_FIELD_PTR  off,  reg
-	add \reg, r25, \off
+	add \reg, gp, \off
 .endm
 
 #else
diff --git a/arch/arc/include/asm/ptrace.h b/arch/arc/include/asm/ptrace.h
index cf90fcd2a628..e9798f46cdc4 100644
--- a/arch/arc/include/asm/ptrace.h
+++ b/arch/arc/include/asm/ptrace.h
@@ -52,11 +52,9 @@ struct pt_regs {
 		};
 		unsigned long event;
 	};
-
-	unsigned long user_r25;
 };
 
-#define MAX_REG_OFFSET offsetof(struct pt_regs, user_r25)
+#define MAX_REG_OFFSET offsetof(struct pt_regs, event)
 
 #else
 
@@ -79,8 +77,6 @@ struct pt_regs {
 
 	unsigned long bta;	/* bta_l1, bta_l2, erbta */
 
-	unsigned long user_r25;
-
 	unsigned long r26;	/* gp */
 	unsigned long fp;
 	unsigned long sp;	/* user/kernel sp depending on where we came from  */
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),
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 10/20] ARC: kernel stack: INIT_THREAD need not setup @init_stack in @ksp

[Vineet Gupta]

There are 2 pointers to kernel mode stack of a task
 - task_struct.stack: base address of stack page (max possible stack top)
 - thread_info.ksp  : runtime stack top in __switch_to

INIT_THREAD was setting up ksp to stack base which was not really needed
 - it would get overwritten with dynamic value on first call to
   __switch_to when init is switched out for the very first time.
 - generic code already does
      init_task.stack = init_stack
   and ARC code uses that to retrieve task's stack base.

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/processor.h | 4 +---
 1 file changed, 1 insertion(+), 3 deletions(-)

diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
index fb844fce1ab6..82ddc929d6e7 100644
--- a/arch/arc/include/asm/processor.h
+++ b/arch/arc/include/asm/processor.h
@@ -33,9 +33,7 @@ struct thread_struct {
 #endif
 };
 
-#define INIT_THREAD  {                          \
-	.ksp = sizeof(init_stack) + (unsigned long) init_stack, \
-}
+#define INIT_THREAD  { }
 
 /* Forward declaration, a strange C thing */
 struct task_struct;
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 11/20] ARC: __switch_to: asm with dwarf ops (vs. inline asm)

[Vineet Gupta]

__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/include/asm/dwarf.h |  32 +++++++----
 arch/arc/kernel/Makefile     |   9 +--
 arch/arc/kernel/ctx_sw.c     | 103 -----------------------------------
 arch/arc/kernel/ctx_sw_asm.S |  72 +++++++++++++-----------
 4 files changed, 61 insertions(+), 155 deletions(-)
 delete mode 100644 arch/arc/kernel/ctx_sw.c

diff --git a/arch/arc/include/asm/dwarf.h b/arch/arc/include/asm/dwarf.h
index 5f4de05bd4ee..a0d5ebe1bc3f 100644
--- a/arch/arc/include/asm/dwarf.h
+++ b/arch/arc/include/asm/dwarf.h
@@ -10,23 +10,31 @@
 
 #ifdef ARC_DW2_UNWIND_AS_CFI
 
-#define CFI_STARTPROC	.cfi_startproc
-#define CFI_ENDPROC	.cfi_endproc
-#define CFI_DEF_CFA	.cfi_def_cfa
-#define CFI_REGISTER	.cfi_register
-#define CFI_REL_OFFSET	.cfi_rel_offset
-#define CFI_UNDEFINED	.cfi_undefined
+#define CFI_STARTPROC		.cfi_startproc
+#define CFI_ENDPROC		.cfi_endproc
+#define CFI_DEF_CFA		.cfi_def_cfa
+#define CFI_DEF_CFA_OFFSET	.cfi_def_cfa_offset
+#define CFI_DEF_CFA_REGISTER	.cfi_def_cfa_register
+#define CFI_OFFSET		.cfi_offset
+#define CFI_REL_OFFSET		.cfi_rel_offset
+#define CFI_REGISTER		.cfi_register
+#define CFI_RESTORE		.cfi_restore
+#define CFI_UNDEFINED		.cfi_undefined
 
 #else
 
 #define CFI_IGNORE	#
 
-#define CFI_STARTPROC	CFI_IGNORE
-#define CFI_ENDPROC	CFI_IGNORE
-#define CFI_DEF_CFA	CFI_IGNORE
-#define CFI_REGISTER	CFI_IGNORE
-#define CFI_REL_OFFSET	CFI_IGNORE
-#define CFI_UNDEFINED	CFI_IGNORE
+#define CFI_STARTPROC		CFI_IGNORE
+#define CFI_ENDPROC		CFI_IGNORE
+#define CFI_DEF_CFA		CFI_IGNORE
+#define CFI_DEF_CFA_OFFSET	CFI_IGNORE
+#define CFI_DEF_CFA_REGISTER	CFI_IGNORE
+#define CFI_OFFSET		CFI_IGNORE
+#define CFI_REL_OFFSET		CFI_IGNORE
+#define CFI_REGISTER		CFI_IGNORE
+#define CFI_RESTORE		CFI_IGNORE
+#define CFI_UNDEFINED		CFI_IGNORE
 
 #endif	/* !ARC_DW2_UNWIND_AS_CFI */
 
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)
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 12/20] ARC: __switch_to: move ksp to thread_info from thread_struct

[Vineet Gupta]

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/include/asm/processor.h   |  3 +--
 arch/arc/include/asm/thread_info.h | 10 +++++-----
 arch/arc/kernel/asm-offsets.c      |  2 +-
 arch/arc/kernel/ctx_sw_asm.S       | 22 ++++++++++------------
 arch/arc/kernel/process.c          |  6 +++---
 5 files changed, 20 insertions(+), 23 deletions(-)

diff --git a/arch/arc/include/asm/processor.h b/arch/arc/include/asm/processor.h
index 82ddc929d6e7..d606658e2fe7 100644
--- a/arch/arc/include/asm/processor.h
+++ b/arch/arc/include/asm/processor.h
@@ -22,7 +22,6 @@
  * struct thread_info
  */
 struct thread_struct {
-	unsigned long ksp;	/* kernel mode stack pointer */
 	unsigned long callee_reg;	/* pointer to callee regs */
 	unsigned long fault_address;	/* dbls as brkpt holder as well */
 #ifdef CONFIG_ARC_DSP_SAVE_RESTORE_REGS
@@ -54,7 +53,7 @@ struct task_struct;
  * Where about of Task's sp, fp, blink when it was last seen in kernel mode.
  * Look in process.c for details of kernel stack layout
  */
-#define TSK_K_ESP(tsk)		(tsk->thread.ksp)
+#define TSK_K_ESP(tsk)		(task_thread_info(tsk)->ksp)
 
 #define TSK_K_REG(tsk, off)	(*((unsigned long *)(TSK_K_ESP(tsk) + \
 					sizeof(struct callee_regs) + off)))
diff --git a/arch/arc/include/asm/thread_info.h b/arch/arc/include/asm/thread_info.h
index 6ba7fe417095..4c530cf131f3 100644
--- a/arch/arc/include/asm/thread_info.h
+++ b/arch/arc/include/asm/thread_info.h
@@ -37,16 +37,16 @@
  */
 struct thread_info {
 	unsigned long flags;		/* low level flags */
+	unsigned long ksp;		/* kernel mode stack top in __switch_to */
 	int preempt_count;		/* 0 => preemptable, <0 => BUG */
-	struct task_struct *task;	/* main task structure */
-	__u32 cpu;			/* current CPU */
+	int cpu;			/* current CPU */
 	unsigned long thr_ptr;		/* TLS ptr */
+	struct task_struct *task;	/* main task structure */
 };
 
 /*
- * macros/functions for gaining access to the thread information structure
- *
- * preempt_count needs to be 1 initially, until the scheduler is functional.
+ * initilaize thread_info for any @tsk
+ *  - this is not related to init_task per se
  */
 #define INIT_THREAD_INFO(tsk)			\
 {						\
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 */
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 13/20] ARC: entry: rework (non-functional)

[Vineet Gupta]

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

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/entry-arcv2.h   | 23 ++++++++------
 arch/arc/include/asm/entry-compact.h |  3 +-
 arch/arc/include/asm/ptrace.h        |  2 +-
 arch/arc/kernel/asm-offsets.c        |  1 +
 arch/arc/kernel/entry.S              | 45 +++++++++++++---------------
 arch/arc/kernel/ptrace.c             |  2 +-
 6 files changed, 39 insertions(+), 37 deletions(-)

diff --git a/arch/arc/include/asm/entry-arcv2.h b/arch/arc/include/asm/entry-arcv2.h
index 858742feab71..a38ed505b3de 100644
--- a/arch/arc/include/asm/entry-arcv2.h
+++ b/arch/arc/include/asm/entry-arcv2.h
@@ -48,14 +48,18 @@
 /*------------------------------------------------------------------------*/
 .macro INTERRUPT_PROLOGUE
 
-	; (A) Before jumping to Interrupt Vector, hardware micro-ops did following:
+	; Before jumping to Interrupt Vector, hardware micro-ops did following:
 	;   1. SP auto-switched to kernel mode stack
 	;   2. STATUS32.Z flag set if in U mode at time of interrupt (U:1,K:0)
 	;   3. Auto save: (mandatory) Push PC and STAT32 on stack
 	;                 hardware does even if CONFIG_ARC_IRQ_NO_AUTOSAVE
-	;   4. Auto save: (optional) r0-r11, blink, LPE,LPS,LPC, JLI,LDI,EI
+	;  4a. Auto save: (optional) r0-r11, blink, LPE,LPS,LPC, JLI,LDI,EI
 	;
-	; (B) Manually saved some regs: r12,r30, sp,fp,gp, ACCL pair
+	; Now
+	;  4b. If Auto-save (optional) not enabled in hw, manually save them
+	;   5. Manually save: r12,r30, sp,fp,gp, ACCL pair
+	;
+	; At the end, SP points to pt_regs
 
 #ifdef CONFIG_ARC_IRQ_NO_AUTOSAVE
 	; carve pt_regs on stack (case #3), PC/STAT32 already on stack
@@ -73,13 +77,14 @@
 /*------------------------------------------------------------------------*/
 .macro EXCEPTION_PROLOGUE
 
-	; (A) Before jumping to Exception Vector, hardware micro-ops did following:
+	; Before jumping to Exception Vector, hardware micro-ops did following:
 	;   1. SP auto-switched to kernel mode stack
 	;   2. STATUS32.Z flag set if in U mode at time of exception (U:1,K:0)
 	;
-	; (B) Manually save the complete reg file below
+	; Now manually save rest of reg file
+	; At the end, SP points to pt_regs
 
-	sub	sp, sp, SZ_PT_REGS	; carve pt_regs
+	sub	sp, sp, SZ_PT_REGS	; carve space for pt_regs
 
 	; _HARD saves r10 clobbered by _SOFT as scratch hence comes first
 
@@ -136,8 +141,8 @@
 
 	ST2	gp, fp, PT_r26		; gp (r26), fp (r27)
 
-	st	r12, [sp, PT_sp + 4]
-	st	r30, [sp, PT_sp + 8]
+	st	r12, [sp, PT_r12]
+	st	r30, [sp, PT_r30]
 
 	; Saving pt_regs->sp correctly requires some extra work due to the way
 	; Auto stack switch works
@@ -244,7 +249,7 @@
 
 	btst	r0, STATUS_U_BIT	; Z flag set if K, used in restoring SP
 
-	ld	r10, [sp, PT_event + 4]
+	ld	r10, [sp, PT_bta]
 	sr	r10, [erbta]
 
 	LD2	r10, r11, PT_ret
diff --git a/arch/arc/include/asm/entry-compact.h b/arch/arc/include/asm/entry-compact.h
index e3383e1cb040..4e2ae82779ed 100644
--- a/arch/arc/include/asm/entry-compact.h
+++ b/arch/arc/include/asm/entry-compact.h
@@ -170,12 +170,13 @@
 	PUSHAX	erbta
 
 	lr	r10, [ecr]
-	st      r10, [sp, PT_event]    /* EV_Trap expects r10 to have ECR */
+	st      r10, [sp, PT_event]
 
 #ifdef CONFIG_ARC_CURR_IN_REG
 	/* gp already saved on stack: now load with "current" */
 	GET_CURR_TASK_ON_CPU   gp
 #endif
+	; OUTPUT: r10 has ECR expected by EV_Trap
 .endm
 
 /*--------------------------------------------------------------
diff --git a/arch/arc/include/asm/ptrace.h b/arch/arc/include/asm/ptrace.h
index e9798f46cdc4..2bf8ea96ea21 100644
--- a/arch/arc/include/asm/ptrace.h
+++ b/arch/arc/include/asm/ptrace.h
@@ -75,7 +75,7 @@ struct pt_regs {
 		unsigned long event;
 	};
 
-	unsigned long bta;	/* bta_l1, bta_l2, erbta */
+	unsigned long bta;	/* erbta */
 
 	unsigned long r26;	/* gp */
 	unsigned long fp;
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))
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 14/20] ARC: entry: ARcompact EV_ProtV to use r10 directly

[Vineet Gupta]

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

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
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 15/20] ARC: entry: EV_MachineCheck dont re-read ECR

[Vineet Gupta]

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

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
 
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 16/20] ARC: entry: Add more common chores to EXCEPTION_PROLOGUE

[Vineet Gupta]

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)

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/entry-arcv2.h   | 12 +++++++++++-
 arch/arc/include/asm/entry-compact.h | 12 +++++++++++-
 arch/arc/kernel/entry-arcv2.S        | 15 ---------------
 arch/arc/kernel/entry-compact.S      | 10 ----------
 arch/arc/kernel/entry.S              | 19 ++-----------------
 5 files changed, 24 insertions(+), 44 deletions(-)

diff --git a/arch/arc/include/asm/entry-arcv2.h b/arch/arc/include/asm/entry-arcv2.h
index a38ed505b3de..11b48ab39154 100644
--- a/arch/arc/include/asm/entry-arcv2.h
+++ b/arch/arc/include/asm/entry-arcv2.h
@@ -75,7 +75,7 @@
 .endm
 
 /*------------------------------------------------------------------------*/
-.macro EXCEPTION_PROLOGUE
+.macro EXCEPTION_PROLOGUE_KEEP_AE
 
 	; Before jumping to Exception Vector, hardware micro-ops did following:
 	;   1. SP auto-switched to kernel mode stack
@@ -104,6 +104,16 @@
 	; OUTPUT: r10 has ECR expected by EV_Trap
 .endm
 
+.macro EXCEPTION_PROLOGUE
+
+	EXCEPTION_PROLOGUE_KEEP_AE	; return ECR in r10
+
+	lr  r0, [efa]
+	mov r1, sp
+
+	FAKE_RET_FROM_EXCPN		; clobbers r9
+.endm
+
 /*------------------------------------------------------------------------
  * This macro saves the registers manually which would normally be autosaved
  * by hardware on taken interrupts. It is used by
diff --git a/arch/arc/include/asm/entry-compact.h b/arch/arc/include/asm/entry-compact.h
index 4e2ae82779ed..a0e760eb35a8 100644
--- a/arch/arc/include/asm/entry-compact.h
+++ b/arch/arc/include/asm/entry-compact.h
@@ -140,7 +140,7 @@
  *
  * After this it is safe to call the "C" handlers
  *-------------------------------------------------------------*/
-.macro EXCEPTION_PROLOGUE
+.macro EXCEPTION_PROLOGUE_KEEP_AE
 
 	/* Need at least 1 reg to code the early exception prologue */
 	PROLOG_FREEUP_REG r9, @ex_saved_reg1
@@ -179,6 +179,16 @@
 	; OUTPUT: r10 has ECR expected by EV_Trap
 .endm
 
+.macro EXCEPTION_PROLOGUE
+
+	EXCEPTION_PROLOGUE_KEEP_AE	; return ECR in r10
+
+	lr  r0, [efa]
+	mov r1, sp
+
+	FAKE_RET_FROM_EXCPN		; clobbers r9
+.endm
+
 /*--------------------------------------------------------------
  * Restore all registers used by system call or Exceptions
  * SP should always be pointing to the next free stack element
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..c218285cff87 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
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]
 
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 17/20] ARC: entry: replace 8 byte OR with 4 byte BSET

[Vineet Gupta]

FAKE_RET_FROM_EXCEPTION drops down to pure kernel mode. It currently has
an 8 byte instruction which can be replaced with 4 byte BSET

This is applicable to both ARCv2 and ARCv3 entr code.

ARCv2 current
------------
00000804 <EV_Trap>:
...
 874:   216a 1280               lr      r9,[status32]
 878:   2146 1809               bic     r9,r9,0x20
 87c:   2105 1f89 8000 0000     or      r9,r9,0x80000000
                  ^^^^^^^^^
 884:   2029 8240               kflag   r9

ARCv2 after
----------
000007e0 <EV_Trap>:
...
 850:   216a 1280               lr      r9,[status32]
 854:   2150 1149               bclr    r9,r9,0x5
 858:   214f 17c9               bset    r9,r9,0x1f
 85c:   2029 8240               kflag   r9

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/entry-arcv2.h | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)

diff --git a/arch/arc/include/asm/entry-arcv2.h b/arch/arc/include/asm/entry-arcv2.h
index 11b48ab39154..f7c9b3915d10 100644
--- a/arch/arc/include/asm/entry-arcv2.h
+++ b/arch/arc/include/asm/entry-arcv2.h
@@ -274,8 +274,8 @@
 
 .macro FAKE_RET_FROM_EXCPN
 	lr      r9, [status32]
-	bic     r9, r9, STATUS_AE_MASK
-	or      r9, r9, STATUS_IE_MASK
+	bclr    r9, r9, STATUS_AE_BIT
+	bset    r9, r9, STATUS_IE_BIT
 	kflag   r9
 .endm
 
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 18/20] ARC: entry: replace 8 byte ADD.ne with 4 byte ADD2.ne

[Vineet Gupta]

ARCv2 current
------------
000007e0 <EV_Trap>:
 7e0:   2482 3c01               sub     sp,sp,112
 7e4:   1c28 3006               std     r0r1,[sp,40]
 7e8:   1c30 3086               std     r2r3,[sp,48]
 7ec:   1c38 3106               std     r4r5,[sp,56]
 7f0:   1c40 3186               std     r6r7,[sp,64]
 7f4:   1c48 3206               std     r8r9,[sp,72]
 7f8:   1c50 3286               std     r10r11,[sp,80]
 7fc:   1c58 37c0               st      blink,[sp,88]
 800:   1c0c 36c0               st      fp,[sp,12]
 804:   1c18 3680               st      gp,[sp,24]
 808:   1c10 3780               st      r30,[sp,16]
 80c:   1c14 3300               st      r12,[sp,20]
 810:   226a 1340               lr      r10,[aux_user_sp]
 814:   22ca 1702               mov.ne  r10,sp
 818:   22c0 1f82 0000 0070     add.ne  r10,r10,0x70
                  ^^^^^^^^^
With fix
--------
000007b4 <EV_Trap>:
 7b4:   2482 3c01               sub     sp,sp,112
 7b8:   1c28 3006               std     r0r1,[sp,40]
 7bc:   1c30 3086               std     r2r3,[sp,48]
 7c0:   1c38 3106               std     r4r5,[sp,56]
 7c4:   1c40 3186               std     r6r7,[sp,64]
 7c8:   1c48 3206               std     r8r9,[sp,72]
 7cc:   1c50 3286               std     r10r11,[sp,80]
 7d0:   1c58 37c0               st      blink,[sp,88]
 7d4:   1c0c 36c0               st      fp,[sp,12]
 7d8:   1c18 3680               st      gp,[sp,24]
 7dc:   1c10 3780               st      r30,[sp,16]
 7e0:   1c14 3300               st      r12,[sp,20]
 7e4:   226a 1340               lr      r10,[aux_user_sp]
 7e8:   22ca 1702               mov.ne  r10,sp
 7ec:   22d5 1722               add2.ne r10,r10,0x1c

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/entry-arcv2.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/arch/arc/include/asm/entry-arcv2.h b/arch/arc/include/asm/entry-arcv2.h
index f7c9b3915d10..a030eae93d35 100644
--- a/arch/arc/include/asm/entry-arcv2.h
+++ b/arch/arc/include/asm/entry-arcv2.h
@@ -167,7 +167,7 @@
 
 	; ISA requires ADD.nz to have same dest and src reg operands
 	mov.nz	r10, sp
-	add.nz	r10, r10, SZ_PT_REGS	; K mode SP
+	add2.nz	r10, r10, SZ_PT_REGS/4	; K mode SP
 
 	st	r10, [sp, PT_sp]	; SP (pt_regs->sp)
 
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 19/20] ARCv2: entry: rearrange pt_regs slightly

[Vineet Gupta]

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/include/asm/entry-arcv2.h | 12 ++++++------
 arch/arc/include/asm/ptrace.h      |  9 +++++----
 arch/arc/kernel/asm-offsets.c      |  7 ++++---
 3 files changed, 15 insertions(+), 13 deletions(-)

diff --git a/arch/arc/include/asm/entry-arcv2.h b/arch/arc/include/asm/entry-arcv2.h
index a030eae93d35..4d13320e0c1b 100644
--- a/arch/arc/include/asm/entry-arcv2.h
+++ b/arch/arc/include/asm/entry-arcv2.h
@@ -149,10 +149,10 @@
  */
 .macro __SAVE_REGFILE_SOFT
 
-	ST2	gp, fp, PT_r26		; gp (r26), fp (r27)
-
-	st	r12, [sp, PT_r12]
+	st	fp,  [sp, PT_fp]	; r27
 	st	r30, [sp, PT_r30]
+	st	r12, [sp, PT_r12]
+	st	r26, [sp, PT_r26]	; gp
 
 	; Saving pt_regs->sp correctly requires some extra work due to the way
 	; Auto stack switch works
@@ -187,10 +187,10 @@
 /*------------------------------------------------------------------------*/
 .macro __RESTORE_REGFILE_SOFT
 
-	LD2	gp, fp, PT_r26		; gp (r26), fp (r27)
-
-	ld	r12, [sp, PT_r12]
+	ld	fp,  [sp, PT_fp]
 	ld	r30, [sp, PT_r30]
+	ld	r12, [sp, PT_r12]
+	ld	r26, [sp, PT_r26]
 
 	; Restore SP (into AUX_USER_SP) only if returning to U mode
 	;  - for K mode, it will be implicitly restored as stack is unwound
diff --git a/arch/arc/include/asm/ptrace.h b/arch/arc/include/asm/ptrace.h
index 2bf8ea96ea21..3a054b695f28 100644
--- a/arch/arc/include/asm/ptrace.h
+++ b/arch/arc/include/asm/ptrace.h
@@ -77,11 +77,10 @@ struct pt_regs {
 
 	unsigned long bta;	/* erbta */
 
-	unsigned long r26;	/* gp */
 	unsigned long fp;
-	unsigned long sp;	/* user/kernel sp depending on where we came from  */
-
-	unsigned long r12, r30;
+	unsigned long r30;
+	unsigned long r12;
+	unsigned long r26;	/* gp */
 
 #ifdef CONFIG_ARC_HAS_ACCL_REGS
 	unsigned long r58, r59;	/* ACCL/ACCH used by FPU / DSP MPY */
@@ -90,6 +89,8 @@ struct pt_regs {
 	unsigned long DSP_CTRL;
 #endif
 
+	unsigned long sp;	/* user/kernel sp depending on entry  */
+
 	/*------- Below list auto saved by h/w -----------*/
 	unsigned long r0, r1, r2, r3, r4, r5, r6, r7, r8, r9, r10, r11;
 
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;
 }
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH 20/20] ARC: pt_regs: create seperate type for ecr

[Vineet Gupta]

Reduces duplication in each ISA specific pt_regs

Signed-off-by: Vineet Gupta <vgupta@kernel.org>
---
 arch/arc/include/asm/ptrace.h  | 45 +++++++++++++---------------------
 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 +++++-----
 arch/arc/mm/fault.c            |  6 ++---
 7 files changed, 32 insertions(+), 44 deletions(-)

diff --git a/arch/arc/include/asm/ptrace.h b/arch/arc/include/asm/ptrace.h
index 3a054b695f28..724e3fe31ed5 100644
--- a/arch/arc/include/asm/ptrace.h
+++ b/arch/arc/include/asm/ptrace.h
@@ -12,6 +12,17 @@
 
 #ifndef __ASSEMBLY__
 
+typedef union {
+	struct {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+		unsigned long state:8, vec:8, cause:8, param:8;
+#else
+		unsigned long param:8, cause:8, vec:8, state:8;
+#endif
+	};
+	unsigned long full;
+} ecr_reg;
+
 /* THE pt_regs: Defines how regs are saved during entry into kernel */
 
 #ifdef CONFIG_ISA_ARCOMPACT
@@ -40,18 +51,7 @@ struct pt_regs {
 	 * 	Last word used by Linux for extra state mgmt (syscall-restart)
 	 * For interrupts, use artificial ECR values to note current prio-level
 	 */
-	union {
-		struct {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-			unsigned long state:8, ecr_vec:8,
-				      ecr_cause:8, ecr_param:8;
-#else
-			unsigned long ecr_param:8, ecr_cause:8,
-				      ecr_vec:8, state:8;
-#endif
-		};
-		unsigned long event;
-	};
+	ecr_reg ecr;
 };
 
 #define MAX_REG_OFFSET offsetof(struct pt_regs, event)
@@ -62,18 +62,7 @@ struct pt_regs {
 
 	unsigned long orig_r0;
 
-	union {
-		struct {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-			unsigned long state:8, ecr_vec:8,
-				      ecr_cause:8, ecr_param:8;
-#else
-			unsigned long ecr_param:8, ecr_cause:8,
-				      ecr_vec:8, state:8;
-#endif
-		};
-		unsigned long event;
-	};
+	ecr_reg ecr;		/* Exception Cause Reg */
 
 	unsigned long bta;	/* erbta */
 
@@ -131,13 +120,13 @@ struct callee_regs {
 /* return 1 if PC in delay slot */
 #define delay_mode(regs) ((regs->status32 & STATUS_DE_MASK) == STATUS_DE_MASK)
 
-#define in_syscall(regs)    ((regs->ecr_vec == ECR_V_TRAP) && !regs->ecr_param)
-#define in_brkpt_trap(regs) ((regs->ecr_vec == ECR_V_TRAP) && regs->ecr_param)
+#define in_syscall(regs)    ((regs->ecr.vec == ECR_V_TRAP) && !regs->ecr.param)
+#define in_brkpt_trap(regs) ((regs->ecr.vec == ECR_V_TRAP) && regs->ecr.param)
 
 #define STATE_SCALL_RESTARTED	0x01
 
-#define syscall_wont_restart(reg) (reg->state |= STATE_SCALL_RESTARTED)
-#define syscall_restartable(reg) !(reg->state &  STATE_SCALL_RESTARTED)
+#define syscall_wont_restart(regs) (regs->ecr.state |= STATE_SCALL_RESTARTED)
+#define syscall_restartable(regs) !(regs->ecr.state &  STATE_SCALL_RESTARTED)
 
 #define current_pt_regs()					\
 ({								\
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 instruction addr */
 	address = current->thread.fault_address;
 
-	vec = regs->ecr_vec;
-	cause_code = regs->ecr_cause;
+	vec = regs->ecr.vec;
+	cause_code = regs->ecr.cause;
 
 	/* For DTLB Miss or ProtV, display the memory involved too */
 	if (vec == ECR_V_DTLB_MISS) {
@@ -154,7 +154,7 @@ static void show_ecr_verbose(struct pt_regs *regs)
 		pr_cont("Misaligned r/w from 0x%08lx\n", address);
 #endif
 	} else if (vec == ECR_V_TRAP) {
-		if (regs->ecr_param == 5)
+		if (regs->ecr.param == 5)
 			pr_cont("gcc generated __builtin_trap\n");
 	} else {
 		pr_cont("Check Programmer's Manual\n");
@@ -184,9 +184,10 @@ void show_regs(struct pt_regs *regs)
 	if (user_mode(regs))
 		show_faulting_vma(regs->ret); /* faulting code, not data */
 
-	pr_info("ECR: 0x%08lx EFA: 0x%08lx ERET: 0x%08lx\nSTAT: 0x%08lx",
-		regs->event, current->thread.fault_address, regs->ret,
-		regs->status32);
+	pr_info("ECR: 0x%08lx EFA: 0x%08lx ERET: 0x%08lx\n",
+		regs->ecr.full, current->thread.fault_address, regs->ret);
+
+	pr_info("STAT32: 0x%08lx", regs->status32);
 
 #define STS_BIT(r, bit)	r->status32 & STATUS_##bit##_MASK ? #bit" " : ""
 
diff --git a/arch/arc/mm/fault.c b/arch/arc/mm/fault.c
index 26e5823c5710..95119a5e7761 100644
--- a/arch/arc/mm/fault.c
+++ b/arch/arc/mm/fault.c
@@ -100,10 +100,10 @@ void do_page_fault(unsigned long address, struct pt_regs *regs)
 	if (faulthandler_disabled() || !mm)
 		goto no_context;
 
-	if (regs->ecr_cause & ECR_C_PROTV_STORE)	/* ST/EX */
+	if (regs->ecr.cause & ECR_C_PROTV_STORE)	/* ST/EX */
 		write = 1;
-	else if ((regs->ecr_vec == ECR_V_PROTV) &&
-	         (regs->ecr_cause == ECR_C_PROTV_INST_FETCH))
+	else if ((regs->ecr.vec == ECR_V_PROTV) &&
+	         (regs->ecr.cause == ECR_C_PROTV_INST_FETCH))
 		exec = 1;
 
 	flags = FAULT_FLAG_DEFAULT;
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

Re: [PATCH 20/20] ARC: pt_regs: create seperate type for ecr

[kernel test robot]

Hi Vineet,

kernel test robot noticed the following build errors:

[auto build test ERROR on vgupta-arc/for-curr]
[cannot apply to vgupta-arc/for-next linus/master v6.5-rc6 next-20230809]
[If your patch is applied to the wrong git tree, kindly drop us a note.
And when submitting patch, we suggest to use '--base' as documented in
https://git-scm.com/docs/git-format-patch#_base_tree_information]

url:    https://github.com/intel-lab-lkp/linux/commits/Vineet-Gupta/ARC-uaccess-remove-arc-specific-out-of-line-handles-for-Os/20230815-085308
base:   https://git.kernel.org/pub/scm/linux/kernel/git/vgupta/arc.git for-curr
patch link:    https://lore.kernel.org/r/20230815004813.555115-21-vgupta%40kernel.org
patch subject: [PATCH 20/20] ARC: pt_regs: create seperate type for ecr
config: arc-randconfig-r043-20230815 (https://download.01.org/0day-ci/archive/20230815/202308151342.ROQ9Urvv-lkp@intel.com/config)
compiler: arc-elf-gcc (GCC) 12.3.0
reproduce: (https://download.01.org/0day-ci/archive/20230815/202308151342.ROQ9Urvv-lkp@intel.com/reproduce)

If you fix the issue in a separate patch/commit (i.e. not just a new version of
the same patch/commit), kindly add following tags
| Reported-by: kernel test robot <lkp@intel.com>
| Closes: https://lore.kernel.org/oe-kbuild-all/202308151342.ROQ9Urvv-lkp@intel.com/

All errors (new ones prefixed by >>):

   In file included from arch/arc/include/asm/ptrace.h:11,
                    from arch/arc/include/asm/bug.h:11,
                    from include/linux/bug.h:5,
                    from include/linux/page-flags.h:10,
                    from kernel/bounds.c:10:
   arch/arc/include/asm/ptrace.h: In function 'regs_get_register':
>> include/linux/stddef.h:16:33: error: 'struct pt_regs' has no member named 'event'
      16 | #define offsetof(TYPE, MEMBER)  __builtin_offsetof(TYPE, MEMBER)
         |                                 ^~~~~~~~~~~~~~~~~~
   include/linux/compiler.h:77:45: note: in definition of macro 'unlikely'
      77 | # define unlikely(x)    __builtin_expect(!!(x), 0)
         |                                             ^
   arch/arc/include/asm/ptrace.h:57:24: note: in expansion of macro 'offsetof'
      57 | #define MAX_REG_OFFSET offsetof(struct pt_regs, event)
         |                        ^~~~~~~~
   arch/arc/include/asm/ptrace.h:164:31: note: in expansion of macro 'MAX_REG_OFFSET'
     164 |         if (unlikely(offset > MAX_REG_OFFSET))
         |                               ^~~~~~~~~~~~~~
   make[3]: *** [scripts/Makefile.build:116: kernel/bounds.s] Error 1 shuffle=716081594
   make[3]: Target 'prepare' not remade because of errors.
   make[2]: *** [Makefile:1287: prepare0] Error 2 shuffle=716081594
   make[2]: Target 'prepare' not remade because of errors.
   make[1]: *** [Makefile:234: __sub-make] Error 2 shuffle=716081594
   make[1]: Target 'prepare' not remade because of errors.
   make: *** [Makefile:234: __sub-make] Error 2 shuffle=716081594
   make: Target 'prepare' not remade because of errors.


vim +16 include/linux/stddef.h

6e218287432472 Richard Knutsson 2006-09-30  14  
^1da177e4c3f41 Linus Torvalds   2005-04-16  15  #undef offsetof
14e83077d55ff4 Rasmus Villemoes 2022-03-23 @16  #define offsetof(TYPE, MEMBER)	__builtin_offsetof(TYPE, MEMBER)
3876488444e712 Denys Vlasenko   2015-03-09  17  

-- 
0-DAY CI Kernel Test Service
https://github.com/intel/lkp-tests/wiki

_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

Re: [PATCH 20/20] ARC: pt_regs: create seperate type for ecr

[Pavel.Kozlov]

Hi Vineet,

I'm testing your updates and ran into the same build issue reported by the build 
robot.
http://lists.infradead.org/pipermail/linux-snps-arc/2023-August/007522.html

> #ifdef CONFIG_ISA_ARCOMPACT
> @@ -40,18 +51,7 @@ struct pt_regs {
>           *        Last word used by Linux for extra state mgmt (syscall-restart)
>           * For interrupts, use artificial ECR values to note current prio-level
>           */
> -       union {
> -               struct {
> -#ifdef CONFIG_CPU_BIG_ENDIAN
> -                       unsigned long state:8, ecr_vec:8,
> -                                     ecr_cause:8, ecr_param:8;
> -#else
> -                       unsigned long ecr_param:8, ecr_cause:8,
> -                                     ecr_vec:8, state:8;
> -#endif
> -               };
> -               unsigned long event;
> -       };
> +       ecr_reg ecr;
> }
>
> #define MAX_REG_OFFSET offsetof(struct pt_regs, event)

This change causes a build issue for ARC700, as the event field has been
removed and the MAX_REG_OFFSET macro hasn't been updated.

Regards,
Pavel

_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH v2 20/20] ARC: pt_regs: create seperate type for ecr

[Vineet Gupta]

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/include/asm/ptrace.h  | 47 +++++++++++++---------------------
 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 +++++-----
 arch/arc/mm/fault.c            |  6 ++---
 7 files changed, 33 insertions(+), 45 deletions(-)

diff --git a/arch/arc/include/asm/ptrace.h b/arch/arc/include/asm/ptrace.h
index 3a054b695f28..4a2b30fb5a98 100644
--- a/arch/arc/include/asm/ptrace.h
+++ b/arch/arc/include/asm/ptrace.h
@@ -12,6 +12,17 @@
 
 #ifndef __ASSEMBLY__
 
+typedef union {
+	struct {
+#ifdef CONFIG_CPU_BIG_ENDIAN
+		unsigned long state:8, vec:8, cause:8, param:8;
+#else
+		unsigned long param:8, cause:8, vec:8, state:8;
+#endif
+	};
+	unsigned long full;
+} ecr_reg;
+
 /* THE pt_regs: Defines how regs are saved during entry into kernel */
 
 #ifdef CONFIG_ISA_ARCOMPACT
@@ -40,21 +51,10 @@ struct pt_regs {
 	 * 	Last word used by Linux for extra state mgmt (syscall-restart)
 	 * For interrupts, use artificial ECR values to note current prio-level
 	 */
-	union {
-		struct {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-			unsigned long state:8, ecr_vec:8,
-				      ecr_cause:8, ecr_param:8;
-#else
-			unsigned long ecr_param:8, ecr_cause:8,
-				      ecr_vec:8, state:8;
-#endif
-		};
-		unsigned long event;
-	};
+	ecr_reg ecr;
 };
 
-#define MAX_REG_OFFSET offsetof(struct pt_regs, event)
+#define MAX_REG_OFFSET offsetof(struct pt_regs, ecr)
 
 #else
 
@@ -62,18 +62,7 @@ struct pt_regs {
 
 	unsigned long orig_r0;
 
-	union {
-		struct {
-#ifdef CONFIG_CPU_BIG_ENDIAN
-			unsigned long state:8, ecr_vec:8,
-				      ecr_cause:8, ecr_param:8;
-#else
-			unsigned long ecr_param:8, ecr_cause:8,
-				      ecr_vec:8, state:8;
-#endif
-		};
-		unsigned long event;
-	};
+	ecr_reg ecr;		/* Exception Cause Reg */
 
 	unsigned long bta;	/* erbta */
 
@@ -131,13 +120,13 @@ struct callee_regs {
 /* return 1 if PC in delay slot */
 #define delay_mode(regs) ((regs->status32 & STATUS_DE_MASK) == STATUS_DE_MASK)
 
-#define in_syscall(regs)    ((regs->ecr_vec == ECR_V_TRAP) && !regs->ecr_param)
-#define in_brkpt_trap(regs) ((regs->ecr_vec == ECR_V_TRAP) && regs->ecr_param)
+#define in_syscall(regs)    ((regs->ecr.vec == ECR_V_TRAP) && !regs->ecr.param)
+#define in_brkpt_trap(regs) ((regs->ecr.vec == ECR_V_TRAP) && regs->ecr.param)
 
 #define STATE_SCALL_RESTARTED	0x01
 
-#define syscall_wont_restart(reg) (reg->state |= STATE_SCALL_RESTARTED)
-#define syscall_restartable(reg) !(reg->state &  STATE_SCALL_RESTARTED)
+#define syscall_wont_restart(regs) (regs->ecr.state |= STATE_SCALL_RESTARTED)
+#define syscall_restartable(regs) !(regs->ecr.state &  STATE_SCALL_RESTARTED)
 
 #define current_pt_regs()					\
 ({								\
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 instruction addr */
 	address = current->thread.fault_address;
 
-	vec = regs->ecr_vec;
-	cause_code = regs->ecr_cause;
+	vec = regs->ecr.vec;
+	cause_code = regs->ecr.cause;
 
 	/* For DTLB Miss or ProtV, display the memory involved too */
 	if (vec == ECR_V_DTLB_MISS) {
@@ -154,7 +154,7 @@ static void show_ecr_verbose(struct pt_regs *regs)
 		pr_cont("Misaligned r/w from 0x%08lx\n", address);
 #endif
 	} else if (vec == ECR_V_TRAP) {
-		if (regs->ecr_param == 5)
+		if (regs->ecr.param == 5)
 			pr_cont("gcc generated __builtin_trap\n");
 	} else {
 		pr_cont("Check Programmer's Manual\n");
@@ -184,9 +184,10 @@ void show_regs(struct pt_regs *regs)
 	if (user_mode(regs))
 		show_faulting_vma(regs->ret); /* faulting code, not data */
 
-	pr_info("ECR: 0x%08lx EFA: 0x%08lx ERET: 0x%08lx\nSTAT: 0x%08lx",
-		regs->event, current->thread.fault_address, regs->ret,
-		regs->status32);
+	pr_info("ECR: 0x%08lx EFA: 0x%08lx ERET: 0x%08lx\n",
+		regs->ecr.full, current->thread.fault_address, regs->ret);
+
+	pr_info("STAT32: 0x%08lx", regs->status32);
 
 #define STS_BIT(r, bit)	r->status32 & STATUS_##bit##_MASK ? #bit" " : ""
 
diff --git a/arch/arc/mm/fault.c b/arch/arc/mm/fault.c
index 26e5823c5710..95119a5e7761 100644
--- a/arch/arc/mm/fault.c
+++ b/arch/arc/mm/fault.c
@@ -100,10 +100,10 @@ void do_page_fault(unsigned long address, struct pt_regs *regs)
 	if (faulthandler_disabled() || !mm)
 		goto no_context;
 
-	if (regs->ecr_cause & ECR_C_PROTV_STORE)	/* ST/EX */
+	if (regs->ecr.cause & ECR_C_PROTV_STORE)	/* ST/EX */
 		write = 1;
-	else if ((regs->ecr_vec == ECR_V_PROTV) &&
-	         (regs->ecr_cause == ECR_C_PROTV_INST_FETCH))
+	else if ((regs->ecr.vec == ECR_V_PROTV) &&
+	         (regs->ecr.cause == ECR_C_PROTV_INST_FETCH))
 		exec = 1;
 
 	flags = FAULT_FLAG_DEFAULT;
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

Re: [PATCH 20/20] ARC: pt_regs: create seperate type for ecr

[Vineet Gupta]

On 8/17/23 05:09, Pavel.Kozlov@synopsys.com wrote:
> Hi Vineet,
>
> I'm testing your updates and ran into the same build issue reported by the build
> robot.
> http://lists.infradead.org/pipermail/linux-snps-arc/2023-August/007522.html
>
>> #define MAX_REG_OFFSET offsetof(struct pt_regs, event)
> This change causes a build issue for ARC700, as the event field has been
> removed and the MAX_REG_OFFSET macro hasn't been updated.

I've posted v2 for 3 patches. Please reapply/retest the whole series.

Thx,
-Vineet


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

Re: [PATCH 16/20] ARC: entry: Add more common chores to EXCEPTION_PROLOGUE

[Pavel.Kozlov]

Hi Vineet,

> Subject: [PATCH 16/20] 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)

I'm observing the ARC700 (nSIM) system freeze after this patch.

...
f0000000.serial: ttyS0 at MMIO 0xf0000000 (irq = 24, base_baud = 3125000) is a 16550A
printk: console [ttyS0] enabled
printk: console [ttyS0] enabled
printk: bootconsole [uart8250] disabled
printk: bootconsole [uart8250] disabled
NET: Registered PF_PACKET protocol family
NET: Registered PF_KEY protocol family
clk: Disabling unused clocks
Freeing unused kernel image (initmem) memory: 2856K
This architecture does not have kernel memory protection.
Run /init as init process

> @@ -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

The same update is also required for the call_do_page_fault wrapper for ARcompact. 

Regards,
Pavel



_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

Re: [PATCH 16/20] ARC: entry: Add more common chores to EXCEPTION_PROLOGUE

[Vineet Gupta]

On 8/18/23 05:56, Pavel.Kozlov@synopsys.com wrote:
> Hi Vineet,
>
>> Subject: [PATCH 16/20] 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)
> I'm observing the ARC700 (nSIM) system freeze after this patch.
>
> ...
> f0000000.serial: ttyS0 at MMIO 0xf0000000 (irq = 24, base_baud = 3125000) is a 16550A
> printk: console [ttyS0] enabled
> printk: console [ttyS0] enabled
> printk: bootconsole [uart8250] disabled
> printk: bootconsole [uart8250] disabled
> NET: Registered PF_PACKET protocol family
> NET: Registered PF_KEY protocol family
> clk: Disabling unused clocks
> Freeing unused kernel image (initmem) memory: 2856K
> This architecture does not have kernel memory protection.
> Run /init as init process
>
>> @@ -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
> The same update is also required for the call_do_page_fault wrapper for ARcompact.

Indeed I missed that part.

-Vineet

_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

[PATCH v2 16/20] ARC: entry: Add more common chores to EXCEPTION_PROLOGUE

[Vineet Gupta]

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/include/asm/entry-arcv2.h   | 12 +++++++++++-
 arch/arc/include/asm/entry-compact.h | 12 +++++++++++-
 arch/arc/kernel/entry-arcv2.S        | 15 ---------------
 arch/arc/kernel/entry-compact.S      | 13 -------------
 arch/arc/kernel/entry.S              | 19 ++-----------------
 5 files changed, 24 insertions(+), 47 deletions(-)

diff --git a/arch/arc/include/asm/entry-arcv2.h b/arch/arc/include/asm/entry-arcv2.h
index a38ed505b3de..11b48ab39154 100644
--- a/arch/arc/include/asm/entry-arcv2.h
+++ b/arch/arc/include/asm/entry-arcv2.h
@@ -75,7 +75,7 @@
 .endm
 
 /*------------------------------------------------------------------------*/
-.macro EXCEPTION_PROLOGUE
+.macro EXCEPTION_PROLOGUE_KEEP_AE
 
 	; Before jumping to Exception Vector, hardware micro-ops did following:
 	;   1. SP auto-switched to kernel mode stack
@@ -104,6 +104,16 @@
 	; OUTPUT: r10 has ECR expected by EV_Trap
 .endm
 
+.macro EXCEPTION_PROLOGUE
+
+	EXCEPTION_PROLOGUE_KEEP_AE	; return ECR in r10
+
+	lr  r0, [efa]
+	mov r1, sp
+
+	FAKE_RET_FROM_EXCPN		; clobbers r9
+.endm
+
 /*------------------------------------------------------------------------
  * This macro saves the registers manually which would normally be autosaved
  * by hardware on taken interrupts. It is used by
diff --git a/arch/arc/include/asm/entry-compact.h b/arch/arc/include/asm/entry-compact.h
index 4e2ae82779ed..a0e760eb35a8 100644
--- a/arch/arc/include/asm/entry-compact.h
+++ b/arch/arc/include/asm/entry-compact.h
@@ -140,7 +140,7 @@
  *
  * After this it is safe to call the "C" handlers
  *-------------------------------------------------------------*/
-.macro EXCEPTION_PROLOGUE
+.macro EXCEPTION_PROLOGUE_KEEP_AE
 
 	/* Need at least 1 reg to code the early exception prologue */
 	PROLOG_FREEUP_REG r9, @ex_saved_reg1
@@ -179,6 +179,16 @@
 	; OUTPUT: r10 has ECR expected by EV_Trap
 .endm
 
+.macro EXCEPTION_PROLOGUE
+
+	EXCEPTION_PROLOGUE_KEEP_AE	; return ECR in r10
+
+	lr  r0, [efa]
+	mov r1, sp
+
+	FAKE_RET_FROM_EXCPN		; clobbers r9
+.endm
+
 /*--------------------------------------------------------------
  * Restore all registers used by system call or Exceptions
  * SP should always be pointing to the next free stack element
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]
 
-- 
2.34.1


_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc

Re: [PATCH 00/20] ARC updates

[Pavel Kozlov]

> Hi,
> 
> This is a pile of miscll improvements/updates sitting in one of my local trees.
> Given the recent warning fix, we coudl also push them out.
> @Alexey, @Shahab: care to give these a spin on hsdk (and test ARC700 build/boot on nSIM if possible).
> 
> Thx,
> -Vineet

The entire "ARC updates" patch series (with all subsequent patches with updates)
has been tested on the HSDK borard for ARCv2 using the glibc and ksefltest test 
framworks and on nSIM for ARCompact using the uclibc tests. No regressions found.

Tested-by: Pavel Kozlov <Pavel.Kozlov@synopsys.com>

Regards,
Pavel

_______________________________________________
linux-snps-arc mailing list
linux-snps-arc@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/linux-snps-arc