summaryrefslogtreecommitdiffstats
path: root/kernel/arch/sparc
diff options
context:
space:
mode:
Diffstat (limited to 'kernel/arch/sparc')
-rw-r--r--kernel/arch/sparc/include/asm/head_64.h4
-rw-r--r--kernel/arch/sparc/include/asm/mmu_64.h3
-rw-r--r--kernel/arch/sparc/include/asm/pgtable_64.h43
-rw-r--r--kernel/arch/sparc/include/asm/tlbflush_64.h3
-rw-r--r--kernel/arch/sparc/include/asm/ttable.h8
-rw-r--r--kernel/arch/sparc/include/asm/uaccess_32.h4
-rw-r--r--kernel/arch/sparc/include/asm/uaccess_64.h66
-rw-r--r--kernel/arch/sparc/kernel/Makefile1
-rw-r--r--kernel/arch/sparc/kernel/cherrs.S14
-rw-r--r--kernel/arch/sparc/kernel/dtlb_prot.S4
-rw-r--r--kernel/arch/sparc/kernel/entry.S17
-rw-r--r--kernel/arch/sparc/kernel/fpu_traps.S11
-rw-r--r--kernel/arch/sparc/kernel/head_64.S60
-rw-r--r--kernel/arch/sparc/kernel/jump_label.c23
-rw-r--r--kernel/arch/sparc/kernel/ktlb.S12
-rw-r--r--kernel/arch/sparc/kernel/misctrap.S12
-rw-r--r--kernel/arch/sparc/kernel/pci.c17
-rw-r--r--kernel/arch/sparc/kernel/rtrap_64.S57
-rw-r--r--kernel/arch/sparc/kernel/signal32.c46
-rw-r--r--kernel/arch/sparc/kernel/signal_32.c41
-rw-r--r--kernel/arch/sparc/kernel/signal_64.c33
-rw-r--r--kernel/arch/sparc/kernel/sigutil_32.c9
-rw-r--r--kernel/arch/sparc/kernel/sigutil_64.c10
-rw-r--r--kernel/arch/sparc/kernel/sparc_ksyms_64.c1
-rw-r--r--kernel/arch/sparc/kernel/spiterrs.S18
-rw-r--r--kernel/arch/sparc/kernel/syscalls.S36
-rw-r--r--kernel/arch/sparc/kernel/tsb.S12
-rw-r--r--kernel/arch/sparc/kernel/urtt_fill.S98
-rw-r--r--kernel/arch/sparc/kernel/utrap.S3
-rw-r--r--kernel/arch/sparc/kernel/vmlinux.lds.S4
-rw-r--r--kernel/arch/sparc/kernel/winfixup.S3
-rw-r--r--kernel/arch/sparc/lib/GENcopy_from_user.S4
-rw-r--r--kernel/arch/sparc/lib/GENcopy_to_user.S4
-rw-r--r--kernel/arch/sparc/lib/GENmemcpy.S48
-rw-r--r--kernel/arch/sparc/lib/Makefile2
-rw-r--r--kernel/arch/sparc/lib/NG2copy_from_user.S8
-rw-r--r--kernel/arch/sparc/lib/NG2copy_to_user.S8
-rw-r--r--kernel/arch/sparc/lib/NG2memcpy.S228
-rw-r--r--kernel/arch/sparc/lib/NG4copy_from_user.S8
-rw-r--r--kernel/arch/sparc/lib/NG4copy_to_user.S8
-rw-r--r--kernel/arch/sparc/lib/NG4memcpy.S294
-rw-r--r--kernel/arch/sparc/lib/NGcopy_from_user.S4
-rw-r--r--kernel/arch/sparc/lib/NGcopy_to_user.S4
-rw-r--r--kernel/arch/sparc/lib/NGmemcpy.S233
-rw-r--r--kernel/arch/sparc/lib/U1copy_from_user.S8
-rw-r--r--kernel/arch/sparc/lib/U1copy_to_user.S8
-rw-r--r--kernel/arch/sparc/lib/U1memcpy.S345
-rw-r--r--kernel/arch/sparc/lib/U3copy_from_user.S8
-rw-r--r--kernel/arch/sparc/lib/U3copy_to_user.S8
-rw-r--r--kernel/arch/sparc/lib/U3memcpy.S227
-rw-r--r--kernel/arch/sparc/lib/copy_in_user.S35
-rw-r--r--kernel/arch/sparc/lib/user_fixup.c71
-rw-r--r--kernel/arch/sparc/mm/fault_64.c6
-rw-r--r--kernel/arch/sparc/mm/hugetlbpage.c37
-rw-r--r--kernel/arch/sparc/mm/init_64.c111
-rw-r--r--kernel/arch/sparc/mm/tlb.c29
-rw-r--r--kernel/arch/sparc/mm/tsb.c63
-rw-r--r--kernel/arch/sparc/mm/ultra.S374
58 files changed, 1882 insertions, 974 deletions
diff --git a/kernel/arch/sparc/include/asm/head_64.h b/kernel/arch/sparc/include/asm/head_64.h
index 10e9dabc4..f0700cfee 100644
--- a/kernel/arch/sparc/include/asm/head_64.h
+++ b/kernel/arch/sparc/include/asm/head_64.h
@@ -15,6 +15,10 @@
#define PTREGS_OFF (STACK_BIAS + STACKFRAME_SZ)
+#define RTRAP_PSTATE (PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV|PSTATE_IE)
+#define RTRAP_PSTATE_IRQOFF (PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV)
+#define RTRAP_PSTATE_AG_IRQOFF (PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV|PSTATE_AG)
+
#define __CHEETAH_ID 0x003e0014
#define __JALAPENO_ID 0x003e0016
#define __SERRANO_ID 0x003e0022
diff --git a/kernel/arch/sparc/include/asm/mmu_64.h b/kernel/arch/sparc/include/asm/mmu_64.h
index 70067ce18..f7de0dbc3 100644
--- a/kernel/arch/sparc/include/asm/mmu_64.h
+++ b/kernel/arch/sparc/include/asm/mmu_64.h
@@ -92,7 +92,8 @@ struct tsb_config {
typedef struct {
spinlock_t lock;
unsigned long sparc64_ctx_val;
- unsigned long huge_pte_count;
+ unsigned long hugetlb_pte_count;
+ unsigned long thp_pte_count;
struct tsb_config tsb_block[MM_NUM_TSBS];
struct hv_tsb_descr tsb_descr[MM_NUM_TSBS];
} mm_context_t;
diff --git a/kernel/arch/sparc/include/asm/pgtable_64.h b/kernel/arch/sparc/include/asm/pgtable_64.h
index 131d36fcd..408b715c9 100644
--- a/kernel/arch/sparc/include/asm/pgtable_64.h
+++ b/kernel/arch/sparc/include/asm/pgtable_64.h
@@ -375,7 +375,7 @@ static inline pgprot_t pgprot_noncached(pgprot_t prot)
#define pgprot_noncached pgprot_noncached
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
-static inline pte_t pte_mkhuge(pte_t pte)
+static inline unsigned long __pte_huge_mask(void)
{
unsigned long mask;
@@ -390,8 +390,19 @@ static inline pte_t pte_mkhuge(pte_t pte)
: "=r" (mask)
: "i" (_PAGE_SZHUGE_4U), "i" (_PAGE_SZHUGE_4V));
- return __pte(pte_val(pte) | mask);
+ return mask;
+}
+
+static inline pte_t pte_mkhuge(pte_t pte)
+{
+ return __pte(pte_val(pte) | __pte_huge_mask());
+}
+
+static inline bool is_hugetlb_pte(pte_t pte)
+{
+ return !!(pte_val(pte) & __pte_huge_mask());
}
+
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
static inline pmd_t pmd_mkhuge(pmd_t pmd)
{
@@ -403,6 +414,11 @@ static inline pmd_t pmd_mkhuge(pmd_t pmd)
return __pmd(pte_val(pte));
}
#endif
+#else
+static inline bool is_hugetlb_pte(pte_t pte)
+{
+ return false;
+}
#endif
static inline pte_t pte_mkdirty(pte_t pte)
@@ -865,6 +881,19 @@ static inline unsigned long pud_pfn(pud_t pud)
void tlb_batch_add(struct mm_struct *mm, unsigned long vaddr,
pte_t *ptep, pte_t orig, int fullmm);
+static void maybe_tlb_batch_add(struct mm_struct *mm, unsigned long vaddr,
+ pte_t *ptep, pte_t orig, int fullmm)
+{
+ /* It is more efficient to let flush_tlb_kernel_range()
+ * handle init_mm tlb flushes.
+ *
+ * SUN4V NOTE: _PAGE_VALID is the same value in both the SUN4U
+ * and SUN4V pte layout, so this inline test is fine.
+ */
+ if (likely(mm != &init_mm) && pte_accessible(mm, orig))
+ tlb_batch_add(mm, vaddr, ptep, orig, fullmm);
+}
+
#define __HAVE_ARCH_PMDP_HUGE_GET_AND_CLEAR
static inline pmd_t pmdp_huge_get_and_clear(struct mm_struct *mm,
unsigned long addr,
@@ -881,15 +910,7 @@ static inline void __set_pte_at(struct mm_struct *mm, unsigned long addr,
pte_t orig = *ptep;
*ptep = pte;
-
- /* It is more efficient to let flush_tlb_kernel_range()
- * handle init_mm tlb flushes.
- *
- * SUN4V NOTE: _PAGE_VALID is the same value in both the SUN4U
- * and SUN4V pte layout, so this inline test is fine.
- */
- if (likely(mm != &init_mm) && pte_accessible(mm, orig))
- tlb_batch_add(mm, addr, ptep, orig, fullmm);
+ maybe_tlb_batch_add(mm, addr, ptep, orig, fullmm);
}
#define set_pte_at(mm,addr,ptep,pte) \
diff --git a/kernel/arch/sparc/include/asm/tlbflush_64.h b/kernel/arch/sparc/include/asm/tlbflush_64.h
index dea1cfa21..a8e192e90 100644
--- a/kernel/arch/sparc/include/asm/tlbflush_64.h
+++ b/kernel/arch/sparc/include/asm/tlbflush_64.h
@@ -8,6 +8,7 @@
#define TLB_BATCH_NR 192
struct tlb_batch {
+ bool huge;
struct mm_struct *mm;
unsigned long tlb_nr;
unsigned long active;
@@ -16,7 +17,7 @@ struct tlb_batch {
void flush_tsb_kernel_range(unsigned long start, unsigned long end);
void flush_tsb_user(struct tlb_batch *tb);
-void flush_tsb_user_page(struct mm_struct *mm, unsigned long vaddr);
+void flush_tsb_user_page(struct mm_struct *mm, unsigned long vaddr, bool huge);
/* TLB flush operations. */
diff --git a/kernel/arch/sparc/include/asm/ttable.h b/kernel/arch/sparc/include/asm/ttable.h
index 71b5a6752..781b9f1db 100644
--- a/kernel/arch/sparc/include/asm/ttable.h
+++ b/kernel/arch/sparc/include/asm/ttable.h
@@ -589,8 +589,8 @@ user_rtt_fill_64bit: \
restored; \
nop; nop; nop; nop; nop; nop; \
nop; nop; nop; nop; nop; \
- ba,a,pt %xcc, user_rtt_fill_fixup; \
- ba,a,pt %xcc, user_rtt_fill_fixup; \
+ ba,a,pt %xcc, user_rtt_fill_fixup_dax; \
+ ba,a,pt %xcc, user_rtt_fill_fixup_mna; \
ba,a,pt %xcc, user_rtt_fill_fixup;
@@ -652,8 +652,8 @@ user_rtt_fill_32bit: \
restored; \
nop; nop; nop; nop; nop; \
nop; nop; nop; \
- ba,a,pt %xcc, user_rtt_fill_fixup; \
- ba,a,pt %xcc, user_rtt_fill_fixup; \
+ ba,a,pt %xcc, user_rtt_fill_fixup_dax; \
+ ba,a,pt %xcc, user_rtt_fill_fixup_mna; \
ba,a,pt %xcc, user_rtt_fill_fixup;
diff --git a/kernel/arch/sparc/include/asm/uaccess_32.h b/kernel/arch/sparc/include/asm/uaccess_32.h
index 64ee103dc..dfb542c7c 100644
--- a/kernel/arch/sparc/include/asm/uaccess_32.h
+++ b/kernel/arch/sparc/include/asm/uaccess_32.h
@@ -328,8 +328,10 @@ static inline unsigned long copy_from_user(void *to, const void __user *from, un
{
if (n && __access_ok((unsigned long) from, n))
return __copy_user((__force void __user *) to, from, n);
- else
+ else {
+ memset(to, 0, n);
return n;
+ }
}
static inline unsigned long __copy_from_user(void *to, const void __user *from, unsigned long n)
diff --git a/kernel/arch/sparc/include/asm/uaccess_64.h b/kernel/arch/sparc/include/asm/uaccess_64.h
index ea6e9a20f..f42851248 100644
--- a/kernel/arch/sparc/include/asm/uaccess_64.h
+++ b/kernel/arch/sparc/include/asm/uaccess_64.h
@@ -98,7 +98,6 @@ struct exception_table_entry {
unsigned int insn, fixup;
};
-void __ret_efault(void);
void __retl_efault(void);
/* Uh, these should become the main single-value transfer routines..
@@ -179,20 +178,6 @@ int __put_user_bad(void);
__gu_ret; \
})
-#define __get_user_nocheck_ret(data, addr, size, type, retval) ({ \
- register unsigned long __gu_val __asm__ ("l1"); \
- switch (size) { \
- case 1: __get_user_asm_ret(__gu_val, ub, addr, retval); break; \
- case 2: __get_user_asm_ret(__gu_val, uh, addr, retval); break; \
- case 4: __get_user_asm_ret(__gu_val, uw, addr, retval); break; \
- case 8: __get_user_asm_ret(__gu_val, x, addr, retval); break; \
- default: \
- if (__get_user_bad()) \
- return retval; \
- } \
- data = (__force type) __gu_val; \
-})
-
#define __get_user_asm(x, size, addr, ret) \
__asm__ __volatile__( \
"/* Get user asm, inline. */\n" \
@@ -214,80 +199,35 @@ __asm__ __volatile__( \
: "=r" (ret), "=r" (x) : "r" (__m(addr)), \
"i" (-EFAULT))
-#define __get_user_asm_ret(x, size, addr, retval) \
-if (__builtin_constant_p(retval) && retval == -EFAULT) \
- __asm__ __volatile__( \
- "/* Get user asm ret, inline. */\n" \
- "1:\t" "ld"#size "a [%1] %%asi, %0\n\n\t" \
- ".section __ex_table,\"a\"\n\t" \
- ".align 4\n\t" \
- ".word 1b,__ret_efault\n\n\t" \
- ".previous\n\t" \
- : "=r" (x) : "r" (__m(addr))); \
-else \
- __asm__ __volatile__( \
- "/* Get user asm ret, inline. */\n" \
- "1:\t" "ld"#size "a [%1] %%asi, %0\n\n\t" \
- ".section .fixup,#alloc,#execinstr\n\t" \
- ".align 4\n" \
- "3:\n\t" \
- "ret\n\t" \
- " restore %%g0, %2, %%o0\n\n\t" \
- ".previous\n\t" \
- ".section __ex_table,\"a\"\n\t" \
- ".align 4\n\t" \
- ".word 1b, 3b\n\n\t" \
- ".previous\n\t" \
- : "=r" (x) : "r" (__m(addr)), "i" (retval))
-
int __get_user_bad(void);
unsigned long __must_check ___copy_from_user(void *to,
const void __user *from,
unsigned long size);
-unsigned long copy_from_user_fixup(void *to, const void __user *from,
- unsigned long size);
static inline unsigned long __must_check
copy_from_user(void *to, const void __user *from, unsigned long size)
{
- unsigned long ret = ___copy_from_user(to, from, size);
-
- if (unlikely(ret))
- ret = copy_from_user_fixup(to, from, size);
-
- return ret;
+ return ___copy_from_user(to, from, size);
}
#define __copy_from_user copy_from_user
unsigned long __must_check ___copy_to_user(void __user *to,
const void *from,
unsigned long size);
-unsigned long copy_to_user_fixup(void __user *to, const void *from,
- unsigned long size);
static inline unsigned long __must_check
copy_to_user(void __user *to, const void *from, unsigned long size)
{
- unsigned long ret = ___copy_to_user(to, from, size);
-
- if (unlikely(ret))
- ret = copy_to_user_fixup(to, from, size);
- return ret;
+ return ___copy_to_user(to, from, size);
}
#define __copy_to_user copy_to_user
unsigned long __must_check ___copy_in_user(void __user *to,
const void __user *from,
unsigned long size);
-unsigned long copy_in_user_fixup(void __user *to, void __user *from,
- unsigned long size);
static inline unsigned long __must_check
copy_in_user(void __user *to, void __user *from, unsigned long size)
{
- unsigned long ret = ___copy_in_user(to, from, size);
-
- if (unlikely(ret))
- ret = copy_in_user_fixup(to, from, size);
- return ret;
+ return ___copy_in_user(to, from, size);
}
#define __copy_in_user copy_in_user
diff --git a/kernel/arch/sparc/kernel/Makefile b/kernel/arch/sparc/kernel/Makefile
index 7cf9c6ea3..fdb13327f 100644
--- a/kernel/arch/sparc/kernel/Makefile
+++ b/kernel/arch/sparc/kernel/Makefile
@@ -21,6 +21,7 @@ CFLAGS_REMOVE_perf_event.o := -pg
CFLAGS_REMOVE_pcr.o := -pg
endif
+obj-$(CONFIG_SPARC64) += urtt_fill.o
obj-$(CONFIG_SPARC32) += entry.o wof.o wuf.o
obj-$(CONFIG_SPARC32) += etrap_32.o
obj-$(CONFIG_SPARC32) += rtrap_32.o
diff --git a/kernel/arch/sparc/kernel/cherrs.S b/kernel/arch/sparc/kernel/cherrs.S
index 4ee1ad420..655628def 100644
--- a/kernel/arch/sparc/kernel/cherrs.S
+++ b/kernel/arch/sparc/kernel/cherrs.S
@@ -214,8 +214,7 @@ do_dcpe_tl1_nonfatal: /* Ok we may use interrupt globals safely. */
subcc %g1, %g2, %g1 ! Next cacheline
bge,pt %icc, 1b
nop
- ba,pt %xcc, dcpe_icpe_tl1_common
- nop
+ ba,a,pt %xcc, dcpe_icpe_tl1_common
do_dcpe_tl1_fatal:
sethi %hi(1f), %g7
@@ -224,8 +223,7 @@ do_dcpe_tl1_fatal:
mov 0x2, %o0
call cheetah_plus_parity_error
add %sp, PTREGS_OFF, %o1
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size do_dcpe_tl1,.-do_dcpe_tl1
.globl do_icpe_tl1
@@ -259,8 +257,7 @@ do_icpe_tl1_nonfatal: /* Ok we may use interrupt globals safely. */
subcc %g1, %g2, %g1
bge,pt %icc, 1b
nop
- ba,pt %xcc, dcpe_icpe_tl1_common
- nop
+ ba,a,pt %xcc, dcpe_icpe_tl1_common
do_icpe_tl1_fatal:
sethi %hi(1f), %g7
@@ -269,8 +266,7 @@ do_icpe_tl1_fatal:
mov 0x3, %o0
call cheetah_plus_parity_error
add %sp, PTREGS_OFF, %o1
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size do_icpe_tl1,.-do_icpe_tl1
.type dcpe_icpe_tl1_common,#function
@@ -456,7 +452,7 @@ __cheetah_log_error:
cmp %g2, 0x63
be c_cee
nop
- ba,pt %xcc, c_deferred
+ ba,a,pt %xcc, c_deferred
.size __cheetah_log_error,.-__cheetah_log_error
/* Cheetah FECC trap handling, we get here from tl{0,1}_fecc
diff --git a/kernel/arch/sparc/kernel/dtlb_prot.S b/kernel/arch/sparc/kernel/dtlb_prot.S
index d668ca149..4087a62f9 100644
--- a/kernel/arch/sparc/kernel/dtlb_prot.S
+++ b/kernel/arch/sparc/kernel/dtlb_prot.S
@@ -25,13 +25,13 @@
/* PROT ** ICACHE line 2: More real fault processing */
ldxa [%g4] ASI_DMMU, %g5 ! Put tagaccess in %g5
+ srlx %g5, PAGE_SHIFT, %g5
+ sllx %g5, PAGE_SHIFT, %g5 ! Clear context ID bits
bgu,pn %xcc, winfix_trampoline ! Yes, perform winfixup
mov FAULT_CODE_DTLB | FAULT_CODE_WRITE, %g4
ba,pt %xcc, sparc64_realfault_common ! Nope, normal fault
nop
nop
- nop
- nop
/* PROT ** ICACHE line 3: Unused... */
nop
diff --git a/kernel/arch/sparc/kernel/entry.S b/kernel/arch/sparc/kernel/entry.S
index 33c02b15f..a83707c83 100644
--- a/kernel/arch/sparc/kernel/entry.S
+++ b/kernel/arch/sparc/kernel/entry.S
@@ -948,7 +948,24 @@ linux_syscall_trace:
cmp %o0, 0
bne 3f
mov -ENOSYS, %o0
+
+ /* Syscall tracing can modify the registers. */
+ ld [%sp + STACKFRAME_SZ + PT_G1], %g1
+ sethi %hi(sys_call_table), %l7
+ ld [%sp + STACKFRAME_SZ + PT_I0], %i0
+ or %l7, %lo(sys_call_table), %l7
+ ld [%sp + STACKFRAME_SZ + PT_I1], %i1
+ ld [%sp + STACKFRAME_SZ + PT_I2], %i2
+ ld [%sp + STACKFRAME_SZ + PT_I3], %i3
+ ld [%sp + STACKFRAME_SZ + PT_I4], %i4
+ ld [%sp + STACKFRAME_SZ + PT_I5], %i5
+ cmp %g1, NR_syscalls
+ bgeu 3f
+ mov -ENOSYS, %o0
+
+ sll %g1, 2, %l4
mov %i0, %o0
+ ld [%l7 + %l4], %l7
mov %i1, %o1
mov %i2, %o2
mov %i3, %o3
diff --git a/kernel/arch/sparc/kernel/fpu_traps.S b/kernel/arch/sparc/kernel/fpu_traps.S
index a6864826a..336d2750f 100644
--- a/kernel/arch/sparc/kernel/fpu_traps.S
+++ b/kernel/arch/sparc/kernel/fpu_traps.S
@@ -100,8 +100,8 @@ do_fpdis:
fmuld %f0, %f2, %f26
faddd %f0, %f2, %f28
fmuld %f0, %f2, %f30
- b,pt %xcc, fpdis_exit
- nop
+ ba,a,pt %xcc, fpdis_exit
+
2: andcc %g5, FPRS_DU, %g0
bne,pt %icc, 3f
fzero %f32
@@ -144,8 +144,8 @@ do_fpdis:
fmuld %f32, %f34, %f58
faddd %f32, %f34, %f60
fmuld %f32, %f34, %f62
- ba,pt %xcc, fpdis_exit
- nop
+ ba,a,pt %xcc, fpdis_exit
+
3: mov SECONDARY_CONTEXT, %g3
add %g6, TI_FPREGS, %g1
@@ -197,8 +197,7 @@ fpdis_exit2:
fp_other_bounce:
call do_fpother
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size fp_other_bounce,.-fp_other_bounce
.align 32
diff --git a/kernel/arch/sparc/kernel/head_64.S b/kernel/arch/sparc/kernel/head_64.S
index f2d30cab5..7eeeb1d5a 100644
--- a/kernel/arch/sparc/kernel/head_64.S
+++ b/kernel/arch/sparc/kernel/head_64.S
@@ -461,9 +461,8 @@ sun4v_chip_type:
subcc %g3, 1, %g3
bne,pt %xcc, 41b
add %g1, 1, %g1
- mov SUN4V_CHIP_SPARC64X, %g4
ba,pt %xcc, 5f
- nop
+ mov SUN4V_CHIP_SPARC64X, %g4
49:
mov SUN4V_CHIP_UNKNOWN, %g4
@@ -548,8 +547,7 @@ sun4u_init:
stxa %g0, [%g7] ASI_DMMU
membar #Sync
- ba,pt %xcc, sun4u_continue
- nop
+ ba,a,pt %xcc, sun4u_continue
sun4v_init:
/* Set ctx 0 */
@@ -560,14 +558,12 @@ sun4v_init:
mov SECONDARY_CONTEXT, %g7
stxa %g0, [%g7] ASI_MMU
membar #Sync
- ba,pt %xcc, niagara_tlb_fixup
- nop
+ ba,a,pt %xcc, niagara_tlb_fixup
sun4u_continue:
BRANCH_IF_ANY_CHEETAH(g1, g7, cheetah_tlb_fixup)
- ba,pt %xcc, spitfire_tlb_fixup
- nop
+ ba,a,pt %xcc, spitfire_tlb_fixup
niagara_tlb_fixup:
mov 3, %g2 /* Set TLB type to hypervisor. */
@@ -639,8 +635,7 @@ niagara_patch:
call hypervisor_patch_cachetlbops
nop
- ba,pt %xcc, tlb_fixup_done
- nop
+ ba,a,pt %xcc, tlb_fixup_done
cheetah_tlb_fixup:
mov 2, %g2 /* Set TLB type to cheetah+. */
@@ -659,8 +654,7 @@ cheetah_tlb_fixup:
call cheetah_patch_cachetlbops
nop
- ba,pt %xcc, tlb_fixup_done
- nop
+ ba,a,pt %xcc, tlb_fixup_done
spitfire_tlb_fixup:
/* Set TLB type to spitfire. */
@@ -782,8 +776,7 @@ setup_trap_table:
call %o1
add %sp, (2047 + 128), %o0
- ba,pt %xcc, 2f
- nop
+ ba,a,pt %xcc, 2f
1: sethi %hi(sparc64_ttable_tl0), %o0
set prom_set_trap_table_name, %g2
@@ -822,8 +815,7 @@ setup_trap_table:
BRANCH_IF_ANY_CHEETAH(o2, o3, 1f)
- ba,pt %xcc, 2f
- nop
+ ba,a,pt %xcc, 2f
/* Disable STICK_INT interrupts. */
1:
@@ -930,47 +922,11 @@ prom_tba: .xword 0
tlb_type: .word 0 /* Must NOT end up in BSS */
.section ".fixup",#alloc,#execinstr
- .globl __ret_efault, __retl_efault, __ret_one, __retl_one
-ENTRY(__ret_efault)
- ret
- restore %g0, -EFAULT, %o0
-ENDPROC(__ret_efault)
-
ENTRY(__retl_efault)
retl
mov -EFAULT, %o0
ENDPROC(__retl_efault)
-ENTRY(__retl_one)
- retl
- mov 1, %o0
-ENDPROC(__retl_one)
-
-ENTRY(__retl_one_fp)
- VISExitHalf
- retl
- mov 1, %o0
-ENDPROC(__retl_one_fp)
-
-ENTRY(__ret_one_asi)
- wr %g0, ASI_AIUS, %asi
- ret
- restore %g0, 1, %o0
-ENDPROC(__ret_one_asi)
-
-ENTRY(__retl_one_asi)
- wr %g0, ASI_AIUS, %asi
- retl
- mov 1, %o0
-ENDPROC(__retl_one_asi)
-
-ENTRY(__retl_one_asi_fp)
- wr %g0, ASI_AIUS, %asi
- VISExitHalf
- retl
- mov 1, %o0
-ENDPROC(__retl_one_asi_fp)
-
ENTRY(__retl_o1)
retl
mov %o1, %o0
diff --git a/kernel/arch/sparc/kernel/jump_label.c b/kernel/arch/sparc/kernel/jump_label.c
index 59bbeff55..07933b9e9 100644
--- a/kernel/arch/sparc/kernel/jump_label.c
+++ b/kernel/arch/sparc/kernel/jump_label.c
@@ -13,19 +13,30 @@
void arch_jump_label_transform(struct jump_entry *entry,
enum jump_label_type type)
{
- u32 val;
u32 *insn = (u32 *) (unsigned long) entry->code;
+ u32 val;
if (type == JUMP_LABEL_JMP) {
s32 off = (s32)entry->target - (s32)entry->code;
+ bool use_v9_branch = false;
+
+ BUG_ON(off & 3);
#ifdef CONFIG_SPARC64
- /* ba,pt %xcc, . + (off << 2) */
- val = 0x10680000 | ((u32) off >> 2);
-#else
- /* ba . + (off << 2) */
- val = 0x10800000 | ((u32) off >> 2);
+ if (off <= 0xfffff && off >= -0x100000)
+ use_v9_branch = true;
#endif
+ if (use_v9_branch) {
+ /* WDISP19 - target is . + immed << 2 */
+ /* ba,pt %xcc, . + off */
+ val = 0x10680000 | (((u32) off >> 2) & 0x7ffff);
+ } else {
+ /* WDISP22 - target is . + immed << 2 */
+ BUG_ON(off > 0x7fffff);
+ BUG_ON(off < -0x800000);
+ /* ba . + off */
+ val = 0x10800000 | (((u32) off >> 2) & 0x3fffff);
+ }
} else {
val = 0x01000000;
}
diff --git a/kernel/arch/sparc/kernel/ktlb.S b/kernel/arch/sparc/kernel/ktlb.S
index ef0d8e9e1..f22bec0db 100644
--- a/kernel/arch/sparc/kernel/ktlb.S
+++ b/kernel/arch/sparc/kernel/ktlb.S
@@ -20,6 +20,10 @@ kvmap_itlb:
mov TLB_TAG_ACCESS, %g4
ldxa [%g4] ASI_IMMU, %g4
+ /* The kernel executes in context zero, therefore we do not
+ * need to clear the context ID bits out of %g4 here.
+ */
+
/* sun4v_itlb_miss branches here with the missing virtual
* address already loaded into %g4
*/
@@ -128,6 +132,10 @@ kvmap_dtlb:
mov TLB_TAG_ACCESS, %g4
ldxa [%g4] ASI_DMMU, %g4
+ /* The kernel executes in context zero, therefore we do not
+ * need to clear the context ID bits out of %g4 here.
+ */
+
/* sun4v_dtlb_miss branches here with the missing virtual
* address already loaded into %g4
*/
@@ -251,6 +259,10 @@ kvmap_dtlb_longpath:
nop
.previous
+ /* The kernel executes in context zero, therefore we do not
+ * need to clear the context ID bits out of %g5 here.
+ */
+
be,pt %xcc, sparc64_realfault_common
mov FAULT_CODE_DTLB, %g4
ba,pt %xcc, winfix_trampoline
diff --git a/kernel/arch/sparc/kernel/misctrap.S b/kernel/arch/sparc/kernel/misctrap.S
index 753b4f031..34b493390 100644
--- a/kernel/arch/sparc/kernel/misctrap.S
+++ b/kernel/arch/sparc/kernel/misctrap.S
@@ -18,8 +18,7 @@ __do_privact:
109: or %g7, %lo(109b), %g7
call do_privact
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size __do_privact,.-__do_privact
.type do_mna,#function
@@ -46,8 +45,7 @@ do_mna:
mov %l5, %o2
call mem_address_unaligned
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size do_mna,.-do_mna
.type do_lddfmna,#function
@@ -65,8 +63,7 @@ do_lddfmna:
mov %l5, %o2
call handle_lddfmna
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size do_lddfmna,.-do_lddfmna
.type do_stdfmna,#function
@@ -84,8 +81,7 @@ do_stdfmna:
mov %l5, %o2
call handle_stdfmna
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size do_stdfmna,.-do_stdfmna
.type breakpoint_trap,#function
diff --git a/kernel/arch/sparc/kernel/pci.c b/kernel/arch/sparc/kernel/pci.c
index badf0951d..9f9614df9 100644
--- a/kernel/arch/sparc/kernel/pci.c
+++ b/kernel/arch/sparc/kernel/pci.c
@@ -994,6 +994,23 @@ void pcibios_set_master(struct pci_dev *dev)
/* No special bus mastering setup handling */
}
+#ifdef CONFIG_PCI_IOV
+int pcibios_add_device(struct pci_dev *dev)
+{
+ struct pci_dev *pdev;
+
+ /* Add sriov arch specific initialization here.
+ * Copy dev_archdata from PF to VF
+ */
+ if (dev->is_virtfn) {
+ pdev = dev->physfn;
+ memcpy(&dev->dev.archdata, &pdev->dev.archdata,
+ sizeof(struct dev_archdata));
+ }
+ return 0;
+}
+#endif /* CONFIG_PCI_IOV */
+
static int __init pcibios_init(void)
{
pci_dfl_cache_line_size = 64 >> 2;
diff --git a/kernel/arch/sparc/kernel/rtrap_64.S b/kernel/arch/sparc/kernel/rtrap_64.S
index d08bdaffd..216948ca4 100644
--- a/kernel/arch/sparc/kernel/rtrap_64.S
+++ b/kernel/arch/sparc/kernel/rtrap_64.S
@@ -14,10 +14,6 @@
#include <asm/visasm.h>
#include <asm/processor.h>
-#define RTRAP_PSTATE (PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV|PSTATE_IE)
-#define RTRAP_PSTATE_IRQOFF (PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV)
-#define RTRAP_PSTATE_AG_IRQOFF (PSTATE_TSO|PSTATE_PEF|PSTATE_PRIV|PSTATE_AG)
-
#ifdef CONFIG_CONTEXT_TRACKING
# define SCHEDULE_USER schedule_user
#else
@@ -242,52 +238,17 @@ rt_continue: ldx [%sp + PTREGS_OFF + PT_V9_G1], %g1
wrpr %g1, %cwp
ba,a,pt %xcc, user_rtt_fill_64bit
-user_rtt_fill_fixup:
- rdpr %cwp, %g1
- add %g1, 1, %g1
- wrpr %g1, 0x0, %cwp
-
- rdpr %wstate, %g2
- sll %g2, 3, %g2
- wrpr %g2, 0x0, %wstate
-
- /* We know %canrestore and %otherwin are both zero. */
-
- sethi %hi(sparc64_kern_pri_context), %g2
- ldx [%g2 + %lo(sparc64_kern_pri_context)], %g2
- mov PRIMARY_CONTEXT, %g1
-
-661: stxa %g2, [%g1] ASI_DMMU
- .section .sun4v_1insn_patch, "ax"
- .word 661b
- stxa %g2, [%g1] ASI_MMU
- .previous
-
- sethi %hi(KERNBASE), %g1
- flush %g1
+user_rtt_fill_fixup_dax:
+ ba,pt %xcc, user_rtt_fill_fixup_common
+ mov 1, %g3
- or %g4, FAULT_CODE_WINFIXUP, %g4
- stb %g4, [%g6 + TI_FAULT_CODE]
- stx %g5, [%g6 + TI_FAULT_ADDR]
+user_rtt_fill_fixup_mna:
+ ba,pt %xcc, user_rtt_fill_fixup_common
+ mov 2, %g3
- mov %g6, %l1
- wrpr %g0, 0x0, %tl
-
-661: nop
- .section .sun4v_1insn_patch, "ax"
- .word 661b
- SET_GL(0)
- .previous
-
- wrpr %g0, RTRAP_PSTATE, %pstate
-
- mov %l1, %g6
- ldx [%g6 + TI_TASK], %g4
- LOAD_PER_CPU_BASE(%g5, %g6, %g1, %g2, %g3)
- call do_sparc64_fault
- add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+user_rtt_fill_fixup:
+ ba,pt %xcc, user_rtt_fill_fixup_common
+ clr %g3
user_rtt_pre_restore:
add %g1, 1, %g1
diff --git a/kernel/arch/sparc/kernel/signal32.c b/kernel/arch/sparc/kernel/signal32.c
index 4eed773a7..77655f0f0 100644
--- a/kernel/arch/sparc/kernel/signal32.c
+++ b/kernel/arch/sparc/kernel/signal32.c
@@ -138,12 +138,24 @@ int copy_siginfo_from_user32(siginfo_t *to, compat_siginfo_t __user *from)
return 0;
}
+/* Checks if the fp is valid. We always build signal frames which are
+ * 16-byte aligned, therefore we can always enforce that the restore
+ * frame has that property as well.
+ */
+static bool invalid_frame_pointer(void __user *fp, int fplen)
+{
+ if ((((unsigned long) fp) & 15) ||
+ ((unsigned long)fp) > 0x100000000ULL - fplen)
+ return true;
+ return false;
+}
+
void do_sigreturn32(struct pt_regs *regs)
{
struct signal_frame32 __user *sf;
compat_uptr_t fpu_save;
compat_uptr_t rwin_save;
- unsigned int psr;
+ unsigned int psr, ufp;
unsigned pc, npc;
sigset_t set;
compat_sigset_t seta;
@@ -158,11 +170,16 @@ void do_sigreturn32(struct pt_regs *regs)
sf = (struct signal_frame32 __user *) regs->u_regs[UREG_FP];
/* 1. Make sure we are not getting garbage from the user */
- if (!access_ok(VERIFY_READ, sf, sizeof(*sf)) ||
- (((unsigned long) sf) & 3))
+ if (invalid_frame_pointer(sf, sizeof(*sf)))
+ goto segv;
+
+ if (get_user(ufp, &sf->info.si_regs.u_regs[UREG_FP]))
+ goto segv;
+
+ if (ufp & 0x7)
goto segv;
- if (get_user(pc, &sf->info.si_regs.pc) ||
+ if (__get_user(pc, &sf->info.si_regs.pc) ||
__get_user(npc, &sf->info.si_regs.npc))
goto segv;
@@ -227,7 +244,7 @@ segv:
asmlinkage void do_rt_sigreturn32(struct pt_regs *regs)
{
struct rt_signal_frame32 __user *sf;
- unsigned int psr, pc, npc;
+ unsigned int psr, pc, npc, ufp;
compat_uptr_t fpu_save;
compat_uptr_t rwin_save;
sigset_t set;
@@ -242,11 +259,16 @@ asmlinkage void do_rt_sigreturn32(struct pt_regs *regs)
sf = (struct rt_signal_frame32 __user *) regs->u_regs[UREG_FP];
/* 1. Make sure we are not getting garbage from the user */
- if (!access_ok(VERIFY_READ, sf, sizeof(*sf)) ||
- (((unsigned long) sf) & 3))
+ if (invalid_frame_pointer(sf, sizeof(*sf)))
goto segv;
- if (get_user(pc, &sf->regs.pc) ||
+ if (get_user(ufp, &sf->regs.u_regs[UREG_FP]))
+ goto segv;
+
+ if (ufp & 0x7)
+ goto segv;
+
+ if (__get_user(pc, &sf->regs.pc) ||
__get_user(npc, &sf->regs.npc))
goto segv;
@@ -307,14 +329,6 @@ segv:
force_sig(SIGSEGV, current);
}
-/* Checks if the fp is valid */
-static int invalid_frame_pointer(void __user *fp, int fplen)
-{
- if ((((unsigned long) fp) & 7) || ((unsigned long)fp) > 0x100000000ULL - fplen)
- return 1;
- return 0;
-}
-
static void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs, unsigned long framesize)
{
unsigned long sp;
diff --git a/kernel/arch/sparc/kernel/signal_32.c b/kernel/arch/sparc/kernel/signal_32.c
index 52aa5e4ce..9c0c8fd0b 100644
--- a/kernel/arch/sparc/kernel/signal_32.c
+++ b/kernel/arch/sparc/kernel/signal_32.c
@@ -60,10 +60,22 @@ struct rt_signal_frame {
#define SF_ALIGNEDSZ (((sizeof(struct signal_frame) + 7) & (~7)))
#define RT_ALIGNEDSZ (((sizeof(struct rt_signal_frame) + 7) & (~7)))
+/* Checks if the fp is valid. We always build signal frames which are
+ * 16-byte aligned, therefore we can always enforce that the restore
+ * frame has that property as well.
+ */
+static inline bool invalid_frame_pointer(void __user *fp, int fplen)
+{
+ if ((((unsigned long) fp) & 15) || !__access_ok((unsigned long)fp, fplen))
+ return true;
+
+ return false;
+}
+
asmlinkage void do_sigreturn(struct pt_regs *regs)
{
+ unsigned long up_psr, pc, npc, ufp;
struct signal_frame __user *sf;
- unsigned long up_psr, pc, npc;
sigset_t set;
__siginfo_fpu_t __user *fpu_save;
__siginfo_rwin_t __user *rwin_save;
@@ -77,10 +89,13 @@ asmlinkage void do_sigreturn(struct pt_regs *regs)
sf = (struct signal_frame __user *) regs->u_regs[UREG_FP];
/* 1. Make sure we are not getting garbage from the user */
- if (!access_ok(VERIFY_READ, sf, sizeof(*sf)))
+ if (invalid_frame_pointer(sf, sizeof(*sf)))
+ goto segv_and_exit;
+
+ if (get_user(ufp, &sf->info.si_regs.u_regs[UREG_FP]))
goto segv_and_exit;
- if (((unsigned long) sf) & 3)
+ if (ufp & 0x7)
goto segv_and_exit;
err = __get_user(pc, &sf->info.si_regs.pc);
@@ -127,7 +142,7 @@ segv_and_exit:
asmlinkage void do_rt_sigreturn(struct pt_regs *regs)
{
struct rt_signal_frame __user *sf;
- unsigned int psr, pc, npc;
+ unsigned int psr, pc, npc, ufp;
__siginfo_fpu_t __user *fpu_save;
__siginfo_rwin_t __user *rwin_save;
sigset_t set;
@@ -135,8 +150,13 @@ asmlinkage void do_rt_sigreturn(struct pt_regs *regs)
synchronize_user_stack();
sf = (struct rt_signal_frame __user *) regs->u_regs[UREG_FP];
- if (!access_ok(VERIFY_READ, sf, sizeof(*sf)) ||
- (((unsigned long) sf) & 0x03))
+ if (invalid_frame_pointer(sf, sizeof(*sf)))
+ goto segv;
+
+ if (get_user(ufp, &sf->regs.u_regs[UREG_FP]))
+ goto segv;
+
+ if (ufp & 0x7)
goto segv;
err = __get_user(pc, &sf->regs.pc);
@@ -178,15 +198,6 @@ segv:
force_sig(SIGSEGV, current);
}
-/* Checks if the fp is valid */
-static inline int invalid_frame_pointer(void __user *fp, int fplen)
-{
- if ((((unsigned long) fp) & 7) || !__access_ok((unsigned long)fp, fplen))
- return 1;
-
- return 0;
-}
-
static inline void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs, unsigned long framesize)
{
unsigned long sp = regs->u_regs[UREG_FP];
diff --git a/kernel/arch/sparc/kernel/signal_64.c b/kernel/arch/sparc/kernel/signal_64.c
index d88beff47..5ee930c48 100644
--- a/kernel/arch/sparc/kernel/signal_64.c
+++ b/kernel/arch/sparc/kernel/signal_64.c
@@ -52,7 +52,7 @@ asmlinkage void sparc64_set_context(struct pt_regs *regs)
unsigned char fenab;
int err;
- flush_user_windows();
+ synchronize_user_stack();
if (get_thread_wsaved() ||
(((unsigned long)ucp) & (sizeof(unsigned long)-1)) ||
(!__access_ok(ucp, sizeof(*ucp))))
@@ -234,6 +234,17 @@ do_sigsegv:
goto out;
}
+/* Checks if the fp is valid. We always build rt signal frames which
+ * are 16-byte aligned, therefore we can always enforce that the
+ * restore frame has that property as well.
+ */
+static bool invalid_frame_pointer(void __user *fp)
+{
+ if (((unsigned long) fp) & 15)
+ return true;
+ return false;
+}
+
struct rt_signal_frame {
struct sparc_stackf ss;
siginfo_t info;
@@ -246,8 +257,8 @@ struct rt_signal_frame {
void do_rt_sigreturn(struct pt_regs *regs)
{
+ unsigned long tpc, tnpc, tstate, ufp;
struct rt_signal_frame __user *sf;
- unsigned long tpc, tnpc, tstate;
__siginfo_fpu_t __user *fpu_save;
__siginfo_rwin_t __user *rwin_save;
sigset_t set;
@@ -261,10 +272,16 @@ void do_rt_sigreturn(struct pt_regs *regs)
(regs->u_regs [UREG_FP] + STACK_BIAS);
/* 1. Make sure we are not getting garbage from the user */
- if (((unsigned long) sf) & 3)
+ if (invalid_frame_pointer(sf))
+ goto segv;
+
+ if (get_user(ufp, &sf->regs.u_regs[UREG_FP]))
goto segv;
- err = get_user(tpc, &sf->regs.tpc);
+ if ((ufp + STACK_BIAS) & 0x7)
+ goto segv;
+
+ err = __get_user(tpc, &sf->regs.tpc);
err |= __get_user(tnpc, &sf->regs.tnpc);
if (test_thread_flag(TIF_32BIT)) {
tpc &= 0xffffffff;
@@ -308,14 +325,6 @@ segv:
force_sig(SIGSEGV, current);
}
-/* Checks if the fp is valid */
-static int invalid_frame_pointer(void __user *fp)
-{
- if (((unsigned long) fp) & 15)
- return 1;
- return 0;
-}
-
static inline void __user *get_sigframe(struct ksignal *ksig, struct pt_regs *regs, unsigned long framesize)
{
unsigned long sp = regs->u_regs[UREG_FP] + STACK_BIAS;
diff --git a/kernel/arch/sparc/kernel/sigutil_32.c b/kernel/arch/sparc/kernel/sigutil_32.c
index 0f6eebe71..e5fe8cef9 100644
--- a/kernel/arch/sparc/kernel/sigutil_32.c
+++ b/kernel/arch/sparc/kernel/sigutil_32.c
@@ -48,6 +48,10 @@ int save_fpu_state(struct pt_regs *regs, __siginfo_fpu_t __user *fpu)
int restore_fpu_state(struct pt_regs *regs, __siginfo_fpu_t __user *fpu)
{
int err;
+
+ if (((unsigned long) fpu) & 3)
+ return -EFAULT;
+
#ifdef CONFIG_SMP
if (test_tsk_thread_flag(current, TIF_USEDFPU))
regs->psr &= ~PSR_EF;
@@ -97,7 +101,10 @@ int restore_rwin_state(__siginfo_rwin_t __user *rp)
struct thread_info *t = current_thread_info();
int i, wsaved, err;
- __get_user(wsaved, &rp->wsaved);
+ if (((unsigned long) rp) & 3)
+ return -EFAULT;
+
+ get_user(wsaved, &rp->wsaved);
if (wsaved > NSWINS)
return -EFAULT;
diff --git a/kernel/arch/sparc/kernel/sigutil_64.c b/kernel/arch/sparc/kernel/sigutil_64.c
index 387834a9c..36aadcbea 100644
--- a/kernel/arch/sparc/kernel/sigutil_64.c
+++ b/kernel/arch/sparc/kernel/sigutil_64.c
@@ -37,7 +37,10 @@ int restore_fpu_state(struct pt_regs *regs, __siginfo_fpu_t __user *fpu)
unsigned long fprs;
int err;
- err = __get_user(fprs, &fpu->si_fprs);
+ if (((unsigned long) fpu) & 7)
+ return -EFAULT;
+
+ err = get_user(fprs, &fpu->si_fprs);
fprs_write(0);
regs->tstate &= ~TSTATE_PEF;
if (fprs & FPRS_DL)
@@ -72,7 +75,10 @@ int restore_rwin_state(__siginfo_rwin_t __user *rp)
struct thread_info *t = current_thread_info();
int i, wsaved, err;
- __get_user(wsaved, &rp->wsaved);
+ if (((unsigned long) rp) & 7)
+ return -EFAULT;
+
+ get_user(wsaved, &rp->wsaved);
if (wsaved > NSWINS)
return -EFAULT;
diff --git a/kernel/arch/sparc/kernel/sparc_ksyms_64.c b/kernel/arch/sparc/kernel/sparc_ksyms_64.c
index a92d5d2c4..51b25325a 100644
--- a/kernel/arch/sparc/kernel/sparc_ksyms_64.c
+++ b/kernel/arch/sparc/kernel/sparc_ksyms_64.c
@@ -27,7 +27,6 @@ EXPORT_SYMBOL(__flushw_user);
EXPORT_SYMBOL_GPL(real_hard_smp_processor_id);
/* from head_64.S */
-EXPORT_SYMBOL(__ret_efault);
EXPORT_SYMBOL(tlb_type);
EXPORT_SYMBOL(sun4v_chip_type);
EXPORT_SYMBOL(prom_root_node);
diff --git a/kernel/arch/sparc/kernel/spiterrs.S b/kernel/arch/sparc/kernel/spiterrs.S
index c357e40ff..4a73009f6 100644
--- a/kernel/arch/sparc/kernel/spiterrs.S
+++ b/kernel/arch/sparc/kernel/spiterrs.S
@@ -85,8 +85,7 @@ __spitfire_cee_trap_continue:
ba,pt %xcc, etraptl1
rd %pc, %g7
- ba,pt %xcc, 2f
- nop
+ ba,a,pt %xcc, 2f
1: ba,pt %xcc, etrap_irq
rd %pc, %g7
@@ -100,8 +99,7 @@ __spitfire_cee_trap_continue:
mov %l5, %o2
call spitfire_access_error
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size __spitfire_access_error,.-__spitfire_access_error
/* This is the trap handler entry point for ECC correctable
@@ -179,8 +177,7 @@ __spitfire_data_access_exception_tl1:
mov %l5, %o2
call spitfire_data_access_exception_tl1
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size __spitfire_data_access_exception_tl1,.-__spitfire_data_access_exception_tl1
.type __spitfire_data_access_exception,#function
@@ -200,8 +197,7 @@ __spitfire_data_access_exception:
mov %l5, %o2
call spitfire_data_access_exception
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size __spitfire_data_access_exception,.-__spitfire_data_access_exception
.type __spitfire_insn_access_exception_tl1,#function
@@ -220,8 +216,7 @@ __spitfire_insn_access_exception_tl1:
mov %l5, %o2
call spitfire_insn_access_exception_tl1
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size __spitfire_insn_access_exception_tl1,.-__spitfire_insn_access_exception_tl1
.type __spitfire_insn_access_exception,#function
@@ -240,6 +235,5 @@ __spitfire_insn_access_exception:
mov %l5, %o2
call spitfire_insn_access_exception
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
.size __spitfire_insn_access_exception,.-__spitfire_insn_access_exception
diff --git a/kernel/arch/sparc/kernel/syscalls.S b/kernel/arch/sparc/kernel/syscalls.S
index bb0008927..c4a1b5c40 100644
--- a/kernel/arch/sparc/kernel/syscalls.S
+++ b/kernel/arch/sparc/kernel/syscalls.S
@@ -158,7 +158,25 @@ linux_syscall_trace32:
add %sp, PTREGS_OFF, %o0
brnz,pn %o0, 3f
mov -ENOSYS, %o0
+
+ /* Syscall tracing can modify the registers. */
+ ldx [%sp + PTREGS_OFF + PT_V9_G1], %g1
+ sethi %hi(sys_call_table32), %l7
+ ldx [%sp + PTREGS_OFF + PT_V9_I0], %i0
+ or %l7, %lo(sys_call_table32), %l7
+ ldx [%sp + PTREGS_OFF + PT_V9_I1], %i1
+ ldx [%sp + PTREGS_OFF + PT_V9_I2], %i2
+ ldx [%sp + PTREGS_OFF + PT_V9_I3], %i3
+ ldx [%sp + PTREGS_OFF + PT_V9_I4], %i4
+ ldx [%sp + PTREGS_OFF + PT_V9_I5], %i5
+
+ cmp %g1, NR_syscalls
+ bgeu,pn %xcc, 3f
+ mov -ENOSYS, %o0
+
+ sll %g1, 2, %l4
srl %i0, 0, %o0
+ lduw [%l7 + %l4], %l7
srl %i4, 0, %o4
srl %i1, 0, %o1
srl %i2, 0, %o2
@@ -170,7 +188,25 @@ linux_syscall_trace:
add %sp, PTREGS_OFF, %o0
brnz,pn %o0, 3f
mov -ENOSYS, %o0
+
+ /* Syscall tracing can modify the registers. */
+ ldx [%sp + PTREGS_OFF + PT_V9_G1], %g1
+ sethi %hi(sys_call_table64), %l7
+ ldx [%sp + PTREGS_OFF + PT_V9_I0], %i0
+ or %l7, %lo(sys_call_table64), %l7
+ ldx [%sp + PTREGS_OFF + PT_V9_I1], %i1
+ ldx [%sp + PTREGS_OFF + PT_V9_I2], %i2
+ ldx [%sp + PTREGS_OFF + PT_V9_I3], %i3
+ ldx [%sp + PTREGS_OFF + PT_V9_I4], %i4
+ ldx [%sp + PTREGS_OFF + PT_V9_I5], %i5
+
+ cmp %g1, NR_syscalls
+ bgeu,pn %xcc, 3f
+ mov -ENOSYS, %o0
+
+ sll %g1, 2, %l4
mov %i0, %o0
+ lduw [%l7 + %l4], %l7
mov %i1, %o1
mov %i2, %o2
mov %i3, %o3
diff --git a/kernel/arch/sparc/kernel/tsb.S b/kernel/arch/sparc/kernel/tsb.S
index be98685c1..d568c8207 100644
--- a/kernel/arch/sparc/kernel/tsb.S
+++ b/kernel/arch/sparc/kernel/tsb.S
@@ -29,13 +29,17 @@
*/
tsb_miss_dtlb:
mov TLB_TAG_ACCESS, %g4
+ ldxa [%g4] ASI_DMMU, %g4
+ srlx %g4, PAGE_SHIFT, %g4
ba,pt %xcc, tsb_miss_page_table_walk
- ldxa [%g4] ASI_DMMU, %g4
+ sllx %g4, PAGE_SHIFT, %g4
tsb_miss_itlb:
mov TLB_TAG_ACCESS, %g4
+ ldxa [%g4] ASI_IMMU, %g4
+ srlx %g4, PAGE_SHIFT, %g4
ba,pt %xcc, tsb_miss_page_table_walk
- ldxa [%g4] ASI_IMMU, %g4
+ sllx %g4, PAGE_SHIFT, %g4
/* At this point we have:
* %g1 -- PAGE_SIZE TSB entry address
@@ -284,6 +288,10 @@ tsb_do_dtlb_fault:
nop
.previous
+ /* Clear context ID bits. */
+ srlx %g5, PAGE_SHIFT, %g5
+ sllx %g5, PAGE_SHIFT, %g5
+
be,pt %xcc, sparc64_realfault_common
mov FAULT_CODE_DTLB, %g4
ba,pt %xcc, winfix_trampoline
diff --git a/kernel/arch/sparc/kernel/urtt_fill.S b/kernel/arch/sparc/kernel/urtt_fill.S
new file mode 100644
index 000000000..5604a2b05
--- /dev/null
+++ b/kernel/arch/sparc/kernel/urtt_fill.S
@@ -0,0 +1,98 @@
+#include <asm/thread_info.h>
+#include <asm/trap_block.h>
+#include <asm/spitfire.h>
+#include <asm/ptrace.h>
+#include <asm/head.h>
+
+ .text
+ .align 8
+ .globl user_rtt_fill_fixup_common
+user_rtt_fill_fixup_common:
+ rdpr %cwp, %g1
+ add %g1, 1, %g1
+ wrpr %g1, 0x0, %cwp
+
+ rdpr %wstate, %g2
+ sll %g2, 3, %g2
+ wrpr %g2, 0x0, %wstate
+
+ /* We know %canrestore and %otherwin are both zero. */
+
+ sethi %hi(sparc64_kern_pri_context), %g2
+ ldx [%g2 + %lo(sparc64_kern_pri_context)], %g2
+ mov PRIMARY_CONTEXT, %g1
+
+661: stxa %g2, [%g1] ASI_DMMU
+ .section .sun4v_1insn_patch, "ax"
+ .word 661b
+ stxa %g2, [%g1] ASI_MMU
+ .previous
+
+ sethi %hi(KERNBASE), %g1
+ flush %g1
+
+ mov %g4, %l4
+ mov %g5, %l5
+ brnz,pn %g3, 1f
+ mov %g3, %l3
+
+ or %g4, FAULT_CODE_WINFIXUP, %g4
+ stb %g4, [%g6 + TI_FAULT_CODE]
+ stx %g5, [%g6 + TI_FAULT_ADDR]
+1:
+ mov %g6, %l1
+ wrpr %g0, 0x0, %tl
+
+661: nop
+ .section .sun4v_1insn_patch, "ax"
+ .word 661b
+ SET_GL(0)
+ .previous
+
+ wrpr %g0, RTRAP_PSTATE, %pstate
+
+ mov %l1, %g6
+ ldx [%g6 + TI_TASK], %g4
+ LOAD_PER_CPU_BASE(%g5, %g6, %g1, %g2, %g3)
+
+ brnz,pn %l3, 1f
+ nop
+
+ call do_sparc64_fault
+ add %sp, PTREGS_OFF, %o0
+ ba,pt %xcc, rtrap
+ nop
+
+1: cmp %g3, 2
+ bne,pn %xcc, 2f
+ nop
+
+ sethi %hi(tlb_type), %g1
+ lduw [%g1 + %lo(tlb_type)], %g1
+ cmp %g1, 3
+ bne,pt %icc, 1f
+ add %sp, PTREGS_OFF, %o0
+ mov %l4, %o2
+ call sun4v_do_mna
+ mov %l5, %o1
+ ba,a,pt %xcc, rtrap
+1: mov %l4, %o1
+ mov %l5, %o2
+ call mem_address_unaligned
+ nop
+ ba,a,pt %xcc, rtrap
+
+2: sethi %hi(tlb_type), %g1
+ mov %l4, %o1
+ lduw [%g1 + %lo(tlb_type)], %g1
+ mov %l5, %o2
+ cmp %g1, 3
+ bne,pt %icc, 1f
+ add %sp, PTREGS_OFF, %o0
+ call sun4v_data_access_exception
+ nop
+ ba,a,pt %xcc, rtrap
+
+1: call spitfire_data_access_exception
+ nop
+ ba,a,pt %xcc, rtrap
diff --git a/kernel/arch/sparc/kernel/utrap.S b/kernel/arch/sparc/kernel/utrap.S
index b7f0f3f3a..c731e8023 100644
--- a/kernel/arch/sparc/kernel/utrap.S
+++ b/kernel/arch/sparc/kernel/utrap.S
@@ -11,8 +11,7 @@ utrap_trap: /* %g3=handler,%g4=level */
mov %l4, %o1
call bad_trap
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
invoke_utrap:
sllx %g3, 3, %g3
diff --git a/kernel/arch/sparc/kernel/vmlinux.lds.S b/kernel/arch/sparc/kernel/vmlinux.lds.S
index f1a2f688b..4a41d412d 100644
--- a/kernel/arch/sparc/kernel/vmlinux.lds.S
+++ b/kernel/arch/sparc/kernel/vmlinux.lds.S
@@ -33,6 +33,10 @@ ENTRY(_start)
jiffies = jiffies_64;
#endif
+#ifdef CONFIG_SPARC64
+ASSERT((swapper_tsb == 0x0000000000408000), "Error: sparc64 early assembler too large")
+#endif
+
SECTIONS
{
#ifdef CONFIG_SPARC64
diff --git a/kernel/arch/sparc/kernel/winfixup.S b/kernel/arch/sparc/kernel/winfixup.S
index 1e67ce958..855019a85 100644
--- a/kernel/arch/sparc/kernel/winfixup.S
+++ b/kernel/arch/sparc/kernel/winfixup.S
@@ -32,8 +32,7 @@ fill_fixup:
rd %pc, %g7
call do_sparc64_fault
add %sp, PTREGS_OFF, %o0
- ba,pt %xcc, rtrap
- nop
+ ba,a,pt %xcc, rtrap
/* Be very careful about usage of the trap globals here.
* You cannot touch %g5 as that has the fault information.
diff --git a/kernel/arch/sparc/lib/GENcopy_from_user.S b/kernel/arch/sparc/lib/GENcopy_from_user.S
index b7d0bd6b1..69a439fa2 100644
--- a/kernel/arch/sparc/lib/GENcopy_from_user.S
+++ b/kernel/arch/sparc/lib/GENcopy_from_user.S
@@ -3,11 +3,11 @@
* Copyright (C) 2007 David S. Miller (davem@davemloft.net)
*/
-#define EX_LD(x) \
+#define EX_LD(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one; \
+ .word 98b, y; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/GENcopy_to_user.S b/kernel/arch/sparc/lib/GENcopy_to_user.S
index 780550e1a..9947427ce 100644
--- a/kernel/arch/sparc/lib/GENcopy_to_user.S
+++ b/kernel/arch/sparc/lib/GENcopy_to_user.S
@@ -3,11 +3,11 @@
* Copyright (C) 2007 David S. Miller (davem@davemloft.net)
*/
-#define EX_ST(x) \
+#define EX_ST(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one; \
+ .word 98b, y; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/GENmemcpy.S b/kernel/arch/sparc/lib/GENmemcpy.S
index 89358ee94..059ea24ad 100644
--- a/kernel/arch/sparc/lib/GENmemcpy.S
+++ b/kernel/arch/sparc/lib/GENmemcpy.S
@@ -4,21 +4,18 @@
*/
#ifdef __KERNEL__
+#include <linux/linkage.h>
#define GLOBAL_SPARE %g7
#else
#define GLOBAL_SPARE %g5
#endif
#ifndef EX_LD
-#define EX_LD(x) x
+#define EX_LD(x,y) x
#endif
#ifndef EX_ST
-#define EX_ST(x) x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x) x
+#define EX_ST(x,y) x
#endif
#ifndef LOAD
@@ -45,6 +42,29 @@
.register %g3,#scratch
.text
+
+#ifndef EX_RETVAL
+#define EX_RETVAL(x) x
+ENTRY(GEN_retl_o4_1)
+ add %o4, %o2, %o4
+ retl
+ add %o4, 1, %o0
+ENDPROC(GEN_retl_o4_1)
+ENTRY(GEN_retl_g1_8)
+ add %g1, %o2, %g1
+ retl
+ add %g1, 8, %o0
+ENDPROC(GEN_retl_g1_8)
+ENTRY(GEN_retl_o2_4)
+ retl
+ add %o2, 4, %o0
+ENDPROC(GEN_retl_o2_4)
+ENTRY(GEN_retl_o2_1)
+ retl
+ add %o2, 1, %o0
+ENDPROC(GEN_retl_o2_1)
+#endif
+
.align 64
.globl FUNC_NAME
@@ -73,8 +93,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
sub %g0, %o4, %o4
sub %o2, %o4, %o2
1: subcc %o4, 1, %o4
- EX_LD(LOAD(ldub, %o1, %g1))
- EX_ST(STORE(stb, %g1, %o0))
+ EX_LD(LOAD(ldub, %o1, %g1),GEN_retl_o4_1)
+ EX_ST(STORE(stb, %g1, %o0),GEN_retl_o4_1)
add %o1, 1, %o1
bne,pt %XCC, 1b
add %o0, 1, %o0
@@ -82,8 +102,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
andn %o2, 0x7, %g1
sub %o2, %g1, %o2
1: subcc %g1, 0x8, %g1
- EX_LD(LOAD(ldx, %o1, %g2))
- EX_ST(STORE(stx, %g2, %o0))
+ EX_LD(LOAD(ldx, %o1, %g2),GEN_retl_g1_8)
+ EX_ST(STORE(stx, %g2, %o0),GEN_retl_g1_8)
add %o1, 0x8, %o1
bne,pt %XCC, 1b
add %o0, 0x8, %o0
@@ -100,8 +120,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
1:
subcc %o2, 4, %o2
- EX_LD(LOAD(lduw, %o1, %g1))
- EX_ST(STORE(stw, %g1, %o1 + %o3))
+ EX_LD(LOAD(lduw, %o1, %g1),GEN_retl_o2_4)
+ EX_ST(STORE(stw, %g1, %o1 + %o3),GEN_retl_o2_4)
bgu,pt %XCC, 1b
add %o1, 4, %o1
@@ -111,8 +131,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
.align 32
90:
subcc %o2, 1, %o2
- EX_LD(LOAD(ldub, %o1, %g1))
- EX_ST(STORE(stb, %g1, %o1 + %o3))
+ EX_LD(LOAD(ldub, %o1, %g1),GEN_retl_o2_1)
+ EX_ST(STORE(stb, %g1, %o1 + %o3),GEN_retl_o2_1)
bgu,pt %XCC, 90b
add %o1, 1, %o1
retl
diff --git a/kernel/arch/sparc/lib/Makefile b/kernel/arch/sparc/lib/Makefile
index 3269b0234..4f2384a42 100644
--- a/kernel/arch/sparc/lib/Makefile
+++ b/kernel/arch/sparc/lib/Makefile
@@ -38,7 +38,7 @@ lib-$(CONFIG_SPARC64) += NG4patch.o NG4copy_page.o NG4clear_page.o NG4memset.o
lib-$(CONFIG_SPARC64) += GENmemcpy.o GENcopy_from_user.o GENcopy_to_user.o
lib-$(CONFIG_SPARC64) += GENpatch.o GENpage.o GENbzero.o
-lib-$(CONFIG_SPARC64) += copy_in_user.o user_fixup.o memmove.o
+lib-$(CONFIG_SPARC64) += copy_in_user.o memmove.o
lib-$(CONFIG_SPARC64) += mcount.o ipcsum.o xor.o hweight.o ffs.o
obj-$(CONFIG_SPARC64) += iomap.o
diff --git a/kernel/arch/sparc/lib/NG2copy_from_user.S b/kernel/arch/sparc/lib/NG2copy_from_user.S
index d5242b8c4..b79a6998d 100644
--- a/kernel/arch/sparc/lib/NG2copy_from_user.S
+++ b/kernel/arch/sparc/lib/NG2copy_from_user.S
@@ -3,19 +3,19 @@
* Copyright (C) 2007 David S. Miller (davem@davemloft.net)
*/
-#define EX_LD(x) \
+#define EX_LD(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_asi;\
+ .word 98b, y; \
.text; \
.align 4;
-#define EX_LD_FP(x) \
+#define EX_LD_FP(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_asi_fp;\
+ .word 98b, y##_fp; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/NG2copy_to_user.S b/kernel/arch/sparc/lib/NG2copy_to_user.S
index 4e962d993..dcec55f25 100644
--- a/kernel/arch/sparc/lib/NG2copy_to_user.S
+++ b/kernel/arch/sparc/lib/NG2copy_to_user.S
@@ -3,19 +3,19 @@
* Copyright (C) 2007 David S. Miller (davem@davemloft.net)
*/
-#define EX_ST(x) \
+#define EX_ST(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_asi;\
+ .word 98b, y; \
.text; \
.align 4;
-#define EX_ST_FP(x) \
+#define EX_ST_FP(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_asi_fp;\
+ .word 98b, y##_fp; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/NG2memcpy.S b/kernel/arch/sparc/lib/NG2memcpy.S
index d5f585df2..c629dbd12 100644
--- a/kernel/arch/sparc/lib/NG2memcpy.S
+++ b/kernel/arch/sparc/lib/NG2memcpy.S
@@ -4,6 +4,7 @@
*/
#ifdef __KERNEL__
+#include <linux/linkage.h>
#include <asm/visasm.h>
#include <asm/asi.h>
#define GLOBAL_SPARE %g7
@@ -32,21 +33,17 @@
#endif
#ifndef EX_LD
-#define EX_LD(x) x
+#define EX_LD(x,y) x
#endif
#ifndef EX_LD_FP
-#define EX_LD_FP(x) x
+#define EX_LD_FP(x,y) x
#endif
#ifndef EX_ST
-#define EX_ST(x) x
+#define EX_ST(x,y) x
#endif
#ifndef EX_ST_FP
-#define EX_ST_FP(x) x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x) x
+#define EX_ST_FP(x,y) x
#endif
#ifndef LOAD
@@ -140,45 +137,110 @@
fsrc2 %x6, %f12; \
fsrc2 %x7, %f14;
#define FREG_LOAD_1(base, x0) \
- EX_LD_FP(LOAD(ldd, base + 0x00, %x0))
+ EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1)
#define FREG_LOAD_2(base, x0, x1) \
- EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
- EX_LD_FP(LOAD(ldd, base + 0x08, %x1));
+ EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1);
#define FREG_LOAD_3(base, x0, x1, x2) \
- EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
- EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
- EX_LD_FP(LOAD(ldd, base + 0x10, %x2));
+ EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1);
#define FREG_LOAD_4(base, x0, x1, x2, x3) \
- EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
- EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
- EX_LD_FP(LOAD(ldd, base + 0x10, %x2)); \
- EX_LD_FP(LOAD(ldd, base + 0x18, %x3));
+ EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x18, %x3), NG2_retl_o2_plus_g1);
#define FREG_LOAD_5(base, x0, x1, x2, x3, x4) \
- EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
- EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
- EX_LD_FP(LOAD(ldd, base + 0x10, %x2)); \
- EX_LD_FP(LOAD(ldd, base + 0x18, %x3)); \
- EX_LD_FP(LOAD(ldd, base + 0x20, %x4));
+ EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x18, %x3), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x20, %x4), NG2_retl_o2_plus_g1);
#define FREG_LOAD_6(base, x0, x1, x2, x3, x4, x5) \
- EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
- EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
- EX_LD_FP(LOAD(ldd, base + 0x10, %x2)); \
- EX_LD_FP(LOAD(ldd, base + 0x18, %x3)); \
- EX_LD_FP(LOAD(ldd, base + 0x20, %x4)); \
- EX_LD_FP(LOAD(ldd, base + 0x28, %x5));
+ EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x18, %x3), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x20, %x4), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x28, %x5), NG2_retl_o2_plus_g1);
#define FREG_LOAD_7(base, x0, x1, x2, x3, x4, x5, x6) \
- EX_LD_FP(LOAD(ldd, base + 0x00, %x0)); \
- EX_LD_FP(LOAD(ldd, base + 0x08, %x1)); \
- EX_LD_FP(LOAD(ldd, base + 0x10, %x2)); \
- EX_LD_FP(LOAD(ldd, base + 0x18, %x3)); \
- EX_LD_FP(LOAD(ldd, base + 0x20, %x4)); \
- EX_LD_FP(LOAD(ldd, base + 0x28, %x5)); \
- EX_LD_FP(LOAD(ldd, base + 0x30, %x6));
+ EX_LD_FP(LOAD(ldd, base + 0x00, %x0), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x08, %x1), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x10, %x2), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x18, %x3), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x20, %x4), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x28, %x5), NG2_retl_o2_plus_g1); \
+ EX_LD_FP(LOAD(ldd, base + 0x30, %x6), NG2_retl_o2_plus_g1);
.register %g2,#scratch
.register %g3,#scratch
.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x) x
+__restore_fp:
+ VISExitHalf
+__restore_asi:
+ retl
+ wr %g0, ASI_AIUS, %asi
+ENTRY(NG2_retl_o2)
+ ba,pt %xcc, __restore_asi
+ mov %o2, %o0
+ENDPROC(NG2_retl_o2)
+ENTRY(NG2_retl_o2_plus_1)
+ ba,pt %xcc, __restore_asi
+ add %o2, 1, %o0
+ENDPROC(NG2_retl_o2_plus_1)
+ENTRY(NG2_retl_o2_plus_4)
+ ba,pt %xcc, __restore_asi
+ add %o2, 4, %o0
+ENDPROC(NG2_retl_o2_plus_4)
+ENTRY(NG2_retl_o2_plus_8)
+ ba,pt %xcc, __restore_asi
+ add %o2, 8, %o0
+ENDPROC(NG2_retl_o2_plus_8)
+ENTRY(NG2_retl_o2_plus_o4_plus_1)
+ add %o4, 1, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG2_retl_o2_plus_o4_plus_1)
+ENTRY(NG2_retl_o2_plus_o4_plus_8)
+ add %o4, 8, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG2_retl_o2_plus_o4_plus_8)
+ENTRY(NG2_retl_o2_plus_o4_plus_16)
+ add %o4, 16, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG2_retl_o2_plus_o4_plus_16)
+ENTRY(NG2_retl_o2_plus_g1_fp)
+ ba,pt %xcc, __restore_fp
+ add %o2, %g1, %o0
+ENDPROC(NG2_retl_o2_plus_g1_fp)
+ENTRY(NG2_retl_o2_plus_g1_plus_64_fp)
+ add %g1, 64, %g1
+ ba,pt %xcc, __restore_fp
+ add %o2, %g1, %o0
+ENDPROC(NG2_retl_o2_plus_g1_plus_64_fp)
+ENTRY(NG2_retl_o2_plus_g1_plus_1)
+ add %g1, 1, %g1
+ ba,pt %xcc, __restore_asi
+ add %o2, %g1, %o0
+ENDPROC(NG2_retl_o2_plus_g1_plus_1)
+ENTRY(NG2_retl_o2_and_7_plus_o4)
+ and %o2, 7, %o2
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG2_retl_o2_and_7_plus_o4)
+ENTRY(NG2_retl_o2_and_7_plus_o4_plus_8)
+ and %o2, 7, %o2
+ add %o4, 8, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG2_retl_o2_and_7_plus_o4_plus_8)
+#endif
+
.align 64
.globl FUNC_NAME
@@ -230,8 +292,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
sub %g0, %o4, %o4 ! bytes to align dst
sub %o2, %o4, %o2
1: subcc %o4, 1, %o4
- EX_LD(LOAD(ldub, %o1, %g1))
- EX_ST(STORE(stb, %g1, %o0))
+ EX_LD(LOAD(ldub, %o1, %g1), NG2_retl_o2_plus_o4_plus_1)
+ EX_ST(STORE(stb, %g1, %o0), NG2_retl_o2_plus_o4_plus_1)
add %o1, 1, %o1
bne,pt %XCC, 1b
add %o0, 1, %o0
@@ -281,11 +343,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
nop
/* fall through for 0 < low bits < 8 */
110: sub %o4, 64, %g2
- EX_LD_FP(LOAD_BLK(%g2, %f0))
-1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
- EX_LD_FP(LOAD_BLK(%o4, %f16))
+ EX_LD_FP(LOAD_BLK(%g2, %f0), NG2_retl_o2_plus_g1)
+1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+ EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
FREG_FROB(f0, f2, f4, f6, f8, f10, f12, f14, f16)
- EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+ EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
FREG_MOVE_8(f16, f18, f20, f22, f24, f26, f28, f30)
subcc %g1, 64, %g1
add %o4, 64, %o4
@@ -296,10 +358,10 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
120: sub %o4, 56, %g2
FREG_LOAD_7(%g2, f0, f2, f4, f6, f8, f10, f12)
-1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
- EX_LD_FP(LOAD_BLK(%o4, %f16))
+1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+ EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
FREG_FROB(f0, f2, f4, f6, f8, f10, f12, f16, f18)
- EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+ EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
FREG_MOVE_7(f18, f20, f22, f24, f26, f28, f30)
subcc %g1, 64, %g1
add %o4, 64, %o4
@@ -310,10 +372,10 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
130: sub %o4, 48, %g2
FREG_LOAD_6(%g2, f0, f2, f4, f6, f8, f10)
-1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
- EX_LD_FP(LOAD_BLK(%o4, %f16))
+1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+ EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
FREG_FROB(f0, f2, f4, f6, f8, f10, f16, f18, f20)
- EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+ EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
FREG_MOVE_6(f20, f22, f24, f26, f28, f30)
subcc %g1, 64, %g1
add %o4, 64, %o4
@@ -324,10 +386,10 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
140: sub %o4, 40, %g2
FREG_LOAD_5(%g2, f0, f2, f4, f6, f8)
-1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
- EX_LD_FP(LOAD_BLK(%o4, %f16))
+1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+ EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
FREG_FROB(f0, f2, f4, f6, f8, f16, f18, f20, f22)
- EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+ EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
FREG_MOVE_5(f22, f24, f26, f28, f30)
subcc %g1, 64, %g1
add %o4, 64, %o4
@@ -338,10 +400,10 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
150: sub %o4, 32, %g2
FREG_LOAD_4(%g2, f0, f2, f4, f6)
-1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
- EX_LD_FP(LOAD_BLK(%o4, %f16))
+1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+ EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
FREG_FROB(f0, f2, f4, f6, f16, f18, f20, f22, f24)
- EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+ EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
FREG_MOVE_4(f24, f26, f28, f30)
subcc %g1, 64, %g1
add %o4, 64, %o4
@@ -352,10 +414,10 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
160: sub %o4, 24, %g2
FREG_LOAD_3(%g2, f0, f2, f4)
-1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
- EX_LD_FP(LOAD_BLK(%o4, %f16))
+1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+ EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
FREG_FROB(f0, f2, f4, f16, f18, f20, f22, f24, f26)
- EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+ EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
FREG_MOVE_3(f26, f28, f30)
subcc %g1, 64, %g1
add %o4, 64, %o4
@@ -366,10 +428,10 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
170: sub %o4, 16, %g2
FREG_LOAD_2(%g2, f0, f2)
-1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
- EX_LD_FP(LOAD_BLK(%o4, %f16))
+1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+ EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
FREG_FROB(f0, f2, f16, f18, f20, f22, f24, f26, f28)
- EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+ EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
FREG_MOVE_2(f28, f30)
subcc %g1, 64, %g1
add %o4, 64, %o4
@@ -380,10 +442,10 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
180: sub %o4, 8, %g2
FREG_LOAD_1(%g2, f0)
-1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
- EX_LD_FP(LOAD_BLK(%o4, %f16))
+1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
+ EX_LD_FP(LOAD_BLK(%o4, %f16), NG2_retl_o2_plus_g1)
FREG_FROB(f0, f16, f18, f20, f22, f24, f26, f28, f30)
- EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+ EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1)
FREG_MOVE_1(f30)
subcc %g1, 64, %g1
add %o4, 64, %o4
@@ -393,10 +455,10 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
nop
190:
-1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3))
+1: EX_ST_FP(STORE_INIT(%g0, %o4 + %g3), NG2_retl_o2_plus_g1)
subcc %g1, 64, %g1
- EX_LD_FP(LOAD_BLK(%o4, %f0))
- EX_ST_FP(STORE_BLK(%f0, %o4 + %g3))
+ EX_LD_FP(LOAD_BLK(%o4, %f0), NG2_retl_o2_plus_g1_plus_64)
+ EX_ST_FP(STORE_BLK(%f0, %o4 + %g3), NG2_retl_o2_plus_g1_plus_64)
add %o4, 64, %o4
bne,pt %xcc, 1b
LOAD(prefetch, %o4 + 64, #one_read)
@@ -423,28 +485,28 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
andn %o2, 0xf, %o4
and %o2, 0xf, %o2
1: subcc %o4, 0x10, %o4
- EX_LD(LOAD(ldx, %o1, %o5))
+ EX_LD(LOAD(ldx, %o1, %o5), NG2_retl_o2_plus_o4_plus_16)
add %o1, 0x08, %o1
- EX_LD(LOAD(ldx, %o1, %g1))
+ EX_LD(LOAD(ldx, %o1, %g1), NG2_retl_o2_plus_o4_plus_16)
sub %o1, 0x08, %o1
- EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE))
+ EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_o4_plus_16)
add %o1, 0x8, %o1
- EX_ST(STORE(stx, %g1, %o1 + GLOBAL_SPARE))
+ EX_ST(STORE(stx, %g1, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_o4_plus_8)
bgu,pt %XCC, 1b
add %o1, 0x8, %o1
73: andcc %o2, 0x8, %g0
be,pt %XCC, 1f
nop
sub %o2, 0x8, %o2
- EX_LD(LOAD(ldx, %o1, %o5))
- EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE))
+ EX_LD(LOAD(ldx, %o1, %o5), NG2_retl_o2_plus_8)
+ EX_ST(STORE(stx, %o5, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_8)
add %o1, 0x8, %o1
1: andcc %o2, 0x4, %g0
be,pt %XCC, 1f
nop
sub %o2, 0x4, %o2
- EX_LD(LOAD(lduw, %o1, %o5))
- EX_ST(STORE(stw, %o5, %o1 + GLOBAL_SPARE))
+ EX_LD(LOAD(lduw, %o1, %o5), NG2_retl_o2_plus_4)
+ EX_ST(STORE(stw, %o5, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_4)
add %o1, 0x4, %o1
1: cmp %o2, 0
be,pt %XCC, 85f
@@ -460,8 +522,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
sub %o2, %g1, %o2
1: subcc %g1, 1, %g1
- EX_LD(LOAD(ldub, %o1, %o5))
- EX_ST(STORE(stb, %o5, %o1 + GLOBAL_SPARE))
+ EX_LD(LOAD(ldub, %o1, %o5), NG2_retl_o2_plus_g1_plus_1)
+ EX_ST(STORE(stb, %o5, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_g1_plus_1)
bgu,pt %icc, 1b
add %o1, 1, %o1
@@ -477,16 +539,16 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
8: mov 64, GLOBAL_SPARE
andn %o1, 0x7, %o1
- EX_LD(LOAD(ldx, %o1, %g2))
+ EX_LD(LOAD(ldx, %o1, %g2), NG2_retl_o2)
sub GLOBAL_SPARE, %g1, GLOBAL_SPARE
andn %o2, 0x7, %o4
sllx %g2, %g1, %g2
1: add %o1, 0x8, %o1
- EX_LD(LOAD(ldx, %o1, %g3))
+ EX_LD(LOAD(ldx, %o1, %g3), NG2_retl_o2_and_7_plus_o4)
subcc %o4, 0x8, %o4
srlx %g3, GLOBAL_SPARE, %o5
or %o5, %g2, %o5
- EX_ST(STORE(stx, %o5, %o0))
+ EX_ST(STORE(stx, %o5, %o0), NG2_retl_o2_and_7_plus_o4_plus_8)
add %o0, 0x8, %o0
bgu,pt %icc, 1b
sllx %g3, %g1, %g2
@@ -506,8 +568,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
1:
subcc %o2, 4, %o2
- EX_LD(LOAD(lduw, %o1, %g1))
- EX_ST(STORE(stw, %g1, %o1 + GLOBAL_SPARE))
+ EX_LD(LOAD(lduw, %o1, %g1), NG2_retl_o2_plus_4)
+ EX_ST(STORE(stw, %g1, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_4)
bgu,pt %XCC, 1b
add %o1, 4, %o1
@@ -517,8 +579,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
.align 32
90:
subcc %o2, 1, %o2
- EX_LD(LOAD(ldub, %o1, %g1))
- EX_ST(STORE(stb, %g1, %o1 + GLOBAL_SPARE))
+ EX_LD(LOAD(ldub, %o1, %g1), NG2_retl_o2_plus_1)
+ EX_ST(STORE(stb, %g1, %o1 + GLOBAL_SPARE), NG2_retl_o2_plus_1)
bgu,pt %XCC, 90b
add %o1, 1, %o1
retl
diff --git a/kernel/arch/sparc/lib/NG4copy_from_user.S b/kernel/arch/sparc/lib/NG4copy_from_user.S
index 2e8ee7ad0..16a286c1a 100644
--- a/kernel/arch/sparc/lib/NG4copy_from_user.S
+++ b/kernel/arch/sparc/lib/NG4copy_from_user.S
@@ -3,19 +3,19 @@
* Copyright (C) 2012 David S. Miller (davem@davemloft.net)
*/
-#define EX_LD(x) \
+#define EX_LD(x, y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_asi;\
+ .word 98b, y; \
.text; \
.align 4;
-#define EX_LD_FP(x) \
+#define EX_LD_FP(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_asi_fp;\
+ .word 98b, y##_fp; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/NG4copy_to_user.S b/kernel/arch/sparc/lib/NG4copy_to_user.S
index be0bf4590..6b0276ffc 100644
--- a/kernel/arch/sparc/lib/NG4copy_to_user.S
+++ b/kernel/arch/sparc/lib/NG4copy_to_user.S
@@ -3,19 +3,19 @@
* Copyright (C) 2012 David S. Miller (davem@davemloft.net)
*/
-#define EX_ST(x) \
+#define EX_ST(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_asi;\
+ .word 98b, y; \
.text; \
.align 4;
-#define EX_ST_FP(x) \
+#define EX_ST_FP(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_asi_fp;\
+ .word 98b, y##_fp; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/NG4memcpy.S b/kernel/arch/sparc/lib/NG4memcpy.S
index 8e13ee1f4..75bb93b14 100644
--- a/kernel/arch/sparc/lib/NG4memcpy.S
+++ b/kernel/arch/sparc/lib/NG4memcpy.S
@@ -4,6 +4,7 @@
*/
#ifdef __KERNEL__
+#include <linux/linkage.h>
#include <asm/visasm.h>
#include <asm/asi.h>
#define GLOBAL_SPARE %g7
@@ -46,22 +47,19 @@
#endif
#ifndef EX_LD
-#define EX_LD(x) x
+#define EX_LD(x,y) x
#endif
#ifndef EX_LD_FP
-#define EX_LD_FP(x) x
+#define EX_LD_FP(x,y) x
#endif
#ifndef EX_ST
-#define EX_ST(x) x
+#define EX_ST(x,y) x
#endif
#ifndef EX_ST_FP
-#define EX_ST_FP(x) x
+#define EX_ST_FP(x,y) x
#endif
-#ifndef EX_RETVAL
-#define EX_RETVAL(x) x
-#endif
#ifndef LOAD
#define LOAD(type,addr,dest) type [addr], dest
@@ -94,6 +92,158 @@
.register %g3,#scratch
.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x) x
+__restore_asi_fp:
+ VISExitHalf
+__restore_asi:
+ retl
+ wr %g0, ASI_AIUS, %asi
+
+ENTRY(NG4_retl_o2)
+ ba,pt %xcc, __restore_asi
+ mov %o2, %o0
+ENDPROC(NG4_retl_o2)
+ENTRY(NG4_retl_o2_plus_1)
+ ba,pt %xcc, __restore_asi
+ add %o2, 1, %o0
+ENDPROC(NG4_retl_o2_plus_1)
+ENTRY(NG4_retl_o2_plus_4)
+ ba,pt %xcc, __restore_asi
+ add %o2, 4, %o0
+ENDPROC(NG4_retl_o2_plus_4)
+ENTRY(NG4_retl_o2_plus_o5)
+ ba,pt %xcc, __restore_asi
+ add %o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5)
+ENTRY(NG4_retl_o2_plus_o5_plus_4)
+ add %o5, 4, %o5
+ ba,pt %xcc, __restore_asi
+ add %o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_4)
+ENTRY(NG4_retl_o2_plus_o5_plus_8)
+ add %o5, 8, %o5
+ ba,pt %xcc, __restore_asi
+ add %o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_8)
+ENTRY(NG4_retl_o2_plus_o5_plus_16)
+ add %o5, 16, %o5
+ ba,pt %xcc, __restore_asi
+ add %o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_16)
+ENTRY(NG4_retl_o2_plus_o5_plus_24)
+ add %o5, 24, %o5
+ ba,pt %xcc, __restore_asi
+ add %o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_24)
+ENTRY(NG4_retl_o2_plus_o5_plus_32)
+ add %o5, 32, %o5
+ ba,pt %xcc, __restore_asi
+ add %o2, %o5, %o0
+ENDPROC(NG4_retl_o2_plus_o5_plus_32)
+ENTRY(NG4_retl_o2_plus_g1)
+ ba,pt %xcc, __restore_asi
+ add %o2, %g1, %o0
+ENDPROC(NG4_retl_o2_plus_g1)
+ENTRY(NG4_retl_o2_plus_g1_plus_1)
+ add %g1, 1, %g1
+ ba,pt %xcc, __restore_asi
+ add %o2, %g1, %o0
+ENDPROC(NG4_retl_o2_plus_g1_plus_1)
+ENTRY(NG4_retl_o2_plus_g1_plus_8)
+ add %g1, 8, %g1
+ ba,pt %xcc, __restore_asi
+ add %o2, %g1, %o0
+ENDPROC(NG4_retl_o2_plus_g1_plus_8)
+ENTRY(NG4_retl_o2_plus_o4)
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4)
+ENTRY(NG4_retl_o2_plus_o4_plus_8)
+ add %o4, 8, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_8)
+ENTRY(NG4_retl_o2_plus_o4_plus_16)
+ add %o4, 16, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_16)
+ENTRY(NG4_retl_o2_plus_o4_plus_24)
+ add %o4, 24, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_24)
+ENTRY(NG4_retl_o2_plus_o4_plus_32)
+ add %o4, 32, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_32)
+ENTRY(NG4_retl_o2_plus_o4_plus_40)
+ add %o4, 40, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_40)
+ENTRY(NG4_retl_o2_plus_o4_plus_48)
+ add %o4, 48, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_48)
+ENTRY(NG4_retl_o2_plus_o4_plus_56)
+ add %o4, 56, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_56)
+ENTRY(NG4_retl_o2_plus_o4_plus_64)
+ add %o4, 64, %o4
+ ba,pt %xcc, __restore_asi
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_64)
+ENTRY(NG4_retl_o2_plus_o4_fp)
+ ba,pt %xcc, __restore_asi_fp
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_8_fp)
+ add %o4, 8, %o4
+ ba,pt %xcc, __restore_asi_fp
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_8_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_16_fp)
+ add %o4, 16, %o4
+ ba,pt %xcc, __restore_asi_fp
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_16_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_24_fp)
+ add %o4, 24, %o4
+ ba,pt %xcc, __restore_asi_fp
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_24_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_32_fp)
+ add %o4, 32, %o4
+ ba,pt %xcc, __restore_asi_fp
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_32_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_40_fp)
+ add %o4, 40, %o4
+ ba,pt %xcc, __restore_asi_fp
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_40_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_48_fp)
+ add %o4, 48, %o4
+ ba,pt %xcc, __restore_asi_fp
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_48_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_56_fp)
+ add %o4, 56, %o4
+ ba,pt %xcc, __restore_asi_fp
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_56_fp)
+ENTRY(NG4_retl_o2_plus_o4_plus_64_fp)
+ add %o4, 64, %o4
+ ba,pt %xcc, __restore_asi_fp
+ add %o2, %o4, %o0
+ENDPROC(NG4_retl_o2_plus_o4_plus_64_fp)
+#endif
.align 64
.globl FUNC_NAME
@@ -124,12 +274,13 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
brz,pt %g1, 51f
sub %o2, %g1, %o2
-1: EX_LD(LOAD(ldub, %o1 + 0x00, %g2))
+
+1: EX_LD(LOAD(ldub, %o1 + 0x00, %g2), NG4_retl_o2_plus_g1)
add %o1, 1, %o1
subcc %g1, 1, %g1
add %o0, 1, %o0
bne,pt %icc, 1b
- EX_ST(STORE(stb, %g2, %o0 - 0x01))
+ EX_ST(STORE(stb, %g2, %o0 - 0x01), NG4_retl_o2_plus_g1_plus_1)
51: LOAD(prefetch, %o1 + 0x040, #n_reads_strong)
LOAD(prefetch, %o1 + 0x080, #n_reads_strong)
@@ -154,43 +305,43 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
brz,pt %g1, .Llarge_aligned
sub %o2, %g1, %o2
-1: EX_LD(LOAD(ldx, %o1 + 0x00, %g2))
+1: EX_LD(LOAD(ldx, %o1 + 0x00, %g2), NG4_retl_o2_plus_g1)
add %o1, 8, %o1
subcc %g1, 8, %g1
add %o0, 8, %o0
bne,pt %icc, 1b
- EX_ST(STORE(stx, %g2, %o0 - 0x08))
+ EX_ST(STORE(stx, %g2, %o0 - 0x08), NG4_retl_o2_plus_g1_plus_8)
.Llarge_aligned:
/* len >= 0x80 && src 8-byte aligned && dest 8-byte aligned */
andn %o2, 0x3f, %o4
sub %o2, %o4, %o2
-1: EX_LD(LOAD(ldx, %o1 + 0x00, %g1))
+1: EX_LD(LOAD(ldx, %o1 + 0x00, %g1), NG4_retl_o2_plus_o4)
add %o1, 0x40, %o1
- EX_LD(LOAD(ldx, %o1 - 0x38, %g2))
+ EX_LD(LOAD(ldx, %o1 - 0x38, %g2), NG4_retl_o2_plus_o4)
subcc %o4, 0x40, %o4
- EX_LD(LOAD(ldx, %o1 - 0x30, %g3))
- EX_LD(LOAD(ldx, %o1 - 0x28, GLOBAL_SPARE))
- EX_LD(LOAD(ldx, %o1 - 0x20, %o5))
- EX_ST(STORE_INIT(%g1, %o0))
+ EX_LD(LOAD(ldx, %o1 - 0x30, %g3), NG4_retl_o2_plus_o4_plus_64)
+ EX_LD(LOAD(ldx, %o1 - 0x28, GLOBAL_SPARE), NG4_retl_o2_plus_o4_plus_64)
+ EX_LD(LOAD(ldx, %o1 - 0x20, %o5), NG4_retl_o2_plus_o4_plus_64)
+ EX_ST(STORE_INIT(%g1, %o0), NG4_retl_o2_plus_o4_plus_64)
add %o0, 0x08, %o0
- EX_ST(STORE_INIT(%g2, %o0))
+ EX_ST(STORE_INIT(%g2, %o0), NG4_retl_o2_plus_o4_plus_56)
add %o0, 0x08, %o0
- EX_LD(LOAD(ldx, %o1 - 0x18, %g2))
- EX_ST(STORE_INIT(%g3, %o0))
+ EX_LD(LOAD(ldx, %o1 - 0x18, %g2), NG4_retl_o2_plus_o4_plus_48)
+ EX_ST(STORE_INIT(%g3, %o0), NG4_retl_o2_plus_o4_plus_48)
add %o0, 0x08, %o0
- EX_LD(LOAD(ldx, %o1 - 0x10, %g3))
- EX_ST(STORE_INIT(GLOBAL_SPARE, %o0))
+ EX_LD(LOAD(ldx, %o1 - 0x10, %g3), NG4_retl_o2_plus_o4_plus_40)
+ EX_ST(STORE_INIT(GLOBAL_SPARE, %o0), NG4_retl_o2_plus_o4_plus_40)
add %o0, 0x08, %o0
- EX_LD(LOAD(ldx, %o1 - 0x08, GLOBAL_SPARE))
- EX_ST(STORE_INIT(%o5, %o0))
+ EX_LD(LOAD(ldx, %o1 - 0x08, GLOBAL_SPARE), NG4_retl_o2_plus_o4_plus_32)
+ EX_ST(STORE_INIT(%o5, %o0), NG4_retl_o2_plus_o4_plus_32)
add %o0, 0x08, %o0
- EX_ST(STORE_INIT(%g2, %o0))
+ EX_ST(STORE_INIT(%g2, %o0), NG4_retl_o2_plus_o4_plus_24)
add %o0, 0x08, %o0
- EX_ST(STORE_INIT(%g3, %o0))
+ EX_ST(STORE_INIT(%g3, %o0), NG4_retl_o2_plus_o4_plus_16)
add %o0, 0x08, %o0
- EX_ST(STORE_INIT(GLOBAL_SPARE, %o0))
+ EX_ST(STORE_INIT(GLOBAL_SPARE, %o0), NG4_retl_o2_plus_o4_plus_8)
add %o0, 0x08, %o0
bne,pt %icc, 1b
LOAD(prefetch, %o1 + 0x200, #n_reads_strong)
@@ -216,17 +367,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
sub %o2, %o4, %o2
alignaddr %o1, %g0, %g1
add %o1, %o4, %o1
- EX_LD_FP(LOAD(ldd, %g1 + 0x00, %f0))
-1: EX_LD_FP(LOAD(ldd, %g1 + 0x08, %f2))
+ EX_LD_FP(LOAD(ldd, %g1 + 0x00, %f0), NG4_retl_o2_plus_o4)
+1: EX_LD_FP(LOAD(ldd, %g1 + 0x08, %f2), NG4_retl_o2_plus_o4)
subcc %o4, 0x40, %o4
- EX_LD_FP(LOAD(ldd, %g1 + 0x10, %f4))
- EX_LD_FP(LOAD(ldd, %g1 + 0x18, %f6))
- EX_LD_FP(LOAD(ldd, %g1 + 0x20, %f8))
- EX_LD_FP(LOAD(ldd, %g1 + 0x28, %f10))
- EX_LD_FP(LOAD(ldd, %g1 + 0x30, %f12))
- EX_LD_FP(LOAD(ldd, %g1 + 0x38, %f14))
+ EX_LD_FP(LOAD(ldd, %g1 + 0x10, %f4), NG4_retl_o2_plus_o4_plus_64)
+ EX_LD_FP(LOAD(ldd, %g1 + 0x18, %f6), NG4_retl_o2_plus_o4_plus_64)
+ EX_LD_FP(LOAD(ldd, %g1 + 0x20, %f8), NG4_retl_o2_plus_o4_plus_64)
+ EX_LD_FP(LOAD(ldd, %g1 + 0x28, %f10), NG4_retl_o2_plus_o4_plus_64)
+ EX_LD_FP(LOAD(ldd, %g1 + 0x30, %f12), NG4_retl_o2_plus_o4_plus_64)
+ EX_LD_FP(LOAD(ldd, %g1 + 0x38, %f14), NG4_retl_o2_plus_o4_plus_64)
faligndata %f0, %f2, %f16
- EX_LD_FP(LOAD(ldd, %g1 + 0x40, %f0))
+ EX_LD_FP(LOAD(ldd, %g1 + 0x40, %f0), NG4_retl_o2_plus_o4_plus_64)
faligndata %f2, %f4, %f18
add %g1, 0x40, %g1
faligndata %f4, %f6, %f20
@@ -235,14 +386,14 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
faligndata %f10, %f12, %f26
faligndata %f12, %f14, %f28
faligndata %f14, %f0, %f30
- EX_ST_FP(STORE(std, %f16, %o0 + 0x00))
- EX_ST_FP(STORE(std, %f18, %o0 + 0x08))
- EX_ST_FP(STORE(std, %f20, %o0 + 0x10))
- EX_ST_FP(STORE(std, %f22, %o0 + 0x18))
- EX_ST_FP(STORE(std, %f24, %o0 + 0x20))
- EX_ST_FP(STORE(std, %f26, %o0 + 0x28))
- EX_ST_FP(STORE(std, %f28, %o0 + 0x30))
- EX_ST_FP(STORE(std, %f30, %o0 + 0x38))
+ EX_ST_FP(STORE(std, %f16, %o0 + 0x00), NG4_retl_o2_plus_o4_plus_64)
+ EX_ST_FP(STORE(std, %f18, %o0 + 0x08), NG4_retl_o2_plus_o4_plus_56)
+ EX_ST_FP(STORE(std, %f20, %o0 + 0x10), NG4_retl_o2_plus_o4_plus_48)
+ EX_ST_FP(STORE(std, %f22, %o0 + 0x18), NG4_retl_o2_plus_o4_plus_40)
+ EX_ST_FP(STORE(std, %f24, %o0 + 0x20), NG4_retl_o2_plus_o4_plus_32)
+ EX_ST_FP(STORE(std, %f26, %o0 + 0x28), NG4_retl_o2_plus_o4_plus_24)
+ EX_ST_FP(STORE(std, %f28, %o0 + 0x30), NG4_retl_o2_plus_o4_plus_16)
+ EX_ST_FP(STORE(std, %f30, %o0 + 0x38), NG4_retl_o2_plus_o4_plus_8)
add %o0, 0x40, %o0
bne,pt %icc, 1b
LOAD(prefetch, %g1 + 0x200, #n_reads_strong)
@@ -270,37 +421,38 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
andncc %o2, 0x20 - 1, %o5
be,pn %icc, 2f
sub %o2, %o5, %o2
-1: EX_LD(LOAD(ldx, %o1 + 0x00, %g1))
- EX_LD(LOAD(ldx, %o1 + 0x08, %g2))
- EX_LD(LOAD(ldx, %o1 + 0x10, GLOBAL_SPARE))
- EX_LD(LOAD(ldx, %o1 + 0x18, %o4))
+1: EX_LD(LOAD(ldx, %o1 + 0x00, %g1), NG4_retl_o2_plus_o5)
+ EX_LD(LOAD(ldx, %o1 + 0x08, %g2), NG4_retl_o2_plus_o5)
+ EX_LD(LOAD(ldx, %o1 + 0x10, GLOBAL_SPARE), NG4_retl_o2_plus_o5)
+ EX_LD(LOAD(ldx, %o1 + 0x18, %o4), NG4_retl_o2_plus_o5)
add %o1, 0x20, %o1
subcc %o5, 0x20, %o5
- EX_ST(STORE(stx, %g1, %o0 + 0x00))
- EX_ST(STORE(stx, %g2, %o0 + 0x08))
- EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x10))
- EX_ST(STORE(stx, %o4, %o0 + 0x18))
+ EX_ST(STORE(stx, %g1, %o0 + 0x00), NG4_retl_o2_plus_o5_plus_32)
+ EX_ST(STORE(stx, %g2, %o0 + 0x08), NG4_retl_o2_plus_o5_plus_24)
+ EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x10), NG4_retl_o2_plus_o5_plus_24)
+ EX_ST(STORE(stx, %o4, %o0 + 0x18), NG4_retl_o2_plus_o5_plus_8)
bne,pt %icc, 1b
add %o0, 0x20, %o0
2: andcc %o2, 0x18, %o5
be,pt %icc, 3f
sub %o2, %o5, %o2
-1: EX_LD(LOAD(ldx, %o1 + 0x00, %g1))
+
+1: EX_LD(LOAD(ldx, %o1 + 0x00, %g1), NG4_retl_o2_plus_o5)
add %o1, 0x08, %o1
add %o0, 0x08, %o0
subcc %o5, 0x08, %o5
bne,pt %icc, 1b
- EX_ST(STORE(stx, %g1, %o0 - 0x08))
+ EX_ST(STORE(stx, %g1, %o0 - 0x08), NG4_retl_o2_plus_o5_plus_8)
3: brz,pt %o2, .Lexit
cmp %o2, 0x04
bl,pn %icc, .Ltiny
nop
- EX_LD(LOAD(lduw, %o1 + 0x00, %g1))
+ EX_LD(LOAD(lduw, %o1 + 0x00, %g1), NG4_retl_o2)
add %o1, 0x04, %o1
add %o0, 0x04, %o0
subcc %o2, 0x04, %o2
bne,pn %icc, .Ltiny
- EX_ST(STORE(stw, %g1, %o0 - 0x04))
+ EX_ST(STORE(stw, %g1, %o0 - 0x04), NG4_retl_o2_plus_4)
ba,a,pt %icc, .Lexit
.Lmedium_unaligned:
/* First get dest 8 byte aligned. */
@@ -309,12 +461,12 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
brz,pt %g1, 2f
sub %o2, %g1, %o2
-1: EX_LD(LOAD(ldub, %o1 + 0x00, %g2))
+1: EX_LD(LOAD(ldub, %o1 + 0x00, %g2), NG4_retl_o2_plus_g1)
add %o1, 1, %o1
subcc %g1, 1, %g1
add %o0, 1, %o0
bne,pt %icc, 1b
- EX_ST(STORE(stb, %g2, %o0 - 0x01))
+ EX_ST(STORE(stb, %g2, %o0 - 0x01), NG4_retl_o2_plus_g1_plus_1)
2:
and %o1, 0x7, %g1
brz,pn %g1, .Lmedium_noprefetch
@@ -322,16 +474,16 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
mov 64, %g2
sub %g2, %g1, %g2
andn %o1, 0x7, %o1
- EX_LD(LOAD(ldx, %o1 + 0x00, %o4))
+ EX_LD(LOAD(ldx, %o1 + 0x00, %o4), NG4_retl_o2)
sllx %o4, %g1, %o4
andn %o2, 0x08 - 1, %o5
sub %o2, %o5, %o2
-1: EX_LD(LOAD(ldx, %o1 + 0x08, %g3))
+1: EX_LD(LOAD(ldx, %o1 + 0x08, %g3), NG4_retl_o2_plus_o5)
add %o1, 0x08, %o1
subcc %o5, 0x08, %o5
srlx %g3, %g2, GLOBAL_SPARE
or GLOBAL_SPARE, %o4, GLOBAL_SPARE
- EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x00))
+ EX_ST(STORE(stx, GLOBAL_SPARE, %o0 + 0x00), NG4_retl_o2_plus_o5_plus_8)
add %o0, 0x08, %o0
bne,pt %icc, 1b
sllx %g3, %g1, %o4
@@ -342,17 +494,17 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
ba,pt %icc, .Lsmall_unaligned
.Ltiny:
- EX_LD(LOAD(ldub, %o1 + 0x00, %g1))
+ EX_LD(LOAD(ldub, %o1 + 0x00, %g1), NG4_retl_o2)
subcc %o2, 1, %o2
be,pn %icc, .Lexit
- EX_ST(STORE(stb, %g1, %o0 + 0x00))
- EX_LD(LOAD(ldub, %o1 + 0x01, %g1))
+ EX_ST(STORE(stb, %g1, %o0 + 0x00), NG4_retl_o2_plus_1)
+ EX_LD(LOAD(ldub, %o1 + 0x01, %g1), NG4_retl_o2)
subcc %o2, 1, %o2
be,pn %icc, .Lexit
- EX_ST(STORE(stb, %g1, %o0 + 0x01))
- EX_LD(LOAD(ldub, %o1 + 0x02, %g1))
+ EX_ST(STORE(stb, %g1, %o0 + 0x01), NG4_retl_o2_plus_1)
+ EX_LD(LOAD(ldub, %o1 + 0x02, %g1), NG4_retl_o2)
ba,pt %icc, .Lexit
- EX_ST(STORE(stb, %g1, %o0 + 0x02))
+ EX_ST(STORE(stb, %g1, %o0 + 0x02), NG4_retl_o2)
.Lsmall:
andcc %g2, 0x3, %g0
@@ -360,22 +512,22 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
andn %o2, 0x4 - 1, %o5
sub %o2, %o5, %o2
1:
- EX_LD(LOAD(lduw, %o1 + 0x00, %g1))
+ EX_LD(LOAD(lduw, %o1 + 0x00, %g1), NG4_retl_o2_plus_o5)
add %o1, 0x04, %o1
subcc %o5, 0x04, %o5
add %o0, 0x04, %o0
bne,pt %icc, 1b
- EX_ST(STORE(stw, %g1, %o0 - 0x04))
+ EX_ST(STORE(stw, %g1, %o0 - 0x04), NG4_retl_o2_plus_o5_plus_4)
brz,pt %o2, .Lexit
nop
ba,a,pt %icc, .Ltiny
.Lsmall_unaligned:
-1: EX_LD(LOAD(ldub, %o1 + 0x00, %g1))
+1: EX_LD(LOAD(ldub, %o1 + 0x00, %g1), NG4_retl_o2)
add %o1, 1, %o1
add %o0, 1, %o0
subcc %o2, 1, %o2
bne,pt %icc, 1b
- EX_ST(STORE(stb, %g1, %o0 - 0x01))
+ EX_ST(STORE(stb, %g1, %o0 - 0x01), NG4_retl_o2_plus_1)
ba,a,pt %icc, .Lexit
.size FUNC_NAME, .-FUNC_NAME
diff --git a/kernel/arch/sparc/lib/NGcopy_from_user.S b/kernel/arch/sparc/lib/NGcopy_from_user.S
index 5d1e4d1ac..9cd42fcbc 100644
--- a/kernel/arch/sparc/lib/NGcopy_from_user.S
+++ b/kernel/arch/sparc/lib/NGcopy_from_user.S
@@ -3,11 +3,11 @@
* Copyright (C) 2006, 2007 David S. Miller (davem@davemloft.net)
*/
-#define EX_LD(x) \
+#define EX_LD(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __ret_one_asi;\
+ .word 98b, y; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/NGcopy_to_user.S b/kernel/arch/sparc/lib/NGcopy_to_user.S
index ff630dcb2..5c358afd4 100644
--- a/kernel/arch/sparc/lib/NGcopy_to_user.S
+++ b/kernel/arch/sparc/lib/NGcopy_to_user.S
@@ -3,11 +3,11 @@
* Copyright (C) 2006, 2007 David S. Miller (davem@davemloft.net)
*/
-#define EX_ST(x) \
+#define EX_ST(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __ret_one_asi;\
+ .word 98b, y; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/NGmemcpy.S b/kernel/arch/sparc/lib/NGmemcpy.S
index 96a14caf6..d88c4ed50 100644
--- a/kernel/arch/sparc/lib/NGmemcpy.S
+++ b/kernel/arch/sparc/lib/NGmemcpy.S
@@ -4,6 +4,7 @@
*/
#ifdef __KERNEL__
+#include <linux/linkage.h>
#include <asm/asi.h>
#include <asm/thread_info.h>
#define GLOBAL_SPARE %g7
@@ -27,15 +28,11 @@
#endif
#ifndef EX_LD
-#define EX_LD(x) x
+#define EX_LD(x,y) x
#endif
#ifndef EX_ST
-#define EX_ST(x) x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x) x
+#define EX_ST(x,y) x
#endif
#ifndef LOAD
@@ -79,6 +76,92 @@
.register %g3,#scratch
.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x) x
+__restore_asi:
+ ret
+ wr %g0, ASI_AIUS, %asi
+ restore
+ENTRY(NG_ret_i2_plus_i4_plus_1)
+ ba,pt %xcc, __restore_asi
+ add %i2, %i5, %i0
+ENDPROC(NG_ret_i2_plus_i4_plus_1)
+ENTRY(NG_ret_i2_plus_g1)
+ ba,pt %xcc, __restore_asi
+ add %i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1)
+ENTRY(NG_ret_i2_plus_g1_minus_8)
+ sub %g1, 8, %g1
+ ba,pt %xcc, __restore_asi
+ add %i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_8)
+ENTRY(NG_ret_i2_plus_g1_minus_16)
+ sub %g1, 16, %g1
+ ba,pt %xcc, __restore_asi
+ add %i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_16)
+ENTRY(NG_ret_i2_plus_g1_minus_24)
+ sub %g1, 24, %g1
+ ba,pt %xcc, __restore_asi
+ add %i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_24)
+ENTRY(NG_ret_i2_plus_g1_minus_32)
+ sub %g1, 32, %g1
+ ba,pt %xcc, __restore_asi
+ add %i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_32)
+ENTRY(NG_ret_i2_plus_g1_minus_40)
+ sub %g1, 40, %g1
+ ba,pt %xcc, __restore_asi
+ add %i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_40)
+ENTRY(NG_ret_i2_plus_g1_minus_48)
+ sub %g1, 48, %g1
+ ba,pt %xcc, __restore_asi
+ add %i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_48)
+ENTRY(NG_ret_i2_plus_g1_minus_56)
+ sub %g1, 56, %g1
+ ba,pt %xcc, __restore_asi
+ add %i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_minus_56)
+ENTRY(NG_ret_i2_plus_i4)
+ ba,pt %xcc, __restore_asi
+ add %i2, %i4, %i0
+ENDPROC(NG_ret_i2_plus_i4)
+ENTRY(NG_ret_i2_plus_i4_minus_8)
+ sub %i4, 8, %i4
+ ba,pt %xcc, __restore_asi
+ add %i2, %i4, %i0
+ENDPROC(NG_ret_i2_plus_i4_minus_8)
+ENTRY(NG_ret_i2_plus_8)
+ ba,pt %xcc, __restore_asi
+ add %i2, 8, %i0
+ENDPROC(NG_ret_i2_plus_8)
+ENTRY(NG_ret_i2_plus_4)
+ ba,pt %xcc, __restore_asi
+ add %i2, 4, %i0
+ENDPROC(NG_ret_i2_plus_4)
+ENTRY(NG_ret_i2_plus_1)
+ ba,pt %xcc, __restore_asi
+ add %i2, 1, %i0
+ENDPROC(NG_ret_i2_plus_1)
+ENTRY(NG_ret_i2_plus_g1_plus_1)
+ add %g1, 1, %g1
+ ba,pt %xcc, __restore_asi
+ add %i2, %g1, %i0
+ENDPROC(NG_ret_i2_plus_g1_plus_1)
+ENTRY(NG_ret_i2)
+ ba,pt %xcc, __restore_asi
+ mov %i2, %i0
+ENDPROC(NG_ret_i2)
+ENTRY(NG_ret_i2_and_7_plus_i4)
+ and %i2, 7, %i2
+ ba,pt %xcc, __restore_asi
+ add %i2, %i4, %i0
+ENDPROC(NG_ret_i2_and_7_plus_i4)
+#endif
+
.align 64
.globl FUNC_NAME
@@ -126,8 +209,8 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
sub %g0, %i4, %i4 ! bytes to align dst
sub %i2, %i4, %i2
1: subcc %i4, 1, %i4
- EX_LD(LOAD(ldub, %i1, %g1))
- EX_ST(STORE(stb, %g1, %o0))
+ EX_LD(LOAD(ldub, %i1, %g1), NG_ret_i2_plus_i4_plus_1)
+ EX_ST(STORE(stb, %g1, %o0), NG_ret_i2_plus_i4_plus_1)
add %i1, 1, %i1
bne,pt %XCC, 1b
add %o0, 1, %o0
@@ -160,7 +243,7 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
and %i4, 0x7, GLOBAL_SPARE
sll GLOBAL_SPARE, 3, GLOBAL_SPARE
mov 64, %i5
- EX_LD(LOAD_TWIN(%i1, %g2, %g3))
+ EX_LD(LOAD_TWIN(%i1, %g2, %g3), NG_ret_i2_plus_g1)
sub %i5, GLOBAL_SPARE, %i5
mov 16, %o4
mov 32, %o5
@@ -178,31 +261,31 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
srlx WORD3, PRE_SHIFT, TMP; \
or WORD2, TMP, WORD2;
-8: EX_LD(LOAD_TWIN(%i1 + %o4, %o2, %o3))
+8: EX_LD(LOAD_TWIN(%i1 + %o4, %o2, %o3), NG_ret_i2_plus_g1)
MIX_THREE_WORDS(%g2, %g3, %o2, %i5, GLOBAL_SPARE, %o1)
LOAD(prefetch, %i1 + %i3, #one_read)
- EX_ST(STORE_INIT(%g2, %o0 + 0x00))
- EX_ST(STORE_INIT(%g3, %o0 + 0x08))
+ EX_ST(STORE_INIT(%g2, %o0 + 0x00), NG_ret_i2_plus_g1)
+ EX_ST(STORE_INIT(%g3, %o0 + 0x08), NG_ret_i2_plus_g1_minus_8)
- EX_LD(LOAD_TWIN(%i1 + %o5, %g2, %g3))
+ EX_LD(LOAD_TWIN(%i1 + %o5, %g2, %g3), NG_ret_i2_plus_g1_minus_16)
MIX_THREE_WORDS(%o2, %o3, %g2, %i5, GLOBAL_SPARE, %o1)
- EX_ST(STORE_INIT(%o2, %o0 + 0x10))
- EX_ST(STORE_INIT(%o3, %o0 + 0x18))
+ EX_ST(STORE_INIT(%o2, %o0 + 0x10), NG_ret_i2_plus_g1_minus_16)
+ EX_ST(STORE_INIT(%o3, %o0 + 0x18), NG_ret_i2_plus_g1_minus_24)
- EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3))
+ EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3), NG_ret_i2_plus_g1_minus_32)
MIX_THREE_WORDS(%g2, %g3, %o2, %i5, GLOBAL_SPARE, %o1)
- EX_ST(STORE_INIT(%g2, %o0 + 0x20))
- EX_ST(STORE_INIT(%g3, %o0 + 0x28))
+ EX_ST(STORE_INIT(%g2, %o0 + 0x20), NG_ret_i2_plus_g1_minus_32)
+ EX_ST(STORE_INIT(%g3, %o0 + 0x28), NG_ret_i2_plus_g1_minus_40)
- EX_LD(LOAD_TWIN(%i1 + %i3, %g2, %g3))
+ EX_LD(LOAD_TWIN(%i1 + %i3, %g2, %g3), NG_ret_i2_plus_g1_minus_48)
add %i1, 64, %i1
MIX_THREE_WORDS(%o2, %o3, %g2, %i5, GLOBAL_SPARE, %o1)
- EX_ST(STORE_INIT(%o2, %o0 + 0x30))
- EX_ST(STORE_INIT(%o3, %o0 + 0x38))
+ EX_ST(STORE_INIT(%o2, %o0 + 0x30), NG_ret_i2_plus_g1_minus_48)
+ EX_ST(STORE_INIT(%o3, %o0 + 0x38), NG_ret_i2_plus_g1_minus_56)
subcc %g1, 64, %g1
bne,pt %XCC, 8b
@@ -211,31 +294,31 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
ba,pt %XCC, 60f
add %i1, %i4, %i1
-9: EX_LD(LOAD_TWIN(%i1 + %o4, %o2, %o3))
+9: EX_LD(LOAD_TWIN(%i1 + %o4, %o2, %o3), NG_ret_i2_plus_g1)
MIX_THREE_WORDS(%g3, %o2, %o3, %i5, GLOBAL_SPARE, %o1)
LOAD(prefetch, %i1 + %i3, #one_read)
- EX_ST(STORE_INIT(%g3, %o0 + 0x00))
- EX_ST(STORE_INIT(%o2, %o0 + 0x08))
+ EX_ST(STORE_INIT(%g3, %o0 + 0x00), NG_ret_i2_plus_g1)
+ EX_ST(STORE_INIT(%o2, %o0 + 0x08), NG_ret_i2_plus_g1_minus_8)
- EX_LD(LOAD_TWIN(%i1 + %o5, %g2, %g3))
+ EX_LD(LOAD_TWIN(%i1 + %o5, %g2, %g3), NG_ret_i2_plus_g1_minus_16)
MIX_THREE_WORDS(%o3, %g2, %g3, %i5, GLOBAL_SPARE, %o1)
- EX_ST(STORE_INIT(%o3, %o0 + 0x10))
- EX_ST(STORE_INIT(%g2, %o0 + 0x18))
+ EX_ST(STORE_INIT(%o3, %o0 + 0x10), NG_ret_i2_plus_g1_minus_16)
+ EX_ST(STORE_INIT(%g2, %o0 + 0x18), NG_ret_i2_plus_g1_minus_24)
- EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3))
+ EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3), NG_ret_i2_plus_g1_minus_32)
MIX_THREE_WORDS(%g3, %o2, %o3, %i5, GLOBAL_SPARE, %o1)
- EX_ST(STORE_INIT(%g3, %o0 + 0x20))
- EX_ST(STORE_INIT(%o2, %o0 + 0x28))
+ EX_ST(STORE_INIT(%g3, %o0 + 0x20), NG_ret_i2_plus_g1_minus_32)
+ EX_ST(STORE_INIT(%o2, %o0 + 0x28), NG_ret_i2_plus_g1_minus_40)
- EX_LD(LOAD_TWIN(%i1 + %i3, %g2, %g3))
+ EX_LD(LOAD_TWIN(%i1 + %i3, %g2, %g3), NG_ret_i2_plus_g1_minus_48)
add %i1, 64, %i1
MIX_THREE_WORDS(%o3, %g2, %g3, %i5, GLOBAL_SPARE, %o1)
- EX_ST(STORE_INIT(%o3, %o0 + 0x30))
- EX_ST(STORE_INIT(%g2, %o0 + 0x38))
+ EX_ST(STORE_INIT(%o3, %o0 + 0x30), NG_ret_i2_plus_g1_minus_48)
+ EX_ST(STORE_INIT(%g2, %o0 + 0x38), NG_ret_i2_plus_g1_minus_56)
subcc %g1, 64, %g1
bne,pt %XCC, 9b
@@ -249,25 +332,25 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
* one twin load ahead, then add 8 back into source when
* we finish the loop.
*/
- EX_LD(LOAD_TWIN(%i1, %o4, %o5))
+ EX_LD(LOAD_TWIN(%i1, %o4, %o5), NG_ret_i2_plus_g1)
mov 16, %o7
mov 32, %g2
mov 48, %g3
mov 64, %o1
-1: EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3))
+1: EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3), NG_ret_i2_plus_g1)
LOAD(prefetch, %i1 + %o1, #one_read)
- EX_ST(STORE_INIT(%o5, %o0 + 0x00)) ! initializes cache line
- EX_ST(STORE_INIT(%o2, %o0 + 0x08))
- EX_LD(LOAD_TWIN(%i1 + %g2, %o4, %o5))
- EX_ST(STORE_INIT(%o3, %o0 + 0x10))
- EX_ST(STORE_INIT(%o4, %o0 + 0x18))
- EX_LD(LOAD_TWIN(%i1 + %g3, %o2, %o3))
- EX_ST(STORE_INIT(%o5, %o0 + 0x20))
- EX_ST(STORE_INIT(%o2, %o0 + 0x28))
- EX_LD(LOAD_TWIN(%i1 + %o1, %o4, %o5))
+ EX_ST(STORE_INIT(%o5, %o0 + 0x00), NG_ret_i2_plus_g1) ! initializes cache line
+ EX_ST(STORE_INIT(%o2, %o0 + 0x08), NG_ret_i2_plus_g1_minus_8)
+ EX_LD(LOAD_TWIN(%i1 + %g2, %o4, %o5), NG_ret_i2_plus_g1_minus_16)
+ EX_ST(STORE_INIT(%o3, %o0 + 0x10), NG_ret_i2_plus_g1_minus_16)
+ EX_ST(STORE_INIT(%o4, %o0 + 0x18), NG_ret_i2_plus_g1_minus_24)
+ EX_LD(LOAD_TWIN(%i1 + %g3, %o2, %o3), NG_ret_i2_plus_g1_minus_32)
+ EX_ST(STORE_INIT(%o5, %o0 + 0x20), NG_ret_i2_plus_g1_minus_32)
+ EX_ST(STORE_INIT(%o2, %o0 + 0x28), NG_ret_i2_plus_g1_minus_40)
+ EX_LD(LOAD_TWIN(%i1 + %o1, %o4, %o5), NG_ret_i2_plus_g1_minus_48)
add %i1, 64, %i1
- EX_ST(STORE_INIT(%o3, %o0 + 0x30))
- EX_ST(STORE_INIT(%o4, %o0 + 0x38))
+ EX_ST(STORE_INIT(%o3, %o0 + 0x30), NG_ret_i2_plus_g1_minus_48)
+ EX_ST(STORE_INIT(%o4, %o0 + 0x38), NG_ret_i2_plus_g1_minus_56)
subcc %g1, 64, %g1
bne,pt %XCC, 1b
add %o0, 64, %o0
@@ -282,20 +365,20 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
mov 32, %g2
mov 48, %g3
mov 64, %o1
-1: EX_LD(LOAD_TWIN(%i1 + %g0, %o4, %o5))
- EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3))
+1: EX_LD(LOAD_TWIN(%i1 + %g0, %o4, %o5), NG_ret_i2_plus_g1)
+ EX_LD(LOAD_TWIN(%i1 + %o7, %o2, %o3), NG_ret_i2_plus_g1)
LOAD(prefetch, %i1 + %o1, #one_read)
- EX_ST(STORE_INIT(%o4, %o0 + 0x00)) ! initializes cache line
- EX_ST(STORE_INIT(%o5, %o0 + 0x08))
- EX_LD(LOAD_TWIN(%i1 + %g2, %o4, %o5))
- EX_ST(STORE_INIT(%o2, %o0 + 0x10))
- EX_ST(STORE_INIT(%o3, %o0 + 0x18))
- EX_LD(LOAD_TWIN(%i1 + %g3, %o2, %o3))
+ EX_ST(STORE_INIT(%o4, %o0 + 0x00), NG_ret_i2_plus_g1) ! initializes cache line
+ EX_ST(STORE_INIT(%o5, %o0 + 0x08), NG_ret_i2_plus_g1_minus_8)
+ EX_LD(LOAD_TWIN(%i1 + %g2, %o4, %o5), NG_ret_i2_plus_g1_minus_16)
+ EX_ST(STORE_INIT(%o2, %o0 + 0x10), NG_ret_i2_plus_g1_minus_16)
+ EX_ST(STORE_INIT(%o3, %o0 + 0x18), NG_ret_i2_plus_g1_minus_24)
+ EX_LD(LOAD_TWIN(%i1 + %g3, %o2, %o3), NG_ret_i2_plus_g1_minus_32)
add %i1, 64, %i1
- EX_ST(STORE_INIT(%o4, %o0 + 0x20))
- EX_ST(STORE_INIT(%o5, %o0 + 0x28))
- EX_ST(STORE_INIT(%o2, %o0 + 0x30))
- EX_ST(STORE_INIT(%o3, %o0 + 0x38))
+ EX_ST(STORE_INIT(%o4, %o0 + 0x20), NG_ret_i2_plus_g1_minus_32)
+ EX_ST(STORE_INIT(%o5, %o0 + 0x28), NG_ret_i2_plus_g1_minus_40)
+ EX_ST(STORE_INIT(%o2, %o0 + 0x30), NG_ret_i2_plus_g1_minus_48)
+ EX_ST(STORE_INIT(%o3, %o0 + 0x38), NG_ret_i2_plus_g1_minus_56)
subcc %g1, 64, %g1
bne,pt %XCC, 1b
add %o0, 64, %o0
@@ -321,28 +404,28 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
andn %i2, 0xf, %i4
and %i2, 0xf, %i2
1: subcc %i4, 0x10, %i4
- EX_LD(LOAD(ldx, %i1, %o4))
+ EX_LD(LOAD(ldx, %i1, %o4), NG_ret_i2_plus_i4)
add %i1, 0x08, %i1
- EX_LD(LOAD(ldx, %i1, %g1))
+ EX_LD(LOAD(ldx, %i1, %g1), NG_ret_i2_plus_i4)
sub %i1, 0x08, %i1
- EX_ST(STORE(stx, %o4, %i1 + %i3))
+ EX_ST(STORE(stx, %o4, %i1 + %i3), NG_ret_i2_plus_i4)
add %i1, 0x8, %i1
- EX_ST(STORE(stx, %g1, %i1 + %i3))
+ EX_ST(STORE(stx, %g1, %i1 + %i3), NG_ret_i2_plus_i4_minus_8)
bgu,pt %XCC, 1b
add %i1, 0x8, %i1
73: andcc %i2, 0x8, %g0
be,pt %XCC, 1f
nop
sub %i2, 0x8, %i2
- EX_LD(LOAD(ldx, %i1, %o4))
- EX_ST(STORE(stx, %o4, %i1 + %i3))
+ EX_LD(LOAD(ldx, %i1, %o4), NG_ret_i2_plus_8)
+ EX_ST(STORE(stx, %o4, %i1 + %i3), NG_ret_i2_plus_8)
add %i1, 0x8, %i1
1: andcc %i2, 0x4, %g0
be,pt %XCC, 1f
nop
sub %i2, 0x4, %i2
- EX_LD(LOAD(lduw, %i1, %i5))
- EX_ST(STORE(stw, %i5, %i1 + %i3))
+ EX_LD(LOAD(lduw, %i1, %i5), NG_ret_i2_plus_4)
+ EX_ST(STORE(stw, %i5, %i1 + %i3), NG_ret_i2_plus_4)
add %i1, 0x4, %i1
1: cmp %i2, 0
be,pt %XCC, 85f
@@ -358,8 +441,8 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
sub %i2, %g1, %i2
1: subcc %g1, 1, %g1
- EX_LD(LOAD(ldub, %i1, %i5))
- EX_ST(STORE(stb, %i5, %i1 + %i3))
+ EX_LD(LOAD(ldub, %i1, %i5), NG_ret_i2_plus_g1_plus_1)
+ EX_ST(STORE(stb, %i5, %i1 + %i3), NG_ret_i2_plus_g1_plus_1)
bgu,pt %icc, 1b
add %i1, 1, %i1
@@ -375,16 +458,16 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
8: mov 64, %i3
andn %i1, 0x7, %i1
- EX_LD(LOAD(ldx, %i1, %g2))
+ EX_LD(LOAD(ldx, %i1, %g2), NG_ret_i2)
sub %i3, %g1, %i3
andn %i2, 0x7, %i4
sllx %g2, %g1, %g2
1: add %i1, 0x8, %i1
- EX_LD(LOAD(ldx, %i1, %g3))
+ EX_LD(LOAD(ldx, %i1, %g3), NG_ret_i2_and_7_plus_i4)
subcc %i4, 0x8, %i4
srlx %g3, %i3, %i5
or %i5, %g2, %i5
- EX_ST(STORE(stx, %i5, %o0))
+ EX_ST(STORE(stx, %i5, %o0), NG_ret_i2_and_7_plus_i4)
add %o0, 0x8, %o0
bgu,pt %icc, 1b
sllx %g3, %g1, %g2
@@ -404,8 +487,8 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
1:
subcc %i2, 4, %i2
- EX_LD(LOAD(lduw, %i1, %g1))
- EX_ST(STORE(stw, %g1, %i1 + %i3))
+ EX_LD(LOAD(lduw, %i1, %g1), NG_ret_i2_plus_4)
+ EX_ST(STORE(stw, %g1, %i1 + %i3), NG_ret_i2_plus_4)
bgu,pt %XCC, 1b
add %i1, 4, %i1
@@ -415,8 +498,8 @@ FUNC_NAME: /* %i0=dst, %i1=src, %i2=len */
.align 32
90:
subcc %i2, 1, %i2
- EX_LD(LOAD(ldub, %i1, %g1))
- EX_ST(STORE(stb, %g1, %i1 + %i3))
+ EX_LD(LOAD(ldub, %i1, %g1), NG_ret_i2_plus_1)
+ EX_ST(STORE(stb, %g1, %i1 + %i3), NG_ret_i2_plus_1)
bgu,pt %XCC, 90b
add %i1, 1, %i1
ret
diff --git a/kernel/arch/sparc/lib/U1copy_from_user.S b/kernel/arch/sparc/lib/U1copy_from_user.S
index ecc5692fa..bb6ff7322 100644
--- a/kernel/arch/sparc/lib/U1copy_from_user.S
+++ b/kernel/arch/sparc/lib/U1copy_from_user.S
@@ -3,19 +3,19 @@
* Copyright (C) 1999, 2000, 2004 David S. Miller (davem@redhat.com)
*/
-#define EX_LD(x) \
+#define EX_LD(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one; \
+ .word 98b, y; \
.text; \
.align 4;
-#define EX_LD_FP(x) \
+#define EX_LD_FP(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_fp;\
+ .word 98b, y; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/U1copy_to_user.S b/kernel/arch/sparc/lib/U1copy_to_user.S
index 9eea392e4..ed92ce739 100644
--- a/kernel/arch/sparc/lib/U1copy_to_user.S
+++ b/kernel/arch/sparc/lib/U1copy_to_user.S
@@ -3,19 +3,19 @@
* Copyright (C) 1999, 2000, 2004 David S. Miller (davem@redhat.com)
*/
-#define EX_ST(x) \
+#define EX_ST(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one; \
+ .word 98b, y; \
.text; \
.align 4;
-#define EX_ST_FP(x) \
+#define EX_ST_FP(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_fp;\
+ .word 98b, y; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/U1memcpy.S b/kernel/arch/sparc/lib/U1memcpy.S
index 3e6209ebb..f30d2ab2c 100644
--- a/kernel/arch/sparc/lib/U1memcpy.S
+++ b/kernel/arch/sparc/lib/U1memcpy.S
@@ -5,6 +5,7 @@
*/
#ifdef __KERNEL__
+#include <linux/linkage.h>
#include <asm/visasm.h>
#include <asm/asi.h>
#define GLOBAL_SPARE g7
@@ -23,21 +24,17 @@
#endif
#ifndef EX_LD
-#define EX_LD(x) x
+#define EX_LD(x,y) x
#endif
#ifndef EX_LD_FP
-#define EX_LD_FP(x) x
+#define EX_LD_FP(x,y) x
#endif
#ifndef EX_ST
-#define EX_ST(x) x
+#define EX_ST(x,y) x
#endif
#ifndef EX_ST_FP
-#define EX_ST_FP(x) x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x) x
+#define EX_ST_FP(x,y) x
#endif
#ifndef LOAD
@@ -78,53 +75,169 @@
faligndata %f7, %f8, %f60; \
faligndata %f8, %f9, %f62;
-#define MAIN_LOOP_CHUNK(src, dest, fdest, fsrc, len, jmptgt) \
- EX_LD_FP(LOAD_BLK(%src, %fdest)); \
- EX_ST_FP(STORE_BLK(%fsrc, %dest)); \
- add %src, 0x40, %src; \
- subcc %len, 0x40, %len; \
- be,pn %xcc, jmptgt; \
- add %dest, 0x40, %dest; \
-
-#define LOOP_CHUNK1(src, dest, len, branch_dest) \
- MAIN_LOOP_CHUNK(src, dest, f0, f48, len, branch_dest)
-#define LOOP_CHUNK2(src, dest, len, branch_dest) \
- MAIN_LOOP_CHUNK(src, dest, f16, f48, len, branch_dest)
-#define LOOP_CHUNK3(src, dest, len, branch_dest) \
- MAIN_LOOP_CHUNK(src, dest, f32, f48, len, branch_dest)
+#define MAIN_LOOP_CHUNK(src, dest, fdest, fsrc, jmptgt) \
+ EX_LD_FP(LOAD_BLK(%src, %fdest), U1_gs_80_fp); \
+ EX_ST_FP(STORE_BLK(%fsrc, %dest), U1_gs_80_fp); \
+ add %src, 0x40, %src; \
+ subcc %GLOBAL_SPARE, 0x40, %GLOBAL_SPARE; \
+ be,pn %xcc, jmptgt; \
+ add %dest, 0x40, %dest; \
+
+#define LOOP_CHUNK1(src, dest, branch_dest) \
+ MAIN_LOOP_CHUNK(src, dest, f0, f48, branch_dest)
+#define LOOP_CHUNK2(src, dest, branch_dest) \
+ MAIN_LOOP_CHUNK(src, dest, f16, f48, branch_dest)
+#define LOOP_CHUNK3(src, dest, branch_dest) \
+ MAIN_LOOP_CHUNK(src, dest, f32, f48, branch_dest)
#define DO_SYNC membar #Sync;
#define STORE_SYNC(dest, fsrc) \
- EX_ST_FP(STORE_BLK(%fsrc, %dest)); \
+ EX_ST_FP(STORE_BLK(%fsrc, %dest), U1_gs_80_fp); \
add %dest, 0x40, %dest; \
DO_SYNC
#define STORE_JUMP(dest, fsrc, target) \
- EX_ST_FP(STORE_BLK(%fsrc, %dest)); \
+ EX_ST_FP(STORE_BLK(%fsrc, %dest), U1_gs_40_fp); \
add %dest, 0x40, %dest; \
ba,pt %xcc, target; \
nop;
-#define FINISH_VISCHUNK(dest, f0, f1, left) \
- subcc %left, 8, %left;\
- bl,pn %xcc, 95f; \
- faligndata %f0, %f1, %f48; \
- EX_ST_FP(STORE(std, %f48, %dest)); \
+#define FINISH_VISCHUNK(dest, f0, f1) \
+ subcc %g3, 8, %g3; \
+ bl,pn %xcc, 95f; \
+ faligndata %f0, %f1, %f48; \
+ EX_ST_FP(STORE(std, %f48, %dest), U1_g3_8_fp); \
add %dest, 8, %dest;
-#define UNEVEN_VISCHUNK_LAST(dest, f0, f1, left) \
- subcc %left, 8, %left; \
- bl,pn %xcc, 95f; \
+#define UNEVEN_VISCHUNK_LAST(dest, f0, f1) \
+ subcc %g3, 8, %g3; \
+ bl,pn %xcc, 95f; \
fsrc2 %f0, %f1;
-#define UNEVEN_VISCHUNK(dest, f0, f1, left) \
- UNEVEN_VISCHUNK_LAST(dest, f0, f1, left) \
+#define UNEVEN_VISCHUNK(dest, f0, f1) \
+ UNEVEN_VISCHUNK_LAST(dest, f0, f1) \
ba,a,pt %xcc, 93f;
.register %g2,#scratch
.register %g3,#scratch
.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x) x
+ENTRY(U1_g1_1_fp)
+ VISExitHalf
+ add %g1, 1, %g1
+ add %g1, %g2, %g1
+ retl
+ add %g1, %o2, %o0
+ENDPROC(U1_g1_1_fp)
+ENTRY(U1_g2_0_fp)
+ VISExitHalf
+ retl
+ add %g2, %o2, %o0
+ENDPROC(U1_g2_0_fp)
+ENTRY(U1_g2_8_fp)
+ VISExitHalf
+ add %g2, 8, %g2
+ retl
+ add %g2, %o2, %o0
+ENDPROC(U1_g2_8_fp)
+ENTRY(U1_gs_0_fp)
+ VISExitHalf
+ add %GLOBAL_SPARE, %g3, %o0
+ retl
+ add %o0, %o2, %o0
+ENDPROC(U1_gs_0_fp)
+ENTRY(U1_gs_80_fp)
+ VISExitHalf
+ add %GLOBAL_SPARE, 0x80, %GLOBAL_SPARE
+ add %GLOBAL_SPARE, %g3, %o0
+ retl
+ add %o0, %o2, %o0
+ENDPROC(U1_gs_80_fp)
+ENTRY(U1_gs_40_fp)
+ VISExitHalf
+ add %GLOBAL_SPARE, 0x40, %GLOBAL_SPARE
+ add %GLOBAL_SPARE, %g3, %o0
+ retl
+ add %o0, %o2, %o0
+ENDPROC(U1_gs_40_fp)
+ENTRY(U1_g3_0_fp)
+ VISExitHalf
+ retl
+ add %g3, %o2, %o0
+ENDPROC(U1_g3_0_fp)
+ENTRY(U1_g3_8_fp)
+ VISExitHalf
+ add %g3, 8, %g3
+ retl
+ add %g3, %o2, %o0
+ENDPROC(U1_g3_8_fp)
+ENTRY(U1_o2_0_fp)
+ VISExitHalf
+ retl
+ mov %o2, %o0
+ENDPROC(U1_o2_0_fp)
+ENTRY(U1_o2_1_fp)
+ VISExitHalf
+ retl
+ add %o2, 1, %o0
+ENDPROC(U1_o2_1_fp)
+ENTRY(U1_gs_0)
+ VISExitHalf
+ retl
+ add %GLOBAL_SPARE, %o2, %o0
+ENDPROC(U1_gs_0)
+ENTRY(U1_gs_8)
+ VISExitHalf
+ add %GLOBAL_SPARE, %o2, %GLOBAL_SPARE
+ retl
+ add %GLOBAL_SPARE, 0x8, %o0
+ENDPROC(U1_gs_8)
+ENTRY(U1_gs_10)
+ VISExitHalf
+ add %GLOBAL_SPARE, %o2, %GLOBAL_SPARE
+ retl
+ add %GLOBAL_SPARE, 0x10, %o0
+ENDPROC(U1_gs_10)
+ENTRY(U1_o2_0)
+ retl
+ mov %o2, %o0
+ENDPROC(U1_o2_0)
+ENTRY(U1_o2_8)
+ retl
+ add %o2, 8, %o0
+ENDPROC(U1_o2_8)
+ENTRY(U1_o2_4)
+ retl
+ add %o2, 4, %o0
+ENDPROC(U1_o2_4)
+ENTRY(U1_o2_1)
+ retl
+ add %o2, 1, %o0
+ENDPROC(U1_o2_1)
+ENTRY(U1_g1_0)
+ retl
+ add %g1, %o2, %o0
+ENDPROC(U1_g1_0)
+ENTRY(U1_g1_1)
+ add %g1, 1, %g1
+ retl
+ add %g1, %o2, %o0
+ENDPROC(U1_g1_1)
+ENTRY(U1_gs_0_o2_adj)
+ and %o2, 7, %o2
+ retl
+ add %GLOBAL_SPARE, %o2, %o0
+ENDPROC(U1_gs_0_o2_adj)
+ENTRY(U1_gs_8_o2_adj)
+ and %o2, 7, %o2
+ add %GLOBAL_SPARE, 8, %GLOBAL_SPARE
+ retl
+ add %GLOBAL_SPARE, %o2, %o0
+ENDPROC(U1_gs_8_o2_adj)
+#endif
+
.align 64
.globl FUNC_NAME
@@ -166,8 +279,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
and %g2, 0x38, %g2
1: subcc %g1, 0x1, %g1
- EX_LD_FP(LOAD(ldub, %o1 + 0x00, %o3))
- EX_ST_FP(STORE(stb, %o3, %o1 + %GLOBAL_SPARE))
+ EX_LD_FP(LOAD(ldub, %o1 + 0x00, %o3), U1_g1_1_fp)
+ EX_ST_FP(STORE(stb, %o3, %o1 + %GLOBAL_SPARE), U1_g1_1_fp)
bgu,pt %XCC, 1b
add %o1, 0x1, %o1
@@ -178,20 +291,20 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
be,pt %icc, 3f
alignaddr %o1, %g0, %o1
- EX_LD_FP(LOAD(ldd, %o1, %f4))
-1: EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f6))
+ EX_LD_FP(LOAD(ldd, %o1, %f4), U1_g2_0_fp)
+1: EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f6), U1_g2_0_fp)
add %o1, 0x8, %o1
subcc %g2, 0x8, %g2
faligndata %f4, %f6, %f0
- EX_ST_FP(STORE(std, %f0, %o0))
+ EX_ST_FP(STORE(std, %f0, %o0), U1_g2_8_fp)
be,pn %icc, 3f
add %o0, 0x8, %o0
- EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f4))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f4), U1_g2_0_fp)
add %o1, 0x8, %o1
subcc %g2, 0x8, %g2
faligndata %f6, %f4, %f0
- EX_ST_FP(STORE(std, %f0, %o0))
+ EX_ST_FP(STORE(std, %f0, %o0), U1_g2_8_fp)
bne,pt %icc, 1b
add %o0, 0x8, %o0
@@ -214,13 +327,13 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
add %g1, %GLOBAL_SPARE, %g1
subcc %o2, %g3, %o2
- EX_LD_FP(LOAD_BLK(%o1, %f0))
+ EX_LD_FP(LOAD_BLK(%o1, %f0), U1_gs_0_fp)
add %o1, 0x40, %o1
add %g1, %g3, %g1
- EX_LD_FP(LOAD_BLK(%o1, %f16))
+ EX_LD_FP(LOAD_BLK(%o1, %f16), U1_gs_0_fp)
add %o1, 0x40, %o1
sub %GLOBAL_SPARE, 0x80, %GLOBAL_SPARE
- EX_LD_FP(LOAD_BLK(%o1, %f32))
+ EX_LD_FP(LOAD_BLK(%o1, %f32), U1_gs_80_fp)
add %o1, 0x40, %o1
/* There are 8 instances of the unrolled loop,
@@ -240,11 +353,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
.align 64
1: FREG_FROB(f0, f2, f4, f6, f8, f10,f12,f14,f16)
- LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+ LOOP_CHUNK1(o1, o0, 1f)
FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
- LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+ LOOP_CHUNK2(o1, o0, 2f)
FREG_FROB(f32,f34,f36,f38,f40,f42,f44,f46,f0)
- LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+ LOOP_CHUNK3(o1, o0, 3f)
ba,pt %xcc, 1b+4
faligndata %f0, %f2, %f48
1: FREG_FROB(f16,f18,f20,f22,f24,f26,f28,f30,f32)
@@ -261,11 +374,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
STORE_JUMP(o0, f48, 56f)
1: FREG_FROB(f2, f4, f6, f8, f10,f12,f14,f16,f18)
- LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+ LOOP_CHUNK1(o1, o0, 1f)
FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
- LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+ LOOP_CHUNK2(o1, o0, 2f)
FREG_FROB(f34,f36,f38,f40,f42,f44,f46,f0, f2)
- LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+ LOOP_CHUNK3(o1, o0, 3f)
ba,pt %xcc, 1b+4
faligndata %f2, %f4, %f48
1: FREG_FROB(f18,f20,f22,f24,f26,f28,f30,f32,f34)
@@ -282,11 +395,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
STORE_JUMP(o0, f48, 57f)
1: FREG_FROB(f4, f6, f8, f10,f12,f14,f16,f18,f20)
- LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+ LOOP_CHUNK1(o1, o0, 1f)
FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
- LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+ LOOP_CHUNK2(o1, o0, 2f)
FREG_FROB(f36,f38,f40,f42,f44,f46,f0, f2, f4)
- LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+ LOOP_CHUNK3(o1, o0, 3f)
ba,pt %xcc, 1b+4
faligndata %f4, %f6, %f48
1: FREG_FROB(f20,f22,f24,f26,f28,f30,f32,f34,f36)
@@ -303,11 +416,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
STORE_JUMP(o0, f48, 58f)
1: FREG_FROB(f6, f8, f10,f12,f14,f16,f18,f20,f22)
- LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+ LOOP_CHUNK1(o1, o0, 1f)
FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
- LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+ LOOP_CHUNK2(o1, o0, 2f)
FREG_FROB(f38,f40,f42,f44,f46,f0, f2, f4, f6)
- LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+ LOOP_CHUNK3(o1, o0, 3f)
ba,pt %xcc, 1b+4
faligndata %f6, %f8, %f48
1: FREG_FROB(f22,f24,f26,f28,f30,f32,f34,f36,f38)
@@ -324,11 +437,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
STORE_JUMP(o0, f48, 59f)
1: FREG_FROB(f8, f10,f12,f14,f16,f18,f20,f22,f24)
- LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+ LOOP_CHUNK1(o1, o0, 1f)
FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
- LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+ LOOP_CHUNK2(o1, o0, 2f)
FREG_FROB(f40,f42,f44,f46,f0, f2, f4, f6, f8)
- LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+ LOOP_CHUNK3(o1, o0, 3f)
ba,pt %xcc, 1b+4
faligndata %f8, %f10, %f48
1: FREG_FROB(f24,f26,f28,f30,f32,f34,f36,f38,f40)
@@ -345,11 +458,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
STORE_JUMP(o0, f48, 60f)
1: FREG_FROB(f10,f12,f14,f16,f18,f20,f22,f24,f26)
- LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+ LOOP_CHUNK1(o1, o0, 1f)
FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
- LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+ LOOP_CHUNK2(o1, o0, 2f)
FREG_FROB(f42,f44,f46,f0, f2, f4, f6, f8, f10)
- LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+ LOOP_CHUNK3(o1, o0, 3f)
ba,pt %xcc, 1b+4
faligndata %f10, %f12, %f48
1: FREG_FROB(f26,f28,f30,f32,f34,f36,f38,f40,f42)
@@ -366,11 +479,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
STORE_JUMP(o0, f48, 61f)
1: FREG_FROB(f12,f14,f16,f18,f20,f22,f24,f26,f28)
- LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+ LOOP_CHUNK1(o1, o0, 1f)
FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
- LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+ LOOP_CHUNK2(o1, o0, 2f)
FREG_FROB(f44,f46,f0, f2, f4, f6, f8, f10,f12)
- LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+ LOOP_CHUNK3(o1, o0, 3f)
ba,pt %xcc, 1b+4
faligndata %f12, %f14, %f48
1: FREG_FROB(f28,f30,f32,f34,f36,f38,f40,f42,f44)
@@ -387,11 +500,11 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
STORE_JUMP(o0, f48, 62f)
1: FREG_FROB(f14,f16,f18,f20,f22,f24,f26,f28,f30)
- LOOP_CHUNK1(o1, o0, GLOBAL_SPARE, 1f)
+ LOOP_CHUNK1(o1, o0, 1f)
FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
- LOOP_CHUNK2(o1, o0, GLOBAL_SPARE, 2f)
+ LOOP_CHUNK2(o1, o0, 2f)
FREG_FROB(f46,f0, f2, f4, f6, f8, f10,f12,f14)
- LOOP_CHUNK3(o1, o0, GLOBAL_SPARE, 3f)
+ LOOP_CHUNK3(o1, o0, 3f)
ba,pt %xcc, 1b+4
faligndata %f14, %f16, %f48
1: FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
@@ -407,53 +520,53 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
FREG_FROB(f30,f32,f34,f36,f38,f40,f42,f44,f46)
STORE_JUMP(o0, f48, 63f)
-40: FINISH_VISCHUNK(o0, f0, f2, g3)
-41: FINISH_VISCHUNK(o0, f2, f4, g3)
-42: FINISH_VISCHUNK(o0, f4, f6, g3)
-43: FINISH_VISCHUNK(o0, f6, f8, g3)
-44: FINISH_VISCHUNK(o0, f8, f10, g3)
-45: FINISH_VISCHUNK(o0, f10, f12, g3)
-46: FINISH_VISCHUNK(o0, f12, f14, g3)
-47: UNEVEN_VISCHUNK(o0, f14, f0, g3)
-48: FINISH_VISCHUNK(o0, f16, f18, g3)
-49: FINISH_VISCHUNK(o0, f18, f20, g3)
-50: FINISH_VISCHUNK(o0, f20, f22, g3)
-51: FINISH_VISCHUNK(o0, f22, f24, g3)
-52: FINISH_VISCHUNK(o0, f24, f26, g3)
-53: FINISH_VISCHUNK(o0, f26, f28, g3)
-54: FINISH_VISCHUNK(o0, f28, f30, g3)
-55: UNEVEN_VISCHUNK(o0, f30, f0, g3)
-56: FINISH_VISCHUNK(o0, f32, f34, g3)
-57: FINISH_VISCHUNK(o0, f34, f36, g3)
-58: FINISH_VISCHUNK(o0, f36, f38, g3)
-59: FINISH_VISCHUNK(o0, f38, f40, g3)
-60: FINISH_VISCHUNK(o0, f40, f42, g3)
-61: FINISH_VISCHUNK(o0, f42, f44, g3)
-62: FINISH_VISCHUNK(o0, f44, f46, g3)
-63: UNEVEN_VISCHUNK_LAST(o0, f46, f0, g3)
-
-93: EX_LD_FP(LOAD(ldd, %o1, %f2))
+40: FINISH_VISCHUNK(o0, f0, f2)
+41: FINISH_VISCHUNK(o0, f2, f4)
+42: FINISH_VISCHUNK(o0, f4, f6)
+43: FINISH_VISCHUNK(o0, f6, f8)
+44: FINISH_VISCHUNK(o0, f8, f10)
+45: FINISH_VISCHUNK(o0, f10, f12)
+46: FINISH_VISCHUNK(o0, f12, f14)
+47: UNEVEN_VISCHUNK(o0, f14, f0)
+48: FINISH_VISCHUNK(o0, f16, f18)
+49: FINISH_VISCHUNK(o0, f18, f20)
+50: FINISH_VISCHUNK(o0, f20, f22)
+51: FINISH_VISCHUNK(o0, f22, f24)
+52: FINISH_VISCHUNK(o0, f24, f26)
+53: FINISH_VISCHUNK(o0, f26, f28)
+54: FINISH_VISCHUNK(o0, f28, f30)
+55: UNEVEN_VISCHUNK(o0, f30, f0)
+56: FINISH_VISCHUNK(o0, f32, f34)
+57: FINISH_VISCHUNK(o0, f34, f36)
+58: FINISH_VISCHUNK(o0, f36, f38)
+59: FINISH_VISCHUNK(o0, f38, f40)
+60: FINISH_VISCHUNK(o0, f40, f42)
+61: FINISH_VISCHUNK(o0, f42, f44)
+62: FINISH_VISCHUNK(o0, f44, f46)
+63: UNEVEN_VISCHUNK_LAST(o0, f46, f0)
+
+93: EX_LD_FP(LOAD(ldd, %o1, %f2), U1_g3_0_fp)
add %o1, 8, %o1
subcc %g3, 8, %g3
faligndata %f0, %f2, %f8
- EX_ST_FP(STORE(std, %f8, %o0))
+ EX_ST_FP(STORE(std, %f8, %o0), U1_g3_8_fp)
bl,pn %xcc, 95f
add %o0, 8, %o0
- EX_LD_FP(LOAD(ldd, %o1, %f0))
+ EX_LD_FP(LOAD(ldd, %o1, %f0), U1_g3_0_fp)
add %o1, 8, %o1
subcc %g3, 8, %g3
faligndata %f2, %f0, %f8
- EX_ST_FP(STORE(std, %f8, %o0))
+ EX_ST_FP(STORE(std, %f8, %o0), U1_g3_8_fp)
bge,pt %xcc, 93b
add %o0, 8, %o0
95: brz,pt %o2, 2f
mov %g1, %o1
-1: EX_LD_FP(LOAD(ldub, %o1, %o3))
+1: EX_LD_FP(LOAD(ldub, %o1, %o3), U1_o2_0_fp)
add %o1, 1, %o1
subcc %o2, 1, %o2
- EX_ST_FP(STORE(stb, %o3, %o0))
+ EX_ST_FP(STORE(stb, %o3, %o0), U1_o2_1_fp)
bne,pt %xcc, 1b
add %o0, 1, %o0
@@ -469,27 +582,27 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
72: andn %o2, 0xf, %GLOBAL_SPARE
and %o2, 0xf, %o2
-1: EX_LD(LOAD(ldx, %o1 + 0x00, %o5))
- EX_LD(LOAD(ldx, %o1 + 0x08, %g1))
+1: EX_LD(LOAD(ldx, %o1 + 0x00, %o5), U1_gs_0)
+ EX_LD(LOAD(ldx, %o1 + 0x08, %g1), U1_gs_0)
subcc %GLOBAL_SPARE, 0x10, %GLOBAL_SPARE
- EX_ST(STORE(stx, %o5, %o1 + %o3))
+ EX_ST(STORE(stx, %o5, %o1 + %o3), U1_gs_10)
add %o1, 0x8, %o1
- EX_ST(STORE(stx, %g1, %o1 + %o3))
+ EX_ST(STORE(stx, %g1, %o1 + %o3), U1_gs_8)
bgu,pt %XCC, 1b
add %o1, 0x8, %o1
73: andcc %o2, 0x8, %g0
be,pt %XCC, 1f
nop
- EX_LD(LOAD(ldx, %o1, %o5))
+ EX_LD(LOAD(ldx, %o1, %o5), U1_o2_0)
sub %o2, 0x8, %o2
- EX_ST(STORE(stx, %o5, %o1 + %o3))
+ EX_ST(STORE(stx, %o5, %o1 + %o3), U1_o2_8)
add %o1, 0x8, %o1
1: andcc %o2, 0x4, %g0
be,pt %XCC, 1f
nop
- EX_LD(LOAD(lduw, %o1, %o5))
+ EX_LD(LOAD(lduw, %o1, %o5), U1_o2_0)
sub %o2, 0x4, %o2
- EX_ST(STORE(stw, %o5, %o1 + %o3))
+ EX_ST(STORE(stw, %o5, %o1 + %o3), U1_o2_4)
add %o1, 0x4, %o1
1: cmp %o2, 0
be,pt %XCC, 85f
@@ -503,9 +616,9 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
sub %g0, %g1, %g1
sub %o2, %g1, %o2
-1: EX_LD(LOAD(ldub, %o1, %o5))
+1: EX_LD(LOAD(ldub, %o1, %o5), U1_g1_0)
subcc %g1, 1, %g1
- EX_ST(STORE(stb, %o5, %o1 + %o3))
+ EX_ST(STORE(stb, %o5, %o1 + %o3), U1_g1_1)
bgu,pt %icc, 1b
add %o1, 1, %o1
@@ -521,16 +634,16 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
8: mov 64, %o3
andn %o1, 0x7, %o1
- EX_LD(LOAD(ldx, %o1, %g2))
+ EX_LD(LOAD(ldx, %o1, %g2), U1_o2_0)
sub %o3, %g1, %o3
andn %o2, 0x7, %GLOBAL_SPARE
sllx %g2, %g1, %g2
-1: EX_LD(LOAD(ldx, %o1 + 0x8, %g3))
+1: EX_LD(LOAD(ldx, %o1 + 0x8, %g3), U1_gs_0_o2_adj)
subcc %GLOBAL_SPARE, 0x8, %GLOBAL_SPARE
add %o1, 0x8, %o1
srlx %g3, %o3, %o5
or %o5, %g2, %o5
- EX_ST(STORE(stx, %o5, %o0))
+ EX_ST(STORE(stx, %o5, %o0), U1_gs_8_o2_adj)
add %o0, 0x8, %o0
bgu,pt %icc, 1b
sllx %g3, %g1, %g2
@@ -548,9 +661,9 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
bne,pn %XCC, 90f
sub %o0, %o1, %o3
-1: EX_LD(LOAD(lduw, %o1, %g1))
+1: EX_LD(LOAD(lduw, %o1, %g1), U1_o2_0)
subcc %o2, 4, %o2
- EX_ST(STORE(stw, %g1, %o1 + %o3))
+ EX_ST(STORE(stw, %g1, %o1 + %o3), U1_o2_4)
bgu,pt %XCC, 1b
add %o1, 4, %o1
@@ -558,9 +671,9 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
mov EX_RETVAL(%o4), %o0
.align 32
-90: EX_LD(LOAD(ldub, %o1, %g1))
+90: EX_LD(LOAD(ldub, %o1, %g1), U1_o2_0)
subcc %o2, 1, %o2
- EX_ST(STORE(stb, %g1, %o1 + %o3))
+ EX_ST(STORE(stb, %g1, %o1 + %o3), U1_o2_1)
bgu,pt %XCC, 90b
add %o1, 1, %o1
retl
diff --git a/kernel/arch/sparc/lib/U3copy_from_user.S b/kernel/arch/sparc/lib/U3copy_from_user.S
index 88ad73d86..db73010a1 100644
--- a/kernel/arch/sparc/lib/U3copy_from_user.S
+++ b/kernel/arch/sparc/lib/U3copy_from_user.S
@@ -3,19 +3,19 @@
* Copyright (C) 1999, 2000, 2004 David S. Miller (davem@redhat.com)
*/
-#define EX_LD(x) \
+#define EX_LD(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one; \
+ .word 98b, y; \
.text; \
.align 4;
-#define EX_LD_FP(x) \
+#define EX_LD_FP(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_fp;\
+ .word 98b, y##_fp; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/U3copy_to_user.S b/kernel/arch/sparc/lib/U3copy_to_user.S
index 845139d75..c4ee858e3 100644
--- a/kernel/arch/sparc/lib/U3copy_to_user.S
+++ b/kernel/arch/sparc/lib/U3copy_to_user.S
@@ -3,19 +3,19 @@
* Copyright (C) 1999, 2000, 2004 David S. Miller (davem@redhat.com)
*/
-#define EX_ST(x) \
+#define EX_ST(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one; \
+ .word 98b, y; \
.text; \
.align 4;
-#define EX_ST_FP(x) \
+#define EX_ST_FP(x,y) \
98: x; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one_fp;\
+ .word 98b, y##_fp; \
.text; \
.align 4;
diff --git a/kernel/arch/sparc/lib/U3memcpy.S b/kernel/arch/sparc/lib/U3memcpy.S
index 491ee69e4..54f98706b 100644
--- a/kernel/arch/sparc/lib/U3memcpy.S
+++ b/kernel/arch/sparc/lib/U3memcpy.S
@@ -4,6 +4,7 @@
*/
#ifdef __KERNEL__
+#include <linux/linkage.h>
#include <asm/visasm.h>
#include <asm/asi.h>
#define GLOBAL_SPARE %g7
@@ -22,21 +23,17 @@
#endif
#ifndef EX_LD
-#define EX_LD(x) x
+#define EX_LD(x,y) x
#endif
#ifndef EX_LD_FP
-#define EX_LD_FP(x) x
+#define EX_LD_FP(x,y) x
#endif
#ifndef EX_ST
-#define EX_ST(x) x
+#define EX_ST(x,y) x
#endif
#ifndef EX_ST_FP
-#define EX_ST_FP(x) x
-#endif
-
-#ifndef EX_RETVAL
-#define EX_RETVAL(x) x
+#define EX_ST_FP(x,y) x
#endif
#ifndef LOAD
@@ -77,6 +74,87 @@
*/
.text
+#ifndef EX_RETVAL
+#define EX_RETVAL(x) x
+__restore_fp:
+ VISExitHalf
+ retl
+ nop
+ENTRY(U3_retl_o2_plus_g2_plus_g1_plus_1_fp)
+ add %g1, 1, %g1
+ add %g2, %g1, %g2
+ ba,pt %xcc, __restore_fp
+ add %o2, %g2, %o0
+ENDPROC(U3_retl_o2_plus_g2_plus_g1_plus_1_fp)
+ENTRY(U3_retl_o2_plus_g2_fp)
+ ba,pt %xcc, __restore_fp
+ add %o2, %g2, %o0
+ENDPROC(U3_retl_o2_plus_g2_fp)
+ENTRY(U3_retl_o2_plus_g2_plus_8_fp)
+ add %g2, 8, %g2
+ ba,pt %xcc, __restore_fp
+ add %o2, %g2, %o0
+ENDPROC(U3_retl_o2_plus_g2_plus_8_fp)
+ENTRY(U3_retl_o2)
+ retl
+ mov %o2, %o0
+ENDPROC(U3_retl_o2)
+ENTRY(U3_retl_o2_plus_1)
+ retl
+ add %o2, 1, %o0
+ENDPROC(U3_retl_o2_plus_1)
+ENTRY(U3_retl_o2_plus_4)
+ retl
+ add %o2, 4, %o0
+ENDPROC(U3_retl_o2_plus_4)
+ENTRY(U3_retl_o2_plus_8)
+ retl
+ add %o2, 8, %o0
+ENDPROC(U3_retl_o2_plus_8)
+ENTRY(U3_retl_o2_plus_g1_plus_1)
+ add %g1, 1, %g1
+ retl
+ add %o2, %g1, %o0
+ENDPROC(U3_retl_o2_plus_g1_plus_1)
+ENTRY(U3_retl_o2_fp)
+ ba,pt %xcc, __restore_fp
+ mov %o2, %o0
+ENDPROC(U3_retl_o2_fp)
+ENTRY(U3_retl_o2_plus_o3_sll_6_plus_0x80_fp)
+ sll %o3, 6, %o3
+ add %o3, 0x80, %o3
+ ba,pt %xcc, __restore_fp
+ add %o2, %o3, %o0
+ENDPROC(U3_retl_o2_plus_o3_sll_6_plus_0x80_fp)
+ENTRY(U3_retl_o2_plus_o3_sll_6_plus_0x40_fp)
+ sll %o3, 6, %o3
+ add %o3, 0x40, %o3
+ ba,pt %xcc, __restore_fp
+ add %o2, %o3, %o0
+ENDPROC(U3_retl_o2_plus_o3_sll_6_plus_0x40_fp)
+ENTRY(U3_retl_o2_plus_GS_plus_0x10)
+ add GLOBAL_SPARE, 0x10, GLOBAL_SPARE
+ retl
+ add %o2, GLOBAL_SPARE, %o0
+ENDPROC(U3_retl_o2_plus_GS_plus_0x10)
+ENTRY(U3_retl_o2_plus_GS_plus_0x08)
+ add GLOBAL_SPARE, 0x08, GLOBAL_SPARE
+ retl
+ add %o2, GLOBAL_SPARE, %o0
+ENDPROC(U3_retl_o2_plus_GS_plus_0x08)
+ENTRY(U3_retl_o2_and_7_plus_GS)
+ and %o2, 7, %o2
+ retl
+ add %o2, GLOBAL_SPARE, %o2
+ENDPROC(U3_retl_o2_and_7_plus_GS)
+ENTRY(U3_retl_o2_and_7_plus_GS_plus_8)
+ add GLOBAL_SPARE, 8, GLOBAL_SPARE
+ and %o2, 7, %o2
+ retl
+ add %o2, GLOBAL_SPARE, %o2
+ENDPROC(U3_retl_o2_and_7_plus_GS_plus_8)
+#endif
+
.align 64
/* The cheetah's flexible spine, oversized liver, enlarged heart,
@@ -126,8 +204,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
and %g2, 0x38, %g2
1: subcc %g1, 0x1, %g1
- EX_LD_FP(LOAD(ldub, %o1 + 0x00, %o3))
- EX_ST_FP(STORE(stb, %o3, %o1 + GLOBAL_SPARE))
+ EX_LD_FP(LOAD(ldub, %o1 + 0x00, %o3), U3_retl_o2_plus_g2_plus_g1_plus_1)
+ EX_ST_FP(STORE(stb, %o3, %o1 + GLOBAL_SPARE), U3_retl_o2_plus_g2_plus_g1_plus_1)
bgu,pt %XCC, 1b
add %o1, 0x1, %o1
@@ -138,20 +216,20 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
be,pt %icc, 3f
alignaddr %o1, %g0, %o1
- EX_LD_FP(LOAD(ldd, %o1, %f4))
-1: EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f6))
+ EX_LD_FP(LOAD(ldd, %o1, %f4), U3_retl_o2_plus_g2)
+1: EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f6), U3_retl_o2_plus_g2)
add %o1, 0x8, %o1
subcc %g2, 0x8, %g2
faligndata %f4, %f6, %f0
- EX_ST_FP(STORE(std, %f0, %o0))
+ EX_ST_FP(STORE(std, %f0, %o0), U3_retl_o2_plus_g2_plus_8)
be,pn %icc, 3f
add %o0, 0x8, %o0
- EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f4))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x8, %f4), U3_retl_o2_plus_g2)
add %o1, 0x8, %o1
subcc %g2, 0x8, %g2
faligndata %f6, %f4, %f2
- EX_ST_FP(STORE(std, %f2, %o0))
+ EX_ST_FP(STORE(std, %f2, %o0), U3_retl_o2_plus_g2_plus_8)
bne,pt %icc, 1b
add %o0, 0x8, %o0
@@ -161,25 +239,25 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
LOAD(prefetch, %o1 + 0x080, #one_read)
LOAD(prefetch, %o1 + 0x0c0, #one_read)
LOAD(prefetch, %o1 + 0x100, #one_read)
- EX_LD_FP(LOAD(ldd, %o1 + 0x000, %f0))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x000, %f0), U3_retl_o2)
LOAD(prefetch, %o1 + 0x140, #one_read)
- EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2), U3_retl_o2)
LOAD(prefetch, %o1 + 0x180, #one_read)
- EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4), U3_retl_o2)
LOAD(prefetch, %o1 + 0x1c0, #one_read)
faligndata %f0, %f2, %f16
- EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6), U3_retl_o2)
faligndata %f2, %f4, %f18
- EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8), U3_retl_o2)
faligndata %f4, %f6, %f20
- EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10), U3_retl_o2)
faligndata %f6, %f8, %f22
- EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12), U3_retl_o2)
faligndata %f8, %f10, %f24
- EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14), U3_retl_o2)
faligndata %f10, %f12, %f26
- EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0), U3_retl_o2)
subcc GLOBAL_SPARE, 0x80, GLOBAL_SPARE
add %o1, 0x40, %o1
@@ -190,26 +268,26 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
.align 64
1:
- EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2), U3_retl_o2_plus_o3_sll_6_plus_0x80)
faligndata %f12, %f14, %f28
- EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4), U3_retl_o2_plus_o3_sll_6_plus_0x80)
faligndata %f14, %f0, %f30
- EX_ST_FP(STORE_BLK(%f16, %o0))
- EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6))
+ EX_ST_FP(STORE_BLK(%f16, %o0), U3_retl_o2_plus_o3_sll_6_plus_0x80)
+ EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6), U3_retl_o2_plus_o3_sll_6_plus_0x40)
faligndata %f0, %f2, %f16
add %o0, 0x40, %o0
- EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8), U3_retl_o2_plus_o3_sll_6_plus_0x40)
faligndata %f2, %f4, %f18
- EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10), U3_retl_o2_plus_o3_sll_6_plus_0x40)
faligndata %f4, %f6, %f20
- EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12), U3_retl_o2_plus_o3_sll_6_plus_0x40)
subcc %o3, 0x01, %o3
faligndata %f6, %f8, %f22
- EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14), U3_retl_o2_plus_o3_sll_6_plus_0x80)
faligndata %f8, %f10, %f24
- EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0), U3_retl_o2_plus_o3_sll_6_plus_0x80)
LOAD(prefetch, %o1 + 0x1c0, #one_read)
faligndata %f10, %f12, %f26
bg,pt %XCC, 1b
@@ -217,29 +295,29 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
/* Finally we copy the last full 64-byte block. */
2:
- EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x008, %f2), U3_retl_o2_plus_o3_sll_6_plus_0x80)
faligndata %f12, %f14, %f28
- EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x010, %f4), U3_retl_o2_plus_o3_sll_6_plus_0x80)
faligndata %f14, %f0, %f30
- EX_ST_FP(STORE_BLK(%f16, %o0))
- EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6))
+ EX_ST_FP(STORE_BLK(%f16, %o0), U3_retl_o2_plus_o3_sll_6_plus_0x80)
+ EX_LD_FP(LOAD(ldd, %o1 + 0x018, %f6), U3_retl_o2_plus_o3_sll_6_plus_0x40)
faligndata %f0, %f2, %f16
- EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x020, %f8), U3_retl_o2_plus_o3_sll_6_plus_0x40)
faligndata %f2, %f4, %f18
- EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x028, %f10), U3_retl_o2_plus_o3_sll_6_plus_0x40)
faligndata %f4, %f6, %f20
- EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x030, %f12), U3_retl_o2_plus_o3_sll_6_plus_0x40)
faligndata %f6, %f8, %f22
- EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x038, %f14), U3_retl_o2_plus_o3_sll_6_plus_0x40)
faligndata %f8, %f10, %f24
cmp %g1, 0
be,pt %XCC, 1f
add %o0, 0x40, %o0
- EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x040, %f0), U3_retl_o2_plus_o3_sll_6_plus_0x40)
1: faligndata %f10, %f12, %f26
faligndata %f12, %f14, %f28
faligndata %f14, %f0, %f30
- EX_ST_FP(STORE_BLK(%f16, %o0))
+ EX_ST_FP(STORE_BLK(%f16, %o0), U3_retl_o2_plus_o3_sll_6_plus_0x40)
add %o0, 0x40, %o0
add %o1, 0x40, %o1
membar #Sync
@@ -259,20 +337,20 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
sub %o2, %g2, %o2
be,a,pt %XCC, 1f
- EX_LD_FP(LOAD(ldd, %o1 + 0x00, %f0))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x00, %f0), U3_retl_o2_plus_g2)
-1: EX_LD_FP(LOAD(ldd, %o1 + 0x08, %f2))
+1: EX_LD_FP(LOAD(ldd, %o1 + 0x08, %f2), U3_retl_o2_plus_g2)
add %o1, 0x8, %o1
subcc %g2, 0x8, %g2
faligndata %f0, %f2, %f8
- EX_ST_FP(STORE(std, %f8, %o0))
+ EX_ST_FP(STORE(std, %f8, %o0), U3_retl_o2_plus_g2_plus_8)
be,pn %XCC, 2f
add %o0, 0x8, %o0
- EX_LD_FP(LOAD(ldd, %o1 + 0x08, %f0))
+ EX_LD_FP(LOAD(ldd, %o1 + 0x08, %f0), U3_retl_o2_plus_g2)
add %o1, 0x8, %o1
subcc %g2, 0x8, %g2
faligndata %f2, %f0, %f8
- EX_ST_FP(STORE(std, %f8, %o0))
+ EX_ST_FP(STORE(std, %f8, %o0), U3_retl_o2_plus_g2_plus_8)
bne,pn %XCC, 1b
add %o0, 0x8, %o0
@@ -292,30 +370,33 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
andcc %o2, 0x8, %g0
be,pt %icc, 1f
nop
- EX_LD(LOAD(ldx, %o1, %o5))
- EX_ST(STORE(stx, %o5, %o1 + %o3))
+ EX_LD(LOAD(ldx, %o1, %o5), U3_retl_o2)
+ EX_ST(STORE(stx, %o5, %o1 + %o3), U3_retl_o2)
add %o1, 0x8, %o1
+ sub %o2, 8, %o2
1: andcc %o2, 0x4, %g0
be,pt %icc, 1f
nop
- EX_LD(LOAD(lduw, %o1, %o5))
- EX_ST(STORE(stw, %o5, %o1 + %o3))
+ EX_LD(LOAD(lduw, %o1, %o5), U3_retl_o2)
+ EX_ST(STORE(stw, %o5, %o1 + %o3), U3_retl_o2)
add %o1, 0x4, %o1
+ sub %o2, 4, %o2
1: andcc %o2, 0x2, %g0
be,pt %icc, 1f
nop
- EX_LD(LOAD(lduh, %o1, %o5))
- EX_ST(STORE(sth, %o5, %o1 + %o3))
+ EX_LD(LOAD(lduh, %o1, %o5), U3_retl_o2)
+ EX_ST(STORE(sth, %o5, %o1 + %o3), U3_retl_o2)
add %o1, 0x2, %o1
+ sub %o2, 2, %o2
1: andcc %o2, 0x1, %g0
be,pt %icc, 85f
nop
- EX_LD(LOAD(ldub, %o1, %o5))
+ EX_LD(LOAD(ldub, %o1, %o5), U3_retl_o2)
ba,pt %xcc, 85f
- EX_ST(STORE(stb, %o5, %o1 + %o3))
+ EX_ST(STORE(stb, %o5, %o1 + %o3), U3_retl_o2)
.align 64
70: /* 16 < len <= 64 */
@@ -326,26 +407,26 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
andn %o2, 0xf, GLOBAL_SPARE
and %o2, 0xf, %o2
1: subcc GLOBAL_SPARE, 0x10, GLOBAL_SPARE
- EX_LD(LOAD(ldx, %o1 + 0x00, %o5))
- EX_LD(LOAD(ldx, %o1 + 0x08, %g1))
- EX_ST(STORE(stx, %o5, %o1 + %o3))
+ EX_LD(LOAD(ldx, %o1 + 0x00, %o5), U3_retl_o2_plus_GS_plus_0x10)
+ EX_LD(LOAD(ldx, %o1 + 0x08, %g1), U3_retl_o2_plus_GS_plus_0x10)
+ EX_ST(STORE(stx, %o5, %o1 + %o3), U3_retl_o2_plus_GS_plus_0x10)
add %o1, 0x8, %o1
- EX_ST(STORE(stx, %g1, %o1 + %o3))
+ EX_ST(STORE(stx, %g1, %o1 + %o3), U3_retl_o2_plus_GS_plus_0x08)
bgu,pt %XCC, 1b
add %o1, 0x8, %o1
73: andcc %o2, 0x8, %g0
be,pt %XCC, 1f
nop
sub %o2, 0x8, %o2
- EX_LD(LOAD(ldx, %o1, %o5))
- EX_ST(STORE(stx, %o5, %o1 + %o3))
+ EX_LD(LOAD(ldx, %o1, %o5), U3_retl_o2_plus_8)
+ EX_ST(STORE(stx, %o5, %o1 + %o3), U3_retl_o2_plus_8)
add %o1, 0x8, %o1
1: andcc %o2, 0x4, %g0
be,pt %XCC, 1f
nop
sub %o2, 0x4, %o2
- EX_LD(LOAD(lduw, %o1, %o5))
- EX_ST(STORE(stw, %o5, %o1 + %o3))
+ EX_LD(LOAD(lduw, %o1, %o5), U3_retl_o2_plus_4)
+ EX_ST(STORE(stw, %o5, %o1 + %o3), U3_retl_o2_plus_4)
add %o1, 0x4, %o1
1: cmp %o2, 0
be,pt %XCC, 85f
@@ -361,8 +442,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
sub %o2, %g1, %o2
1: subcc %g1, 1, %g1
- EX_LD(LOAD(ldub, %o1, %o5))
- EX_ST(STORE(stb, %o5, %o1 + %o3))
+ EX_LD(LOAD(ldub, %o1, %o5), U3_retl_o2_plus_g1_plus_1)
+ EX_ST(STORE(stb, %o5, %o1 + %o3), U3_retl_o2_plus_g1_plus_1)
bgu,pt %icc, 1b
add %o1, 1, %o1
@@ -378,16 +459,16 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
8: mov 64, %o3
andn %o1, 0x7, %o1
- EX_LD(LOAD(ldx, %o1, %g2))
+ EX_LD(LOAD(ldx, %o1, %g2), U3_retl_o2)
sub %o3, %g1, %o3
andn %o2, 0x7, GLOBAL_SPARE
sllx %g2, %g1, %g2
-1: EX_LD(LOAD(ldx, %o1 + 0x8, %g3))
+1: EX_LD(LOAD(ldx, %o1 + 0x8, %g3), U3_retl_o2_and_7_plus_GS)
subcc GLOBAL_SPARE, 0x8, GLOBAL_SPARE
add %o1, 0x8, %o1
srlx %g3, %o3, %o5
or %o5, %g2, %o5
- EX_ST(STORE(stx, %o5, %o0))
+ EX_ST(STORE(stx, %o5, %o0), U3_retl_o2_and_7_plus_GS_plus_8)
add %o0, 0x8, %o0
bgu,pt %icc, 1b
sllx %g3, %g1, %g2
@@ -407,8 +488,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
1:
subcc %o2, 4, %o2
- EX_LD(LOAD(lduw, %o1, %g1))
- EX_ST(STORE(stw, %g1, %o1 + %o3))
+ EX_LD(LOAD(lduw, %o1, %g1), U3_retl_o2_plus_4)
+ EX_ST(STORE(stw, %g1, %o1 + %o3), U3_retl_o2_plus_4)
bgu,pt %XCC, 1b
add %o1, 4, %o1
@@ -418,8 +499,8 @@ FUNC_NAME: /* %o0=dst, %o1=src, %o2=len */
.align 32
90:
subcc %o2, 1, %o2
- EX_LD(LOAD(ldub, %o1, %g1))
- EX_ST(STORE(stb, %g1, %o1 + %o3))
+ EX_LD(LOAD(ldub, %o1, %g1), U3_retl_o2_plus_1)
+ EX_ST(STORE(stb, %g1, %o1 + %o3), U3_retl_o2_plus_1)
bgu,pt %XCC, 90b
add %o1, 1, %o1
retl
diff --git a/kernel/arch/sparc/lib/copy_in_user.S b/kernel/arch/sparc/lib/copy_in_user.S
index 302c0e60d..4c89b486f 100644
--- a/kernel/arch/sparc/lib/copy_in_user.S
+++ b/kernel/arch/sparc/lib/copy_in_user.S
@@ -8,18 +8,33 @@
#define XCC xcc
-#define EX(x,y) \
+#define EX(x,y,z) \
98: x,y; \
.section __ex_table,"a";\
.align 4; \
- .word 98b, __retl_one; \
+ .word 98b, z; \
.text; \
.align 4;
+#define EX_O4(x,y) EX(x,y,__retl_o4_plus_8)
+#define EX_O2_4(x,y) EX(x,y,__retl_o2_plus_4)
+#define EX_O2_1(x,y) EX(x,y,__retl_o2_plus_1)
+
.register %g2,#scratch
.register %g3,#scratch
.text
+__retl_o4_plus_8:
+ add %o4, %o2, %o4
+ retl
+ add %o4, 8, %o0
+__retl_o2_plus_4:
+ retl
+ add %o2, 4, %o0
+__retl_o2_plus_1:
+ retl
+ add %o2, 1, %o0
+
.align 32
/* Don't try to get too fancy here, just nice and
@@ -44,8 +59,8 @@ ENTRY(___copy_in_user) /* %o0=dst, %o1=src, %o2=len */
andn %o2, 0x7, %o4
and %o2, 0x7, %o2
1: subcc %o4, 0x8, %o4
- EX(ldxa [%o1] %asi, %o5)
- EX(stxa %o5, [%o0] %asi)
+ EX_O4(ldxa [%o1] %asi, %o5)
+ EX_O4(stxa %o5, [%o0] %asi)
add %o1, 0x8, %o1
bgu,pt %XCC, 1b
add %o0, 0x8, %o0
@@ -53,8 +68,8 @@ ENTRY(___copy_in_user) /* %o0=dst, %o1=src, %o2=len */
be,pt %XCC, 1f
nop
sub %o2, 0x4, %o2
- EX(lduwa [%o1] %asi, %o5)
- EX(stwa %o5, [%o0] %asi)
+ EX_O2_4(lduwa [%o1] %asi, %o5)
+ EX_O2_4(stwa %o5, [%o0] %asi)
add %o1, 0x4, %o1
add %o0, 0x4, %o0
1: cmp %o2, 0
@@ -70,8 +85,8 @@ ENTRY(___copy_in_user) /* %o0=dst, %o1=src, %o2=len */
82:
subcc %o2, 4, %o2
- EX(lduwa [%o1] %asi, %g1)
- EX(stwa %g1, [%o0] %asi)
+ EX_O2_4(lduwa [%o1] %asi, %g1)
+ EX_O2_4(stwa %g1, [%o0] %asi)
add %o1, 4, %o1
bgu,pt %XCC, 82b
add %o0, 4, %o0
@@ -82,8 +97,8 @@ ENTRY(___copy_in_user) /* %o0=dst, %o1=src, %o2=len */
.align 32
90:
subcc %o2, 1, %o2
- EX(lduba [%o1] %asi, %g1)
- EX(stba %g1, [%o0] %asi)
+ EX_O2_1(lduba [%o1] %asi, %g1)
+ EX_O2_1(stba %g1, [%o0] %asi)
add %o1, 1, %o1
bgu,pt %XCC, 90b
add %o0, 1, %o0
diff --git a/kernel/arch/sparc/lib/user_fixup.c b/kernel/arch/sparc/lib/user_fixup.c
deleted file mode 100644
index ac96ae236..000000000
--- a/kernel/arch/sparc/lib/user_fixup.c
+++ /dev/null
@@ -1,71 +0,0 @@
-/* user_fixup.c: Fix up user copy faults.
- *
- * Copyright (C) 2004 David S. Miller <davem@redhat.com>
- */
-
-#include <linux/compiler.h>
-#include <linux/kernel.h>
-#include <linux/string.h>
-#include <linux/errno.h>
-#include <linux/module.h>
-
-#include <asm/uaccess.h>
-
-/* Calculating the exact fault address when using
- * block loads and stores can be very complicated.
- *
- * Instead of trying to be clever and handling all
- * of the cases, just fix things up simply here.
- */
-
-static unsigned long compute_size(unsigned long start, unsigned long size, unsigned long *offset)
-{
- unsigned long fault_addr = current_thread_info()->fault_address;
- unsigned long end = start + size;
-
- if (fault_addr < start || fault_addr >= end) {
- *offset = 0;
- } else {
- *offset = fault_addr - start;
- size = end - fault_addr;
- }
- return size;
-}
-
-unsigned long copy_from_user_fixup(void *to, const void __user *from, unsigned long size)
-{
- unsigned long offset;
-
- size = compute_size((unsigned long) from, size, &offset);
- if (likely(size))
- memset(to + offset, 0, size);
-
- return size;
-}
-EXPORT_SYMBOL(copy_from_user_fixup);
-
-unsigned long copy_to_user_fixup(void __user *to, const void *from, unsigned long size)
-{
- unsigned long offset;
-
- return compute_size((unsigned long) to, size, &offset);
-}
-EXPORT_SYMBOL(copy_to_user_fixup);
-
-unsigned long copy_in_user_fixup(void __user *to, void __user *from, unsigned long size)
-{
- unsigned long fault_addr = current_thread_info()->fault_address;
- unsigned long start = (unsigned long) to;
- unsigned long end = start + size;
-
- if (fault_addr >= start && fault_addr < end)
- return end - fault_addr;
-
- start = (unsigned long) from;
- end = start + size;
- if (fault_addr >= start && fault_addr < end)
- return end - fault_addr;
-
- return size;
-}
-EXPORT_SYMBOL(copy_in_user_fixup);
diff --git a/kernel/arch/sparc/mm/fault_64.c b/kernel/arch/sparc/mm/fault_64.c
index dbabe5713..e15f33715 100644
--- a/kernel/arch/sparc/mm/fault_64.c
+++ b/kernel/arch/sparc/mm/fault_64.c
@@ -479,14 +479,14 @@ good_area:
up_read(&mm->mmap_sem);
mm_rss = get_mm_rss(mm);
-#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- mm_rss -= (mm->context.huge_pte_count * (HPAGE_SIZE / PAGE_SIZE));
+#if defined(CONFIG_TRANSPARENT_HUGEPAGE)
+ mm_rss -= (mm->context.thp_pte_count * (HPAGE_SIZE / PAGE_SIZE));
#endif
if (unlikely(mm_rss >
mm->context.tsb_block[MM_TSB_BASE].tsb_rss_limit))
tsb_grow(mm, MM_TSB_BASE, mm_rss);
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- mm_rss = mm->context.huge_pte_count;
+ mm_rss = mm->context.hugetlb_pte_count + mm->context.thp_pte_count;
if (unlikely(mm_rss >
mm->context.tsb_block[MM_TSB_HUGE].tsb_rss_limit)) {
if (mm->context.tsb_block[MM_TSB_HUGE].tsb)
diff --git a/kernel/arch/sparc/mm/hugetlbpage.c b/kernel/arch/sparc/mm/hugetlbpage.c
index 131eaf4ad..da1142401 100644
--- a/kernel/arch/sparc/mm/hugetlbpage.c
+++ b/kernel/arch/sparc/mm/hugetlbpage.c
@@ -176,17 +176,31 @@ void set_huge_pte_at(struct mm_struct *mm, unsigned long addr,
pte_t *ptep, pte_t entry)
{
int i;
+ pte_t orig[2];
+ unsigned long nptes;
if (!pte_present(*ptep) && pte_present(entry))
- mm->context.huge_pte_count++;
+ mm->context.hugetlb_pte_count++;
addr &= HPAGE_MASK;
- for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
- set_pte_at(mm, addr, ptep, entry);
+
+ nptes = 1 << HUGETLB_PAGE_ORDER;
+ orig[0] = *ptep;
+ orig[1] = *(ptep + nptes / 2);
+ for (i = 0; i < nptes; i++) {
+ *ptep = entry;
ptep++;
addr += PAGE_SIZE;
pte_val(entry) += PAGE_SIZE;
}
+
+ /* Issue TLB flush at REAL_HPAGE_SIZE boundaries */
+ addr -= REAL_HPAGE_SIZE;
+ ptep -= nptes / 2;
+ maybe_tlb_batch_add(mm, addr, ptep, orig[1], 0);
+ addr -= REAL_HPAGE_SIZE;
+ ptep -= nptes / 2;
+ maybe_tlb_batch_add(mm, addr, ptep, orig[0], 0);
}
pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
@@ -194,19 +208,28 @@ pte_t huge_ptep_get_and_clear(struct mm_struct *mm, unsigned long addr,
{
pte_t entry;
int i;
+ unsigned long nptes;
entry = *ptep;
if (pte_present(entry))
- mm->context.huge_pte_count--;
+ mm->context.hugetlb_pte_count--;
addr &= HPAGE_MASK;
-
- for (i = 0; i < (1 << HUGETLB_PAGE_ORDER); i++) {
- pte_clear(mm, addr, ptep);
+ nptes = 1 << HUGETLB_PAGE_ORDER;
+ for (i = 0; i < nptes; i++) {
+ *ptep = __pte(0UL);
addr += PAGE_SIZE;
ptep++;
}
+ /* Issue TLB flush at REAL_HPAGE_SIZE boundaries */
+ addr -= REAL_HPAGE_SIZE;
+ ptep -= nptes / 2;
+ maybe_tlb_batch_add(mm, addr, ptep, entry, 0);
+ addr -= REAL_HPAGE_SIZE;
+ ptep -= nptes / 2;
+ maybe_tlb_batch_add(mm, addr, ptep, entry, 0);
+
return entry;
}
diff --git a/kernel/arch/sparc/mm/init_64.c b/kernel/arch/sparc/mm/init_64.c
index 3025bd57f..3d3414c14 100644
--- a/kernel/arch/sparc/mm/init_64.c
+++ b/kernel/arch/sparc/mm/init_64.c
@@ -324,18 +324,6 @@ static void __update_mmu_tsb_insert(struct mm_struct *mm, unsigned long tsb_inde
tsb_insert(tsb, tag, tte);
}
-#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
-static inline bool is_hugetlb_pte(pte_t pte)
-{
- if ((tlb_type == hypervisor &&
- (pte_val(pte) & _PAGE_SZALL_4V) == _PAGE_SZHUGE_4V) ||
- (tlb_type != hypervisor &&
- (pte_val(pte) & _PAGE_SZALL_4U) == _PAGE_SZHUGE_4U))
- return true;
- return false;
-}
-#endif
-
void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t *ptep)
{
struct mm_struct *mm;
@@ -358,7 +346,8 @@ void update_mmu_cache(struct vm_area_struct *vma, unsigned long address, pte_t *
spin_lock_irqsave(&mm->context.lock, flags);
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- if (mm->context.huge_pte_count && is_hugetlb_pte(pte))
+ if ((mm->context.hugetlb_pte_count || mm->context.thp_pte_count) &&
+ is_hugetlb_pte(pte))
__update_mmu_tsb_insert(mm, MM_TSB_HUGE, REAL_HPAGE_SHIFT,
address, pte_val(pte));
else
@@ -811,8 +800,10 @@ struct mdesc_mblock {
};
static struct mdesc_mblock *mblocks;
static int num_mblocks;
+static int find_numa_node_for_addr(unsigned long pa,
+ struct node_mem_mask *pnode_mask);
-static unsigned long ra_to_pa(unsigned long addr)
+static unsigned long __init ra_to_pa(unsigned long addr)
{
int i;
@@ -828,8 +819,11 @@ static unsigned long ra_to_pa(unsigned long addr)
return addr;
}
-static int find_node(unsigned long addr)
+static int __init find_node(unsigned long addr)
{
+ static bool search_mdesc = true;
+ static struct node_mem_mask last_mem_mask = { ~0UL, ~0UL };
+ static int last_index;
int i;
addr = ra_to_pa(addr);
@@ -839,13 +833,30 @@ static int find_node(unsigned long addr)
if ((addr & p->mask) == p->val)
return i;
}
- /* The following condition has been observed on LDOM guests.*/
- WARN_ONCE(1, "find_node: A physical address doesn't match a NUMA node"
- " rule. Some physical memory will be owned by node 0.");
- return 0;
+ /* The following condition has been observed on LDOM guests because
+ * node_masks only contains the best latency mask and value.
+ * LDOM guest's mdesc can contain a single latency group to
+ * cover multiple address range. Print warning message only if the
+ * address cannot be found in node_masks nor mdesc.
+ */
+ if ((search_mdesc) &&
+ ((addr & last_mem_mask.mask) != last_mem_mask.val)) {
+ /* find the available node in the mdesc */
+ last_index = find_numa_node_for_addr(addr, &last_mem_mask);
+ numadbg("find_node: latency group for address 0x%lx is %d\n",
+ addr, last_index);
+ if ((last_index < 0) || (last_index >= num_node_masks)) {
+ /* WARN_ONCE() and use default group 0 */
+ WARN_ONCE(1, "find_node: A physical address doesn't match a NUMA node rule. Some physical memory will be owned by node 0.");
+ search_mdesc = false;
+ last_index = 0;
+ }
+ }
+
+ return last_index;
}
-static u64 memblock_nid_range(u64 start, u64 end, int *nid)
+static u64 __init memblock_nid_range(u64 start, u64 end, int *nid)
{
*nid = find_node(start);
start += PAGE_SIZE;
@@ -1169,6 +1180,41 @@ int __node_distance(int from, int to)
return numa_latency[from][to];
}
+static int find_numa_node_for_addr(unsigned long pa,
+ struct node_mem_mask *pnode_mask)
+{
+ struct mdesc_handle *md = mdesc_grab();
+ u64 node, arc;
+ int i = 0;
+
+ node = mdesc_node_by_name(md, MDESC_NODE_NULL, "latency-groups");
+ if (node == MDESC_NODE_NULL)
+ goto out;
+
+ mdesc_for_each_node_by_name(md, node, "group") {
+ mdesc_for_each_arc(arc, md, node, MDESC_ARC_TYPE_FWD) {
+ u64 target = mdesc_arc_target(md, arc);
+ struct mdesc_mlgroup *m = find_mlgroup(target);
+
+ if (!m)
+ continue;
+ if ((pa & m->mask) == m->match) {
+ if (pnode_mask) {
+ pnode_mask->mask = m->mask;
+ pnode_mask->val = m->match;
+ }
+ mdesc_release(md);
+ return i;
+ }
+ }
+ i++;
+ }
+
+out:
+ mdesc_release(md);
+ return -1;
+}
+
static int find_best_numa_node_for_mlgroup(struct mdesc_mlgroup *grp)
{
int i;
@@ -1267,13 +1313,6 @@ static int __init numa_parse_mdesc(void)
int i, j, err, count;
u64 node;
- /* Some sane defaults for numa latency values */
- for (i = 0; i < MAX_NUMNODES; i++) {
- for (j = 0; j < MAX_NUMNODES; j++)
- numa_latency[i][j] = (i == j) ?
- LOCAL_DISTANCE : REMOTE_DISTANCE;
- }
-
node = mdesc_node_by_name(md, MDESC_NODE_NULL, "latency-groups");
if (node == MDESC_NODE_NULL) {
mdesc_release(md);
@@ -1369,10 +1408,18 @@ static int __init numa_parse_sun4u(void)
static int __init bootmem_init_numa(void)
{
+ int i, j;
int err = -1;
numadbg("bootmem_init_numa()\n");
+ /* Some sane defaults for numa latency values */
+ for (i = 0; i < MAX_NUMNODES; i++) {
+ for (j = 0; j < MAX_NUMNODES; j++)
+ numa_latency[i][j] = (i == j) ?
+ LOCAL_DISTANCE : REMOTE_DISTANCE;
+ }
+
if (numa_enabled) {
if (tlb_type == hypervisor)
err = numa_parse_mdesc();
@@ -2832,9 +2879,10 @@ void hugetlb_setup(struct pt_regs *regs)
* the Data-TLB for huge pages.
*/
if (tlb_type == cheetah_plus) {
+ bool need_context_reload = false;
unsigned long ctx;
- spin_lock(&ctx_alloc_lock);
+ spin_lock_irq(&ctx_alloc_lock);
ctx = mm->context.sparc64_ctx_val;
ctx &= ~CTX_PGSZ_MASK;
ctx |= CTX_PGSZ_BASE << CTX_PGSZ0_SHIFT;
@@ -2853,9 +2901,12 @@ void hugetlb_setup(struct pt_regs *regs)
* also executing in this address space.
*/
mm->context.sparc64_ctx_val = ctx;
- on_each_cpu(context_reload, mm, 0);
+ need_context_reload = true;
}
- spin_unlock(&ctx_alloc_lock);
+ spin_unlock_irq(&ctx_alloc_lock);
+
+ if (need_context_reload)
+ on_each_cpu(context_reload, mm, 0);
}
}
#endif
diff --git a/kernel/arch/sparc/mm/tlb.c b/kernel/arch/sparc/mm/tlb.c
index 9df2190c0..3659d37b4 100644
--- a/kernel/arch/sparc/mm/tlb.c
+++ b/kernel/arch/sparc/mm/tlb.c
@@ -67,7 +67,7 @@ void arch_leave_lazy_mmu_mode(void)
}
static void tlb_batch_add_one(struct mm_struct *mm, unsigned long vaddr,
- bool exec)
+ bool exec, bool huge)
{
struct tlb_batch *tb = &get_cpu_var(tlb_batch);
unsigned long nr;
@@ -84,13 +84,21 @@ static void tlb_batch_add_one(struct mm_struct *mm, unsigned long vaddr,
}
if (!tb->active) {
- flush_tsb_user_page(mm, vaddr);
+ flush_tsb_user_page(mm, vaddr, huge);
global_flush_tlb_page(mm, vaddr);
goto out;
}
- if (nr == 0)
+ if (nr == 0) {
tb->mm = mm;
+ tb->huge = huge;
+ }
+
+ if (tb->huge != huge) {
+ flush_tlb_pending();
+ tb->huge = huge;
+ nr = 0;
+ }
tb->vaddrs[nr] = vaddr;
tb->tlb_nr = ++nr;
@@ -104,6 +112,8 @@ out:
void tlb_batch_add(struct mm_struct *mm, unsigned long vaddr,
pte_t *ptep, pte_t orig, int fullmm)
{
+ bool huge = is_hugetlb_pte(orig);
+
if (tlb_type != hypervisor &&
pte_dirty(orig)) {
unsigned long paddr, pfn = pte_pfn(orig);
@@ -129,7 +139,7 @@ void tlb_batch_add(struct mm_struct *mm, unsigned long vaddr,
no_cache_flush:
if (!fullmm)
- tlb_batch_add_one(mm, vaddr, pte_exec(orig));
+ tlb_batch_add_one(mm, vaddr, pte_exec(orig), huge);
}
#ifdef CONFIG_TRANSPARENT_HUGEPAGE
@@ -145,7 +155,7 @@ static void tlb_batch_pmd_scan(struct mm_struct *mm, unsigned long vaddr,
if (pte_val(*pte) & _PAGE_VALID) {
bool exec = pte_exec(*pte);
- tlb_batch_add_one(mm, vaddr, exec);
+ tlb_batch_add_one(mm, vaddr, exec, false);
}
pte++;
vaddr += PAGE_SIZE;
@@ -165,9 +175,9 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr,
if ((pmd_val(pmd) ^ pmd_val(orig)) & _PAGE_PMD_HUGE) {
if (pmd_val(pmd) & _PAGE_PMD_HUGE)
- mm->context.huge_pte_count++;
+ mm->context.thp_pte_count++;
else
- mm->context.huge_pte_count--;
+ mm->context.thp_pte_count--;
/* Do not try to allocate the TSB hash table if we
* don't have one already. We have various locks held
@@ -185,8 +195,9 @@ void set_pmd_at(struct mm_struct *mm, unsigned long addr,
pte_t orig_pte = __pte(pmd_val(orig));
bool exec = pte_exec(orig_pte);
- tlb_batch_add_one(mm, addr, exec);
- tlb_batch_add_one(mm, addr + REAL_HPAGE_SIZE, exec);
+ tlb_batch_add_one(mm, addr, exec, true);
+ tlb_batch_add_one(mm, addr + REAL_HPAGE_SIZE, exec,
+ true);
} else {
tlb_batch_pmd_scan(mm, addr, orig);
}
diff --git a/kernel/arch/sparc/mm/tsb.c b/kernel/arch/sparc/mm/tsb.c
index a06576683..9cdeca0fa 100644
--- a/kernel/arch/sparc/mm/tsb.c
+++ b/kernel/arch/sparc/mm/tsb.c
@@ -27,6 +27,20 @@ static inline int tag_compare(unsigned long tag, unsigned long vaddr)
return (tag == (vaddr >> 22));
}
+static void flush_tsb_kernel_range_scan(unsigned long start, unsigned long end)
+{
+ unsigned long idx;
+
+ for (idx = 0; idx < KERNEL_TSB_NENTRIES; idx++) {
+ struct tsb *ent = &swapper_tsb[idx];
+ unsigned long match = idx << 13;
+
+ match |= (ent->tag << 22);
+ if (match >= start && match < end)
+ ent->tag = (1UL << TSB_TAG_INVALID_BIT);
+ }
+}
+
/* TSB flushes need only occur on the processor initiating the address
* space modification, not on each cpu the address space has run on.
* Only the TLB flush needs that treatment.
@@ -36,6 +50,9 @@ void flush_tsb_kernel_range(unsigned long start, unsigned long end)
{
unsigned long v;
+ if ((end - start) >> PAGE_SHIFT >= 2 * KERNEL_TSB_NENTRIES)
+ return flush_tsb_kernel_range_scan(start, end);
+
for (v = start; v < end; v += PAGE_SIZE) {
unsigned long hash = tsb_hash(v, PAGE_SHIFT,
KERNEL_TSB_NENTRIES);
@@ -76,14 +93,15 @@ void flush_tsb_user(struct tlb_batch *tb)
spin_lock_irqsave(&mm->context.lock, flags);
- base = (unsigned long) mm->context.tsb_block[MM_TSB_BASE].tsb;
- nentries = mm->context.tsb_block[MM_TSB_BASE].tsb_nentries;
- if (tlb_type == cheetah_plus || tlb_type == hypervisor)
- base = __pa(base);
- __flush_tsb_one(tb, PAGE_SHIFT, base, nentries);
-
+ if (!tb->huge) {
+ base = (unsigned long) mm->context.tsb_block[MM_TSB_BASE].tsb;
+ nentries = mm->context.tsb_block[MM_TSB_BASE].tsb_nentries;
+ if (tlb_type == cheetah_plus || tlb_type == hypervisor)
+ base = __pa(base);
+ __flush_tsb_one(tb, PAGE_SHIFT, base, nentries);
+ }
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- if (mm->context.tsb_block[MM_TSB_HUGE].tsb) {
+ if (tb->huge && mm->context.tsb_block[MM_TSB_HUGE].tsb) {
base = (unsigned long) mm->context.tsb_block[MM_TSB_HUGE].tsb;
nentries = mm->context.tsb_block[MM_TSB_HUGE].tsb_nentries;
if (tlb_type == cheetah_plus || tlb_type == hypervisor)
@@ -94,20 +112,21 @@ void flush_tsb_user(struct tlb_batch *tb)
spin_unlock_irqrestore(&mm->context.lock, flags);
}
-void flush_tsb_user_page(struct mm_struct *mm, unsigned long vaddr)
+void flush_tsb_user_page(struct mm_struct *mm, unsigned long vaddr, bool huge)
{
unsigned long nentries, base, flags;
spin_lock_irqsave(&mm->context.lock, flags);
- base = (unsigned long) mm->context.tsb_block[MM_TSB_BASE].tsb;
- nentries = mm->context.tsb_block[MM_TSB_BASE].tsb_nentries;
- if (tlb_type == cheetah_plus || tlb_type == hypervisor)
- base = __pa(base);
- __flush_tsb_one_entry(base, vaddr, PAGE_SHIFT, nentries);
-
+ if (!huge) {
+ base = (unsigned long) mm->context.tsb_block[MM_TSB_BASE].tsb;
+ nentries = mm->context.tsb_block[MM_TSB_BASE].tsb_nentries;
+ if (tlb_type == cheetah_plus || tlb_type == hypervisor)
+ base = __pa(base);
+ __flush_tsb_one_entry(base, vaddr, PAGE_SHIFT, nentries);
+ }
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- if (mm->context.tsb_block[MM_TSB_HUGE].tsb) {
+ if (huge && mm->context.tsb_block[MM_TSB_HUGE].tsb) {
base = (unsigned long) mm->context.tsb_block[MM_TSB_HUGE].tsb;
nentries = mm->context.tsb_block[MM_TSB_HUGE].tsb_nentries;
if (tlb_type == cheetah_plus || tlb_type == hypervisor)
@@ -468,7 +487,7 @@ retry_tsb_alloc:
int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
{
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- unsigned long huge_pte_count;
+ unsigned long total_huge_pte_count;
#endif
unsigned int i;
@@ -477,12 +496,14 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
mm->context.sparc64_ctx_val = 0UL;
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- /* We reset it to zero because the fork() page copying
+ /* We reset them to zero because the fork() page copying
* will re-increment the counters as the parent PTEs are
* copied into the child address space.
*/
- huge_pte_count = mm->context.huge_pte_count;
- mm->context.huge_pte_count = 0;
+ total_huge_pte_count = mm->context.hugetlb_pte_count +
+ mm->context.thp_pte_count;
+ mm->context.hugetlb_pte_count = 0;
+ mm->context.thp_pte_count = 0;
#endif
/* copy_mm() copies over the parent's mm_struct before calling
@@ -498,8 +519,8 @@ int init_new_context(struct task_struct *tsk, struct mm_struct *mm)
tsb_grow(mm, MM_TSB_BASE, get_mm_rss(mm));
#if defined(CONFIG_HUGETLB_PAGE) || defined(CONFIG_TRANSPARENT_HUGEPAGE)
- if (unlikely(huge_pte_count))
- tsb_grow(mm, MM_TSB_HUGE, huge_pte_count);
+ if (unlikely(total_huge_pte_count))
+ tsb_grow(mm, MM_TSB_HUGE, total_huge_pte_count);
#endif
if (unlikely(!mm->context.tsb_block[MM_TSB_BASE].tsb))
diff --git a/kernel/arch/sparc/mm/ultra.S b/kernel/arch/sparc/mm/ultra.S
index b4f4733ab..5d2fd6cd3 100644
--- a/kernel/arch/sparc/mm/ultra.S
+++ b/kernel/arch/sparc/mm/ultra.S
@@ -30,7 +30,7 @@
.text
.align 32
.globl __flush_tlb_mm
-__flush_tlb_mm: /* 18 insns */
+__flush_tlb_mm: /* 19 insns */
/* %o0=(ctx & TAG_CONTEXT_BITS), %o1=SECONDARY_CONTEXT */
ldxa [%o1] ASI_DMMU, %g2
cmp %g2, %o0
@@ -81,7 +81,7 @@ __flush_tlb_page: /* 22 insns */
.align 32
.globl __flush_tlb_pending
-__flush_tlb_pending: /* 26 insns */
+__flush_tlb_pending: /* 27 insns */
/* %o0 = context, %o1 = nr, %o2 = vaddrs[] */
rdpr %pstate, %g7
sllx %o1, 3, %o1
@@ -113,12 +113,14 @@ __flush_tlb_pending: /* 26 insns */
.align 32
.globl __flush_tlb_kernel_range
-__flush_tlb_kernel_range: /* 16 insns */
+__flush_tlb_kernel_range: /* 31 insns */
/* %o0=start, %o1=end */
cmp %o0, %o1
be,pn %xcc, 2f
+ sub %o1, %o0, %o3
+ srlx %o3, 18, %o4
+ brnz,pn %o4, __spitfire_flush_tlb_kernel_range_slow
sethi %hi(PAGE_SIZE), %o4
- sub %o1, %o0, %o3
sub %o3, %o4, %o3
or %o0, 0x20, %o0 ! Nucleus
1: stxa %g0, [%o0 + %o3] ASI_DMMU_DEMAP
@@ -131,6 +133,41 @@ __flush_tlb_kernel_range: /* 16 insns */
retl
nop
nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+
+__spitfire_flush_tlb_kernel_range_slow:
+ mov 63 * 8, %o4
+1: ldxa [%o4] ASI_ITLB_DATA_ACCESS, %o3
+ andcc %o3, 0x40, %g0 /* _PAGE_L_4U */
+ bne,pn %xcc, 2f
+ mov TLB_TAG_ACCESS, %o3
+ stxa %g0, [%o3] ASI_IMMU
+ stxa %g0, [%o4] ASI_ITLB_DATA_ACCESS
+ membar #Sync
+2: ldxa [%o4] ASI_DTLB_DATA_ACCESS, %o3
+ andcc %o3, 0x40, %g0
+ bne,pn %xcc, 2f
+ mov TLB_TAG_ACCESS, %o3
+ stxa %g0, [%o3] ASI_DMMU
+ stxa %g0, [%o4] ASI_DTLB_DATA_ACCESS
+ membar #Sync
+2: sub %o4, 8, %o4
+ brgez,pt %o4, 1b
+ nop
+ retl
+ nop
__spitfire_flush_tlb_mm_slow:
rdpr %pstate, %g1
@@ -285,6 +322,40 @@ __cheetah_flush_tlb_pending: /* 27 insns */
retl
wrpr %g7, 0x0, %pstate
+__cheetah_flush_tlb_kernel_range: /* 31 insns */
+ /* %o0=start, %o1=end */
+ cmp %o0, %o1
+ be,pn %xcc, 2f
+ sub %o1, %o0, %o3
+ srlx %o3, 18, %o4
+ brnz,pn %o4, 3f
+ sethi %hi(PAGE_SIZE), %o4
+ sub %o3, %o4, %o3
+ or %o0, 0x20, %o0 ! Nucleus
+1: stxa %g0, [%o0 + %o3] ASI_DMMU_DEMAP
+ stxa %g0, [%o0 + %o3] ASI_IMMU_DEMAP
+ membar #Sync
+ brnz,pt %o3, 1b
+ sub %o3, %o4, %o3
+2: sethi %hi(KERNBASE), %o3
+ flush %o3
+ retl
+ nop
+3: mov 0x80, %o4
+ stxa %g0, [%o4] ASI_DMMU_DEMAP
+ membar #Sync
+ stxa %g0, [%o4] ASI_IMMU_DEMAP
+ membar #Sync
+ retl
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+
#ifdef DCACHE_ALIASING_POSSIBLE
__cheetah_flush_dcache_page: /* 11 insns */
sethi %hi(PAGE_OFFSET), %g1
@@ -309,19 +380,28 @@ __hypervisor_tlb_tl0_error:
ret
restore
-__hypervisor_flush_tlb_mm: /* 10 insns */
+__hypervisor_flush_tlb_mm: /* 19 insns */
mov %o0, %o2 /* ARG2: mmu context */
mov 0, %o0 /* ARG0: CPU lists unimplemented */
mov 0, %o1 /* ARG1: CPU lists unimplemented */
mov HV_MMU_ALL, %o3 /* ARG3: flags */
mov HV_FAST_MMU_DEMAP_CTX, %o5
ta HV_FAST_TRAP
- brnz,pn %o0, __hypervisor_tlb_tl0_error
+ brnz,pn %o0, 1f
mov HV_FAST_MMU_DEMAP_CTX, %o1
retl
nop
+1: sethi %hi(__hypervisor_tlb_tl0_error), %o5
+ jmpl %o5 + %lo(__hypervisor_tlb_tl0_error), %g0
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
-__hypervisor_flush_tlb_page: /* 11 insns */
+__hypervisor_flush_tlb_page: /* 22 insns */
/* %o0 = context, %o1 = vaddr */
mov %o0, %g2
mov %o1, %o0 /* ARG0: vaddr + IMMU-bit */
@@ -330,12 +410,23 @@ __hypervisor_flush_tlb_page: /* 11 insns */
srlx %o0, PAGE_SHIFT, %o0
sllx %o0, PAGE_SHIFT, %o0
ta HV_MMU_UNMAP_ADDR_TRAP
- brnz,pn %o0, __hypervisor_tlb_tl0_error
+ brnz,pn %o0, 1f
mov HV_MMU_UNMAP_ADDR_TRAP, %o1
retl
nop
+1: sethi %hi(__hypervisor_tlb_tl0_error), %o2
+ jmpl %o2 + %lo(__hypervisor_tlb_tl0_error), %g0
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
-__hypervisor_flush_tlb_pending: /* 16 insns */
+__hypervisor_flush_tlb_pending: /* 27 insns */
/* %o0 = context, %o1 = nr, %o2 = vaddrs[] */
sllx %o1, 3, %g1
mov %o2, %g2
@@ -347,31 +438,57 @@ __hypervisor_flush_tlb_pending: /* 16 insns */
srlx %o0, PAGE_SHIFT, %o0
sllx %o0, PAGE_SHIFT, %o0
ta HV_MMU_UNMAP_ADDR_TRAP
- brnz,pn %o0, __hypervisor_tlb_tl0_error
+ brnz,pn %o0, 1f
mov HV_MMU_UNMAP_ADDR_TRAP, %o1
brnz,pt %g1, 1b
nop
retl
nop
+1: sethi %hi(__hypervisor_tlb_tl0_error), %o2
+ jmpl %o2 + %lo(__hypervisor_tlb_tl0_error), %g0
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
-__hypervisor_flush_tlb_kernel_range: /* 16 insns */
+__hypervisor_flush_tlb_kernel_range: /* 31 insns */
/* %o0=start, %o1=end */
cmp %o0, %o1
be,pn %xcc, 2f
- sethi %hi(PAGE_SIZE), %g3
- mov %o0, %g1
- sub %o1, %g1, %g2
+ sub %o1, %o0, %g2
+ srlx %g2, 18, %g3
+ brnz,pn %g3, 4f
+ mov %o0, %g1
+ sethi %hi(PAGE_SIZE), %g3
sub %g2, %g3, %g2
1: add %g1, %g2, %o0 /* ARG0: virtual address */
mov 0, %o1 /* ARG1: mmu context */
mov HV_MMU_ALL, %o2 /* ARG2: flags */
ta HV_MMU_UNMAP_ADDR_TRAP
- brnz,pn %o0, __hypervisor_tlb_tl0_error
+ brnz,pn %o0, 3f
mov HV_MMU_UNMAP_ADDR_TRAP, %o1
brnz,pt %g2, 1b
sub %g2, %g3, %g2
2: retl
nop
+3: sethi %hi(__hypervisor_tlb_tl0_error), %o2
+ jmpl %o2 + %lo(__hypervisor_tlb_tl0_error), %g0
+ nop
+4: mov 0, %o0 /* ARG0: CPU lists unimplemented */
+ mov 0, %o1 /* ARG1: CPU lists unimplemented */
+ mov 0, %o2 /* ARG2: mmu context == nucleus */
+ mov HV_MMU_ALL, %o3 /* ARG3: flags */
+ mov HV_FAST_MMU_DEMAP_CTX, %o5
+ ta HV_FAST_TRAP
+ brnz,pn %o0, 3b
+ mov HV_FAST_MMU_DEMAP_CTX, %o1
+ retl
+ nop
#ifdef DCACHE_ALIASING_POSSIBLE
/* XXX Niagara and friends have an 8K cache, so no aliasing is
@@ -394,43 +511,6 @@ tlb_patch_one:
retl
nop
- .globl cheetah_patch_cachetlbops
-cheetah_patch_cachetlbops:
- save %sp, -128, %sp
-
- sethi %hi(__flush_tlb_mm), %o0
- or %o0, %lo(__flush_tlb_mm), %o0
- sethi %hi(__cheetah_flush_tlb_mm), %o1
- or %o1, %lo(__cheetah_flush_tlb_mm), %o1
- call tlb_patch_one
- mov 19, %o2
-
- sethi %hi(__flush_tlb_page), %o0
- or %o0, %lo(__flush_tlb_page), %o0
- sethi %hi(__cheetah_flush_tlb_page), %o1
- or %o1, %lo(__cheetah_flush_tlb_page), %o1
- call tlb_patch_one
- mov 22, %o2
-
- sethi %hi(__flush_tlb_pending), %o0
- or %o0, %lo(__flush_tlb_pending), %o0
- sethi %hi(__cheetah_flush_tlb_pending), %o1
- or %o1, %lo(__cheetah_flush_tlb_pending), %o1
- call tlb_patch_one
- mov 27, %o2
-
-#ifdef DCACHE_ALIASING_POSSIBLE
- sethi %hi(__flush_dcache_page), %o0
- or %o0, %lo(__flush_dcache_page), %o0
- sethi %hi(__cheetah_flush_dcache_page), %o1
- or %o1, %lo(__cheetah_flush_dcache_page), %o1
- call tlb_patch_one
- mov 11, %o2
-#endif /* DCACHE_ALIASING_POSSIBLE */
-
- ret
- restore
-
#ifdef CONFIG_SMP
/* These are all called by the slaves of a cross call, at
* trap level 1, with interrupts fully disabled.
@@ -447,7 +527,7 @@ cheetah_patch_cachetlbops:
*/
.align 32
.globl xcall_flush_tlb_mm
-xcall_flush_tlb_mm: /* 21 insns */
+xcall_flush_tlb_mm: /* 24 insns */
mov PRIMARY_CONTEXT, %g2
ldxa [%g2] ASI_DMMU, %g3
srlx %g3, CTX_PGSZ1_NUC_SHIFT, %g4
@@ -469,9 +549,12 @@ xcall_flush_tlb_mm: /* 21 insns */
nop
nop
nop
+ nop
+ nop
+ nop
.globl xcall_flush_tlb_page
-xcall_flush_tlb_page: /* 17 insns */
+xcall_flush_tlb_page: /* 20 insns */
/* %g5=context, %g1=vaddr */
mov PRIMARY_CONTEXT, %g4
ldxa [%g4] ASI_DMMU, %g2
@@ -490,15 +573,20 @@ xcall_flush_tlb_page: /* 17 insns */
retry
nop
nop
+ nop
+ nop
+ nop
.globl xcall_flush_tlb_kernel_range
-xcall_flush_tlb_kernel_range: /* 25 insns */
+xcall_flush_tlb_kernel_range: /* 44 insns */
sethi %hi(PAGE_SIZE - 1), %g2
or %g2, %lo(PAGE_SIZE - 1), %g2
andn %g1, %g2, %g1
andn %g7, %g2, %g7
sub %g7, %g1, %g3
- add %g2, 1, %g2
+ srlx %g3, 18, %g2
+ brnz,pn %g2, 2f
+ add %g2, 1, %g2
sub %g3, %g2, %g3
or %g1, 0x20, %g1 ! Nucleus
1: stxa %g0, [%g1 + %g3] ASI_DMMU_DEMAP
@@ -507,8 +595,25 @@ xcall_flush_tlb_kernel_range: /* 25 insns */
brnz,pt %g3, 1b
sub %g3, %g2, %g3
retry
- nop
- nop
+2: mov 63 * 8, %g1
+1: ldxa [%g1] ASI_ITLB_DATA_ACCESS, %g2
+ andcc %g2, 0x40, %g0 /* _PAGE_L_4U */
+ bne,pn %xcc, 2f
+ mov TLB_TAG_ACCESS, %g2
+ stxa %g0, [%g2] ASI_IMMU
+ stxa %g0, [%g1] ASI_ITLB_DATA_ACCESS
+ membar #Sync
+2: ldxa [%g1] ASI_DTLB_DATA_ACCESS, %g2
+ andcc %g2, 0x40, %g0
+ bne,pn %xcc, 2f
+ mov TLB_TAG_ACCESS, %g2
+ stxa %g0, [%g2] ASI_DMMU
+ stxa %g0, [%g1] ASI_DTLB_DATA_ACCESS
+ membar #Sync
+2: sub %g1, 8, %g1
+ brgez,pt %g1, 1b
+ nop
+ retry
nop
nop
nop
@@ -637,6 +742,52 @@ xcall_fetch_glob_pmu_n4:
retry
+__cheetah_xcall_flush_tlb_kernel_range: /* 44 insns */
+ sethi %hi(PAGE_SIZE - 1), %g2
+ or %g2, %lo(PAGE_SIZE - 1), %g2
+ andn %g1, %g2, %g1
+ andn %g7, %g2, %g7
+ sub %g7, %g1, %g3
+ srlx %g3, 18, %g2
+ brnz,pn %g2, 2f
+ add %g2, 1, %g2
+ sub %g3, %g2, %g3
+ or %g1, 0x20, %g1 ! Nucleus
+1: stxa %g0, [%g1 + %g3] ASI_DMMU_DEMAP
+ stxa %g0, [%g1 + %g3] ASI_IMMU_DEMAP
+ membar #Sync
+ brnz,pt %g3, 1b
+ sub %g3, %g2, %g3
+ retry
+2: mov 0x80, %g2
+ stxa %g0, [%g2] ASI_DMMU_DEMAP
+ membar #Sync
+ stxa %g0, [%g2] ASI_IMMU_DEMAP
+ membar #Sync
+ retry
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+ nop
+
#ifdef DCACHE_ALIASING_POSSIBLE
.align 32
.globl xcall_flush_dcache_page_cheetah
@@ -700,7 +851,7 @@ __hypervisor_tlb_xcall_error:
ba,a,pt %xcc, rtrap
.globl __hypervisor_xcall_flush_tlb_mm
-__hypervisor_xcall_flush_tlb_mm: /* 21 insns */
+__hypervisor_xcall_flush_tlb_mm: /* 24 insns */
/* %g5=ctx, g1,g2,g3,g4,g7=scratch, %g6=unusable */
mov %o0, %g2
mov %o1, %g3
@@ -714,7 +865,7 @@ __hypervisor_xcall_flush_tlb_mm: /* 21 insns */
mov HV_FAST_MMU_DEMAP_CTX, %o5
ta HV_FAST_TRAP
mov HV_FAST_MMU_DEMAP_CTX, %g6
- brnz,pn %o0, __hypervisor_tlb_xcall_error
+ brnz,pn %o0, 1f
mov %o0, %g5
mov %g2, %o0
mov %g3, %o1
@@ -723,9 +874,12 @@ __hypervisor_xcall_flush_tlb_mm: /* 21 insns */
mov %g7, %o5
membar #Sync
retry
+1: sethi %hi(__hypervisor_tlb_xcall_error), %g4
+ jmpl %g4 + %lo(__hypervisor_tlb_xcall_error), %g0
+ nop
.globl __hypervisor_xcall_flush_tlb_page
-__hypervisor_xcall_flush_tlb_page: /* 17 insns */
+__hypervisor_xcall_flush_tlb_page: /* 20 insns */
/* %g5=ctx, %g1=vaddr */
mov %o0, %g2
mov %o1, %g3
@@ -737,42 +891,64 @@ __hypervisor_xcall_flush_tlb_page: /* 17 insns */
sllx %o0, PAGE_SHIFT, %o0
ta HV_MMU_UNMAP_ADDR_TRAP
mov HV_MMU_UNMAP_ADDR_TRAP, %g6
- brnz,a,pn %o0, __hypervisor_tlb_xcall_error
+ brnz,a,pn %o0, 1f
mov %o0, %g5
mov %g2, %o0
mov %g3, %o1
mov %g4, %o2
membar #Sync
retry
+1: sethi %hi(__hypervisor_tlb_xcall_error), %g4
+ jmpl %g4 + %lo(__hypervisor_tlb_xcall_error), %g0
+ nop
.globl __hypervisor_xcall_flush_tlb_kernel_range
-__hypervisor_xcall_flush_tlb_kernel_range: /* 25 insns */
+__hypervisor_xcall_flush_tlb_kernel_range: /* 44 insns */
/* %g1=start, %g7=end, g2,g3,g4,g5,g6=scratch */
sethi %hi(PAGE_SIZE - 1), %g2
or %g2, %lo(PAGE_SIZE - 1), %g2
andn %g1, %g2, %g1
andn %g7, %g2, %g7
sub %g7, %g1, %g3
+ srlx %g3, 18, %g7
add %g2, 1, %g2
sub %g3, %g2, %g3
mov %o0, %g2
mov %o1, %g4
- mov %o2, %g7
+ brnz,pn %g7, 2f
+ mov %o2, %g7
1: add %g1, %g3, %o0 /* ARG0: virtual address */
mov 0, %o1 /* ARG1: mmu context */
mov HV_MMU_ALL, %o2 /* ARG2: flags */
ta HV_MMU_UNMAP_ADDR_TRAP
mov HV_MMU_UNMAP_ADDR_TRAP, %g6
- brnz,pn %o0, __hypervisor_tlb_xcall_error
+ brnz,pn %o0, 1f
mov %o0, %g5
sethi %hi(PAGE_SIZE), %o2
brnz,pt %g3, 1b
sub %g3, %o2, %g3
- mov %g2, %o0
+5: mov %g2, %o0
mov %g4, %o1
mov %g7, %o2
membar #Sync
retry
+1: sethi %hi(__hypervisor_tlb_xcall_error), %g4
+ jmpl %g4 + %lo(__hypervisor_tlb_xcall_error), %g0
+ nop
+2: mov %o3, %g1
+ mov %o5, %g3
+ mov 0, %o0 /* ARG0: CPU lists unimplemented */
+ mov 0, %o1 /* ARG1: CPU lists unimplemented */
+ mov 0, %o2 /* ARG2: mmu context == nucleus */
+ mov HV_MMU_ALL, %o3 /* ARG3: flags */
+ mov HV_FAST_MMU_DEMAP_CTX, %o5
+ ta HV_FAST_TRAP
+ mov %g1, %o3
+ brz,pt %o0, 5b
+ mov %g3, %o5
+ mov HV_FAST_MMU_DEMAP_CTX, %g6
+ ba,pt %xcc, 1b
+ clr %g5
/* These just get rescheduled to PIL vectors. */
.globl xcall_call_function
@@ -809,6 +985,58 @@ xcall_kgdb_capture:
#endif /* CONFIG_SMP */
+ .globl cheetah_patch_cachetlbops
+cheetah_patch_cachetlbops:
+ save %sp, -128, %sp
+
+ sethi %hi(__flush_tlb_mm), %o0
+ or %o0, %lo(__flush_tlb_mm), %o0
+ sethi %hi(__cheetah_flush_tlb_mm), %o1
+ or %o1, %lo(__cheetah_flush_tlb_mm), %o1
+ call tlb_patch_one
+ mov 19, %o2
+
+ sethi %hi(__flush_tlb_page), %o0
+ or %o0, %lo(__flush_tlb_page), %o0
+ sethi %hi(__cheetah_flush_tlb_page), %o1
+ or %o1, %lo(__cheetah_flush_tlb_page), %o1
+ call tlb_patch_one
+ mov 22, %o2
+
+ sethi %hi(__flush_tlb_pending), %o0
+ or %o0, %lo(__flush_tlb_pending), %o0
+ sethi %hi(__cheetah_flush_tlb_pending), %o1
+ or %o1, %lo(__cheetah_flush_tlb_pending), %o1
+ call tlb_patch_one
+ mov 27, %o2
+
+ sethi %hi(__flush_tlb_kernel_range), %o0
+ or %o0, %lo(__flush_tlb_kernel_range), %o0
+ sethi %hi(__cheetah_flush_tlb_kernel_range), %o1
+ or %o1, %lo(__cheetah_flush_tlb_kernel_range), %o1
+ call tlb_patch_one
+ mov 31, %o2
+
+#ifdef DCACHE_ALIASING_POSSIBLE
+ sethi %hi(__flush_dcache_page), %o0
+ or %o0, %lo(__flush_dcache_page), %o0
+ sethi %hi(__cheetah_flush_dcache_page), %o1
+ or %o1, %lo(__cheetah_flush_dcache_page), %o1
+ call tlb_patch_one
+ mov 11, %o2
+#endif /* DCACHE_ALIASING_POSSIBLE */
+
+#ifdef CONFIG_SMP
+ sethi %hi(xcall_flush_tlb_kernel_range), %o0
+ or %o0, %lo(xcall_flush_tlb_kernel_range), %o0
+ sethi %hi(__cheetah_xcall_flush_tlb_kernel_range), %o1
+ or %o1, %lo(__cheetah_xcall_flush_tlb_kernel_range), %o1
+ call tlb_patch_one
+ mov 44, %o2
+#endif /* CONFIG_SMP */
+
+ ret
+ restore
.globl hypervisor_patch_cachetlbops
hypervisor_patch_cachetlbops:
@@ -819,28 +1047,28 @@ hypervisor_patch_cachetlbops:
sethi %hi(__hypervisor_flush_tlb_mm), %o1
or %o1, %lo(__hypervisor_flush_tlb_mm), %o1
call tlb_patch_one
- mov 10, %o2
+ mov 19, %o2
sethi %hi(__flush_tlb_page), %o0
or %o0, %lo(__flush_tlb_page), %o0
sethi %hi(__hypervisor_flush_tlb_page), %o1
or %o1, %lo(__hypervisor_flush_tlb_page), %o1
call tlb_patch_one
- mov 11, %o2
+ mov 22, %o2
sethi %hi(__flush_tlb_pending), %o0
or %o0, %lo(__flush_tlb_pending), %o0
sethi %hi(__hypervisor_flush_tlb_pending), %o1
or %o1, %lo(__hypervisor_flush_tlb_pending), %o1
call tlb_patch_one
- mov 16, %o2
+ mov 27, %o2
sethi %hi(__flush_tlb_kernel_range), %o0
or %o0, %lo(__flush_tlb_kernel_range), %o0
sethi %hi(__hypervisor_flush_tlb_kernel_range), %o1
or %o1, %lo(__hypervisor_flush_tlb_kernel_range), %o1
call tlb_patch_one
- mov 16, %o2
+ mov 31, %o2
#ifdef DCACHE_ALIASING_POSSIBLE
sethi %hi(__flush_dcache_page), %o0
@@ -857,21 +1085,21 @@ hypervisor_patch_cachetlbops:
sethi %hi(__hypervisor_xcall_flush_tlb_mm), %o1
or %o1, %lo(__hypervisor_xcall_flush_tlb_mm), %o1
call tlb_patch_one
- mov 21, %o2
+ mov 24, %o2
sethi %hi(xcall_flush_tlb_page), %o0
or %o0, %lo(xcall_flush_tlb_page), %o0
sethi %hi(__hypervisor_xcall_flush_tlb_page), %o1
or %o1, %lo(__hypervisor_xcall_flush_tlb_page), %o1
call tlb_patch_one
- mov 17, %o2
+ mov 20, %o2
sethi %hi(xcall_flush_tlb_kernel_range), %o0
or %o0, %lo(xcall_flush_tlb_kernel_range), %o0
sethi %hi(__hypervisor_xcall_flush_tlb_kernel_range), %o1
or %o1, %lo(__hypervisor_xcall_flush_tlb_kernel_range), %o1
call tlb_patch_one
- mov 25, %o2
+ mov 44, %o2
#endif /* CONFIG_SMP */
ret