linux_dsm_epyc7002/arch/powerpc/mm/hugetlbpage-book3e.c

206 lines
4.2 KiB
C
Raw Normal View History

/*
* PPC Huge TLB Page Support for Book3E MMU
*
* Copyright (C) 2009 David Gibson, IBM Corporation.
* Copyright (C) 2011 Becky Bruce, Freescale Semiconductor
*
*/
#include <linux/mm.h>
#include <linux/hugetlb.h>
#include <asm/mmu.h>
#ifdef CONFIG_PPC_FSL_BOOK3E
#ifdef CONFIG_PPC64
static inline int tlb1_next(void)
{
struct paca_struct *paca = get_paca();
struct tlb_core_data *tcd;
int this, next;
tcd = paca->tcd_ptr;
this = tcd->esel_next;
next = this + 1;
if (next >= tcd->esel_max)
next = tcd->esel_first;
tcd->esel_next = next;
return this;
}
#else
static inline int tlb1_next(void)
{
int index, ncams;
ncams = mfspr(SPRN_TLB1CFG) & TLBnCFG_N_ENTRY;
powerpc: Replace __get_cpu_var uses This still has not been merged and now powerpc is the only arch that does not have this change. Sorry about missing linuxppc-dev before. V2->V2 - Fix up to work against 3.18-rc1 __get_cpu_var() is used for multiple purposes in the kernel source. One of them is address calculation via the form &__get_cpu_var(x). This calculates the address for the instance of the percpu variable of the current processor based on an offset. Other use cases are for storing and retrieving data from the current processors percpu area. __get_cpu_var() can be used as an lvalue when writing data or on the right side of an assignment. __get_cpu_var() is defined as : __get_cpu_var() always only does an address determination. However, store and retrieve operations could use a segment prefix (or global register on other platforms) to avoid the address calculation. this_cpu_write() and this_cpu_read() can directly take an offset into a percpu area and use optimized assembly code to read and write per cpu variables. This patch converts __get_cpu_var into either an explicit address calculation using this_cpu_ptr() or into a use of this_cpu operations that use the offset. Thereby address calculations are avoided and less registers are used when code is generated. At the end of the patch set all uses of __get_cpu_var have been removed so the macro is removed too. The patch set includes passes over all arches as well. Once these operations are used throughout then specialized macros can be defined in non -x86 arches as well in order to optimize per cpu access by f.e. using a global register that may be set to the per cpu base. Transformations done to __get_cpu_var() 1. Determine the address of the percpu instance of the current processor. DEFINE_PER_CPU(int, y); int *x = &__get_cpu_var(y); Converts to int *x = this_cpu_ptr(&y); 2. Same as #1 but this time an array structure is involved. DEFINE_PER_CPU(int, y[20]); int *x = __get_cpu_var(y); Converts to int *x = this_cpu_ptr(y); 3. Retrieve the content of the current processors instance of a per cpu variable. DEFINE_PER_CPU(int, y); int x = __get_cpu_var(y) Converts to int x = __this_cpu_read(y); 4. Retrieve the content of a percpu struct DEFINE_PER_CPU(struct mystruct, y); struct mystruct x = __get_cpu_var(y); Converts to memcpy(&x, this_cpu_ptr(&y), sizeof(x)); 5. Assignment to a per cpu variable DEFINE_PER_CPU(int, y) __get_cpu_var(y) = x; Converts to __this_cpu_write(y, x); 6. Increment/Decrement etc of a per cpu variable DEFINE_PER_CPU(int, y); __get_cpu_var(y)++ Converts to __this_cpu_inc(y) Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> CC: Paul Mackerras <paulus@samba.org> Signed-off-by: Christoph Lameter <cl@linux.com> [mpe: Fix build errors caused by set/or_softirq_pending(), and rework assignment in __set_breakpoint() to use memcpy().] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2014-10-22 03:23:25 +07:00
index = this_cpu_read(next_tlbcam_idx);
/* Just round-robin the entries and wrap when we hit the end */
if (unlikely(index == ncams - 1))
powerpc: Replace __get_cpu_var uses This still has not been merged and now powerpc is the only arch that does not have this change. Sorry about missing linuxppc-dev before. V2->V2 - Fix up to work against 3.18-rc1 __get_cpu_var() is used for multiple purposes in the kernel source. One of them is address calculation via the form &__get_cpu_var(x). This calculates the address for the instance of the percpu variable of the current processor based on an offset. Other use cases are for storing and retrieving data from the current processors percpu area. __get_cpu_var() can be used as an lvalue when writing data or on the right side of an assignment. __get_cpu_var() is defined as : __get_cpu_var() always only does an address determination. However, store and retrieve operations could use a segment prefix (or global register on other platforms) to avoid the address calculation. this_cpu_write() and this_cpu_read() can directly take an offset into a percpu area and use optimized assembly code to read and write per cpu variables. This patch converts __get_cpu_var into either an explicit address calculation using this_cpu_ptr() or into a use of this_cpu operations that use the offset. Thereby address calculations are avoided and less registers are used when code is generated. At the end of the patch set all uses of __get_cpu_var have been removed so the macro is removed too. The patch set includes passes over all arches as well. Once these operations are used throughout then specialized macros can be defined in non -x86 arches as well in order to optimize per cpu access by f.e. using a global register that may be set to the per cpu base. Transformations done to __get_cpu_var() 1. Determine the address of the percpu instance of the current processor. DEFINE_PER_CPU(int, y); int *x = &__get_cpu_var(y); Converts to int *x = this_cpu_ptr(&y); 2. Same as #1 but this time an array structure is involved. DEFINE_PER_CPU(int, y[20]); int *x = __get_cpu_var(y); Converts to int *x = this_cpu_ptr(y); 3. Retrieve the content of the current processors instance of a per cpu variable. DEFINE_PER_CPU(int, y); int x = __get_cpu_var(y) Converts to int x = __this_cpu_read(y); 4. Retrieve the content of a percpu struct DEFINE_PER_CPU(struct mystruct, y); struct mystruct x = __get_cpu_var(y); Converts to memcpy(&x, this_cpu_ptr(&y), sizeof(x)); 5. Assignment to a per cpu variable DEFINE_PER_CPU(int, y) __get_cpu_var(y) = x; Converts to __this_cpu_write(y, x); 6. Increment/Decrement etc of a per cpu variable DEFINE_PER_CPU(int, y); __get_cpu_var(y)++ Converts to __this_cpu_inc(y) Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> CC: Paul Mackerras <paulus@samba.org> Signed-off-by: Christoph Lameter <cl@linux.com> [mpe: Fix build errors caused by set/or_softirq_pending(), and rework assignment in __set_breakpoint() to use memcpy().] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2014-10-22 03:23:25 +07:00
__this_cpu_write(next_tlbcam_idx, tlbcam_index);
else
powerpc: Replace __get_cpu_var uses This still has not been merged and now powerpc is the only arch that does not have this change. Sorry about missing linuxppc-dev before. V2->V2 - Fix up to work against 3.18-rc1 __get_cpu_var() is used for multiple purposes in the kernel source. One of them is address calculation via the form &__get_cpu_var(x). This calculates the address for the instance of the percpu variable of the current processor based on an offset. Other use cases are for storing and retrieving data from the current processors percpu area. __get_cpu_var() can be used as an lvalue when writing data or on the right side of an assignment. __get_cpu_var() is defined as : __get_cpu_var() always only does an address determination. However, store and retrieve operations could use a segment prefix (or global register on other platforms) to avoid the address calculation. this_cpu_write() and this_cpu_read() can directly take an offset into a percpu area and use optimized assembly code to read and write per cpu variables. This patch converts __get_cpu_var into either an explicit address calculation using this_cpu_ptr() or into a use of this_cpu operations that use the offset. Thereby address calculations are avoided and less registers are used when code is generated. At the end of the patch set all uses of __get_cpu_var have been removed so the macro is removed too. The patch set includes passes over all arches as well. Once these operations are used throughout then specialized macros can be defined in non -x86 arches as well in order to optimize per cpu access by f.e. using a global register that may be set to the per cpu base. Transformations done to __get_cpu_var() 1. Determine the address of the percpu instance of the current processor. DEFINE_PER_CPU(int, y); int *x = &__get_cpu_var(y); Converts to int *x = this_cpu_ptr(&y); 2. Same as #1 but this time an array structure is involved. DEFINE_PER_CPU(int, y[20]); int *x = __get_cpu_var(y); Converts to int *x = this_cpu_ptr(y); 3. Retrieve the content of the current processors instance of a per cpu variable. DEFINE_PER_CPU(int, y); int x = __get_cpu_var(y) Converts to int x = __this_cpu_read(y); 4. Retrieve the content of a percpu struct DEFINE_PER_CPU(struct mystruct, y); struct mystruct x = __get_cpu_var(y); Converts to memcpy(&x, this_cpu_ptr(&y), sizeof(x)); 5. Assignment to a per cpu variable DEFINE_PER_CPU(int, y) __get_cpu_var(y) = x; Converts to __this_cpu_write(y, x); 6. Increment/Decrement etc of a per cpu variable DEFINE_PER_CPU(int, y); __get_cpu_var(y)++ Converts to __this_cpu_inc(y) Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org> CC: Paul Mackerras <paulus@samba.org> Signed-off-by: Christoph Lameter <cl@linux.com> [mpe: Fix build errors caused by set/or_softirq_pending(), and rework assignment in __set_breakpoint() to use memcpy().] Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
2014-10-22 03:23:25 +07:00
__this_cpu_inc(next_tlbcam_idx);
return index;
}
#endif /* !PPC64 */
#endif /* FSL */
static inline int mmu_get_tsize(int psize)
{
return mmu_psize_defs[psize].enc;
}
#if defined(CONFIG_PPC_FSL_BOOK3E) && defined(CONFIG_PPC64)
#include <asm/paca.h>
static inline void book3e_tlb_lock(void)
{
struct paca_struct *paca = get_paca();
unsigned long tmp;
int token = smp_processor_id() + 1;
/*
* Besides being unnecessary in the absence of SMT, this
* check prevents trying to do lbarx/stbcx. on e5500 which
* doesn't implement either feature.
*/
if (!cpu_has_feature(CPU_FTR_SMT))
return;
asm volatile("1: lbarx %0, 0, %1;"
"cmpwi %0, 0;"
"bne 2f;"
"stbcx. %2, 0, %1;"
"bne 1b;"
"b 3f;"
"2: lbzx %0, 0, %1;"
"cmpwi %0, 0;"
"bne 2b;"
"b 1b;"
"3:"
: "=&r" (tmp)
: "r" (&paca->tcd_ptr->lock), "r" (token)
: "memory");
}
static inline void book3e_tlb_unlock(void)
{
struct paca_struct *paca = get_paca();
if (!cpu_has_feature(CPU_FTR_SMT))
return;
isync();
paca->tcd_ptr->lock = 0;
}
#else
static inline void book3e_tlb_lock(void)
{
}
static inline void book3e_tlb_unlock(void)
{
}
#endif
static inline int book3e_tlb_exists(unsigned long ea, unsigned long pid)
{
int found = 0;
mtspr(SPRN_MAS6, pid << 16);
if (mmu_has_feature(MMU_FTR_USE_TLBRSRV)) {
asm volatile(
"li %0,0\n"
"tlbsx. 0,%1\n"
"bne 1f\n"
"li %0,1\n"
"1:\n"
: "=&r"(found) : "r"(ea));
} else {
asm volatile(
"tlbsx 0,%1\n"
"mfspr %0,0x271\n"
"srwi %0,%0,31\n"
: "=&r"(found) : "r"(ea));
}
return found;
}
void book3e_hugetlb_preload(struct vm_area_struct *vma, unsigned long ea,
pte_t pte)
{
unsigned long mas1, mas2;
u64 mas7_3;
unsigned long psize, tsize, shift;
unsigned long flags;
struct mm_struct *mm;
#ifdef CONFIG_PPC_FSL_BOOK3E
int index;
#endif
if (unlikely(is_kernel_addr(ea)))
return;
mm = vma->vm_mm;
psize = vma_mmu_pagesize(vma);
shift = __ilog2(psize);
tsize = shift - 10;
/*
* We can't be interrupted while we're setting up the MAS
* regusters or after we've confirmed that no tlb exists.
*/
local_irq_save(flags);
book3e_tlb_lock();
if (unlikely(book3e_tlb_exists(ea, mm->context.id))) {
book3e_tlb_unlock();
local_irq_restore(flags);
return;
}
#ifdef CONFIG_PPC_FSL_BOOK3E
/* We have to use the CAM(TLB1) on FSL parts for hugepages */
index = tlb1_next();
mtspr(SPRN_MAS0, MAS0_ESEL(index) | MAS0_TLBSEL(1));
#endif
mas1 = MAS1_VALID | MAS1_TID(mm->context.id) | MAS1_TSIZE(tsize);
mas2 = ea & ~((1UL << shift) - 1);
mas2 |= (pte_val(pte) >> PTE_WIMGE_SHIFT) & MAS2_WIMGE_MASK;
mas7_3 = (u64)pte_pfn(pte) << PAGE_SHIFT;
mas7_3 |= (pte_val(pte) >> PTE_BAP_SHIFT) & MAS3_BAP_MASK;
if (!pte_dirty(pte))
mas7_3 &= ~(MAS3_SW|MAS3_UW);
mtspr(SPRN_MAS1, mas1);
mtspr(SPRN_MAS2, mas2);
if (mmu_has_feature(MMU_FTR_USE_PAIRED_MAS)) {
mtspr(SPRN_MAS7_MAS3, mas7_3);
} else {
if (mmu_has_feature(MMU_FTR_BIG_PHYS))
mtspr(SPRN_MAS7, upper_32_bits(mas7_3));
mtspr(SPRN_MAS3, lower_32_bits(mas7_3));
}
asm volatile ("tlbwe");
book3e_tlb_unlock();
local_irq_restore(flags);
}
void flush_hugetlb_page(struct vm_area_struct *vma, unsigned long vmaddr)
{
struct hstate *hstate = hstate_file(vma->vm_file);
unsigned long tsize = huge_page_shift(hstate) - 10;
__flush_tlb_page(vma->vm_mm, vmaddr, tsize, 0);
}