From: "Aneesh Kumar K.V" <aneesh.kumar@linux•vnet.ibm.com>
To: benh@kernel•crashing.org, paulus@samba•org, mpe@ellerman•id.au
Cc: linuxppc-dev@lists•ozlabs.org,
"Aneesh Kumar K.V" <aneesh.kumar@linux•vnet.ibm.com>
Subject: [PATCH V2 57/68] powerpc/mm: pte_frag abstraction
Date: Sat, 9 Apr 2016 11:43:53 +0530 [thread overview]
Message-ID: <1460182444-2468-58-git-send-email-aneesh.kumar@linux.vnet.ibm.com> (raw)
In-Reply-To: <1460182444-2468-1-git-send-email-aneesh.kumar@linux.vnet.ibm.com>
In this patch we make the number of pte fragments per level4 page table
page a variable. Radix level 4 table size is 256 bytes and hence we
can have 256 fragments per level4 page. We don't update the fragment
count in this patch. We need to do performance measurements to find the
right value for fragment count.
Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux•vnet.ibm.com>
---
arch/powerpc/include/asm/book3s/64/hash-4k.h | 2 ++
arch/powerpc/include/asm/book3s/64/hash-64k.h | 4 ++--
arch/powerpc/include/asm/book3s/64/pgtable.h | 6 ++++++
arch/powerpc/mm/hash_utils_64.c | 3 +++
arch/powerpc/mm/pgtable-radix.c | 5 +++++
arch/powerpc/mm/pgtable_64.c | 4 ++++
6 files changed, 22 insertions(+), 2 deletions(-)
diff --git a/arch/powerpc/include/asm/book3s/64/hash-4k.h b/arch/powerpc/include/asm/book3s/64/hash-4k.h
index dcb9d6e94a0c..99f360e0ecb0 100644
--- a/arch/powerpc/include/asm/book3s/64/hash-4k.h
+++ b/arch/powerpc/include/asm/book3s/64/hash-4k.h
@@ -28,6 +28,8 @@
#define H_PAGE_4K_PFN 0x0
#define H_PAGE_THP_HUGE 0x0
#define H_PAGE_COMBO 0x0
+#define H_PTE_FRAG_NR 0
+#define H_PTE_FRAG_SIZE_SHIFT 0
/*
* On all 4K setups, remap_4k_pfn() equates to remap_pfn_range()
*/
diff --git a/arch/powerpc/include/asm/book3s/64/hash-64k.h b/arch/powerpc/include/asm/book3s/64/hash-64k.h
index ad797a8cc615..f4a19914fdbe 100644
--- a/arch/powerpc/include/asm/book3s/64/hash-64k.h
+++ b/arch/powerpc/include/asm/book3s/64/hash-64k.h
@@ -29,12 +29,12 @@
/*
* we support 16 fragments per PTE page of 64K size.
*/
-#define PTE_FRAG_NR 16
+#define H_PTE_FRAG_NR 16
/*
* We use a 2K PTE page fragment and another 2K for storing
* real_pte_t hash index
*/
-#define PTE_FRAG_SIZE_SHIFT 12
+#define H_PTE_FRAG_SIZE_SHIFT 12
#define PTE_FRAG_SIZE (1UL << PTE_FRAG_SIZE_SHIFT)
#ifndef __ASSEMBLY__
diff --git a/arch/powerpc/include/asm/book3s/64/pgtable.h b/arch/powerpc/include/asm/book3s/64/pgtable.h
index 87519ad1c5dc..be8f46853cc9 100644
--- a/arch/powerpc/include/asm/book3s/64/pgtable.h
+++ b/arch/powerpc/include/asm/book3s/64/pgtable.h
@@ -177,6 +177,12 @@ extern unsigned long __pgd_val_bits;
#define PMD_VAL_BITS __pmd_val_bits
#define PUD_VAL_BITS __pud_val_bits
#define PGD_VAL_BITS __pgd_val_bits
+
+extern unsigned long __pte_frag_nr;
+#define PTE_FRAG_NR __pte_frag_nr
+extern unsigned long __pte_frag_size_shift;
+#define PTE_FRAG_SIZE_SHIFT __pte_frag_size_shift
+#define PTE_FRAG_SIZE (1UL << PTE_FRAG_SIZE_SHIFT)
/*
* Pgtable size used by swapper, init in asm code
*/
diff --git a/arch/powerpc/mm/hash_utils_64.c b/arch/powerpc/mm/hash_utils_64.c
index 599c4684e158..3356a48ca479 100644
--- a/arch/powerpc/mm/hash_utils_64.c
+++ b/arch/powerpc/mm/hash_utils_64.c
@@ -869,6 +869,9 @@ void __init hlearly_init_mmu(void)
/*
* initialize page table size
*/
+ __pte_frag_nr = H_PTE_FRAG_NR;
+ __pte_frag_size_shift = H_PTE_FRAG_SIZE_SHIFT;
+
__pte_index_size = H_PTE_INDEX_SIZE;
__pmd_index_size = H_PMD_INDEX_SIZE;
__pud_index_size = H_PUD_INDEX_SIZE;
diff --git a/arch/powerpc/mm/pgtable-radix.c b/arch/powerpc/mm/pgtable-radix.c
index 508453c0eac4..c59dfe706a0f 100644
--- a/arch/powerpc/mm/pgtable-radix.c
+++ b/arch/powerpc/mm/pgtable-radix.c
@@ -328,6 +328,11 @@ void __init rearly_init_mmu(void)
__vmalloc_end = R_VMALLOC_END;
vmemmap = (struct page *)R_VMEMMAP_BASE;
ioremap_bot = IOREMAP_BASE;
+ /*
+ * For now radix also use the same frag size
+ */
+ __pte_frag_nr = H_PTE_FRAG_NR;
+ __pte_frag_size_shift = H_PTE_FRAG_SIZE_SHIFT;
radix_init_page_sizes();
if (!firmware_has_feature(FW_FEATURE_LPAR))
diff --git a/arch/powerpc/mm/pgtable_64.c b/arch/powerpc/mm/pgtable_64.c
index 88b6cfc0a3e3..0e980204c4d5 100644
--- a/arch/powerpc/mm/pgtable_64.c
+++ b/arch/powerpc/mm/pgtable_64.c
@@ -107,6 +107,10 @@ unsigned long __vmalloc_end;
EXPORT_SYMBOL(__vmalloc_end);
struct page *vmemmap;
EXPORT_SYMBOL(vmemmap);
+unsigned long __pte_frag_nr;
+EXPORT_SYMBOL(__pte_frag_nr);
+unsigned long __pte_frag_size_shift;
+EXPORT_SYMBOL(__pte_frag_size_shift);
#endif
unsigned long ioremap_bot;
--
2.5.0
next prev parent reply other threads:[~2016-04-09 6:17 UTC|newest]
Thread overview: 139+ messages / expand[flat|nested] mbox.gz Atom feed top
2016-04-09 6:12 [PATCH V2 00/68] PowerISA 3.0 Radix page table support Aneesh Kumar K.V
2016-04-09 6:12 ` [PATCH V2 01/68] powerpc/cxl: Use REGION_ID instead of opencoding Aneesh Kumar K.V
2016-04-11 0:37 ` Andrew Donnellan
2016-04-13 2:42 ` Aneesh Kumar K.V
2016-04-20 3:03 ` Michael Ellerman
2016-04-20 7:53 ` Aneesh Kumar K.V
2016-04-09 6:12 ` [PATCH V2 02/68] powerpc/mm/nohash: Return correctly from flush_tlb_page Aneesh Kumar K.V
2016-04-20 4:08 ` [V2, " Michael Ellerman
2016-04-20 7:49 ` Aneesh Kumar K.V
2016-04-09 6:12 ` [PATCH V2 03/68] powerpc/mm/nohash: Update non SMP version of flush_tlb_page to handle hugetlb address Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 04/68] powerpc/mm: Use big endian page table for book3s 64 Aneesh Kumar K.V
2016-04-22 11:01 ` Michael Ellerman
2016-04-24 22:29 ` Aneesh Kumar K.V
2016-05-29 11:03 ` Anton Blanchard
2016-05-29 21:27 ` Benjamin Herrenschmidt
2016-05-29 23:08 ` Anton Blanchard
2016-05-30 3:42 ` Michael Ellerman
2016-05-30 5:31 ` Anton Blanchard
2016-05-30 8:42 ` Aneesh Kumar K.V
2016-05-30 11:00 ` Benjamin Herrenschmidt
2016-05-30 14:48 ` Aneesh Kumar K.V
2016-05-30 14:59 ` Segher Boessenkool
2016-04-09 6:13 ` [PATCH V2 05/68] powerpc/mm: use _PAGE_READ to indicate Read access Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 06/68] powerpc/mm/subpage: Clear RWX bit to indicate no access Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 07/68] powerpc/mm: Use pte_user instead of opencoding Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 08/68] powerpc/mm: Replace _PAGE_USER with _PAGE_PRIVILEGED Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 09/68] powerpc/mm: Remove RPN_SHIFT and RPN_SIZE Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 10/68] powerpc/mm: Update _PAGE_KERNEL_RO Aneesh Kumar K.V
2016-11-20 0:43 ` [V2,10/68] " Geoff Levand
2016-11-20 18:03 ` Aneesh Kumar K.V
2016-11-21 0:33 ` Geoff Levand
2016-11-23 10:41 ` Aneesh Kumar K.V
2016-11-24 4:04 ` Geoff Levand
2016-04-09 6:13 ` [PATCH V2 11/68] powerpc/mm: Use helper for finding pte bits mapping I/O area Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 12/68] powerpc/mm: Drop WIMG in favour of new constants Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 13/68] powerpc/mm: Use generic version of pmdp_clear_flush_young Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 14/68] powerpc/mm: Use generic version of ptep_clear_flush_young Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 15/68] powerpc/mm: Move common data structure between radix and hash to book3s 64 generic headers Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 16/68] powerpc/mm/power9: Add partition table format Aneesh Kumar K.V
2016-04-11 0:59 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 17/68] powerpc/mm/hash: Add support for POWER9 hash Aneesh Kumar K.V
2016-04-11 4:55 ` Balbir Singh
2016-04-16 19:06 ` Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 18/68] powerpc/mm: Move hash and no hash code to separate files Aneesh Kumar K.V
2016-04-11 5:14 ` Balbir Singh
2016-04-17 10:20 ` Aneesh Kumar K.V
2016-04-20 6:17 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 19/68] powerpc/mm/book3s: Rename hash specific PTE bits to carry H_ prefix Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 20/68] powerpc/mm: Handle _PTE_NONE_MASK Aneesh Kumar K.V
2016-04-11 6:09 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 21/68] powerpc/mm: Move common pte bits and accessors to book3s/64/pgtable.h Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 22/68] powerpc/mm: Move pte accessors that operate on common pte bits to pgtable.h Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 23/68] powerpc/mm: Make page table size a variable Aneesh Kumar K.V
2016-04-12 1:49 ` Balbir Singh
2016-04-17 10:27 ` Aneesh Kumar K.V
2016-04-20 6:21 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 24/68] powerpc/mm: Move page table index and and vaddr to pgtable.h Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 25/68] powerpc/mm: Move pte related function together Aneesh Kumar K.V
2016-04-14 6:50 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 26/68] powerpc/mm/radix: Add radix pte defines Aneesh Kumar K.V
2016-04-21 4:12 ` Balbir Singh
2016-04-23 8:30 ` Benjamin Herrenschmidt
2016-04-26 1:40 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 27/68] powerpc/mm/radix: Dummy radix_enabled() Aneesh Kumar K.V
2016-04-21 4:27 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 28/68] powerpc/mm: Add radix callbacks to pte accessors Aneesh Kumar K.V
2016-04-21 4:30 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 29/68] powerpc/mm: Move hugetlb and THP related pmd accessors to pgtable.h Aneesh Kumar K.V
2016-04-21 9:53 ` Balbir Singh
2016-04-21 9:59 ` Michael Ellerman
2016-04-21 11:42 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 30/68] powerpc/mm/radix: Add radix callback for pmd accessors Aneesh Kumar K.V
2016-04-21 11:39 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 31/68] powerpc/mm: Abstraction for early init routines Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 32/68] powerpc/mm/radix: Add radix callback " Aneesh Kumar K.V
2016-04-21 12:22 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 33/68] powerpc/mm: Abstraction for vmemmap and map_kernel_page Aneesh Kumar K.V
2016-04-21 12:59 ` Balbir Singh
2016-04-28 6:17 ` Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 34/68] powerpc/mm/radix: Add radix callback for vmemmap and map_kernel page Aneesh Kumar K.V
2016-04-21 13:46 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 35/68] powerpc/mm: Abstraction for switch_mmu_context Aneesh Kumar K.V
2016-04-21 14:12 ` Balbir Singh
2016-04-28 6:13 ` Aneesh Kumar K.V
2016-04-28 6:13 ` Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 36/68] powerpc/mm/radix: Add mmu context handling callback for radix Aneesh Kumar K.V
2016-04-22 6:19 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 37/68] powerpc/mm: Rename mmu_context_hash64.c to mmu_context_book3s64.c Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 38/68] powerpc/mm: Hash linux abstraction for tlbflush routines Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 39/68] powerpc/mm/radix: Add " Aneesh Kumar K.V
2016-04-22 7:20 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 40/68] powerpc/mm/radix: Add MMU_FTR_RADIX Aneesh Kumar K.V
2016-04-22 7:23 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 41/68] powerpc/mm/radix: Use STD_MMU_64 to properly isolate hash related code Aneesh Kumar K.V
2016-04-22 7:32 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 42/68] powerpc/mm/radix: Isolate hash table function from pseries guest code Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 43/68] powerpc/mm/radix: Add checks in slice code to catch radix usage Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 44/68] powerpc/mm/radix: Limit paca allocation in radix Aneesh Kumar K.V
2016-04-22 8:07 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 45/68] powerpc/mm/radix: Pick the address layout for radix config Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 46/68] powerpc/mm/radix: Update secondary PTCR Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 47/68] powerpc/mm: Make a copy of pgalloc.h for 32 and 64 book3s Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 48/68] powerpc/mm: Copy pgalloc (part 2) Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 49/68] powerpc/mm: Revert changes made to nohash pgalloc-64.h Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 50/68] powerpc/mm: Simplify the code dropping 4 level table #ifdef Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 51/68] powerpc/mm: Rename function to indicate we are allocating fragments Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 52/68] powerpc/mm: make 4k and 64k use pte_t for pgtable_t Aneesh Kumar K.V
2016-04-26 2:58 ` Balbir Singh
2016-04-28 6:03 ` Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 53/68] powerpc/mm: Add radix pgalloc details Aneesh Kumar K.V
2016-04-26 3:05 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 54/68] powerpc/mm: Update pte filter for radix Aneesh Kumar K.V
2016-04-26 3:06 ` Balbir Singh
2016-04-09 6:13 ` [PATCH V2 55/68] powerpc/mm: VMALLOC abstraction Aneesh Kumar K.V
2016-04-22 6:52 ` Michael Neuling
2016-04-23 3:29 ` Aneesh Kumar K.V
2016-04-26 6:20 ` Balbir Singh
2016-04-26 4:47 ` Balbir Singh
2016-04-28 6:09 ` Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 56/68] powerpc/radix: update mmu cache Aneesh Kumar K.V
2016-04-26 6:23 ` Balbir Singh
2016-04-09 6:13 ` Aneesh Kumar K.V [this message]
2016-04-09 6:13 ` [PATCH V2 58/68] powerpc/mm: Fix vma_mmu_pagesize for radix Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 59/68] powerpc/mm: Add radix support for hugetlb Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 60/68] powerpc/mm/radix: Make sure swapper pgdir is properly aligned Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 61/68] powerpc/mm/radix: Add hugetlb support 4K page size Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 62/68] powerpc/mm: Drop PTE_ATOMIC_UPDATES from pmd_hugepage_update Aneesh Kumar K.V
2016-04-09 6:13 ` [PATCH V2 63/68] powerpc/mm: THP is only available on hash64 as of now Aneesh Kumar K.V
2016-04-09 6:14 ` [PATCH V2 64/68] powerpc/mm/thp: Abstraction for THP functions Aneesh Kumar K.V
2016-04-09 6:14 ` [PATCH V2 65/68] powerpc/mm/radix: Add radix THP callbacks Aneesh Kumar K.V
2016-04-09 6:14 ` [PATCH V2 66/68] powerpc/mm/radix: Add THP support for 4k linux page size Aneesh Kumar K.V
2016-04-28 4:56 ` [V2, " Michael Ellerman
2016-04-28 6:28 ` Aneesh Kumar K.V
2016-04-09 6:14 ` [PATCH V2 67/68] powerpc/mm/radix: Cputable update for radix Aneesh Kumar K.V
2016-04-28 14:10 ` [V2,67/68] " Michael Ellerman
2016-04-09 6:14 ` [PATCH V2 68/68] powerpc/mm/radix: Use firmware feature to disable radix Aneesh Kumar K.V
2016-04-20 2:59 ` [V2, " Michael Ellerman
2016-04-20 8:21 ` Aneesh Kumar K.V
2016-04-20 11:25 ` Michael Neuling
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1460182444-2468-58-git-send-email-aneesh.kumar@linux.vnet.ibm.com \
--to=aneesh.kumar@linux$(echo .)vnet.ibm.com \
--cc=benh@kernel$(echo .)crashing.org \
--cc=linuxppc-dev@lists$(echo .)ozlabs.org \
--cc=mpe@ellerman$(echo .)id.au \
--cc=paulus@samba$(echo .)org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox