Linux-Fsdevel Archive on lore.kernel.org
 help / Atom feed
* [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
@ 2019-01-17 18:39 Alexandre Ghiti
  2019-01-17 18:39 ` Alexandre Ghiti
                   ` (5 more replies)
  0 siblings, 6 replies; 10+ messages in thread
From: Alexandre Ghiti @ 2019-01-17 18:39 UTC (permalink / raw)
  To: Catalin Marinas, Will Deacon, Benjamin Herrenschmidt,
	Paul Mackerras, Michael Ellerman, Martin Schwidefsky,
	Heiko Carstens, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
	H . Peter Anvin, x86, Alexander Viro, Mike Kravetz,
	linux-arm-kernel, linux-kernel, linuxppc-dev, linux-s390,
	linux-fsdevel, linux-mm
  Cc: hch, linux-riscv, Alexandre Ghiti

From: Alexandre Ghiti <alex@ghiti.fr>

On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
that support gigantic pages, boottime reserved gigantic pages can not be
freed at all. This patchs simply enables the possibility to hand back
those pages to memory allocator.

This commit then renames gigantic_page_supported and
ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
being false does not mean that the system cannot use gigantic pages: it
just means that runtime allocation of gigantic pages is not supported,
one can still allocate boottime gigantic pages if the architecture supports
it.

Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
---

- Compiled on all architectures
- Tested on riscv architecture

 arch/arm64/Kconfig                           |  2 +-
 arch/arm64/include/asm/hugetlb.h             |  7 +++--
 arch/powerpc/include/asm/book3s/64/hugetlb.h |  4 +--
 arch/powerpc/platforms/Kconfig.cputype       |  2 +-
 arch/s390/Kconfig                            |  2 +-
 arch/s390/include/asm/hugetlb.h              |  7 +++--
 arch/x86/Kconfig                             |  2 +-
 arch/x86/include/asm/hugetlb.h               |  7 +++--
 fs/Kconfig                                   |  2 +-
 include/linux/gfp.h                          |  2 +-
 mm/hugetlb.c                                 | 43 +++++++++++++++-------------
 mm/page_alloc.c                              |  4 +--
 12 files changed, 48 insertions(+), 36 deletions(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index a4168d366127..18239cbd7fcd 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -18,7 +18,7 @@ config ARM64
 	select ARCH_HAS_FAST_MULTIPLIER
 	select ARCH_HAS_FORTIFY_SOURCE
 	select ARCH_HAS_GCOV_PROFILE_ALL
-	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
+	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
 	select ARCH_HAS_KCOV
 	select ARCH_HAS_MEMBARRIER_SYNC_CORE
 	select ARCH_HAS_PTE_SPECIAL
diff --git a/arch/arm64/include/asm/hugetlb.h b/arch/arm64/include/asm/hugetlb.h
index fb6609875455..797fc77eabcd 100644
--- a/arch/arm64/include/asm/hugetlb.h
+++ b/arch/arm64/include/asm/hugetlb.h
@@ -65,8 +65,11 @@ extern void set_huge_swap_pte_at(struct mm_struct *mm, unsigned long addr,
 
 #include <asm-generic/hugetlb.h>
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
-static inline bool gigantic_page_supported(void) { return true; }
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
+static inline bool gigantic_page_runtime_allocation_supported(void)
+{
+	return true;
+}
 #endif
 
 #endif /* __ASM_HUGETLB_H */
diff --git a/arch/powerpc/include/asm/book3s/64/hugetlb.h b/arch/powerpc/include/asm/book3s/64/hugetlb.h
index 5b0177733994..7711f0e2c7e5 100644
--- a/arch/powerpc/include/asm/book3s/64/hugetlb.h
+++ b/arch/powerpc/include/asm/book3s/64/hugetlb.h
@@ -32,8 +32,8 @@ static inline int hstate_get_psize(struct hstate *hstate)
 	}
 }
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
-static inline bool gigantic_page_supported(void)
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
+static inline bool gigantic_page_runtime_allocation_supported(void)
 {
 	return true;
 }
diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
index 8c7464c3f27f..779e06bac697 100644
--- a/arch/powerpc/platforms/Kconfig.cputype
+++ b/arch/powerpc/platforms/Kconfig.cputype
@@ -319,7 +319,7 @@ config ARCH_ENABLE_SPLIT_PMD_PTLOCK
 config PPC_RADIX_MMU
 	bool "Radix MMU Support"
 	depends on PPC_BOOK3S_64
-	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
+	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
 	default y
 	help
 	  Enable support for the Power ISA 3.0 Radix style MMU. Currently this
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index ed554b09eb3f..6776eef6a9ae 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -69,7 +69,7 @@ config S390
 	select ARCH_HAS_ELF_RANDOMIZE
 	select ARCH_HAS_FORTIFY_SOURCE
 	select ARCH_HAS_GCOV_PROFILE_ALL
-	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
+	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
 	select ARCH_HAS_KCOV
 	select ARCH_HAS_PTE_SPECIAL
 	select ARCH_HAS_SET_MEMORY
diff --git a/arch/s390/include/asm/hugetlb.h b/arch/s390/include/asm/hugetlb.h
index 2d1afa58a4b6..57c952f5388e 100644
--- a/arch/s390/include/asm/hugetlb.h
+++ b/arch/s390/include/asm/hugetlb.h
@@ -116,7 +116,10 @@ static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot)
 	return pte_modify(pte, newprot);
 }
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
-static inline bool gigantic_page_supported(void) { return true; }
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
+static inline bool gigantic_page_runtime_allocation_supported(void)
+{
+	return true;
+}
 #endif
 #endif /* _ASM_S390_HUGETLB_H */
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 6185d4f33296..a88f5a4311c9 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -23,7 +23,7 @@ config X86_64
 	def_bool y
 	depends on 64BIT
 	# Options that are inherently 64-bit kernel only:
-	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
+	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
 	select ARCH_SUPPORTS_INT128
 	select ARCH_USE_CMPXCHG_LOCKREF
 	select HAVE_ARCH_SOFT_DIRTY
diff --git a/arch/x86/include/asm/hugetlb.h b/arch/x86/include/asm/hugetlb.h
index 7469d321f072..5a5e7119ced4 100644
--- a/arch/x86/include/asm/hugetlb.h
+++ b/arch/x86/include/asm/hugetlb.h
@@ -17,8 +17,11 @@ static inline void arch_clear_hugepage_flags(struct page *page)
 {
 }
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
-static inline bool gigantic_page_supported(void) { return true; }
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
+static inline bool gigantic_page_runtime_allocation_supported(void)
+{
+	return true;
+}
 #endif
 
 #endif /* _ASM_X86_HUGETLB_H */
diff --git a/fs/Kconfig b/fs/Kconfig
index ac474a61be37..4192d1fde0f0 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -207,7 +207,7 @@ config HUGETLB_PAGE
 config MEMFD_CREATE
 	def_bool TMPFS || HUGETLBFS
 
-config ARCH_HAS_GIGANTIC_PAGE
+config ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
 	bool
 
 source "fs/configfs/Kconfig"
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 5f5e25fd6149..79ff86fabd42 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -589,8 +589,8 @@ static inline bool pm_suspended_storage(void)
 /* The below functions must be run on a range from a single zone. */
 extern int alloc_contig_range(unsigned long start, unsigned long end,
 			      unsigned migratetype, gfp_t gfp_mask);
-extern void free_contig_range(unsigned long pfn, unsigned nr_pages);
 #endif
+extern void free_contig_range(unsigned long pfn, unsigned int nr_pages);
 
 #ifdef CONFIG_CMA
 /* CMA stuff */
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 745088810965..9893ba26b3b8 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1035,7 +1035,6 @@ static int hstate_next_node_to_free(struct hstate *h, nodemask_t *nodes_allowed)
 		((node = hstate_next_node_to_free(hs, mask)) || 1);	\
 		nr_nodes--)
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
 static void destroy_compound_gigantic_page(struct page *page,
 					unsigned int order)
 {
@@ -1058,6 +1057,7 @@ static void free_gigantic_page(struct page *page, unsigned int order)
 	free_contig_range(page_to_pfn(page), 1 << order);
 }
 
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
 static int __alloc_gigantic_page(unsigned long start_pfn,
 				unsigned long nr_pages, gfp_t gfp_mask)
 {
@@ -1143,22 +1143,19 @@ static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
 static void prep_new_huge_page(struct hstate *h, struct page *page, int nid);
 static void prep_compound_gigantic_page(struct page *page, unsigned int order);
 
-#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE */
-static inline bool gigantic_page_supported(void) { return false; }
+#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION */
+static inline bool gigantic_page_runtime_allocation_supported(void)
+{
+	return false;
+}
 static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
 		int nid, nodemask_t *nodemask) { return NULL; }
-static inline void free_gigantic_page(struct page *page, unsigned int order) { }
-static inline void destroy_compound_gigantic_page(struct page *page,
-						unsigned int order) { }
 #endif
 
 static void update_and_free_page(struct hstate *h, struct page *page)
 {
 	int i;
 
-	if (hstate_is_gigantic(h) && !gigantic_page_supported())
-		return;
-
 	h->nr_huge_pages--;
 	h->nr_huge_pages_node[page_to_nid(page)]--;
 	for (i = 0; i < pages_per_huge_page(h); i++) {
@@ -2276,13 +2273,20 @@ static int adjust_pool_surplus(struct hstate *h, nodemask_t *nodes_allowed,
 }
 
 #define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages)
-static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
+static int set_max_huge_pages(struct hstate *h, unsigned long count,
 						nodemask_t *nodes_allowed)
 {
 	unsigned long min_count, ret;
 
-	if (hstate_is_gigantic(h) && !gigantic_page_supported())
-		return h->max_huge_pages;
+	if (hstate_is_gigantic(h) &&
+		!gigantic_page_runtime_allocation_supported()) {
+		spin_lock(&hugetlb_lock);
+		if (count > persistent_huge_pages(h)) {
+			spin_unlock(&hugetlb_lock);
+			return -EINVAL;
+		}
+		goto decrease_pool;
+	}
 
 	/*
 	 * Increase the pool size
@@ -2322,6 +2326,7 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
 			goto out;
 	}
 
+decrease_pool:
 	/*
 	 * Decrease the pool size
 	 * First return free pages to the buddy allocator (being careful
@@ -2350,9 +2355,10 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
 			break;
 	}
 out:
-	ret = persistent_huge_pages(h);
+	h->max_huge_pages = persistent_huge_pages(h);
 	spin_unlock(&hugetlb_lock);
-	return ret;
+
+	return 0;
 }
 
 #define HSTATE_ATTR_RO(_name) \
@@ -2404,11 +2410,6 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
 	int err;
 	NODEMASK_ALLOC(nodemask_t, nodes_allowed, GFP_KERNEL | __GFP_NORETRY);
 
-	if (hstate_is_gigantic(h) && !gigantic_page_supported()) {
-		err = -EINVAL;
-		goto out;
-	}
-
 	if (nid == NUMA_NO_NODE) {
 		/*
 		 * global hstate attribute
@@ -2428,7 +2429,9 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
 	} else
 		nodes_allowed = &node_states[N_MEMORY];
 
-	h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed);
+	err = set_max_huge_pages(h, count, nodes_allowed);
+	if (err)
+		goto out;
 
 	if (nodes_allowed != &node_states[N_MEMORY])
 		NODEMASK_FREE(nodes_allowed);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index cde5dac6229a..81b931db85a1 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -8241,8 +8241,9 @@ int alloc_contig_range(unsigned long start, unsigned long end,
 				pfn_max_align_up(end), migratetype);
 	return ret;
 }
+#endif
 
-void free_contig_range(unsigned long pfn, unsigned nr_pages)
+void free_contig_range(unsigned long pfn, unsigned int nr_pages)
 {
 	unsigned int count = 0;
 
@@ -8254,7 +8255,6 @@ void free_contig_range(unsigned long pfn, unsigned nr_pages)
 	}
 	WARN(count != 0, "%d pages are still in use!\n", count);
 }
-#endif
 
 #ifdef CONFIG_MEMORY_HOTPLUG
 /*
-- 
2.16.2

^ permalink raw reply	[flat|nested] 10+ messages in thread

* [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
  2019-01-17 18:39 [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration Alexandre Ghiti
@ 2019-01-17 18:39 ` Alexandre Ghiti
  2019-01-17 20:09 ` Mike Kravetz
                   ` (4 subsequent siblings)
  5 siblings, 0 replies; 10+ messages in thread
From: Alexandre Ghiti @ 2019-01-17 18:39 UTC (permalink / raw)
  To: Catalin Marinas, Will Deacon, Benjamin Herrenschmidt,
	Paul Mackerras, Michael Ellerman, Martin Schwidefsky,
	Heiko Carstens, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
	H . Peter Anvin, x86, Alexander Viro, Mike Kravetz,
	linux-arm-kernel, linux-kernel, linuxppc-dev, linux-s390,
	linux-fsdevel, linux-mm
  Cc: linux-riscv, hch, Alexandre Ghiti

From: Alexandre Ghiti <alex@ghiti.fr>

On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
that support gigantic pages, boottime reserved gigantic pages can not be
freed at all. This patchs simply enables the possibility to hand back
those pages to memory allocator.

This commit then renames gigantic_page_supported and
ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
being false does not mean that the system cannot use gigantic pages: it
just means that runtime allocation of gigantic pages is not supported,
one can still allocate boottime gigantic pages if the architecture supports
it.

Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
---

- Compiled on all architectures
- Tested on riscv architecture

 arch/arm64/Kconfig                           |  2 +-
 arch/arm64/include/asm/hugetlb.h             |  7 +++--
 arch/powerpc/include/asm/book3s/64/hugetlb.h |  4 +--
 arch/powerpc/platforms/Kconfig.cputype       |  2 +-
 arch/s390/Kconfig                            |  2 +-
 arch/s390/include/asm/hugetlb.h              |  7 +++--
 arch/x86/Kconfig                             |  2 +-
 arch/x86/include/asm/hugetlb.h               |  7 +++--
 fs/Kconfig                                   |  2 +-
 include/linux/gfp.h                          |  2 +-
 mm/hugetlb.c                                 | 43 +++++++++++++++-------------
 mm/page_alloc.c                              |  4 +--
 12 files changed, 48 insertions(+), 36 deletions(-)

diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
index a4168d366127..18239cbd7fcd 100644
--- a/arch/arm64/Kconfig
+++ b/arch/arm64/Kconfig
@@ -18,7 +18,7 @@ config ARM64
 	select ARCH_HAS_FAST_MULTIPLIER
 	select ARCH_HAS_FORTIFY_SOURCE
 	select ARCH_HAS_GCOV_PROFILE_ALL
-	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
+	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
 	select ARCH_HAS_KCOV
 	select ARCH_HAS_MEMBARRIER_SYNC_CORE
 	select ARCH_HAS_PTE_SPECIAL
diff --git a/arch/arm64/include/asm/hugetlb.h b/arch/arm64/include/asm/hugetlb.h
index fb6609875455..797fc77eabcd 100644
--- a/arch/arm64/include/asm/hugetlb.h
+++ b/arch/arm64/include/asm/hugetlb.h
@@ -65,8 +65,11 @@ extern void set_huge_swap_pte_at(struct mm_struct *mm, unsigned long addr,
 
 #include <asm-generic/hugetlb.h>
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
-static inline bool gigantic_page_supported(void) { return true; }
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
+static inline bool gigantic_page_runtime_allocation_supported(void)
+{
+	return true;
+}
 #endif
 
 #endif /* __ASM_HUGETLB_H */
diff --git a/arch/powerpc/include/asm/book3s/64/hugetlb.h b/arch/powerpc/include/asm/book3s/64/hugetlb.h
index 5b0177733994..7711f0e2c7e5 100644
--- a/arch/powerpc/include/asm/book3s/64/hugetlb.h
+++ b/arch/powerpc/include/asm/book3s/64/hugetlb.h
@@ -32,8 +32,8 @@ static inline int hstate_get_psize(struct hstate *hstate)
 	}
 }
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
-static inline bool gigantic_page_supported(void)
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
+static inline bool gigantic_page_runtime_allocation_supported(void)
 {
 	return true;
 }
diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
index 8c7464c3f27f..779e06bac697 100644
--- a/arch/powerpc/platforms/Kconfig.cputype
+++ b/arch/powerpc/platforms/Kconfig.cputype
@@ -319,7 +319,7 @@ config ARCH_ENABLE_SPLIT_PMD_PTLOCK
 config PPC_RADIX_MMU
 	bool "Radix MMU Support"
 	depends on PPC_BOOK3S_64
-	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
+	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
 	default y
 	help
 	  Enable support for the Power ISA 3.0 Radix style MMU. Currently this
diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
index ed554b09eb3f..6776eef6a9ae 100644
--- a/arch/s390/Kconfig
+++ b/arch/s390/Kconfig
@@ -69,7 +69,7 @@ config S390
 	select ARCH_HAS_ELF_RANDOMIZE
 	select ARCH_HAS_FORTIFY_SOURCE
 	select ARCH_HAS_GCOV_PROFILE_ALL
-	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
+	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
 	select ARCH_HAS_KCOV
 	select ARCH_HAS_PTE_SPECIAL
 	select ARCH_HAS_SET_MEMORY
diff --git a/arch/s390/include/asm/hugetlb.h b/arch/s390/include/asm/hugetlb.h
index 2d1afa58a4b6..57c952f5388e 100644
--- a/arch/s390/include/asm/hugetlb.h
+++ b/arch/s390/include/asm/hugetlb.h
@@ -116,7 +116,10 @@ static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot)
 	return pte_modify(pte, newprot);
 }
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
-static inline bool gigantic_page_supported(void) { return true; }
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
+static inline bool gigantic_page_runtime_allocation_supported(void)
+{
+	return true;
+}
 #endif
 #endif /* _ASM_S390_HUGETLB_H */
diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 6185d4f33296..a88f5a4311c9 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -23,7 +23,7 @@ config X86_64
 	def_bool y
 	depends on 64BIT
 	# Options that are inherently 64-bit kernel only:
-	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
+	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
 	select ARCH_SUPPORTS_INT128
 	select ARCH_USE_CMPXCHG_LOCKREF
 	select HAVE_ARCH_SOFT_DIRTY
diff --git a/arch/x86/include/asm/hugetlb.h b/arch/x86/include/asm/hugetlb.h
index 7469d321f072..5a5e7119ced4 100644
--- a/arch/x86/include/asm/hugetlb.h
+++ b/arch/x86/include/asm/hugetlb.h
@@ -17,8 +17,11 @@ static inline void arch_clear_hugepage_flags(struct page *page)
 {
 }
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
-static inline bool gigantic_page_supported(void) { return true; }
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
+static inline bool gigantic_page_runtime_allocation_supported(void)
+{
+	return true;
+}
 #endif
 
 #endif /* _ASM_X86_HUGETLB_H */
diff --git a/fs/Kconfig b/fs/Kconfig
index ac474a61be37..4192d1fde0f0 100644
--- a/fs/Kconfig
+++ b/fs/Kconfig
@@ -207,7 +207,7 @@ config HUGETLB_PAGE
 config MEMFD_CREATE
 	def_bool TMPFS || HUGETLBFS
 
-config ARCH_HAS_GIGANTIC_PAGE
+config ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
 	bool
 
 source "fs/configfs/Kconfig"
diff --git a/include/linux/gfp.h b/include/linux/gfp.h
index 5f5e25fd6149..79ff86fabd42 100644
--- a/include/linux/gfp.h
+++ b/include/linux/gfp.h
@@ -589,8 +589,8 @@ static inline bool pm_suspended_storage(void)
 /* The below functions must be run on a range from a single zone. */
 extern int alloc_contig_range(unsigned long start, unsigned long end,
 			      unsigned migratetype, gfp_t gfp_mask);
-extern void free_contig_range(unsigned long pfn, unsigned nr_pages);
 #endif
+extern void free_contig_range(unsigned long pfn, unsigned int nr_pages);
 
 #ifdef CONFIG_CMA
 /* CMA stuff */
diff --git a/mm/hugetlb.c b/mm/hugetlb.c
index 745088810965..9893ba26b3b8 100644
--- a/mm/hugetlb.c
+++ b/mm/hugetlb.c
@@ -1035,7 +1035,6 @@ static int hstate_next_node_to_free(struct hstate *h, nodemask_t *nodes_allowed)
 		((node = hstate_next_node_to_free(hs, mask)) || 1);	\
 		nr_nodes--)
 
-#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
 static void destroy_compound_gigantic_page(struct page *page,
 					unsigned int order)
 {
@@ -1058,6 +1057,7 @@ static void free_gigantic_page(struct page *page, unsigned int order)
 	free_contig_range(page_to_pfn(page), 1 << order);
 }
 
+#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
 static int __alloc_gigantic_page(unsigned long start_pfn,
 				unsigned long nr_pages, gfp_t gfp_mask)
 {
@@ -1143,22 +1143,19 @@ static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
 static void prep_new_huge_page(struct hstate *h, struct page *page, int nid);
 static void prep_compound_gigantic_page(struct page *page, unsigned int order);
 
-#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE */
-static inline bool gigantic_page_supported(void) { return false; }
+#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION */
+static inline bool gigantic_page_runtime_allocation_supported(void)
+{
+	return false;
+}
 static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
 		int nid, nodemask_t *nodemask) { return NULL; }
-static inline void free_gigantic_page(struct page *page, unsigned int order) { }
-static inline void destroy_compound_gigantic_page(struct page *page,
-						unsigned int order) { }
 #endif
 
 static void update_and_free_page(struct hstate *h, struct page *page)
 {
 	int i;
 
-	if (hstate_is_gigantic(h) && !gigantic_page_supported())
-		return;
-
 	h->nr_huge_pages--;
 	h->nr_huge_pages_node[page_to_nid(page)]--;
 	for (i = 0; i < pages_per_huge_page(h); i++) {
@@ -2276,13 +2273,20 @@ static int adjust_pool_surplus(struct hstate *h, nodemask_t *nodes_allowed,
 }
 
 #define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages)
-static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
+static int set_max_huge_pages(struct hstate *h, unsigned long count,
 						nodemask_t *nodes_allowed)
 {
 	unsigned long min_count, ret;
 
-	if (hstate_is_gigantic(h) && !gigantic_page_supported())
-		return h->max_huge_pages;
+	if (hstate_is_gigantic(h) &&
+		!gigantic_page_runtime_allocation_supported()) {
+		spin_lock(&hugetlb_lock);
+		if (count > persistent_huge_pages(h)) {
+			spin_unlock(&hugetlb_lock);
+			return -EINVAL;
+		}
+		goto decrease_pool;
+	}
 
 	/*
 	 * Increase the pool size
@@ -2322,6 +2326,7 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
 			goto out;
 	}
 
+decrease_pool:
 	/*
 	 * Decrease the pool size
 	 * First return free pages to the buddy allocator (being careful
@@ -2350,9 +2355,10 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
 			break;
 	}
 out:
-	ret = persistent_huge_pages(h);
+	h->max_huge_pages = persistent_huge_pages(h);
 	spin_unlock(&hugetlb_lock);
-	return ret;
+
+	return 0;
 }
 
 #define HSTATE_ATTR_RO(_name) \
@@ -2404,11 +2410,6 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
 	int err;
 	NODEMASK_ALLOC(nodemask_t, nodes_allowed, GFP_KERNEL | __GFP_NORETRY);
 
-	if (hstate_is_gigantic(h) && !gigantic_page_supported()) {
-		err = -EINVAL;
-		goto out;
-	}
-
 	if (nid == NUMA_NO_NODE) {
 		/*
 		 * global hstate attribute
@@ -2428,7 +2429,9 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
 	} else
 		nodes_allowed = &node_states[N_MEMORY];
 
-	h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed);
+	err = set_max_huge_pages(h, count, nodes_allowed);
+	if (err)
+		goto out;
 
 	if (nodes_allowed != &node_states[N_MEMORY])
 		NODEMASK_FREE(nodes_allowed);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index cde5dac6229a..81b931db85a1 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -8241,8 +8241,9 @@ int alloc_contig_range(unsigned long start, unsigned long end,
 				pfn_max_align_up(end), migratetype);
 	return ret;
 }
+#endif
 
-void free_contig_range(unsigned long pfn, unsigned nr_pages)
+void free_contig_range(unsigned long pfn, unsigned int nr_pages)
 {
 	unsigned int count = 0;
 
@@ -8254,7 +8255,6 @@ void free_contig_range(unsigned long pfn, unsigned nr_pages)
 	}
 	WARN(count != 0, "%d pages are still in use!\n", count);
 }
-#endif
 
 #ifdef CONFIG_MEMORY_HOTPLUG
 /*
-- 
2.16.2


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
  2019-01-17 18:39 [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration Alexandre Ghiti
  2019-01-17 18:39 ` Alexandre Ghiti
@ 2019-01-17 20:09 ` Mike Kravetz
  2019-01-17 20:09   ` Mike Kravetz
  2019-01-19 23:57 ` Will Deacon
                   ` (3 subsequent siblings)
  5 siblings, 1 reply; 10+ messages in thread
From: Mike Kravetz @ 2019-01-17 20:09 UTC (permalink / raw)
  To: Alexandre Ghiti, Catalin Marinas, Will Deacon,
	Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
	Martin Schwidefsky, Heiko Carstens, Thomas Gleixner, Ingo Molnar,
	Borislav Petkov, H . Peter Anvin, x86, Alexander Viro,
	linux-arm-kernel, linux-kernel, linuxppc-dev, linux-s390,
	linux-fsdevel, linux-mm
  Cc: hch, linux-riscv, Alexandre Ghiti

On 1/17/19 10:39 AM, Alexandre Ghiti wrote:
> From: Alexandre Ghiti <alex@ghiti.fr>
> 
> On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
> that support gigantic pages, boottime reserved gigantic pages can not be
> freed at all. This patchs simply enables the possibility to hand back
> those pages to memory allocator.
> 
> This commit then renames gigantic_page_supported and
> ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
> being false does not mean that the system cannot use gigantic pages: it
> just means that runtime allocation of gigantic pages is not supported,
> one can still allocate boottime gigantic pages if the architecture supports
> it.
> 
> Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>

Thank you for doing this!

Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>

> --- a/include/linux/gfp.h
> +++ b/include/linux/gfp.h
> @@ -589,8 +589,8 @@ static inline bool pm_suspended_storage(void)
>  /* The below functions must be run on a range from a single zone. */
>  extern int alloc_contig_range(unsigned long start, unsigned long end,
>  			      unsigned migratetype, gfp_t gfp_mask);
> -extern void free_contig_range(unsigned long pfn, unsigned nr_pages);
>  #endif
> +extern void free_contig_range(unsigned long pfn, unsigned int nr_pages);

I think nr_pages should be an unsigned long in cma_release() and here
as well, but that is beyond the scope of this patch.  Most callers of
cma_release pass in a truncated unsigned long.  The truncation is unlikely
to cause any issues, just would be nice if types were consistent.  I have
a patch to do that as part of a contiguous allocation series that I will
get back to someday.

> @@ -2350,9 +2355,10 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
>  			break;
>  	}
>  out:
> -	ret = persistent_huge_pages(h);
> +	h->max_huge_pages = persistent_huge_pages(h);
>  	spin_unlock(&hugetlb_lock);
> -	return ret;
> +
> +	return 0;
>  }
>  
>  #define HSTATE_ATTR_RO(_name) \
> @@ -2404,11 +2410,6 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>  	int err;
>  	NODEMASK_ALLOC(nodemask_t, nodes_allowed, GFP_KERNEL | __GFP_NORETRY);
>  
> -	if (hstate_is_gigantic(h) && !gigantic_page_supported()) {
> -		err = -EINVAL;
> -		goto out;
> -	}
> -
>  	if (nid == NUMA_NO_NODE) {
>  		/*
>  		 * global hstate attribute
> @@ -2428,7 +2429,9 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>  	} else
>  		nodes_allowed = &node_states[N_MEMORY];
>  
> -	h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed);
> +	err = set_max_huge_pages(h, count, nodes_allowed);
> +	if (err)
> +		goto out;
>  
>  	if (nodes_allowed != &node_states[N_MEMORY])
>  		NODEMASK_FREE(nodes_allowed);

Yeah!  Those changes causes max_huge_pages to be modified while holding
hugetlb_lock as it should be.
-- 
Mike Kravetz

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
  2019-01-17 20:09 ` Mike Kravetz
@ 2019-01-17 20:09   ` Mike Kravetz
  0 siblings, 0 replies; 10+ messages in thread
From: Mike Kravetz @ 2019-01-17 20:09 UTC (permalink / raw)
  To: Alexandre Ghiti, Catalin Marinas, Will Deacon,
	Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
	Martin Schwidefsky, Heiko Carstens, Thomas Gleixner, Ingo Molnar,
	Borislav Petkov, H . Peter Anvin, x86, Alexander Viro,
	linux-arm-kernel, linux-kernel, linuxppc-dev, linux-s390,
	linux-fsdevel, linux-mm
  Cc: linux-riscv, hch, Alexandre Ghiti

On 1/17/19 10:39 AM, Alexandre Ghiti wrote:
> From: Alexandre Ghiti <alex@ghiti.fr>
> 
> On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
> that support gigantic pages, boottime reserved gigantic pages can not be
> freed at all. This patchs simply enables the possibility to hand back
> those pages to memory allocator.
> 
> This commit then renames gigantic_page_supported and
> ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
> being false does not mean that the system cannot use gigantic pages: it
> just means that runtime allocation of gigantic pages is not supported,
> one can still allocate boottime gigantic pages if the architecture supports
> it.
> 
> Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>

Thank you for doing this!

Reviewed-by: Mike Kravetz <mike.kravetz@oracle.com>

> --- a/include/linux/gfp.h
> +++ b/include/linux/gfp.h
> @@ -589,8 +589,8 @@ static inline bool pm_suspended_storage(void)
>  /* The below functions must be run on a range from a single zone. */
>  extern int alloc_contig_range(unsigned long start, unsigned long end,
>  			      unsigned migratetype, gfp_t gfp_mask);
> -extern void free_contig_range(unsigned long pfn, unsigned nr_pages);
>  #endif
> +extern void free_contig_range(unsigned long pfn, unsigned int nr_pages);

I think nr_pages should be an unsigned long in cma_release() and here
as well, but that is beyond the scope of this patch.  Most callers of
cma_release pass in a truncated unsigned long.  The truncation is unlikely
to cause any issues, just would be nice if types were consistent.  I have
a patch to do that as part of a contiguous allocation series that I will
get back to someday.

> @@ -2350,9 +2355,10 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
>  			break;
>  	}
>  out:
> -	ret = persistent_huge_pages(h);
> +	h->max_huge_pages = persistent_huge_pages(h);
>  	spin_unlock(&hugetlb_lock);
> -	return ret;
> +
> +	return 0;
>  }
>  
>  #define HSTATE_ATTR_RO(_name) \
> @@ -2404,11 +2410,6 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>  	int err;
>  	NODEMASK_ALLOC(nodemask_t, nodes_allowed, GFP_KERNEL | __GFP_NORETRY);
>  
> -	if (hstate_is_gigantic(h) && !gigantic_page_supported()) {
> -		err = -EINVAL;
> -		goto out;
> -	}
> -
>  	if (nid == NUMA_NO_NODE) {
>  		/*
>  		 * global hstate attribute
> @@ -2428,7 +2429,9 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>  	} else
>  		nodes_allowed = &node_states[N_MEMORY];
>  
> -	h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed);
> +	err = set_max_huge_pages(h, count, nodes_allowed);
> +	if (err)
> +		goto out;
>  
>  	if (nodes_allowed != &node_states[N_MEMORY])
>  		NODEMASK_FREE(nodes_allowed);

Yeah!  Those changes causes max_huge_pages to be modified while holding
hugetlb_lock as it should be.
-- 
Mike Kravetz

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
  2019-01-17 18:39 [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration Alexandre Ghiti
  2019-01-17 18:39 ` Alexandre Ghiti
  2019-01-17 20:09 ` Mike Kravetz
@ 2019-01-19 23:57 ` Will Deacon
  2019-02-03 18:17 ` Alex Ghiti
                   ` (2 subsequent siblings)
  5 siblings, 0 replies; 10+ messages in thread
From: Will Deacon @ 2019-01-19 23:57 UTC (permalink / raw)
  To: Alexandre Ghiti
  Cc: Catalin Marinas, Benjamin Herrenschmidt, Paul Mackerras,
	Michael Ellerman, Martin Schwidefsky, Heiko Carstens,
	Thomas Gleixner, Ingo Molnar, Borislav Petkov, H . Peter Anvin,
	x86, Alexander Viro, Mike Kravetz, linux-arm-kernel,
	linux-kernel, linuxppc-dev, linux-s390, linux-fsdevel, linux-mm,
	hch, linux-riscv, Alexandre Ghiti

On Thu, Jan 17, 2019 at 06:39:53PM +0000, Alexandre Ghiti wrote:
> From: Alexandre Ghiti <alex@ghiti.fr>
> 
> On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
> that support gigantic pages, boottime reserved gigantic pages can not be
> freed at all. This patchs simply enables the possibility to hand back
> those pages to memory allocator.
> 
> This commit then renames gigantic_page_supported and
> ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
> being false does not mean that the system cannot use gigantic pages: it
> just means that runtime allocation of gigantic pages is not supported,
> one can still allocate boottime gigantic pages if the architecture supports
> it.
> 
> Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
> ---
> 
> - Compiled on all architectures
> - Tested on riscv architecture
> 
>  arch/arm64/Kconfig                           |  2 +-
>  arch/arm64/include/asm/hugetlb.h             |  7 +++--

The arm64 bits look straightforward enough to me...

Acked-by: Will Deacon <will.deacon@arm.com>

Will

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
  2019-01-17 18:39 [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration Alexandre Ghiti
                   ` (2 preceding siblings ...)
  2019-01-19 23:57 ` Will Deacon
@ 2019-02-03 18:17 ` Alex Ghiti
  2019-02-05 11:23 ` Michael Ellerman
  2019-02-13 11:27 ` Vlastimil Babka
  5 siblings, 0 replies; 10+ messages in thread
From: Alex Ghiti @ 2019-02-03 18:17 UTC (permalink / raw)
  To: akpm, Catalin Marinas, Will Deacon, Benjamin Herrenschmidt,
	Paul Mackerras, Michael Ellerman, Martin Schwidefsky,
	Heiko Carstens, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
	H . Peter Anvin, x86, Alexander Viro, Mike Kravetz,
	linux-arm-kernel, linux-kernel, linuxppc-dev, linux-s390,
	linux-fsdevel, linux-mm
  Cc: linux-riscv, hch

On 1/17/19 1:39 PM, Alexandre Ghiti wrote:
> From: Alexandre Ghiti <alex@ghiti.fr>
>
> On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
> that support gigantic pages, boottime reserved gigantic pages can not be
> freed at all. This patchs simply enables the possibility to hand back
> those pages to memory allocator.
>
> This commit then renames gigantic_page_supported and
> ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
> being false does not mean that the system cannot use gigantic pages: it
> just means that runtime allocation of gigantic pages is not supported,
> one can still allocate boottime gigantic pages if the architecture supports
> it.
>
> Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
> ---
>
> - Compiled on all architectures
> - Tested on riscv architecture
>
>   arch/arm64/Kconfig                           |  2 +-
>   arch/arm64/include/asm/hugetlb.h             |  7 +++--
>   arch/powerpc/include/asm/book3s/64/hugetlb.h |  4 +--
>   arch/powerpc/platforms/Kconfig.cputype       |  2 +-
>   arch/s390/Kconfig                            |  2 +-
>   arch/s390/include/asm/hugetlb.h              |  7 +++--
>   arch/x86/Kconfig                             |  2 +-
>   arch/x86/include/asm/hugetlb.h               |  7 +++--
>   fs/Kconfig                                   |  2 +-
>   include/linux/gfp.h                          |  2 +-
>   mm/hugetlb.c                                 | 43 +++++++++++++++-------------
>   mm/page_alloc.c                              |  4 +--
>   12 files changed, 48 insertions(+), 36 deletions(-)
>
> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> index a4168d366127..18239cbd7fcd 100644
> --- a/arch/arm64/Kconfig
> +++ b/arch/arm64/Kconfig
> @@ -18,7 +18,7 @@ config ARM64
>   	select ARCH_HAS_FAST_MULTIPLIER
>   	select ARCH_HAS_FORTIFY_SOURCE
>   	select ARCH_HAS_GCOV_PROFILE_ALL
> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>   	select ARCH_HAS_KCOV
>   	select ARCH_HAS_MEMBARRIER_SYNC_CORE
>   	select ARCH_HAS_PTE_SPECIAL
> diff --git a/arch/arm64/include/asm/hugetlb.h b/arch/arm64/include/asm/hugetlb.h
> index fb6609875455..797fc77eabcd 100644
> --- a/arch/arm64/include/asm/hugetlb.h
> +++ b/arch/arm64/include/asm/hugetlb.h
> @@ -65,8 +65,11 @@ extern void set_huge_swap_pte_at(struct mm_struct *mm, unsigned long addr,
>   
>   #include <asm-generic/hugetlb.h>
>   
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
> -static inline bool gigantic_page_supported(void) { return true; }
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
> +static inline bool gigantic_page_runtime_allocation_supported(void)
> +{
> +	return true;
> +}
>   #endif
>   
>   #endif /* __ASM_HUGETLB_H */
> diff --git a/arch/powerpc/include/asm/book3s/64/hugetlb.h b/arch/powerpc/include/asm/book3s/64/hugetlb.h
> index 5b0177733994..7711f0e2c7e5 100644
> --- a/arch/powerpc/include/asm/book3s/64/hugetlb.h
> +++ b/arch/powerpc/include/asm/book3s/64/hugetlb.h
> @@ -32,8 +32,8 @@ static inline int hstate_get_psize(struct hstate *hstate)
>   	}
>   }
>   
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
> -static inline bool gigantic_page_supported(void)
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
> +static inline bool gigantic_page_runtime_allocation_supported(void)
>   {
>   	return true;
>   }
> diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
> index 8c7464c3f27f..779e06bac697 100644
> --- a/arch/powerpc/platforms/Kconfig.cputype
> +++ b/arch/powerpc/platforms/Kconfig.cputype
> @@ -319,7 +319,7 @@ config ARCH_ENABLE_SPLIT_PMD_PTLOCK
>   config PPC_RADIX_MMU
>   	bool "Radix MMU Support"
>   	depends on PPC_BOOK3S_64
> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>   	default y
>   	help
>   	  Enable support for the Power ISA 3.0 Radix style MMU. Currently this
> diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
> index ed554b09eb3f..6776eef6a9ae 100644
> --- a/arch/s390/Kconfig
> +++ b/arch/s390/Kconfig
> @@ -69,7 +69,7 @@ config S390
>   	select ARCH_HAS_ELF_RANDOMIZE
>   	select ARCH_HAS_FORTIFY_SOURCE
>   	select ARCH_HAS_GCOV_PROFILE_ALL
> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>   	select ARCH_HAS_KCOV
>   	select ARCH_HAS_PTE_SPECIAL
>   	select ARCH_HAS_SET_MEMORY
> diff --git a/arch/s390/include/asm/hugetlb.h b/arch/s390/include/asm/hugetlb.h
> index 2d1afa58a4b6..57c952f5388e 100644
> --- a/arch/s390/include/asm/hugetlb.h
> +++ b/arch/s390/include/asm/hugetlb.h
> @@ -116,7 +116,10 @@ static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot)
>   	return pte_modify(pte, newprot);
>   }
>   
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
> -static inline bool gigantic_page_supported(void) { return true; }
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
> +static inline bool gigantic_page_runtime_allocation_supported(void)
> +{
> +	return true;
> +}
>   #endif
>   #endif /* _ASM_S390_HUGETLB_H */
> diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
> index 6185d4f33296..a88f5a4311c9 100644
> --- a/arch/x86/Kconfig
> +++ b/arch/x86/Kconfig
> @@ -23,7 +23,7 @@ config X86_64
>   	def_bool y
>   	depends on 64BIT
>   	# Options that are inherently 64-bit kernel only:
> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>   	select ARCH_SUPPORTS_INT128
>   	select ARCH_USE_CMPXCHG_LOCKREF
>   	select HAVE_ARCH_SOFT_DIRTY
> diff --git a/arch/x86/include/asm/hugetlb.h b/arch/x86/include/asm/hugetlb.h
> index 7469d321f072..5a5e7119ced4 100644
> --- a/arch/x86/include/asm/hugetlb.h
> +++ b/arch/x86/include/asm/hugetlb.h
> @@ -17,8 +17,11 @@ static inline void arch_clear_hugepage_flags(struct page *page)
>   {
>   }
>   
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
> -static inline bool gigantic_page_supported(void) { return true; }
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
> +static inline bool gigantic_page_runtime_allocation_supported(void)
> +{
> +	return true;
> +}
>   #endif
>   
>   #endif /* _ASM_X86_HUGETLB_H */
> diff --git a/fs/Kconfig b/fs/Kconfig
> index ac474a61be37..4192d1fde0f0 100644
> --- a/fs/Kconfig
> +++ b/fs/Kconfig
> @@ -207,7 +207,7 @@ config HUGETLB_PAGE
>   config MEMFD_CREATE
>   	def_bool TMPFS || HUGETLBFS
>   
> -config ARCH_HAS_GIGANTIC_PAGE
> +config ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>   	bool
>   
>   source "fs/configfs/Kconfig"
> diff --git a/include/linux/gfp.h b/include/linux/gfp.h
> index 5f5e25fd6149..79ff86fabd42 100644
> --- a/include/linux/gfp.h
> +++ b/include/linux/gfp.h
> @@ -589,8 +589,8 @@ static inline bool pm_suspended_storage(void)
>   /* The below functions must be run on a range from a single zone. */
>   extern int alloc_contig_range(unsigned long start, unsigned long end,
>   			      unsigned migratetype, gfp_t gfp_mask);
> -extern void free_contig_range(unsigned long pfn, unsigned nr_pages);
>   #endif
> +extern void free_contig_range(unsigned long pfn, unsigned int nr_pages);
>   
>   #ifdef CONFIG_CMA
>   /* CMA stuff */
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index 745088810965..9893ba26b3b8 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -1035,7 +1035,6 @@ static int hstate_next_node_to_free(struct hstate *h, nodemask_t *nodes_allowed)
>   		((node = hstate_next_node_to_free(hs, mask)) || 1);	\
>   		nr_nodes--)
>   
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
>   static void destroy_compound_gigantic_page(struct page *page,
>   					unsigned int order)
>   {
> @@ -1058,6 +1057,7 @@ static void free_gigantic_page(struct page *page, unsigned int order)
>   	free_contig_range(page_to_pfn(page), 1 << order);
>   }
>   
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>   static int __alloc_gigantic_page(unsigned long start_pfn,
>   				unsigned long nr_pages, gfp_t gfp_mask)
>   {
> @@ -1143,22 +1143,19 @@ static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
>   static void prep_new_huge_page(struct hstate *h, struct page *page, int nid);
>   static void prep_compound_gigantic_page(struct page *page, unsigned int order);
>   
> -#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE */
> -static inline bool gigantic_page_supported(void) { return false; }
> +#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION */
> +static inline bool gigantic_page_runtime_allocation_supported(void)
> +{
> +	return false;
> +}
>   static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
>   		int nid, nodemask_t *nodemask) { return NULL; }
> -static inline void free_gigantic_page(struct page *page, unsigned int order) { }
> -static inline void destroy_compound_gigantic_page(struct page *page,
> -						unsigned int order) { }
>   #endif
>   
>   static void update_and_free_page(struct hstate *h, struct page *page)
>   {
>   	int i;
>   
> -	if (hstate_is_gigantic(h) && !gigantic_page_supported())
> -		return;
> -
>   	h->nr_huge_pages--;
>   	h->nr_huge_pages_node[page_to_nid(page)]--;
>   	for (i = 0; i < pages_per_huge_page(h); i++) {
> @@ -2276,13 +2273,20 @@ static int adjust_pool_surplus(struct hstate *h, nodemask_t *nodes_allowed,
>   }
>   
>   #define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages)
> -static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
> +static int set_max_huge_pages(struct hstate *h, unsigned long count,
>   						nodemask_t *nodes_allowed)
>   {
>   	unsigned long min_count, ret;
>   
> -	if (hstate_is_gigantic(h) && !gigantic_page_supported())
> -		return h->max_huge_pages;
> +	if (hstate_is_gigantic(h) &&
> +		!gigantic_page_runtime_allocation_supported()) {
> +		spin_lock(&hugetlb_lock);
> +		if (count > persistent_huge_pages(h)) {
> +			spin_unlock(&hugetlb_lock);
> +			return -EINVAL;
> +		}
> +		goto decrease_pool;
> +	}
>   
>   	/*
>   	 * Increase the pool size
> @@ -2322,6 +2326,7 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
>   			goto out;
>   	}
>   
> +decrease_pool:
>   	/*
>   	 * Decrease the pool size
>   	 * First return free pages to the buddy allocator (being careful
> @@ -2350,9 +2355,10 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
>   			break;
>   	}
>   out:
> -	ret = persistent_huge_pages(h);
> +	h->max_huge_pages = persistent_huge_pages(h);
>   	spin_unlock(&hugetlb_lock);
> -	return ret;
> +
> +	return 0;
>   }
>   
>   #define HSTATE_ATTR_RO(_name) \
> @@ -2404,11 +2410,6 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>   	int err;
>   	NODEMASK_ALLOC(nodemask_t, nodes_allowed, GFP_KERNEL | __GFP_NORETRY);
>   
> -	if (hstate_is_gigantic(h) && !gigantic_page_supported()) {
> -		err = -EINVAL;
> -		goto out;
> -	}
> -
>   	if (nid == NUMA_NO_NODE) {
>   		/*
>   		 * global hstate attribute
> @@ -2428,7 +2429,9 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>   	} else
>   		nodes_allowed = &node_states[N_MEMORY];
>   
> -	h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed);
> +	err = set_max_huge_pages(h, count, nodes_allowed);
> +	if (err)
> +		goto out;
>   
>   	if (nodes_allowed != &node_states[N_MEMORY])
>   		NODEMASK_FREE(nodes_allowed);
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index cde5dac6229a..81b931db85a1 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -8241,8 +8241,9 @@ int alloc_contig_range(unsigned long start, unsigned long end,
>   				pfn_max_align_up(end), migratetype);
>   	return ret;
>   }
> +#endif
>   
> -void free_contig_range(unsigned long pfn, unsigned nr_pages)
> +void free_contig_range(unsigned long pfn, unsigned int nr_pages)
>   {
>   	unsigned int count = 0;
>   
> @@ -8254,7 +8255,6 @@ void free_contig_range(unsigned long pfn, unsigned nr_pages)
>   	}
>   	WARN(count != 0, "%d pages are still in use!\n", count);
>   }
> -#endif
>   
>   #ifdef CONFIG_MEMORY_HOTPLUG
>   /*


Hi Andrew,

Can you consider this patch for inclusion in mm tree ? It lacks reviews 
from some
arch maintainers and has been reviewed by Mike Kravetz.
Tell me if I can do something to help,

Thanks,

Alex


^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
  2019-01-17 18:39 [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration Alexandre Ghiti
                   ` (3 preceding siblings ...)
  2019-02-03 18:17 ` Alex Ghiti
@ 2019-02-05 11:23 ` Michael Ellerman
  2019-02-05 11:35   ` Alex Ghiti
  2019-02-13 11:27 ` Vlastimil Babka
  5 siblings, 1 reply; 10+ messages in thread
From: Michael Ellerman @ 2019-02-05 11:23 UTC (permalink / raw)
  To: Alexandre Ghiti, Catalin Marinas, Will Deacon,
	Benjamin Herrenschmidt, Paul Mackerras, Martin Schwidefsky,
	Heiko Carstens, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
	H . Peter Anvin, x86, Alexander Viro, Mike Kravetz,
	linux-arm-kernel, linux-kernel, linuxppc-dev, linux-s390,
	linux-fsdevel, linux-mm
  Cc: linux-riscv, hch, Alexandre Ghiti

Alexandre Ghiti <aghiti@upmem.com> writes:

> From: Alexandre Ghiti <alex@ghiti.fr>
>
> On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
> that support gigantic pages, boottime reserved gigantic pages can not be
> freed at all. This patchs simply enables the possibility to hand back
> those pages to memory allocator.
>
> This commit then renames gigantic_page_supported and
> ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
> being false does not mean that the system cannot use gigantic pages: it
> just means that runtime allocation of gigantic pages is not supported,
> one can still allocate boottime gigantic pages if the architecture supports
> it.
>
> Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
> ---
>
> - Compiled on all architectures
> - Tested on riscv architecture
>
>  arch/arm64/Kconfig                           |  2 +-
>  arch/arm64/include/asm/hugetlb.h             |  7 +++--
>  arch/powerpc/include/asm/book3s/64/hugetlb.h |  4 +--
>  arch/powerpc/platforms/Kconfig.cputype       |  2 +-

The powerpc parts look fine.

Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)

cheers

>  arch/s390/Kconfig                            |  2 +-
>  arch/s390/include/asm/hugetlb.h              |  7 +++--
>  arch/x86/Kconfig                             |  2 +-
>  arch/x86/include/asm/hugetlb.h               |  7 +++--
>  fs/Kconfig                                   |  2 +-
>  include/linux/gfp.h                          |  2 +-
>  mm/hugetlb.c                                 | 43 +++++++++++++++-------------
>  mm/page_alloc.c                              |  4 +--
>  12 files changed, 48 insertions(+), 36 deletions(-)
>
> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
> index a4168d366127..18239cbd7fcd 100644
> --- a/arch/arm64/Kconfig
> +++ b/arch/arm64/Kconfig
> @@ -18,7 +18,7 @@ config ARM64
>  	select ARCH_HAS_FAST_MULTIPLIER
>  	select ARCH_HAS_FORTIFY_SOURCE
>  	select ARCH_HAS_GCOV_PROFILE_ALL
> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>  	select ARCH_HAS_KCOV
>  	select ARCH_HAS_MEMBARRIER_SYNC_CORE
>  	select ARCH_HAS_PTE_SPECIAL
> diff --git a/arch/arm64/include/asm/hugetlb.h b/arch/arm64/include/asm/hugetlb.h
> index fb6609875455..797fc77eabcd 100644
> --- a/arch/arm64/include/asm/hugetlb.h
> +++ b/arch/arm64/include/asm/hugetlb.h
> @@ -65,8 +65,11 @@ extern void set_huge_swap_pte_at(struct mm_struct *mm, unsigned long addr,
>  
>  #include <asm-generic/hugetlb.h>
>  
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
> -static inline bool gigantic_page_supported(void) { return true; }
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
> +static inline bool gigantic_page_runtime_allocation_supported(void)
> +{
> +	return true;
> +}
>  #endif
>  
>  #endif /* __ASM_HUGETLB_H */
> diff --git a/arch/powerpc/include/asm/book3s/64/hugetlb.h b/arch/powerpc/include/asm/book3s/64/hugetlb.h
> index 5b0177733994..7711f0e2c7e5 100644
> --- a/arch/powerpc/include/asm/book3s/64/hugetlb.h
> +++ b/arch/powerpc/include/asm/book3s/64/hugetlb.h
> @@ -32,8 +32,8 @@ static inline int hstate_get_psize(struct hstate *hstate)
>  	}
>  }
>  
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
> -static inline bool gigantic_page_supported(void)
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
> +static inline bool gigantic_page_runtime_allocation_supported(void)
>  {
>  	return true;
>  }
> diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
> index 8c7464c3f27f..779e06bac697 100644
> --- a/arch/powerpc/platforms/Kconfig.cputype
> +++ b/arch/powerpc/platforms/Kconfig.cputype
> @@ -319,7 +319,7 @@ config ARCH_ENABLE_SPLIT_PMD_PTLOCK
>  config PPC_RADIX_MMU
>  	bool "Radix MMU Support"
>  	depends on PPC_BOOK3S_64
> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>  	default y
>  	help
>  	  Enable support for the Power ISA 3.0 Radix style MMU. Currently this
> diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
> index ed554b09eb3f..6776eef6a9ae 100644
> --- a/arch/s390/Kconfig
> +++ b/arch/s390/Kconfig
> @@ -69,7 +69,7 @@ config S390
>  	select ARCH_HAS_ELF_RANDOMIZE
>  	select ARCH_HAS_FORTIFY_SOURCE
>  	select ARCH_HAS_GCOV_PROFILE_ALL
> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>  	select ARCH_HAS_KCOV
>  	select ARCH_HAS_PTE_SPECIAL
>  	select ARCH_HAS_SET_MEMORY
> diff --git a/arch/s390/include/asm/hugetlb.h b/arch/s390/include/asm/hugetlb.h
> index 2d1afa58a4b6..57c952f5388e 100644
> --- a/arch/s390/include/asm/hugetlb.h
> +++ b/arch/s390/include/asm/hugetlb.h
> @@ -116,7 +116,10 @@ static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot)
>  	return pte_modify(pte, newprot);
>  }
>  
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
> -static inline bool gigantic_page_supported(void) { return true; }
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
> +static inline bool gigantic_page_runtime_allocation_supported(void)
> +{
> +	return true;
> +}
>  #endif
>  #endif /* _ASM_S390_HUGETLB_H */
> diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
> index 6185d4f33296..a88f5a4311c9 100644
> --- a/arch/x86/Kconfig
> +++ b/arch/x86/Kconfig
> @@ -23,7 +23,7 @@ config X86_64
>  	def_bool y
>  	depends on 64BIT
>  	# Options that are inherently 64-bit kernel only:
> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>  	select ARCH_SUPPORTS_INT128
>  	select ARCH_USE_CMPXCHG_LOCKREF
>  	select HAVE_ARCH_SOFT_DIRTY
> diff --git a/arch/x86/include/asm/hugetlb.h b/arch/x86/include/asm/hugetlb.h
> index 7469d321f072..5a5e7119ced4 100644
> --- a/arch/x86/include/asm/hugetlb.h
> +++ b/arch/x86/include/asm/hugetlb.h
> @@ -17,8 +17,11 @@ static inline void arch_clear_hugepage_flags(struct page *page)
>  {
>  }
>  
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
> -static inline bool gigantic_page_supported(void) { return true; }
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
> +static inline bool gigantic_page_runtime_allocation_supported(void)
> +{
> +	return true;
> +}
>  #endif
>  
>  #endif /* _ASM_X86_HUGETLB_H */
> diff --git a/fs/Kconfig b/fs/Kconfig
> index ac474a61be37..4192d1fde0f0 100644
> --- a/fs/Kconfig
> +++ b/fs/Kconfig
> @@ -207,7 +207,7 @@ config HUGETLB_PAGE
>  config MEMFD_CREATE
>  	def_bool TMPFS || HUGETLBFS
>  
> -config ARCH_HAS_GIGANTIC_PAGE
> +config ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>  	bool
>  
>  source "fs/configfs/Kconfig"
> diff --git a/include/linux/gfp.h b/include/linux/gfp.h
> index 5f5e25fd6149..79ff86fabd42 100644
> --- a/include/linux/gfp.h
> +++ b/include/linux/gfp.h
> @@ -589,8 +589,8 @@ static inline bool pm_suspended_storage(void)
>  /* The below functions must be run on a range from a single zone. */
>  extern int alloc_contig_range(unsigned long start, unsigned long end,
>  			      unsigned migratetype, gfp_t gfp_mask);
> -extern void free_contig_range(unsigned long pfn, unsigned nr_pages);
>  #endif
> +extern void free_contig_range(unsigned long pfn, unsigned int nr_pages);
>  
>  #ifdef CONFIG_CMA
>  /* CMA stuff */
> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
> index 745088810965..9893ba26b3b8 100644
> --- a/mm/hugetlb.c
> +++ b/mm/hugetlb.c
> @@ -1035,7 +1035,6 @@ static int hstate_next_node_to_free(struct hstate *h, nodemask_t *nodes_allowed)
>  		((node = hstate_next_node_to_free(hs, mask)) || 1);	\
>  		nr_nodes--)
>  
> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
>  static void destroy_compound_gigantic_page(struct page *page,
>  					unsigned int order)
>  {
> @@ -1058,6 +1057,7 @@ static void free_gigantic_page(struct page *page, unsigned int order)
>  	free_contig_range(page_to_pfn(page), 1 << order);
>  }
>  
> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>  static int __alloc_gigantic_page(unsigned long start_pfn,
>  				unsigned long nr_pages, gfp_t gfp_mask)
>  {
> @@ -1143,22 +1143,19 @@ static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
>  static void prep_new_huge_page(struct hstate *h, struct page *page, int nid);
>  static void prep_compound_gigantic_page(struct page *page, unsigned int order);
>  
> -#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE */
> -static inline bool gigantic_page_supported(void) { return false; }
> +#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION */
> +static inline bool gigantic_page_runtime_allocation_supported(void)
> +{
> +	return false;
> +}
>  static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
>  		int nid, nodemask_t *nodemask) { return NULL; }
> -static inline void free_gigantic_page(struct page *page, unsigned int order) { }
> -static inline void destroy_compound_gigantic_page(struct page *page,
> -						unsigned int order) { }
>  #endif
>  
>  static void update_and_free_page(struct hstate *h, struct page *page)
>  {
>  	int i;
>  
> -	if (hstate_is_gigantic(h) && !gigantic_page_supported())
> -		return;
> -
>  	h->nr_huge_pages--;
>  	h->nr_huge_pages_node[page_to_nid(page)]--;
>  	for (i = 0; i < pages_per_huge_page(h); i++) {
> @@ -2276,13 +2273,20 @@ static int adjust_pool_surplus(struct hstate *h, nodemask_t *nodes_allowed,
>  }
>  
>  #define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages)
> -static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
> +static int set_max_huge_pages(struct hstate *h, unsigned long count,
>  						nodemask_t *nodes_allowed)
>  {
>  	unsigned long min_count, ret;
>  
> -	if (hstate_is_gigantic(h) && !gigantic_page_supported())
> -		return h->max_huge_pages;
> +	if (hstate_is_gigantic(h) &&
> +		!gigantic_page_runtime_allocation_supported()) {
> +		spin_lock(&hugetlb_lock);
> +		if (count > persistent_huge_pages(h)) {
> +			spin_unlock(&hugetlb_lock);
> +			return -EINVAL;
> +		}
> +		goto decrease_pool;
> +	}
>  
>  	/*
>  	 * Increase the pool size
> @@ -2322,6 +2326,7 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
>  			goto out;
>  	}
>  
> +decrease_pool:
>  	/*
>  	 * Decrease the pool size
>  	 * First return free pages to the buddy allocator (being careful
> @@ -2350,9 +2355,10 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
>  			break;
>  	}
>  out:
> -	ret = persistent_huge_pages(h);
> +	h->max_huge_pages = persistent_huge_pages(h);
>  	spin_unlock(&hugetlb_lock);
> -	return ret;
> +
> +	return 0;
>  }
>  
>  #define HSTATE_ATTR_RO(_name) \
> @@ -2404,11 +2410,6 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>  	int err;
>  	NODEMASK_ALLOC(nodemask_t, nodes_allowed, GFP_KERNEL | __GFP_NORETRY);
>  
> -	if (hstate_is_gigantic(h) && !gigantic_page_supported()) {
> -		err = -EINVAL;
> -		goto out;
> -	}
> -
>  	if (nid == NUMA_NO_NODE) {
>  		/*
>  		 * global hstate attribute
> @@ -2428,7 +2429,9 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>  	} else
>  		nodes_allowed = &node_states[N_MEMORY];
>  
> -	h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed);
> +	err = set_max_huge_pages(h, count, nodes_allowed);
> +	if (err)
> +		goto out;
>  
>  	if (nodes_allowed != &node_states[N_MEMORY])
>  		NODEMASK_FREE(nodes_allowed);
> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
> index cde5dac6229a..81b931db85a1 100644
> --- a/mm/page_alloc.c
> +++ b/mm/page_alloc.c
> @@ -8241,8 +8241,9 @@ int alloc_contig_range(unsigned long start, unsigned long end,
>  				pfn_max_align_up(end), migratetype);
>  	return ret;
>  }
> +#endif
>  
> -void free_contig_range(unsigned long pfn, unsigned nr_pages)
> +void free_contig_range(unsigned long pfn, unsigned int nr_pages)
>  {
>  	unsigned int count = 0;
>  
> @@ -8254,7 +8255,6 @@ void free_contig_range(unsigned long pfn, unsigned nr_pages)
>  	}
>  	WARN(count != 0, "%d pages are still in use!\n", count);
>  }
> -#endif
>  
>  #ifdef CONFIG_MEMORY_HOTPLUG
>  /*
> -- 
> 2.16.2

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
  2019-02-05 11:23 ` Michael Ellerman
@ 2019-02-05 11:35   ` Alex Ghiti
  0 siblings, 0 replies; 10+ messages in thread
From: Alex Ghiti @ 2019-02-05 11:35 UTC (permalink / raw)
  To: Michael Ellerman
  Cc: linux-riscv, hch, Catalin Marinas, Will Deacon,
	Benjamin Herrenschmidt, Paul Mackerras, Martin Schwidefsky,
	Heiko Carstens, Thomas Gleixner, Ingo Molnar, Borislav Petkov,
	H . Peter Anvin, x86, Alexander Viro, Mike Kravetz,
	linux-arm-kernel, linux-kernel, linuxppc-dev, linux-s390,
	linux-fsdevel, linux-mm

On 2/5/19 6:23 AM, Michael Ellerman wrote:
> Alexandre Ghiti <aghiti@upmem.com> writes:
>
>> From: Alexandre Ghiti <alex@ghiti.fr>
>>
>> On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
>> that support gigantic pages, boottime reserved gigantic pages can not be
>> freed at all. This patchs simply enables the possibility to hand back
>> those pages to memory allocator.
>>
>> This commit then renames gigantic_page_supported and
>> ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
>> being false does not mean that the system cannot use gigantic pages: it
>> just means that runtime allocation of gigantic pages is not supported,
>> one can still allocate boottime gigantic pages if the architecture supports
>> it.
>>
>> Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
>> ---
>>
>> - Compiled on all architectures
>> - Tested on riscv architecture
>>
>>   arch/arm64/Kconfig                           |  2 +-
>>   arch/arm64/include/asm/hugetlb.h             |  7 +++--
>>   arch/powerpc/include/asm/book3s/64/hugetlb.h |  4 +--
>>   arch/powerpc/platforms/Kconfig.cputype       |  2 +-
> The powerpc parts look fine.
>
> Acked-by: Michael Ellerman <mpe@ellerman.id.au> (powerpc)

Thank you Michael,

Alex

>
> cheers
>
>>   arch/s390/Kconfig                            |  2 +-
>>   arch/s390/include/asm/hugetlb.h              |  7 +++--
>>   arch/x86/Kconfig                             |  2 +-
>>   arch/x86/include/asm/hugetlb.h               |  7 +++--
>>   fs/Kconfig                                   |  2 +-
>>   include/linux/gfp.h                          |  2 +-
>>   mm/hugetlb.c                                 | 43 +++++++++++++++-------------
>>   mm/page_alloc.c                              |  4 +--
>>   12 files changed, 48 insertions(+), 36 deletions(-)
>>
>> diff --git a/arch/arm64/Kconfig b/arch/arm64/Kconfig
>> index a4168d366127..18239cbd7fcd 100644
>> --- a/arch/arm64/Kconfig
>> +++ b/arch/arm64/Kconfig
>> @@ -18,7 +18,7 @@ config ARM64
>>   	select ARCH_HAS_FAST_MULTIPLIER
>>   	select ARCH_HAS_FORTIFY_SOURCE
>>   	select ARCH_HAS_GCOV_PROFILE_ALL
>> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
>> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>>   	select ARCH_HAS_KCOV
>>   	select ARCH_HAS_MEMBARRIER_SYNC_CORE
>>   	select ARCH_HAS_PTE_SPECIAL
>> diff --git a/arch/arm64/include/asm/hugetlb.h b/arch/arm64/include/asm/hugetlb.h
>> index fb6609875455..797fc77eabcd 100644
>> --- a/arch/arm64/include/asm/hugetlb.h
>> +++ b/arch/arm64/include/asm/hugetlb.h
>> @@ -65,8 +65,11 @@ extern void set_huge_swap_pte_at(struct mm_struct *mm, unsigned long addr,
>>   
>>   #include <asm-generic/hugetlb.h>
>>   
>> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
>> -static inline bool gigantic_page_supported(void) { return true; }
>> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>> +static inline bool gigantic_page_runtime_allocation_supported(void)
>> +{
>> +	return true;
>> +}
>>   #endif
>>   
>>   #endif /* __ASM_HUGETLB_H */
>> diff --git a/arch/powerpc/include/asm/book3s/64/hugetlb.h b/arch/powerpc/include/asm/book3s/64/hugetlb.h
>> index 5b0177733994..7711f0e2c7e5 100644
>> --- a/arch/powerpc/include/asm/book3s/64/hugetlb.h
>> +++ b/arch/powerpc/include/asm/book3s/64/hugetlb.h
>> @@ -32,8 +32,8 @@ static inline int hstate_get_psize(struct hstate *hstate)
>>   	}
>>   }
>>   
>> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
>> -static inline bool gigantic_page_supported(void)
>> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>> +static inline bool gigantic_page_runtime_allocation_supported(void)
>>   {
>>   	return true;
>>   }
>> diff --git a/arch/powerpc/platforms/Kconfig.cputype b/arch/powerpc/platforms/Kconfig.cputype
>> index 8c7464c3f27f..779e06bac697 100644
>> --- a/arch/powerpc/platforms/Kconfig.cputype
>> +++ b/arch/powerpc/platforms/Kconfig.cputype
>> @@ -319,7 +319,7 @@ config ARCH_ENABLE_SPLIT_PMD_PTLOCK
>>   config PPC_RADIX_MMU
>>   	bool "Radix MMU Support"
>>   	depends on PPC_BOOK3S_64
>> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
>> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>>   	default y
>>   	help
>>   	  Enable support for the Power ISA 3.0 Radix style MMU. Currently this
>> diff --git a/arch/s390/Kconfig b/arch/s390/Kconfig
>> index ed554b09eb3f..6776eef6a9ae 100644
>> --- a/arch/s390/Kconfig
>> +++ b/arch/s390/Kconfig
>> @@ -69,7 +69,7 @@ config S390
>>   	select ARCH_HAS_ELF_RANDOMIZE
>>   	select ARCH_HAS_FORTIFY_SOURCE
>>   	select ARCH_HAS_GCOV_PROFILE_ALL
>> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
>> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>>   	select ARCH_HAS_KCOV
>>   	select ARCH_HAS_PTE_SPECIAL
>>   	select ARCH_HAS_SET_MEMORY
>> diff --git a/arch/s390/include/asm/hugetlb.h b/arch/s390/include/asm/hugetlb.h
>> index 2d1afa58a4b6..57c952f5388e 100644
>> --- a/arch/s390/include/asm/hugetlb.h
>> +++ b/arch/s390/include/asm/hugetlb.h
>> @@ -116,7 +116,10 @@ static inline pte_t huge_pte_modify(pte_t pte, pgprot_t newprot)
>>   	return pte_modify(pte, newprot);
>>   }
>>   
>> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
>> -static inline bool gigantic_page_supported(void) { return true; }
>> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>> +static inline bool gigantic_page_runtime_allocation_supported(void)
>> +{
>> +	return true;
>> +}
>>   #endif
>>   #endif /* _ASM_S390_HUGETLB_H */
>> diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
>> index 6185d4f33296..a88f5a4311c9 100644
>> --- a/arch/x86/Kconfig
>> +++ b/arch/x86/Kconfig
>> @@ -23,7 +23,7 @@ config X86_64
>>   	def_bool y
>>   	depends on 64BIT
>>   	# Options that are inherently 64-bit kernel only:
>> -	select ARCH_HAS_GIGANTIC_PAGE if (MEMORY_ISOLATION && COMPACTION) || CMA
>> +	select ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION if (MEMORY_ISOLATION && COMPACTION) || CMA
>>   	select ARCH_SUPPORTS_INT128
>>   	select ARCH_USE_CMPXCHG_LOCKREF
>>   	select HAVE_ARCH_SOFT_DIRTY
>> diff --git a/arch/x86/include/asm/hugetlb.h b/arch/x86/include/asm/hugetlb.h
>> index 7469d321f072..5a5e7119ced4 100644
>> --- a/arch/x86/include/asm/hugetlb.h
>> +++ b/arch/x86/include/asm/hugetlb.h
>> @@ -17,8 +17,11 @@ static inline void arch_clear_hugepage_flags(struct page *page)
>>   {
>>   }
>>   
>> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
>> -static inline bool gigantic_page_supported(void) { return true; }
>> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>> +static inline bool gigantic_page_runtime_allocation_supported(void)
>> +{
>> +	return true;
>> +}
>>   #endif
>>   
>>   #endif /* _ASM_X86_HUGETLB_H */
>> diff --git a/fs/Kconfig b/fs/Kconfig
>> index ac474a61be37..4192d1fde0f0 100644
>> --- a/fs/Kconfig
>> +++ b/fs/Kconfig
>> @@ -207,7 +207,7 @@ config HUGETLB_PAGE
>>   config MEMFD_CREATE
>>   	def_bool TMPFS || HUGETLBFS
>>   
>> -config ARCH_HAS_GIGANTIC_PAGE
>> +config ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>>   	bool
>>   
>>   source "fs/configfs/Kconfig"
>> diff --git a/include/linux/gfp.h b/include/linux/gfp.h
>> index 5f5e25fd6149..79ff86fabd42 100644
>> --- a/include/linux/gfp.h
>> +++ b/include/linux/gfp.h
>> @@ -589,8 +589,8 @@ static inline bool pm_suspended_storage(void)
>>   /* The below functions must be run on a range from a single zone. */
>>   extern int alloc_contig_range(unsigned long start, unsigned long end,
>>   			      unsigned migratetype, gfp_t gfp_mask);
>> -extern void free_contig_range(unsigned long pfn, unsigned nr_pages);
>>   #endif
>> +extern void free_contig_range(unsigned long pfn, unsigned int nr_pages);
>>   
>>   #ifdef CONFIG_CMA
>>   /* CMA stuff */
>> diff --git a/mm/hugetlb.c b/mm/hugetlb.c
>> index 745088810965..9893ba26b3b8 100644
>> --- a/mm/hugetlb.c
>> +++ b/mm/hugetlb.c
>> @@ -1035,7 +1035,6 @@ static int hstate_next_node_to_free(struct hstate *h, nodemask_t *nodes_allowed)
>>   		((node = hstate_next_node_to_free(hs, mask)) || 1);	\
>>   		nr_nodes--)
>>   
>> -#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE
>>   static void destroy_compound_gigantic_page(struct page *page,
>>   					unsigned int order)
>>   {
>> @@ -1058,6 +1057,7 @@ static void free_gigantic_page(struct page *page, unsigned int order)
>>   	free_contig_range(page_to_pfn(page), 1 << order);
>>   }
>>   
>> +#ifdef CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION
>>   static int __alloc_gigantic_page(unsigned long start_pfn,
>>   				unsigned long nr_pages, gfp_t gfp_mask)
>>   {
>> @@ -1143,22 +1143,19 @@ static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
>>   static void prep_new_huge_page(struct hstate *h, struct page *page, int nid);
>>   static void prep_compound_gigantic_page(struct page *page, unsigned int order);
>>   
>> -#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE */
>> -static inline bool gigantic_page_supported(void) { return false; }
>> +#else /* !CONFIG_ARCH_HAS_GIGANTIC_PAGE_RUNTIME_ALLOCATION */
>> +static inline bool gigantic_page_runtime_allocation_supported(void)
>> +{
>> +	return false;
>> +}
>>   static struct page *alloc_gigantic_page(struct hstate *h, gfp_t gfp_mask,
>>   		int nid, nodemask_t *nodemask) { return NULL; }
>> -static inline void free_gigantic_page(struct page *page, unsigned int order) { }
>> -static inline void destroy_compound_gigantic_page(struct page *page,
>> -						unsigned int order) { }
>>   #endif
>>   
>>   static void update_and_free_page(struct hstate *h, struct page *page)
>>   {
>>   	int i;
>>   
>> -	if (hstate_is_gigantic(h) && !gigantic_page_supported())
>> -		return;
>> -
>>   	h->nr_huge_pages--;
>>   	h->nr_huge_pages_node[page_to_nid(page)]--;
>>   	for (i = 0; i < pages_per_huge_page(h); i++) {
>> @@ -2276,13 +2273,20 @@ static int adjust_pool_surplus(struct hstate *h, nodemask_t *nodes_allowed,
>>   }
>>   
>>   #define persistent_huge_pages(h) (h->nr_huge_pages - h->surplus_huge_pages)
>> -static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
>> +static int set_max_huge_pages(struct hstate *h, unsigned long count,
>>   						nodemask_t *nodes_allowed)
>>   {
>>   	unsigned long min_count, ret;
>>   
>> -	if (hstate_is_gigantic(h) && !gigantic_page_supported())
>> -		return h->max_huge_pages;
>> +	if (hstate_is_gigantic(h) &&
>> +		!gigantic_page_runtime_allocation_supported()) {
>> +		spin_lock(&hugetlb_lock);
>> +		if (count > persistent_huge_pages(h)) {
>> +			spin_unlock(&hugetlb_lock);
>> +			return -EINVAL;
>> +		}
>> +		goto decrease_pool;
>> +	}
>>   
>>   	/*
>>   	 * Increase the pool size
>> @@ -2322,6 +2326,7 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
>>   			goto out;
>>   	}
>>   
>> +decrease_pool:
>>   	/*
>>   	 * Decrease the pool size
>>   	 * First return free pages to the buddy allocator (being careful
>> @@ -2350,9 +2355,10 @@ static unsigned long set_max_huge_pages(struct hstate *h, unsigned long count,
>>   			break;
>>   	}
>>   out:
>> -	ret = persistent_huge_pages(h);
>> +	h->max_huge_pages = persistent_huge_pages(h);
>>   	spin_unlock(&hugetlb_lock);
>> -	return ret;
>> +
>> +	return 0;
>>   }
>>   
>>   #define HSTATE_ATTR_RO(_name) \
>> @@ -2404,11 +2410,6 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>>   	int err;
>>   	NODEMASK_ALLOC(nodemask_t, nodes_allowed, GFP_KERNEL | __GFP_NORETRY);
>>   
>> -	if (hstate_is_gigantic(h) && !gigantic_page_supported()) {
>> -		err = -EINVAL;
>> -		goto out;
>> -	}
>> -
>>   	if (nid == NUMA_NO_NODE) {
>>   		/*
>>   		 * global hstate attribute
>> @@ -2428,7 +2429,9 @@ static ssize_t __nr_hugepages_store_common(bool obey_mempolicy,
>>   	} else
>>   		nodes_allowed = &node_states[N_MEMORY];
>>   
>> -	h->max_huge_pages = set_max_huge_pages(h, count, nodes_allowed);
>> +	err = set_max_huge_pages(h, count, nodes_allowed);
>> +	if (err)
>> +		goto out;
>>   
>>   	if (nodes_allowed != &node_states[N_MEMORY])
>>   		NODEMASK_FREE(nodes_allowed);
>> diff --git a/mm/page_alloc.c b/mm/page_alloc.c
>> index cde5dac6229a..81b931db85a1 100644
>> --- a/mm/page_alloc.c
>> +++ b/mm/page_alloc.c
>> @@ -8241,8 +8241,9 @@ int alloc_contig_range(unsigned long start, unsigned long end,
>>   				pfn_max_align_up(end), migratetype);
>>   	return ret;
>>   }
>> +#endif
>>   
>> -void free_contig_range(unsigned long pfn, unsigned nr_pages)
>> +void free_contig_range(unsigned long pfn, unsigned int nr_pages)
>>   {
>>   	unsigned int count = 0;
>>   
>> @@ -8254,7 +8255,6 @@ void free_contig_range(unsigned long pfn, unsigned nr_pages)
>>   	}
>>   	WARN(count != 0, "%d pages are still in use!\n", count);
>>   }
>> -#endif
>>   
>>   #ifdef CONFIG_MEMORY_HOTPLUG
>>   /*
>> -- 
>> 2.16.2

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
  2019-01-17 18:39 [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration Alexandre Ghiti
                   ` (4 preceding siblings ...)
  2019-02-05 11:23 ` Michael Ellerman
@ 2019-02-13 11:27 ` Vlastimil Babka
  2019-02-13 19:22   ` Alex Ghiti
  5 siblings, 1 reply; 10+ messages in thread
From: Vlastimil Babka @ 2019-02-13 11:27 UTC (permalink / raw)
  To: Alexandre Ghiti, Catalin Marinas, Will Deacon,
	Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
	Martin Schwidefsky, Heiko Carstens, Thomas Gleixner, Ingo Molnar,
	Borislav Petkov, H . Peter Anvin, x86, Alexander Viro,
	Mike Kravetz, linux-arm-kernel, linux-kernel, linuxppc-dev,
	linux-s390, linux-fsdevel, linux-mm
  Cc: linux-riscv, hch, Alexandre Ghiti

On 1/17/19 7:39 PM, Alexandre Ghiti wrote:
> From: Alexandre Ghiti <alex@ghiti.fr>
> 
> On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
> that support gigantic pages, boottime reserved gigantic pages can not be
> freed at all. This patchs simply enables the possibility to hand back
> those pages to memory allocator.
> 
> This commit then renames gigantic_page_supported and
> ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
> being false does not mean that the system cannot use gigantic pages: it
> just means that runtime allocation of gigantic pages is not supported,
> one can still allocate boottime gigantic pages if the architecture supports
> it.
> 
> Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>

I'm fine with the change, but wonder if this can be structured better in a way
which would remove the duplicated "if (MEMORY_ISOLATION && COMPACTION) || CMA"
from all arches, as well as the duplicated
gigantic_page_runtime_allocation_supported()

something like:

- "select ARCH_HAS_GIGANTIC_PAGE" has no conditions, it just says the arch can
support them either at boottime or runtime (but runtime is usable only if other
conditions are met)
- gigantic_page_runtime_allocation_supported() is a function that returns true
if ARCH_HAS_GIGANTIC_PAGE && ((MEMORY_ISOLATION && COMPACTION) || CMA) and
there's a single instance, not per-arch.
- code for freeing gigantic pages can probably still be conditional on
ARCH_HAS_GIGANTIC_PAGE

BTW I wanted also to do something about the "(MEMORY_ISOLATION && COMPACTION) ||
CMA" ugliness itself, i.e. put the common parts behind some new kconfig
(COMPACTION_CORE ?) and expose it better to users, but I can take a stab on that
once the above part is settled.

Vlastimil

^ permalink raw reply	[flat|nested] 10+ messages in thread

* Re: [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration
  2019-02-13 11:27 ` Vlastimil Babka
@ 2019-02-13 19:22   ` Alex Ghiti
  0 siblings, 0 replies; 10+ messages in thread
From: Alex Ghiti @ 2019-02-13 19:22 UTC (permalink / raw)
  To: Vlastimil Babka, Catalin Marinas, Will Deacon,
	Benjamin Herrenschmidt, Paul Mackerras, Michael Ellerman,
	Martin Schwidefsky, Heiko Carstens, Thomas Gleixner, Ingo Molnar,
	Borislav Petkov, H . Peter Anvin, x86, Alexander Viro,
	Mike Kravetz, linux-arm-kernel, linux-kernel, linuxppc-dev,
	linux-s390, linux-fsdevel, linux-mm
  Cc: linux-riscv, hch


On 2/13/19 6:27 AM, Vlastimil Babka wrote:
> On 1/17/19 7:39 PM, Alexandre Ghiti wrote:
>> From: Alexandre Ghiti <alex@ghiti.fr>
>>
>> On systems without CMA or (MEMORY_ISOLATION && COMPACTION) activated but
>> that support gigantic pages, boottime reserved gigantic pages can not be
>> freed at all. This patchs simply enables the possibility to hand back
>> those pages to memory allocator.
>>
>> This commit then renames gigantic_page_supported and
>> ARCH_HAS_GIGANTIC_PAGE to make them more accurate. Indeed, those values
>> being false does not mean that the system cannot use gigantic pages: it
>> just means that runtime allocation of gigantic pages is not supported,
>> one can still allocate boottime gigantic pages if the architecture supports
>> it.
>>
>> Signed-off-by: Alexandre Ghiti <alex@ghiti.fr>
> I'm fine with the change, but wonder if this can be structured better in a way
> which would remove the duplicated "if (MEMORY_ISOLATION && COMPACTION) || CMA"
> from all arches, as well as the duplicated
> gigantic_page_runtime_allocation_supported()


Yeah, totally, we can factorize more than what I did. I prepared a v2 of 
this
patch that does exactly that: remove the triplet from arch specific code
and the duplicated gigantic_page_runtime_allocation_supported.


> something like:
>
> - "select ARCH_HAS_GIGANTIC_PAGE" has no conditions, it just says the arch can
> support them either at boottime or runtime (but runtime is usable only if other
> conditions are met)


And the v2 gets rid of ARCH_HAS_GIGANTIC_PAGE totally since it
is not needed by arch to advertise the fact they support gigantic page,
actually, when selected, it really just means that an arch has the means
to allocate runtime gigantic page: it is equivalent to
(MEMORY_ISOLATION && COMPACTION) || CMA.


> - gigantic_page_runtime_allocation_supported() is a function that returns true
> if ARCH_HAS_GIGANTIC_PAGE && ((MEMORY_ISOLATION && COMPACTION) || CMA) and
> there's a single instance, not per-arch.
> - code for freeing gigantic pages can probably still be conditional on
> ARCH_HAS_GIGANTIC_PAGE
>
> BTW I wanted also to do something about the "(MEMORY_ISOLATION && COMPACTION) ||
> CMA" ugliness itself, i.e. put the common parts behind some new kconfig
> (COMPACTION_CORE ?) and expose it better to users, but I can take a stab on that
> once the above part is settled.
> Vlastimil


I send the v2 right away, if you can take a look Vlastimil, that would 
be great.
Note that Andrew already picked this patch in its tree, I'm not sure how to
proceed.


Thanks for your remarks !


Alex


^ permalink raw reply	[flat|nested] 10+ messages in thread

end of thread, back to index

Thread overview: 10+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2019-01-17 18:39 [PATCH] hugetlb: allow to free gigantic pages regardless of the configuration Alexandre Ghiti
2019-01-17 18:39 ` Alexandre Ghiti
2019-01-17 20:09 ` Mike Kravetz
2019-01-17 20:09   ` Mike Kravetz
2019-01-19 23:57 ` Will Deacon
2019-02-03 18:17 ` Alex Ghiti
2019-02-05 11:23 ` Michael Ellerman
2019-02-05 11:35   ` Alex Ghiti
2019-02-13 11:27 ` Vlastimil Babka
2019-02-13 19:22   ` Alex Ghiti

Linux-Fsdevel Archive on lore.kernel.org

Archives are clonable:
	git clone --mirror https://lore.kernel.org/linux-fsdevel/0 linux-fsdevel/git/0.git

	# If you have public-inbox 1.1+ installed, you may
	# initialize and index your mirror using the following commands:
	public-inbox-init -V2 linux-fsdevel linux-fsdevel/ https://lore.kernel.org/linux-fsdevel \
		linux-fsdevel@vger.kernel.org linux-fsdevel@archiver.kernel.org
	public-inbox-index linux-fsdevel


Newsgroup available over NNTP:
	nntp://nntp.lore.kernel.org/org.kernel.vger.linux-fsdevel


AGPL code for this site: git clone https://public-inbox.org/ public-inbox