linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Ankur Arora <ankur.a.arora@oracle.com>
To: linux-kernel@vger.kernel.org, linux-mm@kvack.org, x86@kernel.org
Cc: torvalds@linux-foundation.org, akpm@linux-foundation.org,
	mike.kravetz@oracle.com, mingo@kernel.org, luto@kernel.org,
	tglx@linutronix.de, bp@alien8.de, peterz@infradead.org,
	ak@linux.intel.com, arnd@arndb.de, jgg@nvidia.com,
	jon.grimm@amd.com, boris.ostrovsky@oracle.com,
	konrad.wilk@oracle.com, joao.m.martins@oracle.com,
	ankur.a.arora@oracle.com
Subject: [PATCH v3 14/21] x86/clear_page: add clear_pages_incoherent()
Date: Mon,  6 Jun 2022 20:37:18 +0000	[thread overview]
Message-ID: <20220606203725.1313715-10-ankur.a.arora@oracle.com> (raw)
In-Reply-To: <20220606202109.1306034-1-ankur.a.arora@oracle.com>

Expose incoherent clearing primitives (clear_pages_movnt(),
clear_pages_clzero()) as alternatives via clear_pages_incoherent().

Fallback to clear_pages() if, X86_FEATURE_MOVNT_SLOW is set and
the CPU does not have X86_FEATURE_CLZERO.

Both these primitives use weakly-ordered stores. To ensure that
callers don't mix accesses to different types of address_spaces,
annotate clear_user_pages_incoherent(), and clear_pages_incoherent()
as taking __incoherent pointers as arguments.

Also add clear_page_make_coherent() which provides the necessary
store fence to make access to these __incoherent regions safe.

Signed-off-by: Ankur Arora <ankur.a.arora@oracle.com>
---
 arch/x86/include/asm/page.h    | 13 +++++++++++++
 arch/x86/include/asm/page_64.h | 34 ++++++++++++++++++++++++++++++++++
 2 files changed, 47 insertions(+)

diff --git a/arch/x86/include/asm/page.h b/arch/x86/include/asm/page.h
index 045eaab08f43..8fc6cc6759b9 100644
--- a/arch/x86/include/asm/page.h
+++ b/arch/x86/include/asm/page.h
@@ -40,6 +40,19 @@ static inline void clear_user_page(void *page, unsigned long vaddr,
 	clear_page(page);
 }
 
+#ifdef __HAVE_ARCH_CLEAR_USER_PAGES_INCOHERENT /* x86_64 */
+/*
+ * clear_pages_incoherent: valid on only __incoherent memory regions.
+ */
+static inline void clear_user_pages_incoherent(__incoherent void *page,
+					       unsigned long vaddr,
+					       struct page *pg,
+					       unsigned int npages)
+{
+	clear_pages_incoherent(page, npages);
+}
+#endif /* __HAVE_ARCH_CLEAR_USER_PAGES_INCOHERENT */
+
 static inline void copy_user_page(void *to, void *from, unsigned long vaddr,
 				  struct page *topage)
 {
diff --git a/arch/x86/include/asm/page_64.h b/arch/x86/include/asm/page_64.h
index e8d4698fda65..78417f63f522 100644
--- a/arch/x86/include/asm/page_64.h
+++ b/arch/x86/include/asm/page_64.h
@@ -69,6 +69,40 @@ static inline void clear_pages(void *page, unsigned int npages)
 			   : "cc", "memory", "rax", "rcx");
 }
 
+#define __HAVE_ARCH_CLEAR_USER_PAGES_INCOHERENT
+/*
+ * clear_pages_incoherent: only allowed on __incoherent memory regions.
+ */
+static inline void clear_pages_incoherent(__incoherent void *page,
+					  unsigned int npages)
+{
+	alternative_call_2(clear_pages_movnt,
+			   clear_pages, X86_FEATURE_MOVNT_SLOW,
+			   clear_pages_clzero, X86_FEATURE_CLZERO,
+			   "=D" (page), "S" ((unsigned long) npages),
+			   "0" (page)
+			   : "cc", "memory", "rax", "rcx");
+}
+
+/*
+ * clear_page_make_coherent: execute the necessary store fence
+ * after which __incoherent regions can be safely accessed.
+ */
+static inline void clear_page_make_coherent(void)
+{
+	/*
+	 * Keep the sfence for oldinstr and clzero separate to guard against
+	 * the possibility that a CPU has both X86_FEATURE_MOVNT_SLOW and
+	 * X86_FEATURE_CLZERO.
+	 *
+	 * The alternatives need to be in the same order as the ones
+	 * in clear_pages_incoherent().
+	 */
+	alternative_2("sfence",
+		      "", X86_FEATURE_MOVNT_SLOW,
+		      "sfence", X86_FEATURE_CLZERO);
+}
+
 void copy_page(void *to, void *from);
 
 #ifdef CONFIG_X86_5LEVEL
-- 
2.31.1



  parent reply	other threads:[~2022-06-06 20:38 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-06 20:20 [PATCH v3 00/21] huge page clearing optimizations Ankur Arora
2022-06-06 20:20 ` [PATCH v3 01/21] mm, huge-page: reorder arguments to process_huge_page() Ankur Arora
2022-06-06 20:20 ` [PATCH v3 02/21] mm, huge-page: refactor process_subpage() Ankur Arora
2022-06-06 20:20 ` [PATCH v3 03/21] clear_page: add generic clear_user_pages() Ankur Arora
2022-06-06 20:20 ` [PATCH v3 04/21] mm, clear_huge_page: support clear_user_pages() Ankur Arora
2022-06-06 20:37 ` [PATCH v3 05/21] mm/huge_page: generalize process_huge_page() Ankur Arora
2022-06-06 20:37 ` [PATCH v3 06/21] x86/clear_page: add clear_pages() Ankur Arora
2022-06-06 20:37 ` [PATCH v3 07/21] x86/asm: add memset_movnti() Ankur Arora
2022-06-06 20:37 ` [PATCH v3 08/21] perf bench: " Ankur Arora
2022-06-06 20:37 ` [PATCH v3 09/21] x86/asm: add clear_pages_movnt() Ankur Arora
2022-06-10 22:11   ` Noah Goldstein
2022-06-10 22:15     ` Noah Goldstein
2022-06-12 11:18       ` Ankur Arora
2022-06-06 20:37 ` [PATCH v3 10/21] x86/asm: add clear_pages_clzero() Ankur Arora
2022-06-06 20:37 ` [PATCH v3 11/21] x86/cpuid: add X86_FEATURE_MOVNT_SLOW Ankur Arora
2022-06-06 20:37 ` [PATCH v3 12/21] sparse: add address_space __incoherent Ankur Arora
2022-06-06 20:37 ` [PATCH v3 13/21] clear_page: add generic clear_user_pages_incoherent() Ankur Arora
2022-06-08  0:01   ` Luc Van Oostenryck
2022-06-12 11:19     ` Ankur Arora
2022-06-06 20:37 ` Ankur Arora [this message]
2022-06-06 20:37 ` [PATCH v3 15/21] mm/clear_page: add clear_page_non_caching_threshold() Ankur Arora
2022-06-06 20:37 ` [PATCH v3 16/21] x86/clear_page: add arch_clear_page_non_caching_threshold() Ankur Arora
2022-06-06 20:37 ` [PATCH v3 17/21] clear_huge_page: use non-cached clearing Ankur Arora
2022-06-06 20:37 ` [PATCH v3 18/21] gup: add FOLL_HINT_BULK, FAULT_FLAG_NON_CACHING Ankur Arora
2022-06-06 20:37 ` [PATCH v3 19/21] gup: hint non-caching if clearing large regions Ankur Arora
2022-06-06 20:37 ` [PATCH v3 20/21] vfio_iommu_type1: specify FOLL_HINT_BULK to pin_user_pages() Ankur Arora
2022-06-06 20:37 ` [PATCH v3 21/21] x86/cpu/intel: set X86_FEATURE_MOVNT_SLOW for Skylake Ankur Arora
2022-06-06 21:53 ` [PATCH v3 00/21] huge page clearing optimizations Linus Torvalds
2022-06-07 15:08   ` Ankur Arora
2022-06-07 17:56     ` Linus Torvalds
2022-06-08 19:24       ` Ankur Arora
2022-06-08 19:39         ` Linus Torvalds
2022-06-08 20:21           ` Ankur Arora
2022-06-08 19:49       ` Matthew Wilcox
2022-06-08 19:51         ` Matthew Wilcox

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=20220606203725.1313715-10-ankur.a.arora@oracle.com \
    --to=ankur.a.arora@oracle.com \
    --cc=ak@linux.intel.com \
    --cc=akpm@linux-foundation.org \
    --cc=arnd@arndb.de \
    --cc=boris.ostrovsky@oracle.com \
    --cc=bp@alien8.de \
    --cc=jgg@nvidia.com \
    --cc=joao.m.martins@oracle.com \
    --cc=jon.grimm@amd.com \
    --cc=konrad.wilk@oracle.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=luto@kernel.org \
    --cc=mike.kravetz@oracle.com \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.org \
    --cc=tglx@linutronix.de \
    --cc=torvalds@linux-foundation.org \
    --cc=x86@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
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