linux-kernel - [PATCH v2 2/2] arm64/mm: add speculative page fault

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [day] [month] [year] [list]

Date:   Fri,  4 May 2018 14:57:49 +0800
From:   Ganesh Mahendran <opensource.ganesh@...il.com>
To:     ldufour@...ux.vnet.ibm.com, catalin.marinas@....com,
        will.deacon@....com, mark.rutland@....com, cpandya@...eaurora.org,
        punit.agrawal@....com
Cc:     linux-arm-kernel@...ts.infradead.org, linux-kernel@...r.kernel.org,
        linux-mm@...ck.org, Ganesh Mahendran <opensource.ganesh@...il.com>
Subject: [PATCH v2 2/2] arm64/mm: add speculative page fault

This patch enables the speculative page fault on the arm64
architecture.

I completed spf porting in 4.9. From the test result,
we can see app launching time improved by about 10% in average.
For the apps which have more than 50 threads, 15% or even more
improvement can be got.

Signed-off-by: Ganesh Mahendran <opensource.ganesh@...il.com>
---
v2:
  move find_vma() to do_page_fault()
  remove IS_ENABLED()
  remove fault != VM_FAULT_SIGSEGV check
  initilize vma = NULL
---
 arch/arm64/mm/fault.c | 29 +++++++++++++++++++++++++----
 1 file changed, 25 insertions(+), 4 deletions(-)

diff --git a/arch/arm64/mm/fault.c b/arch/arm64/mm/fault.c
index 4165485..efd5956 100644
--- a/arch/arm64/mm/fault.c
+++ b/arch/arm64/mm/fault.c
@@ -320,14 +320,12 @@ static void do_bad_area(unsigned long addr, unsigned int esr, struct pt_regs *re
 #define VM_FAULT_BADMAP		0x010000
 #define VM_FAULT_BADACCESS	0x020000
 
-static int __do_page_fault(struct mm_struct *mm, unsigned long addr,
+static int __do_page_fault(struct vm_area_struct *vma, unsigned long addr,
 			   unsigned int mm_flags, unsigned long vm_flags,
 			   struct task_struct *tsk)
 {
-	struct vm_area_struct *vma;
 	int fault;
 
-	vma = find_vma(mm, addr);
 	fault = VM_FAULT_BADMAP;
 	if (unlikely(!vma))
 		goto out;
@@ -371,6 +369,7 @@ static int __kprobes do_page_fault(unsigned long addr, unsigned int esr,
 	int fault, major = 0;
 	unsigned long vm_flags = VM_READ | VM_WRITE;
 	unsigned int mm_flags = FAULT_FLAG_ALLOW_RETRY | FAULT_FLAG_KILLABLE;
+	struct vm_area_struct *vma = NULL;
 
 	if (notify_page_fault(regs, esr))
 		return 0;
@@ -410,6 +409,16 @@ static int __kprobes do_page_fault(unsigned long addr, unsigned int esr,
 	perf_sw_event(PERF_COUNT_SW_PAGE_FAULTS, 1, regs, addr);
 
 	/*
+	 * let's try a speculative page fault without grabbing the
+	 * mmap_sem.
+	 */
+	fault = handle_speculative_fault(mm, addr, mm_flags, &vma);
+	if (fault != VM_FAULT_RETRY) {
+		perf_sw_event(PERF_COUNT_SW_SPF, 1, regs, addr);
+		goto done;
+	}
+
+	/*
 	 * As per x86, we may deadlock here. However, since the kernel only
 	 * validly references user space from well defined areas of the code,
 	 * we can bug out early if this is from code which shouldn't.
@@ -431,7 +440,10 @@ static int __kprobes do_page_fault(unsigned long addr, unsigned int esr,
 #endif
 	}
 
-	fault = __do_page_fault(mm, addr, mm_flags, vm_flags, tsk);
+	if (!vma || !can_reuse_spf_vma(vma, addr))
+		vma = find_vma(mm, addr);
+
+	fault = __do_page_fault(vma, addr, mm_flags, vm_flags, tsk);
 	major |= fault & VM_FAULT_MAJOR;
 
 	if (fault & VM_FAULT_RETRY) {
@@ -454,11 +466,20 @@ static int __kprobes do_page_fault(unsigned long addr, unsigned int esr,
 		if (mm_flags & FAULT_FLAG_ALLOW_RETRY) {
 			mm_flags &= ~FAULT_FLAG_ALLOW_RETRY;
 			mm_flags |= FAULT_FLAG_TRIED;
+
+			/*
+			 * Do not try to reuse this vma and fetch it
+			 * again since we will release the mmap_sem.
+			 */
+			vma = NULL;
+
 			goto retry;
 		}
 	}
 	up_read(&mm->mmap_sem);
 
+done:
+
 	/*
 	 * Handle the "normal" (no error) case first.
 	 */
-- 
1.9.1