public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Rik van Riel <riel@surriel.com>
To: linux-kernel@vger.kernel.org
Cc: kernel-team@fb.com, mingo@kernel.org, peterz@infradead.org,
	luto@kernel.org, x86@kernel.org, efault@gmx.de,
	dave.hansen@intel.com, Rik van Riel <riel@surriel.com>
Subject: [PATCH 09/11] mm,x86: shoot down lazy TLB references at exit_mmap time
Date: Wed,  1 Aug 2018 06:02:53 -0400	[thread overview]
Message-ID: <20180801100255.4278-10-riel@surriel.com> (raw)
In-Reply-To: <20180801100255.4278-1-riel@surriel.com>

Shooting down lazy TLB references to an mm at exit_mmap time ensures
that no users of the mm_struct will be left anywhere in the system,
allowing it to be torn down and freed immediately.

Signed-off-by: Rik van Riel <riel@surriel.com>
Suggested-by: Andy Lutomirski <luto@kernel.org>
Suggested-by: Peter Zijlstra <peterz@infradead.org>
---
 arch/x86/Kconfig                   |  1 +
 arch/x86/include/asm/mmu_context.h |  1 +
 arch/x86/include/asm/tlbflush.h    |  2 ++
 arch/x86/mm/tlb.c                  | 15 +++++++++++++++
 4 files changed, 19 insertions(+)

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 6d4774f203d0..ecdfc6933203 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -75,6 +75,7 @@ config X86
 	select ARCH_MIGHT_HAVE_ACPI_PDC		if ACPI
 	select ARCH_MIGHT_HAVE_PC_PARPORT
 	select ARCH_MIGHT_HAVE_PC_SERIO
+	select ARCH_NO_ACTIVE_MM_REFCOUNTING
 	select ARCH_SUPPORTS_ATOMIC_RMW
 	select ARCH_SUPPORTS_NUMA_BALANCING	if X86_64
 	select ARCH_USE_BUILTIN_BSWAP
diff --git a/arch/x86/include/asm/mmu_context.h b/arch/x86/include/asm/mmu_context.h
index eeeb9289c764..529bf7bc5f75 100644
--- a/arch/x86/include/asm/mmu_context.h
+++ b/arch/x86/include/asm/mmu_context.h
@@ -238,6 +238,7 @@ static inline void arch_exit_mmap(struct mm_struct *mm)
 {
 	paravirt_arch_exit_mmap(mm);
 	ldt_arch_exit_mmap(mm);
+	lazy_tlb_exit_mmap(mm);
 }
 
 #ifdef CONFIG_X86_64
diff --git a/arch/x86/include/asm/tlbflush.h b/arch/x86/include/asm/tlbflush.h
index 511bf5fae8b8..3966a45367cd 100644
--- a/arch/x86/include/asm/tlbflush.h
+++ b/arch/x86/include/asm/tlbflush.h
@@ -538,6 +538,8 @@ extern void arch_tlbbatch_flush(struct arch_tlbflush_unmap_batch *batch);
 	native_flush_tlb_others(mask, info)
 #endif
 
+extern void lazy_tlb_exit_mmap(struct mm_struct *mm);
+
 extern void tlb_flush_remove_tables(struct mm_struct *mm);
 extern void tlb_flush_remove_tables_local(void *arg);
 
diff --git a/arch/x86/mm/tlb.c b/arch/x86/mm/tlb.c
index ea4ef5ceaba2..7b1add904396 100644
--- a/arch/x86/mm/tlb.c
+++ b/arch/x86/mm/tlb.c
@@ -713,6 +713,21 @@ void tlb_flush_remove_tables(struct mm_struct *mm)
 	put_cpu();
 }
 
+/*
+ * At exit or execve time, all other threads of a process have disappeared,
+ * but other CPUs could still be referencing this mm in lazy TLB mode.
+ * Get rid of those references before releasing the mm.
+ */
+void lazy_tlb_exit_mmap(struct mm_struct *mm)
+{
+	int cpu = get_cpu();
+
+	if (cpumask_any_but(mm_cpumask(mm), cpu) < nr_cpu_ids)
+		on_each_cpu_mask(mm_cpumask(mm), leave_mm, NULL, 1);
+
+	put_cpu();
+}
+
 static void do_flush_tlb_all(void *info)
 {
 	count_vm_tlb_event(NR_TLB_REMOTE_FLUSH_RECEIVED);
-- 
2.14.4


  parent reply	other threads:[~2018-08-01 10:03 UTC|newest]

Thread overview: 17+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-08-01 10:02 [PATCH v2 0/11] x86,tlb,mm: more lazy TLB cleanups & optimizations Rik van Riel
2018-08-01 10:02 ` [PATCH 01/11] x86,tlb: clarify memory barrier in switch_mm_irqs_off Rik van Riel
2018-08-01 10:02 ` [PATCH 02/11] smp: use __cpumask_set_cpu in on_each_cpu_cond Rik van Riel
2018-08-01 10:02 ` [PATCH 03/11] smp,cpumask: introduce on_each_cpu_cond_mask Rik van Riel
2018-08-01 10:02 ` [PATCH 04/11] x86,mm: use on_each_cpu_cond for TLB flushes Rik van Riel
2018-08-01 10:02 ` [PATCH 05/11] mm,tlb: turn dummy defines into inline functions Rik van Riel
2018-08-01 10:02 ` [PATCH 06/11] mm,x86: skip cr4 and ldt reload when mm stays the same Rik van Riel
2018-08-01 10:02 ` [PATCH 07/11] x86,mm: remove leave_mm cpu argument Rik van Riel
2018-08-01 10:02 ` [PATCH 08/11] arch,mm: add config variable to skip lazy TLB mm refcounting Rik van Riel
2018-08-01 10:02 ` Rik van Riel [this message]
2018-08-01 10:02 ` [PATCH 10/11] x86,tlb: really leave mm on shootdown Rik van Riel
2018-08-01 10:02 ` [PATCH 11/11] mm,sched: conditionally skip lazy TLB mm refcounting Rik van Riel
2018-08-03 15:56   ` Peter Zijlstra
2018-08-03 16:40     ` Rik van Riel
2018-08-03 17:25       ` Peter Zijlstra
2018-08-03 17:39         ` Rik van Riel
2018-08-06 17:50           ` Peter Zijlstra

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=20180801100255.4278-10-riel@surriel.com \
    --to=riel@surriel.com \
    --cc=dave.hansen@intel.com \
    --cc=efault@gmx.de \
    --cc=kernel-team@fb.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=mingo@kernel.org \
    --cc=peterz@infradead.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