mirror of https://lore.kernel.org/lkml/
 help / color / mirror / Atom feed
From: Andi Kleen <ak@suse.de>
To: zach@vmware.com, patches@x86-64.org, linux-kernel@vger.kernel.org
Subject: [PATCH] [27/34] i386: pte clear optimization
Date: Mon, 30 Apr 2007 17:50:00 +0200 (CEST)	[thread overview]
Message-ID: <20070430155000.7459E151D3@wotan.suse.de> (raw)
In-Reply-To: <20070430549.617098000@suse.de>


From: Zachary Amsden <zach@vmware.com>

When exiting from an address space, no special hypervisor notification of page
table updates needs to occur; direct page table hypervisors, such as Xen,
switch to another address space first (init_mm) and unprotects the page tables
to avoid the cost of trapping to the hypervisor for each pte_clear.  Shadow
mode hypervisors, such as VMI and lhype don't need to do the extra work of
calling through paravirt-ops, and can just directly clear the page table
entries without notifiying the hypervisor, since all the page tables are about
to be freed.

So introduce native_pte_clear functions which bypass any paravirt-ops
notification.  This results in a significant performance win for VMI and
removes some indirect calls from zap_pte_range.

Note the 3-level paging already had a native_pte_clear function, thus
demanding argument conformance and extra args for the 2-level definition.

Signed-off-by: Zachary Amsden <zach@vmware.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Andi Kleen <ak@suse.de>

---

 include/asm-i386/pgtable-2level.h |    5 +++++
 include/asm-i386/pgtable.h        |    2 +-
 2 files changed, 6 insertions(+), 1 deletion(-)

Index: linux/include/asm-i386/pgtable-2level.h
===================================================================
--- linux.orig/include/asm-i386/pgtable-2level.h
+++ linux/include/asm-i386/pgtable-2level.h
@@ -36,6 +36,11 @@ static inline void native_set_pmd(pmd_t 
 #define pte_clear(mm,addr,xp)	do { set_pte_at(mm, addr, xp, __pte(0)); } while (0)
 #define pmd_clear(xp)	do { set_pmd(xp, __pmd(0)); } while (0)
 
+static inline void native_pte_clear(struct mm_struct *mm, unsigned long addr, pte_t *xp)
+{
+	*xp = __pte(0);
+}
+
 static inline pte_t native_ptep_get_and_clear(pte_t *xp)
 {
 	return __pte(xchg(&xp->pte_low, 0));
Index: linux/include/asm-i386/pgtable.h
===================================================================
--- linux.orig/include/asm-i386/pgtable.h
+++ linux/include/asm-i386/pgtable.h
@@ -344,7 +344,7 @@ static inline pte_t ptep_get_and_clear_f
 	pte_t pte;
 	if (full) {
 		pte = *ptep;
-		pte_clear(mm, addr, ptep);
+		native_pte_clear(mm, addr, ptep);
 	} else {
 		pte = ptep_get_and_clear(mm, addr, ptep);
 	}

  parent reply	other threads:[~2007-04-30 15:51 UTC|newest]

Thread overview: 43+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-04-30 15:49 [PATCH] [0/34] x86 candidate patches for review VI: MTRR, APIC, more paravirt, misc Andi Kleen
2007-04-30 15:49 ` [PATCH] [1/34] x86_64: Move mtrr prototypes from proto.h to mtrr.h Andi Kleen
2007-04-30 15:49 ` [PATCH] [2/34] x86: Adds mtrr_save_fixed_ranges() for use in two later patches Andi Kleen
2007-04-30 15:49 ` [PATCH] [3/34] x86: Save the MTRRs of the BSP before booting an AP Andi Kleen
2007-04-30 15:49 ` [PATCH] [4/34] x86: Save and restore the fixed-range MTRRs of the BSP when suspending Andi Kleen
2007-04-30 15:49 ` [PATCH] [5/34] i386: Enable support for fixed-range IORRs to keep RdMem & WrMem in sync Andi Kleen
2007-04-30 15:49 ` [PATCH] [6/34] i386: safe_apic_wait_icr_idle - i386 Andi Kleen
2007-04-30 15:49 ` [PATCH] [7/34] x86_64: safe_apic_wait_icr_idle - x86_64 Andi Kleen
2007-04-30 15:49 ` [PATCH] [8/34] i386: use safe_apic_wait_icr_idle - i386 Andi Kleen
2007-04-30 15:49 ` [PATCH] [9/34] x86_64: use safe_apic_wait_icr_idle in smpboot.c - x86_64 Andi Kleen
2007-04-30 15:49 ` [PATCH] [10/34] i386: use safe_apic_wait_icr_idle in smpboot.c Andi Kleen
2007-04-30 15:49 ` [PATCH] [11/34] x86_64: use safe_apic_wait_icr_idle in smpboot.c - x86_64 Andi Kleen
2007-04-30 15:49 ` [PATCH] [12/34] i386: __send_IPI_dest_field - i386 Andi Kleen
2007-04-30 15:49 ` [PATCH] [13/34] x86_64: __send_IPI_dest_field - x86_64 Andi Kleen
2007-04-30 15:49 ` [PATCH] [14/34] i386: Use safe_apic_wait_icr_idle in safe_apic_wait_icr_idle - i386 Andi Kleen
2007-04-30 15:49 ` [PATCH] [15/34] x86_64: Use safe_apic_wait_icr_idle in __send_IPI_dest_field - x86_64 Andi Kleen
2007-04-30 15:49 ` [PATCH] [16/34] i386: fix mtrr sections Andi Kleen
2007-05-02 17:08   ` Bernhard Kaindl
2007-05-03 11:16     ` Bernhard Kaindl
2007-05-03 11:19       ` Andi Kleen
2007-05-03 15:12         ` Randy Dunlap
2007-04-30 15:49 ` [PATCH] [17/34] x86: PARAVIRT: Add a sched_clock paravirt_op Andi Kleen
2007-04-30 15:49 ` [PATCH] [18/34] i386: PARAVIRT: fix startup_ipi_hook config dependency Andi Kleen
2007-04-30 15:49 ` [PATCH] [19/34] x86_64: move __vgetcpu_mode & __jiffies to the vsyscall_2 zone Andi Kleen
2007-04-30 15:49 ` [PATCH] [20/34] i386: avoid checking for cpu gone when CONFIG_HOTPLUG_CPU not defined Andi Kleen
2007-04-30 15:49 ` [PATCH] [21/34] i386: replace spin_lock_irqsave with spin_lock Andi Kleen
2007-04-30 15:49 ` [PATCH] [22/34] i386: clean up flush_tlb_others fn Andi Kleen
2007-04-30 15:49 ` [PATCH] [23/34] i386: check capability Andi Kleen
2007-04-30 15:49 ` [PATCH] [24/34] x86_64: Auto compute __NR_syscall_max at compile time Andi Kleen
2007-04-30 15:49 ` [PATCH] [25/34] x86_64: skip cache_free_alien() on non NUMA Andi Kleen
2007-04-30 15:49 ` [PATCH] [26/34] i386: remove xtime_lock'ing around cpufreq notifier Andi Kleen
2007-04-30 15:50 ` Andi Kleen [this message]
2007-04-30 15:50 ` [PATCH] [28/34] i386: pte xchg optimization Andi Kleen
2007-05-02 10:52   ` [patches] " Jan Beulich
2007-05-02 11:05     ` Andi Kleen
2007-05-02 19:19       ` Zachary Amsden
2007-04-30 15:50 ` [PATCH] [29/34] i386: pte simplify ops Andi Kleen
2007-04-30 15:50 ` [PATCH] [30/34] i386: convert the monitor thread to use the kthread API Andi Kleen
2007-04-30 16:55   ` James Bottomley
2007-04-30 17:56     ` Andi Kleen
2007-04-30 15:50 ` [PATCH] [31/34] i386: convert to " Andi Kleen
2007-04-30 15:50 ` [PATCH] [32/34] x86_64: unexport cpu_llc_id Andi Kleen
2007-04-30 15:50 ` [PATCH] [34/34] x86_64: ignore vgacon if hardware not present Andi Kleen

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=20070430155000.7459E151D3@wotan.suse.de \
    --to=ak@suse.de \
    --cc=linux-kernel@vger.kernel.org \
    --cc=patches@x86-64.org \
    --cc=zach@vmware.com \
    /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

all inboxes | Powered by JetHome®