LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Andi Kleen <ak@suse.de>
To: Ingo Molnar <mingo@elte.hu>,
	Suresh Siddha <suresh.b.siddha@intel.com>,
	Andi Kleen <ak@suse.de>, "Li, Shaohua" <shaohua.li@intel.com>,
	"Eric W. Biederman" <ebiederm@xmission.com>,
	patches@x86-64.org, linux-kernel@vger.kernel.org
Subject: [PATCH 2.6.21 review I] [8/25] x86_64: optimize & fix APIC mode setup
Date: Sat, 10 Feb 2007 12:50:20 +0100 (CET)	[thread overview]
Message-ID: <20070210115020.A112613DFB@wotan.suse.de> (raw)
In-Reply-To: <200702101250.142420000@suse.de>


From: Ingo Molnar <mingo@elte.hu>

Fix a couple of inconsistencies/problems I found while reviewing the x86_64
genapic code (when I was chasing mysterious eth0 timeouts that would only
trigger if CPU_HOTPLUG is enabled):

 - AMD systems defaulted to the slower flat-physical mode instead
   of the flat-logical mode. The only restriction on AMD systems
   is that they should not use clustered APIC mode.

 - removed the CPU hotplug hacks, switching the default for small
   systems back from phys-flat to logical-flat. The switching to logical
   flat mode on small systems fixed sporadic ethernet driver timeouts i
   was getting on a dual-core Athlon64 system:

    NETDEV WATCHDOG: eth0: transmit timed out
    eth0: Transmit timeout, status 0c 0005 c07f media 80.
    eth0: Tx queue start entry 32  dirty entry 28.
    eth0:  Tx descriptor 0 is 0008a04a. (queue head)
    eth0:  Tx descriptor 1 is 0008a04a.
    eth0:  Tx descriptor 2 is 0008a04a.
    eth0:  Tx descriptor 3 is 0008a04a.
    eth0: link up, 100Mbps, full-duplex, lpa 0xC5E1

 - The use of '<= 8' was a bug by itself (the valid APIC ids
   for logical flat mode go from 0 to 7, not 0 to 8). The new logic
   is to use logical flat mode on both AMD and Intel systems, and
   to only switch to physical mode when logical mode cannot be used.
   If CPU hotplug is racy wrt. APIC shutdown then CPU hotplug needs
   fixing, not the whole IRQ system be made inconsistent and slowed
   down.

 - minor cleanups: simplified some code constructs

build & booted on a couple of AMD and Intel SMP systems.

Signed-off-by: Ingo Molnar <mingo@elte.hu>
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: Suresh Siddha <suresh.b.siddha@intel.com>
Cc: Andi Kleen <ak@suse.de>
Cc: "Li, Shaohua" <shaohua.li@intel.com>
Cc: "Eric W. Biederman" <ebiederm@xmission.com>
Signed-off-by: Andrew Morton <akpm@osdl.org>
---

 arch/x86_64/kernel/genapic.c |   39 +++++++++++++++------------------------
 1 file changed, 15 insertions(+), 24 deletions(-)

Index: linux/arch/x86_64/kernel/genapic.c
===================================================================
--- linux.orig/arch/x86_64/kernel/genapic.c
+++ linux/arch/x86_64/kernel/genapic.c
@@ -32,21 +32,20 @@ extern struct genapic apic_cluster;
 extern struct genapic apic_flat;
 extern struct genapic apic_physflat;
 
-struct genapic *genapic = &apic_flat;
-
+struct genapic __read_mostly *genapic = &apic_flat;
 
 /*
  * Check the APIC IDs in bios_cpu_apicid and choose the APIC mode.
  */
 void __init clustered_apic_check(void)
 {
-	long i;
+	int i;
 	u8 clusters, max_cluster;
 	u8 id;
 	u8 cluster_cnt[NUM_APIC_CLUSTERS];
 	int max_apic = 0;
 
-#if defined(CONFIG_ACPI)
+#ifdef CONFIG_ACPI
 	/*
 	 * Some x86_64 machines use physical APIC mode regardless of how many
 	 * procs/clusters are present (x86_64 ES7000 is an example).
@@ -68,20 +67,17 @@ void __init clustered_apic_check(void)
 		cluster_cnt[APIC_CLUSTERID(id)]++;
 	}
 
-	/* Don't use clustered mode on AMD platforms. */
+	/*
+	 * Don't use clustered mode on AMD platforms, default
+	 * to flat logical mode.
+	 */
  	if (boot_cpu_data.x86_vendor == X86_VENDOR_AMD) {
-		genapic = &apic_physflat;
-#ifndef CONFIG_HOTPLUG_CPU
-		/* In the CPU hotplug case we cannot use broadcast mode
-		   because that opens a race when a CPU is removed.
-		   Stay at physflat mode in this case.
-		   It is bad to do this unconditionally though. Once
-		   we have ACPI platform support for CPU hotplug
-		   we should detect hotplug capablity from ACPI tables and
-		   only do this when really needed. -AK */
-		if (max_apic <= 8)
-			genapic = &apic_flat;
-#endif
+		/*
+		 * Switch to physical flat mode if more than 8 APICs
+		 * (In the case of 8 CPUs APIC ID goes from 0 to 7):
+		 */
+		if (max_apic >= 8)
+			genapic = &apic_physflat;
  		goto print;
  	}
 
@@ -103,14 +99,9 @@ void __init clustered_apic_check(void)
 	 * (We don't use lowest priority delivery + HW APIC IRQ steering, so
 	 * can ignore the clustered logical case and go straight to physical.)
 	 */
-	if (clusters <= 1 && max_cluster <= 8 && cluster_cnt[0] == max_cluster) {
-#ifdef CONFIG_HOTPLUG_CPU
-		/* Don't use APIC shortcuts in CPU hotplug to avoid races */
-		genapic = &apic_physflat;
-#else
+	if (clusters <= 1 && max_cluster <= 8 && cluster_cnt[0] == max_cluster)
 		genapic = &apic_flat;
-#endif
-	} else
+	else
 		genapic = &apic_cluster;
 
 print:

  parent reply	other threads:[~2007-02-10 12:02 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-02-10 11:50 [PATCH 2.6.21 review I] [1/25] x86_64: Add __copy_from_user_nocache Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [2/25] x86_64: Make the NUMA hash function nodemap allocation Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [3/25] i386: Convert i386 PDA code to use %fs Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [4/25] x86: kernel-mode faults pollute current->thead Andi Kleen
2007-02-12  9:32   ` [patches] " Jan Beulich
2007-02-12 16:42     ` Jeff Dike
2007-02-12 17:01       ` Jan Beulich
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [5/25] i386: revert i386-fix-the-verify_quirk_intel_irqbalance Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [6/25] x86_64: revert x86_64-mm-add-genapic_force Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [7/25] x86: revert x86_64-mm-fix-the-irqbalance-quirk-for-e7320-e7520-e7525 Andi Kleen
2007-02-10 11:50 ` Andi Kleen [this message]
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [9/25] x86_64: always use physical delivery mode on > 8 CPUs Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [10/25] x86_64: remove clustered APIC mode Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [11/25] x86: default to physical mode on hotplug CPU kernels Andi Kleen
2007-02-11 11:13   ` Eric W. Biederman
2007-02-12 22:36     ` Andi Kleen
2007-02-12 23:10       ` Eric W. Biederman
2007-02-12 23:51         ` Siddha, Suresh B
2007-02-12 23:43       ` Siddha, Suresh B
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [12/25] x86_64: x86_64-make-the-numa-hash-function-nodemap-allocation fix fix Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [13/25] i386: Fix a typo in an IRQ handler name Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [14/25] x86: Share what's shareable Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [15/25] i386: Only call unreachable_devices() when type 1 is available Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [16/25] i386: Detect and support the E7520 and the 945G/GZ/P/PL Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [17/25] i386: Reserve resources but only when we're sure about them Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [18/25] x86_64: Fix x86_64 ioremap base_address Andi Kleen
2007-02-10 11:58   ` Arjan van de Ven
2007-02-10 12:07     ` Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [19/25] x86: Reject a broken MCFG tables on Asus etc Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [20/25] x86_64: get rid of ARCH_HAVE_XTIME_LOCK Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [21/25] x86_64: a memcpy that tries to reduce cache pressure Andi Kleen
2007-02-12  9:57   ` [patches] " Jan Beulich
2007-02-12 10:25     ` Andi Kleen
2007-02-13 11:27   ` Eric Dumazet
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [22/25] x86_64: use memcpy_uncached_read() in RDMA interrupt handler to reduce packet loss Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [23/25] x86_64: improved iommu documentation Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [24/25] x86_64: do not always end the stack trace with ULONG_MAX Andi Kleen
2007-02-10 11:50 ` [PATCH 2.6.21 review I] [25/25] i386: arch/i386/kernel/e820.c should #include <asm/setup.h 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=20070210115020.A112613DFB@wotan.suse.de \
    --to=ak@suse.de \
    --cc=ebiederm@xmission.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=mingo@elte.hu \
    --cc=patches@x86-64.org \
    --cc=shaohua.li@intel.com \
    --cc=suresh.b.siddha@intel.com \
    --subject='Re: [PATCH 2.6.21 review I] [8/25] x86_64: optimize & fix APIC mode setup' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).