Merge with /pub/scm/linux/kernel/git/torvalds/linux-2.6.git

2026-05-01 15:00:59 -07:00 · 2005-11-09 14:33:22 -08:00
parent ec58ef0328 ad8f76be48
commit e82b3aec8d
2548 changed files with 138980 additions and 95293 deletions
@@ -3642,11 +3642,9 @@ S: Beaverton, OR 97005
 S: USA
 N: Michal Wronski
-E: wrona@mat.uni.torun.pl
+E: Michal.Wronski@motorola.com
 W: http://www.mat.uni.torun.pl/~wrona
 D: POSIX message queues fs (with K. Benedyczak)
-S: ul. Teczowa 23/12
+S: Krakow
 S: 80-680 Gdansk-Sobieszewo
 S: Poland
 N: Frank Xia
@@ -139,9 +139,14 @@ You'll probably want to upgrade.
 Ksymoops
 --------
-If the unthinkable happens and your kernel oopses, you'll need a 2.4
+If the unthinkable happens and your kernel oopses, you may need the
-version of ksymoops to decode the report; see REPORTING-BUGS in the
+ksymoops tool to decode it, but in most cases you don't.
-root of the Linux source for more information.
+In the 2.6 kernel it is generally preferred to build the kernel with
 CONFIG_KALLSYMS so that it produces readable dumps that can be used as-is
 (this also produces better output than ksymoops).
 If for some reason your kernel is not build with CONFIG_KALLSYMS and
 you have no way to rebuild and reproduce the Oops with that option, then
 you can still decode that Oops with ksymoops.
 Module-Init-Tools
 -----------------
@@ -10,7 +10,7 @@ DOCBOOKS := wanbook.xml z8530book.xml mcabook.xml videobook.xml \
 	    kernel-hacking.xml kernel-locking.xml deviceiobook.xml \
 	    procfs-guide.xml writing_usb_driver.xml \
 	    sis900.xml kernel-api.xml journal-api.xml lsm.xml usb.xml \
-	    gadget.xml libata.xml mtdnand.xml librs.xml
+	    gadget.xml libata.xml mtdnand.xml librs.xml rapidio.xml
 ###
 # The build process is as follows (targets):
@@ -306,7 +306,7 @@ an example.
 </para>
 	<sect1><title>Journal Level</title>
 !Efs/jbd/journal.c
-!Efs/jbd/recovery.c
+!Ifs/jbd/recovery.c
 	</sect1>
 	<sect1><title>Transasction Level</title>
 !Efs/jbd/transaction.c	
@@ -118,7 +118,7 @@ X!Ilib/string.c
     </sect1>
     <sect1><title>User Space Memory Access</title>
 !Iinclude/asm-i386/uaccess.h
-!Iarch/i386/lib/usercopy.c
+!Earch/i386/lib/usercopy.c
     </sect1>
     <sect1><title>More Memory Management Functions</title>
 !Iinclude/linux/rmap.h
@@ -174,7 +174,6 @@ X!Ilib/string.c
     <title>The Linux VFS</title>
     <sect1><title>The Filesystem types</title>
 !Iinclude/linux/fs.h
 !Einclude/linux/fs.h
     </sect1>
     <sect1><title>The Directory Cache</title>
 !Efs/dcache.c
@@ -239,9 +238,9 @@ X!Ilib/string.c
     <title>Network device support</title>
     <sect1><title>Driver Support</title>
 !Enet/core/dev.c
-     </sect1>
+!Enet/ethernet/eth.c
-     <sect1><title>8390 Based Network Cards</title>
+!Einclude/linux/etherdevice.h
-!Edrivers/net/8390.c
+!Enet/core/wireless.c
     </sect1>
     <sect1><title>Synchronous PPP</title>
 !Edrivers/net/wan/syncppp.c
@@ -266,7 +265,7 @@ X!Ekernel/module.c
  <chapter id="hardware">
     <title>Hardware Interfaces</title>
     <sect1><title>Interrupt Handling</title>
-!Ikernel/irq/manage.c
+!Ekernel/irq/manage.c
     </sect1>
     <sect1><title>Resources Management</title>
@@ -501,7 +500,7 @@ KAO -->
 !Edrivers/video/modedb.c
     </sect1>
     <sect1><title>Frame Buffer Macintosh Video Mode Database</title>
-!Idrivers/video/macmodes.c
+!Edrivers/video/macmodes.c
     </sect1>
     <sect1><title>Frame Buffer Fonts</title>
        <para>
@@ -0,0 +1,160 @@
 <?xml version="1.0" encoding="UTF-8"?>
 <!DOCTYPE book PUBLIC "-//OASIS//DTD DocBook XML V4.1.2//EN"
        "http://www.oasis-open.org/docbook/xml/4.1.2/docbookx.dtd" [
 	<!ENTITY rapidio SYSTEM "rapidio.xml">
 	]>
 <book id="RapidIO-Guide">
 <bookinfo>
  <title>RapidIO Subsystem Guide</title>
  <authorgroup>
   <author>
    <firstname>Matt</firstname>
    <surname>Porter</surname>
    <affiliation>
     <address>
      <email>mporter@kernel.crashing.org</email>
      <email>mporter@mvista.com</email>
     </address>
    </affiliation>
   </author>
  </authorgroup>
  <copyright>
   <year>2005</year>
   <holder>MontaVista Software, Inc.</holder>
  </copyright>
  <legalnotice>
   <para>
     This documentation is free software; you can redistribute
     it and/or modify it under the terms of the GNU General Public
     License version 2 as published by the Free Software Foundation.
   </para>
   <para>
     This program is distributed in the hope that it will be
     useful, but WITHOUT ANY WARRANTY; without even the implied
     warranty of MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.
     See the GNU General Public License for more details.
   </para>
   <para>
     You should have received a copy of the GNU General Public
     License along with this program; if not, write to the Free
     Software Foundation, Inc., 59 Temple Place, Suite 330, Boston,
     MA 02111-1307 USA
   </para>
   <para>
     For more details see the file COPYING in the source
     distribution of Linux.
   </para>
  </legalnotice>
 </bookinfo>
 <toc></toc>
  <chapter id="intro">
      <title>Introduction</title>
  <para>
 	RapidIO is a high speed switched fabric interconnect with
 	features aimed at the embedded market.  RapidIO provides
 	support for memory-mapped I/O as well as message-based
 	transactions over the switched fabric network. RapidIO has
 	a standardized discovery mechanism not unlike the PCI bus
 	standard that allows simple detection of devices in a
 	network.
  </para>
  <para>
  	This documentation is provided for developers intending
 	to support RapidIO on new architectures, write new drivers,
 	or to understand the subsystem internals.
  </para>
  </chapter>
  <chapter id="bugs">
     <title>Known Bugs and Limitations</title>
     <sect1>
     	<title>Bugs</title>
 	  <para>None. ;)</para>
     </sect1>
     <sect1>
     	<title>Limitations</title>
 	  <para>
 	    <orderedlist>
 	      <listitem><para>Access/management of RapidIO memory regions is not supported</para></listitem>
 	      <listitem><para>Multiple host enumeration is not supported</para></listitem>
 	    </orderedlist>
 	 </para>
     </sect1>
  </chapter>
  <chapter id="drivers">
     	<title>RapidIO driver interface</title>
 	<para>
 		Drivers are provided a set of calls in order
 		to interface with the subsystem to gather info
 		on devices, request/map memory region resources,
 		and manage mailboxes/doorbells.
 	</para>
 	<sect1>
 		<title>Functions</title>
 !Iinclude/linux/rio_drv.h
 !Edrivers/rapidio/rio-driver.c
 !Edrivers/rapidio/rio.c
 	</sect1>
  </chapter>
  <chapter id="internals">
     <title>Internals</title>
     <para>
     This chapter contains the autogenerated documentation of the RapidIO
     subsystem.
     </para>
     <sect1><title>Structures</title>
 !Iinclude/linux/rio.h
     </sect1>
     <sect1><title>Enumeration and Discovery</title>
 !Idrivers/rapidio/rio-scan.c
     </sect1>
     <sect1><title>Driver functionality</title>
 !Idrivers/rapidio/rio.c
 !Idrivers/rapidio/rio-access.c
     </sect1>
     <sect1><title>Device model support</title>
 !Idrivers/rapidio/rio-driver.c
     </sect1>
     <sect1><title>Sysfs support</title>
 !Idrivers/rapidio/rio-sysfs.c
     </sect1>
     <sect1><title>PPC32 support</title>
 !Iarch/ppc/kernel/rio.c
 !Earch/ppc/syslib/ppc85xx_rio.c
 !Iarch/ppc/syslib/ppc85xx_rio.c
     </sect1>
  </chapter>
  <chapter id="credits">
     <title>Credits</title>
 	<para>
 		The following people have contributed to the RapidIO
 		subsystem directly or indirectly:
 		<orderedlist>
 			<listitem><para>Matt Porter<email>mporter@kernel.crashing.org</email></para></listitem>
 			<listitem><para>Randy Vinson<email>rvinson@mvista.com</email></para></listitem>
 			<listitem><para>Dan Malek<email>dan@embeddedalley.com</email></para></listitem>
 		</orderedlist>
 	</para>
 	<para>
 		The following people have contributed to this document:
 		<orderedlist>
 			<listitem><para>Matt Porter<email>mporter@kernel.crashing.org</email></para></listitem>
 		</orderedlist>
 	</para>
  </chapter>
 </book>
@@ -10,14 +10,22 @@
 This guide describes the basics of Message Signaled Interrupts (MSI),
 the advantages of using MSI over traditional interrupt mechanisms,
 and how to enable your driver to use MSI or MSI-X. Also included is
-a Frequently Asked Questions.
+a Frequently Asked Questions (FAQ) section.
 1.1 Terminology
 PCI devices can be single-function or multi-function.  In either case,
 when this text talks about enabling or disabling MSI on a "device
 function," it is referring to one specific PCI device and function and
 not to all functions on a PCI device (unless the PCI device has only
 one function).
 2. Copyright 2003 Intel Corporation
 3. What is MSI/MSI-X?
 Message Signaled Interrupt (MSI), as described in the PCI Local Bus
-Specification Revision 2.3 or latest, is an optional feature, and a
+Specification Revision 2.3 or later, is an optional feature, and a
 required feature for PCI Express devices. MSI enables a device function
 to request service by sending an Inbound Memory Write on its PCI bus to
 the FSB as a Message Signal Interrupt transaction. Because MSI is
@@ -27,7 +35,7 @@ supported.
 A PCI device that supports MSI must also support pin IRQ assertion
 interrupt mechanism to provide backward compatibility for systems that
-do not support MSI. In Systems, which support MSI, the bus driver is
+do not support MSI. In systems which support MSI, the bus driver is
 responsible for initializing the message address and message data of
 the device function's MSI/MSI-X capability structure during device
 initial configuration.
@@ -61,17 +69,17 @@ over the MSI capability structure as described below.
        - MSI and MSI-X both support per-vector masking. Per-vector
 	masking is an optional extension of MSI but a required
-	feature for MSI-X. Per-vector masking provides the kernel
+	feature for MSI-X. Per-vector masking provides the kernel the
-	the ability to mask/unmask MSI when servicing its software
+	ability to mask/unmask a single MSI while running its
-	interrupt service routing handler. If per-vector masking is
+	interrupt service routine. If per-vector masking is
 	not supported, then the device driver should provide the
 	hardware/software synchronization to ensure that the device
 	generates MSI when the driver wants it to do so.
 4. Why use MSI?
-As a benefit the simplification of board design, MSI allows board
+As a benefit to the simplification of board design, MSI allows board
-designers to remove out of band interrupt routing. MSI is another
+designers to remove out-of-band interrupt routing. MSI is another
 step towards a legacy-free environment.
 Due to increasing pressure on chipset and processor packages to
@@ -87,7 +95,7 @@ support. As a result, the PCI Express technology requires MSI
 support for better interrupt performance.
 Using MSI enables the device functions to support two or more
-vectors, which can be configured to target different CPU's to
+vectors, which can be configured to target different CPUs to
 increase scalability.
 5. Configuring a driver to use MSI/MSI-X
@@ -119,13 +127,13 @@ pci_enable_msi() explicitly.
 int pci_enable_msi(struct pci_dev *dev)
-With this new API, any existing device driver, which like to have
+With this new API, a device driver that wants to have MSI
-MSI enabled on its device function, must call this API to enable MSI
+enabled on its device function must call this API to enable MSI.
 A successful call will initialize the MSI capability structure
 with ONE vector, regardless of whether a device function is
 capable of supporting multiple messages. This vector replaces the
-pre-assigned dev->irq with a new MSI vector. To avoid the conflict
+pre-assigned dev->irq with a new MSI vector. To avoid a conflict
-of new assigned vector with existing pre-assigned vector requires
+of the new assigned vector with existing pre-assigned vector requires
 a device driver to call this API before calling request_irq().
 5.2.2 API pci_disable_msi
@@ -137,14 +145,14 @@ when a device driver is unloading. This API restores dev->irq with
 the pre-assigned IOAPIC vector and switches a device's interrupt
 mode to PCI pin-irq assertion/INTx emulation mode.
-Note that a device driver should always call free_irq() on MSI vector
+Note that a device driver should always call free_irq() on the MSI vector
-it has done request_irq() on before calling this API. Failure to do
+that it has done request_irq() on before calling this API. Failure to do
-so results a BUG_ON() and a device will be left with MSI enabled and
+so results in a BUG_ON() and a device will be left with MSI enabled and
 leaks its vector.
 5.2.3 MSI mode vs. legacy mode diagram
-The below diagram shows the events, which switches the interrupt
+The below diagram shows the events which switch the interrupt
 mode on the MSI-capable device function between MSI mode and
 PIN-IRQ assertion mode.
@@ -155,9 +163,9 @@ PIN-IRQ assertion mode.
 	 ------------	pci_disable_msi  ------------------------
-Figure 1.0 MSI Mode vs. Legacy Mode
+Figure 1. MSI Mode vs. Legacy Mode
-In Figure 1.0, a device operates by default in legacy mode. Legacy
+In Figure 1, a device operates by default in legacy mode. Legacy
 in this context means PCI pin-irq assertion or PCI-Express INTx
 emulation. A successful MSI request (using pci_enable_msi()) switches
 a device's interrupt mode to MSI mode. A pre-assigned IOAPIC vector
@@ -166,11 +174,11 @@ assigned MSI vector will replace dev->irq.
 To return back to its default mode, a device driver should always call
 pci_disable_msi() to undo the effect of pci_enable_msi(). Note that a
-device driver should always call free_irq() on MSI vector it has done
+device driver should always call free_irq() on the MSI vector it has
-request_irq() on before calling pci_disable_msi(). Failure to do so
+done request_irq() on before calling pci_disable_msi(). Failure to do
-results a BUG_ON() and a device will be left with MSI enabled and
+so results in a BUG_ON() and a device will be left with MSI enabled and
 leaks its vector. Otherwise, the PCI subsystem restores a device's
-dev->irq with a pre-assigned IOAPIC vector and marks released
+dev->irq with a pre-assigned IOAPIC vector and marks the released
 MSI vector as unused.
 Once being marked as unused, there is no guarantee that the PCI
@@ -178,8 +186,8 @@ subsystem will reserve this MSI vector for a device. Depending on
 the availability of current PCI vector resources and the number of
 MSI/MSI-X requests from other drivers, this MSI may be re-assigned.
-For the case where the PCI subsystem re-assigned this MSI vector
+For the case where the PCI subsystem re-assigns this MSI vector to
-another driver, a request to switching back to MSI mode may result
+another driver, a request to switch back to MSI mode may result
 in being assigned a different MSI vector or a failure if no more
 vectors are available.
@@ -208,12 +216,12 @@ Unlike the function pci_enable_msi(), the function pci_enable_msix()
 does not replace the pre-assigned IOAPIC dev->irq with a new MSI
 vector because the PCI subsystem writes the 1:1 vector-to-entry mapping
 into the field vector of each element contained in a second argument.
-Note that the pre-assigned IO-APIC dev->irq is valid only if the device
+Note that the pre-assigned IOAPIC dev->irq is valid only if the device
-operates in PIN-IRQ assertion mode. In MSI-X mode, any attempt of
+operates in PIN-IRQ assertion mode. In MSI-X mode, any attempt at
 using dev->irq by the device driver to request for interrupt service
 may result unpredictabe behavior.
-For each MSI-X vector granted, a device driver is responsible to call
+For each MSI-X vector granted, a device driver is responsible for calling
 other functions like request_irq(), enable_irq(), etc. to enable
 this vector with its corresponding interrupt service handler. It is
 a device driver's choice to assign all vectors with the same
@@ -224,13 +232,13 @@ service handler.
 The PCI 3.0 specification has implementation notes that MMIO address
 space for a device's MSI-X structure should be isolated so that the
-software system can set different page for controlling accesses to
+software system can set different pages for controlling accesses to the
-the MSI-X structure. The implementation of MSI patch requires the PCI
+MSI-X structure. The implementation of MSI support requires the PCI
 subsystem, not a device driver, to maintain full control of the MSI-X
-table/MSI-X PBA and MMIO address space of the MSI-X table/MSI-X PBA.
+table/MSI-X PBA (Pending Bit Array) and MMIO address space of the MSI-X
-A device driver is prohibited from requesting the MMIO address space
+table/MSI-X PBA.  A device driver is prohibited from requesting the MMIO
-of the MSI-X table/MSI-X PBA. Otherwise, the PCI subsystem will fail
+address space of the MSI-X table/MSI-X PBA. Otherwise, the PCI subsystem
-enabling MSI-X on its hardware device when it calls the function
+will fail enabling MSI-X on its hardware device when it calls the function
 pci_enable_msix().
 5.3.2 Handling MSI-X allocation
@@ -274,9 +282,9 @@ For the case where fewer MSI-X vectors are allocated to a function
 than requested, the function pci_enable_msix() will return the
 maximum number of MSI-X vectors available to the caller. A device
 driver may re-send its request with fewer or equal vectors indicated
-in a return. For example, if a device driver requests 5 vectors, but
+in the return. For example, if a device driver requests 5 vectors, but
-the number of available vectors is 3 vectors, a value of 3 will be a
+the number of available vectors is 3 vectors, a value of 3 will be
-return as a result of pci_enable_msix() call. A function could be
+returned as a result of pci_enable_msix() call. A function could be
 designed for its driver to use only 3 MSI-X table entries as
 different combinations as ABC--, A-B-C, A--CB, etc. Note that this
 patch does not support multiple entries with the same vector. Such
@@ -285,49 +293,46 @@ as ABBCC, AABCC, BCCBA, etc will result as a failure by the function
 pci_enable_msix(). Below are the reasons why supporting multiple
 entries with the same vector is an undesirable solution.
-	- The PCI subsystem can not determine which entry, which
+	- The PCI subsystem cannot determine the entry that
-	  generated the message, to mask/unmask MSI while handling
+	  generated the message to mask/unmask MSI while handling
 	  software driver ISR. Attempting to walk through all MSI-X
 	  table entries (2048 max) to mask/unmask any match vector
 	  is an undesirable solution.
-	- Walk through all MSI-X table entries (2048 max) to handle
+	- Walking through all MSI-X table entries (2048 max) to handle
 	  SMP affinity of any match vector is an undesirable solution.
 5.3.4 API pci_enable_msix
-int pci_enable_msix(struct pci_dev *dev, u32 *entries, int nvec)
+int pci_enable_msix(struct pci_dev *dev, struct msix_entry *entries, int nvec)
 This API enables a device driver to request the PCI subsystem
-for enabling MSI-X messages on its hardware device. Depending on
+to enable MSI-X messages on its hardware device. Depending on
 the availability of PCI vectors resources, the PCI subsystem enables
-either all or nothing.
+either all or none of the requested vectors.
-Argument dev points to the device (pci_dev) structure.
+Argument 'dev' points to the device (pci_dev) structure.
-Argument entries is a pointer of unsigned integer type. The number of
+Argument 'entries' is a pointer to an array of msix_entry structs.
-elements is indicated in argument nvec. The content of each element
+The number of entries is indicated in argument 'nvec'.
-will be mapped to the following struct defined in /driver/pci/msi.h.
+struct msix_entry is defined in /driver/pci/msi.h:
 struct msix_entry {
 	u16 	vector; /* kernel uses to write alloc vector */
 	u16	entry; /* driver uses to specify entry */
 };
-A device driver is responsible for initializing the field entry of
+A device driver is responsible for initializing the field 'entry' of
-each element with unique entry supported by MSI-X table. Otherwise,
+each element with a unique entry supported by MSI-X table. Otherwise,
 -EINVAL will be returned as a result. A successful return of zero
-indicates the PCI subsystem completes initializing each of requested
+indicates the PCI subsystem completed initializing each of the requested
 entries of the MSI-X table with message address and message data.
 Last but not least, the PCI subsystem will write the 1:1
-vector-to-entry mapping into the field vector of each element. A
+vector-to-entry mapping into the field 'vector' of each element. A
-device driver is responsible of keeping track of allocated MSI-X
+device driver is responsible for keeping track of allocated MSI-X
 vectors in its internal data structure.
-Argument nvec is an integer indicating the number of messages
+A return of zero indicates that the number of MSI-X vectors was
 requested.
 A return of zero indicates that the number of MSI-X vectors is
 successfully allocated. A return of greater than zero indicates
 MSI-X vector shortage. Or a return of less than zero indicates
 a failure. This failure may be a result of duplicate entries
@@ -341,12 +346,12 @@ void pci_disable_msix(struct pci_dev *dev)
 This API should always be used to undo the effect of pci_enable_msix()
 when a device driver is unloading. Note that a device driver should
 always call free_irq() on all MSI-X vectors it has done request_irq()
-on before calling this API. Failure to do so results a BUG_ON() and
+on before calling this API. Failure to do so results in a BUG_ON() and
 a device will be left with MSI-X enabled and leaks its vectors.
 5.3.6 MSI-X mode vs. legacy mode diagram
-The below diagram shows the events, which switches the interrupt
+The below diagram shows the events which switch the interrupt
 mode on the MSI-X capable device function between MSI-X mode and
 PIN-IRQ assertion mode (legacy).
@@ -356,22 +361,22 @@ PIN-IRQ assertion mode (legacy).
 	| 	     | ===============>	    |			     |
 	 ------------	pci_disable_msix     ------------------------
-Figure 2.0 MSI-X Mode vs. Legacy Mode
+Figure 2. MSI-X Mode vs. Legacy Mode
-In Figure 2.0, a device operates by default in legacy mode. A
+In Figure 2, a device operates by default in legacy mode. A
 successful MSI-X request (using pci_enable_msix()) switches a
 device's interrupt mode to MSI-X mode. A pre-assigned IOAPIC vector
 stored in dev->irq will be saved by the PCI subsystem; however,
 unlike MSI mode, the PCI subsystem will not replace dev->irq with
 assigned MSI-X vector because the PCI subsystem already writes the 1:1
-vector-to-entry mapping into the field vector of each element
+vector-to-entry mapping into the field 'vector' of each element
 specified in second argument.
 To return back to its default mode, a device driver should always call
 pci_disable_msix() to undo the effect of pci_enable_msix(). Note that
 a device driver should always call free_irq() on all MSI-X vectors it
 has done request_irq() on before calling pci_disable_msix(). Failure
-to do so results a BUG_ON() and a device will be left with MSI-X
+to do so results in a BUG_ON() and a device will be left with MSI-X
 enabled and leaks its vectors. Otherwise, the PCI subsystem switches a
 device function's interrupt mode from MSI-X mode to legacy mode and
 marks all allocated MSI-X vectors as unused.
@@ -383,53 +388,56 @@ MSI/MSI-X requests from other drivers, these MSI-X vectors may be
 re-assigned.
 For the case where the PCI subsystem re-assigned these MSI-X vectors
-to other driver, a request to switching back to MSI-X mode may result
+to other drivers, a request to switch back to MSI-X mode may result
 being assigned with another set of MSI-X vectors or a failure if no
 more vectors are available.
-5.4 Handling function implementng both MSI and MSI-X capabilities
+5.4 Handling function implementing both MSI and MSI-X capabilities
 For the case where a function implements both MSI and MSI-X
 capabilities, the PCI subsystem enables a device to run either in MSI
 mode or MSI-X mode but not both. A device driver determines whether it
 wants MSI or MSI-X enabled on its hardware device. Once a device
-driver requests for MSI, for example, it is prohibited to request for
+driver requests for MSI, for example, it is prohibited from requesting
 MSI-X; in other words, a device driver is not permitted to ping-pong
 between MSI mod MSI-X mode during a run-time.
 5.5 Hardware requirements for MSI/MSI-X support
 MSI/MSI-X support requires support from both system hardware and
 individual hardware device functions.
 5.5.1 System hardware support
 Since the target of MSI address is the local APIC CPU, enabling
-MSI/MSI-X support in Linux kernel is dependent on whether existing
+MSI/MSI-X support in the Linux kernel is dependent on whether existing
-system hardware supports local APIC. Users should verify their
+system hardware supports local APIC. Users should verify that their
-system whether it runs when CONFIG_X86_LOCAL_APIC=y.
+system supports local APIC operation by testing that it runs when
 CONFIG_X86_LOCAL_APIC=y.
 In SMP environment, CONFIG_X86_LOCAL_APIC is automatically set;
 however, in UP environment, users must manually set
 CONFIG_X86_LOCAL_APIC. Once CONFIG_X86_LOCAL_APIC=y, setting
-CONFIG_PCI_MSI enables the VECTOR based scheme and
+CONFIG_PCI_MSI enables the VECTOR based scheme and the option for
-the option for MSI-capable device drivers to selectively enable
+MSI-capable device drivers to selectively enable MSI/MSI-X.
 MSI/MSI-X.
 Note that CONFIG_X86_IO_APIC setting is irrelevant because MSI/MSI-X
 vector is allocated new during runtime and MSI/MSI-X support does not
 depend on BIOS support. This key independency enables MSI/MSI-X
-support on future IOxAPIC free platform.
+support on future IOxAPIC free platforms.
 5.5.2 Device hardware support
 The hardware device function supports MSI by indicating the
 MSI/MSI-X capability structure on its PCI capability list. By
 default, this capability structure will not be initialized by
 the kernel to enable MSI during the system boot. In other words,
 the device function is running on its default pin assertion mode.
 Note that in many cases the hardware supporting MSI have bugs,
-which may result in system hang. The software driver of specific
+which may result in system hangs. The software driver of specific
-MSI-capable hardware is responsible for whether calling
+MSI-capable hardware is responsible for deciding whether to call
 pci_enable_msi or not. A return of zero indicates the kernel
-successfully initializes the MSI/MSI-X capability structure of the
+successfully initialized the MSI/MSI-X capability structure of the
 device function. The device function is now running on MSI/MSI-X mode.
 5.6 How to tell whether MSI/MSI-X is enabled on device function
@@ -439,10 +447,10 @@ pci_enable_msi()/pci_enable_msix() indicates to a device driver that
 its device function is initialized successfully and ready to run in
 MSI/MSI-X mode.
-At the user level, users can use command 'cat /proc/interrupts'
+At the user level, users can use the command 'cat /proc/interrupts'
-to display the vector allocated for a device and its interrupt
+to display the vectors allocated for devices and their interrupt
-MSI/MSI-X mode ("PCI MSI"/"PCI MSIX"). Below shows below MSI mode is
+MSI/MSI-X modes ("PCI-MSI"/"PCI-MSI-X"). Below shows MSI mode is
-enabled on a SCSI Adaptec 39320D Ultra320.
+enabled on a SCSI Adaptec 39320D Ultra320 controller.
           CPU0       CPU1
  0:     324639          0    IO-APIC-edge  timer
@@ -453,8 +461,8 @@ enabled on a SCSI Adaptec 39320D Ultra320.
 15:          1          0    IO-APIC-edge  ide1
 169:          0          0   IO-APIC-level  uhci-hcd
 185:          0          0   IO-APIC-level  uhci-hcd
-193:        138         10         PCI MSI  aic79xx
+193:        138         10         PCI-MSI  aic79xx
-201:         30          0         PCI MSI  aic79xx
+201:         30          0         PCI-MSI  aic79xx
 225:         30          0   IO-APIC-level  aic7xxx
 233:         30          0   IO-APIC-level  aic7xxx
 NMI:          0          0
@@ -490,8 +498,8 @@ target address set as 0xfeexxxxx, as conformed to PCI
 specification 2.3 or latest, then it should work.
 Q4. From the driver point of view, if the MSI is lost because
-of the errors occur during inbound memory write, then it may
+of errors occurring during inbound memory write, then it may
-wait for ever. Is there a mechanism for it to recover?
+wait forever. Is there a mechanism for it to recover?
 A4. Since the target of the transaction is an inbound memory
 write, all transaction termination conditions (Retry,
@@ -772,8 +772,6 @@ RCU pointer/list traversal:
 	list_for_each_entry_rcu
 	list_for_each_continue_rcu	(to be deprecated in favor of new
 					 list_for_each_entry_continue_rcu)
 	hlist_for_each_rcu		(to be deprecated in favor of
 					 hlist_for_each_entry_rcu)
 	hlist_for_each_entry_rcu
 RCU pointer update:
@@ -8,10 +8,9 @@ Compilation of kernel
 ---------------------
  In order to compile ARM Linux, you will need a compiler capable of
-  generating ARM ELF code with GNU extensions.  GCC 2.95.1, EGCS
+  generating ARM ELF code with GNU extensions.  GCC 3.3 is known to be
-  1.1.2, and GCC 3.3 are known to be good compilers.  Fortunately, you
+  a good compiler.  Fortunately, you needn't guess.  The kernel will report
-  needn't guess.  The kernel will report an error if your compiler is
+  an error if your compiler is a recognized offender.
  a recognized offender.
  To build ARM Linux natively, you shouldn't have to alter the ARCH = line
  in the top level Makefile.  However, if you don't have the ARM Linux ELF
@@ -25,7 +25,7 @@
 #include <linux/skbuff.h>
 #include <linux/timer.h>
-#include "connector.h"
+#include <linux/connector.h>
 static struct cb_id cn_test_id = { 0x123, 0x456 };
 static char cn_test_name[] = "cn_test";
@@ -104,7 +104,7 @@ static int cn_test_want_notify(void)
 	req->first = cn_test_id.val + 20;
 	req->range = 10;
-	NETLINK_CB(skb).dst_groups = ctl->group;
+	NETLINK_CB(skb).dst_group = ctl->group;
 	//netlink_broadcast(nls, skb, 0, ctl->group, GFP_ATOMIC);
 	netlink_unicast(nls, skb, 0, 0);
@@ -19,7 +19,6 @@ There are two dm targets available: snapshot and snapshot-origin.
 *) snapshot-origin <origin>
 which will normally have one or more snapshots based on it.
 You must create the snapshot-origin device before you can create snapshots.
 Reads will be mapped directly to the backing device. For each write, the
 original data will be saved in the <COW device> of each snapshot to keep
 its visible content unchanged, at least until the <COW device> fills up.
@@ -27,7 +26,7 @@ its visible content unchanged, at least until the <COW device> fills up.
 *) snapshot <origin> <COW device> <persistent?> <chunksize>
-A snapshot is created of the <origin> block device. Changed chunks of
+A snapshot of the <origin> block device is created. Changed chunks of
 <chunksize> sectors will be stored on the <COW device>.  Writes will
 only go to the <COW device>.  Reads will come from the <COW device> or
 from <origin> for unchanged data.  <COW device> will often be
@@ -37,6 +36,8 @@ the amount of free space and expand the <COW device> before it fills up.
 <persistent?> is P (Persistent) or N (Not persistent - will not survive
 after reboot).
 The difference is that for transient snapshots less metadata must be
 saved on disk - they can be kept in memory by the kernel.
 How this is used by LVM2
@@ -1,5 +1,5 @@
-How to get the Nebula, PCTV and Twinhan DST cards working
+How to get the Nebula, PCTV, FusionHDTV Lite and Twinhan DST cards working
-=========================================================
+==========================================================================
 This class of cards has a bt878a as the PCI interface, and
 require the bttv driver.
@@ -26,27 +26,31 @@ Furthermore you need to enable
 In general you need to load the bttv driver, which will handle the gpio and
 i2c communication for us, plus the common dvb-bt8xx device driver.
-The frontends for Nebula (nxt6000), Pinnacle PCTV (cx24110) and
+The frontends for Nebula (nxt6000), Pinnacle PCTV (cx24110), TwinHan (dst),
-TwinHan (dst) are loaded automatically by the dvb-bt8xx device driver.
+FusionHDTV DVB-T Lite (mt352) and FusionHDTV5 Lite (lgdt330x) are loaded
 automatically by the dvb-bt8xx device driver.
-3a) Nebula / Pinnacle PCTV
+3a) Nebula / Pinnacle PCTV / FusionHDTV Lite
--------------------------
+---------------------------------------------
   $ modprobe bttv (normally bttv is being loaded automatically by kmod)
-   $ modprobe dvb-bt8xx (or just place dvb-bt8xx in /etc/modules for automatic loading)
+   $ modprobe dvb-bt8xx
 (or just place dvb-bt8xx in /etc/modules for automatic loading)
 3b) TwinHan and Clones
 --------------------------
-   $ modprobe bttv i2c_hw=1 card=0x71
+   $ modprobe bttv card=0x71
   $ modprobe dvb-bt8xx
   $ modprobe dst
 The value 0x71 will override the PCI type detection for dvb-bt8xx,
-which  is necessary for TwinHan cards.
+which  is necessary for TwinHan cards. Omission of this parameter might result
 in a system lockup.
-If you're having an older card (blue color circuit) and card=0x71 locks
+If you're having an older card (blue color PCB) and card=0x71 locks up
 your machine, try using 0x68, too. If that does not work, ask on the
 mailing list.
@@ -64,11 +68,47 @@ verbose=0 means complete disabling of messages
 dst_addons takes values 0 and 0x20. A value of 0 means it is a FTA card.
 0x20 means it has a Conditional Access slot.
-The autodected values are determined bythe cards 'response
+The autodetected values are determined by the cards 'response string'
-string' which you can see in your logs e.g.
+which you can see in your logs e.g.
 dst_get_device_id: Recognise [DSTMCI]
 If you need to sent in bug reports on the dst, please do send in a complete
 log with the verbose=4 module parameter. For general usage, the default setting
 of verbose=1 is ideal.
 4) Multiple cards
 --------------------------
 If you happen to be running multiple cards, it would be advisable to load
 the bttv module with the card id. This would help to solve any module loading
 problems that you might face.
 For example, if you have a Twinhan and Clones card along with a FusionHDTV5 Lite
 	$ modprobe bttv card=0x71 card=0x87
 Here the order of the card id is important and should be the same as that of the
 physical order of the cards. Here card=0x71 represents the Twinhan and clones
 and card=0x87 represents Fusion HDTV5 Lite. These arguments can also be
 specified in decimal, rather than hex:
 	$ modprobe bttv card=113 card=135
 Some examples of card-id's
 Pinnacle Sat		0x5e  (94)
 Nebula Digi TV		0x68  (104)
 PC HDTV			0x70  (112)
 Twinhan			0x71  (113)
 FusionHDTV DVB-T Lite	0x80  (128)
 FusionHDTV5 Lite	0x87  (135)
 For a full list of card-id's, see the V4L Documentation within the kernel
 source:  linux/Documentation/video4linux/CARDLIST.bttv
 If you have problems with this please do ask on the mailing list.
 --
 Authors: Richard Walker, Jamie Honan, Michael Hunold, Manu Abraham
@@ -41,6 +41,12 @@ o Frontends drivers:
   - dib3000mb	: DiBcom 3000-MB demodulator
  DVB-S/C/T:
   - dst		: TwinHan DST Frontend
  ATSC:
   - nxt200x		: Nxtwave NXT2002 & NXT2004
   - or51211		: or51211 based (pcHDTV HD2000 card)
   - or51132		: or51132 based (pcHDTV HD3000 card)
   - bcm3510		: Broadcom BCM3510
   - lgdt330x		: LG Electronics DT3302 & DT3303
 o Cards based on the Phillips saa7146 multimedia PCI bridge chip:
@@ -62,6 +68,10 @@ o Cards based on the Conexant Bt8xx PCI bridge:
  - Nebula Electronics DigiTV
  - TwinHan DST
  - Avermedia DVB-T
  - ChainTech digitop DST-1000 DVB-S
  - pcHDTV HD-2000 TV
  - DViCO FusionHDTV DVB-T Lite
  - DViCO FusionHDTV5 Lite
 o Technotrend / Hauppauge DVB USB devices:
  - Nova USB
@@ -83,3 +93,30 @@ o DiBcom DVB-T USB based devices:
  - DiBcom USB2.0 DVB-T reference device (non-public)
 o Experimental support for the analog module of the Siemens DVB-C PCI card
 o Cards based on the Conexant cx2388x PCI bridge:
  - ADS Tech Instant TV DVB-T PCI
  - ATI HDTV Wonder
  - digitalnow DNTV Live! DVB-T
  - DViCO FusionHDTV DVB-T1
  - DViCO FusionHDTV DVB-T Plus
  - DViCO FusionHDTV3 Gold-Q
  - DViCO FusionHDTV3 Gold-T
  - DViCO FusionHDTV5 Gold
  - Hauppauge Nova-T DVB-T
  - KWorld/VStream XPert DVB-T
  - pcHDTV HD3000 HDTV
  - TerraTec Cinergy 1400 DVB-T
  - WinFast DTV1000-T
 o Cards based on the Phillips saa7134 PCI bridge:
  - Medion 7134
  - Pinnacle PCTV 300i DVB-T + PAL
  - LifeView FlyDVB-T DUO
  - Typhoon DVB-T Duo Digital/Analog Cardbus
  - Philips TOUGH DVB-T reference design
  - Philips EUROPA V3 reference design
  - Compro Videomate DVB-T300
  - Compro Videomate DVB-T200
  - AVerMedia AVerTVHD MCE A180
@@ -75,5 +75,22 @@ Ernst Peinlich <e.peinlich@inode.at>
 Peter Beutner <p.beutner@gmx.net>
  for the IR code for the ttusb-dec driver
 Wilson Michaels <wilsonmichaels@earthlink.net>
  for the lgdt330x frontend driver, and various bugfixes
 Michael Krufky <mkrufky@m1k.net>
  for maintaining v4l/dvb inter-tree dependencies
 Taylor Jacob <rtjacob@earthlink.net>
  for the nxt2002 frontend driver
 Jean-Francois Thibert <jeanfrancois@sagetv.com>
  for the nxt2004 frontend driver
 Kirk Lapray <kirk.lapray@gmail.com>
  for the or51211 and or51132 frontend drivers, and
  for merging the nxt2002 and nxt2004 modules into a
  single nxt200x frontend driver.
 (If you think you should be in this list, but you are not, drop a
 line to the DVB mailing list)
@@ -22,7 +22,7 @@ use File::Temp qw/ tempdir /;
 use IO::Handle;
@components = ( "sp8870", "sp887x", "tda10045", "tda10046", "av7110", "dec2000t",
-		"dec2540t", "dec3000s", "vp7041", "dibusb", "nxt2002",
+		"dec2540t", "dec3000s", "vp7041", "dibusb", "nxt2002", "nxt2004",
 		"or51211", "or51132_qam", "or51132_vsb");
 # Check args
@@ -252,6 +252,23 @@ sub nxt2002 {
    $outfile;
 }
 sub nxt2004 {
    my $sourcefile = "AVerTVHD_MCE_A180_Drv_v1.2.2.16.zip";
    my $url = "http://www.aver.com/support/Drivers/$sourcefile";
    my $hash = "111cb885b1e009188346d72acfed024c";
    my $outfile = "dvb-fe-nxt2004.fw";
    my $tmpdir = tempdir(DIR => "/tmp", CLEANUP => 1);
    checkstandard();
    wgetfile($sourcefile, $url);
    unzip($sourcefile, $tmpdir);
    verify("$tmpdir/3xHybrid.sys", $hash);
    extract("$tmpdir/3xHybrid.sys", 465304, 9584, $outfile);
    $outfile;
 }
 sub or51211 {
    my $fwfile = "dvb-fe-or51211.fw";
    my $url = "http://linuxtv.org/downloads/firmware/$fwfile";
@@ -0,0 +1,152 @@
 The Framebuffer Console
 =======================
 	The framebuffer console (fbcon), as its name implies, is a text
 console running on top of the framebuffer device. It has the functionality of
 any standard text console driver, such as the VGA console, with the added
 features that can be attributed to the graphical nature of the framebuffer.
 	 In the x86 architecture, the framebuffer console is optional, and
 some even treat it as a toy. For other architectures, it is the only available
 display device, text or graphical.
 	 What are the features of fbcon?  The framebuffer console supports
 high resolutions, varying font types, display rotation, primitive multihead,
 etc. Theoretically, multi-colored fonts, blending, aliasing, and any feature
 made available by the underlying graphics card are also possible.
 A. Configuration
 	The framebuffer console can be enabled by using your favorite kernel
 configuration tool.  It is under Device Drivers->Graphics Support->Support for
 framebuffer devices->Framebuffer Console Support. Select 'y' to compile
 support statically, or 'm' for module support.  The module will be fbcon.
 	In order for fbcon to activate, at least one framebuffer driver is
 required, so choose from any of the numerous drivers available. For x86
 systems, they almost universally have VGA cards, so vga16fb and vesafb will
 always be available. However, using a chipset-specific driver will give you
 more speed and features, such as the ability to change the video mode
 dynamically.
 	To display the penguin logo, choose any logo available in Logo
 Configuration->Boot up logo.
 	Also, you will need to select at least one compiled-in fonts, but if
 you don't do anything, the kernel configuration tool will select one for you,
 usually an 8x16 font.
 GOTCHA: A common bug report is enabling the framebuffer without enabling the
 framebuffer console.  Depending on the driver, you may get a blanked or
 garbled display, but the system still boots to completion.  If you are
 fortunate to have a driver that does not alter the graphics chip, then you
 will still get a VGA console.
 B. Loading
 Possible scenarios:
 1. Driver and fbcon are compiled statically
 	 Usually, fbcon will automatically take over your console. The notable
 	 exception is vesafb.  It needs to be explicitly activated with the
 	 vga= boot option parameter.
 2. Driver is compiled statically, fbcon is compiled as a module
 	 Depending on the driver, you either get a standard console, or a
 	 garbled display, as mentioned above.  To get a framebuffer console,
 	 do a 'modprobe fbcon'.
 3. Driver is compiled as a module, fbcon is compiled statically
 	 You get your standard console.  Once the driver is loaded with
 	 'modprobe xxxfb', fbcon automatically takes over the console with
 	 the possible exception of using the fbcon=map:n option. See below.
 4. Driver and fbcon are compiled as a module.
 	 You can load them in any order. Once both are loaded, fbcon will take
 	 over the console.
 C. Boot options
         The framebuffer console has several, largely unknown, boot options
         that can change its behavior.
 1. fbcon=font:<name>
        Select the initial font to use. The value 'name' can be any of the
        compiled-in fonts: VGA8x16, 7x14, 10x18, VGA8x8, MINI4x6, RomanLarge,
        SUN8x16, SUN12x22, ProFont6x11, Acorn8x8, PEARL8x8.
 	Note, not all drivers can handle font with widths not divisible by 8,
        such as vga16fb.
 2. fbcon=scrollback:<value>[k]
        The scrollback buffer is memory that is used to preserve display
        contents that has already scrolled past your view.  This is accessed
        by using the Shift-PageUp key combination.  The value 'value' is any
        integer. It defaults to 32KB.  The 'k' suffix is optional, and will
        multiply the 'value' by 1024.
 3. fbcon=map:<0123>
        This is an interesting option. It tells which driver gets mapped to
        which console. The value '0123' is a sequence that gets repeated until
        the total length is 64 which is the number of consoles available. In
        the above example, it is expanded to 012301230123... and the mapping
        will be:
 		tty | 1 2 3 4 5 6 7 8 9 ...
 		fb  | 0 1 2 3 0 1 2 3 0 ...
 		('cat /proc/fb' should tell you what the fb numbers are)
 	One side effect that may be useful is using a map value that exceeds
 	the number of loaded fb drivers. For example, if only one driver is
 	available, fb0, adding fbcon=map:1 tells fbcon not to take over the
 	console.
 	Later on, when you want to map the console the to the framebuffer
 	device, you can use the con2fbmap utility.
 4. fbcon=vc:<n1>-<n2>
 	This option tells fbcon to take over only a range of consoles as
 	specified by the values 'n1' and 'n2'. The rest of the consoles
 	outside the given range will still be controlled by the standard
 	console driver.
 	NOTE: For x86 machines, the standard console is the VGA console which
 	is typically located on the same video card.  Thus, the consoles that
 	are controlled by the VGA console will be garbled.
 4. fbcon=rotate:<n>
        This option changes the orientation angle of the console display. The
        value 'n' accepts the following:
 	      0 - normal orientation (0 degree)
 	      1 - clockwise orientation (90 degrees)
 	      2 - upside down orientation (180 degrees)
 	      3 - counterclockwise orientation (270 degrees)
 	The angle can be changed anytime afterwards by 'echoing' the same
 	numbers to any one of the 2 attributes found in
 	/sys/class/graphics/fb{x}
 		con_rotate     - rotate the display of the active console
 		con_rotate_all - rotate the display of all consoles
 	Console rotation will only become available if Console Rotation
 	Support is compiled in your kernel.
 	NOTE: This is purely console rotation.  Any other applications that
 	use the framebuffer will remain at their 'normal'orientation.
 	Actually, the underlying fb driver is totally ignorant of console
 	rotation.
 ---
 Antonino Daplas <adaplas@pol.net>
@@ -146,10 +146,10 @@ pmipal	Use the protected mode interface for palette changes.
 mtrr:n	setup memory type range registers for the vesafb framebuffer
 	where n:
-	      0 - disabled (equivalent to nomtrr)
+	      0 - disabled (equivalent to nomtrr) (default)
 	      1 - uncachable
 	      2 - write-back
-	      3 - write-combining (default)
+	      3 - write-combining
 	      4 - write-through
 	If you see the following in dmesg, choose the type that matches the
@@ -25,6 +25,13 @@ Who:	Adrian Bunk <bunk@stusta.de>
 ---------------------------
 What:	drivers depending on OBSOLETE_OSS_DRIVER
 When:	January 2006
 Why:	OSS drivers with ALSA replacements
 Who:	Adrian Bunk <bunk@stusta.de>
 ---------------------------
 What:	RCU API moves to EXPORT_SYMBOL_GPL
 When:	April 2006
 Files:	include/linux/rcupdate.h, kernel/rcupdate.c
@@ -60,6 +67,21 @@ Who:	Jody McIntyre <scjody@steamballoon.com>
 ---------------------------
 What:	Video4Linux API 1 ioctls and video_decoder.h from Video devices.
 When:	July 2006
 Why:	V4L1 AP1 was replaced by V4L2 API. during migration from 2.4 to 2.6
 	series. The old API have lots of drawbacks and don't provide enough
 	means to work with all video and audio standards. The newer API is
 	already available on the main drivers and should be used instead.
 	Newer drivers should use v4l_compat_translate_ioctl function to handle
 	old calls, replacing to newer ones.
 	Decoder iocts are using internally to allow video drivers to
 	communicate with video decoders. This should also be improved to allow
 	V4L2 calls being translated into compatible internal ioctls.
 Who:	Mauro Carvalho Chehab <mchehab@brturbo.com.br>
 ---------------------------
 What:	i2c sysfs name change: in1_ref, vid deprecated in favour of cpu0_vid
 When:	November 2005
 Files:	drivers/i2c/chips/adm1025.c, drivers/i2c/chips/adm1026.c
@@ -69,6 +91,22 @@ Who:	Grant Coady <gcoady@gmail.com>
 ---------------------------
 What:	remove EXPORT_SYMBOL(panic_timeout)
 When:	April 2006
 Files:	kernel/panic.c
 Why:	No modular usage in the kernel.
 Who:	Adrian Bunk <bunk@stusta.de>
 ---------------------------
 What:	remove EXPORT_SYMBOL(insert_resource)
 When:	April 2006
 Files:	kernel/resource.c
 Why:	No modular usage in the kernel.
 Who:	Adrian Bunk <bunk@stusta.de>
 ---------------------------
 What:	PCMCIA control ioctl (needed for pcmcia-cs [cardmgr, cardctl])
 When:	November 2005
 Files:	drivers/pcmcia/: pcmcia_ioctl.c
@@ -95,3 +133,10 @@ Why:	This interface has been obsoleted by the new layer3-independent
 	to link against API-compatible library on top of libnfnetlink_queue 
 	instead of the current 'libipq'.
 Who:	Harald Welte <laforge@netfilter.org>
 ---------------------------
 What:	EXPORT_SYMBOL(lookup_hash)
 When:	January 2006
 Why:	Too low-level interface.  Use lookup_one_len or lookup_create instead.
 Who:	Christoph Hellwig <hch@lst.de>
@@ -0,0 +1,173 @@
 RCU-based dcache locking model
 ==============================
 On many workloads, the most common operation on dcache is to look up a
 dentry, given a parent dentry and the name of the child. Typically,
 for every open(), stat() etc., the dentry corresponding to the
 pathname will be looked up by walking the tree starting with the first
 component of the pathname and using that dentry along with the next
 component to look up the next level and so on. Since it is a frequent
 operation for workloads like multiuser environments and web servers,
 it is important to optimize this path.
 Prior to 2.5.10, dcache_lock was acquired in d_lookup and thus in
 every component during path look-up. Since 2.5.10 onwards, fast-walk
 algorithm changed this by holding the dcache_lock at the beginning and
 walking as many cached path component dentries as possible. This
 significantly decreases the number of acquisition of
 dcache_lock. However it also increases the lock hold time
 significantly and affects performance in large SMP machines. Since
 2.5.62 kernel, dcache has been using a new locking model that uses RCU
 to make dcache look-up lock-free.
 The current dcache locking model is not very different from the
 existing dcache locking model. Prior to 2.5.62 kernel, dcache_lock
 protected the hash chain, d_child, d_alias, d_lru lists as well as
 d_inode and several other things like mount look-up. RCU-based changes
 affect only the way the hash chain is protected. For everything else
 the dcache_lock must be taken for both traversing as well as
 updating. The hash chain updates too take the dcache_lock.  The
 significant change is the way d_lookup traverses the hash chain, it
 doesn't acquire the dcache_lock for this and rely on RCU to ensure
 that the dentry has not been *freed*.
 Dcache locking details
 ======================
 For many multi-user workloads, open() and stat() on files are very
 frequently occurring operations. Both involve walking of path names to
 find the dentry corresponding to the concerned file. In 2.4 kernel,
 dcache_lock was held during look-up of each path component. Contention
 and cache-line bouncing of this global lock caused significant
 scalability problems. With the introduction of RCU in Linux kernel,
 this was worked around by making the look-up of path components during
 path walking lock-free.
 Safe lock-free look-up of dcache hash table
 ===========================================
 Dcache is a complex data structure with the hash table entries also
 linked together in other lists. In 2.4 kernel, dcache_lock protected
 all the lists. We applied RCU only on hash chain walking. The rest of
 the lists are still protected by dcache_lock.  Some of the important
 changes are :
 1. The deletion from hash chain is done using hlist_del_rcu() macro
   which doesn't initialize next pointer of the deleted dentry and
   this allows us to walk safely lock-free while a deletion is
   happening.
 2. Insertion of a dentry into the hash table is done using
   hlist_add_head_rcu() which take care of ordering the writes - the
   writes to the dentry must be visible before the dentry is
   inserted. This works in conjunction with hlist_for_each_rcu() while
   walking the hash chain. The only requirement is that all
   initialization to the dentry must be done before
   hlist_add_head_rcu() since we don't have dcache_lock protection
   while traversing the hash chain. This isn't different from the
   existing code.
 3. The dentry looked up without holding dcache_lock by cannot be
   returned for walking if it is unhashed. It then may have a NULL
   d_inode or other bogosity since RCU doesn't protect the other
   fields in the dentry. We therefore use a flag DCACHE_UNHASHED to
   indicate unhashed dentries and use this in conjunction with a
   per-dentry lock (d_lock). Once looked up without the dcache_lock,
   we acquire the per-dentry lock (d_lock) and check if the dentry is
   unhashed. If so, the look-up is failed. If not, the reference count
   of the dentry is increased and the dentry is returned.
 4. Once a dentry is looked up, it must be ensured during the path walk
   for that component it doesn't go away. In pre-2.5.10 code, this was
   done holding a reference to the dentry. dcache_rcu does the same.
   In some sense, dcache_rcu path walking looks like the pre-2.5.10
   version.
 5. All dentry hash chain updates must take the dcache_lock as well as
   the per-dentry lock in that order. dput() does this to ensure that
   a dentry that has just been looked up in another CPU doesn't get
   deleted before dget() can be done on it.
 6. There are several ways to do reference counting of RCU protected
   objects. One such example is in ipv4 route cache where deferred
   freeing (using call_rcu()) is done as soon as the reference count
   goes to zero. This cannot be done in the case of dentries because
   tearing down of dentries require blocking (dentry_iput()) which
   isn't supported from RCU callbacks. Instead, tearing down of
   dentries happen synchronously in dput(), but actual freeing happens
   later when RCU grace period is over. This allows safe lock-free
   walking of the hash chains, but a matched dentry may have been
   partially torn down. The checking of DCACHE_UNHASHED flag with
   d_lock held detects such dentries and prevents them from being
   returned from look-up.
 Maintaining POSIX rename semantics
 ==================================
 Since look-up of dentries is lock-free, it can race against a
 concurrent rename operation. For example, during rename of file A to
 B, look-up of either A or B must succeed.  So, if look-up of B happens
 after A has been removed from the hash chain but not added to the new
 hash chain, it may fail.  Also, a comparison while the name is being
 written concurrently by a rename may result in false positive matches
 violating rename semantics.  Issues related to race with rename are
 handled as described below :
 1. Look-up can be done in two ways - d_lookup() which is safe from
   simultaneous renames and __d_lookup() which is not.  If
   __d_lookup() fails, it must be followed up by a d_lookup() to
   correctly determine whether a dentry is in the hash table or
   not. d_lookup() protects look-ups using a sequence lock
   (rename_lock).
 2. The name associated with a dentry (d_name) may be changed if a
   rename is allowed to happen simultaneously. To avoid memcmp() in
   __d_lookup() go out of bounds due to a rename and false positive
   comparison, the name comparison is done while holding the
   per-dentry lock. This prevents concurrent renames during this
   operation.
 3. Hash table walking during look-up may move to a different bucket as
   the current dentry is moved to a different bucket due to rename.
   But we use hlists in dcache hash table and they are
   null-terminated.  So, even if a dentry moves to a different bucket,
   hash chain walk will terminate. [with a list_head list, it may not
   since termination is when the list_head in the original bucket is
   reached].  Since we redo the d_parent check and compare name while
   holding d_lock, lock-free look-up will not race against d_move().
 4. There can be a theoretical race when a dentry keeps coming back to
   original bucket due to double moves. Due to this look-up may
   consider that it has never moved and can end up in a infinite loop.
   But this is not any worse that theoretical livelocks we already
   have in the kernel.
 Important guidelines for filesystem developers related to dcache_rcu
 ====================================================================
 1. Existing dcache interfaces (pre-2.5.62) exported to filesystem
   don't change. Only dcache internal implementation changes. However
   filesystems *must not* delete from the dentry hash chains directly
   using the list macros like allowed earlier. They must use dcache
   APIs like d_drop() or __d_drop() depending on the situation.
 2. d_flags is now protected by a per-dentry lock (d_lock). All access
   to d_flags must be protected by it.
 3. For a hashed dentry, checking of d_count needs to be protected by
   d_lock.
 Papers and other documentation on dcache locking
 ================================================
 1. Scaling dcache with RCU (http://linuxjournal.com/article.php?sid=7124).
 2. http://lse.sourceforge.net/locking/dcache/dcache.html
@@ -1812,11 +1812,6 @@ it may overflow the messages buffer, but try to get as much of it as
 you can
 if you get an Oops, run ksymoops to decode it so that the
 names of the offending functions are provided. A non-decoded Oops is
 pretty useless
 send a copy of your devfsd configuration file(s)
 send the bug report to me first.
--- a/Show More
+++ b/Show More