Alan Stern [Fri, 16 Jun 2006 21:10:48 +0000 (17:10 -0400)]
[PATCH] Driver Core: Make dev_info and friends print the bus name if there is no driver
This patch (as721) makes dev_info and related macros print the device's
bus name if the device doesn't have a driver, instead of printing just a
blank. If the device isn't on a bus either... well, then it does leave
a blank space. But it will be easier for someone else to change if they
want.
Cc: Matthew Wilcox <matthew@wil.cx> Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
[PATCH] Driver core: add proper symlinks for devices
We need to create the "compatible" symlinks that class_devices used to
create when they were in the class directories so that userspace does
not know anything changed at all.
Yeah, we have a lot of symlinks now, but we should be able to get rid of
them in a year or two... (wishful thinking...)
Kay Sievers [Thu, 15 Jun 2006 13:31:56 +0000 (15:31 +0200)]
[PATCH] Driver core: add generic "subsystem" link to all devices
Like the SUBSYTEM= key we find in the environment of the uevent, this
creates a generic "subsystem" link in sysfs for every device. Userspace
usually doesn't care at all if its a "class" or a "bus" device. This
provides an unified way to determine the subsytem of a device, regardless
of the way the driver core has created it.
Signed-off-by: Kay Sievers <kay.sievers@suse.de> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
[PATCH] Driver core: allow struct device to have a dev_t
This is the first step in moving class_device to being replaced by
struct device. It allows struct device to export a dev_t and makes it
easy to dynamically create and destroy struct device as long as they are
associated with a specific class.
Cc: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Michael Holzheu [Tue, 9 May 2006 10:53:49 +0000 (12:53 +0200)]
[PATCH] Driver Core: Add /sys/hypervisor when needed
To have a home for all hypervisors, this patch creates /sys/hypervisor.
A new config option SYS_HYPERVISOR is introduced, which should to be set
by architecture dependent hypervisors (e.g. s390 or Xen).
Acked-by: Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by: Michael Holzheu <holzheu@de.ibm.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Russell King [Sat, 6 May 2006 07:15:26 +0000 (08:15 +0100)]
[PATCH] Driver Core: Fix platform_device_add to use device_add
platform_device_add() should be using device_add() rather
than device_register() - any platform device passed to
platform_device_add() should have already been initialised,
either by platform_device_alloc() or platform_device_register().
Signed-off-by: Russell King <rmk@arm.linux.org.uk> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
class_device_add needs to check the return value of all the setup it
does. It doesn't handle out of memory well. This is not complete, probably
more needs to be done.
Signed-off-by: Stephen Hemminger <shemminger@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Hansjoerg Lipp [Sat, 22 Apr 2006 16:43:00 +0000 (18:43 +0200)]
[PATCH] i4l gigaset: move sysfs entry to tty class device
Using the class device pointer returned by tty_register_device() with
part 1 of the patch, attach the Gigaset drivers' "cidmode" sysfs entry
to its tty class device, where it can be found more easily by users
who do not know nor care which USB port the device is attached to.
Hansjoerg Lipp [Sat, 22 Apr 2006 16:36:53 +0000 (18:36 +0200)]
[PATCH] TTY: return class device pointer from tty_register_device()
Let tty_register_device() return a pointer to the class device it creates.
This allows registrants to add their own sysfs files under the class
device node.
Brice Goglin [Wed, 21 Jun 2006 03:03:02 +0000 (20:03 -0700)]
[PATCH] add __iowrite64_copy
Introduce __iowrite64_copy. It will be used by the Myri-10G Ethernet
driver to post requests to the NIC. This driver will be submitted soon.
__iowrite64_copy copies to I/O memory in units of 64 bits when possible (on
64 bit architectures). It reverts to __iowrite32_copy on 32 bit
architectures.
* git://git.kernel.org/pub/scm/linux/kernel/git/bcollins/linux1394-2.6: (28 commits)
eth1394: replace __constant_htons by htons
ieee1394: adjust code formatting in highlevel.c
ieee1394: hl_irqs_lock is taken in hardware interrupt context
ieee1394_core: switch to kthread API
ieee1394: sbp2: Kconfig fix
ieee1394: add preprocessor constant for invalid csr address
sbp2: fix deregistration of status fifo address space
[PATCH] eth1394: endian fixes
Fix broken suspend/resume in ohci1394
sbp2: use __attribute__((packed)) for on-the-wire structures
sbp2: provide helptext for CONFIG_IEEE1394_SBP2_PHYS_DMA and mark it experimental
Update feature removal of obsolete raw1394 ISO requests.
sbp2: fix S800 transfers if phys_dma is off
sbp2: remove ohci1394 specific constant
ohci1394: make phys_dma parameter read-only
ohci1394: set address range properties
ieee1394: extend lowlevel API for address range properties
sbp2: log number of supported concurrent logins
sbp2: remove manipulation of inquiry response
ieee1394: save RAM by using a single tlabel for broadcast transactions
...
Linus Torvalds [Wed, 21 Jun 2006 02:44:03 +0000 (19:44 -0700)]
Fix up CFQ scheduler for recent rbtree node shrinkage
The color is now in the low bits of the parent pointer, and initializing
it to 0 happens as part of the whole memset above, so just remove the
unnecessary RB_CLEAR_COLOR.
Herbert Xu [Wed, 21 Jun 2006 00:53:54 +0000 (10:53 +1000)]
[FORCEDETH] Fix xmit_lock/netif_tx_lock after merge
There has been an update to the forcedeth driver that added a few new
uses of xmit_lock which is no longer meant to be used directly. This
patch replaces them with netif_tx_lock_bh.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
Linus Torvalds [Tue, 20 Jun 2006 22:38:12 +0000 (15:38 -0700)]
Merge branch 'rio.b19' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/bird
* 'rio.b19' of git://git.kernel.org/pub/scm/linux/kernel/git/viro/bird:
[PATCH] missing readb/readw in rio
[PATCH] copy_to_user() from iomem is a bad thing
[PATCH] forgotten swap of copyout() arguments
[PATCH] handling rio MEMDUMP
[PATCH] fix rio_copy_to_card() for OLDPCI case
[PATCH] uses of ->Copy() in rioroute are bogus
[PATCH] bogus order of copy_from_user() arguments
[PATCH] rio ->Copy() expects the sourse as first argument
[PATCH] trivial annotations in rio
Linus Torvalds [Tue, 20 Jun 2006 22:10:08 +0000 (15:10 -0700)]
Merge git://git.infradead.org/hdrcleanup-2.6
* git://git.infradead.org/hdrcleanup-2.6: (63 commits)
[S390] __FD_foo definitions.
Switch to __s32 types in joystick.h instead of C99 types for consistency.
Add <sys/types.h> to headers included for userspace in <linux/input.h>
Move inclusion of <linux/compat.h> out of user scope in asm-x86_64/mtrr.h
Remove struct fddi_statistics from user view in <linux/if_fddi.h>
Move user-visible parts of drivers/s390/crypto/z90crypt.h to include/asm-s390
Revert include/media changes: Mauro says those ioctls are only used in-kernel(!)
Include <linux/types.h> and use __uXX types in <linux/cramfs_fs.h>
Use __uXX types in <linux/i2o_dev.h>, include <linux/ioctl.h> too
Remove private struct dx_hash_info from public view in <linux/ext3_fs.h>
Include <linux/types.h> and use __uXX types in <linux/affs_hardblocks.h>
Use __uXX types in <linux/divert.h> for struct divert_blk et al.
Use __u32 for elf_addr_t in <asm-powerpc/elf.h>, not u32. It's user-visible.
Remove PPP_FCS from user view in <linux/ppp_defs.h>, remove __P mess entirely
Use __uXX types in user-visible structures in <linux/nbd.h>
Don't use 'u32' in user-visible struct ip_conntrack_old_tuple.
Use __uXX types for S390 DASD volume label definitions which are user-visible
S390 BIODASDREADCMB ioctl should use __u64 not u64 type.
Remove unneeded inclusion of <linux/time.h> from <linux/ufs_fs.h>
Fix private integer types used in V4L2 ioctls.
...
Manually resolve conflict in include/linux/mtd/physmap.h
Linus Torvalds [Tue, 20 Jun 2006 21:51:22 +0000 (14:51 -0700)]
Merge git://git.infradead.org/~dwmw2/rbtree-2.6
* git://git.infradead.org/~dwmw2/rbtree-2.6:
[RBTREE] Switch rb_colour() et al to en_US spelling of 'color' for consistency
Update UML kernel/physmem.c to use rb_parent() accessor macro
[RBTREE] Update hrtimers to use rb_parent() accessor macro.
[RBTREE] Add explicit alignment to sizeof(long) for struct rb_node.
[RBTREE] Merge colour and parent fields of struct rb_node.
[RBTREE] Remove dead code in rb_erase()
[RBTREE] Update JFFS2 to use rb_parent() accessor macro.
[RBTREE] Update eventpoll.c to use rb_parent() accessor macro.
[RBTREE] Update key.c to use rb_parent() accessor macro.
[RBTREE] Update ext3 to use rb_parent() accessor macro.
[RBTREE] Change rbtree off-tree marking in I/O schedulers.
[RBTREE] Add accessor macros for colour and parent fields of rb_node
Linus Torvalds [Tue, 20 Jun 2006 21:50:31 +0000 (14:50 -0700)]
Merge git://git.infradead.org/mtd-2.6
* git://git.infradead.org/mtd-2.6: (199 commits)
[MTD] NAND: Fix breakage all over the place
[PATCH] NAND: fix remaining OOB length calculation
[MTD] NAND Fixup NDFC merge brokeness
[MTD NAND] S3C2410 driver cleanup
[MTD NAND] s3c24x0 board: Fix clock handling, ensure proper initialisation.
[JFFS2] Check CRC32 on dirent and data nodes each time they're read
[JFFS2] When retiring nextblock, allocate a node_ref for the wasted space
[JFFS2] Mark XATTR support as experimental, for now
[JFFS2] Don't trust node headers before the CRC is checked.
[MTD] Restore MTD_ROM and MTD_RAM types
[MTD] assume mtd->writesize is 1 for NOR flashes
[MTD NAND] Fix s3c2410 NAND driver so it at least _looks_ like it compiles
[MTD] Prepare physmap for 64-bit-resources
[JFFS2] Fix more breakage caused by janitorial meddling.
[JFFS2] Remove stray __exit from jffs2_compressors_exit()
[MTD] Allow alternate JFFS2 mount variant for root filesystem.
[MTD] Disconnect struct mtd_info from ABI
[MTD] replace MTD_RAM with MTD_GENERIC_TYPE
[MTD] replace MTD_ROM with MTD_GENERIC_TYPE
[MTD] remove a forgotten MTD_XIP
...
Linus Torvalds [Tue, 20 Jun 2006 21:49:00 +0000 (14:49 -0700)]
Merge master.kernel.org:/home/rmk/linux-2.6-arm
* master.kernel.org:/home/rmk/linux-2.6-arm: (22 commits)
[ARM] 3559/1: S3C2442: core and serial port
[ARM] 3557/1: S3C24XX: centralise and cleanup uart registration
[ARM] 3558/1: SMDK24XX: LED platform devices
[ARM] 3534/1: add spi support to lubbock platform
[ARM] 3554/1: ARM: Fix dyntick locking
[ARM] 3553/1: S3C24XX: earlier print of cpu idcode info
[ARM] 3552/1: S3C24XX: Move VA of GPIO for low-level debug
[ARM] 3551/1: S3C24XX: PM code failes to compile with CONFIG_DCACHE_WRITETHROUGH
[ARM] 3550/1: OSIRIS: fix serial port map for 1:1
[ARM] 3548/1: Fix the ARMv6 CPU id in compressed/head.S
[ARM] 3335/1: Old-abi Thumb sys_syscall broken
[ARM] 3467/1: [3/3] Support for Philips PNX4008 platform: defconfig
[ARM] 3466/1: [2/3] Support for Philips PNX4008 platform: chip support
[ARM] 3465/1: [1/3] Support for Philips PNX4008 platform: headers
[ARM] 3407/1: lpd7x: documetation update
[ARM] 3406/1: lpd7x: compilation fix for smc91x
[ARM] 3405/1: lpd7a40x: CPLD ssp driver
[ARM] 3404/1: lpd7a40x: AMBA CLCD support
[ARM] 3403/1: lpd7a40x: updated default configurations
[ARM] 3402/1: lpd7a40x: serial driver bug fix
...
Thomas Gleixner [Tue, 20 Jun 2006 18:05:05 +0000 (20:05 +0200)]
[MTD] NAND: Fix breakage all over the place
Following problems are addressed:
- wrong status caused early break out of nand_wait()
- removed the bogus status check in nand_wait() which
is a relict of the abandoned support for interrupted
erase.
- status check moved to the correct place in read_oob
- oob support for syndrom based ecc with strange layouts
- use given offset in the AUTOOOB based oob operations
Partially based on a patch from Vitaly Vool <vwool@ru.mvista.com>
Thanks to Savin Zlobec <savin@epico.si> for tracking down the
status problem.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
In nand_read_page_syndrome/nand_write_page_syndrome the calculation of
the remaining oob length which is not used by the prepad/ecc/postpad
areas is wrong.
Signed-off-by: Vitaly Wool <vwool@ru.mvista.com> Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Andrew Victor [Tue, 20 Jun 2006 18:30:20 +0000 (19:30 +0100)]
[ARM] 3607/1: AT91RM9200 misc fixes
Patch from Andrew Victor
This final patch includes some general fixes.
1. Link in pm.o if CONFIG_PM is enabled. [Should have been included in
patch 3605/1].
2. Use __raw_readl()/__raw_writel() when accessing System Peripheral
registers.
3. Removed some unnecessary includes
Signed-off-by: Andrew Victor <andrew@sanpeople.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Andrew Victor [Tue, 20 Jun 2006 18:30:19 +0000 (19:30 +0100)]
[ARM] 3605/1: AT91RM9200 Power Management
Patch from Andrew Victor
This patch adds the core Power Management support for the AT91RM9200
processor. It will support suspend-to-RAM and standby modes.
The suspend-to-RAM functionality is not 100% complete. The code that
needs to be execute from the internal SRAM to restore the system is
outstanding. For now we just fall through to Standby mode.
The AT91-specific at91_suspend_entering_slow_clock() function will
eventually be replaced by clk_must_disable() once that functionality is
added to mainline clock API.
Patch from David Brownell.
Signed-off-by: Andrew Victor <andrew@sanpeople.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Andrew Victor [Tue, 20 Jun 2006 18:27:34 +0000 (19:27 +0100)]
[ARM] 3603/1: AT91RM9200 remove old files
Patch from Andrew Victor
This patch removes the now unnecessary
include/asm-arm/arch-at91rm9200/pio.h file. The pin configuration is
now handled in the platform-device setup code
(arch/arm/mach-at91rm9200/devices.c).
Signed-off-by: Andrew Victor <andrew@sanpeople.com> Signed-off-by: Russell King <rmk+kernel@arm.linux.org.uk>
Steve Grubb [Mon, 12 Jun 2006 11:48:28 +0000 (07:48 -0400)]
[PATCH] make set_loginuid obey audit_enabled
Hi,
I was doing some testing and noticed that when the audit system was disabled,
I was still getting messages about the loginuid being set. The following patch
makes audit_set_loginuid look at in_syscall to determine if it should create
an audit event. The loginuid will continue to be set as long as there is a context.
Signed-off-by: Steve Grubb <sgrubb@redhat.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Amy Griffis [Fri, 9 Jun 2006 03:19:31 +0000 (23:19 -0400)]
[PATCH] log more info for directory entry change events
When an audit event involves changes to a directory entry, include
a PATH record for the directory itself. A few other notable changes:
- fixed audit_inode_child() hooks in fsnotify_move()
- removed unused flags arg from audit_inode()
- added audit log routines for logging a portion of a string
In this implementation, audit registers inotify watches on the parent
directories of paths specified in audit rules. When audit's inotify
event handler is called, it updates any affected rules based on the
filesystem event. If the parent directory is renamed, removed, or its
filesystem is unmounted, audit removes all rules referencing that
inotify watch.
To keep things simple, this implementation limits location-based
auditing to the directory entries in an existing directory. Given
a path-based rule for /foo/bar/passwd, the following table applies:
passwd modified -- audit event logged
passwd replaced -- audit event logged, rules list updated
bar renamed -- rule removed
foo renamed -- untracked, meaning that the rule now applies to
the new location
Audit users typically want to have many rules referencing filesystem
objects, which can significantly impact filtering performance. This
patch also adds an inode-number-based rule hash to mitigate this
situation.
The patch is relative to the audit git tree:
http://kernel.org/git/?p=linux/kernel/git/viro/audit-current.git;a=summary
and uses the inotify kernel API:
http://lkml.org/lkml/2006/6/1/145
Signed-off-by: Amy Griffis <amy.griffis@hp.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
George C. Wilson [Wed, 24 May 2006 21:09:55 +0000 (16:09 -0500)]
[PATCH] Audit of POSIX Message Queue Syscalls v.2
This patch adds audit support to POSIX message queues. It applies cleanly to
the lspp.b15 branch of Al Viro's git tree. There are new auxiliary data
structures, and collection and emission routines in kernel/auditsc.c. New hooks
in ipc/mqueue.c collect arguments from the syscalls.
I tested the patch by building the examples from the POSIX MQ library tarball.
Build them -lrt, not against the old MQ library in the tarball. Here's the URL:
http://www.geocities.com/wronski12/posix_ipc/libmqueue-4.41.tar.gz
Do auditctl -a exit,always -S for mq_open, mq_timedsend, mq_timedreceive,
mq_notify, mq_getsetattr. mq_unlink has no new hooks. Please see the
corresponding userspace patch to get correct output from auditd for the new
record types.
[fixes folded]
Signed-off-by: George Wilson <ltcgcw@us.ibm.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Linda Knippers [Wed, 17 May 2006 02:03:48 +0000 (22:03 -0400)]
[PATCH] update of IPC audit record cleanup
The following patch addresses most of the issues with the IPC_SET_PERM
records as described in:
https://www.redhat.com/archives/linux-audit/2006-May/msg00010.html
and addresses the comments I received on the record field names.
To summarize, I made the following changes:
1. Changed sys_msgctl() and semctl_down() so that an IPC_SET_PERM
record is emitted in the failure case as well as the success case.
This matches the behavior in sys_shmctl(). I could simplify the
code in sys_msgctl() and semctl_down() slightly but it would mean
that in some error cases we could get an IPC_SET_PERM record
without an IPC record and that seemed odd.
2. No change to the IPC record type, given no feedback on the backward
compatibility question.
3. Removed the qbytes field from the IPC record. It wasn't being
set and when audit_ipc_obj() is called from ipcperms(), the
information isn't available. If we want the information in the IPC
record, more extensive changes will be necessary. Since it only
applies to message queues and it isn't really permission related, it
doesn't seem worth it.
4. Removed the obj field from the IPC_SET_PERM record. This means that
the kern_ipc_perm argument is no longer needed.
5. Removed the spaces and renamed the IPC_SET_PERM field names. Replaced iuid and
igid fields with ouid and ogid in the IPC record.
I tested this with the lspp.22 kernel on an x86_64 box. I believe it
applies cleanly on the latest kernel.
-- ljk
Signed-off-by: Linda Knippers <linda.knippers@hp.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Just a few minor proposed updates. Only the last one will
actually affect behavior. The rest are just misleading
code.
Several AUDIT_SET functions return 'old' value, but only
return value <0 is checked for. So just return 0.
propagate audit_set_rate_limit and audit_set_backlog_limit
error values
In audit_buffer_free, the audit_freelist_count was being
incremented even when we discard the return buffer, so
audit_freelist_count can end up wrong. This could cause
the actual freelist to shrink over time, eventually
threatening to degrate audit performance.
Signed-off-by: Serge E. Hallyn <serue@us.ibm.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Al Viro [Mon, 22 May 2006 05:09:24 +0000 (01:09 -0400)]
[PATCH] fix deadlocks in AUDIT_LIST/AUDIT_LIST_RULES
We should not send a pile of replies while holding audit_netlink_mutex
since we hold the same mutex when we receive commands. As the result,
we can get blocked while sending and sit there holding the mutex while
auditctl is unable to send the next command and get around to receiving
what we'd sent.
Solution: create skb and put them into a queue instead of sending;
once we are done, send what we've got on the list. The former can
be done synchronously while we are handling AUDIT_LIST or AUDIT_LIST_RULES;
we are holding audit_netlink_mutex at that point. The latter is done
asynchronously and without messing with audit_netlink_mutex.
Update kernel documentation to include a description of the inotify
kernel API.
Signed-off-by: Amy Griffis <amy.griffis@hp.com> Acked-by: Robert Love <rml@novell.com> Acked-by: John McCutchan <john@johnmccutchan.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Amy Griffis [Thu, 1 Jun 2006 20:11:05 +0000 (13:11 -0700)]
[PATCH] inotify (4/5): allow watch removal from event handler
Allow callers to remove watches from their event handler via
inotify_remove_watch_locked(). This functionality can be used to
achieve IN_ONESHOT-like functionality for a subset of events in the
mask.
Signed-off-by: Amy Griffis <amy.griffis@hp.com> Acked-by: Robert Love <rml@novell.com> Acked-by: John McCutchan <john@johnmccutchan.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Amy Griffis [Thu, 1 Jun 2006 20:11:03 +0000 (13:11 -0700)]
[PATCH] inotify (3/5): add interfaces to kernel API
Add inotify_init_watch() so caller can use inotify_watch refcounts
before calling inotify_add_watch().
Add inotify_find_watch() to find an existing watch for an (ih,inode)
pair. This is similar to inotify_find_update_watch(), but does not
update the watch's mask if one is found.
Add inotify_rm_watch() to remove a watch via the watch pointer instead
of the watch descriptor.
Signed-off-by: Amy Griffis <amy.griffis@hp.com> Acked-by: Robert Love <rml@novell.com> Acked-by: John McCutchan <john@johnmccutchan.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Amy Griffis [Thu, 1 Jun 2006 20:11:01 +0000 (13:11 -0700)]
[PATCH] inotify (2/5): add name's inode to event handler
When an inotify event includes a dentry name, also include the inode
associated with that name.
Signed-off-by: Amy Griffis <amy.griffis@hp.com> Acked-by: Robert Love <rml@novell.com> Acked-by: John McCutchan <john@johnmccutchan.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
Amy Griffis [Thu, 1 Jun 2006 20:10:59 +0000 (13:10 -0700)]
[PATCH] inotify (1/5): split kernel API from userspace support
The following series of patches introduces a kernel API for inotify,
making it possible for kernel modules to benefit from inotify's
mechanism for watching inodes. With these patches, inotify will
maintain for each caller a list of watches (via an embedded struct
inotify_watch), where each inotify_watch is associated with a
corresponding struct inode. The caller registers an event handler and
specifies for which filesystem events their event handler should be
called per inotify_watch.
Signed-off-by: Amy Griffis <amy.griffis@hp.com> Acked-by: Robert Love <rml@novell.com> Acked-by: John McCutchan <john@johnmccutchan.com> Signed-off-by: Al Viro <viro@zeniv.linux.org.uk>
David S. Miller [Tue, 20 Jun 2006 08:23:32 +0000 (01:23 -0700)]
[SPARC64]: Move over to GENERIC_HARDIRQS.
This is the long overdue conversion of sparc64 over to
the generic IRQ layer.
The kernel image is slightly larger, but the BSS is ~60K
smaller due to the reduced size of struct ino_bucket.
A lot of IRQ implementation details, including ino_bucket,
were moved out of asm-sparc64/irq.h and are now private to
arch/sparc64/kernel/irq.c, and most of the code in irq.c
totally disappeared.
One thing that's different at the moment is IRQ distribution,
we do it at enable_irq() time. If the cpu mask is ALL then
we round-robin using a global rotating cpu counter, else
we pick the first cpu in the mask to support single cpu
targetting. This is similar to what powerpc's XICS IRQ
support code does.
This works fine on my UP SB1000, and the SMP build goes
fine and runs on that machine, but lots of testing on
different setups is needed.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 20 Jun 2006 08:22:35 +0000 (01:22 -0700)]
[SPARC64]: Virtualize IRQ numbers.
Inspired by PowerPC XICS interrupt support code.
All IRQs are virtualized in order to keep NR_IRQS from needing
to be too large. Interrupts on sparc64 are arbitrary 11-bit
values, but we don't need to define NR_IRQS to 2048 if we
virtualize the IRQs.
As PCI and SBUS controller drivers build device IRQs, we divy
out virtual IRQ numbers incrementally starting at 1. Zero is
a special virtual IRQ used for the timer interrupt.
So device drivers all see virtual IRQs, and all the normal
interfaces such as request_irq(), enable_irq(), etc. translate
that into a real IRQ number in order to configure the IRQ.
At this point knowledge of the struct ino_bucket is almost
entirely contained within arch/sparc64/kernel/irq.c There are
a few small bits in the PCI controller drivers that need to
be swept away before we can remove ino_bucket's definition
out of asm-sparc64/irq.h and privately into kernel/irq.c
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 20 Jun 2006 08:21:57 +0000 (01:21 -0700)]
[SPARC64]: Kill ino_bucket->pil
And reuse that struct member for virt_irq, which will
be used in future changesets for the implementation of
mapping between real and virtual IRQ numbers.
This nicely kills off a ton of SBUS and PCI controller
PIL assignment code which is no longer necessary.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 20 Jun 2006 08:21:29 +0000 (01:21 -0700)]
[SPARC]: Kill __irq_itoa().
This ugly hack was long overdue to die.
It was a way to print out Sparc interrupts in a more freindly format,
since IRQ numbers were arbitrary opaque 32-bit integers which vectored
into PIL levels. These 32-bit integers were not necessarily in the
0-->NR_IRQS range, but the PILs they vectored to were.
The idea now is that we will increase NR_IRQS a little bit and use a
virtual<-->real IRQ number mapping scheme similar to PowerPC.
That makes this IRQ printing hack irrelevant, and furthermore only a
handful of drivers actually used __irq_itoa() making it even less
useful.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 20 Jun 2006 08:20:30 +0000 (01:20 -0700)]
[SPARC64]: bp->pil can never be zero
Only pil0_dummy_bucket had a pil of zero and we just killed that
off, so we can delete all special case code that used bp->pil==0
as a way to identify a dummy bucket.
Signed-off-by: David S. Miller <davem@davemloft.net>
David S. Miller [Tue, 20 Jun 2006 08:20:00 +0000 (01:20 -0700)]
[SPARC64]: Send all device interrupts via one PIL.
This is the first in a series of cleanups that will hopefully
allow a seamless attempt at using the generic IRQ handling
infrastructure in the Linux kernel.
Define PIL_DEVICE_IRQ and vector all device interrupts through
there.
Get rid of the ugly pil0_dummy_{bucket,desc}, instead vector
the timer interrupt directly to a specific handler since the
timer interrupt is the only event that will be signaled on
PIL 14.
The irq_worklist is now in the per-cpu trap_block[].
Signed-off-by: David S. Miller <davem@davemloft.net>
Bob Breuer [Tue, 20 Jun 2006 07:36:56 +0000 (00:36 -0700)]
[SPARC]: Fix iommu_flush_iotlb end address
Fix the calculation of the end address when flushing iotlb entries to
ram. This bug has been a cause of esp dma errors, and it affects
HyperSPARC systems much worse than SuperSPARC systems.
Signed-off-by: Bob Breuer <breuerr@mc.net> Signed-off-by: David S. Miller <davem@davemloft.net>
Herbert Xu [Tue, 20 Jun 2006 06:57:59 +0000 (23:57 -0700)]
[NET]: Prevent multiple qdisc runs
Having two or more qdisc_run's contend against each other is bad because
it can induce packet reordering if the packets have to be requeued. It
appears that this is an unintended consequence of relinquinshing the queue
lock while transmitting. That in turn is needed for devices that spend a
lot of time in their transmit routine.
There are no advantages to be had as devices with queues are inherently
single-threaded (the loopback device is not but then it doesn't have a
queue).
Even if you were to add a queue to a parallel virtual device (e.g., bolt
a tbf filter in front of an ipip tunnel device), you would still want to
process the queue in sequence to ensure that the packets are ordered
correctly.
The solution here is to steal a bit from net_device to prevent this.
BTW, as qdisc_restart is no longer used by anyone as a module inside the
kernel (IIRC it used to with netif_wake_queue), I have not exported the
new __qdisc_run function.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Signed-off-by: David S. Miller <davem@davemloft.net>
projects / linux-2.6 / log