Thomas Graf [Wed, 20 Apr 2005 05:35:07 +0000 (22:35 -0700)]
[RTNETLINK]: Protocol family wildcard dumping for routing rules
Be kind to userspace and don't force them to hardcode protocol
families just to have it changed again once we support routing
rules for more than one protocol family.
Signed-off-by: Thomas Graf <tgraf@suug.ch> Signed-off-by: David S. Miller <davem@davemloft.net>
Herbert Xu [Wed, 20 Apr 2005 05:32:22 +0000 (22:32 -0700)]
[IPV6]: Replace bogus instances of inet->recverr
While looking at this problem I noticed that IPv6 was sometimes
looking at inet->recverr which is bogus. Here is a patch to
correct that and use np->recverr.
Signed-off-by: Herbert Xu <herbert@gondor.apana.org.au> Acked-by: Hideaki YOSHIFUJI <yoshfuji@linux-ipv6.org> Signed-off-by: David S. Miller <davem@davemloft.net>
ARM26 define FIRST_USER_ADDRESS as PAGE_SIZE (beyond the machine vectors when
they are mapped low), and use that definition in place of locally defined
MIN_MAP_ADDR. Previously, ARM26 permitted user mappings at 0 if the machine
vectors were mapped high; but that's inconsistent with ARM, and
FIRST_USER_ADDRESS would then have to be determined at runtime. Let's fix it
at PAGE_SIZE throughout the architecture.
ARM define FIRST_USER_ADDRESS as PAGE_SIZE (beyond the machine vectors when
they are mapped low), and use that definition in place of locally defined
MIN_MAP_ADDR.
Remove use of FIRST_USER_PGD_NR from sys_mincore: it's inconsistent (no other
syscall refers to it), unnecessary (sys_mincore loops over vmas further down)
and incorrect (misses user addresses in ARM's first pgd).
[PATCH] freepgt: free_pgtables from FIRST_USER_ADDRESS
The patches to free_pgtables by vma left problems on any architectures which
leave some user address page table entries unencapsulated by vma. Andi has
fixed the 32-bit vDSO on x86_64 to use a vma. Now fix arm (and arm26), whose
first PAGE_SIZE is reserved (perhaps) for machine vectors.
Our calls to free_pgtables must not touch that area, and exit_mmap's
BUG_ON(nr_ptes) must allow that arm's get_pgd_slow may (or may not) have
allocated an extra page table, which its free_pgd_slow would free later.
FIRST_USER_PGD_NR has misled me and others: until all the arches define
FIRST_USER_ADDRESS instead, a hack in mmap.c to derive one from t'other. This
patch fixes the bugs, the remaining patches just clean it up.
Once we're strict about clearing away page tables, hugetlb_prefault can assume
there are no page tables left within its range. Since the other arches
continue if !pte_none here, let i386 do the same.
ia64 and sparc64 hurriedly had to introduce their own variants of
pgd_addr_end, to leapfrog over the holes in their virtual address spaces which
the final clear_page_range suddenly presented when converted from pgd_index to
pgd_addr_end. But now that free_pgtables respects the vma list, those holes
are never presented, and the arch variants can go.
ia64 and ppc64 had hugetlb_free_pgtables functions which were no longer being
called, and it wasn't obvious what to do about them.
The ppc64 case turns out to be easy: the associated tables are noted elsewhere
and freed later, safe to either skip its hugetlb areas or go through the
motions of freeing nothing. Since ia64 does need a special case, restore to
ppc64 the special case of skipping them.
The ia64 hugetlb case has been broken since pgd_addr_end went in, though it
probably appeared to work okay if you just had one such area; in fact it's
been broken much longer if you consider a long munmap spanning from another
region into the hugetlb region.
In the ia64 hugetlb region, more virtual address bits are available than in
the other regions, yet the page tables are structured the same way: the page
at the bottom is larger. Here we need to scale down each addr before passing
it to the standard free_pgd_range. Was about to write a hugely_scaled_down
macro, but found htlbpage_to_page already exists for just this purpose. Fixed
off-by-one in ia64 is_hugepage_only_range.
Uninline free_pgd_range to make it available to ia64. Make sure the
vma-gathering loop in free_pgtables cannot join a hugepage_only_range to any
other (safe to join huges? probably but don't bother).
There's only one usage of MM_VM_SIZE(mm) left, and it's a troublesome macro
because mm doesn't contain the (32-bit emulation?) info needed. But it too is
only needed because we ignore the end from the vma list.
We could make flush_pgtables return that end, or unmap_vmas. Choose the
latter, since it's a natural fit with unmap_mapping_range_vma needing to know
its restart addr. This does make more than minimal change, but if unmap_vmas
had returned the end before, this is how we'd have done it, rather than
storing the break_addr in zap_details.
unmap_vmas used to return count of vmas scanned, but that's just debug which
hasn't been useful in a while; and if we want the map_count 0 on exit check
back, it can easily come from the final remove_vm_struct loop.
Recent woes with some arches needing their own pgd_addr_end macro; and 4-level
clear_page_range regression since 2.6.10's clear_page_tables; and its
long-standing well-known inefficiency in searching throughout the higher-level
page tables for those few entries to clear and free: all can be blamed on
ignoring the list of vmas when we free page tables.
Replace exit_mmap's clear_page_range of the total user address space by
free_pgtables operating on the mm's vma list; unmap_region use it in the same
way, giving floor and ceiling beyond which it may not free tables. This
brings lmbench fork/exec/sh numbers back to 2.6.10 (unless preempt is enabled,
in which case latency fixes spoil unmap_vmas throughput).
Beware: the do_mmap_pgoff driver failure case must now use unmap_region
instead of zap_page_range, since a page table might have been allocated, and
can only be freed while it is touched by some vma.
Move free_pgtables from mmap.c to memory.c, where its lower levels are adapted
from the clear_page_range levels. (Most of free_pgtables' old code was
actually for a non-existent case, prev not properly set up, dating from before
hch gave us split_vma.) Pass mmu_gather** in the public interfaces, since we
might want to add latency lockdrops later; but no attempt to do so yet, going
by vma should itself reduce latency.
But what if is_hugepage_only_range? Those ia64 and ppc64 cases need careful
examination: put that off until a later patch of the series.
What of x86_64's 32bit vdso page __map_syscall32 maps outside any vma?
And the range to sparc64's flush_tlb_pgtables? It's less clear to me now that
we need to do more than is done here - every PMD_SIZE ever occupied will be
flushed, do we really have to flush every PGDIR_SIZE ever partially occupied?
A shame to complicate it unnecessarily.
Special thanks to David Miller for time spent repairing my ceilings.
I can't use list.h, since sk_buff doesn't have a list_head but instead
has two struct sk_buff pointers, and I want to avoid any extra memory
allocation.
send outgoing packets in order
Signed-off-by: Ed L. Cashin <ecashin@coraid.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
The current <linux/debugfs.h> include file is a little fragile in that
it is not self-contained and hence may cause compile warnings or
errors depending on the files included before it, the kernel config
and the architecture. This patch makes things a little more robust by:
- including <linux/types.h> to get definitions of u32, mode_t, and so on.
- forward declaring struct file_operations.
- including <linux/err.h> when CONFIG_DEBUG_FS is not set
The last change is particularly useful, as a kernel developer is
likely to build with debugfs always enabled and never see the build
breakage cased if debugfs is disabled.
Signed-off-by: Roland Dreier <roland@topspin.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Robert Schwebel [Tue, 19 Apr 2005 04:57:32 +0000 (21:57 -0700)]
[PATCH] export platform_add_devices
platform_add_devices can be used from within modules, so it should be
exported. This can for example happen if you have hotpluggable firmware in
an FPGA on a system on chip processor; in our case the FPGA is probed for
devices and the FPGA base code registers the devices it has found with the
kernel.
(akpm: I think this is reasonable from a licensing POV: it's unlikely that
anyone would be interested in merging such specialised modules into mainline,
and it's a GPL export).
Signed-off-by: Robert Schwebel <r.schwebel@pengutronix.de> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
[PATCH] add TIMEOUT to firmware_class hotplug event
On Tue, 2005-03-15 at 09:25 +0100, Hannes Reinecke wrote:
> The current implementation of the firmware class breaks a fundamental
> assumption in udevd: that the physical device can be initialised fully
> prior to executing the next event for that device.
Here we add a TIMEOUT value to the hotplug environment of the firmware
requesting event. I will adapt udevd not to wait for anything else, if
it finds a TIMEOUT key.
Signed-off-by: Kay Sievers <kay.sievers@vrfy.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Jean Delvare [Tue, 19 Apr 2005 04:16:59 +0000 (21:16 -0700)]
[PATCH] I2C: Fix incorrect sysfs file permissions in it87 and via686a drivers
The it87 and via686a hardware monitoring drivers each create a sysfs
file named "alarms" in R/W mode, while they should really create it in
read-only mode. Since we don't provide a store function for these files,
write attempts to these files will do something undefined (I guess) and
bad (I am sure). My own try resulted in a locked terminal (where I
attempted the write) and a 100% CPU load until next reboot.
As a side note, wouldn't it make sense to check, when creating sysfs
files, that readable files have a non-NULL show method, and writable
files have a non-NULL store method? I know drivers are not supposed to
do stupid things, but there is already a BUG_ON for several conditions
in sysfs_create_file, so maybe we could add two more?
Signed-off-by: Jean Delvare <khali@linux-fr.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Alan Stern [Tue, 19 Apr 2005 00:39:30 +0000 (17:39 -0700)]
[PATCH] USB: USB API documentation modification
In response to complaints about excessive latency in the uhci-hcd driver
I'm planning to convert it to a top-half/bottom-half design. It turns out
that to do this, the USB API has to be modified slightly since the driver
will not be able to meet one of the guarantees in the current API. This
patch changes some kerneldoc, specifying the weaker guarantee.
Signed-off-by: Alan Stern <stern@rowland.harvard.edu> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Tue, 19 Apr 2005 00:39:30 +0000 (17:39 -0700)]
[PATCH] USB: OHCI on Compaq Aramada 7400
This adds a quirk to the OHCI driver that lets it work with an old
Compaq implementation. It also removes some needless strings from
the non-debug version of the driver.
Signed-off-by: Chris Clayton <chris_clayton@f1internet.com> Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Phil Dibowitz [Tue, 19 Apr 2005 00:39:27 +0000 (17:39 -0700)]
[PATCH] Fix GO_SLOW delay
This patch changes the delay for the US_FL_GO_SLOW patch from 110us to 125.
Some delays need this extra delay includign Jan De Luyck's drive which spawned
the original increase from 110 to 110us. 125 is a microframe, so this delay
seems to make sense more than just be a random delay (thanks to David Brownell
for pointing that out after my original patch).
drivers/usb/net/usbnet.c: In function `skb_return':
drivers/usb/net/usbnet.c:429: warning: int format, different type arg (arg 3)
drivers/usb/net/usbnet.c:429: warning: int format, different type arg (arg 3)
David Brownell [Tue, 19 Apr 2005 00:39:25 +0000 (17:39 -0700)]
[PATCH] USB: revert "fix" to usb_set_interface()
This reverts a recent change to usb_set_interface(). The change worked
around a quirk in certain devices, but doing this in usbcore creates
needless regressions for other devices. More appropriate fixes won't
put such handling in usbcore.
Basically it's tricky to do a full software reset of USB device state, since
the devices don't all act the same. This adds a note to the kerneldoc for
the usb_reset_configuration() call to highlight the quirk this was working
around: endpoint data toggles not being reset.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David Brownell [Tue, 19 Apr 2005 00:39:23 +0000 (17:39 -0700)]
[PATCH] USB: hcd suspend uses pm_message_t
This patch includes minor "sparse -Wbitwise" updates for the PCI based
HCDs. Almost all of them involve just changing the second parameter of the
suspend() method to a pm_message_t ... the others relate to how the EHCI
code walks in-memory data structures. (There's a minor bug fixed there too
... affecting the big-endian sysfs async schedule dump.)
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Index: gregkh-2.6/drivers/usb/core/hcd.h
===================================================================
David Brownell [Tue, 19 Apr 2005 00:39:22 +0000 (17:39 -0700)]
[PATCH] usb suspend updates (interface suspend)
This is the first of a few installments of PM API updates to match the
recent switch to "pm_message_t". This installment primarily affects
USB device drivers (for USB interfaces), and it changes the handful of
drivers which currently implement suspend methods:
- <linux/usb.h> and usbcore, signature change
- Some drivers only changed the signature, net effect this just
shuts up "sparse -Wbitwise":
* hid-core
* stir4200
- Two network drivers did that, and also grew slightly more
featureful suspend code ... they now properly shut down
their activities. (As should stir4200...)
* pegasus
* usbnet
Note that the Wake-On-Lan (WOL) support in pegasus doesn't yet work; looks
to me like it's missing a request to turn it on, vs just configuring it.
The ASIX code in usbnet also has WOL hooks that are ready to use; untested.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
Index: gregkh-2.6/drivers/net/irda/stir4200.c
===================================================================
David Brownell [Tue, 19 Apr 2005 00:39:22 +0000 (17:39 -0700)]
[PATCH] usb resume fixes
This has a variety of updates to the shared suspend/resume code for
PCI based USB host controllers.
- Cope with pm_message_t replacing the target system state.
This is actually a loss of functionality; PCI D1 and D2
states will no longer be used, and it's no longer knowable
that D3cold is on the way so power will be lost.
- Most importantly, some of the resume paths are reworked and
cleaned up. They're now an exact mirror of suspend paths,
and more care is taken to ensure the hardware is reactivated
before the hardware re-enables interrupts.
Plus comment and diagnostic cleanups; there are some nasty cases here
especially combined with swsusp, now they're somewhat commented.
Signed-off-by: David Brownell <dbrownell@users.sourceforge.net> Signed-off-by: Andrew Morton <akpm@osdl.org> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
diff -puN drivers/usb/core/hcd-pci.c~usb-resume-fixes drivers/usb/core/hcd-pci.c
Larry Battraw [Tue, 19 Apr 2005 00:39:20 +0000 (17:39 -0700)]
[PATCH] USB: visor Tapwave Zodiac support patch
Here's a tiny patch to add support for the Tapwave Zodiac (for
2.6.11.6). I've been meaning to send it in for a while but kept
upgrading my kernel and losing the changes :-) I own the device and it
works fine with the latest pilot-link beta.
From: Larry Battraw <lbattraw@insightbb.com> Signed-off-by: Greg Kroah-Hartman <gregkh@suse.de>
David S. Miller [Mon, 18 Apr 2005 22:13:15 +0000 (15:13 -0700)]
[PATCH] sparc64: Fix stat
Like Alpha, sparc64's struct stat was defined before we had the
nanosecond et al. fields added. So like Alpha I have to cons up a
struct stat64 to get this stuff. I'll work on the glibc bits soon.
Also, we were forgetting to fill in the nanosecond fields in the sparc
compat stat64 syscalls.
Signed-off-by: David S. Miller <davem@davemloft.net> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
aic7xxx: convert to SPI transport class Domain Validation
Now that we export all the parameters, this is easy to do.
It also means that we can dump about 2000 lines of code that
were dedicated to doing this internally.
Additionally, this removes all the aic7xxx driver abuse
of SCSI timers which were embedded in the DV routines.
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
This is just a simplistic patch to export all of the
aic7xxx internal transport parameters via the SPI
transport class. It doesn't actually alter the way the
driver works at all.
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
CC [M] drivers/scsi/qla2xxx/qla_attr.o
drivers/scsi/qla2xxx/qla_attr.c: In function `qla2x00_sysfs_write_fw_dump':
drivers/scsi/qla2xxx/qla_attr.c:65: warning: implicit declaration of function `vfree'
drivers/scsi/qla2xxx/qla_attr.c:83: warning: implicit declaration of function `vmalloc'
drivers/scsi/qla2xxx/qla_attr.c:83: warning: cast to pointer from integer of different size
Also remove spurious inclusion of linux/version.h
Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
Don't use cmd->request->nr_hw_segments as it may not be initialized
(SG_IO in particular bypasses anything that initializes this and just
uses scsi_do_req to insert a scsi_request directly on the head of the
queue) and a bogus value here can trip up the checks to make sure that
the number of segments will fit in the queue ring buffer, resulting in
commands that are never completed.
Fix up several issues with PCI DMA mapping and failure to check return
values on the mappings.
Make the check for space in the ring buffer happen after the DMA mapping
is done since any checks done before the mapping has taken place are
bogus.
Doug Ledford <dledford@redhat.com>.
Signed-off-by: Andrew Vasquez <andrew.vasquez@qlogic.com> Signed-off-by: James Bottomley <James.Bottomley@SteelEye.com>
David Howells [Mon, 18 Apr 2005 17:54:51 +0000 (10:54 -0700)]
[PATCH] Add 32-bit compatibility for NFSv4 mount
This adds 32-bit compatibility for mounting an NFSv4 mount on a 64-bit
kernel (such as happens with PPC64).
The problem is that the mount data for the NFS4 mount process includes
auxilliary data pointers, probably because the NFS4 mount data may
conceivably exceed PAGE_SIZE in size - thus breaking against the hard
limit imposed by sys_mount().
Signed-Off-By: David Howells <dhowells@redhat.com> Signed-off-by: Linus Torvalds <torvalds@osdl.org>
projects / mirror_ubuntu-focal-kernel.git / log