]> git.neil.brown.name Git - history.git/log
history.git
23 years agoMissed <linux/init.h> in CONFIG_SWAP changes.
Linus Torvalds [Sun, 3 Nov 2002 02:54:31 +0000 (18:54 -0800)]
Missed <linux/init.h> in CONFIG_SWAP changes.

23 years ago[PATCH] page_alloc.c uClinux bits
Christoph Hellwig [Sun, 3 Nov 2002 02:46:07 +0000 (18:46 -0800)]
[PATCH] page_alloc.c uClinux bits

nommu ports need to set a reference on all allocated pages.

23 years ago[PATCH] filemap.c bits for uClinux
Christoph Hellwig [Sun, 3 Nov 2002 02:46:01 +0000 (18:46 -0800)]
[PATCH] filemap.c bits for uClinux

Without CONFIG_MMU we don't need to stub out mmap().  Move one function
around inside filemap.c to need just one ifdef.

23 years ago[PATCH] make swap code conditional
Christoph Hellwig [Sun, 3 Nov 2002 02:45:55 +0000 (18:45 -0800)]
[PATCH] make swap code conditional

Make the swap code conditional on CONFIG_SWAP.  This is mostly for
uClinux, but !CONFIG_SWAP compiles and boots fine for i386, too -
the only problem I've seen is that X doesn't starts, it's probably
shm-related, thus it's disabled unconditionally for "normal" arches.

The patch makes three files in mm/ conditional on CONFIG_SWAP, reorganzies
include/linux/swap.h big time to provide stubs for the !CONFIG_SWAP case,
moves the remaining /proc/swaps code to swapfile.c and cleans up some
more MM code to compile fine without CONFIG_SWAP

23 years ago[PATCH] add CONFIG_MMU and CONFIG_SWAP
Christoph Hellwig [Sun, 3 Nov 2002 02:45:48 +0000 (18:45 -0800)]
[PATCH] add CONFIG_MMU and CONFIG_SWAP

Now that m68knommu and v850 are merged we need all other architectures
to define CONFIG_SWAP and CONFIG_MMU so that we can make code
conditional on it.

23 years ago[PATCH] Make x86-64 compile
Andi Kleen [Sat, 2 Nov 2002 06:05:31 +0000 (22:05 -0800)]
[PATCH] Make x86-64 compile

Quick patch to make x86-64 in 2.4.45 compile with the epoll changes.

23 years agoSet command length for the START_STOP command.
Linus Torvalds [Sat, 2 Nov 2002 05:58:10 +0000 (21:58 -0800)]
Set command length for the START_STOP command.

23 years agoMerge http://gkernel.bkbits.net/minitramfs-2.5
Linus Torvalds [Sat, 2 Nov 2002 05:36:53 +0000 (21:36 -0800)]
Merge http://gkernel.bkbits.net/minitramfs-2.5
into home.transmeta.com:/home/torvalds/v2.5/linux

23 years ago[PATCH] isapnp: fix typo in isapnp_proc_done when CONFIG_PROC_FS is not set
Arnaldo Carvalho de Melo [Sat, 2 Nov 2002 05:30:00 +0000 (21:30 -0800)]
[PATCH] isapnp: fix typo in isapnp_proc_done when CONFIG_PROC_FS is not set

23 years agoMerge bk://bk.arm.linux.org.uk
Linus Torvalds [Sat, 2 Nov 2002 05:12:01 +0000 (21:12 -0800)]
Merge bk://bk.arm.linux.org.uk
into home.transmeta.com:/home/torvalds/v2.5/linux

23 years ago[SERIAL] [PARISC] add support for GSC serial
Russell King [Sat, 2 Nov 2002 17:10:09 +0000 (17:10 +0000)]
[SERIAL] [PARISC] add support for GSC serial
Add discovery for PA-RISC 16550 serial ports.

Patch from Matthew Wilcox.

23 years ago[SERIAL] Set port->type to unknown only when using autoconfig
Russell King [Sat, 2 Nov 2002 16:52:58 +0000 (16:52 +0000)]
[SERIAL] Set port->type to unknown only when using autoconfig

23 years ago[SERIAL] Fix build errors and warnings
Russell King [Sat, 2 Nov 2002 16:47:53 +0000 (16:47 +0000)]
[SERIAL] Fix build errors and warnings
- Remove last usage of EVT_WRITE_WAKEUP
- Remove unused label + kfree()
- scratch3 now known as id3

23 years ago[SERIAL] Tidy up 8250 port type detection
Russell King [Sat, 2 Nov 2002 15:26:37 +0000 (15:26 +0000)]
[SERIAL] Tidy up 8250 port type detection
The original port detection code was one large function with many
tests without any clear structure.

This cset cleans this up, splitting out the tests for 8250/16450
vs 16550 vs 16550A vs later ports, and adds comments about why
we test what we're testing.

23 years ago[SERIAL] Make ALPHA_KLUDGE_MCR more generic for bluetooth modems, etc
Russell King [Sat, 2 Nov 2002 14:36:42 +0000 (14:36 +0000)]
[SERIAL] Make ALPHA_KLUDGE_MCR more generic for bluetooth modems, etc
In addition to the Alpha OUT1/OUT2 kludge, devices like Bluetooth
modems connected to serial ports make use of the modem control lines
in non-standard ways.  Therefore, we implement a more flexible way to
allow the modem control outputs to be forced to particular values
irrespective of the normal usage of these signals.

23 years ago[SERIAL] Fix up formatting of serial names
Russell King [Sat, 2 Nov 2002 14:11:39 +0000 (14:11 +0000)]
[SERIAL] Fix up formatting of serial names
The tty layer's tty_name requires formatting codes in driver->name
for the devfs and non-devfs cases.

23 years ago[SERIAL] Fix up 8250 IRQ chain handling
Russell King [Sat, 2 Nov 2002 12:00:14 +0000 (12:00 +0000)]
[SERIAL] Fix up 8250 IRQ chain handling
- Unlink port from chain if request_irq fails
- Allow port to define whether we use shared IRQs
  (this is required for PCMCIA serial cards where the serial
   card shares the same IRQ as the socket status IRQ.)
- Convert remaining ASYNC_ flags to UPF_ flags

23 years ago[SERIAL] Fix two incorrect uses of __FUNCTION__
Russell King [Sat, 2 Nov 2002 11:32:57 +0000 (11:32 +0000)]
[SERIAL] Fix two incorrect uses of __FUNCTION__

23 years ago[SERIAL] Remove struct pci_board from init_fn
Russell King [Sat, 2 Nov 2002 11:24:58 +0000 (11:24 +0000)]
[SERIAL] Remove struct pci_board from init_fn
Traditionally, we allocated the private array of port parameters based on
the detected board->num_ports, and then called the init_fn with the board
pointer (which points into the global table.)  Some init_fn implementations
modify num_ports, which therefore affects the global table.  Unfortunately,
this means that if the init_fn increases num_ports (because we have two
almost identical cards, the first with a smaller number of ports than the
second), we will overwrite memory which hasn't been allocated to us.

This cset no longer passes the board pointer into the init_fn, but instead
allows the init_fn to return the number of ports found, or zero to use the
default value in board->num_ports, or negative error number.

23 years ago[SERIAL] Rename uart_event() to uart_write_wakeup()
Russell King [Sat, 2 Nov 2002 10:59:49 +0000 (10:59 +0000)]
[SERIAL] Rename uart_event() to uart_write_wakeup()
uart_event() only has one purpose, which is to wake up any pending
writers via the line discipline.  Rename it to reflect its real
functionality, and drop EVT_WRITE_WAKEUP.

23 years ago[PATCH] Orlov block allocator for ext3
Theodore Y. Ts'o [Sat, 2 Nov 2002 04:57:43 +0000 (20:57 -0800)]
[PATCH] Orlov block allocator for ext3

Here's the ext3 version.

23 years ago[PATCH] Fixup Orlov block allocator for ext2
Theodore Y. Ts'o [Sat, 2 Nov 2002 04:57:39 +0000 (20:57 -0800)]
[PATCH] Fixup Orlov block allocator for ext2

I finally had time to look at the Orlov patches, and found a memory
leak; sbi->s_debts wasn't getting freed when the filesystem was
getting unmounted, or in the error path.

This patch also makes the following cleanups/changes:

1) Use sbi->s_debts instead of sbi->debts --- all other fields in
struct ext2_sb_info are prefixed by "s_", so this makes things
consistent.

2) Add support for a new inode flag, EXT2_TOPDIR_FL, which tells tells
the Orlov allocator to treat that directory as the top of
directory hierarchies, so that new subdirectories created in
that directory should be spread apart.  System administrators
should set this flag on directories like /usr/src, /usr/home, etc.

3) Add a mount-time flag, -o oldalloc, which forces the use of the old
inode (pre-Orlov) allocator.  This makes it easier to do
comparison benchmarks, and in case people want to use the old
algorithm.

23 years agoRemove last vestiges of ide-cd.c "struct packet_command". It's been
Linus Torvalds [Sat, 2 Nov 2002 04:47:01 +0000 (20:47 -0800)]
Remove last vestiges of ide-cd.c "struct packet_command". It's been
empty for a while now, and nothing uses it.

23 years agoAdd a "cmd_len" parameter to the request, so that device drivers
Linus Torvalds [Sat, 2 Nov 2002 03:55:44 +0000 (19:55 -0800)]
Add a "cmd_len" parameter to the request, so that device drivers
don't have to try to figure it out for themselves.

Make ide-cd.c use it.

23 years agoMerge redhat.com:/home/jgarzik/repo/linus-2.5
Jeff Garzik [Fri, 1 Nov 2002 20:55:13 +0000 (15:55 -0500)]
Merge redhat.com:/home/jgarzik/repo/linus-2.5
into redhat.com:/home/jgarzik/repo/minitramfs-2.5

23 years agoNeeds <linux/smp_lock.h> for "in_atomic()" test.
Linus Torvalds [Fri, 1 Nov 2002 12:51:14 +0000 (04:51 -0800)]
Needs <linux/smp_lock.h> for "in_atomic()" test.

23 years agoKill stupid bug in initramfs that prevented it from working.
Jeff Garzik [Fri, 1 Nov 2002 12:36:40 +0000 (07:36 -0500)]
Kill stupid bug in initramfs that prevented it from working.
(thanks to Al Viro for his patience, I owe him one)

23 years agoMerge bk://gkernel@bkbits.net/minitramfs-2.5
Jeff Garzik [Fri, 1 Nov 2002 10:26:41 +0000 (05:26 -0500)]
Merge bk://gkernel@bkbits.net/minitramfs-2.5
into redhat.com:/garz/repo/minitramfs-2.5

23 years agoMerge bk://bk.arm.linux.org.uk
Linus Torvalds [Fri, 1 Nov 2002 08:30:03 +0000 (00:30 -0800)]
Merge bk://bk.arm.linux.org.uk
into penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/linux

23 years ago[ARM] 2.5.45 updates
Russell King [Fri, 1 Nov 2002 23:59:33 +0000 (23:59 +0000)]
[ARM] 2.5.45 updates
- Convert to new configuration language and fix up HEX config symbols
- Fix build errors for !CONFIG_KALLSYMS case
- Add module list to oopsen.
- Add POLLREMOVE

23 years agoMinimal initramfs support (based on Al Viro's work).
Jeff Garzik [Fri, 1 Nov 2002 07:02:04 +0000 (02:02 -0500)]
Minimal initramfs support (based on Al Viro's work).

23 years ago[PATCH] Double x86 initialise fix.
Dave Jones [Fri, 1 Nov 2002 06:22:47 +0000 (22:22 -0800)]
[PATCH] Double x86 initialise fix.

For many moons, we've been executing identify_cpu()
on the boot processor twice on SMP kernels.
This is harmless, but has a few downsides..
- Extra cruft in bootlog/dmesg
- Spawns one too many timers for the mcheck handler
- possibly other wasteful things..

This seems to do the right thing here, and has been
acked (after some minor squibbles) on l-k.

23 years ago[PATCH] max_cpus overflow.
Dave Jones [Fri, 1 Nov 2002 06:22:36 +0000 (22:22 -0800)]
[PATCH] max_cpus overflow.

Without the below patch, my HT 2-way prints out
"CPUS Done 4294967295" on boot, which whilst amusing
is somewhat exaggerated.

23 years ago[PATCH] Clean up capabilities printing.
Dave Jones [Fri, 1 Nov 2002 06:22:26 +0000 (22:22 -0800)]
[PATCH] Clean up capabilities printing.

The machine check initialisation prints some blurb
which makes the capabilities dumping a little untidy.
By initialising it slightly later, we get something that
looks a lot better.

23 years ago[PATCH] another kmap imbalance in 2.4.x/2.5.x RPC
Trond Myklebust [Fri, 1 Nov 2002 06:14:29 +0000 (22:14 -0800)]
[PATCH] another kmap imbalance in 2.4.x/2.5.x RPC

  We've uncovered yet another kmap imbalance in the new RPC code.  This
looks like it might be the last one (my debugging printks have been
unable to unearth any more).  One line fix + 4 line comment appended.

23 years agoFix compile warning in slab.c
Linus Torvalds [Fri, 1 Nov 2002 05:34:44 +0000 (21:34 -0800)]
Fix compile warning in slab.c

23 years agoMerge http://linux-acpi.bkbits.net/linux-acpi
Linus Torvalds [Fri, 1 Nov 2002 05:03:37 +0000 (21:03 -0800)]
Merge http://linux-acpi.bkbits.net/linux-acpi
into penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/linux

23 years agoMerge bk://nevyn.them.org:5000
Linus Torvalds [Fri, 1 Nov 2002 04:56:34 +0000 (20:56 -0800)]
Merge bk://nevyn.them.org:5000
into penguin.transmeta.com:/home/penguin/torvalds/repositories/kernel/linux

23 years agoMake presense of old/style EH routines cause warnings, not
Linus Torvalds [Fri, 1 Nov 2002 04:42:50 +0000 (20:42 -0800)]
Make presense of old/style EH routines cause warnings, not
a compile failure.

23 years ago[PATCH] linux-2.5.45_notsc-warning_A0
John Stultz [Fri, 1 Nov 2002 04:06:50 +0000 (20:06 -0800)]
[PATCH] linux-2.5.45_notsc-warning_A0

This is a minor cleanup that removes two instances of CONFIG_X86_TSC
(lets the compiler optimize it out), and adds a warning message should
anyone pass "notsc" to a kernel compiled w/ CONFIG_X86_TSC (which
ignores it).

This is basically a forward port of a patch I got into 2.4 a while back.

23 years ago[PATCH] fixes for building kernel 2.5.45 using Intel compiler
Jun Nakajima [Fri, 1 Nov 2002 04:05:02 +0000 (20:05 -0800)]
[PATCH] fixes for building kernel 2.5.45 using Intel compiler

I like Alan's idea: IGNLABEL("HmacRxAccepted")

23 years ago[PATCH] driverfs topology cleanup
Andrew Morton [Fri, 1 Nov 2002 04:02:41 +0000 (20:02 -0800)]
[PATCH] driverfs topology cleanup

From Matthew Dobson.

This final patch from Matthew cleans up a few leftovers which were noted
after the code had been reviewed and tested a bit in the -mm patchsets.

1) Update register_XXX and arch_register_XXX functions to return int
   instead of void.  Functions calling these functions should know if
   they completed successfully to take appropriate further registration
   action, or not bother.

2) Drop some pointless error checking in the arch_register_XXX
   functions.

23 years ago[PATCH] create node_online_map
Andrew Morton [Fri, 1 Nov 2002 04:02:31 +0000 (20:02 -0800)]
[PATCH] create node_online_map

From Matthew Dobson.

Create and use node_online_map.

This patch creates a node_online_map, much like cpu_online_map.  It
also creates the standard helper functions, ie: node_online(),
num_online_nodes(), node_set_online(), node_set_offline().

This is used by driverFS topology to keep track of which Nodes
are in the system and online.

23 years ago[PATCH] create memblk_online_map
Andrew Morton [Fri, 1 Nov 2002 04:02:20 +0000 (20:02 -0800)]
[PATCH] create memblk_online_map

From Matthew Dobson.

Create and use memblk_online_map.

This patch creates a memblk_online_map, much like cpu_online_map.  It
also creates the standard helper functions, ie: memblk_online(),
num_online_memblks(), memblk_set_online(), memblk_set_offline().

This is used by driverFS topology to keep track of which memory blocks
are in the system and online.

23 years ago[PATCH] NUMA meminfo for driverfs Topology
Andrew Morton [Fri, 1 Nov 2002 04:02:08 +0000 (20:02 -0800)]
[PATCH] NUMA meminfo for driverfs Topology

From Matthew Dobson.

Create nodeX/meminfo files for DriverFS Topology.

This patch adds code to DriverFS Topology to expose per-node memory
statistics.  This information is exposed via: cat nodeX/meminfo

The patch also adds 2 helper functions to gather per-node memory info.

23 years ago[PATCH] i386 driverfs Topology
Andrew Morton [Fri, 1 Nov 2002 04:01:58 +0000 (20:01 -0800)]
[PATCH] i386 driverfs Topology

From Matthew Dobson.

Update/Create i386 specific files for DriverFS Topology.

This patch creates the i386 specific files/functions/structures to
implement driverfs Topology.  These structures have the generic
CPU/MemBlk/Node structures embedded in them.

This patch also creates the arch-specific initialization routine to
instantiate the topology.

23 years ago[PATCH] Update/Create core files for DriverFS Topology.
Andrew Morton [Fri, 1 Nov 2002 04:01:48 +0000 (20:01 -0800)]
[PATCH] Update/Create core files for DriverFS Topology.

From Matthew Dobson.

Update/Create core files for DriverFS Topology.

This patch creates the generic structures that are (will be) embedded in
the per-arch structures.  Also creates calls to register these generic
structures (CPUs, MemBlks, & Nodes).

Note that without arch-specific structures in which to embed these
structures, and an arch-specific initialization routine, these
functions/structures remain unused.

23 years ago[PATCH] page accounting atomicity fix
Andrew Morton [Fri, 1 Nov 2002 04:01:38 +0000 (20:01 -0800)]
[PATCH] page accounting atomicity fix

The global page accounting functions are currently using "+=" against a
ulong.  But this can happen at interrupt time as well, and "+=" is not
atomic against interrupt-time modification of the same word.

Change it to use local_irq_save()

23 years ago[PATCH] ext3 build fix
Andrew Morton [Fri, 1 Nov 2002 04:01:27 +0000 (20:01 -0800)]
[PATCH] ext3 build fix

Patch from Andreas Dilger <adilger@clusterfs.com>

If you have ext2_fs.h and ext3_fs.h included at the same time, we are
missing one define used in ext3_should_journal_data() and
ext3_should_order_data() as part of the test_opt() macro (which
defaults to the ext2 definition if both are present).

23 years ago[PATCH] disable PF_MEMALLOC for interrupt allocations
Andrew Morton [Fri, 1 Nov 2002 04:01:17 +0000 (20:01 -0800)]
[PATCH] disable PF_MEMALLOC for interrupt allocations

If a task is running in state PF_MEMALLOC and is interrupted, the page
allocator will treat allocations by the interrupt handler as
PF_MEMALLOC as well.

Detect that, and stop it.  Also tidies things up a little in there.

23 years ago[PATCH] improved space efficiency in dcache
Andrew Morton [Fri, 1 Nov 2002 04:01:08 +0000 (20:01 -0800)]
[PATCH] improved space efficiency in dcache

Currently we are storing filenames which are 16-chars or less
inside struct dentry itself and then separately allocating
larger names.

But this leaves spare space in the dentry - the dentry slab cache
is using cacheline alignment.   In my build, struct dentry is 112
bytes so there are at least an additional 16 bytes in there.

And the number of files which have names in the 16-32 char range
will be significant.

So Manfred's patch changes the dcache code to utilise _all_ the space
between the last member of the dentry and the start of the next cacheline.

23 years ago[PATCH] faster wakeups in the pipe code
Andrew Morton [Fri, 1 Nov 2002 04:00:56 +0000 (20:00 -0800)]
[PATCH] faster wakeups in the pipe code

Use prepare_to-wait/finish_wait to avoid bouncing the waitqueue
spinlock between the waker and the wakee all the time.

23 years ago[PATCH] use bd_claim in the raw driver
Andrew Morton [Fri, 1 Nov 2002 04:00:46 +0000 (20:00 -0800)]
[PATCH] use bd_claim in the raw driver

Currently if you use the raw driver against /dev/hda1 while it has an
fs mounted, everything gets confused and the kernel locks up or oopses.

Al said to use bd_claim().  This prevents the raw driver from being
able to open the device under these circumstances.

There is value in being able to read the raw device while there's a
filesystem mounted: to poke around at the disk without interacting with
the kernel's caching.  You can still do that, with an O_DIRECT open of
/dev/hda1.

23 years ago[PATCH] direct-io build fix
Andrew Morton [Fri, 1 Nov 2002 04:00:37 +0000 (20:00 -0800)]
[PATCH] direct-io build fix

The arg to ZERO_PAGE() doesn't get evaluated on ia32 :(

23 years agoACPI: Interpreter update to (20021101)
Andy Grover [Fri, 1 Nov 2002 03:46:01 +0000 (19:46 -0800)]
ACPI: Interpreter update to (20021101)
- Fix namespace ordering, which was causing breakage
- Enhance some error messages

23 years ago[PATCH] 2.5.45 UCLinux merge M680x0 mmuless arch and include/asm
Alan Cox [Fri, 1 Nov 2002 03:44:49 +0000 (19:44 -0800)]
[PATCH] 2.5.45 UCLinux merge M680x0 mmuless arch and include/asm

This is the M68K mmuless port code maintained by Greg Ungerer and used
by Snapgear for all sorts of bits of hardware. No shared files included
in this chunk of patches. This matches 2.5.45-uc1 which is the stuff that
has been in my tree for a bit combined with Kconfig changes

23 years ago[PATCH] UCLINUX (forgot one) - header file for binfmt_flat
Alan Cox [Fri, 1 Nov 2002 03:44:22 +0000 (19:44 -0800)]
[PATCH] UCLINUX (forgot one) - header file for binfmt_flat

23 years ago[PATCH] UCLINUX "flat" binary loader
Alan Cox [Fri, 1 Nov 2002 03:44:12 +0000 (19:44 -0800)]
[PATCH] UCLINUX "flat" binary loader

This is much like the old m68k amiga/st loader world. Its a relocating
loader with optional compression. Shared libraries and A5 sharing stuff
is all handled by the userspace/libs

23 years ago[PATCH] add machine ident for the V850 from NEC
Alan Cox [Fri, 1 Nov 2002 03:44:02 +0000 (19:44 -0800)]
[PATCH] add machine ident for the V850 from NEC

23 years ago[PATCH] arch specific files for MMUless NEC v850 port
Alan Cox [Fri, 1 Nov 2002 03:43:49 +0000 (19:43 -0800)]
[PATCH] arch specific files for MMUless NEC v850 port

This is the V850 support maintained by Miles Bader at NEC

Again this is just arch and include/asm- files. It matches 2.5.45-uc1 + v850
fixes which is close to what was in my old tree (some clean up done since then)
plus Kconfig

23 years ago[PATCH] UCLINUX ethernet driver for Coldfire onboard 100Mbit ethernet
Alan Cox [Fri, 1 Nov 2002 03:43:32 +0000 (19:43 -0800)]
[PATCH] UCLINUX ethernet driver for Coldfire onboard 100Mbit ethernet

23 years ago[PATCH] Kconfig and makfile goo for new net drivers
Alan Cox [Fri, 1 Nov 2002 03:43:19 +0000 (19:43 -0800)]
[PATCH] Kconfig and makfile goo for new net drivers

23 years ago[PATCH] UCLINUX ethernet driver for 68360 on board ethernet
Alan Cox [Fri, 1 Nov 2002 03:43:09 +0000 (19:43 -0800)]
[PATCH] UCLINUX ethernet driver for 68360 on board ethernet

23 years ago[PATCH] UCLinux generic memory mapped MTD driver
Alan Cox [Fri, 1 Nov 2002 03:42:58 +0000 (19:42 -0800)]
[PATCH] UCLinux generic memory mapped MTD driver

23 years ago[PATCH] MAINTAINERS entries for UcLinux platforms
Alan Cox [Fri, 1 Nov 2002 03:42:46 +0000 (19:42 -0800)]
[PATCH] MAINTAINERS entries for UcLinux platforms

23 years ago[PATCH] fix i810 printk error
Alan Cox [Fri, 1 Nov 2002 03:42:36 +0000 (19:42 -0800)]
[PATCH] fix i810 printk error

23 years ago[PATCH] some trident needs longer delays to power up codecs
Alan Cox [Fri, 1 Nov 2002 03:42:26 +0000 (19:42 -0800)]
[PATCH] some trident needs longer delays to power up codecs

23 years ago[PATCH] make rxrpc build with gcc 2.x
Alan Cox [Fri, 1 Nov 2002 03:42:16 +0000 (19:42 -0800)]
[PATCH] make rxrpc build with gcc 2.x

23 years ago[PATCH] update MAINTAINERS
Alan Cox [Fri, 1 Nov 2002 03:42:06 +0000 (19:42 -0800)]
[PATCH] update MAINTAINERS

Update DIGI entries
Update IRDA entries
Add Voyager
Put Device mapper under D

23 years ago[PATCH] header update to match m68k pci change
Alan Cox [Fri, 1 Nov 2002 03:41:56 +0000 (19:41 -0800)]
[PATCH] header update to match m68k pci change

23 years ago[PATCH] add includes for voyager interrupt chip
Alan Cox [Fri, 1 Nov 2002 03:41:46 +0000 (19:41 -0800)]
[PATCH] add includes for voyager interrupt chip

23 years ago[PATCH] make afs build with gcc 2.9x
Alan Cox [Fri, 1 Nov 2002 03:41:35 +0000 (19:41 -0800)]
[PATCH] make afs build with gcc 2.9x

23 years ago[PATCH] update the u14-34f driver to maintainer updates
Alan Cox [Fri, 1 Nov 2002 03:41:24 +0000 (19:41 -0800)]
[PATCH] update the u14-34f driver to maintainer updates

(With bios changes merged)

23 years ago[PATCH] ditto for sym53c8xx
Alan Cox [Fri, 1 Nov 2002 03:41:13 +0000 (19:41 -0800)]
[PATCH] ditto for sym53c8xx

23 years ago[PATCH] ditto error handling for qla1280
Alan Cox [Fri, 1 Nov 2002 03:41:02 +0000 (19:41 -0800)]
[PATCH] ditto error handling for qla1280

23 years ago[PATCH] restore missing error handlers for ncr53c8xx
Alan Cox [Fri, 1 Nov 2002 03:40:52 +0000 (19:40 -0800)]
[PATCH] restore missing error handlers for ncr53c8xx

Again so we dont forget to fix this

23 years ago[PATCH] restore inia100.h abort/reset
Alan Cox [Fri, 1 Nov 2002 03:40:42 +0000 (19:40 -0800)]
[PATCH] restore inia100.h abort/reset

So we dont forget to actually fix the lack of error handling

23 years ago[PATCH] eata update from maintainer
Alan Cox [Fri, 1 Nov 2002 03:40:32 +0000 (19:40 -0800)]
[PATCH] eata update from maintainer

23 years ago[PATCH] remove more tqueue.h
Alan Cox [Fri, 1 Nov 2002 03:40:22 +0000 (19:40 -0800)]
[PATCH] remove more tqueue.h

23 years ago[PATCH] fix filters types on winbond 840
Alan Cox [Fri, 1 Nov 2002 03:40:11 +0000 (19:40 -0800)]
[PATCH] fix filters types on winbond 840

23 years ago[PATCH] move de4x5 to new pci api
Alan Cox [Fri, 1 Nov 2002 03:40:01 +0000 (19:40 -0800)]
[PATCH] move de4x5 to new pci api

23 years ago[PATCH] without this tulip doesn't build non modular
Alan Cox [Fri, 1 Nov 2002 03:39:51 +0000 (19:39 -0800)]
[PATCH] without this tulip doesn't build non modular

23 years ago[PATCH] kill stupid search and destroy error
Alan Cox [Fri, 1 Nov 2002 03:39:40 +0000 (19:39 -0800)]
[PATCH] kill stupid search and destroy error

23 years ago[PATCH] warning fixes
Alan Cox [Fri, 1 Nov 2002 03:39:30 +0000 (19:39 -0800)]
[PATCH] warning fixes

These are from inl going back to 32bit

23 years ago[PATCH] use longer delays on 3c509
Alan Cox [Fri, 1 Nov 2002 03:39:19 +0000 (19:39 -0800)]
[PATCH] use longer delays on 3c509

Som boards dont seem to be in spec

23 years ago[PATCH] add voyager specific extra key map
Alan Cox [Fri, 1 Nov 2002 03:39:08 +0000 (19:39 -0800)]
[PATCH] add voyager specific extra key map

23 years ago[PATCH] fix warnings
Alan Cox [Fri, 1 Nov 2002 03:38:57 +0000 (19:38 -0800)]
[PATCH] fix warnings

23 years ago[PATCH] device mapper 64/32bit ioctl maps for S/390x
Alan Cox [Fri, 1 Nov 2002 03:38:47 +0000 (19:38 -0800)]
[PATCH] device mapper 64/32bit ioctl maps for S/390x

23 years ago[PATCH] device mapper ioctls for ppc64
Alan Cox [Fri, 1 Nov 2002 03:38:37 +0000 (19:38 -0800)]
[PATCH] device mapper ioctls for ppc64

23 years ago[PATCH] kill pcibios in ppc32
Alan Cox [Fri, 1 Nov 2002 03:38:27 +0000 (19:38 -0800)]
[PATCH] kill pcibios in ppc32

23 years ago[PATCH] 32/64bit device mapper ioctl maps for mips64
Alan Cox [Fri, 1 Nov 2002 03:38:17 +0000 (19:38 -0800)]
[PATCH] 32/64bit device mapper ioctl maps for mips64

23 years ago[PATCH] kill pcibios in m68k pci code
Alan Cox [Fri, 1 Nov 2002 03:38:05 +0000 (19:38 -0800)]
[PATCH] kill pcibios in m68k pci code

23 years ago[PATCH] kill pcibios in m6k8
Alan Cox [Fri, 1 Nov 2002 03:37:53 +0000 (19:37 -0800)]
[PATCH] kill pcibios in m6k8

23 years ago[PATCH] kill debug printk we dont need now
Alan Cox [Fri, 1 Nov 2002 03:37:43 +0000 (19:37 -0800)]
[PATCH] kill debug printk we dont need now

23 years ago[PATCH] Alpha escaped pcibios...
Alan Cox [Fri, 1 Nov 2002 03:37:31 +0000 (19:37 -0800)]
[PATCH] Alpha escaped pcibios...

23 years ago[PATCH] 3ware driver update for 2.5.46, sync cache, 64-bit, etc.
Adam Radford [Fri, 1 Nov 2002 03:37:21 +0000 (19:37 -0800)]
[PATCH] 3ware driver update for 2.5.46, sync cache, 64-bit, etc.

23 years ago[PATCH] death of ->rq_dev
Alexander Viro [Fri, 1 Nov 2002 03:26:43 +0000 (19:26 -0800)]
[PATCH] death of ->rq_dev

RIP.  It's not used anymore, so we kill assignments to it and the
field itself.  That was the last serious use of kdev_t in block drivers.

23 years ago[PATCH] remaining uses of ->rq_dev
Alexander Viro [Fri, 1 Nov 2002 03:26:31 +0000 (19:26 -0800)]
[PATCH] remaining uses of ->rq_dev

* in floppy drivers and ps2esdi - replaced with use of ->rq_disk;
floppy drivers already have information about "type" part of device number
- they had stored that information in floppy_open(), so all we need is a
"which drive" part.  And that is available from ->rq_disk.
* DAC960: changed ->private_data initialization - instead of pointing
it to controller (which we already have as disk->queue.queuedata) we simply
store a disk number in there.  That had simplified ->revalidate_disk() and
allowed to kill the use of ->rq_dev in request handler.

23 years ago[PATCH] uses of ->rq_dev in printks
Alexander Viro [Fri, 1 Nov 2002 03:26:20 +0000 (19:26 -0800)]
[PATCH] uses of ->rq_dev in printks

* killed the uses of ->rq_dev in printks - several drivers.

23 years ago[PATCH] generic uses of ->rq_dev
Alexander Viro [Fri, 1 Nov 2002 03:26:08 +0000 (19:26 -0800)]
[PATCH] generic uses of ->rq_dev

* uses in ll_rw_blk.c and elevator.c eliminated - ->rq_disk has
everything we need.