More merges all over the map - ppc, scsi, USB, kbuild, input drivers etc.
And both Al and Andrew have been busy again.
This also introduces the support for non-100 Hz internal kernel times on
x86, while still exporting the old interface to user space (ie anybody who
exported raw jiffies before should be exporting "clock_t", which on x86
continues to be a 100 Hz clock, regardless of whatever the internal kernel
frequency is).
Right now the x86 timer frequency is set to 1kHz, but that's just another
random number. It could be a config option if people really care, but I'd
rather just have people argue for or against specific internal frequencies
and we'll find something most people are happy with. It's easy to change
without user space even noticing now.
The other thing that we should sort out eventually is the unified naming
for disk devices, now that both IDE and SCSI are starting to have some
support for driverfs. Let's make sure that we _can_ have sane ways of
accessing a disk, without having to care whether it is IDE or SCSI or
anything else.
Linus
----
Summary of changes from v2.5.24 to v2.5.25
============================================
<[email protected]>:
o APM compile fix, "stime" update broke it
<[email protected]>:
o PPC32: Fixes and cleanups for PPC40x processors
o PPC32: Update handling of the interrupt controller on the PPC405
o PPC32: more PPC40x cleanup (remove CONFIG_PIN_TLB, add comments)
<[email protected]>:
o Initial initio a100 driver DMA mapping changes + selected cleanups
o inia100.c
o i60uscsi.c
<jejb@mulgrave.(none)>:
o [SCSI 53c700] add specimen block layer tcq
o make Scsi_Cmnd and Scsi_Request.request be a pointer to the block
layer request instead of a copy.
o Initial support (mid-layer) for generic TCQ
o [SCSI mid-layer]
o [SCSI mid-layer] bug fix two missing blk_queue_end_tag()s
o export open_softirq
<[email protected]>:
o kbuild: Fix warnings and other buglets
<[email protected]>:
o Fix note sections in ELF core dumps
<[email protected]>:
o USB: bluetty.c allocation bug fix
<[email protected]>:
o pegasus & rtl8150
<[email protected]>:
o kbuild: Add "make help" support
<[email protected]>:
o linux-2.5.22-driverfs.patch
<[email protected]>:
o softscsi patch
o rewrite find_vma_prev
Alexander Viro <[email protected]>:
o devpts cleanup
o (md.c) block device size cleanups
o cdrom.c cleanups
o kdev_t crapectomy
o raid kdev_t cleanups (part 1)
o raid ->diskop() splitup
o raid kdev_t cleanups - part 2
o md_import_device() cleanup
o raid kdev_t cleanups - part 3
o ex_dev switched to dev_t
o assorted kdev_t cleanups in filesystems
o >i_dev switched to dev_t
Andrew Morton <[email protected]>:
o handle BIO allocation failures in swap_writepage()
o Fix 3c59x driver for some 3c566B's
o per-cpu buffer_head cache
o Remove ext2's buffer_head cache
o Remove ext3's buffer_head cache
o debug check for leaked blockdev buffers
o misc cleanups and fixes
o pdflush cleanup
o remove swap_get_block()
o shmem fixes
o add new list_splice_init()
o set TASK_RUNNING in cond_resched()
o set TASK_RUNNING in yield()
o check for O_DIRECT capability in open(), not write()
o set_page_dirty() in mark_dirty_kiobuf()
o resurrect __GFP_HIGH
o Use __GFP_HIGH in mpage_writepages()
o always update page->flags atomically
o suppress more allocation failure warnings
o fix a writeback race
o combine generic_writepages() and mpage_writepages()
o ext3 truncate fix
o JBD commit callback capability
o fix invalidate_inode_pages2() race
o debug: check page refcount in __free_pages_ok()
o reduce lock contention in try_to_free_buffers()
o Use names, not numbers for pagefault types
Anton Altaparmakov <[email protected]>:
o NTFS: 2.0.11 - Initial preparations for fake inode based attribute
i/o
o NTFS: 2.0.12 - Initial cleanup of address space operations
following 2.0.11 changes
o NTFS: 2.0.13 - Use iget5_locked() in preparation for fake inodes
and small cleanups
o NTFS: 2.0.14 - Run list merging code cleanup, minor locking
changes, typo fixes
David Brownell <[email protected]>:
o ohci-hcd cardbus unplug
o Re: [linux-usb-devel] unending timeouts (patch for 2.5.22 oops)
Greg Kroah-Hartman <[email protected]>:
o USB: picked a uhci driver to go forward with
o USB: removed unused Config.help entries from the host controller
file
o USB: removed file ops from usb device structure Moved the file ops
and minor number stuff out of the usb structure, Now
usb_register_dev() and usb_deregister_dev() must be called if you
want to use the USB major number.
o USB: added drivers/usb/core/file.c to the kernel-api documentation
o USB: Fixups due to the changes in struct usb_device for
file_operations and minor number handling
[email protected] <[email protected]>:
o fix SCSI driverfs for IDE panic on boot
Kai Germaschewski <[email protected]>:
o kbuild: Assorted cleanups
o kbuild: Prepare LDFLAGS for general use
o kbuild: Rename ld flags for vmlinux to LDFLAGS_vmlinux
o kbuild: Put flags for ld into LDFLAGS
o kbuild: Put flags for objcopy into OBJCOPYFLAGS
o kbuild: clean up arch/i386/boot, part 1
o kbuild: clean up arch/i386/boot, part 2
o kbuild: clean up arch/i386/boot, part 3
o kbuild: clean up arch/i386/boot, part 4
o kbuild: Fix calling of make_times_h and gentbl
o kbuild: Provide shipped versions of the keymap files
Linus Torvalds <[email protected]>:
o DRI CVS merge: separate out driver-ioctl's into driver headers
o Make in-kernel HZ be 1000 on x86, retaining user-level 100 HZ
clock_t
o Make ramfs/driverfs maintain directory nlink counts
o Fix more places where we exported our internal time to user space.
Convert to standard clock_t.
o Radeon DRI merge
o x86 "make clean" missed some new targets
o Disable ReiserFS bh usage count testing for now
Matthew Dharm <[email protected]>:
o usb-storage: code cleanup, small fixes
o usb-storage: Code consolidation of error paths
o usb-storage: remove timer
Paul Mackerras <[email protected]>:
o PPC32: add CONFIG_DEBUG_SPINLOCK
o PPC32: Fixes for bugs in exception handling in copy_to_user and
clear_user
o PPC32: Add struct page * argument to copy/clear_user_page
o PPC32: fixes for I/O mappings on CHRP machines
o PPC32: update for scheduler changes (switch_to,
prepare/finish_arch_*)
o PPC32: translate addresses in the Open Firmware device tree
correctly
o PPC32 compile fix: add missing parenthesis
o PPC32: fix compilation with current binutils
o PPC32: work around the fact that the PPC601 doesn't implement the
MSR.RI bit
o PPC32: fix some minor compile warnings
o PPC32: fix handling of machine checks
o PPC32: eliminate some compile warnings on the EP405 board
o PPC32: define USER_HZ to be 100 (HZ is still 100 for now)
o PPC32: fix compile error by removing extraneous declarations
(ppc_ksyms.c)
Paul Menage <[email protected]>:
o Shift BKL into ->statfs()
Pavel Machek <[email protected]>:
o suspend-to-disk documentation updates
Petr Vandrovec <[email protected]>:
o 2.5.24 matroxfb memory corruption
o 2.5.24 matroxfb off by one error
o drivers/ide/probe.c leaks memory
[email protected] <[email protected]>:
o sound/oss/maestro.c
Tom Rini <[email protected]>:
o PPC32: Fix building of most of the zImage targets
o PPC32: Fix the rule for using config files from arch/ppc/configs/
To accommodate the change to scripts/Configure which made it look
in /boot/config-`uname -r` before looking at
arch/$(ARCH)/defconfig, change the rule for using a config from
arch/ppc/configs to copy to .config instead of
arch/$(ARCH)/defconfig
o PPC32: Minor PReP OpenPIC fixes from Troy Benjegerdes and Leigh
Brown
o Allow the data cache to be turned off on MPC8260 systems
o PPC32: Move per-board MPC8xx ethernet defines to their respective
board header
o PPC32: Fix dependancies in arch/ppc/boot
o PPC32: Minor fixes with respect to the MPC7450 and MPC7455
Tom Rini <[email protected]>:
o PPC32: Update the OpenPIC code to only require a 'linux_irq_offset'
param
o PPC32: check the binutils version and make sure it is new enough
o PPC32: use the proper instruction mnemonics for altivec
instructions
Vojtech Pavlik <[email protected]>:
o Update the gameport drivers to Dave Jones's tree
o Update the iforce driver to the latest revision (it now lives in a
separate directory), add twiddler and guillemot drivers.
o Add new serio modules for PS/2 AUX/KBD
o Add keyboard, mouse and touchscreen drivers
o Update the input handler modules to latest versions
o Makefile/config.in changes to reflect the new drivers
o Fix button assignments for Saturn and PSX pads
o Handle input-only keyboard interfaces
o Handle slowly responding PS/2 mice
o Use time_after() where applicable
o Minor cleanup in evdev.c
o Minor fixes to make the whole thing compile on latest 2.5 and
kbuild2
o Add vortex anf fm801 gameport drivers, remove obsolete pcigame
driver
o Fix psmouse.c - it needs tqueue.h
2.5.25: 2 problems found
1) Options under Input Device Support and under Character Devices: Mice
are duplicates from a user's point of view. Fortunately they have different
names (CONFIG_PSMOUSE vs CONFIG_MOUSE_PS2).
Unfortunately they both seem to be used...did a recursive grep.
2) This is a compilation error, which has been in 2.5 since 3 kernels ago or
so. Reported twice on LKML by me, as well as another person. No reply.
mpparse.c: In function `mp_parse_prt':
mpparse.c:1080: warning: implicit declaration of function `mp_find_ioapic'
mpparse.c:1083: `mp_ioapic_routing' undeclared (first use in this function)
mpparse.c:1083: (Each undeclared identifier is reported only once
mpparse.c:1083: for each function it appears in.)
mpparse.c:1107: warning: implicit declaration of function
`io_apic_set_pci_routing'
make[1]: *** [mpparse.o] Error 1
make[1]: Leaving directory `/usr/src/linux-2.5.24/arch/i386/kernel'
make: *** [arch/i386/kernel] Error 2
occurs with:
LOCAL_APIC on, IOAPIC off, ACPI on, uniprocessor machine.
Linus Torvalds writes:
> Right now the x86 timer frequency is set to 1kHz, but that's just another
> random number. It could be a config option if people really care, but I'd
> rather just have people argue for or against specific internal frequencies
> and we'll find something most people are happy with. It's easy to change
> without user space even noticing now.
You could go by the CPU choice that already exists:
386 --> 100
486 --> 200
Pentium --> 400
PPro --> 800
P-III,P4 --> 1600
On Sat, 6 Jul 2002 09:54, Linus Torvalds wrote:
> More merges all over the map - ppc, scsi, USB, kbuild, input drivers etc.
Found a few typos in the input changes. Fixup patch attached.
Brad
--
http://conf.linux.org.au. 22-25Jan2003. Perth, Australia. Birds in Black.
On Fri, 05 Jul 2002, Linus Torvalds wrote:
> More merges all over the map - ppc, scsi, USB, kbuild, input drivers etc.
Did the LVM guys (are you listening?) tell anything if they were about
to go fix the current 2.5 LVM breakage? Or does EVMS work on 2.5 instead?
> 2.5.25: 2 problems found
>
> 1) Options under Input Device Support and under Character Devices: Mice
> are duplicates from a user's point of view. Fortunately they have different
> names (CONFIG_PSMOUSE vs CONFIG_MOUSE_PS2).
> Unfortunately they both seem to be used...did a recursive grep.
The input api PS/2 mouse driver is meant to replace the old driver which
is in pc_keyb.c. Unfortunely most keyboard drivers are based around
pc_keyb.c so it has to stay for now. Plus the console system is designed
around pc_keyb.c. This will also change as the console system is migrated
toward the input api. So to any keyboard maintainers please port your
drivers over to the input api before they break when the console system
breaks.
You patch is right except for drivers/input/gameport/Config.help. That was
right before your patch.
. ---
|o_o |
|:_/ | Give Micro$oft the Bird!!!!
// \ \ Use Linux!!!!
(| | )
/'\_ _/`\
\___)=(___/
On Sat, 6 Jul 2002, Brad Hards wrote:
> On Sat, 6 Jul 2002 09:54, Linus Torvalds wrote:
> > More merges all over the map - ppc, scsi, USB, kbuild, input drivers etc.
> Found a few typos in the input changes. Fixup patch attached.
>
> Brad
>
> --
> http://conf.linux.org.au. 22-25Jan2003. Perth, Australia. Birds in Black.
On Sun, 7 Jul 2002 03:48, James Simmons wrote:
> You patch is right except for drivers/input/gameport/Config.help. That was
> right before your patch.
OK. I assumed that Config.in was the authoritative source, and Config.help
should match. Looks like (based on the Makefile) that I probably should have
changed it to be CONFIG_GAMEPORT_EMU10K1 in the Config.in. In fact, that
driver won't ever be built with vanilla 2.5.25, with or without the previous
patch
BK already has the previous change applied, so this in an incremental diff:
<bk text>
EMU10K1 build fix
Revert the CONFIG_INPUT_EMU10K1 change to drivers/input/gameport/Config.help,
and change drivers/input/gameport/Config.in to use CONFIG_GAMEPORT_EMU10K1
</bk text>
Brad
--
http://conf.linux.org.au. 22-25Jan2003. Perth, Australia. Birds in Black.
On Fri 05 Jul at 16:54:20 -0700 [email protected] done said:
>
> The other thing that we should sort out eventually is the unified naming
> for disk devices, now that both IDE and SCSI are starting to have some
> support for driverfs. Let's make sure that we _can_ have sane ways of
> accessing a disk, without having to care whether it is IDE or SCSI or
> anything else.
...and a way that scales better when naming large numbers of disks than
disk_name()'s stuff like:
sprintf(buf, "sd%c%c", 'a' + unit / 26, 'a' + unit % 26);
t.
On Sat, Jul 06, 2002 at 03:54:12PM +0200, Matthias Andree wrote:
> On Fri, 05 Jul 2002, Linus Torvalds wrote:
>
> > More merges all over the map - ppc, scsi, USB, kbuild, input drivers etc.
>
> Did the LVM guys (are you listening?) tell anything if they were about
> to go fix the current 2.5 LVM breakage? Or does EVMS work on 2.5 instead?
I'll say this yet again:
Heinz Mauelshagen is maintaining LVM1.0.x on 2.4 kernels. This is for
bug fixes only, no new features will be added.
Alasdair Kergon, Patrick Caulfield and myself are working on the more
generic device-mapper driver for both 2.4/2.5. Initially we have
concentrated on 2.4, this driver is now very stable IMO (I would
certainly trust my data to it in preference to LVM1).
I will post a URL to the 2.5 patch at some point this week.
There is no intention to maintain the broken design that is LVM1 in
the 2.5 series - we do not have the spare resources to waste.
- Joe
On Mon, 8 Jul 2002, Joe Thornber wrote:
> On Sat, Jul 06, 2002 at 03:54:12PM +0200, Matthias Andree wrote:
> > On Fri, 05 Jul 2002, Linus Torvalds wrote:
> >
> > > More merges all over the map - ppc, scsi, USB, kbuild, input drivers etc.
> >
> > Did the LVM guys (are you listening?) tell anything if they were about
> > to go fix the current 2.5 LVM breakage? Or does EVMS work on 2.5 instead?
>
> I'll say this yet again:
>
> Heinz Mauelshagen is maintaining LVM1.0.x on 2.4 kernels. This is for
> bug fixes only, no new features will be added.
>
> Alasdair Kergon, Patrick Caulfield and myself are working on the more
> generic device-mapper driver for both 2.4/2.5. Initially we have
> concentrated on 2.4, this driver is now very stable IMO (I would
> certainly trust my data to it in preference to LVM1).
>
> I will post a URL to the 2.5 patch at some point this week.
>
> There is no intention to maintain the broken design that is LVM1 in
> the 2.5 series - we do not have the spare resources to waste.
All right. So how about removing it from the tree? It's broken; it
won't be fixed; it's abandoned by maintainers (and $DEITY witness,
there is a lot of very good reasons for that); there's nobody who
would be able and willing to pick it. What's the point of keeping
the damn thing in the tree?
Could you (or Heinz) submit the patch removing it from 2.5?
On Sun, Jul 07, 2002 at 12:22:48PM +1000, Brad Hards wrote:
> On Sun, 7 Jul 2002 03:48, James Simmons wrote:
> > You patch is right except for drivers/input/gameport/Config.help. That was
> > right before your patch.
> OK. I assumed that Config.in was the authoritative source, and Config.help
> should match. Looks like (based on the Makefile) that I probably should have
> changed it to be CONFIG_GAMEPORT_EMU10K1 in the Config.in. In fact, that
> driver won't ever be built with vanilla 2.5.25, with or without the previous
> patch
>
> BK already has the previous change applied, so this in an incremental diff:
>
> <bk text>
> EMU10K1 build fix
>
> Revert the CONFIG_INPUT_EMU10K1 change to drivers/input/gameport/Config.help,
> and change drivers/input/gameport/Config.in to use CONFIG_GAMEPORT_EMU10K1
> </bk text>
Thanks, applied.
--
Vojtech Pavlik
SuSE Labs