22 Oct, 2005
40 commits
-
fixup.S needs to specify .level and use correct LDREG macro.
New binutils has a bug where it doesn't "promote" from PA1.0 to PA1.1
correctly when using ",s" completer.remove use of __LP64__ in assembly.h and add some white space.
Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
Specify sr4 when flushing kernel space (we could equally well use sr5-7,
but must not use sr0).Signed-off-by: Matthew Wilcox
Signed-off-by: Kyle McMartin
-
replace use of "0" with "%r0" since PA 1.1 I/D flush ops only take a
general register and not an immediate value for the index field.
This just forces the code to always be PA 1.1 "clean".From: Joel Soete
Signed-off-by: Grant GrundlerSigned-off-by: Kyle McMartin
-
2.6.12-rc2-pa3 fix copy_user_page_asm to NOT access past end of page.
My bad. /o\
Lamont confirmed that instructions following a conditional
branch are *alway* executed regardless if the branch is taken or not.
Unless they are nullified (which was missing in this case).He also noted:
Conditional branches nullify on forward taken branch, and on
non-taken backward branch. Note that .+4 is a backwards branch.This makes alot more sense than the giberish in the PA20 arch book.
Compiles and boots on both 64-bit (a500) and 32-bit (j6k).
Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
add netpoll support
Patch by Sven SchnelleSigned-off-by: Sven Schnelle
Signed-off-by: Helge DellerSigned-off-by: Kyle McMartin
-
remove the spurious do_softirq calls from entry.S
With these in we were calling do_softirq twice; plus the calls in
entry.S took no account of nesting.Signed-off-by: James Bottomley
Signed-off-by: Kyle McMartin
-
drivers/infiniband depends on definition of pgprot_noncached() macro.
Someone else will have to fix it's wrong.Signed-off-by: Grant Grundler
-
Add IDs for some other STI graphics cards found on HP PA-RISC
machines.Signed-off-by: Kyle McMartin
-
add ECANCELED - SuSv3 wants one L. IB/SDP actually returns this error.
Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
Fix the alloc_slabmgmt panic
Hopefully this should also fix a lot of other intermittent kernel bugs.
The problem has been around since 2.6.9-rc2-pa6 when we allowed
floating point registers to be used in kernel code. The essence of
the problem is that gcc prefers to use floating point for integer
divides and multiples. Further, it can rely on the values in the no
clobber fp regs being correct across a function call. Unfortunately,
our task switch function only saves the integer no clobber registers,
not the fp ones, so if gcc makes a function call to any function in
the kernel which could sleep, the values it is relying on in any no
clobber floating point register may be lost. In the case of
alloc_slabmgmt, the value of the page offset is being stored in %fr12
across a call to kmem_getpages(), which sleeps if no pages are
available. Thus, the offset can be trashed and the slab code can end
up with a completely bogus address leading to corruption.Kudos to Randolph who came up with the program to trip this problem at
will and thus allowed it to be tracked and fixed.Signed-off-by: James Bottomley
Signed-off-by: Kyle McMartin
-
* arch/parisc/kernel/process.c (sys_clone): Use 5 args, and process
CLONE_PARENT_SETTID, CLONE_CHILD_SETTID, CLONE_CHILD_CLEARTID.
(copy_thread): First cut at CLONE_SETTLS.Signed-off-by: Carlos O'Donell
Signed-off-by: Kyle McMartin
-
2.6.14-rc2-pa1: enable 100/250/1000HZ itimer selection
Signed-off-by: Grant Grundler
Mark floppy as broken. Thanks Joel
Signed-off-by: Matthew Wilcox
make DISCONTIGMEM depend on 64BIT
Signed-off-by: Thibaut VARENE
Signed-off-by: Kyle McMartin
-
Prefix changed in debian, include "gnu" in the commandline.
Signed-off-by: Carlos O'Donell
Ensure the compiler version is new enough (>= 3.3)
Signed-off-by: Randolph Chung
Signed-off-by: Kyle McMartin
-
Document history of PDC_NARROW a bit as it will still show
up in an older kernel's .config file.Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
2.6.12-rc2-pa2 kfree cleanups from Jesper Juhl
Signed-off-by: Jesper Juhl
Signed-off-by: Grant GrundlerSigned-off-by: Kyle McMartin
-
1) cleanup whitespace and handle proc_mkdir() failures.
From: Christophe Lucas2) rename "dino" entry to "pcxl_dma"...also seems like it's in
the wrong location.3) don't dump resource bitmap in /proc/pcxl_dma output
4) fixup compiler warning
Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
Generate a more informative message when a resource does not have
a parent.Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
Get rid of some unnecessary includes
Remove a layer of macro indirection around pdc_console_device
Delete pdc_console_die() as it is unused
Avoid double-printing on panic by clearing CON_PRINTBUFFER rather than
setting con_start to be log_end
Make con_start and log_end static againSigned-off-by: Matthew Wilcox
Signed-off-by: Kyle McMartin
-
2.6.13-rc6-pa2
use force_sigsegv() if we have a problem setting up a frame. This is
required to prevent SIGSEGV loops.Signed-off-by: Randolph Chung
Signed-off-by: Kyle McMartin
-
We're using fp regs now in the kernel, so we want to print them
on stack dumpSigned-off-by: Thibaut VARENE
Signed-off-by: Kyle McMartin
-
Signed-off-by: Matthew Wilcox
Signed-off-by: Kyle McMartin
-
Disable use of fpregs in pa_memcpy, and turn on the
-mdisable-fpregs flag.Signed-off-by: Randolph Chung
Signed-off-by: Kyle McMartin
-
git commit 976ecd12b8144d066a23fe97c6fbfc1ac8470af7 changed our locking
characteristics, and put the onus of spin_lock_init on superio.c.Signed-off-by: Kyle McMartin
-
PA20 arch book (page 7-52 and 7-55) indicate a "sync" is required after
the FDC "to enforce instruction ordering". And we want to make
sure FIC is executed after FDC has retired.Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
minor cleanup: qualify constant with "UL"
Acked-by: "Hmamouche, Youssef"
Signed-off-by: Grant GrundlerSigned-off-by: Kyle McMartin
-
2.6.12-rc4-pa3 s/__LP64__/CONFIG_64BIT/ and fixup config.h usage
Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
2.6.12-rc1-pa6 use work queue in LED/LCD driver instead of tasklet.
Main advantage is it allows use of msleep() in the led_LCD_driver to
"atomically" perform two MMIO writes (CMD, then DATA).
Lead to nice cleanup of the main led_work_func() and led_LCD_driver().
Kudos to David for being persistent.From: David Pye
Signed-off-by: Grant GrundlerSigned-off-by: Kyle McMartin
-
add syscall entries for ioprio_set/get as per Jens Axboe.
Signed-off-by: Jens Axboe
Signed-off-by: Grant GrundlerSigned-off-by: Kyle McMartin
-
Optimize ext2_find_next_zero_bit. Gives about 25% perf improvement with a
rsync test with ext3.Signed-off-by: Randolph Chung
fix ext3 performance - ext2_find_next_zero() was culprit.
Kudos to jejb for pointing out the the possibility that ext2_test_bit
and ext2_find_next_zero() may in fact not be enumerating bits in
the bitmap because of endianess. Took sparc64 implementation and
adapted it to our tree. I suspect the real problem is ffz() wants
an unsigned long and was getting garbage in the top half of the
unsigned int. Not confirmed but that's what I suspect.Signed-off-by: Grant Grundler
Fix find_next_bit for 32-bit
Make masking consistent for bitopsFrom: Joel Soete
Signed-off-by: Randolph ChungAdd back incorrectly removed ext2_find_first_zero_bit definition
Signed-off-by: James Bottomley
Fixup bitops.h to use volatile for *_bit() ops
Based on this email thread:
http://marc.theaimsgroup.com/?t=108826637900003In a nutshell:
*_bit() want use of volatile.
__*_bit() are "relaxed" and don't use spinlock or volatile.other minor changes:
o replaces hweight64() macro with alias to generic_hweight64() (Joel Soete)
o cleanup ext2* macros so (a) it's obvious what the XOR magic is about
and (b) one version that works for both 32/64-bit.
o replace 2 uses of CONFIG_64BIT with __LP64__. bitops.h used both.
I think header files that might go to user space should use
something userspace will know about (__LP64__).Signed-off-by: Grant Grundler
Move SHIFT_PER_LONG to standard location for BITS_PER_LONG (asm/types.h)
and ditch the second definition of BITS_PER_LONG in bitops.hSigned-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
update comment about CAFL_STRIDE
Signed-off-by: Kyle McMartin
Fixed a bug in parisc_setup_cache_timing() which caused it to calculate
a poor value for parisc_cache_flush_threshold.Thanks to Joel Soete for spotting the bug.
Thanks to James Bottomley for pointing out the clean way to fix this.Signed-off-by: Stuart Brady
Signed-off-by: Kyle McMartin
-
Add support for changing unaligned trap behaviour on a
per-thread basis.Signed-off-by: Kyle McMartin
-
convert some bl calls to b,l or bv to use longer offsets
Signed-off-by: Randolph Chung
Signed-off-by: Kyle McMartin
-
o Added a control for the input source (which can be either
"line" or "mic")o Mute the speaker/line-out/headphone outputs by default.
o Increased the buffer size from 10 pages to 16.
Signed-off-by: Stuart Brady
ALSA Harmony was resetting the capture position when
preparing the capture substream, which it shouldn't do.
This should fix the problem.Signed-off-by: Stuart Brady
ALSA Harmony should no longer play junk (left in the buffers
from a previous stream) at the start of a new stream.Implement the monitor mixer channel for ALSA Harmony.
Also prevent snd_harmony_volume_get from returning negative values.
Signed-off-by: Stuart Brady
Use the graveyard/silence buffers in ALSA Harmony.
Signed-off-by: Stuart Brady
Signed-off-by: Kyle McMartin
-
export profile_pc() symbol - oprofile needs it when built as a module.
Signed-off-by: Grant Grundler
Take into account nullified insn and lock functions for profiling
This is needed at the end of functions; it is typical that the return
branch nullifies the next insn, which is in the next function. This
causes profiling data to show up against the "wrong" function.We also count lock times against the locker. This is consistent with
other architectures.Signed-off-by: Randolph Chung
Signed-off-by: Kyle McMartin
-
Fix some whitespace issues
Reorganise parisc_device probe routine to be a little less convoluted
Use ->hpa.start instead of ->hpaSigned-off-by: Matthew Wilcox
Signed-off-by: Kyle McMartin
-
Neaten up the CONFIG_PA20 ifdefs
More merge fixes, this time for SMP
Signed-off-by: Matthew Wilcox
Prettify the CONFIG_DEBUG_SPINLOCK __SPIN_LOCK_UNLOCKED initializers.
Clean up some warnings with CONFIG_DEBUG_SPINLOCK enabled.
Fix build with spinlock debugging turned on. Patch is cleaner like this,
too.Remove mandatory 16-byte alignment requirement on PA2.0 processors by
using the ldcw,CO completer. Provides a nice insn savings.Signed-off-by: Kyle McMartin
-
move pa_tlb_lock and it's primary consumers to tlb_flush.h
Future step will be to move spinlock_t definition out of system.h.Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
2.6.12-rc4-pa3 : first pass at making sure use of RFI conforms to
PA 2.0 arch pages F-4 and F-5, PA 1.1 Arch page 3-19 and 3-20.The discussion revolves around all the rules for clearing PSW Q-bit.
The hard part is meeting all the rules for "relied upon translation"..align directive is used to guarantee the critical sequence ends more than
8 instructions (32 bytes) from the end of page.Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
add || STI_CONSOLE to some of the basic FONTs. May need to get at
least one of them to default to "Y" for parisc.Signed-off-by: Grant Grundler
Signed-off-by: Kyle McMartin
-
Fix card-mode Dino crashes on 725 (and probably other Snake) systems.
Dino was coming up in fatal mode after a warm reboot. Resetting Dino
brings it out of fatal mode, so do that if the status register indicates
we're in fatal mode. Since this was never observed on any later systems,
I presume firmware does this for us on those.Signed-off-by: Matthew Wilcox
Add debug statements in the cfg_read and cfg_write functions
Fix debug statements from the IRQ overhaul last winter
Rename dino_driver_callback() to dino_probe()Signed-off-by: Matthew Wilcox
Signed-off-by: Kyle McMartin
