head_64.S maps up to 512 MiB, but that is not necessarity true for
other entry paths, such as Xen.
Thus, co-locate the setting of max_pfn_mapped with the code to
actually set up the page tables in head_64.S. The 32-bit code is
already so co-located. (The Xen code already sets max_pfn_mapped
correctly for its own use case.)
-v2:
Yinghai fixed the following bug in this patch:
|
| max_pfn_mapped is in .bss section, so we need to set that
| after bss get cleared. Without that we crash on bootup.
|
| That is safe because Xen does not call x86_64_start_kernel().
|
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Fixed-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
LKML-Reference: <4CB6AB24.9020504@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Xen requires that all pages containing pagetable entries to be mapped
read-only. If pages used for the initial pagetable are already mapped
then we can change the mapping to RO. However, if they are initially
unmapped, we need to make sure that when they are later mapped, they
are also mapped RO.
We do this by knowing that the kernel pagetable memory is pre-allocated
in the range e820_table_start - e820_table_end, so any pfn within this
range should be mapped read-only. However, the pagetable setup code
early_ioremaps the pages to write their entries, so we must make sure
that mappings created in the early_ioremap fixmap area are mapped RW.
(Those mappings are removed before the pages are presented to Xen
as pagetable pages.)
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
LKML-Reference: <4CB63A80.8060702@goop.org>
Cc: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Reason for merge:
Forward-port urgent change to arch/x86/mm/srat_64.c to the memblock tree.
Resolved Conflicts:
arch/x86/mm/srat_64.c
Originally-by: Yinghai Lu <yinghai@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
The Xen setup code needs to call memblock_x86_reserve_range() very early,
so allow it to initialize the memblock subsystem before doing so. The
second memblock_init() is ignored.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Yinghai Lu <yinghai@kernel.org>
Cc: Benjamin Herrenschmidt <benh@kernel.crashing.org>
LKML-Reference: <4CACFDAD.3090900@goop.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Russ reported SGI UV is broken recently. He said:
| The SRAT table shows that memory range is spread over two nodes.
|
| SRAT: Node 0 PXM 0 100000000-800000000
| SRAT: Node 1 PXM 1 800000000-1000000000
| SRAT: Node 0 PXM 0 1000000000-1080000000
|
|Previously, the kernel early_node_map[] would show three entries
|with the proper node.
|
|[ 0.000000] 0: 0x00100000 -> 0x00800000
|[ 0.000000] 1: 0x00800000 -> 0x01000000
|[ 0.000000] 0: 0x01000000 -> 0x01080000
|
|The problem is recent community kernel early_node_map[] shows
|only two entries with the node 0 entry overlapping the node 1
|entry.
|
| 0: 0x00100000 -> 0x01080000
| 1: 0x00800000 -> 0x01000000
After looking at the changelog, Found out that it has been broken for a while by
following commit
|commit 8716273cae
|Author: David Rientjes <rientjes@google.com>
|Date: Fri Sep 25 15:20:04 2009 -0700
|
| x86: Export srat physical topology
Before that commit, register_active_regions() is called for every SRAT memory
entry right away.
Use nodememblk_range[] instead of nodes[] in order to make sure we
capture the actual memory blocks registered with each node. nodes[]
contains an extended range which spans all memory regions associated
with a node, but that does not mean that all the memory in between are
included.
Reported-by: Russ Anderson <rja@sgi.com>
Tested-by: Russ Anderson <rja@sgi.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4CB27BDF.5000800@kernel.org>
Acked-by: David Rientjes <rientjes@google.com>
Cc: <stable@kernel.org> 2.6.33 .34 .35 .36
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
This fixes possible cases of not collecting valid error info in
the MCE error thresholding groups on F10h hardware.
The current code contains a subtle problem of checking only the
Valid bit of MSR0000_0413 (which is MC4_MISC0 - DRAM
thresholding group) in its first iteration and breaking out if
the bit is cleared.
But (!), this MSR contains an offset value, BlkPtr[31:24], which
points to the remaining MSRs in this thresholding group which
might contain valid information too. But if we bail out only
after we checked the valid bit in the first MSR and not the
block pointer too, we miss that other information.
The thing is, MC4_MISC0[BlkPtr] is not predicated on
MCi_STATUS[MiscV] or MC4_MISC0[Valid] and should be checked
prior to iterating over the MCI_MISCj thresholding group,
irrespective of the MC4_MISC0[Valid] setting.
Signed-off-by: Borislav Petkov <borislav.petkov@amd.com>
Cc: <stable@kernel.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When the feature PTS is not supported by CPU, the sysfile
package_power_limit_count for package should not be
generated.
This patch is used for fixing missing { and }.
The patch is not complete as there are other error handling
problems in this function - but that can wait until the
merge window.
Signed-off-by: Jin Dongming <jin.dongming@np.css.fujitsu.com>
Reviewed-by: Fenghua Yu <fenghua.yu@initel.com>
Acked-by: Jean Delvare <khali@linux-fr.org>
Cc: Brown Len <len.brown@intel.com>
Cc: Guenter Roeck <guenter.roeck@ericsson.com>
Cc: Hidetoshi Seto <seto.hidetoshi@jp.fujitsu.com>
Cc: lm-sensors@lm-sensors.org <lm-sensors@lm-sensors.org>
LKML-Reference: <4C7625D1.4060201@np.css.fujitsu.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* 'v2.6.36-rc6-urgent-fixes' of git://xenbits.xen.org/people/sstabellini/linux-pvhvm:
xen: do not initialize PV timers on HVM if !xen_have_vector_callback
xen: do not set xenstored_ready before xenbus_probe on hvm
Fold it into memblock_x86_find_in_range(), and change bad_addr_size()
to check_reserve_memblock().
So whole memblock_x86_find_in_range_size() code is more readable.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4CAA4DEC.4000401@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Originally the only early reserved range that is overlapped with high
pages is "KVA RAM", but we already do remove that from the active ranges.
However, It turns out Xen could have that kind of overlapping to support memory
ballooning.x
So we need to make add_highpage_with_active_regions() to subtract
memblock reserved just like low ram; this is the proper design anyway.
In this patch, refactering get_freel_all_memory_range() to make it can
be used by add_highpage_with_active_regions(). Also we don't need to
remove "KVA RAM" from active ranges.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4CABB183.1040607@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Cai Qian found crashkernel is broken with the x86 memblock changes.
1. crashkernel=128M@32M always reported that range is used, even if
the first kernel is small and does not usethat range
2. we always got following report when using "kexec -p"
Could not find a free area of memory of a000 bytes...
locate_hole failed
The root cause is that generic memblock_find_in_range() will try to
allocate from the top of the range, whereas the kexec code was written
assuming that allocation was always near the bottom and that it could
blindly extend memory upward. Unfortunately the kexec code doesn't
have a system for requesting the range that it really needs, so this
is subject to probabilistic failures.
This patch hacks around the problem by limiting the target range
heuristically to below the traditional bzImage max range. This number
is arbitrary and not always correct, and a much better result would be
obtained by having kexec communicate this number based on the kernel
header information and any appropriate command line options.
Reported-and-Bisected-by: CAI Qian <caiqian@redhat.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4CABAF2A.5090501@kernel.org>
Cc: Vivek Goyal <vgoyal@redhat.com>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
* 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
perf trace scripting: Fix extern struct definitions
perf ui hist browser: Fix segfault on 'a' for annotate
perf tools: Fix build breakage
perf, x86: Handle in flight NMIs on P4 platform
oprofile, ARM: Release resources on failure
oprofile: Add Support for Intel CPU Family 6 / Model 29
With all the recent module loading cleanups, we've minimized the code
that sits under module_mutex, fixing various deadlocks and making it
possible to do most of the module loading in parallel.
However, that whole conversion totally missed the rather obscure code
that adds a new module to the list for BUG() handling. That code was
doubly obscure because (a) the code itself lives in lib/bugs.c (for
dubious reasons) and (b) it gets called from the architecture-specific
"module_finalize()" rather than from generic code.
Calling it from arch-specific code makes no sense what-so-ever to begin
with, and is now actively wrong since that code isn't protected by the
module loading lock any more.
So this commit moves the "module_bug_{finalize,cleanup}()" calls away
from the arch-specific code, and into the generic code - and in the
process protects it with the module_mutex so that the list operations
are now safe.
Future fixups:
- move the module list handling code into kernel/module.c where it
belongs.
- get rid of 'module_bug_list' and just use the regular list of modules
(called 'modules' - imagine that) that we already create and maintain
for other reasons.
Reported-and-tested-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Rusty Russell <rusty@rustcorp.com.au>
Cc: Adrian Bunk <bunk@kernel.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: stable@kernel.org
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
if !xen_have_vector_callback do not initialize PV timer unconditionally
because we still don't know how many cpus are available and if there is
more than one we won't be able to receive the timer interrupts on
cpu > 0.
This patch fixes an hang at boot when Xen does not support vector
callbacks and the guest has multiple vcpus.
Signed-off-by: Stefano Stabellini <stefano.stabellini@eu.citrix.com>
Acked-by: Jeremy Fitzhardinge <jeremy@goop.org>
create_irq() returns -1 if the interrupt allocation failed, but the
code checks for irq == 0.
Use create_irq_nr() instead.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Venkatesh Pallipadi <venki@google.com>
LKML-Reference: <alpine.LFD.2.00.1009282310360.2416@localhost6.localdomain6>
Cc: stable@kernel.org
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
free_irq_cfg() is not freeing the cpumask_vars in irq_cfg. Fixing this
triggers a use after free caused by the fact that copying struct
irq_cfg is done with memcpy, which copies the pointer not the cpumask.
Fix both places.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: Yinghai Lu <yhlu.kernel@gmail.com>
LKML-Reference: <alpine.LFD.2.00.1009282052570.2416@localhost6.localdomain6>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Cc: stable@kernel.org
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
If acpi_evaluate_object() function call doesn't fail, we must kfree()
output.buffer before returning from pcc_cpufreq_do_osc().
Signed-off-by: Pekka Enberg <penberg@cs.helsinki.fi>
Acked-by: David Rientjes <rientjes@google.com>
Signed-off-by: Dave Jones <davej@redhat.com>
acpi_perf_data is a percpu pointer but was missing __percpu markup.
Add it.
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Dave Jones <davej@redhat.com>
cpu_cstate_entry is a percpu pointer
but was missing __percpu markup.
Signed-off-by: Namhyung Kim <namhyung@gmail.com>
Acked-by: Tejun Heo <tj@kernel.org>
Signed-off-by: Len Brown <len.brown@intel.com>
After uncapping the CPUID level, we need to also re-run the CPU
feature detection code.
This resolves kernel bugzilla 16322.
Reported-by: boris64 <bugzilla.kernel.org@boris64.net>
Cc: <stable@kernel.org> v2.6.29..2.6.35
LKML-Reference: <tip-@git.kernel.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
* 'x86/urgent' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86: Avoid 'constant_test_bit()' misoptimization due to cast to non-volatile
* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86/amd-iommu: Fix rounding-bug in __unmap_single
x86/amd-iommu: Work around S3 BIOS bug
x86/amd-iommu: Set iommu configuration flags in enable-loop
x86, setup: Fix earlyprintk=serial,0x3f8,115200
x86, setup: Fix earlyprintk=serial,ttyS0,115200
* 'perf-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
perf, x86: Catch spurious interrupts after disabling counters
tracing/x86: Don't use mcount in kvmclock.c
tracing/x86: Don't use mcount in pvclock.c
While debugging bit_spin_lock() hang, it was tracked down to gcc-4.4
misoptimization of non-inlined constant_test_bit() due to non-volatile
addr when 'const volatile unsigned long *addr' cast to 'unsigned long *'
with subsequent unconditional jump to pause (and not to the test) leading
to hang.
Compiling with gcc-4.3 or disabling CONFIG_OPTIMIZE_INLINING yields inlined
constant_test_bit() and correct jump, thus working around the kernel bug.
Other arches than asm-x86 may implement this slightly differently;
2.6.29 mitigates the misoptimization by changing the function prototype
(commit c4295fbb60) but probably fixing the issue
itself is better.
Signed-off-by: Alexander Chumachenko <ledest@gmail.com>
Signed-off-by: Michael Shigorin <mike@osdn.org.ua>
Acked-by: Linus Torvalds <torvalds@linux-foundation.org>
Signed-off-by: H. Peter Anvin <hpa@zytor.com>
Using cpuid_eax() to determine feature availability on other than
the current CPU is invalid. And feature availability should also be
checked in the hotplug code path.
Signed-off-by: Jan Beulich <jbeulich@novell.com>
Cc: Rudolf Marek <r.marek@assembler.cz>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Signed-off-by: Guenter Roeck <guenter.roeck@ericsson.com>
In the __unmap_single function the dma_addr is rounded down
to a page boundary before the dma pages are unmapped. The
address is later also used to flush the TLB entries for that
mapping. But without the offset into the dma page the amount
of pages to flush might be miscalculated in the TLB flushing
path. This patch fixes this bug by using the original
address to flush the TLB.
Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch adds a workaround for an IOMMU BIOS problem to
the AMD IOMMU driver. The result of the bug is that the
IOMMU does not execute commands anymore when the system
comes out of the S3 state resulting in system failure. The
bug in the BIOS is that is does not restore certain hardware
specific registers correctly. This workaround reads out the
contents of these registers at boot time and restores them
on resume from S3. The workaround is limited to the specific
IOMMU chipset where this problem occurs.
Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
This patch moves the setting of the configuration and
feature flags out out the acpi table parsing path and moves
it into the iommu-enable path. This is needed to reliably
fix resume-from-s3.
Cc: stable@kernel.org
Signed-off-by: Joerg Roedel <joerg.roedel@amd.com>
The guest can use the paravirt clock in kvmclock.c which is used
by sched_clock(), which in turn is used by the tracing mechanism
for timestamps, which leads to infinite recursion.
Disable mcount/tracing for kvmclock.o.
Cc: stable@kernel.org
Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Cc: Avi Kivity <avi@redhat.com>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
When using a paravirt clock, pvclock.c can be used by sched_clock(),
which in turn is used by the tracing mechanism for timestamps,
which leads to infinite recursion.
Disable mcount/tracing for pvclock.o.
Cc: stable@kernel.org
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
LKML-Reference: <4C9A9A3F.4040201@goop.org>
Signed-off-by: Steven Rostedt <rostedt@goodmis.org>
earlyprintk can take and I/O port, so we need to handle this case in
the setup code too, otherwise 0x3f8 will be treated as a baud rate.
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4C7B05A6.4010801@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Torsten reported that there is garbage output,
after commit 8fee13a48e (x86,
setup: enable early console output from the decompressor)
It turns out we missed the offset for that case.
Reported-by: Torsten Kaiser <just.for.lkml@googlemail.com>
Signed-off-by: Yinghai Lu <yinghai@kernel.org>
LKML-Reference: <4C7B0578.8090807@kernel.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
This patch adds CPU type detection for dunnington processor (Family 6
/ Model 29) to be identified as core 2 family cpu type (wikipedia
source).
I tested oprofile on Intel(R) Xeon(R) CPU E7440 reporting itself as
model 29, and it runs without an issue.
Spec:
http://www.intel.com/Assets/en_US/PDF/specupdate/320336.pdf
Signed-off-by: Jiri Olsa <jolsa@redhat.com>
Acked-by: Andi Kleen <ak@linux.intel.com>
Cc: stable@kernel.org
Signed-off-by: Robert Richter <robert.richter@amd.com>
We used to have a hypercall which reloaded the entire GDT, then we
switched to one which loaded a single entry (to match the IDT code).
Some comments were not updated, so fix them.
Signed-off-by: Rusty Russell <rusty@rustcorp.com.au>
Reported by: Eviatar Khen <eviatarkhen@gmail.com>
* 'x86-fixes-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/linux-2.6-tip:
x86: hpet: Work around hardware stupidity
x86, build: Disable -fPIE when compiling with CONFIG_CC_STACKPROTECTOR=y
x86, cpufeature: Suppress compiler warning with gcc 3.x
x86, UV: Fix initialization of max_pnode
Lengths and types of breakpoints are encoded in a half byte
into CPU registers. However when we extract these values
and store them, we add a high half byte part to them: 0x40 to the
length and 0x80 to the type.
When that gets reloaded to the CPU registers, the high part
is masked.
While making the instruction breakpoints available for perf,
I zapped that high part on instruction breakpoint encoding
and that broke the arch -> generic translation used by ptrace
instruction breakpoints. Writing dr7 to set an inst breakpoint
was then failing.
There is no apparent reason for these high parts so we could get
rid of them altogether. That's an invasive change though so let's
do that later and for now fix the problem by restoring that inst
breakpoint high part encoding in this sole patch.
Reported-by: Kelvie Wong <kelvie@ieee.org>
Signed-off-by: Frederic Weisbecker <fweisbec@gmail.com>
Cc: Prasad <prasad@linux.vnet.ibm.com>
Cc: Mahesh Salgaonkar <mahesh@linux.vnet.ibm.com>
Cc: Will Deacon <will.deacon@arm.com>
This patch adds CPU type detection for the Intel Celeron 540, which is
part of the Core 2 family according to Wikipedia; the family and ID pair
is absent from the Volume 3B table referenced in the source code
comments. I have tested this patch on an Intel Celeron 540 machine
reporting itself as Family 6 Model 22, and OProfile runs on the machine
without issue.
Spec:
http://download.intel.com/design/mobile/SPECUPDT/317667.pdf
Signed-off-by: Patrick Simmons <linuxrocks123@netscape.net>
Acked-by: Andi Kleen <ak@linux.intel.com>
Acked-by: Arnd Bergmann <arnd@arndb.de>
Cc: stable@kernel.org
Signed-off-by: Robert Richter <robert.richter@amd.com>
In commit d4d6715, we reopened an old hole for a 64-bit ptracer touching a
32-bit tracee in system call entry. A %rax value set via ptrace at the
entry tracing stop gets used whole as a 32-bit syscall number, while we
only check the low 32 bits for validity.
Fix it by truncating %rax back to 32 bits after syscall_trace_enter,
in addition to testing the full 64 bits as has already been added.
Reported-by: Ben Hawkes <hawkes@sota.gen.nz>
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
On 64 bits, we always, by necessity, jump through the system call
table via %rax. For 32-bit system calls, in theory the system call
number is stored in %eax, and the code was testing %eax for a valid
system call number. At one point we loaded the stored value back from
the stack to enforce zero-extension, but that was removed in checkin
d4d6715016. An actual 32-bit process
will not be able to introduce a non-zero-extended number, but it can
happen via ptrace.
Instead of re-introducing the zero-extension, test what we are
actually going to use, i.e. %rax. This only adds a handful of REX
prefixes to the code.
Reported-by: Ben Hawkes <hawkes@sota.gen.nz>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Cc: <stable@kernel.org>
Cc: Roland McGrath <roland@redhat.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
compat_alloc_user_space() expects the caller to independently call
access_ok() to verify the returned area. A missing call could
introduce problems on some architectures.
This patch incorporates the access_ok() check into
compat_alloc_user_space() and also adds a sanity check on the length.
The existing compat_alloc_user_space() implementations are renamed
arch_compat_alloc_user_space() and are used as part of the
implementation of the new global function.
This patch assumes NULL will cause __get_user()/__put_user() to either
fail or access userspace on all architectures. This should be
followed by checking the return value of compat_access_user_space()
for NULL in the callers, at which time the access_ok() in the callers
can also be removed.
Reported-by: Ben Hawkes <hawkes@sota.gen.nz>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
Acked-by: Benjamin Herrenschmidt <benh@kernel.crashing.org>
Acked-by: Chris Metcalf <cmetcalf@tilera.com>
Acked-by: David S. Miller <davem@davemloft.net>
Acked-by: Ingo Molnar <mingo@elte.hu>
Acked-by: Thomas Gleixner <tglx@linutronix.de>
Acked-by: Tony Luck <tony.luck@intel.com>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Arnd Bergmann <arnd@arndb.de>
Cc: Fenghua Yu <fenghua.yu@intel.com>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Heiko Carstens <heiko.carstens@de.ibm.com>
Cc: Helge Deller <deller@gmx.de>
Cc: James Bottomley <jejb@parisc-linux.org>
Cc: Kyle McMartin <kyle@mcmartin.ca>
Cc: Martin Schwidefsky <schwidefsky@de.ibm.com>
Cc: Paul Mackerras <paulus@samba.org>
Cc: Ralf Baechle <ralf@linux-mips.org>
Cc: <stable@kernel.org>
This more or less reverts commits 08be979 (x86: Force HPET
readback_cmp for all ATI chipsets) and 30a564be (x86, hpet: Restrict
read back to affected ATI chipsets) to the status of commit 8da854c
(x86, hpet: Erratum workaround for read after write of HPET
comparator).
The delta to commit 8da854c is mostly comments and the change from
WARN_ONCE to printk_once as we know the call path of this function
already.
This needs really in depth explanation:
First of all the HPET design is a complete failure. Having a counter
compare register which generates an interrupt on matching values
forces the software to do at least one superfluous readback of the
counter register.
While it is nice in theory to program "absolute" time events it is
practically useless because the timer runs at some absurd frequency
which can never be matched to real world units. So we are forced to
calculate a relative delta and this forces a readout of the actual
counter value, adding the delta and programming the compare
register. When the delta is small enough we run into the danger that
we program a compare value which is already in the past. Due to the
compare for equal nature of HPET we need to read back the counter
value after writing the compare rehgister (btw. this is necessary for
absolute timeouts as well) to make sure that we did not miss the timer
event. We try to work around that by setting the minimum delta to a
value which is larger than the theoretical time which elapses between
the counter readout and the compare register write, but that's only
true in theory. A NMI or SMI which hits between the readout and the
write can easily push us beyond that limit. This would result in
waiting for the next HPET timer interrupt until the 32bit wraparound
of the counter happens which takes about 306 seconds.
So we designed the next event function to look like:
match = read_cnt() + delta;
write_compare_ref(match);
return read_cnt() < match ? 0 : -ETIME;
At some point we got into trouble with certain ATI chipsets. Even the
above "safe" procedure failed. The reason was that the write to the
compare register was delayed probably for performance reasons. The
theory was that they wanted to avoid the synchronization of the write
with the HPET clock, which is understandable. So the write does not
hit the compare register directly instead it goes to some intermediate
register which is copied to the real compare register in sync with the
HPET clock. That opens another window for hitting the dreaded "wait
for a wraparound" problem.
To work around that "optimization" we added a read back of the compare
register which either enforced the update of the just written value or
just delayed the readout of the counter enough to avoid the issue. We
unfortunately never got any affirmative info from ATI/AMD about this.
One thing is sure, that we nuked the performance "optimization" that
way completely and I'm pretty sure that the result is worse than
before some HW folks came up with those.
Just for paranoia reasons I added a check whether the read back
compare register value was the same as the value we wrote right
before. That paranoia check triggered a couple of years after it was
added on an Intel ICH9 chipset. Venki added a workaround (commit
8da854c) which was reading the compare register twice when the first
check failed. We considered this to be a penalty in general and
restricted the readback (thus the wasted CPU cycles) to the known to
be affected ATI chipsets.
This turned out to be a utterly wrong decision. 2.6.35 testers
experienced massive problems and finally one of them bisected it down
to commit 30a564be which spured some further investigation.
Finally we got confirmation that the write to the compare register can
be delayed by up to two HPET clock cycles which explains the problems
nicely. All we can do about this is to go back to Venki's initial
workaround in a slightly modified version.
Just for the record I need to say, that all of this could have been
avoided if hardware designers and of course the HPET committee would
have thought about the consequences for a split second. It's out of my
comprehension why designing a working timer is so hard. There are two
ways to achieve it:
1) Use a counter wrap around aware compare_reg <= counter_reg
implementation instead of the easy compare_reg == counter_reg
Downsides:
- It needs more silicon.
- It needs a readout of the counter to apply a relative
timeout. This is necessary as the counter does not run in
any useful (and adjustable) frequency and there is no
guarantee that the counter which is used for timer events is
the same which is used for reading the actual time (and
therefor for calculating the delta)
Upsides:
- None
2) Use a simple down counter for relative timer events
Downsides:
- Absolute timeouts are not possible, which is not a problem
at all in the context of an OS and the expected
max. latencies/jitter (also see Downsides of #1)
Upsides:
- It needs less or equal silicon.
- It works ALWAYS
- It is way faster than a compare register based solution (One
write versus one write plus at least one and up to four
reads)
I would not be so grumpy about all of this, if I would not have been
ignored for many years when pointing out these flaws to various
hardware folks. I really hate timers (at least those which seem to be
designed by janitors).
Though finally we got a reasonable explanation plus a solution and I
want to thank all the folks involved in chasing it down and providing
valuable input to this.
Bisected-by: Nix <nix@esperi.org.uk>
Reported-by: Artur Skawina <art.08.09@gmail.com>
Reported-by: Damien Wyart <damien.wyart@free.fr>
Reported-by: John Drescher <drescherjm@gmail.com>
Cc: Venkatesh Pallipadi <venki@google.com>
Cc: Ingo Molnar <mingo@elte.hu>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Arjan van de Ven <arjan@linux.intel.com>
Cc: Andreas Herrmann <andreas.herrmann3@amd.com>
Cc: Borislav Petkov <borislav.petkov@amd.com>
Cc: stable@kernel.org
Acked-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
The arch/x86/Makefile uses scripts/gcc-x86_$(BITS)-has-stack-protector.sh
to check if cc1 supports -fstack-protector. When -fPIE is passed to cc1,
these scripts fail causing stack protection to be disabled even when it
is available.
This fix is similar to commit c47efe5548
Reported-by: Kai Dietrich <mail@cleeus.de>
Signed-off-by: Magnus Granberg <zorry@gentoo.org>
LKML-Reference: <20100913101319.748A1148E216@opensource.dyc.edu>
Signed-off-by: Anthony G. Basile <basile@opensource.dyc.edu>
Cc: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: H. Peter Anvin <hpa@linux.intel.com>
