On AMD SMM protected memory is part of the address map, but handled
internally like an MTRR. That leads to large pages getting split
internally which has some performance implications. Check for the
AMD TSEG MSR and split the large page mapping on that area
explicitely if it is part of the direct mapping.
There is also SMM ASEG, but it is in the first 1MB and already covered by
the earlier split first page patch.
Idea for this came from an earlier patch by Andreas Herrmann
On a RevF dual Socket Opteron system kernbench shows a clear
improvement from this:
(together with the earlier patches in this series, especially the
split first 2MB patch)
[lower is better]
no split stddev split stddev delta
Elapsed Time 87.146 (0.727516) 84.296 (1.09098) -3.2%
User Time 274.537 (4.05226) 273.692 (3.34344) -0.3%
System Time 34.907 (0.42492) 34.508 (0.26832) -1.1%
Percent CPU 322.5 (38.3007) 326.5 (44.5128) +1.2%
=> About 3.2% improvement in elapsed time for kernbench.
With GB pages on AMD Fam1h the impact of splitting is much higher of course,
since it would split two full GB pages (together with the first
1MB split patch) instead of two 2MB pages. I could not benchmark
a clear difference in kernbench on gbpages, so I kept it disabled
for that case
That was only limited benchmarking of course, so if someone
was interested in running more tests for the gbpages case
that could be revisited (contributions welcome)
I didn't bother implementing this for 32bit because it is very
unlikely the 32bit lowmem mapping overlaps into the TSEG near 4GB
and the 2MB low split is already handled for both.
[ mingo@elte.hu: do it on gbpages kernels too, there's no clear reason
why it shouldnt help there. ]
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Intel recommends to not use large pages for the first 1MB
of the physical memory because there are fixed size MTRRs there
which cause splitups in the TLBs.
On AMD doing so is also a good idea.
The implementation is a little different between 32bit and 64bit.
On 32bit I just taught the initial page table set up about this
because it was very simple to do. This also has the advantage
that the risk of a prefetch ever seeing the page even
if it only exists for a short time is minimized.
On 64bit that is not quite possible, so use set_memory_4k() a little
later (in check_bugs) instead.
Signed-off-by: Andi Kleen <ak@suse.de>
Acked-by: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add a new function to force split large pages into 4k pages.
This is needed for some followup optimizations.
I had to add a new field to cpa_data to pass down the information
that try_preserve_large_page should not run.
Right now no set_page_4k() because I didn't need it and all the
specialized users I have in mind would be more comfortable with
pure addresses. I also didn't export it because it's unlikely
external code needs it.
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When end_pfn is not aligned to 2MB (or 1GB) then the kernel might
map more memory than end_pfn. Account this in max_pfn_mapped.
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Even on 32bit 2MB pages can map more memory than is in the true
max_low_pfn if end_pfn is not highmem and not aligned to 2MB.
Add a end_pfn_map similar to x86-64 that accounts for this
fact. This is important for code that really needs to know about
all mapping aliases.
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: andreas.herrmann3@amd.com
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Currently they are in .text.head because the rest of head_64.S.
.text.head is not removed as init data, but the early exception handlers
should be because they are not needed after early boot of the BP.
So move them over.
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The early exception handlers are currently set up using a macro
recursion. There is only one user left. Replace the macro with a
standard loop in place.
Noop patch, just a cleanup.
[ tglx@linutronix.de: simplified ]
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
All of early setup runs with interrupts disabled, so there is no
need to set up early exception handlers for vectors >= 32
This saves some minor text size.
Signed-off-by: Andi Kleen <ak@suse.de>
Cc: mingo@elte.hu
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
* Ingo Molnar (mingo@elte.hu) wrote:
>
> * Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca> wrote:
>
> > The shadow vmap for DEBUG_RODATA kernel text modification uses
> > virt_to_page to get the pages from the pointer address.
> >
> > However, I think vmalloc_to_page would be required in case the page is
> > used for modules.
> >
> > Since only the core kernel text is marked read-only, use
> > kernel_text_address() to make sure we only shadow map the core kernel
> > text, not modules.
>
> actually, i think we should mark module text readonly too.
>
Yes, but in the meantime, the x86 tree would need this patch to make
kprobes work correctly on modules.
I suspect that without this fix, with the enhanced hotplug and kprobes
patch, kprobes will use text_poke to insert breakpoints in modules
(vmalloced pages used), which will map the wrong pages and corrupt
random kernel locations instead of updating the correct page.
Work that would write protect the module pages should clearly be done,
but it can come in a later time. We have to make sure we interact
correctly with the page allocation debugging, as an example.
Here is the patch against x86.git 2.6.25-rc5 :
The shadow vmap for DEBUG_RODATA kernel text modification uses virt_to_page to
get the pages from the pointer address.
However, I think vmalloc_to_page would be required in case the page is used for
modules.
Since only the core kernel text is marked read-only, use kernel_text_address()
to make sure we only shadow map the core kernel text, not modules.
Signed-off-by: Mathieu Desnoyers <mathieu.desnoyers@polymtl.ca>
CC: akpm@linux-foundation.org
Signed-off-by: Ingo Molnar <mingo@elte.hu>
vSMP detection: access pci config space early in boot to detect if the
system is a vSMPowered box, and cache the result in a flag, so that
is_vsmp_box() retrieves the value of the flag always.
Signed-off-by: Ravikiran Thirumalai <kiran@scalex86.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The sysenter path tries to enable interrupts immediately. Unfortunately
this doesn't work in a paravirt environment, because not enough kernel
state has been set up at that point (namely, pointing %fs to the kernel
percpu data segment). To fix this, defer ENABLE_INTERRUPTS until after
the kernel state has been set up.
Unfortunately this means that we're running with interrupts disabled
for a while without calling the IRQ tracing code, but that can't be
called without setting up %fs either.
Signed-off-by: Jeremy Fitzhardinge <jeremy.fitzhardinge@citrix.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
This patch does clean up relocate_kernel_(32|64).S a bit by getting rid
of local PAGE_ALIGNED macro. We should use well-known PAGE_SIZE instead
Signed-off-by: Cyrill Gorcunov <gorcunov@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Allow the maximum number of nodes in an x86_64 system to
be configurable. This patch does NOT change the default value
but allows the value to be a config option.
Signed-off-by: Jack Steiner <steiner@sgi.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arch/x86/math-emu/reg_ld_str.c:380: warning: 'l[0]' may be used uninitialized in this function
arch/x86/math-emu/reg_ld_str.c:380: warning: 'l[1]' may be used uninitialized in this function
I can't actually spot the bug here. There's one obvious place, but fixing
that didn't shut the warning up.
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
arch/x86/math-emu/fpu_entry.c:555: warning: 'entry_sel_off.empty' is used uninitialized in this function
Presumably it's harmless, but I'll sleep better at night knowing that we
initialised it.
Cc: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Make the PAT related printks in ioremap pr_debug.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Bug fixes for reserve_memtype() call in __ioremap and pci_mmap_page_range().
If reserve_memtype returns non-zero, then it is an error and subsequent free is
not required. Requested and returned prot value check should be done when
reserve_memtype returns success.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
make known_pat_cpu to think amd k8 and fam10h is ok too.
also make tom2 below to be WRBACK
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Fix double help section in PAT Kconfig. Thanks to Randy Dunlap for catching
this bug.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Adds debug prints at critical code. Adds enough info in dmesg to allow us to
do effective first round of analysis of any issues that may result due to PAT
patch series.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Introduce ioremap_wc for wc remap.
(generic wrapper is in a later patch)
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add a set_memory_wc interface(), similar to set_memory_uc interface.
Callers has to call set_memory_uc, set_memory_wb and
set_memory_wc, set_memory_wb as pairs.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Add reserve_memtype and free_memtype wrapper for pci_mmap_page_range. Free
is called on unmap, but identity map continues to be mapped as per
pci_mmap_page_range request, until next request for the same region calls
ioremap_change_attr(), which will go through without conflict. This way of
mapping is identical to one used in ioremap/iounmap.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use reserve_memtype and free_memtype interfaces in set_memory_uc/set_memory_wb
interfaces to avoid aliasing.
Usage model of set_memory_uc and set_memory_wb is for RAM memory and users
will first call set_memory_uc and call set_memory_wb after use to reset the
attribute.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use reserve_memtype and free_memtype interfaces in ioremap/iounmap to avoid
aliasing.
If there is an existing alias for the region, inherit the memory type from
the alias. If there are conflicting aliases for the entire region, then fail
ioremap.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Make ioremap_change_attr() non-static and use prot_val in place of ioremap_mode.
This interface is used in subsequent PAT patches.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Sets up pat_init() infrastructure.
PAT MSR has following setting.
PAT
|PCD
||PWT
|||
000 WB _PAGE_CACHE_WB
001 WC _PAGE_CACHE_WC
010 UC- _PAGE_CACHE_UC_MINUS
011 UC _PAGE_CACHE_UC
We are effectively changing WT from boot time setting to WC.
UC_MINUS is used to provide backward compatibility to existing /dev/mem
users(X).
reserve_memtype and free_memtype are new interfaces for maintaining alias-free
mapping. It is currently implemented in a simple way with a linked list and
not optimized. reserve and free tracks the effective memory type, as a result
of PAT and MTRR setting rather than what is actually requested in PAT.
pat_init piggy backs on mtrr_init as the rules for setting both pat and mtrr
are same.
Signed-off-by: Venkatesh Pallipadi <venkatesh.pallipadi@intel.com>
Signed-off-by: Suresh Siddha <suresh.b.siddha@intel.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Initializing to zero is generally bad idea, I hope it is right for
__init data, too.
Signed-off-by: Pavel Machek <pavel@suse.cz>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
do simple memtest after init_memory_mapping
use find_e820_area_size to find all ram range that is not reserved.
and do some simple bits test to find some bad ram.
if find some bad ram, use reserve_early to exclude that range.
Signed-off-by: Yinghai Lu <yhlu.kernel@gmail.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
After an experimental cleanup of <linux/percpu.h>, these files were
exposed as invoking kmalloc() without including <linux/slab.h>.
Signed-off-by: Robert P. J. Day <rpjday@crashcourse.ca>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
I was trying to get the address of instruction to be executed
next after the kprobed instruction. But regs->eip in post_handler()
contains value which is useless to the user. It's pre-corrected value.
This value is difficult to use without access to resume_execution(), which
is not exported anyway.
I moved the invocation of post_handler() to *after* resume_execution().
Now regs->eip contains meaningful value in post_handler().
I do not think this change breaks any backward-compatibility.
To make meaning of the old value, post_handler() would need access to
resume_execution() which is not exported. I have difficulty to believe
that previous, uncorrected, regs->eip can be meaningfully used in
post_handler().
Signed-off-by: Yakov Lerner <iler.ml@gmail.com>
Acked-by: Ananth N Mavinakayanahalli <ananth@in.ibm.com>
Acked-by: Masami Hiramatsu <mhiramat@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Use force_sig in handle_vm86_trap like other machine traps do.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The PT_DTRACE flag is meaningless and obsolete.
Don't touch it.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
The previous "x86_64 ia32 ptrace vs -ENOSYS" fix only covered
the int $0x80 system call entries. This does the same fix
for the sysenter and syscall instruction paths.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When we're stopped at syscall entry tracing, ptrace can change the %rax
value from -ENOSYS to something else. If no system call is actually made
because the syscall number (now in orig_rax) is bad, then we now always
reset %rax to -ENOSYS again.
This changes it to leave the return value alone after entry tracing.
That way, the %rax value set by ptrace is there to be seen in user mode
(or in syscall exit tracing). This is consistent with what the 32-bit
kernel does.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
When we're stopped at syscall entry tracing, ptrace can change the %eax
value from -ENOSYS to something else. If no system call is actually made
because the syscall number (now in orig_eax) is bad, then the %eax value
set by ptrace should be returned to the user. But, instead it gets reset
to -ENOSYS again. This is a regression from the native 32-bit kernel.
This change fixes it by leaving the return value alone after entry tracing.
Signed-off-by: Roland McGrath <roland@redhat.com>
Signed-off-by: Ingo Molnar <mingo@elte.hu>
Glauber Costa