Commit Graph

57 Commits

Author SHA1 Message Date
Peter Maydell fadc1cbe85 Add MemTxAttrs to the IOTLB
Add a MemTxAttrs field to the IOTLB, and allow target-specific
code to set it via a new tlb_set_page_with_attrs() function;
pass the attributes through to the device when making IO accesses.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
2015-04-26 16:49:24 +01:00
Peter Maydell e469b22ffd Make CPU iotlb a structure rather than a plain hwaddr
Make the CPU iotlb a structure rather than a plain hwaddr;
this will allow us to add transaction attributes to it.

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
2015-04-26 16:49:23 +01:00
Peter Maydell 3b64349539 memory: Replace io_mem_read/write with memory_region_dispatch_read/write
Rather than retaining io_mem_read/write as simple wrappers around
the memory_region_dispatch_read/write functions, make the latter
public and change all the callers to use them, since we need to
touch all the callsites anyway to add MemTxAttrs and MemTxResult
support. Delete io_mem_read and io_mem_write entirely.

(All the callers currently pass MEMTXATTRS_UNSPECIFIED
and convert the return value back to bool or ignore it.)

Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
2015-04-26 16:49:23 +01:00
Paolo Bonzini 9d82b5a792 exec: make iotlb RCU-friendly
After the previous patch, TLBs will be flushed on every change to
the memory mapping.  This patch augments that with synchronization
of the MemoryRegionSections referred to in the iotlb array.

With this change, it is guaranteed that iotlb_to_region will access
the correct memory map, even once the TLB will be accessed outside
the BQL.

Reviewed-by: Fam Zheng <famz@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2015-02-16 17:30:19 +01:00
Leon Alrae 55e9409366 softmmu: provide softmmu access type enum
New MIPS features depend on the access type and enum is more convenient than
using the numbers directly.

Signed-off-by: Leon Alrae <leon.alrae@imgtec.com>
Reviewed-by: Thomas Huth <thuth@linux.vnet.ibm.com>
2014-11-03 11:48:34 +00:00
Xin Tong 88e89a57f9 implementing victim TLB for QEMU system emulated TLB
QEMU system mode page table walks are expensive. Taken by running QEMU
qemu-system-x86_64 system mode on Intel PIN , a TLB miss and walking a
4-level page tables in guest Linux OS takes ~450 X86 instructions on
average.

QEMU system mode TLB is implemented using a directly-mapped hashtable.
This structure suffers from conflict misses. Increasing the
associativity of the TLB may not be the solution to conflict misses as
all the ways may have to be walked in serial.

A victim TLB is a TLB used to hold translations evicted from the
primary TLB upon replacement. The victim TLB lies between the main TLB
and its refill path. Victim TLB is of greater associativity (fully
associative in this patch). It takes longer to lookup the victim TLB,
but its likely better than a full page table walk. The memory
translation path is changed as follows :

Before Victim TLB:
1. Inline TLB lookup
2. Exit code cache on TLB miss.
3. Check for unaligned, IO accesses
4. TLB refill.
5. Do the memory access.
6. Return to code cache.

After Victim TLB:
1. Inline TLB lookup
2. Exit code cache on TLB miss.
3. Check for unaligned, IO accesses
4. Victim TLB lookup.
5. If victim TLB misses, TLB refill
6. Do the memory access.
7. Return to code cache

The advantage is that victim TLB can offer more associativity to a
directly mapped TLB and thus potentially fewer page table walks while
still keeping the time taken to flush within reasonable limits.
However, placing a victim TLB before the refill path increase TLB
refill path as the victim TLB is consulted before the TLB refill. The
performance results demonstrate that the pros outweigh the cons.

some performance results taken on SPECINT2006 train
datasets and kernel boot and qemu configure script on an
Intel(R) Xeon(R) CPU  E5620  @ 2.40GHz Linux machine are shown in the
Google Doc link below.

https://docs.google.com/spreadsheets/d/1eiItzekZwNQOal_h-5iJmC4tMDi051m9qidi5_nwvH4/edit?usp=sharing

In summary, victim TLB improves the performance of qemu-system-x86_64 by
11% on average on SPECINT2006, kernelboot and qemu configscript and with
highest improvement of in 26% in 456.hmmer. And victim TLB does not result
in any performance degradation in any of the measured benchmarks. Furthermore,
the implemented victim TLB is architecture independent and is expected to
benefit other architectures in QEMU as well.

Although there are measurement fluctuations, the performance
improvement is very significant and by no means in the range of
noises.

Signed-off-by: Xin Tong <trent.tong@gmail.com>
Message-id: 1407202523-23553-1-git-send-email-trent.tong@gmail.com
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
2014-09-01 17:43:06 +01:00
Paolo Bonzini 58ed270df9 softmmu: move softmmu_template.h out of include/
It is only included in cputlb.c now.

Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2014-06-05 16:10:33 +02:00
Paolo Bonzini 022c62cbbc exec: move include files to include/exec/
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
2012-12-19 08:31:31 +01:00
Yeongkyoon Lee fdbb84d133 tcg: Add extended GETPC mechanism for MMU helpers with ldst optimization
Add GETPC_EXT which is used by MMU helpers to selectively calculate the code
address of accessing guest memory when called from a qemu_ld/st optimized code
or a C function. Currently, it supports only i386 and x86-64 hosts.

Signed-off-by: Yeongkyoon Lee <yeongkyoon.lee@samsung.com>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2012-11-03 09:44:20 +00:00
Avi Kivity a8170e5e97 Rename target_phys_addr_t to hwaddr
target_phys_addr_t is unwieldly, violates the C standard (_t suffixes are
reserved) and its purpose doesn't match the name (most target_phys_addr_t
addresses are not target specific).  Replace it with a finger-friendly,
standards conformant hwaddr.

Outstanding patchsets can be fixed up with the command

  git rebase -i --exec 'find -name "*.[ch]"
                        | xargs s/target_phys_addr_t/hwaddr/g' origin

Signed-off-by: Avi Kivity <avi@redhat.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2012-10-23 08:58:25 -05:00
Blue Swirl 89c33337fd Remove unused CONFIG_TCG_PASS_AREG0 and dead code
Now that CONFIG_TCG_PASS_AREG0 is enabled for all targets,
remove dead code and support for !CONFIG_TCG_PASS_AREG0 case.

Remove dyngen-exec.h and all references to it. Although included by
hw/spapr_hcall.c, it does not seem to use it.

Remove unused HELPER_CFLAGS.

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
2012-09-15 17:51:14 +00:00
Stefan Weil b065927a02 w64: Fix data types in softmmu*.h
w64 requires uintptr_t.

Signed-off-by: Stefan Weil <sw@weilnetz.de>
2012-04-15 21:25:17 +02:00
Blue Swirl 2050396801 Use uintptr_t for various op related functions
Use uintptr_t instead of void * or unsigned long in
several op related functions, env->mem_io_pc and
GETPC() macro.

Reviewed-by: Stefan Weil <sw@weilnetz.de>
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2012-04-14 14:23:37 +00:00
Blue Swirl e141ab52d2 softmmu templates: optionally pass CPUState to memory access functions
Optionally, make memory access helpers take a parameter for CPUState
instead of relying on global env.

On most targets, perform simple moves to reorder registers. On i386,
switch from regparm(3) calling convention to standard stack-based
version.

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2012-03-18 12:21:52 +00:00
Blue Swirl 6a18ae2d29 i386: Remove REGPARM
Use stack based calling convention (GCC default) for interfacing with
generated code instead of register based convention (regparm(3)).

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2012-03-18 12:21:48 +00:00
Avi Kivity 37ec01d433 memory: dispatch directly via MemoryRegion
Instead of indirecting via io_mem_region, dispatch directly
through the MemoryRegion obtained from the iotlb or phys_page_find().

Signed-off-by: Avi Kivity <avi@redhat.com>
2012-03-08 19:06:11 +02:00
Avi Kivity aa102231f0 memory: store section indices in iotlb instead of io indices
A step towards eliminating io indices.

Signed-off-by: Avi Kivity <avi@redhat.com>
2012-03-08 17:06:55 +02:00
Avi Kivity 11c7ef0c73 Remove IO_MEM_SHIFT
We no longer use any of the lower bits of a ram_addr, so we might as well
use them for the io table index.  This increases the number of potential
I/O handlers by a factor of 8.

Signed-off-by: Avi Kivity <avi@redhat.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
2012-01-04 13:34:50 +02:00
Avi Kivity 0e0df1e24d Convert IO_MEM_{RAM,ROM,UNASSIGNED,NOTDIRTY} to MemoryRegions
Convert the fixed-address IO_MEM_RAM, IO_MEM_ROM, IO_MEM_UNASSIGNED,
and IO_MEM_NOTDIRTY io handlers to MemoryRegions.  These aren't real
regions, since they are never added to the memory hierarchy, but they
allow reuse of the dispatch functionality.

Signed-off-by: Avi Kivity <avi@redhat.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
2012-01-04 13:34:50 +02:00
Avi Kivity 1d393fa2d1 Avoid range comparisons on io index types
The code sometimes uses range comparisons on io indexes (e.g.
index =< IO_MEM_ROM).  Avoid these as they make moving to objects harder.

Signed-off-by: Avi Kivity <avi@redhat.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
2012-01-04 13:34:49 +02:00
Avi Kivity acbbec5d43 memory: move mmio access to functions
Currently mmio access goes directly to the io_mem_{read,write} arrays.
In preparation for eliminating them, add indirection via a function.

Signed-off-by: Avi Kivity <avi@redhat.com>
Reviewed-by: Richard Henderson <rth@twiddle.net>
2012-01-04 13:34:49 +02:00
Blue Swirl bccd9ec5f0 softmmu_header: pass CPUState to tlb_fill
Pass CPUState pointer to tlb_fill() instead of architecture local
cpu_single_env hacks.

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2011-10-01 09:31:26 +00:00
Blue Swirl efbf29b681 Document softmmu templates
Add some comments to describe each file.

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2011-10-01 09:31:08 +00:00
Paul Brook 355b194369 Split TLB addend and target_phys_addr_t
Historically the qemu tlb "addend" field was used for both RAM and IO accesses,
so needed to be able to hold both host addresses (unsigned long) and guest
physical addresses (target_phys_addr_t).  However since the introduction of
the iotlb field it has only been used for RAM accesses.

This means we can change the type of addend to unsigned long, and remove
associated hacks in the big-endian TCG backends.

We can also remove the host dependence from target_phys_addr_t.

Signed-off-by: Paul Brook <paul@codesourcery.com>
2010-04-05 00:28:53 +01:00
Blue Swirl 29e922b61f Compile qemu-timer only once
Arrange various declarations so that also non-CPU code can access
them, adjust users.

Move CPU specific code to cpus.c.

Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2010-03-29 19:24:00 +00:00
Edgar E. Iglesias 794401ca9c softmmu: Dont clobber retaddr in slow_ldx().
When splitting up unaligned IO accesses, ld calls slow_ld which was
clobbering retaddr.

AFAIK the problem only shows up when running emulations with -icount
that may abort TB execution on IO accesses.

Signed-off-by: Edgar E. Iglesias <edgar.iglesias@gmail.com>
2010-01-28 22:46:13 +01:00
Anthony Liguori c227f0995e Revert "Get rid of _t suffix"
In the very least, a change like this requires discussion on the list.

The naming convention is goofy and it causes a massive merge problem.  Something
like this _must_ be presented on the list first so people can provide input
and cope with it.

This reverts commit 99a0949b72.

Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2009-10-01 16:12:16 -05:00
malc 99a0949b72 Get rid of _t suffix
Some not so obvious bits, slirp and Xen were left alone for the time
being.

Signed-off-by: malc <av1474@comtv.ru>
2009-10-01 22:45:02 +04:00
Anthony Liguori 4a1418e07b Unbreak large mem support by removing kqemu
kqemu introduces a number of restrictions on the i386 target.  The worst is that
it prevents large memory from working in the default build.

Furthermore, kqemu is fundamentally flawed in a number of ways.  It relies on
the TSC as a time source which will not be reliable on a multiple processor
system in userspace.  Since most modern processors are multicore, this severely
limits the utility of kqemu.

kvm is a viable alternative for people looking to accelerate qemu and has the
benefit of being supported by the upstream Linux kernel.  If someone can
implement work arounds to remove the restrictions introduced by kqemu, I'm
happy to avoid and/or revert this patch.

N.B. kqemu will still function in the 0.11 series but this patch removes it from
the 0.12 series.

Paul, please Ack or Nack this patch.

Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>
2009-08-24 08:02:55 -05:00
Blue Swirl 8167ee8839 Update to a hopefully more future proof FSF address
Signed-off-by: Blue Swirl <blauwirbel@gmail.com>
2009-07-16 20:47:01 +00:00
blueswir1 640f42e4e9 kqemu: merge CONFIG_KQEMU and USE_KQEMU
Basically a recursive ":%s/USE_KQEMU/CONFIG_KQEMU/g".

Signed-off-by: Paul Bolle <pebolle@tiscali.nl>



git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@7189 c046a42c-6fe2-441c-8c8c-71466251a162
2009-04-19 10:18:01 +00:00
aurel32 fad6cb1a56 Update FSF address in GPL/LGPL boilerplate
The attached patch updates the FSF address in the GPL/LGPL boilerplate
in most GPL/LGPLed files, and also in COPYING.LIB.

Signed-off-by: Stuart Brady <stuart.brady@gmail.com>
Signed-off-by: Aurelien Jarno <aurelien@aurel32.net>

git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@6162 c046a42c-6fe2-441c-8c8c-71466251a162
2009-01-04 22:05:52 +00:00
aliguori db8886d3fd Set mem_io_vaddr on io_read (Jan Kiszka)
Analogously to write accesses, we have to save the memory address also
on read accesses in order to support read watchpoints.

Signed-off-by: Jan Kiszka <jan.kiszka@siemens.com>
Signed-off-by: Anthony Liguori <aliguori@us.ibm.com>



git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@5739 c046a42c-6fe2-441c-8c8c-71466251a162
2008-11-18 20:09:43 +00:00
pbrook 2e70f6efa8 Add instruction counter.
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@4799 c046a42c-6fe2-441c-8c8c-71466251a162
2008-06-29 01:03:05 +00:00
pbrook 0f459d16c3 Clean up MMIO TLB handling.
The IO index is now stored in its own field, instead of being wedged
into the vaddr field.  This eliminates the ROMD and watchpoint host
pointer weirdness.  The IO index space is expanded by 1 bit, and
several additional bits are made available in the TLB vaddr field.


git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@4704 c046a42c-6fe2-441c-8c8c-71466251a162
2008-06-09 00:20:13 +00:00
bellard d656469f44 use simpler REGPARM convention - make CPUTLBEntry size a power of two
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@3935 c046a42c-6fe2-441c-8c8c-71466251a162
2008-01-31 09:22:27 +00:00
balrog 6c41b2723f Don't compare '\0' against pointers.
Add a note from Fabrice in slow_st template.


git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@3669 c046a42c-6fe2-441c-8c8c-71466251a162
2007-11-17 12:12:29 +00:00
balrog 7221fa98d3 Check permissions for the last byte first in unaligned slow_st accesses (patch from TeLeMan).
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@3665 c046a42c-6fe2-441c-8c8c-71466251a162
2007-11-17 09:53:42 +00:00
j_mayer 6ebbf39000 Replace is_user variable with mmu_idx in softmmu core,
allowing support of more than 2 mmu access modes.
Add backward compatibility is_user variable in targets code when needed.
Implement per target cpu_mmu_index function, avoiding duplicated code
  and #ifdef TARGET_xxx in softmmu core functions.
Implement per target mmu modes definitions. As an example, add PowerPC
  hypervisor mode definition and Alpha executive and kernel modes definitions.
Optimize PowerPC case, precomputing mmu_idx when MSR register changes
  and using the same definition in code translation code.


git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@3384 c046a42c-6fe2-441c-8c8c-71466251a162
2007-10-14 07:07:08 +00:00
ths 3b46e62427 find -type f | xargs sed -i 's/[\t ]*$//g' # Yes, again. Note the star in the regex.
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@3177 c046a42c-6fe2-441c-8c8c-71466251a162
2007-09-17 08:09:54 +00:00
ths 5fafdf24ef find -type f | xargs sed -i 's/[\t ]$//g' # on most files
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@3173 c046a42c-6fe2-441c-8c8c-71466251a162
2007-09-16 21:08:06 +00:00
bellard f1c85677fc added last_io_time field
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@1752 c046a42c-6fe2-441c-8c8c-71466251a162
2006-02-08 22:41:53 +00:00
bellard a64d4718f1 MIPS unaligned accesses exceptions (Daniel Jacobowitz)
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@1688 c046a42c-6fe2-441c-8c8c-71466251a162
2005-12-05 19:57:57 +00:00
bellard 84b7b8e778 PAGE_EXEC support in TLBs
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@1676 c046a42c-6fe2-441c-8c8c-71466251a162
2005-11-28 21:19:04 +00:00
bellard 98699967b8 use TARGET_PAGE_SIZE (Paul Brook)
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@1659 c046a42c-6fe2-441c-8c8c-71466251a162
2005-11-26 10:29:22 +00:00
bellard 108c49b8a2 allow more than 32 bit of physical memory
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@1526 c046a42c-6fe2-441c-8c8c-71466251a162
2005-07-24 12:55:09 +00:00
bellard c27004ec78 64 bit target support
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@1189 c046a42c-6fe2-441c-8c8c-71466251a162
2005-01-03 23:35:10 +00:00
bellard b769d8fef6 removed access_type hack
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@1095 c046a42c-6fe2-441c-8c8c-71466251a162
2004-10-03 15:07:13 +00:00
bellard a4193c8a4b support for opaque data on memory I/Os
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@874 c046a42c-6fe2-441c-8c8c-71466251a162
2004-06-03 14:01:43 +00:00
bellard d720b93d0b precise self modifying code support
git-svn-id: svn://svn.savannah.nongnu.org/qemu/trunk@745 c046a42c-6fe2-441c-8c8c-71466251a162
2004-04-25 17:57:43 +00:00