move away TCG-only code, make it compile only on TCG.
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
[claudio: moved the prototypes from hw/core/cpu.h to exec/cpu-all.h]
Signed-off-by: Claudio Fontana <cfontana@suse.de>
Message-Id: <20210204163931.7358-4-cfontana@suse.de>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Pages can't be both write and executable at the same time on Apple
Silicon. macOS provides public API to switch write protection [1] for
JIT applications, like TCG.
1. https://developer.apple.com/documentation/apple_silicon/porting_just-in-time_compilers_to_apple_silicon
Tested-by: Alexander Graf <agraf@csgraf.de>
Signed-off-by: Roman Bolshakov <r.bolshakov@yadro.com>
Message-Id: <20210113032806.18220-1-r.bolshakov@yadro.com>
[rth: Inline the qemu_thread_jit_* functions;
drop the MAP_JIT change for a follow-on patch.]
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
tb_gen_code() is only called within TCG accelerator, declare it locally.
Signed-off-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Message-Id: <20210117164813.4101761-4-f4bug@amsat.org>
[rth: Adjust vs changed tb_flush_jmp_cache patch.]
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The cpu_exec_step_atomic() function is called with the cpu->running
clear and proceeds to run target code without setting this flag. If
this target code generates an exception then handle_cpu_signal() will
unnecessarily abort. For example if atomic code generates a memory
protection fault.
This patch at least sets and clears this running flag, and adds some
assertions to help detect other cases.
Signed-off-by: Douglas Crosher <dtc-ubuntu@scieneer.com>
Message-Id: <a272c656-f7c5-019d-1cc0-499b8f80f2fc@scieneer.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This produces a small pc-relative displacement within the
generated code to the TB structure that preceeds it.
Reviewed-by: Joelle van Dyne <j@getutm.app>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Pass both rx and rw addresses to tb_target_set_jmp_target.
Reviewed-by: Joelle van Dyne <j@getutm.app>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Add two helper functions, using a global variable to hold
the displacement. The displacement is currently always 0,
so no change in behaviour.
Begin using the functions in tcg common code only.
Reviewed-by: Joelle van Dyne <j@getutm.app>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
LLVM/Clang, supports runtime checks for forward-edge Control-Flow
Integrity (CFI).
CFI on indirect function calls (cfi-icall) ensures that, in indirect
function calls, the function called is of the right signature for the
pointer type defined at compile time.
For this check to work, the code must always respect the function
signature when using function pointer, the function must be defined
at compile time, and be compiled with link-time optimization.
This rules out, for example, shared libraries that are dynamically loaded
(given that functions are not known at compile time), and code that is
dynamically generated at run-time.
This patch:
1) Introduces the CONFIG_CFI flag to support cfi in QEMU
2) Introduces a decorator to allow the definition of "sensitive"
functions, where a non-instrumented function may be called at runtime
through a pointer. The decorator will take care of disabling cfi-icall
checks on such functions, when cfi is enabled.
3) Marks functions currently in QEMU that exhibit such behavior,
in particular:
- The function in TCG that calls pre-compiled TBs
- The function in TCI that interprets instructions
- Functions in the plugin infrastructures that jump to callbacks
- Functions in util that directly call a signal handler
Signed-off-by: Daniele Buono <dbuono@linux.vnet.ibm.com>
Acked-by: Alex Bennée <alex.bennee@linaro.org
Message-Id: <20201204230615.2392-3-dbuono@linux.vnet.ibm.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
This will let us simplify the code that initializes CPU class
methods, when we move cpu_exec_*() to a separate struct.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Claudio Fontana <cfontana@suse.de>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20201212155530.23098-11-cfontana@suse.de>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Move invocation of CPUClass.cpu_exec_*() to separate helpers,
to make it easier to refactor that code later.
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Signed-off-by: Claudio Fontana <cfontana@suse.de>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20201212155530.23098-10-cfontana@suse.de>
Signed-off-by: Eduardo Habkost <ehabkost@redhat.com>
Since commit efc6c070ac ("configure: Add a test for the
minimum compiler version") the minimum compiler version
required for GCC is 4.8.
We can safely remove the special case for GCC 4.6 introduced
in commit 0448f5f8b8 ("cpu-exec: Fix compiler warning
(-Werror=clobbered)").
No change for Clang as we don't know.
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Marc-André Lureau <marcandre.lureau@redhat.com>
Message-Id: <20201210134752.780923-3-marcandre.lureau@redhat.com>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
cpu-exec tries to execute TB without caching when current
icount budget is over. But sometimes refilled budget is big
enough to try executing cached blocks.
This patch checks that instruction budget is big enough
for next block execution instead of just running cpu_exec_nocache.
It halves the number of calls of cpu_exec_nocache function
during tested OS boot scenario.
Signed-off-by: Pavel Dovgalyuk <pavel.dovgalyuk@ispras.ru>
Message-Id: <160741865825.348476.7169239332367828943.stgit@pasha-ThinkPad-X280>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Interrupt poll is not a real interrupt event. It is needed only for
thread safety. This interrupt is used for i386 and converted
to hardware interrupt by cpu_handle_interrupt function.
Therefore it is not needed to be recorded, because hardware
interrupt will be recorded after converting.
Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgalyuk@ispras.ru>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
--
v4 changes:
- Condition check refactoring (suggested by Alex Bennée)
Message-Id: <160174517124.12451.12983410242461131737.stgit@pasha-ThinkPad-X280>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
refactoring of cpus.c continues with cpu timer state extraction.
cpu-timers: responsible for the softmmu cpu timers state,
including cpu clocks and ticks.
icount: counts the TCG instructions executed. As such it is specific to
the TCG accelerator. Therefore, it is built only under CONFIG_TCG.
One complication is due to qtest, which uses an icount field to warp time
as part of qtest (qtest_clock_warp).
In order to solve this problem, provide a separate counter for qtest.
This requires fixing assumptions scattered in the code that
qtest_enabled() implies icount_enabled(), checking each specific case.
Signed-off-by: Claudio Fontana <cfontana@suse.de>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
[remove redundant initialization with qemu_spice_init]
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
[fix lingering calls to icount_get]
Signed-off-by: Claudio Fontana <cfontana@suse.de>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
clang's C11 atomic_fetch_*() functions only take a C11 atomic type
pointer argument. QEMU uses direct types (int, etc) and this causes a
compiler error when a QEMU code calls these functions in a source file
that also included <stdatomic.h> via a system header file:
$ CC=clang CXX=clang++ ./configure ... && make
../util/async.c:79:17: error: address argument to atomic operation must be a pointer to _Atomic type ('unsigned int *' invalid)
Avoid using atomic_*() names in QEMU's atomic.h since that namespace is
used by <stdatomic.h>. Prefix QEMU's APIs with 'q' so that atomic.h
and <stdatomic.h> can co-exist. I checked /usr/include on my machine and
searched GitHub for existing "qatomic_" users but there seem to be none.
This patch was generated using:
$ git grep -h -o '\<atomic\(64\)\?_[a-z0-9_]\+' include/qemu/atomic.h | \
sort -u >/tmp/changed_identifiers
$ for identifier in $(</tmp/changed_identifiers); do
sed -i "s%\<$identifier\>%q$identifier%g" \
$(git grep -I -l "\<$identifier\>")
done
I manually fixed line-wrap issues and misaligned rST tables.
Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Acked-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20200923105646.47864-1-stefanha@redhat.com>
When single-stepping with a debugger attached to QEMU, and when an
interrupt is raised, the debugger misses the first instruction after
the interrupt.
Tested-by: Luc Michel <luc.michel@greensocs.com>
Reviewed-by: Luc Michel <luc.michel@greensocs.com>
Buglink: https://bugs.launchpad.net/qemu/+bug/757702
Message-Id: <20200717163029.2737546-1-richard.henderson@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
When single-stepping with a debugger attached to QEMU, and when an
exception is raised, the debugger misses the first instruction after the
exception:
$ qemu-system-aarch64 -M virt -display none -cpu cortex-a53 -s -S
$ aarch64-linux-gnu-gdb
GNU gdb (GDB) 9.2
[...]
(gdb) tar rem :1234
Remote debugging using :1234
warning: No executable has been specified and target does not support
determining executable automatically. Try using the "file" command.
0x0000000000000000 in ?? ()
(gdb) # writing nop insns to 0x200 and 0x204
(gdb) set *0x200 = 0xd503201f
(gdb) set *0x204 = 0xd503201f
(gdb) # 0x0 address contains 0 which is an invalid opcode.
(gdb) # The CPU should raise an exception and jump to 0x200
(gdb) si
0x0000000000000204 in ?? ()
With this commit, the same run steps correctly on the first instruction
of the exception vector:
(gdb) si
0x0000000000000200 in ?? ()
Buglink: https://bugs.launchpad.net/qemu/+bug/757702
Signed-off-by: Luc Michel <luc.michel@greensocs.com>
Message-Id: <20200716193947.3058389-1-luc.michel@greensocs.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The bug describes a race whereby cpu_exec_step_atomic can acquire a TB
which is invalidated by a tb_flush before we execute it. This doesn't
affect the other cpu_exec modes as a tb_flush by it's nature can only
occur on a quiescent system. The race was described as:
B2. tcg_cpu_exec => cpu_exec => tb_find => tb_gen_code
B3. tcg_tb_alloc obtains a new TB
C3. TB obtained with tb_lookup__cpu_state or tb_gen_code
(same TB as B2)
A3. start_exclusive critical section entered
A4. do_tb_flush is called, TB memory freed/re-allocated
A5. end_exclusive exits critical section
B2. tcg_cpu_exec => cpu_exec => tb_find => tb_gen_code
B3. tcg_tb_alloc reallocates TB from B2
C4. start_exclusive critical section entered
C5. cpu_tb_exec executes the TB code that was free in A4
The simplest fix is to widen the exclusive period to include the TB
lookup. As a result we can drop the complication of checking we are in
the exclusive region before we end it.
Cc: Yifan <me@yifanlu.com>
Buglink: https://bugs.launchpad.net/qemu/+bug/1863025
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20200214144952.15502-1-alex.bennee@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
We currently search both the root and the tcg/ directories for tcg
files:
$ git grep '#include "tcg/' | wc -l
28
$ git grep '#include "tcg[^/]' | wc -l
94
To simplify the preprocessor search path, unify by expliciting the
tcg/ directory.
Patch created mechanically by running:
$ for x in \
tcg.h tcg-mo.h tcg-op.h tcg-opc.h \
tcg-op-gvec.h tcg-gvec-desc.h; do \
sed -i "s,#include \"$x\",#include \"tcg/$x\"," \
$(git grep -l "#include \"$x\""); \
done
Acked-by: David Gibson <david@gibson.dropbear.id.au> (ppc parts)
Reviewed-by: Paolo Bonzini <pbonzini@redhat.com>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Stefan Weil <sw@weilnetz.de>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@redhat.com>
Message-Id: <20200101112303.20724-2-philmd@redhat.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
qemu_log_lock() now returns a handle and qemu_log_unlock() receives a
handle to unlock. This allows for changing the handle during logging
and ensures the lock() and unlock() are for the same file.
Also in target/tilegx/translate.c removed the qemu_log_lock()/unlock()
calls (and the log("\n")), since the translator can longjmp out of the
loop if it attempts to translate an instruction in an inaccessible page.
Signed-off-by: Robert Foley <robert.foley@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20191118211528.3221-5-robert.foley@linaro.org>
To capture all memory accesses we need hook into all the various
helper functions that are involved in memory operations as well as the
injected inline helper calls. A later commit will allow us to resolve
the actual guest HW addresses by replaying the lookup.
Signed-off-by: Emilio G. Cota <cota@braap.org>
[AJB: drop haddr handling, just deal in vaddr]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Suggested-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
[AJB: moved inside start/end_exclusive fns + cleanup]
Signed-off-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Most of IO instructions can be executed only at the end of the block in
icount mode. Therefore translator can set cpu_can_io flag when translating
the last instruction.
But when the blocks are chained, then this flag is not reset and may
remain set at the beginning of the next block.
This patch resets the flag at the entry of any translation block,
making I/O operations impossible by default.
Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgaluk@ispras.ru>
--
v2 changes:
- reset can_do_io at the start of every TB (suggested by Paolo Bonzini)
Message-Id: <156404428943.18669.15747009371169578935.stgit@pasha-Precision-3630-Tower>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
No header includes qemu-common.h after this commit, as prescribed by
qemu-common.h's file comment.
Signed-off-by: Markus Armbruster <armbru@redhat.com>
Message-Id: <20190523143508.25387-5-armbru@redhat.com>
[Rebased with conflicts resolved automatically, except for
include/hw/arm/xlnx-zynqmp.h hw/arm/nrf51_soc.c hw/arm/msf2-soc.c
block/qcow2-refcount.c block/qcow2-cluster.c block/qcow2-cache.c
target/arm/cpu.h target/lm32/cpu.h target/m68k/cpu.h target/mips/cpu.h
target/moxie/cpu.h target/nios2/cpu.h target/openrisc/cpu.h
target/riscv/cpu.h target/tilegx/cpu.h target/tricore/cpu.h
target/unicore32/cpu.h target/xtensa/cpu.h; bsd-user/main.c and
net/tap-bsd.c fixed up]
Amusingly, we had already ignored the comment to keep this value
at the end of CPUState. This restores the minimum negative offset
from TCG_AREG0 for code generation.
For the couple of uses within qom/cpu.c, without NEED_CPU_H, add
a pointer from the CPUState object to the IcountDecr object within
CPUNegativeOffsetState.
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
In commit f7b78602fd we added the CPU cluster number to the
cflags field of the TB hash; this included adding it to the value
kept in tb->cflags, since we pass that field directly into the hash
calculation in some places. Unfortunately we forgot to check whether
other parts of the code were doing comparisons against tb->cflags
that would need to be updated.
It turns out that there is exactly one such place: the
tb_lookup__cpu_state() function checks whether the TB it has
found in the tb_jmp_cache has a tb->cflags matching the cf_mask
that is passed in. The tb->cflags has the cluster_index in it
but the cf_mask does not.
Hoist the "add cluster index to the cf_mask" code up from
tb_htable_lookup() to tb_lookup__cpu_state() so it can be considered
in the "did this TB match in the jmp cache" condition, as well as
when we do the full hash lookup by physical PC, flags, etc.
(tb_htable_lookup() is only called from tb_lookup__cpu_state(),
so this change doesn't require any further knock-on changes.)
Fixes: f7b78602fd ("accel/tcg: Add cluster number to TCG TB hash")
Tested-by: Cleber Rosa <crosa@redhat.com>
Tested-by: Mark Cave-Ayland <mark.cave-ayland@ilande.co.uk>
Reported-by: Howard Spoelstra <hsp.cat7@gmail.com>
Reported-by: Cleber Rosa <crosa@redhat.com>
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20190205151810.571-1-peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Just like we do in cpu_exec().
Reported-by: Max Filippov <jcmvbkbc@gmail.com>
Tested-by: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
We forgot to add this check in faa9372c07 ("translate-all:
introduce assert_no_pages_locked", 2018-06-15); we only added
it after returning from a longjmp in cpu_exec_step_atomic. Fix it.
Signed-off-by: Emilio G. Cota <cota@braap.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
It's either "GNU *Library* General Public version 2" or "GNU Lesser
General Public version *2.1*", but there was no "version 2.0" of the
"Lesser" library. So assume that version 2.1 is meant here.
Cc: Richard Henderson <rth@twiddle.net>
Signed-off-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <1548252536-6242-5-git-send-email-thuth@redhat.com>
Signed-off-by: Laurent Vivier <laurent@vivier.eu>
Include the cluster number in the hash we use to look
up TBs. This is important because a TB that is valid
for one cluster at a given physical address and set
of CPU flags is not necessarily valid for another:
the two clusters may have different views of physical
memory, or may have different CPU features (eg FPU
present or absent).
We put the cluster number in the high 8 bits of the
TB cflags. This gives us up to 256 clusters, which should
be enough for anybody. If we ever need more, or need
more bits in cflags for other purposes, we could make
tb_hash_func() take more data (and expand qemu_xxhash7()
to qemu_xxhash8()).
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alistair Francis <alistair.francis@wdc.com>
Reviewed-by: Edgar E. Iglesias <edgar.iglesias@xilinx.com>
Message-id: 20190121152218.9592-4-peter.maydell@linaro.org
Rather than test NOCHAIN before linking, do not emit the
goto_tb opcode at all. We already do this for goto_ptr.
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
When we support execution from non-RAM MMIO regions, get_page_addr_code()
will return -1 to indicate that there is no RAM at the requested address.
Handle this in the cpu-exec TB hashtable lookup code, treating it as
"no match found".
Note that the call to get_page_addr_code() in tb_lookup_cmp() needs
no changes -- a return of -1 will already correctly result in the
function returning false.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Emilio G. Cota <cota@braap.org>
Tested-by: Cédric Le Goater <clg@kaod.org>
Message-id: 20180710160013.26559-3-peter.maydell@linaro.org
Use mmap_lock in user-mode to protect TCG state and the page descriptors.
In !user-mode, each vCPU has its own TCG state, so no locks needed.
Per-page locks are used to protect the page descriptors.
Per-TB locks are used in both modes to protect TB jumps.
Some notes:
- tb_lock is removed from notdirty_mem_write by passing a
locked page_collection to tb_invalidate_phys_page_fast.
- tcg_tb_lookup/remove/insert/etc have their own internal lock(s),
so there is no need to further serialize access to them.
- do_tb_flush is run in a safe async context, meaning no other
vCPU threads are running. Therefore acquiring mmap_lock there
is just to please tools such as thread sanitizer.
- Not visible in the diff, but tb_invalidate_phys_page already
has an assert_memory_lock.
- cpu_io_recompile is !user-only, so no mmap_lock there.
- Added mmap_unlock()'s before all siglongjmp's that could
be called in user-mode while mmap_lock is held.
+ Added an assert for !have_mmap_lock() after returning from
the longjmp in cpu_exec, just like we do in cpu_exec_step_atomic.
Performance numbers before/after:
Host: AMD Opteron(tm) Processor 6376
ubuntu 17.04 ppc64 bootup+shutdown time
700 +-+--+----+------+------------+-----------+------------*--+-+
| + + + + + *B |
| before ***B*** ** * |
|tb lock removal ###D### *** |
600 +-+ *** +-+
| ** # |
| *B* #D |
| *** * ## |
500 +-+ *** ### +-+
| * *** ### |
| *B* # ## |
| ** * #D# |
400 +-+ ** ## +-+
| ** ### |
| ** ## |
| ** # ## |
300 +-+ * B* #D# +-+
| B *** ### |
| * ** #### |
| * *** ### |
200 +-+ B *B #D# +-+
| #B* * ## # |
| #* ## |
| + D##D# + + + + |
100 +-+--+----+------+------------+-----------+------------+--+-+
1 8 16 Guest CPUs 48 64
png: https://imgur.com/HwmBHXe
debian jessie aarch64 bootup+shutdown time
90 +-+--+-----+-----+------------+------------+------------+--+-+
| + + + + + + |
| before ***B*** B |
80 +tb lock removal ###D### **D +-+
| **### |
| **## |
70 +-+ ** # +-+
| ** ## |
| ** # |
60 +-+ *B ## +-+
| ** ## |
| *** #D |
50 +-+ *** ## +-+
| * ** ### |
| **B* ### |
40 +-+ **** # ## +-+
| **** #D# |
| ***B** ### |
30 +-+ B***B** #### +-+
| B * * # ### |
| B ###D# |
20 +-+ D ##D## +-+
| D# |
| + + + + + + |
10 +-+--+-----+-----+------------+------------+------------+--+-+
1 8 16 Guest CPUs 48 64
png: https://imgur.com/iGpGFtv
The gains are high for 4-8 CPUs. Beyond that point, however, unrelated
lock contention significantly hurts scalability.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This applies to both user-mode and !user-mode emulation.
Instead of relying on a global lock, protect the list of incoming
jumps with tb->jmp_lock. This lock also protects tb->cflags,
so update all tb->cflags readers outside tb->jmp_lock to use
atomic reads via tb_cflags().
In order to find the destination TB (and therefore its jmp_lock)
from the origin TB, we introduce tb->jmp_dest[].
I considered not using a linked list of jumps, which simplifies
code and makes the struct smaller. However, it unnecessarily increases
memory usage, which results in a performance decrease. See for
instance these numbers booting+shutting down debian-arm:
Time (s) Rel. err (%) Abs. err (s) Rel. slowdown (%)
------------------------------------------------------------------------------
before 20.88 0.74 0.154512 0.
after 20.81 0.38 0.079078 -0.33524904
GTree 21.02 0.28 0.058856 0.67049808
GHashTable + xxhash 21.63 1.08 0.233604 3.5919540
Using a hash table or a binary tree to keep track of the jumps
doesn't really pay off, not only due to the increased memory usage,
but also because most TBs have only 0 or 1 jumps to them. The maximum
number of jumps when booting debian-arm that I measured is 35, but
as we can see in the histogram below a TB with that many incoming jumps
is extremely rare; the average TB has 0.80 incoming jumps.
n_jumps: 379208; avg jumps/tb: 0.801099
dist: [0.0,1.0)|▄█▁▁▁▁▁▁▁▁▁▁▁ ▁▁▁▁▁▁ ▁▁▁ ▁▁▁ ▁|[34.0,35.0]
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Use the recently-gained QHT feature of returning the matching TB if it
already exists. This allows us to get rid of the lookup we perform
right after acquiring tb_lock.
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
The appended adds assertions to make sure we do not longjmp with page
locks held. Note that user-mode has nothing to check, since page_locks
are !user-mode only.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This paves the way for enabling scalable parallel generation of TCG code.
Instead of tracking TBs with a single binary search tree (BST), use a
BST for each TCG region, protecting it with a lock. This is as scalable
as it gets, since each TCG thread operates on a separate region.
The core of this change is the introduction of struct tcg_region_tree,
which contains a pointer to a GTree and an associated lock to serialize
accesses to it. We then allocate an array of tcg_region_tree's, adding
the appropriate padding to avoid false sharing based on
qemu_dcache_linesize.
Given a tc_ptr, we first find the corresponding region_tree. This
is done by special-casing the first and last regions first, since they
might be of size != region.size; otherwise we just divide the offset
by region.stride. I was worried about this division (several dozen
cycles of latency), but profiling shows that this is not a fast path.
Note that region.stride is not required to be a power of two; it
is only required to be a multiple of the host's page size.
Note that with this design we can also provide consistent snapshots
about all region trees at once; for instance, tcg_tb_foreach
acquires/releases all region_tree locks before/after iterating over them.
For this reason we now drop tb_lock in dump_exec_info().
As an alternative I considered implementing a concurrent BST, but this
can be tricky to get right, offers no consistent snapshots of the BST,
and performance and scalability-wise I don't think it could ever beat
having separate GTrees, given that our workload is insert-mostly (all
concurrent BST designs I've seen focus, understandably, on making
lookups fast, which comes at the expense of convoluted, non-wait-free
insertions/removals).
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
qht_lookup now uses the default cmp function. qht_lookup_custom is defined
to retain the old behaviour, that is a cmp function is explicitly provided.
qht_insert will gain use of the default cmp in the next patch.
Note that we move qht_lookup_custom's @func to be the last argument,
which makes the new qht_lookup as simple as possible.
Instead of this (i.e. keeping @func 2nd):
0000000000010750 <qht_lookup>:
10750: 89 d1 mov %edx,%ecx
10752: 48 89 f2 mov %rsi,%rdx
10755: 48 8b 77 08 mov 0x8(%rdi),%rsi
10759: e9 22 ff ff ff jmpq 10680 <qht_lookup_custom>
1075e: 66 90 xchg %ax,%ax
We get:
0000000000010740 <qht_lookup>:
10740: 48 8b 4f 08 mov 0x8(%rdi),%rcx
10744: e9 37 ff ff ff jmpq 10680 <qht_lookup_custom>
10749: 0f 1f 80 00 00 00 00 nopl 0x0(%rax)
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Usually the logging of the CPU state produced by -d cpu is sufficient
to diagnose problems, but sometimes you want to see the state of
the floating point registers as well. We don't want to enable that
by default as it adds a lot of extra data to the log; instead,
allow it to be optionally enabled via -d fpu.
Signed-off-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-id: 20180510130024.31678-1-peter.maydell@linaro.org
In icount mode, instructions that access io memory spaces in the middle
of the translation block invoke TB recompilation. After recompilation,
such instructions become last in the TB and are allowed to access io
memory spaces.
When the code includes instruction like i386 'xchg eax, 0xffffd080'
which accesses APIC, QEMU goes into an infinite loop of the recompilation.
This instruction includes two memory accesses - one read and one write.
After the first access, APIC calls cpu_report_tpr_access, which restores
the CPU state to get the current eip. But cpu_restore_state_from_tb
resets the cpu->can_do_io flag which makes the second memory access invalid.
Therefore the second memory access causes a recompilation of the block.
Then these operations repeat again and again.
This patch moves resetting cpu->can_do_io flag from
cpu_restore_state_from_tb to cpu_loop_exit* functions.
It also adds a parameter for cpu_restore_state which controls restoring
icount. There is no need to restore icount when we only query CPU state
without breaking the TB. Restoring it in such cases leads to the
incorrect flow of the virtual time.
In most cases new parameter is true (icount should be recalculated).
But there are two cases in i386 and openrisc when the CPU state is only
queried without the need to break the TB. This patch fixes both of
these cases.
Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgaluk@ispras.ru>
Message-Id: <20180409091320.12504.35329.stgit@pasha-VirtualBox>
[rth: Make can_do_io setting unconditional; move from cpu_exec;
make cpu_loop_exit_{noexc,restore} call cpu_loop_exit.]
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Function cpu_handle_interrupt calls cc->cpu_exec_interrupt to process
pending hardware interrupts. Under the hood cpu_exec_interrupt uses
cpu->exception_index to pass information to the internal function which
is usually common for exception and interrupt processing.
But this value is not reset after return and may be processed again
by cpu_handle_exception. This does not happen due to overwriting
the exception_index at the end of cpu_handle_interrupt.
But this branch may also overwrite the valid exception_index in some cases.
Therefore this patch:
1. resets exception_index just after the call to cpu_exec_interrupt
2. prevents overwriting the meaningful value of exception_index
Signed-off-by: Pavel Dovgalyuk <pavel.dovgaluk@ispras.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20180227095140.1060.61357.stgit@pasha-VirtualBox>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Signed-off-by: Pavel Dovgalyuk <Pavel.Dovgaluk@ispras.ru>
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
Message-Id: <20171217055023.29225-1-pbonzini@redhat.com>
[rth: Also change the Chain logging in helper_lookup_tb_ptr.]
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Claudio Fontana
Eduardo Habkost
Roman Bolshakov
Philippe Mathieu-Daudé
Douglas Crosher
Richard Henderson
Daniele Buono
Philippe Mathieu-Daudé
Pavel Dovgalyuk
Stefan Hajnoczi
Luc Michel
Alex Bennée
Robert Foley
Emilio G. Cota
Pavel Dovgalyuk
Markus Armbruster
Peter Maydell
Thomas Huth
Paolo Bonzini