During 'qemu-img compare', when we are checking that an allocated
portion of one file is all zeros, we don't need to waste time
computing how many additional sectors after the first non-zero
byte are also non-zero. Create a new helper find_nonzero() to do
the check for a first non-zero sector, and rebase
check_empty_sectors() to use it.
The new interface intentionally uses bytes in its interface, even
though it still crawls the buffer a sector at a time; it is robust
to a partial sector at the end of the buffer.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Compare the following images with all-zero contents:
$ truncate --size 1M A
$ qemu-img create -f qcow2 -o preallocation=off B 1G
$ qemu-img create -f qcow2 -o preallocation=metadata C 1G
On my machine, the difference is noticeable for pre-patch speeds,
with more than an order of magnitude in difference caused by the
choice of preallocation in the qcow2 file:
$ time ./qemu-img compare -f raw -F qcow2 A B
Warning: Image size mismatch!
Images are identical.
real 0m0.014s
user 0m0.007s
sys 0m0.007s
$ time ./qemu-img compare -f raw -F qcow2 A C
Warning: Image size mismatch!
Images are identical.
real 0m0.341s
user 0m0.144s
sys 0m0.188s
Why? Because bdrv_is_allocated() returns false for image B but
true for image C, throwing away the fact that both images know
via lseek(SEEK_HOLE) that the entire image still reads as zero.
From there, qemu-img ends up calling bdrv_pread() for every byte
of the tail, instead of quickly looking for the next allocation.
The solution: use block_status instead of is_allocated, giving:
$ time ./qemu-img compare -f raw -F qcow2 A C
Warning: Image size mismatch!
Images are identical.
real 0m0.014s
user 0m0.011s
sys 0m0.003s
which is on par with the speeds for no pre-allocation.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Reviewed-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
As long as we are querying the status for a chunk smaller than
the known image size, we are guaranteed that a successful return
will have set pnum to a non-zero size (pnum is zero only for
queries beyond the end of the file). Use that to slightly
simplify the calculation of the current chunk size being compared.
Likewise, we don't have to shrink the amount of data operated on
until we know we have to read the file, and therefore have to fit
in the bounds of our buffer. Also, note that 'total_sectors_over'
is equivalent to 'progress_base'.
With these changes in place, sectors_to_process() is now dead code,
and can be removed.
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are gradually moving away from sector-based interfaces, towards
byte-based. In the common case, allocation is unlikely to ever use
values that are not naturally sector-aligned, but it is possible
that byte-based values will let us be more precise about allocation
at the end of an unaligned file that can do byte-based access.
Changing the name of the function from bdrv_get_block_status_above()
to bdrv_block_status_above() ensures that the compiler enforces that
all callers are updated. Likewise, since it a byte interface allows
an offset mapping that might not be sector aligned, split the mapping
out of the return value and into a pass-by-reference parameter. For
now, the io.c layer still assert()s that all uses are sector-aligned,
but that can be relaxed when a later patch implements byte-based
block status in the drivers.
For the most part this patch is just the addition of scaling at the
callers followed by inverse scaling at bdrv_block_status(), plus
updates for the new split return interface. But some code,
particularly bdrv_block_status(), gets a lot simpler because it no
longer has to mess with sectors. Likewise, mirror code no longer
computes s->granularity >> BDRV_SECTOR_BITS, and can therefore drop
an assertion about alignment because the loop no longer depends on
alignment (never mind that we don't really have a driver that
reports sub-sector alignments, so it's not really possible to test
the effect of sub-sector mirroring). Fix a neighboring assertion to
use is_power_of_2 while there.
For ease of review, bdrv_get_block_status() was tackled separately.
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are gradually converting to byte-based interfaces, as they are
easier to reason about than sector-based. Convert another internal
type (no semantic change), and rename it to match the corresponding
public function rename.
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are gradually converting to byte-based interfaces, as they are
easier to reason about than sector-based. Convert another internal
function (no semantic change).
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are gradually converting to byte-based interfaces, as they are
easier to reason about than sector-based. Convert another internal
type (no semantic change), and rename it to match the corresponding
public function rename.
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are gradually converting to byte-based interfaces, as they are
easier to reason about than sector-based. Convert another internal
function (no semantic change); and as with its public counterpart,
rename to bdrv_co_block_status() and split the offset return, to
make the compiler enforce that we catch all uses. For now, we
assert that callers and the return value still use aligned data,
but ultimately, this will be the function where we hand off to a
byte-based driver callback, and will eventually need to add logic
to ensure we round calls according to the driver's
request_alignment then touch up the result handed back to the
caller, to start permitting a caller to pass unaligned offsets.
Note that we are now prepared to accepts 'bytes' larger than INT_MAX;
this is okay as long as we clamp things internally before violating
any 32-bit limits, and makes no difference to how a client will
use the information (clients looping over the entire file must
already be prepared for consecutive calls to return the same status,
as drivers are already free to return shorter-than-maximal status
due to any other convenient split points, such as when the L2 table
crosses cluster boundaries in qcow2).
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are gradually moving away from sector-based interfaces, towards
byte-based. In the common case, allocation is unlikely to ever use
values that are not naturally sector-aligned, but it is possible
that byte-based values will let us be more precise about allocation
at the end of an unaligned file that can do byte-based access.
Changing the name of the function from bdrv_get_block_status() to
bdrv_block_status() ensures that the compiler enforces that all
callers are updated. For now, the io.c layer still assert()s that
all callers are sector-aligned, but that can be relaxed when a later
patch implements byte-based block status in the drivers.
There was an inherent limitation in returning the offset via the
return value: we only have room for BDRV_BLOCK_OFFSET_MASK bits, which
means an offset can only be mapped for sector-aligned queries (or,
if we declare that non-aligned input is at the same relative position
modulo 512 of the answer), so the new interface also changes things to
return the offset via output through a parameter by reference rather
than mashed into the return value. We'll have some glue code that
munges between the two styles until we finish converting all uses.
For the most part this patch is just the addition of scaling at the
callers followed by inverse scaling at bdrv_block_status(), coupled
with the tweak in calling convention. But some code, particularly
bdrv_is_allocated(), gets a lot simpler because it no longer has to
mess with sectors.
For ease of review, bdrv_get_block_status_above() will be tackled
separately.
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are gradually converting to byte-based interfaces, as they are
easier to reason about than sector-based. Continue by converting
an internal function (no semantic change), and simplifying its
caller accordingly.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are gradually converting to byte-based interfaces, as they are
easier to reason about than sector-based. Change the internal
loop iteration of zeroing a device to track by bytes instead of
sectors (although we are still guaranteed that we iterate by steps
that are sector-aligned).
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
We are gradually converting to byte-based interfaces, as they are
easier to reason about than sector-based. Convert another internal
function (no semantic change), and rename it to is_zero() in the
process.
Signed-off-by: Eric Blake <eblake@redhat.com>
Reviewed-by: Fam Zheng <famz@redhat.com>
Reviewed-by: John Snow <jsnow@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
In the process of converting sector-based interfaces to bytes,
I'm finding it easier to represent a byte count as a 64-bit
integer at the block layer (even if we are internally capped
by SIZE_MAX or even INT_MAX for individual transactions, it's
still nicer to not have to worry about truncation/overflow
issues on as many variables). Update the signature of
bdrv_round_to_clusters() to uniformly use int64_t, matching
the signature already chosen for bdrv_is_allocated and the
fact that off_t is also a signed type, then adjust clients
according to the required fallout (even where the result could
now exceed 32 bits, no client is directly assigning the result
into a 32-bit value without breaking things into a loop first).
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Not all callers care about which BDS owns the mapping for a given
range of the file, or where the zeroes lie within that mapping. In
particular, bdrv_is_allocated() cares more about finding the
largest run of allocated data from the guest perspective, whether
or not that data is consecutive from the host perspective, and
whether or not the data reads as zero. Therefore, doing subsequent
refinements such as checking how much of the format-layer
allocation also satisfies BDRV_BLOCK_ZERO at the protocol layer is
wasted work - in the best case, it just costs extra CPU cycles
during a single bdrv_is_allocated(), but in the worst case, it
results in a smaller *pnum, and forces callers to iterate through
more status probes when visiting the entire file for even more
extra CPU cycles.
This patch only optimizes the block layer (no behavior change when
want_zero is true, but skip unnecessary effort when it is false).
Then when subsequent patches tweak the driver callback to be
byte-based, we can also pass this hint through to the driver.
Tweak BdrvCoGetBlockStatusData to declare arguments in parameter
order, rather than mixing things up (minimizing padding is not
necessary here).
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Not all callers care about which BDS owns the mapping for a given
range of the file. This patch merely simplifies the callers by
consolidating the logic in the common call point, while guaranteeing
a non-NULL file to all the driver callbacks, for no semantic change.
The only caller that does not care about pnum is bdrv_is_allocated,
as invoked by vvfat; we can likewise add assertions that the rest
of the stack does not have to worry about a NULL pnum.
Furthermore, this will also set the stage for a future cleanup: when
a caller does not care about which BDS owns an offset, it would be
nice to allow the driver to optimize things to not have to return
BDRV_BLOCK_OFFSET_VALID in the first place. In the case of fragmented
allocation (for example, it's fairly easy to create a qcow2 image
where consecutive guest addresses are not at consecutive host
addresses), the current contract requires bdrv_get_block_status()
to clamp *pnum to the limit where host addresses are no longer
consecutive, but allowing a NULL file means that *pnum could be
set to the full length of known-allocated data.
Signed-off-by: Eric Blake <eblake@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
This changes test case 191 to include a backing image that has
backing_fmt set in the image file, but is referenced by node name in the
qemu command line.
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Reviewed-by: Eric Blake <eblake@redhat.com>
When referring to a backing file of an image via node name
bdrv_open_backing_file would add the 'driver' option to the option list
filling it with the backing format driver. This breaks construction of
the backing chain via -blockdev, as bdrv_open_inherit reports an error
if both 'reference' and 'options' are provided.
$ qemu-img create -f raw /tmp/backing.raw 64M
$ qemu-img create -f qcow2 -F raw -b /tmp/backing.raw /tmp/test.qcow2
$ qemu-system-x86_64 \
-blockdev driver=file,filename=/tmp/backing.raw,node-name=backing \
-blockdev driver=qcow2,file.driver=file,file.filename=/tmp/test.qcow2,node-name=root,backing=backing
qemu-system-x86_64: -blockdev driver=qcow2,file.driver=file,file.filename=/tmp/test.qcow2,node-name=root,backing=backing: Could not open backing file: Cannot reference an existing block device with additional options or a new filename
Signed-off-by: Peter Krempa <pkrempa@redhat.com>
Signed-off-by: Kevin Wolf <kwolf@redhat.com>
Two or more threads might race while invalidating the same TB. We currently
do not check for this at all despite taking tb_lock, which means we would
wrongly invalidate the same TB more than once. This bug has actually been
hit by users: I recently saw a report on IRC, although I have yet to see
the corresponding test case.
Fix this by using qht_remove as the synchronization point; if it fails,
that means the TB has already been invalidated, and therefore there
is nothing left to do in tb_phys_invalidate.
Note that this solution works now that we still have tb_lock, and will
continue working once we remove tb_lock.
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Message-Id: <1508445114-4717-1-git-send-email-cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This is identical for each target. So, move the initialization to
common code. Move the variable itself out of tcg_ctx and name it
cpu_env to minimize changes within targets.
This also means we can remove tcg_global_reg_new_{ptr,i32,i64},
since there are no longer global-register temps created by targets.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This enables parallel TCG code generation. However, we do not take
advantage of it yet since tb_lock is still held during tb_gen_code.
In user-mode we use a single TCG context; see the documentation
added to tcg_region_init for the rationale.
Note that targets do not need any conversion: targets initialize a
TCGContext (e.g. defining TCG globals), and after this initialization
has finished, the context is cloned by the vCPU threads, each of
them keeping a separate copy.
TCG threads claim one entry in tcg_ctxs[] by atomically increasing
n_tcg_ctxs. Do not be too annoyed by the subsequent atomic_read's
of that variable and tcg_ctxs; they are there just to play nice with
analysis tools such as thread sanitizer.
Note that we do not allocate an array of contexts (we allocate
an array of pointers instead) because when tcg_context_init
is called, we do not know yet how many contexts we'll use since
the bool behind qemu_tcg_mttcg_enabled() isn't set yet.
Previous patches folded some TCG globals into TCGContext. The non-const
globals remaining are only set at init time, i.e. before the TCG
threads are spawned. Here is a list of these set-at-init-time globals
under tcg/:
Only written by tcg_context_init:
- indirect_reg_alloc_order
- tcg_op_defs
Only written by tcg_target_init (called from tcg_context_init):
- tcg_target_available_regs
- tcg_target_call_clobber_regs
- arm: arm_arch, use_idiv_instructions
- i386: have_cmov, have_bmi1, have_bmi2, have_lzcnt,
have_movbe, have_popcnt
- mips: use_movnz_instructions, use_mips32_instructions,
use_mips32r2_instructions, got_sigill (tcg_target_detect_isa)
- ppc: have_isa_2_06, have_isa_3_00, tb_ret_addr
- s390: tb_ret_addr, s390_facilities
- sparc: qemu_ld_trampoline, qemu_st_trampoline (build_trampolines),
use_vis3_instructions
Only written by tcg_prologue_init:
- 'struct jit_code_entry one_entry'
- aarch64: tb_ret_addr
- arm: tb_ret_addr
- i386: tb_ret_addr, guest_base_flags
- ia64: tb_ret_addr
- mips: tb_ret_addr, bswap32_addr, bswap32u_addr, bswap64_addr
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This is groundwork for supporting multiple TCG contexts.
The naive solution here is to split code_gen_buffer statically
among the TCG threads; this however results in poor utilization
if translation needs are different across TCG threads.
What we do here is to add an extra layer of indirection, assigning
regions that act just like pages do in virtual memory allocation.
(BTW if you are wondering about the chosen naming, I did not want
to use blocks or pages because those are already heavily used in QEMU).
We use a global lock to serialize allocations as well as statistics
reporting (we now export the size of the used code_gen_buffer with
tcg_code_size()). Note that for the allocator we could just use
a counter and atomic_inc; however, that would complicate the gathering
of tcg_code_size()-like stats. So given that the region operations are
not a fast path, a lock seems the most reasonable choice.
The effectiveness of this approach is clear after seeing some numbers.
I used the bootup+shutdown of debian-arm with '-tb-size 80' as a benchmark.
Note that I'm evaluating this after enabling per-thread TCG (which
is done by a subsequent commit).
* -smp 1, 1 region (entire buffer):
qemu: flush code_size=83885014 nb_tbs=154739 avg_tb_size=357
qemu: flush code_size=83884902 nb_tbs=153136 avg_tb_size=363
qemu: flush code_size=83885014 nb_tbs=152777 avg_tb_size=364
qemu: flush code_size=83884950 nb_tbs=150057 avg_tb_size=373
qemu: flush code_size=83884998 nb_tbs=150234 avg_tb_size=373
qemu: flush code_size=83885014 nb_tbs=154009 avg_tb_size=360
qemu: flush code_size=83885014 nb_tbs=151007 avg_tb_size=370
qemu: flush code_size=83885014 nb_tbs=151816 avg_tb_size=367
That is, 8 flushes.
* -smp 8, 32 regions (80/32 MB per region) [i.e. this patch]:
qemu: flush code_size=76328008 nb_tbs=141040 avg_tb_size=356
qemu: flush code_size=75366534 nb_tbs=138000 avg_tb_size=361
qemu: flush code_size=76864546 nb_tbs=140653 avg_tb_size=361
qemu: flush code_size=76309084 nb_tbs=135945 avg_tb_size=375
qemu: flush code_size=74581856 nb_tbs=132909 avg_tb_size=375
qemu: flush code_size=73927256 nb_tbs=135616 avg_tb_size=360
qemu: flush code_size=78629426 nb_tbs=142896 avg_tb_size=365
qemu: flush code_size=76667052 nb_tbs=138508 avg_tb_size=368
Again, 8 flushes. Note how buffer utilization is not 100%, but it
is close. Smaller region sizes would yield higher utilization,
but we want region allocation to be rare (it acquires a lock), so
we do not want to go too small.
* -smp 8, static partitioning of 8 regions (10 MB per region):
qemu: flush code_size=21936504 nb_tbs=40570 avg_tb_size=354
qemu: flush code_size=11472174 nb_tbs=20633 avg_tb_size=370
qemu: flush code_size=11603976 nb_tbs=21059 avg_tb_size=365
qemu: flush code_size=23254872 nb_tbs=41243 avg_tb_size=377
qemu: flush code_size=28289496 nb_tbs=52057 avg_tb_size=358
qemu: flush code_size=43605160 nb_tbs=78896 avg_tb_size=367
qemu: flush code_size=45166552 nb_tbs=82158 avg_tb_size=364
qemu: flush code_size=63289640 nb_tbs=116494 avg_tb_size=358
qemu: flush code_size=51389960 nb_tbs=93937 avg_tb_size=362
qemu: flush code_size=59665928 nb_tbs=107063 avg_tb_size=372
qemu: flush code_size=38380824 nb_tbs=68597 avg_tb_size=374
qemu: flush code_size=44884568 nb_tbs=79901 avg_tb_size=376
qemu: flush code_size=50782632 nb_tbs=90681 avg_tb_size=374
qemu: flush code_size=39848888 nb_tbs=71433 avg_tb_size=372
qemu: flush code_size=64708840 nb_tbs=119052 avg_tb_size=359
qemu: flush code_size=49830008 nb_tbs=90992 avg_tb_size=362
qemu: flush code_size=68372408 nb_tbs=123442 avg_tb_size=368
qemu: flush code_size=33555560 nb_tbs=59514 avg_tb_size=378
qemu: flush code_size=44748344 nb_tbs=80974 avg_tb_size=367
qemu: flush code_size=37104248 nb_tbs=67609 avg_tb_size=364
That is, 20 flushes. Note how a static partitioning approach uses
the code buffer poorly, leading to many unnecessary flushes.
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 helpers require the address and size to be page-aligned, so
do that before calling them.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Groundwork for supporting multiple TCG contexts.
While at it, also allocate temps_used directly as a bitmap of the
required size, instead of using a bitmap of TCG_MAX_TEMPS via
TCGTempSet.
Performance-wise we lose about 1.12% in a translation-heavy workload
such as booting+shutting down debian-arm:
Performance counter stats for 'taskset -c 0 arm-softmmu/qemu-system-arm \
-machine type=virt -nographic -smp 1 -m 4096 \
-netdev user,id=unet,hostfwd=tcp::2222-:22 \
-device virtio-net-device,netdev=unet \
-drive file=die-on-boot.qcow2,id=myblock,index=0,if=none \
-device virtio-blk-device,drive=myblock \
-kernel kernel.img -append console=ttyAMA0 root=/dev/vda1 \
-name arm,debug-threads=on -smp 1' (10 runs):
exec time (s) Relative slowdown wrt original (%)
---------------------------------------------------------------
original 20.213321616 0.
tcg_malloc 20.441130078 1.1270214
TCGContext 20.477846517 1.3086662
g_malloc 20.780527895 2.8061013
The other two alternatives shown in the table are:
- TCGContext: embed temps[TCG_MAX_TEMPS] and TCGTempSet used_temps
in TCGContext. This is simple enough but it isn't faster than using
tcg_malloc; moreover, it wastes memory.
- g_malloc: allocate/deallocate both temps and used_temps every time
tcg_optimize is executed.
Suggested-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This is groundwork for supporting multiple TCG contexts.
To avoid scalability issues when profiling info is enabled, this patch
makes the profiling info counters distributed via the following changes:
1) Consolidate profile info into its own struct, TCGProfile, which
TCGContext also includes. Note that tcg_table_op_count is brought
into TCGProfile after dropping the tcg_ prefix.
2) Iterate over the TCG contexts in the system to obtain the total counts.
This change also requires updating the accessors to TCGProfile fields to
use atomic_read/set whenever there may be conflicting accesses (as defined
in C11) to them.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Groundwork for supporting multiple TCG contexts.
Note that having n_tcg_ctxs is unnecessary. However, it is
convenient to have it, since it will simplify iterating over the
array: we'll have just a for loop instead of having to iterate
over a NULL-terminated array (which would require n+1 elems)
or having to check with ifdef's for usermode/softmmu.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Groundwork for supporting multiple TCG contexts.
Reviewed-by: Richard Henderson <rth@twiddle.net>
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>
Groundwork for supporting multiple TCG contexts.
The core of this patch is this change to tcg/tcg.h:
> -extern TCGContext tcg_ctx;
> +extern TCGContext tcg_init_ctx;
> +extern TCGContext *tcg_ctx;
Note that for now we set *tcg_ctx to whatever TCGContext is passed
to tcg_context_init -- in this case &tcg_init_ctx.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Groundwork for supporting multiple TCG contexts.
Reviewed-by: Richard Henderson <rth@twiddle.net>
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>
Since commit 6e3b2bfd6 ("tcg: allocate TB structs before the
corresponding translated code") we are not fully utilizing
code_gen_buffer for translated code, and therefore are
incorrectly reporting the amount of translated code as well as
the average host TB size. Address this by:
- Making the conscious choice of misreporting the total translated code;
doing otherwise would mislead users into thinking "-tb-size" is not
honoured.
- Expanding tb_tree_stats to accurately count the bytes of translated code on
the host, and using this for reporting the average tb host size,
as well as the expansion ratio.
In the future we might want to consider reporting the accurate numbers for
the total translated code, together with a "bookkeeping/overhead" field to
account for the TB structs.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
We don't really free anything in this function anymore; we just remove
the TB from the binary search tree.
Suggested-by: Alex Bennée <alex.bennee@linaro.org>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This is a prerequisite for supporting multiple TCG contexts, since
we will have threads generating code in separate regions of
code_gen_buffer.
For this we need a new field (.size) in struct tb_tc to keep
track of the size of the translated code. This field uses a size_t
to avoid adding a hole to the struct, although really an unsigned
int would have been enough.
The comparison function we use is optimized for the common case:
insertions. Profiling shows that upon booting debian-arm, 98%
of comparisons are between existing tb's (i.e. a->size and b->size
are both !0), which happens during insertions (and removals, but
those are rare). The remaining cases are lookups. From reading the glib
sources we see that the first key is always the lookup key. However,
the code does not assume this to always be the case because this
behaviour is not guaranteed in the glib docs. However, we embed
this knowledge in the code as a branch hint for the compiler.
Note that tb_free does not free space in the code_gen_buffer anymore,
since we cannot easily know whether the tb is the last one inserted
in code_gen_buffer. The next patch in this series renames tb_free
to tb_remove to reflect this.
Performance-wise, lookups in tb_find_pc are the same as before:
O(log n). However, insertions are O(log n) instead of O(1), which
results in a small slowdown when booting debian-arm:
Performance counter stats for 'build/arm-softmmu/qemu-system-arm \
-machine type=virt -nographic -smp 1 -m 4096 \
-netdev user,id=unet,hostfwd=tcp::2222-:22 \
-device virtio-net-device,netdev=unet \
-drive file=img/arm/jessie-arm32.qcow2,id=myblock,index=0,if=none \
-device virtio-blk-device,drive=myblock \
-kernel img/arm/aarch32-current-linux-kernel-only.img \
-append console=ttyAMA0 root=/dev/vda1 \
-name arm,debug-threads=on -smp 1' (10 runs):
- Before:
8048.598422 task-clock (msec) # 0.931 CPUs utilized ( +- 0.28% )
16,974 context-switches # 0.002 M/sec ( +- 0.12% )
0 cpu-migrations # 0.000 K/sec
10,125 page-faults # 0.001 M/sec ( +- 1.23% )
35,144,901,879 cycles # 4.367 GHz ( +- 0.14% )
<not supported> stalled-cycles-frontend
<not supported> stalled-cycles-backend
65,758,252,643 instructions # 1.87 insns per cycle ( +- 0.33% )
10,871,298,668 branches # 1350.707 M/sec ( +- 0.41% )
192,322,212 branch-misses # 1.77% of all branches ( +- 0.32% )
8.640869419 seconds time elapsed ( +- 0.57% )
- After:
8146.242027 task-clock (msec) # 0.923 CPUs utilized ( +- 1.23% )
17,016 context-switches # 0.002 M/sec ( +- 0.40% )
0 cpu-migrations # 0.000 K/sec
18,769 page-faults # 0.002 M/sec ( +- 0.45% )
35,660,956,120 cycles # 4.378 GHz ( +- 1.22% )
<not supported> stalled-cycles-frontend
<not supported> stalled-cycles-backend
65,095,366,607 instructions # 1.83 insns per cycle ( +- 1.73% )
10,803,480,261 branches # 1326.192 M/sec ( +- 1.95% )
195,601,289 branch-misses # 1.81% of all branches ( +- 0.39% )
8.828660235 seconds time elapsed ( +- 0.38% )
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Now that we have curr_cflags, we can include CF_USE_ICOUNT
early and then remove it as necessary.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
These flags are used by target/*/translate.c,
and affect code generation.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Now that all code generation has been converted to check CF_PARALLEL, we can
generate !CF_PARALLEL code without having yet set !parallel_cpus --
and therefore without having to be in the exclusive region during
cpu_exec_step_atomic.
While at it, merge cpu_exec_step into cpu_exec_step_atomic.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thereby decoupling the resulting translated code from the current state
of the system.
The tb->cflags field is not passed to tcg generation functions. So
we add a field to TCGContext, storing there a copy of tb->cflags.
Most architectures have <= 32 registers, which results in a 4-byte hole
in TCGContext. Use this hole for the new field.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thereby decoupling the resulting translated code from the current state
of the system.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thereby decoupling the resulting translated code from the current state
of the system.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thereby decoupling the resulting translated code from the current state
of the system.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thereby decoupling the resulting translated code from the current state
of the system.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thereby decoupling the resulting translated code from the current state
of the system.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thereby decoupling the resulting translated code from the current state
of the system.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Thereby decoupling the resulting translated code from the current state
of the system.
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Convert all existing readers of tb->cflags to tb_cflags, so that we
use atomic_read and therefore avoid undefined behaviour in C11.
Note that the remaining setters/getters of the field are protected
by tb_lock, and therefore do not need conversion.
Luckily all readers access the field via 'tb->cflags' (so no foo.cflags,
bar->cflags in the code base), which makes the conversion easily
scriptable:
FILES=$(git grep 'tb->cflags' target include/exec/gen-icount.h \
accel/tcg/translator.c | cut -f1 -d':' | sort | uniq)
perl -pi -e 's/([^.>])tb->cflags/$1tb_cflags(tb)/g' $FILES
perl -pi -e 's/([a-z->.]*)(->|\.)tb->cflags/tb_cflags($1$2tb)/g' $FILES
Then manually fixed the few errors that checkpatch reported.
Compile-tested for all targets.
Suggested-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
We were generating code during tb_invalidate_phys_page_range,
check_watchpoint, cpu_io_recompile, and (seemingly) discarding
the TB, assuming that it would magically be picked up during
the next iteration through the cpu_exec loop.
Instead, record the desired cflags in CPUState so that we request
the proper TB so that there is no more magic.
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
This will enable us to decouple code translation from the value
of parallel_cpus at any given time. It will also help us minimize
TB flushes when generating code via EXCP_ATOMIC.
Note that the declaration of parallel_cpus is brought to exec-all.h
to be able to define there the "curr_cflags" inline.
Signed-off-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Using the offset of a temporary, relative to TCGContext, rather than
its index means that we don't use 0. That leaves offset 0 free for
a NULL representation without having to leave index 0 unused.
Reviewed-by: Philippe Mathieu-Daudé <f4bug@amsat.org>
Reviewed-by: Emilio G. Cota <cota@braap.org>
Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
