Don't embed the root of the visit into the stack of current
containers being visited. That way, we no longer get confused
on whether the first visit of a dictionary is to the dictionary
itself or to one of the members of the dictionary, based on
whether the caller passed name=NULL; and makes the QMP Input
visitor like other visitors where the value of 'name' is now
ignored on the root visit. (We may someday want to revisit
the rules on what 'name' should be on a top-level visit,
rather than just ignoring it; but that would be the topic of
another patch).
An audit of all qmp_input_visitor_new() call sites shows that
there were only two places where callers had previously been
visiting to a QDict with a non-NULL name to bypass a call to
visit_start_struct(), and those were fixed in prior patches.
Backports commit ce140b176920b5b65184020735a3c65ed3e9aeda from qemu
Commit e8316d7 mistakenly passed consume=true within
qmp_input_optional() when checking if an optional member was
present, but the mistake was silently ignored since the code
happily let us extract a member more than once. Fix
qmp_input_optional() to not consume anything, then tighten up
the input visitor to ensure that a member is consumed exactly
once (all generated code follows this pattern; and the new
assert will catch any hand-written code that tries to visit
the same key more than once).
Backports commit e5826a2fd727f0be54a81083f31fe02a275465cd from qemu
The following uses of a QMP input visitor should be strict
(that is, excess keys in QDict input should be flagged if not
converted to QAPI):
- Testsuite code unrelated to explicitly testing non-strict
mode (test-qmp-commands, test-visitor-serialization); since
we want more code to be strict by default, having more tests
of strict mode doesn't hurt
- Code used for cloning QAPI objects (replay-input.c,
qemu-sockets.c); we are reparsing a QObject just barely
produced by the qmp output visitor and which therefore should
not have any garbage, so while it is extra work to be strict,
it validates that our clone is correct [note that a later patch
series will simplify these two uses by creating an actual
clone visitor that is much more efficient than a
generate/reparse cycle]
- qmp_object_add(), which calls into user_creatable_add_type().
Since command line parsing for '-object' uses the same
user_creatable_add_type() through the OptsVisitor, and that is
always strict, we want to ensure that any nested dictionaries
would be treated the same in QMP and from the command line (I
don't actually know if such nested dictionaries exist). Note
that on this code change, strictness only matters for nested
dictionaries (if even possible), since we already flag excess
input at the top level during an earlier object_property_set()
on an unknown key, whether from QemuOpts:
$ ./x86_64-softmmu/qemu-system-x86_64 -nographic -nodefaults -qmp stdio -object secret,id=sec0,data=letmein,format=raw,foo=bar
qemu-system-x86_64: -object secret,id=sec0,data=letmein,format=raw,foo=bar: Property '.foo' not found
or from QMP:
$ ./x86_64-softmmu/qemu-system-x86_64 -nographic -nodefaults -qmp stdio
{"QMP": {"version": {"qemu": {"micro": 93, "minor": 5, "major": 2}, "package": ""}, "capabilities": []}}
{"execute":"qmp_capabilities"}
{"return": {}}
{"execute":"object-add","arguments":{"qom-type":"secret","id":"sec0","props":{"format":"raw","data":"letmein","foo":"bar"}}}
{"error": {"class": "GenericError", "desc": "Property '.foo' not found"}}
The only remaining uses of non-strict input visits are:
- QMP 'qom-set' (which eventually executes
object_property_set_qobject()) - mark it as something to revisit
in the future (I didn't want to spend any more time on this patch
auditing if we have any QOM dictionary properties that might be
impacted, and couldn't easily prove whether this code path is
shared with anything else).
- test-qmp-input-visitor: explicit tests of non-strict mode. If
we later get rid of users that don't need strictness, then this
test should be merged with test-qmp-input-strict
Backports relevant parts of commit 240f64b6dc3346d044d7beb7cc3a53668ce47384 from qemu
Rather than having two separate ways to create a QMP input
visitor, where the safer approach has the more verbose name,
it is better to consolidate things into a single function
where the caller must explicitly choose whether to be strict
or to ignore excess input. This patch is the strictly
mechanical conversion; the next patch will then audit which
uses can be made stricter.
Backports commit fc471c18d5d2ec713d5a019f9530398675494bc8 from qemu
Management of the top of stack was a bit verbose; creating a
temporary variable and adding some comments makes the existing
code more legible before the next few patches improve things.
No semantic changes other than asserting that we are always
visiting a QObject, and not a NULL value. In particular, the
check for 'name && qobject_type(qobj) == QTYPE_QDICT)' is a
bit overkill (a dict visit should always have a name); a later
patch revisits that, while this patch is only changing one
layer of indentation due to dropping 'if (qobj)'.
Backports commit b471d012e5d7bec1d2272738141e121b5581fcdf from qemu
Our existing input visitors were not very consistent on errors in a
function taking 'TYPE **obj'. These are start_struct(),
start_alternate(), type_str(), and type_any(). next_list() is
similar, but can't fail (see commit 08f9541). While all of them set
'*obj' to allocated storage on success, it was not obvious whether
'*obj' was guaranteed safe on failure, or whether it was left
uninitialized. But a future patch wants to guarantee that
visit_type_FOO() does not leak a partially-constructed obj back to
the caller; it is easier to implement this if we can reliably state
that input visitors assign '*obj' regardless of success or failure,
and that on failure *obj is NULL. Add assertions to enforce
consistency in the final setting of err vs. *obj.
The opts-visitor start_struct() doesn't set an error, but it
also was doing a weird check for 0 size; all callers pass in
non-zero size if obj is non-NULL.
The testsuite has at least one spot where we no longer need
to pre-initialize a variable prior to a visit; valgrind confirms
that the test is still fine with the cleanup.
A later patch will document the design constraint implemented
here.
Backports commit e58d695e6c3a5cfa0aa2fc91b87ade017ef28b05 from qemu
By sticking the next pointer first, we don't need a union with
64-bit padding for smaller types. On 32-bit platforms, this
can reduce the size of uint8List from 16 bytes (or 12, depending
on whether 64-bit ints can tolerate 4-byte alignment) down to 8.
It has no effect on 64-bit platforms (where alignment still
dictates a 16-byte struct); but fewer anonymous unions is still
a win in my book.
It requires visit_next_list() to gain a size parameter, to know
what size element to allocate; comparable to the size parameter
of visit_start_struct().
I debated about going one step further, to allow for fewer casts,
by doing:
typedef GenericList GenericList;
struct GenericList {
GenericList *next;
};
struct FooList {
GenericList base;
Foo *value;
};
so that you convert to 'GenericList *' by '&foolist->base', and
back by 'container_of(generic, GenericList, base)' (as opposed to
the existing '(GenericList *)foolist' and '(FooList *)generic').
But doing that would require hoisting the declaration of
GenericList prior to inclusion of qapi-types.h, rather than its
current spot in visitor.h; it also makes iteration a bit more
verbose through 'foolist->base.next' instead of 'foolist->next'.
Note that for lists of objects, the 'value' payload is still
hidden behind a boxed pointer. Someday, it would be nice to do:
struct FooList {
FooList *next;
Foo value;
};
for one less level of malloc for each list element. This patch
is a step in that direction (now that 'next' is no longer at a
fixed non-zero offset within the struct, we can store more than
just a pointer's-worth of data as the value payload), but the
actual conversion would be a task for another series, as it will
touch a lot of code.
Backports commit e65d89bf1a4484e0db0f3dc820a8b209f2fb1e8b from qemu
We have three classes of QAPI visitors: input, output, and dealloc.
Currently, all implementations of these visitors have one thing in
common based on their visitor type: the implementation used for the
visit_type_enum() callback. But since we plan to add more such
common behavior, in relation to documenting and further refining
the semantics, it makes more sense to have the visitor
implementations advertise which class they belong to, so the common
qapi-visit-core code can use that information in multiple places.
A later patch will better document the types of visitors directly
in visitor.h.
For this patch, knowing the class of a visitor implementation lets
us make input_type_enum() and output_type_enum() become static
functions, by replacing the callback function Visitor.type_enum()
with the simpler enum member Visitor.type. Share a common
assertion in qapi-visit-core as part of the refactoring.
Move comments in opts-visitor.c to match the refactored layout.
Backports commit 983f52d4b3f86fb9dc9f8b142132feb5a8723016 from qemu
QEMU 2.6 added support for the XSAVE family of instructions, which
includes the XSETBV instruction which allows setting the XCR0
register.
But, when booting Linux kernels with XSAVE support enabled, I was
getting very early crashes where the instruction pointer was set
to 0x3. I tracked it down to a jump instruction generated by this:
gen_jmp_im(s->pc - pc_start);
where s->pc is pointing to the instruction after XSETBV and pc_start
is pointing _at_ XSETBV. Subtract the two and you get 0x3. Whoops.
The fix is to replace this typo with the pattern found everywhere
else in the file when folks want to end the translation buffer.
Richard Henderson confirmed that this is a bug and that this is the
correct fix.
Backports commit 502c8e86ea07294067578292c6d402601c196019 from qemu
Recent versions of GCC report the following error when compiling
target-mips/helper.c:
qemu/target-mips/helper.c:542:9: warning: ‘memset’ used with length
equal to number of elements without multiplication by element size
[-Wmemset-elt-size]
Backports commit a525decfaa3449f1458ea2d7a06320cf46aebf3f from qemu
Commit b00c72180c36 ("target-mips: add PC, XNP reg numbers to RDHWR")
changed the rdhwr helpers to use check_hwrena() to check the register
being accessed is enabled in CP0_HWREna when used from user mode. If
that check fails an EXCP_RI exception is raised at the host PC
calculated with GETPC().
However check_hwrena() may not be fully inlined as the
do_raise_exception() part of it is common regardless of the arguments.
This causes GETPC() to calculate the address in the call in the helper
instead of the generated code calling the helper. No TB will be found
and the EPC reported with the resulting guest RI exception points to the
beginning of the TB instead of the RDHWR instruction.
We can't reliably force check_hwrena() to be inlined, and converting it
to a macro would be ugly, so instead pass the host PC in as an argument,
with each rdhwr helper passing GETPC(). This should avoid any dependence
on compiler behaviour, and in practice seems to ensure the full inlining
of check_hwrena() on x86_64.
This issue causes failures when running a MIPS KVM (trap & emulate)
guest in a MIPS QEMU TCG guest, as the inner guest kernel will do a
RDHWR of counter, which is disabled in the outer guest's CP0_HWREna by
KVM so it can emulate the inner guest's counter. The emulation fails and
the RI exception is passed to the inner guest.
Backports commit d96391c1ffeb30a0afa695c86579517c69d9a889 from qemu
For KVM to use Transparent Huge Pages (THP) we have to ensure that the
alignment of the userspace address of the KVM memory slot and the IPA
that the guest sees for a memory region have the same offset from the 2M
huge page size boundary.
One way to achieve this is to always align the IPA region at a 2M
boundary and ensure that the mmap alignment is also at 2M.
Unfortunately, we were only doing this for __arm__, not for __aarch64__,
so add this simple condition.
This fixes a performance regression using KVM/ARM on AArch64 platforms
that showed a performance penalty of more than 50%, introduced by the
following commit:
9fac18f (oslib: allocate PROT_NONE pages on top of RAM, 2015-09-10)
We were only lucky before the above commit, because we were allocating
large regions and naturally getting a 2M alignment on those allocations
then.
Backports commit ee1e0f8e5d3682c561edcdceccff72b9d9b16d8b from qemu
The TCG code is quite performance sensitive, but at the same time can
also be quite tricky. That is why asserts that can be enabled with the
--enable-debug-tcg configure option.
This used to work the following way:
| #include "config.h"
|
| ...
|
| #if !defined(CONFIG_DEBUG_TCG) && !defined(NDEBUG)
| /* define it to suppress various consistency checks (faster) */
| #define NDEBUG
| #endif
|
| ...
|
| #include <assert.h>
Since commit 757e725b (tcg: Clean up includes) "config.h" as been
replaced by "qemu/osdep.h" which itself includes <assert.h>. As a
consequence the assertions are always enabled, even when using
--disable-debug-tcg, causing a performance regression, especially on
targets with many registers. For instance on qemu-system-ppc the
speed difference is about 15%.
tcg_debug_assert is controlled directly by CONFIG_DEBUG_TCG and already
uses in some places. This patch replaces all the calls to assert into
calss to tcg_debug_assert.
Backports commit eabb7b91b36b202b4dac2df2d59d698e3aff197a from qemu
Accoding the chapter 7.6 Trap Processing of the SPARC Architecture Manual v9,
the Trap Based Address Register is not modified as a trap is taken.
This fix allows booting FreeBSD-10.3-RELEASE-sparc64.
Backports commit de5f1077446ca455342db149737bdc395a7b9882 from qemu
ldstub [addr], reg incorrectly reads a signed byte from memory which causes
problems in the 32-bit Solaris mutex code. Here the byte value being read is
0xff which is incorrectly sign-extended to 0xffffffff before being written back
to the target register causing lock detection to behave incorrectly.
This fixes the intermittent hangs and MUTEX_HELD warnings issued to the
console when running 32-bit Solaris images under qemu-system-sparc.
With thanks to Joseph Dery for providing a condensed test image to consistently
reproduce the problem on demand, and Martin Husemann for allowing me access to
real hardware for comparison.
Backports commit 4553e10360a0713e31647220ed396942f9a6fca0 from qemu
Since 5e5f07e08 "TCG: Move translation block variables
to new context inside tcg_ctx: tb_ctx" on Feb 1 2013, compilation
of usermode + TB_DEBUG_CHECK has been broken. Fix it.
Backports commit 7e6bd36d61129feb7f667cb09ffec1b7b54b971c from qemu
Xiao Guangrong ran kvm-unit-tests on an actual machine with PKU and
found that it fails:
test pte.p pte.user pde.p pde.user pde.a pde.pse pkru.wd pkey=1 user write efer.nx cr4.pke: FAIL: error code 27 expected 7
Dump mapping: address: 0x123400000000
------L4: 2ebe007
------L3: 2ebf007
------L2: 8000000020000a5
(All failures are combinations of "pde.user pde.p pkru.wd pkey=1",
plus either "pde.pse" or "pte.p pte.user", plus one of "user cr0.wp",
"cr0.wp" or "user", plus unimportant bits such as accessed/dirty or
efer.nx).
So PFEC.PKEY is set even if the ordinary check failed (which it did
because pde.w is zero). Adjust QEMU to match behavior of silicon.
Backports commit 44d066a2f770ee9d61fd1c2a609bdf2a994dfdf7 from qemu
The MIPS TCG backend is the only one to have
tcg_target_reg_alloc_order[] elements of type TCGReg rather than int.
This resulted in commit 91478cefaaf2 ("tcg: Allocate indirect_base
temporaries in a different order") breaking the build on MIPS since the
type differed from indirect_reg_alloc_order[]:
tcg/tcg.c:1725:44: error: pointer type mismatch in conditional expression [-Werror]
order = rev ? indirect_reg_alloc_order : tcg_target_reg_alloc_order;
^
Make it an array of ints to fix the build and match other architectures.
Backports commit 2dc7553d0c0a3915c649e1a91b0f0be70b4674b3 from qemu
In order to simplify arguments of function, introduce a new struct
named X86CPUTopoInfo.
Backports commit ed256144cd6f0ca2ff59fc3fc8dca547506f433b from qemu
Move the architecture agnostic function prototypes for exec.c out of
cputlb.h to exec-all.h. This allows hiding of the arch specific
cputlb.h from exec.c which should be getting close to having no
architecture specifics. Prepares support for multi-arch, which will have
a minimal cpu.h that services exec.c but not cputlb.h.
Backports commit dfccc7602374c9fd3b083208b552d62daa244811 from qemu
To prepare for multi-arch, cputlb.c should only have awareness of one
single architecture. This means it should not have access to the full
CPU lists which may be heterogeneous. Instead, push the CPU_LOOP() up
to the one and only caller in exec.c.
Backports commit 9a13565d52bfd321934fb44ee004bbaf5f5913a8 from qemu
The last two arguments to these functions are the last and first bit to
check relative to the base. The code was using incorrectly the first
bit and the number of bits. Fix this in cpu_physical_memory_get_dirty
and cpu_physical_memory_all_dirty. This requires a few changes in the
iteration; change the code in cpu_physical_memory_set_dirty_range to
match.
Backports commit 88c73d16ad1b6c22a2ab082064d0d521f756296a from qemu
The __atomic primitives have been available since GCC 4.7 and provide
a richer interface for describing memory ordering requirements. As a
bonus by using the primitives instead of hand-rolled functions we can
use tools such as the ThreadSanitizer which need the use of well
defined APIs for its analysis.
If we have __ATOMIC defines we exclusively use the __atomic primitives
for all our atomic access. Otherwise we fall back to the mixture of
__sync and hand-rolled barrier cases.
Backports commit a0aa44b488b3601415d55041e4619aef5f3a4ba8 from qemu
__atomic_thread_fence does not include a compiler barrier; in the
C++11 memory model, fences take effect in combination with other
atomic operations. GCC implements this by making __atomic_load and
__atomic_store access memory as if the pointer was volatile, and
leaves no trace whatsoever of acquire and release fences in the
compiler's intermediate representation.
In QEMU, we want memory barriers to act on all memory, but at the same
time we would like to use __atomic_thread_fence for portability reasons.
Add compiler barriers manually around the __atomic_thread_fence.
Backports commit 3bbf572345c65813f86a8fc434ea1b23beb08e16 from qemu
Although accesses to ram_list.dirty_memory[] use atomics so multiple
threads can safely dirty the bitmap, the data structure is not fully
thread-safe yet.
This patch handles the RAM hotplug case where ram_list.dirty_memory[] is
grown. ram_list.dirty_memory[] is change from a regular bitmap to an
RCU array of pointers to fixed-size bitmap blocks. Threads can continue
accessing bitmap blocks while the array is being extended. See the
comments in the code for an in-depth explanation of struct
DirtyMemoryBlocks.
I have tested that live migration with virtio-blk dataplane works.
Backports commit 5b82b703b69acc67b78b98a5efc897a3912719eb from qemu
Move the ALIAS tag from VTCR_EL2 to VTCR so that we migrate the
64-bit version, as is usual. (This has no particular effect now
unless the guest wrote to the high RES0 bits of VTCR_EL2.)
Add a comment about why it's OK that we don't have the various
accessor functions that the EL1 TCR regdefs do.
Backports commit bf06c1123a427fefc2cf9cf8019578eafc19eb6f from qemu
The regdefs for the ESR_EL2 and ESR_EL3 system registers should not
be marked as ARM_CP_ALIAS, because these are the master copies; the
DFSR regdef in vmsa_pmsa_cp_reginfo[] is marked as an alias.
Remove the ALIAS tags so that these registers are correctly migrated.
Backports commit 094a7d0b9d10812d06be2c5c19288cee4603c693 from qemu
The regdef for SCTRL_EL3 was incorrectly marked as being an
ARM_CP_ALIAS, with the remark that this was because the 32-bit
definition would take care of reset and migration. However the
intention for banked registers as documented in the comment in
add_cpreg_to_hashtable() is:
* 2) If ARMv8 is enabled then we can count on a 64-bit version
* taking care of the secure bank. This requires that separate
* 32 and 64-bit definitions are provided.
and so it marks the 32-bit secure banked version as an alias.
This results in the sctlr_s/sctlr_el[3] field never being reset
or migrated for a 64-bit CPU with EL3 enabled.
Fix this by removing the ARM_CP_ALIAS annotation from SCTLR_EL3.
Since this means it now needs a real reset value, move the regdef
into the same place that we define the 32-bit SCTLR.
Backports commit e24fdd238a159d830a9a65dd9b08f80fba9b9e06 from qemu
MIPS Release 6 and MIPS SIMD Architecture make it mandatory to have IEEE
754-2008 FPU which is indicated by CP1 FIR.HAS2008, FCSR.ABS2008 and
FCSR.NAN2008 bits set to 1.
In QEMU we still keep these bits cleared as there is no 2008-NaN support.
However, this now causes problems preventing from running R6 Linux with
the v4.5 kernel. Kernel refuses to execute 2008-NaN ELFs on a CPU
whose FPU does not support 2008-NaN encoding:
(...)
VFS: Mounted root (ext4 filesystem) readonly on device 8:0.
devtmpfs: mounted
Freeing unused kernel memory: 256K (ffffffff806f0000 - ffffffff80730000)
request_module: runaway loop modprobe binfmt-464c
Starting init: /sbin/init exists but couldn't execute it (error -8)
request_module: runaway loop modprobe binfmt-464c
Starting init: /bin/sh exists but couldn't execute it (error -8)
Kernel panic - not syncing: No working init found. Try passing init= option to kernel. See Linux Documentation/init.txt for guidance.
Therefore always indicate presence of 2008-NaN support in R6 as well as in
R5+MSA CPUs, even though this feature is not yet supported by MIPS in QEMU.
Backports commit ba5c79f26221c0fd7139c883a34a4e75d993f732 from qemu
There is no particular reason to keep these functions in the header.
Suggested by Paolo.
Backports commit 99affd1d5bd4e396ecda50e53dfbc5147fa1313d from qemu
The MAAR register is a read/write register included in Release 5
of the architecture that defines the accessibility attributes of
physical address regions. In particular, MAAR defines whether an
instruction fetch or data load can speculatively access a memory
region within the physical address bounds specified by MAAR.
As QEMU doesn't do speculative access, hence this patch only
provides ability to access the registers.
Backports commit f6d4dd810983fdf3d1c9fb81838167efef63d1c8 from qemu
Indicate that in the MIPS64R6-generic CPU the memory-mapped
Global Configuration Register Space is implemented.
Backports commit a9a95061715ca09abff56a3f239f704c410912c2 from qemu
Physical base address for the memory-mapped Coherency Manager Global
Configuration Register space.
The MIPS default location for the GCR_BASE address is 0x1FBF_8.
This register only exists if Config3 CMGCR is set to one.
Backports commit c870e3f52cac0c8a4a1377398327c4ff20d49d41 from qemu
To avoid cluttering the code with #ifdef legs we wrap up the print
statements into a tlb_debug() macro. As access to the virtual TLB can
get quite heavy defining DEBUG_TLB_LOG will ensure all the logs go to
the qemu_log target of CPU_LOG_MMU instead of stderr. This remains
compile time optional as these debug statements haven't been considered
for usefulness for user visible logging.
I've also removed DEBUG_TLB_CHECK which wasn't used.
Backports commit 8526e1f4e418443a4d6ed0714487e47d45ef9c98 from qemu
qemu-log: dfilter-ise exec, out_asm, op and opt_op
This ensures the code generation debug code will honour -dfilter if set.
For the "exec" tracing I've added a new inline macro for efficiency's
sake.
Backports commit d977e1c2dbc9e63454b2000f91954d02543bf43b from qemu
When debugging big programs or system emulation sometimes you want both
the verbosity of cpu,exec et all but don't want to generate lots of logs
for unneeded stuff. This patch adds a new option -dfilter which allows
you to specify interesting address ranges in the form:
-dfilter 0x8000..0x8fff,0xffffffc000080000+0x200,...
Then logging code can use the new qemu_log_in_addr_range() function to
decide if it will output logging information for the given range.
Backports commit 3514552e04388d8e7686bcf89efd022e892acb5b from qemu
