Use pow2floor() to round down to the nearest power of 2,
rather than an inline calculation.
Backports commit 6554f5c03793bb8a3d5dedcebf758a1694fa186c from qemu
Introduces reusable definitions for CPU affinity masks/shifts and gets rid
of hardcoded magic numbers.
Backports commit 0f4a9e45ec35811ee250ac232d84d3c6d4fcd7fc from qemu
There is an error in arm_excp_unmasked() function:
bitwise operator & is used with integer and bool operands
causing an incorrect zeroed result.
The patch fixes it.
Backports commit 771842585f3119f69641ed90a97d56eb9ed6f5ae from qemu
There is an error in functions aarch64_sync_32_to_64() and
aarch64_sync_64_to_32() with mapping of registers between AArch32 and
AArch64. This commit fixes the mapping to match the v8 ARM ARM
section D1.20.1 (table D1-77).
Backports commit 3a9148d0bdcee990fbe86759b9b1f5723c1d7fbc from qemu
All of these hw_errors are fatal and indicate something wrong with
QEMU implementation.
Convert to g_assert_not_reached.
Backports commit 8f6fd322f6e25995629a1a07b56bc5b91fb947ca from qemu
For the A64 instruction set, the semihosting call instruction
is 'HLT 0xf000'. Wire this up to call do_arm_semihosting()
if semihosting is enabled.
Backports commit 8012c84ff92a36d05dfe61af9b24dd01a7ea25e4 from qemu
The 64-bit A64 semihosting API has some pervasive changes from
the 32-bit version:
* all parameter blocks are arrays of 64-bit values, not 32-bit
* the semihosting call number is passed in W0
* the return value is a 64-bit value in X0
Implement the necessary handling for this widening.
Backports relevant parts of commit faacc041619581c566c21ed87aa1933420731282 from qemu
Print semihosting debugging information before the
do_arm_semihosting() call so that angel_SWIreason_ReportException,
which causes the function to not return, gets the same debug prints as
other semihosting calls. Also print out the semihosting call number.
Backports commit 205ace55ffff77964e50af08c99639ec47db53f6 from qemu
Softmmu unaligned load/stores currently goes through through the slow
path for two reasons:
- to support unaligned access on host with strict alignement
- to correctly handle accesses crossing pages
x86 is only concerned by the second reason. Unaligned accesses are
avoided by compilers, but are not uncommon. We therefore would like
to see them going through the fast path, if they don't cross pages.
For that we can use the fact that two adjacent TLB entries can't contain
the same page. Therefore accessing the TLB entry corresponding to the
first byte, but comparing its content to page address of the last byte
ensures that we don't cross pages. We can do this check without adding
more instructions in the TLB code (but increasing its length by one
byte) by using the LEA instruction to combine the existing move with the
size addition.
On an x86-64 host, this gives a 3% boot time improvement for a powerpc
guest and 4% for an x86-64 guest.
Backports commit 8cc580f6a0d8c0e2f590c1472cf5cd8e51761760 from qemu
Implement the AArch64 TLBI operations which take an intermediate
physical address and invalidate stage 2 translations.
Backports commit cea66e91212164e02ad1d245c2371f7e8eb59e7f from qemu
Now we have the ability to flush the TLB only for specific MMU indexes,
update the AArch64 TLB maintenance instruction implementations to only
flush the parts of the TLB they need to, rather than doing full flushes.
We take the opportunity to remove some duplicate functions (the per-asid
tlb ops work like the non-per-asid ones because we don't support
flushing a TLB only by ASID) and to bring the function names in line
with the architectural TLBI operation names.
Backports commit fd3ed969227f54f08f87d9eb6de2d4e48e99279b from qemu
Move the two regdefs for TLBI ALLE1 and TLBI ALLE1IS down so that the
whole set of AArch64 TLBI regdefs is arranged in numeric order.
Backports commit 83ddf975777cc23337b7ef92e83b1b9c949396f3 from qemu
Guest CPU TLB maintenance operations may be sufficiently
specialized to only need to flush TLB entries corresponding
to a particular MMU index. Implement cputlb functions for
this, to avoid the inefficiency of flushing TLB entries
which we don't need to.
Backports commit d7a74a9d4a68e27b3a8ceda17bb95cb0a23d8e4d from qemu
Apply the correct conditions in the ats_access() function for
the ATS12NSO* address translation operations:
* succeed at EL2 or EL3
* normal UNDEF trap from NS EL1
* trap to EL3 from S EL1 (only possible if EL3 is AArch64)
(This change means they're now available in our EL3-supporting
CPUs when they would previously always UNDEF.)
Backports commit 87562e4f4a2bdd028eef3549ce9cb4e7c83cb0bf from qemu
Some coprocessor register access functions need to be able
to report "trap to EL3 with an 'uncategorized' syndrome";
add the necessary CPAccessResult enum and handling for it.
I don't currently know of any registers that need to trap
to EL2 with the 'uncategorized' syndrome, but adding the
_EL2 enum as well is trivial and fills in what would
otherwise be an odd gap in the handling.
Backports commit e76157264da20b85698b09fa5eb8e02e515e232c from qemu
Wire up the AArch64 EL2 and EL3 address translation operations
(AT S12E1*, AT S12E0*, AT S1E2*, AT S1E3*), and correct some
errors in the ats_write64() function in previously unused code
that would have done the wrong kind of lookup for accesses from
EL3 when SCR.NS==0.
Backports commit 2a47df953202e1f226aa045ea974427c4540a167 from qemu
For EL2 stage 1 translations, there is no TTBR1. We were already
handling this for 64-bit EL2; add the code to take the 'no TTBR1'
code path for 64-bit EL2 as well.
Backports commit d0a2cbceb2aa20d64d53e1c20c7d26a78ade8382 from qemu
We already implemented ACTLR_EL1; add the missing ACTLR_EL2 and
ACTLR_EL3, for consistency.
Since we don't currently have any CPUs that need the EL2/EL3
versions to reset to non-zero values, implement as RAZ/WI.
Backports commit 834a6c6920316d39aaf0e68ac936c0a3ad164815 from qemu
The AFSR registers are implementation dependent auxiliary fault
status registers. We already implemented a RAZ/WI AFSR0_EL1 and
AFSR_EL1; add the missing AFSR{0,1}_EL{2,3} for consistency.
Backports commit 37cd6c2478196623ca28526627ca8c69afe0d654 from qemu
The AMAIR registers are for providing auxiliary implementation
defined memory attributes. We already implemented a RAZ/WI
AMAIR_EL1; add the EL2 and EL3 versions for consistency.
Backports commit 2179ef958c81480b841ffa0aab5e265688ffd2b0 from qemu
Add the AArch64 registers MAIR_EL3 and TPIDR_EL3, which are the only
two which we had implemented the 32-bit Secure equivalents of but
not the 64-bit Secure versions.
Backports commit 4cfb8ad896a6f85953038bd913ce3d82d347013d from qemu
apic_internal.h relies on cpu.h having been included (for the
X86CPU type); include it directly rather than relying on it
being pulled in via one of the other includes like timer.h.
Backports commit 20fbcfdd58ea47607a5755979d43f8c48ac93f08 from qemu
Move the muldiv64() function from qemu-common.h to host-utils.h.
This puts it together with all the other arithmetic functions
where we provide a version with __int128_t and a fallback
without, and allows headers which need muldiv64() to avoid
including qemu-common.h.
We don't include host-utils from qemu-common.h, to avoid dragging
more things into qemu-common.h than it already has; in practice
everywhere that needs muldiv64() can get it via qemu/timer.h.
Backports commit 49caffe0cc95a9d0dc344e3328be8197f3536cf8 from qemu
Add a header comment to osdep.h, explaining what the header is for
and some rules to avoid circular-include difficulties.
Backports commit 03557b9abaee78e9d1ef5cd236d32a7b3e75e6f8 from qemu
qemu-common.h has some system header includes and fixups for
things that might be missing. This is really an OS dependency
and belongs in osdep.h, so move it across.
Backports commit bfe7e449f14313f646da621288ca2fd12223414f from qemu
qemu-common.h includes some fixups for things the Win32
headers don't define or define weirdly. These really
belong in os-win32.h, so move them there.
Backports commit 1aad8104f3b69206da1f868639e1f69c26f6d482 from qemu
Add documentation comments for various utility string functions
which we have implemented in util/cutils.c:
pstrcpy()
strpadcpy()
pstrcat()
strstart()
stristart()
qemu_strnlen()
qemu_strsep()
Backports commit ab6036630865eff8bb12dd51dfa6921b4607fc81 from qemu
Rather than rolling custom concatenate-strings macros for the
QEMU_BUILD_BUG_ON macro to use, use the glue() macro we already
have (since it's now available to us in this header).
Backports commit 24134c4e9126bf505b612e901c63a102fc471083 from qemu
osdep.h has a few things which are really compiler specific;
move them to compiler.h, and include compiler.h from osdep.h.
Backports commit 4912086865083a008f4fb73173fd0ddf2206c4d9 from qemu
qemu_printf is an ancient remnant which has been a simple #define to
printf for over a decade, and is used in only a few places. Expand
it out in those places and remove the #define.
Backports commit 71baf787d8fa2a5d186f22d8154069fd212be37f from qemu
There was a complicated subtractive arithmetic for determining the
padding on the CPUTLBEntry structure. Simplify this with a union.
Backports commit b4a4b8d0e0767c85946fd8fc404643bf5766351a from qemu
The LWL/LDL instructions mask the GPR with a mask depending on the
address alignement. It is currently computed by doing:
mask = 0x7fffffffffffffffull >> (t1 ^ 63)
It's simpler to generate it by doing:
mask = ~(-1 << t1)
It uses one TCG instruction less, and it avoids a 32/64-bit constant
loading which can take a few instructions on RISC hosts.
Backports commit eb02cc3f89013612cb05df23b5441741e902bbd2 from qemu
As full specification of P5600 is available, mips32r5-generic should
be renamed to P5600 and corrected as its intention.
Correct PRid and detail of configuration.
Features which are not currently supported are described as FIXME.
Fix Config.MM bit location
Backports commit aff2bc6dc6d839caf6df0900437cc2cc9e180605 from qemu
If EL3 is AArch32, then the secure physical timer is accessed via
banking of the registers used for the non-secure physical timer.
Implement this banking.
Note that the access controls for the AArch32 banked registers
remain the same as the physical-timer checks; they are not the
same as the controls on the AArch64 secure timer registers.
Backports commit 9ff9dd3c875956523bb4c19ca712e5d05aab3c65 from qemu
On CPUs with EL3, there are two physical timers, one for Secure and one
for Non-secure. Implement this extra timer and the AArch64 registers
which access it.
Backports commit b4d3978c2fdf944e428a46d2850dbd950b6fbe78 from qemu
It's easy to accidentally define two cpregs which both try
to reset the same underlying state field (for instance a
clash between an AArch64 EL3 definition and an AArch32
banked register definition). if the two definitions disagree
about the reset value then the result is dependent on which
one happened to be reached last in the hashtable enumeration.
Add a consistency check to detect and assert in these cases:
after reset, we run a second pass where we check that the
reset operation doesn't change the value of the register.
Backports commit 49a661910c1374858602a3002b67115893673c25 from qemu
Rename gt_cnt_reset to gt_timer_reset as the function really
resets the timers and not the counters. Move the registration
from counter regs to timer regs.
Backports commit d57b9ee84f6b2786f025712609edb259d0de086d from qemu
This is legal; the MemoryRegion will simply unreference all the
existing subregions and possibly bring them down with it as well.
However, it requires a bit of care to avoid an infinite loop.
Finalizing a memory region cannot trigger an address space update,
but memory_region_del_subregion errs on the side of caution and
might trigger a spurious update: avoid that by resetting mr->enabled
first.
Backports commit 91232d98da2bfe042d4c5744076b488880de3040 from qemu
The MIPS TCG backend implements qemu_ld with 64-bit targets using the v0
register (base) as a temporary to load the upper half of the QEMU TLB
comparator (see line 5 below), however this happens before the input
address is used (line 8 to mask off the low bits for the TLB
comparison, and line 12 to add the host-guest offset). If the input
address (addrl) also happens to have been placed in v0 (as in the second
column below), it gets clobbered before it is used.
addrl in t2 addrl in v0
1 srl a0,t2,0x7 srl a0,v0,0x7
2 andi a0,a0,0x1fe0 andi a0,a0,0x1fe0
3 addu a0,a0,s0 addu a0,a0,s0
4 lw at,9136(a0) lw at,9136(a0) set TCG_TMP0 (at)
5 lw v0,9140(a0) lw v0,9140(a0) set base (v0)
6 li t9,-4093 li t9,-4093
7 lw a0,9160(a0) lw a0,9160(a0) set addend (a0)
8 and t9,t9,t2 and t9,t9,v0 use addrl
9 bne at,t9,0x836d8c8 bne at,t9,0x836d838 use TCG_TMP0
10 nop nop
11 bne v0,t8,0x836d8c8 bne v0,a1,0x836d838 use base
12 addu v0,a0,t2 addu v0,a0,v0 use addrl, addend
13 lw t0,0(v0) lw t0,0(v0)
Fix by using TCG_TMP0 (at) as the temporary instead of v0 (base),
pushing the load on line 5 forward into the delay slot of the low
comparison (line 10). The early load of the addend on line 7 also needs
pushing even further for 64-bit targets, or it will clobber a0 before
we're done with it. The output for 32-bit targets is unaffected.
srl a0,v0,0x7
andi a0,a0,0x1fe0
addu a0,a0,s0
lw at,9136(a0)
-lw v0,9140(a0) load high comparator
li t9,-4093
-lw a0,9160(a0) load addend
and t9,t9,v0
bne at,t9,0x836d838
- nop
+ lw at,9140(a0) load high comparator
+lw a0,9160(a0) load addend
-bne v0,a1,0x836d838
+bne at,a1,0x836d838
addu v0,a0,v0
lw t0,0(v0)
Backports commit 33fca8589cf2aa7bf91564e6a8f26b3ba0910541 from qemu
When a function returns a null pointer on error and only on error, you
can do
if (!foo(foos, errp)) {
... handle error ...
}
instead of the more cumbersome
Error *err = NULL;
if (!foo(foos, &err)) {
error_propagate(errp, err);
... handle error ...
}
A StringProperty's getter, however, may return null on success! We
then fail to call visit_type_str().
Screwed up in 6a146eb, v1.1.
Fails tests/qom-test in my current, heavily hacked QAPI branch. No
reproducer for master known (but I didn't look hard).
Backports commit a479b21c111a87a50203a7413c4e5ec419fc88dd from qemu
In semihosting mode the SDBBP 1 instructions should trigger UHI syscall,
but in QEMU this does not happen for recently added microMIPS R6.
Consequently bare metal microMIPS R6 programs supporting UHI will not run.
Backports commit 060ebfef1a09b58fb219b3769b72efb407515bf1 from qemu
The add2 code in the tcg_out_addsub2 function doesn't take into account
the case where rl == al == bl. In that case we can't compute the carry
after the addition. As it corresponds to a multiplication by 2, the
carry bit is the bit 31.
While this is a corner case, this prevents x86-64 guests to boot on a
MIPS host.
Backports commit c99d69694af4ed15b33e3f7c2e3ef6972c14358d from qemu
Commit 2b7ec66f fixed TCGMemOp masking following the MO_AMASK addition,
but two cases were forgotten in the TCG S390 backend.
Backports commit 3c8691f568f49bf623dcb2850464d4156d95e61b from qemu
Commit 2b7ec66f fixed TCGMemOp masking following the MO_AMASK addition,
but two cases were forgotten in the TCG MIPS backend.
Backports commit 4214a8cb7c15ec43d4b2a43ebf248b273a0f4d45 from qemu
For 32-bit guest, we load a 32-bit address from the TLB, so there is no
need to compensate for the low or high part. This fixes 32-bit guests on
big-endian hosts.
Backports commit e72c4fb81db52be881c9356f1c60e0a7817d2d32 from qemu
Correct computation of vector offsets for EXCP_EXT_INTERRUPT.
For instance, if Cause.IV is 0 the vector offset should be 0x180.
Simplify the finding vector number logic for the Vectored Interrupts.
Backports commit da52a4dfcc4864fd2260ec4eab331f75b1f0240b from qemu
When a LWL, LWR, LDL or LDR instruction triggers a page fault, QEMU
currently reports the aligned address in CP0 BadVAddr, while the Windows
NT kernel expects the unaligned address.
This patch adds a byte access with the unaligned address at the
beginning of the LWL/LWR/LDL/LDR instructions to possibly trigger a page
fault and fill the QEMU TLB.
Backports commit 908680c6441ac468f4871d513f42be396ea0d264 from qemu
Fix Debug Mode flag clearing, and when DERET is placed between LL and SC
do not make SC fail.
Backports commit fe87c2b36ae9c1c9a5279f3891f3bce1b573baa0 from qemu
When syncing the task ASID with EntryHi, correctly or the value instead
of assigning it.
Backports commit 6a973e6b6584221bed89a01e755b88e58b496652 from qemu
MSACSR.Cause bits are needed to be cleared before a vector floating-point
instructions.
FEXDO.df, FEXUPL.df and FEXUPR.df were missed out.
Backports commit d4f4f0d5d9e74c19614479592c8bc865d92773d0 from qemu
Fix core configuration for MIPS64R6-generic to make it as close as
I6400.
I6400 core has 48-bit of Virtual Address available (SEGBITS).
MIPS SIMD Architecture is available.
Rearrange order of bits to match the specification.
Backports commit 4dc89b782095d7a0b919fafd7b1322b3cb1279f1 from qemu
W10 insider has a bug where it ignores CPUID level and interprets
CPUID.(EAX=07H, ECX=0H) incorrectly, because CPUID in fact returned
CPUID.(EAX=04H, ECX=0H); this resulted in execution of unsupported
instructions.
While it's a Windows bug, there is no reason to emulate incorrect level.
I used http://instlatx64.atw.hu/ as a source of CPUID and checked that
it matches Penryn Xeon X5472, Westmere Xeon W3520, SandyBridge i5-2540M,
and Haswell i5-4670T.
kvm64 and qemu64 were bumped to 0xD to allow all available features for
them (and to avoid the same Windows bug).
Backports commit 3046bb5debc8153a542acb1df93b2a1a85527a15 from qemu.
With the Intel microcode update that removed HLE and RTM, there will be
different kinds of Haswell and Broadwell CPUs out there: some that still
have the HLE and RTM features, and some that don't have the HLE and RTM
features. On both cases people may be willing to use the pc-*-2.3
machine-types.
So, to cover both cases, introduce Haswell-noTSX and Broadwell-noTSX CPU
models, for hosts that have Haswell and Broadwell CPUs without TSX support.
Backports commit a356850b80b3d13b2ef737dad2acb05e6da03753 from qemu
ARAT signals that the APIC timer does not stop in power saving states.
As our APICs are emulated, it's fine to expose this feature to guests,
at least when asking for KVM host features or with CPU types that
include the flag. The exact model number that introduced the feature is
not known, but reports can be found that it's at least available since
Sandy Bridge.
Backports commit 28b8e4d0bf93ba176b4b7be819d537383c5a9060 from qemu
This patch denies crossing the boundary of the pages in the replay mode,
because it can cause an exception. Do it only when boundary is
crossed by the first instruction in the block.
If current instruction already crossed the bound - it's ok,
because an exception hasn't stopped this code.
Backports commit 5b9efc39aee90bbd343793e942bf8f582a0c9e4f from qemu
TCG generates optimized code for i386 repz instructions in single step mode.
It means that when ecx becomes 0, execution of the string instruction breaks
immediately without an additional iteration for ecx==0 (which will only check
ecx and set the flags). Omitting this iteration leads to different
instructions counting in singlestep mode and in normal execution.
This patch disables optimization of this last iteration for icount mode
which should be deterministic.
Backport commit c4d4525c38cd93cc5d1a743976eb25ac571d435f from qemu
This patch simplifies the AES code, by directly accessing the newly added
S-Box, InvS-Box and InvMixColumns tables instead of recreating them by
using the AES_Te and AES_Td tables.
Backports commit 9551ea6991cfb7c777f7943ad69b30d0a4fadac3 from qemu
These represent xsave-related capabilities of the processor, and KVM may
or may not support them.
Add feature bits so that they are considered by "-cpu ...,enforce", and use
the new feature work instead of calling kvm_arch_get_supported_cpuid.
Bit 3 (XSAVES) is not migratables because it requires saving MSR_IA32_XSS.
Neither KVM nor any commonly available hardware supports it anyway.
Backports commit 0bb0b2d2fe7f645ddaf1f0ff40ac669c9feb4aa1 from qemu
also backports 18cd2c17b5370369a886155c001da0a7f54bbcca
With this, object_property_add_alias() callers can safely free the
target property name, like what already happens with the 'name' argument
to all object_property_add*() functions.
Backports commit 1590d266d96b3f9b42443d6388dfc38f527ac2d8 from qemu
The SCTLR_EL3 cpreg definition was implicitly resetting the
register state to 0, which is both wrong and clashes with
the reset done via the SCTLR definition (since sctlr[3]
is unioned with sctlr_s). This went unnoticed until recently,
when an unrelated change (commit a903c449b41f105aa) happened to
perturb the order of enumeration through the cpregs hashtable for
reset such that the erroneous reset happened after the correct one
rather than before it. Fix this by marking SCTLR_EL3 as an alias,
so its reset is left up to the AArch32 view.
Backports commit e46e1a74ef482f1ef773e750df9654ef4442ca29 from qemu
Remove un-needed usages of ENV_GET_CPU() by converting the APIs to use
CPUState pointers and retrieving the env_ptr as minimally needed.
Scripted conversion for target-* change:
for I in target-*/cpu.h; do
sed -i \
's/\(^int cpu_[^_]*_exec(\)[^ ][^ ]* \*s);$/\1CPUState *cpu);/' \
$I;
done
Backports commit ea3e9847408131abc840240bd61e892d28459452 from qemu
The callers (most of them in target-foo/cpu.c) to this function all
have the cpu pointer handy. Just pass it to avoid an ENV_GET_CPU() from
core code (in exec.c).
Backports commit 4bad9e392e788a218967167a38ce2ae7a32a6231 from qemu
The sole caller of this function navigates the cpu->env_ptr only for
this function to take it back the cpu pointer straight away. Pass in
cpu pointer instead and grab the env pointer locally in the function.
Removes a core code usage of ENV_GET_CPU().
Backports commit 3d57f7893c90d911d786cb2c622b0926fc808b57 from qemu
All of the core-code usages of this API have the cpu pointer handy so
pass it in. There are only 3 architecture specific usages (2 of which
are commented out) which can just use ENV_GET_CPU() locally to get the
cpu pointer. The reduces core code usage of the CPU env, which brings
us closer to common-obj'ing these core files.
Backports commit bbd77c180d7ff1b04a7661bb878939b2e1d23798 from qemu
QOM objects are already zero-filled when instantiated, there's no need
to explicitly set numa_node to 0.
Backports commit 199fc85acd0571902eeefef6ea861b8ba4c8201f from qemu
To prepare for a generic internal cipher API, move the
built-in AES implementation into the crypto/ directory
Backports commit 6f2945cde60545aae7f31ab9d5ef29531efbc94f from qemu
Introduce a new crypto/ directory that will (eventually) contain
all the cryptographic related code. This initially defines a
wrapper for initializing gnutls and for computing hashes with
gnutls. The former ensures that gnutls is guaranteed to be
initialized exactly once in QEMU regardless of CLI args. The
block quorum code currently fails to initialize gnutls so it
only works by luck, if VNC server TLS is not requested. The
hash APIs avoids the need to litter the rest of the code with
preprocessor checks and simplifies callers by allocating the
correct amount of memory for the requested hash.
Backports commit ddbb0d09661f5fce21b335ba9aea8202d189b98e from qemu
Make sure to not modify the branch target. This ensure that the
branch target is not corrupted during partial retranslation.
Backports commit cd3b29b745b0ff393b2d37317837bc726b8dacc8 from qemu
The TSC frequency fits comfortably in an int when expressed in kHz,
but it may overflow when converted to Hz. In this case,
tsc-frequency returns a negative value because x86_cpuid_get_tsc_freq
does a 32-bit multiplication before assigning to int64_t.
For simplicity just make tsc_khz a 64-bit value.
Backports commit 06ef227e5158cca6710e6c268d6a7f65a5e2811b from qemu
Currently the "host" page size alignment API is really aligning to both
host and target page sizes. There is the qemu_real_page_size which can
be used for the actual host page size but it's missing a mask and ALIGN
macro as provided for qemu_page_size. Complete the API. This allows
system level code that cares about the host page size to use a
consistent alignment interface without having to un-needingly align to
the target page size. This also reduces system level code dependency
on the cpu specific TARGET_PAGE_SIZE.
Backports commit 4e51361d79289aee2985dfed472f8d87bd53a8df from qemu
Apart from the MSR, the smi field of struct kvm_vcpu_events has to be
translated into the corresponding CPUX86State fields. Also,
memory transaction flags depend on SMM state, so pull it from struct
kvm_run on every exit from KVM to userspace.
Backports relevant parts of commit fc12d72e10828ca6ff75f2ad432b741f07a10cef from qemu
Loading the BIOS in the mac99 machine is interesting, because there is a
PROM in the middle of the BIOS region (from 16K to 32K). Before memory
region accesses were clamped, when QEMU was asked to load a BIOS from
0xfff00000 to 0xffffffff it would put even those 16K from the BIOS file
into the region. This is weird because those 16K were not actually
visible between 0xfff04000 and 0xfff07fff. However, it worked.
After clamping was added, this also worked. In this case, the
cpu_physical_memory_write_rom_internal function split the write in
three parts: the first 16K were copied, the PROM area (second 16K) were
ignored, then the rest was copied.
Problems then started with commit 965eb2f (exec: do not clamp accesses
to MMIO regions, 2015-06-17). Clamping accesses is not done for MMIO
regions because they can overlap wildly, and MMIO registers can be
expected to perform full-width accesses based only on their address
(with no respect for adjacent registers that could decode to completely
different MemoryRegions). However, this lack of clamping also applied
to the PROM area! cpu_physical_memory_write_rom_internal thus failed
to copy the third range above, i.e. only copied the first 16K of the BIOS.
In effect, address_space_translate is expecting _something else_ to do
the clamping for MMIO regions if the incoming length is large. This
"something else" is memory_access_size in the case of address_space_rw,
so use the same logic in cpu_physical_memory_write_rom_internal.
Backports commit b242e0e0e2969c044a318e56f7988bbd84de1f63 from qemu
Including qemu-common.h from other header files is generally a bad
idea, because it means it's very easy to end up with a circular
dependency. For instance, if we wanted to include memory.h from
qom/cpu.h we'd end up with this loop:
memory.h -> qemu-common.h -> cpu.h -> cpu-qom.h -> qom/cpu.h -> memory.h
Remove the include from memory.h. This requires us to fix up a few
other files which were inadvertently getting declarations indirectly
through memory.h.
The biggest change is splitting the fprintf_function typedef out
into its own header so other headers can get at it without having
to include qemu-common.h.
Backports commit fba0a593b2809ecdda68650952cf3d3332ac1990 from qemu
This introduces the memory region property "global_locking". It is true
by default. By setting it to false, a device model can request BQL-free
dispatching of region accesses to its r/w handlers. The actual BQL
break-up will be provided in a separate patch.
Backports commit 196ea13104f802c508e57180b2a0d2b3418989a3 from qemu
This makes it more consistent with all other core code files, which
either just rely on qemu-common.h inclusion or precede cpu.h with
qemu-common.h.
cpu-all.h should not be included in addition to cpu.h. Remove it.
Backports commit 94beb661bd90bcb477eed6d3b07aced988c40163 from qemu
These are not Architecture specific in any way so move them out of
cpu-defs.h. tb-hash.h is an appropriate place as a leading user and
their strong relationship to TB hashing and caching.
Backports commit 41da4bd6420afd1209c408974920f63ff9c658e1 from qemu
This is one of very few things in exec-all with a genuine CPU
architecture dependency. Move these hashing helpers to a new
header to trim exec-all.h down to a near architecture-agnostic
header.
The defs are only used by cpu-exec and translate-all which are both
arch-obj's so the new tb-hash.h has no core code usage.
Backports commit e1b89321bafea9fb33d87852fc91fee579d17dfe from qemu
These exception indicies are generic and don't have any reliance on the
per-arch cpu.h defs. Move them to cpu-all.h so they can be used by core
code that does not have access to cpu-defs.h.
Backports commit 9e0dc48c9f05505b53cb28f860456a0648e56ddf from qemu
Intel C Compiler version 15.0.3.187 Build 20150407 doesn't support
'|' function for non floating-point simd operands.
Define VEC_OR macro which uses _mm_or_si128 supported
both in icc and gcc on x86 platform.
Backports commit 34664507c7f038842f20a2c787915680b1fabba2 from qemu
The usages of this define are pure TCG and there is no architecture
specific variation of the value. Localise it to the TCG engine to
remove another architecture agnostic piece from cpu-defs.h.
This follows on from a28177820a868eafda8fab007561cc19f41941f4 where
temp_buf was moved out of the CPU_COMMON obsoleting the need for
the super early definition.
Backports commit 6e0b07306d1793e8402dd218d2e38a7377b5fc27 from qemu
Implement the YIELD instruction in the ARM and Thumb translators to
actually yield control back to the top level loop rather than being
a simple no-op. (We already do this for A64.)
Backports commit c87e5a61c2b3024116f52f7e68273f864ff7ab82 from qemu
Currently we use DISAS_WFE for both WFE and YIELD instructions.
This is functionally correct because at the moment both of them
are implemented as "yield this CPU back to the top level loop so
another CPU has a chance to run". However it's rather confusing
that YIELD ends up calling HELPER(wfe), and if we ever want to
implement real behaviour for WFE and SEV it's likely to trip us up.
Split out the yield codepath to use DISAS_YIELD and a new
HELPER(yield) function, and have HELPER(wfe) call HELPER(yield).
Backports commit 049e24a191c212d9468db84169197887f2c91586 from qemu
TLBI ALLE1IS is an operation that does invalidate TLB entries on all PEs
in the same Inner Sharable domain, not just on the current CPU. So we
must use tlbiall_is_write() here.
Backports commit 2a6332d968297266dbabf9d33f959e3a5efdd0f9 from qemu
Remove is_mem as it is never tested anymore since:
commit bfa50bc2638d877cf2900712b7503be22e8811cb
Backports commit 805167adcb900fa7b2b114d639c418f5313d0b42 from qemu
When taking an exception print the content of the exception link
register. This is useful especially for synchronous exceptions because
in that case this registers holds the address of the instruction that
generated the exception.
Backports commit b21ab1fc217b4a2b8f2f85d16bdd8510a7817a34 from qemu
Define a new CPU definition supporting MIPS32 Release 6 ISA and
microMIPS32 Release 6 ISA.
Backports commit 4b3bcd016d83cc75f6a495c1db54b6c77f037adc from qemu
Signal a Reserved Instruction exception for removed instruction encoding
in microMIPS Release 6.
Backports commit 9e8f441a7e094c0dc33a1c8f521d9e5bcfc1b4da from qemu
64-bit paired-single (PS) floating point data type is optional in the
pre-Release 6.
It has to raise RI exception when PS type is not implemented. (FIR.PS = 0)
(The PS data type is removed in the Release 6.)
Loongson-2E and Loongson-2F don't have any implementation field in
FCSR0(FIR) but do support PS data format, therefore for these cores RI will
not be signalled regardless of PS bit.
Backports commit e29c962804c4dd3fabd44e703aa87eec555ed910 from qemu
Refactor those instructions in order to reuse them for microMIPS32
Release 6.
Rearrange gen_move_low32().
Backports commit 1f1b4c008e250f870719ed38fbd0bcc14322fc01 from qemu
Remove it except for two things in qerror.h:
* Two #include to be cleaned up separately to avoid cluttering this
patch.
* The QERR_ macros. Mark as obsolete.
Backports commit 4629ed1e98961bbe678db68ef5f4342ff174a6c3 from qemu
These macros expand into error class enumeration constant, comma,
string. Unclean. Has been that way since commit 13f59ae.
The error class is always ERROR_CLASS_GENERIC_ERROR since the previous
commit.
* Prepend every use of a QERR_ macro by ERROR_CLASS_GENERIC_ERROR, and
delete it from the QERR_ macro. No change after preprocessing.
* Rewrite error_set(ERROR_CLASS_GENERIC_ERROR, ...) into
error_setg(...). Again, no change after preprocessing.
Backports commit c6bd8c706a799eb0fece99f468aaa22b818036f3 from qemu
Error classes other than ERROR_CLASS_GENERIC_ERROR should not be used
in new code. Hiding them in QERR_ macros makes new uses hard to spot.
Fortunately, there's just one such macro left. Eliminate it with this
coccinelle semantic patch:
@@
expression EP, E;
@@
-error_set(EP, QERR_DEVICE_NOT_FOUND, E)
+error_set(EP, ERROR_CLASS_DEVICE_NOT_FOUND, "Device '%s' not found", E)
Backports commit 75158ebbe259f0bd8bf435e8f4827a43ec89c877 from qemu
Now that qbool is fixed, let's fix getting and setting a bool
value to a qdict member to also use C99 bool rather than int.
I audited all callers to ensure that the changed return type
will not cause any changed semantics.
Backports commit 34acbc95229f9f841bde83691a5af949c15e105b from qemu
We require a C99 compiler, so let's use 'bool' instead of 'int'
when dealing with boolean values. There are few enough clients
to fix them all in one pass.
Backports commit fc48ffc39ed1060856475e4320d5896f26c945e8 from qemu
The action to potentially switch sp register is not occurring at the correct
point in the interrupt entry or exception exit sequences.
For the interrupt entry case the sp on entry is used to create the stack
exception frame - but this may well be the user stack pointer, since we
haven't done the switch yet. Re-order the flow to switch the sp regs then
use the current sp to create the exception frame.
For the return from exception case the code is unwinding the sp after
switching sp registers. But it should always unwind the supervisor sp
first, then carry out any required sp switch.
Note that these problems don't effect operation unless the user sp bit is
set in the CACR register. Only a single sp is used in the default power up
state. Previously Linux only used this single sp mode. But modern versions
of Linux use the user sp mode now, so we need correct behavior for Linux
to work.
Backports commit 0c8ff723bd29e5c8b2ca989f857ae5c37ec49c4e from qemu
Fill out the code support for the move to/from usp instructions. They are
being decoded, but there is no code to support there actions. So add it.
Current versions of Linux running on the ColdFire 5208 use these instructions.
Backports commit 2a8327e8a8288e301a2f01bc3ca2d465a3a4ca78 from qemu
Introduce a CPU model for the Cortex R5 processor. ARMv7 with MPU,
and both thumb and ARM div instructions.
Also implement dummy ATCM and BTCM. These CPs are defined for R5 but
don't have a lot of meaning in QEMU yet. Raz them so the guest can
proceed if they are read. The TCM registers will return a size of 0,
indicating no TCM.
Backports commit d6a6b13ea1dfeb25c43a648e94cfe4395906f1da from qemu
Unified MPU only. Uses ARM architecture major revision to switch
between PMSAv5 and v7 when ARM_FEATURE_MPU is set. PMSA v6 remains
unsupported and is asserted against.
Backports commit f6bda88ff839e2adefe4959b7def420b90703855 from qemu
Define the arm CP registers for PMSAv7 and their accessor functions.
RGNR serves as a shared index that indexes into arrays storing the
DRBAR, DRSR and DRACR registers. DRBAR and friends have to be VMSDd
separately from the CP interface using a new PMSA specific VMSD
subsection.
Backports commit 6cb0b013a1fa421cdfb83257cd33f855cc90649a from qemu
Define the MPUIR register for MPU supporting ARMv6 and onwards.
Currently we only support unified MPU.
The size of the unified MPU is defined via the number of "dregions".
So just a single config is added to specify this size. (When split MPU
is implemented we will add an extra iregions config).
Backports commit 3281af8114c6b8ead02f08b58e3c36895c1ea047 from qemu
cp_reg_reset() is called from g_hash_table_foreach() which does not
define a specific ordering of the hash table iteration. Thus doing reset
for registers marked as ALIAS would give an ambiguous result when
resetvalue is different for original and alias registers. Exit
cp_reg_reset() early when passed an alias register. Then clean up alias
register definitions from needless resetvalue and resetfn.
In particular, this fixes a bug in the handling of the PMCR register,
which had different resetvalues for its 32 and 64-bit views.
Backports commit b061a82b8afcc45ce09d770d9c0acdf429401054 from qemu
This patch adds the Cortex-M4 CPU. The M4 is basically the same as
the M3, the main differences being the DSP instructions and an
optional FPU. Only no-FPU cortex-M4 is implemented here, cortex-M4F
is not because the core target-arm code doesn't support the M-profile
FPU model yet.
Backports commit ba890a9b2509a0087bb7eafddae02ea5ecbb7bb4 from qemu
The tb_check_watchpoint function currently assumes that all memory
access is done either directly through the TCG code or through an
helper which knows its return address. This is obviously wrong as the
helpers use cpu_ldxx/stxx_data functions to access the memory.
Instead of aborting in that case, don't try to retranslate the code, but
assume that the CPU state (and especially the program counter) has been
saved before calling the helper. Then invalidate the TB based on this
address.
Backports commit 8d302e76755b8157373073d7107e31b0b13f80c1 from qemu
To avoid to many #ifdef in target code, provide a tlb_vaddr_to_host for
both user and softmmu modes. In the first case the function always
succeed and just call the g2h function.
Backports commit 2e83c496261c799b0fe6b8e18ac80cdc0a5c97ce from qemu
The architecture defines that when taking an exception trying to
access a coprocessor register, the "preferred return address" for
the exception is the address of the instruction that caused the
exception. Correct an off-by-4 error which meant we were returning
the address after the instruction for traps which happened because
of a failure of a runtime access-check function on an AArch32
register. (Traps caused by translate-time checkable permissions
failures had the correct address, as did traps on AArch64 registers.)
This fixes https://bugs.launchpad.net/qemu/+bug/1463338
Backports commit 3977ee5d7a9f2e3664dd8b233f3224694e23b62b from qemu
For processors that support MPUs, add a property to de-feature it. This
is similar to the implementation of the EL3 feature.
The processor definition in init sets ARM_FEATURE_MPU if it can support
an MPU. post_init exposes the property, defaulting to true. If cleared
by the instantiator, ARM_FEATURE_MPU is then removed at realize time.
This is to support R profile processors that may or may-not have an MPU
configured.
Backports commit 8f325f568fbd0158cd413e7d637573ba90b3eaab from qemu
Add a boolean for indicating uniprocessors with MP extensions. This
drives the U bit in MPIDR. Prepares support for Cortex-R5.
Backports commit a8e81b319d1ae1224cc7059877dcdf04a5aad59d from qemu
Currently, the return code for get_phys_addr is overloaded for both
success/fail and FSR value return. This doesn't handle the case where
there is an error with a 0 FSR. This case exists in PMSAv7.
So rework get_phys_addr and friends to return a success/failure boolean
return code and populate the FSR via a caller provided uint32_t
pointer.
Backports commit b7cc4e82f04a1c5b218a657f677a2fdd1e1c2889 from qemu
V6+ PMSA and VMSA share some common registers that are currently
in the VMSA definition block. Split them out into a new def that can
be shared to PMSA.
Backports commit 8e5d75c950a1241f6e1243c37f28cd58f68fedc9 from qemu
These registers are VMSA specific so they should be conditional on
VMSA (i.e. !MPU).
Backports commit 5e5cf9e35f25f9f932a6ce25107c11b67b426a43 from qemu
If doing a PMSA (MPU) system do not define the VMSA specific TLBTR CP.
The def is done separately from VMSA registers group as it is affected
by both the OMAP/STRONGARM RW errata and the MIDR backgrounding.
Backports commit 8085ce63c5967d200f1241b6c0a189371993c5df from qemu
When we're using KVM, the kernel's internal idea of the MPIDR
affinity fields must match the values we tell it for the guest
vcpu cluster configuration in the device tree. Since at the moment
the kernel doesn't support letting userspace tell it the correct
affinity fields to use, we must read the kernel's view and
reflect that back in the device tree.
Backports commit eb5e1d3c85dffe677da2550d211f9304a7d5ba3b from qemu
According to ARMv8 ARM, there are additional aliases to MIDR system register in
AArch32 state. So add them to the list.
Backports commit ac00c79ff6635ae9fd732ff357ada0d05e795500 from qemu
According to ARM Cortex-A53/A57 TRM, REVIDR reset value should be zero. So let
REVIDR reset value be specified by CPU model and correct it for Cortex-A53/A57.
Backports commit 13b72b2b9aa7ab7ee129e38e9587acd6a1b9a932 from qemu
This isn't used by any of the code. In fact it looks like it was never
used as it came in with ARMv7 support.
Backports commit a79e0218e0ae27c9cdd2648bd46e5a916c903cc2 from qemu
Since ARMv7 with LPAE support, a supersection short translation table
descriptor has had extended base address fields which hold bits 39:32 of
translated address. These fields are IMPDEF in ARMv6 and ARMv7 without
LPAE support.
Backports commit 4e42a6ca37e39e56725518851f4388e46bd91129 from qemu
The old ARMv5-style page table format includes a kind of second level
descriptor named the "extended small page" format, whose primary purpose
is to allow specification of the TEX memory attribute bits on a 4K page.
This exists on ARMv6 and also (as an implementation extension) on XScale
CPUs; it's UNPREDICTABLE on v5.
We were mishandling this in two ways:
(1) we weren't implementing it for v6 (probably never noticed because
Linux will use the new-style v6 page table format there)
(2) we were not correctly setting the page_size, which is 4K, not 1K
The latter bug went unnoticed for years because the only thing which
the page_size affects is which TLB entries get flushed when the guest
does a TLB invalidate on an address in the page, and prior to commit
2f0d8631b7 we were doing a full TLB flush very frequently due to Linux's
habit of writing the SCTLR pointlessly a lot.
(We can assume that after commit 2f0d8631b7 the bug went unnoticed
for a year because nobody's actually using the Zaurus/XScale emulation...)
Report the correct page size for these descriptors, and permit them
on ARMv6 CPUs. This fixes a problem where a kernel image for Zaurus
can boot the kernel OK but gets random segfaults when it tries to
run userspace programs.
Backports commit fc1891c74ae122a9dc7854f38bae7db03cd911e6 from qemu
PABITS are not hardcoded to 36 bits and we do not model 59 PABITS (which is
the architectural limit) in QEMU.
Backports commit 28b027d5b63c7550c7390041d6dd50948c8f55b8 from qemu
Implement MTHC0 and MFHC0 instructions. In MIPS32 they are used to access
upper word of extended to 64-bits CP0 registers.
In MIPS64, when CP0 destination register specified is the EntryLo0 or
EntryLo1, bits 1:0 of the GPR appear at bits 31:30 of EntryLo0 or
EntryLo1. This is to compensate for RI and XI, which were shifted to bits
63:62 by MTC0 to EntryLo0 or EntryLo1. Therefore creating separate
functions for EntryLo0 and EntryLo1.
Backports commit 5204ea79ea739b557f47fc4db96c94edcb33a5d6 from qemu
CP0.PageGrain.ELPA enables support for large physical addresses. This field
is encoded as follows:
0: Large physical address support is disabled.
1: Large physical address support is enabled.
If this bit is a 1, the following changes occur to coprocessor 0 registers:
- The PFNX field of the EntryLo0 and EntryLo1 registers is writable and
concatenated with the PFN field to form the full page frame number.
- Access to optional COP0 registers with PA extension, LLAddr, TagLo is
defined.
P5600 can operate in 32-bit or 40-bit Physical Address Mode. Therefore if
XPA is disabled (CP0.PageGrain.ELPA = 0) then assume 32-bit Address Mode.
In MIPS64 assume 36 as default PABITS (when CP0.PageGrain.ELPA = 0).
env->PABITS value is constant and indicates maximum PABITS available on
a core, whereas env->PAMask is calculated from env->PABITS and is also
affected by CP0.PageGrain.ELPA.
Backports commit e117f52636d04502fab28bd3abe93347c29f39a5 from qemu
Update tlb->PFN to contain PFN concatenated with PFNX. PFNX is 0 if large
physical address is not supported.
Backports commit cd0d45c40133ef8b409aede5ad8a99aeaf6a70fe from qemu
Extend EntryLo0, EntryLo1, LLAddr and TagLo from 32 to 64 bits in MIPS32.
Introduce gen_move_low32() function which moves low 32 bits from 64-bit
temp to GPR; it sign extends 32-bit value on MIPS64 and truncates on
MIPS32.
Backports commit 284b731a6ae47b9ebabb9613e753c4d83cf75dd3 from qemu
ERETNC is identical to ERET except that an ERETNC will not clear the LLbit
that is set by execution of an LL instruction, and thus when placed between
an LL and SC sequence, will never cause the SC to fail.
Presence of ERETNC is denoted by the Config5.LLB.
Backports commit ce9782f40ac16660ea9437bfaa2c9c34d5ed8110 from qemu
MIPS SIMD Architecture vector loads and stores require misalignment support.
MSA Memory access should work as an atomic operation. Therefore, it has to
check validity of all addresses for a vector store access if it is spanning
into two pages.
Separating helper functions for each data format as format is known in
translation.
To use mmu_idx from cpu_mmu_index() instead of calculating it from hflag.
Removing save_cpu_state() call in translation because it is able to use
cpu_restore_state() on fault as GETRA() is passed.
Backports commit adc370a48fd26b92188fa4848dfb088578b1936c from qemu
Release 6 requires misaligned memory access support for all ordinary memory
access instructions (for example, LW/SW, LWC1/SWC1).
However misaligned support is not provided for certain special memory accesses
such as atomics (for example, LL/SC).
Backports commit be3a8c53b4f18bcc51a462d977cc61a0f46ebb1c from qemu
This relatively small architectural feature adds the following:
FIR.FREP: Read-only. If FREP=1, then Config5.FRE and Config5.UFE are
available.
Config5.FRE: When enabled all single-precision FP arithmetic instructions,
LWC1/LWXC1/MTC1, SWC1/SWXC1/MFC1 cause a Reserved Instructions
exception.
Config5.UFE: Allows user to write/read Config5.FRE using CTC1/CFC1
instructions.
Enable the feature in MIPS64R6-generic CPU.
Backports commit 7c979afd11b09a16634699dd6344e3ba10c9677e from qemu
Move the "Tests" group of functions so that gen_load_fpr32() and
gen_store_fpr32() can use generate_exception().
Backports commit eab9944c7801525737626fa45cddaf00932dd2c8 from qemu
Different CPUs can be in SMM or not at the same time, thus they
will see different things where the chipset places SMRAM.
Backports commit 2001d0cd6d55e5efa9956fa8ff8b89034d6a4329 from qemu
An SMI should definitely wake up a processor in halted state!
This lets OVMF boot with SMM on multiprocessor systems, although
it halts very soon after that with a "CpuIndex != BspIndex"
assertion failure.
Backports commit a9bad65d2c1f61af74ce2ff43238d4b20bf81c3a from qemu
Because the limit field's bits 31:20 is 1, G should be 1.
VMX actually enforces this, let's do it for completeness
in QEMU as well.
Backports commit b4854f1384176d897747de236f426d020668fa3c from qemu
QEMU is not blocking NMIs on entry to SMM. Implementing this has to
cover a few corner cases, because:
- NMIs can then be enabled by an IRET instruction and there
is no mechanism to _set_ the "NMIs masked" flag on exit from SMM:
"A special case can occur if an SMI handler nests inside an NMI handler
and then another NMI occurs. [...] When the processor enters SMM while
executing an NMI handler, the processor saves the SMRAM state save map
but does not save the attribute to keep NMI interrupts disabled.
- However, there is some hidden state, because "If NMIs were blocked
before the SMI occurred [and no IRET is executed while in SMM], they
are blocked after execution of RSM." This is represented by the new
HF2_SMM_INSIDE_NMI_MASK bit. If it is zero, NMIs are _unblocked_
on exit from RSM.
Backports commit 9982f74bad70479939491b69522da047a3be5a0d from qemu
In order to do this, stop using the cpu_in*/out* helpers, and instead
access address_space_io directly.
cpu_in* and cpu_out* remain for usage in the monitor, in qtest, and
in Xen.
Backports commit 3f7d84648607cc0fcb3812bb4b88978e2a7aa24f from qemu
These include page table walks, SVM accesses and SMM state save accesses.
The bulk of the patch is obtained with
sed -i 's/\(\<[a-z_]*_phys\(_notdirty\)\?\>(cs\)->as,/x86_\1,/'
Backports commit b216aa6c0fcbaa8ff4128969c14594896a5485a4 from qemu
mr->terminates alone doesn't guarantee that we are looking at a RAM region.
mr->ram_addr also has to be checked, in order to distinguish RAM and I/O
regions.
So, do the following:
1) add a new define RAM_ADDR_INVALID, and test it in the assertions
instead of mr->terminates
2) IOMMU regions were not setting mr->ram_addr to a bogus value, initialize
it in the instance_init function so that the new assertions would fire
for IOMMU regions as well.
Backports commit ec05ec26f940564b1e07bf88857035ec27e21dd8 from qemu
The cpu_physical_memory_reset_dirty() function is sometimes used
together with cpu_physical_memory_get_dirty(). This is not atomic since
two separate accesses to the dirty memory bitmap are made.
Turn cpu_physical_memory_reset_dirty() and
cpu_physical_memory_clear_dirty_range_type() into the atomic
cpu_physical_memory_test_and_clear_dirty().
Backports commit 03eebc9e3246b9b3f5925aa41f7dfd7c1e467875 from qemu
Use set_bit_atomic() and bitmap_set_atomic() so that multiple threads
can dirty memory without race conditions.
Backports commit d114875b9a1c21162f69a12d72f69a22e7bab376 from qemu
Most of the time, not all bitmaps have to be marked as dirty;
do not do anything if the interesting ones are already dirty.
Previously, any clean bitmap would have cause all the bitmaps to be
marked dirty.
In fact, unless running TCG most of the time bitmap operations need
not be done at all, because memory_region_is_logging returns zero.
In this case, skip the call to cpu_physical_memory_range_includes_clean
altogether as well.
With this patch, cpu_physical_memory_set_dirty_range is called
unconditionally, so there need not be anymore a separate call to
xen_modified_memory.
Backports commit e87f7778b64d4a6a78e16c288c7fdc6c15317d5f from qemu
cpu_physical_memory_set_dirty_lebitmap unconditionally syncs the
DIRTY_MEMORY_CODE bitmap. This however is unused unless TCG is
enabled.
Backports commit 9460dee4b2258e3990906fb34099481c8334c267 from qemu
The new bitmap_test_and_clear_atomic() function clears a range and
returns whether or not the bits were set.
Backports commit 36546e5b803f6e363906607307f27c489441fd15 from qemu
Use atomic_or() for atomic bitmaps where several threads may set bits at
the same time. This avoids the race condition between threads loading
an element, bitwise ORing, and then storing the element.
When setting all bits in a word we can avoid atomic ops and instead just
use an smp_mb() at the end.
Most bitmap users don't need atomicity so introduce new functions.
Backports commit 9f02cfc84b85929947b32fe1674fbc6a429f332a from qemu
While it is obvious that cpu_physical_memory_get_dirty returns true even if
a single page is dirty, the same is not true for cpu_physical_memory_get_clean;
one would expect that it returns true only if all the pages are clean, but
it actually looks for even one clean page. (By contrast, the caller of that
function, cpu_physical_memory_range_includes_clean, has a good name).
To clarify, rename the function to cpu_physical_memory_all_dirty and return
true if _all_ the pages are dirty. This is the opposite of the previous
meaning, because "all are 1" is the same as "not (any is 0)", so we have to
modify cpu_physical_memory_range_includes_clean as well
Backports commit 72b47e79cef36ed6ffc718f10e21001d7ec2a66f from qemu
is_cpu_write_access is only set if tb_invalidate_phys_page_range is called
from tb_invalidate_phys_page_fast, and hence from notdirty_mem_write.
However:
- the code bitmap can be built directly in tb_invalidate_phys_page_fast
(unconditionally, since is_cpu_write_access would always be passed as 1);
- the virtual address is not needed to mark the page as "not containing
code" (dirty code bitmap = 1), so we can also remove that use of
is_cpu_write_access. For calls of tb_invalidate_phys_page_range
that do not come from notdirty_mem_write, the next call to
notdirty_mem_write will notice that the page does not contain code
anymore, and will fix up the TLB entry.
The parameter needs to remain in order to guard accesses to cpu->mem_io_pc.
Backports commit fc377bcf617a48233a99a9fe0a26247c38b5cb76 from qemu
These days modification of the TLB is done in notdirty_mem_write,
so the virtual address and env pointer as unnecessary.
The new name of the function, tlb_unprotect_code, is consistent with
tlb_protect_code.
Backports commit 9564f52da7eb061326956ed9a468935e3352512d from qemu
Remove them from the sundry exec-all.h header, since they are only used by
the TCG runtime in exec.c and user-exec.c.
Backports commit 1652b974766401743879d78f796f44b8929b0787 from qemu
DIRTY_MEMORY_CODE is only needed for TCG. By adding it directly to
mr->dirty_log_mask, we avoid testing for TCG everywhere a region is
checked for the enabled/disabled state of dirty logging.
Backports commit 677e7805cf95f3b2bca8baf0888d1ebed7f0c606 from qemu
When the dirty log mask will also cover other bits than DIRTY_MEMORY_VGA,
some listeners may be interested in the overall zero/non-zero value of
the dirty log mask; others may be interested in the value of single bits.
For this reason, always call log_start/log_stop if bits have respectively
appeared or disappeared, and pass the old and new values of the dirty log
mask so that listeners can distinguish the kinds of change.
For example, KVM checks if dirty logging used to be completely disabled
(in log_start) or is now completely disabled (in log_stop). On the
other hand, Xen has to check manually if DIRTY_MEMORY_VGA changed,
since that is the only bit it cares about.
Backports commit b2dfd71c4843a762f2befe702adb249cf55baf66 from qemu
For now memory regions only track DIRTY_MEMORY_VGA individually, but
this will change soon. To support this, split memory_region_is_logging
in two functions: one that returns a given bit from dirty_log_mask,
and one that returns the entire mask. memory_region_is_logging gets an
extra parameter so that the compiler flags misuse.
While VGA-specific users (including the Xen listener!) will want to keep
checking that bit, KVM and vhost check for "any bit except migration"
(because migration is handled via the global start/stop listener
callbacks).
Backports commit 2d1a35bef0ed96b3f23535e459c552414ccdbafd from qemu
phys_page_set_level is writing zeroes to a struct that has just been
filled in by phys_map_node_alloc. Instead, tell phys_map_node_alloc
whether to fill in the page "as a leaf" or "as a non-leaf".
memcpy is faster than struct assignment, which copies each bitfield
individually. A compiler bug (https://gcc.gnu.org/PR66391), and
small memcpys like this one are special-cased anyway, and optimized
to a register move, so just use the memcpy.
This cuts the cost of phys_page_set_level from 25% to 5% when
booting qboot.
Backports commit db94604b20278c1dc227a04e4c564d80230e6c3f from qemu
At 8k per TLB (for 64-bit host or target), 8 or more modes
make the TLBs bigger than 64k, and some RISC TCG backends do
not like that. On the affected hosts, cut the TLB size in
half---there is still a measurable speedup on PPC with the
next patch.
Backports commit 1de29aef17a7d70dbc04a7fe51e18942e3ebe313 from qemu
This will be used to size the TLB when more than 8 MMU modes are
used by the target. Limitations come from the limited size of
the immediate fields (which sometimes, as in the case of Aarch64,
extend to instructions that shift the immediate).
Backports commit 006f8638c62bca2b0caf609485f47fa5e14d8a3c from qemu
Existing definition triggers the following when using clang
-fsanitize=undefined:
hw/intc/apic_common.c:314:55: runtime error: left shift of 1048575 by 12
places cannot be represented in type 'int'
Fix it so we won't try to shift a 1 to the sign bit of a signed integer.
Backports commit 458cf469f4a1cb520b07092f5537c5a6d2389d23 from qemu
The ARMCPRegInfo arrays v8_el3_no_el2_cp_reginfo and v8_el2_cp_reginfo
are actually used on non-v8 CPUs as well. Remove the incorrect v8_
prefix from their names.
Backports commit 4771cd01daaccb2a8929fa04c88c608e378cf814 from qemu
