target/mips: Remove tswap() calls in semihosting uhi_fstat_cb()

In preparation of heterogeneous emulation where cores with
different endianness can run concurrently, we need to remove
the tswap() calls -- which use a fixed per-binary endianness.

Get the endianness of the UHI CPU accessed using
mips_env_is_bigendian() and replace the tswap() calls
by bswap() ones when necessary.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-Id: <20241211230357.97036-3-philmd@linaro.org>

target/xtensa: Remove tswap() calls in semihosting simcall() helper

In preparation of heterogeneous emulation where cores with
different endianness can run concurrently, replace the pair
of cpu_memory_rw_debug() + tswap() calls by put/get_user_u32()
ones, which still do the same under the hood, but simplify the
code maintenance (having less sites to do endianness code
conversion).

Suggested-by: Peter Maydell <peter.maydell@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <6e1e69d8-a9f3-4a30-83c8-84c5647578d5@linaro.org>

accel/tcg: Un-inline translator_is_same_page()

Remove the single target-specific definition used in
"exec/translator.h" (TARGET_PAGE_MASK) by un-inlining
is_same_page().
Rename the method as translator_is_same_page() and
improve its documentation.
Use it in translator_use_goto_tb().

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241218154145.71353-1-philmd@linaro.org>

accel/tcg: Include missing 'exec/translation-block.h' header

TB compile flags, tb_page_addr_t type, tb_cflags() and few
other methods are defined in "exec/translation-block.h".

All these files don't include "exec/translation-block.h" but
include "exec/exec-all.h" which include it. Explicitly include
"exec/translation-block.h" to be able to remove it from
"exec/exec-all.h" later when it won't be necessary. Otherwise
we'd get errors such:

  accel/tcg/internal-target.h:59:20: error: a parameter list without types is only allowed in a function definition
     59 | void tb_lock_page0(tb_page_addr_t);
        |                    ^
  accel/tcg/tb-hash.h:64:23: error: unknown type name 'tb_page_addr_t'
     64 | uint32_t tb_hash_func(tb_page_addr_t phys_pc, vaddr pc,
        |                       ^
  accel/tcg/tcg-accel-ops.c:62:36: error: use of undeclared identifier 'CF_CLUSTER_SHIFT'
     62 |     cflags = cpu->cluster_index << CF_CLUSTER_SHIFT;
        |                                    ^
  accel/tcg/watchpoint.c:102:47: error: use of undeclared identifier 'CF_NOIRQ'
    102 |                     cpu->cflags_next_tb = 1 | CF_NOIRQ | curr_cflags(cpu);
        |                                               ^
  target/i386/helper.c:536:28: error: use of undeclared identifier 'CF_PCREL'
    536 |     if (tcg_cflags_has(cs, CF_PCREL)) {
        |                            ^
  target/rx/cpu.c:51:21: error: incomplete definition of type 'struct TranslationBlock'
     51 |     cpu->env.pc = tb->pc;
        |                   ~~^
  system/physmem.c:2977:9: error: call to undeclared function 'tb_invalidate_phys_range'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
   2977 |         tb_invalidate_phys_range(addr, addr + length - 1);
        |         ^
  plugins/api.c:96:12: error: call to undeclared function 'tb_cflags'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
     96 |     return tb_cflags(tcg_ctx->gen_tb) & CF_MEMI_ONLY;
        |            ^

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241114011310.3615-5-philmd@linaro.org>

accel/tcg: Move tcg_cflags_has/set() to 'exec/translation-block.h'

The TranslationBlock flags are defined in 'exec/translation-block.h'.
tcg_cflags_has/set() use them, it is more logical to declare them in
the same place. Move them there too.

Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212144430.66224-2-philmd@linaro.org>

accel/tcg: Restrict curr_cflags() declaration to 'internal-common.h'

curr_cflags() is only used within accel/tcg/,
move its declaration to accel/tcg/internal-common.h.

Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241216214030.59393-1-philmd@linaro.org>

qemu/coroutine: Include missing 'qemu/atomic.h' header

Commit 944f3d5dd21 ("coroutine: Add qemu_co_mutex_assert_locked")
added an inline method which uses qatomic_read(), itself declared
in "qemu/atomic.h". Explicitly include it now to avoid issue when
refactoring unrelated headers.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
Message-Id: <20241217141326.98947-3-philmd@linaro.org>

exec/translation-block: Include missing 'qemu/atomic.h' header

When moving tb_cflags() in commit 88d4b5138a8 ("tcg: Make
tb_cflags() usable from target-agnostic code") we forgot to
include "qemu/atomic.h", which declares qatomic_read().
Explicitly include it now to avoid issue when refactoring
unrelated headers.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
Message-Id: <20241217141326.98947-2-philmd@linaro.org>

accel/tcg: Declare cpu_loop_exit_requested() in 'exec/cpu-common.h'

Move cpu_loop_exit_requested() declaration to "exec/cpu-common.h",
along with the related cpu_loop_exit_noexc(), cpu_loop_exit(),
cpu_loop_exit_atomic() and cpu_loop_exit_restore() methods.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Message-Id: <20241217140648.98538-1-philmd@linaro.org>

exec/cpu-all: Include 'cpu.h' earlier so MMU_USER_IDX is always defined

Include "cpu.h" earlier to get the MMU_USER_IDX definition soon
enough and avoid when refactoring unrelated headers:

  In file included from include/exec/translator.h:271,
                   from ../../accel/tcg/translator.c:13:
  include/exec/cpu-all.h: In function ‘cpu_mmu_index’:
  include/exec/cpu-all.h:274:12: error: ‘MMU_USER_IDX’ undeclared (first use in this function)
    274 |     return MMU_USER_IDX;
        |            ^~~~~~~~~~~~
  include/exec/cpu-all.h:274:12: note: each undeclared identifier is reported only once for each function it appears in
  ninja: build stopped: subcommand failed.

We need to forward-declare cpu_mmu_index() to avoid on user emulation:

  In file included from include/exec/cpu-all.h:263,
                   from include/exec/translator.h:271,
                   from ../../accel/tcg/translator.c:13:
  ../../target/sparc/cpu.h: In function ‘cpu_get_tb_cpu_state’:
  ../../target/sparc/cpu.h:757:13: error: implicit declaration of function ‘cpu_mmu_index’ [-Werror=implicit-function-declaration]
    757 |     flags = cpu_mmu_index(env_cpu(env), false);
        |             ^~~~~~~~~~~~~

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241218155202.71931-5-philmd@linaro.org>

target/sparc: Move sparc_restore_state_to_opc() to cpu.c

Most targets define their restore_state_to_opc() handler in cpu.c.
In order to keep SPARC aligned, move sparc_restore_state_to_opc()
from translate.c to cpu.c.

Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-Id: <20241115152053.66442-4-philmd@linaro.org>
[PMD: Move definitions to new target/sparc/translate.h]

target/sparc: Uninline cpu_get_tb_cpu_state()

Required to compile accel/tcg/translate-all.c once for softmmu targets.
The function gets quite big for some targets so uninlining makes sense.

Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Anton Johansson<anjo@rev.ng>
Message-Id: <20240119144024.14289-14-anjo@rev.ng>
[PMD: Only take SPARC part]
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

target/loongarch: Declare loongarch_cpu_dump_state() locally

loongarch_cpu_dump_state() is not used outside of cpu.c,
no need to expose its prototype.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241115152053.66442-3-philmd@linaro.org>

user: Move various declarations out of 'exec/exec-all.h'

Move various declarations related to user emulation
from "exec/exec-all.h" to "user/cpu_loop.h".

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-19-philmd@linaro.org>

user: Declare cpu_loop() once in 'user/cpu_loop.h'

Declare cpu_loop() once in "user/cpu_loop.h".
bsd-user gets the G_NORETURN attribute.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-18-philmd@linaro.org>

user: Move 'linux-user/cpu_loop-common.h' -> 'user/cpu_loop.h'

"linux-user/cpu_loop-common.h" is generic enough to be used by
bsd-user, so rename it as "user/cpu_loop.h".

Mechanical change running:

  $ sed -i -e 's,cpu_loop-common.h,user/cpu_loop.h,' \
                $(git grep -l cpu_loop-common.h)

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-17-philmd@linaro.org>

user: Forward declare target_cpu_copy_regs structure

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-16-philmd@linaro.org>

accel/tcg: Move user-related declarations out of 'exec/cpu-all.h' (4/4)

Move declarations related to page protection under user
emulation from "exec/cpu-all.h" to "user/page-protection.h".

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-15-philmd@linaro.org>

accel/tcg: Move user-related declarations out of 'exec/cpu-all.h' (3/4)

Move declarations related to page protection under user
emulation from "exec/cpu-all.h" to "user/page-protection.h".

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-14-philmd@linaro.org>

accel/tcg: Move user-related declarations out of 'exec/cpu-all.h' (2/4)

Move declarations related to page protection under user
emulation from "exec/cpu-all.h" to "user/page-protection.h".

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-13-philmd@linaro.org>

accel/tcg: Move user-related declarations out of 'exec/cpu-all.h' (1/4)

Move declarations related to page protection under user
emulation from "exec/cpu-all.h" to "user/page-protection.h".

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-12-philmd@linaro.org>

accel/tcg: Really restrict cpu_io_recompile() to system emulation

Commit 38fc4b11e03 ("accel/tcg: Restrict cpu_io_recompile() to
system emulation") inadvertently restricted cpu_io_recompile()
to SoftMMU. Correct to restrict to system emulation.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241216160514.56630-1-philmd@linaro.org>

accel/tcg: Move TranslationBlock declarations to 'tb-internal.h'

Move declarations related to TranslationBlock out of the
generic "internal-target.h" to "tb-internal.h".

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-11-philmd@linaro.org>

accel/tcg: Un-inline log_pc()

log_pc() is only used within cpu-exec.c, no need to
expose it via "internal-target.h".

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-10-philmd@linaro.org>

accel/tcg: Move 'exec/translate-all.h' -> 'tb-internal.h'

"exec/translate-all.h" is only useful to TCG accelerator,
so move it to accel/tcg/, after renaming it 'tb-internal.h'.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-9-philmd@linaro.org>

system: Remove unnecessary 'exec/translate-all.h' include

At this point "exec/translate-all.h" only declare
tb_check_watchpoint(), which isn't used by any of
cpu-target.c or system/physmem.c, so remove its
inclusion.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-8-philmd@linaro.org>

accel/tcg: Move page_[un]protect() to 'user/page-protection.h'

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-7-philmd@linaro.org>

accel/tcg: Use tb_page_addr_t type in page_unprotect()

Match with the page_protect() prototype, use a tb_page_addr_t
argument instead of target_ulong.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-6-philmd@linaro.org>

accel/tcg: Declare mmap_[un]lock() in 'exec/page-protection.h'

Move mmap_lock(), mmap_unlock() declarations and the
WITH_MMAP_LOCK_GUARD() definition to 'exec/page-protection.h'.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-5-philmd@linaro.org>

include: Include missing 'qemu/clang-tsa.h' header

The next commit will remove "qemu/clang-tsa.h" of "exec/exec-all.h",
however the following files indirectly include it:

  $ git grep -L qemu/clang-tsa.h $(git grep -wl TSA_NO_TSA)
  block/create.c
  include/block/block_int-common.h
  tests/unit/test-bdrv-drain.c
  tests/unit/test-block-iothread.c
  util/qemu-thread-posix.c

Explicitly include it so we can process with the removal in the
next commit.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-4-philmd@linaro.org>

exec/ram_addr: Include missing 'exec/hwaddr.h' and 'exec/cpu-common.h'

'hwaddr' is defined in "exec/hwaddr.h", 'ram_addr_t' in
"exec/cpu-common.h". Include these headers in order to
avoid when refactoring unrelated headers:

  In file included from ../../hw/s390x/s390-virtio-ccw.c:17:
  include/sysemu/physmem-target.h:37:24: error: unknown type name 'hwaddr'
     37 |     (MemoryRegion *mr, hwaddr offset, hwaddr length, unsigned client);
        |                        ^
  In file included from ../../hw/s390x/s390-virtio-ccw.c:16:
  include/exec/ram_addr.h:52:36: error: unknown type name 'ram_addr_t'
     52 | RAMBlock *qemu_ram_alloc_from_file(ram_addr_t size, MemoryRegion *mr,
        |                                    ^

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241218155202.71931-4-philmd@linaro.org>

target/arm/mte: Restrict 'exec/ram_addr.h' to system emulation

"exec/ram_addr.h" contains system specific declarations.
Restrict its inclusion to sysemu to avoid build errors
when refactoring.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241114011310.3615-10-philmd@linaro.org>

target/arm/cpu: Restrict cpu_untagged_addr() to user emulation

Move the #endif guard where it belongs to restrict
the cpu_untagged_addr() implementation to user
emulation.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241114011310.3615-11-philmd@linaro.org>

user: Introduce 'user/guest-host.h' header

Extract all declarations related to 'guest from/to host'
address translation to a new "user/guest-host.h" header.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-2-philmd@linaro.org>

linux-user/aarch64: Include missing 'user/abitypes.h' header

arm_set_mte_tcf0() uses the abi_long type which is defined
in "user/abitypes.h". Include it in order to avoid when
refactoring:

  In file included from ../../target/arm/gdbstub64.c:28:
  ../linux-user/aarch64/mte_user_helper.h:30:42: error: unknown type name ‘abi_long’; did you mean ‘u_long’?
     30 | void arm_set_mte_tcf0(CPUArchState *env, abi_long value);
        |                                          ^~~~~~~~

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241212185341.2857-3-philmd@linaro.org>

exec/translation-block: Include missing 'exec/vaddr.h' header

'vaddr' type is declared in "exec/vaddr.h".
"exec/translation-block.h" uses this type without including
the corresponding header. It works because this header is
indirectly included, but won't work when the other headers
are refactored:

  include/exec/translation-block.h:56:5: error: unknown type name 'vaddr'
     56 |     vaddr pc;
        |     ^

Explitly include "exec/vaddr.h" to avoid such problem in a
few commits.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241114011310.3615-4-philmd@linaro.org>

exec/cpu-defs: Remove unnecessary headers

"exec/cpu-defs.h" should be kept as minimal as possible;
besides these includes don't seem necessary. Remove them.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Tested-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241114011310.3615-3-philmd@linaro.org>

exec/cpu-all: Include missing 'exec/cpu-defs.h' header

TARGET_PAGE_BITS is defined in each target "cpu-param.h",
itself included by "exec/cpu-defs.h".
Include the latter in order to avoid when refactoring:

  In file included from ../../system/watchpoint.c:23:
  include/exec/cpu-all.h:356:19: error: use of undeclared identifier 'TARGET_PAGE_BITS'
    356 | QEMU_BUILD_BUG_ON(TLB_FLAGS_MASK & TLB_SLOW_FLAGS_MASK);
        |                   ^

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241114011310.3615-2-philmd@linaro.org>

accel/tcg: Have tlb_vaddr_to_host() use vaddr type

abi_ptr is expected to be used in user emulation.
tlb_vaddr_to_host() uses it, but can be used in
system emulation. Replace the type by 'vaddr' which
is equivalent on user emulation but also works on
system.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Pierrick Bouvier <pierrick.bouvier@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241114011310.3615-13-philmd@linaro.org>

accel/tcg: Include missing 'exec/tswap.h' header in translator.c

translator.c indirectly gets "exec/tswap.h" declarations via
"exec/cpu-all.h". Include it directly to be able to remove the
former from the latter, otherwise we get:

  accel/tcg/translator.c:433:15: error: call to undeclared function 'tswap16'; ISO C99 and later do not support implicit function declarations [-Wimplicit-function-declaration]
    433 |         tgt = tswap16(raw);
        |               ^

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241211230357.97036-4-philmd@linaro.org>

tcg/tci: Include missing 'disas/dis-asm.h' header

"disas/dis-asm.h" defines bfd_vma and disassemble_info,
include it in order to avoid (when refactoring other
headers):

  tcg/tci.c:1066:20: error: unknown type name 'bfd_vma'
  int print_insn_tci(bfd_vma addr, disassemble_info *info)
                     ^
  tcg/tci.c:1066:34: error: unknown type name 'disassemble_info'
  int print_insn_tci(bfd_vma addr, disassemble_info *info)
                                   ^

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20241218155202.71931-3-philmd@linaro.org>

system: Move 'exec/confidential-guest-support.h' to system/

"exec/confidential-guest-support.h" is specific to system
emulation, so move it under the system/ namespace.
Mechanical change doing:

  $ sed -i \
    -e 's,exec/confidential-guest-support.h,sysemu/confidential-guest-support.h,' \
        $(git grep -l exec/confidential-guest-support.h)

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Zhao Liu <zhao1.liu@intel.com>
Message-Id: <20241218155913.72288-2-philmd@linaro.org>

include: Rename sysemu/ -> system/

Headers in include/sysemu/ are not only related to system
*emulation*, they are also used by virtualization. Rename
as system/ which is clearer.

Files renamed manually then mechanical change using sed tool.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Tested-by: Lei Yang <leiyang@redhat.com>
Message-Id: <20241203172445.28576-1-philmd@linaro.org>

target/i386/sev: Reduce system specific declarations

"system/confidential-guest-support.h" is not needed,
remove it. Reorder #ifdef'ry to reduce declarations
exposed on user emulation.

Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Thomas Huth <thuth@redhat.com>
Reviewed-by: Zhao Liu <zhao1.liu@intel.com>
Message-Id: <20241218155913.72288-3-philmd@linaro.org>

qom: Create system containers explicitly

Always explicitly create QEMU system containers upfront.

Root containers will be created when trying to fetch the root object the
1st time.  They are:

  /objects
  /chardevs
  /backend

Machine sub-containers will be created only until machine is being
initialized.  They are:

  /machine/unattached
  /machine/peripheral
  /machine/peripheral-anon

Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241121192202.4155849-8-peterx@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

hw/ppc: Explicitly create the drc container

QEMU will start to not rely on implicit creations of containers soon.  Make
PPC drc devices follow by explicitly create the container whenever a drc
device is realized, dropping container_get() calls.

No functional change intended.

Cc: Nicholas Piggin <npiggin@gmail.com>
Cc: Daniel Henrique Barboza <danielhb413@gmail.com>
Cc: Harsh Prateek Bora <harshpb@linux.ibm.com>
Cc: qemu-ppc@nongnu.org
Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241121192202.4155849-7-peterx@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

ppc/e500: Avoid abuse of container_get()

container_get() is going to become strict on not allowing to return a
non-container.

Switch the e500 user to use object_resolve_path_component() explicitly.

Cc: Bharat Bhushan <r65777@freescale.com>
Cc: qemu-ppc@nongnu.org
Reviewed-by: Cédric Le Goater <clg@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-ID: <20241121192202.4155849-6-peterx@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

tests: Explicitly create containers in test_qom_partial_path()

Drop one use of container_get(), instead switch to the explicit function to
create a container.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-ID: <20241121192202.4155849-5-peterx@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

tests: Fix test-qdev-global-props on anonymous qdev realize()

test-qdev-global-props creates a few subprocesses and test things based on
qdev realize().  One thing was overlooked since the start, that anonymous
creations of qdev (then realize() the device) requires the machine object's
presence, as all these devices need to be attached to QOM tree, by default
to path "/machine/unattached".

The test didn't crash simply because container_get() has an implicit
semantic to silently create any missing container, hence what happened here
is container_get() (when running these tests) will try to create containers
at QOM path "/machine" on the fly.  That's probably unexpected by the test,
but worked like charm before.

We're going to fix device_set_realized() soon, but before that make the
test case prepared, by creating the machine object on its own.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-ID: <20241121192202.4155849-4-peterx@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

qom: New object_property_add_new_container()

To move towards explicit creations of containers, starting that by
providing a helper for creating container objects.

Signed-off-by: Peter Xu <peterx@redhat.com>
Reviewed-by: Markus Armbruster <armbru@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Message-ID: <20241121192202.4155849-3-peterx@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

qom: Add TYPE_CONTAINER macro

Provide a macro for the container type across QEMU source tree, rather than
hard code it every time.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Peter Xu <peterx@redhat.com>
Message-ID: <20241121192202.4155849-2-peterx@redhat.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

log: Add separate debug option for logging invalid memory accesses

Currently -d guest_errors enables logging of different invalid actions
by the guest such as misusing hardware, accessing missing features or
invalid memory areas. The memory access logging can be quite verbose
which obscures the other messages enabled by this debug switch so
separate it by adding a new -d invalid_mem option to make it possible
to control it independently of other guest error logs.

Signed-off-by: BALATON Zoltan <balaton@eik.bme.hu>
Reviewed-by: Peter Maydell <peter.maydell@linaro.org>
Message-ID: <1bb0d0e91ba14aca13056df3b0a774f89cbf966c.1730549443.git.balaton@eik.bme.hu>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

hvf: arm: Ignore writes to CNTP_CTL_EL0

MacOS unconditionally disables interrupts of the physical timer on boot
and then continues to use the virtual one. We don't really want to support
a full physical timer emulation, so let's just ignore those writes.

Signed-off-by: Alexander Graf <graf@amazon.com>
Signed-off-by: Phil Dennis-Jordan <phil@philjordan.eu>
Reviewed-by: Akihiko Odaki <akihiko.odaki@daynix.com>
Message-ID: <20230830161425.91946-5-graf@amazon.com>
Signed-off-by: Philippe Mathieu-Daudé <philmd@linaro.org>

Merge tag 'pull-loongarch-20241219' of https://gitlab.com/bibo-mao/qemu into staging

loongarch queue

# -----BEGIN PGP SIGNATURE-----
#
# iHUEABYKAB0WIQQNhkKjomWfgLCz0aQfewwSUazn0QUCZ2PKBQAKCRAfewwSUazn
# 0QAZAQCxbLnvzOb9TPORlg5w0n/xFaKCL7dJbJE4WjlM7dhLkAEA5G8JVoP5Ju2B
# mcK7wbymyXNX1ocsukL/JM2JavHS+AI=
# =JoSk
# -----END PGP SIGNATURE-----
# gpg: Signature made Thu 19 Dec 2024 02:23:49 EST
# gpg:                using EDDSA key 0D8642A3A2659F80B0B3D1A41F7B0C1251ACE7D1
# gpg: Good signature from "bibo mao <maobibo@loongson.cn>" [unknown]
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: 7044 3A00 19C0 E97A 31C7  13C4 8E86 8FB7 A176 9D4C
#      Subkey fingerprint: 0D86 42A3 A265 9F80 B0B3  D1A4 1F7B 0C12 51AC E7D1

* tag 'pull-loongarch-20241219' of https://gitlab.com/bibo-mao/qemu:
  hw/intc/loongarch_extioi: Code cleanup about loongarch_extioi
  hw/intc/loongarch_extioi: Add pre_save interface
  hw/intc/loongarch_extioi: Inherit from loongarch_extioi_common
  hw/intc/loongarch_extioi: Add common file loongarch_extioi_common
  hw/intc/loongarch_extioi: Add unrealize interface
  hw/intc/loongarch_extioi: Add common realize interface
  hw/intc/loongarch_extioi: Rename LoongArchExtIOI with LoongArchExtIOICommonState
  include: Rename LoongArchExtIOI with LoongArchExtIOICommonState
  include: Move struct LoongArchExtIOI to header file loongarch_extioi_common
  include: Add loongarch_extioi_common header file
  hw/intc/loongarch_pch: Code cleanup about loongarch_pch_pic
  hw/intc/loongarch_pch: Add pre_save and post_load interfaces
  hw/intc/loongarch_pch: Inherit from loongarch_pic_common
  hw/intc/loongarch_pch: Move some functions to file loongarch_pic_common
  hw/intc/loongarch_pch: Rename LoongArchPCHPIC with LoongArchPICCommonState
  hw/intc/loongarch_pch: Merge instance_init() into realize()
  include: Move struct LoongArchPCHPIC to loongarch_pic_common header file
  include: Add loongarch_pic_common header file

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

hw/intc/loongarch_extioi: Code cleanup about loongarch_extioi

Remove definition about LoongArchExtIOI and LOONGARCH_EXTIOI, and
replace them with LoongArchExtIOICommonState and macro
LOONGARCH_EXTIOI_COMMON separately. Also remove unnecessary header
files.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_extioi: Add pre_save interface

Add vmstate pre_save interface, which can be used extioi kvm driver
in future.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_extioi: Inherit from loongarch_extioi_common

Set TYPE_LOONGARCH_EXTIOI inherit from TYPE_LOONGARCH_EXTIOI_COMMON
object, it shares vmsate and property of TYPE_LOONGARCH_EXTIOI_COMMON,
and has its own realize() function.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_extioi: Add common file loongarch_extioi_common

Add new common file loongarch_extioi_common.c, and move vmstate
and property structure to common file.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_extioi: Add unrealize interface

For loongarch extioi emulation driver, add unrealize interface and
remove instance_finalize interface and move the code to unrealize
interface.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_extioi: Add common realize interface

Add common realize function, it is only to check validity of property.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_extioi: Rename LoongArchExtIOI with LoongArchExtIOICommonState

With some structure such as vmstate and property, rename LoongArchExtIOI
with LoongArchExtIOICommonState, these common structure will be moved
to common file.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

include: Rename LoongArchExtIOI with LoongArchExtIOICommonState

Rename structure LoongArchExtIOI with LoongArchExtIOICommonState,
since it is defined in file loongarch_extioi_common.h

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

include: Move struct LoongArchExtIOI to header file loongarch_extioi_common

Move definiton of structure LoongArchExtIOI from header file loongarch_extioi.h
to file loongarch_extioi_common.h.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

include: Add loongarch_extioi_common header file

Add common header file include/hw/intc/loongarch_extioi_common.h, and
move some macro definition from include/hw/intc/loongarch_extioi.h to
the common header file.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_pch: Code cleanup about loongarch_pch_pic

Remove definition about LoongArchPCHPIC and LOONGARCH_PCH_PIC, and
replace them with LoongArchPICCommonState and LOONGARCH_PIC_COMMON
separately. Also remove unnecessary header files.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_pch: Add pre_save and post_load interfaces

Add vmstate pre_save and post_load interfaces, which can be used
by pic kvm driver in future.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_pch: Inherit from loongarch_pic_common

Set TYPE_LOONGARCH_PIC inherit from TYPE_LOONGARCH_PIC_COMMON object,
it shares vmsate and property of TYPE_LOONGARCH_PIC_COMMON, and has
its own realize() function.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_pch: Move some functions to file loongarch_pic_common

Move some common functions to file loongarch_pic_common.c, the common
functions include loongarch_pic_common_realize(), property structure
loongarch_pic_common_properties and vmstate structure
vmstate_loongarch_pic_common.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_pch: Rename LoongArchPCHPIC with LoongArchPICCommonState

With pic vmstate, rename structure name vmstate_loongarch_pch_pic with
vmstate_loongarch_pic_common, and with pic property rename
loongarch_pch_pic_properties with loongarch_pic_common_properties.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

hw/intc/loongarch_pch: Merge instance_init() into realize()

Memory region is created in instance_init(), merge it into function
realize(). There is no special class_init() for loongarch_pch object.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

include: Move struct LoongArchPCHPIC to loongarch_pic_common header file

Move structure LoongArchPCHPIC from header file loongarch_pch_pic.h
to file loongarch_pic_common.h, and rename structure name with
LoongArchPICCommonState.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

include: Add loongarch_pic_common header file

Add common header file hw/intc/loongarch_pic_common.h, and move
some macro definition from hw/intc/loongarch_pch_pic.h to the common
header file.

Signed-off-by: Bibo Mao <maobibo@loongson.cn>
Reviewed-by: Song Gao <gaosong@loongson.cn>

Merge tag 'qga-pull-2024-12-18' of https://github.com/kostyanf14/qemu into staging

qga-pull-2024-12-18

# -----BEGIN PGP SIGNATURE-----
#
# iQIzBAABCAAdFiEEwsLBCepDxjwUI+uE711egWG6hOcFAmdit4wACgkQ711egWG6
# hOfQ0xAAjByk0U6cTjIOTgNb3GDEE65CnL/dqiriOtmTPdbmVT4WqkQjPNG8REge
# lA0eDr7oBLWn/Yku7nqMjn4IcnBop92h9jL3RpQNeIbvPdQcpaPEORCCulky58l0
# 0hM/SnhBmRomZ2CwWu4mvl4bKRzn7WCg7mdPbBZEtvAfNdiUGOH5nwdUoJKtcPTj
# OeEOTxcN+FeYNyMRqCUJ4S44OAIh7WMbi4h/0gxufU8KFGYbQphk4xITIBFXBfIf
# oRxqwSMTLXG7nzegfad4HGX2pFUR6Omtmlgn3Cvh8UF8xEJIJIk1mA6Wd2R2am9j
# pdyKwo25K6FSkSvH43cH/IXTaQVvfCLIOOUA7rlAhfWrDsvRuisfh75T9jMOYH5S
# /jB9Vjd8E85qpTwQD2HuphJMzFRpWEeOeMJ2jWHWuMj+d3lHrR3hAmpGFBy7Xmxb
# KyVkUBgaulqoPQqt7C1urNxHjr6FoWDU1AqIAYQdqOJ5TLh/j5o6oz2ZJDOT+RiG
# NZLhTJWtjwA3Rbu1BcF2eyYv20cqeHVAFhS5tJdXOoHyqoJ7Hf+4G56zhXgjxfV9
# 60W3ETQwUPrrUtmb+qDZzmEagSQjj3UpMGEQFWXHKx4zBBXYeYU9O0Xvy04RGh+L
# PV4Y35yPegQ69GbRYUAagXk8Rh3Gbh2l4iFQz+QYCvd1b2CxYFg=
# =nY8c
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 18 Dec 2024 06:52:44 EST
# gpg:                using RSA key C2C2C109EA43C63C1423EB84EF5D5E8161BA84E7
# gpg: Good signature from "Kostiantyn Kostiuk (Upstream PR sign) <kkostiuk@redhat.com>" [unknown]
# gpg: WARNING: This key is not certified with a trusted signature!
# gpg:          There is no indication that the signature belongs to the owner.
# Primary key fingerprint: C2C2 C109 EA43 C63C 1423  EB84 EF5D 5E81 61BA 84E7

* tag 'qga-pull-2024-12-18' of https://github.com/kostyanf14/qemu:
  qga: Don't access global variable in run_agent_once()
  qemu-ga-win: Fix a typo error
  qga: skip bind mounts in fs list

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'pull-request-2024-12-18' of https://gitlab.com/thuth/qemu into staging

* Lots of functional test improvements (clean-ups, don't fail on
  temporary download errors, etc.)
* Convert some more avocado tests to the functional framework
* Disallow building with libnfs v6 due to an API breakage

# -----BEGIN PGP SIGNATURE-----
#
# iQJFBAABCAAvFiEEJ7iIR+7gJQEY8+q5LtnXdP5wLbUFAmdirOIRHHRodXRoQHJl
# ZGhhdC5jb20ACgkQLtnXdP5wLbU0NRAAke8X0B6OOD+99lY5nc7Hrh7N1m+sw5Lw
# TVwIpxdhxU11vgdlCodfdoVJCV1NGVHwkR57lLNr+bdspWDBBwlmUWn0+t2QCXGe
# oyQsV+boznsjG9pan6v6DcU/gOu7/7ZydhJi+M8Msf8ah0lcn/otAdC4ZFB93JLh
# 6xPnj69y8HomCW+wMyXl7WTjcWX0wQFzweEYY8p7X7p1rtjYyseiZlRjNAvPgTMI
# jznZ6v9/qU54xR9RnKdW+0m1Qu06nx26Wz+ZBlvrJS1Llloe23X9+LY1tDD0Xh1D
# 9P0v9PuaBWRRF+UjVjl37LMyn9h1aaKFKBoWQiKMbyvOVr4ncobjRgN8r5kdNxDP
# FZ/fA1GiX8O3foN9uB9JLKd6Hl49LAqQSPzAneEc3pfQLH3NdAjPxJDbJH5fyMa7
# qVOQC0Bdy8+2kCxFfKbemrwDOFcyq1fVYcADPDZySjMiPnwFJ1Qpni1tXY1PZ+Tl
# Q18AsFJanyAAn7L+8R3Yl54983SuR5eXIFxO+Tq9mw1V1V2h+Cm09HGcS8y5bxFG
# Xh+jhMsMB98NFLR87W6olwl57gKllSbTYuGtiz9TrbnuT/THhUJ0k/B76L7C9HWE
# ZefkFxC5Zy8jrcz3pgarO+19V+eXg5rwGtEngRQrji/3cY5CbK7Jeh5nvZQeASpb
# nZ/gJ/gC8Gs=
# =SWw6
# -----END PGP SIGNATURE-----
# gpg: Signature made Wed 18 Dec 2024 06:07:14 EST
# gpg:                using RSA key 27B88847EEE0250118F3EAB92ED9D774FE702DB5
# gpg:                issuer "thuth@redhat.com"
# gpg: Good signature from "Thomas Huth <th.huth@gmx.de>" [full]
# gpg:                 aka "Thomas Huth <thuth@redhat.com>" [full]
# gpg:                 aka "Thomas Huth <huth@tuxfamily.org>" [full]
# gpg:                 aka "Thomas Huth <th.huth@posteo.de>" [unknown]
# Primary key fingerprint: 27B8 8847 EEE0 2501 18F3  EAB9 2ED9 D774 FE70 2DB5

* tag 'pull-request-2024-12-18' of https://gitlab.com/thuth/qemu: (38 commits)
  meson.build: Disallow libnfs v6 to fix the broken macOS build
  tests/functional: Convert the hotplug_cpu avocado test
  tests/functional: Convert the intel_iommu avocado test
  tests/functional: Add a helper function for retrieving the hostfwd port
  tests/functional: Convert the arm virt avocado test
  tests/functional: Convert the quanta-gsj avocado test
  MAINTAINERS: add myself as reviewer for functional test suite
  tests/functional: ignore errors when caching assets, except for 404
  tests/functional: skip tests if assets are not available
  tests/functional: remove now unused 'run_cmd' helper
  tests/functional: replace 'run_cmd' with subprocess helpers
  tests/functional: drop back compat imports from utils.py
  tests/functional: convert tests to new uncompress helper
  tests/functional: add 'uncompress' to QemuBaseTest
  tests/functional: add a generalized uncompress helper
  tests/functional: convert tests to new archive_extract helper
  tests/functional: add 'archive_extract' to QemuBaseTest
  tests/functional: add a generalized archive_extract
  tests/functional: let cpio_extract accept filenames
  tests/functional: add common deb_extract helper
  ...

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

Merge tag 'pull-target-arm-20241217' of https://git.linaro.org/people/pmaydell/qemu-arm into staging

target-arm queue:
 * remove a line of redundant code
 * convert various TCG helper fns to use 'fpst' alias
 * Use float_status in helper_fcvtx_f64_to_f32
 * Use float_status in helper_vfp_fcvt{ds,sd}
 * Implement FEAT_XS
 * hw/intc/arm_gicv3_its: Zero initialize local DTEntry etc structs
 * tests/functional: update sbsa-ref firmware used in test

# -----BEGIN PGP SIGNATURE-----
#
# iQJNBAABCAA3FiEE4aXFk81BneKOgxXPPCUl7RQ2DN4FAmdhsj0ZHHBldGVyLm1h
# eWRlbGxAbGluYXJvLm9yZwAKCRA8JSXtFDYM3vldEACY/BJeDSfE3X6fi2qiVF+p
# hs2NaeiM6HS3+/1oPlPz7sW87cGV58hPW0vOCjZYlfw+Afv2wuL9zb2C7IV+PJOn
# Vy1L/tBezYvgBwi1+bRvLpCge3UKdbRNa6aowGmLnvcnQuZ6lOOcmQ8Jq/a5W1TM
# xOrvPYp0FbcfGXLcDLIluCozupsq4aJsh0gWayzr9zm2tWnzMAhb/GQLuCmLLn35
# pUiAI209xU393AOfdCpAmCGjDCqcqbjHpz0AqrIPtOwaDO3hRlJIMw1eGk2dS0BD
# R0vZG+WFSQMV972reVoTT83W7NmIeeqhZgeKDv2R347EHsbCKt5PIUpWX2zYvk+H
# rfCltOxFVWvJ0e33b9opk2/GfgckkTGw+6ZYMIdMm8UxABrdEaHcWCV0qMyLk2JH
# 4EKYlmKeR8yqkPRDbGkRVANEIDxRBmL96kN6il0wSM742y7UMwMzP3C344Jg0tf/
# AhGrwjsuRW8oeEZMgk9Z0i/J6q3CNxRQSVGGQtYsEt8fs5OXLltLrXEX+aZNF5ua
# ry5SfLWlwIR+0AO4oNaJqJYNQArqhzDUgsY4ryzrueZnaaMShobMn0AP2H3+/l4X
# W3wlIqOQ97ivk0Snc9WpDQyhPPOZuj4LN4IkTkodHu7+eoMrkvojf/BVc5kku2VL
# dE224ctbEbKsbydwubSVfQ==
# =Z/7q
# -----END PGP SIGNATURE-----
# gpg: Signature made Tue 17 Dec 2024 12:17:49 EST
# gpg:                using RSA key E1A5C593CD419DE28E8315CF3C2525ED14360CDE
# gpg:                issuer "peter.maydell@linaro.org"
# gpg: Good signature from "Peter Maydell <peter.maydell@linaro.org>" [full]
# gpg:                 aka "Peter Maydell <pmaydell@gmail.com>" [full]
# gpg:                 aka "Peter Maydell <pmaydell@chiark.greenend.org.uk>" [full]
# gpg:                 aka "Peter Maydell <peter@archaic.org.uk>" [unknown]
# Primary key fingerprint: E1A5 C593 CD41 9DE2 8E83  15CF 3C25 25ED 1436 0CDE

* tag 'pull-target-arm-20241217' of https://git.linaro.org/people/pmaydell/qemu-arm:
  tests/functional: update sbsa-ref firmware used in test
  hw/intc/arm_gicv3_its: Zero initialize local DTEntry etc structs
  tests/tcg/aarch64: add system test for FEAT_XS
  target/arm: Enable FEAT_XS for the max cpu
  target/arm: Add decodetree entry for DSB nXS variant
  target/arm: Add ARM_CP_ADD_TLBI_NXS type flag to TLBI insns
  target/arm: Add ARM_CP_ADD_TLBI_NXS type flag for NXS insns
  target/arm: Implement fine-grained-trap handling for FEAT_XS
  target/arm: Use float_status in helper_vfp_fcvt{ds,sd}
  target/arm: Use float_status in helper_fcvtx_f64_to_f32
  target/arm: Convert neon_helper.c to use env alias
  target/arm: Convert vec_helper.c to use env alias
  target/arm: Convert sme_helper.c to fpst alias
  target/arm: Convert sve_helper.c to fpst alias
  target/arm: Convert neon_helper.c to fpst alias
  target/arm: Convert vec_helper.c to fpst alias
  target/arm: Convert helper-a64.c to fpst alias
  target/arm: Convert vfp_helper.c to fpst alias
  target/arm: remove redundant code

Signed-off-by: Stefan Hajnoczi <stefanha@redhat.com>

qga: Don't access global variable in run_agent_once()

The run_agent_once() function is already given GAState via an
argument. There's no need to access the global ga_state variable
which points to the argument anyways (thanks to
initialize_agent()). Worse, some parts of the function use the
argument and the other use the global variable.  Stick with the
function argument.

Signed-off-by: Michal Privoznik <mprivozn@redhat.com>
Reviewed-by: Ján Tomko <jtomko@redhat.com>
Reviewed-by: Konstantin Kostiuk <kkostiuk@redhat.com>
Link: https://lore.kernel.org/r/8ae7f5d5032b14a5b956fe8aaf47bae5ca401699.1733414906.git.mprivozn@redhat.com
Signed-off-by: Konstantin Kostiuk <kkostiuk@redhat.com>

qemu-ga-win: Fix a typo error

There is a typo error for api 'guest-get-osinfo',
the win2025's version in WIN_10_0_SERVER_VERSION_MATRIX
should be adjusted.

Signed-off-by: Dehan Meng <demeng@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Konstantin Kostiuk <kkostiuk@redhat.com>
Link: https://lore.kernel.org/r/20241210054616.260386-1-demeng@redhat.com
Signed-off-by: Konstantin Kostiuk <kkostiuk@redhat.com>

qga: skip bind mounts in fs list

The filesystem list in build_fs_mount_list should skip bind mounts.
This because we end up in locking situations when doing fsFreeze. Like
mentioned in [1] and [2].

Next to that, the build_fs_mount_list call did a fallback via
build_fs_mount_list_from_mtab if mountinfo did not exist.
There it skipped bind mounts, but this is broken for newer OS.
This as mounts does not return the path of the bind mount but the
underlying dev/partition, so S_ISDIR will never return true in
dev_major_minor call.

This patch simply checks the existing devmajor:devminor tuple in the
mounts, and if it already exists, this means we have the same devices
mounted again, a bind mount. So skip this.

Same approach is used in open-vm-tools [3].

[1]: https://gitlab.com/qemu-project/qemu/-/issues/592
[2]: https://gitlab.com/qemu-project/qemu/-/issues/520
[3]: https://github.com/vmware/open-vm-tools/commit/d58847b497e212737007958c945af1df22a8ab58

Signed-off-by: Jean-Louis Dupond <jean-louis@dupond.be>
Reviewed-by: Konstantin Kostiuk <kkostiuk@redhat.com>
Link: https://lore.kernel.org/r/20241002100634.162499-2-jean-louis@dupond.be
Signed-off-by: Konstantin Kostiuk <kkostiuk@redhat.com>

meson.build: Disallow libnfs v6 to fix the broken macOS build

The macOS builds in our CI (and possibly other very recent distros)
are currently broken since the update to libnfs version 6 there.
That version apparently comes with a big API breakage. v5.0.3 was
the final release of the old API (see the libnfs commit here:
https://github.com/sahlberg/libnfs/commit/4379837 ).

Disallow version 6.x for now to get the broken CI job working
again. Once somebody had enough time to adapt our code in
block/nfs.c, we can revert this change again.

Message-ID: <20241218065157.209020-1-thuth@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: Convert the hotplug_cpu avocado test

Since we don't have ssh support in the functional test framework yet,
simply use the serial console for this test instead. It's also
sufficient to only boot into an initrd here, no need to fire up a
full-blown guest, so the test now finishes much faster.

While we're at it, also unplug the CPU now and check that it is gone
in the guest.

Message-ID: <20241217142020.155776-1-thuth@redhat.com>
Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: Convert the intel_iommu avocado test

Convert the intel_iommu test to the new functional framework.
This test needs some changes since we neither support the old 'LinuxTest'
class in the functional framework yet, nor a way to use SSH for running
commands in the guest. So we now directly download a Fedora kernel and
initrd and set up the serial console for executing the commands and for
looking for the results. Instead of configuring the cloud image via
cloud-init, we now simply mount the file system manually from an initrd
rescue shell.

While the old test was exercising the network with a "dnf install"
command (which is not the best option for the CI since this depends
on third party servers), the new code is now setting up a little
HTTP server in the guest and transfers a file from the guest to the
host instead.

The test should now run much faster and more reliable (since we
don't depend on the third party servers for "dnf install" anymore),
so we can also drop the @skipUnless decorator now.

Message-ID: <20241217121550.141072-3-thuth@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: Add a helper function for retrieving the hostfwd port

It's just a wrapper around get_info_usernet_hostfwd_port from the
qemu module that is also calling the right monitor command for
retrieving the information from QEMU.

Message-ID: <20241217121550.141072-2-thuth@redhat.com>
Acked-by: Michael S. Tsirkin <mst@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: Convert the arm virt avocado test

Straight forward conversion, basically just the hashsums needed
to be updated to sha256 now.

Message-ID: <20241206102358.1186644-7-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: Convert the quanta-gsj avocado test

Straight forward conversion, basically just the hashsums needed
to be updated to sha256 now.

Message-ID: <20241206102358.1186644-6-thuth@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

MAINTAINERS: add myself as reviewer for functional test suite

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-33-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: ignore errors when caching assets, except for 404

We see periodic errors caching assets due to a combination of transient
networking and server problems. With the previous patch to skip running
a test when it has missing assets, we can now treat most cache download
errors as non-fatal.

Only HTTP 404 is retained as fatal, since it is a strong indicator of
a fully broken test rather than a transient error.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-32-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: skip tests if assets are not available

If downloading of assets has been disabled, then skip running a
test if the assets it has registered are not already downloaded.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-31-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: remove now unused 'run_cmd' helper

All usage has been replaced by direct 'subprocess' helpers.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-30-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: replace 'run_cmd' with subprocess helpers

The 'run_cmd' helper is re-implementing a convenient helper that
already exists in the form of the 'run' and 'check_call' methods
provided by 'subprocess'.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-29-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: drop back compat imports from utils.py

Now that all tests are converted over to the higher level wrapper
functions, the back compat imports from utils.py are redundant.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-28-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: convert tests to new uncompress helper

Replace use of lzma_uncompress and gzip_uncompress with the
new uncompress helper.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-27-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: add 'uncompress' to QemuBaseTest

This helper wrappers utils.uncompress, forcing the use of the scratch
directory, to ensure any uncompressed files are cleaned at test
termination.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-26-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: add a generalized uncompress helper

There are many types of compression that the tests deal with, and
it makes sense to have a single helper 'uncompress' that can deal
with all.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-25-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: convert tests to new archive_extract helper

Replace use of utils.archive_extract and extract_from_deb with the
new archive_extract helper.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-24-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: add 'archive_extract' to QemuBaseTest

This helper wrappers archive.archive_extract, forcing the use of the
scratch directory, to ensure any extracted files are cleaned at test
termination. If a specific member is requested, then the path to the
extracted file is also returned.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-23-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: add a generalized archive_extract

There are many types of archives that the tests deal with. Provide
a generalized 'archive_extract' that can detect the format and
delegate to the appropriate helper for extraction. This ensures
that all archive extraction code follows the same design pattern.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-22-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: let cpio_extract accept filenames

Currently cpio_extract differs from tar_extract/zip_extract
in that it only allows a file-like object as input. Adapt it
to also support filenames.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-21-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: add common deb_extract helper

This mirrors the existing archive_extract, cpio_extract and zip_extract
helpers

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-20-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: add common zip_extract helper

This mirrors the existing archive_extract and cpio_extract helpers

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-19-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>

tests/functional: move uncompress handling into new uncompress.py file

More uncompress related code will be added shortly, so having a
separate file makes more sense.

The utils.py imports the functions from archive.py, so that
existing callers don't need to be modified. This avoids
redundant code churn until later in the series when all
calls will be adapted for other reasons.

Reviewed-by: Thomas Huth <thuth@redhat.com>
Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Message-ID: <20241217155953.3950506-18-berrange@redhat.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>