From: Peter Maydell <peter.maydell@linaro.org>
To: Richard Henderson <richard.henderson@linaro.org>
Cc: QEMU Developers <qemu-devel@nongnu.org>,
Eduardo Habkost <ehabkost@redhat.com>
Subject: Re: [PATCH v3 50/66] hw/core/cpu: Move cpu properties to cpu-sysemu.c
Date: Thu, 19 Aug 2021 16:26:10 +0100 [thread overview]
Message-ID: <CAFEAcA9Wdriiz-8Km2s9c3cZCVnov3Z2OY6wpP6o3e6cA2qRxw@mail.gmail.com> (raw)
In-Reply-To: <20210818191920.390759-51-richard.henderson@linaro.org>
On Wed, 18 Aug 2021 at 21:09, Richard Henderson
<richard.henderson@linaro.org> wrote:
>
> The comment in cpu-common.c is absolutely correct, we can't
> rely on the ifdef in a file built once. This was only "working"
> because we used ifndef.
>
> Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Fixes: 1b36e4f5a5de585
which moved the properties out of cpu.c and into cpu-common.c
with the remark "There's no reason to keep the property list
separate from the CPU class code" despite there being a big
fat warning comment saying why it can't go in a compiled-once
source file !
Is there a reason to prefer this patch over just reverting
1b36e4f5a5de585 ?
thanks
-- PMM
(patch below for context)
> ---
> hw/core/cpu-common.h | 17 +++++++++++++++++
> hw/core/cpu-common.c | 18 ++----------------
> hw/core/cpu-sysemu.c | 24 ++++++++++++++++++++++++
> hw/core/cpu-user.c | 22 ++++++++++++++++++++++
> hw/core/meson.build | 4 ++++
> 5 files changed, 69 insertions(+), 16 deletions(-)
> create mode 100644 hw/core/cpu-common.h
> create mode 100644 hw/core/cpu-user.c
>
> diff --git a/hw/core/cpu-common.h b/hw/core/cpu-common.h
> new file mode 100644
> index 0000000000..9049f74e67
> --- /dev/null
> +++ b/hw/core/cpu-common.h
> @@ -0,0 +1,17 @@
> +/*
> + * QEMU CPU model
> + *
> + * Copyright (c) 2014 SUSE LINUX Products GmbH
> + *
> + * SPDX-License-Identifier: GPL-2.0-or-later
> + *
> + * This work is licensed under the terms of the GNU GPL, version 2 or later.
> + * See the COPYING file in the top-level directory.
> + */
> +
> +#ifndef HW_CORE_CPU_COMMON_H
> +#define HW_CORE_CPU_COMMON_H
> +
> +void cpu_class_init_props(DeviceClass *dc);
> +
> +#endif
> diff --git a/hw/core/cpu-common.c b/hw/core/cpu-common.c
> index e2f5a64604..6406ea79f0 100644
> --- a/hw/core/cpu-common.c
> +++ b/hw/core/cpu-common.c
> @@ -34,6 +34,7 @@
> #include "hw/qdev-properties.h"
> #include "trace/trace-root.h"
> #include "qemu/plugin.h"
> +#include "cpu-common.h"
>
> CPUState *cpu_by_arch_id(int64_t id)
> {
> @@ -257,21 +258,6 @@ static int64_t cpu_common_get_arch_id(CPUState *cpu)
> return cpu->cpu_index;
> }
>
> -static Property cpu_common_props[] = {
> -#ifndef CONFIG_USER_ONLY
> - /* Create a memory property for softmmu CPU object,
> - * so users can wire up its memory. (This can't go in hw/core/cpu.c
> - * because that file is compiled only once for both user-mode
> - * and system builds.) The default if no link is set up is to use
> - * the system address space.
> - */
> - DEFINE_PROP_LINK("memory", CPUState, memory, TYPE_MEMORY_REGION,
> - MemoryRegion *),
> -#endif
> - DEFINE_PROP_BOOL("start-powered-off", CPUState, start_powered_off, false),
> - DEFINE_PROP_END_OF_LIST(),
> -};
> -
> static void cpu_class_init(ObjectClass *klass, void *data)
> {
> DeviceClass *dc = DEVICE_CLASS(klass);
> @@ -286,7 +272,7 @@ static void cpu_class_init(ObjectClass *klass, void *data)
> dc->realize = cpu_common_realizefn;
> dc->unrealize = cpu_common_unrealizefn;
> dc->reset = cpu_common_reset;
> - device_class_set_props(dc, cpu_common_props);
> + cpu_class_init_props(dc);
> /*
> * Reason: CPUs still need special care by board code: wiring up
> * IRQs, adding reset handlers, halting non-first CPUs, ...
> diff --git a/hw/core/cpu-sysemu.c b/hw/core/cpu-sysemu.c
> index 00253f8929..5a62960f3c 100644
> --- a/hw/core/cpu-sysemu.c
> +++ b/hw/core/cpu-sysemu.c
> @@ -20,8 +20,11 @@
>
> #include "qemu/osdep.h"
> #include "qapi/error.h"
> +#include "exec/memory.h"
> #include "hw/core/cpu.h"
> #include "hw/core/sysemu-cpu-ops.h"
> +#include "hw/qdev-properties.h"
> +#include "cpu-common.h"
>
> bool cpu_paging_enabled(const CPUState *cpu)
> {
> @@ -143,3 +146,24 @@ GuestPanicInformation *cpu_get_crash_info(CPUState *cpu)
> }
> return res;
> }
> +
> +/*
> + * This can't go in hw/core/cpu-common.c because that file is compiled only
> + * once for both user-mode and system builds.
> + */
> +static Property cpu_sysemu_props[] = {
> + /*
> + * Create a memory property for softmmu CPU object, so users can wire
> + * up its memory. The default if no link is set up is to use the
> + * system address space.
> + */
> + DEFINE_PROP_LINK("memory", CPUState, memory, TYPE_MEMORY_REGION,
> + MemoryRegion *),
> + DEFINE_PROP_BOOL("start-powered-off", CPUState, start_powered_off, false),
> + DEFINE_PROP_END_OF_LIST(),
> +};
> +
> +void cpu_class_init_props(DeviceClass *dc)
> +{
> + device_class_set_props(dc, cpu_sysemu_props);
> +}
> diff --git a/hw/core/cpu-user.c b/hw/core/cpu-user.c
> new file mode 100644
> index 0000000000..62037e8669
> --- /dev/null
> +++ b/hw/core/cpu-user.c
> @@ -0,0 +1,22 @@
> +/*
> + * QEMU CPU model (user-only emulation specific)
> + *
> + * Copyright (c) 2021 Linaro, Ltd.
> + *
> + * SPDX-License-Identifier: GPL-2.0-or-later
> + *
> + * This work is licensed under the terms of the GNU GPL, version 2 or later.
> + * See the COPYING file in the top-level directory.
> + */
> +
> +#include "qemu/osdep.h"
> +#include "hw/core/cpu.h"
> +#include "cpu-common.h"
> +
> +/*
> + * This can't go in hw/core/cpu-common.c because that file is compiled only
> + * once for both user-mode and system builds.
> + */
> +void cpu_class_init_props(DeviceClass *dc)
> +{
> +}
> diff --git a/hw/core/meson.build b/hw/core/meson.build
> index 18f44fb7c2..4b3de4e247 100644
> --- a/hw/core/meson.build
> +++ b/hw/core/meson.build
> @@ -43,3 +43,7 @@ specific_ss.add(when: 'CONFIG_SOFTMMU', if_true: files(
> 'machine-qmp-cmds.c',
> 'numa.c',
> ))
> +
> +specific_ss.add(when: 'CONFIG_USER_ONLY', if_true: files(
> + 'cpu-user.c'
> +))
> --
next prev parent reply other threads:[~2021-08-19 15:27 UTC|newest]
Thread overview: 108+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-08-18 19:18 [PATCH v3 00/66] Unaligned access for user-only Richard Henderson
2021-08-18 19:18 ` [PATCH v3 01/66] util: Suppress -Wstringop-overflow in qemu_thread_start Richard Henderson
2021-08-19 15:13 ` Peter Maydell
2021-08-18 19:18 ` [PATCH v3 02/66] hw/core: Make do_unaligned_access noreturn Richard Henderson
2021-08-19 6:15 ` Alistair Francis
2021-08-18 19:18 ` [PATCH v3 03/66] hw/core: Make do_unaligned_access available to user-only Richard Henderson
2021-08-18 19:18 ` [PATCH v3 04/66] target/alpha: Implement do_unaligned_access for user-only Richard Henderson
2021-08-18 19:18 ` [PATCH v3 05/66] target/arm: " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 06/66] target/hppa: " Richard Henderson
2021-08-19 15:32 ` Peter Maydell
2021-08-18 19:18 ` [PATCH v3 07/66] target/microblaze: Do not set MO_ALIGN " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 08/66] target/mips: Implement do_unaligned_access " Richard Henderson
2021-08-19 15:34 ` Peter Maydell
2021-08-18 19:18 ` [PATCH v3 09/66] target/ppc: Move SPR_DSISR setting to powerpc_excp Richard Henderson
2021-08-19 15:39 ` Peter Maydell
2021-08-19 19:13 ` Richard Henderson
2021-08-18 19:18 ` [PATCH v3 10/66] target/ppc: Set fault address in ppc_cpu_do_unaligned_access Richard Henderson
2021-08-19 15:41 ` Peter Maydell
2021-08-18 19:18 ` [PATCH v3 11/66] target/ppc: Implement do_unaligned_access for user-only Richard Henderson
2021-08-19 15:44 ` Peter Maydell
2021-08-18 19:18 ` [PATCH v3 12/66] target/riscv: " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 13/66] target/s390x: " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 14/66] target/sh4: Set fault address in superh_cpu_do_unaligned_access Richard Henderson
2021-08-18 19:18 ` [PATCH v3 15/66] target/sh4: Implement do_unaligned_access for user-only Richard Henderson
2021-08-19 15:46 ` Peter Maydell
2021-08-19 19:21 ` Richard Henderson
2021-08-18 19:18 ` [PATCH v3 16/66] target/sparc: Remove DEBUG_UNALIGNED Richard Henderson
2021-08-18 19:18 ` [PATCH v3 17/66] target/sparc: Split out build_sfsr Richard Henderson
2021-08-18 19:18 ` [PATCH v3 18/66] target/sparc: Set fault address in sparc_cpu_do_unaligned_access Richard Henderson
2021-08-18 19:18 ` [PATCH v3 19/66] target/sparc: Implement do_unaligned_access for user-only Richard Henderson
2021-08-18 19:18 ` [PATCH v3 20/66] target/xtensa: " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 21/66] accel/tcg: Report unaligned atomics " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 22/66] accel/tcg: Drop signness in tracing in cputlb.c Richard Henderson
2021-08-18 21:14 ` Philippe Mathieu-Daudé
2021-08-18 19:18 ` [PATCH v3 23/66] tcg: Expand MO_SIZE to 3 bits Richard Henderson
2021-08-19 6:17 ` Alistair Francis
2021-08-18 19:18 ` [PATCH v3 24/66] tcg: Rename TCGMemOpIdx to MemOpIdx Richard Henderson
2021-08-19 6:17 ` Alistair Francis
2021-08-18 19:18 ` [PATCH v3 25/66] tcg: Split out MemOpIdx to exec/memopidx.h Richard Henderson
2021-08-18 19:18 ` [PATCH v3 26/66] trace/mem: Pass MemOpIdx to trace_mem_get_info Richard Henderson
2021-08-19 15:49 ` Peter Maydell
2021-08-18 19:18 ` [PATCH v3 27/66] accel/tcg: Pass MemOpIdx to atomic_trace_*_post Richard Henderson
2021-08-18 19:18 ` [PATCH v3 28/66] plugins: Reorg arguments to qemu_plugin_vcpu_mem_cb Richard Henderson
2021-08-30 21:42 ` Philippe Mathieu-Daudé
2021-08-18 19:18 ` [PATCH v3 29/66] trace: Split guest_mem_before Richard Henderson
2021-08-18 19:18 ` [PATCH v3 30/66] target/arm: Use MO_128 for 16 byte atomics Richard Henderson
2021-08-18 19:18 ` [PATCH v3 31/66] target/i386: " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 32/66] target/ppc: " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 33/66] target/s390x: " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 34/66] target/hexagon: Implement cpu_mmu_index Richard Henderson
2021-08-18 19:18 ` [PATCH v3 35/66] accel/tcg: Add cpu_{ld,st}*_mmu interfaces Richard Henderson
2021-08-19 15:57 ` Peter Maydell
2021-08-18 19:18 ` [PATCH v3 36/66] accel/tcg: Move cpu_atomic decls to exec/cpu_ldst.h Richard Henderson
2021-08-18 19:18 ` [PATCH v3 37/66] target/mips: Use cpu_*_data_ra for msa load/store Richard Henderson
2021-08-18 19:18 ` [PATCH v3 38/66] target/mips: Use 8-byte memory ops " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 39/66] target/s390x: Use cpu_*_mmu instead of helper_*_mmu Richard Henderson
2021-08-18 19:18 ` [PATCH v3 40/66] target/sparc: " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 41/66] target/arm: " Richard Henderson
2021-08-18 19:18 ` [PATCH v3 42/66] tcg: Move helper_*_mmu decls to tcg/tcg-ldst.h Richard Henderson
2021-08-18 19:18 ` [PATCH v3 43/66] tcg: Add helper_unaligned_{ld, st} for user-only sigbus Richard Henderson
2021-08-19 15:58 ` Peter Maydell
2021-08-18 19:18 ` [PATCH v3 44/66] tcg/i386: Support raising sigbus for user-only Richard Henderson
2021-08-19 16:02 ` Peter Maydell
2021-08-18 19:18 ` [PATCH v3 45/66] tests/tcg/multiarch: Add sigbus.c Richard Henderson
2021-08-19 16:04 ` Peter Maydell
2021-08-18 19:19 ` [PATCH v3 46/66] linux-user: Split out do_prctl and subroutines Richard Henderson
2021-08-19 16:06 ` Peter Maydell
2021-08-19 19:30 ` Richard Henderson
2021-08-18 19:19 ` [PATCH v3 47/66] linux-user: Disable more prctl subcodes Richard Henderson
2021-08-18 19:19 ` [PATCH v3 48/66] hw/core/cpu: Re-sort the non-pointers to the end of CPUClass Richard Henderson
2021-08-18 21:17 ` Philippe Mathieu-Daudé
2021-08-18 19:19 ` [PATCH v3 49/66] linux-user: Add code for PR_GET/SET_UNALIGN Richard Henderson
2021-08-18 19:19 ` [PATCH v3 50/66] hw/core/cpu: Move cpu properties to cpu-sysemu.c Richard Henderson
2021-08-19 15:26 ` Peter Maydell [this message]
2021-08-19 16:52 ` Eduardo Habkost
2021-08-18 19:19 ` [PATCH v3 51/66] hw/core/cpu: Add prctl-unalign-sigbus property for user-only Richard Henderson
2021-08-18 19:19 ` [PATCH v3 52/66] target/alpha: Reorg fp memory operations Richard Henderson
2021-08-18 21:21 ` Philippe Mathieu-Daudé
2021-08-18 19:19 ` [PATCH v3 53/66] target/alpha: Reorg integer " Richard Henderson
2021-08-20 9:29 ` Peter Maydell
2021-08-18 19:19 ` [PATCH v3 54/66] target/alpha: Implement prctl_unalign_sigbus Richard Henderson
2021-08-18 19:19 ` [PATCH v3 55/66] target/hppa: " Richard Henderson
2021-08-18 19:19 ` [PATCH v3 56/66] target/sh4: " Richard Henderson
2021-08-18 19:19 ` [PATCH v3 57/66] accel/tcg/user-exec: Convert DEBUG_SIGNAL to tracepoint Richard Henderson
2021-08-18 21:22 ` Philippe Mathieu-Daudé
2021-08-18 19:19 ` [PATCH v3 58/66] include/exec: Move cpu_signal_handler declaration Richard Henderson
2021-08-18 21:23 ` Philippe Mathieu-Daudé
2021-08-19 6:18 ` Alistair Francis
2021-08-18 19:19 ` [PATCH v3 59/66] accel/tcg: Handle SIGBUS in handle_cpu_signal Richard Henderson
2021-08-20 9:34 ` Peter Maydell
2021-08-22 7:48 ` Richard Henderson
2021-08-18 19:19 ` [PATCH v3 60/66] tcg/aarch64: Support raising sigbus for user-only Richard Henderson
2021-08-20 9:46 ` Peter Maydell
2021-08-18 19:19 ` [PATCH v3 61/66] tcg/ppc: " Richard Henderson
2021-08-20 10:11 ` Peter Maydell
2021-08-18 19:19 ` [PATCH v3 62/66] tcg/s390: " Richard Henderson
2021-08-20 10:12 ` Peter Maydell
2021-08-18 19:19 ` [PATCH v3 63/66] tcg/tci: " Richard Henderson
2021-08-20 10:14 ` Peter Maydell
2021-08-22 7:59 ` Richard Henderson
2021-08-22 12:32 ` Peter Maydell
2021-08-22 17:09 ` Richard Henderson
2021-08-18 19:19 ` [PATCH v3 64/66] tcg: Canonicalize alignment flags in MemOp Richard Henderson
2021-08-18 21:24 ` Philippe Mathieu-Daudé
2021-08-18 19:19 ` [PATCH v3 65/66] tcg/riscv: Support raising sigbus for user-only Richard Henderson
2021-08-18 19:19 ` [PATCH v3 66/66] tcg/riscv: Remove add with zero on user-only memory access Richard Henderson
2021-08-30 21:29 ` Philippe Mathieu-Daudé
2021-08-30 22:38 ` Alistair Francis
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=CAFEAcA9Wdriiz-8Km2s9c3cZCVnov3Z2OY6wpP6o3e6cA2qRxw@mail.gmail.com \
--to=peter.maydell@linaro.org \
--cc=ehabkost@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=richard.henderson@linaro.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).