qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Claudio Fontana <cfontana@suse.de>
To: "Peter Maydell" <peter.maydell@linaro.org>,
	"Philippe Mathieu-Daudé" <philmd@redhat.com>,
	"Richard Henderson" <richard.henderson@linaro.org>,
	"Alex Bennée" <alex.bennee@linaro.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>,
	Roman Bolshakov <r.bolshakov@yadro.com>,
	Claudio Fontana <cfontana@suse.de>,
	Eduardo Habkost <ehabkost@redhat.com>,
	qemu-devel@nongnu.org
Subject: [RFC v8 40/44] accel: add double dispatch mechanism for class initialization
Date: Tue, 16 Mar 2021 19:36:58 +0100	[thread overview]
Message-ID: <20210316183702.10216-41-cfontana@suse.de> (raw)
In-Reply-To: <20210316183702.10216-1-cfontana@suse.de>

while on x86 all CPU classes can use the same set of TCGCPUOps,
on ARM the right accel behavior depends on the type of the CPU.

So we need a way to specialize the accel behavior according to
the CPU. Therefore, add a second initialization, after the
accel_cpu->cpu_class_init, that allows to do this.

Signed-off-by: Claudio Fontana <cfontana@suse.de>
Cc: Paolo Bonzini <pbonzini@redhat.com>
---
 include/hw/core/cpu.h     |  6 ++++++
 accel/accel-common.c      | 12 ++++++++++++
 target/i386/tcg/tcg-cpu.c |  8 +++++++-
 3 files changed, 25 insertions(+), 1 deletion(-)

diff --git a/include/hw/core/cpu.h b/include/hw/core/cpu.h
index c68bc3ba8a..d45f78290e 100644
--- a/include/hw/core/cpu.h
+++ b/include/hw/core/cpu.h
@@ -192,6 +192,12 @@ struct CPUClass {
 
     /* when TCG is not available, this pointer is NULL */
     struct TCGCPUOps *tcg_ops;
+
+    /*
+     * if not NULL, this is called in order for the CPUClass to initialize
+     * class data that depends on the accelerator, see accel/accel-common.c.
+     */
+    void (*init_accel_cpu)(struct AccelCPUClass *accel_cpu, CPUClass *cc);
 };
 
 /*
diff --git a/accel/accel-common.c b/accel/accel-common.c
index d77c09d7b5..2103397d5e 100644
--- a/accel/accel-common.c
+++ b/accel/accel-common.c
@@ -54,10 +54,22 @@ static void accel_init_cpu_int_aux(ObjectClass *klass, void *opaque)
     CPUClass *cc = CPU_CLASS(klass);
     AccelCPUClass *accel_cpu = opaque;
 
+    /*
+     * double dispatch. The first callback allows the accel cpu
+     * to run initializations for the CPU,
+     * the second one allows the CPU to customize the accel cpu
+     * behavior according to the CPU.
+     *
+     * The second is currently only used by TCG, to specialize the
+     * TCGCPUOps depending on the CPU type.
+     */
     cc->accel_cpu = accel_cpu;
     if (accel_cpu->cpu_class_init) {
         accel_cpu->cpu_class_init(cc);
     }
+    if (cc->init_accel_cpu) {
+        cc->init_accel_cpu(accel_cpu, cc);
+    }
 }
 
 /* initialize the arch-specific accel CpuClass interfaces */
diff --git a/target/i386/tcg/tcg-cpu.c b/target/i386/tcg/tcg-cpu.c
index e311f52855..ba39531aa5 100644
--- a/target/i386/tcg/tcg-cpu.c
+++ b/target/i386/tcg/tcg-cpu.c
@@ -69,11 +69,17 @@ static struct TCGCPUOps x86_tcg_ops = {
 #endif /* !CONFIG_USER_ONLY */
 };
 
-static void tcg_cpu_class_init(CPUClass *cc)
+static void tcg_cpu_init_ops(AccelCPUClass *accel_cpu, CPUClass *cc)
 {
+    /* for x86, all cpus use the same set of operations */
     cc->tcg_ops = &x86_tcg_ops;
 }
 
+static void tcg_cpu_class_init(CPUClass *cc)
+{
+    cc->init_accel_cpu = tcg_cpu_init_ops;
+}
+
 /*
  * TCG-specific defaults that override all CPU models when using TCG
  */
-- 
2.26.2



  parent reply	other threads:[~2021-03-16 19:30 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-03-16 18:36 [RFC v8 00/44] arm cleanup experiment for kvm-only build Claudio Fontana
2021-03-16 18:36 ` [RFC v8 01/44] target/arm: move translate modules to tcg/ Claudio Fontana
2021-03-16 18:36 ` [RFC v8 02/44] target/arm: move helpers " Claudio Fontana
2021-03-16 18:36 ` [RFC v8 03/44] arm: tcg: only build under CONFIG_TCG Claudio Fontana
2021-03-16 18:36 ` [RFC v8 04/44] target/arm: tcg: add sysemu and user subsirs Claudio Fontana
2021-03-16 18:36 ` [RFC v8 05/44] target/arm: only build psci for TCG Claudio Fontana
2021-03-16 18:36 ` [RFC v8 06/44] target/arm: split off cpu-sysemu.c Claudio Fontana
2021-03-16 18:36 ` [RFC v8 07/44] target/arm: move physical address translation to cpu-mmu Claudio Fontana
2021-03-16 18:36 ` [RFC v8 08/44] target/arm: cpu-mmu: fix comment style Claudio Fontana
2021-03-16 18:36 ` [RFC v8 09/44] target/arm: split cpregs from tcg/helper.c Claudio Fontana
2021-03-16 18:36 ` [RFC v8 10/44] target/arm: cpregs: fix style (mostly just comments) Claudio Fontana
2021-03-16 18:36 ` [RFC v8 11/44] target/arm: move cpu definitions to common cpu module Claudio Fontana
2021-03-16 18:36 ` [RFC v8 12/44] target/arm: only perform TCG cpu and machine inits if TCG enabled Claudio Fontana
2021-03-16 18:36 ` [RFC v8 13/44] target/arm: kvm: add stubs for some helpers Claudio Fontana
2021-03-16 18:36 ` [RFC v8 14/44] target/arm: move cpsr_read, cpsr_write to cpu_common Claudio Fontana
2021-03-16 18:36 ` [RFC v8 15/44] target/arm: add temporary stub for arm_rebuild_hflags Claudio Fontana
2021-03-16 18:36 ` [RFC v8 16/44] target/arm: split vfp state setting from tcg helpers Claudio Fontana
2021-03-16 18:36 ` [RFC v8 17/44] target/arm: move arm_mmu_idx* to cpu-mmu Claudio Fontana
2021-03-16 18:36 ` [RFC v8 18/44] target/arm: move sve_zcr_len_for_el to common_cpu Claudio Fontana
2021-03-16 18:36 ` [RFC v8 19/44] target/arm: move arm_sctlr away from tcg helpers Claudio Fontana
2021-03-16 18:36 ` [RFC v8 20/44] target/arm: move arm_cpu_list to common_cpu Claudio Fontana
2021-03-16 18:36 ` [RFC v8 21/44] target/arm: move aarch64_sync_32_to_64 (and vv) to cpu code Claudio Fontana
2021-03-16 18:36 ` [RFC v8 22/44] target/arm: split a15 cpu model and 32bit class functions to cpu32.c Claudio Fontana
2021-03-16 18:36 ` [RFC v8 23/44] target/arm: move sve_exception_el out of TCG helpers Claudio Fontana
2021-03-16 18:36 ` [RFC v8 24/44] target/arm: refactor exception and cpu code Claudio Fontana
2021-03-16 18:36 ` [RFC v8 25/44] target/arm: cpu: fix style Claudio Fontana
2021-03-16 18:36 ` [RFC v8 26/44] target/arm: wrap call to aarch64_sve_change_el in tcg_enabled() Claudio Fontana
2021-03-16 18:36 ` [RFC v8 27/44] target/arm: remove kvm include file for PSCI and arm-powerctl Claudio Fontana
2021-03-16 18:36 ` [RFC v8 28/44] target/arm: move kvm-const.h, kvm.c, kvm64.c, kvm_arm.h to kvm/ Claudio Fontana
2021-03-16 18:36 ` [RFC v8 29/44] target/arm: cleanup cpu includes Claudio Fontana
2021-03-16 18:36 ` [RFC v8 30/44] target/arm: remove broad "else" statements when checking accels Claudio Fontana
2021-03-16 18:36 ` [RFC v8 31/44] tests/qtest: skip bios-tables-test test_acpi_oem_fields_virt for KVM Claudio Fontana
2021-03-16 18:36 ` [RFC v8 32/44] tests: restrict TCG-only arm-cpu-features tests to TCG builds Claudio Fontana
2021-03-16 18:36 ` [RFC v8 33/44] tests: do not run test-hmp on all machines for ARM KVM-only Claudio Fontana
2021-03-16 18:36 ` [RFC v8 34/44] tests: device-introspect-test: cope with ARM TCG-only devices Claudio Fontana
2021-03-16 18:36 ` [RFC v8 35/44] tests: do not run qom-test on all machines for ARM KVM-only Claudio Fontana
2021-03-16 18:36 ` [RFC v8 36/44] Revert "target/arm: Restrict v8M IDAU to TCG" Claudio Fontana
2021-03-16 18:36 ` [RFC v8 37/44] target/arm: create kvm cpu accel class Claudio Fontana
2021-03-16 18:36 ` [RFC v8 38/44] target/arm: move kvm cpu properties setting to kvm-cpu Claudio Fontana
2021-03-16 18:36 ` [RFC v8 39/44] accel: move call to accel_init_interfaces Claudio Fontana
2021-03-16 18:36 ` Claudio Fontana [this message]
2021-03-16 18:36 ` [RFC v8 41/44] target/arm: add tcg cpu accel class Claudio Fontana
2021-03-16 18:37 ` [RFC v8 42/44] target/arm: move TCG gt timer creation code in tcg/ Claudio Fontana
2021-03-16 18:37 ` [RFC v8 43/44] target/arm: cpu-sve: new module Claudio Fontana
2021-03-16 18:37 ` [RFC v8 44/44] target/arm: cpu-sve: split TCG and KVM functionality Claudio Fontana

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=20210316183702.10216-41-cfontana@suse.de \
    --to=cfontana@suse.de \
    --cc=alex.bennee@linaro.org \
    --cc=ehabkost@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=philmd@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=r.bolshakov@yadro.com \
    --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).