All of lore.kernel.org
 help / color / mirror / Atom feed
From: "Philippe Mathieu-Daudé" <philmd@oss.qualcomm.com>
To: qemu-devel@nongnu.org
Cc: "Mark Cave-Ayland" <mark.cave-ayland@ilande.co.uk>,
	"Peter Maydell" <peter.maydell@linaro.org>,
	"Paolo Bonzini" <pbonzini@redhat.com>,
	qemu-ppc@nongnu.org, qemu-s390x@nongnu.org,
	"Richard Henderson" <richard.henderson@linaro.org>,
	"Pierrick Bouvier" <pierrick.bouvier@oss.qualcomm.com>,
	qemu-riscv@nongnu.org, qemu-arm@nongnu.org,
	"Philippe Mathieu-Daudé" <philmd@oss.qualcomm.com>
Subject: [PATCH v3 07/11] accel/tcg: Add do_interrupt() wrapper for targets to manage BQL
Date: Wed,  2 Sep 2026 17:20:38 +0200	[thread overview]
Message-ID: <20260902152044.31291-8-philmd@oss.qualcomm.com> (raw)
In-Reply-To: <20260902152044.31291-1-philmd@oss.qualcomm.com>

Add TCGCPUOps::do_interrupt() as a wrapper that targets can
implement to control BQL locking independently. The existing
do_interrupt_locked() callback remains for BQL-held contexts.
This allows each target to decide when to acquire the BQL.

Signed-off-by: Philippe Mathieu-Daudé <philmd@oss.qualcomm.com>
---
 include/accel/tcg/cpu-ops.h | 14 ++++++++++++++
 accel/tcg/cpu-exec.c        | 12 ++++++++----
 2 files changed, 22 insertions(+), 4 deletions(-)

diff --git a/include/accel/tcg/cpu-ops.h b/include/accel/tcg/cpu-ops.h
index 2e97f79a373..e2e67f796f6 100644
--- a/include/accel/tcg/cpu-ops.h
+++ b/include/accel/tcg/cpu-ops.h
@@ -169,6 +169,20 @@ struct TCGCPUOps {
      */
     vaddr (*untagged_addr)(CPUState *cs, vaddr addr);
 #else
+    /**
+     * @do_interrupt: Deliver a pending exception/interrupt to the CPU
+     *
+     * Called when cs->exception_index contains the exception code to deliver.
+     * Updates CPU architectural state (usually before executing a guest
+     * exception handler).
+     *
+     * Called from cpu_handle_exception(). Implementations must acquire the
+     * BQL before modifying CPU state and hold it for the duration of the
+     * handler (this ensures safe access to shared CPU and device state during
+     * exception delivery, which may be cross-vCPU).
+     */
+    void (*do_interrupt)(CPUState *cpu);
+
     /**
      * @do_interrupt_locked: Deliver a pending exception/interrupt to the CPU
      * @cpu: cpu context
diff --git a/accel/tcg/cpu-exec.c b/accel/tcg/cpu-exec.c
index 8d9c203f222..4deb7e00571 100644
--- a/accel/tcg/cpu-exec.c
+++ b/accel/tcg/cpu-exec.c
@@ -727,9 +727,13 @@ static inline bool cpu_handle_exception(CPUState *cpu, int *ret)
     if (replay_exception()) {
         const TCGCPUOps *tcg_ops = cpu->cc->tcg_ops;
 
-        bql_lock();
-        tcg_ops->do_interrupt_locked(cpu);
-        bql_unlock();
+        if (tcg_ops->do_interrupt) {
+            tcg_ops->do_interrupt(cpu);
+        } else {
+            bql_lock();
+            tcg_ops->do_interrupt_locked(cpu);
+            bql_unlock();
+        }
         cpu->exception_index = -1;
 
         if (unlikely(cpu_single_stepping(cpu))) {
@@ -1059,7 +1063,7 @@ bool tcg_exec_realizefn(CPUState *cpu, Error **errp)
         assert(tcg_ops->cpu_exec_halt);
         assert(tcg_ops->cpu_exec_interrupt);
         assert(tcg_ops->cpu_exec_reset);
-        assert(tcg_ops->do_interrupt_locked);
+        assert(tcg_ops->do_interrupt || tcg_ops->do_interrupt_locked);
         assert(tcg_ops->pointer_wrap);
 #endif /* !CONFIG_USER_ONLY */
         assert(tcg_ops->translate_code);
-- 
2.53.0



  parent reply	other threads:[~2026-09-02 15:22 UTC|newest]

Thread overview: 28+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2026-09-02 15:20 [PATCH v3 00/11] accel/tcg: Push BQL down into per-target do_interrupt handlers Philippe Mathieu-Daudé
2026-09-02 15:20 ` [PATCH v3 01/11] target/arm: Call arm*_cpu_do_interrupt directly instead of via TCGCPUOps Philippe Mathieu-Daudé
2026-09-02 19:52   ` Pierrick Bouvier
2026-09-08 16:09   ` Peter Maydell
2026-09-13  2:42   ` Richard Henderson
2026-09-02 15:20 ` [PATCH v3 03/11] target/i386: Rename fake_do_interrupt to fake_user_exception Philippe Mathieu-Daudé
2026-09-13  2:44   ` Richard Henderson
2026-09-02 15:20 ` [PATCH v3 04/11] accel/tcg: Document cpu_exec_interrupt() callback contract Philippe Mathieu-Daudé
2026-09-08 16:25   ` Peter Maydell
2026-09-02 15:20 ` [PATCH v3 05/11] accel/tcg: Document do_interrupt() " Philippe Mathieu-Daudé
2026-09-02 19:54   ` Pierrick Bouvier
2026-09-02 20:48     ` Philippe Mathieu-Daudé
2026-09-13  3:30   ` Richard Henderson
2026-09-02 15:20 ` [PATCH v3 06/11] accel/tcg: Rename do_interrupt() to do_interrupt_locked() Philippe Mathieu-Daudé
2026-09-13  3:31   ` Richard Henderson
2026-09-02 15:20 ` Philippe Mathieu-Daudé [this message]
2026-09-13  3:50   ` [PATCH v3 07/11] accel/tcg: Add do_interrupt() wrapper for targets to manage BQL Richard Henderson
2026-09-02 15:20 ` [PATCH v3 08/11] target/hexagon: Move to do_interrupt() (BQL acquired internally) Philippe Mathieu-Daudé
2026-09-03  0:15   ` Brian Cain
2026-09-13  3:55   ` Richard Henderson
2026-09-02 15:20 ` [RFC PATCH v3 09/11] targets: Move interrupt handlers to do_interrupt() Philippe Mathieu-Daudé
2026-09-13  4:00   ` Richard Henderson
2026-09-02 15:20 ` [RFC PATCH v3 10/11] targets: Move BQL locking into do_interrupt() handlers Philippe Mathieu-Daudé
2026-09-02 15:20 ` [PATCH v3 11/11] accel/tcg: Remove do_interrupt_locked() callback Philippe Mathieu-Daudé
2026-09-13  4:00   ` Richard Henderson
2026-09-02 15:33 ` [PATCH v3 00/11] accel/tcg: Push BQL down into per-target do_interrupt handlers Philippe Mathieu-Daudé
     [not found] ` <20260902152044.31291-3-philmd@oss.qualcomm.com>
2026-09-02 19:52   ` [PATCH v3 02/11] target/s390x: Restrict interrupt handlers to system mode Pierrick Bouvier
2026-09-13  2:43   ` Richard Henderson

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=20260902152044.31291-8-philmd@oss.qualcomm.com \
    --to=philmd@oss.qualcomm.com \
    --cc=mark.cave-ayland@ilande.co.uk \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=pierrick.bouvier@oss.qualcomm.com \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-ppc@nongnu.org \
    --cc=qemu-riscv@nongnu.org \
    --cc=qemu-s390x@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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.