All of lore.kernel.org
 help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: qemu-s390x@nongnu.org, qemu-devel@nongnu.org
Cc: Christian Borntraeger <borntraeger@de.ibm.com>,
	Cornelia Huck <cohuck@redhat.com>,
	Richard Henderson <rth@twiddle.net>,
	Alexander Graf <agraf@suse.de>, Thomas Huth <thuth@redhat.com>,
	David Hildenbrand <david@redhat.com>
Subject: [Qemu-devel] [PATCH v2 07/18] s390x/tcg: tolerate wrong wakeups due to floating interrupts
Date: Wed, 17 Jan 2018 18:40:36 +0100	[thread overview]
Message-ID: <20180117174047.6382-8-david@redhat.com> (raw)
In-Reply-To: <20180117174047.6382-1-david@redhat.com>

This is a prparation for floating interrupt support and only applies to
MTTCG, single threaded TCG works just fine. If a floating interrupt wakes
up a VCPU and the CPU thinks it can run (clearing cs->halted), at
the point where the interrupt would be delivered, already another VCPU
might have picked up the interrupt, resulting in a wakeup without an
interrupt (executing wrong code).

It is wrong to let the VCPU continue to execute (the WAIT PSW). Instead,
we have to put the VCPU back to sleep.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 target/s390x/excp_helper.c | 5 +++++
 1 file changed, 5 insertions(+)

diff --git a/target/s390x/excp_helper.c b/target/s390x/excp_helper.c
index a18842ccbd..eeffb49f63 100644
--- a/target/s390x/excp_helper.c
+++ b/target/s390x/excp_helper.c
@@ -503,6 +503,11 @@ bool s390_cpu_exec_interrupt(CPUState *cs, int interrupt_request)
             s390_cpu_do_interrupt(cs);
             return true;
         }
+        if (env->psw.mask & PSW_MASK_WAIT) {
+            /* Woken up because of a floating interrupt but it has already
+             * been delivered. Go back to sleep. */
+            cpu_interrupt(CPU(cpu), CPU_INTERRUPT_HALT);
+        }
     }
     return false;
 }
-- 
2.14.3

  parent reply	other threads:[~2018-01-17 17:41 UTC|newest]

Thread overview: 31+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-01-17 17:40 [Qemu-devel] [PATCH v2 00/18] s390x: flic rework, tcg flic support and STSI David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 01/18] s390x/tcg: deliver multiple interrupts in a row David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 02/18] s390x/flic: simplify flic initialization David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 03/18] s390x/tcg: simplify lookup of flic David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 04/18] s390x/tcg: simplify machine check handling David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 05/18] s390x/flic: factor out injection of floating interrupts David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 06/18] s390x/flic: no need to call s390_io_interrupt() from flic David Hildenbrand
2018-01-17 17:40 ` David Hildenbrand [this message]
2018-01-25 10:22   ` [Qemu-devel] [PATCH v2 07/18] s390x/tcg: tolerate wrong wakeups due to floating interrupts Cornelia Huck
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 08/18] s390x/flic: make floating interrupts on TCG actually floating David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 09/18] s390x/tcg: implement TEST PENDING INTERRUPTION David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 10/18] s390x/flic: implement qemu_s390_clear_io_flic() David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 11/18] s390x/flic: optimize CPU wakeup for TCG David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 12/18] s390x: fix size + content of STSI blocks David Hildenbrand
2018-01-25 10:26   ` Cornelia Huck
2018-01-25 13:15     ` David Hildenbrand
2018-01-25 13:51       ` Cornelia Huck
2018-01-29  9:42         ` [Qemu-devel] [qemu-s390x] " David Hildenbrand
2018-01-29  9:48           ` Cornelia Huck
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 13/18] s390x/tcg: STSI overhaul David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 14/18] s390x/tcg: remove SMP warning David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 15/18] configure: s390x supports mttcg now David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 16/18] s390x/tcg: cache the qemu flic in a central function David Hildenbrand
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 17/18] s390x/kvm: cache the kvm " David Hildenbrand
2018-01-25 11:03   ` Christian Borntraeger
2018-01-17 17:40 ` [Qemu-devel] [PATCH v2 18/18] s390x/flic: cache the common flic class " David Hildenbrand
2018-01-25 10:29 ` [Qemu-devel] [PATCH v2 00/18] s390x: flic rework, tcg flic support and STSI Cornelia Huck
2018-01-25 12:42 ` Cornelia Huck
2018-01-25 12:47   ` Cornelia Huck
2018-01-25 13:11     ` David Hildenbrand
2018-01-25 13:13       ` Thomas Huth

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=20180117174047.6382-8-david@redhat.com \
    --to=david@redhat.com \
    --cc=agraf@suse.de \
    --cc=borntraeger@de.ibm.com \
    --cc=cohuck@redhat.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-s390x@nongnu.org \
    --cc=rth@twiddle.net \
    --cc=thuth@redhat.com \
    /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.