qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: Peter Maydell <peter.maydell@linaro.org>, qemu-devel@nongnu.org
Cc: "Thomas Huth" <thuth@redhat.com>,
	"David Hildenbrand" <david@redhat.com>,
	"Cornelia Huck" <cohuck@redhat.com>,
	"Richard Henderson" <richard.henderson@linaro.org>,
	qemu-s390x@nongnu.org, "Paolo Bonzini" <pbonzini@redhat.com>,
	"Alex Bennée" <alex.bennee@linaro.org>
Subject: [PULL 31/31] s390x/tcg: MVCL: Exit to main loop if requested
Date: Thu, 10 Oct 2019 13:33:56 +0200	[thread overview]
Message-ID: <20191010113356.5017-32-david@redhat.com> (raw)
In-Reply-To: <20191010113356.5017-1-david@redhat.com>

MVCL is interruptible and we should check for interrupts and process
them after writing back the variables to the registers. Let's check
for any exit requests and exit to the main loop. Introduce a new helper
function for that: cpu_loop_exit_requested().

When booting Fedora 30, I can see a handful of these exits and it seems
to work reliable. Also, Richard explained why this works correctly even
when MVCL is called via EXECUTE:

    (1) TB with EXECUTE runs, at address Ae
        - env->psw_addr stored with Ae.
        - helper_ex() runs, memory address Am computed
          from D2a(X2a,B2a) or from psw.addr+RI2.
        - env->ex_value stored with memory value modified by R1a

    (2) TB of executee runs,
        - env->ex_value stored with 0.
        - helper_mvcl() runs, using and updating R1b, R1b+1, R2b, R2b+1.

    (3a) helper_mvcl() completes,
         - TB of executee continues, psw.addr += ilen.
         - Next instruction is the one following EXECUTE.

    (3b) helper_mvcl() exits to main loop,
         - cpu_loop_exit_restore() unwinds psw.addr = Ae.
         - Next instruction is the EXECUTE itself...
         - goto 1.

As the PoP mentiones that an interruptible instruction called via EXECUTE
should avoid modifying storage/registers that are used by EXECUTE itself,
it is fine to retrigger EXECUTE.

Cc: Alex Bennée <alex.bennee@linaro.org>
Cc: Peter Maydell <peter.maydell@linaro.org>
Cc: Paolo Bonzini <pbonzini@redhat.com>
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
Signed-off-by: David Hildenbrand <david@redhat.com>
---
 include/exec/exec-all.h   | 17 +++++++++++++++++
 target/s390x/mem_helper.c | 11 ++++++++++-
 2 files changed, 27 insertions(+), 1 deletion(-)

diff --git a/include/exec/exec-all.h b/include/exec/exec-all.h
index 49db07ba0b..04795c49bf 100644
--- a/include/exec/exec-all.h
+++ b/include/exec/exec-all.h
@@ -72,6 +72,23 @@ void QEMU_NORETURN cpu_loop_exit(CPUState *cpu);
 void QEMU_NORETURN cpu_loop_exit_restore(CPUState *cpu, uintptr_t pc);
 void QEMU_NORETURN cpu_loop_exit_atomic(CPUState *cpu, uintptr_t pc);
 
+/**
+ * cpu_loop_exit_requested:
+ * @cpu: The CPU state to be tested
+ *
+ * Indicate if somebody asked for a return of the CPU to the main loop
+ * (e.g., via cpu_exit() or cpu_interrupt()).
+ *
+ * This is helpful for architectures that support interruptible
+ * instructions. After writing back all state to registers/memory, this
+ * call can be used to check if it makes sense to return to the main loop
+ * or to continue executing the interruptible instruction.
+ */
+static inline bool cpu_loop_exit_requested(CPUState *cpu)
+{
+    return (int32_t)atomic_read(&cpu_neg(cpu)->icount_decr.u32) < 0;
+}
+
 #if !defined(CONFIG_USER_ONLY)
 void cpu_reloading_memory_map(void);
 /**
diff --git a/target/s390x/mem_helper.c b/target/s390x/mem_helper.c
index 4254548935..2325767f17 100644
--- a/target/s390x/mem_helper.c
+++ b/target/s390x/mem_helper.c
@@ -1015,6 +1015,7 @@ uint32_t HELPER(mvcl)(CPUS390XState *env, uint32_t r1, uint32_t r2)
     uint64_t srclen = env->regs[r2 + 1] & 0xffffff;
     uint64_t src = get_address(env, r2);
     uint8_t pad = env->regs[r2 + 1] >> 24;
+    CPUState *cs = env_cpu(env);
     S390Access srca, desta;
     uint32_t cc, cur_len;
 
@@ -1065,7 +1066,15 @@ uint32_t HELPER(mvcl)(CPUS390XState *env, uint32_t r1, uint32_t r2)
         env->regs[r1 + 1] = deposit64(env->regs[r1 + 1], 0, 24, destlen);
         set_address_zero(env, r1, dest);
 
-        /* TODO: Deliver interrupts. */
+        /*
+         * MVCL is interruptible. Return to the main loop if requested after
+         * writing back all state to registers. If no interrupt will get
+         * injected, we'll end up back in this handler and continue processing
+         * the remaining parts.
+         */
+        if (destlen && unlikely(cpu_loop_exit_requested(cs))) {
+            cpu_loop_exit_restore(cs, ra);
+        }
     }
     return cc;
 }
-- 
2.21.0



  parent reply	other threads:[~2019-10-10 11:59 UTC|newest]

Thread overview: 38+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-10 11:33 [PULL 00/31] s390x/tcg update David Hildenbrand
2019-10-10 11:33 ` [PULL 01/31] s390x/mmu: Drop debug logging from MMU code David Hildenbrand
2019-10-10 11:33 ` [PULL 02/31] s390x/mmu: Move DAT protection handling out of mmu_translate_asce() David Hildenbrand
2019-10-10 11:33 ` [PULL 03/31] s390x/mmu: Inject DAT exceptions from a single place David Hildenbrand
2019-10-10 11:33 ` [PULL 04/31] s390x/mmu: Inject PGM_ADDRESSING on bogus table addresses David Hildenbrand
2019-10-10 11:33 ` [PULL 05/31] s390x/mmu: Use TARGET_PAGE_MASK in mmu_translate_pte() David Hildenbrand
2019-10-10 11:33 ` [PULL 06/31] s390x/mmu: DAT table definition overhaul David Hildenbrand
2019-10-10 11:33 ` [PULL 07/31] s390x/mmu: Convert to non-recursive page table walk David Hildenbrand
2019-10-10 11:33 ` [PULL 08/31] s390x/mmu: Add EDAT2 translation support David Hildenbrand
2019-10-10 11:33 ` [PULL 09/31] s390x/mmu: Implement ESOP-2 and access-exception-fetch/store-indication facility David Hildenbrand
2019-10-10 11:33 ` [PULL 10/31] s390x/mmu: Implement Instruction-Execution-Protection Facility David Hildenbrand
2019-10-10 11:33 ` [PULL 11/31] s390x/cpumodel: Prepare for changes of QEMU model David Hildenbrand
2019-10-10 11:33 ` [PULL 12/31] s390x/cpumodel: Add new TCG features to QEMU cpu model David Hildenbrand
2019-10-10 11:33 ` [PULL 13/31] target/s390x: Add ilen to unwind data David Hildenbrand
2019-10-10 11:33 ` [PULL 14/31] target/s390x: Remove ilen parameter from tcg_s390_program_interrupt David Hildenbrand
2019-10-10 11:33 ` [PULL 15/31] target/s390x: Remove ilen parameter from s390_program_interrupt David Hildenbrand
2019-10-10 11:33 ` [PULL 16/31] target/s390x: Use tcg_s390_program_interrupt in TCG helpers David Hildenbrand
2019-10-10 11:33 ` [PULL 17/31] target/s390x: Push trigger_pgm_exception lower in s390_cpu_tlb_fill David Hildenbrand
2019-10-10 11:33 ` [PULL 18/31] target/s390x: Handle tec " David Hildenbrand
2019-10-10 11:33 ` [PULL 19/31] target/s390x: Return exception from mmu_translate_real David Hildenbrand
2019-10-17 11:57   ` Peter Maydell
2019-10-17 12:05     ` Peter Maydell
2019-10-17 12:13       ` David Hildenbrand
2019-10-10 11:33 ` [PULL 20/31] target/s390x: Remove exc argument to mmu_translate_asce David Hildenbrand
2019-10-10 11:33 ` [PULL 21/31] target/s390x: Return exception from mmu_translate David Hildenbrand
2019-10-10 11:33 ` [PULL 22/31] target/s390x: Return exception from translate_pages David Hildenbrand
2019-10-10 11:33 ` [PULL 23/31] target/s390x: Remove fail variable from s390_cpu_tlb_fill David Hildenbrand
2019-10-10 11:33 ` [PULL 24/31] target/s390x: Simplify helper_lra David Hildenbrand
2019-10-10 11:33 ` [PULL 25/31] target/s390x: Rely on unwinding in s390_cpu_tlb_fill David Hildenbrand
2019-10-10 11:33 ` [PULL 26/31] target/s390x: Rely on unwinding in s390_cpu_virt_mem_rw David Hildenbrand
2019-10-10 11:33 ` [PULL 27/31] target/s390x: Remove ILEN_AUTO David Hildenbrand
2019-10-10 11:33 ` [PULL 28/31] target/s390x: Remove ilen argument from trigger_access_exception David Hildenbrand
2019-10-10 11:33 ` [PULL 29/31] target/s390x: Remove ilen argument from trigger_pgm_exception David Hildenbrand
2019-10-10 11:33 ` [PULL 30/31] target/s390x: Remove ILEN_UNWIND David Hildenbrand
2019-10-10 11:33 ` David Hildenbrand [this message]
2019-10-10 12:33 ` [PULL 00/31] s390x/tcg update no-reply
2019-10-10 13:11   ` David Hildenbrand
2019-10-14 11:26 ` Peter Maydell

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=20191010113356.5017-32-david@redhat.com \
    --to=david@redhat.com \
    --cc=alex.bennee@linaro.org \
    --cc=cohuck@redhat.com \
    --cc=pbonzini@redhat.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-s390x@nongnu.org \
    --cc=richard.henderson@linaro.org \
    --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 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).