qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Thomas Huth <thuth@redhat.com>
To: Peter Maydell <peter.maydell@linaro.org>, qemu-devel@nongnu.org
Cc: Ilya Leoshkevich <iii@linux.ibm.com>,
	David Hildenbrand <david@redhat.com>
Subject: [PULL 05/24] target/s390x: Fix R[NOX]SBG with T=1
Date: Mon, 20 Mar 2023 14:03:11 +0100	[thread overview]
Message-ID: <20230320130330.406378-6-thuth@redhat.com> (raw)
In-Reply-To: <20230320130330.406378-1-thuth@redhat.com>

From: Ilya Leoshkevich <iii@linux.ibm.com>

RXSBG usage in the "filetests" test from the wasmtime testsuite makes
tcg_reg_alloc_op() attempt to temp_load() a TEMP_VAL_DEAD temporary,
causing an assertion failure:

    0x01000a70:  ec14 b040 3057  rxsbg    %r1, %r4, 0xb0, 0x40, 0x30

    OP after optimization and liveness analysis:
     ---- 0000000001000a70 0000000000000004 0000000000000006
     rotl_i64 tmp2,r4,$0x30                   dead: 1 2  pref=0xffff
     and_i64 tmp2,tmp2,$0x800000000000ffff    dead: 1  pref=0xffff
    [xor_i64 tmp3,tmp3,tmp2                   dead: 1 2  pref=0xffff]
     and_i64 cc_dst,tmp3,$0x800000000000ffff  sync: 0  dead: 0 1 2  pref=0xffff
     mov_i64 psw_addr,$0x1000a76              sync: 0  dead: 0 1  pref=0xffff
     mov_i32 cc_op,$0x6                       sync: 0  dead: 0 1  pref=0xffff
     call lookup_tb_ptr,$0x6,$1,tmp8,env      dead: 1  pref=none
     goto_ptr tmp8                            dead: 0
     set_label $L0
     exit_tb $0x7fffe809d183

    ../tcg/tcg.c:3865: tcg fatal error

The reason is that tmp3 does not have an initial value, which confuses
the register allocator. This also affects the correctness of the
results.

Fix by assigning R1 to it.

Exposed by commit e2e641fa3d5 ("tcg: Change default temp lifetime to
TEMP_TB").

Fixes: d6c6372e186e ("target-s390: Implement R[NOX]SBG")
Signed-off-by: Ilya Leoshkevich <iii@linux.ibm.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Message-Id: <20230316172205.281369-2-iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
---
 target/s390x/tcg/translate.c | 4 ++++
 1 file changed, 4 insertions(+)

diff --git a/target/s390x/tcg/translate.c b/target/s390x/tcg/translate.c
index 7832cf02a6..6758d9f47a 100644
--- a/target/s390x/tcg/translate.c
+++ b/target/s390x/tcg/translate.c
@@ -3697,11 +3697,15 @@ static DisasJumpType op_rosbg(DisasContext *s, DisasOps *o)
     int i3 = get_field(s, i3);
     int i4 = get_field(s, i4);
     int i5 = get_field(s, i5);
+    TCGv_i64 orig_out;
     uint64_t mask;
 
     /* If this is a test-only form, arrange to discard the result.  */
     if (i3 & 0x80) {
+        tcg_debug_assert(o->out != NULL);
+        orig_out = o->out;
         o->out = tcg_temp_new_i64();
+        tcg_gen_mov_i64(o->out, orig_out);
     }
 
     i3 &= 63;
-- 
2.31.1



  parent reply	other threads:[~2023-03-20 13:06 UTC|newest]

Thread overview: 30+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2023-03-20 13:03 [PULL 00/24] s390x and misc patches for 8.0-rc1 Thomas Huth
2023-03-20 13:03 ` [PULL 01/24] MAINTAINERS: Mark the Nios II CPU as orphan Thomas Huth
2023-03-20 13:03 ` [PULL 02/24] target/s390x: Fix LPSW Thomas Huth
2023-03-20 13:03 ` [PULL 03/24] target/s390x: Implement Early Exception Recognition Thomas Huth
2023-03-20 13:03 ` [PULL 04/24] tests/tcg/s390x: Add PSW modification tests Thomas Huth
2023-03-20 13:03 ` Thomas Huth [this message]
2023-03-20 13:03 ` [PULL 06/24] tests/tcg/s390x: Add rxsbg.c Thomas Huth
2023-03-20 13:03 ` [PULL 07/24] target/s390x: Fix EXECUTE of relative long instructions Thomas Huth
2023-03-20 13:03 ` [PULL 08/24] tests/tcg/s390x: Add ex-relative-long.c Thomas Huth
2023-03-20 13:03 ` [PULL 09/24] target/s390x: Handle branching to odd addresses Thomas Huth
2023-03-20 13:03 ` [PULL 10/24] target/s390x: Handle EXECUTE of " Thomas Huth
2023-03-20 13:03 ` [PULL 11/24] target/s390x: Handle LGRL from non-aligned addresses Thomas Huth
2023-03-20 13:03 ` [PULL 12/24] target/s390x: Handle LRL and LGFRL " Thomas Huth
2023-03-20 13:03 ` [PULL 13/24] target/s390x: Handle LLGFRL " Thomas Huth
2023-03-20 13:03 ` [PULL 14/24] target/s390x: Handle CRL and CGFRL with " Thomas Huth
2023-03-20 13:03 ` [PULL 15/24] target/s390x: Handle CGRL and CLGRL " Thomas Huth
2023-03-20 13:03 ` [PULL 16/24] target/s390x: Handle CLRL and CLGFRL " Thomas Huth
2023-03-20 13:03 ` [PULL 17/24] target/s390x: Handle STRL to " Thomas Huth
2023-03-20 13:03 ` [PULL 18/24] target/s390x: Handle STGRL " Thomas Huth
2023-03-20 13:03 ` [PULL 19/24] target/s390x: Update do_unaligned_access() comment Thomas Huth
2023-03-20 13:03 ` [PULL 20/24] tests/tcg/s390x: Test unaligned accesses Thomas Huth
2023-03-20 13:03 ` [PULL 21/24] target/s390x/tcg/mem_helper: Remove bad assert() statement Thomas Huth
2023-03-20 13:03 ` [PULL 22/24] tests/unit/test-blockjob: Disable complete_in_standby test Thomas Huth
2023-03-20 13:03 ` [PULL 23/24] qemu/osdep: Switch position of "extern" and "G_NORETURN" Thomas Huth
2023-03-20 13:03 ` [PULL 24/24] replace TABs with spaces Thomas Huth
2023-03-20 14:02 ` [PULL 00/24] s390x and misc patches for 8.0-rc1 Thomas Huth
2023-03-20 15:10   ` Peter Maydell
2023-03-20 15:27     ` Philippe Mathieu-Daudé
2023-03-20 15:36       ` Thomas Huth
2023-03-20 17:50 ` 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=20230320130330.406378-6-thuth@redhat.com \
    --to=thuth@redhat.com \
    --cc=david@redhat.com \
    --cc=iii@linux.ibm.com \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.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).