qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
* [risu PATCH v2 0/4] Add support for s390x to RISU
@ 2023-09-05 11:49 Thomas Huth
  2023-09-05 11:49 ` [risu PATCH v2 1/4] s390x: Add basic s390x support to the C code Thomas Huth
                   ` (3 more replies)
  0 siblings, 4 replies; 8+ messages in thread
From: Thomas Huth @ 2023-09-05 11:49 UTC (permalink / raw)
  To: Peter Maydell
  Cc: qemu-s390x, qemu-devel, Richard Henderson, Ilya Leoshkevich,
	David Hildenbrand

 Hi Peter!

Here are some patches that add basic support for s390x to RISU.
It's still quite limited, e.g. no support for load/store memory
operations yet, but the basics with simple 16-bit or 32-bit
instructions work already fine.

v2:
- Removed the code to avoid r14 (return address) and r15 (stack pointer)
  since it is not necessary anymore since commit ad82a069e8d6a
- Initialize the floating point registers in test_s390x.S, too
- Added Acked-bys and Reviewed-bys from v1

Thomas Huth (4):
  s390x: Add basic s390x support to the C code
  s390x: Add simple s390x.risu file
  s390x: Add basic risugen perl module for s390x
  s390x: Update the configure script for s390x support

 configure            |   4 +-
 risu_reginfo_s390x.c | 140 ++++++++++++++++++++++++++++++++
 risu_reginfo_s390x.h |  23 ++++++
 risu_s390x.c         |  48 +++++++++++
 risugen_s390x.pm     | 186 +++++++++++++++++++++++++++++++++++++++++++
 s390x.risu           |  48 +++++++++++
 test_s390x.S         |  51 ++++++++++++
 7 files changed, 499 insertions(+), 1 deletion(-)
 create mode 100644 risu_reginfo_s390x.c
 create mode 100644 risu_reginfo_s390x.h
 create mode 100644 risu_s390x.c
 create mode 100644 risugen_s390x.pm
 create mode 100644 s390x.risu
 create mode 100644 test_s390x.S

-- 
2.39.3



^ permalink raw reply	[flat|nested] 8+ messages in thread

* [risu PATCH v2 1/4] s390x: Add basic s390x support to the C code
  2023-09-05 11:49 [risu PATCH v2 0/4] Add support for s390x to RISU Thomas Huth
@ 2023-09-05 11:49 ` Thomas Huth
  2023-09-12 16:50   ` Peter Maydell
  2023-09-05 11:49 ` [risu PATCH v2 2/4] s390x: Add simple s390x.risu file Thomas Huth
                   ` (2 subsequent siblings)
  3 siblings, 1 reply; 8+ messages in thread
From: Thomas Huth @ 2023-09-05 11:49 UTC (permalink / raw)
  To: Peter Maydell
  Cc: qemu-s390x, qemu-devel, Richard Henderson, Ilya Leoshkevich,
	David Hildenbrand

With these changes, it is now possible to compile the "risu" binary
for s390x hosts.

Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
---
 risu_reginfo_s390x.c | 140 +++++++++++++++++++++++++++++++++++++++++++
 risu_reginfo_s390x.h |  23 +++++++
 risu_s390x.c         |  48 +++++++++++++++
 test_s390x.S         |  51 ++++++++++++++++
 4 files changed, 262 insertions(+)
 create mode 100644 risu_reginfo_s390x.c
 create mode 100644 risu_reginfo_s390x.h
 create mode 100644 risu_s390x.c
 create mode 100644 test_s390x.S

diff --git a/risu_reginfo_s390x.c b/risu_reginfo_s390x.c
new file mode 100644
index 0000000..1c6aa0c
--- /dev/null
+++ b/risu_reginfo_s390x.c
@@ -0,0 +1,140 @@
+/******************************************************************************
+ * Copyright 2023 Red Hat Inc.
+ * All rights reserved. This program and the accompanying materials
+ * are made available under the terms of the Eclipse Public License v1.0
+ * which accompanies this distribution, and is available at
+ * http://www.eclipse.org/legal/epl-v10.html
+ *
+ * Contributors:
+ *     Thomas Huth - initial implementation
+ *****************************************************************************/
+
+#include <stdio.h>
+#include <ucontext.h>
+#include <string.h>
+#include <math.h>
+#include <stdlib.h>
+#include <sys/user.h>
+
+#include "risu.h"
+#include "risu_reginfo_s390x.h"
+
+
+const struct option * const arch_long_opts;
+const char * const arch_extra_help;
+
+void process_arch_opt(int opt, const char *arg)
+{
+    abort();
+}
+
+void arch_init(void)
+{
+}
+
+int reginfo_size(struct reginfo *ri)
+{
+    return sizeof(*ri);
+}
+
+/* reginfo_init: initialize with a ucontext */
+void reginfo_init(struct reginfo *ri, ucontext_t *uc)
+{
+    int i;
+
+    memset(ri, 0, sizeof(*ri));
+
+    ri->faulting_insn = *((uint32_t *) uc->uc_mcontext.psw.addr);
+    ri->psw_mask = uc->uc_mcontext.psw.mask;
+    ri->psw_addr = uc->uc_mcontext.psw.addr - image_start_address;
+
+    for (i = 0; i < 16; i++) {
+        ri->gregs[i] = uc->uc_mcontext.gregs[i];
+    }
+
+    memcpy(&ri->fpregs, &uc->uc_mcontext.fpregs, sizeof(fpregset_t));
+}
+
+/* reginfo_is_eq: compare the reginfo structs, returns nonzero if equal */
+int reginfo_is_eq(struct reginfo *m, struct reginfo *a)
+{
+    int i;
+
+    if (m->psw_mask != a->psw_mask || m->psw_addr != a->psw_addr) {
+        return 0;
+    }
+
+    for (i = 0; i < 16; i++) {
+        if (m->gregs[i] != a->gregs[i]) {
+            return 0;
+        }
+    }
+
+    if (memcmp(&m->fpregs, &a->fpregs, sizeof(fpregset_t))) {
+        return 0;
+    }
+
+    return 1;
+}
+
+/* reginfo_dump: print state to a stream, returns nonzero on success */
+int reginfo_dump(struct reginfo *ri, FILE * f)
+{
+    int i;
+
+    fprintf(f, "  faulting insn 0x%x\n", ri->faulting_insn);
+    fprintf(f, "  PSW mask      0x%" PRIx64 "\n", ri->psw_mask);
+    fprintf(f, "  PSW addr offs 0x%" PRIx64 "\n\n", ri->psw_addr);
+
+    for (i = 0; i < 16/2; i++) {
+        fprintf(f, "\tr%d: %16lx\tr%02d: %16lx\n", i, ri->gregs[i],
+                i + 8, ri->gregs[i + 8]);
+    }
+    fprintf(f, "\n");
+
+    for (i = 0; i < 16/2; i++) {
+        fprintf(f, "\tf%d: %16lx\tf%02d: %16lx\n",
+                i, *(uint64_t *)&ri->fpregs.fprs[i],
+                i + 8, *(uint64_t *)&ri->fpregs.fprs[i + 8]);
+    }
+    fprintf(f, "\tFPC: %8x\n\n", ri->fpregs.fpc);
+
+    return !ferror(f);
+}
+
+int reginfo_dump_mismatch(struct reginfo *m, struct reginfo *a, FILE *f)
+{
+    int i;
+
+    if (m->psw_mask != a->psw_mask) {
+        fprintf(f, "Mismatch: PSW mask master: [%016lx] - PSW mask apprentice: [%016lx]\n",
+                m->psw_mask, a->psw_mask);
+    }
+
+    if (m->psw_addr != a->psw_addr) {
+        fprintf(f, "Mismatch: PSW addr offset master: [%016lx] - PSW addr offset apprentice: [%016lx]\n",
+                m->psw_addr, a->psw_addr);
+    }
+
+    for (i = 0; i < 16; i++) {
+        if (m->gregs[i] != a->gregs[i]) {
+            fprintf(f, "Mismatch: r%d master: [%016lx] - r%d apprentice: [%016lx]\n",
+                    i, m->gregs[i], i, a->gregs[i]);
+        }
+    }
+
+    for (i = 0; i < 16; i++) {
+        if (*(uint64_t *)&m->fpregs.fprs[i] != *(uint64_t *)&a->fpregs.fprs[i]) {
+            fprintf(f, "Mismatch: f%d master: [%016lx] - f%d apprentice: [%016lx]\n",
+                    i, *(uint64_t *)&m->fpregs.fprs[i],
+                    i, *(uint64_t *)&a->fpregs.fprs[i]);
+        }
+    }
+
+    if (m->fpregs.fpc != a->fpregs.fpc) {
+        fprintf(f, "Mismatch: FPC master: [%08x] - FPC apprentice: [%08x]\n",
+                m->fpregs.fpc, a->fpregs.fpc);
+    }
+
+    return !ferror(f);
+}
diff --git a/risu_reginfo_s390x.h b/risu_reginfo_s390x.h
new file mode 100644
index 0000000..b55a11d
--- /dev/null
+++ b/risu_reginfo_s390x.h
@@ -0,0 +1,23 @@
+/******************************************************************************
+ * Copyright 2023 Red Hat Inc.
+ * All rights reserved. This program and the accompanying materials
+ * are made available under the terms of the Eclipse Public License v1.0
+ * which accompanies this distribution, and is available at
+ * http://www.eclipse.org/legal/epl-v10.html
+ *
+ * Contributors:
+ *     Thomas Huth - initial implementation
+ *****************************************************************************/
+
+#ifndef RISU_REGINFO_S390X_H
+#define RISU_REGINFO_S390X_H
+
+struct reginfo {
+    uint32_t faulting_insn;
+    uint64_t psw_mask;
+    uint64_t psw_addr;
+    gregset_t gregs;
+    fpregset_t fpregs;
+};
+
+#endif /* RISU_REGINFO_S390X_H */
diff --git a/risu_s390x.c b/risu_s390x.c
new file mode 100644
index 0000000..4a83869
--- /dev/null
+++ b/risu_s390x.c
@@ -0,0 +1,48 @@
+/******************************************************************************
+ * Copyright 2023 Red Hat Inc.
+ * All rights reserved. This program and the accompanying materials
+ * are made available under the terms of the Eclipse Public License v1.0
+ * which accompanies this distribution, and is available at
+ * http://www.eclipse.org/legal/epl-v10.html
+ *
+ * Contributors:
+ *     Thomas Huth - initial implementation
+ *****************************************************************************/
+
+#include <sys/user.h>
+
+#include "risu.h"
+
+void advance_pc(void *vuc)
+{
+    /*
+     * Note: The PSW address already points to the next instruction
+     * after we get a SIGILL, so we must not advance it here!
+     */
+    // ucontext_t *uc = (ucontext_t *) vuc;
+    // uc->uc_mcontext.psw.addr += 4;
+}
+
+void set_ucontext_paramreg(void *vuc, uint64_t value)
+{
+    ucontext_t *uc = vuc;
+    uc->uc_mcontext.gregs[0] = value;
+}
+
+uint64_t get_reginfo_paramreg(struct reginfo *ri)
+{
+    return ri->gregs[0];
+}
+
+RisuOp get_risuop(struct reginfo *ri)
+{
+    uint32_t insn = ri->faulting_insn;
+    uint32_t op = insn & 0xff;
+    uint32_t key = insn & ~0xff;
+    return (key != 0x835a0f00) ? OP_SIGILL : op;
+}
+
+uintptr_t get_pc(struct reginfo *ri)
+{
+   return ri->psw_addr;
+}
diff --git a/test_s390x.S b/test_s390x.S
new file mode 100644
index 0000000..59f365f
--- /dev/null
+++ b/test_s390x.S
@@ -0,0 +1,51 @@
+/*****************************************************************************
+ * Copyright 2023 Red Hat Inc.
+ * All rights reserved. This program and the accompanying materials
+ * are made available under the terms of the Eclipse Public License v1.0
+ * which accompanies this distribution, and is available at
+ * http://www.eclipse.org/legal/epl-v10.html
+ *
+ * Contributors:
+ *     Thomas Huth - initial implementation
+ *****************************************************************************/
+
+    /* Initialise the general purpose registers */
+    lgfi %r0, 0
+    lgfi %r1, 0x1111111
+    lgfi %r2, 0x2222222
+    lgfi %r3, 0x3333333
+    lgfi %r4, 0x4444444
+    lgfi %r5, 0x5555555
+    lgfi %r6, 0x6666666
+    lgfi %r7, 0x7777777
+    lgfi %r8, 0x8888888
+    lgfi %r9, 0x9999999
+    lgfi %r10, 0xaaaaaaa
+    lgfi %r11, 0xbbbbbbb
+    lgfi %r12, 0xccccccc
+    lgfi %r13, 0xddddddd
+    lgfi %r14, 0xeeeeeee
+    lgfi %r15, 0xfffffff
+
+    /* Initialize floating point registers */
+    ldgr %f0,%r0
+    ldgr %f1,%r1
+    ldgr %f2,%r2
+    ldgr %f3,%r3
+    ldgr %f4,%r4
+    ldgr %f5,%r5
+    ldgr %f6,%r6
+    ldgr %f7,%r7
+    ldgr %f8,%r8
+    ldgr %f9,%r9
+    ldgr %f10,%r10
+    ldgr %f11,%r11
+    ldgr %f12,%r12
+    ldgr %f13,%r13
+    ldgr %f14,%r14
+    ldgr %f15,%r15
+
+    /* do compare */
+    .int 0x835a0f00
+    /* exit test */
+    .int 0x835a0f01
-- 
2.39.3



^ permalink raw reply related	[flat|nested] 8+ messages in thread

* [risu PATCH v2 2/4] s390x: Add simple s390x.risu file
  2023-09-05 11:49 [risu PATCH v2 0/4] Add support for s390x to RISU Thomas Huth
  2023-09-05 11:49 ` [risu PATCH v2 1/4] s390x: Add basic s390x support to the C code Thomas Huth
@ 2023-09-05 11:49 ` Thomas Huth
  2023-09-05 11:49 ` [risu PATCH v2 3/4] s390x: Add basic risugen perl module for s390x Thomas Huth
  2023-09-05 11:50 ` [risu PATCH v2 4/4] s390x: Update the configure script for s390x support Thomas Huth
  3 siblings, 0 replies; 8+ messages in thread
From: Thomas Huth @ 2023-09-05 11:49 UTC (permalink / raw)
  To: Peter Maydell
  Cc: qemu-s390x, qemu-devel, Richard Henderson, Ilya Leoshkevich,
	David Hildenbrand

This only adds a limited set of s390x instructions for initial testing.
More instructions will be added later.

Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
Signed-off-by: Thomas Huth <thuth@redhat.com>
---
 s390x.risu | 48 ++++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 48 insertions(+)
 create mode 100644 s390x.risu

diff --git a/s390x.risu b/s390x.risu
new file mode 100644
index 0000000..3ad7015
--- /dev/null
+++ b/s390x.risu
@@ -0,0 +1,48 @@
+###############################################################################
+# Copyright 2023 Red Hat Inc.
+# All rights reserved. This program and the accompanying materials
+# are made available under the terms of the Eclipse Public License v1.0
+# which accompanies this distribution, and is available at
+# http://www.eclipse.org/legal/epl-v10.html
+#
+# Contributors:
+#     Thomas Huth - initial implementation
+###############################################################################
+
+.mode s390x
+
+# format:RR Add (register + register, 32 bit)
+AR Z 00011010 r1:4 r2:4
+
+# format:RRE Add (register + register, 64 bit)
+AGR Z 10111001 00001000 00000000 r1:4 r2:4
+
+# format:RRE Add (register + register, 32 bit to 64 bit)
+AGFR Z 10111001 00011000 00000000 r1:4 r2:4
+
+# format:RRF-a Add (three registers, 32 bit)
+ARK STFLE45 10111001 11111000 r3:4 0000 r1:4 r2:4
+
+# format:RRF-a Add (three registers, 64 bit)
+AGRK STFLE45 10111001 11101000 r3:4 0000 r1:4 r2:4
+
+
+# format:RRE Add Halfword Immediate (32 bit)
+AHI Z 10100111 r1:4 1010 i2:16
+
+# format:RI Add Halfword Immediate (64 bit)
+AGHI Z 10100111 r1:4 1011 i2:16
+
+
+# format:RR Add Logical (32 bit)
+ALR Z 00011110 r1:4 r2:4
+
+# format:RRE Add Logical (64 bit)
+ALGR Z 10111001 00001010 00000000 r1:4 r2:4
+
+# format:RRE Add Logical (32 bit to 64 bit)
+ALGFR Z 10111001 00011010 00000000 r1:4 r2:4
+
+
+# format:RRF-c Population Count
+POPCNT STFLE45 10111001 11100001 m3:4 0000 r1:4 r2:4
-- 
2.39.3



^ permalink raw reply related	[flat|nested] 8+ messages in thread

* [risu PATCH v2 3/4] s390x: Add basic risugen perl module for s390x
  2023-09-05 11:49 [risu PATCH v2 0/4] Add support for s390x to RISU Thomas Huth
  2023-09-05 11:49 ` [risu PATCH v2 1/4] s390x: Add basic s390x support to the C code Thomas Huth
  2023-09-05 11:49 ` [risu PATCH v2 2/4] s390x: Add simple s390x.risu file Thomas Huth
@ 2023-09-05 11:49 ` Thomas Huth
  2023-09-05 11:50 ` [risu PATCH v2 4/4] s390x: Update the configure script for s390x support Thomas Huth
  3 siblings, 0 replies; 8+ messages in thread
From: Thomas Huth @ 2023-09-05 11:49 UTC (permalink / raw)
  To: Peter Maydell
  Cc: qemu-s390x, qemu-devel, Richard Henderson, Ilya Leoshkevich,
	David Hildenbrand

This implements support for simple 16-bit and 32-bit instructions.
Support for 48-bit instructions and support for load/store memory
instructions is not implemented yet.

Signed-off-by: Thomas Huth <thuth@redhat.com>
---
 risugen_s390x.pm | 186 +++++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 186 insertions(+)
 create mode 100644 risugen_s390x.pm

diff --git a/risugen_s390x.pm b/risugen_s390x.pm
new file mode 100644
index 0000000..f0d03c4
--- /dev/null
+++ b/risugen_s390x.pm
@@ -0,0 +1,186 @@
+#!/usr/bin/perl -w
+###############################################################################
+# Copyright 2023 Red Hat Inc.
+# All rights reserved. This program and the accompanying materials
+# are made available under the terms of the Eclipse Public License v1.0
+# which accompanies this distribution, and is available at
+# http://www.eclipse.org/legal/epl-v10.html
+#
+# Contributors:
+#     Thomas Huth - initial implementation (based on risugen_ppc64.pm etc.)
+###############################################################################
+
+# risugen -- generate a test binary file for use with risu
+# See 'risugen --help' for usage information.
+package risugen_s390x;
+
+use strict;
+use warnings;
+
+use risugen_common;
+
+require Exporter;
+
+our @ISA    = qw(Exporter);
+our @EXPORT = qw(write_test_code);
+
+my $periodic_reg_random = 1;
+
+# Maximum alignment restriction permitted for a memory op.
+my $MAXALIGN = 64;
+
+sub write_mov_ri($$$)
+{
+    my ($r, $imm_h, $imm_l) = @_;
+
+    # LGFI
+    insn16(0xc0 << 8 | $r << 4 | 0x1);
+    insn32($imm_l);
+    # IIHF r,imm_high
+    insn16(0xc0 << 8 | $r << 4 | 0x8);
+    insn32($imm_h);
+}
+
+sub write_mov_fp($$)
+{
+    my ($r, $imm) = @_;
+
+    write_mov_ri(0, ~$imm, $imm);
+    # LDGR
+    insn32(0xb3c1 << 16 | $r << 4);
+}
+
+sub write_random_regdata()
+{
+    # Floating point registers
+    for (my $i = 0; $i < 16; $i++) {
+        write_mov_fp($i, rand(0xffffffff));
+    }
+
+    # Load FPC (via r0)
+    write_mov_ri(0, 0, (rand(0xffffffff) & 0xfcfcff77));
+    insn32(0xb3840000);
+
+    # general purpose registers
+    for (my $i = 0; $i < 16; $i++) {
+        write_mov_ri($i, rand(0xffffffff), rand(0xffffffff));
+    }
+}
+
+my $OP_COMPARE = 0;        # compare registers
+my $OP_TESTEND = 1;        # end of test, stop
+
+sub write_random_register_data()
+{
+    write_random_regdata();
+    write_risuop($OP_COMPARE);
+}
+
+sub gen_one_insn($$)
+{
+    # Given an instruction-details array, generate an instruction
+    my $constraintfailures = 0;
+
+    INSN: while(1) {
+        my ($forcecond, $rec) = @_;
+        my $insn = int(rand(0xffffffff));
+        my $insnname = $rec->{name};
+        my $insnwidth = $rec->{width};
+        my $fixedbits = $rec->{fixedbits};
+        my $fixedbitmask = $rec->{fixedbitmask};
+        my $constraint = $rec->{blocks}{"constraints"};
+        my $memblock = $rec->{blocks}{"memory"};
+
+        $insn &= ~$fixedbitmask;
+        $insn |= $fixedbits;
+
+        if (defined $constraint) {
+            # user-specified constraint: evaluate in an environment
+            # with variables set corresponding to the variable fields.
+            my $v = eval_with_fields($insnname, $insn, $rec, "constraints", $constraint);
+            if (!$v) {
+                $constraintfailures++;
+                if ($constraintfailures > 10000) {
+                    print "10000 consecutive constraint failures for $insnname constraints string:\n$constraint\n";
+                    exit (1);
+                }
+                next INSN;
+            }
+        }
+
+        # OK, we got a good one
+        $constraintfailures = 0;
+
+        my $basereg;
+
+        if (defined $memblock) {
+            die "memblock handling has not been implemented yet."
+        }
+
+        if ($insnwidth == 16) {
+            insn16(($insn >> 16) & 0xffff);
+        } else {
+            insn32($insn);
+        }
+
+        return;
+    }
+}
+
+sub write_risuop($)
+{
+    my ($op) = @_;
+    insn32(0x835a0f00 | $op);
+}
+
+sub write_test_code($)
+{
+    my ($params) = @_;
+
+    my $condprob = $params->{ 'condprob' };
+    my $numinsns = $params->{ 'numinsns' };
+    my $outfile = $params->{ 'outfile' };
+
+    my %insn_details = %{ $params->{ 'details' } };
+    my @keys = @{ $params->{ 'keys' } };
+
+    set_endian(1);
+
+    open_bin($outfile);
+
+    # convert from probability that insn will be conditional to
+    # probability of forcing insn to unconditional
+    $condprob = 1 - $condprob;
+
+    # TODO better random number generator?
+    srand(0);
+
+    print "Generating code using patterns: @keys...\n";
+    progress_start(78, $numinsns);
+
+    if (grep { defined($insn_details{$_}->{blocks}->{"memory"}) } @keys) {
+        write_memblock_setup();
+    }
+
+    # memblock setup doesn't clean its registers, so this must come afterwards.
+    write_random_register_data();
+
+    for my $i (1..$numinsns) {
+        my $insn_enc = $keys[int rand (@keys)];
+        #dump_insn_details($insn_enc, $insn_details{$insn_enc});
+        my $forcecond = (rand() < $condprob) ? 1 : 0;
+        gen_one_insn($forcecond, $insn_details{$insn_enc});
+        write_risuop($OP_COMPARE);
+        # Rewrite the registers periodically. This avoids the tendency
+        # for the VFP registers to decay to NaNs and zeroes.
+        if ($periodic_reg_random && ($i % 100) == 0) {
+            write_random_register_data();
+        }
+        progress_update($i);
+    }
+    write_risuop($OP_TESTEND);
+    progress_end();
+    close_bin();
+}
+
+1;
-- 
2.39.3



^ permalink raw reply related	[flat|nested] 8+ messages in thread

* [risu PATCH v2 4/4] s390x: Update the configure script for s390x support
  2023-09-05 11:49 [risu PATCH v2 0/4] Add support for s390x to RISU Thomas Huth
                   ` (2 preceding siblings ...)
  2023-09-05 11:49 ` [risu PATCH v2 3/4] s390x: Add basic risugen perl module for s390x Thomas Huth
@ 2023-09-05 11:50 ` Thomas Huth
  2023-09-12 17:02   ` Peter Maydell
  3 siblings, 1 reply; 8+ messages in thread
From: Thomas Huth @ 2023-09-05 11:50 UTC (permalink / raw)
  To: Peter Maydell
  Cc: qemu-s390x, qemu-devel, Richard Henderson, Ilya Leoshkevich,
	David Hildenbrand

Auto-detect s390x hosts and add s390x information to the help text.

Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
Signed-off-by: Thomas Huth <thuth@redhat.com>
---
 configure | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)

diff --git a/configure b/configure
index ca2d7db..2f7c580 100755
--- a/configure
+++ b/configure
@@ -58,6 +58,8 @@ guess_arch() {
         ARCH="m68k"
     elif check_define __powerpc64__ ; then
         ARCH="ppc64"
+    elif check_define __s390x__ ; then
+        ARCH="s390x"
     else
         echo "This cpu is not supported by risu. Try -h. " >&2
         exit 1
@@ -139,7 +141,7 @@ Some influential environment variables:
                prefixed with the given string.
 
   ARCH         force target architecture instead of trying to detect it.
-               Valid values=[arm|aarch64|ppc64|ppc64le|m68k]
+               Valid values=[arm|aarch64|m68k|ppc64|ppc64le|s390x]
 
   CC           C compiler command
   CFLAGS       C compiler flags
-- 
2.39.3



^ permalink raw reply related	[flat|nested] 8+ messages in thread

* Re: [risu PATCH v2 1/4] s390x: Add basic s390x support to the C code
  2023-09-05 11:49 ` [risu PATCH v2 1/4] s390x: Add basic s390x support to the C code Thomas Huth
@ 2023-09-12 16:50   ` Peter Maydell
  2023-09-13  8:50     ` Thomas Huth
  0 siblings, 1 reply; 8+ messages in thread
From: Peter Maydell @ 2023-09-12 16:50 UTC (permalink / raw)
  To: Thomas Huth
  Cc: qemu-s390x, qemu-devel, Richard Henderson, Ilya Leoshkevich,
	David Hildenbrand

On Tue, 5 Sept 2023 at 12:50, Thomas Huth <thuth@redhat.com> wrote:
>
> With these changes, it is now possible to compile the "risu" binary
> for s390x hosts.
>
> Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
> Signed-off-by: Thomas Huth <thuth@redhat.com>



> +/* reginfo_init: initialize with a ucontext */
> +void reginfo_init(struct reginfo *ri, ucontext_t *uc)
> +{
> +    int i;
> +
> +    memset(ri, 0, sizeof(*ri));
> +
> +    ri->faulting_insn = *((uint32_t *) uc->uc_mcontext.psw.addr);

Here we extract the faulting instruction, assuming it to
be a 32-bit insn starting at uc_mcontext.psw.addr...

> +    ri->psw_mask = uc->uc_mcontext.psw.mask;
> +    ri->psw_addr = uc->uc_mcontext.psw.addr - image_start_address;
> +
> +    for (i = 0; i < 16; i++) {
> +        ri->gregs[i] = uc->uc_mcontext.gregs[i];
> +    }
> +
> +    memcpy(&ri->fpregs, &uc->uc_mcontext.fpregs, sizeof(fpregset_t));
> +}

> +void advance_pc(void *vuc)
> +{
> +    /*
> +     * Note: The PSW address already points to the next instruction
> +     * after we get a SIGILL, so we must not advance it here!
> +     */
> +    // ucontext_t *uc = (ucontext_t *) vuc;
> +    // uc->uc_mcontext.psw.addr += 4;

...but here we say that psw.addr points to the instruction
*after* the faulting one.

These don't seem like they can both be correct?

> +}

thanks
-- PMM


^ permalink raw reply	[flat|nested] 8+ messages in thread

* Re: [risu PATCH v2 4/4] s390x: Update the configure script for s390x support
  2023-09-05 11:50 ` [risu PATCH v2 4/4] s390x: Update the configure script for s390x support Thomas Huth
@ 2023-09-12 17:02   ` Peter Maydell
  0 siblings, 0 replies; 8+ messages in thread
From: Peter Maydell @ 2023-09-12 17:02 UTC (permalink / raw)
  To: Thomas Huth
  Cc: qemu-s390x, qemu-devel, Richard Henderson, Ilya Leoshkevich,
	David Hildenbrand

On Tue, 5 Sept 2023 at 12:50, Thomas Huth <thuth@redhat.com> wrote:
>
> Auto-detect s390x hosts and add s390x information to the help text.
>
> Reviewed-by: Philippe Mathieu-Daudé <philmd@linaro.org>
> Signed-off-by: Thomas Huth <thuth@redhat.com>
> ---
>  configure | 4 +++-
>  1 file changed, 3 insertions(+), 1 deletion(-)
>
> diff --git a/configure b/configure
> index ca2d7db..2f7c580 100755
> --- a/configure
> +++ b/configure
> @@ -58,6 +58,8 @@ guess_arch() {
>          ARCH="m68k"
>      elif check_define __powerpc64__ ; then
>          ARCH="ppc64"
> +    elif check_define __s390x__ ; then
> +        ARCH="s390x"
>      else
>          echo "This cpu is not supported by risu. Try -h. " >&2
>          exit 1
> @@ -139,7 +141,7 @@ Some influential environment variables:
>                 prefixed with the given string.
>
>    ARCH         force target architecture instead of trying to detect it.
> -               Valid values=[arm|aarch64|ppc64|ppc64le|m68k]
> +               Valid values=[arm|aarch64|m68k|ppc64|ppc64le|s390x]
>
>    CC           C compiler command
>    CFLAGS       C compiler flags
> --
> 2.39.3

You could also add s390 to the list of archs that
build-all-archs handles:

diff --git a/build-all-archs b/build-all-archs
index e5dcfc8..e89851b 100755
--- a/build-all-archs
+++ b/build-all-archs
@@ -91,7 +91,8 @@ program_exists() {
 for triplet in i386-linux-gnu i686-linux-gnu x86_64-linux-gnu \
                    aarch64-linux-gnu arm-linux-gnueabihf \
                    m68k-linux-gnu \
-                   powerpc64le-linux-gnu powerpc64-linux-gnu ; do
+                   powerpc64le-linux-gnu powerpc64-linux-gnu \
+                   s390x-linux-gnu ; do

     if ! program_exists "${triplet}-gcc"; then
         echo "Skipping ${triplet}: no compiler found"


(That script checks that all the architectures can at least
build, and that risugen on the foo.risu file works.)

thanks
-- PMM


^ permalink raw reply related	[flat|nested] 8+ messages in thread

* Re: [risu PATCH v2 1/4] s390x: Add basic s390x support to the C code
  2023-09-12 16:50   ` Peter Maydell
@ 2023-09-13  8:50     ` Thomas Huth
  0 siblings, 0 replies; 8+ messages in thread
From: Thomas Huth @ 2023-09-13  8:50 UTC (permalink / raw)
  To: Peter Maydell
  Cc: qemu-s390x, qemu-devel, Richard Henderson, Ilya Leoshkevich,
	David Hildenbrand

On 12/09/2023 18.50, Peter Maydell wrote:
> On Tue, 5 Sept 2023 at 12:50, Thomas Huth <thuth@redhat.com> wrote:
>>
>> With these changes, it is now possible to compile the "risu" binary
>> for s390x hosts.
>>
>> Acked-by: Ilya Leoshkevich <iii@linux.ibm.com>
>> Signed-off-by: Thomas Huth <thuth@redhat.com>
> 
> 
> 
>> +/* reginfo_init: initialize with a ucontext */
>> +void reginfo_init(struct reginfo *ri, ucontext_t *uc)
>> +{
>> +    int i;
>> +
>> +    memset(ri, 0, sizeof(*ri));
>> +
>> +    ri->faulting_insn = *((uint32_t *) uc->uc_mcontext.psw.addr);
> 
> Here we extract the faulting instruction, assuming it to
> be a 32-bit insn starting at uc_mcontext.psw.addr...
> 
>> +    ri->psw_mask = uc->uc_mcontext.psw.mask;
>> +    ri->psw_addr = uc->uc_mcontext.psw.addr - image_start_address;
>> +
>> +    for (i = 0; i < 16; i++) {
>> +        ri->gregs[i] = uc->uc_mcontext.gregs[i];
>> +    }
>> +
>> +    memcpy(&ri->fpregs, &uc->uc_mcontext.fpregs, sizeof(fpregset_t));
>> +}
> 
>> +void advance_pc(void *vuc)
>> +{
>> +    /*
>> +     * Note: The PSW address already points to the next instruction
>> +     * after we get a SIGILL, so we must not advance it here!
>> +     */
>> +    // ucontext_t *uc = (ucontext_t *) vuc;
>> +    // uc->uc_mcontext.psw.addr += 4;
> 
> ...but here we say that psw.addr points to the instruction
> *after* the faulting one.
> 
> These don't seem like they can both be correct?

Drat, you're right, thanks for the hint! The code in reginfo_init() is 
wrong, it takes the instruction after the illegal one. It was just working 
(wrongly, but without crashing) by accident here since the OP_TESTEND 
instruction was preceeded by an OP_COMPARE instruction :-/

To get the address (and thus opcode) of the faulting instruction, I have to 
use siginfo_t->si_addr instead of using the ucontext_t structure. I will fix 
it in the next version.

  Thomas



^ permalink raw reply	[flat|nested] 8+ messages in thread

end of thread, other threads:[~2023-09-13  8:50 UTC | newest]

Thread overview: 8+ messages (download: mbox.gz follow: Atom feed
-- links below jump to the message on this page --
2023-09-05 11:49 [risu PATCH v2 0/4] Add support for s390x to RISU Thomas Huth
2023-09-05 11:49 ` [risu PATCH v2 1/4] s390x: Add basic s390x support to the C code Thomas Huth
2023-09-12 16:50   ` Peter Maydell
2023-09-13  8:50     ` Thomas Huth
2023-09-05 11:49 ` [risu PATCH v2 2/4] s390x: Add simple s390x.risu file Thomas Huth
2023-09-05 11:49 ` [risu PATCH v2 3/4] s390x: Add basic risugen perl module for s390x Thomas Huth
2023-09-05 11:50 ` [risu PATCH v2 4/4] s390x: Update the configure script for s390x support Thomas Huth
2023-09-12 17:02   ` Peter Maydell

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).