On Thu, Mar 31, 2022 at 10:15 PM David Miller <dmiller...@gmail.com> wrote:
>
> Hi,
>
> There is some issue with instruction sub/alt encodings not matching,
> but I worked around it easily.
>
> I'm dropping the updated patch for the tests in here.
> I know I should resend the entire patch series as a higher version
> really, and will do so.
> I'm hoping someone can tell me if it's ok to use .insn vrr in place
> of vri(-d) as it doesn't match vri.
> [https://sourceware.org/binutils/docs-2.37/as/s390-Formats.html]
>
> .insn doesn't deal with sub encodings and there is no good alternative
> that I know of.
>
> example:
>
> /* vri-d as vrr */
> asm volatile(".insn vrr, 0xE70000000086, %[v1], %[v2], %[v3], 0, %[I],
> 0\n"
> : [v1] "=v" (v1->v)
> : [v2] "v" (v2->v)
> , [v3] "v" (v3->v)
> , [I] "i" (I & 7));
>
> Patch is attached
>
>
> Thanks
> - David Miller
>
>
> On Thu, Mar 31, 2022 at 2:26 PM David Miller <dmiller...@gmail.com> wrote:
> >
> > Sorry,
> > Didn't notice this, as it was on v4 patch emails.
> > I assume since there is no other follow up after a week,
> > CI jobs are not being updated and I should change samples to use .insn.
> > I will try to get this out tomorrow.
> >
> > Thanks,
> > - David Miller
> >
> > On Wed, Mar 23, 2022 at 1:13 PM Thomas Huth <th...@redhat.com> wrote:
> > >
> > > On 22/03/2022 11.31, Thomas Huth wrote:
> > > > On 22/03/2022 09.53, David Hildenbrand wrote:
> > > >> On 22.03.22 01:04, David Miller wrote:
> > > > [...]
> > > >>> diff --git a/tests/tcg/s390x/Makefile.target
> > > >>> b/tests/tcg/s390x/Makefile.target
> > > >>> index 8c9b6a13ce..921a056dd1 100644
> > > >>> --- a/tests/tcg/s390x/Makefile.target
> > > >>> +++ b/tests/tcg/s390x/Makefile.target
> > > >>> @@ -16,6 +16,14 @@ TESTS+=shift
> > > >>> TESTS+=trap
> > > >>> TESTS+=signals-s390x
> > > >>> +VECTOR_TESTS=vxeh2_vs
> > > >>> +VECTOR_TESTS+=vxeh2_vcvt
> > > >>> +VECTOR_TESTS+=vxeh2_vlstr
> > > >>> +
> > > >>> +TESTS+=$(VECTOR_TESTS)
> > > >>> +
> > > >>> +$(VECTOR_TESTS): CFLAGS+=-march=z15 -O2
> > > >>
> > > >> @Thomas, will that survive our test framework already, or do we have to
> > > >> wait for the debain11 changes?
> > > >
> > > > Alex' update to the container has already been merged:
> > > >
> > > > https://gitlab.com/qemu-project/qemu/-/commit/89767579cad2e371b
> > > >
> > > > ... and seems like it's working in Travis on s390x, too:
> > > >
> > > > https://app.travis-ci.com/github/huth/qemu/jobs/564188977#L12797
> > > >
> > > > ... so it seems like it should be OK now (considering that we drop
> > > > support
> > > > for the old Ubuntu version 18.04 in QEMU 7.1, too).
> > >
> > > Looks like I spoke a little bit too soon - some of the CI pipelines are
> > > still using Debian 10 for running the TCG tests, and they are failing with
> > > these patches applied:
> > >
> > > https://gitlab.com/thuth/qemu/-/jobs/2238422870#L3499
> > >
> > > Thus we either need to update the CI jobs to use Debian 11, or use
> > > handcrafted instruction opcodes here again...
> > >
> > > Thomas
> > >
From bb6bf2f9529c4d76db9a9eff2ff7fa1235657103 Mon Sep 17 00:00:00 2001
From: David Miller <dmiller...@gmail.com>
Date: Mon, 21 Mar 2022 16:58:57 -0400
Subject: [PATCH v5 10/11] tests/tcg/s390x: Tests for Vector Enhancements
Facility 2
Signed-off-by: David Miller <dmiller...@gmail.com>
---
tests/tcg/s390x/Makefile.target | 8 ++
tests/tcg/s390x/vx.h | 19 +++++
tests/tcg/s390x/vxeh2_vcvt.c | 88 ++++++++++++++++++++
tests/tcg/s390x/vxeh2_vlstr.c | 139 ++++++++++++++++++++++++++++++++
tests/tcg/s390x/vxeh2_vs.c | 95 ++++++++++++++++++++++
5 files changed, 349 insertions(+)
create mode 100644 tests/tcg/s390x/vx.h
create mode 100644 tests/tcg/s390x/vxeh2_vcvt.c
create mode 100644 tests/tcg/s390x/vxeh2_vlstr.c
create mode 100644 tests/tcg/s390x/vxeh2_vs.c
diff --git a/tests/tcg/s390x/Makefile.target b/tests/tcg/s390x/Makefile.target
index 8c9b6a13ce..921a056dd1 100644
--- a/tests/tcg/s390x/Makefile.target
+++ b/tests/tcg/s390x/Makefile.target
@@ -16,6 +16,14 @@ TESTS+=shift
TESTS+=trap
TESTS+=signals-s390x
+VECTOR_TESTS=vxeh2_vs
+VECTOR_TESTS+=vxeh2_vcvt
+VECTOR_TESTS+=vxeh2_vlstr
+
+TESTS+=$(VECTOR_TESTS)
+
+$(VECTOR_TESTS): CFLAGS+=-march=z15 -O2
+
ifneq ($(HAVE_GDB_BIN),)
GDB_SCRIPT=$(SRC_PATH)/tests/guest-debug/run-test.py
diff --git a/tests/tcg/s390x/vx.h b/tests/tcg/s390x/vx.h
new file mode 100644
index 0000000000..2e66f8b714
--- /dev/null
+++ b/tests/tcg/s390x/vx.h
@@ -0,0 +1,19 @@
+#ifndef QEMU_TESTS_S390X_VX_H
+#define QEMU_TESTS_S390X_VX_H
+
+typedef union S390Vector {
+ uint64_t d[2]; /* doubleword */
+ uint32_t w[4]; /* word */
+ uint16_t h[8]; /* halfword */
+ uint8_t b[16]; /* byte */
+ float f[4]; /* float32 */
+ double fd[2]; /* float64 */
+ __uint128_t v;
+} S390Vector;
+
+#define ES8 0
+#define ES16 1
+#define ES32 2
+#define ES64 3
+
+#endif
\ No newline at end of file
diff --git a/tests/tcg/s390x/vxeh2_vcvt.c b/tests/tcg/s390x/vxeh2_vcvt.c
new file mode 100644
index 0000000000..2e46841ab5
--- /dev/null
+++ b/tests/tcg/s390x/vxeh2_vcvt.c
@@ -0,0 +1,88 @@
+/*
+ * vxeh2_vcvt: vector-enhancements facility 2 vector convert *
+ */
+#include <stdint.h>
+#include "vx.h"
+
+#define M_S 8
+#define M4_XxC 4
+#define M4_def M4_XxC
+
+static inline void vcfps(S390Vector *v1, S390Vector *v2,
+ const uint8_t m3, const uint8_t m4, const uint8_t m5)
+{
+ asm volatile(".insn vrr, 0xE700000000C3, %[v1], %[v2], 0, %[m3], %[m4], %[m5]\n"
+ : [v1] "=v" (v1->v)
+ : [v2] "v" (v2->v)
+ , [m3] "i" (m3)
+ , [m4] "i" (m4)
+ , [m5] "i" (m5));
+}
+
+static inline void vcfpl(S390Vector *v1, S390Vector *v2,
+ const uint8_t m3, const uint8_t m4, const uint8_t m5)
+{
+ asm volatile(".insn vrr, 0xE700000000C1, %[v1], %[v2], 0, %[m3], %[m4], %[m5]\n"
+ : [v1] "=v" (v1->v)
+ : [v2] "v" (v2->v)
+ , [m3] "i" (m3)
+ , [m4] "i" (m4)
+ , [m5] "i" (m5));
+}
+
+static inline void vcsfp(S390Vector *v1, S390Vector *v2,
+ const uint8_t m3, const uint8_t m4, const uint8_t m5)
+{
+ asm volatile(".insn vrr, 0xE700000000C2, %[v1], %[v2], 0, %[m3], %[m4], %[m5]\n"
+ : [v1] "=v" (v1->v)
+ : [v2] "v" (v2->v)
+ , [m3] "i" (m3)
+ , [m4] "i" (m4)
+ , [m5] "i" (m5));
+}
+
+static inline void vclfp(S390Vector *v1, S390Vector *v2,
+ const uint8_t m3, const uint8_t m4, const uint8_t m5)
+{
+ asm volatile(".insn vrr, 0xE700000000C0, %[v1], %[v2], 0, %[m3], %[m4], %[m5]\n"
+ : [v1] "=v" (v1->v)
+ : [v2] "v" (v2->v)
+ , [m3] "i" (m3)
+ , [m4] "i" (m4)
+ , [m5] "i" (m5));
+}
+
+int main(int argc, char *argv[])
+{
+ S390Vector vd;
+ S390Vector vs_i32 = { .w[0] = 1, .w[1] = 64, .w[2] = 1024, .w[3] = -10 };
+ S390Vector vs_u32 = { .w[0] = 2, .w[1] = 32, .w[2] = 4096, .w[3] = 8888 };
+ S390Vector vs_f32 = { .f[0] = 3.987, .f[1] = 5.123,
+ .f[2] = 4.499, .f[3] = 0.512 };
+
+ vd.d[0] = vd.d[1] = 0;
+ vcfps(&vd, &vs_i32, 2, M4_def, 0);
+ if (1 != vd.f[0] || 1024 != vd.f[2] || 64 != vd.f[1] || -10 != vd.f[3]) {
+ return 1;
+ }
+
+ vd.d[0] = vd.d[1] = 0;
+ vcfpl(&vd, &vs_u32, 2, M4_def, 0);
+ if (2 != vd.f[0] || 4096 != vd.f[2] || 32 != vd.f[1] || 8888 != vd.f[3]) {
+ return 1;
+ }
+
+ vd.d[0] = vd.d[1] = 0;
+ vcsfp(&vd, &vs_f32, 2, M4_def, 0);
+ if (4 != vd.w[0] || 4 != vd.w[2] || 5 != vd.w[1] || 1 != vd.w[3]) {
+ return 1;
+ }
+
+ vd.d[0] = vd.d[1] = 0;
+ vclfp(&vd, &vs_f32, 2, M4_def, 0);
+ if (4 != vd.w[0] || 4 != vd.w[2] || 5 != vd.w[1] || 1 != vd.w[3]) {
+ return 1;
+ }
+
+ return 0;
+}
diff --git a/tests/tcg/s390x/vxeh2_vlstr.c b/tests/tcg/s390x/vxeh2_vlstr.c
new file mode 100644
index 0000000000..770691a4e8
--- /dev/null
+++ b/tests/tcg/s390x/vxeh2_vlstr.c
@@ -0,0 +1,139 @@
+/*
+ * vxeh2_vlstr: vector-enhancements facility 2 vector load/store reversed *
+ */
+#include <stdint.h>
+#include "vx.h"
+
+#define vtst(v1, v2) \
+ if (v1.d[0] != v2.d[0] || v1.d[1] != v2.d[1]) { \
+ return 1; \
+ }
+
+static inline void vler(S390Vector *v1, const void *va, uint8_t m3)
+{
+ asm volatile(".insn vrx, 0xE60000000007, %[v1], 0(%[va]), %[m3]\n"
+ : [v1] "+v" (v1->v)
+ : [va] "d" (va)
+ , [m3] "i" (m3)
+ : "memory");
+}
+
+static inline void vster(S390Vector *v1, const void *va, uint8_t m3)
+{
+ asm volatile(".insn vrx, 0xE6000000000F, %[v1], 0(%[va]), %[m3]\n"
+ : [va] "+d" (va)
+ : [v1] "v" (v1->v)
+ , [m3] "i" (m3)
+ : "memory");
+}
+
+static inline void vlbr(S390Vector *v1, void *va, const uint8_t m3)
+{
+ asm volatile(".insn vrx, 0xE60000000006, %[v1], 0(%[va]), %[m3]\n"
+ : [v1] "+v" (v1->v)
+ : [va] "d" (va)
+ , [m3] "i" (m3)
+ : "memory");
+}
+
+static inline void vstbr(S390Vector *v1, void *va, const uint8_t m3)
+{
+ asm volatile(".insn vrx, 0xE6000000000E, %[v1], 0(%[va]), %[m3]\n"
+ : [va] "+d" (va)
+ : [v1] "v" (v1->v)
+ , [m3] "i" (m3)
+ : "memory");
+}
+
+
+static inline void vlebrh(S390Vector *v1, void *va, const uint8_t m3)
+{
+ asm volatile(".insn vrx, 0xE60000000001, %[v1], 0(%[va]), %[m3]\n"
+ : [v1] "+v" (v1->v)
+ : [va] "d" (va)
+ , [m3] "i" (m3)
+ : "memory");
+}
+
+static inline void vstebrh(S390Vector *v1, void *va, const uint8_t m3)
+{
+ asm volatile(".insn vrx, 0xE60000000009, %[v1], 0(%[va]), %[m3]\n"
+ : [va] "+d" (va)
+ : [v1] "v" (v1->v)
+ , [m3] "i" (m3)
+ : "memory");
+}
+
+static inline void vllebrz(S390Vector *v1, void *va, const uint8_t m3)
+{
+ asm volatile(".insn vrx, 0xE60000000004, %[v1], 0(%[va]), %[m3]\n"
+ : [v1] "+v" (v1->v)
+ : [va] "d" (va)
+ , [m3] "i" (m3)
+ : "memory");
+}
+
+static inline void vlbrrep(S390Vector *v1, void *va, const uint8_t m3)
+{
+ asm volatile(".insn vrx, 0xE60000000005, %[v1], 0(%[va]), %[m3]\n"
+ : [v1] "+v" (v1->v)
+ : [va] "d" (va)
+ , [m3] "i" (m3)
+ : "memory");
+}
+
+int main(int argc, char *argv[])
+{
+ S390Vector vd = { .d[0] = 0, .d[1] = 0 };
+ S390Vector vs = { .d[0] = 0x8FEEDDCCBBAA9988ull,
+ .d[1] = 0x7766554433221107ull };
+
+ const S390Vector vt_v_er16 = {
+ .h[0] = 0x1107, .h[1] = 0x3322, .h[2] = 0x5544, .h[3] = 0x7766,
+ .h[4] = 0x9988, .h[5] = 0xBBAA, .h[6] = 0xDDCC, .h[7] = 0x8FEE };
+
+ const S390Vector vt_v_br16 = {
+ .h[0] = 0xEE8F, .h[1] = 0xCCDD, .h[2] = 0xAABB, .h[3] = 0x8899,
+ .h[4] = 0x6677, .h[5] = 0x4455, .h[6] = 0x2233, .h[7] = 0x0711 };
+
+ int ix;
+ uint64_t ss64 = 0xFEEDFACE0BADBEEFull, sd64 = 0;
+
+ vler(&vd, &vs, ES16);
+ vtst(vd, vt_v_er16);
+
+ vster(&vs, &vd, ES16);
+ vtst(vd, vt_v_er16);
+
+ vlbr(&vd, &vs, ES16);
+ vtst(vd, vt_v_br16);
+
+ vstbr(&vs, &vd, ES16);
+ vtst(vd, vt_v_br16);
+
+ vlebrh(&vd, &ss64, 5);
+ if (0xEDFE != vd.h[5]) {
+ return 1;
+ }
+
+ vstebrh(&vs, (uint8_t *)&sd64 + 4, 7);
+ if (0x0000000007110000ull != sd64) {
+ return 1;
+ }
+
+ vllebrz(&vd, (uint8_t *)&ss64 + 3, 2);
+ for (ix = 0; ix < 4; ix++) {
+ if (vd.w[ix] != (ix != 1 ? 0 : 0xBEAD0BCE)) {
+ return 1;
+ }
+ }
+
+ vlbrrep(&vd, (uint8_t *)&ss64 + 4, 1);
+ for (ix = 0; ix < 8; ix++) {
+ if (0xAD0B != vd.h[ix]) {
+ return 1;
+ }
+ }
+
+ return 0;
+}
diff --git a/tests/tcg/s390x/vxeh2_vs.c b/tests/tcg/s390x/vxeh2_vs.c
new file mode 100644
index 0000000000..78f5c9a8be
--- /dev/null
+++ b/tests/tcg/s390x/vxeh2_vs.c
@@ -0,0 +1,95 @@
+/*
+ * vxeh2_vs: vector-enhancements facility 2 vector shift
+ */
+#include <stdint.h>
+#include "vx.h"
+
+#define vtst(v1, v2) \
+ if (v1.d[0] != v2.d[0] || v1.d[1] != v2.d[1]) { \
+ return 1; \
+ }
+
+static inline void vsl(S390Vector *v1, S390Vector *v2, S390Vector *v3)
+{
+ asm volatile(".insn vrr, 0xE70000000074, %[v1], %[v2], %[v3], 0,0,0\n"
+ : [v1] "=v" (v1->v)
+ : [v2] "v" (v2->v)
+ , [v3] "v" (v3->v));
+}
+
+static inline void vsra(S390Vector *v1, S390Vector *v2, S390Vector *v3)
+{
+ asm volatile(".insn vrr, 0xE7000000007E, %[v1], %[v2], %[v3], 0,0,0\n"
+ : [v1] "=v" (v1->v)
+ : [v2] "v" (v2->v)
+ , [v3] "v" (v3->v));
+}
+
+static inline void vsrl(S390Vector *v1, S390Vector *v2, S390Vector *v3)
+{
+ asm volatile(".insn vrr, 0xE7000000007C, %[v1], %[v2], %[v3], 0,0,0\n"
+ : [v1] "=v" (v1->v)
+ : [v2] "v" (v2->v)
+ , [v3] "v" (v3->v));
+}
+
+static inline void vsld(S390Vector *v1, S390Vector *v2,
+ S390Vector *v3, const uint8_t I)
+{
+ /* vri-d as vrr */
+ asm volatile(".insn vrr, 0xE70000000086, %[v1], %[v2], %[v3], 0, %[I], 0\n"
+ : [v1] "=v" (v1->v)
+ : [v2] "v" (v2->v)
+ , [v3] "v" (v3->v)
+ , [I] "i" (I & 7));
+}
+
+static inline void vsrd(S390Vector *v1, S390Vector *v2,
+ S390Vector *v3, const uint8_t I)
+{
+ /* vri-d as vrr */
+ asm volatile(".insn vrr, 0xE70000000087, %[v1], %[v2], %[v3], 0, %[I], 0\n"
+ : [v1] "=v" (v1->v)
+ : [v2] "v" (v2->v)
+ , [v3] "v" (v3->v)
+ , [I] "i" (I & 7));
+}
+
+int main(int argc, char *argv[])
+{
+ const S390Vector vt_vsl = { .d[0] = 0x7FEDBB32D5AA311Dull,
+ .d[1] = 0xBB65AA10912220C0ull };
+ const S390Vector vt_vsra = { .d[0] = 0xF1FE6E7399AA5466ull,
+ .d[1] = 0x0E762A5188221044ull };
+ const S390Vector vt_vsrl = { .d[0] = 0x11FE6E7399AA5466ull,
+ .d[1] = 0x0E762A5188221044ull };
+ const S390Vector vt_vsld = { .d[0] = 0x7F76EE65DD54CC43ull,
+ .d[1] = 0xBB32AA2199108838ull };
+ const S390Vector vt_vsrd = { .d[0] = 0x0E060802040E000Aull,
+ .d[1] = 0x0C060802040E000Aull };
+ S390Vector vs = { .d[0] = 0x8FEEDDCCBBAA9988ull,
+ .d[1] = 0x7766554433221107ull };
+ S390Vector vd = { .d[0] = 0, .d[1] = 0 };
+ S390Vector vsi = { .d[0] = 0, .d[1] = 0 };
+
+ for (int ix = 0; ix < 16; ix++) {
+ vsi.b[ix] = (1 + (5 ^ ~ix)) & 7;
+ }
+
+ vsl(&vd, &vs, &vsi);
+ vtst(vd, vt_vsl);
+
+ vsra(&vd, &vs, &vsi);
+ vtst(vd, vt_vsra);
+
+ vsrl(&vd, &vs, &vsi);
+ vtst(vd, vt_vsrl);
+
+ vsld(&vd, &vs, &vsi, 3);
+ vtst(vd, vt_vsld);
+
+ vsrd(&vd, &vs, &vsi, 15);
+ vtst(vd, vt_vsrd);
+
+ return 0;
+}
--
2.32.0
