From: Cornelia Huck <cohuck@redhat.com>
To: Peter Maydell <peter.maydell@linaro.org>
Cc: qemu-s390x@nongnu.org, qemu-devel@nongnu.org,
David Hildenbrand <david@redhat.com>,
Cornelia Huck <cohuck@redhat.com>
Subject: [Qemu-devel] [PULL 13/27] s390x/tcg: Fix TEST DATA CLASS instructions
Date: Mon, 4 Mar 2019 13:01:56 +0100 [thread overview]
Message-ID: <20190304120210.31500-14-cohuck@redhat.com> (raw)
In-Reply-To: <20190304120210.31500-1-cohuck@redhat.com>
From: David Hildenbrand <david@redhat.com>
Let's detect normal and denormal ("subnormal") numbers reliably. Also
test for quiet NaN's. As only one class is possible, test common cases
first.
While at it, use a better check to test for the mask bits in the data
class mask. The data class mask has 12 bits, whereby bit 0 is the
leftmost bit and bit 11 the rightmost bit. In the PoP an easy to read
table with the numbers is provided for the VECTOR FP TEST DATA CLASS
IMMEDIATE instruction, the table for TEST DATA CLASS is more confusing
as it is based on 64 bit values.
Factor the checks out into separate functions, as they will also be
needed for floating point vector instructions. We can use a makro to
generate the functions.
Signed-off-by: David Hildenbrand <david@redhat.com>
Message-Id: <20190218122710.23639-2-david@redhat.com>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Cornelia Huck <cohuck@redhat.com>
---
target/s390x/fpu_helper.c | 85 ++++++++++++++++-----------------------
1 file changed, 35 insertions(+), 50 deletions(-)
diff --git a/target/s390x/fpu_helper.c b/target/s390x/fpu_helper.c
index e921172bc424..0e9247bf7ef4 100644
--- a/target/s390x/fpu_helper.c
+++ b/target/s390x/fpu_helper.c
@@ -645,67 +645,52 @@ uint64_t HELPER(msdb)(CPUS390XState *env, uint64_t f1,
return ret;
}
+/* The rightmost bit has the number 11. */
+static inline uint16_t dcmask(int bit, bool neg)
+{
+ return 1 << (11 - bit - neg);
+}
+
+#define DEF_FLOAT_DCMASK(_TYPE) \
+static uint16_t _TYPE##_dcmask(CPUS390XState *env, _TYPE f1) \
+{ \
+ const bool neg = _TYPE##_is_neg(f1); \
+ \
+ /* Sorted by most common cases - only one class is possible */ \
+ if (_TYPE##_is_normal(f1)) { \
+ return dcmask(2, neg); \
+ } else if (_TYPE##_is_zero(f1)) { \
+ return dcmask(0, neg); \
+ } else if (_TYPE##_is_denormal(f1)) { \
+ return dcmask(4, neg); \
+ } else if (_TYPE##_is_infinity(f1)) { \
+ return dcmask(6, neg); \
+ } else if (_TYPE##_is_quiet_nan(f1, &env->fpu_status)) { \
+ return dcmask(8, neg); \
+ } \
+ /* signaling nan, as last remaining case */ \
+ return dcmask(10, neg); \
+}
+DEF_FLOAT_DCMASK(float32)
+DEF_FLOAT_DCMASK(float64)
+DEF_FLOAT_DCMASK(float128)
+
/* test data class 32-bit */
uint32_t HELPER(tceb)(CPUS390XState *env, uint64_t f1, uint64_t m2)
{
- float32 v1 = f1;
- int neg = float32_is_neg(v1);
- uint32_t cc = 0;
-
- if ((float32_is_zero(v1) && (m2 & (1 << (11-neg)))) ||
- (float32_is_infinity(v1) && (m2 & (1 << (5-neg)))) ||
- (float32_is_any_nan(v1) && (m2 & (1 << (3-neg)))) ||
- (float32_is_signaling_nan(v1, &env->fpu_status) &&
- (m2 & (1 << (1-neg))))) {
- cc = 1;
- } else if (m2 & (1 << (9-neg))) {
- /* assume normalized number */
- cc = 1;
- }
- /* FIXME: denormalized? */
- return cc;
+ return (m2 & float32_dcmask(env, f1)) != 0;
}
/* test data class 64-bit */
uint32_t HELPER(tcdb)(CPUS390XState *env, uint64_t v1, uint64_t m2)
{
- int neg = float64_is_neg(v1);
- uint32_t cc = 0;
-
- if ((float64_is_zero(v1) && (m2 & (1 << (11-neg)))) ||
- (float64_is_infinity(v1) && (m2 & (1 << (5-neg)))) ||
- (float64_is_any_nan(v1) && (m2 & (1 << (3-neg)))) ||
- (float64_is_signaling_nan(v1, &env->fpu_status) &&
- (m2 & (1 << (1-neg))))) {
- cc = 1;
- } else if (m2 & (1 << (9-neg))) {
- /* assume normalized number */
- cc = 1;
- }
- /* FIXME: denormalized? */
- return cc;
+ return (m2 & float64_dcmask(env, v1)) != 0;
}
/* test data class 128-bit */
-uint32_t HELPER(tcxb)(CPUS390XState *env, uint64_t ah,
- uint64_t al, uint64_t m2)
-{
- float128 v1 = make_float128(ah, al);
- int neg = float128_is_neg(v1);
- uint32_t cc = 0;
-
- if ((float128_is_zero(v1) && (m2 & (1 << (11-neg)))) ||
- (float128_is_infinity(v1) && (m2 & (1 << (5-neg)))) ||
- (float128_is_any_nan(v1) && (m2 & (1 << (3-neg)))) ||
- (float128_is_signaling_nan(v1, &env->fpu_status) &&
- (m2 & (1 << (1-neg))))) {
- cc = 1;
- } else if (m2 & (1 << (9-neg))) {
- /* assume normalized number */
- cc = 1;
- }
- /* FIXME: denormalized? */
- return cc;
+uint32_t HELPER(tcxb)(CPUS390XState *env, uint64_t ah, uint64_t al, uint64_t m2)
+{
+ return (m2 & float128_dcmask(env, make_float128(ah, al))) != 0;
}
/* square root 32-bit */
--
2.17.2
next prev parent reply other threads:[~2019-03-04 12:03 UTC|newest]
Thread overview: 29+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-03-04 12:01 [Qemu-devel] [PULL 00/27] s390x updates Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 01/27] s390x: Use cpu_to_be64 in SIGP STORE ADDITIONAL STATUS Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 02/27] s390x: use a QEMU-style typedef + name for SIGP save area struct Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 03/27] s390x/tcg: Save vregs to extended mchk save area Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 04/27] s390x/vfio-ap: Implement hot plug/unplug of vfio-ap device Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 05/27] s390x/vfio-ap: document " Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 06/27] s390x/tcg: RXE has an optional M3 field Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 07/27] s390x/tcg: Simplify disassembler operands initialization Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 08/27] s390x/tcg: Clarify terminology in vec_reg_offset() Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 09/27] s390x/tcg: Factor out vec_full_reg_offset() Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 10/27] s390x/tcg: Factor out gen_addi_and_wrap_i64() from get_address() Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 11/27] s390x/tcg: Implement LOAD LENGTHENED short HFP to long HFP Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 12/27] s390x/tcg: Implement LOAD COUNT TO BLOCK BOUNDARY Cornelia Huck
2019-03-04 12:01 ` Cornelia Huck [this message]
2019-03-04 12:01 ` [Qemu-devel] [PULL 14/27] s390x/tcg: Fix rounding from float128 to uint64_t/uint32_t Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 15/27] s390x/tcg: Factor out conversion of softfloat exceptions Cornelia Huck
2019-03-04 12:01 ` [Qemu-devel] [PULL 16/27] s390x/tcg: Fix parts of IEEE exception handling Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 17/27] s390x/tcg: Hide IEEE underflows in some scenarios Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 18/27] s390x/tcg: Refactor SET FPC AND SIGNAL handling Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 19/27] s390x/tcg: Fix simulated-IEEE exceptions Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 20/27] s390x/tcg: Handle SET FPC AND LOAD FPC 3-bit BFP rounding modes Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 21/27] s390x/tcg: Check for exceptions in SET BFP ROUNDING MODE Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 22/27] s390x/tcg: Refactor saving/restoring the bfp rounding mode Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 23/27] s390x/tcg: Prepare for IEEE-inexact-exception control (XxC) Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 24/27] s390x/tcg: Implement XxC and checks for most FP instructions Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 25/27] s390x/tcg: Implement rounding mode and XxC for LOAD ROUNDED Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 26/27] s390x/tcg: Handle all rounding modes overwritten by BFP instructions Cornelia Huck
2019-03-04 12:02 ` [Qemu-devel] [PULL 27/27] s390x: Add floating-point extension facility to "qemu" cpu model Cornelia Huck
2019-03-04 14:29 ` [Qemu-devel] [PULL 00/27] s390x updates 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=20190304120210.31500-14-cohuck@redhat.com \
--to=cohuck@redhat.com \
--cc=david@redhat.com \
--cc=peter.maydell@linaro.org \
--cc=qemu-devel@nongnu.org \
--cc=qemu-s390x@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).