From: LemonBoy <thatlemon@gmail.com>
To: qemu-devel@nongnu.org
Cc: LemonBoy <thatlemon@gmail.com>,
richard.henderson@linaro.org, qemu-ppc@nongnu.org,
david@gibson.dropbear.id.au
Subject: [PATCH v2 3/4] ppc/translate: Delay NaN checking after comparison
Date: Tue, 10 Nov 2020 11:53:20 +0100 [thread overview]
Message-ID: <20201110105321.25889-4-thatlemon@gmail.com> (raw)
In-Reply-To: <20201110105321.25889-1-thatlemon@gmail.com>
Since we always perform a comparison between the two operands avoid
checking for NaN unless the result states they're unordered.
Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Giuseppe Musacchio <thatlemon@gmail.com>
---
target/ppc/fpu_helper.c | 82 +++++++++++++++++++++--------------------
1 file changed, 42 insertions(+), 40 deletions(-)
diff --git a/target/ppc/fpu_helper.c b/target/ppc/fpu_helper.c
index c74c080c17..315959f681 100644
--- a/target/ppc/fpu_helper.c
+++ b/target/ppc/fpu_helper.c
@@ -2475,25 +2475,6 @@ static inline void do_scalar_cmp(CPUPPCState *env, ppc_vsr_t *xa, ppc_vsr_t *xb,
helper_reset_fpstatus(env);
- if (float64_is_signaling_nan(xa->VsrD(0), &env->fp_status) ||
- float64_is_signaling_nan(xb->VsrD(0), &env->fp_status)) {
- vxsnan_flag = true;
- if (fpscr_ve == 0 && ordered) {
- vxvc_flag = true;
- }
- } else if (float64_is_quiet_nan(xa->VsrD(0), &env->fp_status) ||
- float64_is_quiet_nan(xb->VsrD(0), &env->fp_status)) {
- if (ordered) {
- vxvc_flag = true;
- }
- }
- if (vxsnan_flag) {
- float_invalid_op_vxsnan(env, GETPC());
- }
- if (vxvc_flag) {
- float_invalid_op_vxvc(env, 0, GETPC());
- }
-
switch (float64_compare(xa->VsrD(0), xb->VsrD(0), &env->fp_status)) {
case float_relation_less:
cc |= CRF_LT;
@@ -2506,6 +2487,27 @@ static inline void do_scalar_cmp(CPUPPCState *env, ppc_vsr_t *xa, ppc_vsr_t *xb,
break;
case float_relation_unordered:
cc |= CRF_SO;
+
+ if (float64_is_signaling_nan(xa->VsrD(0), &env->fp_status) ||
+ float64_is_signaling_nan(xb->VsrD(0), &env->fp_status)) {
+ vxsnan_flag = true;
+ if (fpscr_ve == 0 && ordered) {
+ vxvc_flag = true;
+ }
+ } else if (float64_is_quiet_nan(xa->VsrD(0), &env->fp_status) ||
+ float64_is_quiet_nan(xb->VsrD(0), &env->fp_status)) {
+ if (ordered) {
+ vxvc_flag = true;
+ }
+ }
+
+ if (vxsnan_flag) {
+ float_invalid_op_vxsnan(env, GETPC());
+ }
+ if (vxvc_flag) {
+ float_invalid_op_vxvc(env, 0, GETPC());
+ }
+
break;
}
@@ -2536,27 +2538,6 @@ static inline void do_scalar_cmpq(CPUPPCState *env, ppc_vsr_t *xa,
helper_reset_fpstatus(env);
- if (float128_is_signaling_nan(xa->f128, &env->fp_status) ||
- float128_is_signaling_nan(xb->f128, &env->fp_status)) {
- vxsnan_flag = true;
- cc = CRF_SO;
- if (fpscr_ve == 0 && ordered) {
- vxvc_flag = true;
- }
- } else if (float128_is_quiet_nan(xa->f128, &env->fp_status) ||
- float128_is_quiet_nan(xb->f128, &env->fp_status)) {
- cc = CRF_SO;
- if (ordered) {
- vxvc_flag = true;
- }
- }
- if (vxsnan_flag) {
- float_invalid_op_vxsnan(env, GETPC());
- }
- if (vxvc_flag) {
- float_invalid_op_vxvc(env, 0, GETPC());
- }
-
switch (float128_compare(xa->f128, xb->f128, &env->fp_status)) {
case float_relation_less:
cc |= CRF_LT;
@@ -2569,6 +2550,27 @@ static inline void do_scalar_cmpq(CPUPPCState *env, ppc_vsr_t *xa,
break;
case float_relation_unordered:
cc |= CRF_SO;
+
+ if (float128_is_signaling_nan(xa->f128, &env->fp_status) ||
+ float128_is_signaling_nan(xb->f128, &env->fp_status)) {
+ vxsnan_flag = true;
+ if (fpscr_ve == 0 && ordered) {
+ vxvc_flag = true;
+ }
+ } else if (float128_is_quiet_nan(xa->f128, &env->fp_status) ||
+ float128_is_quiet_nan(xb->f128, &env->fp_status)) {
+ if (ordered) {
+ vxvc_flag = true;
+ }
+ }
+
+ if (vxsnan_flag) {
+ float_invalid_op_vxsnan(env, GETPC());
+ }
+ if (vxvc_flag) {
+ float_invalid_op_vxvc(env, 0, GETPC());
+ }
+
break;
}
--
2.27.0
next prev parent reply other threads:[~2020-11-10 10:57 UTC|newest]
Thread overview: 9+ messages / expand[flat|nested] mbox.gz Atom feed top
2020-11-10 10:53 [PATCH v2 0/4] ppc/translate: Fix unordered f64/f128 comparisons LemonBoy
2020-11-10 10:53 ` [PATCH v2 1/4] " LemonBoy
2020-11-12 21:26 ` Richard Henderson
2020-11-10 10:53 ` [PATCH v2 2/4] ppc/translate: Turn the helper macros into functions LemonBoy
2020-11-12 21:27 ` Richard Henderson
2020-11-10 10:53 ` LemonBoy [this message]
2020-11-12 21:28 ` [PATCH v2 3/4] ppc/translate: Delay NaN checking after comparison Richard Henderson
2020-11-10 10:53 ` [PATCH v2 4/4] ppc/translate: Raise exceptions after setting the cc LemonBoy
2020-11-12 21:28 ` Richard Henderson
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=20201110105321.25889-4-thatlemon@gmail.com \
--to=thatlemon@gmail.com \
--cc=david@gibson.dropbear.id.au \
--cc=qemu-devel@nongnu.org \
--cc=qemu-ppc@nongnu.org \
--cc=richard.henderson@linaro.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).