qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: David Gibson <david@gibson.dropbear.id.au>
To: peter.maydell@linaro.org, groug@kaod.org
Cc: Richard Henderson <richard.henderson@linaro.org>,
	qemu-ppc@nongnu.org, qemu-devel@nongnu.org,
	David Gibson <david@gibson.dropbear.id.au>
Subject: [PULL 03/42] target/ppc: reduce usage of fpscr_set_rounding_mode
Date: Thu,  3 Jun 2021 18:21:52 +1000	[thread overview]
Message-ID: <20210603082231.601214-4-david@gibson.dropbear.id.au> (raw)
In-Reply-To: <20210603082231.601214-1-david@gibson.dropbear.id.au>

From: "Bruno Larsen (billionai)" <bruno.larsen@eldorado.org.br>

It is preferable to store the current rounding mode and retore from that
than recalculating from fpscr, so we changed the behavior of do_fri and
VSX_ROUND to do it like that.

Suggested-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Bruno Larsen (billionai) <bruno.larsen@eldorado.org.br>
Reviewed-by: Richard Henderson <richard.henderson@linaro.org>
Message-Id: <20210521201759.85475-4-bruno.larsen@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
---
 target/ppc/fpu_helper.c | 8 +++++---
 1 file changed, 5 insertions(+), 3 deletions(-)

diff --git a/target/ppc/fpu_helper.c b/target/ppc/fpu_helper.c
index 44315fca0b..a4a283df2b 100644
--- a/target/ppc/fpu_helper.c
+++ b/target/ppc/fpu_helper.c
@@ -822,6 +822,7 @@ static inline uint64_t do_fri(CPUPPCState *env, uint64_t arg,
                               int rounding_mode)
 {
     CPU_DoubleU farg;
+    FloatRoundMode old_rounding_mode = get_float_rounding_mode(&env->fp_status);
 
     farg.ll = arg;
 
@@ -834,8 +835,7 @@ static inline uint64_t do_fri(CPUPPCState *env, uint64_t arg,
                       float_flag_inexact;
         set_float_rounding_mode(rounding_mode, &env->fp_status);
         farg.ll = float64_round_to_int(farg.d, &env->fp_status);
-        /* Restore rounding mode from FPSCR */
-        fpscr_set_rounding_mode(env);
+        set_float_rounding_mode(old_rounding_mode, &env->fp_status);
 
         /* fri* does not set FPSCR[XX] */
         if (!inexact) {
@@ -3136,8 +3136,10 @@ void helper_##op(CPUPPCState *env, ppc_vsr_t *xt, ppc_vsr_t *xb)       \
 {                                                                      \
     ppc_vsr_t t = *xt;                                                 \
     int i;                                                             \
+    FloatRoundMode curr_rounding_mode;                                 \
                                                                        \
     if (rmode != FLOAT_ROUND_CURRENT) {                                \
+        curr_rounding_mode = get_float_rounding_mode(&env->fp_status); \
         set_float_rounding_mode(rmode, &env->fp_status);               \
     }                                                                  \
                                                                        \
@@ -3160,7 +3162,7 @@ void helper_##op(CPUPPCState *env, ppc_vsr_t *xt, ppc_vsr_t *xb)       \
      * mode from FPSCR                                                 \
      */                                                                \
     if (rmode != FLOAT_ROUND_CURRENT) {                                \
-        fpscr_set_rounding_mode(env);                                  \
+        set_float_rounding_mode(curr_rounding_mode, &env->fp_status);  \
         env->fp_status.float_exception_flags &= ~float_flag_inexact;   \
     }                                                                  \
                                                                        \
-- 
2.31.1



  parent reply	other threads:[~2021-06-03  8:24 UTC|newest]

Thread overview: 45+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-06-03  8:21 [PULL 00/42] ppc-for-6.1 queue 20210603 David Gibson
2021-06-03  8:21 ` [PULL 01/42] target/ppc: cleaned error_report from ppc_store_sdr1 David Gibson
2021-06-03  8:21 ` [PULL 02/42] target/ppc: moved ppc_store_lpcr and ppc_store_msr to cpu.c David Gibson
2021-06-03  8:21 ` David Gibson [this message]
2021-06-03  8:21 ` [PULL 04/42] target/ppc: removed unnecessary inclusion of helper-proto.h David Gibson
2021-06-03  8:21 ` [PULL 05/42] spapr: Don't hijack current_machine->boot_order David Gibson
2021-06-03  8:21 ` [PULL 06/42] spapr: Fix EEH capability issue on KVM guest for PCI passthru David Gibson
2021-06-03  8:21 ` [PULL 07/42] spapr: nvdimm: Forward declare and move the definitions David Gibson
2021-06-03  8:21 ` [PULL 08/42] spapr: nvdimm: Fix the persistent-memory root node name in device tree David Gibson
2021-06-03  8:21 ` [PULL 09/42] target/ppc: fold ppc_store_ptcr into it's only caller David Gibson
2021-06-03  8:21 ` [PULL 10/42] spapr: Remove stale comment about power-saving LPCR bits David Gibson
2021-06-03  8:22 ` [PULL 11/42] spapr: Set LPCR to current AIL mode when starting a new CPU David Gibson
2021-06-03  8:22 ` [PULL 12/42] target/ppc: used ternary operator when registering MAS David Gibson
2021-06-03  8:22 ` [PULL 13/42] target/ppc: added ifdefs around TCG-only code David Gibson
2021-06-03  8:22 ` [PULL 14/42] target/ppc: created tcg-stub.c file David Gibson
2021-06-03  8:22 ` [PULL 15/42] target/ppc: updated meson.build to support disable-tcg David Gibson
2021-06-03  8:22 ` [PULL 16/42] target/ppc: remove ppc_cpu_dump_statistics David Gibson
2021-06-03  8:22 ` [PULL 17/42] target/ppc: removed mentions to DO_PPC_STATISTICS David Gibson
2021-06-03  8:22 ` [PULL 18/42] monitor: removed cpustats command David Gibson
2021-06-03  8:22 ` [PULL 19/42] ppc/pef.c: initialize cgs->ready in kvmppc_svm_init() David Gibson
2021-06-03  8:22 ` [PULL 20/42] hw/core/cpu: removed cpu_dump_statistics function David Gibson
2021-06-03  8:22 ` [PULL 21/42] HMP: added info cpustats to removed_features.rst David Gibson
2021-06-03  8:22 ` [PULL 22/42] target/ppc: removed GEN_OPCODE decision tree David Gibson
2021-06-03  8:22 ` [PULL 23/42] target/ppc: removed all mentions to PPC_DUMP_CPU David Gibson
2021-06-03  8:22 ` [PULL 24/42] target/ppc: overhauled and moved logic of storing fpscr David Gibson
2021-06-03  8:22 ` [PULL 25/42] target/ppc: powerpc_excp: Move lpes code to where it is used David Gibson
2021-06-03  8:22 ` [PULL 26/42] target/ppc: powerpc_excp: Remove dump_syscall_vectored David Gibson
2021-06-03  8:22 ` [PULL 27/42] target/ppc: powerpc_excp: Consolidade TLB miss code David Gibson
2021-06-03  8:22 ` [PULL 28/42] target/ppc: Introduce macros to check isa extensions David Gibson
2021-06-03  8:22 ` [PULL 29/42] target/ppc: Move page crossing check to ppc_tr_translate_insn David Gibson
2021-06-03  8:22 ` [PULL 30/42] target/ppc: Add infrastructure for prefixed insns David Gibson
2021-06-03  8:22 ` [PULL 31/42] target/ppc: Move ADDI, ADDIS to decodetree, implement PADDI David Gibson
2021-06-03  8:22 ` [PULL 32/42] target/ppc: Implement PNOP David Gibson
2021-06-03  8:22 ` [PULL 33/42] target/ppc: Move D/DS/X-form integer loads to decodetree David Gibson
2021-06-03  8:22 ` [PULL 34/42] target/ppc: Implement prefixed integer load instructions David Gibson
2021-06-03  8:22 ` [PULL 35/42] target/ppc: Move D/DS/X-form integer stores to decodetree David Gibson
2021-06-03  8:22 ` [PULL 36/42] target/ppc: Implement prefixed integer store instructions David Gibson
2021-06-03  8:22 ` [PULL 37/42] target/ppc: Implement setbc/setbcr/stnbc/setnbcr instructions David Gibson
2021-06-03  8:22 ` [PULL 38/42] target/ppc: Implement cfuged instruction David Gibson
2021-06-03  8:22 ` [PULL 39/42] target/ppc: Implement vcfuged instruction David Gibson
2021-06-03  8:22 ` [PULL 40/42] target/ppc: Move addpcis to decodetree David Gibson
2021-06-03  8:22 ` [PULL 41/42] target/ppc: Move cmp/cmpi/cmpl/cmpli " David Gibson
2021-06-03  8:22 ` [PULL 42/42] target/ppc: fix single-step exception regression David Gibson
2021-06-03  9:01 ` [PULL 00/42] ppc-for-6.1 queue 20210603 no-reply
2021-06-03 10:36 ` 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=20210603082231.601214-4-david@gibson.dropbear.id.au \
    --to=david@gibson.dropbear.id.au \
    --cc=groug@kaod.org \
    --cc=peter.maydell@linaro.org \
    --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).