qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: peter.maydell@linaro.org, "Lluís Vilanova" <vilanova@ac.upc.edu>,
	"Richard Henderson" <rth@twiddle.net>
Subject: [Qemu-devel] [PULL 18/32] target/arm: [tcg] Port to tb_start
Date: Wed,  6 Sep 2017 09:05:58 -0700	[thread overview]
Message-ID: <20170906160612.22769-19-richard.henderson@linaro.org> (raw)
In-Reply-To: <20170906160612.22769-1-richard.henderson@linaro.org>

From: Lluís Vilanova <vilanova@ac.upc.edu>

Incrementally paves the way towards using the generic instruction translation
loop.

Signed-off-by: Lluís Vilanova <vilanova@ac.upc.edu>
Reviewed-by: Richard Henderson <rth@twiddle.net>
Reviewed-by: Alex Benneé <alex.benee@linaro.org>
Message-Id: <150002364681.22386.1701754996184325808.stgit@frigg.lan>
[rth: Adjust for tb_start interface change.]
Signed-off-by: Richard Henderson <rth@twiddle.net>
---
 target/arm/translate.c | 82 +++++++++++++++++++++++++++-----------------------
 1 file changed, 44 insertions(+), 38 deletions(-)

diff --git a/target/arm/translate.c b/target/arm/translate.c
index a95c183cee..3138a23e0c 100644
--- a/target/arm/translate.c
+++ b/target/arm/translate.c
@@ -11893,6 +11893,49 @@ static int arm_tr_init_disas_context(DisasContextBase *dcbase,
     return max_insns;
 }
 
+static void arm_tr_tb_start(DisasContextBase *dcbase, CPUState *cpu)
+{
+    DisasContext *dc = container_of(dcbase, DisasContext, base);
+
+    /* A note on handling of the condexec (IT) bits:
+     *
+     * We want to avoid the overhead of having to write the updated condexec
+     * bits back to the CPUARMState for every instruction in an IT block. So:
+     * (1) if the condexec bits are not already zero then we write
+     * zero back into the CPUARMState now. This avoids complications trying
+     * to do it at the end of the block. (For example if we don't do this
+     * it's hard to identify whether we can safely skip writing condexec
+     * at the end of the TB, which we definitely want to do for the case
+     * where a TB doesn't do anything with the IT state at all.)
+     * (2) if we are going to leave the TB then we call gen_set_condexec()
+     * which will write the correct value into CPUARMState if zero is wrong.
+     * This is done both for leaving the TB at the end, and for leaving
+     * it because of an exception we know will happen, which is done in
+     * gen_exception_insn(). The latter is necessary because we need to
+     * leave the TB with the PC/IT state just prior to execution of the
+     * instruction which caused the exception.
+     * (3) if we leave the TB unexpectedly (eg a data abort on a load)
+     * then the CPUARMState will be wrong and we need to reset it.
+     * This is handled in the same way as restoration of the
+     * PC in these situations; we save the value of the condexec bits
+     * for each PC via tcg_gen_insn_start(), and restore_state_to_opc()
+     * then uses this to restore them after an exception.
+     *
+     * Note that there are no instructions which can read the condexec
+     * bits, and none which can write non-static values to them, so
+     * we don't need to care about whether CPUARMState is correct in the
+     * middle of a TB.
+     */
+
+    /* Reset the conditional execution bits immediately. This avoids
+       complications trying to do it at the end of the block.  */
+    if (dc->condexec_mask || dc->condexec_cond) {
+        TCGv_i32 tmp = tcg_temp_new_i32();
+        tcg_gen_movi_i32(tmp, 0);
+        store_cpu_field(tmp, condexec_bits);
+    }
+}
+
 /* generate intermediate code for basic block 'tb'.  */
 void gen_intermediate_code(CPUState *cs, TranslationBlock *tb)
 {
@@ -11932,45 +11975,8 @@ void gen_intermediate_code(CPUState *cs, TranslationBlock *tb)
     gen_tb_start(tb);
 
     tcg_clear_temp_count();
+    arm_tr_tb_start(&dc->base, cs);
 
-    /* A note on handling of the condexec (IT) bits:
-     *
-     * We want to avoid the overhead of having to write the updated condexec
-     * bits back to the CPUARMState for every instruction in an IT block. So:
-     * (1) if the condexec bits are not already zero then we write
-     * zero back into the CPUARMState now. This avoids complications trying
-     * to do it at the end of the block. (For example if we don't do this
-     * it's hard to identify whether we can safely skip writing condexec
-     * at the end of the TB, which we definitely want to do for the case
-     * where a TB doesn't do anything with the IT state at all.)
-     * (2) if we are going to leave the TB then we call gen_set_condexec()
-     * which will write the correct value into CPUARMState if zero is wrong.
-     * This is done both for leaving the TB at the end, and for leaving
-     * it because of an exception we know will happen, which is done in
-     * gen_exception_insn(). The latter is necessary because we need to
-     * leave the TB with the PC/IT state just prior to execution of the
-     * instruction which caused the exception.
-     * (3) if we leave the TB unexpectedly (eg a data abort on a load)
-     * then the CPUARMState will be wrong and we need to reset it.
-     * This is handled in the same way as restoration of the
-     * PC in these situations; we save the value of the condexec bits
-     * for each PC via tcg_gen_insn_start(), and restore_state_to_opc()
-     * then uses this to restore them after an exception.
-     *
-     * Note that there are no instructions which can read the condexec
-     * bits, and none which can write non-static values to them, so
-     * we don't need to care about whether CPUARMState is correct in the
-     * middle of a TB.
-     */
-
-    /* Reset the conditional execution bits immediately. This avoids
-       complications trying to do it at the end of the block.  */
-    if (dc->condexec_mask || dc->condexec_cond)
-      {
-        TCGv_i32 tmp = tcg_temp_new_i32();
-        tcg_gen_movi_i32(tmp, 0);
-        store_cpu_field(tmp, condexec_bits);
-      }
     do {
         dc->base.num_insns++;
         dc->insn_start_idx = tcg_op_buf_count();
-- 
2.13.5

  parent reply	other threads:[~2017-09-06 16:06 UTC|newest]

Thread overview: 36+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2017-09-06 16:05 [Qemu-devel] [PULL 00/32] tcg generic translate loop Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 01/32] tcg: Add generic DISAS_NORETURN Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 02/32] target/i386: Use generic DISAS_* enumerators Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 03/32] target/arm: Use DISAS_NORETURN Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 04/32] target: [tcg] Use a generic enum for DISAS_ values Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 05/32] target/arm: Delay check for magic kernel page Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 06/32] tcg: Add generic translation framework Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 07/32] target/i386: [tcg] Port to DisasContextBase Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 08/32] target/i386: [tcg] Port to init_disas_context Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 09/32] target/i386: [tcg] Port to insn_start Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 10/32] target/i386: [tcg] Port to breakpoint_check Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 11/32] target/i386: [tcg] Port to translate_insn Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 12/32] target/i386: [tcg] Port to tb_stop Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 13/32] target/i386: [tcg] Port to disas_log Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 14/32] target/i386: [tcg] Port to generic translation framework Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 15/32] target/arm: [tcg] Port to DisasContextBase Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 16/32] target/arm: [tcg] Port to init_disas_context Richard Henderson
2017-09-06 16:05 ` [Qemu-devel] [PULL 17/32] target/arm: [tcg, a64] " Richard Henderson
2017-09-06 16:05 ` Richard Henderson [this message]
2017-09-06 16:05 ` [Qemu-devel] [PULL 19/32] target/arm: [tcg] Port to insn_start Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 20/32] target/arm: [tcg, a64] " Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 21/32] target/arm: [tcg, a64] Port to breakpoint_check Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 22/32] target/arm: [tcg] Port to translate_insn Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 23/32] target/arm: [tcg, a64] " Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 24/32] target/arm: [tcg] Port to tb_stop Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 25/32] target/arm: [tcg,a64] " Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 26/32] target/arm: [tcg] Port to disas_log Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 27/32] target/arm: [tcg,a64] " Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 28/32] target/arm: [tcg] Port to generic translation framework Richard Henderson
2017-09-12 16:12   ` Laurent Desnogues
2017-09-12 16:37     ` Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 29/32] target/arm: [a64] Move page and ss checks to init_disas_context Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 30/32] target/arm: Move ss check " Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 31/32] target/arm: Split out thumb_tr_translate_insn Richard Henderson
2017-09-06 16:06 ` [Qemu-devel] [PULL 32/32] target/arm: Perform per-insn cross-page check only for Thumb Richard Henderson
2017-09-07 14:25 ` [Qemu-devel] [PULL 00/32] tcg generic translate loop 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=20170906160612.22769-19-richard.henderson@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=peter.maydell@linaro.org \
    --cc=qemu-devel@nongnu.org \
    --cc=rth@twiddle.net \
    --cc=vilanova@ac.upc.edu \
    /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).