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@linaro.org,
	David Gibson <david@gibson.dropbear.id.au>,
	qemu-ppc@nongnu.org, qemu-devel@nongnu.org,
	Matheus Ferst <matheus.ferst@eldorado.org.br>
Subject: [PULL 29/48] target/ppc: Move single-step check to ppc_tr_tb_stop
Date: Wed, 19 May 2021 22:51:29 +1000	[thread overview]
Message-ID: <20210519125148.27720-30-david@gibson.dropbear.id.au> (raw)
In-Reply-To: <20210519125148.27720-1-david@gibson.dropbear.id.au>

From: Richard Henderson <richard.henderson@linaro.org>

When single-stepping, force max_insns to 1 in init_disas
so that we exit the translation loop immediately.

Combine the single-step checks in tb_stop, and give the
gdb exception priority over the cpu exception, just as
we already do in gen_lookup_and_goto_ptr.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
Signed-off-by: Matheus Ferst <matheus.ferst@eldorado.org.br>
Message-Id: <20210517205025.3777947-6-matheus.ferst@eldorado.org.br>
Signed-off-by: David Gibson <david@gibson.dropbear.id.au>
---
 target/ppc/translate.c | 33 +++++++++++++++++++--------------
 1 file changed, 19 insertions(+), 14 deletions(-)

diff --git a/target/ppc/translate.c b/target/ppc/translate.c
index 80cd11b3f8..05e3c0417a 100644
--- a/target/ppc/translate.c
+++ b/target/ppc/translate.c
@@ -8992,7 +8992,6 @@ static void ppc_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
     DisasContext *ctx = container_of(dcbase, DisasContext, base);
     CPUPPCState *env = cs->env_ptr;
     uint32_t hflags = ctx->base.tb->flags;
-    int bound;
 
     ctx->spr_cb = env->spr_cb;
     ctx->pr = (hflags >> HFLAGS_PR) & 1;
@@ -9032,8 +9031,12 @@ static void ppc_tr_init_disas_context(DisasContextBase *dcbase, CPUState *cs)
         ctx->singlestep_enabled |= GDBSTUB_SINGLE_STEP;
     }
 
-    bound = -(ctx->base.pc_first | TARGET_PAGE_MASK) / 4;
-    ctx->base.max_insns = MIN(ctx->base.max_insns, bound);
+    if (ctx->singlestep_enabled & (CPU_SINGLE_STEP | GDBSTUB_SINGLE_STEP)) {
+        ctx->base.max_insns = 1;
+    } else {
+        int bound = -(ctx->base.pc_first | TARGET_PAGE_MASK) / 4;
+        ctx->base.max_insns = MIN(ctx->base.max_insns, bound);
+    }
 }
 
 static void ppc_tr_tb_start(DisasContextBase *db, CPUState *cs)
@@ -9087,14 +9090,6 @@ static void ppc_tr_translate_insn(DisasContextBase *dcbase, CPUState *cs)
     handler->count++;
 #endif
 
-    /* Check trace mode exceptions */
-    if (unlikely(ctx->singlestep_enabled & CPU_SINGLE_STEP &&
-                 (ctx->base.pc_next <= 0x100 || ctx->base.pc_next > 0xF00) &&
-                 ctx->base.is_jmp != DISAS_NORETURN)) {
-        uint32_t excp = gen_prep_dbgex(ctx);
-        gen_exception_nip(ctx, excp, ctx->base.pc_next);
-    }
-
     if (tcg_check_temp_count()) {
         qemu_log("Opcode %02x %02x %02x %02x (%08x) leaked "
                  "temporaries\n", opc1(ctx->opcode), opc2(ctx->opcode),
@@ -9107,6 +9102,7 @@ static void ppc_tr_tb_stop(DisasContextBase *dcbase, CPUState *cs)
     DisasContext *ctx = container_of(dcbase, DisasContext, base);
     DisasJumpType is_jmp = ctx->base.is_jmp;
     target_ulong nip = ctx->base.pc_next;
+    int sse;
 
     if (is_jmp == DISAS_NORETURN) {
         /* We have already exited the TB. */
@@ -9114,7 +9110,8 @@ static void ppc_tr_tb_stop(DisasContextBase *dcbase, CPUState *cs)
     }
 
     /* Honor single stepping. */
-    if (unlikely(ctx->base.singlestep_enabled)) {
+    sse = ctx->singlestep_enabled & (CPU_SINGLE_STEP | GDBSTUB_SINGLE_STEP);
+    if (unlikely(sse)) {
         switch (is_jmp) {
         case DISAS_TOO_MANY:
         case DISAS_EXIT_UPDATE:
@@ -9127,8 +9124,16 @@ static void ppc_tr_tb_stop(DisasContextBase *dcbase, CPUState *cs)
         default:
             g_assert_not_reached();
         }
-        gen_debug_exception(ctx);
-        return;
+
+        if (sse & GDBSTUB_SINGLE_STEP) {
+            gen_debug_exception(ctx);
+            return;
+        }
+        /* else CPU_SINGLE_STEP... */
+        if (nip <= 0x100 || nip > 0xf00) {
+            gen_exception(ctx, gen_prep_dbgex(ctx));
+            return;
+        }
     }
 
     switch (is_jmp) {
-- 
2.31.1



  parent reply	other threads:[~2021-05-19 13:18 UTC|newest]

Thread overview: 52+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-05-19 12:51 [PULL 00/48] ppc-for-6.1 queue 20210519 David Gibson
2021-05-19 12:51 ` [PULL 01/48] hw/ppc/spapr.c: Extract MMU mode error reporting into a function David Gibson
2021-05-19 12:51 ` [PULL 02/48] hw/ppc/spapr.c: Make sure the host supports the selected MMU mode David Gibson
2021-05-19 12:51 ` [PULL 03/48] target/ppc: Fold gen_*_xer into their callers David Gibson
2021-05-19 12:51 ` [PULL 04/48] target/ppc: renamed SPR registration functions David Gibson
2021-05-19 12:51 ` [PULL 05/48] target/ppc: move SPR R/W callbacks to translate.c David Gibson
2021-05-19 12:51 ` [PULL 06/48] hw/ppc: moved hcalls that depend on softmmu David Gibson
2021-05-19 12:51 ` [PULL 07/48] target/ppc: moved function out of mmu-hash64 David Gibson
2021-05-19 12:51 ` [PULL 08/48] target/ppc: moved ppc_store_lpcr to misc_helper.c David Gibson
2021-05-19 12:51 ` [PULL 09/48] hw/ppc: moved has_spr to cpu.h David Gibson
2021-05-19 12:51 ` [PULL 10/48] target/ppc: turned SPR R/W callbacks not static David Gibson
2021-05-19 12:51 ` [PULL 11/48] target/ppc: isolated cpu init from translation logic David Gibson
2021-05-19 12:51 ` [PULL 12/48] target/ppc: created ppc_{store, get}_vscr for generic vscr usage David Gibson
2021-05-19 12:51 ` [PULL 13/48] target/ppc: updated vscr manipulation in machine.c David Gibson
2021-05-19 12:51 ` [PULL 14/48] target/ppc: moved ppc_store_sdr1 to cpu.c David Gibson
2021-05-19 12:51 ` [PULL 15/48] target/ppc: moved ppc_cpu_dump_state to cpu_init.c David Gibson
2021-05-19 12:51 ` [PULL 16/48] target/ppc: Add cia field to DisasContext David Gibson
2021-05-19 12:51 ` [PULL 17/48] target/ppc: Split out decode_legacy David Gibson
2021-05-19 12:51 ` [PULL 18/48] target/ppc: Move DISAS_NORETURN setting into gen_exception* David Gibson
2021-05-19 12:51 ` [PULL 19/48] target/ppc: Remove special case for POWERPC_SYSCALL David Gibson
2021-05-19 12:51 ` [PULL 20/48] target/ppc: Remove special case for POWERPC_EXCP_TRAP David Gibson
2021-05-19 12:51 ` [PULL 21/48] target/ppc: Simplify gen_debug_exception David Gibson
2021-05-19 12:51 ` [PULL 22/48] target/ppc: Introduce DISAS_{EXIT,CHAIN}{,_UPDATE} David Gibson
2021-05-19 12:51 ` [PULL 23/48] target/ppc: Replace POWERPC_EXCP_SYNC with DISAS_EXIT David Gibson
2021-05-19 12:51 ` [PULL 24/48] target/ppc: Remove unnecessary gen_io_end calls David Gibson
2021-05-19 12:51 ` [PULL 25/48] target/ppc: Introduce gen_icount_io_start David Gibson
2021-05-19 12:51 ` [PULL 26/48] target/ppc: Replace POWERPC_EXCP_STOP with DISAS_EXIT_UPDATE David Gibson
2021-05-19 12:51 ` [PULL 27/48] target/ppc: Replace POWERPC_EXCP_BRANCH with DISAS_NORETURN David Gibson
2021-05-19 12:51 ` [PULL 28/48] target/ppc: Remove DisasContext.exception David Gibson
2021-05-19 12:51 ` David Gibson [this message]
2021-05-19 12:51 ` [PULL 30/48] target/ppc: Tidy exception vs exit_tb David Gibson
2021-05-19 12:51 ` [PULL 31/48] target/ppc: Mark helper_raise_exception* as noreturn David Gibson
2021-05-19 12:51 ` [PULL 32/48] target/ppc: Use translator_loop_temp_check David Gibson
2021-05-19 12:51 ` [PULL 33/48] target/ppc: Fix load endianness for lxvwsx/lxvdsx David Gibson
2021-05-19 12:51 ` [PULL 34/48] target/ppc: Introduce prot_for_access_type David Gibson
2021-05-19 12:51 ` [PULL 35/48] target/ppc: Use MMUAccessType in mmu-radix64.c David Gibson
2021-05-19 12:51 ` [PULL 36/48] target/ppc: Use MMUAccessType in mmu-hash64.c David Gibson
2021-05-19 12:51 ` [PULL 37/48] target/ppc: Use MMUAccessType in mmu-hash32.c David Gibson
2021-05-19 12:51 ` [PULL 38/48] target/ppc: Rename access_type to type in mmu_helper.c David Gibson
2021-05-19 12:51 ` [PULL 39/48] target/ppc: Use MMUAccessType " David Gibson
2021-05-19 12:51 ` [PULL 40/48] target/ppc: Remove type argument from check_prot David Gibson
2021-05-19 12:51 ` [PULL 41/48] target/ppc: Remove type argument from ppc6xx_tlb_pte_check David Gibson
2021-05-19 12:51 ` [PULL 42/48] target/ppc: Remove type argument from ppc6xx_tlb_check David Gibson
2021-05-19 12:51 ` [PULL 43/48] target/ppc: Remove type argument from get_bat_6xx_tlb David Gibson
2021-05-19 12:51 ` [PULL 44/48] target/ppc: Remove type argument from mmu40x_get_physical_address David Gibson
2021-05-19 12:51 ` [PULL 45/48] target/ppc: Remove type argument from mmubooke_check_tlb David Gibson
2021-05-19 12:51 ` [PULL 46/48] target/ppc: Remove type argument from mmubooke_get_physical_address David Gibson
2021-05-19 12:51 ` [PULL 47/48] target/ppc: Remove type argument from mmubooke206_check_tlb David Gibson
2021-05-19 12:51 ` [PULL 48/48] target/ppc: Remove type argument for mmubooke206_get_physical_address David Gibson
2021-05-19 13:46 ` [PULL 00/48] ppc-for-6.1 queue 20210519 no-reply
2021-05-20  0:42   ` David Gibson
2021-05-20  9:00 ` 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=20210519125148.27720-30-david@gibson.dropbear.id.au \
    --to=david@gibson.dropbear.id.au \
    --cc=groug@kaod.org \
    --cc=matheus.ferst@eldorado.org.br \
    --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).