qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Richard Henderson <richard.henderson@linaro.org>
To: qemu-devel@nongnu.org
Cc: qemu-arm@nongnu.org
Subject: [PATCH 02/71] linux-user/aarch64: Introduce sve_vq_cached
Date: Thu,  2 Jun 2022 14:47:44 -0700	[thread overview]
Message-ID: <20220602214853.496211-3-richard.henderson@linaro.org> (raw)
In-Reply-To: <20220602214853.496211-1-richard.henderson@linaro.org>

Add an interface function to extract the digested vector length
rather than the raw zcr_el[1] value.  This fixes an incorrect
return from do_prctl_set_vl where we didn't take into account
the set of vector lengths supported by the cpu.

Signed-off-by: Richard Henderson <richard.henderson@linaro.org>
---
v2: Add sve_vq_cached rather than directly access hflags.
---
 linux-user/aarch64/target_prctl.h | 20 +++++++++++++-------
 target/arm/cpu.h                  | 11 +++++++++++
 linux-user/aarch64/signal.c       |  4 ++--
 3 files changed, 26 insertions(+), 9 deletions(-)

diff --git a/linux-user/aarch64/target_prctl.h b/linux-user/aarch64/target_prctl.h
index 3f5a5d3933..fdd973e07d 100644
--- a/linux-user/aarch64/target_prctl.h
+++ b/linux-user/aarch64/target_prctl.h
@@ -10,7 +10,7 @@ static abi_long do_prctl_get_vl(CPUArchState *env)
 {
     ARMCPU *cpu = env_archcpu(env);
     if (cpu_isar_feature(aa64_sve, cpu)) {
-        return ((cpu->env.vfp.zcr_el[1] & 0xf) + 1) * 16;
+        return sve_vq_cached(env) * 16;
     }
     return -TARGET_EINVAL;
 }
@@ -25,18 +25,24 @@ static abi_long do_prctl_set_vl(CPUArchState *env, abi_long arg2)
      */
     if (cpu_isar_feature(aa64_sve, env_archcpu(env))
         && arg2 >= 0 && arg2 <= 512 * 16 && !(arg2 & 15)) {
-        ARMCPU *cpu = env_archcpu(env);
         uint32_t vq, old_vq;
 
-        old_vq = (env->vfp.zcr_el[1] & 0xf) + 1;
-        vq = MAX(arg2 / 16, 1);
-        vq = MIN(vq, cpu->sve_max_vq);
+        old_vq = sve_vq_cached(env);
 
+        /*
+         * Bound the value of arg2, so that we know that it fits into
+         * the 4-bit field in ZCR_EL1.  Rely on the hflags rebuild to
+         * sort out the length supported by the cpu.
+         */
+        vq = MAX(arg2 / 16, 1);
+        vq = MIN(vq, ARM_MAX_VQ);
+        env->vfp.zcr_el[1] = vq - 1;
+        arm_rebuild_hflags(env);
+
+        vq = sve_vq_cached(env);
         if (vq < old_vq) {
             aarch64_sve_narrow_vq(env, vq);
         }
-        env->vfp.zcr_el[1] = vq - 1;
-        arm_rebuild_hflags(env);
         return vq * 16;
     }
     return -TARGET_EINVAL;
diff --git a/target/arm/cpu.h b/target/arm/cpu.h
index 015ce12fe2..830d358d46 100644
--- a/target/arm/cpu.h
+++ b/target/arm/cpu.h
@@ -3286,6 +3286,17 @@ static inline int cpu_mmu_index(CPUARMState *env, bool ifetch)
     return EX_TBFLAG_ANY(env->hflags, MMUIDX);
 }
 
+/**
+ * sve_vq_cached
+ * @env: the cpu context
+ *
+ * Return the VL cached within env->hflags, in units of quadwords.
+ */
+static inline int sve_vq_cached(CPUARMState *env)
+{
+    return EX_TBFLAG_A64(env->hflags, VL) + 1;
+}
+
 static inline bool bswap_code(bool sctlr_b)
 {
 #ifdef CONFIG_USER_ONLY
diff --git a/linux-user/aarch64/signal.c b/linux-user/aarch64/signal.c
index 7de4c96eb9..30e89f67c8 100644
--- a/linux-user/aarch64/signal.c
+++ b/linux-user/aarch64/signal.c
@@ -315,7 +315,7 @@ static int target_restore_sigframe(CPUARMState *env,
 
         case TARGET_SVE_MAGIC:
             if (cpu_isar_feature(aa64_sve, env_archcpu(env))) {
-                vq = (env->vfp.zcr_el[1] & 0xf) + 1;
+                vq = sve_vq_cached(env);
                 sve_size = QEMU_ALIGN_UP(TARGET_SVE_SIG_CONTEXT_SIZE(vq), 16);
                 if (!sve && size == sve_size) {
                     sve = (struct target_sve_context *)ctx;
@@ -434,7 +434,7 @@ static void target_setup_frame(int usig, struct target_sigaction *ka,
 
     /* SVE state needs saving only if it exists.  */
     if (cpu_isar_feature(aa64_sve, env_archcpu(env))) {
-        vq = (env->vfp.zcr_el[1] & 0xf) + 1;
+        vq = sve_vq_cached(env);
         sve_size = QEMU_ALIGN_UP(TARGET_SVE_SIG_CONTEXT_SIZE(vq), 16);
         sve_ofs = alloc_sigframe_space(sve_size, &layout);
     }
-- 
2.34.1



  parent reply	other threads:[~2022-06-02 21:52 UTC|newest]

Thread overview: 109+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-06-02 21:47 [PATCH 00/71] target/arm: Scalable Matrix Extension Richard Henderson
2022-06-02 21:47 ` [PATCH 01/71] target/arm: Rename TBFLAG_A64 ZCR_LEN to VL Richard Henderson
2022-06-02 21:47 ` Richard Henderson [this message]
2022-06-06 10:31   ` [PATCH 02/71] linux-user/aarch64: Introduce sve_vq_cached Peter Maydell
2022-06-02 21:47 ` [PATCH 03/71] target/arm: Remove route_to_el2 check from sve_exception_el Richard Henderson
2022-06-06 12:13   ` Peter Maydell
2022-06-02 21:47 ` [PATCH 04/71] target/arm: Remove fp checks " Richard Henderson
2022-06-06 12:23   ` Peter Maydell
2022-06-02 21:47 ` [PATCH 05/71] target/arm: Add el_is_in_host Richard Henderson
2022-06-02 21:47 ` [PATCH 06/71] target/arm: Use el_is_in_host for sve_zcr_len_for_el Richard Henderson
2022-06-02 21:47 ` [PATCH 07/71] target/arm: Use el_is_in_host for sve_exception_el Richard Henderson
2022-06-06 12:24   ` Peter Maydell
2022-06-02 21:47 ` [PATCH 08/71] target/arm: Hoist arm_is_el2_enabled check in sve_exception_el Richard Henderson
2022-06-06 12:27   ` Peter Maydell
2022-06-02 21:47 ` [PATCH 09/71] target/arm: Do not use aarch64_sve_zcr_get_valid_len in reset Richard Henderson
2022-06-06 12:27   ` Peter Maydell
2022-06-02 21:47 ` [PATCH 10/71] target/arm: Merge aarch64_sve_zcr_get_valid_len into caller Richard Henderson
2022-06-06 12:30   ` Peter Maydell
2022-06-02 21:47 ` [PATCH 11/71] target/arm: Use uint32_t instead of bitmap for sve vq's Richard Henderson
2022-06-02 21:47 ` [PATCH 12/71] target/arm: Rename sve_zcr_len_for_el to sve_vqm1_for_el Richard Henderson
2022-06-02 21:47 ` [PATCH 13/71] target/arm: Split out load/store primitives to sve_ldst_internal.h Richard Henderson
2022-06-02 21:47 ` [PATCH 14/71] target/arm: Export sve contiguous ldst support functions Richard Henderson
2022-06-02 21:47 ` [PATCH 15/71] target/arm: Move expand_pred_b to vec_internal.h Richard Henderson
2022-06-02 21:47 ` [PATCH 16/71] target/arm: Use expand_pred_b in mve_helper.c Richard Henderson
2022-06-02 21:47 ` [PATCH 17/71] target/arm: Move expand_pred_h to vec_internal.h Richard Henderson
2022-06-02 21:48 ` [PATCH 18/71] target/arm: Export bfdotadd from vec_helper.c Richard Henderson
2022-06-02 21:48 ` [PATCH 19/71] target/arm: Add isar_feature_aa64_sme Richard Henderson
2022-06-06 12:31   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 20/71] target/arm: Add ID_AA64SMFR0_EL1 Richard Henderson
2022-06-06 13:05   ` Peter Maydell
2022-06-06 16:19     ` Richard Henderson
2022-06-02 21:48 ` [PATCH 21/71] target/arm: Implement TPIDR2_EL0 Richard Henderson
2022-06-06 13:18   ` Peter Maydell
2022-06-06 14:38     ` Richard Henderson
2022-06-02 21:48 ` [PATCH 22/71] target/arm: Add SMEEXC_EL to TB flags Richard Henderson
2022-06-06 13:25   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 23/71] target/arm: Add syn_smetrap Richard Henderson
2022-06-06 13:28   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 24/71] target/arm: Add ARM_CP_SME Richard Henderson
2022-06-06 13:32   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 25/71] target/arm: Add SVCR Richard Henderson
2022-06-06 13:40   ` Peter Maydell
2022-06-06 14:41     ` Richard Henderson
2022-06-02 21:48 ` [PATCH 26/71] target/arm: Add SMCR_ELx Richard Henderson
2022-06-06 13:42   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 27/71] target/arm: Add SMIDR_EL1, SMPRI_EL1, SMPRIMAP_EL2 Richard Henderson
2022-06-06 15:55   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 28/71] target/arm: Add PSTATE.{SM,ZA} to TB flags Richard Henderson
2022-06-06 15:58   ` Peter Maydell
2022-06-06 16:50     ` Richard Henderson
2022-06-02 21:48 ` [PATCH 29/71] target/arm: Add the SME ZA storage to CPUARMState Richard Henderson
2022-06-06 16:13   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 30/71] target/arm: Implement SMSTART, SMSTOP Richard Henderson
2022-06-06 16:50   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 31/71] target/arm: Move error for sve%d property to arm_cpu_sve_finalize Richard Henderson
2022-06-07  8:44   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 32/71] target/arm: Create ARMVQMap Richard Henderson
2022-06-07  8:45   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 33/71] target/arm: Generalize cpu_arm_{get,set}_vq Richard Henderson
2022-06-07  8:48   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 34/71] target/arm: Generalize cpu_arm_{get, set}_default_vec_len Richard Henderson
2022-06-07  8:49   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 35/71] target/arm: Move arm_cpu_*_finalize to internals.h Richard Henderson
2022-06-07  8:50   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 36/71] target/arm: Unexport aarch64_add_*_properties Richard Henderson
2022-06-07  8:51   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 37/71] target/arm: Add cpu properties for SME Richard Henderson
2022-06-07  9:47   ` Peter Maydell
2022-06-07 14:45     ` Richard Henderson
2022-06-02 21:48 ` [PATCH 38/71] target/arm: Introduce sve_vqm1_for_el_sm Richard Henderson
2022-06-07  9:54   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 39/71] target/arm: Add SVL to TB flags Richard Henderson
2022-06-07  9:58   ` Peter Maydell
2022-06-07 14:49     ` Richard Henderson
2022-06-02 21:48 ` [PATCH 40/71] target/arm: Move pred_{full, gvec}_reg_{offset, size} to translate-a64.h Richard Henderson
2022-06-07  9:58   ` Peter Maydell
2022-06-02 21:48 ` [PATCH 41/71] target/arm: Add infrastructure for disas_sme Richard Henderson
2022-06-07 10:03   ` Peter Maydell
2022-06-07 14:52     ` Richard Henderson
2022-06-02 21:48 ` [PATCH 42/71] target/arm: Trap AdvSIMD usage when Streaming SVE is active Richard Henderson
2022-06-02 21:48 ` [PATCH 43/71] target/arm: Implement SME RDSVL, ADDSVL, ADDSPL Richard Henderson
2022-06-02 21:48 ` [PATCH 44/71] target/arm: Implement SME ZERO Richard Henderson
2022-06-02 21:48 ` [PATCH 45/71] target/arm: Implement SME MOVA Richard Henderson
2022-06-02 21:48 ` [PATCH 46/71] target/arm: Implement SME LD1, ST1 Richard Henderson
2022-06-02 21:48 ` [PATCH 47/71] target/arm: Export unpredicated ld/st from translate-sve.c Richard Henderson
2022-06-02 21:48 ` [PATCH 48/71] target/arm: Implement SME LDR, STR Richard Henderson
2022-06-02 21:48 ` [PATCH 49/71] target/arm: Implement SME ADDHA, ADDVA Richard Henderson
2022-06-02 21:48 ` [PATCH 50/71] target/arm: Implement FMOPA, FMOPS (non-widening) Richard Henderson
2022-06-02 21:48 ` [PATCH 51/71] target/arm: Implement BFMOPA, BFMOPS Richard Henderson
2022-06-02 21:48 ` [PATCH 52/71] target/arm: Implement FMOPA, FMOPS (widening) Richard Henderson
2022-06-02 21:48 ` [PATCH 53/71] target/arm: Implement SME integer outer product Richard Henderson
2022-06-02 21:48 ` [PATCH 54/71] target/arm: Implement PSEL Richard Henderson
2022-06-02 21:48 ` [PATCH 55/71] target/arm: Implement REVD Richard Henderson
2022-06-02 21:48 ` [PATCH 56/71] target/arm: Implement SCLAMP, UCLAMP Richard Henderson
2022-06-02 21:48 ` [PATCH 57/71] target/arm: Reset streaming sve state on exception boundaries Richard Henderson
2022-06-02 21:48 ` [PATCH 58/71] target/arm: Enable SME for -cpu max Richard Henderson
2022-06-02 21:48 ` [PATCH 59/71] linux-user/aarch64: Clear tpidr2_el0 if CLONE_SETTLS Richard Henderson
2022-06-02 21:48 ` [PATCH 60/71] linux-user/aarch64: Reset PSTATE.SM on syscalls Richard Henderson
2022-06-02 21:48 ` [PATCH 61/71] linux-user/aarch64: Add SM bit to SVE signal context Richard Henderson
2022-06-02 21:48 ` [PATCH 62/71] linux-user/aarch64: Tidy target_restore_sigframe error return Richard Henderson
2022-06-02 21:48 ` [PATCH 63/71] linux-user/aarch64: Do not allow duplicate or short sve records Richard Henderson
2022-06-02 21:48 ` [PATCH 64/71] linux-user/aarch64: Verify extra record lock succeeded Richard Henderson
2022-06-02 21:48 ` [PATCH 65/71] linux-user/aarch64: Move sve record checks into restore Richard Henderson
2022-06-02 21:48 ` [PATCH 66/71] linux-user/aarch64: Implement SME signal handling Richard Henderson
2022-06-02 21:48 ` [PATCH 67/71] linux-user: Rename sve prctls Richard Henderson
2022-06-02 21:48 ` [PATCH 68/71] linux-user/aarch64: Implement PR_SME_GET_VL, PR_SME_SET_VL Richard Henderson
2022-06-02 21:48 ` [PATCH 69/71] target/arm: Only set ZEN in reset if SVE present Richard Henderson
2022-06-02 21:48 ` [PATCH 70/71] target/arm: Enable SME for user-only Richard Henderson
2022-06-02 21:48 ` [PATCH 71/71] linux-user/aarch64: Add SME related hwcap entries 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=20220602214853.496211-3-richard.henderson@linaro.org \
    --to=richard.henderson@linaro.org \
    --cc=qemu-arm@nongnu.org \
    --cc=qemu-devel@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).