qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: Tsuneo Saito <tsnsaito@gmail.com>
To: qemu-devel@nongnu.org
Cc: Tsuneo Saito <tsnsaito@gmail.com>
Subject: [Qemu-devel] [PATCH 6/7] SPARC64: implement MMU miss traps on nonfaulting loads
Date: Fri, 22 Jul 2011 00:16:32 +0900	[thread overview]
Message-ID: <1311261393-47400-7-git-send-email-tsnsaito@gmail.com> (raw)
In-Reply-To: <1311261393-47400-1-git-send-email-tsnsaito@gmail.com>

Nonfaulting loads should raise fast_data_access_MMU_miss traps as
normal loads do.  It is up to the guest OS kernel that detect MMU misses
on nonfaulting load instructions and make them complete without signaling.

Signed-off-by: Tsuneo Saito <tsnsaito@gmail.com>
---
 target-sparc/op_helper.c |   36 ++++++++++++++++++++----------------
 1 files changed, 20 insertions(+), 16 deletions(-)

diff --git a/target-sparc/op_helper.c b/target-sparc/op_helper.c
index 3b7f9ca..8962e38 100644
--- a/target-sparc/op_helper.c
+++ b/target-sparc/op_helper.c
@@ -2567,24 +2567,30 @@ uint64_t helper_ld_asi(target_ulong addr, int asi, int size, int sign)
     helper_check_align(addr, size - 1);
     addr = asi_address_mask(env, asi, addr);
 
-    switch (asi) {
-    case 0x82: // Primary no-fault
-    case 0x8a: // Primary no-fault LE
-    case 0x83: // Secondary no-fault
-    case 0x8b: // Secondary no-fault LE
-        {
-            /* secondary space access has lowest asi bit equal to 1 */
-            int access_mmu_idx = ( asi & 1 ) ? MMU_KERNEL_IDX
-                                             : MMU_KERNEL_SECONDARY_IDX;
+    /* process nonfaulting loads first */
+    if ((asi & 0xf6) == 0x82) {
+        int mmu_idx;
+
+        /* secondary space access has lowest asi bit equal to 1 */
+        if (env->pstate & PS_PRIV) {
+            mmu_idx = (asi & 1) ? MMU_KERNEL_SECONDARY_IDX : MMU_KERNEL_IDX;
+        } else {
+            mmu_idx = (asi & 1) ? MMU_USER_SECONDARY_IDX : MMU_USER_IDX;
+        }
 
-            if (cpu_get_phys_page_nofault(env, addr, access_mmu_idx) == -1ULL) {
+        if (cpu_get_phys_page_nofault(env, addr, mmu_idx) == -1ULL) {
 #ifdef DEBUG_ASI
-                dump_asi("read ", last_addr, asi, size, ret);
+            dump_asi("read ", last_addr, asi, size, ret);
 #endif
-                return 0;
-            }
+            /* env->exception_index is set in get_physical_address_data(). */
+            raise_exception(env->exception_index);
         }
-        // Fall through
+
+        /* convert nonfaulting load ASIs to normal load ASIs */
+        asi &= ~0x02;
+    }
+
+    switch (asi) {
     case 0x10: // As if user primary
     case 0x11: // As if user secondary
     case 0x18: // As if user primary LE
@@ -2862,8 +2868,6 @@ uint64_t helper_ld_asi(target_ulong addr, int asi, int size, int sign)
     case 0x1d: // Bypass, non-cacheable LE
     case 0x88: // Primary LE
     case 0x89: // Secondary LE
-    case 0x8a: // Primary no-fault LE
-    case 0x8b: // Secondary no-fault LE
         switch(size) {
         case 2:
             ret = bswap16(ret);
-- 
1.7.5.4

  parent reply	other threads:[~2011-07-21 15:19 UTC|newest]

Thread overview: 9+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2011-07-21 15:16 [Qemu-devel] [PATCH 0/7] SPARC64: fix nonfaulting load on softmmu Tsuneo Saito
2011-07-21 15:16 ` [Qemu-devel] [PATCH 1/7] SPARC64: TTE bits cleanup Tsuneo Saito
2011-07-21 15:16 ` [Qemu-devel] [PATCH 2/7] SPARC64: SFSR cleanup and fix Tsuneo Saito
2011-07-21 15:16 ` [Qemu-devel] [PATCH 3/7] SPARC64: introduce a convenience function for getting physical addresses Tsuneo Saito
2011-07-21 15:16 ` [Qemu-devel] [PATCH 4/7] SPARC64: split cpu_get_phys_page_debug() from cpu_get_phys_page_nofault() Tsuneo Saito
2011-07-21 15:16 ` [Qemu-devel] [PATCH 5/7] SPARC64: fix fault status overwritten on nonfaulting load Tsuneo Saito
2011-07-21 15:16 ` Tsuneo Saito [this message]
2011-07-21 15:16 ` [Qemu-devel] [PATCH 7/7] SPARC64: implement addtional MMU faults related to " Tsuneo Saito
2011-07-21 20:15 ` [Qemu-devel] [PATCH 0/7] SPARC64: fix nonfaulting load on softmmu Blue Swirl

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=1311261393-47400-7-git-send-email-tsnsaito@gmail.com \
    --to=tsnsaito@gmail.com \
    --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).