qemu-devel.nongnu.org archive mirror
 help / color / mirror / Atom feed
From: David Hildenbrand <david@redhat.com>
To: qemu-devel@nongnu.org
Cc: Thomas Huth <thuth@redhat.com>,
	Janosch Frank <frankja@linux.ibm.com>,
	David Hildenbrand <david@redhat.com>,
	Cornelia Huck <cohuck@redhat.com>,
	Halil Pasic <pasic@linux.ibm.com>,
	Christian Borntraeger <borntraeger@de.ibm.com>,
	qemu-s390x@nongnu.org, Richard Henderson <rth@twiddle.net>
Subject: [PATCH v2 4/7] s390x/mmu: Inject PGM_ADDRESSING on boguous table addresses
Date: Wed, 25 Sep 2019 14:52:33 +0200	[thread overview]
Message-ID: <20190925125236.4043-5-david@redhat.com> (raw)
In-Reply-To: <20190925125236.4043-1-david@redhat.com>

Let's document how it works and inject PGM_ADDRESSING if reading of
table entries fails.

Signed-off-by: David Hildenbrand <david@redhat.com>
---
 target/s390x/mmu_helper.c | 28 ++++++++++++++++++++++++----
 1 file changed, 24 insertions(+), 4 deletions(-)

diff --git a/target/s390x/mmu_helper.c b/target/s390x/mmu_helper.c
index f6ae444655..c9fde78614 100644
--- a/target/s390x/mmu_helper.c
+++ b/target/s390x/mmu_helper.c
@@ -93,6 +93,24 @@ target_ulong mmu_real2abs(CPUS390XState *env, target_ulong raddr)
     return raddr;
 }
 
+static inline int read_table_entry(hwaddr gaddr, uint64_t *entry)
+{
+    /*
+     * According to the PoP, these table addresses are "unpredictably real
+     * or absolute". Also, "it is unpredictable whether the address wraps
+     * or an addressing exception is recognized".
+     *
+     * We treat them as absolute addresses and don't wrap them.
+     */
+    if (unlikely(address_space_read(&address_space_memory, gaddr,
+                 MEMTXATTRS_UNSPECIFIED, (uint8_t *)entry, sizeof(*entry)) !=
+                 MEMTX_OK)) {
+        return -EFAULT;
+    }
+    *entry = be64_to_cpu(*entry);
+    return 0;
+}
+
 /* Decode page table entry (normal 4KB page) */
 static int mmu_translate_pte(CPUS390XState *env, target_ulong vaddr,
                              uint64_t asc, uint64_t pt_entry,
@@ -118,7 +136,6 @@ static int mmu_translate_segment(CPUS390XState *env, target_ulong vaddr,
                                  target_ulong *raddr, int *flags, int rw,
                                  bool exc)
 {
-    CPUState *cs = env_cpu(env);
     uint64_t origin, offs, pt_entry;
 
     if (st_entry & SEGMENT_ENTRY_RO) {
@@ -134,7 +151,9 @@ static int mmu_translate_segment(CPUS390XState *env, target_ulong vaddr,
     /* Look up 4KB page entry */
     origin = st_entry & SEGMENT_ENTRY_ORIGIN;
     offs  = (vaddr & VADDR_PX) >> 9;
-    pt_entry = ldq_phys(cs->as, origin + offs);
+    if (read_table_entry(origin + offs, &pt_entry)) {
+        return PGM_ADDRESSING;
+    }
     return mmu_translate_pte(env, vaddr, asc, pt_entry, raddr, flags, rw, exc);
 }
 
@@ -144,7 +163,6 @@ static int mmu_translate_region(CPUS390XState *env, target_ulong vaddr,
                                 target_ulong *raddr, int *flags, int rw,
                                 bool exc)
 {
-    CPUState *cs = env_cpu(env);
     uint64_t origin, offs, new_entry;
     const int pchks[4] = {
         PGM_SEGMENT_TRANS, PGM_REG_THIRD_TRANS,
@@ -154,7 +172,9 @@ static int mmu_translate_region(CPUS390XState *env, target_ulong vaddr,
     origin = entry & REGION_ENTRY_ORIGIN;
     offs = (vaddr >> (17 + 11 * level / 4)) & 0x3ff8;
 
-    new_entry = ldq_phys(cs->as, origin + offs);
+    if (read_table_entry(origin + offs, &new_entry)) {
+        return PGM_ADDRESSING;
+    }
 
     if ((new_entry & REGION_ENTRY_INV) != 0) {
         return pchks[level / 4];
-- 
2.21.0



  parent reply	other threads:[~2019-09-25 13:27 UTC|newest]

Thread overview: 24+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-09-25 12:52 [PATCH v2 0/7] s390x/mmu: DAT translation rewrite David Hildenbrand
2019-09-25 12:52 ` [PATCH v2 1/7] s390x/mmu: Drop debug logging from MMU code David Hildenbrand
2019-09-25 13:28   ` Thomas Huth
2019-09-25 19:11   ` Richard Henderson
2019-09-25 12:52 ` [PATCH v2 2/7] s390x/mmu: Move DAT protection handling out of mmu_translate_asce() David Hildenbrand
2019-09-25 17:01   ` Thomas Huth
2019-09-25 19:14   ` Richard Henderson
2019-09-25 12:52 ` [PATCH v2 3/7] s390x/mmu: Inject DAT exceptions from a single place David Hildenbrand
2019-09-25 17:05   ` Thomas Huth
2019-09-25 19:14   ` Richard Henderson
2019-09-25 12:52 ` David Hildenbrand [this message]
2019-09-25 17:12   ` [PATCH v2 4/7] s390x/mmu: Inject PGM_ADDRESSING on boguous table addresses Thomas Huth
2019-09-25 19:25   ` Richard Henderson
2019-09-25 19:36     ` David Hildenbrand
2019-09-25 12:52 ` [PATCH v2 5/7] s390x/mmu: Use TARGET_PAGE_MASK in mmu_translate_pte() David Hildenbrand
2019-09-25 17:15   ` Thomas Huth
2019-09-25 19:26   ` Richard Henderson
2019-09-25 12:52 ` [PATCH v2 6/7] s390x/mmu: DAT table definition overhaul David Hildenbrand
2019-09-26  7:35   ` Thomas Huth
2019-09-26  7:38     ` David Hildenbrand
2019-09-26  7:52       ` Thomas Huth
2019-09-26  7:59         ` David Hildenbrand
2019-09-26  8:07           ` Thomas Huth
2019-09-25 12:52 ` [PATCH v2 7/7] s390x/mmu: Convert to non-recursive page table walk David Hildenbrand

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=20190925125236.4043-5-david@redhat.com \
    --to=david@redhat.com \
    --cc=borntraeger@de.ibm.com \
    --cc=cohuck@redhat.com \
    --cc=frankja@linux.ibm.com \
    --cc=pasic@linux.ibm.com \
    --cc=qemu-devel@nongnu.org \
    --cc=qemu-s390x@nongnu.org \
    --cc=rth@twiddle.net \
    --cc=thuth@redhat.com \
    /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).