From: Claudio Imbrenda <imbrenda@linux.ibm.com>
To: linux-kernel@vger.kernel.org
Cc: kvm@vger.kernel.org, linux-s390@vger.kernel.org,
borntraeger@de.ibm.com, frankja@linux.ibm.com, david@kernel.org,
seiden@linux.ibm.com, nrb@linux.ibm.com,
schlameuss@linux.ibm.com, gra@linux.ibm.com
Subject: [PATCH v2 6/6] KVM: s390: Fix potential races in dat skey functions
Date: Fri, 14 Aug 2026 18:33:55 +0200 [thread overview]
Message-ID: <20260814163355.227191-7-imbrenda@linux.ibm.com> (raw)
In-Reply-To: <20260814163355.227191-1-imbrenda@linux.ibm.com>
When dat_cond_set_storage_key() finds a large page, it will
conditionally set the storage key in absolute memory using
large_crste_to_phys() to get the absolute address.
There is a race window between dat_entry_walk() and
large_crste_to_phys(): the large page could have been split
concurrently, and large_crste_to_phys() might be called with a crste
that does not designate a large page, leading to crashes.
Similar issues were also present in dat_set_storage_key().
dat_get_storage_key() and dat_reset_reference_bit() did instead check
for a potential concurrent splitting of the large page, but then
handled it incorrectly.
Fix by performing a READ_ONCE on the crste pointer, checking and using
the result, instead of dereferencing the pointer again. In case a race
is detacted, try dat_entry_walk() again.
Fixes: 8e03e8316eb2 ("KVM: s390: KVM page table management functions: storage keys")
Signed-off-by: Claudio Imbrenda <imbrenda@linux.ibm.com>
---
arch/s390/kvm/dat.c | 36 ++++++++++++++++++++++++++++--------
1 file changed, 28 insertions(+), 8 deletions(-)
diff --git a/arch/s390/kvm/dat.c b/arch/s390/kvm/dat.c
index 7e5dd5a1eb1e..b4c318ebd91e 100644
--- a/arch/s390/kvm/dat.c
+++ b/arch/s390/kvm/dat.c
@@ -620,17 +620,20 @@ int dat_get_storage_key(union asce asce, gfn_t gfn, union skey *skey)
union pte *ptep;
int rc;
+again:
skey->skey = 0;
rc = dat_entry_walk(NULL, gfn, asce, DAT_WALK_ANY, TABLE_TYPE_PAGE_TABLE, &crstep, &ptep);
if (rc)
return rc;
if (!ptep) {
- union crste crste;
+ union crste crste = READ_ONCE(*crstep);
- crste = READ_ONCE(*crstep);
- if (!crste.h.fc || !crste.s.fc1.pr)
+ if (!crste_leaf(crste))
+ goto again;
+ if (!crste.s.fc1.pr)
return 0;
+
skey->skey = page_get_storage_key(large_crste_to_phys(crste, gfn));
return 0;
}
@@ -661,13 +664,20 @@ int dat_set_storage_key(struct kvm_s390_mmu_cache *mc, union asce asce, gfn_t gf
union pte *ptep;
int rc;
+again:
rc = dat_entry_walk(mc, gfn, asce, DAT_WALK_LEAF_ALLOC, TABLE_TYPE_PAGE_TABLE,
&crstep, &ptep);
if (rc)
return rc;
if (!ptep) {
- page_set_storage_key(large_crste_to_phys(*crstep, gfn), skey.skey, !nq);
+ union crste crste = READ_ONCE(*crstep);
+
+ /* A large page has been split concurrently, try again */
+ if (!crste_leaf(crste))
+ goto again;
+
+ page_set_storage_key(large_crste_to_phys(crste, gfn), skey.skey, !nq);
return 0;
}
@@ -717,15 +727,22 @@ int dat_cond_set_storage_key(struct kvm_s390_mmu_cache *mmc, union asce asce, gf
union pte *ptep;
int rc;
+again:
rc = dat_entry_walk(mmc, gfn, asce, DAT_WALK_LEAF_ALLOC, TABLE_TYPE_PAGE_TABLE,
&crstep, &ptep);
if (rc)
return rc;
if (!ptep) {
+ union crste crste = READ_ONCE(*crstep);
+
+ /* A large page has been split concurrently, try again */
+ if (!crste_leaf(crste))
+ goto again;
if (!oldkey)
oldkey = &prev;
- return page_cond_set_storage_key(large_crste_to_phys(*crstep, gfn), skey, oldkey,
+
+ return page_cond_set_storage_key(large_crste_to_phys(crste, gfn), skey, oldkey,
nq, mr, mc);
}
@@ -767,7 +784,7 @@ int dat_reset_reference_bit(union asce asce, gfn_t gfn, union skey *skey)
int rc;
skey->skey = 0;
-
+again:
rc = dat_entry_walk(NULL, gfn, asce, DAT_WALK_ANY, TABLE_TYPE_PAGE_TABLE, &crstep, &ptep);
if (rc)
return rc;
@@ -775,9 +792,12 @@ int dat_reset_reference_bit(union asce asce, gfn_t gfn, union skey *skey)
if (!ptep) {
union crste crste = READ_ONCE(*crstep);
- if (!crste.h.fc || !crste.s.fc1.pr)
+ /* A large page has been split concurrently, try again */
+ if (!crste_leaf(crste))
+ goto again;
+ if (!crste.s.fc1.pr)
return 0;
- skey->skey = page_reset_referenced(large_crste_to_phys(*crstep, gfn)) << 1;
+ skey->skey = page_reset_referenced(large_crste_to_phys(crste, gfn)) << 1;
return 0;
}
old = pgste_get_lock(ptep);
--
2.55.0
next prev parent reply other threads:[~2026-08-14 16:34 UTC|newest]
Thread overview: 13+ messages / expand[flat|nested] mbox.gz Atom feed top
2026-08-14 16:33 [PATCH v2 0/6] KVM: s390: Even more misc fixes Claudio Imbrenda
2026-08-14 16:33 ` [PATCH v2 1/6] KVM: s390: Fix dirty marking in adapter_indicators_set*() Claudio Imbrenda
2026-08-14 16:42 ` sashiko-bot
2026-08-14 16:33 ` [PATCH v2 2/6] KVM: s390: Fix _gaccess_shadow_fault() Claudio Imbrenda
2026-08-14 16:59 ` sashiko-bot
2026-08-14 16:33 ` [PATCH v2 3/6] KVM: s390: Refactor dat_set_slot() Claudio Imbrenda
2026-08-14 16:46 ` sashiko-bot
2026-08-14 16:33 ` [PATCH v2 4/6] KVM: s390: Move all code into kvm_arch_prepare_memory_region() Claudio Imbrenda
2026-08-14 16:46 ` sashiko-bot
2026-08-14 16:33 ` [PATCH v2 5/6] KVM: s390: Add missing srcu in kvm_s390_set_irq_state() Claudio Imbrenda
2026-08-14 16:43 ` sashiko-bot
2026-08-14 16:33 ` Claudio Imbrenda [this message]
2026-08-14 16:46 ` [PATCH v2 6/6] KVM: s390: Fix potential races in dat skey functions sashiko-bot
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=20260814163355.227191-7-imbrenda@linux.ibm.com \
--to=imbrenda@linux.ibm.com \
--cc=borntraeger@de.ibm.com \
--cc=david@kernel.org \
--cc=frankja@linux.ibm.com \
--cc=gra@linux.ibm.com \
--cc=kvm@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-s390@vger.kernel.org \
--cc=nrb@linux.ibm.com \
--cc=schlameuss@linux.ibm.com \
--cc=seiden@linux.ibm.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