From: Paolo Bonzini <pbonzini@redhat.com>
To: qemu-devel@nongnu.org
Subject: [Qemu-devel] [PULL 16/33] target-i386: fix kernel accesses with SMAP and CPL = 3
Date: Thu, 5 Jun 2014 16:22:06 +0200 [thread overview]
Message-ID: <1401978143-11896-17-git-send-email-pbonzini@redhat.com> (raw)
In-Reply-To: <1401978143-11896-1-git-send-email-pbonzini@redhat.com>
With SMAP, implicit kernel accesses from user mode always behave as
if AC=0. To do this, kernel mode is not anymore a separate MMU mode.
Instead, KERNEL_IDX is renamed to KSMAP_IDX and the kernel mode accessors
wrap KSMAP_IDX and KNOSMAP_IDX.
Signed-off-by: Paolo Bonzini <pbonzini@redhat.com>
---
target-i386/cpu.h | 15 +++++++++++----
target-i386/helper.c | 8 ++++----
target-i386/seg_helper.c | 18 ++++++++++++++++++
3 files changed, 33 insertions(+), 8 deletions(-)
diff --git a/target-i386/cpu.h b/target-i386/cpu.h
index f88b675..b80df66 100644
--- a/target-i386/cpu.h
+++ b/target-i386/cpu.h
@@ -1153,17 +1153,24 @@ static inline CPUX86State *cpu_init(const char *cpu_model)
#define cpudef_setup x86_cpudef_setup
/* MMU modes definitions */
-#define MMU_MODE0_SUFFIX _kernel
+#define MMU_MODE0_SUFFIX _ksmap
#define MMU_MODE1_SUFFIX _user
#define MMU_MODE2_SUFFIX _knosmap /* SMAP disabled or CPL<3 && AC=1 */
-#define MMU_KERNEL_IDX 0
+#define MMU_KSMAP_IDX 0
#define MMU_USER_IDX 1
#define MMU_KNOSMAP_IDX 2
-static inline int cpu_mmu_index (CPUX86State *env)
+static inline int cpu_mmu_index(CPUX86State *env)
{
return (env->hflags & HF_CPL_MASK) == 3 ? MMU_USER_IDX :
((env->hflags & HF_SMAP_MASK) && (env->eflags & AC_MASK))
- ? MMU_KNOSMAP_IDX : MMU_KERNEL_IDX;
+ ? MMU_KNOSMAP_IDX : MMU_KSMAP_IDX;
+}
+
+static inline int cpu_mmu_index_kernel(CPUX86State *env)
+{
+ return !(env->hflags & HF_SMAP_MASK) ? MMU_KNOSMAP_IDX :
+ ((env->hflags & HF_CPL_MASK) < 3 && (env->eflags & AC_MASK))
+ ? MMU_KNOSMAP_IDX : MMU_KSMAP_IDX;
}
#define CC_DST (env->cc_dst)
diff --git a/target-i386/helper.c b/target-i386/helper.c
index 88bbe65..6d8e350 100644
--- a/target-i386/helper.c
+++ b/target-i386/helper.c
@@ -648,7 +648,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
}
break;
- case MMU_KERNEL_IDX:
+ case MMU_KSMAP_IDX:
if (is_write1 != 2 && (env->cr[4] & CR4_SMAP_MASK) &&
(ptep & PG_USER_MASK)) {
goto do_fault_protect;
@@ -710,7 +710,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
}
break;
- case MMU_KERNEL_IDX:
+ case MMU_KSMAP_IDX:
if (is_write1 != 2 && (env->cr[4] & CR4_SMAP_MASK) &&
(ptep & PG_USER_MASK)) {
goto do_fault_protect;
@@ -765,7 +765,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
}
break;
- case MMU_KERNEL_IDX:
+ case MMU_KSMAP_IDX:
if (is_write1 != 2 && (env->cr[4] & CR4_SMAP_MASK) &&
(pde & PG_USER_MASK)) {
goto do_fault_protect;
@@ -822,7 +822,7 @@ int x86_cpu_handle_mmu_fault(CPUState *cs, vaddr addr,
}
break;
- case MMU_KERNEL_IDX:
+ case MMU_KSMAP_IDX:
if (is_write1 != 2 && (env->cr[4] & CR4_SMAP_MASK) &&
(ptep & PG_USER_MASK)) {
goto do_fault_protect;
diff --git a/target-i386/seg_helper.c b/target-i386/seg_helper.c
index 31c5508..934cc2b 100644
--- a/target-i386/seg_helper.c
+++ b/target-i386/seg_helper.c
@@ -34,6 +34,24 @@
# define LOG_PCALL_STATE(cpu) do { } while (0)
#endif
+#ifndef CONFIG_USER_ONLY
+#define CPU_MMU_INDEX (cpu_mmu_index_kernel(env))
+#define MEMSUFFIX _kernel
+#define DATA_SIZE 1
+#include "exec/cpu_ldst_template.h"
+
+#define DATA_SIZE 2
+#include "exec/cpu_ldst_template.h"
+
+#define DATA_SIZE 4
+#include "exec/cpu_ldst_template.h"
+
+#define DATA_SIZE 8
+#include "exec/cpu_ldst_template.h"
+#undef CPU_MMU_INDEX
+#undef MEMSUFFIX
+#endif
+
/* return non zero if error */
static inline int load_segment(CPUX86State *env, uint32_t *e1_ptr,
uint32_t *e2_ptr, int selector)
--
1.8.3.1
next prev parent reply other threads:[~2014-06-05 14:23 UTC|newest]
Thread overview: 35+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-06-05 14:21 [Qemu-devel] [PULL 00/33] softmmu cleanups and target-i386 paging fixes Paolo Bonzini
2014-06-05 14:21 ` [Qemu-devel] [PULL 01/33] cputlb: Fix regression with TCG interpreter (bug 1310324) Paolo Bonzini
2014-06-05 14:21 ` [Qemu-devel] [PULL 02/33] nseries: clean up coding style Paolo Bonzini
2014-06-05 14:21 ` [Qemu-devel] [PULL 03/33] hw: use ld_p/st_p instead of ld_raw/st_raw Paolo Bonzini
2014-06-05 14:21 ` [Qemu-devel] [PULL 04/33] softmmu: start introducing SOFTMMU_CODE_ACCESS in softmmu_header.h Paolo Bonzini
2014-06-05 14:21 ` [Qemu-devel] [PULL 05/33] softmmu: move MMUSUFFIX under SOFTMMU_CODE_ACCESS Paolo Bonzini
2014-06-05 14:21 ` [Qemu-devel] [PULL 06/33] softmmu: move definition of CPU_MMU_INDEX to inclusion site, drop ACCESS_TYPE Paolo Bonzini
2014-06-05 14:21 ` [Qemu-devel] [PULL 07/33] softmmu: make do_unaligned_access a method of CPU Paolo Bonzini
2014-06-05 14:21 ` [Qemu-devel] [PULL 08/33] softmmu: move ALIGNED_ONLY to cpu.h Paolo Bonzini
2014-06-05 14:21 ` [Qemu-devel] [PULL 09/33] softmmu: commonize helper definitions Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 10/33] softmmu: move softmmu_template.h out of include/ Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 11/33] target-arm: move arm_*_code to a separate file Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 12/33] softmmu: introduce cpu_ldst.h Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 13/33] softmmu: move all load/store functions to cpu_ldst.h Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 14/33] target-i386: rename KSMAP to KNOSMAP Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 15/33] target-i386: move check_io helpers to seg_helper.c Paolo Bonzini
2014-06-05 14:22 ` Paolo Bonzini [this message]
2014-06-05 14:22 ` [Qemu-devel] [PULL 17/33] target-i386: simplify SMAP handling in MMU_KSMAP_IDX Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 18/33] target-i386: fix coding standards in x86_cpu_handle_mmu_fault Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 19/33] target-i386: commonize checks for 2MB and 4KB pages Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 20/33] target-i386: commonize checks for 4MB " Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 21/33] target-i386: commonize checks for PAE and non-PAE Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 22/33] target-i386: tweak handling of PG_NX_MASK Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 23/33] target-i386: introduce do_check_protect label Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 24/33] target-i386: introduce support for 1 GB pages Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 25/33] target-i386: set correct error code for reserved bit access Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 26/33] target-i386: test reserved PS bit on PML4Es Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 27/33] target-i386: raise page fault for reserved physical address bits Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 28/33] target-i386: simplify pte/vaddr calculation Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 29/33] target-i386: unify reserved bits and NX bit check Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 30/33] target-i386: raise page fault for reserved bits in large pages Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 31/33] target-i386: support long addresses for 4MB pages (PSE-36) Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 32/33] target-i386: fix protection bits in the TLB for SMEP Paolo Bonzini
2014-06-05 14:22 ` [Qemu-devel] [PULL 33/33] target-i386: cleanup x86_cpu_get_phys_page_debug Paolo Bonzini
2014-06-05 20:50 ` [Qemu-devel] [PULL 00/33] softmmu cleanups and target-i386 paging fixes 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=1401978143-11896-17-git-send-email-pbonzini@redhat.com \
--to=pbonzini@redhat.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).