From: "Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
To: Dave Hansen <dave.hansen@linux.intel.com>,
Andy Lutomirski <luto@kernel.org>,
Peter Zijlstra <peterz@infradead.org>
Cc: x86@kernel.org, Kostya Serebryany <kcc@google.com>,
Andrey Ryabinin <ryabinin.a.a@gmail.com>,
Andrey Konovalov <andreyknvl@gmail.com>,
Alexander Potapenko <glider@google.com>,
Taras Madan <tarasmadan@google.com>,
Dmitry Vyukov <dvyukov@google.com>,
"H . J . Lu" <hjl.tools@gmail.com>,
Andi Kleen <ak@linux.intel.com>,
Rick Edgecombe <rick.p.edgecombe@intel.com>,
Bharata B Rao <bharata@amd.com>,
Jacob Pan <jacob.jun.pan@linux.intel.com>,
Ashok Raj <ashok.raj@intel.com>,
Linus Torvalds <torvalds@linux-foundation.org>,
linux-mm@kvack.org, linux-kernel@vger.kernel.org,
"Kirill A. Shutemov" <kirill.shutemov@linux.intel.com>
Subject: [PATCHv15 07/17] x86/mm: Reduce untagged_addr() overhead for systems without LAM
Date: Tue, 24 Jan 2023 01:04:50 +0300 [thread overview]
Message-ID: <20230123220500.21077-8-kirill.shutemov@linux.intel.com> (raw)
In-Reply-To: <20230123220500.21077-1-kirill.shutemov@linux.intel.com>
Use alternatives to reduce untagged_addr() overhead.
Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
---
arch/x86/include/asm/disabled-features.h | 8 ++++-
arch/x86/include/asm/uaccess.h | 41 +++++++++++++++++-------
2 files changed, 37 insertions(+), 12 deletions(-)
diff --git a/arch/x86/include/asm/disabled-features.h b/arch/x86/include/asm/disabled-features.h
index c44b56f7ffba..3f0c31044f02 100644
--- a/arch/x86/include/asm/disabled-features.h
+++ b/arch/x86/include/asm/disabled-features.h
@@ -75,6 +75,12 @@
# define DISABLE_CALL_DEPTH_TRACKING (1 << (X86_FEATURE_CALL_DEPTH & 31))
#endif
+#ifdef CONFIG_ADDRESS_MASKING
+# define DISABLE_LAM 0
+#else
+# define DISABLE_LAM (1 << (X86_FEATURE_LAM & 31))
+#endif
+
#ifdef CONFIG_INTEL_IOMMU_SVM
# define DISABLE_ENQCMD 0
#else
@@ -115,7 +121,7 @@
#define DISABLED_MASK10 0
#define DISABLED_MASK11 (DISABLE_RETPOLINE|DISABLE_RETHUNK|DISABLE_UNRET| \
DISABLE_CALL_DEPTH_TRACKING)
-#define DISABLED_MASK12 0
+#define DISABLED_MASK12 (DISABLE_LAM)
#define DISABLED_MASK13 0
#define DISABLED_MASK14 0
#define DISABLED_MASK15 0
diff --git a/arch/x86/include/asm/uaccess.h b/arch/x86/include/asm/uaccess.h
index fd9182951084..6450a2723bcd 100644
--- a/arch/x86/include/asm/uaccess.h
+++ b/arch/x86/include/asm/uaccess.h
@@ -9,6 +9,7 @@
#include <linux/kasan-checks.h>
#include <linux/mm_types.h>
#include <linux/string.h>
+#include <linux/mmap_lock.h>
#include <asm/asm.h>
#include <asm/page.h>
#include <asm/smap.h>
@@ -30,26 +31,44 @@ static inline bool pagefault_disabled(void);
* Magic with the 'sign' allows to untag userspace pointer without any branches
* while leaving kernel addresses intact.
*/
-static inline unsigned long __untagged_addr(unsigned long addr,
- unsigned long mask)
+static inline unsigned long __untagged_addr(unsigned long addr)
{
- long sign = addr >> 63;
+ long sign;
+
+ /*
+ * Refer tlbstate_untag_mask directly to avoid RIP-relative relocation
+ * in alternative instructions. The relocation gets wrong when gets
+ * copied to the target place.
+ */
+ asm (ALTERNATIVE("",
+ "sar $63, %[sign]\n\t" /* user_ptr ? 0 : -1UL */
+ "or %%gs:tlbstate_untag_mask, %[sign]\n\t"
+ "and %[sign], %[addr]\n\t", X86_FEATURE_LAM)
+ : [addr] "+r" (addr), [sign] "=r" (sign)
+ : "m" (tlbstate_untag_mask), "[sign]" (addr));
- addr &= mask | sign;
return addr;
}
#define untagged_addr(addr) ({ \
- u64 __addr = (__force u64)(addr); \
- __addr = __untagged_addr(__addr, current_untag_mask()); \
- (__force __typeof__(addr))__addr; \
+ unsigned long __addr = (__force unsigned long)(addr); \
+ (__force __typeof__(addr))__untagged_addr(__addr); \
})
+static inline unsigned long __untagged_addr_remote(struct mm_struct *mm,
+ unsigned long addr)
+{
+ long sign = addr >> 63;
+
+ mmap_assert_locked(mm);
+ addr &= (mm)->context.untag_mask | sign;
+
+ return addr;
+}
+
#define untagged_addr_remote(mm, addr) ({ \
- u64 __addr = (__force u64)(addr); \
- mmap_assert_locked(mm); \
- __addr = __untagged_addr(__addr, (mm)->context.untag_mask); \
- (__force __typeof__(addr))__addr; \
+ unsigned long __addr = (__force unsigned long)(addr); \
+ (__force __typeof__(addr))__untagged_addr_remote(mm, __addr); \
})
#else
--
2.39.1
next prev parent reply other threads:[~2023-01-23 22:05 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2023-01-23 22:04 [PATCHv15 00/17] Linear Address Masking enabling Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 01/17] x86/mm: Rework address range check in get_user() and put_user() Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 02/17] x86: Allow atomic MM_CONTEXT flags setting Kirill A. Shutemov
2023-02-09 14:06 ` Alexander Potapenko
2023-01-23 22:04 ` [PATCHv15 03/17] x86: CPUID and CR3/CR4 flags for Linear Address Masking Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 04/17] x86/mm: Handle LAM on context switch Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 05/17] mm: Introduce untagged_addr_remote() Kirill A. Shutemov
2023-02-09 14:08 ` Alexander Potapenko
2023-01-23 22:04 ` [PATCHv15 06/17] x86/uaccess: Provide untagged_addr() and remove tags before address check Kirill A. Shutemov
2023-02-09 14:10 ` Alexander Potapenko
2023-02-10 15:18 ` Alexander Potapenko
2023-01-23 22:04 ` Kirill A. Shutemov [this message]
2023-01-23 22:04 ` [PATCHv15 08/17] x86/mm: Provide arch_prctl() interface for LAM Kirill A. Shutemov
2023-02-09 14:12 ` Alexander Potapenko
2023-01-23 22:04 ` [PATCHv15 09/17] mm: Expose untagging mask in /proc/$PID/status Kirill A. Shutemov
2023-02-09 14:13 ` Alexander Potapenko
2023-02-10 12:29 ` Alexander Potapenko
2023-01-23 22:04 ` [PATCHv15 10/17] iommu/sva: Replace pasid_valid() helper with mm_valid_pasid() Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 11/17] x86/mm/iommu/sva: Make LAM and SVA mutually exclusive Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 12/17] selftests/x86/lam: Add malloc and tag-bits test cases for linear-address masking Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 13/17] selftests/x86/lam: Add mmap and SYSCALL " Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 14/17] selftests/x86/lam: Add io_uring " Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 15/17] selftests/x86/lam: Add inherit " Kirill A. Shutemov
2023-01-23 22:04 ` [PATCHv15 16/17] selftests/x86/lam: Add ARCH_FORCE_TAGGED_SVA " Kirill A. Shutemov
2023-01-23 22:05 ` [PATCHv15 17/17] selftests/x86/lam: Add test cases for LAM vs thread creation Kirill A. Shutemov
2023-01-24 2:07 ` [PATCHv15 00/17] Linear Address Masking enabling Linus Torvalds
2023-01-24 11:01 ` Kirill A. Shutemov
2023-01-26 14:42 ` Kirill A. Shutemov
2023-01-26 18:31 ` Linus Torvalds
2023-03-13 3:23 ` Binbin Wu
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=20230123220500.21077-8-kirill.shutemov@linux.intel.com \
--to=kirill.shutemov@linux.intel.com \
--cc=ak@linux.intel.com \
--cc=andreyknvl@gmail.com \
--cc=ashok.raj@intel.com \
--cc=bharata@amd.com \
--cc=dave.hansen@linux.intel.com \
--cc=dvyukov@google.com \
--cc=glider@google.com \
--cc=hjl.tools@gmail.com \
--cc=jacob.jun.pan@linux.intel.com \
--cc=kcc@google.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-mm@kvack.org \
--cc=luto@kernel.org \
--cc=peterz@infradead.org \
--cc=rick.p.edgecombe@intel.com \
--cc=ryabinin.a.a@gmail.com \
--cc=tarasmadan@google.com \
--cc=torvalds@linux-foundation.org \
--cc=x86@kernel.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