public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: ira.weiny@intel.com
To: Dave Hansen <dave.hansen@linux.intel.com>,
	"H. Peter Anvin" <hpa@zytor.com>,
	Dan Williams <dan.j.williams@intel.com>
Cc: Ira Weiny <ira.weiny@intel.com>,
	Fenghua Yu <fenghua.yu@intel.com>,
	Rick Edgecombe <rick.p.edgecombe@intel.com>,
	linux-kernel@vger.kernel.org
Subject: [PATCH V8 14/44] x86/pkeys: Introduce pks_write_pkrs()
Date: Thu, 27 Jan 2022 09:54:35 -0800	[thread overview]
Message-ID: <20220127175505.851391-15-ira.weiny@intel.com> (raw)
In-Reply-To: <20220127175505.851391-1-ira.weiny@intel.com>

From: Ira Weiny <ira.weiny@intel.com>

Writing to MSR's is inefficient.  Even though the underlying
WRMSR(MSR_IA32_PKRS) is not serializing (see below), writing to the MSR
unnecessarily should be avoided.  This is especially true when the value
of the PKS protections is unlikely to change from the default often.

Introduce pks_write_pkrs() which avoids writing the MSR if the pkrs
value has not changed for the CPU.  Do this by utilizing a per-cpu
cache.  Protect the use of the cached value from preemption by
restricting the use of pks_write_pkrs() to non-preemptable context.
Further restrict it's use to callers which have checked X86_FEATURE_PKS.

The initial value of the MSR is preserved on INIT.  While unlikely, the
PKS_INIT_VALUE may be 0 someday which would prevent pks_write_pkrs()
from updating the MSR.  Keep the MSR write in pks_setup() to ensure the
MSR is initialized at least one time.  Then call pks_write_pkrs() to set
up the per-cache value to ensure it is in sync with the MSR.

It should be noted that the underlying WRMSR(MSR_IA32_PKRS) is not
serializing but still maintains ordering properties similar to WRPKRU.
The current SDM section on PKRS needs updating but should be the same as
that of WRPKRU.  So to quote from the WRPKRU text:

	WRPKRU will never execute transiently. Memory accesses affected
	by PKRU register will not execute (even transiently) until all
	prior executions of WRPKRU have completed execution and updated
	the PKRU register.

Suggested-by: Dave Hansen <dave.hansen@intel.com>
Signed-off-by: Ira Weiny <ira.weiny@intel.com>

---
Changes for V8
	From Thomas
		Remove get/put_cpu_ptr() and make this a 'lower level
		call.  This makes it preemption unsafe but it is called
		mostly where preemption is already disabled.  Add this
		as a predicate of the call and those calls which need to
		can disable preemption.
		Add lockdep assert for preemption
	Ensure MSR gets written even if the PKS_INIT_VALUE is 0.
	Completely re-write the commit message.
	s/write_pkrs/pks_write_pkrs/
	Split this off into a singular patch

Changes for V7
	Create a dynamic pkrs_initial_value in early init code.
	Clean up comments
	Add comment to macro guard
---
 arch/x86/mm/pkeys.c | 41 +++++++++++++++++++++++++++++++++++++++++
 1 file changed, 41 insertions(+)

diff --git a/arch/x86/mm/pkeys.c b/arch/x86/mm/pkeys.c
index a5b5b86e97ce..3dce99ef4127 100644
--- a/arch/x86/mm/pkeys.c
+++ b/arch/x86/mm/pkeys.c
@@ -209,15 +209,56 @@ u32 pkey_update_pkval(u32 pkval, int pkey, u32 accessbits)
 
 #ifdef CONFIG_ARCH_ENABLE_SUPERVISOR_PKEYS
 
+static DEFINE_PER_CPU(u32, pkrs_cache);
+
+/*
+ * pks_write_pkrs() - Write the pkrs of the current CPU
+ * @new_pkrs: New value to write to the current CPU register
+ *
+ * Optimizes the MSR writes by maintaining a per cpu cache.
+ *
+ * Context: must be called with preemption disabled
+ * Context: must only be called if PKS is enabled
+ *
+ * It should also be noted that the underlying WRMSR(MSR_IA32_PKRS) is not
+ * serializing but still maintains ordering properties similar to WRPKRU.
+ * The current SDM section on PKRS needs updating but should be the same as
+ * that of WRPKRU.  Quote from the WRPKRU text:
+ *
+ *     WRPKRU will never execute transiently. Memory accesses
+ *     affected by PKRU register will not execute (even transiently)
+ *     until all prior executions of WRPKRU have completed execution
+ *     and updated the PKRU register.
+ */
+static inline void pks_write_pkrs(u32 new_pkrs)
+{
+	u32 pkrs = __this_cpu_read(pkrs_cache);
+
+	lockdep_assert_preemption_disabled();
+
+	if (pkrs != new_pkrs) {
+		__this_cpu_write(pkrs_cache, new_pkrs);
+		wrmsrl(MSR_IA32_PKRS, new_pkrs);
+	}
+}
+
 /*
  * PKS is independent of PKU and either or both may be supported on a CPU.
+ *
+ * Context: must be called with preemption disabled
  */
 void pks_setup(void)
 {
 	if (!cpu_feature_enabled(X86_FEATURE_PKS))
 		return;
 
+	/*
+	 * If the PKS_INIT_VALUE is 0 then pks_write_pkrs() could fail to
+	 * initialize the MSR.  Do a single write here to ensure the MSR is
+	 * written at least one time.
+	 */
 	wrmsrl(MSR_IA32_PKRS, PKS_INIT_VALUE);
+	pks_write_pkrs(PKS_INIT_VALUE);
 	cr4_set_bits(X86_CR4_PKS);
 }
 
-- 
2.31.1


  parent reply	other threads:[~2022-01-27 17:55 UTC|newest]

Thread overview: 145+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2022-01-27 17:54 [PATCH V8 00/44] PKS/PMEM: Add Stray Write Protection ira.weiny
2022-01-27 17:54 ` [PATCH V8 01/44] entry: Create an internal irqentry_exit_cond_resched() call ira.weiny
2022-01-27 17:54 ` [PATCH V8 02/44] Documentation/protection-keys: Clean up documentation for User Space pkeys ira.weiny
2022-01-28 22:39   ` Dave Hansen
2022-02-01 23:49     ` Ira Weiny
2022-02-01 23:54       ` Dave Hansen
2022-01-27 17:54 ` [PATCH V8 03/44] x86/pkeys: Create pkeys_common.h ira.weiny
2022-01-28 22:43   ` Dave Hansen
2022-02-02  1:00     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 04/44] x86/pkeys: Add additional PKEY helper macros ira.weiny
2022-01-28 22:47   ` Dave Hansen
2022-02-02 20:21     ` Ira Weiny
2022-02-02 20:26       ` Dave Hansen
2022-02-02 20:28         ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 05/44] x86/fpu: Refactor arch_set_user_pkey_access() ira.weiny
2022-01-28 22:50   ` Dave Hansen
2022-02-02 20:22     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 06/44] mm/pkeys: Add Kconfig options for PKS ira.weiny
2022-01-28 22:54   ` Dave Hansen
2022-01-28 23:10     ` Ira Weiny
2022-01-28 23:51       ` Dave Hansen
2022-02-04 19:08         ` Ira Weiny
2022-02-09  5:34           ` Ira Weiny
2022-02-14 19:20             ` Dave Hansen
2022-02-14 23:03               ` Ira Weiny
2022-01-29  0:06   ` Dave Hansen
2022-02-04 19:14     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 07/44] x86/pkeys: Add PKS CPU feature bit ira.weiny
2022-01-28 23:05   ` Dave Hansen
2022-02-04 19:21     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 08/44] x86/fault: Adjust WARN_ON for PKey fault ira.weiny
2022-01-28 23:10   ` Dave Hansen
2022-02-04 20:06     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 09/44] x86/pkeys: Enable PKS on cpus which support it ira.weiny
2022-01-28 23:18   ` Dave Hansen
2022-01-28 23:41     ` Ira Weiny
2022-01-28 23:53       ` Dave Hansen
2022-01-27 17:54 ` [PATCH V8 10/44] Documentation/pkeys: Add initial PKS documentation ira.weiny
2022-01-28 23:57   ` Dave Hansen
2022-01-27 17:54 ` [PATCH V8 11/44] mm/pkeys: Define static PKS key array and default values ira.weiny
2022-01-29  0:02   ` Dave Hansen
2022-02-04 23:54     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 12/44] mm/pkeys: Define PKS page table macros ira.weiny
2022-01-27 17:54 ` [PATCH V8 13/44] mm/pkeys: Add initial PKS Test code ira.weiny
2022-01-31 19:30   ` Edgecombe, Rick P
2022-02-09 23:44     ` Ira Weiny
2022-01-27 17:54 ` ira.weiny [this message]
2022-01-29  0:12   ` [PATCH V8 14/44] x86/pkeys: Introduce pks_write_pkrs() Dave Hansen
2022-01-29  0:16     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 15/44] x86/pkeys: Preserve the PKS MSR on context switch ira.weiny
2022-01-29  0:22   ` Dave Hansen
2022-02-11  6:10     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 16/44] mm/pkeys: Introduce pks_mk_readwrite() ira.weiny
2022-01-31 23:10   ` Edgecombe, Rick P
2022-02-18  2:22     ` Ira Weiny
2022-02-01 17:40   ` Dave Hansen
2022-02-18  4:39     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 17/44] mm/pkeys: Introduce pks_mk_noaccess() ira.weiny
2022-01-27 17:54 ` [PATCH V8 18/44] x86/fault: Add a PKS test fault hook ira.weiny
2022-01-31 19:56   ` Edgecombe, Rick P
2022-02-11 20:40     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 19/44] mm/pkeys: PKS Testing, add pks_mk_*() tests ira.weiny
2022-02-01 17:45   ` Dave Hansen
2022-02-18  5:34     ` Ira Weiny
2022-02-18 15:28       ` Dave Hansen
2022-02-18 17:25         ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 20/44] mm/pkeys: Add PKS test for context switching ira.weiny
2022-02-01 17:43   ` Edgecombe, Rick P
2022-02-22 21:42     ` Ira Weiny
2022-02-01 17:47   ` Edgecombe, Rick P
2022-02-01 19:52     ` Edgecombe, Rick P
2022-02-18  6:03       ` Ira Weiny
2022-02-18  6:02     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 21/44] x86/entry: Add auxiliary pt_regs space ira.weiny
2022-01-27 17:54 ` [PATCH V8 22/44] entry: Pass pt_regs to irqentry_exit_cond_resched() ira.weiny
2022-01-27 17:54 ` [PATCH V8 23/44] entry: Add architecture auxiliary pt_regs save/restore calls ira.weiny
2022-01-27 17:54 ` [PATCH V8 24/44] x86/entry: Define arch_{save|restore}_auxiliary_pt_regs() ira.weiny
2022-01-27 17:54 ` [PATCH V8 25/44] x86/pkeys: Preserve PKRS MSR across exceptions ira.weiny
2022-01-27 17:54 ` [PATCH V8 26/44] x86/fault: Print PKS MSR on fault ira.weiny
2022-02-01 18:13   ` Edgecombe, Rick P
2022-02-18  6:01     ` Ira Weiny
2022-02-18 17:28       ` Edgecombe, Rick P
2022-02-18 20:20         ` Dave Hansen
2022-02-18 20:54           ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 27/44] mm/pkeys: Add PKS exception test ira.weiny
2022-01-27 17:54 ` [PATCH V8 28/44] mm/pkeys: Introduce pks_update_exception() ira.weiny
2022-01-27 17:54 ` [PATCH V8 29/44] mm/pkeys: Introduce PKS fault callbacks ira.weiny
2022-01-27 17:54 ` [PATCH V8 30/44] mm/pkeys: Test setting a PKS key in a custom fault callback ira.weiny
2022-02-01  0:55   ` Edgecombe, Rick P
2022-03-01 15:39     ` Ira Weiny
2022-02-01 17:42   ` Edgecombe, Rick P
2022-02-11 20:44     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 31/44] mm/pkeys: Add pks_available() ira.weiny
2022-01-27 17:54 ` [PATCH V8 32/44] memremap_pages: Add Kconfig for DEVMAP_ACCESS_PROTECTION ira.weiny
2022-02-04 15:49   ` Dan Williams
2022-01-27 17:54 ` [PATCH V8 33/44] memremap_pages: Introduce pgmap_protection_available() ira.weiny
2022-02-04 16:19   ` Dan Williams
2022-02-28 16:59     ` Ira Weiny
2022-03-01 15:56       ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 34/44] memremap_pages: Introduce a PGMAP_PROTECTION flag ira.weiny
2022-01-27 17:54 ` [PATCH V8 35/44] memremap_pages: Introduce devmap_protected() ira.weiny
2022-01-27 17:54 ` [PATCH V8 36/44] memremap_pages: Reserve a PKS PKey for eventual use by PMEM ira.weiny
2022-02-01 18:35   ` Edgecombe, Rick P
2022-02-04 17:12     ` Dan Williams
2022-02-05  5:40       ` Ira Weiny
2022-02-05  8:19         ` Dan Williams
2022-02-06 18:14           ` Dan Williams
2022-02-08 22:48           ` Ira Weiny
2022-02-08 23:22             ` Dan Williams
2022-02-08 23:42               ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 37/44] memremap_pages: Set PKS PKey in PTEs if PGMAP_PROTECTIONS is requested ira.weiny
2022-02-04 17:41   ` Dan Williams
2022-03-01 18:15     ` Ira Weiny
2022-01-27 17:54 ` [PATCH V8 38/44] memremap_pages: Define pgmap_mk_{readwrite|noaccess}() calls ira.weiny
2022-02-04 18:35   ` Dan Williams
2022-02-05  0:09     ` Ira Weiny
2022-02-05  0:19       ` Dan Williams
2022-02-05  0:25         ` Dan Williams
2022-02-05  0:27           ` Dan Williams
2022-02-05  5:55             ` Ira Weiny
2022-02-05  6:28               ` Dan Williams
2022-02-22 22:05     ` Ira Weiny
2022-01-27 17:55 ` [PATCH V8 39/44] memremap_pages: Add memremap.pks_fault_mode ira.weiny
2022-02-01  1:16   ` Edgecombe, Rick P
2022-03-02  0:20     ` Ira Weiny
2022-02-04 19:01   ` Dan Williams
2022-03-02  2:00     ` Ira Weiny
2022-01-27 17:55 ` [PATCH V8 40/44] memremap_pages: Add pgmap_protection_flag_invalid() ira.weiny
2022-02-01  1:37   ` Edgecombe, Rick P
2022-03-02  2:01     ` Ira Weiny
2022-02-04 19:18   ` Dan Williams
2022-01-27 17:55 ` [PATCH V8 41/44] kmap: Ensure kmap works for devmap pages ira.weiny
2022-02-04 21:07   ` Dan Williams
2022-03-01 19:45     ` Ira Weiny
2022-03-01 19:50       ` Ira Weiny
2022-03-01 20:05       ` Dan Williams
2022-03-01 23:03         ` Ira Weiny
2022-01-27 17:55 ` [PATCH V8 42/44] dax: Stray access protection for dax_direct_access() ira.weiny
2022-02-04  5:19   ` Dan Williams
2022-03-01 18:13     ` Ira Weiny
2022-01-27 17:55 ` [PATCH V8 43/44] nvdimm/pmem: Enable stray access protection ira.weiny
2022-02-04 21:10   ` Dan Williams
2022-03-01 18:18     ` Ira Weiny
2022-01-27 17:55 ` [PATCH V8 44/44] devdax: " ira.weiny
2022-02-04 21:12   ` Dan Williams

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=20220127175505.851391-15-ira.weiny@intel.com \
    --to=ira.weiny@intel.com \
    --cc=dan.j.williams@intel.com \
    --cc=dave.hansen@linux.intel.com \
    --cc=fenghua.yu@intel.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=rick.p.edgecombe@intel.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