All of lore.kernel.org
 help / color / mirror / Atom feed
From: Tony Luck <tony.luck@intel.com>
To: Sean Christopherson <seanjc@google.com>,
	Jarkko Sakkinen <jarkko@kernel.org>,
	Dave Hansen <dave.hansen@intel.com>
Cc: Cathy Zhang <cathy.zhang@intel.com>,
	linux-sgx@vger.kernel.org, Tony Luck <tony.luck@intel.com>
Subject: [PATCH v8 4/7] x86/sgx: Add SGX infrastructure to recover from poison
Date: Fri,  1 Oct 2021 09:47:21 -0700	[thread overview]
Message-ID: <20211001164724.220532-5-tony.luck@intel.com> (raw)
In-Reply-To: <20211001164724.220532-1-tony.luck@intel.com>

Provide a recovery function sgx_memory_failure(). If the poison was
consumed synchronously then send a SIGBUS. Note that the virtual
address of the access is not included with the SIGBUS as is the case
for poison outside of SGX enclaves. This doesn't matter as addresses
of code/data inside an enclave is of little to no use to code executing
outside the (now dead) enclave.

Poison found in a free page results in the page being moved from the
free list to the poison page list.

Signed-off-by: Tony Luck <tony.luck@intel.com>
---
 arch/x86/kernel/cpu/sgx/main.c | 77 ++++++++++++++++++++++++++++++++++
 1 file changed, 77 insertions(+)

diff --git a/arch/x86/kernel/cpu/sgx/main.c b/arch/x86/kernel/cpu/sgx/main.c
index 653bace26100..398c9749e4d1 100644
--- a/arch/x86/kernel/cpu/sgx/main.c
+++ b/arch/x86/kernel/cpu/sgx/main.c
@@ -682,6 +682,83 @@ bool arch_is_platform_page(u64 paddr)
 }
 EXPORT_SYMBOL_GPL(arch_is_platform_page);
 
+static struct sgx_epc_page *sgx_paddr_to_page(u64 paddr)
+{
+	struct sgx_epc_section *section;
+
+	section = xa_load(&sgx_epc_address_space, paddr);
+	if (!section)
+		return NULL;
+
+	return &section->pages[PFN_DOWN(paddr - section->phys_addr)];
+}
+
+/*
+ * Called in process context to handle a hardware reported
+ * error in an SGX EPC page.
+ * If the MF_ACTION_REQUIRED bit is set in flags, then the
+ * context is the task that consumed the poison data. Otherwise
+ * this is called from a kernel thread unrelated to the page.
+ */
+int arch_memory_failure(unsigned long pfn, int flags)
+{
+	struct sgx_epc_page *page = sgx_paddr_to_page(pfn << PAGE_SHIFT);
+	struct sgx_epc_section *section;
+	struct sgx_numa_node *node;
+
+	/*
+	 * mm/memory-failure.c calls this routine for all errors
+	 * where there isn't a "struct page" for the address. But that
+	 * includes other address ranges besides SGX.
+	 */
+	if (!page)
+		return -ENXIO;
+
+	/*
+	 * If poison was consumed synchronously. Send a SIGBUS to
+	 * the task. Hardware has already exited the SGX enclave and
+	 * will not allow re-entry to an enclave that has a memory
+	 * error. The signal may help the task understand why the
+	 * enclave is broken.
+	 */
+	if (flags & MF_ACTION_REQUIRED)
+		force_sig(SIGBUS);
+
+	section = &sgx_epc_sections[page->section];
+	node = section->node;
+
+	spin_lock(&node->lock);
+
+	/* Already poisoned? Nothing more to do */
+	if (page->poison)
+		goto out;
+
+	page->poison = 1;
+
+	/*
+	 * If flags is zero, then the page is on a free list.
+	 * Move it to the poison page list.
+	 */
+	if (!page->flags) {
+		list_del(&page->list);
+		list_add(&page->list, &sgx_poison_page_list);
+		goto out;
+	}
+
+	/*
+	 * TBD: Add additional plumbing to enable pre-emptive
+	 * action for asynchronous poison notification. Until
+	 * then just hope that the poison:
+	 * a) is not accessed - sgx_free_epc_page() will deal with it
+	 *    when the user gives it back
+	 * b) results in a recoverable machine check rather than
+	 *    a fatal one
+	 */
+out:
+	spin_unlock(&node->lock);
+	return 0;
+}
+
 /**
  * A section metric is concatenated in a way that @low bits 12-31 define the
  * bits 12-31 of the metric and @high bits 0-19 define the bits 32-51 of the
-- 
2.31.1


  parent reply	other threads:[~2021-10-01 16:47 UTC|newest]

Thread overview: 187+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2021-07-08 18:14 [PATCH 0/4] Basic recovery for machine checks inside SGX Tony Luck
2021-07-08 18:14 ` [PATCH 1/4] x86/sgx: Track phase and type of SGX EPC pages Tony Luck
2021-07-09 18:08   ` Jarkko Sakkinen
2021-07-09 18:09     ` Jarkko Sakkinen
2021-07-14 20:42   ` Reinette Chatre
2021-07-14 20:59     ` Luck, Tony
2021-07-14 21:21       ` Reinette Chatre
2021-07-14 23:08         ` Sean Christopherson
2021-07-14 23:39           ` Luck, Tony
2021-07-15 15:33             ` Sean Christopherson
2021-07-08 18:14 ` [PATCH 2/4] x86/sgx: Add basic infrastructure to recover from errors in SGX memory Tony Luck
2021-07-08 18:14 ` [PATCH 3/4] x86/sgx: Hook sgx_memory_failure() into mainline code Tony Luck
2021-07-08 18:14 ` [PATCH 4/4] x86/sgx: Add hook to error injection address validation Tony Luck
2021-07-19 18:20 ` [PATCH v2 0/6] Basic recovery for machine checks inside SGX Tony Luck
2021-07-19 18:20   ` [PATCH v2 1/6] x86/sgx: Provide indication of life-cycle of EPC pages Tony Luck
2021-07-19 18:28     ` Dave Hansen
2021-07-27  2:04     ` Sakkinen, Jarkko
2021-07-19 18:20   ` [PATCH v2 2/6] x86/sgx: Add infrastructure to identify SGX " Tony Luck
2021-07-19 18:20   ` [PATCH v2 3/6] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-07-27  2:08     ` Sakkinen, Jarkko
2021-07-19 18:20   ` [PATCH v2 4/6] x86/sgx: Add SGX infrastructure to recover from poison Tony Luck
2021-07-19 18:20   ` [PATCH v2 5/6] x86/sgx: Hook sgx_memory_failure() into mainline code Tony Luck
2021-07-19 18:20   ` [PATCH v2 6/6] x86/sgx: Add hook to error injection address validation Tony Luck
2021-07-27  1:54   ` [PATCH v2 0/6] Basic recovery for machine checks inside SGX Sakkinen, Jarkko
2021-07-28 20:46   ` [PATCH v3 0/7] " Tony Luck
2021-07-28 20:46     ` [PATCH v3 1/7] x86/sgx: Provide indication of life-cycle of EPC pages Tony Luck
2021-07-28 22:12       ` Dave Hansen
2021-07-28 22:57         ` Luck, Tony
2021-07-28 23:12           ` Dave Hansen
2021-07-28 23:32             ` Sean Christopherson
2021-07-28 23:48               ` Luck, Tony
2021-07-29  0:07                 ` Sean Christopherson
2021-07-29  0:42                   ` Luck, Tony
2021-07-30  0:34           ` Jarkko Sakkinen
2021-07-30  0:33         ` Jarkko Sakkinen
2021-07-28 20:46     ` [PATCH v3 2/7] x86/sgx: Add infrastructure to identify SGX " Tony Luck
2021-07-28 22:19       ` Dave Hansen
2021-07-30  0:38         ` Jarkko Sakkinen
2021-07-30 16:46           ` Sean Christopherson
2021-07-30 16:50             ` Dave Hansen
2021-07-30 18:44               ` Luck, Tony
2021-07-30 20:35                 ` Dave Hansen
2021-07-30 23:35                   ` Luck, Tony
2021-08-03 21:34                     ` Matthew Wilcox
2021-08-03 23:49                       ` Luck, Tony
2021-08-02  8:52                 ` Jarkko Sakkinen
2021-08-02  8:51               ` Jarkko Sakkinen
2021-08-02  8:48             ` Jarkko Sakkinen
2021-07-28 20:46     ` [PATCH v3 3/7] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-07-30  0:42       ` Jarkko Sakkinen
2021-07-28 20:46     ` [PATCH v3 4/7] x86/sgx: Add SGX infrastructure to recover from poison Tony Luck
2021-07-28 22:29       ` Dave Hansen
2021-07-28 23:00         ` Sean Christopherson
2021-07-28 20:46     ` [PATCH v3 5/7] x86/sgx: Hook sgx_memory_failure() into mainline code Tony Luck
2021-07-28 20:46     ` [PATCH v3 6/7] x86/sgx: Add hook to error injection address validation Tony Luck
2021-07-28 20:46     ` [PATCH v3 7/7] x86/sgx: Add documentation for SGX memory errors Tony Luck
2021-08-27 19:55     ` [PATCH v4 0/6] Basic recovery for machine checks inside SGX Tony Luck
2021-08-27 19:55       ` [PATCH v4 1/6] x86/sgx: Provide indication of life-cycle of EPC pages Tony Luck
2021-09-01  3:55         ` Jarkko Sakkinen
2021-08-27 19:55       ` [PATCH v4 2/6] x86/sgx: Add infrastructure to identify SGX " Tony Luck
2021-09-01  4:30         ` Jarkko Sakkinen
2021-08-27 19:55       ` [PATCH v4 3/6] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-08-27 19:55       ` [PATCH v4 4/6] x86/sgx: Add SGX infrastructure to recover from poison Tony Luck
2021-08-27 19:55       ` [PATCH v4 5/6] x86/sgx: Hook sgx_memory_failure() into mainline code Tony Luck
2021-09-03  6:12         ` Jarkko Sakkinen
2021-09-03  6:56           ` Jarkko Sakkinen
2021-09-06 18:51             ` Luck, Tony
2021-09-07 14:07               ` Jarkko Sakkinen
2021-09-07 14:13                 ` Dave Hansen
2021-09-07 15:07                   ` Luck, Tony
2021-09-07 15:03                 ` Luck, Tony
2021-09-07 15:08                   ` Jarkko Sakkinen
2021-09-07 17:46                     ` Luck, Tony
2021-09-08  0:59                       ` Luck, Tony
2021-09-08 16:49                         ` Dave Hansen
2021-09-08  2:29                       ` Jarkko Sakkinen
2021-08-27 19:55       ` [PATCH v4 6/6] x86/sgx: Add hook to error injection address validation Tony Luck
2021-08-27 20:28       ` [PATCH v4 0/6] Basic recovery for machine checks inside SGX Borislav Petkov
2021-08-27 20:43         ` Sean Christopherson
2021-09-01  2:06       ` Jarkko Sakkinen
2021-09-01 14:48         ` Luck, Tony
2021-09-17 21:38       ` [PATCH v5 0/7] " Tony Luck
2021-09-17 21:38         ` [PATCH v5 1/7] x86/sgx: Provide indication of life-cycle of EPC pages Tony Luck
2021-09-21 21:28           ` Jarkko Sakkinen
2021-09-21 21:34             ` Luck, Tony
2021-09-22  5:17               ` Jarkko Sakkinen
2021-09-21 22:15             ` Dave Hansen
2021-09-22  5:27               ` Jarkko Sakkinen
2021-09-17 21:38         ` [PATCH v5 2/7] x86/sgx: Add infrastructure to identify SGX " Tony Luck
2021-09-21 20:23           ` Dave Hansen
2021-09-21 20:50             ` Luck, Tony
2021-09-21 22:32               ` Dave Hansen
2021-09-21 23:48                 ` Luck, Tony
2021-09-21 23:50                   ` Dave Hansen
2021-09-17 21:38         ` [PATCH v5 3/7] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-09-17 21:38         ` [PATCH v5 4/7] x86/sgx: Add SGX infrastructure to recover from poison Tony Luck
2021-09-17 21:38         ` [PATCH v5 5/7] x86/sgx: Hook arch_memory_failure() into mainline code Tony Luck
2021-09-17 21:38         ` [PATCH v5 6/7] x86/sgx: Add hook to error injection address validation Tony Luck
2021-09-17 21:38         ` [PATCH v5 7/7] x86/sgx: Add check for SGX pages to ghes_do_memory_failure() Tony Luck
2021-09-22 18:21         ` [PATCH v6 0/7] Basic recovery for machine checks inside SGX Tony Luck
2021-09-22 18:21           ` [PATCH v6 1/7] x86/sgx: Provide indication of life-cycle of EPC pages Tony Luck
2021-09-23 20:21             ` Jarkko Sakkinen
2021-09-23 20:24               ` Jarkko Sakkinen
2021-09-23 20:46                 ` Luck, Tony
2021-09-23 22:11                   ` Luck, Tony
2021-09-28  2:13                     ` Jarkko Sakkinen
2021-09-22 18:21           ` [PATCH v6 2/7] x86/sgx: Add infrastructure to identify SGX " Tony Luck
2021-09-22 18:21           ` [PATCH v6 3/7] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-09-22 18:21           ` [PATCH v6 4/7] x86/sgx: Add SGX infrastructure to recover from poison Tony Luck
2021-09-22 18:21           ` [PATCH v6 5/7] x86/sgx: Hook arch_memory_failure() into mainline code Tony Luck
2021-09-22 18:21           ` [PATCH v6 6/7] x86/sgx: Add hook to error injection address validation Tony Luck
2021-09-22 18:21           ` [PATCH v6 7/7] x86/sgx: Add check for SGX pages to ghes_do_memory_failure() Tony Luck
2021-09-27 21:34           ` [PATCH v7 0/7] Basic recovery for machine checks inside SGX Tony Luck
2021-09-27 21:34             ` [PATCH v7 1/7] x86/sgx: Add new sgx_epc_page flag bit to mark in-use pages Tony Luck
2021-09-28  2:28               ` Jarkko Sakkinen
2021-09-27 21:34             ` [PATCH v7 2/7] x86/sgx: Add infrastructure to identify SGX EPC pages Tony Luck
2021-09-28  2:30               ` Jarkko Sakkinen
2021-09-27 21:34             ` [PATCH v7 3/7] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-09-28  2:46               ` Jarkko Sakkinen
2021-09-28 15:41                 ` Luck, Tony
2021-09-28 20:11                   ` Jarkko Sakkinen
2021-09-28 20:53                     ` Luck, Tony
2021-09-30 14:40                       ` Jarkko Sakkinen
2021-09-30 18:02                         ` Luck, Tony
2021-09-27 21:34             ` [PATCH v7 4/7] x86/sgx: Add SGX infrastructure to recover from poison Tony Luck
2021-09-27 21:34             ` [PATCH v7 5/7] x86/sgx: Hook arch_memory_failure() into mainline code Tony Luck
2021-09-27 21:34             ` [PATCH v7 6/7] x86/sgx: Add hook to error injection address validation Tony Luck
2021-09-27 21:34             ` [PATCH v7 7/7] x86/sgx: Add check for SGX pages to ghes_do_memory_failure() Tony Luck
2021-10-01 16:47             ` [PATCH v8 0/7] Basic recovery for machine checks inside SGX Tony Luck
2021-10-01 16:47               ` [PATCH v8 1/7] x86/sgx: Add new sgx_epc_page flag bit to mark in-use pages Tony Luck
2021-10-01 16:47               ` [PATCH v8 2/7] x86/sgx: Add infrastructure to identify SGX EPC pages Tony Luck
2021-10-01 16:47               ` [PATCH v8 3/7] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-10-04 23:24                 ` Jarkko Sakkinen
2021-10-01 16:47               ` Tony Luck [this message]
2021-10-04 23:30                 ` [PATCH v8 4/7] x86/sgx: Add SGX infrastructure to recover from poison Jarkko Sakkinen
2021-10-01 16:47               ` [PATCH v8 5/7] x86/sgx: Hook arch_memory_failure() into mainline code Tony Luck
2021-10-01 16:47               ` [PATCH v8 6/7] x86/sgx: Add hook to error injection address validation Tony Luck
2021-10-01 16:47               ` [PATCH v8 7/7] x86/sgx: Add check for SGX pages to ghes_do_memory_failure() Tony Luck
2021-10-04 21:56               ` [PATCH v8 0/7] Basic recovery for machine checks inside SGX Reinette Chatre
2021-10-11 18:59               ` [PATCH v9 " Tony Luck
2021-10-11 18:59                 ` [PATCH v9 1/7] x86/sgx: Add new sgx_epc_page flag bit to mark in-use pages Tony Luck
2021-10-15 22:57                   ` Sean Christopherson
2021-10-11 18:59                 ` [PATCH v9 2/7] x86/sgx: Add infrastructure to identify SGX EPC pages Tony Luck
2021-10-22 10:43                   ` kernel test robot
2021-10-22 10:43                     ` kernel test robot
2021-10-11 18:59                 ` [PATCH v9 3/7] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-10-15 23:07                   ` Sean Christopherson
2021-10-15 23:32                     ` Luck, Tony
2021-10-11 18:59                 ` [PATCH v9 4/7] x86/sgx: Add SGX infrastructure to recover from poison Tony Luck
2021-10-15 23:10                   ` Sean Christopherson
2021-10-15 23:19                     ` Luck, Tony
2021-10-11 18:59                 ` [PATCH v9 5/7] x86/sgx: Hook arch_memory_failure() into mainline code Tony Luck
2021-10-12 16:49                   ` Jarkko Sakkinen
2021-10-11 18:59                 ` [PATCH v9 6/7] x86/sgx: Add hook to error injection address validation Tony Luck
2021-10-12 16:50                   ` Jarkko Sakkinen
2021-10-11 18:59                 ` [PATCH v9 7/7] x86/sgx: Add check for SGX pages to ghes_do_memory_failure() Tony Luck
2021-10-12 16:51                   ` Jarkko Sakkinen
2021-10-12 16:48                 ` [PATCH v9 0/7] Basic recovery for machine checks inside SGX Jarkko Sakkinen
2021-10-12 17:57                   ` Luck, Tony
2021-10-18 20:25                 ` [PATCH v10 " Tony Luck
2021-10-18 20:25                   ` [PATCH v10 1/7] x86/sgx: Add new sgx_epc_page flag bit to mark free pages Tony Luck
2021-10-18 20:25                   ` [PATCH v10 2/7] x86/sgx: Add infrastructure to identify SGX EPC pages Tony Luck
2021-11-19 12:25                     ` kernel test robot
2021-11-19 12:25                       ` kernel test robot
2021-10-18 20:25                   ` [PATCH v10 3/7] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-10-18 20:25                   ` [PATCH v10 4/7] x86/sgx: Add SGX infrastructure to recover from poison Tony Luck
2021-10-18 20:25                   ` [PATCH v10 5/7] x86/sgx: Hook arch_memory_failure() into mainline code Tony Luck
2021-10-20  9:06                     ` Naoya Horiguchi
2021-10-20 17:04                       ` Luck, Tony
2021-10-18 20:25                   ` [PATCH v10 6/7] x86/sgx: Add hook to error injection address validation Tony Luck
2021-10-18 20:25                   ` [PATCH v10 7/7] x86/sgx: Add check for SGX pages to ghes_do_memory_failure() Tony Luck
2021-10-26 22:00                   ` [PATCH v11 0/7] Basic recovery for machine checks inside SGX Tony Luck
2021-10-26 22:00                     ` [PATCH v11 1/7] x86/sgx: Add new sgx_epc_page flag bit to mark free pages Tony Luck
2021-11-15 20:22                       ` [tip: x86/sgx] " tip-bot2 for Tony Luck
2021-10-26 22:00                     ` [PATCH v11 2/7] x86/sgx: Add infrastructure to identify SGX EPC pages Tony Luck
2021-11-15 20:22                       ` [tip: x86/sgx] " tip-bot2 for Tony Luck
2021-10-26 22:00                     ` [PATCH v11 3/7] x86/sgx: Initial poison handling for dirty and free pages Tony Luck
2021-11-15 20:22                       ` [tip: x86/sgx] " tip-bot2 for Tony Luck
2021-10-26 22:00                     ` [PATCH v11 4/7] x86/sgx: Add SGX infrastructure to recover from poison Tony Luck
2021-11-15 20:22                       ` [tip: x86/sgx] " tip-bot2 for Tony Luck
2021-10-26 22:00                     ` [PATCH v11 5/7] x86/sgx: Hook arch_memory_failure() into mainline code Tony Luck
2021-11-15 20:22                       ` [tip: x86/sgx] " tip-bot2 for Tony Luck
2021-10-26 22:00                     ` [PATCH v11 6/7] x86/sgx: Add hook to error injection address validation Tony Luck
2021-11-15 20:22                       ` [tip: x86/sgx] " tip-bot2 for Tony Luck
2021-10-26 22:00                     ` [PATCH v11 7/7] x86/sgx: Add check for SGX pages to ghes_do_memory_failure() Tony Luck
2021-10-29 18:39                       ` Rafael J. Wysocki
2021-11-15 20:22                       ` [tip: x86/sgx] " tip-bot2 for Tony Luck

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=20211001164724.220532-5-tony.luck@intel.com \
    --to=tony.luck@intel.com \
    --cc=cathy.zhang@intel.com \
    --cc=dave.hansen@intel.com \
    --cc=jarkko@kernel.org \
    --cc=linux-sgx@vger.kernel.org \
    --cc=seanjc@google.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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.