[PATCHv14 0/9] mm, x86/cc, efi: Implement support for unaccepted memory

[PATCHv14 0/9] mm, x86/cc, efi: Implement support for unaccepted memory

[Kirill A. Shutemov]

UEFI Specification version 2.9 introduces the concept of memory
acceptance: some Virtual Machine platforms, such as Intel TDX or AMD
SEV-SNP, requiring memory to be accepted before it can be used by the
guest. Accepting happens via a protocol specific for the Virtual
Machine platform.

Accepting memory is costly and it makes VMM allocate memory for the
accepted guest physical address range. It's better to postpone memory
acceptance until memory is needed. It lowers boot time and reduces
memory overhead.

The kernel needs to know what memory has been accepted. Firmware
communicates this information via memory map: a new memory type --
EFI_UNACCEPTED_MEMORY -- indicates such memory.

Range-based tracking works fine for firmware, but it gets bulky for
the kernel: e820 has to be modified on every page acceptance. It leads
to table fragmentation, but there's a limited number of entries in the
e820 table

Another option is to mark such memory as usable in e820 and track if the
range has been accepted in a bitmap. One bit in the bitmap represents
2MiB in the address space: one 4k page is enough to track 64GiB or
physical address space.

In the worst-case scenario -- a huge hole in the middle of the
address space -- It needs 256MiB to handle 4PiB of the address
space.

Any unaccepted memory that is not aligned to 2M gets accepted upfront.

The approach lowers boot time substantially. Boot to shell is ~2.5x
faster for 4G TDX VM and ~4x faster for 64G.

TDX-specific code isolated from the core of unaccepted memory support. It
supposed to help to plug-in different implementation of unaccepted memory
such as SEV-SNP.

-- Fragmentation study --

Vlastimil and Mel were concern about effect of unaccepted memory on
fragmentation prevention measures in page allocator. I tried to evaluate
it, but it is tricky. As suggested I tried to run multiple parallel kernel
builds and follow how often kmem:mm_page_alloc_extfrag gets hit.

See results in the v9 of the patchset[1][2]

[1] https://lore.kernel.org/all/20230330114956.20342-1-kirill.shutemov@linux.intel.com
[2] https://lore.kernel.org/all/20230416191940.ex7ao43pmrjhru2p@box.shutemov.name

--

The tree can be found here:

https://github.com/intel/tdx.git guest-unaccepted-memory

v14:
 - Fix error handling in arch_accept_memory() (Tom);
 - Address Borislav's feedback:
   + code restructure;
   + added/adjusted comments;
v13:
 - Fix few boot issues discovered by 0day;
 - Simplify tdx_accept_memory(): no need in MAP_GPA hypercall;
 - Update commit message for the first patch;
 - Add Reviewed-bys from Tom and Ard;
v12:
 - Re-initialize 'unaccepted_table' variable from decompressor to cover some
   boot scenarios;
 - Add missing memblock_reserve() for the unaccepted memory configuration
   table (Mika);
 - Add efi.unaccepted into efi_tables (Tom);
 - Do not build tdx-shared.o for !TDX (Tom);
 - Typo fix (Liam)
 - Whitespace fix;
 - Reviewed-bys from Liam, Tom and Ard;
v11:
 - Restructure the code to make it less x86-specific (suggested by Ard):
   + use EFI configuration table instead of zero-page to pass down bitmap;
   + do not imply 1bit == 2M in bitmap;
   + move bulk of the code under driver/firmware/efi;
 - The bitmap only covers unaccpeted memory now. All memory that is not covered
   by the bitmap assumed accepted;
 - Reviewed-by from Ard;
v10:
 - Restructure code around zones_with_unaccepted_pages static brach to avoid
   unnecessary function calls (Suggested by Vlastimil);
 - Drop mentions of PageUnaccepted();
 - Drop patches that add fake unaccepted memory support and sysfs handle to
   accept memory manually;
 - Add Reviewed-by from Vlastimil;
v9:
 - Accept memory up to high watermark when kernel runs out of free memory;
 - Treat unaccepted memory as unusable in __zone_watermark_unusable_free();
 - Per-zone unaccepted memory accounting;
 - All pages on unaccepted list are MAX_ORDER now;
 - accept_memory=eager in cmdline to pre-accept memory during the boot;
 - Implement fake unaccepted memory;
 - Sysfs handle to accept memory manually;
 - Drop PageUnaccepted();
 - Rename unaccepted_pages static key to zones_with_unaccepted_pages;
v8:
 - Rewrite core-mm support for unaccepted memory (patch 02/14);
 - s/UnacceptedPages/Unaccepted/ in meminfo;
 - Drop arch/x86/boot/compressed/compiler.h;
 - Fix build errors;
 - Adjust commit messages and comments;
 - Reviewed-bys from Dave and Borislav;
 - Rebased to tip/master.
v7:
 - Rework meminfo counter to use PageUnaccepted() and move to generic code;
 - Fix range_contains_unaccepted_memory() on machines without unaccepted memory;
 - Add Reviewed-by from David;
v6:
 - Fix load_unaligned_zeropad() on machine with unaccepted memory;
 - Clear PageUnaccepted() on merged pages, leaving it only on head;
 - Clarify error handling in allocate_e820();
 - Fix build with CONFIG_UNACCEPTED_MEMORY=y, but without TDX;
 - Disable kexec at boottime instead of build conflict;
 - Rebased to tip/master;
 - Spelling fixes;
 - Add Reviewed-by from Mike and David;
v5:
 - Updates comments and commit messages;
   + Explain options for unaccepted memory handling;
 - Expose amount of unaccepted memory in /proc/meminfo
 - Adjust check in page_expected_state();
 - Fix error code handling in allocate_e820();
 - Centralize __pa()/__va() definitions in the boot stub;
 - Avoid includes from the main kernel in the boot stub;
 - Use an existing hole in boot_param for unaccepted_memory, instead of adding
   to the end of the structure;
 - Extract allocate_unaccepted_memory() form allocate_e820();
 - Complain if there's unaccepted memory, but kernel does not support it;
 - Fix vmstat counter;
 - Split up few preparatory patches;
 - Random readability adjustments;
v4:
 - PageBuddyUnaccepted() -> PageUnaccepted;
 - Use separate page_type, not shared with offline;
 - Rework interface between core-mm and arch code;
 - Adjust commit messages;
 - Ack from Mike;
Kirill A. Shutemov (9):
  mm: Add support for unaccepted memory
  efi/x86: Get full memory map in allocate_e820()
  efi/libstub: Implement support for unaccepted memory
  x86/boot/compressed: Handle unaccepted memory
  efi: Add unaccepted memory support
  efi/unaccepted: Avoid load_unaligned_zeropad() stepping into
    unaccepted memory
  x86/tdx: Make _tdx_hypercall() and __tdx_module_call() available in
    boot stub
  x86/tdx: Refactor try_accept_one()
  x86/tdx: Add unaccepted memory support

 arch/x86/Kconfig                              |   2 +
 arch/x86/boot/compressed/Makefile             |   3 +-
 arch/x86/boot/compressed/efi.h                |  10 +
 arch/x86/boot/compressed/error.c              |  19 ++
 arch/x86/boot/compressed/error.h              |   1 +
 arch/x86/boot/compressed/kaslr.c              |  40 +++-
 arch/x86/boot/compressed/mem.c                |  83 +++++++
 arch/x86/boot/compressed/misc.c               |   6 +
 arch/x86/boot/compressed/misc.h               |  10 +
 arch/x86/boot/compressed/tdx-shared.c         |   2 +
 arch/x86/coco/tdx/Makefile                    |   2 +-
 arch/x86/coco/tdx/tdx-shared.c                |  71 ++++++
 arch/x86/coco/tdx/tdx.c                       | 102 +-------
 arch/x86/include/asm/efi.h                    |   2 +
 arch/x86/include/asm/shared/tdx.h             |  53 +++++
 arch/x86/include/asm/tdx.h                    |  19 --
 arch/x86/include/asm/unaccepted_memory.h      |  24 ++
 arch/x86/platform/efi/efi.c                   |   3 +
 drivers/base/node.c                           |   7 +
 drivers/firmware/efi/Kconfig                  |  14 ++
 drivers/firmware/efi/Makefile                 |   1 +
 drivers/firmware/efi/efi.c                    |  26 ++
 drivers/firmware/efi/libstub/Makefile         |   2 +
 drivers/firmware/efi/libstub/bitmap.c         |  41 ++++
 drivers/firmware/efi/libstub/efistub.h        |   6 +
 drivers/firmware/efi/libstub/find.c           |  43 ++++
 .../firmware/efi/libstub/unaccepted_memory.c  | 222 ++++++++++++++++++
 drivers/firmware/efi/libstub/x86-stub.c       |  39 +--
 drivers/firmware/efi/unaccepted_memory.c      | 147 ++++++++++++
 fs/proc/meminfo.c                             |   5 +
 include/linux/efi.h                           |  13 +-
 include/linux/mm.h                            |  19 ++
 include/linux/mmzone.h                        |   8 +
 mm/memblock.c                                 |   9 +
 mm/mm_init.c                                  |   7 +
 mm/page_alloc.c                               | 173 ++++++++++++++
 mm/vmstat.c                                   |   3 +
 37 files changed, 1089 insertions(+), 148 deletions(-)
 create mode 100644 arch/x86/boot/compressed/mem.c
 create mode 100644 arch/x86/boot/compressed/tdx-shared.c
 create mode 100644 arch/x86/coco/tdx/tdx-shared.c
 create mode 100644 arch/x86/include/asm/unaccepted_memory.h
 create mode 100644 drivers/firmware/efi/libstub/bitmap.c
 create mode 100644 drivers/firmware/efi/libstub/find.c
 create mode 100644 drivers/firmware/efi/libstub/unaccepted_memory.c
 create mode 100644 drivers/firmware/efi/unaccepted_memory.c

-- 
2.39.3

[PATCHv14 1/9] mm: Add support for unaccepted memory

[Kirill A. Shutemov]

UEFI Specification version 2.9 introduces the concept of memory
acceptance. Some Virtual Machine platforms, such as Intel TDX or AMD
SEV-SNP, require memory to be accepted before it can be used by the
guest. Accepting happens via a protocol specific to the Virtual Machine
platform.

There are several ways kernel can deal with unaccepted memory:

 1. Accept all the memory during the boot. It is easy to implement and
    it doesn't have runtime cost once the system is booted. The downside
    is very long boot time.

    Accept can be parallelized to multiple CPUs to keep it manageable
    (i.e. via DEFERRED_STRUCT_PAGE_INIT), but it tends to saturate
    memory bandwidth and does not scale beyond the point.

 2. Accept a block of memory on the first use. It requires more
    infrastructure and changes in page allocator to make it work, but
    it provides good boot time.

    On-demand memory accept means latency spikes every time kernel steps
    onto a new memory block. The spikes will go away once workload data
    set size gets stabilized or all memory gets accepted.

 3. Accept all memory in background. Introduce a thread (or multiple)
    that gets memory accepted proactively. It will minimize time the
    system experience latency spikes on memory allocation while keeping
    low boot time.

    This approach cannot function on its own. It is an extension of #2:
    background memory acceptance requires functional scheduler, but the
    page allocator may need to tap into unaccepted memory before that.

    The downside of the approach is that these threads also steal CPU
    cycles and memory bandwidth from the user's workload and may hurt
    user experience.

Implement #1 and #2 for now. #2 is the default. Some workloads may want
to use #1 with accept_memory=eager in kernel command line. #3 can be
implemented later based on user's demands.

Support of unaccepted memory requires a few changes in core-mm code:

  - memblock accepts memory on allocation. It serves early boot memory
    allocations and doesn't limit them to pre-accepted pool of memory.

  - page allocator accepts memory on the first allocation of the page.
    When kernel runs out of accepted memory, it accepts memory until the
    high watermark is reached. It helps to minimize fragmentation.

EFI code will provide two helpers if the platform supports unaccepted
memory:

 - accept_memory() makes a range of physical addresses accepted.

 - range_contains_unaccepted_memory() checks anything within the range
   of physical addresses requires acceptance.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Mike Rapoport <rppt@linux.ibm.com>	# memblock
Reviewed-by: Vlastimil Babka <vbabka@suse.cz>
---
 drivers/base/node.c    |   7 ++
 fs/proc/meminfo.c      |   5 ++
 include/linux/mm.h     |  19 +++++
 include/linux/mmzone.h |   8 ++
 mm/memblock.c          |   9 +++
 mm/mm_init.c           |   7 ++
 mm/page_alloc.c        | 173 +++++++++++++++++++++++++++++++++++++++++
 mm/vmstat.c            |   3 +
 8 files changed, 231 insertions(+)

diff --git a/drivers/base/node.c b/drivers/base/node.c
index b46db17124f3..655975946ef6 100644
--- a/drivers/base/node.c
+++ b/drivers/base/node.c
@@ -448,6 +448,9 @@ static ssize_t node_read_meminfo(struct device *dev,
 			     "Node %d ShmemPmdMapped: %8lu kB\n"
 			     "Node %d FileHugePages: %8lu kB\n"
 			     "Node %d FilePmdMapped: %8lu kB\n"
+#endif
+#ifdef CONFIG_UNACCEPTED_MEMORY
+			     "Node %d Unaccepted:     %8lu kB\n"
 #endif
 			     ,
 			     nid, K(node_page_state(pgdat, NR_FILE_DIRTY)),
@@ -477,6 +480,10 @@ static ssize_t node_read_meminfo(struct device *dev,
 			     nid, K(node_page_state(pgdat, NR_SHMEM_PMDMAPPED)),
 			     nid, K(node_page_state(pgdat, NR_FILE_THPS)),
 			     nid, K(node_page_state(pgdat, NR_FILE_PMDMAPPED))
+#endif
+#ifdef CONFIG_UNACCEPTED_MEMORY
+			     ,
+			     nid, K(sum_zone_node_page_state(nid, NR_UNACCEPTED))
 #endif
 			    );
 	len += hugetlb_report_node_meminfo(buf, len, nid);
diff --git a/fs/proc/meminfo.c b/fs/proc/meminfo.c
index b43d0bd42762..8dca4d6d96c7 100644
--- a/fs/proc/meminfo.c
+++ b/fs/proc/meminfo.c
@@ -168,6 +168,11 @@ static int meminfo_proc_show(struct seq_file *m, void *v)
 		    global_zone_page_state(NR_FREE_CMA_PAGES));
 #endif
 
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	show_val_kb(m, "Unaccepted:     ",
+		    global_zone_page_state(NR_UNACCEPTED));
+#endif
+
 	hugetlb_report_meminfo(m);
 
 	arch_report_meminfo(m);
diff --git a/include/linux/mm.h b/include/linux/mm.h
index 27ce77080c79..d9174d464348 100644
--- a/include/linux/mm.h
+++ b/include/linux/mm.h
@@ -3816,4 +3816,23 @@ madvise_set_anon_name(struct mm_struct *mm, unsigned long start,
 }
 #endif
 
+#ifdef CONFIG_UNACCEPTED_MEMORY
+
+bool range_contains_unaccepted_memory(phys_addr_t start, phys_addr_t end);
+void accept_memory(phys_addr_t start, phys_addr_t end);
+
+#else
+
+static inline bool range_contains_unaccepted_memory(phys_addr_t start,
+						    phys_addr_t end)
+{
+	return false;
+}
+
+static inline void accept_memory(phys_addr_t start, phys_addr_t end)
+{
+}
+
+#endif
+
 #endif /* _LINUX_MM_H */
diff --git a/include/linux/mmzone.h b/include/linux/mmzone.h
index a4889c9d4055..6c1c2fc13017 100644
--- a/include/linux/mmzone.h
+++ b/include/linux/mmzone.h
@@ -143,6 +143,9 @@ enum zone_stat_item {
 	NR_ZSPAGES,		/* allocated in zsmalloc */
 #endif
 	NR_FREE_CMA_PAGES,
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	NR_UNACCEPTED,
+#endif
 	NR_VM_ZONE_STAT_ITEMS };
 
 enum node_stat_item {
@@ -910,6 +913,11 @@ struct zone {
 	/* free areas of different sizes */
 	struct free_area	free_area[MAX_ORDER + 1];
 
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	/* Pages to be accepted. All pages on the list are MAX_ORDER */
+	struct list_head	unaccepted_pages;
+#endif
+
 	/* zone flags, see below */
 	unsigned long		flags;
 
diff --git a/mm/memblock.c b/mm/memblock.c
index 3feafea06ab2..50b921119600 100644
--- a/mm/memblock.c
+++ b/mm/memblock.c
@@ -1436,6 +1436,15 @@ phys_addr_t __init memblock_alloc_range_nid(phys_addr_t size,
 		 */
 		kmemleak_alloc_phys(found, size, 0);
 
+	/*
+	 * Some Virtual Machine platforms, such as Intel TDX or AMD SEV-SNP,
+	 * require memory to be accepted before it can be used by the
+	 * guest.
+	 *
+	 * Accept the memory of the allocated buffer.
+	 */
+	accept_memory(found, found + size);
+
 	return found;
 }
 
diff --git a/mm/mm_init.c b/mm/mm_init.c
index 7f7f9c677854..1cfc08e25f93 100644
--- a/mm/mm_init.c
+++ b/mm/mm_init.c
@@ -1375,6 +1375,10 @@ static void __meminit zone_init_free_lists(struct zone *zone)
 		INIT_LIST_HEAD(&zone->free_area[order].free_list[t]);
 		zone->free_area[order].nr_free = 0;
 	}
+
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	INIT_LIST_HEAD(&zone->unaccepted_pages);
+#endif
 }
 
 void __meminit init_currently_empty_zone(struct zone *zone,
@@ -1960,6 +1964,9 @@ static void __init deferred_free_range(unsigned long pfn,
 		return;
 	}
 
+	/* Accept chunks smaller than MAX_ORDER upfront */
+	accept_memory(PFN_PHYS(pfn), PFN_PHYS(pfn + nr_pages));
+
 	for (i = 0; i < nr_pages; i++, page++, pfn++) {
 		if (pageblock_aligned(pfn))
 			set_pageblock_migratetype(page, MIGRATE_MOVABLE);
diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index 47421bedc12b..d239fba3f31c 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -387,6 +387,12 @@ EXPORT_SYMBOL(nr_node_ids);
 EXPORT_SYMBOL(nr_online_nodes);
 #endif
 
+static bool page_contains_unaccepted(struct page *page, unsigned int order);
+static void accept_page(struct page *page, unsigned int order);
+static bool try_to_accept_memory(struct zone *zone, unsigned int order);
+static inline bool has_unaccepted_memory(void);
+static bool __free_unaccepted(struct page *page);
+
 int page_group_by_mobility_disabled __read_mostly;
 
 #ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT
@@ -1481,6 +1487,13 @@ void __free_pages_core(struct page *page, unsigned int order)
 
 	atomic_long_add(nr_pages, &page_zone(page)->managed_pages);
 
+	if (page_contains_unaccepted(page, order)) {
+		if (order == MAX_ORDER && __free_unaccepted(page))
+			return;
+
+		accept_page(page, order);
+	}
+
 	/*
 	 * Bypass PCP and place fresh pages right to the tail, primarily
 	 * relevant for memory onlining.
@@ -3159,6 +3172,9 @@ static inline long __zone_watermark_unusable_free(struct zone *z,
 	if (!(alloc_flags & ALLOC_CMA))
 		unusable_free += zone_page_state(z, NR_FREE_CMA_PAGES);
 #endif
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	unusable_free += zone_page_state(z, NR_UNACCEPTED);
+#endif
 
 	return unusable_free;
 }
@@ -3458,6 +3474,11 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 				       gfp_mask)) {
 			int ret;
 
+			if (has_unaccepted_memory()) {
+				if (try_to_accept_memory(zone, order))
+					goto try_this_zone;
+			}
+
 #ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT
 			/*
 			 * Watermark failed for this zone, but see if we can
@@ -3510,6 +3531,11 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 
 			return page;
 		} else {
+			if (has_unaccepted_memory()) {
+				if (try_to_accept_memory(zone, order))
+					goto try_this_zone;
+			}
+
 #ifdef CONFIG_DEFERRED_STRUCT_PAGE_INIT
 			/* Try again if zone has deferred pages */
 			if (deferred_pages_enabled()) {
@@ -7215,3 +7241,150 @@ bool has_managed_dma(void)
 	return false;
 }
 #endif /* CONFIG_ZONE_DMA */
+
+#ifdef CONFIG_UNACCEPTED_MEMORY
+
+/* Counts number of zones with unaccepted pages. */
+static DEFINE_STATIC_KEY_FALSE(zones_with_unaccepted_pages);
+
+static bool lazy_accept = true;
+
+static int __init accept_memory_parse(char *p)
+{
+	if (!strcmp(p, "lazy")) {
+		lazy_accept = true;
+		return 0;
+	} else if (!strcmp(p, "eager")) {
+		lazy_accept = false;
+		return 0;
+	} else {
+		return -EINVAL;
+	}
+}
+early_param("accept_memory", accept_memory_parse);
+
+static bool page_contains_unaccepted(struct page *page, unsigned int order)
+{
+	phys_addr_t start = page_to_phys(page);
+	phys_addr_t end = start + (PAGE_SIZE << order);
+
+	return range_contains_unaccepted_memory(start, end);
+}
+
+static void accept_page(struct page *page, unsigned int order)
+{
+	phys_addr_t start = page_to_phys(page);
+
+	accept_memory(start, start + (PAGE_SIZE << order));
+}
+
+static bool try_to_accept_memory_one(struct zone *zone)
+{
+	unsigned long flags;
+	struct page *page;
+	bool last;
+
+	if (list_empty(&zone->unaccepted_pages))
+		return false;
+
+	spin_lock_irqsave(&zone->lock, flags);
+	page = list_first_entry_or_null(&zone->unaccepted_pages,
+					struct page, lru);
+	if (!page) {
+		spin_unlock_irqrestore(&zone->lock, flags);
+		return false;
+	}
+
+	list_del(&page->lru);
+	last = list_empty(&zone->unaccepted_pages);
+
+	__mod_zone_freepage_state(zone, -MAX_ORDER_NR_PAGES, MIGRATE_MOVABLE);
+	__mod_zone_page_state(zone, NR_UNACCEPTED, -MAX_ORDER_NR_PAGES);
+	spin_unlock_irqrestore(&zone->lock, flags);
+
+	accept_page(page, MAX_ORDER);
+
+	__free_pages_ok(page, MAX_ORDER, FPI_TO_TAIL);
+
+	if (last)
+		static_branch_dec(&zones_with_unaccepted_pages);
+
+	return true;
+}
+
+static bool try_to_accept_memory(struct zone *zone, unsigned int order)
+{
+	long to_accept;
+	int ret = false;
+
+	/* How much to accept to get to high watermark? */
+	to_accept = high_wmark_pages(zone) -
+		    (zone_page_state(zone, NR_FREE_PAGES) -
+		    __zone_watermark_unusable_free(zone, order, 0));
+
+	/* Accept at least one page */
+	do {
+		if (!try_to_accept_memory_one(zone))
+			break;
+		ret = true;
+		to_accept -= MAX_ORDER_NR_PAGES;
+	} while (to_accept > 0);
+
+	return ret;
+}
+
+static inline bool has_unaccepted_memory(void)
+{
+	return static_branch_unlikely(&zones_with_unaccepted_pages);
+}
+
+static bool __free_unaccepted(struct page *page)
+{
+	struct zone *zone = page_zone(page);
+	unsigned long flags;
+	bool first = false;
+
+	if (!lazy_accept)
+		return false;
+
+	spin_lock_irqsave(&zone->lock, flags);
+	first = list_empty(&zone->unaccepted_pages);
+	list_add_tail(&page->lru, &zone->unaccepted_pages);
+	__mod_zone_freepage_state(zone, MAX_ORDER_NR_PAGES, MIGRATE_MOVABLE);
+	__mod_zone_page_state(zone, NR_UNACCEPTED, MAX_ORDER_NR_PAGES);
+	spin_unlock_irqrestore(&zone->lock, flags);
+
+	if (first)
+		static_branch_inc(&zones_with_unaccepted_pages);
+
+	return true;
+}
+
+#else
+
+static bool page_contains_unaccepted(struct page *page, unsigned int order)
+{
+	return false;
+}
+
+static void accept_page(struct page *page, unsigned int order)
+{
+}
+
+static bool try_to_accept_memory(struct zone *zone, unsigned int order)
+{
+	return false;
+}
+
+static inline bool has_unaccepted_memory(void)
+{
+	return false;
+}
+
+static bool __free_unaccepted(struct page *page)
+{
+	BUILD_BUG();
+	return false;
+}
+
+#endif /* CONFIG_UNACCEPTED_MEMORY */
diff --git a/mm/vmstat.c b/mm/vmstat.c
index c28046371b45..282349cabf01 100644
--- a/mm/vmstat.c
+++ b/mm/vmstat.c
@@ -1180,6 +1180,9 @@ const char * const vmstat_text[] = {
 	"nr_zspages",
 #endif
 	"nr_free_cma",
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	"nr_unaccepted",
+#endif
 
 	/* enum numa_stat_item counters */
 #ifdef CONFIG_NUMA
-- 
2.39.3

[PATCHv14 2/9] efi/x86: Get full memory map in allocate_e820()

[Kirill A. Shutemov]

Currently allocate_e820() is only interested in the size of map and size
of memory descriptor to determine how many e820 entries the kernel
needs.

UEFI Specification version 2.9 introduces a new memory type --
unaccepted memory. To track unaccepted memory kernel needs to allocate
a bitmap. The size of the bitmap is dependent on the maximum physical
address present in the system. A full memory map is required to find
the maximum address.

Modify allocate_e820() to get a full memory map.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Borislav Petkov <bp@suse.de>
Acked-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
---
 drivers/firmware/efi/libstub/x86-stub.c | 26 +++++++++++--------------
 1 file changed, 11 insertions(+), 15 deletions(-)

diff --git a/drivers/firmware/efi/libstub/x86-stub.c b/drivers/firmware/efi/libstub/x86-stub.c
index a0bfd31358ba..fff81843169c 100644
--- a/drivers/firmware/efi/libstub/x86-stub.c
+++ b/drivers/firmware/efi/libstub/x86-stub.c
@@ -681,28 +681,24 @@ static efi_status_t allocate_e820(struct boot_params *params,
 				  struct setup_data **e820ext,
 				  u32 *e820ext_size)
 {
-	unsigned long map_size, desc_size, map_key;
+	struct efi_boot_memmap *map;
 	efi_status_t status;
-	__u32 nr_desc, desc_version;
+	__u32 nr_desc;
 
-	/* Only need the size of the mem map and size of each mem descriptor */
-	map_size = 0;
-	status = efi_bs_call(get_memory_map, &map_size, NULL, &map_key,
-			     &desc_size, &desc_version);
-	if (status != EFI_BUFFER_TOO_SMALL)
-		return (status != EFI_SUCCESS) ? status : EFI_UNSUPPORTED;
+	status = efi_get_memory_map(&map, false);
+	if (status != EFI_SUCCESS)
+		return status;
 
-	nr_desc = map_size / desc_size + EFI_MMAP_NR_SLACK_SLOTS;
-
-	if (nr_desc > ARRAY_SIZE(params->e820_table)) {
-		u32 nr_e820ext = nr_desc - ARRAY_SIZE(params->e820_table);
+	nr_desc = map->map_size / map->desc_size;
+	if (nr_desc > ARRAY_SIZE(params->e820_table) - EFI_MMAP_NR_SLACK_SLOTS) {
+		u32 nr_e820ext = nr_desc - ARRAY_SIZE(params->e820_table) +
+			EFI_MMAP_NR_SLACK_SLOTS;
 
 		status = alloc_e820ext(nr_e820ext, e820ext, e820ext_size);
-		if (status != EFI_SUCCESS)
-			return status;
 	}
 
-	return EFI_SUCCESS;
+	efi_bs_call(free_pool, map);
+	return status;
 }
 
 struct exit_boot_struct {
-- 
2.39.3

[PATCHv14 3/9] efi/libstub: Implement support for unaccepted memory

[Kirill A. Shutemov]

UEFI Specification version 2.9 introduces the concept of memory
acceptance: Some Virtual Machine platforms, such as Intel TDX or AMD
SEV-SNP, requiring memory to be accepted before it can be used by the
guest. Accepting happens via a protocol specific for the Virtual
Machine platform.

Accepting memory is costly and it makes VMM allocate memory for the
accepted guest physical address range. It's better to postpone memory
acceptance until memory is needed. It lowers boot time and reduces
memory overhead.

The kernel needs to know what memory has been accepted. Firmware
communicates this information via memory map: a new memory type --
EFI_UNACCEPTED_MEMORY -- indicates such memory.

Range-based tracking works fine for firmware, but it gets bulky for
the kernel: e820 (or whatever the arch uses) has to be modified on every
page acceptance. It leads to table fragmentation and there's a limited
number of entries in the e820 table.

Another option is to mark such memory as usable in e820 and track if the
range has been accepted in a bitmap. One bit in the bitmap represents a
naturally aligned power-2-sized region of address space -- unit.

For x86, unit size is 2MiB: 4k of the bitmap is enough to track 64GiB or
physical address space.

In the worst-case scenario -- a huge hole in the middle of the
address space -- It needs 256MiB to handle 4PiB of the address
space.

Any unaccepted memory that is not aligned to unit_size gets accepted
upfront.

The bitmap is allocated and constructed in the EFI stub and passed down
to the kernel via EFI configuration table. allocate_e820() allocates the
bitmap if unaccepted memory is present, according to the size of
unaccepted region.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
---
 arch/x86/boot/compressed/Makefile             |   1 +
 arch/x86/boot/compressed/mem.c                |   9 +
 arch/x86/include/asm/efi.h                    |   2 +
 drivers/firmware/efi/Kconfig                  |  14 ++
 drivers/firmware/efi/efi.c                    |   1 +
 drivers/firmware/efi/libstub/Makefile         |   2 +
 drivers/firmware/efi/libstub/bitmap.c         |  41 ++++
 drivers/firmware/efi/libstub/efistub.h        |   6 +
 drivers/firmware/efi/libstub/find.c           |  43 ++++
 .../firmware/efi/libstub/unaccepted_memory.c  | 222 ++++++++++++++++++
 drivers/firmware/efi/libstub/x86-stub.c       |  13 +
 include/linux/efi.h                           |  12 +-
 12 files changed, 365 insertions(+), 1 deletion(-)
 create mode 100644 arch/x86/boot/compressed/mem.c
 create mode 100644 drivers/firmware/efi/libstub/bitmap.c
 create mode 100644 drivers/firmware/efi/libstub/find.c
 create mode 100644 drivers/firmware/efi/libstub/unaccepted_memory.c

diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
index 6b6cfe607bdb..cc4978123c30 100644
--- a/arch/x86/boot/compressed/Makefile
+++ b/arch/x86/boot/compressed/Makefile
@@ -107,6 +107,7 @@ endif
 
 vmlinux-objs-$(CONFIG_ACPI) += $(obj)/acpi.o
 vmlinux-objs-$(CONFIG_INTEL_TDX_GUEST) += $(obj)/tdx.o $(obj)/tdcall.o
+vmlinux-objs-$(CONFIG_UNACCEPTED_MEMORY) += $(obj)/mem.o
 
 vmlinux-objs-$(CONFIG_EFI) += $(obj)/efi.o
 vmlinux-objs-$(CONFIG_EFI_MIXED) += $(obj)/efi_mixed.o
diff --git a/arch/x86/boot/compressed/mem.c b/arch/x86/boot/compressed/mem.c
new file mode 100644
index 000000000000..67594fcb11d9
--- /dev/null
+++ b/arch/x86/boot/compressed/mem.c
@@ -0,0 +1,9 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include "error.h"
+
+void arch_accept_memory(phys_addr_t start, phys_addr_t end)
+{
+	/* Platform-specific memory-acceptance call goes here */
+	error("Cannot accept memory");
+}
diff --git a/arch/x86/include/asm/efi.h b/arch/x86/include/asm/efi.h
index 419280d263d2..8b4be7cecdb8 100644
--- a/arch/x86/include/asm/efi.h
+++ b/arch/x86/include/asm/efi.h
@@ -31,6 +31,8 @@ extern unsigned long efi_mixed_mode_stack_pa;
 
 #define ARCH_EFI_IRQ_FLAGS_MASK	X86_EFLAGS_IF
 
+#define EFI_UNACCEPTED_UNIT_SIZE PMD_SIZE
+
 /*
  * The EFI services are called through variadic functions in many cases. These
  * functions are implemented in assembler and support only a fixed number of
diff --git a/drivers/firmware/efi/Kconfig b/drivers/firmware/efi/Kconfig
index 043ca31c114e..231f1c70d1db 100644
--- a/drivers/firmware/efi/Kconfig
+++ b/drivers/firmware/efi/Kconfig
@@ -269,6 +269,20 @@ config EFI_COCO_SECRET
 	  virt/coco/efi_secret module to access the secrets, which in turn
 	  allows userspace programs to access the injected secrets.
 
+config UNACCEPTED_MEMORY
+	bool
+	depends on EFI_STUB
+	help
+	   Some Virtual Machine platforms, such as Intel TDX, require
+	   some memory to be "accepted" by the guest before it can be used.
+	   This mechanism helps prevent malicious hosts from making changes
+	   to guest memory.
+
+	   UEFI specification v2.9 introduced EFI_UNACCEPTED_MEMORY memory type.
+
+	   This option adds support for unaccepted memory and makes such memory
+	   usable by the kernel.
+
 config EFI_EMBEDDED_FIRMWARE
 	bool
 	select CRYPTO_LIB_SHA256
diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c
index abeff7dc0b58..7dce06e419c5 100644
--- a/drivers/firmware/efi/efi.c
+++ b/drivers/firmware/efi/efi.c
@@ -843,6 +843,7 @@ static __initdata char memory_type_name[][13] = {
 	"MMIO Port",
 	"PAL Code",
 	"Persistent",
+	"Unaccepted",
 };
 
 char * __init efi_md_typeattr_format(char *buf, size_t size,
diff --git a/drivers/firmware/efi/libstub/Makefile b/drivers/firmware/efi/libstub/Makefile
index 3abb2b357482..16d64a34d1e1 100644
--- a/drivers/firmware/efi/libstub/Makefile
+++ b/drivers/firmware/efi/libstub/Makefile
@@ -96,6 +96,8 @@ CFLAGS_arm32-stub.o		:= -DTEXT_OFFSET=$(TEXT_OFFSET)
 zboot-obj-$(CONFIG_RISCV)	:= lib-clz_ctz.o lib-ashldi3.o
 lib-$(CONFIG_EFI_ZBOOT)		+= zboot.o $(zboot-obj-y)
 
+lib-$(CONFIG_UNACCEPTED_MEMORY) += unaccepted_memory.o bitmap.o find.o
+
 extra-y				:= $(lib-y)
 lib-y				:= $(patsubst %.o,%.stub.o,$(lib-y))
 
diff --git a/drivers/firmware/efi/libstub/bitmap.c b/drivers/firmware/efi/libstub/bitmap.c
new file mode 100644
index 000000000000..5c9bba0d549b
--- /dev/null
+++ b/drivers/firmware/efi/libstub/bitmap.c
@@ -0,0 +1,41 @@
+#include <linux/bitmap.h>
+
+void __bitmap_set(unsigned long *map, unsigned int start, int len)
+{
+	unsigned long *p = map + BIT_WORD(start);
+	const unsigned int size = start + len;
+	int bits_to_set = BITS_PER_LONG - (start % BITS_PER_LONG);
+	unsigned long mask_to_set = BITMAP_FIRST_WORD_MASK(start);
+
+	while (len - bits_to_set >= 0) {
+		*p |= mask_to_set;
+		len -= bits_to_set;
+		bits_to_set = BITS_PER_LONG;
+		mask_to_set = ~0UL;
+		p++;
+	}
+	if (len) {
+		mask_to_set &= BITMAP_LAST_WORD_MASK(size);
+		*p |= mask_to_set;
+	}
+}
+
+void __bitmap_clear(unsigned long *map, unsigned int start, int len)
+{
+	unsigned long *p = map + BIT_WORD(start);
+	const unsigned int size = start + len;
+	int bits_to_clear = BITS_PER_LONG - (start % BITS_PER_LONG);
+	unsigned long mask_to_clear = BITMAP_FIRST_WORD_MASK(start);
+
+	while (len - bits_to_clear >= 0) {
+		*p &= ~mask_to_clear;
+		len -= bits_to_clear;
+		bits_to_clear = BITS_PER_LONG;
+		mask_to_clear = ~0UL;
+		p++;
+	}
+	if (len) {
+		mask_to_clear &= BITMAP_LAST_WORD_MASK(size);
+		*p &= ~mask_to_clear;
+	}
+}
diff --git a/drivers/firmware/efi/libstub/efistub.h b/drivers/firmware/efi/libstub/efistub.h
index 67d5a20802e0..8659a01664b8 100644
--- a/drivers/firmware/efi/libstub/efistub.h
+++ b/drivers/firmware/efi/libstub/efistub.h
@@ -1133,4 +1133,10 @@ const u8 *__efi_get_smbios_string(const struct efi_smbios_record *record,
 void efi_remap_image(unsigned long image_base, unsigned alloc_size,
 		     unsigned long code_size);
 
+efi_status_t allocate_unaccepted_bitmap(__u32 nr_desc,
+					struct efi_boot_memmap *map);
+void process_unaccepted_memory(u64 start, u64 end);
+void accept_memory(phys_addr_t start, phys_addr_t end);
+void arch_accept_memory(phys_addr_t start, phys_addr_t end);
+
 #endif
diff --git a/drivers/firmware/efi/libstub/find.c b/drivers/firmware/efi/libstub/find.c
new file mode 100644
index 000000000000..4e7740d28987
--- /dev/null
+++ b/drivers/firmware/efi/libstub/find.c
@@ -0,0 +1,43 @@
+// SPDX-License-Identifier: GPL-2.0-only
+#include <linux/bitmap.h>
+#include <linux/math.h>
+#include <linux/minmax.h>
+
+/*
+ * Common helper for find_next_bit() function family
+ * @FETCH: The expression that fetches and pre-processes each word of bitmap(s)
+ * @MUNGE: The expression that post-processes a word containing found bit (may be empty)
+ * @size: The bitmap size in bits
+ * @start: The bitnumber to start searching at
+ */
+#define FIND_NEXT_BIT(FETCH, MUNGE, size, start)				\
+({										\
+	unsigned long mask, idx, tmp, sz = (size), __start = (start);		\
+										\
+	if (unlikely(__start >= sz))						\
+		goto out;							\
+										\
+	mask = MUNGE(BITMAP_FIRST_WORD_MASK(__start));				\
+	idx = __start / BITS_PER_LONG;						\
+										\
+	for (tmp = (FETCH) & mask; !tmp; tmp = (FETCH)) {			\
+		if ((idx + 1) * BITS_PER_LONG >= sz)				\
+			goto out;						\
+		idx++;								\
+	}									\
+										\
+	sz = min(idx * BITS_PER_LONG + __ffs(MUNGE(tmp)), sz);			\
+out:										\
+	sz;									\
+})
+
+unsigned long _find_next_bit(const unsigned long *addr, unsigned long nbits, unsigned long start)
+{
+	return FIND_NEXT_BIT(addr[idx], /* nop */, nbits, start);
+}
+
+unsigned long _find_next_zero_bit(const unsigned long *addr, unsigned long nbits,
+					 unsigned long start)
+{
+	return FIND_NEXT_BIT(~addr[idx], /* nop */, nbits, start);
+}
diff --git a/drivers/firmware/efi/libstub/unaccepted_memory.c b/drivers/firmware/efi/libstub/unaccepted_memory.c
new file mode 100644
index 000000000000..ca61f4733ea5
--- /dev/null
+++ b/drivers/firmware/efi/libstub/unaccepted_memory.c
@@ -0,0 +1,222 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/efi.h>
+#include <asm/efi.h>
+#include "efistub.h"
+
+struct efi_unaccepted_memory *unaccepted_table;
+
+efi_status_t allocate_unaccepted_bitmap(__u32 nr_desc,
+					struct efi_boot_memmap *map)
+{
+	efi_guid_t unaccepted_table_guid = LINUX_EFI_UNACCEPTED_MEM_TABLE_GUID;
+	u64 unaccepted_start = ULLONG_MAX, unaccepted_end = 0, bitmap_size;
+	efi_status_t status;
+	int i;
+
+	/* Check if the table is already installed */
+	unaccepted_table = get_efi_config_table(unaccepted_table_guid);
+	if (unaccepted_table) {
+		if (unaccepted_table->version != 1) {
+			efi_err("Unknown version of unaccepted memory table\n");
+			return EFI_UNSUPPORTED;
+		}
+		return EFI_SUCCESS;
+	}
+
+	/* Check if there's any unaccepted memory and find the max address */
+	for (i = 0; i < nr_desc; i++) {
+		efi_memory_desc_t *d;
+		unsigned long m = (unsigned long)map->map;
+
+		d = efi_early_memdesc_ptr(m, map->desc_size, i);
+		if (d->type != EFI_UNACCEPTED_MEMORY)
+			continue;
+
+		unaccepted_start = min(unaccepted_start, d->phys_addr);
+		unaccepted_end = max(unaccepted_end,
+				     d->phys_addr + d->num_pages * PAGE_SIZE);
+	}
+
+	if (unaccepted_start == ULLONG_MAX)
+		return EFI_SUCCESS;
+
+	unaccepted_start = round_down(unaccepted_start,
+				      EFI_UNACCEPTED_UNIT_SIZE);
+	unaccepted_end = round_up(unaccepted_end, EFI_UNACCEPTED_UNIT_SIZE);
+
+	/*
+	 * If unaccepted memory is present, allocate a bitmap to track what
+	 * memory has to be accepted before access.
+	 *
+	 * One bit in the bitmap represents 2MiB in the address space:
+	 * A 4k bitmap can track 64GiB of physical address space.
+	 *
+	 * In the worst case scenario -- a huge hole in the middle of the
+	 * address space -- It needs 256MiB to handle 4PiB of the address
+	 * space.
+	 *
+	 * The bitmap will be populated in setup_e820() according to the memory
+	 * map after efi_exit_boot_services().
+	 */
+	bitmap_size = DIV_ROUND_UP(unaccepted_end - unaccepted_start,
+				   EFI_UNACCEPTED_UNIT_SIZE * BITS_PER_BYTE);
+
+	status = efi_bs_call(allocate_pool, EFI_LOADER_DATA,
+			     sizeof(*unaccepted_table) + bitmap_size,
+			     (void **)&unaccepted_table);
+	if (status != EFI_SUCCESS) {
+		efi_err("Failed to allocate unaccepted memory config table\n");
+		return status;
+	}
+
+	unaccepted_table->version = 1;
+	unaccepted_table->unit_size = EFI_UNACCEPTED_UNIT_SIZE;
+	unaccepted_table->phys_base = unaccepted_start;
+	unaccepted_table->size = bitmap_size;
+	memset(unaccepted_table->bitmap, 0, bitmap_size);
+
+	status = efi_bs_call(install_configuration_table,
+			     &unaccepted_table_guid, unaccepted_table);
+	if (status != EFI_SUCCESS) {
+		efi_bs_call(free_pool, unaccepted_table);
+		efi_err("Failed to install unaccepted memory config table!\n");
+	}
+
+	return status;
+}
+
+/*
+ * The accepted memory bitmap only works at unit_size granularity.  Take
+ * unaligned start/end addresses and either:
+ *  1. Accepts the memory immediately and in its entirety
+ *  2. Accepts unaligned parts, and marks *some* aligned part unaccepted
+ *
+ * The function will never reach the bitmap_set() with zero bits to set.
+ */
+void process_unaccepted_memory(u64 start, u64 end)
+{
+	u64 unit_size = unaccepted_table->unit_size;
+	u64 unit_mask = unaccepted_table->unit_size - 1;
+	u64 bitmap_size = unaccepted_table->size;
+
+	/*
+	 * Ensure that at least one bit will be set in the bitmap by
+	 * immediately accepting all regions under 2*unit_size.  This is
+	 * imprecise and may immediately accept some areas that could
+	 * have been represented in the bitmap.  But, results in simpler
+	 * code below
+	 *
+	 * Consider case like this (assuming unit_size == 2MB):
+	 *
+	 * | 4k | 2044k |    2048k   |
+	 * ^ 0x0        ^ 2MB        ^ 4MB
+	 *
+	 * Only the first 4k has been accepted. The 0MB->2MB region can not be
+	 * represented in the bitmap. The 2MB->4MB region can be represented in
+	 * the bitmap. But, the 0MB->4MB region is <2*unit_size and will be
+	 * immediately accepted in its entirety.
+	 */
+	if (end - start < 2 * unit_size) {
+		arch_accept_memory(start, end);
+		return;
+	}
+
+	/*
+	 * No matter how the start and end are aligned, at least one unaccepted
+	 * unit_size area will remain to be marked in the bitmap.
+	 */
+
+	/* Immediately accept a <unit_size piece at the start: */
+	if (start & unit_mask) {
+		arch_accept_memory(start, round_up(start, unit_size));
+		start = round_up(start, unit_size);
+	}
+
+	/* Immediately accept a <unit_size piece at the end: */
+	if (end & unit_mask) {
+		arch_accept_memory(round_down(end, unit_size), end);
+		end = round_down(end, unit_size);
+	}
+
+	/*
+	 * Accept part of the range that before phys_base and cannot be recorded
+	 * into the bitmap.
+	 */
+	if (start < unaccepted_table->phys_base) {
+		arch_accept_memory(start,
+				   min(unaccepted_table->phys_base, end));
+		start = unaccepted_table->phys_base;
+	}
+
+	/* Nothing to record */
+	if (end < unaccepted_table->phys_base)
+		return;
+
+	/* Translate to offsets from the beginning of the bitmap */
+	start -= unaccepted_table->phys_base;
+	end -= unaccepted_table->phys_base;
+
+	/* Accept memory that doesn't fit into bitmap */
+	if (end > bitmap_size * unit_size * BITS_PER_BYTE) {
+		unsigned long phys_start, phys_end;
+
+		phys_start = bitmap_size * unit_size * BITS_PER_BYTE +
+			     unaccepted_table->phys_base;
+		phys_end = end + unaccepted_table->phys_base;
+
+		arch_accept_memory(phys_start, phys_end);
+		end = bitmap_size * unit_size * BITS_PER_BYTE;
+	}
+
+	/*
+	 * 'start' and 'end' are now both unit_size-aligned.
+	 * Record the range as being unaccepted:
+	 */
+	bitmap_set(unaccepted_table->bitmap,
+		   start / unit_size, (end - start) / unit_size);
+}
+
+void accept_memory(phys_addr_t start, phys_addr_t end)
+{
+	unsigned long range_start, range_end;
+	unsigned long bitmap_size;
+	u64 unit_size;
+
+	if (!unaccepted_table)
+		return;
+
+	unit_size = unaccepted_table->unit_size;
+
+	/*
+	 * Only care for the part of the range that is represented
+	 * in the bitmap.
+	 */
+	if (start < unaccepted_table->phys_base)
+		start = unaccepted_table->phys_base;
+	if (end < unaccepted_table->phys_base)
+		return;
+
+	/* Translate to offsets from the beginning of the bitmap */
+	start -= unaccepted_table->phys_base;
+	end -= unaccepted_table->phys_base;
+
+	/* Make sure not to overrun the bitmap */
+	if (end > unaccepted_table->size * unit_size * BITS_PER_BYTE)
+		end = unaccepted_table->size * unit_size * BITS_PER_BYTE;
+
+	range_start = start / unit_size;
+	bitmap_size = DIV_ROUND_UP(end, unit_size);
+
+	for_each_set_bitrange_from(range_start, range_end,
+				   unaccepted_table->bitmap, bitmap_size) {
+		unsigned long phys_start, phys_end;
+
+		phys_start = range_start * unit_size + unaccepted_table->phys_base;
+		phys_end = range_end * unit_size + unaccepted_table->phys_base;
+
+		arch_accept_memory(phys_start, phys_end);
+		bitmap_clear(unaccepted_table->bitmap,
+			     range_start, range_end - range_start);
+	}
+}
diff --git a/drivers/firmware/efi/libstub/x86-stub.c b/drivers/firmware/efi/libstub/x86-stub.c
index fff81843169c..8d17cee8b98e 100644
--- a/drivers/firmware/efi/libstub/x86-stub.c
+++ b/drivers/firmware/efi/libstub/x86-stub.c
@@ -613,6 +613,16 @@ setup_e820(struct boot_params *params, struct setup_data *e820ext, u32 e820ext_s
 			e820_type = E820_TYPE_PMEM;
 			break;
 
+		case EFI_UNACCEPTED_MEMORY:
+			if (!IS_ENABLED(CONFIG_UNACCEPTED_MEMORY)) {
+				efi_warn_once(
+"The system has unaccepted memory,  but kernel does not support it\nConsider enabling CONFIG_UNACCEPTED_MEMORY\n");
+				continue;
+			}
+			e820_type = E820_TYPE_RAM;
+			process_unaccepted_memory(d->phys_addr,
+						  d->phys_addr + PAGE_SIZE * d->num_pages);
+			break;
 		default:
 			continue;
 		}
@@ -697,6 +707,9 @@ static efi_status_t allocate_e820(struct boot_params *params,
 		status = alloc_e820ext(nr_e820ext, e820ext, e820ext_size);
 	}
 
+	if (IS_ENABLED(CONFIG_UNACCEPTED_MEMORY) && status == EFI_SUCCESS)
+		status = allocate_unaccepted_bitmap(nr_desc, map);
+
 	efi_bs_call(free_pool, map);
 	return status;
 }
diff --git a/include/linux/efi.h b/include/linux/efi.h
index 7aa62c92185f..29cc622910da 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -108,7 +108,8 @@ typedef	struct {
 #define EFI_MEMORY_MAPPED_IO_PORT_SPACE	12
 #define EFI_PAL_CODE			13
 #define EFI_PERSISTENT_MEMORY		14
-#define EFI_MAX_MEMORY_TYPE		15
+#define EFI_UNACCEPTED_MEMORY		15
+#define EFI_MAX_MEMORY_TYPE		16
 
 /* Attribute values: */
 #define EFI_MEMORY_UC		((u64)0x0000000000000001ULL)	/* uncached */
@@ -417,6 +418,7 @@ void efi_native_runtime_setup(void);
 #define LINUX_EFI_MOK_VARIABLE_TABLE_GUID	EFI_GUID(0xc451ed2b, 0x9694, 0x45d3,  0xba, 0xba, 0xed, 0x9f, 0x89, 0x88, 0xa3, 0x89)
 #define LINUX_EFI_COCO_SECRET_AREA_GUID		EFI_GUID(0xadf956ad, 0xe98c, 0x484c,  0xae, 0x11, 0xb5, 0x1c, 0x7d, 0x33, 0x64, 0x47)
 #define LINUX_EFI_BOOT_MEMMAP_GUID		EFI_GUID(0x800f683f, 0xd08b, 0x423a,  0xa2, 0x93, 0x96, 0x5c, 0x3c, 0x6f, 0xe2, 0xb4)
+#define LINUX_EFI_UNACCEPTED_MEM_TABLE_GUID	EFI_GUID(0xd5d1de3c, 0x105c, 0x44f9,  0x9e, 0xa9, 0xbc, 0xef, 0x98, 0x12, 0x00, 0x31)
 
 #define RISCV_EFI_BOOT_PROTOCOL_GUID		EFI_GUID(0xccd15fec, 0x6f73, 0x4eec,  0x83, 0x95, 0x3e, 0x69, 0xe4, 0xb9, 0x40, 0xbf)
 
@@ -534,6 +536,14 @@ struct efi_boot_memmap {
 	efi_memory_desc_t	map[];
 };
 
+struct efi_unaccepted_memory {
+	u32 version;
+	u32 unit_size;
+	u64 phys_base;
+	u64 size;
+	unsigned long bitmap[];
+};
+
 /*
  * Architecture independent structure for describing a memory map for the
  * benefit of efi_memmap_init_early(), and for passing context between
-- 
2.39.3

[PATCHv14 4/9] x86/boot/compressed: Handle unaccepted memory

[Kirill A. Shutemov]

The firmware will pre-accept the memory used to run the stub. But, the
stub is responsible for accepting the memory into which it decompresses
the main kernel. Accept memory just before decompression starts.

The stub is also responsible for choosing a physical address in which to
place the decompressed kernel image. The KASLR mechanism will randomize
this physical address. Since the accepted memory region is relatively
small, KASLR would be quite ineffective if it only used the pre-accepted
area (EFI_CONVENTIONAL_MEMORY). Ensure that KASLR randomizes among the
entire physical address space by also including EFI_UNACCEPTED_MEMORY.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Liam Merwick <liam.merwick@oracle.com>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
---
 arch/x86/boot/compressed/efi.h   | 10 ++++++++
 arch/x86/boot/compressed/kaslr.c | 40 +++++++++++++++++++++----------
 arch/x86/boot/compressed/mem.c   | 41 ++++++++++++++++++++++++++++++++
 arch/x86/boot/compressed/misc.c  |  6 +++++
 arch/x86/boot/compressed/misc.h  | 10 ++++++++
 5 files changed, 95 insertions(+), 12 deletions(-)

diff --git a/arch/x86/boot/compressed/efi.h b/arch/x86/boot/compressed/efi.h
index 7db2f41b54cd..866c0af8b5b9 100644
--- a/arch/x86/boot/compressed/efi.h
+++ b/arch/x86/boot/compressed/efi.h
@@ -16,6 +16,7 @@ typedef guid_t efi_guid_t __aligned(__alignof__(u32));
 #define ACPI_TABLE_GUID				EFI_GUID(0xeb9d2d30, 0x2d88, 0x11d3,  0x9a, 0x16, 0x00, 0x90, 0x27, 0x3f, 0xc1, 0x4d)
 #define ACPI_20_TABLE_GUID			EFI_GUID(0x8868e871, 0xe4f1, 0x11d3,  0xbc, 0x22, 0x00, 0x80, 0xc7, 0x3c, 0x88, 0x81)
 #define EFI_CC_BLOB_GUID			EFI_GUID(0x067b1f5f, 0xcf26, 0x44c5, 0x85, 0x54, 0x93, 0xd7, 0x77, 0x91, 0x2d, 0x42)
+#define LINUX_EFI_UNACCEPTED_MEM_TABLE_GUID	EFI_GUID(0xd5d1de3c, 0x105c, 0x44f9,  0x9e, 0xa9, 0xbc, 0xef, 0x98, 0x12, 0x00, 0x31)
 
 #define EFI32_LOADER_SIGNATURE	"EL32"
 #define EFI64_LOADER_SIGNATURE	"EL64"
@@ -32,6 +33,7 @@ typedef	struct {
 } efi_table_hdr_t;
 
 #define EFI_CONVENTIONAL_MEMORY		 7
+#define EFI_UNACCEPTED_MEMORY		15
 
 #define EFI_MEMORY_MORE_RELIABLE \
 				((u64)0x0000000000010000ULL)	/* higher reliability */
@@ -104,6 +106,14 @@ struct efi_setup_data {
 	u64 reserved[8];
 };
 
+struct efi_unaccepted_memory {
+	u32 version;
+	u32 unit_size;
+	u64 phys_base;
+	u64 size;
+	unsigned long bitmap[];
+};
+
 static inline int efi_guidcmp (efi_guid_t left, efi_guid_t right)
 {
 	return memcmp(&left, &right, sizeof (efi_guid_t));
diff --git a/arch/x86/boot/compressed/kaslr.c b/arch/x86/boot/compressed/kaslr.c
index 454757fbdfe5..9193acf0e9cd 100644
--- a/arch/x86/boot/compressed/kaslr.c
+++ b/arch/x86/boot/compressed/kaslr.c
@@ -672,6 +672,33 @@ static bool process_mem_region(struct mem_vector *region,
 }
 
 #ifdef CONFIG_EFI
+
+/*
+ * Only EFI_CONVENTIONAL_MEMORY and EFI_UNACCEPTED_MEMORY (if supported) are
+ * guaranteed to be free.
+ *
+ * Pick free memory more conservatively than the EFI spec allows: according to
+ * the spec, EFI_BOOT_SERVICES_{CODE|DATA} are also free memory and thus
+ * available to place the kernel image into, but in practice there's firmware
+ * where using that memory leads to crashes. Buggy vendor EFI code registers
+ * for an event that triggers on SetVirtualAddressMap(). The handler assumes
+ * that EFI_BOOT_SERVICES_DATA memory has not been touched by loader yet, which
+ * is probably true for Windows.
+ *
+ * Preserve EFI_BOOT_SERVICES_* regions until after SetVirtualAddressMap().
+ */
+static inline bool memory_type_is_free(efi_memory_desc_t *md)
+{
+	if (md->type == EFI_CONVENTIONAL_MEMORY)
+		return true;
+
+	if (IS_ENABLED(CONFIG_UNACCEPTED_MEMORY) &&
+	    md->type == EFI_UNACCEPTED_MEMORY)
+		    return true;
+
+	return false;
+}
+
 /*
  * Returns true if we processed the EFI memmap, which we prefer over the E820
  * table if it is available.
@@ -716,18 +743,7 @@ process_efi_entries(unsigned long minimum, unsigned long image_size)
 	for (i = 0; i < nr_desc; i++) {
 		md = efi_early_memdesc_ptr(pmap, e->efi_memdesc_size, i);
 
-		/*
-		 * Here we are more conservative in picking free memory than
-		 * the EFI spec allows:
-		 *
-		 * According to the spec, EFI_BOOT_SERVICES_{CODE|DATA} are also
-		 * free memory and thus available to place the kernel image into,
-		 * but in practice there's firmware where using that memory leads
-		 * to crashes.
-		 *
-		 * Only EFI_CONVENTIONAL_MEMORY is guaranteed to be free.
-		 */
-		if (md->type != EFI_CONVENTIONAL_MEMORY)
+		if (!memory_type_is_free(md))
 			continue;
 
 		if (efi_soft_reserve_enabled() &&
diff --git a/arch/x86/boot/compressed/mem.c b/arch/x86/boot/compressed/mem.c
index 67594fcb11d9..69038ed7a310 100644
--- a/arch/x86/boot/compressed/mem.c
+++ b/arch/x86/boot/compressed/mem.c
@@ -1,9 +1,50 @@
 // SPDX-License-Identifier: GPL-2.0-only
 
 #include "error.h"
+#include "misc.h"
 
 void arch_accept_memory(phys_addr_t start, phys_addr_t end)
 {
 	/* Platform-specific memory-acceptance call goes here */
 	error("Cannot accept memory");
 }
+
+bool init_unaccepted_memory(void)
+{
+	guid_t guid = LINUX_EFI_UNACCEPTED_MEM_TABLE_GUID;
+	struct efi_unaccepted_memory *table;
+	unsigned long cfg_table_pa;
+	unsigned int cfg_table_len;
+	enum efi_type et;
+	int ret;
+
+	et = efi_get_type(boot_params);
+	if (et == EFI_TYPE_NONE)
+		return false;
+
+	ret = efi_get_conf_table(boot_params, &cfg_table_pa, &cfg_table_len);
+	if (ret) {
+		warn("EFI config table not found.");
+		return false;
+	}
+
+	table = (void *)efi_find_vendor_table(boot_params, cfg_table_pa,
+					      cfg_table_len, guid);
+	if (!table)
+		return false;
+
+	if (table->version != 1)
+		error("Unknown version of unaccepted memory table\n");
+
+	/*
+	 * In many cases unaccepted_table is already set by EFI stub, but it
+	 * has to be initialized again to cover cases when the table is not
+	 * allocated by EFI stub or EFI stub copied the kernel image with
+	 * efi_relocate_kernel() before the variable is set.
+	 *
+	 * It must be initialized before the first usage of accept_memory().
+	 */
+	unaccepted_table = table;
+
+	return true;
+}
diff --git a/arch/x86/boot/compressed/misc.c b/arch/x86/boot/compressed/misc.c
index 014ff222bf4b..94b7abcf624b 100644
--- a/arch/x86/boot/compressed/misc.c
+++ b/arch/x86/boot/compressed/misc.c
@@ -455,6 +455,12 @@ asmlinkage __visible void *extract_kernel(void *rmode, memptr heap,
 #endif
 
 	debug_putstr("\nDecompressing Linux... ");
+
+	if (init_unaccepted_memory()) {
+		debug_putstr("Accepting memory... ");
+		accept_memory(__pa(output), __pa(output) + needed_size);
+	}
+
 	__decompress(input_data, input_len, NULL, NULL, output, output_len,
 			NULL, error);
 	entry_offset = parse_elf(output);
diff --git a/arch/x86/boot/compressed/misc.h b/arch/x86/boot/compressed/misc.h
index 2f155a0e3041..964fe903a1cd 100644
--- a/arch/x86/boot/compressed/misc.h
+++ b/arch/x86/boot/compressed/misc.h
@@ -247,4 +247,14 @@ static inline unsigned long efi_find_vendor_table(struct boot_params *bp,
 }
 #endif /* CONFIG_EFI */
 
+#ifdef CONFIG_UNACCEPTED_MEMORY
+bool init_unaccepted_memory(void);
+#else
+static inline bool init_unaccepted_memory(void) { return false; }
+#endif
+
+/* Defined in EFI stub */
+extern struct efi_unaccepted_memory *unaccepted_table;
+void accept_memory(phys_addr_t start, phys_addr_t end);
+
 #endif /* BOOT_COMPRESSED_MISC_H */
-- 
2.39.3

[PATCHv14 5/9] efi: Add unaccepted memory support

[Kirill A. Shutemov]

efi_config_parse_tables() reserves memory that holds unaccepted memory
configuration table so it won't be reused by page allocator.

Core-mm requires few helpers to support unaccepted memory:

 - accept_memory() checks the range of addresses against the bitmap and
   accept memory if needed.

 - range_contains_unaccepted_memory() checks if anything within the
   range requires acceptance.

Architectural code has to provide efi_get_unaccepted_table() that
returns pointer to the unaccepted memory configuration table.

arch_accept_memory() handles arch-specific part of memory acceptance.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
---
 arch/x86/platform/efi/efi.c              |   3 +
 drivers/firmware/efi/Makefile            |   1 +
 drivers/firmware/efi/efi.c               |  25 +++++
 drivers/firmware/efi/unaccepted_memory.c | 112 +++++++++++++++++++++++
 include/linux/efi.h                      |   1 +
 5 files changed, 142 insertions(+)
 create mode 100644 drivers/firmware/efi/unaccepted_memory.c

diff --git a/arch/x86/platform/efi/efi.c b/arch/x86/platform/efi/efi.c
index f3f2d87cce1b..e9f99c56f3ce 100644
--- a/arch/x86/platform/efi/efi.c
+++ b/arch/x86/platform/efi/efi.c
@@ -96,6 +96,9 @@ static const unsigned long * const efi_tables[] = {
 #ifdef CONFIG_EFI_COCO_SECRET
 	&efi.coco_secret,
 #endif
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	&efi.unaccepted,
+#endif
 };
 
 u64 efi_setup;		/* efi setup_data physical address */
diff --git a/drivers/firmware/efi/Makefile b/drivers/firmware/efi/Makefile
index b51f2a4c821e..e489fefd23da 100644
--- a/drivers/firmware/efi/Makefile
+++ b/drivers/firmware/efi/Makefile
@@ -41,3 +41,4 @@ obj-$(CONFIG_EFI_CAPSULE_LOADER)	+= capsule-loader.o
 obj-$(CONFIG_EFI_EARLYCON)		+= earlycon.o
 obj-$(CONFIG_UEFI_CPER_ARM)		+= cper-arm.o
 obj-$(CONFIG_UEFI_CPER_X86)		+= cper-x86.o
+obj-$(CONFIG_UNACCEPTED_MEMORY)		+= unaccepted_memory.o
diff --git a/drivers/firmware/efi/efi.c b/drivers/firmware/efi/efi.c
index 7dce06e419c5..d817e7afd266 100644
--- a/drivers/firmware/efi/efi.c
+++ b/drivers/firmware/efi/efi.c
@@ -50,6 +50,9 @@ struct efi __read_mostly efi = {
 #ifdef CONFIG_EFI_COCO_SECRET
 	.coco_secret		= EFI_INVALID_TABLE_ADDR,
 #endif
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	.unaccepted		= EFI_INVALID_TABLE_ADDR,
+#endif
 };
 EXPORT_SYMBOL(efi);
 
@@ -605,6 +608,9 @@ static const efi_config_table_type_t common_tables[] __initconst = {
 #ifdef CONFIG_EFI_COCO_SECRET
 	{LINUX_EFI_COCO_SECRET_AREA_GUID,	&efi.coco_secret,	"CocoSecret"	},
 #endif
+#ifdef CONFIG_UNACCEPTED_MEMORY
+	{LINUX_EFI_UNACCEPTED_MEM_TABLE_GUID,	&efi.unaccepted,	"Unaccepted"	},
+#endif
 #ifdef CONFIG_EFI_GENERIC_STUB
 	{LINUX_EFI_SCREEN_INFO_TABLE_GUID,	&screen_info_table			},
 #endif
@@ -759,6 +765,25 @@ int __init efi_config_parse_tables(const efi_config_table_t *config_tables,
 		}
 	}
 
+	if (IS_ENABLED(CONFIG_UNACCEPTED_MEMORY) &&
+	    efi.unaccepted != EFI_INVALID_TABLE_ADDR) {
+		struct efi_unaccepted_memory *unaccepted;
+
+		unaccepted = early_memremap(efi.unaccepted, sizeof(*unaccepted));
+		if (unaccepted) {
+			unsigned long size;
+
+			if (unaccepted->version == 1) {
+				size = sizeof(*unaccepted) + unaccepted->size;
+				memblock_reserve(efi.unaccepted, size);
+			} else {
+				efi.unaccepted = EFI_INVALID_TABLE_ADDR;
+			}
+
+			early_memunmap(unaccepted, sizeof(*unaccepted));
+		}
+	}
+
 	return 0;
 }
 
diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
new file mode 100644
index 000000000000..08a9a843550a
--- /dev/null
+++ b/drivers/firmware/efi/unaccepted_memory.c
@@ -0,0 +1,112 @@
+// SPDX-License-Identifier: GPL-2.0-only
+
+#include <linux/efi.h>
+#include <linux/memblock.h>
+#include <linux/spinlock.h>
+#include <asm/unaccepted_memory.h>
+
+/* Protects unaccepted memory bitmap */
+static DEFINE_SPINLOCK(unaccepted_memory_lock);
+
+/*
+ * accept_memory() -- Consult bitmap and accept the memory if needed.
+ *
+ * Only memory that is explicitly marked as unaccepted in the bitmap requires
+ * an action. All the remaining memory is implicitly accepted and doesn't need
+ * acceptance.
+ *
+ * No need to accept:
+ *  - anything if the system has no unaccepted table;
+ *  - memory that is below phys_base;
+ *  - memory that is above the memory that addressable by the bitmap;
+ */
+void accept_memory(phys_addr_t start, phys_addr_t end)
+{
+	struct efi_unaccepted_memory *unaccepted;
+	unsigned long range_start, range_end;
+	unsigned long flags;
+	u64 unit_size;
+
+	unaccepted = efi_get_unaccepted_table();
+	if (!unaccepted)
+		return;
+
+	unit_size = unaccepted->unit_size;
+
+	/*
+	 * Only care for the part of the range that is represented
+	 * in the bitmap.
+	 */
+	if (start < unaccepted->phys_base)
+		start = unaccepted->phys_base;
+	if (end < unaccepted->phys_base)
+		return;
+
+	/* Translate to offsets from the beginning of the bitmap */
+	start -= unaccepted->phys_base;
+	end -= unaccepted->phys_base;
+
+	/* Make sure not to overrun the bitmap */
+	if (end > unaccepted->size * unit_size * BITS_PER_BYTE)
+		end = unaccepted->size * unit_size * BITS_PER_BYTE;
+
+	range_start = start / unit_size;
+
+	spin_lock_irqsave(&unaccepted_memory_lock, flags);
+	for_each_set_bitrange_from(range_start, range_end, unaccepted->bitmap,
+				   DIV_ROUND_UP(end, unit_size)) {
+		unsigned long phys_start, phys_end;
+		unsigned long len = range_end - range_start;
+
+		phys_start = range_start * unit_size + unaccepted->phys_base;
+		phys_end = range_end * unit_size + unaccepted->phys_base;
+
+		arch_accept_memory(phys_start, phys_end);
+		bitmap_clear(unaccepted->bitmap, range_start, len);
+	}
+	spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
+}
+
+bool range_contains_unaccepted_memory(phys_addr_t start, phys_addr_t end)
+{
+	struct efi_unaccepted_memory *unaccepted;
+	unsigned long flags;
+	bool ret = false;
+	u64 unit_size;
+
+	unaccepted = efi_get_unaccepted_table();
+	if (!unaccepted)
+		return false;
+
+	unit_size = unaccepted->unit_size;
+
+	/*
+	 * Only care for the part of the range that is represented
+	 * in the bitmap.
+	 */
+	if (start < unaccepted->phys_base)
+		start = unaccepted->phys_base;
+	if (end < unaccepted->phys_base)
+		return false;
+
+	/* Translate to offsets from the beginning of the bitmap */
+	start -= unaccepted->phys_base;
+	end -= unaccepted->phys_base;
+
+	/* Make sure not to overrun the bitmap */
+	if (end > unaccepted->size * unit_size * BITS_PER_BYTE)
+		end = unaccepted->size * unit_size * BITS_PER_BYTE;
+
+	spin_lock_irqsave(&unaccepted_memory_lock, flags);
+	while (start < end) {
+		if (test_bit(start / unit_size, unaccepted->bitmap)) {
+			ret = true;
+			break;
+		}
+
+		start += unit_size;
+	}
+	spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
+
+	return ret;
+}
diff --git a/include/linux/efi.h b/include/linux/efi.h
index 29cc622910da..9864f9c00da2 100644
--- a/include/linux/efi.h
+++ b/include/linux/efi.h
@@ -646,6 +646,7 @@ extern struct efi {
 	unsigned long			tpm_final_log;		/* TPM2 Final Events Log table */
 	unsigned long			mokvar_table;		/* MOK variable config table */
 	unsigned long			coco_secret;		/* Confidential computing secret table */
+	unsigned long			unaccepted;		/* Unaccepted memory table */
 
 	efi_get_time_t			*get_time;
 	efi_set_time_t			*set_time;
-- 
2.39.3

[PATCHv14 6/9] efi/unaccepted: Avoid load_unaligned_zeropad() stepping into unaccepted memory

[Kirill A. Shutemov]

load_unaligned_zeropad() can lead to unwanted loads across page boundaries.
The unwanted loads are typically harmless. But, they might be made to
totally unrelated or even unmapped memory. load_unaligned_zeropad()
relies on exception fixup (#PF, #GP and now #VE) to recover from these
unwanted loads.

But, this approach does not work for unaccepted memory. For TDX, a load
from unaccepted memory will not lead to a recoverable exception within
the guest. The guest will exit to the VMM where the only recourse is to
terminate the guest.

There are two parts to fix this issue and comprehensively avoid access
to unaccepted memory. Together these ensure that an extra "guard" page
is accepted in addition to the memory that needs to be used.

1. Implicitly extend the range_contains_unaccepted_memory(start, end)
   checks up to end+unit_size if 'end' is aligned on a unit_size
   boundary.
2. Implicitly extend accept_memory(start, end) to end+unit_size if 'end'
   is aligned on a unit_size boundary.

Side note: This leads to something strange. Pages which were accepted
	   at boot, marked by the firmware as accepted and will never
	   _need_ to be accepted might be on unaccepted_pages list
	   This is a cue to ensure that the next page is accepted
	   before 'page' can be used.

This is an actual, real-world problem which was discovered during TDX
testing.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
---
 drivers/firmware/efi/unaccepted_memory.c | 35 ++++++++++++++++++++++++
 1 file changed, 35 insertions(+)

diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
index 08a9a843550a..853f7dc3c21d 100644
--- a/drivers/firmware/efi/unaccepted_memory.c
+++ b/drivers/firmware/efi/unaccepted_memory.c
@@ -46,6 +46,34 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
 	start -= unaccepted->phys_base;
 	end -= unaccepted->phys_base;
 
+	/*
+	 * load_unaligned_zeropad() can lead to unwanted loads across page
+	 * boundaries. The unwanted loads are typically harmless. But, they
+	 * might be made to totally unrelated or even unmapped memory.
+	 * load_unaligned_zeropad() relies on exception fixup (#PF, #GP and now
+	 * #VE) to recover from these unwanted loads.
+	 *
+	 * But, this approach does not work for unaccepted memory. For TDX, a
+	 * load from unaccepted memory will not lead to a recoverable exception
+	 * within the guest. The guest will exit to the VMM where the only
+	 * recourse is to terminate the guest.
+	 *
+	 * There are two parts to fix this issue and comprehensively avoid
+	 * access to unaccepted memory. Together these ensure that an extra
+	 * "guard" page is accepted in addition to the memory that needs to be
+	 * used:
+	 *
+	 * 1. Implicitly extend the range_contains_unaccepted_memory(start, end)
+	 *    checks up to end+unit_size if 'end' is aligned on a unit_size
+	 *    boundary.
+	 *
+	 * 2. Implicitly extend accept_memory(start, end) to end+unit_size if
+	 *    'end' is aligned on a unit_size boundary. (immediately following
+	 *    this comment)
+	 */
+	if (!(end % unit_size))
+		end += unit_size;
+
 	/* Make sure not to overrun the bitmap */
 	if (end > unaccepted->size * unit_size * BITS_PER_BYTE)
 		end = unaccepted->size * unit_size * BITS_PER_BYTE;
@@ -93,6 +121,13 @@ bool range_contains_unaccepted_memory(phys_addr_t start, phys_addr_t end)
 	start -= unaccepted->phys_base;
 	end -= unaccepted->phys_base;
 
+	/*
+	 * Also consider the unaccepted state of the *next* page. See fix #1 in
+	 * the comment on load_unaligned_zeropad() in accept_memory().
+	 */
+	if (!(end % unit_size))
+		end += unit_size;
+
 	/* Make sure not to overrun the bitmap */
 	if (end > unaccepted->size * unit_size * BITS_PER_BYTE)
 		end = unaccepted->size * unit_size * BITS_PER_BYTE;
-- 
2.39.3

[PATCHv14 7/9] x86/tdx: Make _tdx_hypercall() and __tdx_module_call() available in boot stub

[Kirill A. Shutemov]

Memory acceptance requires a hypercall and one or multiple module calls.

Make helpers for the calls available in boot stub. It has to accept
memory where kernel image and initrd are placed.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com>
---
 arch/x86/coco/tdx/tdx.c           | 32 -------------------
 arch/x86/include/asm/shared/tdx.h | 51 +++++++++++++++++++++++++++++++
 arch/x86/include/asm/tdx.h        | 19 ------------
 3 files changed, 51 insertions(+), 51 deletions(-)

diff --git a/arch/x86/coco/tdx/tdx.c b/arch/x86/coco/tdx/tdx.c
index e146b599260f..e6f4c2758a68 100644
--- a/arch/x86/coco/tdx/tdx.c
+++ b/arch/x86/coco/tdx/tdx.c
@@ -14,20 +14,6 @@
 #include <asm/insn-eval.h>
 #include <asm/pgtable.h>
 
-/* TDX module Call Leaf IDs */
-#define TDX_GET_INFO			1
-#define TDX_GET_VEINFO			3
-#define TDX_GET_REPORT			4
-#define TDX_ACCEPT_PAGE			6
-#define TDX_WR				8
-
-/* TDCS fields. To be used by TDG.VM.WR and TDG.VM.RD module calls */
-#define TDCS_NOTIFY_ENABLES		0x9100000000000010
-
-/* TDX hypercall Leaf IDs */
-#define TDVMCALL_MAP_GPA		0x10001
-#define TDVMCALL_REPORT_FATAL_ERROR	0x10003
-
 /* MMIO direction */
 #define EPT_READ	0
 #define EPT_WRITE	1
@@ -51,24 +37,6 @@
 
 #define TDREPORT_SUBTYPE_0	0
 
-/*
- * Wrapper for standard use of __tdx_hypercall with no output aside from
- * return code.
- */
-static inline u64 _tdx_hypercall(u64 fn, u64 r12, u64 r13, u64 r14, u64 r15)
-{
-	struct tdx_hypercall_args args = {
-		.r10 = TDX_HYPERCALL_STANDARD,
-		.r11 = fn,
-		.r12 = r12,
-		.r13 = r13,
-		.r14 = r14,
-		.r15 = r15,
-	};
-
-	return __tdx_hypercall(&args);
-}
-
 /* Called from __tdx_hypercall() for unrecoverable failure */
 noinstr void __tdx_hypercall_failed(void)
 {
diff --git a/arch/x86/include/asm/shared/tdx.h b/arch/x86/include/asm/shared/tdx.h
index 2631e01f6e0f..1ff0ee822961 100644
--- a/arch/x86/include/asm/shared/tdx.h
+++ b/arch/x86/include/asm/shared/tdx.h
@@ -10,6 +10,20 @@
 #define TDX_CPUID_LEAF_ID	0x21
 #define TDX_IDENT		"IntelTDX    "
 
+/* TDX module Call Leaf IDs */
+#define TDX_GET_INFO			1
+#define TDX_GET_VEINFO			3
+#define TDX_GET_REPORT			4
+#define TDX_ACCEPT_PAGE			6
+#define TDX_WR				8
+
+/* TDCS fields. To be used by TDG.VM.WR and TDG.VM.RD module calls */
+#define TDCS_NOTIFY_ENABLES		0x9100000000000010
+
+/* TDX hypercall Leaf IDs */
+#define TDVMCALL_MAP_GPA		0x10001
+#define TDVMCALL_REPORT_FATAL_ERROR	0x10003
+
 #ifndef __ASSEMBLY__
 
 /*
@@ -37,8 +51,45 @@ struct tdx_hypercall_args {
 u64 __tdx_hypercall(struct tdx_hypercall_args *args);
 u64 __tdx_hypercall_ret(struct tdx_hypercall_args *args);
 
+/*
+ * Wrapper for standard use of __tdx_hypercall with no output aside from
+ * return code.
+ */
+static inline u64 _tdx_hypercall(u64 fn, u64 r12, u64 r13, u64 r14, u64 r15)
+{
+	struct tdx_hypercall_args args = {
+		.r10 = TDX_HYPERCALL_STANDARD,
+		.r11 = fn,
+		.r12 = r12,
+		.r13 = r13,
+		.r14 = r14,
+		.r15 = r15,
+	};
+
+	return __tdx_hypercall(&args);
+}
+
+
 /* Called from __tdx_hypercall() for unrecoverable failure */
 void __tdx_hypercall_failed(void);
 
+/*
+ * Used in __tdx_module_call() to gather the output registers' values of the
+ * TDCALL instruction when requesting services from the TDX module. This is a
+ * software only structure and not part of the TDX module/VMM ABI
+ */
+struct tdx_module_output {
+	u64 rcx;
+	u64 rdx;
+	u64 r8;
+	u64 r9;
+	u64 r10;
+	u64 r11;
+};
+
+/* Used to communicate with the TDX module */
+u64 __tdx_module_call(u64 fn, u64 rcx, u64 rdx, u64 r8, u64 r9,
+		      struct tdx_module_output *out);
+
 #endif /* !__ASSEMBLY__ */
 #endif /* _ASM_X86_SHARED_TDX_H */
diff --git a/arch/x86/include/asm/tdx.h b/arch/x86/include/asm/tdx.h
index 28d889c9aa16..234197ec17e4 100644
--- a/arch/x86/include/asm/tdx.h
+++ b/arch/x86/include/asm/tdx.h
@@ -20,21 +20,6 @@
 
 #ifndef __ASSEMBLY__
 
-/*
- * Used to gather the output registers values of the TDCALL and SEAMCALL
- * instructions when requesting services from the TDX module.
- *
- * This is a software only structure and not part of the TDX module/VMM ABI.
- */
-struct tdx_module_output {
-	u64 rcx;
-	u64 rdx;
-	u64 r8;
-	u64 r9;
-	u64 r10;
-	u64 r11;
-};
-
 /*
  * Used by the #VE exception handler to gather the #VE exception
  * info from the TDX module. This is a software only structure
@@ -55,10 +40,6 @@ struct ve_info {
 
 void __init tdx_early_init(void);
 
-/* Used to communicate with the TDX module */
-u64 __tdx_module_call(u64 fn, u64 rcx, u64 rdx, u64 r8, u64 r9,
-		      struct tdx_module_output *out);
-
 void tdx_get_ve_info(struct ve_info *ve);
 
 bool tdx_handle_virt_exception(struct pt_regs *regs, struct ve_info *ve);
-- 
2.39.3

[PATCHv14 8/9] x86/tdx: Refactor try_accept_one()

[Kirill A. Shutemov]

Rework try_accept_one() to return accepted size instead of modifying
'start' inside the helper. It makes 'start' in-only argument and
streamlines code on the caller side.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Suggested-by: Borislav Petkov <bp@alien8.de>
Reviewed-by: Dave Hansen <dave.hansen@linux.intel.com>
---
 arch/x86/coco/tdx/tdx.c | 38 +++++++++++++++++++-------------------
 1 file changed, 19 insertions(+), 19 deletions(-)

diff --git a/arch/x86/coco/tdx/tdx.c b/arch/x86/coco/tdx/tdx.c
index e6f4c2758a68..0d5fe6e24e45 100644
--- a/arch/x86/coco/tdx/tdx.c
+++ b/arch/x86/coco/tdx/tdx.c
@@ -713,18 +713,18 @@ static bool tdx_cache_flush_required(void)
 	return true;
 }
 
-static bool try_accept_one(phys_addr_t *start, unsigned long len,
-			  enum pg_level pg_level)
+static unsigned long try_accept_one(phys_addr_t start, unsigned long len,
+				    enum pg_level pg_level)
 {
 	unsigned long accept_size = page_level_size(pg_level);
 	u64 tdcall_rcx;
 	u8 page_size;
 
-	if (!IS_ALIGNED(*start, accept_size))
-		return false;
+	if (!IS_ALIGNED(start, accept_size))
+		return 0;
 
 	if (len < accept_size)
-		return false;
+		return 0;
 
 	/*
 	 * Pass the page physical address to the TDX module to accept the
@@ -743,15 +743,14 @@ static bool try_accept_one(phys_addr_t *start, unsigned long len,
 		page_size = 2;
 		break;
 	default:
-		return false;
+		return 0;
 	}
 
-	tdcall_rcx = *start | page_size;
+	tdcall_rcx = start | page_size;
 	if (__tdx_module_call(TDX_ACCEPT_PAGE, tdcall_rcx, 0, 0, 0, NULL))
-		return false;
+		return 0;
 
-	*start += accept_size;
-	return true;
+	return accept_size;
 }
 
 /*
@@ -788,21 +787,22 @@ static bool tdx_enc_status_changed(unsigned long vaddr, int numpages, bool enc)
 	 */
 	while (start < end) {
 		unsigned long len = end - start;
+		unsigned long accept_size;
 
 		/*
 		 * Try larger accepts first. It gives chance to VMM to keep
-		 * 1G/2M SEPT entries where possible and speeds up process by
-		 * cutting number of hypercalls (if successful).
+		 * 1G/2M Secure EPT entries where possible and speeds up
+		 * process by cutting number of hypercalls (if successful).
 		 */
 
-		if (try_accept_one(&start, len, PG_LEVEL_1G))
-			continue;
-
-		if (try_accept_one(&start, len, PG_LEVEL_2M))
-			continue;
-
-		if (!try_accept_one(&start, len, PG_LEVEL_4K))
+		accept_size = try_accept_one(start, len, PG_LEVEL_1G);
+		if (!accept_size)
+			accept_size = try_accept_one(start, len, PG_LEVEL_2M);
+		if (!accept_size)
+			accept_size = try_accept_one(start, len, PG_LEVEL_4K);
+		if (!accept_size)
 			return false;
+		start += accept_size;
 	}
 
 	return true;
-- 
2.39.3

[PATCHv14 9/9] x86/tdx: Add unaccepted memory support

[Kirill A. Shutemov]

Hookup TDX-specific code to accept memory.

Accepting the memory is done with ACCEPT_PAGE module call on every page
in the range. MAP_GPA hypercall is not required as the unaccepted memory
is considered private already.

Extract the part of tdx_enc_status_changed() that does memory acceptance
in a new helper. Move the helper tdx-shared.c. It is going to be used by
both main kernel and decompressor.

Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
---
 arch/x86/Kconfig                         |  2 +
 arch/x86/boot/compressed/Makefile        |  2 +-
 arch/x86/boot/compressed/error.c         | 19 +++++++
 arch/x86/boot/compressed/error.h         |  1 +
 arch/x86/boot/compressed/mem.c           | 35 +++++++++++-
 arch/x86/boot/compressed/tdx-shared.c    |  2 +
 arch/x86/coco/tdx/Makefile               |  2 +-
 arch/x86/coco/tdx/tdx-shared.c           | 71 ++++++++++++++++++++++++
 arch/x86/coco/tdx/tdx.c                  | 70 +----------------------
 arch/x86/include/asm/shared/tdx.h        |  2 +
 arch/x86/include/asm/unaccepted_memory.h | 24 ++++++++
 11 files changed, 160 insertions(+), 70 deletions(-)
 create mode 100644 arch/x86/boot/compressed/tdx-shared.c
 create mode 100644 arch/x86/coco/tdx/tdx-shared.c
 create mode 100644 arch/x86/include/asm/unaccepted_memory.h

diff --git a/arch/x86/Kconfig b/arch/x86/Kconfig
index 53bab123a8ee..5c72067c06d4 100644
--- a/arch/x86/Kconfig
+++ b/arch/x86/Kconfig
@@ -884,9 +884,11 @@ config INTEL_TDX_GUEST
 	bool "Intel TDX (Trust Domain Extensions) - Guest Support"
 	depends on X86_64 && CPU_SUP_INTEL
 	depends on X86_X2APIC
+	depends on EFI_STUB
 	select ARCH_HAS_CC_PLATFORM
 	select X86_MEM_ENCRYPT
 	select X86_MCE
+	select UNACCEPTED_MEMORY
 	help
 	  Support running as a guest under Intel TDX.  Without this support,
 	  the guest kernel can not boot or run under TDX.
diff --git a/arch/x86/boot/compressed/Makefile b/arch/x86/boot/compressed/Makefile
index cc4978123c30..b13a58021086 100644
--- a/arch/x86/boot/compressed/Makefile
+++ b/arch/x86/boot/compressed/Makefile
@@ -106,7 +106,7 @@ ifdef CONFIG_X86_64
 endif
 
 vmlinux-objs-$(CONFIG_ACPI) += $(obj)/acpi.o
-vmlinux-objs-$(CONFIG_INTEL_TDX_GUEST) += $(obj)/tdx.o $(obj)/tdcall.o
+vmlinux-objs-$(CONFIG_INTEL_TDX_GUEST) += $(obj)/tdx.o $(obj)/tdcall.o $(obj)/tdx-shared.o
 vmlinux-objs-$(CONFIG_UNACCEPTED_MEMORY) += $(obj)/mem.o
 
 vmlinux-objs-$(CONFIG_EFI) += $(obj)/efi.o
diff --git a/arch/x86/boot/compressed/error.c b/arch/x86/boot/compressed/error.c
index c881878e56d3..5313c5cb2b80 100644
--- a/arch/x86/boot/compressed/error.c
+++ b/arch/x86/boot/compressed/error.c
@@ -22,3 +22,22 @@ void error(char *m)
 	while (1)
 		asm("hlt");
 }
+
+/* EFI libstub  provides vsnprintf() */
+#ifdef CONFIG_EFI_STUB
+void panic(const char *fmt, ...)
+{
+	static char buf[1024];
+	va_list args;
+	int len;
+
+	va_start(args, fmt);
+	len = vsnprintf(buf, sizeof(buf), fmt, args);
+	va_end(args);
+
+	if (len && buf[len - 1] == '\n')
+		buf[len - 1] = '\0';
+
+	error(buf);
+}
+#endif
diff --git a/arch/x86/boot/compressed/error.h b/arch/x86/boot/compressed/error.h
index 1de5821184f1..86fe33b93715 100644
--- a/arch/x86/boot/compressed/error.h
+++ b/arch/x86/boot/compressed/error.h
@@ -6,5 +6,6 @@
 
 void warn(char *m);
 void error(char *m) __noreturn;
+void panic(const char *fmt, ...) __noreturn __cold;
 
 #endif /* BOOT_COMPRESSED_ERROR_H */
diff --git a/arch/x86/boot/compressed/mem.c b/arch/x86/boot/compressed/mem.c
index 69038ed7a310..f04b29f3572f 100644
--- a/arch/x86/boot/compressed/mem.c
+++ b/arch/x86/boot/compressed/mem.c
@@ -2,11 +2,44 @@
 
 #include "error.h"
 #include "misc.h"
+#include "tdx.h"
+#include <asm/shared/tdx.h>
+
+/*
+ * accept_memory() and process_unaccepted_memory() called from EFI stub which
+ * runs before decompresser and its early_tdx_detect().
+ *
+ * Enumerate TDX directly from the early users.
+ */
+static bool early_is_tdx_guest(void)
+{
+	static bool once;
+	static bool is_tdx;
+
+	if (!IS_ENABLED(CONFIG_INTEL_TDX_GUEST))
+		return false;
+
+	if (!once) {
+		u32 eax, sig[3];
+
+		cpuid_count(TDX_CPUID_LEAF_ID, 0, &eax,
+			    &sig[0], &sig[2],  &sig[1]);
+		is_tdx = !memcmp(TDX_IDENT, sig, sizeof(sig));
+		once = true;
+	}
+
+	return is_tdx;
+}
 
 void arch_accept_memory(phys_addr_t start, phys_addr_t end)
 {
 	/* Platform-specific memory-acceptance call goes here */
-	error("Cannot accept memory");
+	if (early_is_tdx_guest()) {
+		if (!tdx_accept_memory(start, end))
+			panic("TDX: Failed to accept memory\n");
+	} else {
+		error("Cannot accept memory: unknown platform\n");
+	}
 }
 
 bool init_unaccepted_memory(void)
diff --git a/arch/x86/boot/compressed/tdx-shared.c b/arch/x86/boot/compressed/tdx-shared.c
new file mode 100644
index 000000000000..5ac43762fe13
--- /dev/null
+++ b/arch/x86/boot/compressed/tdx-shared.c
@@ -0,0 +1,2 @@
+#include "error.h"
+#include "../../coco/tdx/tdx-shared.c"
diff --git a/arch/x86/coco/tdx/Makefile b/arch/x86/coco/tdx/Makefile
index 46c55998557d..2c7dcbf1458b 100644
--- a/arch/x86/coco/tdx/Makefile
+++ b/arch/x86/coco/tdx/Makefile
@@ -1,3 +1,3 @@
 # SPDX-License-Identifier: GPL-2.0
 
-obj-y += tdx.o tdcall.o
+obj-y += tdx.o tdx-shared.o tdcall.o
diff --git a/arch/x86/coco/tdx/tdx-shared.c b/arch/x86/coco/tdx/tdx-shared.c
new file mode 100644
index 000000000000..ef20ddc37b58
--- /dev/null
+++ b/arch/x86/coco/tdx/tdx-shared.c
@@ -0,0 +1,71 @@
+#include <asm/tdx.h>
+#include <asm/pgtable.h>
+
+static unsigned long try_accept_one(phys_addr_t start, unsigned long len,
+				    enum pg_level pg_level)
+{
+	unsigned long accept_size = page_level_size(pg_level);
+	u64 tdcall_rcx;
+	u8 page_size;
+
+	if (!IS_ALIGNED(start, accept_size))
+		return 0;
+
+	if (len < accept_size)
+		return 0;
+
+	/*
+	 * Pass the page physical address to the TDX module to accept the
+	 * pending, private page.
+	 *
+	 * Bits 2:0 of RCX encode page size: 0 - 4K, 1 - 2M, 2 - 1G.
+	 */
+	switch (pg_level) {
+	case PG_LEVEL_4K:
+		page_size = 0;
+		break;
+	case PG_LEVEL_2M:
+		page_size = 1;
+		break;
+	case PG_LEVEL_1G:
+		page_size = 2;
+		break;
+	default:
+		return 0;
+	}
+
+	tdcall_rcx = start | page_size;
+	if (__tdx_module_call(TDX_ACCEPT_PAGE, tdcall_rcx, 0, 0, 0, NULL))
+		return 0;
+
+	return accept_size;
+}
+
+bool tdx_accept_memory(phys_addr_t start, phys_addr_t end)
+{
+	/*
+	 * For shared->private conversion, accept the page using
+	 * TDX_ACCEPT_PAGE TDX module call.
+	 */
+	while (start < end) {
+		unsigned long len = end - start;
+		unsigned long accept_size;
+
+		/*
+		 * Try larger accepts first. It gives chance to VMM to keep
+		 * 1G/2M Secure EPT entries where possible and speeds up
+		 * process by cutting number of hypercalls (if successful).
+		 */
+
+		accept_size = try_accept_one(start, len, PG_LEVEL_1G);
+		if (!accept_size)
+			accept_size = try_accept_one(start, len, PG_LEVEL_2M);
+		if (!accept_size)
+			accept_size = try_accept_one(start, len, PG_LEVEL_4K);
+		if (!accept_size)
+			return false;
+		start += accept_size;
+	}
+
+	return true;
+}
diff --git a/arch/x86/coco/tdx/tdx.c b/arch/x86/coco/tdx/tdx.c
index 0d5fe6e24e45..a9c4ba6c5c5d 100644
--- a/arch/x86/coco/tdx/tdx.c
+++ b/arch/x86/coco/tdx/tdx.c
@@ -713,46 +713,6 @@ static bool tdx_cache_flush_required(void)
 	return true;
 }
 
-static unsigned long try_accept_one(phys_addr_t start, unsigned long len,
-				    enum pg_level pg_level)
-{
-	unsigned long accept_size = page_level_size(pg_level);
-	u64 tdcall_rcx;
-	u8 page_size;
-
-	if (!IS_ALIGNED(start, accept_size))
-		return 0;
-
-	if (len < accept_size)
-		return 0;
-
-	/*
-	 * Pass the page physical address to the TDX module to accept the
-	 * pending, private page.
-	 *
-	 * Bits 2:0 of RCX encode page size: 0 - 4K, 1 - 2M, 2 - 1G.
-	 */
-	switch (pg_level) {
-	case PG_LEVEL_4K:
-		page_size = 0;
-		break;
-	case PG_LEVEL_2M:
-		page_size = 1;
-		break;
-	case PG_LEVEL_1G:
-		page_size = 2;
-		break;
-	default:
-		return 0;
-	}
-
-	tdcall_rcx = start | page_size;
-	if (__tdx_module_call(TDX_ACCEPT_PAGE, tdcall_rcx, 0, 0, 0, NULL))
-		return 0;
-
-	return accept_size;
-}
-
 /*
  * Inform the VMM of the guest's intent for this physical page: shared with
  * the VMM or private to the guest.  The VMM is expected to change its mapping
@@ -777,33 +737,9 @@ static bool tdx_enc_status_changed(unsigned long vaddr, int numpages, bool enc)
 	if (_tdx_hypercall(TDVMCALL_MAP_GPA, start, end - start, 0, 0))
 		return false;
 
-	/* private->shared conversion  requires only MapGPA call */
-	if (!enc)
-		return true;
-
-	/*
-	 * For shared->private conversion, accept the page using
-	 * TDX_ACCEPT_PAGE TDX module call.
-	 */
-	while (start < end) {
-		unsigned long len = end - start;
-		unsigned long accept_size;
-
-		/*
-		 * Try larger accepts first. It gives chance to VMM to keep
-		 * 1G/2M Secure EPT entries where possible and speeds up
-		 * process by cutting number of hypercalls (if successful).
-		 */
-
-		accept_size = try_accept_one(start, len, PG_LEVEL_1G);
-		if (!accept_size)
-			accept_size = try_accept_one(start, len, PG_LEVEL_2M);
-		if (!accept_size)
-			accept_size = try_accept_one(start, len, PG_LEVEL_4K);
-		if (!accept_size)
-			return false;
-		start += accept_size;
-	}
+	/* shared->private conversion requires memory to be accepted before use */
+	if (enc)
+		return tdx_accept_memory(start, end);
 
 	return true;
 }
diff --git a/arch/x86/include/asm/shared/tdx.h b/arch/x86/include/asm/shared/tdx.h
index 1ff0ee822961..19228beb4894 100644
--- a/arch/x86/include/asm/shared/tdx.h
+++ b/arch/x86/include/asm/shared/tdx.h
@@ -91,5 +91,7 @@ struct tdx_module_output {
 u64 __tdx_module_call(u64 fn, u64 rcx, u64 rdx, u64 r8, u64 r9,
 		      struct tdx_module_output *out);
 
+bool tdx_accept_memory(phys_addr_t start, phys_addr_t end);
+
 #endif /* !__ASSEMBLY__ */
 #endif /* _ASM_X86_SHARED_TDX_H */
diff --git a/arch/x86/include/asm/unaccepted_memory.h b/arch/x86/include/asm/unaccepted_memory.h
new file mode 100644
index 000000000000..572514e36fde
--- /dev/null
+++ b/arch/x86/include/asm/unaccepted_memory.h
@@ -0,0 +1,24 @@
+#ifndef _ASM_X86_UNACCEPTED_MEMORY_H
+#define _ASM_X86_UNACCEPTED_MEMORY_H
+
+#include <linux/efi.h>
+#include <asm/tdx.h>
+
+static inline void arch_accept_memory(phys_addr_t start, phys_addr_t end)
+{
+	/* Platform-specific memory-acceptance call goes here */
+	if (cpu_feature_enabled(X86_FEATURE_TDX_GUEST)) {
+		if (!tdx_accept_memory(start, end))
+			panic("TDX: Failed to accept memory\n");
+	} else {
+		panic("Cannot accept memory: unknown platform\n");
+	}
+}
+
+static inline struct efi_unaccepted_memory *efi_get_unaccepted_table(void)
+{
+	if (efi.unaccepted == EFI_INVALID_TABLE_ADDR)
+		return NULL;
+	return __va(efi.unaccepted);
+}
+#endif
-- 
2.39.3

Re: [PATCHv14 0/9] mm, x86/cc, efi: Implement support for unaccepted memory

[Borislav Petkov]

On Tue, Jun 06, 2023 at 05:26:28PM +0300, Kirill A. Shutemov wrote:
> v14:
>  - Fix error handling in arch_accept_memory() (Tom);
>  - Address Borislav's feedback:
>    + code restructure;
>    + added/adjusted comments;

In file included from arch/x86/coco/tdx/tdx-shared.c:1:
./arch/x86/include/asm/tdx.h: In function ‘tdx_kvm_hypercall’:
./arch/x86/include/asm/tdx.h:70:17: error: ‘ENODEV’ undeclared (first use in this function)
   70 |         return -ENODEV;
      |                 ^~~~~~
./arch/x86/include/asm/tdx.h:70:17: note: each undeclared identifier is reported only once for each function it appears in
make[4]: *** [scripts/Makefile.build:252: arch/x86/coco/tdx/tdx-shared.o] Error 1
make[3]: *** [scripts/Makefile.build:494: arch/x86/coco/tdx] Error 2
make[2]: *** [scripts/Makefile.build:494: arch/x86/coco] Error 2
make[2]: *** Waiting for unfinished jobs....
make[1]: *** [scripts/Makefile.build:494: arch/x86] Error 2
make[1]: *** Waiting for unfinished jobs....
make: *** [Makefile:2026: .] Error 2

Not enough build tests ran?

$ grep INTEL_TDX_GUEST .config
CONFIG_INTEL_TDX_GUEST=y
$ grep KVM_GUEST .config
$

Why does that tdx_kvm_hypercall() thing even depend on CONFIG_KVM_GUEST?

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCHv14 0/9] mm, x86/cc, efi: Implement support for unaccepted memory

[Borislav Petkov]

On Tue, Jun 06, 2023 at 06:16:06PM +0200, Borislav Petkov wrote:
> Why does that tdx_kvm_hypercall() thing even depend on CONFIG_KVM_GUEST?

Amending with this, obviously:

---
diff --git a/arch/x86/include/asm/tdx.h b/arch/x86/include/asm/tdx.h
index 234197ec17e4..603e6d1e9d4a 100644
--- a/arch/x86/include/asm/tdx.h
+++ b/arch/x86/include/asm/tdx.h
@@ -5,6 +5,8 @@
 
 #include <linux/init.h>
 #include <linux/bits.h>
+
+#include <asm/errno.h>
 #include <asm/ptrace.h>
 #include <asm/shared/tdx.h>
 

-- 
Regards/Gruss,
    Boris.

https://people.kernel.org/tglx/notes-about-netiquette

Re: [PATCHv14 0/9] mm, x86/cc, efi: Implement support for unaccepted memory

[Kirill A. Shutemov]

On Tue, Jun 06, 2023 at 06:16:06PM +0200, Borislav Petkov wrote:
> On Tue, Jun 06, 2023 at 05:26:28PM +0300, Kirill A. Shutemov wrote:
> > v14:
> >  - Fix error handling in arch_accept_memory() (Tom);
> >  - Address Borislav's feedback:
> >    + code restructure;
> >    + added/adjusted comments;
> 
> In file included from arch/x86/coco/tdx/tdx-shared.c:1:
> ./arch/x86/include/asm/tdx.h: In function ‘tdx_kvm_hypercall’:
> ./arch/x86/include/asm/tdx.h:70:17: error: ‘ENODEV’ undeclared (first use in this function)
>    70 |         return -ENODEV;
>       |                 ^~~~~~
> ./arch/x86/include/asm/tdx.h:70:17: note: each undeclared identifier is reported only once for each function it appears in
> make[4]: *** [scripts/Makefile.build:252: arch/x86/coco/tdx/tdx-shared.o] Error 1
> make[3]: *** [scripts/Makefile.build:494: arch/x86/coco/tdx] Error 2
> make[2]: *** [scripts/Makefile.build:494: arch/x86/coco] Error 2
> make[2]: *** Waiting for unfinished jobs....
> make[1]: *** [scripts/Makefile.build:494: arch/x86] Error 2
> make[1]: *** Waiting for unfinished jobs....
> make: *** [Makefile:2026: .] Error 2
> 
> Not enough build tests ran?

Hm. I've got a lot of reports from 0day, but never this one.

Kai has posted patch that fixes it already:

https://lore.kernel.org/all/20230606034000.380270-1-kai.huang@intel.com/

> $ grep INTEL_TDX_GUEST .config
> CONFIG_INTEL_TDX_GUEST=y
> $ grep KVM_GUEST .config
> $
> 
> Why does that tdx_kvm_hypercall() thing even depend on CONFIG_KVM_GUEST?

Because nobody uses it otherwise. For instance, Hyper-V guest will no need
this KVM glue.

-- 
  Kiryl Shutsemau / Kirill A. Shutemov

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Mel Gorman]

On Tue, Jun 06, 2023 at 05:26:33PM +0300, Kirill A. Shutemov wrote:
> efi_config_parse_tables() reserves memory that holds unaccepted memory
> configuration table so it won't be reused by page allocator.
> 
> Core-mm requires few helpers to support unaccepted memory:
> 
>  - accept_memory() checks the range of addresses against the bitmap and
>    accept memory if needed.
> 
>  - range_contains_unaccepted_memory() checks if anything within the
>    range requires acceptance.
> 
> Architectural code has to provide efi_get_unaccepted_table() that
> returns pointer to the unaccepted memory configuration table.
> 
> arch_accept_memory() handles arch-specific part of memory acceptance.
> 
> Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
> Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>

By and large, this looks ok from the page allocator perspective as the
checks for unaccepted are mostly after watermark checks. However, if you
look in the initial fast path, you'll see this

        /* 
         * Forbid the first pass from falling back to types that fragment
         * memory until all local zones are considered.
         */     
        alloc_flags |= alloc_flags_nofragment(ac.preferred_zoneref->zone, gfp);

While checking watermarks should be fine from a functional perspective and
the fast paths are unaffected, there is a risk of premature fragmentation
until all memory has been accepted. Meeting watermarks does not necessarily
mean that fragmentation is avoided as pageblocks can get mixed while still
meeting watermarks. This will be tricky to detect in most cases so it may
be worth considering augmenting the watermark checks with a check in this
block for unaccepted memory

        /*
         * It's possible on a UMA machine to get through all zones that are
         * fragmented. If avoiding fragmentation, reset and try again.  */
        if (no_fallback) {
                alloc_flags &= ~ALLOC_NOFRAGMENT;
		goto retry;
	}

I think the watermark checks will still be needed because hypothetically
if 100% of allocations were MIGRATE_UNMOVABLE then there never would be
a request that fragments memory and memory would not be accepted.

It's not a blocker to the series, simply a suggestion for follow-on
work.

-- 
Mel Gorman
SUSE Labs

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Kirill A. Shutemov]

On Mon, Jul 03, 2023 at 02:25:18PM +0100, Mel Gorman wrote:
> On Tue, Jun 06, 2023 at 05:26:33PM +0300, Kirill A. Shutemov wrote:
> > efi_config_parse_tables() reserves memory that holds unaccepted memory
> > configuration table so it won't be reused by page allocator.
> > 
> > Core-mm requires few helpers to support unaccepted memory:
> > 
> >  - accept_memory() checks the range of addresses against the bitmap and
> >    accept memory if needed.
> > 
> >  - range_contains_unaccepted_memory() checks if anything within the
> >    range requires acceptance.
> > 
> > Architectural code has to provide efi_get_unaccepted_table() that
> > returns pointer to the unaccepted memory configuration table.
> > 
> > arch_accept_memory() handles arch-specific part of memory acceptance.
> > 
> > Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> > Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
> > Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
> 
> By and large, this looks ok from the page allocator perspective as the
> checks for unaccepted are mostly after watermark checks. However, if you
> look in the initial fast path, you'll see this
> 
>         /* 
>          * Forbid the first pass from falling back to types that fragment
>          * memory until all local zones are considered.
>          */     
>         alloc_flags |= alloc_flags_nofragment(ac.preferred_zoneref->zone, gfp);
> 
> While checking watermarks should be fine from a functional perspective and
> the fast paths are unaffected, there is a risk of premature fragmentation
> until all memory has been accepted. Meeting watermarks does not necessarily
> mean that fragmentation is avoided as pageblocks can get mixed while still
> meeting watermarks.

Could you elaborate on this scenario?

Current code checks the watermark, if it is met, try rmqueue().

If rmqueue() fails anyway, try to accept more pages and retry the zone if
it is successful.

I'm not sure how we can get to the 'if (no_fallback) {' case with any
unaccepted memory in the allowed zones.

I see that there's preferred_zoneref and spread_dirty_pages cases, but
unaccepted memory seems change nothing for them.

Hm?

-- 
  Kiryl Shutsemau / Kirill A. Shutemov

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Mel Gorman]

On Tue, Jul 04, 2023 at 05:37:40PM +0300, Kirill A. Shutemov wrote:
> On Mon, Jul 03, 2023 at 02:25:18PM +0100, Mel Gorman wrote:
> > On Tue, Jun 06, 2023 at 05:26:33PM +0300, Kirill A. Shutemov wrote:
> > > efi_config_parse_tables() reserves memory that holds unaccepted memory
> > > configuration table so it won't be reused by page allocator.
> > > 
> > > Core-mm requires few helpers to support unaccepted memory:
> > > 
> > >  - accept_memory() checks the range of addresses against the bitmap and
> > >    accept memory if needed.
> > > 
> > >  - range_contains_unaccepted_memory() checks if anything within the
> > >    range requires acceptance.
> > > 
> > > Architectural code has to provide efi_get_unaccepted_table() that
> > > returns pointer to the unaccepted memory configuration table.
> > > 
> > > arch_accept_memory() handles arch-specific part of memory acceptance.
> > > 
> > > Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> > > Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
> > > Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
> > 
> > By and large, this looks ok from the page allocator perspective as the
> > checks for unaccepted are mostly after watermark checks. However, if you
> > look in the initial fast path, you'll see this
> > 
> >         /* 
> >          * Forbid the first pass from falling back to types that fragment
> >          * memory until all local zones are considered.
> >          */     
> >         alloc_flags |= alloc_flags_nofragment(ac.preferred_zoneref->zone, gfp);
> > 
> > While checking watermarks should be fine from a functional perspective and
> > the fast paths are unaffected, there is a risk of premature fragmentation
> > until all memory has been accepted. Meeting watermarks does not necessarily
> > mean that fragmentation is avoided as pageblocks can get mixed while still
> > meeting watermarks.
> 
> Could you elaborate on this scenario?
> 
> Current code checks the watermark, if it is met, try rmqueue().
> 
> If rmqueue() fails anyway, try to accept more pages and retry the zone if
> it is successful.
> 
> I'm not sure how we can get to the 'if (no_fallback) {' case with any
> unaccepted memory in the allowed zones.
> 

Lets take an extreme example and assume that the low watermark is lower
than 2MB (one pageblock). Just before the watermark is reached (free
count between 1MB and 2MB), it is unlikely that all free pages are within
pageblocks of the same migratetype (e.g. MIGRATE_MOVABLE). If there is an
allocation near the watermark of a different type (e.g. MIGRATE_UNMOVABLE)
then the page allocation could fallback to a different pageblock and now
it is mixed.  It's a condition that is only obvious if you are explicitly
checking for it via tracepoints.  This can happen in the normal case, but
unaccepted memory makes it worse because the "pageblock mixing" could have
been avoided if the "no_fallback" case accepted at least one new pageblock
instead of mixing pageblocks.

That is an extreme example but the same logic applies when the free
count is at or near MIGRATE_TYPES*pageblock_nr_pages as it is not
guaranteed that the pageblocks with free pages are a migratetype that
matches the allocation request.

Hence, it may be more robust from a fragmentation perspective if
ALLOC_NOFRAGMENT requests accept memory if it is available and retries
before clearing ALLOC_NOFRAGMENT and mixing pageblocks before the watermarks
are reached.

> I see that there's preferred_zoneref and spread_dirty_pages cases, but
> unaccepted memory seems change nothing for them.
> 

preferred_zoneref is about premature zone exhaustion and
spread_dirty_pages is about avoiding premature stalls on a node/zone due
to an imbalance in the number of pages waiting for writeback to
complete. There is an arguement to be made that they also should accept
memory but it's less clear how much of a problem this is. Both are very
obvious when they "fail" and likely are covered by the existing
watermark checks. Premature pageblock mixing is more subtle as the final
impact (root cause of a premature THP allocation failure) is harder to
detect.

-- 
Mel Gorman
SUSE Labs

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Michael Roth]

On Tue, Jun 06, 2023 at 05:26:33PM +0300, Kirill A. Shutemov wrote:
> efi_config_parse_tables() reserves memory that holds unaccepted memory
> configuration table so it won't be reused by page allocator.
> 
> Core-mm requires few helpers to support unaccepted memory:
> 
>  - accept_memory() checks the range of addresses against the bitmap and
>    accept memory if needed.
> 
>  - range_contains_unaccepted_memory() checks if anything within the
>    range requires acceptance.
> 
> Architectural code has to provide efi_get_unaccepted_table() that
> returns pointer to the unaccepted memory configuration table.
> 
> arch_accept_memory() handles arch-specific part of memory acceptance.
> 
> Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
> Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
> ---
>  arch/x86/platform/efi/efi.c              |   3 +
>  drivers/firmware/efi/Makefile            |   1 +
>  drivers/firmware/efi/efi.c               |  25 +++++
>  drivers/firmware/efi/unaccepted_memory.c | 112 +++++++++++++++++++++++
>  include/linux/efi.h                      |   1 +
>  5 files changed, 142 insertions(+)
>  create mode 100644 drivers/firmware/efi/unaccepted_memory.c
> 
> diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
> new file mode 100644
> index 000000000000..08a9a843550a
> --- /dev/null
> +++ b/drivers/firmware/efi/unaccepted_memory.c
> @@ -0,0 +1,112 @@
> +// SPDX-License-Identifier: GPL-2.0-only
> +
> +#include <linux/efi.h>
> +#include <linux/memblock.h>
> +#include <linux/spinlock.h>
> +#include <asm/unaccepted_memory.h>
> +
> +/* Protects unaccepted memory bitmap */
> +static DEFINE_SPINLOCK(unaccepted_memory_lock);
> +
> +/*
> + * accept_memory() -- Consult bitmap and accept the memory if needed.
> + *
> + * Only memory that is explicitly marked as unaccepted in the bitmap requires
> + * an action. All the remaining memory is implicitly accepted and doesn't need
> + * acceptance.
> + *
> + * No need to accept:
> + *  - anything if the system has no unaccepted table;
> + *  - memory that is below phys_base;
> + *  - memory that is above the memory that addressable by the bitmap;
> + */
> +void accept_memory(phys_addr_t start, phys_addr_t end)
> +{
> +	struct efi_unaccepted_memory *unaccepted;
> +	unsigned long range_start, range_end;
> +	unsigned long flags;
> +	u64 unit_size;
> +
> +	unaccepted = efi_get_unaccepted_table();
> +	if (!unaccepted)
> +		return;
> +
> +	unit_size = unaccepted->unit_size;
> +
> +	/*
> +	 * Only care for the part of the range that is represented
> +	 * in the bitmap.
> +	 */
> +	if (start < unaccepted->phys_base)
> +		start = unaccepted->phys_base;
> +	if (end < unaccepted->phys_base)
> +		return;
> +
> +	/* Translate to offsets from the beginning of the bitmap */
> +	start -= unaccepted->phys_base;
> +	end -= unaccepted->phys_base;
> +
> +	/* Make sure not to overrun the bitmap */
> +	if (end > unaccepted->size * unit_size * BITS_PER_BYTE)
> +		end = unaccepted->size * unit_size * BITS_PER_BYTE;
> +
> +	range_start = start / unit_size;
> +
> +	spin_lock_irqsave(&unaccepted_memory_lock, flags);
> +	for_each_set_bitrange_from(range_start, range_end, unaccepted->bitmap,
> +				   DIV_ROUND_UP(end, unit_size)) {
> +		unsigned long phys_start, phys_end;
> +		unsigned long len = range_end - range_start;
> +
> +		phys_start = range_start * unit_size + unaccepted->phys_base;
> +		phys_end = range_end * unit_size + unaccepted->phys_base;
> +
> +		arch_accept_memory(phys_start, phys_end);
> +		bitmap_clear(unaccepted->bitmap, range_start, len);
> +	}
> +	spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
> +}

While testing SNP guests running today's tip/master (ef19bc9dddc3) I ran
into what seems to be fairly significant lock contention due to the
unaccepted_memory_lock spinlock above, which results in a constant stream
of soft-lockups until the workload gets all its memory accepted/faulted
in if the guest has around 16+ vCPUs.

I've included the guest dmesg traces I was seeing below.

In this case I was running a 32 vCPU guest with 200GB of memory running on
a 256 thread EPYC (Milan) system, and can trigger the above situation fairly
reliably by running the following workload in a freshly-booted guests:

  stress --vm 32 --vm-bytes 5G --vm-keep

Scaling up the number of stress threads and vCPUs should make it easier
to reproduce.

Other than unresponsiveness/lockup messages until the memory is accepted,
the guest seems to continue running fine, but for large guests where
unaccepted memory is more likely to be useful, it seems like it could be
an issue, especially when consider 100+ vCPU guests.

-Mike

[  105.753073] watchdog: BUG: soft lockup - CPU#3 stuck for 27s! [stress:1590]
[  105.756109] Modules linked in: btrfs(E) blake2b_generic(E) raid10(E) raid456(E) async_raid6_recov(E) async_memcpy(E) async_pq(E) async_xor(E) async_tx(E) xor(E) raid6_pq(E) libcrc32c(E) raid1(E) raid0(E) multipath(E) linear(E) crc32_pclmul(E) virtio_net(E) i2c_i801(E) net_failover(E) i2c_smbus(E) psmouse(E) failover(E) virtio_scsi(E) lpc_ich(E)
[  105.771644] irq event stamp: 392274
[  105.773852] hardirqs last  enabled at (392273): [<ffffffff9fbddf9a>] _raw_spin_unlock_irqrestore+0x5a/0x70
[  105.780058] hardirqs last disabled at (392274): [<ffffffff9fbddca9>] _raw_spin_lock_irqsave+0x69/0x70
[  105.785486] softirqs last  enabled at (392158): [<ffffffff9fbdf663>] __do_softirq+0x2a3/0x357
[  105.790627] softirqs last disabled at (392149): [<ffffffff9ecd0f5c>] irq_exit_rcu+0x8c/0xb0
[  105.795972] CPU: 3 PID: 1590 Comm: stress Tainted: G            EL     6.6.0-rc5-snp-guest-v10n+ #1
[  105.802416] Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS unknown 2/2/2022
[  105.807040] RIP: 0010:_raw_spin_unlock_irqrestore+0x45/0x70
[  105.810327] Code: b1 b1 18 ff 4c 89 e7 e8 79 e7 18 ff 81 e3 00 02 00 00 75 26 9c 58 0f 1f 40 00 f6 c4 02 75 22 48 85 db 74 06 fb 0f 1f 44 00 00 <65> ff 0d fc 5d 46 60 5b 41 5c 5d c3 cc cc cc cc e8 c6 9f 28 ff eb
[  105.818749] RSP: 0000:ffffc9000585faa8 EFLAGS: 00000206
[  105.821155] RAX: 0000000000000046 RBX: 0000000000000200 RCX: 00000000000028ce
[  105.825258] RDX: ffffffff9f8c7830 RSI: ffffffff9fbddf9a RDI: ffffffff9fbddf9a
[  105.828287] RBP: ffffc9000585fab8 R08: 0000000000000000 R09: 0000000000000000
[  105.831322] R10: 000000ffffffffff R11: 0000000000000000 R12: ffffffffa091a160
[  105.834713] R13: 00000000000028cd R14: ffff88807c459030 R15: ffff88807c459018
[  105.838230] FS:  00007fa4a8ee8740(0000) GS:ffff88b1aff80000(0000) knlGS:0000000000000000
[  105.841677] CS:  0010 DS: 0000 ES: 0000 CR0: 0000000080050033
[  105.844115] CR2: 00007fa3719ec010 CR3: 000800010f7c2003 CR4: 0000000000770ef0
[  105.847160] PKRU: 55555554
[  105.848490] Call Trace:
[  105.849634]  <IRQ>
[  105.850517]  ? show_regs+0x68/0x70
[  105.851986]  ? watchdog_timer_fn+0x1dc/0x240
[  105.853868]  ? __pfx_watchdog_timer_fn+0x10/0x10
[  105.855827]  ? __hrtimer_run_queues+0x1c3/0x340
[  105.857821]  ? hrtimer_interrupt+0x109/0x240
[  105.859673]  ? __sysvec_apic_timer_interrupt+0x67/0x170
[  105.861941]  ? sysvec_apic_timer_interrupt+0x7b/0x90
[  105.864048]  </IRQ>
[  105.864981]  <TASK>
[  105.865901]  ? asm_sysvec_apic_timer_interrupt+0x1b/0x20
[  105.868110]  ? accept_memory+0x150/0x170
[  105.869795]  ? _raw_spin_unlock_irqrestore+0x5a/0x70
[  105.871885]  ? _raw_spin_unlock_irqrestore+0x5a/0x70
[  105.873966]  ? _raw_spin_unlock_irqrestore+0x45/0x70
[  105.876047]  accept_memory+0x150/0x170
[  105.877665]  try_to_accept_memory+0x134/0x1b0
[  105.879532]  get_page_from_freelist+0xa3e/0x1370
[  105.881479]  ? lock_acquire+0xd8/0x2b0
[  105.883045]  __alloc_pages+0x1b7/0x390
[  105.884619]  __folio_alloc+0x1b/0x50
[  105.886151]  ? policy_node+0x57/0x70
[  105.887673]  vma_alloc_folio+0xa6/0x360
[  105.889325]  do_pte_missing+0x1a5/0x8b0
[  105.890948]  __handle_mm_fault+0x75e/0xda0
[  105.892693]  handle_mm_fault+0xe1/0x2d0
[  105.894304]  do_user_addr_fault+0x1ce/0xa40
[  105.896050]  ? exit_to_user_mode_prepare+0xa4/0x230
[  105.898113]  exc_page_fault+0x84/0x200
[  105.899693]  asm_exc_page_fault+0x27/0x30
[  105.901405] RIP: 0033:0x55ebd4602cf0
[  105.902926] Code: 8b 54 24 0c 31 c0 85 d2 0f 94 c0 89 04 24 41 83 fd 02 0f 8f 3e 02 00 00 48 85 ed 48 89 d8 7e 1b 66 2e 0f 1f 84 00 00 00 00 00 <c6> 00 5a 4c 01 f8 48 89 c2 48 29 da 48 39 d5 7f ef 49 83 fc 00 0f
[  105.910671] RSP: 002b:00007ffebfede470 EFLAGS: 00010206
[  105.912897] RAX: 00007fa3719ec010 RBX: 00007fa3683ff010 RCX: 00007fa3683ff010
[  105.915859] RDX: 00000000095ed000 RSI: 0000000140001000 RDI: 0000000000000000
[  105.918853] RBP: 0000000140000000 R08: 00000000ffffffff R09: 0000000000000000
[  105.921836] R10: 0000000000000022 R11: 0000000000000246 R12: ffffffffffffffff
[  105.924810] R13: 0000000000000002 R14: fffffffffffff000 R15: 0000000000001000
[  105.927771]  </TASK>
[  110.367774] watchdog: BUG: soft lockup - CPU#15 stuck for 32s! [stress:1594]
[  110.369415] watchdog: BUG: soft lockup - CPU#13 stuck for 36s! [stress:1602]

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Kirill A. Shutemov]

On Tue, Oct 10, 2023 at 04:05:18PM -0500, Michael Roth wrote:
> On Tue, Jun 06, 2023 at 05:26:33PM +0300, Kirill A. Shutemov wrote:
> > efi_config_parse_tables() reserves memory that holds unaccepted memory
> > configuration table so it won't be reused by page allocator.
> > 
> > Core-mm requires few helpers to support unaccepted memory:
> > 
> >  - accept_memory() checks the range of addresses against the bitmap and
> >    accept memory if needed.
> > 
> >  - range_contains_unaccepted_memory() checks if anything within the
> >    range requires acceptance.
> > 
> > Architectural code has to provide efi_get_unaccepted_table() that
> > returns pointer to the unaccepted memory configuration table.
> > 
> > arch_accept_memory() handles arch-specific part of memory acceptance.
> > 
> > Signed-off-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
> > Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
> > Reviewed-by: Tom Lendacky <thomas.lendacky@amd.com>
> > ---
> >  arch/x86/platform/efi/efi.c              |   3 +
> >  drivers/firmware/efi/Makefile            |   1 +
> >  drivers/firmware/efi/efi.c               |  25 +++++
> >  drivers/firmware/efi/unaccepted_memory.c | 112 +++++++++++++++++++++++
> >  include/linux/efi.h                      |   1 +
> >  5 files changed, 142 insertions(+)
> >  create mode 100644 drivers/firmware/efi/unaccepted_memory.c
> > 
> > diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
> > new file mode 100644
> > index 000000000000..08a9a843550a
> > --- /dev/null
> > +++ b/drivers/firmware/efi/unaccepted_memory.c
> > @@ -0,0 +1,112 @@
> > +// SPDX-License-Identifier: GPL-2.0-only
> > +
> > +#include <linux/efi.h>
> > +#include <linux/memblock.h>
> > +#include <linux/spinlock.h>
> > +#include <asm/unaccepted_memory.h>
> > +
> > +/* Protects unaccepted memory bitmap */
> > +static DEFINE_SPINLOCK(unaccepted_memory_lock);
> > +
> > +/*
> > + * accept_memory() -- Consult bitmap and accept the memory if needed.
> > + *
> > + * Only memory that is explicitly marked as unaccepted in the bitmap requires
> > + * an action. All the remaining memory is implicitly accepted and doesn't need
> > + * acceptance.
> > + *
> > + * No need to accept:
> > + *  - anything if the system has no unaccepted table;
> > + *  - memory that is below phys_base;
> > + *  - memory that is above the memory that addressable by the bitmap;
> > + */
> > +void accept_memory(phys_addr_t start, phys_addr_t end)
> > +{
> > +	struct efi_unaccepted_memory *unaccepted;
> > +	unsigned long range_start, range_end;
> > +	unsigned long flags;
> > +	u64 unit_size;
> > +
> > +	unaccepted = efi_get_unaccepted_table();
> > +	if (!unaccepted)
> > +		return;
> > +
> > +	unit_size = unaccepted->unit_size;
> > +
> > +	/*
> > +	 * Only care for the part of the range that is represented
> > +	 * in the bitmap.
> > +	 */
> > +	if (start < unaccepted->phys_base)
> > +		start = unaccepted->phys_base;
> > +	if (end < unaccepted->phys_base)
> > +		return;
> > +
> > +	/* Translate to offsets from the beginning of the bitmap */
> > +	start -= unaccepted->phys_base;
> > +	end -= unaccepted->phys_base;
> > +
> > +	/* Make sure not to overrun the bitmap */
> > +	if (end > unaccepted->size * unit_size * BITS_PER_BYTE)
> > +		end = unaccepted->size * unit_size * BITS_PER_BYTE;
> > +
> > +	range_start = start / unit_size;
> > +
> > +	spin_lock_irqsave(&unaccepted_memory_lock, flags);
> > +	for_each_set_bitrange_from(range_start, range_end, unaccepted->bitmap,
> > +				   DIV_ROUND_UP(end, unit_size)) {
> > +		unsigned long phys_start, phys_end;
> > +		unsigned long len = range_end - range_start;
> > +
> > +		phys_start = range_start * unit_size + unaccepted->phys_base;
> > +		phys_end = range_end * unit_size + unaccepted->phys_base;
> > +
> > +		arch_accept_memory(phys_start, phys_end);
> > +		bitmap_clear(unaccepted->bitmap, range_start, len);
> > +	}
> > +	spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
> > +}
> 
> While testing SNP guests running today's tip/master (ef19bc9dddc3) I ran
> into what seems to be fairly significant lock contention due to the
> unaccepted_memory_lock spinlock above, which results in a constant stream
> of soft-lockups until the workload gets all its memory accepted/faulted
> in if the guest has around 16+ vCPUs.
> 
> I've included the guest dmesg traces I was seeing below.
> 
> In this case I was running a 32 vCPU guest with 200GB of memory running on
> a 256 thread EPYC (Milan) system, and can trigger the above situation fairly
> reliably by running the following workload in a freshly-booted guests:
> 
>   stress --vm 32 --vm-bytes 5G --vm-keep
> 
> Scaling up the number of stress threads and vCPUs should make it easier
> to reproduce.
> 
> Other than unresponsiveness/lockup messages until the memory is accepted,
> the guest seems to continue running fine, but for large guests where
> unaccepted memory is more likely to be useful, it seems like it could be
> an issue, especially when consider 100+ vCPU guests.

Okay, sorry for delay. It took time to reproduce it with TDX.

I will look what can be done.

-- 
  Kiryl Shutsemau / Kirill A. Shutemov

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Kirill A. Shutemov]

On Fri, Oct 13, 2023 at 03:33:58PM +0300, Kirill A. Shutemov wrote:
> > While testing SNP guests running today's tip/master (ef19bc9dddc3) I ran
> > into what seems to be fairly significant lock contention due to the
> > unaccepted_memory_lock spinlock above, which results in a constant stream
> > of soft-lockups until the workload gets all its memory accepted/faulted
> > in if the guest has around 16+ vCPUs.
> > 
> > I've included the guest dmesg traces I was seeing below.
> > 
> > In this case I was running a 32 vCPU guest with 200GB of memory running on
> > a 256 thread EPYC (Milan) system, and can trigger the above situation fairly
> > reliably by running the following workload in a freshly-booted guests:
> > 
> >   stress --vm 32 --vm-bytes 5G --vm-keep
> > 
> > Scaling up the number of stress threads and vCPUs should make it easier
> > to reproduce.
> > 
> > Other than unresponsiveness/lockup messages until the memory is accepted,
> > the guest seems to continue running fine, but for large guests where
> > unaccepted memory is more likely to be useful, it seems like it could be
> > an issue, especially when consider 100+ vCPU guests.
> 
> Okay, sorry for delay. It took time to reproduce it with TDX.
> 
> I will look what can be done.

Could you check if the patch below helps?

diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
index 853f7dc3c21d..591da3f368fa 100644
--- a/drivers/firmware/efi/unaccepted_memory.c
+++ b/drivers/firmware/efi/unaccepted_memory.c
@@ -8,6 +8,14 @@
 /* Protects unaccepted memory bitmap */
 static DEFINE_SPINLOCK(unaccepted_memory_lock);
 
+struct accept_range {
+	struct list_head list;
+	unsigned long start;
+	unsigned long end;
+};
+
+static LIST_HEAD(accepting_list);
+
 /*
  * accept_memory() -- Consult bitmap and accept the memory if needed.
  *
@@ -24,6 +32,7 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
 {
 	struct efi_unaccepted_memory *unaccepted;
 	unsigned long range_start, range_end;
+	struct accept_range range, *entry;
 	unsigned long flags;
 	u64 unit_size;
 
@@ -80,7 +89,25 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
 
 	range_start = start / unit_size;
 
+	range.start = start;
+	range.end = end;
+retry:
 	spin_lock_irqsave(&unaccepted_memory_lock, flags);
+
+	list_for_each_entry(entry, &accepting_list, list) {
+		if (entry->end < start)
+			continue;
+		if (entry->start > end)
+			continue;
+		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
+
+		/* Somebody else accepting the range */
+		cpu_relax();
+		goto retry;
+	}
+
+	list_add(&range.list, &accepting_list);
+
 	for_each_set_bitrange_from(range_start, range_end, unaccepted->bitmap,
 				   DIV_ROUND_UP(end, unit_size)) {
 		unsigned long phys_start, phys_end;
@@ -89,9 +116,15 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
 		phys_start = range_start * unit_size + unaccepted->phys_base;
 		phys_end = range_end * unit_size + unaccepted->phys_base;
 
+		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
+
 		arch_accept_memory(phys_start, phys_end);
+
+		spin_lock_irqsave(&unaccepted_memory_lock, flags);
 		bitmap_clear(unaccepted->bitmap, range_start, len);
 	}
+
+	list_del(&range.list);
 	spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
 }
 
-- 
  Kiryl Shutsemau / Kirill A. Shutemov

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Vlastimil Babka]

On 10/13/23 18:22, Kirill A. Shutemov wrote:
> On Fri, Oct 13, 2023 at 03:33:58PM +0300, Kirill A. Shutemov wrote:
>> > While testing SNP guests running today's tip/master (ef19bc9dddc3) I ran
>> > into what seems to be fairly significant lock contention due to the
>> > unaccepted_memory_lock spinlock above, which results in a constant stream
>> > of soft-lockups until the workload gets all its memory accepted/faulted
>> > in if the guest has around 16+ vCPUs.
>> > 
>> > I've included the guest dmesg traces I was seeing below.
>> > 
>> > In this case I was running a 32 vCPU guest with 200GB of memory running on
>> > a 256 thread EPYC (Milan) system, and can trigger the above situation fairly
>> > reliably by running the following workload in a freshly-booted guests:
>> > 
>> >   stress --vm 32 --vm-bytes 5G --vm-keep
>> > 
>> > Scaling up the number of stress threads and vCPUs should make it easier
>> > to reproduce.
>> > 
>> > Other than unresponsiveness/lockup messages until the memory is accepted,
>> > the guest seems to continue running fine, but for large guests where
>> > unaccepted memory is more likely to be useful, it seems like it could be
>> > an issue, especially when consider 100+ vCPU guests.
>> 
>> Okay, sorry for delay. It took time to reproduce it with TDX.
>> 
>> I will look what can be done.
> 
> Could you check if the patch below helps?
> 
> diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
> index 853f7dc3c21d..591da3f368fa 100644
> --- a/drivers/firmware/efi/unaccepted_memory.c
> +++ b/drivers/firmware/efi/unaccepted_memory.c
> @@ -8,6 +8,14 @@
>  /* Protects unaccepted memory bitmap */
>  static DEFINE_SPINLOCK(unaccepted_memory_lock);
>  
> +struct accept_range {
> +	struct list_head list;
> +	unsigned long start;
> +	unsigned long end;
> +};
> +
> +static LIST_HEAD(accepting_list);
> +
>  /*
>   * accept_memory() -- Consult bitmap and accept the memory if needed.
>   *
> @@ -24,6 +32,7 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
>  {
>  	struct efi_unaccepted_memory *unaccepted;
>  	unsigned long range_start, range_end;
> +	struct accept_range range, *entry;
>  	unsigned long flags;
>  	u64 unit_size;
>  
> @@ -80,7 +89,25 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
>  
>  	range_start = start / unit_size;
>  
> +	range.start = start;
> +	range.end = end;
> +retry:
>  	spin_lock_irqsave(&unaccepted_memory_lock, flags);
> +
> +	list_for_each_entry(entry, &accepting_list, list) {
> +		if (entry->end < start)
> +			continue;
> +		if (entry->start > end)
> +			continue;
> +		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
> +
> +		/* Somebody else accepting the range */
> +		cpu_relax();

Should this be rather cond_resched()? I think cpu_relax() isn't enough to
prevent soft lockups.
Although IIUC hitting this should be rare, as the contending tasks will pick
different ranges via try_to_accept_memory_one(), right?

> +		goto retry;
> +	}
> +
> +	list_add(&range.list, &accepting_list);
> +
>  	for_each_set_bitrange_from(range_start, range_end, unaccepted->bitmap,
>  				   DIV_ROUND_UP(end, unit_size)) {
>  		unsigned long phys_start, phys_end;
> @@ -89,9 +116,15 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
>  		phys_start = range_start * unit_size + unaccepted->phys_base;
>  		phys_end = range_end * unit_size + unaccepted->phys_base;
>  
> +		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
> +
>  		arch_accept_memory(phys_start, phys_end);
> +
> +		spin_lock_irqsave(&unaccepted_memory_lock, flags);
>  		bitmap_clear(unaccepted->bitmap, range_start, len);
>  	}
> +
> +	list_del(&range.list);
>  	spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
>  }
>  

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Kirill A. Shutemov]

On Fri, Oct 13, 2023 at 06:44:45PM +0200, Vlastimil Babka wrote:
> On 10/13/23 18:22, Kirill A. Shutemov wrote:
> > On Fri, Oct 13, 2023 at 03:33:58PM +0300, Kirill A. Shutemov wrote:
> >> > While testing SNP guests running today's tip/master (ef19bc9dddc3) I ran
> >> > into what seems to be fairly significant lock contention due to the
> >> > unaccepted_memory_lock spinlock above, which results in a constant stream
> >> > of soft-lockups until the workload gets all its memory accepted/faulted
> >> > in if the guest has around 16+ vCPUs.
> >> > 
> >> > I've included the guest dmesg traces I was seeing below.
> >> > 
> >> > In this case I was running a 32 vCPU guest with 200GB of memory running on
> >> > a 256 thread EPYC (Milan) system, and can trigger the above situation fairly
> >> > reliably by running the following workload in a freshly-booted guests:
> >> > 
> >> >   stress --vm 32 --vm-bytes 5G --vm-keep
> >> > 
> >> > Scaling up the number of stress threads and vCPUs should make it easier
> >> > to reproduce.
> >> > 
> >> > Other than unresponsiveness/lockup messages until the memory is accepted,
> >> > the guest seems to continue running fine, but for large guests where
> >> > unaccepted memory is more likely to be useful, it seems like it could be
> >> > an issue, especially when consider 100+ vCPU guests.
> >> 
> >> Okay, sorry for delay. It took time to reproduce it with TDX.
> >> 
> >> I will look what can be done.
> > 
> > Could you check if the patch below helps?
> > 
> > diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
> > index 853f7dc3c21d..591da3f368fa 100644
> > --- a/drivers/firmware/efi/unaccepted_memory.c
> > +++ b/drivers/firmware/efi/unaccepted_memory.c
> > @@ -8,6 +8,14 @@
> >  /* Protects unaccepted memory bitmap */
> >  static DEFINE_SPINLOCK(unaccepted_memory_lock);
> >  
> > +struct accept_range {
> > +	struct list_head list;
> > +	unsigned long start;
> > +	unsigned long end;
> > +};
> > +
> > +static LIST_HEAD(accepting_list);
> > +
> >  /*
> >   * accept_memory() -- Consult bitmap and accept the memory if needed.
> >   *
> > @@ -24,6 +32,7 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
> >  {
> >  	struct efi_unaccepted_memory *unaccepted;
> >  	unsigned long range_start, range_end;
> > +	struct accept_range range, *entry;
> >  	unsigned long flags;
> >  	u64 unit_size;
> >  
> > @@ -80,7 +89,25 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
> >  
> >  	range_start = start / unit_size;
> >  
> > +	range.start = start;
> > +	range.end = end;
> > +retry:
> >  	spin_lock_irqsave(&unaccepted_memory_lock, flags);
> > +
> > +	list_for_each_entry(entry, &accepting_list, list) {
> > +		if (entry->end < start)
> > +			continue;
> > +		if (entry->start > end)
> > +			continue;
> > +		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
> > +
> > +		/* Somebody else accepting the range */
> > +		cpu_relax();
> 
> Should this be rather cond_resched()? I think cpu_relax() isn't enough to
> prevent soft lockups.

Right. For some reason, I thought we cannot call cond_resched() from
atomic context (we sometimes get there from atomic context), but we can.

> Although IIUC hitting this should be rare, as the contending tasks will pick
> different ranges via try_to_accept_memory_one(), right?

Yes, it should be rare.

Generally, with exception of memblock, we accept all memory with MAX_ORDER
chunks. As long as unit_size <= MAX_ORDER page allocator should never
trigger the conflict as the caller owns full range to accept.

I will test the idea with larger unit_size to see how it behaves.

-- 
  Kiryl Shutsemau / Kirill A. Shutemov

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Tom Lendacky]

On 10/13/23 11:22, Kirill A. Shutemov wrote:
> On Fri, Oct 13, 2023 at 03:33:58PM +0300, Kirill A. Shutemov wrote:
>>> While testing SNP guests running today's tip/master (ef19bc9dddc3) I ran
>>> into what seems to be fairly significant lock contention due to the
>>> unaccepted_memory_lock spinlock above, which results in a constant stream
>>> of soft-lockups until the workload gets all its memory accepted/faulted
>>> in if the guest has around 16+ vCPUs.
>>>
>>> I've included the guest dmesg traces I was seeing below.
>>>
>>> In this case I was running a 32 vCPU guest with 200GB of memory running on
>>> a 256 thread EPYC (Milan) system, and can trigger the above situation fairly
>>> reliably by running the following workload in a freshly-booted guests:
>>>
>>>    stress --vm 32 --vm-bytes 5G --vm-keep
>>>
>>> Scaling up the number of stress threads and vCPUs should make it easier
>>> to reproduce.
>>>
>>> Other than unresponsiveness/lockup messages until the memory is accepted,
>>> the guest seems to continue running fine, but for large guests where
>>> unaccepted memory is more likely to be useful, it seems like it could be
>>> an issue, especially when consider 100+ vCPU guests.
>>
>> Okay, sorry for delay. It took time to reproduce it with TDX.
>>
>> I will look what can be done.
> 
> Could you check if the patch below helps?
> 
> diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
> index 853f7dc3c21d..591da3f368fa 100644
> --- a/drivers/firmware/efi/unaccepted_memory.c
> +++ b/drivers/firmware/efi/unaccepted_memory.c
> @@ -8,6 +8,14 @@
>   /* Protects unaccepted memory bitmap */
>   static DEFINE_SPINLOCK(unaccepted_memory_lock);
>   
> +struct accept_range {
> +	struct list_head list;
> +	unsigned long start;
> +	unsigned long end;
> +};
> +
> +static LIST_HEAD(accepting_list);
> +
>   /*
>    * accept_memory() -- Consult bitmap and accept the memory if needed.
>    *
> @@ -24,6 +32,7 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
>   {
>   	struct efi_unaccepted_memory *unaccepted;
>   	unsigned long range_start, range_end;
> +	struct accept_range range, *entry;
>   	unsigned long flags;
>   	u64 unit_size;
>   
> @@ -80,7 +89,25 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
>   
>   	range_start = start / unit_size;
>   
> +	range.start = start;
> +	range.end = end;
> +retry:
>   	spin_lock_irqsave(&unaccepted_memory_lock, flags);
> +
> +	list_for_each_entry(entry, &accepting_list, list) {
> +		if (entry->end < start)
> +			continue;
> +		if (entry->start > end)

Should this be a >= check since start and end are page aligned values?

> +			continue;
> +		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
> +
> +		/* Somebody else accepting the range */
> +		cpu_relax();
> +		goto retry;

Could you set some kind of flag here so that ...

> +	}
> +

... once you get here, that means that area was accepted and removed from 
the list, so I think you could just drop the lock and exit now, right? 
Because at that point the bitmap will have been updated and you wouldn't 
be accepting any memory anyway?

Thanks,
Tom

> +	list_add(&range.list, &accepting_list);
> +
>   	for_each_set_bitrange_from(range_start, range_end, unaccepted->bitmap,
>   				   DIV_ROUND_UP(end, unit_size)) {
>   		unsigned long phys_start, phys_end;
> @@ -89,9 +116,15 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
>   		phys_start = range_start * unit_size + unaccepted->phys_base;
>   		phys_end = range_end * unit_size + unaccepted->phys_base;
>   
> +		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
> +
>   		arch_accept_memory(phys_start, phys_end);
> +
> +		spin_lock_irqsave(&unaccepted_memory_lock, flags);
>   		bitmap_clear(unaccepted->bitmap, range_start, len);
>   	}
> +
> +	list_del(&range.list);
>   	spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
>   }
>   

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Kirill A. Shutemov]

On Fri, Oct 13, 2023 at 12:45:20PM -0500, Tom Lendacky wrote:
> On 10/13/23 11:22, Kirill A. Shutemov wrote:
> > On Fri, Oct 13, 2023 at 03:33:58PM +0300, Kirill A. Shutemov wrote:
> > > > While testing SNP guests running today's tip/master (ef19bc9dddc3) I ran
> > > > into what seems to be fairly significant lock contention due to the
> > > > unaccepted_memory_lock spinlock above, which results in a constant stream
> > > > of soft-lockups until the workload gets all its memory accepted/faulted
> > > > in if the guest has around 16+ vCPUs.
> > > > 
> > > > I've included the guest dmesg traces I was seeing below.
> > > > 
> > > > In this case I was running a 32 vCPU guest with 200GB of memory running on
> > > > a 256 thread EPYC (Milan) system, and can trigger the above situation fairly
> > > > reliably by running the following workload in a freshly-booted guests:
> > > > 
> > > >    stress --vm 32 --vm-bytes 5G --vm-keep
> > > > 
> > > > Scaling up the number of stress threads and vCPUs should make it easier
> > > > to reproduce.
> > > > 
> > > > Other than unresponsiveness/lockup messages until the memory is accepted,
> > > > the guest seems to continue running fine, but for large guests where
> > > > unaccepted memory is more likely to be useful, it seems like it could be
> > > > an issue, especially when consider 100+ vCPU guests.
> > > 
> > > Okay, sorry for delay. It took time to reproduce it with TDX.
> > > 
> > > I will look what can be done.
> > 
> > Could you check if the patch below helps?
> > 
> > diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
> > index 853f7dc3c21d..591da3f368fa 100644
> > --- a/drivers/firmware/efi/unaccepted_memory.c
> > +++ b/drivers/firmware/efi/unaccepted_memory.c
> > @@ -8,6 +8,14 @@
> >   /* Protects unaccepted memory bitmap */
> >   static DEFINE_SPINLOCK(unaccepted_memory_lock);
> > +struct accept_range {
> > +	struct list_head list;
> > +	unsigned long start;
> > +	unsigned long end;
> > +};
> > +
> > +static LIST_HEAD(accepting_list);
> > +
> >   /*
> >    * accept_memory() -- Consult bitmap and accept the memory if needed.
> >    *
> > @@ -24,6 +32,7 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
> >   {
> >   	struct efi_unaccepted_memory *unaccepted;
> >   	unsigned long range_start, range_end;
> > +	struct accept_range range, *entry;
> >   	unsigned long flags;
> >   	u64 unit_size;
> > @@ -80,7 +89,25 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
> >   	range_start = start / unit_size;
> > +	range.start = start;
> > +	range.end = end;
> > +retry:
> >   	spin_lock_irqsave(&unaccepted_memory_lock, flags);
> > +
> > +	list_for_each_entry(entry, &accepting_list, list) {
> > +		if (entry->end < start)
> > +			continue;
> > +		if (entry->start > end)
> 
> Should this be a >= check since start and end are page aligned values?

Right. Good catch.

> > +			continue;
> > +		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
> > +
> > +		/* Somebody else accepting the range */
> > +		cpu_relax();
> > +		goto retry;
> 
> Could you set some kind of flag here so that ...
> 
> > +	}
> > +
> 
> ... once you get here, that means that area was accepted and removed from
> the list, so I think you could just drop the lock and exit now, right?
> Because at that point the bitmap will have been updated and you wouldn't be
> accepting any memory anyway?

No. Consider the case if someone else accept part of the range you are
accepting.

I guess we can check if the range on the list covers what we are accepting
fully, but it complication. Checking bitmap at this point is cheap enough:
we already hold the lock.

-- 
  Kiryl Shutsemau / Kirill A. Shutemov

Re: [PATCHv14 5/9] efi: Add unaccepted memory support

[Kirill A. Shutemov]

On Fri, Oct 13, 2023 at 08:27:28PM +0300, Kirill A. Shutemov wrote:
> I will test the idea with larger unit_size to see how it behaves.

It indeed uncovered an issue. We need to record ranges on accepting_list
in unit_size granularity. Otherwise, we fail to stop parallel accept
requests to the same unit_size block if they don't overlap on physical
addresses.

Updated patch:

diff --git a/drivers/firmware/efi/unaccepted_memory.c b/drivers/firmware/efi/unaccepted_memory.c
index 853f7dc3c21d..8af0306c8e5c 100644
--- a/drivers/firmware/efi/unaccepted_memory.c
+++ b/drivers/firmware/efi/unaccepted_memory.c
@@ -5,9 +5,17 @@
 #include <linux/spinlock.h>
 #include <asm/unaccepted_memory.h>
 
-/* Protects unaccepted memory bitmap */
+/* Protects unaccepted memory bitmap and accepting_list */
 static DEFINE_SPINLOCK(unaccepted_memory_lock);
 
+struct accept_range {
+	struct list_head list;
+	unsigned long start;
+	unsigned long end;
+};
+
+static LIST_HEAD(accepting_list);
+
 /*
  * accept_memory() -- Consult bitmap and accept the memory if needed.
  *
@@ -24,6 +32,7 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
 {
 	struct efi_unaccepted_memory *unaccepted;
 	unsigned long range_start, range_end;
+	struct accept_range range, *entry;
 	unsigned long flags;
 	u64 unit_size;
 
@@ -78,20 +87,58 @@ void accept_memory(phys_addr_t start, phys_addr_t end)
 	if (end > unaccepted->size * unit_size * BITS_PER_BYTE)
 		end = unaccepted->size * unit_size * BITS_PER_BYTE;
 
-	range_start = start / unit_size;
-
+	range.start = start / unit_size;
+	range.end = DIV_ROUND_UP(end, unit_size);
+retry:
 	spin_lock_irqsave(&unaccepted_memory_lock, flags);
+
+	/*
+	 * Check if anybody works on accepting the same range of the memory.
+	 *
+	 * The check with unit_size granularity. It is crucial to catch all
+	 * accept requests to the same unit_size block, even if they don't
+	 * overlap on physical address level.
+	 */
+	list_for_each_entry(entry, &accepting_list, list) {
+		if (entry->end < range.start)
+			continue;
+		if (entry->start >= range.end)
+			continue;
+
+		/*
+		 * Somebody else accepting the range. Or at least part of it.
+		 *
+		 * Drop the lock and retry until it is complete.
+		 */
+		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
+		cond_resched();
+		goto retry;
+	}
+
+	/*
+	 * Register that the range is about to be accepted.
+	 * Make sure nobody else will accept it.
+	 */
+	list_add(&range.list, &accepting_list);
+
+	range_start = range.start;
 	for_each_set_bitrange_from(range_start, range_end, unaccepted->bitmap,
-				   DIV_ROUND_UP(end, unit_size)) {
+				   range.end) {
 		unsigned long phys_start, phys_end;
 		unsigned long len = range_end - range_start;
 
 		phys_start = range_start * unit_size + unaccepted->phys_base;
 		phys_end = range_end * unit_size + unaccepted->phys_base;
 
+		spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
+
 		arch_accept_memory(phys_start, phys_end);
+
+		spin_lock_irqsave(&unaccepted_memory_lock, flags);
 		bitmap_clear(unaccepted->bitmap, range_start, len);
 	}
+
+	list_del(&range.list);
 	spin_unlock_irqrestore(&unaccepted_memory_lock, flags);
 }
 
-- 
  Kiryl Shutsemau / Kirill A. Shutemov