Skip to contentSkip to conversation

x86: enable SWIOTLB if system has SRAT memory regions above MAX_DMA32_PFN

PATCH New Igor Mammedov 2 participants 5 messages Dec 4, 2015 – Dec 6, 2015

Subsystems (3)
lore
Share & tools
lore
#0 lore raw

v3 cover x86: enable SWIOTLB if system has SRAT memory regions above MAX_DMA32_PFN

changes since v2:
- moved 'max_possible_pfn' tracking hunk to the 1st patch
- make 'max_possible_pfn' 64-bit
- simplify condition to oneliner as suggested by Ingo
changes since v1:
- reduce # of #ifdefs by introducing max_possible_pfn global variable
- don't check 'acpi_no_memhotplug=1' for disabling SWIOTLB initialization, since existing 'no_iommu' kernel option could be used to the same effect.
- split into 2 patches
- 1st adds max_possible_pfn,
- 2nd fixes bug enabling SWIOTLB when it's needed

when memory hotplug enabled system is booted with less than 4GB of RAM and then later more RAM is hotplugged 32-bit devices stop functioning with following error:

nommu_map_single: overflow 327b4f8c0+1522 of device mask ffffffff

the reason for this is that if x86_64 system were booted with RAM less than 4GB, it doesn't enable SWIOTLB and when memory is hotplugged beyond MAX_DMA32_PFN, devices that expect 32-bit addresses can't handle 64-bit addresses.

Fix it by tracking max possible PFN when parsing memory affinity structures from SRAT ACPI table and enable SWIOTLB if there is hotpluggable memory regions beyond MAX_DMA32_PFN.

It fixes KVM guests when they use emulated devices (reproduces with ata_piix, e1000 and usb devices,
RHBZ: 1275941, 1275977, 1271527)
It also fixes the HyperV, VMWare with emulated devices which are affected by this issue as well.

ref to v2:
https://lkml.org/lkml/2015/12/4/151
ref to v1:
https://lkml.org/lkml/2015/11/30/594

Igor Mammedov (2):
x86: introduce max_possible_pfn
x86_64: enable SWIOTLB if system has SRAT memory regions above

    MAX_DMA32_PFN
 arch/x86/kernel/pci-swiotlb.c | 2 +-
 arch/x86/kernel/setup.c       | 2 ++
 arch/x86/mm/srat.c            | 2 ++
 include/linux/bootmem.h       | 4 ++++
 mm/bootmem.c                  | 1 +
 mm/nobootmem.c                | 1 +
 6 files changed, 11 insertions(+), 1 deletion(-)
Signature · 1 line
1.8.3.1
#1 lore raw

PATCH v3 1/2 x86: introduce max_possible_pfn

max_possible_pfn will be used for tracking max possible PFN for memory that isn't present in E820 table and could be hotplugged later.

By default max_possible_pfn is initialized with max_pfn, but later it could be updated with highest PFN of hotpluggable memory ranges declared in ACPI SRAT table if any present.

---
v3:
- make 'max_possible_pfn' 64-bit
- simplify condition to oneliner as suggested by Ingo

---
 arch/x86/kernel/setup.c | 2 ++
 arch/x86/mm/srat.c      | 2 ++
 include/linux/bootmem.h | 4 ++++
 mm/bootmem.c            | 1 +
 mm/nobootmem.c          | 1 +
 5 files changed, 10 insertions(+)
arch/x86/kernel/setup.c +2 -0
index 29db25f..16a8465 100644
@@ -1048,6 +1048,8 @@ void __init setup_arch(char **cmdline_p)
 	if (mtrr_trim_uncached_memory(max_pfn))
 		max_pfn = e820_end_of_ram_pfn();
 
+	max_possible_pfn = max_pfn;
+
 #ifdef CONFIG_X86_32
 	/* max_low_pfn get updated here */
 	find_low_pfn_range();
arch/x86/mm/srat.c +2 -0
index c2aea63..b5f8218 100644
@@ -203,6 +203,8 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma)
 		pr_warn("SRAT: Failed to mark hotplug range [mem %#010Lx-%#010Lx] in memblock\n",
 			(unsigned long long)start, (unsigned long long)end - 1);
 
+	max_possible_pfn = max(max_possible_pfn, PFN_UP(end - 1));
+
 	return 0;
 out_err_bad_srat:
 	bad_srat();
include/linux/bootmem.h +4 -0
index f589222..35b22f9 100644
@@ -19,6 +19,10 @@ extern unsigned long min_low_pfn;
  * highest page
  */
 extern unsigned long max_pfn;
+/*
+ * highest possible page
+ */
+extern unsigned long long max_possible_pfn;
 
 #ifndef CONFIG_NO_BOOTMEM
 /*
mm/bootmem.c +1 -0
index 3b63807..91e32bc 100644
@@ -33,6 +33,7 @@ EXPORT_SYMBOL(contig_page_data);
 unsigned long max_low_pfn;
 unsigned long min_low_pfn;
 unsigned long max_pfn;
+unsigned long long max_possible_pfn;
 
 bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata;
 
mm/nobootmem.c +1 -0
index e57cf24..99feb2b 100644
@@ -31,6 +31,7 @@ EXPORT_SYMBOL(contig_page_data);
 unsigned long max_low_pfn;
 unsigned long min_low_pfn;
 unsigned long max_pfn;
+unsigned long long max_possible_pfn;
 
 static void * __init __alloc_memory_core_early(int nid, u64 size, u64 align,
 					u64 goal, u64 limit)
Signature · 1 line
1.8.3.1
#2 lore raw

[tip:x86/mm] x86/mm: Introduce max_possible_pfn

x86/mm: Introduce max_possible_pfn
Commit-ID:  8dd3303001976aa8583bf20f6b93590c74114308
Gitweb:     http://git.kernel.org/tip/8dd3303001976aa8583bf20f6b93590c74114308
Author:     Igor Mammedov <[email protected]>
AuthorDate: Fri, 4 Dec 2015 14:07:05 +0100
Committer:  Ingo Molnar <[email protected]>
CommitDate: Sun, 6 Dec 2015 12:46:31 +0100

x86/mm: Introduce max_possible_pfn

max_possible_pfn will be used for tracking max possible PFN for memory that isn't present in E820 table and could be hotplugged later.

By default max_possible_pfn is initialized with max_pfn, but later it could be updated with highest PFN of hotpluggable memory ranges declared in ACPI SRAT table if any present.

---
 arch/x86/kernel/setup.c | 2 ++
 arch/x86/mm/srat.c      | 2 ++
 include/linux/bootmem.h | 4 ++++
 mm/bootmem.c            | 1 +
 mm/nobootmem.c          | 1 +
 5 files changed, 10 insertions(+)
arch/x86/kernel/setup.c +2 -0
index 29db25f..16a8465 100644
@@ -1048,6 +1048,8 @@ void __init setup_arch(char **cmdline_p)
 	if (mtrr_trim_uncached_memory(max_pfn))
 		max_pfn = e820_end_of_ram_pfn();
 
+	max_possible_pfn = max_pfn;
+
 #ifdef CONFIG_X86_32
 	/* max_low_pfn get updated here */
 	find_low_pfn_range();
arch/x86/mm/srat.c +2 -0
index c2aea63..b5f8218 100644
@@ -203,6 +203,8 @@ acpi_numa_memory_affinity_init(struct acpi_srat_mem_affinity *ma)
 		pr_warn("SRAT: Failed to mark hotplug range [mem %#010Lx-%#010Lx] in memblock\n",
 			(unsigned long long)start, (unsigned long long)end - 1);
 
+	max_possible_pfn = max(max_possible_pfn, PFN_UP(end - 1));
+
 	return 0;
 out_err_bad_srat:
 	bad_srat();
include/linux/bootmem.h +4 -0
index f589222..35b22f9 100644
@@ -19,6 +19,10 @@ extern unsigned long min_low_pfn;
  * highest page
  */
 extern unsigned long max_pfn;
+/*
+ * highest possible page
+ */
+extern unsigned long long max_possible_pfn;
 
 #ifndef CONFIG_NO_BOOTMEM
 /*
mm/bootmem.c +1 -0
index 3b63807..91e32bc 100644
@@ -33,6 +33,7 @@ EXPORT_SYMBOL(contig_page_data);
 unsigned long max_low_pfn;
 unsigned long min_low_pfn;
 unsigned long max_pfn;
+unsigned long long max_possible_pfn;
 
 bootmem_data_t bootmem_node_data[MAX_NUMNODES] __initdata;
 
mm/nobootmem.c +1 -0
index e57cf24..99feb2b 100644
@@ -31,6 +31,7 @@ EXPORT_SYMBOL(contig_page_data);
 unsigned long max_low_pfn;
 unsigned long min_low_pfn;
 unsigned long max_pfn;
+unsigned long long max_possible_pfn;
 
 static void * __init __alloc_memory_core_early(int nid, u64 size, u64 align,
 					u64 goal, u64 limit)
#3 lore raw

PATCH v3 2/2 x86_64: enable SWIOTLB if system has SRAT memory regions above MAX_DMA32_PFN

In reply to Igor Mammedov

when memory hotplug enabled system is booted with less than 4GB of RAM and then later more RAM is hotplugged 32-bit devices stop functioning with following error:

nommu_map_single: overflow 327b4f8c0+1522 of device mask ffffffff

the reason for this is that if x86_64 system were booted with RAM less than 4GB, it doesn't enable SWIOTLB and when memory is hotplugged beyond MAX_DMA32_PFN, devices that expect 32-bit addresses can't handle 64-bit addresses.

Fix it by tracking max possible PFN when parsing memory affinity structures from SRAT ACPI table and enable SWIOTLB if there is hotpluggable memory regions beyond MAX_DMA32_PFN.

It fixes KVM guests when they use emulated devices (reproduces with ata_piix, e1000 and usb devices,
RHBZ: 1275941, 1275977, 1271527)
It also fixes the HyperV, VMWare with emulated devices which are affected by this issue as well.

---
v3:
- moved 'max_possible_pfn' tracking hunk to the patch that introduces it.

---
 arch/x86/kernel/pci-swiotlb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/pci-swiotlb.c +1 -1
index adf0392..7c577a1 100644
@@ -88,7 +88,7 @@ int __init pci_swiotlb_detect_4gb(void)
 {
 	/* don't initialize swiotlb if iommu=off (no_iommu=1) */
 #ifdef CONFIG_X86_64
-	if (!no_iommu && max_pfn > MAX_DMA32_PFN)
+	if (!no_iommu && max_possible_pfn > MAX_DMA32_PFN)
 		swiotlb = 1;
 #endif
 	return swiotlb;
Signature · 1 line
1.8.3.1
#4 lore raw

[tip:x86/mm] x86/mm/64: Enable SWIOTLB if system has SRAT memory regions above MAX_DMA32_PFN

x86/mm/64: Enable SWIOTLB if system has SRAT memory regions above MAX_DMA32_PFN
Commit-ID:  ec941c5ffede4d788b9fc008f9eeca75b9e964f5
Gitweb:     http://git.kernel.org/tip/ec941c5ffede4d788b9fc008f9eeca75b9e964f5
Author:     Igor Mammedov <[email protected]>
AuthorDate: Fri, 4 Dec 2015 14:07:06 +0100
Committer:  Ingo Molnar <[email protected]>
CommitDate: Sun, 6 Dec 2015 12:46:31 +0100

x86/mm/64: Enable SWIOTLB if system has SRAT memory regions above MAX_DMA32_PFN

when memory hotplug enabled system is booted with less than 4GB of RAM and then later more RAM is hotplugged 32-bit devices stop functioning with following error:

nommu_map_single: overflow 327b4f8c0+1522 of device mask ffffffff

the reason for this is that if x86_64 system were booted with RAM less than 4GB, it doesn't enable SWIOTLB and when memory is hotplugged beyond MAX_DMA32_PFN, devices that expect 32-bit addresses can't handle 64-bit addresses.

Fix it by tracking max possible PFN when parsing memory affinity structures from SRAT ACPI table and enable SWIOTLB if there is hotpluggable memory regions beyond MAX_DMA32_PFN.

It fixes KVM guests when they use emulated devices (reproduces with ata_piix, e1000 and usb devices,
RHBZ: 1275941, 1275977, 1271527)

It also fixes the HyperV, VMWare with emulated devices which are affected by this issue as well.

---
 arch/x86/kernel/pci-swiotlb.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/kernel/pci-swiotlb.c +1 -1
index adf0392..7c577a1 100644
@@ -88,7 +88,7 @@ int __init pci_swiotlb_detect_4gb(void)
 {
 	/* don't initialize swiotlb if iommu=off (no_iommu=1) */
 #ifdef CONFIG_X86_64
-	if (!no_iommu && max_pfn > MAX_DMA32_PFN)
+	if (!no_iommu && max_possible_pfn > MAX_DMA32_PFN)
 		swiotlb = 1;
 #endif
 	return swiotlb;