The patch below does not apply to the 4.9-stable tree.
If someone wants it applied there, or to any other stable or longterm
tree, then please email the backport, including the original git commit
id to <stable(a)vger.kernel.org>.
thanks,
greg k-h
------------------ original commit in Linus's tree ------------------
From 1d0e84806047f38027d7572adb4702ef7c09b317 Mon Sep 17 00:00:00 2001
From: Sean Christopherson <seanjc(a)google.com>
Date: Thu, 31 Mar 2022 22:13:59 +0000
Subject: [PATCH] KVM: x86/mmu: Resolve nx_huge_pages when kvm.ko is loaded
Resolve nx_huge_pages to true/false when kvm.ko is loaded, leaving it as
-1 is technically undefined behavior when its value is read out by
param_get_bool(), as boolean values are supposed to be '0' or '1'.
Alternatively, KVM could define a custom getter for the param, but the
auto value doesn't depend on the vendor module in any way, and printing
"auto" would be unnecessarily unfriendly to the user.
In addition to fixing the undefined behavior, resolving the auto value
also fixes the scenario where the auto value resolves to N and no vendor
module is loaded. Previously, -1 would result in Y being printed even
though KVM would ultimately disable the mitigation.
Rename the existing MMU module init/exit helpers to clarify that they're
invoked with respect to the vendor module, and add comments to document
why KVM has two separate "module init" flows.
=========================================================================
UBSAN: invalid-load in kernel/params.c:320:33
load of value 255 is not a valid value for type '_Bool'
CPU: 6 PID: 892 Comm: tail Not tainted 5.17.0-rc3+ #799
Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 0.0.0 02/06/2015
Call Trace:
<TASK>
dump_stack_lvl+0x34/0x44
ubsan_epilogue+0x5/0x40
__ubsan_handle_load_invalid_value.cold+0x43/0x48
param_get_bool.cold+0xf/0x14
param_attr_show+0x55/0x80
module_attr_show+0x1c/0x30
sysfs_kf_seq_show+0x93/0xc0
seq_read_iter+0x11c/0x450
new_sync_read+0x11b/0x1a0
vfs_read+0xf0/0x190
ksys_read+0x5f/0xe0
do_syscall_64+0x3b/0xc0
entry_SYSCALL_64_after_hwframe+0x44/0xae
</TASK>
=========================================================================
Fixes: b8e8c8303ff2 ("kvm: mmu: ITLB_MULTIHIT mitigation")
Cc: stable(a)vger.kernel.org
Reported-by: Bruno Goncalves <bgoncalv(a)redhat.com>
Reported-by: Jan Stancek <jstancek(a)redhat.com>
Signed-off-by: Sean Christopherson <seanjc(a)google.com>
Message-Id: <20220331221359.3912754-1-seanjc(a)google.com>
Signed-off-by: Paolo Bonzini <pbonzini(a)redhat.com>
diff --git a/arch/x86/include/asm/kvm_host.h b/arch/x86/include/asm/kvm_host.h
index d23e80a56eb8..0d37ba442de3 100644
--- a/arch/x86/include/asm/kvm_host.h
+++ b/arch/x86/include/asm/kvm_host.h
@@ -1585,8 +1585,9 @@ static inline int kvm_arch_flush_remote_tlb(struct kvm *kvm)
#define kvm_arch_pmi_in_guest(vcpu) \
((vcpu) && (vcpu)->arch.handling_intr_from_guest)
-int kvm_mmu_module_init(void);
-void kvm_mmu_module_exit(void);
+void kvm_mmu_x86_module_init(void);
+int kvm_mmu_vendor_module_init(void);
+void kvm_mmu_vendor_module_exit(void);
void kvm_mmu_destroy(struct kvm_vcpu *vcpu);
int kvm_mmu_create(struct kvm_vcpu *vcpu);
diff --git a/arch/x86/kvm/mmu/mmu.c b/arch/x86/kvm/mmu/mmu.c
index 8f19ea752704..f9080ee50ffa 100644
--- a/arch/x86/kvm/mmu/mmu.c
+++ b/arch/x86/kvm/mmu/mmu.c
@@ -6237,12 +6237,24 @@ static int set_nx_huge_pages(const char *val, const struct kernel_param *kp)
return 0;
}
-int kvm_mmu_module_init(void)
+/*
+ * nx_huge_pages needs to be resolved to true/false when kvm.ko is loaded, as
+ * its default value of -1 is technically undefined behavior for a boolean.
+ */
+void kvm_mmu_x86_module_init(void)
{
- int ret = -ENOMEM;
-
if (nx_huge_pages == -1)
__set_nx_huge_pages(get_nx_auto_mode());
+}
+
+/*
+ * The bulk of the MMU initialization is deferred until the vendor module is
+ * loaded as many of the masks/values may be modified by VMX or SVM, i.e. need
+ * to be reset when a potentially different vendor module is loaded.
+ */
+int kvm_mmu_vendor_module_init(void)
+{
+ int ret = -ENOMEM;
/*
* MMU roles use union aliasing which is, generally speaking, an
@@ -6290,7 +6302,7 @@ void kvm_mmu_destroy(struct kvm_vcpu *vcpu)
mmu_free_memory_caches(vcpu);
}
-void kvm_mmu_module_exit(void)
+void kvm_mmu_vendor_module_exit(void)
{
mmu_destroy_caches();
percpu_counter_destroy(&kvm_total_used_mmu_pages);
diff --git a/arch/x86/kvm/x86.c b/arch/x86/kvm/x86.c
index 0c0ca599a353..de49a88df1c2 100644
--- a/arch/x86/kvm/x86.c
+++ b/arch/x86/kvm/x86.c
@@ -8926,7 +8926,7 @@ int kvm_arch_init(void *opaque)
}
kvm_nr_uret_msrs = 0;
- r = kvm_mmu_module_init();
+ r = kvm_mmu_vendor_module_init();
if (r)
goto out_free_percpu;
@@ -8974,7 +8974,7 @@ void kvm_arch_exit(void)
cancel_work_sync(&pvclock_gtod_work);
#endif
kvm_x86_ops.hardware_enable = NULL;
- kvm_mmu_module_exit();
+ kvm_mmu_vendor_module_exit();
free_percpu(user_return_msrs);
kmem_cache_destroy(x86_emulator_cache);
#ifdef CONFIG_KVM_XEN
@@ -12986,3 +12986,19 @@ EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_vmgexit_enter);
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_vmgexit_exit);
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_vmgexit_msr_protocol_enter);
EXPORT_TRACEPOINT_SYMBOL_GPL(kvm_vmgexit_msr_protocol_exit);
+
+static int __init kvm_x86_init(void)
+{
+ kvm_mmu_x86_module_init();
+ return 0;
+}
+module_init(kvm_x86_init);
+
+static void __exit kvm_x86_exit(void)
+{
+ /*
+ * If module_init() is implemented, module_exit() must also be
+ * implemented to allow module unload.
+ */
+}
+module_exit(kvm_x86_exit);
This patch fixes a memory corruption that occurred in the
nand_scan() path for Hynix nand device.
On boot, for Hynix nand device will panic at a weird place:
| Unable to handle kernel NULL pointer dereference at virtual
address 00000070
| [00000070] *pgd=00000000
| Internal error: Oops: 5 [#1] PREEMPT SMP ARM
| Modules linked in:
| CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.17.0-01473-g13ae1769cfb0
#38
| Hardware name: Generic DT based system
| PC is at nandc_set_reg+0x8/0x1c
| LR is at qcom_nandc_command+0x20c/0x5d0
| pc : [<c088b74c>] lr : [<c088d9c8>] psr: 00000113
| sp : c14adc50 ip : c14ee208 fp : c0cc970c
| r10: 000000a3 r9 : 00000000 r8 : 00000040
| r7 : c16f6a00 r6 : 00000090 r5 : 00000004 r4 :c14ee040
| r3 : 00000000 r2 : 0000000b r1 : 00000000 r0 :c14ee040
| Flags: nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
| Control: 10c5387d Table: 8020406a DAC: 00000051
| Register r0 information: slab kmalloc-2k start c14ee000 pointer offset
64 size 2048
| Process swapper/0 (pid: 1, stack limit = 0x(ptrval))
| nandc_set_reg from qcom_nandc_command+0x20c/0x5d0
| qcom_nandc_command from nand_readid_op+0x198/0x1e8
| nand_readid_op from hynix_nand_has_valid_jedecid+0x30/0x78
| hynix_nand_has_valid_jedecid from hynix_nand_init+0xb8/0x454
| hynix_nand_init from nand_scan_with_ids+0xa30/0x14a8
| nand_scan_with_ids from qcom_nandc_probe+0x648/0x7b0
| qcom_nandc_probe from platform_probe+0x58/0xac
The problem is that the nand_scan()'s qcom_nand_attach_chip callback
is updating the nandc->max_cwperpage from 1 to 4 or 8 based on page size.
This causes the sg_init_table of clear_bam_transaction() in the driver's
qcom_nandc_command() to memset much more than what was initially
allocated by alloc_bam_transaction().
This patch will update nandc->max_cwperpage 1 to 4 or 8 based on page
size in qcom_nand_attach_chip call back after freeing the previously
allocated memory for bam txn as per nandc->max_cwperpage = 1 and then
again allocating bam txn as per nandc->max_cwperpage = 4 or 8 based on
page size in qcom_nand_attach_chip call back itself.
Cc: stable(a)vger.kernel.org
Fixes: 6a3cec64f18c ("mtd: rawnand: qcom: convert driver to nand_scan()")
Reported-by: Konrad Dybcio <konrad.dybcio(a)somainline.org>
Signed-off-by: Md Sadre Alam <quic_mdalam(a)quicinc.com>
Signed-off-by: Sricharan R <quic_srichara(a)quicinc.com>
---
Changes in V4:
* Incorporated "commit log wrong" comment from Mani
* Updated commit log
Changes in V3:
* Incorporated "Fixes tags are missing" comment from Miquèl
* Added Fixes tag Fixes:6a3cec64f18c ("mtd: rawnand: qcom: convert driver to nand_scan()")
* Incorporated "stable tag missing" comment from Miquèl
* Added stable tag Cc: stable(a)vger.kernel.org
* Incorporated "Reported-by tag missing" comment from Mani
* Added Reported-by tag Reported-by: Konrad Dybcio <konrad.dybcio(a)somainline.org>
Changes in V2:
* Incorporated "alloc_bam_transaction inside qcom_nand_attach_chip" suggestion from Mani
* Freed previously alloacted memory for bam txn before updating max_cwperpage inside
qcom_nand_attach_chip().
* Moved alloc_bam_transaction() inside qcom_nand_attach_chip(). after upding max_cwperpage
4 or 8 based on page size.
drivers/mtd/nand/raw/qcom_nandc.c | 24 +++++++++++++-----------
1 file changed, 13 insertions(+), 11 deletions(-)
diff --git a/drivers/mtd/nand/raw/qcom_nandc.c b/drivers/mtd/nand/raw/qcom_nandc.c
index 1a77542..048b255 100644
--- a/drivers/mtd/nand/raw/qcom_nandc.c
+++ b/drivers/mtd/nand/raw/qcom_nandc.c
@@ -2651,10 +2651,23 @@ static int qcom_nand_attach_chip(struct nand_chip *chip)
ecc->engine_type = NAND_ECC_ENGINE_TYPE_ON_HOST;
mtd_set_ooblayout(mtd, &qcom_nand_ooblayout_ops);
+ /* Free the initially allocated BAM transaction for reading the ONFI params */
+ if (nandc->props->is_bam)
+ free_bam_transaction(nandc);
nandc->max_cwperpage = max_t(unsigned int, nandc->max_cwperpage,
cwperpage);
+ /* Now allocate the BAM transaction based on updated max_cwperpage */
+ if (nandc->props->is_bam) {
+ nandc->bam_txn = alloc_bam_transaction(nandc);
+ if (!nandc->bam_txn) {
+ dev_err(nandc->dev,
+ "failed to allocate bam transaction\n");
+ return -ENOMEM;
+ }
+ }
+
/*
* DATA_UD_BYTES varies based on whether the read/write command protects
* spare data with ECC too. We protect spare data by default, so we set
@@ -2955,17 +2968,6 @@ static int qcom_nand_host_init_and_register(struct qcom_nand_controller *nandc,
if (ret)
return ret;
- if (nandc->props->is_bam) {
- free_bam_transaction(nandc);
- nandc->bam_txn = alloc_bam_transaction(nandc);
- if (!nandc->bam_txn) {
- dev_err(nandc->dev,
- "failed to allocate bam transaction\n");
- nand_cleanup(chip);
- return -ENOMEM;
- }
- }
-
ret = mtd_device_parse_register(mtd, probes, NULL, NULL, 0);
if (ret)
nand_cleanup(chip);
--
2.7.4
This patch fixes a memory corruption that occurred in the
nand_scan() path for Hynix nand device.
On boot, for Hynix nand device will panic at a weird place:
| Unable to handle kernel NULL pointer dereference at virtual
address 00000070
| [00000070] *pgd=00000000
| Internal error: Oops: 5 [#1] PREEMPT SMP ARM
| Modules linked in:
| CPU: 0 PID: 1 Comm: swapper/0 Not tainted 5.17.0-01473-g13ae1769cfb0
#38
| Hardware name: Generic DT based system
| PC is at nandc_set_reg+0x8/0x1c
| LR is at qcom_nandc_command+0x20c/0x5d0
| pc : [<c088b74c>] lr : [<c088d9c8>] psr: 00000113
| sp : c14adc50 ip : c14ee208 fp : c0cc970c
| r10: 000000a3 r9 : 00000000 r8 : 00000040
| r7 : c16f6a00 r6 : 00000090 r5 : 00000004 r4 :c14ee040
| r3 : 00000000 r2 : 0000000b r1 : 00000000 r0 :c14ee040
| Flags: nzcv IRQs on FIQs on Mode SVC_32 ISA ARM Segment none
| Control: 10c5387d Table: 8020406a DAC: 00000051
| Register r0 information: slab kmalloc-2k start c14ee000 pointer offset
64 size 2048
| Process swapper/0 (pid: 1, stack limit = 0x(ptrval))
| nandc_set_reg from qcom_nandc_command+0x20c/0x5d0
| qcom_nandc_command from nand_readid_op+0x198/0x1e8
| nand_readid_op from hynix_nand_has_valid_jedecid+0x30/0x78
| hynix_nand_has_valid_jedecid from hynix_nand_init+0xb8/0x454
| hynix_nand_init from nand_scan_with_ids+0xa30/0x14a8
| nand_scan_with_ids from qcom_nandc_probe+0x648/0x7b0
| qcom_nandc_probe from platform_probe+0x58/0xac
The problem is that the nand_scan()'s qcom_nand_attach_chip callback
is updating the nandc->max_cwperpage from 1 to 4.This causes the
sg_init_table of clear_bam_transaction() in the driver's
qcom_nandc_command() to memset much more than what was initially
allocated by alloc_bam_transaction().
This patch will update nandc->max_cwperpage 1 to 4 after nand_scan()
returns, and remove updating nandc->max_cwperpage from
qcom_nand_attach_chip call back.
Cc: stable(a)vger.kernel.org
Fixes: 6a3cec64f18c ("mtd: rawnand: qcom: convert driver to nand_scan()")
Reported-by: Konrad Dybcio <konrad.dybcio(a)somainline.org>
Signed-off-by: Md Sadre Alam <quic_mdalam(a)quicinc.com>
Signed-off-by: Sricharan R <quic_srichara(a)quicinc.com>
---
[V3]
* Updated commit message Fixes, Cc, Reported-by
drivers/mtd/nand/raw/qcom_nandc.c | 24 +++++++++++++-----------
1 file changed, 13 insertions(+), 11 deletions(-)
diff --git a/drivers/mtd/nand/raw/qcom_nandc.c b/drivers/mtd/nand/raw/qcom_nandc.c
index 1a77542..048b255 100644
--- a/drivers/mtd/nand/raw/qcom_nandc.c
+++ b/drivers/mtd/nand/raw/qcom_nandc.c
@@ -2651,10 +2651,23 @@ static int qcom_nand_attach_chip(struct nand_chip *chip)
ecc->engine_type = NAND_ECC_ENGINE_TYPE_ON_HOST;
mtd_set_ooblayout(mtd, &qcom_nand_ooblayout_ops);
+ /* Free the initially allocated BAM transaction for reading the ONFI params */
+ if (nandc->props->is_bam)
+ free_bam_transaction(nandc);
nandc->max_cwperpage = max_t(unsigned int, nandc->max_cwperpage,
cwperpage);
+ /* Now allocate the BAM transaction based on updated max_cwperpage */
+ if (nandc->props->is_bam) {
+ nandc->bam_txn = alloc_bam_transaction(nandc);
+ if (!nandc->bam_txn) {
+ dev_err(nandc->dev,
+ "failed to allocate bam transaction\n");
+ return -ENOMEM;
+ }
+ }
+
/*
* DATA_UD_BYTES varies based on whether the read/write command protects
* spare data with ECC too. We protect spare data by default, so we set
@@ -2955,17 +2968,6 @@ static int qcom_nand_host_init_and_register(struct qcom_nand_controller *nandc,
if (ret)
return ret;
- if (nandc->props->is_bam) {
- free_bam_transaction(nandc);
- nandc->bam_txn = alloc_bam_transaction(nandc);
- if (!nandc->bam_txn) {
- dev_err(nandc->dev,
- "failed to allocate bam transaction\n");
- nand_cleanup(chip);
- return -ENOMEM;
- }
- }
-
ret = mtd_device_parse_register(mtd, probes, NULL, NULL, 0);
if (ret)
nand_cleanup(chip);
--
2.7.4