Skip to contentSkip to conversation

FAILED: patch "[PATCH] MIPS: kvm: Fix ioctl error handling." failed to apply to 4.4-stable tree

PATCH [email protected] 2 participants 6 messages Mar 7, 2016 – Mar 12, 2016

Subsystems (3)
lore
Share & tools
lore

[email protected]

author maintainer
#0 lore raw

The patch below does not apply to the 4.4-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 <[email protected]>.

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

Quoted 1 line

From 887349f69f37e71e2a8bfbd743831625a0b2ff51 Mon Sep 17 00:00:00 2001

From: "Michael S. Tsirkin" <[email protected]>
Date: Sun, 28 Feb 2016 17:35:59 +0200
Subject: [PATCH] MIPS: kvm: Fix ioctl error handling.

Calling return copy_to_user(...) or return copy_from_user in an ioctl will not do the right thing if there's a pagefault:
copy_to_user/copy_from_user return the number of bytes not copied in this case.

Fix up kvm on mips to do

	return copy_to_user(...)) ?  -EFAULT : 0;
and
	return copy_from_user(...)) ?  -EFAULT : 0;

everywhere.

Patchwork: https://patchwork.linux-mips.org/patch/12709/

arch/mips/kvm/mips.c +2 -2
index 8bc3977576e6..3110447ab1e9 100644
@@ -702,7 +702,7 @@ static int kvm_mips_get_reg(struct kvm_vcpu *vcpu,
 	} else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
 		void __user *uaddr = (void __user *)(long)reg->addr;
 
-		return copy_to_user(uaddr, vs, 16);
+		return copy_to_user(uaddr, vs, 16) ? -EFAULT : 0;
 	} else {
 		return -EINVAL;
 	}
@@ -732,7 +732,7 @@ static int kvm_mips_set_reg(struct kvm_vcpu *vcpu,
 	} else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
 		void __user *uaddr = (void __user *)(long)reg->addr;
 
-		return copy_from_user(vs, uaddr, 16);
+		return copy_from_user(vs, uaddr, 16) ? -EFAULT : 0;
 	} else {
 		return -EINVAL;
 	}
#1 lore raw

Hi Greg,

Quoted 5 lines · On Mon, Mar 07, 2016 at 12:39:08PM -0800, [email protected] wrote:

The patch below does not apply to the 4.4-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 <[email protected]>.

I don't see any conflicts when this is applied to 4.4.4. Asside from the hashes the cherry-picked commit is identical. Is it conflicting with some other stable patch since then or something?

Thanks
James

Quoted 57 lines

thanks,

greg k-h

------------------ original commit in Linus's tree ------------------

From 887349f69f37e71e2a8bfbd743831625a0b2ff51 Mon Sep 17 00:00:00 2001
From: "Michael S. Tsirkin" <[email protected]>
Date: Sun, 28 Feb 2016 17:35:59 +0200
Subject: [PATCH] MIPS: kvm: Fix ioctl error handling.

Calling return copy_to_user(...) or return copy_from_user in an ioctl will not do the right thing if there's a pagefault:
copy_to_user/copy_from_user return the number of bytes not copied in this case.

Fix up kvm on mips to do

	return copy_to_user(...)) ?  -EFAULT : 0;
and
	return copy_from_user(...)) ?  -EFAULT : 0;

everywhere.

Patchwork: https://patchwork.linux-mips.org/patch/12709/

arch/mips/kvm/mips.c +2 -2
index 8bc3977576e6..3110447ab1e9 100644
@@ -702,7 +702,7 @@ static int kvm_mips_get_reg(struct kvm_vcpu *vcpu,
 	} else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
 		void __user *uaddr = (void __user *)(long)reg->addr;
 
-		return copy_to_user(uaddr, vs, 16);
+		return copy_to_user(uaddr, vs, 16) ? -EFAULT : 0;
 	} else {
 		return -EINVAL;
 	}
@@ -732,7 +732,7 @@ static int kvm_mips_set_reg(struct kvm_vcpu *vcpu,
 	} else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
 		void __user *uaddr = (void __user *)(long)reg->addr;
 
-		return copy_from_user(vs, uaddr, 16);
+		return copy_from_user(vs, uaddr, 16) ? -EFAULT : 0;
 	} else {
 		return -EINVAL;
 	}
author maintainer
#2 lore raw
Quoted 4 lines · James Hogan

Hi Greg,

Quoted 1 line · On Mon, Mar 07, 2016 at 12:39:08PM -0800, [email protected] wrote:
Quoted 8 lines
Quoted 4 lines

The patch below does not apply to the 4.4-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 <[email protected]>.

I don't see any conflicts when this is applied to 4.4.4. Asside from the hashes the cherry-picked commit is identical. Is it conflicting with some other stable patch since then or something?

Try using patch, not 'cherry-pick' and see if it still applies.

thanks,

greg k-h

#3 lore raw
Quoted 10 lines · Greg KH
Quoted 9 lines · James Hogan

Hi Greg,

Quoted 5 lines · On Mon, Mar 07, 2016 at 12:39:08PM -0800, [email protected] wrote:

The patch below does not apply to the 4.4-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 <[email protected]>.

Quoted 5 lines
Quoted 3 lines

I don't see any conflicts when this is applied to 4.4.4. Asside from the hashes the cherry-picked commit is identical. Is it conflicting with some other stable patch since then or something?

Try using patch, not 'cherry-pick' and see if it still applies.

$ git reset --hard HEAD~
HEAD is now at c252409a688a Linux 4.4.4
$ git show 887349f69f37e71e2a8bfbd743831625a0b2ff51 | patch -p1
patching file arch/mips/kvm/mips.c
$ git diff | cat
arch/mips/kvm/mips.c +2 -2
index b9b803facdbf..2683d04fdda5 100644
@@ -702,7 +702,7 @@ static int kvm_mips_get_reg(struct kvm_vcpu *vcpu,
        } else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
                void __user *uaddr = (void __user *)(long)reg->addr;
 
-               return copy_to_user(uaddr, vs, 16);
+               return copy_to_user(uaddr, vs, 16) ? -EFAULT : 0;
        } else {
                return -EINVAL;
        }
@@ -732,7 +732,7 @@ static int kvm_mips_set_reg(struct kvm_vcpu *vcpu,
        } else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
                void __user *uaddr = (void __user *)(long)reg->addr;
 
-               return copy_from_user(vs, uaddr, 16);
+               return copy_from_user(vs, uaddr, 16) ? -EFAULT : 0;
        } else {
                return -EINVAL;
        }

Cheers
James

author maintainer
#4 lore raw
Quoted 37 lines · James Hogan
Quoted 10 lines · Greg KH
Quoted 9 lines · James Hogan

Hi Greg,

Quoted 5 lines · On Mon, Mar 07, 2016 at 12:39:08PM -0800, [email protected] wrote:

The patch below does not apply to the 4.4-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 <[email protected]>.

Quoted 5 lines
Quoted 3 lines

I don't see any conflicts when this is applied to 4.4.4. Asside from the hashes the cherry-picked commit is identical. Is it conflicting with some other stable patch since then or something?

Try using patch, not 'cherry-pick' and see if it still applies.

$ git reset --hard HEAD~
HEAD is now at c252409a688a Linux 4.4.4
$ git show 887349f69f37e71e2a8bfbd743831625a0b2ff51 | patch -p1
patching file arch/mips/kvm/mips.c
$ git diff | cat
arch/mips/kvm/mips.c +1 -1
index b9b803facdbf..2683d04fdda5 100644
@@ -702,7 +702,7 @@ static int kvm_mips_get_reg(struct kvm_vcpu *vcpu,
        } else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
                void __user *uaddr = (void __user *)(long)reg->addr;
 
-               return copy_to_user(uaddr, vs, 16);
+               return copy_to_user(uaddr, vs, 16) ? -EFAULT : 0;
        } else {
                return -EINVAL;
        }

arch/mips/kvm/mips.c static int kvm_mips_set_reg(struct kvm_vcpu *vcpu, · line 738

Quoted 8 lines
arch/mips/kvm/mips.c +1 -1
@@ -732,7 +732,7 @@ static int kvm_mips_set_reg(struct kvm_vcpu *vcpu,
        } else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
                void __user *uaddr = (void __user *)(long)reg->addr;
 
-               return copy_from_user(vs, uaddr, 16);
+               return copy_from_user(vs, uaddr, 16) ? -EFAULT : 0;
        } else {
                return -EINVAL;
        }

Fails for me:

~/linux/stable/linux-4.4.y $ git show 887349f69f37e71e2a8bfbd743831625a0b2ff51 | patch -p1 patching file arch/mips/kvm/mips.c
Reversed (or previously applied) patch detected! Assume -R? [n]
Apply anyway? [n]
Skipping patch.
2 out of 2 hunks ignored -- saving rejects to file arch/mips/kvm/mips.c.rej

Is this an issue with 4.4.5 perhaps?

Ah, do we have duplicate patches here? Look at commit 0178fd7dcc4451fcb90bec5e91226586962478d2 which is in 4.4.5 now, it's this same patch...

#5 lore raw
Quoted 57 lines · Greg KH
Quoted 37 lines · James Hogan
Quoted 10 lines · Greg KH
Quoted 9 lines · James Hogan

Hi Greg,

Quoted 5 lines · On Mon, Mar 07, 2016 at 12:39:08PM -0800, [email protected] wrote:

The patch below does not apply to the 4.4-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 <[email protected]>.

Quoted 5 lines
Quoted 3 lines

I don't see any conflicts when this is applied to 4.4.4. Asside from the hashes the cherry-picked commit is identical. Is it conflicting with some other stable patch since then or something?

Try using patch, not 'cherry-pick' and see if it still applies.

$ git reset --hard HEAD~
HEAD is now at c252409a688a Linux 4.4.4
$ git show 887349f69f37e71e2a8bfbd743831625a0b2ff51 | patch -p1
patching file arch/mips/kvm/mips.c
$ git diff | cat
arch/mips/kvm/mips.c +1 -1
index b9b803facdbf..2683d04fdda5 100644
@@ -702,7 +702,7 @@ static int kvm_mips_get_reg(struct kvm_vcpu *vcpu,
        } else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
                void __user *uaddr = (void __user *)(long)reg->addr;
 
-               return copy_to_user(uaddr, vs, 16);
+               return copy_to_user(uaddr, vs, 16) ? -EFAULT : 0;
        } else {
                return -EINVAL;
        }

arch/mips/kvm/mips.c static int kvm_mips_set_reg(struct kvm_vcpu *vcpu, · line 738

Quoted 8 lines
arch/mips/kvm/mips.c +1 -1
@@ -732,7 +732,7 @@ static int kvm_mips_set_reg(struct kvm_vcpu *vcpu,
        } else if ((reg->id & KVM_REG_SIZE_MASK) == KVM_REG_SIZE_U128) {
                void __user *uaddr = (void __user *)(long)reg->addr;
 
-               return copy_from_user(vs, uaddr, 16);
+               return copy_from_user(vs, uaddr, 16) ? -EFAULT : 0;
        } else {
                return -EINVAL;
        }

Fails for me:

~/linux/stable/linux-4.4.y $ git show 887349f69f37e71e2a8bfbd743831625a0b2ff51 | patch -p1 patching file arch/mips/kvm/mips.c
Reversed (or previously applied) patch detected! Assume -R? [n]
Apply anyway? [n]
Skipping patch.
2 out of 2 hunks ignored -- saving rejects to file arch/mips/kvm/mips.c.rej

Quoted 5 lines

Is this an issue with 4.4.5 perhaps?

Ah, do we have duplicate patches here? Look at commit 0178fd7dcc4451fcb90bec5e91226586962478d2 which is in 4.4.5 now, it's this same patch...

Ah yes, that'd be it. Both Ralf and Paolo applied it.

Thanks
James