[3.13.y-ckt stable] Patch "KVM: MIPS: Don't leak FPU/DSP to guest" has been added to staging queue
This is a note to let you know that I have just added a patch titled KVM: MIPS: Don't leak FPU/DSP to guest to the linux-3.13.y-queue branch of the 3.13.y-ckt extended stable tree which can be found at: http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.13.y-queue This patch is scheduled to be released in version 3.13.11-ckt18. If you, or anyone else, feels it should not be added to this tree, please reply to this email. For more information about the 3.13.y-ckt tree, see https://wiki.ubuntu.com/Kernel/Dev/ExtendedStable Thanks. -Kamal -- >From adb94d141d17042e7eee5118f4f6358bfa61ffd9 Mon Sep 17 00:00:00 2001 From: James Hogan Date: Wed, 4 Feb 2015 17:06:37 + Subject: KVM: MIPS: Don't leak FPU/DSP to guest commit f798217dfd038af981a18bbe4bc57027a08bb182 upstream. The FPU and DSP are enabled via the CP0 Status CU1 and MX bits by kvm_mips_set_c0_status() on a guest exit, presumably in case there is active state that needs saving if pre-emption occurs. However neither of these bits are cleared again when returning to the guest. This effectively gives the guest access to the FPU/DSP hardware after the first guest exit even though it is not aware of its presence, allowing FP instructions in guest user code to intermittently actually execute instead of trapping into the guest OS for emulation. It will then read & manipulate the hardware FP registers which technically belong to the user process (e.g. QEMU), or are stale from another user process. It can also crash the guest OS by causing an FP exception, for which a guest exception handler won't have been registered. First lets save and disable the FPU (and MSA) state with lose_fpu(1) before entering the guest. This simplifies the problem, especially for when guest FPU/MSA support is added in the future, and prevents FR=1 FPU state being live when the FR bit gets cleared for the guest, which according to the architecture causes the contents of the FPU and vector registers to become UNPREDICTABLE. We can then safely remove the enabling of the FPU in kvm_mips_set_c0_status(), since there should never be any active FPU or MSA state to save at pre-emption, which should plug the FPU leak. DSP state is always live rather than being lazily restored, so for that it is simpler to just clear the MX bit again when re-entering the guest. Signed-off-by: James Hogan Cc: Paolo Bonzini Cc: Ralf Baechle Cc: Sanjay Lal Cc: Gleb Natapov Cc: kvm@vger.kernel.org Cc: linux-m...@linux-mips.org Signed-off-by: Paolo Bonzini [ luis: backported to 3.16: files rename: - locore.S -> kvm_locore.S - mips.c -> kvm_mips.c ] Signed-off-by: Luis Henriques Signed-off-by: Kamal Mostafa --- arch/mips/kvm/kvm_locore.S | 2 +- arch/mips/kvm/kvm_mips.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/mips/kvm/kvm_locore.S b/arch/mips/kvm/kvm_locore.S index bbace09..03a2db5 100644 --- a/arch/mips/kvm/kvm_locore.S +++ b/arch/mips/kvm/kvm_locore.S @@ -428,7 +428,7 @@ __kvm_mips_return_to_guest: /* Setup status register for running guest in UM */ .setat or v1, v1, (ST0_EXL | KSU_USER | ST0_IE) - and v1, v1, ~ST0_CU0 + and v1, v1, ~(ST0_CU0 | ST0_MX) .setnoat mtc0v1, CP0_STATUS ehb diff --git a/arch/mips/kvm/kvm_mips.c b/arch/mips/kvm/kvm_mips.c index 4d058a7..bdc5eeb 100644 --- a/arch/mips/kvm/kvm_mips.c +++ b/arch/mips/kvm/kvm_mips.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include @@ -424,6 +425,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) vcpu->mmio_needed = 0; } + lose_fpu(1); + local_irq_disable(); /* Check if we have any exceptions/interrupts pending */ kvm_mips_deliver_interrupts(vcpu, @@ -1028,9 +1031,6 @@ void kvm_mips_set_c0_status(void) { uint32_t status = read_c0_status(); - if (cpu_has_fpu) - status |= (ST0_CU1); - if (cpu_has_dsp) status |= (ST0_MX); -- 1.9.1 -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
[3.16.y-ckt stable] Patch "KVM: MIPS: Don't leak FPU/DSP to guest" has been added to staging queue
This is a note to let you know that I have just added a patch titled KVM: MIPS: Don't leak FPU/DSP to guest to the linux-3.16.y-queue branch of the 3.16.y-ckt extended stable tree which can be found at: http://kernel.ubuntu.com/git?p=ubuntu/linux.git;a=shortlog;h=refs/heads/linux-3.16.y-queue This patch is scheduled to be released in version 3.16.7-ckt8. If you, or anyone else, feels it should not be added to this tree, please reply to this email. For more information about the 3.16.y-ckt tree, see https://wiki.ubuntu.com/Kernel/Dev/ExtendedStable Thanks. -Luis -- >From b4db76b6e5898d4f4389cc944e5262788fa90d8c Mon Sep 17 00:00:00 2001 From: James Hogan Date: Wed, 4 Feb 2015 17:06:37 + Subject: KVM: MIPS: Don't leak FPU/DSP to guest commit f798217dfd038af981a18bbe4bc57027a08bb182 upstream. The FPU and DSP are enabled via the CP0 Status CU1 and MX bits by kvm_mips_set_c0_status() on a guest exit, presumably in case there is active state that needs saving if pre-emption occurs. However neither of these bits are cleared again when returning to the guest. This effectively gives the guest access to the FPU/DSP hardware after the first guest exit even though it is not aware of its presence, allowing FP instructions in guest user code to intermittently actually execute instead of trapping into the guest OS for emulation. It will then read & manipulate the hardware FP registers which technically belong to the user process (e.g. QEMU), or are stale from another user process. It can also crash the guest OS by causing an FP exception, for which a guest exception handler won't have been registered. First lets save and disable the FPU (and MSA) state with lose_fpu(1) before entering the guest. This simplifies the problem, especially for when guest FPU/MSA support is added in the future, and prevents FR=1 FPU state being live when the FR bit gets cleared for the guest, which according to the architecture causes the contents of the FPU and vector registers to become UNPREDICTABLE. We can then safely remove the enabling of the FPU in kvm_mips_set_c0_status(), since there should never be any active FPU or MSA state to save at pre-emption, which should plug the FPU leak. DSP state is always live rather than being lazily restored, so for that it is simpler to just clear the MX bit again when re-entering the guest. Signed-off-by: James Hogan Cc: Paolo Bonzini Cc: Ralf Baechle Cc: Sanjay Lal Cc: Gleb Natapov Cc: kvm@vger.kernel.org Cc: linux-m...@linux-mips.org Signed-off-by: Paolo Bonzini [ luis: backported to 3.16: files rename: - locore.S -> kvm_locore.S - mips.c -> kvm_mips.c ] Signed-off-by: Luis Henriques --- arch/mips/kvm/kvm_locore.S | 2 +- arch/mips/kvm/kvm_mips.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/mips/kvm/kvm_locore.S b/arch/mips/kvm/kvm_locore.S index 033ac343e72c..17376cd838e6 100644 --- a/arch/mips/kvm/kvm_locore.S +++ b/arch/mips/kvm/kvm_locore.S @@ -428,7 +428,7 @@ __kvm_mips_return_to_guest: /* Setup status register for running guest in UM */ .setat or v1, v1, (ST0_EXL | KSU_USER | ST0_IE) - and v1, v1, ~ST0_CU0 + and v1, v1, ~(ST0_CU0 | ST0_MX) .setnoat mtc0v1, CP0_STATUS ehb diff --git a/arch/mips/kvm/kvm_mips.c b/arch/mips/kvm/kvm_mips.c index f3c56a182fd8..d84f96e51349 100644 --- a/arch/mips/kvm/kvm_mips.c +++ b/arch/mips/kvm/kvm_mips.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include @@ -413,6 +414,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) vcpu->mmio_needed = 0; } + lose_fpu(1); + local_irq_disable(); /* Check if we have any exceptions/interrupts pending */ kvm_mips_deliver_interrupts(vcpu, @@ -1028,9 +1031,6 @@ void kvm_mips_set_c0_status(void) { uint32_t status = read_c0_status(); - if (cpu_has_fpu) - status |= (ST0_CU1); - if (cpu_has_dsp) status |= (ST0_MX); -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH] KVM: MIPS: Don't leak FPU/DSP to guest
On Tue, Feb 10, 2015 at 09:01:07AM +0100, Paolo Bonzini wrote: > -BEGIN PGP SIGNED MESSAGE- > Hash: SHA1 > > > > On 09/02/2015 23:58, James Hogan wrote: > >> First lets save and disable the FPU (and MSA) state with > >> lose_fpu(1) > > > > Please don't apply this patch yet. lose_fpu() uses function > > symbols which aren't exported for modules to use yet, so that'll > > need fixing first or KVM won't build as a module. > > Well, too late. :) > > James/Ralf, should I revert, or can that be fixed during the RC period? Okay no problem. I have patches ready so I'll submit today. Sorry about that! Cheers James > > Paolo > -BEGIN PGP SIGNATURE- > Version: GnuPG v2 > > iQEcBAEBAgAGBQJU2brAAAoJEL/70l94x66DdXIIAImur1pdSKYWw1+FzZH+H8Xo > 86j9EfptORk554o0a62LG9dOTY+5sJfAV9CoB7Q+8IfdLDKxpk1sLjMkiS0E0EWU > 2ilQfjYEXLTgCW38p03ype4m6g4uSfT16dnizrwnUviFk/EvVgCWHy88tA3+Vfn/ > WgoxcXkd+hguyNaLR2oAVqyNhAETLTo4kQQqKwGbXFXf0GLno44pj7bJprCR/jlO > 4+sUzuV5dno/GI6z8dyMmASo0QEy+IoXJ+aSw+IoRED9nlBMAS4+7uD4XfocGpca > En5KmXVnyJoazgV3Y6w2ymS606S0JNGRcOzqr8ZbOHtjJmAsZxjuVxP6PVzZqQg= > =ozzu > -END PGP SIGNATURE- signature.asc Description: Digital signature
Re: [PATCH] KVM: MIPS: Don't leak FPU/DSP to guest
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 09/02/2015 23:58, James Hogan wrote: >> First lets save and disable the FPU (and MSA) state with >> lose_fpu(1) > > Please don't apply this patch yet. lose_fpu() uses function > symbols which aren't exported for modules to use yet, so that'll > need fixing first or KVM won't build as a module. Well, too late. :) James/Ralf, should I revert, or can that be fixed during the RC period? Paolo -BEGIN PGP SIGNATURE- Version: GnuPG v2 iQEcBAEBAgAGBQJU2brAAAoJEL/70l94x66DdXIIAImur1pdSKYWw1+FzZH+H8Xo 86j9EfptORk554o0a62LG9dOTY+5sJfAV9CoB7Q+8IfdLDKxpk1sLjMkiS0E0EWU 2ilQfjYEXLTgCW38p03ype4m6g4uSfT16dnizrwnUviFk/EvVgCWHy88tA3+Vfn/ WgoxcXkd+hguyNaLR2oAVqyNhAETLTo4kQQqKwGbXFXf0GLno44pj7bJprCR/jlO 4+sUzuV5dno/GI6z8dyMmASo0QEy+IoXJ+aSw+IoRED9nlBMAS4+7uD4XfocGpca En5KmXVnyJoazgV3Y6w2ymS606S0JNGRcOzqr8ZbOHtjJmAsZxjuVxP6PVzZqQg= =ozzu -END PGP SIGNATURE- -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html
Re: [PATCH] KVM: MIPS: Don't leak FPU/DSP to guest
Hi Paolo, On Wed, Feb 04, 2015 at 05:06:37PM +, James Hogan wrote: > The FPU and DSP are enabled via the CP0 Status CU1 and MX bits by > kvm_mips_set_c0_status() on a guest exit, presumably in case there is > active state that needs saving if pre-emption occurs. However neither of > these bits are cleared again when returning to the guest. > > This effectively gives the guest access to the FPU/DSP hardware after > the first guest exit even though it is not aware of its presence, > allowing FP instructions in guest user code to intermittently actually > execute instead of trapping into the guest OS for emulation. It will > then read & manipulate the hardware FP registers which technically > belong to the user process (e.g. QEMU), or are stale from another user > process. It can also crash the guest OS by causing an FP exception, for > which a guest exception handler won't have been registered. > > First lets save and disable the FPU (and MSA) state with lose_fpu(1) Please don't apply this patch yet. lose_fpu() uses function symbols which aren't exported for modules to use yet, so that'll need fixing first or KVM won't build as a module. Thanks James > before entering the guest. This simplifies the problem, especially for > when guest FPU/MSA support is added in the future, and prevents FR=1 FPU > state being live when the FR bit gets cleared for the guest, which > according to the architecture causes the contents of the FPU and vector > registers to become UNPREDICTABLE. > > We can then safely remove the enabling of the FPU in > kvm_mips_set_c0_status(), since there should never be any active FPU or > MSA state to save at pre-emption, which should plug the FPU leak. > > DSP state is always live rather than being lazily restored, so for that > it is simpler to just clear the MX bit again when re-entering the guest. > > Signed-off-by: James Hogan > Cc: Paolo Bonzini > Cc: Ralf Baechle > Cc: Sanjay Lal > Cc: Gleb Natapov > Cc: kvm@vger.kernel.org > Cc: linux-m...@linux-mips.org > Cc: # v3.10+: 044f0f03eca0: MIPS: KVM: Deliver guest > interrupts > Cc: # v3.10+ > --- > arch/mips/kvm/locore.S | 2 +- > arch/mips/kvm/mips.c | 6 +++--- > 2 files changed, 4 insertions(+), 4 deletions(-) > > diff --git a/arch/mips/kvm/locore.S b/arch/mips/kvm/locore.S > index d7279c03c517..4a68b176d6e4 100644 > --- a/arch/mips/kvm/locore.S > +++ b/arch/mips/kvm/locore.S > @@ -434,7 +434,7 @@ __kvm_mips_return_to_guest: > /* Setup status register for running guest in UM */ > .setat > or v1, v1, (ST0_EXL | KSU_USER | ST0_IE) > - and v1, v1, ~ST0_CU0 > + and v1, v1, ~(ST0_CU0 | ST0_MX) > .setnoat > mtc0v1, CP0_STATUS > ehb > diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c > index dd133ccecec4..270bbd41769e 100644 > --- a/arch/mips/kvm/mips.c > +++ b/arch/mips/kvm/mips.c > @@ -15,6 +15,7 @@ > #include > #include > #include > +#include > #include > #include > #include > @@ -379,6 +380,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct > kvm_run *run) > vcpu->mmio_needed = 0; > } > > + lose_fpu(1); > + > local_irq_disable(); > /* Check if we have any exceptions/interrupts pending */ > kvm_mips_deliver_interrupts(vcpu, > @@ -987,9 +990,6 @@ static void kvm_mips_set_c0_status(void) > { > uint32_t status = read_c0_status(); > > - if (cpu_has_fpu) > - status |= (ST0_CU1); > - > if (cpu_has_dsp) > status |= (ST0_MX); > > -- > 2.0.5 > signature.asc Description: Digital signature
[PATCH] KVM: MIPS: Don't leak FPU/DSP to guest
The FPU and DSP are enabled via the CP0 Status CU1 and MX bits by kvm_mips_set_c0_status() on a guest exit, presumably in case there is active state that needs saving if pre-emption occurs. However neither of these bits are cleared again when returning to the guest. This effectively gives the guest access to the FPU/DSP hardware after the first guest exit even though it is not aware of its presence, allowing FP instructions in guest user code to intermittently actually execute instead of trapping into the guest OS for emulation. It will then read & manipulate the hardware FP registers which technically belong to the user process (e.g. QEMU), or are stale from another user process. It can also crash the guest OS by causing an FP exception, for which a guest exception handler won't have been registered. First lets save and disable the FPU (and MSA) state with lose_fpu(1) before entering the guest. This simplifies the problem, especially for when guest FPU/MSA support is added in the future, and prevents FR=1 FPU state being live when the FR bit gets cleared for the guest, which according to the architecture causes the contents of the FPU and vector registers to become UNPREDICTABLE. We can then safely remove the enabling of the FPU in kvm_mips_set_c0_status(), since there should never be any active FPU or MSA state to save at pre-emption, which should plug the FPU leak. DSP state is always live rather than being lazily restored, so for that it is simpler to just clear the MX bit again when re-entering the guest. Signed-off-by: James Hogan Cc: Paolo Bonzini Cc: Ralf Baechle Cc: Sanjay Lal Cc: Gleb Natapov Cc: kvm@vger.kernel.org Cc: linux-m...@linux-mips.org Cc: # v3.10+: 044f0f03eca0: MIPS: KVM: Deliver guest interrupts Cc: # v3.10+ --- arch/mips/kvm/locore.S | 2 +- arch/mips/kvm/mips.c | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-) diff --git a/arch/mips/kvm/locore.S b/arch/mips/kvm/locore.S index d7279c03c517..4a68b176d6e4 100644 --- a/arch/mips/kvm/locore.S +++ b/arch/mips/kvm/locore.S @@ -434,7 +434,7 @@ __kvm_mips_return_to_guest: /* Setup status register for running guest in UM */ .setat or v1, v1, (ST0_EXL | KSU_USER | ST0_IE) - and v1, v1, ~ST0_CU0 + and v1, v1, ~(ST0_CU0 | ST0_MX) .setnoat mtc0v1, CP0_STATUS ehb diff --git a/arch/mips/kvm/mips.c b/arch/mips/kvm/mips.c index dd133ccecec4..270bbd41769e 100644 --- a/arch/mips/kvm/mips.c +++ b/arch/mips/kvm/mips.c @@ -15,6 +15,7 @@ #include #include #include +#include #include #include #include @@ -379,6 +380,8 @@ int kvm_arch_vcpu_ioctl_run(struct kvm_vcpu *vcpu, struct kvm_run *run) vcpu->mmio_needed = 0; } + lose_fpu(1); + local_irq_disable(); /* Check if we have any exceptions/interrupts pending */ kvm_mips_deliver_interrupts(vcpu, @@ -987,9 +990,6 @@ static void kvm_mips_set_c0_status(void) { uint32_t status = read_c0_status(); - if (cpu_has_fpu) - status |= (ST0_CU1); - if (cpu_has_dsp) status |= (ST0_MX); -- 2.0.5 -- To unsubscribe from this list: send the line "unsubscribe kvm" in the body of a message to majord...@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html