linux-kernel - Re: [PATCH v3 2/2] arch/powerpc/kvm: kvmppc_hv

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <87v8itt1y8.fsf@mpe.ellerman.id.au>
Date:   Wed, 22 Mar 2023 23:17:35 +1100
From:   Michael Ellerman <mpe@...erman.id.au>
To:     Kautuk Consul <kconsul@...ux.vnet.ibm.com>,
        Nicholas Piggin <npiggin@...il.com>,
        Christophe Leroy <christophe.leroy@...roup.eu>,
        Fabiano Rosas <farosas@...ux.ibm.com>,
        Sathvika Vasireddy <sv@...ux.ibm.com>,
        Alexey Kardashevskiy <aik@...abs.ru>
Cc:     linuxppc-dev@...ts.ozlabs.org, linux-kernel@...r.kernel.org,
        Kautuk Consul <kconsul@...ux.vnet.ibm.com>
Subject: Re: [PATCH v3 2/2] arch/powerpc/kvm: kvmppc_hv_entry: remove r4
 argument

Kautuk Consul <kconsul@...ux.vnet.ibm.com> writes:
> kvmppc_hv_entry is called from only 2 locations within
> book3s_hv_rmhandlers.S. Both of those locations set r4
> as HSTATE_KVM_VCPU(r13) before calling kvmppc_hv_entry.
> So, shift the r4 load instruction to kvmppc_hv_entry and
> thus modify the calling convention of this function.
>
> Signed-off-by: Kautuk Consul <kconsul@...ux.vnet.ibm.com>
> ---
>  arch/powerpc/kvm/book3s_hv_rmhandlers.S | 11 ++++++-----
>  1 file changed, 6 insertions(+), 5 deletions(-)
>
> diff --git a/arch/powerpc/kvm/book3s_hv_rmhandlers.S b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> index b81ba4ee0521..b61f0b2c677b 100644
> --- a/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> +++ b/arch/powerpc/kvm/book3s_hv_rmhandlers.S
> @@ -85,7 +85,7 @@ _GLOBAL_TOC(kvmppc_hv_entry_trampoline)
>  	RFI_TO_KERNEL
>  
>  kvmppc_call_hv_entry:
> -	ld	r4, HSTATE_KVM_VCPU(r13)
> +	/* Enter guest. */
>  	bl	kvmppc_hv_entry
>  
>  	/* Back from guest - restore host state and return to caller */
> @@ -352,9 +352,7 @@ kvm_secondary_got_guest:
>  	mtspr	SPRN_LDBAR, r0
>  	isync
>  63:
> -	/* Order load of vcpu after load of vcore */
> -	lwsync
> -	ld	r4, HSTATE_KVM_VCPU(r13)
> +	/* Enter guest. */
>  	bl	kvmppc_hv_entry
>  
>  	/* Back from the guest, go back to nap */
> @@ -506,7 +504,6 @@ SYM_INNER_LABEL(kvmppc_hv_entry, SYM_L_LOCAL)
>  
>  	/* Required state:
>  	 *
> -	 * R4 = vcpu pointer (or NULL)
>  	 * MSR = ~IR|DR
>  	 * R13 = PACA
>  	 * R1 = host R1
> @@ -524,6 +521,10 @@ SYM_INNER_LABEL(kvmppc_hv_entry, SYM_L_LOCAL)
>  	li	r6, KVM_GUEST_MODE_HOST_HV
>  	stb	r6, HSTATE_IN_GUEST(r13)
>  
> +	/* Order load of vcpu after load of vcore */

Just copying the comment here doesn't work. It doesn't make sense on its
own here, because the VCORE is loaded (again) a few lines below (536).
So as written this comment seems backward vs the code.

The comment would need to expand to explain that the barrier is for the
case where we came from kvm_secondary_got_guest.

> +	lwsync
> +	ld	r4, HSTATE_KVM_VCPU(r13)
> +
>  #ifdef CONFIG_KVM_BOOK3S_HV_P8_TIMING
>  	/* Store initial timestamp */
>  	cmpdi	r4, 0


But as Nick says I don't think it's worth investing effort in small
tweaks to this code. The risk of introducing bugs is too high for such a
small improvement to the code.

Thanks for trying, but I think this asm code is best left more or less
alone unless we find actual bugs in it - or unless we can make
substantial improvements to it, which would be rewriting in C, or at
least converting to a fully call/return style rather than the current
forest of labels.

I will take patch 1 though, as that's an obvious cleanup and poses no
risk (famous last words :D).

cheers