From: Mark Rutland <mark.rutland@arm.com>
To: Alexandru Elisei <alexandru.elisei@arm.com>
Cc: Andre Przywara <andre.przywara@arm.com>,
kvm@vger.kernel.org, pbonzini@redhat.com, drjones@redhat.com,
maz@kernel.org, vladimir.murzin@arm.com
Subject: Re: [kvm-unit-tests PATCH v3 06/18] arm/arm64: psci: Don't run C code without stack or vectors
Date: Mon, 6 Jan 2020 11:41:49 +0000 [thread overview]
Message-ID: <20200106114149.GB9630@lakrids.cambridge.arm.com> (raw)
In-Reply-To: <61ea7391-7e65-4548-17b6-7dbd977fa394@arm.com>
On Mon, Jan 06, 2020 at 10:41:55AM +0000, Alexandru Elisei wrote:
> On 1/2/20 6:11 PM, Andre Przywara wrote:
> > On Tue, 31 Dec 2019 16:09:37 +0000
> > Alexandru Elisei <alexandru.elisei@arm.com> wrote:
> >> +.global asm_cpu_psci_cpu_die
> >> +asm_cpu_psci_cpu_die:
> >> + ldr r0, =PSCI_0_2_FN_CPU_OFF
> >> + hvc #0
> >> + b .
> > I am wondering if this implementation is actually too simple. Both
> > the current implementation and the kernel clear at least the first
> > three arguments to 0.
> > I failed to find a requirement for doing this (nothing in the SMCCC
> > or the PSCI spec), but I guess it would make sense when looking at
> > forward compatibility.
>
> The SMC calling convention only specifies the values for the arguments that are
> used by a function, not the values for all possible arguments. kvm-unit-tests sets
> the other arguments to 0 because the function prototype that does the actual SMC
> call takes 4 arguments. The value 0 is a random value that was chosen for those
> unused parameters. For example, it could have been a random number on each call.
That's correct.
A caller can leave arbitrary values in non-argument registers, in the
same manner as a caller of an AAPCS function. The callee should not
consume those values as they are not arguments.
> Let me put it another way. Suggesting that unused arguments should be set to 0 is
> the same as suggesting that normal C function that adheres to procedure call
> standard for arm64 should always have 8 arguments, and for a particular function
> that doesn't use all of them, they should be set to 0 by the caller.
Heh, same rationale. :)
> @Mark Rutland has worked on the SMC implementation for the Linux kernel, if he
> wants to chime in on this.
>
> Thanks,
> Alex
> > At the very least it's a change in behaviour (ignoring the missing printf).
> > So shall we just clear r1, r2 and r3 here? (Same for arm64 below)
There's no need to zero non-argument registers, and it could potentially
mask bugs in callees, so I don't think it's a good idea to do so.
If you really want to test that the callee is robust and correct, it
would be better to randomize the content of non-argument regsiters to
fuzz the callee.
Thanks,
Mark.
next prev parent reply other threads:[~2020-01-06 11:41 UTC|newest]
Thread overview: 47+ messages / expand[flat|nested] mbox.gz Atom feed top
2019-12-31 16:09 [kvm-unit-tests PATCH v3 00/18] arm/arm64: Various fixes Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 01/18] lib: arm/arm64: Remove unnecessary dcache maintenance operations Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 02/18] lib: arm: Add proper data synchronization barriers for TLBIs Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 03/18] lib: Add WRITE_ONCE and READ_ONCE implementations in compiler.h Alexandru Elisei
2020-01-02 18:03 ` Andre Przywara
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 04/18] lib: arm/arm64: Use WRITE_ONCE to update the translation tables Alexandru Elisei
2020-01-02 18:06 ` Andre Przywara
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 05/18] lib: arm/arm64: Remove unused CPU_OFF parameter Alexandru Elisei
2020-01-02 18:11 ` Andre Przywara
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 06/18] arm/arm64: psci: Don't run C code without stack or vectors Alexandru Elisei
2020-01-02 18:11 ` Andre Przywara
2020-01-03 15:31 ` Andrew Jones
2020-01-06 11:02 ` Alexandru Elisei
2020-01-06 13:03 ` Andrew Jones
2020-01-06 14:03 ` Alexandru Elisei
2020-01-06 10:41 ` Alexandru Elisei
2020-01-06 11:17 ` Andre Przywara
2020-01-06 11:28 ` Alexandru Elisei
2020-01-06 11:36 ` Mark Rutland
2020-01-06 11:41 ` Mark Rutland [this message]
2020-01-06 13:17 ` Andrew Jones
2020-01-06 14:12 ` Alexandru Elisei
2020-01-06 15:20 ` Andrew Jones
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 07/18] lib: arm/arm64: Add missing include for alloc_page.h in pgtable.h Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 08/18] lib: arm: Implement flush_tlb_all Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 09/18] lib: arm/arm64: Teach mmu_clear_user about block mappings Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 10/18] arm/arm64: selftest: Add prefetch abort test Alexandru Elisei
2020-01-06 9:24 ` Andrew Jones
2020-01-06 11:03 ` Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 11/18] arm64: timer: Write to ICENABLER to disable timer IRQ Alexandru Elisei
2020-01-03 13:36 ` Andre Przywara
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 12/18] arm64: timer: EOIR the interrupt after masking the timer Alexandru Elisei
2020-01-03 13:36 ` Andre Przywara
2020-01-06 11:35 ` Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 13/18] arm64: timer: Test behavior when timer disabled or masked Alexandru Elisei
2020-01-03 13:37 ` Andre Przywara
2020-01-06 13:22 ` Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 14/18] lib: arm/arm64: Refuse to disable the MMU with non-identity stack pointer Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 15/18] arm/arm64: Perform dcache clean + invalidate after turning MMU off Alexandru Elisei
2020-01-03 16:49 ` Andre Przywara
2020-01-06 14:27 ` Alexandru Elisei
2020-01-06 16:28 ` Andrew Jones
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 16/18] arm: cstart64.S: Downgrade TLBI to non-shareable in asm_mmu_enable Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 17/18] arm/arm64: Invalidate TLB before enabling MMU Alexandru Elisei
2019-12-31 16:09 ` [kvm-unit-tests PATCH v3 18/18] arm: cstart64.S: Remove icache invalidation from asm_mmu_enable Alexandru Elisei
2020-01-06 9:28 ` [kvm-unit-tests PATCH v3 00/18] arm/arm64: Various fixes Andrew Jones
2020-01-09 10:01 ` Alexandru Elisei
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=20200106114149.GB9630@lakrids.cambridge.arm.com \
--to=mark.rutland@arm.com \
--cc=alexandru.elisei@arm.com \
--cc=andre.przywara@arm.com \
--cc=drjones@redhat.com \
--cc=kvm@vger.kernel.org \
--cc=maz@kernel.org \
--cc=pbonzini@redhat.com \
--cc=vladimir.murzin@arm.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).