From: Julien Grall <julien.grall@linaro.org>
To: Ian Campbell <ian.campbell@citrix.com>, xen-devel@lists.xen.org
Cc: tim@xen.org, vijay.kilari@gmail.com, stefano.stabellini@eu.citrix.com
Subject: Re: [PATCH RFC 5/9] xen: arm: Defer setting of VTCR_EL2 until after CPUs are up
Date: Wed, 30 Jul 2014 18:11:27 +0100 [thread overview]
Message-ID: <53D9273F.2060404@linaro.org> (raw)
In-Reply-To: <c97b356cddccf5905c335d63e7f9b4a8432a7d48.1406728037.git.ian.campbell@citrix.com>
Hi Ian,
On 07/30/2014 02:47 PM, Ian Campbell wrote:
> Currently we retain the hardcoded values but soon we will want to
> calculate the correct values based upon the CPU properties common to all processors, which are only available once they are all up.
NIT: I think it misses a newline on the second line. It looks strange to
see the first line short and not the second.
> Signed-off-by: Ian Campbell <ian.campbell@citrix.com>
> ---
> xen/arch/arm/p2m.c | 16 +++++++++++++---
> xen/arch/arm/setup.c | 4 ++--
> xen/arch/arm/smpboot.c | 2 --
> 3 files changed, 15 insertions(+), 7 deletions(-)
>
> diff --git a/xen/arch/arm/p2m.c b/xen/arch/arm/p2m.c
> index 705b29b..225d125 100644
> --- a/xen/arch/arm/p2m.c
> +++ b/xen/arch/arm/p2m.c
> @@ -1059,8 +1059,17 @@ err:
> return page;
> }
>
> +static void setup_virt_paging_one(void *data)
I would add __init.
> +{
> + unsigned long val = (unsigned long)data;
VTCR_EL2 is a 32 bit register. I would use uint32_t for the variable type.
> + WRITE_SYSREG32(val, VTCR_EL2);
> + isb();
> +}
> +
> void __cpuinit setup_virt_paging(void)
This can become __init now.
> {
> + unsigned long val;
> +
> /* Setup Stage 2 address translation */
> /* SH0=11 (Inner-shareable)
> * ORGN0=IRGN0=01 (Normal memory, Write-Back Write-Allocate Cacheable)
> @@ -1070,11 +1079,12 @@ void __cpuinit setup_virt_paging(void)
> * PS=010 == 40 bits
> */
> #ifdef CONFIG_ARM_32
> - WRITE_SYSREG32(0x80003558, VTCR_EL2);
> + val = 0x80003558;
> #else
> - WRITE_SYSREG32(0x80023558, VTCR_EL2);
> + val = 0x80023558;
> #endif
> - isb();
> + setup_virt_paging_one((void *)val);
> + smp_call_function(setup_virt_paging_one, (void *)val, 1);
> }
> /*
> diff --git a/xen/arch/arm/setup.c b/xen/arch/arm/setup.c
> index 446b4dc..ddaef7e 100644
> --- a/xen/arch/arm/setup.c
> +++ b/xen/arch/arm/setup.c
> @@ -788,8 +788,6 @@ void __init start_xen(unsigned long boot_phys_offset,
>
> gic_init();
>
> - setup_virt_paging();
> -
> p2m_vmid_allocator_init();
>
> softirq_init();
> @@ -838,6 +836,8 @@ void __init start_xen(unsigned long boot_phys_offset,
>
> do_initcalls();
>
> + setup_virt_paging();
> +
Assuming there is no CPU hotplug, which IIRC is not yet support, this is
only depends on the SMP bring up. Can we move this before do_initcalls?
So if we need to initialize some code that is relying on VTCR_EL2, it
will be possible.
I was mostly thinking about IOMMU drivers where a similar trick will be
necessary sooner or later.
Regards,
--
Julien Grall
next prev parent reply other threads:[~2014-07-30 17:11 UTC|newest]
Thread overview: 30+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-07-30 13:44 [PATCH RFC 0/9] xen: arm: support for > 40-bit physical addressing Ian Campbell
2014-07-30 13:47 ` [PATCH RFC 1/9] xen: arm: rename p2m->first_level to p2m->root Ian Campbell
2014-07-30 13:47 ` [PATCH RFC 2/9] xen: arm: Implement variable levels in dump_pt_walk Ian Campbell
2014-07-30 16:40 ` Julien Grall
2014-07-30 13:47 ` [PATCH RFC 3/9] xen: arm: handle concatenated root tables " Ian Campbell
2014-07-30 16:58 ` Julien Grall
2014-09-04 14:40 ` Ian Campbell
2014-09-08 20:54 ` Julien Grall
2014-07-30 13:47 ` [PATCH RFC 4/9] xen: arm: move setup_virt_paging to p2m.c Ian Campbell
2014-07-30 17:00 ` Julien Grall
2014-07-30 13:47 ` [PATCH RFC 5/9] xen: arm: Defer setting of VTCR_EL2 until after CPUs are up Ian Campbell
2014-07-30 17:11 ` Julien Grall [this message]
2014-09-04 14:50 ` Ian Campbell
2014-07-30 13:47 ` [PATCH RFC 6/9] xen: arm: handle variable p2m levels in p2m_lookup Ian Campbell
2014-07-31 11:14 ` Julien Grall
2014-09-04 14:54 ` Ian Campbell
2014-07-30 13:47 ` [PATCH RFC 7/9] xen: arm: handle variable p2m levels in apply_p2m_changes Ian Campbell
2014-07-31 15:38 ` Julien Grall
2014-08-11 7:00 ` Vijay Kilari
2014-08-26 9:11 ` Vijay Kilari
2014-09-04 14:01 ` Ian Campbell
2014-07-30 13:47 ` [PATCH RFC 8/9] xen: arm: support for up to 48-bit physical addressing on arm64 Ian Campbell
2014-08-07 15:33 ` Julien Grall
2014-07-30 13:47 ` [PATCH RFC 9/9] xen: arm: support for up to 48-bit IPA " Ian Campbell
2014-08-07 15:49 ` Julien Grall
2014-07-30 16:06 ` [PATCH RFC 1/9] xen: arm: rename p2m->first_level to p2m->root Julien Grall
2014-07-30 16:19 ` Ian Campbell
2014-07-30 16:23 ` Julien Grall
2014-07-31 8:22 ` [PATCH RFC 0/9] xen: arm: support for > 40-bit physical addressing Vijay Kilari
2014-07-31 8:54 ` Ian Campbell
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=53D9273F.2060404@linaro.org \
--to=julien.grall@linaro.org \
--cc=ian.campbell@citrix.com \
--cc=stefano.stabellini@eu.citrix.com \
--cc=tim@xen.org \
--cc=vijay.kilari@gmail.com \
--cc=xen-devel@lists.xen.org \
/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).