From: Julien Grall <julien.grall@arm.com>
To: Wei Chen <Wei.Chen@arm.com>, xen-devel@lists.xen.org
Cc: Kaly.Xin@arm.com, nd@arm.com, sstabellini@kernel.org,
steve.capper@arm.com
Subject: Re: [PATCH v2 16/19] xen/arm: Introduce a helper to synchronize SError
Date: Thu, 30 Mar 2017 19:37:56 +0100 [thread overview]
Message-ID: <98b9a340-6f0d-528a-db14-eb991cb836c4@arm.com> (raw)
In-Reply-To: <4796c228-69fa-9bd7-49f3-63b46d8ad4b7@arm.com>
On 30/03/17 19:32, Julien Grall wrote:
> On 30/03/17 19:28, Julien Grall wrote:
>> Hi Wei,
>>
>> On 30/03/17 10:13, Wei Chen wrote:
>>> +void synchronize_serror(void)
>>
>> Sorry for been late in the party. Looking at the way you use the
>> function, you execute depending on the behavior chosen by the user when
>> an SErrors happen. This behavior will not change at runtime, so always
>> checking the option chosen in the hot path does not sound very efficient.
>>
>> I would recommend to look at ALTERNATIVE and streamline (dsb sy, isb).
>> I.e
>>
>> ALTERNATIVE("dsb sy; isb", "nop; nop", ...) or the invert depending of
>> the place.
>
> To complete what I was suggestion, you could define:
>
> /* Synchronize SError unless the feature is selected */
> #define SYNCHRONIZE_SERROR(feat) ALTERNATIVE("dsb sy; isb", "nop; nop");
Or even:
/*
* Synchronize SError unless the feature is selected.
* This is relying on the SErrors are currently unmasked.
*/
#define SYNCHRONIZE_SERROR(feat) \
do { \
ASSERT(cpus_have_cap(feat) && local_abort_is_enabled());\
ALTERNATIVE("dsb sy; isb", "nop; nop"); \
while (0)
The ASSERT is here to check that we have abort enabled. Otherwise, doing
the synchronization would be pointless.
Note that the function local_abort_is_enabled is not implemented. But it
is easy to write it. Looking how local_irq_is_enabled was implemented.
Cheers,
--
Julien Grall
_______________________________________________
Xen-devel mailing list
Xen-devel@lists.xen.org
https://lists.xen.org/xen-devel
next prev parent reply other threads:[~2017-03-30 18:37 UTC|newest]
Thread overview: 57+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-03-30 9:13 [PATCH v2 00/19] Provide a command line option to choose how to handle SErrors Wei Chen
2017-03-30 9:13 ` [PATCH v2 01/19] xen/arm: Save ESR_EL2 to avoid using mismatched value in syndrome check Wei Chen
2017-03-30 13:31 ` Julien Grall
2017-03-31 3:26 ` Wei Chen
2017-03-30 9:13 ` [PATCH v2 02/19] xen/arm: Remove vwfi while setting HCR_EL2 in init_traps Wei Chen
2017-03-30 17:05 ` Julien Grall
2017-03-30 22:29 ` Stefano Stabellini
2017-03-31 5:58 ` Wei Chen
2017-03-31 8:34 ` Julien Grall
2017-03-30 9:13 ` [PATCH v2 03/19] xen/arm: Move parse_vwfi from trap.c to domain.c Wei Chen
2017-03-30 9:13 ` [PATCH v2 04/19] xen/arm: Restore HCR_EL2 register Wei Chen
2017-03-30 17:07 ` Julien Grall
2017-03-30 22:03 ` Stefano Stabellini
2017-03-31 2:10 ` Wei Chen
2017-03-31 8:39 ` Julien Grall
2017-03-31 8:59 ` Wei Chen
2017-03-30 9:13 ` [PATCH v2 05/19] xen/arm: Avoid setting/clearing HCR_RW at every context switch Wei Chen
2017-03-30 17:12 ` Julien Grall
2017-03-30 21:21 ` Stefano Stabellini
2017-03-30 9:13 ` [PATCH v2 06/19] xen/arm: Save HCR_EL2 when a guest took the SError Wei Chen
2017-03-30 9:13 ` [PATCH v2 07/19] xen/arm: Introduce a virtual abort injection helper Wei Chen
2017-03-30 17:20 ` Julien Grall
2017-03-30 21:24 ` Stefano Stabellini
2017-03-31 5:25 ` Wei Chen
2017-03-30 9:13 ` [PATCH v2 08/19] xen/arm: Introduce a command line parameter for SErrors/Aborts Wei Chen
2017-03-30 17:39 ` Julien Grall
2017-03-31 5:28 ` Wei Chen
2017-03-30 9:13 ` [PATCH v2 09/19] xen/arm: Introduce a initcall to update cpu_hwcaps by serror_op Wei Chen
2017-03-30 17:51 ` Julien Grall
2017-03-30 18:02 ` Julien Grall
2017-03-30 21:28 ` Stefano Stabellini
2017-03-31 8:50 ` Julien Grall
2017-03-30 9:13 ` [PATCH v2 10/19] xen/arm64: Use alternative to skip the check of pending serrors Wei Chen
2017-03-30 9:13 ` [PATCH v2 11/19] xen/arm32: " Wei Chen
2017-03-30 18:06 ` Julien Grall
2017-03-30 21:29 ` Stefano Stabellini
2017-03-31 5:33 ` Wei Chen
2017-03-30 9:13 ` [PATCH v2 12/19] xen/arm: Move macro VABORT_GEN_BY_GUEST to common header Wei Chen
2017-03-30 21:36 ` Stefano Stabellini
2017-03-31 5:35 ` Wei Chen
2017-03-30 9:13 ` [PATCH v2 13/19] xen/arm: Introduce new helpers to handle guest/hyp SErrors Wei Chen
2017-03-30 9:13 ` [PATCH v2 14/19] xen/arm: Replace do_trap_guest_serror with new helpers Wei Chen
2017-03-30 9:13 ` [PATCH v2 15/19] xen/arm: Unmask the Abort/SError bit in the exception entries Wei Chen
2017-03-30 9:13 ` [PATCH v2 16/19] xen/arm: Introduce a helper to synchronize SError Wei Chen
2017-03-30 18:28 ` Julien Grall
2017-03-30 18:32 ` Julien Grall
2017-03-30 18:37 ` Julien Grall [this message]
2017-03-31 5:51 ` Wei Chen
2017-03-31 10:55 ` Wei Chen
2017-03-31 11:06 ` Julien Grall
2017-03-31 11:09 ` Wei Chen
2017-03-30 9:13 ` [PATCH v2 17/19] xen/arm: Isolate the SError between the context switch of 2 vCPUs Wei Chen
2017-03-30 21:49 ` Stefano Stabellini
2017-03-30 22:00 ` Julien Grall
2017-03-31 5:52 ` Wei Chen
2017-03-30 9:13 ` [PATCH v2 18/19] xen/arm: Prevent slipping hypervisor SError to guest Wei Chen
2017-03-30 9:13 ` [PATCH v2 19/19] xen/arm: Handle guest external abort as guest SError Wei Chen
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=98b9a340-6f0d-528a-db14-eb991cb836c4@arm.com \
--to=julien.grall@arm.com \
--cc=Kaly.Xin@arm.com \
--cc=Wei.Chen@arm.com \
--cc=nd@arm.com \
--cc=sstabellini@kernel.org \
--cc=steve.capper@arm.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).