Linux MIPS Architecture development
 help / color / mirror / Atom feed
From: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com>
To: James Hogan <james.hogan@imgtec.com>
Cc: Ralf Baechle <ralf@linux-mips.org>, <linux-mips@linux-mips.org>
Subject: Re: [PATCH 7/9] MIPS: uprobes: Flush icache via kernel address
Date: Thu, 22 Sep 2016 15:27:41 -0700	[thread overview]
Message-ID: <57E45ADD.6000202@imgtec.com> (raw)
In-Reply-To: <7C466D41-C786-48E0-9BFB-1024D6F9AFFC@imgtec.com>

On 09/22/2016 03:13 PM, James Hogan wrote:
> well it'll do a protected dcache flush (i.e. using CACHEE with EVA). Would kmap/kunmap or variants (fixed to work with aliasing dcache) be able to take care of colouring / further flushing?

We should flush kernel D-cache and user I-cache in any cache aliasing 
system. I was wrong - a fixed HIGHMEM doesn't do any difference 
actually, because page may be located in directly addressed memory (all 
HIGHMEM stuff is irrelevant in this case, kmap returns a lowmem address).

>
> In any case, simply changing to the user_ one is a no-op compared to leaving as is where patch 9 would probably break it on EVA by making it operate only on kernel addresses.

EVA or not has no difference here - kernel address can still be a 
different color to user address.

And keeping kernel I-cache flush does break it really, not EVA.

- Leonid.

WARNING: multiple messages have this Message-ID (diff)
From: Leonid Yegoshin <Leonid.Yegoshin@imgtec.com>
To: James Hogan <james.hogan@imgtec.com>
Cc: Ralf Baechle <ralf@linux-mips.org>, linux-mips@linux-mips.org
Subject: Re: [PATCH 7/9] MIPS: uprobes: Flush icache via kernel address
Date: Thu, 22 Sep 2016 15:27:41 -0700	[thread overview]
Message-ID: <57E45ADD.6000202@imgtec.com> (raw)
Message-ID: <20160922222741.NZc9MSWFZ1QvYwLyP0EyebAvBfir6IhGmV6GKY-NieA@z> (raw)
In-Reply-To: <7C466D41-C786-48E0-9BFB-1024D6F9AFFC@imgtec.com>

On 09/22/2016 03:13 PM, James Hogan wrote:
> well it'll do a protected dcache flush (i.e. using CACHEE with EVA). Would kmap/kunmap or variants (fixed to work with aliasing dcache) be able to take care of colouring / further flushing?

We should flush kernel D-cache and user I-cache in any cache aliasing 
system. I was wrong - a fixed HIGHMEM doesn't do any difference 
actually, because page may be located in directly addressed memory (all 
HIGHMEM stuff is irrelevant in this case, kmap returns a lowmem address).

>
> In any case, simply changing to the user_ one is a no-op compared to leaving as is where patch 9 would probably break it on EVA by making it operate only on kernel addresses.

EVA or not has no difference here - kernel address can still be a 
different color to user address.

And keeping kernel I-cache flush does break it really, not EVA.

- Leonid.

  reply	other threads:[~2016-09-22 22:27 UTC|newest]

Thread overview: 51+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-09-01 16:30 [PATCH 0/9] MIPS: General EVA fixes & cleanups James Hogan
2016-09-01 16:30 ` [PATCH 1/9] MIPS: traps: 64bit kernels should read CP0_EBase 64bit James Hogan
2016-09-01 16:30   ` James Hogan
2016-09-21 13:08   ` Ralf Baechle
2016-09-21 15:01     ` Matt Redfearn
2016-09-21 15:01       ` Matt Redfearn
2016-10-02 10:30       ` Maciej W. Rozycki
2016-10-05 15:56         ` James Hogan
2016-10-05 15:56           ` James Hogan
2016-10-06 16:18           ` Maciej W. Rozycki
2016-10-06 18:05             ` James Hogan
2016-10-06 18:05               ` James Hogan
2016-10-06 19:56               ` Maciej W. Rozycki
2016-10-06 20:19                 ` James Hogan
2016-10-06 20:19                   ` James Hogan
2016-10-06 22:41                   ` Maciej W. Rozycki
2016-10-06 22:50                     ` James Hogan
2016-10-06 22:50                       ` James Hogan
2016-10-06 23:07                       ` Maciej W. Rozycki
2016-10-07 15:35             ` David Daney
2016-10-07 15:41               ` David Daney
2016-10-07 17:39                 ` Maciej W. Rozycki
2016-09-01 16:30 ` [PATCH 2/9] MIPS: traps: Convert ebase to KSeg0 James Hogan
2016-09-01 16:30   ` James Hogan
2016-09-01 16:30 ` [PATCH 3/9] MIPS: traps: Ensure full EBase is written James Hogan
2016-09-01 16:30   ` James Hogan
2016-09-21 13:19   ` Ralf Baechle
2016-09-01 16:30 ` [PATCH 4/9] MIPS: c-r4k: Drop bc_wback_inv() from icache flush James Hogan
2016-09-01 16:30   ` James Hogan
2016-09-01 16:30 ` [PATCH 5/9] MIPS: c-r4k: Split user/kernel flush_icache_range() James Hogan
2016-09-01 16:30   ` James Hogan
2016-09-01 16:30 ` [PATCH 6/9] MIPS: cacheflush: Use __flush_icache_user_range() James Hogan
2016-09-01 16:30   ` James Hogan
2016-09-01 16:30 ` [PATCH 7/9] MIPS: uprobes: Flush icache via kernel address James Hogan
2016-09-01 16:30   ` James Hogan
2016-09-21 13:26   ` Ralf Baechle
2016-09-21 18:15     ` Leonid Yegoshin
2016-09-21 18:15       ` Leonid Yegoshin
2016-09-22 21:15       ` James Hogan
2016-09-22 21:15         ` James Hogan
2016-09-22 21:38         ` Leonid Yegoshin
2016-09-22 21:38           ` Leonid Yegoshin
2016-09-22 21:42           ` Leonid Yegoshin
2016-09-22 21:42             ` Leonid Yegoshin
2016-09-22 22:13           ` James Hogan
2016-09-22 22:27             ` Leonid Yegoshin [this message]
2016-09-22 22:27               ` Leonid Yegoshin
2016-09-23  7:10               ` James Hogan
2016-09-01 16:30 ` [PATCH 8/9] MIPS: KVM: Use __local_flush_icache_user_range() James Hogan
2016-09-01 16:30 ` [PATCH 9/9] MIPS: c-r4k: Fix flush_icache_range() for EVA James Hogan
2016-09-01 16:30   ` James Hogan

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=57E45ADD.6000202@imgtec.com \
    --to=leonid.yegoshin@imgtec.com \
    --cc=james.hogan@imgtec.com \
    --cc=linux-mips@linux-mips.org \
    --cc=ralf@linux-mips.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