From: rokhanna@nvidia.com (Rohit Khanna)
To: linux-arm-kernel@lists.infradead.org
Subject: [PATCH] arm64: alternative:flush cache with unpatched code
Date: Tue, 22 May 2018 18:08:03 +0000 [thread overview]
Message-ID: <1527012507683.32009@nvidia.com> (raw)
In-Reply-To: <f2db11aa-1a88-0447-6139-435536de2c61@arm.com>
Thanks Suzuki, I have modified the patch and will send it for review.
Thanks
Rohit
________________________________________
From: Suzuki K Poulose <Suzuki.Poulose@arm.com>
Sent: Tuesday, May 22, 2018 8:09 AM
To: Rohit Khanna; catalin.marinas at arm.com; robin.murphy at arm.com; mark.rutland at arm.com
Cc: Bo Yan; Alexander Van Brunt; linux-arm-kernel at lists.infradead.org
Subject: Re: [PATCH] arm64: alternative:flush cache with unpatched code
Rohit,
On 22/05/18 02:27, Rohit Khanna wrote:
> In the current implementation, __apply_alternatives patches
> flush_icache_range and then executes it without invalidating the icache.
> Thus, icache can contain some of the old instructions for
> flush_icache_range. This can cause unpredictable behavior as during
> execution we can get a mix of old and new instructions for
> flush_icache_range.
>
> This patch :
>
> 1. Adds a new function flush_cache_kernel_range for flushing kernel
> memory range. This function is not patched during boot and can be safely
> used to flush cache during code patching.
>
> 2. Modifies __apply_alternatives so that it uses
> flush_cache_kernel_range to flush the cache range after patching code.
>
> Signed-off-by: Rohit Khanna <rokhanna@nvidia.com>
> ---
> arch/arm64/include/asm/cacheflush.h | 1 +
> arch/arm64/kernel/alternative.c | 2 +-
> arch/arm64/mm/cache.S | 42 +++++++++++++++++++++++++++++++++++++
> 3 files changed, 44 insertions(+), 1 deletion(-)
>
> diff --git a/arch/arm64/include/asm/cacheflush.h b/arch/arm64/include/asm/cacheflush.h
> index 0094c6653b06..54692fabdf74 100644
> --- a/arch/arm64/include/asm/cacheflush.h
> +++ b/arch/arm64/include/asm/cacheflush.h
> @@ -73,6 +73,7 @@
> */
> extern void flush_icache_range(unsigned long start, unsigned long end);
> extern int invalidate_icache_range(unsigned long start, unsigned long end);
> +extern void flush_cache_kernel_range(unsigned long start, unsigned long end);
> extern void __flush_dcache_area(void *addr, size_t len);
> extern void __inval_dcache_area(void *addr, size_t len);
> extern void __clean_dcache_area_poc(void *addr, size_t len);
> diff --git a/arch/arm64/kernel/alternative.c b/arch/arm64/kernel/alternative.c
> index 5c4bce4ac381..a5408a3c297a 100644
> --- a/arch/arm64/kernel/alternative.c
> +++ b/arch/arm64/kernel/alternative.c
> @@ -155,7 +155,7 @@ static void __apply_alternatives(void *alt_region, bool use_linear_alias)
>
> alt_cb(alt, origptr, updptr, nr_inst);
>
> - flush_icache_range((uintptr_t)origptr,
> + flush_cache_kernel_range((uintptr_t)origptr,
> (uintptr_t)(origptr + nr_inst));
> }
> }
> diff --git a/arch/arm64/mm/cache.S b/arch/arm64/mm/cache.S
> index 30334d81b021..4dd09352a044 100644
> --- a/arch/arm64/mm/cache.S
> +++ b/arch/arm64/mm/cache.S
> @@ -81,6 +81,48 @@ ENDPROC(flush_icache_range)
> ENDPROC(__flush_cache_user_range)
>
> /*
> + * flush_cache_kernel_range(start,end)
> + *
> + * Ensure that the I and D caches are coherent within specified kernel
> + * region.
> + * This is typically used when code has been written to a kernel memory
> + * region and will be executed.
> + *
> + * NOTE - This macro cannot have "alternatives" applied to it as its
> + * used to update alternatives
> + *
> + * - start - virtual start address of region
> + * - end - virtual end address of region
> + */
> +ENTRY(flush_cache_kernel_range)
> + dcache_line_size x2, x3
...
> + icache_line_size x2, x3
You must use raw_{d,i}cache_line_size helpers above to avoid
using hot-patched code. The above helpers are patched if you
have mismatched cache line sizes.
Suzuki
next prev parent reply other threads:[~2018-05-22 18:08 UTC|newest]
Thread overview: 20+ messages / expand[flat|nested] mbox.gz Atom feed top
2018-05-22 1:27 [PATCH] arm64: alternative:flush cache with unpatched code Rohit Khanna
2018-05-22 15:09 ` Suzuki K Poulose
2018-05-22 18:08 ` Rohit Khanna [this message]
-- strict thread matches above, loose matches on Subject: below --
2018-05-22 18:07 Rohit Khanna
2018-05-23 9:06 ` Will Deacon
2018-05-29 18:11 Rohit Khanna
2018-05-30 9:00 ` Will Deacon
2018-05-31 17:45 ` Rohit Khanna
2018-06-04 9:16 ` Will Deacon
2018-06-04 19:34 ` Alexander Van Brunt
2018-06-05 16:55 ` Will Deacon
2018-06-05 17:07 ` Alexander Van Brunt
2018-06-06 15:44 ` Will Deacon
2018-06-06 16:16 ` Alexander Van Brunt
2018-05-31 20:37 Rohit Khanna
2018-06-01 9:03 ` Mark Rutland
2018-06-01 19:52 ` Rohit Khanna
2018-06-01 21:43 ` Rohit Khanna
2018-06-04 9:01 ` Mark Rutland
2018-06-02 0:39 Rohit Khanna
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=1527012507683.32009@nvidia.com \
--to=rokhanna@nvidia.com \
--cc=linux-arm-kernel@lists.infradead.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).