linuxppc-dev.lists.ozlabs.org archive mirror
 help / color / mirror / Atom feed
From: Nicholas Piggin <npiggin@gmail.com>
To: Michael Ellerman <mpe@ellerman.id.au>
Cc: <linuxppc-dev@ozlabs.org>,
	haokexin@gmail.com, aneesh.kumar@linux.vnet.ibm.com
Subject: Re: [PATCH v3 18/21] powerpc: Add option to use jump label for cpu_has_feature()
Date: Thu, 28 Jul 2016 17:51:54 +1000	[thread overview]
Message-ID: <20160728175154.6967091c@roar.ozlabs.ibm.com> (raw)
In-Reply-To: <1469629097-30859-18-git-send-email-mpe@ellerman.id.au>

On Thu, 28 Jul 2016 00:18:14 +1000
Michael Ellerman <mpe@ellerman.id.au> wrote:

> From: Kevin Hao <haokexin@gmail.com>
> 
> We do binary patching of asm code using CPU features, which is a
> one-time operation, done during early boot. However checks of CPU
> features in C code are currently done at run time, even though the set
> of CPU features can never change after boot.
> 
> We can optimise this by using jump labels to implement
> cpu_has_feature(), meaning checks in C code are binary patched into a
> single nop or branch.
> 
> For a C sequence along the lines of:
> 
>     if (cpu_has_feature(FOO))
>          return 2;
> 
> The generated code before is roughly:
> 
>     ld      r9,-27640(r2)
>     ld      r9,0(r9)
>     lwz     r9,32(r9)
>     cmpwi   cr7,r9,0
>     bge     cr7, 1f
>     li      r3,2
>     blr
> 1:  ...
> 
> After (true):
>     nop
>     li      r3,2
>     blr
> 
> After (false):
>     b	1f
>     li      r3,2
>     blr
> 1:  ...
> 
> Signed-off-by: Kevin Hao <haokexin@gmail.com>
> Signed-off-by: Aneesh Kumar K.V <aneesh.kumar@linux.vnet.ibm.com>
> Signed-off-by: Michael Ellerman <mpe@ellerman.id.au>
> ---
>  arch/powerpc/include/asm/cpu_has_feature.h | 22
> ++++++++++++++++++++++ arch/powerpc/include/asm/cputable.h        |
> 6 ++++++ arch/powerpc/kernel/cputable.c             | 20
> ++++++++++++++++++++ arch/powerpc/lib/feature-fixups.c          |  1 +
>  4 files changed, 49 insertions(+)
> 
> v3: Rename MAX_CPU_FEATURES as we already have a #define with that
> name. Define NUM_CPU_FTR_KEYS as a constant.
>     Rename the array to cpu_feature_keys.
>     Use the kconfig we added to guard it.
>     Rewrite the change log.
> 
> diff --git a/arch/powerpc/include/asm/cpu_has_feature.h
> b/arch/powerpc/include/asm/cpu_has_feature.h index
> ad296b2f1d84..18e60e61bea9 100644 ---
> a/arch/powerpc/include/asm/cpu_has_feature.h +++
> b/arch/powerpc/include/asm/cpu_has_feature.h @@ -11,10 +11,32 @@
> static inline bool __cpu_has_feature(unsigned long feature)
> (CPU_FTRS_POSSIBLE & cur_cpu_spec->cpu_features & feature)); }
>  
> +#ifdef CONFIG_JUMP_LABEL_FEATURE_CHECKS
> +#include <linux/jump_label.h>
> +
> +#define NUM_CPU_FTR_KEYS	64
> +
> +extern struct static_key_true cpu_feature_keys[NUM_CPU_FTR_KEYS];
> +
> +static __always_inline bool cpu_has_feature(unsigned long feature)
> +{
> +	int i;
> +
> +	if (CPU_FTRS_ALWAYS & feature)
> +		return true;
> +
> +	if (!(CPU_FTRS_POSSIBLE & feature))
> +		return false;
> +
> +	i = __builtin_ctzl(feature);
> +	return static_branch_likely(&cpu_feature_keys[i]);

Just a reminder to add a BUILD_BUG_ON(!__builtin_constant_p(feature));
for this.

  reply	other threads:[~2016-07-28  7:52 UTC|newest]

Thread overview: 42+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2016-07-27 14:17 [PATCH v3 01/21] powerpc/mm: Add mmu_early_init_devtree() Michael Ellerman
2016-07-27 14:17 ` [PATCH v3 02/21] powerpc/mm: Move disable_radix handling into mmu_early_init_devtree() Michael Ellerman
2016-07-28  3:14   ` Balbir Singh
2016-07-27 14:17 ` [PATCH v3 03/21] powerpc/mm: Do hash device tree scanning earlier Michael Ellerman
2016-07-28 12:40   ` [PATCH v4] " Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 04/21] powerpc/mm: Do radix " Michael Ellerman
2016-07-28  3:48   ` Balbir Singh
2016-07-28  8:11     ` Michael Ellerman
2016-07-28 11:55       ` Balbir Singh
2016-07-27 14:18 ` [PATCH v3 05/21] powerpc/64: Do feature patching before MMU init Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 06/21] powerpc/kernel: Check features don't change after patching Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 07/21] powerpc/mm: Make MMU_FTR_RADIX a MMU family feature Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 08/21] powerpc/kernel: Convert mmu_has_feature() to returning bool Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 09/21] powerpc/kernel: Convert cpu_has_feature() " Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 10/21] powerpc/mm: Define radix_enabled() in one place & use static inline Michael Ellerman
2016-07-28  7:46   ` Nicholas Piggin
2016-07-29 11:42     ` Michael Ellerman
2016-07-29 12:54       ` Balbir Singh
2016-07-30 10:08         ` Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 11/21] powerpc/mm: Add __cpu/__mmu_has_feature() Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 12/21] powerpc/mm: Convert early cpu/mmu feature check to use the new helpers Michael Ellerman
2016-07-27 21:37   ` Benjamin Herrenschmidt
2016-07-28 11:24     ` Michael Ellerman
2016-07-27 21:42   ` Benjamin Herrenschmidt
2016-07-28  7:49   ` Nicholas Piggin
2016-07-28 13:04     ` Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 13/21] jump_label: Make it possible for arches to invoke jump_label_init() earlier Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 14/21] powerpc: Call jump_label_init() in apply_feature_fixups() Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 15/21] powerpc: Remove mfvtb() Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 16/21] powerpc: Move cpu_has_feature() to a separate file Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 17/21] powerpc: Add kconfig option to use jump labels for cpu/mmu_has_feature() Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 18/21] powerpc: Add option to use jump label for cpu_has_feature() Michael Ellerman
2016-07-28  7:51   ` Nicholas Piggin [this message]
2016-07-27 14:18 ` [PATCH v3 19/21] powerpc: Add option to use jump label for mmu_has_feature() Michael Ellerman
2016-07-28  7:52   ` Nicholas Piggin
2016-08-08  0:59   ` Anton Blanchard
2016-08-08  7:35     ` Anton Blanchard
2016-07-27 14:18 ` [PATCH v3 20/21] powerpc/mm: Catch usage of cpu/mmu_has_feature() before jump label init Michael Ellerman
2016-07-27 14:18 ` [PATCH v3 21/21] powerpc/jump_label: Annotate jump label assembly Michael Ellerman
2016-07-28  7:56   ` Nicholas Piggin
     [not found] ` <1469629097-30859-14-git-send-email-mpe__30163.7288918302$1469630223$gmane$org@ellerman.id.au>
2016-08-13 23:55   ` [PATCH v3 14/21] powerpc: Call jump_label_init() in apply_feature_fixups() Andreas Schwab
2016-08-14  4:44     ` Michael Ellerman

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=20160728175154.6967091c@roar.ozlabs.ibm.com \
    --to=npiggin@gmail.com \
    --cc=aneesh.kumar@linux.vnet.ibm.com \
    --cc=haokexin@gmail.com \
    --cc=linuxppc-dev@ozlabs.org \
    --cc=mpe@ellerman.id.au \
    /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).