All of lore.kernel.org
 help / color / mirror / Atom feed
From: Andrzej Hajda <a.hajda@samsung.com>
To: Mike Turquette <mturquette@linaro.org>, linux-mm@kvack.org
Cc: Andrzej Hajda <a.hajda@samsung.com>,
	Marek Szyprowski <m.szyprowski@samsung.com>,
	Kyungmin Park <kyungmin.park@samsung.com>,
	linux-kernel@vger.kernel.org, andi@firstfloor.org, andi@lisas.de,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Andrew Morton <akpm@linux-foundation.org>,
	sboyd@codeaurora.org
Subject: Re: [PATCH 3/5] clk: convert clock name allocations to kstrdup_const
Date: Tue, 13 Jan 2015 08:57:39 +0100	[thread overview]
Message-ID: <54B4CFF3.5060100@samsung.com> (raw)
In-Reply-To: <20150112231104.20842.5239@quantum>

On 01/13/2015 12:11 AM, Mike Turquette wrote:
> Quoting Andrzej Hajda (2015-01-12 01:18:41)
>> Clock subsystem frequently performs duplication of strings located
>> in read-only memory section. Replacing kstrdup by kstrdup_const
>> allows to avoid such operations.
>>
>> Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
> Looks OK to me. Is there an easy trick to measuring the number of string
> duplications saved short of instrumenting your code with a counter?

I have just added pr_err in kstrdup_const:

diff --git a/mm/util.c b/mm/util.c
index c96fc4b..32a97b2 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -56,8 +56,10 @@ EXPORT_SYMBOL(kstrdup);
 
 const char *kstrdup_const(const char *s, gfp_t gfp)
 {
-       if (is_kernel_rodata((unsigned long)s))
+       if (is_kernel_rodata((unsigned long)s)) {
+               pr_err("%s: %pS:%s\n", __func__,
__builtin_return_address(0), s);
                return s;
+       }
 
        return kstrdup(s, gfp);
 }

Probably printk buffer size should be increased:
CONFIG_LOG_BUF_SHIFT=17

Regards
Andrzej

>
> Regards,
> Mike
>
>> ---
>>  drivers/clk/clk.c | 12 ++++++------
>>  1 file changed, 6 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c
>> index f4963b7..27e644a 100644
>> --- a/drivers/clk/clk.c
>> +++ b/drivers/clk/clk.c
>> @@ -2048,7 +2048,7 @@ struct clk *clk_register(struct device *dev, struct clk_hw *hw)
>>                 goto fail_out;
>>         }
>>  
>> -       clk->name = kstrdup(hw->init->name, GFP_KERNEL);
>> +       clk->name = kstrdup_const(hw->init->name, GFP_KERNEL);
>>         if (!clk->name) {
>>                 pr_err("%s: could not allocate clk->name\n", __func__);
>>                 ret = -ENOMEM;
>> @@ -2075,7 +2075,7 @@ struct clk *clk_register(struct device *dev, struct clk_hw *hw)
>>  
>>         /* copy each string name in case parent_names is __initdata */
>>         for (i = 0; i < clk->num_parents; i++) {
>> -               clk->parent_names[i] = kstrdup(hw->init->parent_names[i],
>> +               clk->parent_names[i] = kstrdup_const(hw->init->parent_names[i],
>>                                                 GFP_KERNEL);
>>                 if (!clk->parent_names[i]) {
>>                         pr_err("%s: could not copy parent_names\n", __func__);
>> @@ -2090,10 +2090,10 @@ struct clk *clk_register(struct device *dev, struct clk_hw *hw)
>>  
>>  fail_parent_names_copy:
>>         while (--i >= 0)
>> -               kfree(clk->parent_names[i]);
>> +               kfree_const(clk->parent_names[i]);
>>         kfree(clk->parent_names);
>>  fail_parent_names:
>> -       kfree(clk->name);
>> +       kfree_const(clk->name);
>>  fail_name:
>>         kfree(clk);
>>  fail_out:
>> @@ -2112,10 +2112,10 @@ static void __clk_release(struct kref *ref)
>>  
>>         kfree(clk->parents);
>>         while (--i >= 0)
>> -               kfree(clk->parent_names[i]);
>> +               kfree_const(clk->parent_names[i]);
>>  
>>         kfree(clk->parent_names);
>> -       kfree(clk->name);
>> +       kfree_const(clk->name);
>>         kfree(clk);
>>  }
>>  
>> -- 
>> 1.9.1
>>

--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org.  For more info on Linux MM,
see: http://www.linux-mm.org/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>

WARNING: multiple messages have this Message-ID (diff)
From: Andrzej Hajda <a.hajda@samsung.com>
To: Mike Turquette <mturquette@linaro.org>, linux-mm@kvack.org
Cc: Andrzej Hajda <a.hajda@samsung.com>,
	Marek Szyprowski <m.szyprowski@samsung.com>,
	Kyungmin Park <kyungmin.park@samsung.com>,
	linux-kernel@vger.kernel.org, andi@firstfloor.org, andi@lisas.de,
	Alexander Viro <viro@zeniv.linux.org.uk>,
	Andrew Morton <akpm@linux-foundation.org>,
	sboyd@codeaurora.org
Subject: Re: [PATCH 3/5] clk: convert clock name allocations to kstrdup_const
Date: Tue, 13 Jan 2015 08:57:39 +0100	[thread overview]
Message-ID: <54B4CFF3.5060100@samsung.com> (raw)
In-Reply-To: <20150112231104.20842.5239@quantum>

On 01/13/2015 12:11 AM, Mike Turquette wrote:
> Quoting Andrzej Hajda (2015-01-12 01:18:41)
>> Clock subsystem frequently performs duplication of strings located
>> in read-only memory section. Replacing kstrdup by kstrdup_const
>> allows to avoid such operations.
>>
>> Signed-off-by: Andrzej Hajda <a.hajda@samsung.com>
> Looks OK to me. Is there an easy trick to measuring the number of string
> duplications saved short of instrumenting your code with a counter?

I have just added pr_err in kstrdup_const:

diff --git a/mm/util.c b/mm/util.c
index c96fc4b..32a97b2 100644
--- a/mm/util.c
+++ b/mm/util.c
@@ -56,8 +56,10 @@ EXPORT_SYMBOL(kstrdup);
 
 const char *kstrdup_const(const char *s, gfp_t gfp)
 {
-       if (is_kernel_rodata((unsigned long)s))
+       if (is_kernel_rodata((unsigned long)s)) {
+               pr_err("%s: %pS:%s\n", __func__,
__builtin_return_address(0), s);
                return s;
+       }
 
        return kstrdup(s, gfp);
 }

Probably printk buffer size should be increased:
CONFIG_LOG_BUF_SHIFT=17

Regards
Andrzej

>
> Regards,
> Mike
>
>> ---
>>  drivers/clk/clk.c | 12 ++++++------
>>  1 file changed, 6 insertions(+), 6 deletions(-)
>>
>> diff --git a/drivers/clk/clk.c b/drivers/clk/clk.c
>> index f4963b7..27e644a 100644
>> --- a/drivers/clk/clk.c
>> +++ b/drivers/clk/clk.c
>> @@ -2048,7 +2048,7 @@ struct clk *clk_register(struct device *dev, struct clk_hw *hw)
>>                 goto fail_out;
>>         }
>>  
>> -       clk->name = kstrdup(hw->init->name, GFP_KERNEL);
>> +       clk->name = kstrdup_const(hw->init->name, GFP_KERNEL);
>>         if (!clk->name) {
>>                 pr_err("%s: could not allocate clk->name\n", __func__);
>>                 ret = -ENOMEM;
>> @@ -2075,7 +2075,7 @@ struct clk *clk_register(struct device *dev, struct clk_hw *hw)
>>  
>>         /* copy each string name in case parent_names is __initdata */
>>         for (i = 0; i < clk->num_parents; i++) {
>> -               clk->parent_names[i] = kstrdup(hw->init->parent_names[i],
>> +               clk->parent_names[i] = kstrdup_const(hw->init->parent_names[i],
>>                                                 GFP_KERNEL);
>>                 if (!clk->parent_names[i]) {
>>                         pr_err("%s: could not copy parent_names\n", __func__);
>> @@ -2090,10 +2090,10 @@ struct clk *clk_register(struct device *dev, struct clk_hw *hw)
>>  
>>  fail_parent_names_copy:
>>         while (--i >= 0)
>> -               kfree(clk->parent_names[i]);
>> +               kfree_const(clk->parent_names[i]);
>>         kfree(clk->parent_names);
>>  fail_parent_names:
>> -       kfree(clk->name);
>> +       kfree_const(clk->name);
>>  fail_name:
>>         kfree(clk);
>>  fail_out:
>> @@ -2112,10 +2112,10 @@ static void __clk_release(struct kref *ref)
>>  
>>         kfree(clk->parents);
>>         while (--i >= 0)
>> -               kfree(clk->parent_names[i]);
>> +               kfree_const(clk->parent_names[i]);
>>  
>>         kfree(clk->parent_names);
>> -       kfree(clk->name);
>> +       kfree_const(clk->name);
>>         kfree(clk);
>>  }
>>  
>> -- 
>> 1.9.1
>>


  reply	other threads:[~2015-01-13  7:58 UTC|newest]

Thread overview: 35+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2015-01-12  9:18 [PATCH 0/5] kstrdup optimization Andrzej Hajda
2015-01-12  9:18 ` Andrzej Hajda
2015-01-12  9:18 ` [PATCH 1/5] mm/util: add kstrdup_const Andrzej Hajda
2015-01-12  9:18   ` Andrzej Hajda
2015-01-12 17:13   ` Joe Perches
2015-01-12 17:13     ` Joe Perches
2015-01-12  9:18 ` [PATCH 2/5] kernfs: convert node name allocation to kstrdup_const Andrzej Hajda
2015-01-12  9:18   ` Andrzej Hajda
2015-01-14 14:13   ` Tejun Heo
2015-01-14 14:13     ` Tejun Heo
2015-01-14 14:37   ` [PATCH 2.5/5] kernfs: remove KERNFS_STATIC_NAME Tejun Heo
2015-01-14 14:37     ` Tejun Heo
2015-01-12  9:18 ` [PATCH 3/5] clk: convert clock name allocations to kstrdup_const Andrzej Hajda
2015-01-12  9:18   ` Andrzej Hajda
2015-01-12 23:11   ` Mike Turquette
2015-01-13  7:57     ` Andrzej Hajda [this message]
2015-01-13  7:57       ` Andrzej Hajda
2015-01-12  9:18 ` [PATCH 4/5] mm/slab: convert cache " Andrzej Hajda
2015-01-12  9:18   ` Andrzej Hajda
2015-01-12  9:18 ` [PATCH 5/5] fs/namespace: convert devname allocation " Andrzej Hajda
2015-01-12  9:18   ` Andrzej Hajda
2015-01-12 20:45 ` [PATCH 0/5] kstrdup optimization Geert Uytterhoeven
2015-01-12 20:45   ` Geert Uytterhoeven
2015-01-13 23:48   ` Andrew Morton
2015-01-13 23:48     ` Andrew Morton
2015-01-14  0:10   ` Craig Milo Rogers
2015-01-14  0:10     ` Craig Milo Rogers
2015-01-14  0:17     ` Andrew Morton
2015-01-14  0:17       ` Andrew Morton
2015-01-13 23:37 ` Andrew Morton
2015-01-13 23:37   ` Andrew Morton
2015-01-14  8:06   ` Andrzej Hajda
2015-01-14  8:06     ` Andrzej Hajda
2015-01-14 14:12   ` Tejun Heo
2015-01-14 14:12     ` Tejun Heo

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=54B4CFF3.5060100@samsung.com \
    --to=a.hajda@samsung.com \
    --cc=akpm@linux-foundation.org \
    --cc=andi@firstfloor.org \
    --cc=andi@lisas.de \
    --cc=kyungmin.park@samsung.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --cc=m.szyprowski@samsung.com \
    --cc=mturquette@linaro.org \
    --cc=sboyd@codeaurora.org \
    --cc=viro@zeniv.linux.org.uk \
    /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 an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.