From: Eduardo Valentin <edubezval@gmail.com>
To: Viresh Kumar <viresh.kumar@linaro.org>
Cc: linux-pm@vger.kernel.org, linaro-kernel@lists.linaro.org,
rui.zhang@intel.com
Subject: Re: [PATCH 24/26] cpu_cooling: Store frequencies in descending order
Date: Tue, 2 Dec 2014 19:21:30 -0400 [thread overview]
Message-ID: <20141202232128.GA3645@developer> (raw)
In-Reply-To: <ecb90bfbe1dd69ad73210630ce0a2f988d6210f1.1417167599.git.viresh.kumar@linaro.org>
[-- Attachment #1: Type: text/plain, Size: 4720 bytes --]
Viresh,
On Fri, Nov 28, 2014 at 03:14:18PM +0530, Viresh Kumar wrote:
> CPUFreq framework *doesn't* guarantee that frequencies present in cpufreq table
> will be in ascending or descending order. But cpu_cooling somehow assumes that.
>
> Probably because most of current users are creating this list from DT, which is
> done with the help of OPP layer. And OPP layer creates the list in ascending
> order of frequencies.
>
> But cpu_cooling can be used for other platforms too, which don't have
> frequencies arranged in any order.
>
> This patch tries to fix this issue by creating another list of valid frequencies
> in descending order. Care is also taken to throw warnings for duplicate entries.
>
> Later patches would use it to simplify code.
>
> Signed-off-by: Viresh Kumar <viresh.kumar@linaro.org>
> ---
> drivers/thermal/cpu_cooling.c | 41 ++++++++++++++++++++++++++++++++++++++++-
> 1 file changed, 40 insertions(+), 1 deletion(-)
>
> diff --git a/drivers/thermal/cpu_cooling.c b/drivers/thermal/cpu_cooling.c
> index 07414c5..9a4a323 100644
> --- a/drivers/thermal/cpu_cooling.c
> +++ b/drivers/thermal/cpu_cooling.c
> @@ -65,6 +65,7 @@ struct cpufreq_cooling_device {
> unsigned int cpufreq_state;
> unsigned int cpufreq_val;
> unsigned int max_level;
> + unsigned int *freq_table; /* In descending order */
> struct cpumask allowed_cpus;
> struct list_head head;
> };
> @@ -321,6 +322,20 @@ static struct notifier_block thermal_cpufreq_notifier_block = {
> .notifier_call = cpufreq_thermal_notifier,
> };
>
> +static unsigned int find_next_max(struct cpufreq_frequency_table *table,
> + unsigned int prev_max)
> +{
> + struct cpufreq_frequency_table *pos;
> + unsigned int max = 0;
> +
> + cpufreq_for_each_valid_entry(pos, table) {
> + if (pos->frequency > max && pos->frequency < prev_max)
What happens if, for some random reason, the cpufreq table is in
ascending order and this function is called with prev_max == (unsigned
int) -1 ? What would be the returned max?
> + max = pos->frequency;
> + }
> +
> + return max;
> +}
> +
> /**
> * __cpufreq_cooling_register - helper function to create cpufreq cooling device
> * @np: a valid struct device_node to the cooling device device tree node
> @@ -343,6 +358,7 @@ __cpufreq_cooling_register(struct device_node *np,
> struct cpufreq_cooling_device *cpufreq_dev;
> char dev_name[THERMAL_NAME_LENGTH];
> struct cpufreq_frequency_table *pos, *table;
> + unsigned int freq, i;
>
> table = cpufreq_frequency_get_table(cpumask_first(clip_cpus));
> if (!table) {
> @@ -358,6 +374,14 @@ __cpufreq_cooling_register(struct device_node *np,
> cpufreq_for_each_valid_entry(pos, table)
> cpufreq_dev->max_level++;
>
> + cpufreq_dev->freq_table = kmalloc(sizeof(*cpufreq_dev->freq_table) *
> + cpufreq_dev->max_level, GFP_KERNEL);
> + if (!cpufreq_dev->freq_table) {
> + return ERR_PTR(-ENOMEM);
> + cool_dev = ERR_PTR(-ENOMEM);
> + goto free_cdev;
> + }
> +
> /* max_level is an index, not a counter */
> cpufreq_dev->max_level--;
>
> @@ -366,7 +390,7 @@ __cpufreq_cooling_register(struct device_node *np,
> cpufreq_dev->id = idr_alloc(&cpufreq_idr, NULL, 0, 0, GFP_KERNEL);
> if (unlikely(cpufreq_dev->id < 0)) {
> cool_dev = ERR_PTR(cpufreq_dev->id);
> - goto free_cdev;
> + goto free_table;
> }
>
> snprintf(dev_name, sizeof(dev_name), "thermal-cpufreq-%d",
> @@ -377,6 +401,18 @@ __cpufreq_cooling_register(struct device_node *np,
> if (IS_ERR(cool_dev))
> goto remove_idr;
>
> + /* Fill freq-table in descending order of frequencies */
> + for (i = 0, freq = -1; i <= cpufreq_dev->max_level; i++) {
> + freq = find_next_max(table, freq);
> + cpufreq_dev->freq_table[i] = freq;
> +
> + /* Warn for duplicate entries */
> + if (!freq)
> + pr_warn("%s: table has duplicate entries\n", __func__);
> + else
> + pr_debug("%s: freq:%u KHz\n", __func__, freq);
> + }
> +
> cpufreq_dev->cool_dev = cool_dev;
> INIT_LIST_HEAD(&cpufreq_dev->head);
>
> @@ -394,6 +430,8 @@ __cpufreq_cooling_register(struct device_node *np,
>
> remove_idr:
> idr_remove(&cpufreq_idr, cpufreq_dev->id);
> +free_table:
> + kfree(cpufreq_dev->freq_table);
> free_cdev:
> kfree(cpufreq_dev);
>
> @@ -467,6 +505,7 @@ void cpufreq_cooling_unregister(struct thermal_cooling_device *cdev)
>
> thermal_cooling_device_unregister(cpufreq_dev->cool_dev);
> idr_remove(&cpufreq_idr, cpufreq_dev->id);
> + kfree(cpufreq_dev->freq_table);
> kfree(cpufreq_dev);
> }
> EXPORT_SYMBOL_GPL(cpufreq_cooling_unregister);
> --
> 2.0.3.693.g996b0fd
>
[-- Attachment #2: Digital signature --]
[-- Type: application/pgp-signature, Size: 473 bytes --]
next prev parent reply other threads:[~2014-12-02 23:21 UTC|newest]
Thread overview: 55+ messages / expand[flat|nested] mbox.gz Atom feed top
2014-11-28 9:43 [PATCH 00/26] thermal: cpu_cooling: Fixes and cleanups Viresh Kumar
2014-11-28 9:43 ` [PATCH 01/26] thermal: db8500: pass cpu_present_mask to cpufreq_cooling_register() Viresh Kumar
2014-11-28 9:43 ` [PATCH 02/26] thermal: imx: " Viresh Kumar
2014-11-28 9:43 ` [PATCH 03/26] thermal: exynos: " Viresh Kumar
2014-11-28 9:43 ` [PATCH 04/26] thermal: exynos: Handle -EPROBE_DEFER properly Viresh Kumar
2014-12-02 23:08 ` Eduardo Valentin
2014-11-28 9:43 ` [PATCH 05/26] cpu_cooling: random comment fixups Viresh Kumar
2014-12-02 23:09 ` Eduardo Valentin
2014-11-28 9:44 ` [PATCH 06/26] cpu_cooling: fix doc comment over struct cpufreq_cooling_device Viresh Kumar
2014-11-28 9:44 ` [PATCH 07/26] cpu_cooling: Add comment to clarify relation between cooling state and frequency Viresh Kumar
2014-11-28 9:44 ` [PATCH 08/26] cpu_cooling: Pass variable instead of its type to sizeof() Viresh Kumar
2014-12-02 15:26 ` Javi Merino
2014-12-02 23:07 ` Eduardo Valentin
2014-12-03 4:38 ` Viresh Kumar
2014-11-28 9:44 ` [PATCH 09/26] cpu_cooling: no need to set cpufreq_state to zero Viresh Kumar
2014-11-28 9:44 ` [PATCH 10/26] cpu_cooling: no need to set cpufreq_dev to NULL Viresh Kumar
2014-11-28 9:44 ` [PATCH 11/26] cpu_cooling: propagate error returned by idr_alloc() Viresh Kumar
2014-12-02 15:35 ` Javi Merino
2014-12-03 4:36 ` Viresh Kumar
2014-12-02 23:03 ` Eduardo Valentin
2014-11-28 9:44 ` [PATCH 12/26] cpu_cooling: Don't match min/max frequencies for all CPUs on cooling register Viresh Kumar
2014-11-28 9:44 ` [PATCH 13/26] cpu_cooling: don't iterate over all allowed_cpus to update cpufreq policy Viresh Kumar
2014-11-28 9:44 ` [PATCH 14/26] cpu_cooling: Don't check is_cpufreq_valid() Viresh Kumar
2014-11-28 9:44 ` [PATCH 15/26] cpu_cooling: do error handling at the bottom in __cpufreq_cooling_register() Viresh Kumar
2014-12-02 15:45 ` Javi Merino
2014-11-28 9:44 ` [PATCH 16/26] cpu_cooling: Drop useless locking around idr_alloc/idr_remove Viresh Kumar
2014-12-02 15:53 ` Javi Merino
2014-12-02 23:05 ` Eduardo Valentin
2014-12-03 9:32 ` Viresh Kumar
2014-11-28 9:44 ` [PATCH 17/26] cpu_cooling: Merge cpufreq_apply_cooling() into cpufreq_set_cur_state() Viresh Kumar
2014-11-28 9:44 ` [PATCH 18/26] cpu_cooling: Merge get_cpu_frequency() " Viresh Kumar
2014-11-28 9:44 ` [PATCH 19/26] cpu_cooling: find max level during device registration Viresh Kumar
2014-12-02 23:39 ` Eduardo Valentin
2014-12-03 4:57 ` Viresh Kumar
2014-12-03 13:40 ` Eduardo Valentin
2014-11-28 9:44 ` [PATCH 20/26] cpu_cooling: get_property() doesn't need to support GET_MAXL anymore Viresh Kumar
2014-11-28 9:44 ` [PATCH 21/26] cpu_cooling: create list of cpufreq_cooling_devices Viresh Kumar
2014-12-02 23:12 ` Eduardo Valentin
2014-11-28 9:44 ` [PATCH 22/26] cpu_cooling: use cpufreq_dev_list instead of cpufreq_dev_count Viresh Kumar
2014-11-28 9:44 ` [PATCH 23/26] cpu_cooling: Pass 'cpufreq_dev' to get_property() Viresh Kumar
2014-11-28 9:44 ` [PATCH 24/26] cpu_cooling: Store frequencies in descending order Viresh Kumar
2014-12-02 23:21 ` Eduardo Valentin [this message]
2014-12-03 4:52 ` Viresh Kumar
2014-12-03 13:41 ` Eduardo Valentin
[not found] ` <CAKohponw7E9yyvjhP97CzjtcFcq3N+5ysde7DR5q+Nm0s=bKAw@mail.gmail.com>
2014-12-03 14:00 ` Eduardo Valentin
2014-11-28 9:44 ` [PATCH 25/26] cpu_cooling: Use cpufreq_dev->freq_table for finding level/freq Viresh Kumar
2014-12-02 23:36 ` Eduardo Valentin
2014-12-03 5:10 ` Viresh Kumar
2014-12-03 13:32 ` Eduardo Valentin
2014-11-28 9:44 ` [PATCH 26/26] cpu_cooling: update copyright tags Viresh Kumar
2014-12-02 19:41 ` Eduardo Valentin
2014-12-03 4:34 ` Viresh Kumar
2014-12-04 4:41 ` amit daniel kachhap
2014-11-28 13:26 ` [PATCH 00/26] thermal: cpu_cooling: Fixes and cleanups Eduardo Valentin
2014-11-28 13:41 ` Viresh Kumar
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=20141202232128.GA3645@developer \
--to=edubezval@gmail.com \
--cc=linaro-kernel@lists.linaro.org \
--cc=linux-pm@vger.kernel.org \
--cc=rui.zhang@intel.com \
--cc=viresh.kumar@linaro.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).