Re: [PATCH v3 2/2] PM / devfreq: mediatek: Introduce MediaTek CCI devfreq driver
From: Johnson Wang
Date: Wed Apr 27 2022 - 06:39:16 EST
On Wed, 2022-04-27 at 17:25 +0800, kernel test robot wrote:
> Hi Johnson,
>
> Thank you for the patch! Perhaps something to improve:
>
> [auto build test WARNING on robh/for-next]
> [also build test WARNING on linus/master v5.18-rc4 next-20220427]
> [If your patch is applied to the wrong git tree, kindly drop us a
> note.
> And when submitting patch, we suggest to use '--base' as documented
> in
>
https://urldefense.com/v3/__https://git-scm.com/docs/git-format-patch__;!!CTRNKA9wMg0ARbw!wdyoWXNLBcYM97vMuNFQXZ9BaajEp-Kmh5-xrvU2Rlmb0o-b9tRvCD0cPzbLMS_s9zEZ$
> ]
>
> url:
> https://urldefense.com/v3/__https://github.com/intel-lab-lkp/linux/commits/Johnson-Wang/Introduce-MediaTek-CCI-devfreq-driver/20220425-205820__;!!CTRNKA9wMg0ARbw!wdyoWXNLBcYM97vMuNFQXZ9BaajEp-Kmh5-xrvU2Rlmb0o-b9tRvCD0cPzbLMc1U_tqz$
>
> base:
> https://urldefense.com/v3/__https://git.kernel.org/pub/scm/linux/kernel/git/robh/linux.git__;!!CTRNKA9wMg0ARbw!wdyoWXNLBcYM97vMuNFQXZ9BaajEp-Kmh5-xrvU2Rlmb0o-b9tRvCD0cPzbLMUzTprof$
> for-next
> config: hexagon-allyesconfig (
> https://urldefense.com/v3/__https://download.01.org/0day-ci/archive/20220427/202204271737.oAuTwqZH-lkp@xxxxxxxxx/config__;!!CTRNKA9wMg0ARbw!wdyoWXNLBcYM97vMuNFQXZ9BaajEp-Kmh5-xrvU2Rlmb0o-b9tRvCD0cPzbLMaVRzbSL$
> )
> compiler: clang version 15.0.0 (
> https://urldefense.com/v3/__https://github.com/llvm/llvm-project__;!!CTRNKA9wMg0ARbw!wdyoWXNLBcYM97vMuNFQXZ9BaajEp-Kmh5-xrvU2Rlmb0o-b9tRvCD0cPzbLMRqw5IY-$
> $ 1cddcfdc3c683b393df1a5c9063252eb60e52818)
> reproduce (this is a W=1 build):
> wget
> https://urldefense.com/v3/__https://raw.githubusercontent.com/intel/lkp-tests/master/sbin/make.cross__;!!CTRNKA9wMg0ARbw!wdyoWXNLBcYM97vMuNFQXZ9BaajEp-Kmh5-xrvU2Rlmb0o-b9tRvCD0cPzbLMQLiD-i9$
> -O ~/bin/make.cross
> chmod +x ~/bin/make.cross
> #
> https://urldefense.com/v3/__https://github.com/intel-lab-lkp/linux/commit/98b34c0587837b0e5b880b11a52433f8f0eee19f__;!!CTRNKA9wMg0ARbw!wdyoWXNLBcYM97vMuNFQXZ9BaajEp-Kmh5-xrvU2Rlmb0o-b9tRvCD0cPzbLMU5yd7Y2$
>
> git remote add linux-review
> https://urldefense.com/v3/__https://github.com/intel-lab-lkp/linux__;!!CTRNKA9wMg0ARbw!wdyoWXNLBcYM97vMuNFQXZ9BaajEp-Kmh5-xrvU2Rlmb0o-b9tRvCD0cPzbLMW4ldtnH$
>
> git fetch --no-tags linux-review Johnson-Wang/Introduce-
> MediaTek-CCI-devfreq-driver/20220425-205820
> git checkout 98b34c0587837b0e5b880b11a52433f8f0eee19f
> # save the config file
> mkdir build_dir && cp config build_dir/.config
> COMPILER_INSTALL_PATH=$HOME/0day COMPILER=clang make.cross
> W=1 O=build_dir ARCH=hexagon SHELL=/bin/bash block/partitions/
> drivers/devfreq/ drivers/iio/imu/ drivers/misc/lkdtm/
>
> If you fix the issue, kindly add following tag as appropriate
> Reported-by: kernel test robot <lkp@xxxxxxxxx>
>
> All warnings (new ones prefixed by >>):
>
> drivers/devfreq/mtk-cci-devfreq.c:372:16: error: no member named
> 'parent_type' in 'struct devfreq_passive_data'
> passive_data->parent_type = CPUFREQ_PARENT_DEV;
> ~~~~~~~~~~~~ ^
> drivers/devfreq/mtk-cci-devfreq.c:372:30: error: use of undeclared
> identifier 'CPUFREQ_PARENT_DEV'
> passive_data->parent_type = CPUFREQ_PARENT_DEV;
> ^
> > > drivers/devfreq/mtk-cci-devfreq.c:379:4: warning: format
> > > specifies type 'int' but the argument has type 'long' [-Wformat]
>
> PTR_ERR(drv->devfreq));
> ^~~~~~~~~~~~~~~~~~~~~
> include/linux/dev_printk.h:144:65: note: expanded from macro
> 'dev_err'
> dev_printk_index_wrap(_dev_err, KERN_ERR, dev,
> dev_fmt(fmt), ##__VA_ARGS__)
> ~~~
> ^~~~~~~~~~~
> include/linux/dev_printk.h:110:23: note: expanded from macro
> 'dev_printk_index_wrap'
> _p_func(dev, fmt,
> ##__VA_ARGS__); \
> ~~~ ^~~~~~~~~~~
> 1 warning and 2 errors generated.
>
>
> vim +379 drivers/devfreq/mtk-cci-devfreq.c
>
> 255
> 256 static int mtk_ccifreq_probe(struct platform_device
> *pdev)
> 257 {
> 258 struct device *dev = &pdev->dev;
> 259 struct mtk_ccifreq_drv *drv;
> 260 struct devfreq_passive_data *passive_data;
> 261 struct dev_pm_opp *opp;
> 262 unsigned long rate, opp_volt;
> 263 int ret;
> 264
> 265 drv = devm_kzalloc(dev, sizeof(*drv),
> GFP_KERNEL);
> 266 if (!drv)
> 267 return -ENOMEM;
> 268
> 269 drv->dev = dev;
> 270 drv->soc_data = (const struct
> mtk_ccifreq_platform_data *)
> 271 of_device_get_match_dat
> a(&pdev->dev);
> 272 mutex_init(&drv->reg_lock);
> 273 platform_set_drvdata(pdev, drv);
> 274
> 275 drv->cci_clk = devm_clk_get(dev, "cci");
> 276 if (IS_ERR(drv->cci_clk)) {
> 277 ret = PTR_ERR(drv->cci_clk);
> 278 return dev_err_probe(dev, ret,
> 279 "failed to get cci
> clk: %d\n", ret);
> 280 }
> 281
> 282 drv->inter_clk = devm_clk_get(dev,
> "intermediate");
> 283 if (IS_ERR(drv->inter_clk)) {
> 284 ret = PTR_ERR(drv->inter_clk);
> 285 dev_err_probe(dev, ret,
> 286 "failed to get
> intermediate clk: %d\n", ret);
> 287 goto out_free_resources;
> 288 }
> 289
> 290 drv->proc_reg =
> devm_regulator_get_optional(dev, "proc");
> 291 if (IS_ERR(drv->proc_reg)) {
> 292 ret = PTR_ERR(drv->proc_reg);
> 293 dev_err_probe(dev, ret,
> 294 "failed to get proc
> regulator: %d\n", ret);
> 295 goto out_free_resources;
> 296 }
> 297
> 298 ret = regulator_enable(drv->proc_reg);
> 299 if (ret) {
> 300 dev_err(dev, "failed to enable proc
> regulator\n");
> 301 goto out_free_resources;
> 302 }
> 303
> 304 drv->sram_reg = regulator_get_optional(dev,
> "sram");
> 305 if (IS_ERR(drv->sram_reg))
> 306 drv->sram_reg = NULL;
> 307 else {
> 308 ret = regulator_enable(drv->sram_reg);
> 309 if (ret) {
> 310 dev_err(dev, "failed to enable
> sram regulator\n");
> 311 goto out_free_resources;
> 312 }
> 313 }
> 314
> 315 /*
> 316 * We assume min voltage is 0 and tracking
> target voltage using
> 317 * min_volt_shift for each iteration.
> 318 * The retry_max is 3 times of expeted
> iteration count.
> 319 */
> 320 drv->vtrack_max = 3 * DIV_ROUND_UP(max(drv-
> >soc_data->sram_max_volt,
> 321 drv-
> >soc_data->proc_max_volt),
> 322 drv-
> >soc_data->min_volt_shift);
> 323
> 324 ret = clk_prepare_enable(drv->cci_clk);
> 325 if (ret)
> 326 goto out_free_resources;
> 327
> 328 ret = clk_prepare_enable(drv->inter_clk);
> 329 if (ret)
> 330 goto out_disable_cci_clk;
> 331
> 332 ret = dev_pm_opp_of_add_table(dev);
> 333 if (ret) {
> 334 dev_err(dev, "failed to add opp table:
> %d\n", ret);
> 335 goto out_disable_inter_clk;
> 336 }
> 337
> 338 rate = clk_get_rate(drv->inter_clk);
> 339 opp = dev_pm_opp_find_freq_ceil(dev, &rate);
> 340 if (IS_ERR(opp)) {
> 341 ret = PTR_ERR(opp);
> 342 dev_err(dev, "failed to get
> intermediate opp: %d\n", ret);
> 343 goto out_remove_opp_table;
> 344 }
> 345 drv->inter_voltage =
> dev_pm_opp_get_voltage(opp);
> 346 dev_pm_opp_put(opp);
> 347
> 348 rate = U32_MAX;
> 349 opp = dev_pm_opp_find_freq_floor(drv->dev,
> &rate);
> 350 if (IS_ERR(opp)) {
> 351 dev_err(dev, "failed to get opp\n");
> 352 ret = PTR_ERR(opp);
> 353 goto out_remove_opp_table;
> 354 }
> 355
> 356 opp_volt = dev_pm_opp_get_voltage(opp);
> 357 dev_pm_opp_put(opp);
> 358 ret = mtk_ccifreq_set_voltage(drv, opp_volt);
> 359 if (ret) {
> 360 dev_err(dev, "failed to scale to
> highest voltage %lu in proc_reg\n",
> 361 opp_volt);
> 362 goto out_remove_opp_table;
> 363 }
> 364
> 365 passive_data = devm_kzalloc(dev, sizeof(struct
> devfreq_passive_data),
> 366 GFP_KERNEL);
> 367 if (!passive_data) {
> 368 ret = -ENOMEM;
> 369 goto out_remove_opp_table;
> 370 }
> 371
> 372 passive_data->parent_type = CPUFREQ_PARENT_DEV;
> 373 drv->devfreq = devm_devfreq_add_device(dev,
> &mtk_ccifreq_profile,
> 374 DEVFREQ_
> GOV_PASSIVE,
> 375 passive_
> data);
> 376 if (IS_ERR(drv->devfreq)) {
> 377 ret = -EPROBE_DEFER;
> 378 dev_err(dev, "failed to add devfreq
> device: %d\n",
> > 379 PTR_ERR(drv->devfreq));
> 380 goto out_remove_opp_table;
> 381 }
> 382
> 383 drv->opp_nb.notifier_call =
> mtk_ccifreq_opp_notifier;
> 384 ret = dev_pm_opp_register_notifier(dev, &drv-
> >opp_nb);
> 385 if (ret) {
> 386 dev_err(dev, "failed to register opp
> notifier: %d\n", ret);
> 387 goto out_remove_devfreq_device;
> 388 }
> 389 return 0;
> 390
> 391 out_remove_devfreq_device:
> 392 devm_devfreq_remove_device(dev, drv->devfreq);
> 393
> 394 out_remove_opp_table:
> 395 dev_pm_opp_of_remove_table(dev);
> 396
> 397 out_disable_inter_clk:
> 398 clk_disable_unprepare(drv->inter_clk);
> 399
> 400 out_disable_cci_clk:
> 401 clk_disable_unprepare(drv->cci_clk);
> 402
> 403 out_free_resources:
> 404 if (regulator_is_enabled(drv->proc_reg))
> 405 regulator_disable(drv->proc_reg);
> 406 if (drv->sram_reg && regulator_is_enabled(drv-
> >sram_reg))
> 407 regulator_disable(drv->sram_reg);
> 408
> 409 if (!IS_ERR(drv->proc_reg))
> 410 regulator_put(drv->proc_reg);
> 411 if (!IS_ERR(drv->sram_reg))
> 412 regulator_put(drv->sram_reg);
> 413 if (!IS_ERR(drv->cci_clk))
> 414 clk_put(drv->cci_clk);
> 415 if (!IS_ERR(drv->inter_clk))
> 416 clk_put(drv->inter_clk);
> 417
> 418 return ret;
> 419 }
> 420
>
Hi "kernel test robot",
Thanks for your review.
This patch is based on chanwoo/devfreq-testing[1]
[1]
https://git.kernel.org/pub/scm/linux/kernel/git/chanwoo/linux.git/log/?h=devfreq-testing
I will follow your suggestion to use '--base' in the next version.
BRs,
Johnson Wang