From mboxrd@z Thu Jan 1 00:00:00 1970 From: hkallweit1@gmail.com (Heiner Kallweit) Date: Wed, 4 Oct 2017 20:50:03 +0200 Subject: [PATCH 3/5] firmware: arm_scpi: pre-populate dvfs info in scpi_probe In-Reply-To: <6ece57aa-81e0-c984-5eac-a0b54aa0d625@arm.com> References: <0e5959a7-6826-1909-1509-f6627a9688ed@arm.com> <106745da-1027-5760-de51-7a74c033964b@gmail.com> <6ece57aa-81e0-c984-5eac-a0b54aa0d625@arm.com> Message-ID: <486a06bc-6bb8-d8e8-29b3-bd959afd2326@gmail.com> To: linux-arm-kernel@lists.infradead.org List-Id: linux-arm-kernel.lists.infradead.org Am 04.10.2017 um 12:10 schrieb Sudeep Holla: > > > On 03/10/17 19:19, Heiner Kallweit wrote: >> Am 03.10.2017 um 18:18 schrieb Sudeep Holla: >>> >>> >>> On 03/10/17 17:00, Heiner Kallweit wrote: > > [...] > >>>>>> >>>>>> Remove does for (i = 0; i < MAX_DVFS_DOMAINS && info->dvfs[i]; i++) { >>>>>> and therefore stops at the first unpopulated domain and doesn't free >>>>>> the memory for further populated domains. I'll provide a patch for >>>>>> it. >>>>>> >>>>> >>>>> Does that mean you are re-introducing scpi_remove ? I kind of liked >>>>> removing it. >>>>> >>>> Sorry for the confusion. Then I'll go with the original approach and >>>> just make sure that errors whilst populating dvfs info are ignored. >>>> >>> >>> Indeed. >>> >>> If you are fine with the below fixup, then I can add it myself. Let me know. >>> >> Fine with me. > > Thanks for confirming, pushed now[1]. Can you please double check if > possible. > Compiles and works fine here. By the way, I have three more small patches which I think are improvements (YMMV). Will submit them later today.