From mboxrd@z Thu Jan 1 00:00:00 1970 From: Dmitry Osipenko Subject: Re: tegra124-jetson-tk1: sata doesnt work since 5.2 Date: Wed, 3 Jun 2020 01:49:57 +0300 Message-ID: <56700c1d-ffb0-e44e-ea8e-48a848ccf3bf@gmail.com> References: <20200319074401.GA4116@Red> <20200531193111.GA15331@Red> <20200602081958.GA21773@Red> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Return-path: In-Reply-To: <20200602081958.GA21773@Red> Content-Language: en-US Sender: linux-tegra-owner-u79uwXL29TY76Z2rM5mHXA@public.gmane.org To: LABBE Corentin Cc: thierry.reding-Re5JQEeQqe8AvxtiuMwx3w@public.gmane.org, jonathanh-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org, pgaikwad-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org, pdeschrijver-DDmLM1+adcrQT0dZR+AlfA@public.gmane.org, mturquette-rdvid1DuHRBWk0Htik3J/w@public.gmane.org, sboyd-DgEjT+Ai2ygdnm+yROfE0A@public.gmane.org, axboe-tSWWG44O7X1aa/9Udqfwiw@public.gmane.org, linux-tegra-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-kernel-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-clk-u79uwXL29TY76Z2rM5mHXA@public.gmane.org, linux-ide-u79uwXL29TY76Z2rM5mHXA@public.gmane.org List-Id: linux-tegra@vger.kernel.org 02.06.2020 11:19, LABBE Corentin пишет: ... >> Sounds like PLLE needs to be enabled for SATA, but ahci-tegra driver >> doesn't do that. >> >> Could you please try this change: >> >> --- >8 --- >> diff --git a/drivers/clk/tegra/clk-tegra124.c >> b/drivers/clk/tegra/clk-tegra124.c >> index e931319dcc9d..7dbc14652830 100644 >> --- a/drivers/clk/tegra/clk-tegra124.c >> +++ b/drivers/clk/tegra/clk-tegra124.c >> @@ -1330,6 +1330,7 @@ static struct tegra_clk_init_table >> common_init_table[] __initdata = { >> { TEGRA124_CLK_I2S3_SYNC, TEGRA124_CLK_CLK_MAX, 24576000, 0 }, >> { TEGRA124_CLK_I2S4_SYNC, TEGRA124_CLK_CLK_MAX, 24576000, 0 }, >> { TEGRA124_CLK_VIMCLK_SYNC, TEGRA124_CLK_CLK_MAX, 24576000, 0 }, >> + { TEGRA124_CLK_PLL_E, TEGRA124_CLK_CLK_MAX, 0, 1 }, >> /* must be the last entry */ >> { TEGRA124_CLK_CLK_MAX, TEGRA124_CLK_CLK_MAX, 0, 0 }, >> }; >> --- >8 --- > > This patch alone does not fix the issue. I'd try to enable all the clocks and voltage regulators that are enabled by the USB driver. Also won't hurt to check the clock rate configurations with the enabled / disabled USB, try to replicate the working configuration.