From mboxrd@z Thu Jan 1 00:00:00 1970 From: Thierry Reding Subject: Re: [PATCH 1/4] clk: tegra: Enable fuse clock on Tegra124 Date: Fri, 4 Oct 2019 14:18:12 +0200 Message-ID: <20191004121812.GB227112@ulmo> References: <20191001211346.104400-1-swarren@wwwdotorg.org> <20191002110454.GJ3716706@ulmo> <6a48d716-2312-4623-f47a-a53ac2ece83c@wwwdotorg.org> Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============3056316276139130121==" Return-path: In-Reply-To: <6a48d716-2312-4623-f47a-a53ac2ece83c@wwwdotorg.org> List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Sender: "linux-arm-kernel" Errors-To: linux-arm-kernel-bounces+linux-arm-kernel=m.gmane.org@lists.infradead.org To: Stephen Warren Cc: Prashant Gaikwad , Stephen Boyd , Peter De Schrijver , linux-clk@vger.kernel.org, Jonathan Hunter , linux-tegra@vger.kernel.org, Michael Turquette , linux-arm-kernel@lists.infradead.org List-Id: linux-tegra@vger.kernel.org --===============3056316276139130121== Content-Type: multipart/signed; micalg=pgp-sha256; protocol="application/pgp-signature"; boundary="eAbsdosE1cNLO4uF" Content-Disposition: inline --eAbsdosE1cNLO4uF Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Content-Transfer-Encoding: quoted-printable On Wed, Oct 02, 2019 at 02:59:03PM -0600, Stephen Warren wrote: > On 10/2/19 5:04 AM, Thierry Reding wrote: > > On Tue, Oct 01, 2019 at 03:13:43PM -0600, Stephen Warren wrote: > > > From: Stephen Warren > > >=20 > > > For a little over a year, U-Boot has configured the flow controller to > > > perform automatic RAM re-repair on off->on power transitions of the C= PU > > > rail1]. This is mandatory for correct operation of Tegra124. However,= RAM > > > re-repair relies on certain clocks, which the kernel must enable and > > > leave running. The fuse clock is one of those clocks. Enable this clo= ck > > > so that LP1 power mode (system suspend) operates correctly. > > >=20 > > > [1] 3cc7942a4ae5 ARM: tegra: implement RAM repair > > >=20 > > > Reported-by: Jonathan Hunter > > > Cc: stable@vger.kernel.org > > > Signed-off-by: Stephen Warren > > > --- > > > drivers/clk/tegra/clk-tegra124.c | 1 + > > > 1 file changed, 1 insertion(+) > > >=20 > > > diff --git a/drivers/clk/tegra/clk-tegra124.c b/drivers/clk/tegra/clk= -tegra124.c > > > index 0224fdc4766f..f53f6315c646 100644 > > > --- a/drivers/clk/tegra/clk-tegra124.c > > > +++ b/drivers/clk/tegra/clk-tegra124.c > > > @@ -1291,6 +1291,7 @@ static struct tegra_clk_init_table common_init_= table[] __initdata =3D { > > > }; > > > static struct tegra_clk_init_table tegra124_init_table[] __initdata= =3D { > > > + { TEGRA124_CLK_FUSE, -1, 0, 1 }, > >=20 > > I think the correct way to do this these days is to mark the clock as > > CRITICAL. Not sure if there's an easy way to do that given that the > > clock init table doesn't allow storing flags. > >=20 > > Do you have any good ideas on how to achieve this with the critical flag > > instead of forcing the refcount to 1? > >=20 > > Perhaps something like the below would work? > > ... >=20 > The following works for me; does this seem like a reasonable approach? It > does set the critical flag for all SoCs, including any that don't require > RAM re-repair. I'm not sure which do; I know it's more than just Tegra124, > but I'm not sure how far back/forward the requirement goes. >=20 > > diff --git a/drivers/clk/tegra/clk-tegra-periph.c b/drivers/clk/tegra/c= lk-tegra-periph.c > > index 1ed85f120a1b..76dd91eebd13 100644 > > --- a/drivers/clk/tegra/clk-tegra-periph.c > > +++ b/drivers/clk/tegra/clk-tegra-periph.c > > @@ -785,7 +785,7 @@ static struct tegra_periph_init_data gate_clks[] = =3D { > > GATE("ahbdma", "hclk", 33, 0, tegra_clk_ahbdma, 0), > > GATE("apbdma", "pclk", 34, 0, tegra_clk_apbdma, 0), > > GATE("kbc", "clk_32k", 36, TEGRA_PERIPH_ON_APB | TEGRA_PERIPH_N= O_RESET, tegra_clk_kbc, 0), > > - GATE("fuse", "clk_m", 39, TEGRA_PERIPH_ON_APB, tegra_clk_fuse, = 0), > > + GATE("fuse", "clk_m", 39, TEGRA_PERIPH_ON_APB, tegra_clk_fuse, = CLK_IS_CRITICAL), > > GATE("fuse_burn", "clk_m", 39, TEGRA_PERIPH_ON_APB, tegra_clk_f= use_burn, 0), > > GATE("kfuse", "clk_m", 40, TEGRA_PERIPH_ON_APB, tegra_clk_kfuse= , 0), > > GATE("apbif", "clk_m", 107, TEGRA_PERIPH_ON_APB, tegra_clk_apbi= f, 0), It's probably fine to do this. The patch I proposed would've restricted the change to just Tegra124. But if we need this on other generations, I don't think the extra complexity is justified, especially since I can't imagine that the FUSE clock remaining always on would consume a lot of extra power. Thierry --eAbsdosE1cNLO4uF Content-Type: application/pgp-signature; name="signature.asc" -----BEGIN PGP SIGNATURE----- iQIzBAABCAAdFiEEiOrDCAFJzPfAjcif3SOs138+s6EFAl2XOIQACgkQ3SOs138+ s6HlwxAAh5ESXhxNQcvDEspLmSsT6O8TIw5aMmfrdeUldnIoOilb5YUh4dp7CixY /+pOnbUDDYbNaYlViZ65/xqJ5/C0+iDn7Kxs4t/AN5W7HNVeUL6p5ltd/iRNIpF7 eiyePXcChZOmYc98HtSsos+2cOMfQ6vyvpDN90tETJZrPvrxp66osXSy2YaECWqy SE6DeN4eJof5bnY+t8EiiXK2rfIcuKMNG9axtYuDcNWlvG+tIrNUbcF2GaajESXe yfxAjlFH2CKfhQ0ju7DIxxkfFhtStvm+1qnW4oUPEzC4LIRRTRBHDFWOmrPnU0UH DKvVQxWSfanaNNFSCBTsT6J+qpVR0N78MlBgWeM9DCQq4fy0lcrozLQhyuyQHeh2 OyXmhfFSm53UilGeFG84kBgu2QfDlWjFMLg4rndPIqCBlVoowbPu6u62Dolxpfp7 ejH7+jGz3ftckFJblStXDw5AwR7x0Aa8X4xYqzrkbDcVXw7R5Mynn6vWYxGurWRd M4I2Lozh/HqPHAIkSvHujI7GxLxa5fVez9kUPMjIyWB6eIE7yE2u33NEQ86FFzlr x0hOLwNhnXhhAZXeKqP6N6wpCTVBX9AVZEGjjRaQKAXIaJWhBbqV5r5PNYGJIPvq AUqYyl/uWZ1GqdJSoLVS42uhIz7D+CsHVGAZfAm7PDpvMg2ol3Q= =C3XR -----END PGP SIGNATURE----- --eAbsdosE1cNLO4uF-- --===============3056316276139130121== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ linux-arm-kernel mailing list linux-arm-kernel@lists.infradead.org http://lists.infradead.org/mailman/listinfo/linux-arm-kernel --===============3056316276139130121==--