From mboxrd@z Thu Jan 1 00:00:00 1970 From: Stefano Babic Date: Tue, 20 Aug 2013 09:21:46 +0200 Subject: [U-Boot] Enabling L2 cache on mx53 In-Reply-To: <201308192231.28535.marex@denx.de> References: <201308192151.18036.marex@denx.de> <201308192231.28535.marex@denx.de> Message-ID: <5213190A.2020807@denx.de> List-Id: MIME-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit To: u-boot@lists.denx.de Hi Marek, On 19/08/2013 22:31, Marek Vasut wrote: > Dear Fabio Estevam, > >> Hi Marek, >> >> On Mon, Aug 19, 2013 at 4:51 PM, Marek Vasut wrote: >>> L2CC on MX53 is enabled by setting just the L2ON and C bits in CP15, >>> there's no configuration. Not even Linux enables the L2CC on MX53, so if >>> it's not on in U- Boot, then it's not on at all (and that sucks). >> >> This is what I have done: >> >> --- a/arch/arm/cpu/armv7/mx5/ >> lowlevel_init.S >> +++ b/arch/arm/cpu/armv7/mx5/lowlevel_init.S >> @@ -45,6 +45,11 @@ >> #endif >> >> mcr 15, 1, r0, c9, c0, 2 >> + >> + /* enable L2 cache */ >> + mrc 15, 0, r0, c1, c0, 1 >> + orr r0, r0, #(1 << 1) /* enable l2 cache */ >> + mcr 15, 0, r0, c1, c0, 1 >> .endm /* init_l2cc */ >> >> /* AIPS setup - Only setup MPROTx registers. >> >> Anything else I am missing? > > Try profiling the RX routine, maybe it's looping somewhere there for too long. > But the RX routine belongs to the FEC driver that it is used by all i.MXes. Is there maybe a problem with the phy and the negotiated speed is less than expected ? Best regards, Stefano Babic -- ===================================================================== DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: +49-8142-66989-53 Fax: +49-8142-66989-80 Email: sbabic at denx.de =====================================================================