From: Benjamin Herrenschmidt <benh@kernel.crashing.org>
To: Tirumala Reddy Marri <tmarri@amcc.com>
Cc: linuxppc-dev@ozlabs.org, tmarri@macc.com
Subject: RE: [PATCH] Adding PCI-E support for 460SX based redwood board.
Date: Tue, 01 Dec 2009 06:38:12 +1100 [thread overview]
Message-ID: <1259609892.2076.96.camel@pasglop> (raw)
In-Reply-To: <AC5E1C3367E37D44970B81A6ADD1DA2C08642D5B@SDCEXCHANGE01.ad.amcc.com>
On Mon, 2009-11-30 at 10:52 -0800, Tirumala Reddy Marri wrote:
> Hi Ben,
> Could you please review the patch I sent .
> Thanks,
> Marri
Hi !
The original message never made it. This one did but the patch is line
wrapped. Please fix your mailer and re-send.
Cheers,
Ben.
> -----Original Message-----
> From: tmarri@amcc.com [mailto:tmarri@amcc.com]
> Sent: Wednesday, November 25, 2009 3:49 PM
> To: linuxppc-dev@ozlabs.org
> Cc: tmarri@macc.com; Tirumala Reddy Marri
> Subject: [PATCH] Adding PCI-E support for 460SX based redwood board.
>
> From: Tirumala Marri <tmarri@amcc.com>
>
> This patch would add PCI-E support for AMCC 460SX processor based
> redwood board.
>
> Signed-off-by: Tirumala Marri <tmarri@amcc.com>
>
> ---
> arch/powerpc/boot/dts/redwood.dts | 122
> +++++++++++++++++++++++++++++++++++++
> arch/powerpc/sysdev/ppc4xx_pci.c | 119
> ++++++++++++++++++++++++++++++++++++
> arch/powerpc/sysdev/ppc4xx_pci.h | 58 +++++++++++++++++
> 3 files changed, 299 insertions(+), 0 deletions(-)
>
> diff --git a/arch/powerpc/boot/dts/redwood.dts
> b/arch/powerpc/boot/dts/redwood.dts
> index ad402c4..9eeec28 100644
> --- a/arch/powerpc/boot/dts/redwood.dts
> +++ b/arch/powerpc/boot/dts/redwood.dts
> @@ -233,6 +233,128 @@
> has-inverted-stacr-oc;
> has-new-stacr-staopc;
> };
> + PCIE0: pciex@d00000000 {
> + device_type = "pci";
> + #interrupt-cells = <1>;
> + #size-cells = <2>;
> + #address-cells = <3>;
> + compatible = "ibm,plb-pciex-460sx",
> "ibm,plb-pciex";
> + primary;
> + port = <0x0>; /* port number */
> + reg = <0x0000000d 0x00000000 0x20000000
> /* Config space access */
> + 0x0000000c 0x10000000
> 0x00001000>; /* Registers */
> + dcr-reg = <0x100 0x020>;
> + sdr-base = <0x300>;
> +
> + /* Outbound ranges, one memory and one
> IO,
> + * later cannot be changed
> + */
> + ranges = <0x02000000 0x00000000
> 0x80000000 0x0000000e 0x00000000 0x00000000 0x80000000
> + 0x01000000 0x00000000
> 0x00000000 0x0000000f 0x80000000 0x00000000 0x00010000>;
> +
> + /* Inbound 2GB range starting at 0 */
> + dma-ranges = <0x42000000 0x0 0x0 0x0 0x0
> 0x0 0x80000000>;
> +
> + /* This drives busses 10 to 0x1f */
> + bus-range = <0x10 0x1f>;
> +
> + /* Legacy interrupts (note the weird
> polarity, the bridge seems
> + * to invert PCIe legacy interrupts).
> + * We are de-swizzling here because the
> numbers are actually for
> + * port of the root complex virtual P2P
> bridge. But I want
> + * to avoid putting a node for it in the
> tree, so the numbers
> + * below are basically de-swizzled
> numbers.
> + * The real slot is on idsel 0, so the
> swizzling is 1:1
> + */
> + interrupt-map-mask = <0x0 0x0 0x0 0x7>;
> + interrupt-map = <
> + 0x0 0x0 0x0 0x1 &UIC3 0x0 0x4 /*
> swizzled int A */
> + 0x0 0x0 0x0 0x2 &UIC3 0x1 0x4 /*
> swizzled int B */
> + 0x0 0x0 0x0 0x3 &UIC3 0x2 0x4 /*
> swizzled int C */
> + 0x0 0x0 0x0 0x4 &UIC3 0x3 0x4 /*
> swizzled int D */>;
> + };
> +
> + PCIE1: pciex@d20000000 {
> + device_type = "pci";
> + #interrupt-cells = <1>;
> + #size-cells = <2>;
> + #address-cells = <3>;
> + compatible = "ibm,plb-pciex-460sx",
> "ibm,plb-pciex";
> + primary;
> + port = <0x1>; /* port number */
> + reg = <0x0000000d 0x20000000 0x20000000
> /* Config space access */
> + 0x0000000c 0x10001000
> 0x00001000>; /* Registers */
> + dcr-reg = <0x120 0x020>;
> + sdr-base = <0x340>;
> +
> + /* Outbound ranges, one memory and one
> IO,
> + * later cannot be changed
> + */
> + ranges = <0x02000000 0x00000000
> 0x80000000 0x0000000e 0x80000000 0x00000000 0x80000000
> + 0x01000000 0x00000000
> 0x00000000 0x0000000f 0x80010000 0x00000000 0x00010000>;
> +
> + /* Inbound 2GB range starting at 0 */
> + dma-ranges = <0x42000000 0x0 0x0 0x0 0x0
> 0x0 0x80000000>;
> +
> + /* This drives busses 10 to 0x1f */
> + bus-range = <0x20 0x2f>;
> +
> + /* Legacy interrupts (note the weird
> polarity, the bridge seems
> + * to invert PCIe legacy interrupts).
> + * We are de-swizzling here because the
> numbers are actually for
> + * port of the root complex virtual P2P
> bridge. But I want
> + * to avoid putting a node for it in the
> tree, so the numbers
> + * below are basically de-swizzled
> numbers.
> + * The real slot is on idsel 0, so the
> swizzling is 1:1
> + */
> + interrupt-map-mask = <0x0 0x0 0x0 0x7>;
> + interrupt-map = <
> + 0x0 0x0 0x0 0x1 &UIC3 0x4 0x4 /*
> swizzled int A */
> + 0x0 0x0 0x0 0x2 &UIC3 0x5 0x4 /*
> swizzled int B */
> + 0x0 0x0 0x0 0x3 &UIC3 0x6 0x4 /*
> swizzled int C */
> + 0x0 0x0 0x0 0x4 &UIC3 0x7 0x4 /*
> swizzled int D */>;
> + };
> +
> + PCIE2: pciex@d40000000 {
> + device_type = "pci";
> + #interrupt-cells = <1>;
> + #size-cells = <2>;
> + #address-cells = <3>;
> + compatible = "ibm,plb-pciex-460sx",
> "ibm,plb-pciex";
> + primary;
> + port = <0x2>; /* port number */
> + reg = <0x0000000d 0x40000000 0x20000000
> /* Config space access */
> + 0x0000000c 0x10002000
> 0x00001000>; /* Registers */
> + dcr-reg = <0x140 0x020>;
> + sdr-base = <0x370>;
> +
> + /* Outbound ranges, one memory and one
> IO,
> + * later cannot be changed
> + */
> + ranges = <0x02000000 0x00000000
> 0x80000000 0x0000000f 0x00000000 0x00000000 0x80000000
> + 0x01000000 0x00000000
> 0x00000000 0x0000000f 0x80020000 0x00000000 0x00010000>;
> +
> + /* Inbound 2GB range starting at 0 */
> + dma-ranges = <0x42000000 0x0 0x0 0x0 0x0
> 0x0 0x80000000>;
> +
> + /* This drives busses 10 to 0x1f */
> + bus-range = <0x30 0x3f>;
> +
> + /* Legacy interrupts (note the weird
> polarity, the bridge seems
> + * to invert PCIe legacy interrupts).
> + * We are de-swizzling here because the
> numbers are actually for
> + * port of the root complex virtual P2P
> bridge. But I want
> + * to avoid putting a node for it in the
> tree, so the numbers
> + * below are basically de-swizzled
> numbers.
> + * The real slot is on idsel 0, so the
> swizzling is 1:1
> + */
> + interrupt-map-mask = <0x0 0x0 0x0 0x7>;
> + interrupt-map = <
> + 0x0 0x0 0x0 0x1 &UIC3 0x8 0x4 /*
> swizzled int A */
> + 0x0 0x0 0x0 0x2 &UIC3 0x9 0x4 /*
> swizzled int B */
> + 0x0 0x0 0x0 0x3 &UIC3 0xa 0x4 /*
> swizzled int C */
> + 0x0 0x0 0x0 0x4 &UIC3 0xb 0x4 /*
> swizzled int D */>;
> + };
>
> };
>
> diff --git a/arch/powerpc/sysdev/ppc4xx_pci.c
> b/arch/powerpc/sysdev/ppc4xx_pci.c
> index 6ff9d71..64cd020 100644
> --- a/arch/powerpc/sysdev/ppc4xx_pci.c
> +++ b/arch/powerpc/sysdev/ppc4xx_pci.c
> @@ -972,6 +972,123 @@ static struct ppc4xx_pciex_hwops
> ppc460ex_pcie_hwops __initdata =
> .setup_utl = ppc460ex_pciex_init_utl,
> };
>
> +static int __init ppc460sx_pciex_core_init(struct device_node *np)
> +{
> + /* HSS drive amplitude */
> + mtdcri(SDR0, PESDR0_460SX_HSSL0DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR0_460SX_HSSL1DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR0_460SX_HSSL2DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR0_460SX_HSSL3DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR0_460SX_HSSL4DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR0_460SX_HSSL5DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR0_460SX_HSSL6DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR0_460SX_HSSL7DAMP, 0xB9843211);
> +
> + mtdcri(SDR0, PESDR1_460SX_HSSL0DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR1_460SX_HSSL1DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR1_460SX_HSSL2DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR1_460SX_HSSL3DAMP, 0xB9843211);
> +
> + mtdcri(SDR0, PESDR2_460SX_HSSL0DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR2_460SX_HSSL1DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR2_460SX_HSSL2DAMP, 0xB9843211);
> + mtdcri(SDR0, PESDR2_460SX_HSSL3DAMP, 0xB9843211);
> +
> + /* HSS TX pre-emphasis */
> + mtdcri(SDR0, PESDR0_460SX_HSSL0COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR0_460SX_HSSL1COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR0_460SX_HSSL2COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR0_460SX_HSSL3COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR0_460SX_HSSL4COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR0_460SX_HSSL5COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR0_460SX_HSSL6COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR0_460SX_HSSL7COEFA, 0xDCB98987);
> +
> + mtdcri(SDR0, PESDR1_460SX_HSSL0COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR1_460SX_HSSL1COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR1_460SX_HSSL2COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR1_460SX_HSSL3COEFA, 0xDCB98987);
> +
> + mtdcri(SDR0, PESDR2_460SX_HSSL0COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR2_460SX_HSSL1COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR2_460SX_HSSL2COEFA, 0xDCB98987);
> + mtdcri(SDR0, PESDR2_460SX_HSSL3COEFA, 0xDCB98987);
> +
> + /* HSS TX calibration control */
> + mtdcri(SDR0, PESDR0_460SX_HSSL1CALDRV, 0x22222222);
> + mtdcri(SDR0, PESDR1_460SX_HSSL1CALDRV, 0x22220000);
> + mtdcri(SDR0, PESDR2_460SX_HSSL1CALDRV, 0x22220000);
> +
> + /* HSS TX slew control */
> + mtdcri(SDR0, PESDR0_460SX_HSSSLEW, 0xFFFFFFFF);
> + mtdcri(SDR0, PESDR1_460SX_HSSSLEW, 0xFFFF0000);
> + mtdcri(SDR0, PESDR2_460SX_HSSSLEW, 0xFFFF0000);
> +
> + udelay(100);
> +
> + /* De-assert PLLRESET */
> + dcri_clrset(SDR0, PESDR0_PLLLCT2, 0x00000100, 0);
> +
> + /* Reset DL, UTL, GPL before configuration */
> + mtdcri(SDR0, PESDR0_460SX_RCSSET,
> + PESDRx_RCSSET_RSTDL | PESDRx_RCSSET_RSTGU);
> + mtdcri(SDR0, PESDR1_460SX_RCSSET,
> + PESDRx_RCSSET_RSTDL | PESDRx_RCSSET_RSTGU);
> + mtdcri(SDR0, PESDR2_460SX_RCSSET,
> + PESDRx_RCSSET_RSTDL | PESDRx_RCSSET_RSTGU);
> +
> + udelay(100);
> +
> + /*
> + * If bifurcation is not enabled, u-boot would have disabled the
> + * third PCIe port
> + */
> + if (((mfdcri(SDR0, PESDR1_460SX_HSSCTLSET) & 0x00000001) ==
> + 0x00000001)) {
> + printk(KERN_INFO "PCI: PCIE bifurcation setup
> successfully.\n");
> + printk(KERN_INFO "PCI: Total 3 PCIE ports are
> present\n");
> + return 3;
> + }
> +
> + printk(KERN_INFO "PCI: Total 2 PCIE ports are present\n");
> + return 2;
> +}
> +
> +static int ppc460sx_pciex_init_port_hw(struct ppc4xx_pciex_port *port)
> +{
> +
> + if (port->endpoint)
> + dcri_clrset(SDR0, port->sdr_base + PESDRn_UTLSET2,
> + 0x01000000, 0);
> + else
> + dcri_clrset(SDR0, port->sdr_base + PESDRn_UTLSET2,
> + 0, 0x01000000);
> +
> + /*Gen-1*/
> + mtdcri(SDR0, port->sdr_base + PESDRn_460SX_RCEI, 0x08000000);
> +
> + dcri_clrset(SDR0, port->sdr_base + PESDRn_RCSSET,
> + (PESDRx_RCSSET_RSTGU | PESDRx_RCSSET_RSTDL),
> + PESDRx_RCSSET_RSTPYN);
> +
> + port->has_ibpre = 1;
> +
> + return 0;
> +}
> +
> +static int ppc460sx_pciex_init_utl(struct ppc4xx_pciex_port *port)
> +{
> + /* Max 128 Bytes */
> + out_be32 (port->utl_base + PEUTL_PBBSZ, 0x00000000);
> + return 0;
> +}
> +
> +static struct ppc4xx_pciex_hwops ppc460sx_pcie_hwops __initdata = {
> + .core_init = ppc460sx_pciex_core_init,
> + .port_init_hw = ppc460sx_pciex_init_port_hw,
> + .setup_utl = ppc460sx_pciex_init_utl,
> +};
> +
> #endif /* CONFIG_44x */
>
> #ifdef CONFIG_40x
> @@ -1087,6 +1204,8 @@ static int __init
> ppc4xx_pciex_check_core_init(struct device_node *np)
> }
> if (of_device_is_compatible(np, "ibm,plb-pciex-460ex"))
> ppc4xx_pciex_hwops = &ppc460ex_pcie_hwops;
> + if (of_device_is_compatible(np, "ibm,plb-pciex-460sx"))
> + ppc4xx_pciex_hwops = &ppc460sx_pcie_hwops;
> #endif /* CONFIG_44x */
> #ifdef CONFIG_40x
> if (of_device_is_compatible(np, "ibm,plb-pciex-405ex"))
> diff --git a/arch/powerpc/sysdev/ppc4xx_pci.h
> b/arch/powerpc/sysdev/ppc4xx_pci.h
> index d04e40b..56d9e5d 100644
> --- a/arch/powerpc/sysdev/ppc4xx_pci.h
> +++ b/arch/powerpc/sysdev/ppc4xx_pci.h
> @@ -324,6 +324,64 @@
> #define PESDR0_460EX_IHS2 0x036D
>
> /*
> + * 460SX addtional DCRs
> + */
> +#define PESDRn_460SX_RCEI 0x02
> +
> +#define PESDR0_460SX_HSSL0DAMP 0x320
> +#define PESDR0_460SX_HSSL1DAMP 0x321
> +#define PESDR0_460SX_HSSL2DAMP 0x322
> +#define PESDR0_460SX_HSSL3DAMP 0x323
> +#define PESDR0_460SX_HSSL4DAMP 0x324
> +#define PESDR0_460SX_HSSL5DAMP 0x325
> +#define PESDR0_460SX_HSSL6DAMP 0x326
> +#define PESDR0_460SX_HSSL7DAMP 0x327
> +
> +#define PESDR1_460SX_HSSL0DAMP 0x354
> +#define PESDR1_460SX_HSSL1DAMP 0x355
> +#define PESDR1_460SX_HSSL2DAMP 0x356
> +#define PESDR1_460SX_HSSL3DAMP 0x357
> +
> +#define PESDR2_460SX_HSSL0DAMP 0x384
> +#define PESDR2_460SX_HSSL1DAMP 0x385
> +#define PESDR2_460SX_HSSL2DAMP 0x386
> +#define PESDR2_460SX_HSSL3DAMP 0x387
> +
> +#define PESDR0_460SX_HSSL0COEFA 0x328
> +#define PESDR0_460SX_HSSL1COEFA 0x329
> +#define PESDR0_460SX_HSSL2COEFA 0x32A
> +#define PESDR0_460SX_HSSL3COEFA 0x32B
> +#define PESDR0_460SX_HSSL4COEFA 0x32C
> +#define PESDR0_460SX_HSSL5COEFA 0x32D
> +#define PESDR0_460SX_HSSL6COEFA 0x32E
> +#define PESDR0_460SX_HSSL7COEFA 0x32F
> +
> +#define PESDR1_460SX_HSSL0COEFA 0x358
> +#define PESDR1_460SX_HSSL1COEFA 0x359
> +#define PESDR1_460SX_HSSL2COEFA 0x35A
> +#define PESDR1_460SX_HSSL3COEFA 0x35B
> +
> +#define PESDR2_460SX_HSSL0COEFA 0x388
> +#define PESDR2_460SX_HSSL1COEFA 0x389
> +#define PESDR2_460SX_HSSL2COEFA 0x38A
> +#define PESDR2_460SX_HSSL3COEFA 0x38B
> +
> +#define PESDR0_460SX_HSSL1CALDRV 0x339
> +#define PESDR1_460SX_HSSL1CALDRV 0x361
> +#define PESDR2_460SX_HSSL1CALDRV 0x391
> +
> +#define PESDR0_460SX_HSSSLEW 0x338
> +#define PESDR1_460SX_HSSSLEW 0x360
> +#define PESDR2_460SX_HSSSLEW 0x390
> +
> +#define PESDR0_460SX_HSSCTLSET 0x31E
> +#define PESDR1_460SX_HSSCTLSET 0x352
> +#define PESDR2_460SX_HSSCTLSET 0x382
> +
> +#define PESDR0_460SX_RCSSET 0x304
> +#define PESDR1_460SX_RCSSET 0x344
> +#define PESDR2_460SX_RCSSET 0x374
> +/*
> * Of the above, some are common offsets from the base
> */
> #define PESDRn_UTLSET1 0x00
next prev parent reply other threads:[~2009-11-30 19:38 UTC|newest]
Thread overview: 5+ messages / expand[flat|nested] mbox.gz Atom feed top
[not found] <1259192937-8591-1-git-send-email-tmarri@amcc.com>
2009-11-30 18:52 ` [PATCH] Adding PCI-E support for 460SX based redwood board Tirumala Reddy Marri
2009-11-30 19:38 ` Benjamin Herrenschmidt [this message]
2009-11-30 22:58 tmarri
2009-12-18 14:03 ` Josh Boyer
[not found] <1259615732-20272-1-git-send-email-tmarr@amcc.com>
2009-12-10 2:22 ` Tirumala Reddy Marri
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=1259609892.2076.96.camel@pasglop \
--to=benh@kernel.crashing.org \
--cc=linuxppc-dev@ozlabs.org \
--cc=tmarri@amcc.com \
--cc=tmarri@macc.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).