January 2011 - U-Boot

[U-Boot] [PATCH] asm-offsets: generate bd_t size
by Mike Frysinger 10 Jan '11

10 Jan '11

Some ports set up the board info structure at the same time as the global data structure, and largely keep them together. So generate a define for the board info struct too. Signed-off-by: Mike Frysinger <vapier(a)gentoo.org> --- lib/asm-offsets.c | 3 +++ 1 files changed, 3 insertions(+), 0 deletions(-) diff --git a/lib/asm-offsets.c b/lib/asm-offsets.c index 2209561..6f253a6 100644 --- a/lib/asm-offsets.c +++ b/lib/asm-offsets.c @@ -25,5 +25,8 @@ int main(void) DEFINE(GENERATED_GBL_DATA_SIZE, (sizeof(struct global_data)+15) & ~15); + DEFINE(GENERATED_BD_INFO_SIZE, + (sizeof(struct bd_info)+15) & ~15); + return 0; } -- 1.7.3.4

3 7

[U-Boot] Art of England Magazine - March 2011 edition - Introduction AdRate Offer until 15th Jan
by Art of England Magazine 10 Jan '11

10 Jan '11

Hi there Happy New year from all of us over here! We're pleased to inform you that your organisation has been invited to participate in the exclusive Art of England Magazine, March 2011 edition, deadline for which is now just 4 days away. Now seven years old, the Art of England Magazine offers you an effective monthly targeted distribution throughout the UK and abroad, providing you with a lucrative reach to the elusive art and luxury market sectors (full distribution details available here). Participants range from art galleries, dealers, auctioneers, investors and associated professions, through to those who wish to contact this financially capable demographic for their products and services, enjoying the appropriate business development from our sought-after readership. The editorial content too, as you might imagine, is suitably exclusive, and commensurate with the subject matter (current edition available via website below). Regarding your organisation and your potential participation, we're pleased to tell you that the Special Introductory Rate for booking for the coming March 2011 Edition has just been extended slightly, and until 15th January 2011 we would like to offer you the extended reduced 'trial' rate of just £950.00 total for a full A4 page advertisement (list price £1250.00, usual rate card available on request), or £500.00 for a half page, or just £275.00 for a quarter! Once you have seen the effectiveness of this sales and PR channel, we would, of course, hope that you might elect to appear regularly later in subsequent editions. To book for this attractive trial rate offer for the March 2011 Edition, simply reply to this email with your preferred ad size no later than 15th January 2011 to begin reaching this high net worth audience now available to your organisation. Thanks for your time, and we hope to be welcoming you aboard very shortly! Kind regards Deni Parkes for Art of England Magazine T: +44 (0)845 80 55 173 F: +44 (0)207 183 4752 W: www.artofengland.uk.com Media Kit available on request Contracted exclusively by: SKS/Enhanced Media & Communications Ltd 1 Farnham Road, Guildford, Surrey GU2 4RG UK ** You were subscribed to this offer list through one of our extensive, but discerning luxury partner networks. ** If you feel you have received this email in error, simply REPLY and SEND with the subject line intact. Thanks.

1 0

[U-Boot] Hello
by mercy lutunku 10 Jan '11

10 Jan '11

Hello, my name is Mercy, i am interested in having good relationship with you. i hope you will welcome my friendship request, I will tell you more about me and send my pictures to you as soon as you respond to me, i will be waiting for your mail soon. Yours Mercy

1 0

[U-Boot] [PATCH V3] mpc83xx: fix pcie configuration space read/write
by Baidu Boy 10 Jan '11

10 Jan '11

This patch fix a problem for the pcie enumeration when the mpc83xx pcie controller is connected with switch or we use both of the two pcie controller Signed-off-by: Baidu Boy <liucai.lfn(a)gmail.com> --- Changes for V2: - Avoid line wrap in the patch Changes for V3 - Add space between ) and { arch/powerpc/cpu/mpc83xx/pcie.c | 8 ++++++-- 1 files changed, 6 insertions(+), 2 deletions(-) diff --git a/arch/powerpc/cpu/mpc83xx/pcie.c b/arch/powerpc/cpu/mpc83xx/pcie.c index 46a706d..8429848 100644 --- a/arch/powerpc/cpu/mpc83xx/pcie.c +++ b/arch/powerpc/cpu/mpc83xx/pcie.c @@ -46,13 +46,15 @@ static struct { #endif }; +static u8 pcie_index = 0; + #ifdef CONFIG_83XX_GENERIC_PCIE_REGISTER_HOSES static int mpc83xx_pcie_remap_cfg(struct pci_controller *hose, pci_dev_t dev) { int bus = PCI_BUS(dev) - hose->first_busno; immap_t *immr = (immap_t *)CONFIG_SYS_IMMR; - pex83xx_t *pex = &immr->pciexp[bus]; + pex83xx_t *pex = &immr->pciexp[pcie_index]; struct pex_outbound_window *out_win = &pex->bridge.pex_outbound_win[0]; u8 devfn = PCI_DEV(dev) << 3 | PCI_FUNC(dev); u32 dev_base = bus << 24 | devfn << 16; @@ -324,6 +326,8 @@ void mpc83xx_pcie_init(int num_buses, struct pci_region **reg) num_buses = ARRAY_SIZE(mpc83xx_pcie_cfg_space); } - for (i = 0; i < num_buses; i++) + for (i = 0; i < num_buses; i++) { + pcie_index = i; mpc83xx_pcie_init_bus(i, reg[i]); + } } -- 1.7.3.1.msysgit.0

1 0

[U-Boot] [PATCH] PPC4xx: Reduce NAND TLB window size on Canyonlands
by Felix Radensky 10 Jan '11

10 Jan '11

16MiB NAND TLB window is way too big. Reduce it to 1KiB. Signed-off-by: Felix Radensky <felix(a)embedded-sol.com> --- board/amcc/canyonlands/init.S | 2 +- 1 files changed, 1 insertions(+), 1 deletions(-) diff --git a/board/amcc/canyonlands/init.S b/board/amcc/canyonlands/init.S index 680feaa..8b7bfce 100644 --- a/board/amcc/canyonlands/init.S +++ b/board/amcc/canyonlands/init.S @@ -80,7 +80,7 @@ tlbtab: #if !defined(CONFIG_ARCHES) /* TLB-entry for NAND */ - tlbentry(CONFIG_SYS_NAND_ADDR, SZ_16M, CONFIG_SYS_NAND_ADDR, 4, AC_RWX | SA_IG) + tlbentry(CONFIG_SYS_NAND_ADDR, SZ_1K, CONFIG_SYS_NAND_ADDR, 4, AC_RWX | SA_IG) /* TLB-entry for CPLD */ tlbentry(CONFIG_SYS_BCSR_BASE, SZ_1K, CONFIG_SYS_BCSR_BASE, 4, AC_RW | SA_IG) -- 1.5.4.3

2 1

[U-Boot] [PATCH] ppc4xx: Remove PCI support from lwmon5
by Stefan Roese 10 Jan '11

10 Jan '11

PCI is not used at all on lwmon5. So lets remove it. It saves space and reduces boot time a bit (approx. 50ms). Signed-off-by: Stefan Roese <sr(a)denx.de> --- include/configs/lwmon5.h | 17 ----------------- 1 files changed, 0 insertions(+), 17 deletions(-) diff --git a/include/configs/lwmon5.h b/include/configs/lwmon5.h index aedf495..a1ead70 100644 --- a/include/configs/lwmon5.h +++ b/include/configs/lwmon5.h @@ -440,7 +440,6 @@ #define CONFIG_CMD_MII #define CONFIG_CMD_NET #define CONFIG_CMD_NFS -#define CONFIG_CMD_PCI #define CONFIG_CMD_PING #define CONFIG_CMD_REGINFO #define CONFIG_CMD_SDRAM @@ -488,22 +487,6 @@ #define CONFIG_MX_CYCLIC 1 /* enable mdc/mwc commands */ #define CONFIG_VERSION_VARIABLE 1 /* include version env variable */ -/* - * PCI stuff - */ -/* General PCI */ -#define CONFIG_PCI /* include pci support */ -#undef CONFIG_PCI_PNP /* do (not) pci plug-and-play */ -#define CONFIG_PCI_SCAN_SHOW /* show pci devices on startup */ -#define CONFIG_SYS_PCI_TARGBASE 0x80000000 /* PCIaddr mapped to CONFIG_SYS_PCI_MEMBASE*/ - -/* Board-specific PCI */ -#define CONFIG_SYS_PCI_TARGET_INIT -#define CONFIG_SYS_PCI_MASTER_INIT - -#define CONFIG_SYS_PCI_SUBSYS_VENDORID 0x10e8 /* AMCC */ -#define CONFIG_SYS_PCI_SUBSYS_ID 0xcafe /* Whatever */ - #ifndef DEBUG #define CONFIG_HW_WATCHDOG 1 /* Use external HW-Watchdog */ #endif -- 1.7.3.4

1 1

[U-Boot] [PATCH] cfi_flash: avoid flash_verbose when possible
by Mike Frysinger 10 Jan '11

10 Jan '11

The flash_verbose logic is only used by the CFI MTD layer, so if we aren't using that, disable the logic completely. Signed-off-by: Mike Frysinger <vapier(a)gentoo.org> --- drivers/mtd/cfi_flash.c | 6 ++++++ 1 files changed, 6 insertions(+), 0 deletions(-) diff --git a/drivers/mtd/cfi_flash.c b/drivers/mtd/cfi_flash.c index 39c235e..b05264b 100644 --- a/drivers/mtd/cfi_flash.c +++ b/drivers/mtd/cfi_flash.c @@ -63,7 +63,11 @@ */ static uint flash_offset_cfi[2] = { FLASH_OFFSET_CFI, FLASH_OFFSET_CFI_ALT }; +#ifdef CONFIG_FLASH_CFI_MTD static uint flash_verbose = 1; +#else +#define flash_verbose 1 +#endif flash_info_t flash_info[CFI_MAX_FLASH_BANKS]; /* FLASH chips info */ @@ -2016,10 +2020,12 @@ ulong flash_get_size (phys_addr_t base, int banknum) return (info->size); } +#ifdef CONFIG_FLASH_CFI_MTD void flash_set_verbose(uint v) { flash_verbose = v; } +#endif /*----------------------------------------------------------------------- */ -- 1.7.3.4

3 2

[U-Boot] NAND: mxc_nand reads/writes only first 512 bytes of each page
by Stefano Babic 10 Jan '11

10 Jan '11

Hi, My target is a mx35 board (not yet in mainline). According to the manual and checking the register set and their functionality, the MX.35 has the same nand controller as the i.MX25 processor (MXC_NFC_V1_1 in the mxc_nand driver). Probably there is the same issue with the i.MX25, too. The storage is a Samsung 2GiB (8-bit), 2KB page, sector size 256KiB, and CONFIG_SYS_NAND_LARGEPAGE is set. The flash is recognized correctly. However, after reading/writing, only the first 512 bytes of each page are correct. The rest is filled with random bytes. Under Linux (2.6.37) everything works flawlessy, however the driver in Linux is quite different as what we have in u-boot, and this makes a comparison quite difficult. Any hint how can I better investigate this issue ? Best regards, Stefano Babic -- ===================================================================== DENX Software Engineering GmbH, MD: Wolfgang Denk & Detlev Zundel HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany Phone: +49-8142-66989-0 Fax: +49-8142-66989-80 Email: office(a)denx.de =====================================================================

2 2

[U-Boot] [PATCH] sf: punt unused spi_flash_region struct
by Mike Frysinger 10 Jan '11

10 Jan '11

Signed-off-by: Mike Frysinger <vapier(a)gentoo.org> --- include/spi_flash.h | 5 ----- 1 files changed, 0 insertions(+), 5 deletions(-) diff --git a/include/spi_flash.h b/include/spi_flash.h index 1f8ba29..89cc3a7 100644 --- a/include/spi_flash.h +++ b/include/spi_flash.h @@ -26,11 +26,6 @@ #include <spi.h> #include <linux/types.h> -struct spi_flash_region { - unsigned int count; - unsigned int size; -}; - struct spi_flash { struct spi_slave *spi; -- 1.7.4.rc1

1 0

[U-Boot] [PATCH] Blackfin: adi boards: enable CONFIG_MONITOR_IS_IN_RAM
by Mike Frysinger 10 Jan '11

10 Jan '11

Our monitor is always in RAM, so enable this define for the CFI layer. Signed-off-by: Mike Frysinger <vapier(a)gentoo.org> --- include/configs/bf506f-ezkit.h | 1 + include/configs/bfin_adi_common.h | 1 + 2 files changed, 2 insertions(+), 0 deletions(-) diff --git a/include/configs/bf506f-ezkit.h b/include/configs/bf506f-ezkit.h index 9464029..0beac78 100644 --- a/include/configs/bf506f-ezkit.h +++ b/include/configs/bf506f-ezkit.h @@ -65,6 +65,7 @@ #define CONFIG_SYS_MAX_FLASH_BANKS 1 #define CONFIG_SYS_MAX_FLASH_SECT 71 #define CONFIG_CMD_FLASH +#define CONFIG_MONITOR_IS_IN_RAM /* diff --git a/include/configs/bfin_adi_common.h b/include/configs/bfin_adi_common.h index 9016a26..a9c66be 100644 --- a/include/configs/bfin_adi_common.h +++ b/include/configs/bfin_adi_common.h @@ -300,5 +300,6 @@ #endif #define CONFIG_BFIN_SPI_GPIO_CS /* Only matters if BFIN_SPI is enabled */ #define CONFIG_LZMA +#define CONFIG_MONITOR_IS_IN_RAM #endif -- 1.7.4.rc1

1 0