July 2013 - U-Boot

[U-Boot] [PULL] u-boot-usb/master
by Marek Vasut 12 Jul '13

12 Jul '13

The following changes since commit 225fd8c5d4556547896a5d32ee092a258f3df638: cmd_bootm.c: Make bootz handle BOOTM_STATE_FINDOTHER itself (2013-07-10 09:15:15 -0400) are available in the git repository at: git://git.denx.de/u-boot-usb.git master for you to fetch changes up to ecc8edbf6995558d8a47b43ac6645840c98a7b95: usb: pxa27x_udc: fix compiler warnings (2013-07-10 20:18:57 +0200) ---------------------------------------------------------------- Mike Dunn (1): usb: pxa27x_udc: fix compiler warnings Łukasz Majewski (1): dfu: Update DFU's authorship history drivers/usb/gadget/f_dfu.c | 7 +++++++ drivers/usb/gadget/pxa27x_udc.c | 14 +++++--------- 2 files changed, 12 insertions(+), 9 deletions(-)

2 1

[U-Boot] [PATCH 0/4] Correct some remaining bootm problems
by Simon Glass 12 Jul '13

12 Jul '13

The tortuous refactoring of bootm and fit_image_load() has thrown up four more issues: - Support for bootm (without arguments) on many OSes is broken since the OS functions do not handle the PREP stage, which is now required - In the case of loading a FIT image with a bootm-selected kernel configuration, this is not used for ramdisk and fdt but must be manually specified for each. - The Elf image loader arguments are out by 1 - There is an extra OK message in the case of loading an uncompressed kernel This series corrects these problems. Verification was done using some sandbox tests written for the occassion. These will be sent to the list during the next merge window (or earlier if anyone wants to try them). For the record the tests run are: $ ./test/image/test-fit.py -u b/sandbox/u-boot FIT Tests ========= Kernel load Kernel + FDT load Kernel + FDT + Ramdisk load Kernel + FDT + Ramdisk load, Config 2 Tests passed Caveat: this is only a sanity check - test coverage is poor $ ./test/image/test-legacy.py -u b/sandbox/u-boot Legacy Image Tests ================== Testing 22 image types ['netbsd', 'lynxos', 'rtems', 'ose', 'plan9', 'vxworks', 'qnx', 'integrity'] netbsd: False netbsd: True lynxos: False lynxos: True rtems: False rtems: True ose: False ose: True plan9: False plan9: True vxworks: False vxworks: True qnx: False qnx: True integrity: False integrity: True Tests passed Caveat: this is only a sanity check - test coverage is poor Simon Glass (4): bootm: Remove extra OK message blackfin: x86: bootm: Handle PREP stage of bootm bootm: Use selected configuration for ramdisk and fdt bootm: Correct the arguments for the ELF image loader arch/blackfin/lib/boot.c | 2 ++ arch/x86/lib/bootm.c | 2 ++ common/cmd_bootm.c | 20 ++++++++++++++++++-- common/cmd_elf.c | 6 +++--- common/image-fdt.c | 4 ++-- common/image-fit.c | 6 +++++- common/image.c | 4 ++-- include/image.h | 7 ++++--- 8 files changed, 38 insertions(+), 13 deletions(-) -- 1.8.3

3 6

[U-Boot] Pull request: nand flash
by Scott Wood 12 Jul '13

12 Jul '13

Sorry for the lateness, but this patch fixes a BUG() regression with CONFIG_MXC_NAND_HWECC. The following changes since commit 225fd8c5d4556547896a5d32ee092a258f3df638: cmd_bootm.c: Make bootz handle BOOTM_STATE_FINDOTHER itself (2013-07-10 09:15:15 -0400) are available in the git repository at: git://git.denx.de/u-boot-nand-flash.git master for you to fetch changes up to 1c903694370259738abccc67ff0d6e5eaf7886f9: mtd: mxc_nand: Fix crash after MTD resync (2013-07-12 16:12:33 -0500) ---------------------------------------------------------------- Marek Vasut (1): mtd: mxc_nand: Fix crash after MTD resync drivers/mtd/nand/mxc_nand.c | 10 ++++------ 1 file changed, 4 insertions(+), 6 deletions(-)

2 1

[U-Boot] SPL size issues on OMAP4
by Michael Cashwell 12 Jul '13

12 Jul '13

Greetings, I've been absent for a while and couldn't find a way to search the list archives so I apologize if this has already been discussed… I've been fighting the SPL binary growing too large on OMAP4 (using custom configs and features). It's annoying that too large just fails to run with no build or runtime notice. But that's a different issue. My main issue is that in looking through the map for SPL I've repeatedly found code that I don't need and have a pretty good handle on that. My issue is that code that is compiled but eliminated because it's not called leaves behind all of its anonymous strings ("like this"). In my latest build I have the following sections that are all anonymous strings: 0x4030b638 0x232 arch/arm/cpu/armv7/omap-common/libomap-common.o 0x4030b8b5 0x19 arch/arm/cpu/armv7/omap4/libomap4.o 0x4030b9ad 0xbe common/spl/libspl.o 0x4030ba6b 0x57 drivers/gpio/libgpio.o 0x4030bac2 0x44c drivers/i2c/libi2c.o 0x4030bf0e 0x302 drivers/mmc/libmmc.o 0x4030c27e 0x15 drivers/serial/libserial.o 0x4030c293 0x145 drivers/spi/libspi.o 0x4030c4d8 0x53 lib/libgeneric.o with more than half being unreferenced. This is a big deal when you only have about 25K of SRAM for code and data. In searching the web it seems that dead code and data are removed using --gc-sections, -ffunction-sections and -fdata-sections which u-boot is using correctly. But it seems that gcc puts all anonymous strings into the same section (.rodata or .rodata.str1.1 if string merging is on) which prevents them from participating in the stripping process. Interestingly, it puts its own __func__ strings into separate sections and they are eliminated if the function go away. It just doesn't do this for plain "" strings. Grrr. I was shocked to find gcc posts asking about this more than 13 years ago with barely any traction at all. Given that embarrassing history I'm not hopefully that the gcc folks will ever address this. Is there a work around I haven't thought of? I'm thinking along the lines of disabling all printfs in SPL in the hope that will take the strings away (since many are some sort of debug / progress message). Any thoughts or guidance would be appreciated. -Mike

3 3

[U-Boot] RAMBOOT
by Saridakis, Dean (US SSA) 12 Jul '13

12 Jul '13

I have a requirement that would be solved nicely by U-Boot booting U-Boot (and perhaps even another time). Is this a reasonable thing to do? I'm working with the mpc85xx CPU, P5020DS board and have gotten somewhat familiar with the design & flow running the P5020_NAND configuration (which is RAMBOOTed). I took a quick stab at build after reading README.ramboot-ppc85xx which failed -- think I've got config define dependencies to work through yet. Guess I'm somewhat skeptical this will work with all the TLB & LAW set-up that'd get redone along the way. Any advice would be appreciated. Thanks, Dean

1 0

[U-Boot] [PATCH] Fix memory stomper in DFU. Loop NULL-initted one past allocated array size.
by mboards＠prograde.net 12 Jul '13

12 Jul '13

From: Michael Cashwell <mboards(a)prograde.net> The memory layout arranged itself such that a long-standing memory stomper in a DFU prepare callback used during USB registration mangled the malloc heap enough to cause my board to panic much later in a call to free(). Since it hadn't happened before but was repeatable I decided to investigate before it vanished again. The actual stomp happened in this line after the for loop: f_dfu->function[i] = NULL; git blame says this code was introduced here: b819ddbf (Lukasz Majewski 2012-08-06 14:41:06 +0200 587) I'm not sure if the function[] array actually needs a NULL entry at the end. If so then this patch is the right fix. If it really always knows the last array index and doesn't need the NULL then removing the offending assignment would be better. Not knowing makes this patch safer. Signed-off-by: Michael Cashwell <mboards(a)prograde.net> --- drivers/usb/gadget/f_dfu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/drivers/usb/gadget/f_dfu.c b/drivers/usb/gadget/f_dfu.c index a322ae5..b24de09 100644 --- a/drivers/usb/gadget/f_dfu.c +++ b/drivers/usb/gadget/f_dfu.c @@ -589,7 +589,7 @@ static int dfu_prepare_function(struct f_dfu *f_dfu, int n) struct usb_interface_descriptor *d; int i = 0; - f_dfu->function = calloc(sizeof(struct usb_descriptor_header *), n); + f_dfu->function = calloc(sizeof(struct usb_descriptor_header *), n + 1); if (!f_dfu->function) goto enomem; -- 1.7.9.5

2 1

[U-Boot] [ANN] v2013.07 _not_ on Monday
by Tom Rini 12 Jul '13

12 Jul '13

Hey all, Looking things over, right now: - We've got one more bootm/related fixup posted today (I've confirmed it works) - The OMAP3 SPL issue (I need to confirm just how bad it is) - The exynos4 clock issue Given the above, and that I'm build testing a rather large set of changes right now, I'm _not_ going to do the next release on Monday. If all of my local tests pass, I will tag -rc3 today. Thanks! -- Tom

2 1

[U-Boot] regression:Exynos4: Clock calculation fix
by Lukasz Majewski 12 Jul '13

12 Jul '13

Hi Tom, I've checked out the origin/master HEAD: SHA1: 225fd8c5d4556547896a5d32ee092a258f3df638 It turns out that the Exynos4 Trats serial console output is broken because of APLL clock frequency change. Responsible change [*]: Exynos5: clock: Update the equation to calculate PLL output frequency SHA1: 234370cab4b2f096e095fe8f3284fd39740a4023 The above change is already at mainline (v2013.07-rc2) Minkyu has provided a correction patch [**]: [U-Boot] [PATCH] arm: exynos: fix clock calculation Unfortunately, this fixed the problem partially (and has been pulled by Albert already). I've just sent the fix for the Exynos4210 based boards [***]: http://patchwork.ozlabs.org/patch/258775/ [U-Boot] arm:exynos:fix: Fix clock calculation for Exynos4210 based targets. The solution was tested at Exynos4210 Trats board. I hope, that we will manage to fix things before final v2013.07 release. Clock change is crucial, and I'm afraid that Minkyu and Rajeshwari will read this mail on Monday. Can we wait for final release until then? For now, change [*] is at v2013.07, but fixes [**] and [***] aren't. I would opt for giving the interested party time to test all three patches together before we release the final v2013.07. -- Best regards, Lukasz Majewski Samsung R&D Institute Poland (SRPOL) | Linux Platform Group

1 0

[U-Boot] [PATCH v2 2/5] bootm: Disable interrupts only when loading
by Simon Glass 12 Jul '13

12 Jul '13

With the move of the interrupt code to earlier in the sequence, we exposed a problem where the interrupts are disabled at each bootm stage. This is not correct - it should be done only once. Let's disable interrupts in the LOAD stage. Put the code in a function for clarity. Also, bootz lost its interrupt code altogether, so reinstate it. Signed-off-by: Simon Glass <sjg(a)chromium.org> --- Changes in v2: None common/cmd_bootm.c | 67 +++++++++++++++++++++++++++++++++++------------------- 1 file changed, 43 insertions(+), 24 deletions(-) diff --git a/common/cmd_bootm.c b/common/cmd_bootm.c index 652513a..0c88be1 100644 --- a/common/cmd_bootm.c +++ b/common/cmd_bootm.c @@ -539,6 +539,42 @@ static int boot_selected_os(int argc, char * const argv[], int state, } /** + * bootm_disable_interrupts() - Disable interrupts in preparation for load/boot + * + * @return interrupt flag (0 if interrupts were disabled, non-zero if they were + * enabled) + */ +static ulong bootm_disable_interrupts(void) +{ + ulong iflag; + + /* + * We have reached the point of no return: we are going to + * overwrite all exception vector code, so we cannot easily + * recover from any failures any more... + */ + iflag = disable_interrupts(); +#ifdef CONFIG_NETCONSOLE + /* Stop the ethernet stack if NetConsole could have left it up */ + eth_halt(); +#endif + +#if defined(CONFIG_CMD_USB) + /* + * turn off USB to prevent the host controller from writing to the + * SDRAM while Linux is booting. This could happen (at least for OHCI + * controller), because the HCCA (Host Controller Communication Area) + * lies within the SDRAM and the host controller writes continously to + * this area (as busmaster!). The HccaFrameNumber is for example + * updated every 1 ms within the HCCA structure in SDRAM! For more + * details see the OpenHCI specification. + */ + usb_stop(); +#endif + return iflag; +} + +/** * Execute selected states of the bootm command. * * Note the arguments to this state must be the first argument, Any 'bootm' @@ -588,34 +624,11 @@ static int do_bootm_states(cmd_tbl_t *cmdtp, int flag, int argc, argc = 0; /* consume the args */ } - /* - * We have reached the point of no return: we are going to - * overwrite all exception vector code, so we cannot easily - * recover from any failures any more... - */ - iflag = disable_interrupts(); -#ifdef CONFIG_NETCONSOLE - /* Stop the ethernet stack if NetConsole could have left it up */ - eth_halt(); -#endif - -#if defined(CONFIG_CMD_USB) - /* - * turn off USB to prevent the host controller from writing to the - * SDRAM while Linux is booting. This could happen (at least for OHCI - * controller), because the HCCA (Host Controller Communication Area) - * lies within the SDRAM and the host controller writes continously to - * this area (as busmaster!). The HccaFrameNumber is for example - * updated every 1 ms within the HCCA structure in SDRAM! For more - * details see the OpenHCI specification. - */ - usb_stop(); -#endif - /* Load the OS */ if (!ret && (states & BOOTM_STATE_LOADOS)) { ulong load_end; + iflag = bootm_disable_interrupts(); ret = bootm_load_os(images, &load_end, 0); if (ret && ret != BOOTM_ERR_OVERLAP) goto err; @@ -1774,6 +1787,12 @@ int do_bootz(cmd_tbl_t *cmdtp, int flag, int argc, char * const argv[]) if (bootz_start(cmdtp, flag, argc, argv, &images)) return 1; + /* + * We are doing the BOOTM_STATE_LOADOS state ourselves, so must + * disable interrupts ourselves + */ + bootm_disable_interrupts(); + ret = do_bootm_states(cmdtp, flag, argc, argv, BOOTM_STATE_OS_FAKE_GO | BOOTM_STATE_OS_GO, &images, 1); -- 1.8.3

4 24

[U-Boot] [PULL] : Please pull u-boot-imx
by Stefano Babic 12 Jul '13

12 Jul '13

Hi Alber, please pull these fixes from u-boot-imx, thanks. The following changes since commit d6c6d127c5b948ec381fad5b24a2bc5497720644: vf610twr: Remove SoC name from U-Boot prompt (2013-06-28 16:59:08 +0200) are available in the git repository at: git://www.denx.de/git/u-boot-imx.git master for you to fetch changes up to ab94cd491faf3e7b0a3b934f5817b15997bcb315: net: fec: Avoid MX28 bus sync issue (2013-07-12 09:29:32 +0200) ---------------------------------------------------------------- Marek Vasut (3): m28evk: add trimffs to nand command net: fec: Remove bogus flush_dcache_range() call net: fec: Avoid MX28 bus sync issue drivers/net/fec_mxc.c | 23 ++++++++++++++++++++++- include/configs/m28evk.h | 1 + 2 files changed, 23 insertions(+), 1 deletion(-) -- ===================================================================== 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(a)denx.de =====================================================================

2 3