September 2009 - U-Boot

Re: [U-Boot] U-Boot praise
by Ben Warren 23 Sep '09

23 Sep '09

Jeff Palmer wrote: > Thank you for your work on the defragmenting udp packets and tftp > block size updates. They are working well. > > It took us a little time to compile/test/update everything to the > newer version of u-boot but I'm glad we did and the large tftp > tranfers to africa are running much better. > It's always great to hear about success stories like this. Thanks and congratulations! I really had very little to do with this feature. It was all Alessandro and Robin, so kudos to them! > Thanks again. > > PS. We are looking for a new software/hardware engineer to bring on > board in our Florida office so if you know of any guru's looking for > an opportunity, let me know. > -- > ------------------------------------------------------------------------ > *Jeffery Palmer > **Head Project Development* > *jpalmer(a)gsat.us** > (m) +1.954.709.7232 > **www.gsat.us* > > > * > * regards, Ben

1 0

[U-Boot] [PATCH 0/7] ppc/p4080: infrastructure patches
by Kumar Gala 23 Sep '09

23 Sep '09

This patch series includes the underpinings for the p4080 support. Its mostly focussed on the differenes in cpu/mpc85xx code between a PQ3/85xx platform and the p4080/CoreNet platforms.

4 19

[U-Boot] Blackfin: UART on SPORT
by Andreas Pretzsch 23 Sep '09

23 Sep '09

In Linux, there is support for a UART on a SPORT, see linux-2.6.x/drivers/serial/bfin_sport_uart.[ch]. It can also be used as console. In U-Boot, I cannot find similar code, therefore I'd like to ask if anybody ported/implemented this for U-Boot already ? If not, is there any objection not to port this code to U-Boot, given that someone (which might be me) has some spare time for this ? Thanks in advance, A. Pretzsch -- carpe noctem engineering Ingenieurbuero fuer Hard- & Software-Entwicklung Andreas Pretzsch Dipl.-Ing. (FH) Andreas Pretzsch Tel. 0731/98588-800 Marlene-Dietrich-Strasse 5 Fax: 0731/98588-801 89231 Neu-Ulm, Germany email: apr(a)cn-eng.de

2 1

[U-Boot] tftp packet failure counter reset
by Jeffery Palmer 23 Sep '09

23 Sep '09

I do large transfers via tftp, and since the timeout counter never resets, they often fail since the failures are counted throughout the entire transfer. By resetting the counter to 0 on a successful packet, this issue is fixed tftp.c: } else { if (((TftpBlock - 1) % 10) == 0) { putc ('#'); } else if ((TftpBlock % (10 * HASHES_PER_LINE)) == 0) { puts ("\n\t "); } + //Reset timeout count since we received a good packet + TftpTimeoutCount = 0; } if (TftpState == STATE_RRQ) _________________________________________________________________ Hotmail: Free, trusted and rich email service. http://clk.atdmt.com/GBL/go/171222984/direct/01/

3 3

[U-Boot] [PATCH] ubifs: Add support for looking up directory and relative symlinks
by Simon Kagstrom 23 Sep '09

23 Sep '09

ubifs: Add support for looking up directory and relative symlinks This patch adds support for resolving symlinks to directories as well as relative symlinks. Symlinks are now always resolved during file lookup, so the load stage no longer needs to special-case them. Signed-off-by: Simon Kagstrom <simon.kagstrom(a)netinsight.net> --- fs/ubifs/ubifs.c | 68 ++++++++++++++++++++++++++++++++++------------------- 1 files changed, 43 insertions(+), 25 deletions(-) diff --git a/fs/ubifs/ubifs.c b/fs/ubifs/ubifs.c index 8ede188..b03ed0b 100644 --- a/fs/ubifs/ubifs.c +++ b/fs/ubifs/ubifs.c @@ -379,9 +379,11 @@ static unsigned long ubifs_findfile(struct super_block *sb, char *filename) int ret; char *next; char fpath[128]; + char symlinkpath[128]; char *name = fpath; unsigned long root_inum = 1; unsigned long inum; + int symlink_count = 0; /* Don't allow symlink recursion */ strcpy(fpath, filename); @@ -397,6 +399,9 @@ static unsigned long ubifs_findfile(struct super_block *sb, char *filename) return inum; for (;;) { + struct inode *inode; + struct ubifs_inode *ui; + /* Extract the actual part from the pathname. */ next = strchr(name, '/'); if (next) { @@ -406,18 +411,48 @@ static unsigned long ubifs_findfile(struct super_block *sb, char *filename) } ret = ubifs_finddir(sb, name, root_inum, &inum); + if (!ret) + return 0; + inode = ubifs_iget(sb, inum); + + if (!inode) + return 0; + ui = ubifs_inode(inode); + + if ((inode->i_mode & S_IFMT) == S_IFLNK) { + char link_name[64]; + char buf[128]; + + /* We have some sort of symlink recursion, bail out */ + if (symlink_count++ > 8) { + printf("Symlink recursion, aborting\n"); + return 0; + } + memcpy(link_name, ui->data, ui->data_len); + link_name[ui->data_len] = '\0'; + + if (link_name[0] == '/') { + /* Absolute path, redo everything without + * the leading slash */ + next = name = link_name + 1; + root_inum = 1; + continue; + } + /* Relative to cur dir */ + sprintf(buf, "%s%s", + link_name, next == NULL ? "" : next); + memcpy(symlinkpath, buf, sizeof(buf)); + next = name = symlinkpath; + continue; + } /* * Check if directory with this name exists */ /* Found the node! */ - if (!next || *next == '\0') { - if (ret) - return inum; - - break; - } + if (!next || *next == '\0') + return inum; root_inum = inum; name = next; @@ -614,10 +649,10 @@ int ubifs_load(char *filename, u32 addr, u32 size) int err = 0; int i; int count; - char link_name[64]; - struct ubifs_inode *ui; c->ubi = ubi_open_volume(c->vi.ubi_num, c->vi.vol_id, UBI_READONLY); + /* ubifs_findfile will resolve symlinks, so we know that we get + * the real file here */ inum = ubifs_findfile(ubifs_sb, filename); if (!inum) { err = -1; @@ -635,23 +670,6 @@ int ubifs_load(char *filename, u32 addr, u32 size) } /* - * Check for symbolic link - */ - ui = ubifs_inode(inode); - if (((inode->i_mode & S_IFMT) == S_IFLNK) && ui->data_len) { - memcpy(link_name, ui->data, ui->data_len); - link_name[ui->data_len] = '\0'; - printf("%s is linked to %s!\n", filename, link_name); - ubifs_iput(inode); - - /* - * Now we have the "real" filename, call ubifs_load() - * again (recursive call) to load this file instead - */ - return ubifs_load(link_name, addr, size); - } - - /* * If no size was specified or if size bigger than filesize * set size to filesize */ -- 1.6.0.4

2 1

[U-Boot] [PATCH] ppc4xx: Fix PCIE PLL lock on 440SPe Yucca board
by Rupjyoti Sarmah 23 Sep '09

23 Sep '09

u-boot reports a PCIE PLL lock error at boot time on Yucca board, and left PCIe nonfunctional. This is fixed by making u-boot function ppc4xx_init_pcie() to wait 300 uS after negating reset before the first check of the PLL lock. Signed-off-by: Rupjyoti Sarmah <rsarmah(a)amcc.com> --- diff --git a/cpu/ppc4xx/4xx_pcie.c b/cpu/ppc4xx/4xx_pcie.c index 07fbb0e..78fdb4e 100644 --- a/cpu/ppc4xx/4xx_pcie.c +++ b/cpu/ppc4xx/4xx_pcie.c @@ -374,28 +374,35 @@ int ppc4xx_init_pcie(void) /* Set PLL clock receiver to LVPECL */ SDR_WRITE(PESDR0_PLLLCT1, SDR_READ(PESDR0_PLLLCT1) | 1 << 28); - if (check_error()) + if (check_error()) { + printf("ERROR: failed to set PCIe reference clock receiver --" + "PESDR0_PLLLCT1 = 0x%08x\n", SDR_READ(PESDR0_PLLLCT1)); + return -1; + } + + /* Did resistance calibration work? */ + if (!(SDR_READ(PESDR0_PLLLCT2) & 0x10000)) { + printf("ERROR: PCIe resistance calibration failed --" + "PESDR0_PLLLCT2 = 0x%08x\n", SDR_READ(PESDR0_PLLLCT2)); - if (!(SDR_READ(PESDR0_PLLLCT2) & 0x10000)) - { - printf("PCIE: PESDR_PLLCT2 resistance calibration failed (0x%08x)\n", - SDR_READ(PESDR0_PLLLCT2)); return -1; } /* De-assert reset of PCIe PLL, wait for lock */ SDR_WRITE(PESDR0_PLLLCT1, SDR_READ(PESDR0_PLLLCT1) & ~(1 << 24)); - udelay(3); + udelay(300); /* 300 uS is maximum time lock should take */ while (time_out) { if (!(SDR_READ(PESDR0_PLLLCT3) & 0x10000000)) { time_out--; - udelay(1); + udelay(20); /* Wait 20 uS more if needed */ } else break; } if (!time_out) { - printf("PCIE: VCO output not locked\n"); + printf("ERROR: PCIe PLL VCO output not locked to ref clock --" + "PESDR0_PLLLCTS=0x%08x\n", SDR_READ(PESDR0_PLLLCT3)); + return -1; } return 0; -- 1.5.6.3

2 1

[U-Boot] [PATCH] ppc4xx: Make DDR2 timing for intip more robust
by Dirk Eibach 23 Sep '09

23 Sep '09

DDR2 timing for intip was on the edge for some of the available chips for this board. Now it is verfied to work with all of them. Signed-off-by: Dirk Eibach <eibach(a)gdsys.de> --- include/configs/intip.h | 20 ++++++++++---------- 1 files changed, 10 insertions(+), 10 deletions(-) diff --git a/include/configs/intip.h b/include/configs/intip.h index 4f7bc7e..19f12fa 100644 --- a/include/configs/intip.h +++ b/include/configs/intip.h @@ -172,7 +172,7 @@ #define CONFIG_SYS_SDRAM_R3BAS 0x00000000 #define CONFIG_SYS_SDRAM_PLBADDULL 0x00000000 #define CONFIG_SYS_SDRAM_PLBADDUHB 0x00000008 -#define CONFIG_SYS_SDRAM_CONF1LL 0x80001C80 +#define CONFIG_SYS_SDRAM_CONF1LL 0x80001C00 #define CONFIG_SYS_SDRAM_CONF1HB 0x80001C80 #define CONFIG_SYS_SDRAM_CONFPATHB 0x10a68000 @@ -181,7 +181,7 @@ #define CONFIG_SYS_SDRAM0_MB1CF 0x00000000 #define CONFIG_SYS_SDRAM0_MB2CF 0x00000000 #define CONFIG_SYS_SDRAM0_MB3CF 0x00000000 -#define CONFIG_SYS_SDRAM0_MCOPT1 0x05122000 +#define CONFIG_SYS_SDRAM0_MCOPT1 0x05120000 #define CONFIG_SYS_SDRAM0_MCOPT2 0x00000000 #define CONFIG_SYS_SDRAM0_MODT0 0x00000000 #define CONFIG_SYS_SDRAM0_MODT1 0x00000000 @@ -193,7 +193,7 @@ #define CONFIG_SYS_SDRAM0_INITPLR1 0x81900400 #define CONFIG_SYS_SDRAM0_INITPLR2 0x81020000 #define CONFIG_SYS_SDRAM0_INITPLR3 0x81030000 -#define CONFIG_SYS_SDRAM0_INITPLR4 0x81010000 +#define CONFIG_SYS_SDRAM0_INITPLR4 0x81010002 #define CONFIG_SYS_SDRAM0_INITPLR5 0xE4000542 #define CONFIG_SYS_SDRAM0_INITPLR6 0x81900400 #define CONFIG_SYS_SDRAM0_INITPLR7 0x8A880000 @@ -201,21 +201,21 @@ #define CONFIG_SYS_SDRAM0_INITPLR9 0x8A880000 #define CONFIG_SYS_SDRAM0_INITPLR10 0x8A880000 #define CONFIG_SYS_SDRAM0_INITPLR11 0x81000442 -#define CONFIG_SYS_SDRAM0_INITPLR12 0x81010380 -#define CONFIG_SYS_SDRAM0_INITPLR13 0x81010000 +#define CONFIG_SYS_SDRAM0_INITPLR12 0x81010382 +#define CONFIG_SYS_SDRAM0_INITPLR13 0x81010002 #define CONFIG_SYS_SDRAM0_INITPLR14 0x00000000 #define CONFIG_SYS_SDRAM0_INITPLR15 0x00000000 #define CONFIG_SYS_SDRAM0_RQDC 0x80000038 -#define CONFIG_SYS_SDRAM0_RFDC 0x003F0000 -#define CONFIG_SYS_SDRAM0_RDCC 0x80000000 +#define CONFIG_SYS_SDRAM0_RFDC 0x00000257 +#define CONFIG_SYS_SDRAM0_RDCC 0x40000000 #define CONFIG_SYS_SDRAM0_DLCR 0x00000000 #define CONFIG_SYS_SDRAM0_CLKTR 0x40000000 -#define CONFIG_SYS_SDRAM0_WRDTR 0x84000800 +#define CONFIG_SYS_SDRAM0_WRDTR 0x84000823 #define CONFIG_SYS_SDRAM0_SDTR1 0x80201000 #define CONFIG_SYS_SDRAM0_SDTR2 0x32204232 -#define CONFIG_SYS_SDRAM0_SDTR3 0x090B0D15 +#define CONFIG_SYS_SDRAM0_SDTR3 0x090C0D15 #define CONFIG_SYS_SDRAM0_MMODE 0x00000442 -#define CONFIG_SYS_SDRAM0_MEMODE 0x00000000 +#define CONFIG_SYS_SDRAM0_MEMODE 0x00000002 #define CONFIG_SYS_MBYTES_SDRAM 256 /* 256MB */ -- 1.5.6.5

2 1

[U-Boot] [PATCH 1/1] Fix bug in jumptable call stubs for SPARC.
by Daniel Hellstrom 23 Sep '09

23 Sep '09

From: Sergey Mironov <ierton(a)gmail.com> Hello Wolfgang, Sergey has resent the patch with a sign-off and I have updated the u-boot-sparc.git repository with the new patch, for SPARC standalone application support. Thank you, Daniel Hellstrom Signed-off-by: Sergey Mironov <ierton(a)gmail.com> Signed-off-by: Daniel Hellstrom <daniel(a)gaisler.com> --- examples/standalone/stubs.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/examples/standalone/stubs.c b/examples/standalone/stubs.c index 339bbf9..ce3371d 100644 --- a/examples/standalone/stubs.c +++ b/examples/standalone/stubs.c @@ -181,9 +181,9 @@ gd_t *global_data; " or %%g1, %%g7, %%g1\n" \ " ld [%%g1], %%g1\n" \ " ld [%%g1 + %1], %%g1\n" \ -" call %%g1\n" \ +" jmp %%g1\n" \ " nop\n" \ - : : "i"(offsetof(gd_t, jt)), "i"(XF_ ## x) : "g1" ); + : : "i"(offsetof(gd_t, jt)), "i"(XF_ ## x * sizeof(void *)) : "g1" ); #else #error stubs definition missing for this architecture -- 1.5.4

1 0

[U-Boot] [PATCH] vxworks: Add CONFIG_VXWORKS_PREBOOT
by Niklaus Giger 23 Sep '09

23 Sep '09

The option CONFIG_VXWORKS_PREBOOT allows a board specific vxworks_preboot to be run just before jumping into the vxWorks images. This can be used to alter a register which is used differently by U-boot and vxWorks. Signed-off-by: Niklaus Giger <niklaus.giger(a)netstal.com> --- common/cmd_elf.c | 4 ++++ include/vxworks.h | 4 ++++ 2 files changed, 8 insertions(+), 0 deletions(-) diff --git a/common/cmd_elf.c b/common/cmd_elf.c index bf7dd63..4e36680 100644 --- a/common/cmd_elf.c +++ b/common/cmd_elf.c @@ -213,6 +213,10 @@ int do_bootvx (cmd_tbl_t *cmdtp, int flag, int argc, char *argv[]) (char *) bootaddr); printf ("## Starting vxWorks at 0x%08lx ...\n", addr); +#ifdef CONFIG_VXWORKS_PREBOOT + vxworks_preboot(); +#endif + ((void (*)(void)) addr) (); puts ("## vxWorks terminated\n"); diff --git a/include/vxworks.h b/include/vxworks.h index 1633904..df2b580 100644 --- a/include/vxworks.h +++ b/include/vxworks.h @@ -50,4 +50,8 @@ int do_bootvx(cmd_tbl_t *cmdtp, int flag, int argc, char *argv[]); #define CONFIG_SYS_VXWORKS_SERVERNAME "srv" #endif +#ifdef CONFIG_VXWORKS_PREBOOT + void vxworks_preboot(void); +#endif + #endif -- 1.6.3.3

5 5

Re: [U-Boot] [PATCH] SPARC standalone app fix
by Сергей Миронов 23 Sep '09

23 Sep '09

Hi Daniel. Here is result of command git format-patch -s 22c3885a59674d078dbd18e1e32702c4b27ec582^ -- Thanks, Sergey. >From 22c3885a59674d078dbd18e1e32702c4b27ec582 Mon Sep 17 00:00:00 2001 From: Sergey Mironov <smironov(a)module.ru> Date: Wed, 16 Sep 2009 10:55:21 +0400 Subject: [PATCH] Fix bug in jumptable call stubs for SPARC. Seems, that i am the first who tested it. Signed-off-by: Sergey Mironov <ierton(a)gmail.com> --- examples/standalone/stubs.c | 4 ++-- 1 files changed, 2 insertions(+), 2 deletions(-) diff --git a/examples/standalone/stubs.c b/examples/standalone/stubs.c index a8cb954..08c4017 100644 --- a/examples/standalone/stubs.c +++ b/examples/standalone/stubs.c @@ -181,9 +181,9 @@ gd_t *global_data; " or %%g1, %%g7, %%g1\n" \ " ld [%%g1], %%g1\n" \ " ld [%%g1 + %1], %%g1\n" \ -" call %%g1\n" \ +" jmp %%g1\n" \ " nop\n" \ - : : "i"(offsetof(gd_t, jt)), "i"(XF_ ## x) : "g1" ); + : : "i"(offsetof(gd_t, jt)), "i"(XF_ ## x * sizeof(void *)) : "g1" ); #else #error stubs definition missing for this architecture -- 1.6.0.3 2009/9/23 Daniel Hellstrom <daniel(a)gaisler.com>: > Hi Sergey, > > Can you resend your patch, this time signed off (the -s options to > git-commit) so that your name will propagate correctly through the git > repos. > > Thanks, > Daniel Hellstrom > > > Daniel Hellstrom wrote: > >> Hi Sergey, >> >> Thank you for your work. I have updated the sparc repository with your >> patch. And, yes, you are probably the first one to use the standalone app >> feature :) >> >> Please CC me next time. >> >> Thanks, >> Daniel >> >> Сергей Миронов wrote: >> >> >>> >>> Hi! Thanks for the beautiful project! Ive tried to launch standalone >>> app 'hello_world' on my SPARC leon3 board (ml-402) and found that >>> application says 'boom' and halts. So here is the patch to solve the >>> problem. Seems, that i am the first who ever launched Sparc'ish >>> standalone code :) >>> >>> >>> >>> >> >> _______________________________________________ >> U-Boot mailing list >> U-Boot(a)lists.denx.de >> http://lists.denx.de/mailman/listinfo/u-boot >> >> >> > > -- С Уважением, Сергей.

1 0