February 2013 - flashrom - mail.coreboot.org

Add timeout check to SPI bitbang bus request
by Carl-Daniel Hailfinger April 29, 2014

April 29, 2014

Hi Iain, your Intel 82540EM which hangs on a flashrom run should spew "Timeout!" messages with this patch. It would be nice if you could tell us how long it waits for each chip probe so I can adjust the timeout value. Please note that this patch will _not_ fix the underlying issue, but I believe that an endless loop on possibly broken hardware is worse than aborting. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Index: flashrom-bitbang_timeout/flash.h =================================================================== --- flashrom-bitbang_timeout/flash.h (Revision 1185) +++ flashrom-bitbang_timeout/flash.h (Arbeitskopie) @@ -35,6 +35,9 @@ #define ERROR_PTR ((void*)-1) +/* Error codes */ +#define TIMEOUT_ERROR -101 + typedef unsigned long chipaddr; int register_shutdown(void (*function) (void *data), void *data); Index: flashrom-bitbang_timeout/bitbang_spi.c =================================================================== --- flashrom-bitbang_timeout/bitbang_spi.c (Revision 1185) +++ flashrom-bitbang_timeout/bitbang_spi.c (Arbeitskopie) @@ -53,10 +53,11 @@ return bitbang_spi_master->get_miso(); } -static void bitbang_spi_request_bus(void) +static int bitbang_spi_request_bus(void) { if (bitbang_spi_master->request_bus) - bitbang_spi_master->request_bus(); + return bitbang_spi_master->request_bus(); + return 0; } static void bitbang_spi_release_bus(void) @@ -86,7 +87,7 @@ bitbang_spi_master = master; bitbang_spi_half_period = halfperiod; - /* FIXME: Run bitbang_spi_request_bus here or in programmer init? */ + /* FIXME: Run bitbang_spi_request_bus here or per command? */ bitbang_spi_set_cs(1); bitbang_spi_set_sck(0); bitbang_spi_set_mosi(0); @@ -131,13 +132,15 @@ int bitbang_spi_send_command(unsigned int writecnt, unsigned int readcnt, const unsigned char *writearr, unsigned char *readarr) { - int i; + int i, ret; /* FIXME: Run bitbang_spi_request_bus here or in programmer init? * Requesting and releasing the SPI bus is handled in here to allow the * programmer to use its own SPI engine for native accesses. */ - bitbang_spi_request_bus(); + ret = bitbang_spi_request_bus(); + if (ret) + return ret; bitbang_spi_set_cs(0); for (i = 0; i < writecnt; i++) bitbang_spi_readwrite_byte(writearr[i]); Index: flashrom-bitbang_timeout/nicintel_spi.c =================================================================== --- flashrom-bitbang_timeout/nicintel_spi.c (Revision 1185) +++ flashrom-bitbang_timeout/nicintel_spi.c (Arbeitskopie) @@ -58,6 +58,8 @@ // #define FL_BUSY 30 // #define FL_ER 31 +#define MAX_REQUEST_LOOPS 10000 + uint8_t *nicintel_spibar; const struct pcidev_status nics_intel_spi[] = { @@ -68,16 +70,27 @@ {}, }; -static void nicintel_request_spibus(void) +static void nicintel_release_spibus(void); + +static int nicintel_request_spibus(void) { uint32_t tmp; + int i = 0; tmp = pci_mmio_readl(nicintel_spibar + FLA); tmp |= 1 << FL_REQ; pci_mmio_writel(tmp, nicintel_spibar + FLA); /* Wait until we are allowed to use the SPI bus. */ - while (!(pci_mmio_readl(nicintel_spibar + FLA) & (1 << FL_GNT))) ; + while (!(pci_mmio_readl(nicintel_spibar + FLA) & (1 << FL_GNT))) { + if (i++ > MAX_REQUEST_LOOPS) { + nicintel_release_spibus(); + msg_perr("%s: Timeout! Aborting.\n", __func__); + return TIMEOUT_ERROR; + } + } + + return 0; } static void nicintel_release_spibus(void) Index: flashrom-bitbang_timeout/mcp6x_spi.c =================================================================== --- flashrom-bitbang_timeout/mcp6x_spi.c (Revision 1185) +++ flashrom-bitbang_timeout/mcp6x_spi.c (Arbeitskopie) @@ -40,22 +40,37 @@ #define MCP6X_SPI_REQUEST 0 #define MCP6X_SPI_GRANT 8 +#define MAX_REQUEST_LOOPS 10000 + void *mcp6x_spibar = NULL; /* Cached value of last GPIO state. */ static uint8_t mcp_gpiostate; -static void mcp6x_request_spibus(void) +static void mcp6x_release_spibus(void); + +static int mcp6x_request_spibus(void) { + int i = 0; + mcp_gpiostate = mmio_readb(mcp6x_spibar + 0x530); mcp_gpiostate |= 1 << MCP6X_SPI_REQUEST; mmio_writeb(mcp_gpiostate, mcp6x_spibar + 0x530); /* Wait until we are allowed to use the SPI bus. */ - while (!(mmio_readw(mcp6x_spibar + 0x530) & (1 << MCP6X_SPI_GRANT))) ; + while (!(mmio_readw(mcp6x_spibar + 0x530) & (1 << MCP6X_SPI_GRANT))) { + if (i++ > MAX_REQUEST_LOOPS) { + /* Update the cache. */ + mcp_gpiostate = mmio_readb(mcp6x_spibar + 0x530); + mcp6x_release_spibus(); + msg_perr("%s: Timeout! Aborting.\n", __func__); + return TIMEOUT_ERROR; + } + } /* Update the cache. */ mcp_gpiostate = mmio_readb(mcp6x_spibar + 0x530); + return 0; } static void mcp6x_release_spibus(void) Index: flashrom-bitbang_timeout/programmer.h =================================================================== --- flashrom-bitbang_timeout/programmer.h (Revision 1185) +++ flashrom-bitbang_timeout/programmer.h (Arbeitskopie) @@ -131,7 +131,7 @@ void (*set_sck) (int val); void (*set_mosi) (int val); int (*get_miso) (void); - void (*request_bus) (void); + int (*request_bus) (void); void (*release_bus) (void); }; -- http://www.hailfinger.org/

4 15

Flashrom stuck using 100% cpu while probing for a chip
by GatoLoko April 28, 2014

April 28, 2014

I've tried to run flashrom -V to see if it could identify the chip on my new board (AsRock M3A-GLAN), and it got stuck while probing for the first chip. I've tried commenting out that specific chip (AMIC A25L05PT) and then it gets stuck on the next one. Then commented out all AMIC chips, and it got stuck on the next one (the first ATMEL on the list) I've tried with an older flashrom (0.9.2-r1141). It skipped some AMD chips, then got stuck on AMIC again. Talked with "tlt" and "stefanct" on irc and they asked for a strace and a gdb backtrace. Strace produced no results. Gdb log is attached. "stefanct" tried and reproduced it on a machine with nicintel_spi, but warned it was with an unsupported nic: http://paste.flashrom.org/view.php?id=738 -- GatoLoko

4 11

Large FWH support
by Kyösti Mälkki Sept. 14, 2013

Sept. 14, 2013

Hi I came across two problems using 2 MiB FWH flash on two different boards with old ICH4 and ICH6 chipsets: First, ICH3/4/5 do not implement fwh_idsel= parameter. If vendor bios had the default map of 1 MiB, it did not even detect 2 MiB flash chips. For this I have patches prepared already. Before: (idsel_failure.txt) Probing for SST SST49LF008C, 1024 kB: probe_82802ab: id1 0xbf, id2 0x5c Probing for SST SST49LF016C, 2048 kB: probe_82802ab: id1 0xff, id2 0xff, id1 parity violation, id1 is normal flash content, id2 is normal flash content After: (idsel_ok.txt) Probing for SST SST49LF008C, 1024 kB: probe_82802ab: id1 0xbf, id2 0x5c Probing for SST SST49LF016C, 2048 kB: probe_82802ab: id1 0xbf, id2 0x5c Found SST flash chip "SST49LF016C" (2048 kB, FWH) at physical address 0xffe00000. Second, a board with ICH6 had all except one 512 kiB range of flash disabled by vendor bios and it would not detect 1MiB and 2MiB flash chips. NOTE: the board actually had 512kiB flash for these logs, but you can see where ther problem is. Before: (decode-disabled.txt) 0xfff80000/0xffb80000 FWH decode enabled 0xfff00000/0xffb00000 FWH decode disabled 0xffe80000/0xffa80000 FWH decode disabled 0xffe00000/0xffa00000 FWH decode disabled ... Maximum FWH chip size: 0x80000 bytes ... Probing for SST SST49LF008C, 1024 kB: Chip size 1024 kB is bigger than supported size 512 kB of chipset/board/programmer for FWH interface, probe/read/erase/write may fail. probe_82802ab: id1 0xff, id2 0xff, id1 parity violation, id1 is normal flash content, id2 is normal flash content I used the following as a work-around for ICH6, but this should be derived from fwh_idsel= parameter: $ setpci -s 0:1f.0 0xd8.W=0xf0c0 After: (decode-enabled.txt) 0xfff80000/0xffb80000 FWH decode enabled 0xfff00000/0xffb00000 FWH decode enabled 0xffe80000/0xffa80000 FWH decode enabled 0xffe00000/0xffa00000 FWH decode enabled Maximum FWH chip size: 0x200000 bytes Regards, Kyösti Mälkki

2 7

[PATCH] CONFIG_DEFAULT setting
by Carl-Daniel Hailfinger Sept. 14, 2013

Sept. 14, 2013

Allow easy disabling of all programmers except selected ones by setting CONFIG_DEFAULT=no. Setting CONFIG_DEFAULT=yes has no effect. Usage example: # make CONFIG_DUMMY=yes CONFIG_NIC3COM=yes CONFIG_DEFAULT=no This will disable all programmers except dummy and nic3com. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Index: flashrom-Makefile_config_default/Makefile =================================================================== --- flashrom-Makefile_config_default/Makefile (Revision 1623) +++ flashrom-Makefile_config_default/Makefile (Arbeitskopie) @@ -52,6 +52,10 @@ # Compilation will fail for unspecified values. CONFIG_DEFAULT_PROGRAMMER ?= PROGRAMMER_INVALID +# This parameter picks the default setting for CONFIG_ variables of programmers not specified explicitly +# on the command line. +CONFIG_DEFAULT ?= yes + # If your compiler spits out excessive warnings, run make WARNERROR=no # You shouldn't have to change this flag. WARNERROR ?= yes @@ -322,65 +326,65 @@ SVNDEF := -D'FLASHROM_VERSION="$(VERSION)"' # Always enable internal/onboard support for now. -CONFIG_INTERNAL ?= yes +CONFIG_INTERNAL ?= $(CONFIG_DEFAULT) # Always enable serprog for now. Needs to be disabled on Windows. -CONFIG_SERPROG ?= yes +CONFIG_SERPROG ?= $(CONFIG_DEFAULT) # RayeR SPIPGM hardware support -CONFIG_RAYER_SPI ?= yes +CONFIG_RAYER_SPI ?= $(CONFIG_DEFAULT) # PonyProg2000 SPI hardware support -CONFIG_PONY_SPI ?= yes +CONFIG_PONY_SPI ?= $(CONFIG_DEFAULT) # Always enable 3Com NICs for now. -CONFIG_NIC3COM ?= yes +CONFIG_NIC3COM ?= $(CONFIG_DEFAULT) # Enable NVIDIA graphics cards. Note: write and erase do not work properly. -CONFIG_GFXNVIDIA ?= yes +CONFIG_GFXNVIDIA ?= $(CONFIG_DEFAULT) # Always enable SiI SATA controllers for now. -CONFIG_SATASII ?= yes +CONFIG_SATASII ?= $(CONFIG_DEFAULT) # Highpoint (HPT) ATA/RAID controller support. # IMPORTANT: This code is not yet working! CONFIG_ATAHPT ?= no # Always enable FT2232 SPI dongles for now. -CONFIG_FT2232_SPI ?= yes +CONFIG_FT2232_SPI ?= $(CONFIG_DEFAULT) # Always enable dummy tracing for now. -CONFIG_DUMMY ?= yes +CONFIG_DUMMY ?= $(CONFIG_DEFAULT) # Always enable Dr. Kaiser for now. -CONFIG_DRKAISER ?= yes +CONFIG_DRKAISER ?= $(CONFIG_DEFAULT) # Always enable Realtek NICs for now. -CONFIG_NICREALTEK ?= yes +CONFIG_NICREALTEK ?= $(CONFIG_DEFAULT) # Disable National Semiconductor NICs until support is complete and tested. CONFIG_NICNATSEMI ?= no # Always enable Intel NICs for now. -CONFIG_NICINTEL ?= yes +CONFIG_NICINTEL ?= $(CONFIG_DEFAULT) # Always enable SPI on Intel NICs for now. -CONFIG_NICINTEL_SPI ?= yes +CONFIG_NICINTEL_SPI ?= $(CONFIG_DEFAULT) # Always enable SPI on OGP cards for now. -CONFIG_OGP_SPI ?= yes +CONFIG_OGP_SPI ?= $(CONFIG_DEFAULT) # Always enable Bus Pirate SPI for now. -CONFIG_BUSPIRATE_SPI ?= yes +CONFIG_BUSPIRATE_SPI ?= $(CONFIG_DEFAULT) # Disable Dediprog SF100 until support is complete and tested. CONFIG_DEDIPROG ?= no # Always enable Marvell SATA controllers for now. -CONFIG_SATAMV ?= yes +CONFIG_SATAMV ?= $(CONFIG_DEFAULT) # Enable Linux spidev interface by default. We disable it on non-Linux targets. -CONFIG_LINUX_SPI ?= yes +CONFIG_LINUX_SPI ?= $(CONFIG_DEFAULT) # Disable wiki printing by default. It is only useful if you have wiki access. CONFIG_PRINT_WIKI ?= no -- http://www.hailfinger.org/

2 2

[PATCH] Remove exit call from physmap_common
by Niklas Söderlund Sept. 14, 2013

Sept. 14, 2013

Only call path where exit is reached is from physmap(). Callers of physmap() that where not prepared to handle the ERROR_PTR return value have been handled. Signed-off-by: Niklas Söderlund <niso(a)kth.se> --- chipset_enable.c | 6 ++++++ drkaiser.c | 2 ++ gfxnvidia.c | 2 ++ ichspi.c | 2 ++ it85spi.c | 2 ++ mcp6x_spi.c | 2 ++ nicintel_spi.c | 2 ++ ogp_spi.c | 2 ++ physmap.c | 12 +++--------- satasii.c | 2 ++ sb600spi.c | 2 ++ 11 files changed, 27 insertions(+), 9 deletions(-) diff --git a/chipset_enable.c b/chipset_enable.c index fa79511..b1738ea 100644 --- a/chipset_enable.c +++ b/chipset_enable.c @@ -494,6 +494,8 @@ static int enable_flash_tunnelcreek(struct pci_dev *dev, const char *name) /* Map RCBA to virtual memory */ rcrb = physmap("ICH RCRB", tmp, 0x4000); + if (rcrb == ERROR_PTR) + return ERROR_FATAL; /* Test Boot BIOS Strap Status */ bnt = mmio_readl(rcrb + 0x3410); @@ -565,6 +567,8 @@ static int enable_flash_ich_dc_spi(struct pci_dev *dev, const char *name, /* Map RCBA to virtual memory */ rcrb = physmap("ICH RCRB", tmp, 0x4000); + if (rcrb == ERROR_PTR) + return ERROR_FATAL; gcs = mmio_readl(rcrb + 0x3410); msg_pdbg("GCS = 0x%x: ", gcs); @@ -1235,6 +1239,8 @@ static int get_flashbase_sc520(struct pci_dev *dev, const char *name) /* 1. Map MMCR */ mmcr = physmap("Elan SC520 MMCR", 0xfffef000, getpagesize()); + if (mmcr == ERROR_PTR) + return -1; /* 2. Scan PAR0 (0x88) - PAR15 (0xc4) for * BOOTCS region (PARx[31:29] = 100b)e diff --git a/drkaiser.c b/drkaiser.c index 83b8b3d..71981e8 100644 --- a/drkaiser.c +++ b/drkaiser.c @@ -82,6 +82,8 @@ int drkaiser_init(void) /* Map 128kB flash memory window. */ drkaiser_bar = physmap("Dr. Kaiser PC-Waechter flash memory", addr, DRKAISER_MEMMAP_SIZE); + if (drkaiser_bar == ERROR_PTR) + return 1; if (register_shutdown(drkaiser_shutdown, NULL)) return 1; diff --git a/gfxnvidia.c b/gfxnvidia.c index 9232677..9600675 100644 --- a/gfxnvidia.c +++ b/gfxnvidia.c @@ -102,6 +102,8 @@ int gfxnvidia_init(void) msg_pinfo("Detected NVIDIA I/O base address: 0x%x.\n", io_base_addr); nvidia_bar = physmap("NVIDIA", io_base_addr, GFXNVIDIA_MEMMAP_SIZE); + if (nvidia_bar == ERROR_PTR) + return 1; /* Must be done before rpci calls. */ if (register_shutdown(gfxnvidia_shutdown, NULL)) diff --git a/ichspi.c b/ichspi.c index bbe2b06..f6b77ad 100644 --- a/ichspi.c +++ b/ichspi.c @@ -1836,6 +1836,8 @@ int via_init_spi(struct pci_dev *dev, uint32_t mmio_base) int i; ich_spibar = physmap("VIA SPI MMIO registers", mmio_base, 0x70); + if (ich_spibar == ERROR_PTR) + return ERROR_FATAL; /* Do we really need no write enable? Like the LPC one at D17F0 0x40 */ /* Not sure if it speaks all these bus protocols. */ diff --git a/it85spi.c b/it85spi.c index 0b074eb..e15a287 100644 --- a/it85spi.c +++ b/it85spi.c @@ -262,6 +262,8 @@ static int it85xx_spi_common_init(struct superio s) * Major TODO here, and it will be a lot of work. */ base = (chipaddr)physmap("it85 communication", 0xFFFFF000, 0x1000); + if (base == ERROR_PTR) + return 1; msg_pdbg("%s():%d base=0x%08x\n", __func__, __LINE__, (unsigned int)base); ce_high = (unsigned char *)(base + 0xE00); /* 0xFFFFFE00 */ diff --git a/mcp6x_spi.c b/mcp6x_spi.c index ac40557..438e661 100644 --- a/mcp6x_spi.c +++ b/mcp6x_spi.c @@ -149,6 +149,8 @@ int mcp6x_spi_init(int want_spi) } /* Map the BAR. Bytewise/wordwise access at 0x530 and 0x540. */ mcp6x_spibar = physmap("NVIDIA MCP6x SPI", mcp6x_spibaraddr, 0x544); + if (mcp6x_spibar == ERROR_PTR) + return 1; #if 0 /* FIXME: Run the physunmap in a shutdown function. */ diff --git a/nicintel_spi.c b/nicintel_spi.c index 72b470b..6dc40f5 100644 --- a/nicintel_spi.c +++ b/nicintel_spi.c @@ -178,6 +178,8 @@ int nicintel_spi_init(void) nicintel_spibar = physmap("Intel Gigabit NIC w/ SPI flash", io_base_addr, MEMMAP_SIZE); + if (nicintel_spibar == ERROR_PTR) + return 1; /* Automatic restore of EECD on shutdown is not possible because EECD * does not only contain FLASH_WRITES_DISABLED|FLASH_WRITES_ENABLED, * but other bits with side effects as well. Those other bits must be diff --git a/ogp_spi.c b/ogp_spi.c index 9b596bf..b3963c2 100644 --- a/ogp_spi.c +++ b/ogp_spi.c @@ -138,6 +138,8 @@ int ogp_spi_init(void) return 1; ogp_spibar = physmap("OGP registers", io_base_addr, 4096); + if (ogp_spibar == ERROR_PTR) + return 1; if (register_shutdown(ogp_spi_shutdown, NULL)) return 1; diff --git a/physmap.c b/physmap.c index adce315..35dcfb8 100644 --- a/physmap.c +++ b/physmap.c @@ -211,13 +211,11 @@ void physunmap(void *virt_addr, size_t len) } #endif -#define PHYSMAP_NOFAIL 0 -#define PHYSMAP_MAYFAIL 1 #define PHYSMAP_RW 0 #define PHYSMAP_RO 1 static void *physmap_common(const char *descr, unsigned long phys_addr, - size_t len, int mayfail, int readonly) + size_t len, int readonly) { void *virt_addr; @@ -261,8 +259,6 @@ static void *physmap_common(const char *descr, unsigned long phys_addr, "and reboot, or reboot into\n" "single user mode.\n"); #endif - if (!mayfail) - exit(3); } return virt_addr; @@ -270,14 +266,12 @@ static void *physmap_common(const char *descr, unsigned long phys_addr, void *physmap(const char *descr, unsigned long phys_addr, size_t len) { - return physmap_common(descr, phys_addr, len, PHYSMAP_NOFAIL, - PHYSMAP_RW); + return physmap_common(descr, phys_addr, len, PHYSMAP_RW); } void *physmap_try_ro(const char *descr, unsigned long phys_addr, size_t len) { - return physmap_common(descr, phys_addr, len, PHYSMAP_MAYFAIL, - PHYSMAP_RO); + return physmap_common(descr, phys_addr, len, PHYSMAP_RO); } #if defined(__i386__) || defined(__x86_64__) diff --git a/satasii.c b/satasii.c index 838fff4..9f9bb7f 100644 --- a/satasii.c +++ b/satasii.c @@ -97,6 +97,8 @@ int satasii_init(void) } sii_bar = physmap("SATA SiI registers", addr, SATASII_MEMMAP_SIZE) + reg_offset; + if (sii_bar == ERROR_PTR) + return 1; /* Check if ROM cycle are OK. */ if ((id != 0x0680) && (!(pci_mmio_readl(sii_bar) & (1 << 26)))) diff --git a/sb600spi.c b/sb600spi.c index fe60aa9..a4ca20b 100644 --- a/sb600spi.c +++ b/sb600spi.c @@ -226,6 +226,8 @@ int sb600_probe_spi(struct pci_dev *dev) /* Physical memory has to be mapped at page (4k) boundaries. */ sb600_spibar = physmap("SB600 SPI registers", tmp & 0xfffff000, 0x1000); + if (sb600_spibar == ERROR_PTR) + return ERROR_FATAL; /* The low bits of the SPI base address are used as offset into * the mapped page. */ -- 1.7.12.4

2 1

[PATCH] Remove exit call from sys_physmap_*
by Niklas Söderlund Sept. 13, 2013

Sept. 13, 2013

All callers are prepared to handle error if ERROR_PTR is returned. Signed-off-by: Niklas Söderlund <niso(a)kth.se> --- physmap.c | 15 ++++++--------- 1 file changed, 6 insertions(+), 9 deletions(-) diff --git a/physmap.c b/physmap.c index 5aa9874..adce315 100644 --- a/physmap.c +++ b/physmap.c @@ -173,12 +173,11 @@ static void *sys_physmap_rw_uncached(unsigned long phys_addr, size_t len) /* Open the memory device UNCACHED. Important for MMIO. */ if (-1 == (fd_mem = open(MEM_DEV, O_RDWR | O_SYNC))) { perror("Critical error: open(" MEM_DEV ")"); - exit(2); + return ERROR_PTR; } } - virt_addr = mmap(NULL, len, PROT_WRITE | PROT_READ, MAP_SHARED, - fd_mem, (off_t)phys_addr); + virt_addr = mmap(NULL, len, PROT_WRITE | PROT_READ, MAP_SHARED, fd_mem, (off_t)phys_addr); return MAP_FAILED == virt_addr ? ERROR_PTR : virt_addr; } @@ -192,14 +191,12 @@ static void *sys_physmap_ro_cached(unsigned long phys_addr, size_t len) if (-1 == fd_mem_cached) { /* Open the memory device CACHED. */ if (-1 == (fd_mem_cached = open(MEM_DEV, O_RDWR))) { - msg_perr("Critical error: open(" MEM_DEV "): %s", - strerror(errno)); - exit(2); + msg_perr("Critical error: open(" MEM_DEV "): %s", strerror(errno)); + return ERROR_PTR; } } - virt_addr = mmap(NULL, len, PROT_READ, MAP_SHARED, - fd_mem_cached, (off_t)phys_addr); + virt_addr = mmap(NULL, len, PROT_READ, MAP_SHARED, fd_mem_cached, (off_t)phys_addr); return MAP_FAILED == virt_addr ? ERROR_PTR : virt_addr; } @@ -209,7 +206,7 @@ void physunmap(void *virt_addr, size_t len) msg_pspew("Not unmapping zero size at %p\n", virt_addr); return; } - + munmap(virt_addr, len); } #endif -- 1.7.12.3

2 1

[PATCH] Add git support to the Makefile
by Carl-Daniel Hailfinger Aug. 21, 2013

Aug. 21, 2013

flashrom relies on svn to export tarballs and only prints the last svn revision in the version message even if flashrom is in a git tree. Print extended version information (version, svn revision, git hash, repository URL). Allow exporting from git trees. This should make git users first-class citizens like svn users. Side note: If anyone has an idea how to make the export target more beautiful (80 column limit etc.), please tell me. Signed-off-by: Carl-Daniel Hailfinger <c-d.hailfinger.devel.2006(a)gmx.net> Index: flashrom-gitfriendly/Makefile =================================================================== --- flashrom-gitfriendly/Makefile (Revision 1145) +++ flashrom-gitfriendly/Makefile (Arbeitskopie) @@ -97,13 +97,15 @@ # of the checked out flashrom files. # Note to packagers: Any tree exported with "make export" or "make tarball" # will not require subversion. The downloadable snapshots are already exported. -SVNVERSION := $(shell LC_ALL=C svnversion -cn . 2>/dev/null | sed -e "s/.*://" -e "s/$[0-9]*$.*/\1/" | grep "[0-9]" || LC_ALL=C svn info . 2>/dev/null | awk '/^Revision:/ {print $$2 }' | grep "[0-9]" || LC_ALL=C git svn info . 2>/dev/null | awk '/^Revision:/ {print $$2 }' | grep "[0-9]" || echo unknown) +SCMVERSION := $(shell LC_ALL=C svnversion -cn . 2>/dev/null | sed -e "s/.*://" -e "s/$[0-9]*$.*/\1/" | grep "[0-9]" || LC_ALL=C svn info . 2>/dev/null | awk '/^Revision:/ {print $$2 }' | grep "[0-9]" || LC_ALL=C git svn info . 2>/dev/null | awk '/^Revision:/ {print $$2 }' | grep "[0-9]" || echo unknown) +SCMEXTVERSION := $(shell LC_ALL=C git rev-parse HEAD 2>/dev/null | sed 's/^./-\0/') +SCMURL := $(shell LC_ALL=C svn info 2>/dev/null | grep URL: | sed 's/.*URL:[[:blank:]]*//' | grep ^. || LC_ALL=C git remote show origin 2>/dev/null |grep URL: | sed 's/.*URL:[[:blank:]]*//' | grep ^. || echo unknown ) RELEASE := 0.9.2 -VERSION := $(RELEASE)-r$(SVNVERSION) +VERSION := $(RELEASE)-r$(SCMVERSION)$(SCMEXTVERSION) RELEASENAME ?= $(VERSION) -SVNDEF := -D'FLASHROM_VERSION="$(VERSION)"' +SCMDEF := -D'FLASHROM_VERSION="$(VERSION)"' -D'FLASHROM_SCMURL="$(SCMURL)"' # Always enable internal/onboard support for now. CONFIG_INTERNAL ?= yes @@ -310,7 +312,7 @@ TAROPTIONS = $(shell LC_ALL=C tar --version|grep -q GNU && echo "--owner=root --group=root") %.o: %.c .features - $(CC) -MMD $(CFLAGS) $(CPPFLAGS) $(FEATURE_CFLAGS) $(SVNDEF) -o $@ -c $< + $(CC) -MMD $(CFLAGS) $(CPPFLAGS) $(FEATURE_CFLAGS) $(SCMDEF) -o $@ -c $< # Make sure to add all names of generated binaries here. # This includes all frontends and libflashrom. @@ -420,9 +422,12 @@ export: @rm -rf $(EXPORTDIR)/flashrom-$(RELEASENAME) - @svn export -r BASE . $(EXPORTDIR)/flashrom-$(RELEASENAME) - @sed "s/^SVNVERSION.*/SVNVERSION := $(SVNVERSION)/" Makefile >$(EXPORTDIR)/flashrom-$(RELEASENAME)/Makefile - @LC_ALL=C svn log >$(EXPORTDIR)/flashrom-$(RELEASENAME)/ChangeLog + @svn export -r BASE . $(EXPORTDIR)/flashrom-$(RELEASENAME) 2>/dev/null || git checkout-index -a -f --prefix=$(EXPORTDIR)/flashrom-$(RELEASENAME)/ + @# If SCMVERSION or SCMEXTVERSION contain a slash or SCMURL contains a hash, this will explode + @sed "s/^SCMVERSION.*/SCMVERSION := $(SCMVERSION)/;s/^SCMEXTVERSION.*/SCMEXTVERSION := $(SCMEXTVERSION)/;s#^SCMURL.*#SCMURL := $(SCMURL)#" Makefile >$(EXPORTDIR)/flashrom-$(RELEASENAME)/Makefile + @# ChangeLog should be in English, but we want UTF-8 for names. + @( LC_ALL=en_US.UTF-8 svn log 2>/dev/null || LC_ALL=en_US.UTF-8 git log 2>/dev/null || echo "Unable to extract log from SCM" ) >$(EXPORTDIR)/flashrom-$(RELEASENAME)/ChangeLog + @rm -f $(EXPORTDIR)/flashrom-$(RELEASENAME)/.gitignore $(EXPORTDIR)/flashrom-$(RELEASENAME)/.gitattributes $(EXPORTDIR)/flashrom-$(RELEASENAME)/.gitmodules @echo Exported $(EXPORTDIR)/flashrom-$(RELEASENAME)/ tarball: export Index: flashrom-gitfriendly/flashrom.c =================================================================== --- flashrom-gitfriendly/flashrom.c (Revision 1145) +++ flashrom-gitfriendly/flashrom.c (Arbeitskopie) @@ -1355,6 +1355,7 @@ void print_version(void) { msg_ginfo("flashrom v%s", flashrom_version); + msg_ginfo(" from " FLASHROM_SCMURL ","); print_sysinfo(); } @@ -1362,7 +1363,6 @@ { msg_ginfo("flashrom is free software, get the source code at " "http://www.flashrom.org\n"); - msg_ginfo("\n"); } int selfcheck(void) -- http://www.hailfinger.org/

5 11

Re: [flashrom] [coreboot] Dualbios on GA-MA770-UD3
by Carl-Daniel Hailfinger Aug. 10, 2013

Aug. 10, 2013

Hi Vadim, thanks for your mail. I have added the flashrom mailing list in CC: because we should develop a generic way to handle DualBIOS and similar techniques. On 09.04.2010 08:12, Vadim Girlin wrote: > I'm going to try coreboot on Gigabyte GA-MA770-UD3. > It's AMD 770 (RX780 / SB700). > > My motherboard supports hardware dual bios - with two chips on it. > I'm going to try flashing backup chip and boot from it. It seems to be > possible - I've tested it (flashing at least). Chips on this board could > be switched by changing bit 0 at undocumented register EF on LDN 7 of > IT8720. I can use slightly patched flashrom for accessing any chip I > want without any problems. And this is tested many times. > > My idea is to use backup chip for debugging - that always keeps my > chance to reboot from main bios chip. And removes the need for > desoldering etc. > > And second problem is that original bios is checking second chip - and > trying to recover it by flashing the bios from main chip on reboot? > rewriting coreboot. AFAICS this could be solved by including some > signatures etc. It seems to be easy to find out. May be someone is > working on this? > > BTW I can send the patch for flashrom - but I think that with > information I mentioned above somebody could make it much better than my > ugly hack. I hope the regs should be the same for all latest Gigabyte > MBs using IT8720/18 > It would be great if you could send that patch. It will help us make a flashrom design decision that works for all boards with multiple flash chips. By the way, some of us have good contacts at ITE, so we can ask ITE for details about the undocumented registers. Regards, Carl-Daniel -- http://www.hailfinger.org/

5 12

[RFC] Layout handling...
by Carl-Daniel Hailfinger July 12, 2013

July 12, 2013

Hi, before we change flashrom to work with current layout requirements, we should summarize the features we need/want, and then decide how to handle the individual layout regions internally. Terms: "write strategy" describes a combination of erase+write commands with a given block size, touching some blocks. "read protected" describes a region which can not be read. "write protected" describes a region which can not be written. "write once" describes a region which can be written exactly once in the chip lifetime. If you add a feature to the list, please give it a nickname so we know which feature people are referring to. Some of the features below might not be desirable, but I want to list them anyway so we can explicitly declare them as unsupported if needed. fullwrite-unrestricted: Write a chip-sized image, no special read/write restrictions of the chip, no layout file needs to be specified. Default write case right now. fullwrite-noread: Write a chip-sized image, reading anything from the chip is not possible. Many DVD writers fall in that category. No verification, violates our reliability guarantees. fullwrite-postread: Write a chip-sized image, reading anything from the chip is only possible after write. Chips which are read-locked until a full erase/write fall in that category. Do those exist as standalone flash chips or only integrated into processors? fullwrite-partialread: Write a chip-sized image, reading is only possible in some chip regions. Only partial verification, violates our reliability guarantees. fullwrite-partialwrite: Write a chip-sized image, but writing is only possible in some regions. This is obviously a conflict unless the image has the same contents as the chip in the write-protected regions and there is a possible write strategy for the whole image which does not touch the write-protected regions. Should flashrom always refuse this scenario, or only refuse it in case of conflicts? partialwrite-unrestricted: Write only parts of an image, the rest of the chip contents is kept, no special read/write restrictions. partialwrite-partialread: Write only parts of an image, the rest of the chip contents is kept, reading is only possible in some chip regions. If no read-protected regions are written and a suitable write strategy exists, should flashrom warn? If a read-protected region is written, should flashrom warn/refuse due to reliability requirements? partialwrite-partialwrite: Write only parts of an image, the rest of the chip contents is kept, writing is only possible in some chip regions. If no write-protected regions are written and a suitable write strategy exists, should flashrom warn? flashrom will refuse to write a write-protected region. fullread-unrestricted: Read the full chip, no special read restrictions of the chip. partialread-unrestricted: Read only parts of a chip, no special read restrictions of the chip. partialread-partialread: Read only parts of a chip, some regions are read-protected. flashrom should refuse to read any read-protected regions. partialread-imagefiller: If only parts of a chip are read and the read image has full chip size, what should be used as filler for unread regions? 0x00 or 0xff? partialread-layout-imagesize: If only parts of a chip are read, should the read image still have full chip size with all read regions filled in at the respective locations? partialread-layout-split: If only parts of a chip are read, should it be possible to write each region (or a combination thereof) to a separate image file, and would that mapping be specified in the layout file? partialwrite-layout-split: If only parts of a chip are written, should it be possible to collect each part of the new image from a separate image file, and would that mapping be specified in the layout file? readwrite-protection-time: Which read protection and write protection times exist? Temporary lock until unlock, temporary lock until chip soft reset, temporary lock until chip/programmer hard reset (powerdown or reset line), permanent eternal lock. readwrite-protection-type: Which read protection and write protection types exist? Programmer lock (e.g. Intel SPI), hardware chip lock (WP pin), software chip lock (chip command). readwrite-protection-interaction: How should we express this situation: A region is write-locked with a software chip lock, but to remove that software chip lock, a hardware chip lock has to be disabled first, then the software chip lock can be disabled. partialaccess-crash: Some regions in the chip are really off-limits and will cause an unrecoverable error (system crash) when accessed (read or write). That seems to be the case for some EC/flash interactions. Comments? Any forks of flashrom (e.g. chromium) which need infrastructure features not mentioned above? Regards, Carl-Daniel -- http://www.hailfinger.org/

5 12

Re: [flashrom] USB 3 PCI Card support (request to add device)
by Philipp Post June 13, 2013

June 13, 2013

I have added some flashrom and linux logs which might be helpfull: The USB 3 PCI card has the following system properties: DRIVER=xhci_hcd PCI_CLASS=C0330 PCI_ID=1033:0194 PCI_SUBSYS_ID=1033:0194 PCI_SLOT_NAME=0000:03:00.0 MODALIAS=pci:v00001033d00000194sv00001033sd00000194bc0Csc03i30 There is also a PCI NIC Intel 82541PI Gigabit Ethernet Controller (8086:107c, BDF 02:0a.0) and a PCI soundcard installed. Tried to read the flash rom out of the 82541PI Intel NIC without problems. /sys/bus/pci/devices/0000:02:0a.0 of the Intel NIC also gives access to "rom" which the USB 3 and the soundcard do not. I am not sure if that means that the rom is not accessible at all or if Ubuntu just does not know how. Thanks in advance, brgds Philipp -----Ursprüngliche Nachricht----- Von: Philipp Post [mailto:Post.Philipp@googlemail.com] Gesendet: Donnerstag, 24. Januar 2013 19:23 An: 'flashrom(a)flashrom.org' Betreff: USB 3 PCI Card support (request to add device) Good day, following configuration: - USB 3 PCI (not PCI-Express) card with RENESAS uPD720200A(Revision 4) USB chip and ATMEL AT25F512B (ATMEL 1138, 25F512B, SSH) rom chip in - IBM NetVista 8305 Desktop PC - built year 2003 The issue with the card is that it needs a flash rom update to work with Windows Vista and up, otherwise the device does not work at all in these OS ("device cannot start"). Windows XP with Renesas drivers and Linux Ubuntu 11.3 out of the box works fine. RENESAS has a rom update for the chip and a flash update tool which however does not work with PCI cards, but just chips soldered to mainboards. At least this seems to me the reason why it does not work. Chip is recognized but the RENESAS tool says "Can not load / access firmware". The tool has the option to change the device / vendor id as well via .ini file. The card used is the only PCI USB 3 card on the market and basically works very well when it works - with max PCI speed of course, which means approx. double speed of USB 2. The card exists in hardware version 1 (RENESAS chip uPD720200A) and 2 (RENESAS chip uPD720202). I do just have access to hardware version 1. The name of the manufacturer behind the card is unknown to me, but it could be an ADDONICS product. I do not have a technical drawing of the card and I fear I can not obtain one but I do have the card itself. My question: would it be possible to incorporate this card into flashrom? Would be great if someone could do it as I do not have the necessary skills to do it myself. In case you need further information, please let me know. Thanks in advance, brgds Philipp Links and further reading: -------------------------- Card Information from different sellers (same card): http://www.addonics.com/products/ad2u3pci.php http://eu.startech.com/Cards-Adapters/USB-3.0/Cards/2-Port-PCI-SuperSpeed-US B-3-Card-Adapter~PCIUSB3S2 http://www.lindy.de/premium-usb-3-karte-2-port-pci/51082.html ATMEL rom specification (contains info how to write rom images): www.atmel.com/Images/doc3689.pdf Renesas USB chip information: http://www.renesas.eu/products/soc/usb_assp/product/upd720200a/index.jsp RENESAS rom update 4.0.2.0 with Windows updater (32 + 64 Bit) and drivers: http://www.station-drivers.com/page/renesas.htm We need: "Firmware Version 4.0.2.0 pour uPD720200a" RENESAS MS DOS flash update tool (which did not do the job): http://www.computerbase.de/forum/attachment.php?attachmentid=283282&d=133832 3319 IBM NetVista 8305 Hardware Maintenance Manual: http://download.lenovo.com/ibmdl/pub/pc/pccbbs/netvista_pdf/24p2969.pdf Hardware Ids: Renesas Electronics USB 3.0 Host Controller: PCI\VEN_1033&DEV_0194&SUBSYS_01941033&REV_04\5&25043F3&0&0048F0 Renesas Electronics USB 3.0 Root Hub NUSB3\ROOT_HUB30\6&2D374EF3&2

2 6