From dave at meadorresearch.com Fri Mar 13 01:28:39 2009 From: dave at meadorresearch.com (Dave Meador) Date: Thu, 12 Mar 2009 18:28:39 -0700 Subject: P&E driver compile failure Message-ID: <49B9B6C7.5020705@meadorresearch.com> Hello, I tried to compile the P&E driver provided with the G++ Lite coldfire-4.3-43-m68k-linux-gnu package and it failed with the following output. The kernel that I am using is 2.6.27.5-37.fc9.i686, and I have the kernel-devel-* and kernel-headers-* rpms installed. [user at Europa unit_cfz_install_ver_332_920]$ sudo ./setup.sh cleaned old version install loading cache ./config.cache checking for cpu architecture... i386 checking for WinDriver root directory... /usr/local/PEMicrocomputerSystems/windriver checking for linux kernel source... found at /lib/modules/2.6.27.5-37.fc9.i686/build checking for lib directory... ln -sf $(ROOT_DIR)/lib/$(SHARED_OBJECT).so /usr/lib/$(SHARED_OBJECT).so checking which directories to include... -I/lib/modules/2.6.27.5-37.fc9.i686/source/include -I/lib/modules/2.6.27.5-37.fc9.i686/source/include/asm/mach-default checking linux kernel version... 2.6.27.5-37.fc9.i686 checking for modules installation directory... /lib/modules/2.6.27.5-37.fc9.i686/kernel/drivers/misc checking for gcc kernel version... 4 checking output directory... LINUX.2.6.27.5-37.fc9.i686.i386 checking target... LINUX.2.6.27.5-37.fc9.i686.i386/windrvr6.ko checking for usb support... yes checking for regparm kernel option... 1 checking for right linked object... windrvr_gcc_v3_regparm.a checking for modpost location... /lib/modules/2.6.27.5-37.fc9.i686/build/scripts/mod/modpost checking for udev support... no creating ./config.status creating makefile rm -f core LINUX.2.6.27.5-37.fc9.i686.i386/windrvr6.ko LINUX.2.6.27.5-37.fc9.i686.i386/* cc -c -O2 -Wall -DLINUX -D__KERNEL__ -DMODULE -DWINDRIVER_KERNEL -DLINUX_USB_SUPPORT -mpreferred-stack-boundary=2 -mregparm=3 -nostdinc -iwithprefix include -Wstrict-prototypes -Wno-trigraphs -fno-common -pipe -O -I/lib/modules/2.6.27.5-37.fc9.i686/source/include -I/lib/modules/2.6.27.5-37.fc9.i686/source/include/asm/mach-default -I/usr/local/PEMicrocomputerSystems/windriver/include -I/usr/local/PEMicrocomputerSystems/windriver -fno-strict-aliasing -D"KBUILD_STR(s)=#s" -D"KBUILD_MODNAME=KBUILD_STR(windrvr6)" -D"KBUILD_BASENAME=KBUILD_STR(windrvr6)"-include /lib/modules/2.6.27.5-37.fc9.i686/build/include/linux/autoconf.h -c -o LINUX.2.6.27.5-37.fc9.i686.i386/linux_wrappers.o linux_wrappers.c linux_wrappers.c:306: error: unknown field 'nopage' specified in initializer linux_wrappers.c:306: warning: initialization from incompatible pointer type linux_wrappers.c: In function 'build_dma_list_26': linux_wrappers.c:1791: error: 'struct scatterlist' has no member named 'page' linux_wrappers.c:1798: error: 'struct scatterlist' has no member named 'page' linux_wrappers.c: In function 'LINUX_free_sg_dma': linux_wrappers.c:1954: error: 'struct scatterlist' has no member named 'page' linux_wrappers.c:1955: error: 'struct scatterlist' has no member named 'page' linux_wrappers.c:1956: error: 'struct scatterlist' has no member named 'page' make: *** [LINUX.2.6.27.5-37.fc9.i686.i386/linux_wrappers.o] Error 1 cc -c -O2 -Wall -DLINUX -D__KERNEL__ -DMODULE -DWINDRIVER_KERNEL -DLINUX_USB_SUPPORT -mpreferred-stack-boundary=2 -mregparm=3 -nostdinc -iwithprefix include -Wstrict-prototypes -Wno-trigraphs -fno-common -pipe -O -I/lib/modules/2.6.27.5-37.fc9.i686/source/include -I/lib/modules/2.6.27.5-37.fc9.i686/source/include/asm/mach-default -I/usr/local/PEMicrocomputerSystems/windriver/include -I/usr/local/PEMicrocomputerSystems/windriver -fno-strict-aliasing -D"KBUILD_STR(s)=#s" -D"KBUILD_MODNAME=KBUILD_STR(windrvr6)" -D"KBUILD_BASENAME=KBUILD_STR(windrvr6)"-include /lib/modules/2.6.27.5-37.fc9.i686/build/include/linux/autoconf.h -c -o LINUX.2.6.27.5-37.fc9.i686.i386/linux_wrappers.o linux_wrappers.c linux_wrappers.c:306: error: unknown field 'nopage' specified in initializer linux_wrappers.c:306: warning: initialization from incompatible pointer type linux_wrappers.c: In function 'build_dma_list_26': linux_wrappers.c:1791: error: 'struct scatterlist' has no member named 'page' linux_wrappers.c:1798: error: 'struct scatterlist' has no member named 'page' linux_wrappers.c: In function 'LINUX_free_sg_dma': linux_wrappers.c:1954: error: 'struct scatterlist' has no member named 'page' linux_wrappers.c:1955: error: 'struct scatterlist' has no member named 'page' linux_wrappers.c:1956: error: 'struct scatterlist' has no member named 'page' make: *** [LINUX.2.6.27.5-37.fc9.i686.i386/linux_wrappers.o] Error 1 Waiting 15 seconds for windriver's kernel module to load... ********* ********** ********** ******** ********* ********* ********** ********** ******** ********* ********* ********** ********** ******** ********* ********* P&E Install has failed******** ********* ********* ********** ********** ******** ********* ********* ********** ********** ******** ********* ********* ********** ********** ******** ********* Please advise, Dave From rich.g.nht at googlemail.com Mon Mar 16 11:27:00 2009 From: rich.g.nht at googlemail.com (Rich G) Date: Mon, 16 Mar 2009 11:27:00 +0000 Subject: Problem using -pg with large binaries and Sourcery G++ Lite 4.3-43 Message-ID: Hi I've been trying to compile using Sourcery G++ Lite 4.3-43 with profiling support enabled with -pg but I keep getting the following error from gcc when trying to compile large binaries (with and without static): relocation truncated to fit: R_68K_PC16 against `.data' This seems similar to the gcc bug shown at http://www.mail-archive.com/gcc-bugs at gcc.gnu.org/msg222205.html -pg works ok when creating smaller binaries. Is the bug in the link above relevant to my problem ? Any info appreciated Rich -------------- next part -------------- An HTML attachment was scrubbed... URL: From rich.g.nht at googlemail.com Mon Mar 16 11:36:58 2009 From: rich.g.nht at googlemail.com (Rich G) Date: Mon, 16 Mar 2009 11:36:58 +0000 Subject: Problem using -pg with large binaries and Sourcery G++ Lite 4.3-43 In-Reply-To: References: Message-ID: Simple testcase - create simple helloworld app and compile with/without -static with -pg Thanks Rich On 16/03/2009, Rich G wrote: > Hi > > I've been trying to compile using Sourcery G++ Lite 4.3-43 with profiling > support enabled with -pg but I keep getting the following error from gcc > when trying to compile large binaries (with and without static): > > relocation truncated to fit: R_68K_PC16 against `.data' > > This seems similar to the gcc bug shown at > http://www.mail-archive.com/gcc-bugs at gcc.gnu.org/msg222205.html > > -pg works ok when creating smaller binaries. > > Is the bug in the link above relevant to my problem ? > > Any info appreciated > > Rich > From maxim at codesourcery.com Mon Mar 16 11:41:48 2009 From: maxim at codesourcery.com (Maxim Kuvyrkov) Date: Mon, 16 Mar 2009 14:41:48 +0300 Subject: [coldfire-gnu-discuss] Re: Problem using -pg with large binaries and Sourcery G++ Lite 4.3-43 In-Reply-To: References: Message-ID: <49BE3AFC.7020607@codesourcery.com> Rich G wrote: > Simple testcase - create simple helloworld app and compile > with/without -static with -pg Thanks, I can reproduce your problem. I'll try to look into it sometime soon. -- Maxim From dan at codesourcery.com Mon Mar 16 16:29:40 2009 From: dan at codesourcery.com (Daniel Jacobowitz) Date: Mon, 16 Mar 2009 12:29:40 -0400 Subject: [coldfire-gnu-discuss] P&E driver compile failure In-Reply-To: <49B9B6C7.5020705@meadorresearch.com> References: <49B9B6C7.5020705@meadorresearch.com> Message-ID: <20090316162939.GJ30257@caradoc.them.org> On Thu, Mar 12, 2009 at 06:28:39PM -0700, Dave Meador wrote: > Hello, > > I tried to compile the P&E driver provided with the G++ Lite > coldfire-4.3-43-m68k-linux-gnu package and it failed with the following > output. The kernel that I am using is 2.6.27.5-37.fc9.i686, and I have > the kernel-devel-* and kernel-headers-* rpms installed. Hi Dave, This kernel is too new for the P&E drivers. You'll have to check with P&E directly about an update; they might have newer drivers or a workaround, but the latest published drivers still have this limitation. -- Daniel Jacobowitz CodeSourcery From stefanomologni at bertronic.it Tue Mar 17 13:34:15 2009 From: stefanomologni at bertronic.it (Stefano Mologni) Date: Tue, 17 Mar 2009 14:34:15 +0100 Subject: Application crash Message-ID: Good morning, I have some problems while debug with codesourcery. In particular after the follow command m68k-elf-gdb RTOSDemo.elf is all ok, but when I try to connect with the hardware with the following command: target remote | m68k-elf-sprite pe://USBMultilink m52235evb the "m68k-elf-sprite" application crash and fail. Everyone have a suggestion for me? Thanks, Stefano Mologni Progettista HW/FW Ufficio R&D ----------------------------------------- BERTRONIC S.r.l. via Silvio Pellico, 26 24040 - STEZZANO (Bg) Tel. +39 035-4379308 Fax. +39 035-4379309 stefanomologni at bertronic.it www.bertronic.it ----------------------------------------- P Rispetta l'ambiente. Hai davvero bisogno di stampare questa mail? --------------------------------------------------------------------- Le informazioni e gli allegati contenute in questa e-mail sono considerate confidenziali e possono essere riservate. Qualora non foste il destinatario, siete pregati di distruggere questo messaggio e notificarci il problema immediatamente. In ogni caso, non dovrete spedire a terze parti questa e-mail. Vi ricordiamo che la diffusione, l'utilizzo e/o la conservazione dei dati ricevuti per errore costituiscono violazione alle disposizioni del D.L. n.196/2003 denominato "Codice in materia di protezione dei dati personali" -------------- next part -------------- An HTML attachment was scrubbed... URL: From braun at stzedn.de Thu Mar 19 11:23:32 2009 From: braun at stzedn.de (Nathan Braun) Date: Thu, 19 Mar 2009 12:23:32 +0100 Subject: M52259DEMO BDM driver problem Message-ID: <49C22B34.9020402@stzedn.de> Hello, I recently got a M52259 demo kit and wanted to test it. Unfortunately the m68k-elf-sprite is not able to connect to the tblcf: device, although it seems being installed correctly . Do you have any hints what might be going wrong? I attached a debug log below. Thanks and best regards Nathan Braun GNU gdb (Sourcery G++ Lite 4.3-54) 6.8.50.20080821-cvs Copyright (C) 2008 Free Software Foundation, Inc. License GPLv3+: GNU GPL version 3 or later This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Type "show copying" and "show warranty" for details. This GDB was configured as "--host=i686-mingw32 --target=m68k-elf". For bug reporting instructions, please see: . (gdb) target remote | m68k-elf-sprite -g tblcf: m52235evb Remote debugging using | m68k-elf-sprite -g tblcf: m52235evb m68k-elf-sprite: CodeSourcery ColdFire Debug Sprite (Sourcery G++ Lite 4.3-54) m68k-elf-sprite: Doing I/O to stdin/stdout m68k-elf-sprite: Memory [0x0,+0x40000) flash (cfm) m68k-elf-sprite: Memory [0x20000000,+0x8000) ram m68k-elf-sprite: Trying to load DLL 'libusb0.dll' m68k-elf-sprite: libusb DLL version 0.1.12.1 m68k-elf-sprite: libusb driver version 0.1.12.1 m68k-elf-sprite: Found 1 TBLCF devices, connecting to #0 m68k-elf-sprite: error: Error claiming interface (-22, unknown reason) m68k-elf-sprite: debug comm:input 15 bytes +$qSupported#37 m68k-elf-sprite: debug comm:Not in sync. Ignoring '+' m68k-elf-sprite: Got packet: 'qSupported' m68k-elf-sprite: debug comm:Send ack m68k-elf-sprite: Sent response: 'E.fatal.Error claiming interface (-22, unknown reason)' warning: Remote failure reply: E.fatal.Error claiming interface (-22, unknown reason) m68k-elf-sprite: debug comm:input 1 bytes + m68k-elf-sprite: debug comm:Got ack Remote communication error: No error. From maxim at codesourcery.com Fri Mar 20 10:47:57 2009 From: maxim at codesourcery.com (Maxim Kuvyrkov) Date: Fri, 20 Mar 2009 13:47:57 +0300 Subject: [coldfire-gnu-discuss] Application crash In-Reply-To: References: Message-ID: <49C3745D.3090504@codesourcery.com> Stefano Mologni wrote: > Good morning, > > I have some problems while debug with codesourcery. In particular after the follow command > > m68k-elf-gdb RTOSDemo.elf > > is all ok, but when I try to connect with the hardware with the following command: > > target remote | m68k-elf-sprite pe://USBMultilink m52235evb > > the ?m68k-elf-sprite? application crash and fail. Stefano, What version of SourceryG++ Lite are you using? How exactly the sprite fails? -- Maxim From maxim at codesourcery.com Fri Mar 20 10:53:22 2009 From: maxim at codesourcery.com (Maxim Kuvyrkov) Date: Fri, 20 Mar 2009 13:53:22 +0300 Subject: [coldfire-gnu-discuss] M52259DEMO BDM driver problem In-Reply-To: <49C22B34.9020402@stzedn.de> References: <49C22B34.9020402@stzedn.de> Message-ID: <49C375A2.9050102@codesourcery.com> Nathan Braun wrote: > Hello, > > I recently got a M52259 demo kit and wanted to test it. Unfortunately the > m68k-elf-sprite is not able to connect to the tblcf: device, although it seems > being installed correctly . > > Do you have any hints what might be going wrong? I attached a debug log below. Hello Nathan, Which OS are you using? Are there any other programs in the background that might be using TBLCF? Regards, -- Maxim From maxim at codesourcery.com Fri Mar 27 13:51:12 2009 From: maxim at codesourcery.com (Maxim Kuvyrkov) Date: Fri, 27 Mar 2009 16:51:12 +0300 Subject: [coldfire-gnu-discuss] M52259DEMO BDM driver problem In-Reply-To: <49C375A2.9050102@codesourcery.com> References: <49C22B34.9020402@stzedn.de> <49C375A2.9050102@codesourcery.com> Message-ID: <49CCD9D0.3060806@codesourcery.com> Nathan Braun wrote: > Hi Maxim, > > I noticed something on the M52259demo, that might eventually be related > to the problem: even when connected over USB the target power LED is not > lit. I think the OSBDM requires a special command to power the MCU. > CodeWarrior obviously does it. Ah, this is an OSBDM board. Nathan, we are working on adding support for OSBDM probe to our debug sprite. Please follow the release notes in our toolchains for the announcement. Regards, -- Maxim From maxim at codesourcery.com Fri Mar 27 13:55:23 2009 From: maxim at codesourcery.com (Maxim Kuvyrkov) Date: Fri, 27 Mar 2009 16:55:23 +0300 Subject: [coldfire-gnu-discuss] Application crash In-Reply-To: <49C3745D.3090504@codesourcery.com> References: <49C3745D.3090504@codesourcery.com> Message-ID: <49CCDACB.5080101@codesourcery.com> Stefano Mologni wrote: > Maxim Kuvyrkov wrote: >> Stefano Mologni wrote: >>> Good morning, >>> >>> I have some problems while debug with codesourcery. In particular >>> after the follow command >>> >>> m68k-elf-gdb RTOSDemo.elf >>> >>> is all ok, but when I try to connect with the hardware with the >>> following command: >>> >>> target remote | m68k-elf-sprite pe://USBMultilink >>> m52235evb >>> >>> the ?m68k-elf-sprite? application crash and fail. >> >> Stefano, >> >> What version of SourceryG++ Lite are you using? How exactly the sprite >> fails? > Thanks. > The exact version is 6.8.50.20080821. The sprite fail without any > message in console, only the Windows message sounds like "The > application have encountered an unknown error." Stefano, Can you share a testcase and describe the exact steps to reproduce the crash? Thanks, -- Maxim