Re: Kernel builds failing with lots of "failed to retrieve array bounds" errors
On 28/05/2012 20:45, Dimitry Andric wrote: Note, in r236149 I have pulled in a change from upstream clang, which should fix the root cause of the "failed to retrieve array bounds" messages. Indeed, I updated & rebuilt yesterday & message no-longer appeared. Thanks Sevan ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"
Re: Kernel builds failing with lots of "failed to retrieve array bounds" errors
On 2012-05-24 22:50, Sevan / Venture37 wrote: > On 24/05/2012 20:21, Dimitry Andric wrote: >> I've seen these too, and it seems clang produces debug info which >> ctfconvert can't handle, for some reason. However, in my case, the >> kernel build doesn't abort at all, it continues and all the object files >> seem to work just fine. > > Updated to r235926 & kernel build completed successfully. Note, in r236149 I have pulled in a change from upstream clang, which should fix the root cause of the "failed to retrieve array bounds" messages. For flexible array members (e.g. arrays at the end of a struct with no specified size), clang didn't produce correct debug information. This confused ctfconvert, causing it to bail out. ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"
Re: Kernel builds failing with lots of "failed to retrieve array bounds" errors
On 24/05/2012 20:21, Dimitry Andric wrote: I've seen these too, and it seems clang produces debug info which ctfconvert can't handle, for some reason. However, in my case, the kernel build doesn't abort at all, it continues and all the object files seem to work just fine. Updated to r235926 & kernel build completed successfully. Sevan ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"
Re: Kernel builds failing with lots of "failed to retrieve array bounds" errors
On 2012-05-24 19:13, Sevan / Venture37 wrote: > I'm unable to build the generic kernel, seeing lots of "failed to > retrieve array bounds" errors (129 to be exact) starting with ERROR: > scsi_all.c: die 43574: failed to retrieve array bounds & stoping at I've seen these too, and it seems clang produces debug info which ctfconvert can't handle, for some reason. However, in my case, the kernel build doesn't abort at all, it continues and all the object files seem to work just fine. I don't know much much about the dtrace/ctfconvert stuff though, so I will have to ask somebody else to step up to investigate, and hopefully fix it. :) ... > Userland was built & installed earlier this morning WITH_CLANG_IS_CC > defined in src.conf, and the make.conf is as follows > STRIP= > CFLAGS+=-fno-omit-frame-pointer Ah, I think that it works for me, because I don't define STRIP to empty. Just as an experiment, can you try commenting that setting, and do a clean build of your kernel? ___ freebsd-current@freebsd.org mailing list http://lists.freebsd.org/mailman/listinfo/freebsd-current To unsubscribe, send any mail to "freebsd-current-unsubscr...@freebsd.org"
Re: Kernel builds failing
It seems [EMAIL PROTECTED] wrote: > > For the last two days, my kernel builds have been failing with: > > > > linking kernel.debug > > ata-all.o: In function `ataioctl': > > /usr/src/sys/dev/ata/ata-all.c(.text+0x791): undefined reference to > > `atapi_queue _cmd' > > *** Error code 1 I'm looking at how to solve this, commit to follow soon... -Søren To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message
RE: Kernel builds failing
Hi Adam, Søren (CC'd to [EMAIL PROTECTED]) On 31-Aug-01 Adam Kranzel wrote: > Hi... > > For the last two days, my kernel builds have been failing with: > > linking kernel.debug > ata-all.o: In function `ataioctl': > /usr/src/sys/dev/ata/ata-all.c(.text+0x791): undefined reference to > `atapi_queue _cmd' > *** Error code 1 > > Stop in /usr/obj/usr/src/sys/CHESHIRE. > *** Error code 1 > > Stop in /usr/src. > *** Error code 1 > > Stop in /usr/src. > > > Any ideas? > > thanks > -Adam > > To Unsubscribe: send mail to [EMAIL PROTECTED] > with "unsubscribe freebsd-current" in the body of the message I had this problem to, I went for a snoop and found the following; >From my kenel config file; device ata device atadisk # ATA disk drives #device atapicd # ATAPI CDROM drives #device atapifd # ATAPI floppy drives #device atapist # ATAPI tape drives I don't have an ata cd, floppy or tape in the machine so I commented them out - I'm assuming you've done the same. >From /usr/src/sys/conf/files; dev/ata/atapi-all.c optional atapicd dev/ata/atapi-all.c optional atapifd dev/ata/atapi-all.c optional atapist the atapi-all.c file contains atapi_queue_cmd(), in other words, with the above configuration, atapi-all.c is never compiled, so the function isn't there. Workaround: add one of the devices to get it compiled, ata[cd,fd,st] Søren, looks like your last ata-all.c commit has broken this, sos 2001/08/30 02:47:17 PDT Modified files: sys/dev/ata ata-all.c Log: Add support for sending ATAPI commands via ioctl. Revision ChangesPath 1.118 +42 -1 src/sys/dev/ata/ata-all.c Cheers, -- Kent Ibbetson [EMAIL PROTECTED] To Unsubscribe: send mail to [EMAIL PROTECTED] with "unsubscribe freebsd-current" in the body of the message