From: Howell, D. P <dav...@in...> - 2002-04-05 00:11:33
|
The attached patch is to the latest 2.4/arch/i386/config.in to correct a problem that we saw after I integrated 4.0.1-3 from the 3/19 CVS tree. The diffs are for the current CVS version. Please take a look at the description and let me know if this makes sense. Another item from the Configure.help, to define help for the CONFIG_DUMP_COMPRESSION entries, reword as appropriate. --- /tmp/Configure.help Thu Apr 4 18:58:51 2002 +++ /tmp/Configure.help.new Thu Apr 4 19:00:17 2002 @@ -3675,6 +3675,16 @@ program to open up /dev/dump to perform dump configuration. More information about dumping is at http://lkcd.sourceforge.net/. +Kernel crash dumps (version 2) +CONFIG_DUMP_COMPRESS_RLE + Compress the dump memory image written to the dump device using + RLE compression (default). + +Kernel crash dumps (version 2) +CONFIG_DUMP_COMPRESS_GZIP + Compress the dump memory image written to the dump device using + GZIP compression. + Kernel support for ELF binaries CONFIG_BINFMT_ELF ELF (Executable and Linkable Format) is a format for libraries and -----Original Message----- From: Howell, David P Sent: Thursday, April 04, 2002 12:13 PM To: Howell, David P; 'va...@in...'; george anzinger Cc: dev...@ca... Subject: RE: [CarrierLinux Developer] Configure problem DUMP_COMPRESS In arch/i386/config.in I just made the following change: --- config.in Fri Mar 22 14:56:01 2002 +++ config.fix.in Thu Apr 4 11:54:36 2002 @@ -406,12 +411,8 @@ comment 'Kernel hacking' tristate 'Linux Kernel Crash Dump (LKCD) Support' CONFIG_DUMP -if [ "$CONFIG_DUMP" = "y" ]; then - dep_bool ' LKCD RLE compression' CONFIG_DUMP_COMPRESS_RLE $CONFIG_DUMP - dep_bool ' LKCD GZIP compression' CONFIG_DUMP_COMPRESS_GZIP $CONFIG_DUMP -fi -if [ "$CONFIG_DUMP" = "m" ]; then - dep_tristate ' LKCD RLE compression' CONFIG_DUMP_COMPRESS_RLE $CONFIG_DUMP +if [ "$CONFIG_DUMP" != "n" ]; then + dep_tristate ' LKCD RLE compression' CONFIG_DUMP_COMPRESS_RLE $CONFIG_DUMP dep_tristate ' LKCD GZIP compression' CONFIG_DUMP_COMPRESS_GZIP $CONFIG_DUMP fi This gets CONFIG_DUMP_COMPRESS_RLE and CONFIG_DUMP_COMPRESS_GZIP defined and gets rid of the extra set of screen entries for these that was making the xconfig screen busy and confusing for them. One deficit, some logic changed in that the values for the compression variables could be "y" | "m" | "n", where in the original if CONFIG_DUMP were "y" then the values for compression could have only been "y" | "n", not "m"; this is not an problem if CONFIG_DUMP is "m". If this is an issue we could add logic to catch and fix this case, but for now they should have appropriate default values (i.e. matching CONFIG_DUMP). Does this work as you understand it needs to, Vamsi? Let me know and if so I'll submit it for TLT 1.1 and to the lkcd mailing list. Thanks, Dave Howell -----Original Message----- From: Howell, David P [mailto:dav...@in...] Sent: Thursday, April 04, 2002 11:14 AM To: 'va...@in...'; george anzinger Cc: dev...@ca... Subject: RE: [CarrierLinux Developer] Configure problem DUMP_COMPRESS I just looked at the latest CVS tree on sourceforge.net/projects/lkcd and we match the latest CVS version for 2.4/arch/i386/config.in. This is a bug and needs to be fixed, I'll take a look at getting it fixed and submit a fix back to the lkcd mailing list. Dave Howell -----Original Message----- From: Vamsi Krishna S . [mailto:va...@in...] Sent: Thursday, April 04, 2002 3:25 AM To: george anzinger Cc: dev...@ca... Subject: Re: [CarrierLinux Developer] Configure problem DUMP_COMPRESS These CONFIG options are from LKCD. Look for "elif" construct in arch/i386/config.in and replace it with fi;if. This was fixed some time ago in the LKCD cvs. I don't know which version of LKCD you are using but you need to something similar to: - elif [ "$CONFIG_DUMP" = "m" ]; then + fi + if [ "$CONFIG_DUMP" = "m" ]; then -- Vamsi Krishna S. Linux Technology Center, IBM Software Lab, Bangalore. Ph: +91 80 5262355 Extn: 3959 Internet: va...@in... On Wed, Apr 03, 2002 at 08:26:31PM -0800, george anzinger wrote: > I suppose this is unique to xconfig, but still something is wrong. This > is the latest kernel with all patches thru high-res-timers. Anybody > recognize those two CONFIG options. > > -g > > nice -10 make xconfig > rm -f include/asm > ( cd include ; ln -sf asm-i386 asm) > make -C scripts kconfig.tk > make[1]: Entering directory `/usr/src/linux-2.4.17-TLT-plus-hir/scripts' > gcc -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -c -o tkparse.o > tkparse.c > gcc -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -c -o tkcond.o > tkcond.c > gcc -Wall -Wstrict-prototypes -O2 -fomit-frame-pointer -c -o tkgen.o > tkgen.c > gcc -o tkparse tkparse.o tkcond.o tkgen.o > cat header.tk >> ./kconfig.tk > ./tkparse < ../arch/i386/config.in >> kconfig.tk > echo "set defaults \"arch/i386/defconfig\"" >> kconfig.tk > echo "set ARCH \"i386\"" >> kconfig.tk > cat tail.tk >> kconfig.tk > chmod 755 kconfig.tk > make[1]: Leaving directory `/usr/src/linux-2.4.17-TLT-plus-hir/scripts' > wish -f scripts/kconfig.tk > ERROR - Attempting to write value for unconfigured variable > (CONFIG_DUMP_COMPRESS_RLE). > ERROR - Attempting to write value for unconfigured variable > (CONFIG_DUMP_COMPRESS_GZIP). > > -- > George Anzinger ge...@mv... > High-res-timers: http://sourceforge.net/projects/high-res-timers/ > Real time sched: http://sourceforge.net/projects/rtsched/ > Preemption patch: http://www.kernel.org/pub/linux/kernel/people/rml > _______________________________________________ > Developer mailing list > Dev...@ca... > http://www.carrierlinux.org/mailman/listinfo/developer _______________________________________________ Developer mailing list Dev...@ca... http://www.carrierlinux.org/mailman/listinfo/developer _______________________________________________ Developer mailing list Dev...@ca... http://www.carrierlinux.org/mailman/listinfo/developer |