summaryrefslogtreecommitdiffstats
path: root/loader
Commit message (Collapse)AuthorAgeFilesLines
* sparc64 is a lib64 archDennis Gilmore2010-03-051-1/+1
| | | | make sure sparc64 has lib64 paths set that same as other 64 bit arches that use lib64
* It's necessary to give each vfprintf invocation a fresh va_list (#568235)Ales Kozumplik2010-02-251-3/+6
| | | | | | | Commit 3fb12e4eafebfa3885c25b9ab93e9c35f6847425 changed the way variable argument list is handled in log.c this causes SIGSEGV on x86_64. I should see 'man vfprintf' the next time.
* logging: make loader say 'loader' (#563009).Ales Kozumplik2010-02-241-10/+10
|
* Make loader log into syslog (so remote logging works for it as well) (#524980)Ales Kozumplik2010-02-242-15/+34
|
* Revert "Add back hald for Xorg input device queries (#553780)"Chris Lumens2010-02-231-6/+0
| | | | | This reverts commit 829bd5980b4c386de48b1a8274b499fe97c1b7dc. X was the only thing still using hal, and it no longer is.
* Fix return values for dasd_settle_all() in linuxrc.s390 (#558881).David Cantrell2010-02-121-4/+3
| | | | Also remove one call to dasd_settle_all(), which is not necessary.
* Correct references to lcs and ctcm devices (#561816).David Cantrell2010-02-111-7/+7
| | | | Followup to 84904cdfc5b6822ac62c39694b07eb323b59b2f3.
* Use lsznet.raw from s390utils package (#563548).David Cantrell2010-02-114-398/+4
| | | | | | The lsznet.raw and controlunits.sh files are now part of the s390-tools project, which is in the s390utils package. Remove our copies and make sure we use the ones from the s390utils package.
* Revert "Write ARP=no to ifcfg file when VSWITCH=1 is set on s390x (#561926)."David Cantrell2010-02-111-1/+0
| | | | | | | | | This reverts commit 3b90bea90eb3c65f34e358babcb866b1514a4994. ARP=no did not actually fix the problem with VSWITCH=1, nor did ARP=yes. The problem was found to be in the /lib/udev/ccw_init file, which is part of initscripts. The ifup script is not able to handle the asynchronous bring up of s390x network devices.
* Use /sys/devices/lcs instead of /sys/devices/cu3088 (#561816).David Cantrell2010-02-111-6/+6
| | | | | | The kernel change is in place and the cu3088 sysfs path is going away at some point. For CTC and LCS devices, we need to use the /sys/devices/lcs path.
* Wait for all DASDs to be online after autodetection (#558881).David Cantrell2010-02-111-0/+12
| | | | | | | | | | | | | | | | | | | | When a user boots with cio_ignore=all,!0.0.0009 on s390x, it tells the kernel to only bring device 0.0.0009 online before booting. Device 0.0.0009 is the 3270 console. The idea behind cio_ignore is to prevent long bootups when the kernel will see thousands of devices and have to bring up each one and gather stats. With cio_ignore, you get in to Linux faster and then specifically enable the devices you want. We support this in anaconda via linuxrc.s390. Your CMS conf file can specify the devices or device ranges to use during installation. If you don't specify that information, linuxrc.s390 prompts you. If you don't know what information to provide, you can just press Enter and the entire system is scanned. When a DASD scan happens and each one is brought online, make sure we wait until the devices enter the 'active' or 'unformatted' states. The other states indicate we should continue waiting for the device to come up.
* Prompt user for install method when going back to STEP_METHOD.David Cantrell2010-02-111-0/+1
| | | | | | | | In STEP_DRIVER, if the user selects Back from the dialog saying no devices of that type were found, we go back to STEP_LANG. The problem is that loaderData->method isn't reset to 1 in STEP_DRIVER when we go back to STEP_METHOD. That causes us to skip over the method selection and since dir is still -1, we go back up to STEP_LANG.
* Improve reboot modes in init.c and shutdown.c.Ales Kozumplik2010-02-103-26/+50
| | | | | | | | Stop the kernel correctly on HALT (so that we dont see 'kernel panic, they killed init' on i386 and on s390 one gets easy access to manual IPL). Introduce a new reboot method that does allow us to see the backtrace and doesn't scroll the screen up with useless unmount info.
* Be more explicit in which libraries we link with.Chris Lumens2010-02-101-1/+1
| | | | This is required by the latest build system's linker changes.
* Merge branch 'no-instdata'Chris Lumens2010-02-081-0/+4
|\
| * Move the isHeadless attribute onto the Anaconda class.Chris Lumens2010-02-041-0/+4
| | | | | | | | | | This requires having loader add --headless to the anaconda arguments automatically on s390, instead of having anaconda figure that out later.
* | Write ARP=no to ifcfg file when VSWITCH=1 is set on s390x (#561926).David Cantrell2010-02-051-0/+1
|/ | | | | | | | | | | When using LAYER2=1 VSWITCH=1 in your CMS conf file on s390x, the ifcfg file written by anaconda will result in a non-working network connection on reboot after installation. When LAYER2=1, we already prevent writing of the HWADDR to the ifcfg file, but when VSWITCH=1, we need to write ARP=no to the ifcfg file as well. Make sure to carry over VSWITCH in linuxrc.s390, but also make sure we don't write it to the final ifcfg file on the target system.
* Don't log the size of what we're unpacking anymore.Chris Lumens2010-02-011-5/+1
|
* Update =~ regexps in lsznet.raw for bash-4.1 (#558537).David Cantrell2010-01-281-1/+1
| | | | Same problem that was present in linuxrc.s390.
* Add 5 second ping delay for gateway and dns test on s390x (#536815).David Cantrell2010-01-271-3/+3
|
* Update =~ regexps in linuxrc.s390 for bash-4.1 (#558537).David Cantrell2010-01-271-6/+6
| | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | | GNU bash 4.1 changes the behavior of the =~ operator in the [[ command. From the bash FAQ: E14) Why does quoting the pattern argument to the regular expression matching conditional operator (=~) cause regexp matching to stop working? In versions of bash prior to bash-3.2, the effect of quoting the regular expression argument to the [[ command's =~ operator was not specified. The practical effect was that double-quoting the pattern argument required backslashes to quote special pattern characters, which interfered with the backslash processing performed by double-quoted word expansion and was inconsistent with how the == shell pattern matching operator treated quoted characters. In bash-3.2, the shell was changed to internally quote characters in single- and double-quoted string arguments to the =~ operator, which suppresses the special meaning of the characters special to regular expression processing (`.', `[', `\', `(', `), `*', `+', `?', `{', `|', `^', and `$') and forces them to be matched literally. This is consistent with how the `==' pattern matching operator treats quoted portions of its pattern argument. Since the treatment of quoted string arguments was changed, several issues have arisen, chief among them the problem of white space in pattern arguments and the differing treatment of quoted strings between bash-3.1 and bash-3.2. Both problems may be solved by using a shell variable to hold the pattern. Since word splitting is not performed when expanding shell variables in all operands of the [[ command, this allows users to quote patterns as they wish when assigning the variable, then expand the values to a single string that may contain whitespace. The first problem may be solved by using backslashes or any other quoting mechanism to escape the white space in the patterns. Bash-4.0 introduces the concept of a `compatibility level', controlled by several options to the `shopt' builtin. If the `compat31' option is enabled, bash reverts to the bash-3.1 behavior with respect to quoting the rhs of the =~ operator.
* init, fixes a bug in getSyslog() causing a SEGVAles Kozumplik2010-01-251-1/+1
| | | | The error could be triggered, e.g. by appending 'stage2=hd:label="fedora"' to the boot line.
* Prevent init from telling us its story if the shutdown was planned.Ales Kozumplik2010-01-221-0/+1
| | | | | | | | dcantrell's recent change to the shutdown mechanism exposed a situation when shutdown is correctly handled but the init's exit handler still gets called and dumps its callstack, that's not desirable because (in case of an exception in python code for example) part of the useful information on the screen is scrolled out of view.
* Introduces rsylogd to anaconda (part of #524980)Ales Kozumplik2010-01-202-157/+83
| | | | | | | A proper syslog daemon allows for remote logging that includes installed system's syslog. Removes the init.c code that simulated syslog activity until now.
* Fix compile problem from 65a3c05.Ales Kozumplik2010-01-203-5/+2
|
* Remove unnecessary free from the rpmextract error handlerMartin Sivak2010-01-201-1/+0
| | | | | | libarchive lacks proper documentation of its memory management, but it seems that it takes care of freeing the memory itself. So remove the bogus free from our code.
* Fix SIGSEGV in dlabel feature (#556390)Martin Sivak2010-01-202-7/+11
| | | | | | | | g_slist_alloc is only used to create a new entry in the list and it is not needed to initialize the list (it actually inserted an empty item at the beginning of the list). Also use strdup to store a copy of the device name (+ cleanup code)
* Handle reboot better on s390 (#533198)David Cantrell2010-01-191-31/+8
| | | | | | | | Do not install a handler for SIGINT on s390 since you can't really press Ctrl+Alt+Del there. Only handle SIGUSR1 and SIGUSR2 as handled in linuxrc.s390 (which comes from loader which comes from iutil.py). Always make sure we unmount filesystems and kill off processes, then reboot if that's requested or exit and display the halt message.
* Reset network setting input counters for IPv4 and IPv6 (#553761).David Cantrell2010-01-191-0/+2
| | | | | | | | | | | | | In loader, the manual TCP/IP configuration loop prompts the user then validates the input. Each time a particular piece of input validates, the counter is incremented. If the counter does not equal the right now at the end of the loop, we tell the user to re-enter the information correctly. But the second time around, if any input was valid the first time, it will be counted again, throwing off the count we check at the end of the loop. This patch just resets the IPv4 and IPv6 counters if we detected invalid input.
* Fix reading dasd status sysfs attribute (#536803).David Cantrell2010-01-191-1/+2
|
* Fix whitespace error that was introduced.Peter Jones2010-01-191-1/+1
|
* setStage2LocFromCmdline() shouldn't strdup so much.Peter Jones2010-01-191-34/+33
| | | | | | setStage2LocFromCmdline() doesn't need to do so much allocation just to use strtok(), which should be cleansed with fire anyway. Get rid of both.
* s390 CHPID types must be treated in hex for lookup table to work (#552844)Steffen Maier2010-01-182-4/+4
|
* Get rid of "stage2param" in parseCmdLineFlags(); it is unused.Peter Jones2010-01-151-4/+2
| | | | | stage2param is set in parseCmdLineFlags, but it's not actually used for anything. Take it out.
* Load all scsi_dh_* modules, since they can't be modprobe by aliases...Peter Jones2010-01-151-0/+51
| | | | | | Load all scsi_dh_* modules. These know how to bind to scsi targets, and which devices to bind to, so this will really only effect people with those targets.
* reIPL code cleanup in loaderDavid Cantrell2010-01-151-3/+2
| | | | | | | | Hans' fixes from this message: https://www.redhat.com/archives/anaconda-devel-list/2009-December/msg00473.html Keeping the code consistent across master, rhel6-branch, and rhel5-branch.
* Look for the SSH config file in /etc/ssh on s390 as well (#555691).Chris Lumens2010-01-151-1/+1
|
* Changed the architecture check from __ppc64__ to __powerpc64__ (#555669)Martin Gracik2010-01-152-2/+2
| | | | | | Wrong check for the architecture resulted in having no lib64 directories in LD_LIBRARY_PATH, and that's why anaconda couldn't find the libudev library.
* Fix the blkid infinite loop. (#555601)Martin Sivak2010-01-151-1/+1
|
* Testing mode was removed.Radek Vykydal2010-01-151-5/+3
|
* Fix path mistakes in dasd_settle() in loader/linuxrc.s390David Cantrell2010-01-131-3/+3
| | | | Fix some stupid path mistakes in dasd_settle().
* Poll DASD status for 'online' or 'unformatted' (#536803)David Cantrell2010-01-131-0/+24
| | | | | | Give DASD devices some time to enter the online or unformatted state before calling udevadm settle. Max duration of loop is suggestion from IBM.
* Add back hald for Xorg input device queries (#553780)David Cantrell2010-01-131-0/+8
| | | | | This reverts commit 13a27933b4b651317deec74cc80daaab5041d206. Xorg is asking hal for input device information.
* Fix implicit declaration of things in sys/stat.h.Chris Lumens2010-01-122-0/+2
|
* st_size is off64_t on i386, off_t on others.David Cantrell2010-01-081-0/+4
|
* fstat->st_size is a long unsigned int, not a long long unsigned int.David Cantrell2010-01-081-1/+1
|
* Use libarchive and rpm pkg-config files during build.David Cantrell2010-01-081-5/+4
|
* Use the updated DriverDisc code in loaderMartin Sivak2010-01-082-0/+47
| | | | | | - enable dlabel on RHEL by default - use glib's linked lists to traverse through available DDs - run depmod if the .rundepmod trigger file is present
* Backport the RHEL5 DriverDisc functionalityMartin Sivak2010-01-082-40/+316
| | | | And adapt it to use glib and better string handling functions
* Add rpm extraction routines (use librpm and libarchive)Martin Sivak2010-01-083-2/+374
|
generated by cgit (git 2.34.1) at 2025-08-30 16:34:01 +0000