Commit message (Collapse) | Author | Age | Files | Lines | |
---|---|---|---|---|---|
* | Tlb miss handling updates: need to fill in the processor id only once | Kanoj Sarcar | 2000-06-29 | 2 | -0/+3 |
| | | | | during bootup. | ||||
* | Add in the CONFIG_NUMA option for IP27s (distinct from DISCONTIGMEM which | Kanoj Sarcar | 2000-06-27 | 3 | -0/+3 |
| | | | | | might also be used by platforms with memory holes, but uniform access times). CONFIG_NUMA special code will probably make its way into generic kernel. | ||||
* | We don't need CONFIG_SGI for IP27 builds, specially when the SMP kernel | Kanoj Sarcar | 2000-06-27 | 3 | -3/+2 |
| | | | | | does not even compile ... restoring the IP27 specific config files to their previous versions. | ||||
* | Explain pgd/pmd allocation failure strategies. Create a global bad-pmd-table, | Kanoj Sarcar | 2000-06-27 | 3 | -47/+20 |
| | | | | | | | distinct from invalid-pmd-table for use during pmd allocation failure (so as to differentiate pgd_bad and pgd_none). The bad page table must have invalid pte entries to catch any user references to the range of virtual addresses it covers. | ||||
* | Fix another minor typo - get the global/special page symbols to point | Kanoj Sarcar | 2000-06-27 | 1 | -1/+1 |
| | | | | to their rightful locations. | ||||
* | Cleanup CONFIG_SGI_IP35. | Ralf Baechle | 2000-06-27 | 4 | -0/+5 |
| | |||||
* | Delete pmd_present/pgd_present. Explain pgd/pmd/pte allocation and | Kanoj Sarcar | 2000-06-26 | 3 | -7/+7 |
| | | | | | initialization. Make sure to allocate only one page for the page table (prevent memory leaks), since only one page is freed up. | ||||
* | Merge with Linux 2.4.0-test2. | Ralf Baechle | 2000-06-25 | 3 | -3/+0 |
| | |||||
* | Fix typo to make sure the page directory/tables get aligned on to page | Kanoj Sarcar | 2000-06-24 | 1 | -1/+1 |
| | | | | boundaries. | ||||
* | Sys32_fcntl from sparc to prevent procmail from confusing the kernel. | Kanoj Sarcar | 2000-06-23 | 2 | -1/+62 |
| | |||||
* | Fix struct stat64 in the 32-bit kernel and struct stat in the 64-bit | Ralf Baechle | 2000-06-20 | 2 | -30/+7 |
| | | | | | kernel to match each other and the the glibc definition. The glibc part of this change has been sent to Andreas. | ||||
* | Merge with 2.4.0-test1-ac21 + pile of MIPS cleanups to make merging | Ralf Baechle | 2000-06-19 | 10 | -50/+43 |
| | | | | | possible. Chainsawed RM200 kernel to compile again. Jazz machine status unknown. | ||||
* | Merge with 2.4.0-test1. | Ralf Baechle | 2000-06-16 | 2 | -12/+52 |
| | |||||
* | Merge with 2.3.99-pre9. | Ralf Baechle | 2000-06-15 | 3 | -3/+0 |
| | |||||
* | One more 32bit call. | Kanoj Sarcar | 2000-06-14 | 2 | -1/+45 |
| | |||||
* | The tlb mod/load/store exception handlers _have_ to go to do_page_fault, | Kanoj Sarcar | 2000-06-13 | 3 | -159/+15 |
| | | | | | and can not do with pte bit twiddling without grabbing page_table_lock. Reinstate most of the old code, after disposing of an extra jump. | ||||
* | Fix a possible race in the initialization code. | Kanoj Sarcar | 2000-06-13 | 1 | -8/+5 |
| | |||||
* | Support to put the lowlevel exception handlers in node local memory. | Kanoj Sarcar | 2000-06-13 | 4 | -0/+28 |
| | |||||
* | When doing kernel text replication, make sure we do not scribble on | Kanoj Sarcar | 2000-06-13 | 2 | -21/+27 |
| | | | | | top of the replicated kernel text thinking it is unused memory. This lets us get to multiuser on a replicated kernel text system. | ||||
* | Add in a build/config option for kernel text replication. | Kanoj Sarcar | 2000-06-12 | 4 | -0/+6 |
| | |||||
* | Fix the code that puts in (possibly different) tlblo0/tlblo1 entries | Kanoj Sarcar | 2000-06-12 | 1 | -3/+5 |
| | | | | for the mapped kernel. Compute the "pfn" field properly. | ||||
* | Have the slave bootstrap code actually look at the text and data server | Kanoj Sarcar | 2000-06-09 | 2 | -13/+14 |
| | | | | | nasids that the master nasid found for it. Sanitize the replication procedures. | ||||
* | Add in the hooks to do kernel text replication. Now start debugging this | Kanoj Sarcar | 2000-06-09 | 3 | -1/+124 |
| | | | | code. | ||||
* | Each node needs to have a structure that the master nasid fills up to | Kanoj Sarcar | 2000-06-09 | 1 | -1/+0 |
| | | | | | tell it where to get the kernel data from (ie, which nasid holds the kernel data). Remove debug cruft from head.S. | ||||
* | Need to have the nasid of the boot/master node for future uses. | Kanoj Sarcar | 2000-06-09 | 1 | -0/+2 |
| | |||||
* | Step one of trying to remove the assumption that nasid 0 is present | Kanoj Sarcar | 2000-06-08 | 1 | -3/+33 |
| | | | | from the mapped kernel code. | ||||
* | SMP Mapped kernel fixes to go along with the new layout. Use properly | Kanoj Sarcar | 2000-06-08 | 1 | -1/+1 |
| | | | | | massaged bootstrap address so that the PRM launches the slaves into legal code. | ||||
* | Mapped kernel changes: create elf sections as close as possible to | Kanoj Sarcar | 2000-06-08 | 4 | -33/+30 |
| | | | | | IRIX, to get the same kind of loader behavior. Read comments in mapped_kernel.h for more details. | ||||
* | Reboot was doing halt and halt was crashing the machine, fixed. | Ralf Baechle | 2000-06-07 | 1 | -6/+13 |
| | |||||
* | *** empty log message *** | Ralf Baechle | 2000-06-06 | 1 | -12/+13 |
| | |||||
* | Get the mapped kernel working for SMP kernels. While launching slaves, | Kanoj Sarcar | 2000-06-06 | 1 | -3/+4 |
| | | | | | pass the proper address for the slave bootstrap routine based on where the kernel is compiled at. | ||||
* | Get some type definitions into their rightful places for wider usage. | Kanoj Sarcar | 2000-06-06 | 1 | -4/+5 |
| | |||||
* | MAke sure the mapped kernel does not loose its prom command line. | Kanoj Sarcar | 2000-06-06 | 1 | -1/+0 |
| | |||||
* | Entrylo1 must at least be marked global for the tlb entry mapping the | Kanoj Sarcar | 2000-06-06 | 1 | -1/+2 |
| | | | | mapped kernel text/data for tlb translation to work properly. | ||||
* | The tlbwired register is now set up in the head.S code. | Kanoj Sarcar | 2000-06-06 | 3 | -365/+115 |
| | |||||
* | Initial rudimentary support for mapped kernel. The kernel text and data | Kanoj Sarcar | 2000-06-06 | 6 | -89/+381 |
| | | | | | | are placed in the cksseg area, and such behavior is tuned thru a make option (off by default). Basis for future kernel text replication and partitioning work/study. | ||||
* | Launch the CPUs one by one until we can make sure that we | Ulf Carlsson | 2000-06-06 | 1 | -7/+11 |
| | | | | | add PCI devices from different PCI buses to the list of PCI devices in the same order every time we boot the system. | ||||
* | Delete unused reference to linker generated symbol. | Kanoj Sarcar | 2000-06-05 | 1 | -2/+0 |
| | |||||
* | Further cleanups to the ld script files not to do more than needed. | Kanoj Sarcar | 2000-06-05 | 4 | -12/+4 |
| | | | | Replace uses of _ftext with the more generic _stext. | ||||
* | Clean up generation of unused symbols. | Kanoj Sarcar | 2000-06-05 | 2 | -6/+0 |
| | |||||
* | Fix the address twiddling. For some reason, the code that I was expecting | Kanoj Sarcar | 2000-06-05 | 1 | -11/+10 |
| | | | | | was not getting generated (coding problem or cpp/as problem, no point tracking this down). | ||||
* | The master processor now waits for all enabled cpus to come into the | Kanoj Sarcar | 2000-06-02 | 1 | -1/+1 |
| | | | | kernel before progressing (as opposed to waiting for _all_ slave cpus). | ||||
* | Most of the kernel arrays/data structures and low level procedures | Kanoj Sarcar | 2000-06-02 | 3 | -27/+28 |
| | | | | | now accept the kernel assigned cpuid (instead of the prom assigned id) as input/index. The only exception is the early boot up code. | ||||
* | Set smp_num_cpus in the IP27 specific code. Getting ready to handle | Kanoj Sarcar | 2000-06-02 | 2 | -3/+1 |
| | | | | PROM-disabled cpus. | ||||
* | Fix some type definitions, add a new field in the per-processor data | Kanoj Sarcar | 2000-06-02 | 1 | -1/+1 |
| | | | | | area to store the PROM assigned cpuid that is used by most of the IP27 kernel and prom data strucutures. | ||||
* | Add the node offset to the IOC3 to the PCI resource | Ulf Carlsson | 2000-06-01 | 1 | -0/+4 |
| | | | | | address for all IOC3 devices. We will now be able to reach IOC3 cards on other nodes than the first one. | ||||
* | Now that __kernel_clock_t is 64 bits, we need to make sure we have a | Kanoj Sarcar | 2000-06-01 | 2 | -1/+30 |
| | | | | | | 32 bit version of the sys_times() call (like sparc64, ia64), else the kernel overwrites user space it is not supposed to. Compilations could not be done because of this. | ||||
* | Only enter do_timer from the first CPU. | Ulf Carlsson | 2000-05-31 | 1 | -16/+16 |
| | | | | | Per processor interrupts at 100 Hz on each processor. Adjust do_gettimeofday to get info from first CPU. | ||||
* | Remove all IOC3 hacks. IOC3 now allocates it's resources the same way | Ralf Baechle | 2000-05-30 | 2 | -71/+79 |
| | | | | | that all others do. Further cleanup to make ioc3-eth.c start looking like a driver. | ||||
* | Don't make Origin SMP as experimental. | Ralf Baechle | 2000-05-30 | 1 | -1/+1 |
| |