Messages in this thread | | | From | Martin Schwidefsky <> | Subject | [PATCH] 2.5.38 s390 fixes. | Date | Tue, 24 Sep 2002 19:26:02 +0200 |
| |
Hi Linus, I put together a set of patches for s/390 again. There are 24 patches, each starting with a small description. To get a useful s390/s390x kernel at least the first 5 patches are needed.
The list of patches and some comments:
01_minimal: Minimal patch to get the kernel without any device drivers compiled and working again. 02_syscalls: Add the new system calls. 03_partition: Get the bloody ibm partition code to work. 04_config: I removed some unnecessary config options and restructure the configuration menu a bit. 05_dasd: I picked up Al Viros changes to the dasd driver and made it even better. 06_xpram: No comment. 07_tasklet: Get rid of the bottom halves in the s/390 device drivers. 08_bitops: A bug fix for the (unused) option ALIGN_CS == 1. 09_emu31: Latest 31 bit emulation fixes. 10_vmlinux: Unify the two linker scripts. 11_preempt: Add preemption support. It still complains about about "scheduling while atomic" though. 12_inline: Inline all tcpip checksum functions and optimize xchg. 13_diag44: Make use of the diag 0x44 in 64 bit spinlock code. The diag 0x44 yields a virtual cpu under VM and LPAR. 14_time: I removed the dependency on the boot cpu in the timer interrupt. This makes it easier with the timer patch and with switching cpus on and off. 15_beauty: Some beautification. 16_fpu: Optimization for loading/storing of fpu registers. Got rid of another stupid file in arch/s390/kernel. 17_ptrace: Make ptrace readable. 18_quiesce: Add signal quiesce support. 19_syncisc: Fix and simplify synchronous i/o functions. 20_ending: Simplify s390_process_IRQ. 21_chpid: Do the right thing if channel paths are not available. 22_boot: Reworked boot sequence. 23_boot_common:Remove s390 specific init call from init/main.c. 22_boot and 23_boot_common are related. 24_proc_misc: /proc/interrupts does not make sense on s390. There are 65536 i/o and 65536 external interrupts. We do not really want to have a file with #cpus * 128K zeros.
blue skies, Martin.
- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/
| |