* [PATCH 1/2] um: Use generic io.h @ 2012-01-25 17:23 Richard Weinberger 2012-01-25 17:23 ` [PATCH 2/2] um: Serve io_remap_pfn_range() Richard Weinberger 2012-01-30 18:23 ` [PATCH 1/2] um: Use generic io.h Al Viro 0 siblings, 2 replies; 17+ messages in thread From: Richard Weinberger @ 2012-01-25 17:23 UTC (permalink / raw) To: user-mode-linux-devel; +Cc: linux-kernel, Richard Weinberger There is no need to implement our own basic io functions. All we need exists already in asm-generic/io.h Signed-off-by: Richard Weinberger <richard@nod.at> --- arch/um/include/asm/io.h | 53 +--------------------------------------------- 1 files changed, 1 insertions(+), 52 deletions(-) diff --git a/arch/um/include/asm/io.h b/arch/um/include/asm/io.h index 44e8b8c..cff48f0 100644 --- a/arch/um/include/asm/io.h +++ b/arch/um/include/asm/io.h @@ -1,57 +1,6 @@ #ifndef __UM_IO_H #define __UM_IO_H -#include "asm/page.h" - -#define IO_SPACE_LIMIT 0xdeadbeef /* Sure hope nothing uses this */ - -static inline int inb(unsigned long i) { return(0); } -static inline void outb(char c, unsigned long i) { } - -/* - * Change virtual addresses to physical addresses and vv. - * These are pretty trivial - */ -static inline unsigned long virt_to_phys(volatile void * address) -{ - return __pa((void *) address); -} - -static inline void * phys_to_virt(unsigned long address) -{ - return __va(address); -} - -/* - * Convert a physical pointer to a virtual kernel pointer for /dev/mem - * access - */ -#define xlate_dev_mem_ptr(p) __va(p) - -/* - * Convert a virtual cached pointer to an uncached pointer - */ -#define xlate_dev_kmem_ptr(p) p - -static inline void writeb(unsigned char b, volatile void __iomem *addr) -{ - *(volatile unsigned char __force *) addr = b; -} -static inline void writew(unsigned short b, volatile void __iomem *addr) -{ - *(volatile unsigned short __force *) addr = b; -} -static inline void writel(unsigned int b, volatile void __iomem *addr) -{ - *(volatile unsigned int __force *) addr = b; -} -static inline void writeq(unsigned int b, volatile void __iomem *addr) -{ - *(volatile unsigned long long __force *) addr = b; -} -#define __raw_writeb writeb -#define __raw_writew writew -#define __raw_writel writel -#define __raw_writeq writeq +#include <asm-generic/io.h> #endif -- 1.7.7 ^ permalink raw reply related [flat|nested] 17+ messages in thread
* [PATCH 2/2] um: Serve io_remap_pfn_range() 2012-01-25 17:23 [PATCH 1/2] um: Use generic io.h Richard Weinberger @ 2012-01-25 17:23 ` Richard Weinberger 2012-01-30 18:23 ` [PATCH 1/2] um: Use generic io.h Al Viro 1 sibling, 0 replies; 17+ messages in thread From: Richard Weinberger @ 2012-01-25 17:23 UTC (permalink / raw) To: user-mode-linux-devel; +Cc: linux-kernel, Richard Weinberger At some places io_remap_pfn_range() is needed. UML has to serve it like all other archs do. Signed-off-by: Richard Weinberger <richard@nod.at> --- arch/um/include/asm/pgtable.h | 2 ++ 1 files changed, 2 insertions(+), 0 deletions(-) diff --git a/arch/um/include/asm/pgtable.h b/arch/um/include/asm/pgtable.h index 41474fb..6a3f984 100644 --- a/arch/um/include/asm/pgtable.h +++ b/arch/um/include/asm/pgtable.h @@ -69,6 +69,8 @@ extern unsigned long end_iomem; #define PAGE_KERNEL __pgprot(_PAGE_PRESENT | _PAGE_RW | _PAGE_DIRTY | _PAGE_ACCESSED) #define PAGE_KERNEL_EXEC __pgprot(__PAGE_KERNEL_EXEC) +#define io_remap_pfn_range remap_pfn_range + /* * The i386 can't do page protection for execute, and considers that the same * are read. -- 1.7.7 ^ permalink raw reply related [flat|nested] 17+ messages in thread
* Re: [PATCH 1/2] um: Use generic io.h 2012-01-25 17:23 [PATCH 1/2] um: Use generic io.h Richard Weinberger 2012-01-25 17:23 ` [PATCH 2/2] um: Serve io_remap_pfn_range() Richard Weinberger @ 2012-01-30 18:23 ` Al Viro 2012-01-30 19:10 ` Richard Weinberger 2012-01-30 19:25 ` Al Viro 1 sibling, 2 replies; 17+ messages in thread From: Al Viro @ 2012-01-30 18:23 UTC (permalink / raw) To: Richard Weinberger; +Cc: user-mode-linux-devel, linux-kernel On Wed, Jan 25, 2012 at 06:23:53PM +0100, Richard Weinberger wrote: > There is no need to implement our own basic io functions. > All we need exists already in asm-generic/io.h Then just do git rm arch/um/include/asm/io.h echo 'generic-y += io.h' >>arch/um/include/asm/Kbuild git add arch/um/include/asm/Kbuild and be done with that. We are already using generic-y for a bunch of headers there... ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH 1/2] um: Use generic io.h 2012-01-30 18:23 ` [PATCH 1/2] um: Use generic io.h Al Viro @ 2012-01-30 19:10 ` Richard Weinberger 2012-01-30 19:25 ` Al Viro 1 sibling, 0 replies; 17+ messages in thread From: Richard Weinberger @ 2012-01-30 19:10 UTC (permalink / raw) To: Al Viro; +Cc: user-mode-linux-devel, linux-kernel Am 30.01.2012 19:23, schrieb Al Viro: > On Wed, Jan 25, 2012 at 06:23:53PM +0100, Richard Weinberger wrote: >> There is no need to implement our own basic io functions. >> All we need exists already in asm-generic/io.h > > Then just do > > git rm arch/um/include/asm/io.h > echo 'generic-y += io.h'>>arch/um/include/asm/Kbuild > git add arch/um/include/asm/Kbuild > > and be done with that. We are already using generic-y for a bunch of > headers there... Will do. Thanks, //richard ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH 1/2] um: Use generic io.h 2012-01-30 18:23 ` [PATCH 1/2] um: Use generic io.h Al Viro 2012-01-30 19:10 ` Richard Weinberger @ 2012-01-30 19:25 ` Al Viro 2012-01-30 19:32 ` Al Viro 2012-02-03 23:17 ` [PATCH] um: Cleanup headers files Richard Weinberger 1 sibling, 2 replies; 17+ messages in thread From: Al Viro @ 2012-01-30 19:25 UTC (permalink / raw) To: Richard Weinberger; +Cc: user-mode-linux-devel, linux-kernel On Mon, Jan 30, 2012 at 06:23:33PM +0000, Al Viro wrote: > On Wed, Jan 25, 2012 at 06:23:53PM +0100, Richard Weinberger wrote: > > There is no need to implement our own basic io functions. > > All we need exists already in asm-generic/io.h > > Then just do > > git rm arch/um/include/asm/io.h > echo 'generic-y += io.h' >>arch/um/include/asm/Kbuild > git add arch/um/include/asm/Kbuild > > and be done with that. We are already using generic-y for a bunch of > headers there... While we are at it, param.h and pci.h appear to be candidates for the same treatment, along with delay.h and auxvec.h. mutex.h as well... BTW, we probably can kill arch/um/include/asm/asm-offsets.h - everything in there is using generated/asm-offsets.h directly. FWIW, I'd probably kill current.h as well - replace get_current() in arch/um/kernel/process.c with get_current_pid() (i.e. task_pid_nr(current)), kill HOST_TASK_PID and switch arch/x86/um/bugs_32.c to use of get_current_pid() and we can switch to asm-generic/current.h just fine. ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH 1/2] um: Use generic io.h 2012-01-30 19:25 ` Al Viro @ 2012-01-30 19:32 ` Al Viro 2012-01-30 19:57 ` Richard Weinberger 2012-02-03 23:17 ` [PATCH] um: Cleanup headers files Richard Weinberger 1 sibling, 1 reply; 17+ messages in thread From: Al Viro @ 2012-01-30 19:32 UTC (permalink / raw) To: Richard Weinberger; +Cc: user-mode-linux-devel, linux-kernel On Mon, Jan 30, 2012 at 07:25:52PM +0000, Al Viro wrote: > While we are at it, param.h and pci.h appear to be candidates for the > same treatment, along with delay.h and auxvec.h. mutex.h as well... > BTW, we probably can kill arch/um/include/asm/asm-offsets.h - everything > in there is using generated/asm-offsets.h directly. > > FWIW, I'd probably kill current.h as well - replace get_current() in > arch/um/kernel/process.c with get_current_pid() (i.e. task_pid_nr(current)), > kill HOST_TASK_PID and switch arch/x86/um/bugs_32.c to use of get_current_pid() > and we can switch to asm-generic/current.h just fine. BTW, while looking at those headers - why do we need asm/user.h pulled from arch/um/include/asm/ptrace-generic.h? ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH 1/2] um: Use generic io.h 2012-01-30 19:32 ` Al Viro @ 2012-01-30 19:57 ` Richard Weinberger 0 siblings, 0 replies; 17+ messages in thread From: Richard Weinberger @ 2012-01-30 19:57 UTC (permalink / raw) To: Al Viro; +Cc: user-mode-linux-devel, linux-kernel Am 30.01.2012 20:32, schrieb Al Viro: > On Mon, Jan 30, 2012 at 07:25:52PM +0000, Al Viro wrote: > >> While we are at it, param.h and pci.h appear to be candidates for the >> same treatment, along with delay.h and auxvec.h. mutex.h as well... >> BTW, we probably can kill arch/um/include/asm/asm-offsets.h - everything >> in there is using generated/asm-offsets.h directly. >> >> FWIW, I'd probably kill current.h as well - replace get_current() in >> arch/um/kernel/process.c with get_current_pid() (i.e. task_pid_nr(current)), >> kill HOST_TASK_PID and switch arch/x86/um/bugs_32.c to use of get_current_pid() >> and we can switch to asm-generic/current.h just fine. > > BTW, while looking at those headers - why do we need asm/user.h pulled from > arch/um/include/asm/ptrace-generic.h? Hmmm, looks like a relict. Let's remove it. Thanks, //richard ^ permalink raw reply [flat|nested] 17+ messages in thread
* [PATCH] um: Cleanup headers files 2012-01-30 19:25 ` Al Viro 2012-01-30 19:32 ` Al Viro @ 2012-02-03 23:17 ` Richard Weinberger 2012-02-04 0:14 ` Al Viro 1 sibling, 1 reply; 17+ messages in thread From: Richard Weinberger @ 2012-02-03 23:17 UTC (permalink / raw) To: viro; +Cc: user-mode-linux-devel, linux-kernel, Richard Weinberger Many header files can be replaced by generic variants. Signed-off-by: Richard Weinberger <richard@nod.at> --- arch/um/include/asm/Kbuild | 3 +- arch/um/include/asm/auxvec.h | 4 -- arch/um/include/asm/current.h | 13 ------- arch/um/include/asm/delay.h | 18 ---------- arch/um/include/asm/io.h | 57 ------------------------------- arch/um/include/asm/mutex.h | 9 ----- arch/um/include/asm/param.h | 20 ----------- arch/um/include/asm/pci.h | 6 --- arch/um/include/shared/common-offsets.h | 2 - arch/um/include/shared/kern_util.h | 2 +- arch/um/kernel/process.c | 4 +- arch/x86/um/bugs_32.c | 4 +-- 12 files changed, 6 insertions(+), 136 deletions(-) delete mode 100644 arch/um/include/asm/auxvec.h delete mode 100644 arch/um/include/asm/current.h delete mode 100644 arch/um/include/asm/delay.h delete mode 100644 arch/um/include/asm/io.h delete mode 100644 arch/um/include/asm/mutex.h delete mode 100644 arch/um/include/asm/param.h delete mode 100644 arch/um/include/asm/pci.h diff --git a/arch/um/include/asm/Kbuild b/arch/um/include/asm/Kbuild index 451f451..00b6fe8 100644 --- a/arch/um/include/asm/Kbuild +++ b/arch/um/include/asm/Kbuild @@ -1,3 +1,4 @@ generic-y += bug.h cputime.h device.h emergency-restart.h futex.h hardirq.h generic-y += hw_irq.h irq_regs.h kdebug.h percpu.h sections.h topology.h xor.h -generic-y += ftrace.h +generic-y += ftrace.h io.h param.h param.h pci.h delay.h mutex.h auxvec.h +generic-y += current.h diff --git a/arch/um/include/asm/auxvec.h b/arch/um/include/asm/auxvec.h deleted file mode 100644 index 1e5e1c2..0000000 --- a/arch/um/include/asm/auxvec.h +++ /dev/null @@ -1,4 +0,0 @@ -#ifndef __UM_AUXVEC_H -#define __UM_AUXVEC_H - -#endif diff --git a/arch/um/include/asm/current.h b/arch/um/include/asm/current.h deleted file mode 100644 index c2191d9..0000000 --- a/arch/um/include/asm/current.h +++ /dev/null @@ -1,13 +0,0 @@ -/* - * Copyright (C) 2000 - 2007 Jeff Dike (jdike@{addtoit,linux.intel}.com) - * Licensed under the GPL - */ - -#ifndef __UM_CURRENT_H -#define __UM_CURRENT_H - -#include "linux/thread_info.h" - -#define current (current_thread_info()->task) - -#endif diff --git a/arch/um/include/asm/delay.h b/arch/um/include/asm/delay.h deleted file mode 100644 index 8a5576d..0000000 --- a/arch/um/include/asm/delay.h +++ /dev/null @@ -1,18 +0,0 @@ -#ifndef __UM_DELAY_H -#define __UM_DELAY_H - -/* Undefined on purpose */ -extern void __bad_udelay(void); -extern void __bad_ndelay(void); - -extern void __udelay(unsigned long usecs); -extern void __ndelay(unsigned long usecs); -extern void __delay(unsigned long loops); - -#define udelay(n) ((__builtin_constant_p(n) && (n) > 20000) ? \ - __bad_udelay() : __udelay(n)) - -#define ndelay(n) ((__builtin_constant_p(n) && (n) > 20000) ? \ - __bad_ndelay() : __ndelay(n)) - -#endif diff --git a/arch/um/include/asm/io.h b/arch/um/include/asm/io.h deleted file mode 100644 index 44e8b8c..0000000 --- a/arch/um/include/asm/io.h +++ /dev/null @@ -1,57 +0,0 @@ -#ifndef __UM_IO_H -#define __UM_IO_H - -#include "asm/page.h" - -#define IO_SPACE_LIMIT 0xdeadbeef /* Sure hope nothing uses this */ - -static inline int inb(unsigned long i) { return(0); } -static inline void outb(char c, unsigned long i) { } - -/* - * Change virtual addresses to physical addresses and vv. - * These are pretty trivial - */ -static inline unsigned long virt_to_phys(volatile void * address) -{ - return __pa((void *) address); -} - -static inline void * phys_to_virt(unsigned long address) -{ - return __va(address); -} - -/* - * Convert a physical pointer to a virtual kernel pointer for /dev/mem - * access - */ -#define xlate_dev_mem_ptr(p) __va(p) - -/* - * Convert a virtual cached pointer to an uncached pointer - */ -#define xlate_dev_kmem_ptr(p) p - -static inline void writeb(unsigned char b, volatile void __iomem *addr) -{ - *(volatile unsigned char __force *) addr = b; -} -static inline void writew(unsigned short b, volatile void __iomem *addr) -{ - *(volatile unsigned short __force *) addr = b; -} -static inline void writel(unsigned int b, volatile void __iomem *addr) -{ - *(volatile unsigned int __force *) addr = b; -} -static inline void writeq(unsigned int b, volatile void __iomem *addr) -{ - *(volatile unsigned long long __force *) addr = b; -} -#define __raw_writeb writeb -#define __raw_writew writew -#define __raw_writel writel -#define __raw_writeq writeq - -#endif diff --git a/arch/um/include/asm/mutex.h b/arch/um/include/asm/mutex.h deleted file mode 100644 index 458c1f7..0000000 --- a/arch/um/include/asm/mutex.h +++ /dev/null @@ -1,9 +0,0 @@ -/* - * Pull in the generic implementation for the mutex fastpath. - * - * TODO: implement optimized primitives instead, or leave the generic - * implementation in place, or pick the atomic_xchg() based generic - * implementation. (see asm-generic/mutex-xchg.h for details) - */ - -#include <asm-generic/mutex-dec.h> diff --git a/arch/um/include/asm/param.h b/arch/um/include/asm/param.h deleted file mode 100644 index e44f4e6..0000000 --- a/arch/um/include/asm/param.h +++ /dev/null @@ -1,20 +0,0 @@ -#ifndef _UM_PARAM_H -#define _UM_PARAM_H - -#define EXEC_PAGESIZE 4096 - -#ifndef NOGROUP -#define NOGROUP (-1) -#endif - -#define MAXHOSTNAMELEN 64 /* max length of hostname */ - -#ifdef __KERNEL__ -#define HZ CONFIG_HZ -#define USER_HZ 100 /* .. some user interfaces are in "ticks" */ -#define CLOCKS_PER_SEC (USER_HZ) /* frequency at which times() counts */ -#else -#define HZ 100 -#endif - -#endif diff --git a/arch/um/include/asm/pci.h b/arch/um/include/asm/pci.h deleted file mode 100644 index b44cf59..0000000 --- a/arch/um/include/asm/pci.h +++ /dev/null @@ -1,6 +0,0 @@ -#ifndef __UM_PCI_H -#define __UM_PCI_H - -#define PCI_DMA_BUS_IS_PHYS (1) - -#endif diff --git a/arch/um/include/shared/common-offsets.h b/arch/um/include/shared/common-offsets.h index d7fe563..40db8f7 100644 --- a/arch/um/include/shared/common-offsets.h +++ b/arch/um/include/shared/common-offsets.h @@ -2,8 +2,6 @@ DEFINE(KERNEL_MADV_REMOVE, MADV_REMOVE); -OFFSET(HOST_TASK_PID, task_struct, pid); - DEFINE(UM_KERN_PAGE_SIZE, PAGE_SIZE); DEFINE(UM_KERN_PAGE_MASK, PAGE_MASK); DEFINE(UM_KERN_PAGE_SHIFT, PAGE_SHIFT); diff --git a/arch/um/include/shared/kern_util.h b/arch/um/include/shared/kern_util.h index 0f14838..d7eccd2 100644 --- a/arch/um/include/shared/kern_util.h +++ b/arch/um/include/shared/kern_util.h @@ -48,7 +48,7 @@ extern void do_uml_exitcalls(void); * GFP_ATOMIC. */ extern int __cant_sleep(void); -extern void *get_current(void); +extern pid_t get_current_pid(void); extern int copy_from_user_proc(void *to, void *from, int size); extern int cpu(void); extern char *uml_strdup(const char *string); diff --git a/arch/um/kernel/process.c b/arch/um/kernel/process.c index 69f2490..9376c24 100644 --- a/arch/um/kernel/process.c +++ b/arch/um/kernel/process.c @@ -126,9 +126,9 @@ void exit_thread(void) { } -void *get_current(void) +pid_t get_current_pid(void) { - return current; + return task_pid_nr(current); } /* diff --git a/arch/x86/um/bugs_32.c b/arch/x86/um/bugs_32.c index a1fba5f..82752bc 100644 --- a/arch/x86/um/bugs_32.c +++ b/arch/x86/um/bugs_32.c @@ -13,8 +13,6 @@ static int host_has_cmov = 1; static jmp_buf cmov_test_return; -#define TASK_PID(task) *((int *) &(((char *) (task))[HOST_TASK_PID])) - static void cmov_sigill_test_handler(int sig) { host_has_cmov = 0; @@ -51,7 +49,7 @@ void arch_examine_signal(int sig, struct uml_pt_regs *regs) * This is testing for a cmov (0x0f 0x4x) instruction causing a * SIGILL in init. */ - if ((sig != SIGILL) || (TASK_PID(get_current()) != 1)) + if ((sig != SIGILL) || get_current_pid() != 1) return; if (copy_from_user_proc(tmp, (void *) UPT_IP(regs), 2)) { -- 1.7.7.3 ^ permalink raw reply related [flat|nested] 17+ messages in thread
* Re: [PATCH] um: Cleanup headers files 2012-02-03 23:17 ` [PATCH] um: Cleanup headers files Richard Weinberger @ 2012-02-04 0:14 ` Al Viro 2012-02-04 0:26 ` Richard Weinberger ` (2 more replies) 0 siblings, 3 replies; 17+ messages in thread From: Al Viro @ 2012-02-04 0:14 UTC (permalink / raw) To: Richard Weinberger; +Cc: user-mode-linux-devel, linux-kernel On Sat, Feb 04, 2012 at 12:17:43AM +0100, Richard Weinberger wrote: > Many header files can be replaced by generic variants. Umm... I suspect that for auxvec.h the right thing is to simply remove it - it's never used on uml builds. FWIW, I've rebased my local queue to current and pushed it to kernel.org again; it's on git.kernel.org/pub/scm/linux/kernel/git/viro/um-headers.git ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] um: Cleanup headers files 2012-02-04 0:14 ` Al Viro @ 2012-02-04 0:26 ` Richard Weinberger 2012-02-04 0:44 ` Richard Weinberger 2012-02-04 14:05 ` Richard Weinberger 2 siblings, 0 replies; 17+ messages in thread From: Richard Weinberger @ 2012-02-04 0:26 UTC (permalink / raw) To: Al Viro; +Cc: user-mode-linux-devel, linux-kernel [-- Attachment #1: Type: text/plain, Size: 393 bytes --] Am 04.02.2012 01:14, schrieb Al Viro: > On Sat, Feb 04, 2012 at 12:17:43AM +0100, Richard Weinberger wrote: >> Many header files can be replaced by generic variants. > > FWIW, I've rebased my local queue to current and pushed it to kernel.org > again; it's on git.kernel.org/pub/scm/linux/kernel/git/viro/um-headers.git Hmm ok, you did already the cleanup... Thanks, //richard [-- Attachment #2: OpenPGP digital signature --] [-- Type: application/pgp-signature, Size: 490 bytes --] ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] um: Cleanup headers files 2012-02-04 0:14 ` Al Viro 2012-02-04 0:26 ` Richard Weinberger @ 2012-02-04 0:44 ` Richard Weinberger 2012-02-12 20:27 ` Al Viro 2012-02-04 14:05 ` Richard Weinberger 2 siblings, 1 reply; 17+ messages in thread From: Richard Weinberger @ 2012-02-04 0:44 UTC (permalink / raw) To: Al Viro; +Cc: user-mode-linux-devel, linux-kernel [-- Attachment #1: Type: text/plain, Size: 595 bytes --] Am 04.02.2012 01:14, schrieb Al Viro: > On Sat, Feb 04, 2012 at 12:17:43AM +0100, Richard Weinberger wrote: >> Many header files can be replaced by generic variants. > > Umm... I suspect that for auxvec.h the right thing is to simply > remove it - it's never used on uml builds. > > FWIW, I've rebased my local queue to current and pushed it to kernel.org > again; it's on git.kernel.org/pub/scm/linux/kernel/git/viro/um-headers.git So, I pulled your work and merged it into my shiny new UML tree: git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml.git Thanks, //richard [-- Attachment #2: OpenPGP digital signature --] [-- Type: application/pgp-signature, Size: 490 bytes --] ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] um: Cleanup headers files 2012-02-04 0:44 ` Richard Weinberger @ 2012-02-12 20:27 ` Al Viro 2012-02-12 21:01 ` Richard Weinberger 0 siblings, 1 reply; 17+ messages in thread From: Al Viro @ 2012-02-12 20:27 UTC (permalink / raw) To: Richard Weinberger; +Cc: user-mode-linux-devel, linux-kernel On Sat, Feb 04, 2012 at 01:44:42AM +0100, Richard Weinberger wrote: > Am 04.02.2012 01:14, schrieb Al Viro: > > On Sat, Feb 04, 2012 at 12:17:43AM +0100, Richard Weinberger wrote: > >> Many header files can be replaced by generic variants. > > > > Umm... I suspect that for auxvec.h the right thing is to simply > > remove it - it's never used on uml builds. > > > > FWIW, I've rebased my local queue to current and pushed it to kernel.org > > again; it's on git.kernel.org/pub/scm/linux/kernel/git/viro/um-headers.git > > So, I pulled your work and merged it into my shiny new UML tree: > git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml.git Hmm... I would really, really recommend you to clean the git history in that tree up. As it is, you have * merge of that branch from um-headers * um: Switch to large mcmodel on x86_64 * um: Serve io_remap_pfn_range() * 205 merges from Linus, from mid-April to late January. * mismerge from Linus on April 14; the diff between the result of that merge and its mainline parent is diff --git a/drivers/platform/x86/intel_pmic_gpio.c b/drivers/platform/x86/intel_pmic_gpio.c index 753bdb3..464bb3f 100644 --- a/drivers/platform/x86/intel_pmic_gpio.c +++ b/drivers/platform/x86/intel_pmic_gpio.c @@ -208,8 +208,8 @@ static struct irq_chip pmic_irqchip = { .name = "PMIC-GPIO", .irq_mask = pmic_irq_mask, .irq_unmask = pmic_irq_unmask, - .irq_set_type = pmic_irq_type, <<<<<<< HEAD - .irq_bus_lock = pmic_irq_buslock, + .irq_set_type = pmic_irq_type, + .irq_bus_lock = pmic_bus_lock, .irq_bus_sync_unlock = pmic_bus_sync_unlock, }; and that mangling is still in your tree * a bunch of cherry-picked (or manually applied?) commits from mainline, done on April 12-13 * 23 more backmerges * a _lot_ of cherry-picks circa March 12 * 91 more backmerges * backmerge with a conflict; note that result of merge is identical to the mainline parent of it (August 2010) * 5 more backmerges (once a week or so) * your "um: call free_irq() only on enabled channels", applied by Linus (over mail, presumably) a couple of weeks later * 7 more backmerges * your variant of removal of unused variable in apic/es7000_32.c; that's what caused conflict resolved by Aug 2010 merge. As it is, you have mainline with mangled intel_pmic_gpio.c plus two recent um commits plus the merge from um-headers. _Please_, replace that with mainline + the same two commits + merge. And don't do the compulsive backmerging like that... ^ permalink raw reply related [flat|nested] 17+ messages in thread
* Re: [PATCH] um: Cleanup headers files 2012-02-12 20:27 ` Al Viro @ 2012-02-12 21:01 ` Richard Weinberger 2012-02-12 21:20 ` Al Viro 0 siblings, 1 reply; 17+ messages in thread From: Richard Weinberger @ 2012-02-12 21:01 UTC (permalink / raw) To: Al Viro; +Cc: user-mode-linux-devel, linux-kernel Am 12.02.2012 21:27, schrieb Al Viro: >> So, I pulled your work and merged it into my shiny new UML tree: >> git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml.git > > Hmm... I would really, really recommend you to clean the git history in that > tree up. As it is, you have > * merge of that branch from um-headers > * um: Switch to large mcmodel on x86_64 > * um: Serve io_remap_pfn_range() > * 205 merges from Linus, from mid-April to late January. > * mismerge from Linus on April 14; the diff between the result Okay, I'll create a new tree. This one is definitely screwed. Sorry for the inconvenience. Thanks, //richard ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] um: Cleanup headers files 2012-02-12 21:01 ` Richard Weinberger @ 2012-02-12 21:20 ` Al Viro 2012-02-12 21:26 ` Richard Weinberger 0 siblings, 1 reply; 17+ messages in thread From: Al Viro @ 2012-02-12 21:20 UTC (permalink / raw) To: Richard Weinberger; +Cc: user-mode-linux-devel, linux-kernel On Sun, Feb 12, 2012 at 10:01:49PM +0100, Richard Weinberger wrote: > Am 12.02.2012 21:27, schrieb Al Viro: > >>So, I pulled your work and merged it into my shiny new UML tree: > >>git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml.git > > > >Hmm... I would really, really recommend you to clean the git history in that > >tree up. As it is, you have > > * merge of that branch from um-headers > > * um: Switch to large mcmodel on x86_64 > > * um: Serve io_remap_pfn_range() > > * 205 merges from Linus, from mid-April to late January. > > * mismerge from Linus on April 14; the diff between the result > > Okay, I'll create a new tree. This one is definitely screwed. > Sorry for the inconvenience. Just create a new branch in the old one, populate it, verify that the tree is what you want it to be, switch to your regular branch, git reset --hard <the branch you've just built>, then git push --force It's not worse than removal of old tree and replacing it with new one, and it's less work... ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] um: Cleanup headers files 2012-02-12 21:20 ` Al Viro @ 2012-02-12 21:26 ` Richard Weinberger 2012-02-12 21:28 ` Al Viro 0 siblings, 1 reply; 17+ messages in thread From: Richard Weinberger @ 2012-02-12 21:26 UTC (permalink / raw) To: Al Viro; +Cc: user-mode-linux-devel, linux-kernel Am 12.02.2012 22:20, schrieb Al Viro: > On Sun, Feb 12, 2012 at 10:01:49PM +0100, Richard Weinberger wrote: >> Am 12.02.2012 21:27, schrieb Al Viro: >>>> So, I pulled your work and merged it into my shiny new UML tree: >>>> git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml.git >>> >>> Hmm... I would really, really recommend you to clean the git history in that >>> tree up. As it is, you have >>> * merge of that branch from um-headers >>> * um: Switch to large mcmodel on x86_64 >>> * um: Serve io_remap_pfn_range() >>> * 205 merges from Linus, from mid-April to late January. >>> * mismerge from Linus on April 14; the diff between the result >> >> Okay, I'll create a new tree. This one is definitely screwed. >> Sorry for the inconvenience. > > Just create a new branch in the old one, populate it, verify that the > tree is what you want it to be, switch to your regular branch, > git reset --hard<the branch you've just built>, then git push --force > It's not worse than removal of old tree and replacing it with new one, > and it's less work... It looks like my master branch is screwed. :⨪\ Anyway, I'll fix it! Thanks, //richard ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] um: Cleanup headers files 2012-02-12 21:26 ` Richard Weinberger @ 2012-02-12 21:28 ` Al Viro 0 siblings, 0 replies; 17+ messages in thread From: Al Viro @ 2012-02-12 21:28 UTC (permalink / raw) To: Richard Weinberger; +Cc: user-mode-linux-devel, linux-kernel On Sun, Feb 12, 2012 at 10:26:06PM +0100, Richard Weinberger wrote: > Am 12.02.2012 22:20, schrieb Al Viro: > >On Sun, Feb 12, 2012 at 10:01:49PM +0100, Richard Weinberger wrote: > >>Am 12.02.2012 21:27, schrieb Al Viro: > >>>>So, I pulled your work and merged it into my shiny new UML tree: > >>>>git://git.kernel.org/pub/scm/linux/kernel/git/rw/uml.git > >>> > >>>Hmm... I would really, really recommend you to clean the git history in that > >>>tree up. As it is, you have > >>> * merge of that branch from um-headers > >>> * um: Switch to large mcmodel on x86_64 > >>> * um: Serve io_remap_pfn_range() > >>> * 205 merges from Linus, from mid-April to late January. > >>> * mismerge from Linus on April 14; the diff between the result > >> > >>Okay, I'll create a new tree. This one is definitely screwed. > >>Sorry for the inconvenience. > > > >Just create a new branch in the old one, populate it, verify that the > >tree is what you want it to be, switch to your regular branch, > >git reset --hard<the branch you've just built>, then git push --force > >It's not worse than removal of old tree and replacing it with new one, > >and it's less work... > > It looks like my master branch is screwed. :???\ > Anyway, I'll fix it! FWIW, #korg in um-headers.git should be about what you had in #master sans that mismerge in intel_pmic_gpio.c and with sane history. Feel free to pick it... ^ permalink raw reply [flat|nested] 17+ messages in thread
* Re: [PATCH] um: Cleanup headers files 2012-02-04 0:14 ` Al Viro 2012-02-04 0:26 ` Richard Weinberger 2012-02-04 0:44 ` Richard Weinberger @ 2012-02-04 14:05 ` Richard Weinberger 2 siblings, 0 replies; 17+ messages in thread From: Richard Weinberger @ 2012-02-04 14:05 UTC (permalink / raw) To: Al Viro; +Cc: user-mode-linux-devel, linux-kernel [-- Attachment #1: Type: text/plain, Size: 703 bytes --] Am 04.02.2012 01:14, schrieb Al Viro: > On Sat, Feb 04, 2012 at 12:17:43AM +0100, Richard Weinberger wrote: >> Many header files can be replaced by generic variants. > > Umm... I suspect that for auxvec.h the right thing is to simply > remove it - it's never used on uml builds. > > FWIW, I've rebased my local queue to current and pushed it to kernel.org > again; it's on git.kernel.org/pub/scm/linux/kernel/git/viro/um-headers.git BTW: You changed a lot in line.c. Have you seen this thread? http://permalink.gmane.org/gmane.linux.uml.devel/13329 At first I thought moving to tty_port solved all problems but later I've realized that this is not the case. Thanks, //richard [-- Attachment #2: OpenPGP digital signature --] [-- Type: application/pgp-signature, Size: 490 bytes --] ^ permalink raw reply [flat|nested] 17+ messages in thread
end of thread, other threads:[~2012-02-12 21:29 UTC | newest] Thread overview: 17+ messages (download: mbox.gz / follow: Atom feed) -- links below jump to the message on this page -- 2012-01-25 17:23 [PATCH 1/2] um: Use generic io.h Richard Weinberger 2012-01-25 17:23 ` [PATCH 2/2] um: Serve io_remap_pfn_range() Richard Weinberger 2012-01-30 18:23 ` [PATCH 1/2] um: Use generic io.h Al Viro 2012-01-30 19:10 ` Richard Weinberger 2012-01-30 19:25 ` Al Viro 2012-01-30 19:32 ` Al Viro 2012-01-30 19:57 ` Richard Weinberger 2012-02-03 23:17 ` [PATCH] um: Cleanup headers files Richard Weinberger 2012-02-04 0:14 ` Al Viro 2012-02-04 0:26 ` Richard Weinberger 2012-02-04 0:44 ` Richard Weinberger 2012-02-12 20:27 ` Al Viro 2012-02-12 21:01 ` Richard Weinberger 2012-02-12 21:20 ` Al Viro 2012-02-12 21:26 ` Richard Weinberger 2012-02-12 21:28 ` Al Viro 2012-02-04 14:05 ` Richard Weinberger
This is a public inbox, see mirroring instructions for how to clone and mirror all data and code used for this inbox; as well as URLs for NNTP newsgroup(s).