- 11 Jul, 2004 40 commits
-
-
Martin Schwidefsky authored
From: Martin Schwidefsky <schwidefsky@de.ibm.com> From: Ulrich Weigand <uweigand@de.ibm.com> s390 core changes: - Add signo between signal frame and the signal return instruction on the user stack for backtrace over signal handlers. - Add hfp floating point exceptions. - Use a single function for region, segment and page translation exceptions. - Discard SIGTRAP for single stepped instructions if the trapping instruction is repeated (normal memory faults) or if another signal is delivered anyway. Signed-off-by:
Martin Schwidefsky <schwidefsky@de.ibm.com> Signed-off-by:
Andrew Morton <akpm@osdl.org> Signed-off-by:
Linus Torvalds <torvalds@osdl.org>
-
Andrea Arcangeli authored
I believe reading the i_size from memory multiple times can generate fs corruption. The "offset" and the "end_index" were not coherent. this is writepages and it runs w/o the i_sem, so the i_size can change from under us anytime. If a parallel write happens while writepages run, the i_size could advance from 4095 to 4100. With the current 2.6 code that could translate in end_index = 0 and offset = 4. That's broken because end_index and offset could be not coherent. Either end_index=1 and offset =4, or end_index = 0 and offset = 4095. When they lose coherency the memset can zeroout actual data. The below patch fixes that (it's at least a theoretical bug). I don't really expect this tiny race to fix the bug in practice after the more serious bugs we covered yesterday didn't fix it (more likely the compiler will get involved into the equation soon ;). This is also an optimization for 32bit archs that needs special locking to read 64bit i_size coherenty. This patch also arranges for mpage_writepages() to always zero out the file's final page between i_size and the end of the file's final block. This is a best-effort correctness thing to deal with errant applications which write into the mmapped page beyond the underlying file's EOF. Signed-off-by:
-
Andrea Arcangeli authored
Signed-off-by:
-
Andrea Arcangeli authored
Fix a data loss bug in mpage_writepages(), triggerable under extreme memory pressure on ext2, JFS, hfs and hfsplus: The bug is the marking of the bh clean despite we could still run into the "confused" path. After that the confused path really becomes confused and it writes nothing and fs corruption triggers silenty (the reugular writepage only writes bh that are marked dirty, it never attempts to submit_bh anything marked clean). The mpage-writepage code must never mark the bh clean as far as it wants to still fallback in the regular writepage which depends on the bh to be dirty (i.e. the "goto confused" path). This could only triggers with memory pressure (it also needs buffer_heads_over_limit == 0, and that is frequent under mm pressure). Thanks a lot to Chris for his fine debugging that localized the problem in the writepage code. Signed-off-by:
-
Pawel Sikora authored
CC [M] sound/oss/ad1889.o sound/oss/ad1889.c: In function `ad1889_ac97_init': sound/oss/ad1889.c:854: warning: comparison is always false due to limited range of data type Signed-off-by:
-
Hugh Dickins authored
Nick has tracked scheduling-while-atomic errors to shmem's fragile kmap avoidance: the root error appears to lie deeper, but rework that fragility. Plus I've been indicted for war crimes at the end of shmem_swp_entry: my apologia scorned, so now hide the evidence. Signed-off-by:
Hugh Dickins <hugh@veritas.com> Signed-off-by:
-
Paul Mundt authored
This just updates the defconfig against current BK. Signed-off-by:
Paul Mundt <lethal@linux-sh.org> Signed-off-by:
-
Paul Mundt authored
With the recent changes to init/main.c, we need an asm-sh64/setup.h. This pulls out the definitions from arch/sh64/kernel/setup.c and moves them to setup.h instead. Signed-off-by:
-
Paul Mundt authored
resource.h was falling a bit behind, so we use the asm-sh version instead. This exposed another compile issue with init_task.c, which in turn needed linux/mqueue.h. Signed-off-by:
-
Paul Mundt authored
With the recent cpumask changes, some things ended up being broken on sh64. This fixes them.. Signed-off-by:
-
Andrey Panin authored
Signed-off-by:
Andrey Panin <pazke@donpac.ru> Signed-off-by:
-
Andrey Panin authored
Signed-off-by:
-
Andrey Panin authored
Signed-off-by:
-
Andrey Panin authored
Signed-off-by:
-
Andrew Morton authored
drivers/net/hamradio/Kconfig:116:warning: 'select' used by config symbol 'BAYCOM_SER_FDX' refer to undefined symbol 'CRC16' drivers/net/hamradio/Kconfig:136:warning: 'select' used by config symbol 'BAYCOM_SER_HDX' refer to undefined symbol 'CRC16' drivers/net/hamradio/Kconfig:154:warning: 'select' used by config symbol 'BAYCOM_PAR' refer to undefined symbol 'CRC16' drivers/net/hamradio/Kconfig:169:warning: 'select' used by config symbol 'BAYCOM_EPP' refer to undefined symbol 'CRC16' net/irda/Kconfig:8:warning: 'select' used by config symbol 'IRDA' refer to undefined symbol 'CRC16' drivers/net/Kconfig:1749:warning: 'select' used by config symbol 'VIA_VELOCITY' refer to undefined symbol 'CRC16' drivers/net/Kconfig:2465:warning: 'select' used by config symbol 'PPP_ASYNC' refer to undefined symbol 'CRC16' drivers/isdn/hisax/Kconfig:403:warning: 'select' used by config symbol 'HISAX_ST5481' refer to undefined symbol 'CRC16' drivers/isdn/hisax/Kconfig:7:warning: 'select' used by config symbol 'ISDN_DRV_HISAX' refer to undefined symbol 'CRC16' drivers/isdn/tpam/Kconfig:7:warning: 'select' used by config symbol 'ISDN_DRV_TPAM' refer to undefined symbol 'CRC16' Signed-off-by:
-
Andrey Panin authored
As pointed by Thomas Sailer, crc16.c module contains CRC16-CCITT (x^16 + x^12 + x^5 + 1) implementation, not IBM CRC16 (x^16 + x^15 + x^2 + 1) one. Looks like we need to rename it accordingly and this patchset does exactly this. Signed-off-by:
-
Andrey Panin authored
This patch removes dmi_broken global variable which is not used anymore. Signed-off-by:
-
Andrey Panin authored
This patch moves Toshiba ACPI sleep workaround out of dmi_scan.c Signed-off-by:
-
Aneesh Kumar authored
Cc: Richard Henderson <rth@twiddle.net> Signed-off-by:
-
Andrew Morton authored
This little debugging __get_user is in fact happening inside a spinlock. It was never very useful, and has caused problems for some architectures in the past. Let's just remove it. Signed-off-by:
-
Andrew Morton authored
Mika Kukkonen <mika@osdl.org> says: CC mm/filemap.o mm/filemap.c: In function `filemap_nopage': mm/filemap.c:1161: warning: comparison of unsigned expression < 0 is always false The pagefault readaround code is currently doing purely readahead. Signed-off-by:
-
Mika Kukkonen authored
CC [M] fs/isofs/namei.o In file included from fs/isofs/namei.c:10: include/linux/iso_fs.h:280: warning: `inline' is not at beginning of declaration Signed-off-by:
-
Mika Kukkonen authored
CC fs/dquot.o fs/dquot.c:208: warning: type qualifiers ignored on function return type Once again with extra gcc warnings enabled. Every user of the function is expecting unsigned value, not int in first place, and I think the const is just misplaced. Signed-off-by:
-
Mika Kukkonen authored
CC mm/slab.o mm/slab.c: In function `kmem_cache_create': mm/slab.c:1129: warning: comparison of unsigned expression < 0 is always false This comes from the fact that 'align' is size_t and so unsigned. Just to be sure, I did $ grep __kernel_size_t include/*/posix_types.h and yes, every arch defines that to be unsigned. Signed-off-by:
-
Randy Dunlap authored
Use %Zd to eliminate a compiler warning in printk. Signed-off-by:
-
Andy Whitcroft authored
As the comments in mmzone.h indicate is_highmem() is designed to reduce the proliferation of the constant ZONE_HIGHMEM. This patch updates references to ZONE_HIGHMEM to use is_highmem(). None appear to be on critical paths. Signed-off-by:
Andy Whitcroft <apw@shadowen.org> Signed-off-by:
-
Rusty Russell authored
(OK from maintainer Geert Uytterhoeven <geert@linux-m68k.org>) From: a.othieno@bluewin.ch (Arthur Othieno) AFAICS, CONFIG_PCI_CONSOLE isn't used elsewhere: hubb@mars:linux-2.6.2$ grep -r PCI_CONSOLE * | grep -v defconfig drivers/video/console/Kconfig:config PCI_CONSOLE hubb@mars:linux-2.6.2$ Signed-off-by:
-
Rusty Russell authored
From: <a.othieno@bluewin.ch> (Arthur Othieno) via scripts/checkincludes.pl: Signed-off-by:
-
Rusty Russell authored
From: Chris Heath <chris@heathens.co.nz> Here's a trivial patch that removes an unused-variable warning in ftape. Signed-off-by:
-
Rusty Russell authored
From: David Gibson <david@gibson.dropbear.id.au> This patch excludes the .pc directory from the same things that SCCS/BitKeeper/.svn/CVS files are excluded from. The .pc directory is used for backup/reference files by quilt, a patch mangling system conceptually derived from akpm's patch scripts. Excluding the .pc directory is handy, because otherwise old versions of files found in there tend to end up at the front of the TAGS index. Signed-off-by:
-
Rusty Russell authored
fixes a cut'n'paste error... Signed-off-by:
Rusty Trivial Russell <trivial@rustcorp.com.au> Signed-off-by:
Adam Lackorzynski <adam@os.inf.tu-dresden.de> Signed-off-by:
-
Rusty Russell authored
From: <adobriyan@mail.ru> Fix missing bracket when parameter to be documented is a pointer to function. int (* resume (struct usb_interface *intf); Signed-off-by:
-
Rusty Russell authored
From: <adobriyan@mail.ru> Imagine you have a code similar to struct foo { union { struct tcphdr *th; } h; union { struct iphdr *iph; } nh; }; kernel-doc in it's current state will happily eat everything from first '{' to last '}' and nobody will see parameter 'h' in documentation (look at include/linux/skbuff.h:struct sk_buff for real world example). So, fix the greedy regexp. Signed-off-by: -
Mika Kukkonen authored
CC [M] fs/smbfs/proc.o fs/smbfs/proc.c: In function `smb_proc_readdir_long': fs/smbfs/proc.c:2313: warning: comparison of unsigned expression < 0 is always false fs/smbfs/proc.c:2467: warning: comparison of unsigned expression < 0 is always false The first one is pretty dangerous looking, as smb_proc_readdir_long() can return several negative error values and all those are converted to unsigned and then erronously pass the test on line 2313. Chris Wright gave it a quick look and we did not see immediately if this can be remotely exploited, but it looks pretty scary. The second warning on line 2467 is just extra so I just removed it. Signed-off-by:
-
Chris Wright authored
Fixup another round of sparse warnings of the type: warning: Using plain integer as NULL pointer Acked by Stephen. From: Mika Kukkonen <mika@osdl.org> Signed-off-by:
Chris Wright <chrisw@osdl.org> Signed-off-by:
-
Hirofumi Ogawa authored
Signed-off-by:
OGAWA Hirofumi <hirofumi@mail.parknet.co.jp> Signed-off-by:
-
Jens Axboe authored
Arjan (sensibly) put a might_sleep() in mempool_alloc() and it caught a bad cfq usage. Signed-off-by:
Jens Axboe <axboe@suse.de> Signed-off-by:
-
Andrew Morton authored
Signed-off-by:
-
Keith Owens authored
PPC small data area base symbols shift between kallsyms phases 1 and 2, which makes the kallsyms data unstable. Exclude them from the kallsyms list. Signed-off-by:
Keith Owens <kaos@sgi.com> Signed-off-by:
-
Christoph Hellwig authored
The <driver>_config function called just before clear DEV_CONFIG, and DEV_STALE_LINK isn't set anywhere else. This is a copy & paste thing found in all old pcmcia drivers. The patch has been sent to the isdn list ages ago but I didn't get any response. Signed-off-by:
-
