From bc22c17e12c130dc929218a95aa347e0f3fd05dc Mon Sep 17 00:00:00 2001 From: Alain Knaff Date: Sun, 4 Jan 2009 22:46:16 +0100 Subject: bzip2/lzma: library support for gzip, bzip2 and lzma decompression Impact: Replaces inflate.c with a wrapper around zlib_inflate; new library code This is the first part of the bzip2/lzma patch The bzip patch is based on an idea by Christian Ludwig, includes support for compressing the kernel with bzip2 or lzma rather than gzip. Both compressors give smaller sizes than gzip. Lzma's decompresses faster than bzip2. It also supports ramdisks and initramfs' compressed using these two compressors. The functionality has been successfully used for a couple of years by the udpcast project This version applies to "tip" kernel 2.6.28 This part contains: - changed inflate.c to accomodate rest of patch - implementation of bzip2 compression (not used at this stage yet) - implementation of lzma compression (not used at this stage yet) - Makefile routines to support bzip2 and lzma kernel compression Signed-off-by: Alain Knaff Signed-off-by: H. Peter Anvin --- include/linux/decompress/bunzip2.h | 10 +++++ include/linux/decompress/generic.h | 30 +++++++++++++ include/linux/decompress/inflate.h | 13 ++++++ include/linux/decompress/mm.h | 87 ++++++++++++++++++++++++++++++++++++++ include/linux/decompress/unlzma.h | 12 ++++++ 5 files changed, 152 insertions(+) create mode 100644 include/linux/decompress/bunzip2.h create mode 100644 include/linux/decompress/generic.h create mode 100644 include/linux/decompress/inflate.h create mode 100644 include/linux/decompress/mm.h create mode 100644 include/linux/decompress/unlzma.h (limited to 'include/linux') diff --git a/include/linux/decompress/bunzip2.h b/include/linux/decompress/bunzip2.h new file mode 100644 index 00000000000..115272137a9 --- /dev/null +++ b/include/linux/decompress/bunzip2.h @@ -0,0 +1,10 @@ +#ifndef DECOMPRESS_BUNZIP2_H +#define DECOMPRESS_BUNZIP2_H + +int bunzip2(unsigned char *inbuf, int len, + int(*fill)(void*, unsigned int), + int(*flush)(void*, unsigned int), + unsigned char *output, + int *pos, + void(*error)(char *x)); +#endif diff --git a/include/linux/decompress/generic.h b/include/linux/decompress/generic.h new file mode 100644 index 00000000000..f847f514f78 --- /dev/null +++ b/include/linux/decompress/generic.h @@ -0,0 +1,30 @@ +#ifndef DECOMPRESS_GENERIC_H +#define DECOMPRESS_GENERIC_H + +/* Minimal chunksize to be read. + *Bzip2 prefers at least 4096 + *Lzma prefers 0x10000 */ +#define COMPR_IOBUF_SIZE 4096 + +typedef int (*decompress_fn) (unsigned char *inbuf, int len, + int(*fill)(void*, unsigned int), + int(*writebb)(void*, unsigned int), + unsigned char *output, + int *posp, + void(*error)(char *x)); + +/* inbuf - input buffer + *len - len of pre-read data in inbuf + *fill - function to fill inbuf if empty + *writebb - function to write out outbug + *posp - if non-null, input position (number of bytes read) will be + * returned here + * + *If len != 0, the inbuf is initialized (with as much data), and fill + *should not be called + *If len = 0, the inbuf is allocated, but empty. Its size is IOBUF_SIZE + *fill should be called (repeatedly...) to read data, at most IOBUF_SIZE + */ + + +#endif diff --git a/include/linux/decompress/inflate.h b/include/linux/decompress/inflate.h new file mode 100644 index 00000000000..f9b06ccc3e5 --- /dev/null +++ b/include/linux/decompress/inflate.h @@ -0,0 +1,13 @@ +#ifndef INFLATE_H +#define INFLATE_H + +/* Other housekeeping constants */ +#define INBUFSIZ 4096 + +int gunzip(unsigned char *inbuf, int len, + int(*fill)(void*, unsigned int), + int(*flush)(void*, unsigned int), + unsigned char *output, + int *pos, + void(*error_fn)(char *x)); +#endif diff --git a/include/linux/decompress/mm.h b/include/linux/decompress/mm.h new file mode 100644 index 00000000000..12ff8c3f1d0 --- /dev/null +++ b/include/linux/decompress/mm.h @@ -0,0 +1,87 @@ +/* + * linux/compr_mm.h + * + * Memory management for pre-boot and ramdisk uncompressors + * + * Authors: Alain Knaff + * + */ + +#ifndef DECOMPR_MM_H +#define DECOMPR_MM_H + +#ifdef STATIC + +/* Code active when included from pre-boot environment: */ + +/* A trivial malloc implementation, adapted from + * malloc by Hannu Savolainen 1993 and Matthias Urlichs 1994 + */ +static unsigned long malloc_ptr; +static int malloc_count; + +static void *malloc(int size) +{ + void *p; + + if (size < 0) + error("Malloc error"); + if (!malloc_ptr) + malloc_ptr = free_mem_ptr; + + malloc_ptr = (malloc_ptr + 3) & ~3; /* Align */ + + p = (void *)malloc_ptr; + malloc_ptr += size; + + if (free_mem_end_ptr && malloc_ptr >= free_mem_end_ptr) + error("Out of memory"); + + malloc_count++; + return p; +} + +static void free(void *where) +{ + malloc_count--; + if (!malloc_count) + malloc_ptr = free_mem_ptr; +} + +#define large_malloc(a) malloc(a) +#define large_free(a) free(a) + +#define set_error_fn(x) + +#define INIT + +#else /* STATIC */ + +/* Code active when compiled standalone for use when loading ramdisk: */ + +#include +#include +#include +#include + +/* Use defines rather than static inline in order to avoid spurious + * warnings when not needed (indeed large_malloc / large_free are not + * needed by inflate */ + +#define malloc(a) kmalloc(a, GFP_KERNEL) +#define free(a) kfree(a) + +#define large_malloc(a) vmalloc(a) +#define large_free(a) vfree(a) + +static void(*error)(char *m); +#define set_error_fn(x) error = x; + +#define INIT __init +#define STATIC + +#include + +#endif /* STATIC */ + +#endif /* DECOMPR_MM_H */ diff --git a/include/linux/decompress/unlzma.h b/include/linux/decompress/unlzma.h new file mode 100644 index 00000000000..7796538f1bf --- /dev/null +++ b/include/linux/decompress/unlzma.h @@ -0,0 +1,12 @@ +#ifndef DECOMPRESS_UNLZMA_H +#define DECOMPRESS_UNLZMA_H + +int unlzma(unsigned char *, int, + int(*fill)(void*, unsigned int), + int(*flush)(void*, unsigned int), + unsigned char *output, + int *posp, + void(*error)(char *x) + ); + +#endif -- cgit v1.2.3 From 889c92d21db40be0b7d22a59395060237895bb85 Mon Sep 17 00:00:00 2001 From: "H. Peter Anvin" Date: Thu, 8 Jan 2009 15:14:17 -0800 Subject: bzip2/lzma: centralize format detection Centralize the compression format detection to a common routine in the lib directory, and use it for both initramfs and initrd. Signed-off-by: H. Peter Anvin --- include/linux/decompress/generic.h | 3 +++ 1 file changed, 3 insertions(+) (limited to 'include/linux') diff --git a/include/linux/decompress/generic.h b/include/linux/decompress/generic.h index f847f514f78..6dfb856327b 100644 --- a/include/linux/decompress/generic.h +++ b/include/linux/decompress/generic.h @@ -26,5 +26,8 @@ typedef int (*decompress_fn) (unsigned char *inbuf, int len, *fill should be called (repeatedly...) to read data, at most IOBUF_SIZE */ +/* Utility routine to detect the decompression method */ +decompress_fn decompress_method(const unsigned char *inbuf, int len, + const char **name); #endif -- cgit v1.2.3 From 704126ad81b8cb7d3d70adb9ecb143f4d3fb38af Mon Sep 17 00:00:00 2001 From: Yu Zhao Date: Sun, 4 Jan 2009 16:28:52 +0800 Subject: VT-d: handle Invalidation Queue Error to avoid system hang When hardware detects any error with a descriptor from the invalidation queue, it stops fetching new descriptors from the queue until software clears the Invalidation Queue Error bit in the Fault Status register. Following fix handles the IQE so the kernel won't be trapped in an infinite loop. Signed-off-by: Yu Zhao Signed-off-by: David Woodhouse --- include/linux/intel-iommu.h | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/intel-iommu.h b/include/linux/intel-iommu.h index c4f6c101dbc..d2e3cbfba14 100644 --- a/include/linux/intel-iommu.h +++ b/include/linux/intel-iommu.h @@ -194,6 +194,7 @@ static inline void dmar_writeq(void __iomem *addr, u64 val) /* FSTS_REG */ #define DMA_FSTS_PPF ((u32)2) #define DMA_FSTS_PFO ((u32)1) +#define DMA_FSTS_IQE (1 << 4) #define dma_fsts_fault_record_index(s) (((s) >> 8) & 0xff) /* FRCD_REG, 32 bits access */ @@ -328,7 +329,7 @@ extern int qi_flush_iotlb(struct intel_iommu *iommu, u16 did, u64 addr, unsigned int size_order, u64 type, int non_present_entry_flush); -extern void qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu); +extern int qi_submit_sync(struct qi_desc *desc, struct intel_iommu *iommu); extern void *intel_alloc_coherent(struct device *, size_t, dma_addr_t *, gfp_t); extern void intel_free_coherent(struct device *, size_t, void *, dma_addr_t); -- cgit v1.2.3 From 5ca431f9ae8db8c6edb9c64bebe6d6521077afd6 Mon Sep 17 00:00:00 2001 From: Eric Leblond Date: Wed, 18 Feb 2009 15:29:23 +0100 Subject: netfilter: nfnetlink_log: fix per-rule qthreshold override In NFLOG the per-rule qthreshold should overrides per-instance only it is set. With current code, the per-rule qthreshold is 1 if not set and it overrides the per-instance qthreshold. This patch modifies the default xt_NFLOG threshold from 1 to 0. Thus a value of 0 means there is no per-rule setting and the instance parameter has to apply. Signed-off-by: Eric Leblond Signed-off-by: Patrick McHardy --- include/linux/netfilter/xt_NFLOG.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/netfilter/xt_NFLOG.h b/include/linux/netfilter/xt_NFLOG.h index cdcd0ed58f7..4b36aeb46a1 100644 --- a/include/linux/netfilter/xt_NFLOG.h +++ b/include/linux/netfilter/xt_NFLOG.h @@ -2,7 +2,7 @@ #define _XT_NFLOG_TARGET #define XT_NFLOG_DEFAULT_GROUP 0x1 -#define XT_NFLOG_DEFAULT_THRESHOLD 1 +#define XT_NFLOG_DEFAULT_THRESHOLD 0 #define XT_NFLOG_MASK 0x0 -- cgit v1.2.3 From 8fed43684174b68f04d01d1210fd00536af790df Mon Sep 17 00:00:00 2001 From: Bartlomiej Zolnierkiewicz Date: Wed, 25 Feb 2009 20:28:24 +0100 Subject: ide: fix refcounting in device drivers During host driver module removal del_gendisk() results in a final put on drive->gendev and freeing the drive by drive_release_dev(). Convert device drivers from using struct kref to use struct device so device driver's object holds reference on ->gendev and prevents drive from prematurely going away. Also fix ->remove methods to not erroneously drop reference on a host driver by using only put_device() instead of ide*_put(). Reported-by: Stanislaw Gruszka Tested-by: Stanislaw Gruszka Signed-off-by: Bartlomiej Zolnierkiewicz --- include/linux/ide.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/ide.h b/include/linux/ide.h index 194da5a4b0d..fe235b65207 100644 --- a/include/linux/ide.h +++ b/include/linux/ide.h @@ -663,7 +663,7 @@ typedef struct ide_drive_s ide_drive_t; #define to_ide_device(dev) container_of(dev, ide_drive_t, gendev) #define to_ide_drv(obj, cont_type) \ - container_of(obj, struct cont_type, kref) + container_of(obj, struct cont_type, dev) #define ide_drv_g(disk, cont_type) \ container_of((disk)->private_data, struct cont_type, driver) -- cgit v1.2.3 From 1e42807918d17e8c93bf14fbb74be84b141334c1 Mon Sep 17 00:00:00 2001 From: Jens Axboe Date: Mon, 23 Feb 2009 09:03:10 +0100 Subject: block: reduce stack footprint of blk_recount_segments() blk_recalc_rq_segments() requires a request structure passed in, which we don't have from blk_recount_segments(). So the latter allocates one on the stack, using > 400 bytes of stack for that. This can cause us to spill over one page of stack from ext4 at least: 0) 4560 400 blk_recount_segments+0x43/0x62 1) 4160 32 bio_phys_segments+0x1c/0x24 2) 4128 32 blk_rq_bio_prep+0x2a/0xf9 3) 4096 32 init_request_from_bio+0xf9/0xfe 4) 4064 112 __make_request+0x33c/0x3f6 5) 3952 144 generic_make_request+0x2d1/0x321 6) 3808 64 submit_bio+0xb9/0xc3 7) 3744 48 submit_bh+0xea/0x10e 8) 3696 368 ext4_mb_init_cache+0x257/0xa6a [ext4] 9) 3328 288 ext4_mb_regular_allocator+0x421/0xcd9 [ext4] 10) 3040 160 ext4_mb_new_blocks+0x211/0x4b4 [ext4] 11) 2880 336 ext4_ext_get_blocks+0xb61/0xd45 [ext4] 12) 2544 96 ext4_get_blocks_wrap+0xf2/0x200 [ext4] 13) 2448 80 ext4_da_get_block_write+0x6e/0x16b [ext4] 14) 2368 352 mpage_da_map_blocks+0x7e/0x4b3 [ext4] 15) 2016 352 ext4_da_writepages+0x2ce/0x43c [ext4] 16) 1664 32 do_writepages+0x2d/0x3c 17) 1632 144 __writeback_single_inode+0x162/0x2cd 18) 1488 96 generic_sync_sb_inodes+0x1e3/0x32b 19) 1392 16 sync_sb_inodes+0xe/0x10 20) 1376 48 writeback_inodes+0x69/0xb3 21) 1328 208 balance_dirty_pages_ratelimited_nr+0x187/0x2f9 22) 1120 224 generic_file_buffered_write+0x1d4/0x2c4 23) 896 176 __generic_file_aio_write_nolock+0x35f/0x393 24) 720 80 generic_file_aio_write+0x6c/0xc8 25) 640 80 ext4_file_write+0xa9/0x137 [ext4] 26) 560 320 do_sync_write+0xf0/0x137 27) 240 48 vfs_write+0xb3/0x13c 28) 192 64 sys_write+0x4c/0x74 29) 128 128 system_call_fastpath+0x16/0x1b Split the segment counting out into a __blk_recalc_rq_segments() helper to avoid allocating an onstack request just for checking the physical segment count. Signed-off-by: Jens Axboe --- include/linux/blkdev.h | 2 ++ 1 file changed, 2 insertions(+) (limited to 'include/linux') diff --git a/include/linux/blkdev.h b/include/linux/blkdev.h index dcaa0fd84b0..465d6babc84 100644 --- a/include/linux/blkdev.h +++ b/include/linux/blkdev.h @@ -708,6 +708,8 @@ struct req_iterator { }; /* This should not be used directly - use rq_for_each_segment */ +#define for_each_bio(_bio) \ + for (; _bio; _bio = _bio->bi_next) #define __rq_for_each_bio(_bio, rq) \ if ((rq->bio)) \ for (_bio = (rq)->bio; _bio; _bio = _bio->bi_next) -- cgit v1.2.3 From 5170836679185357dc1b7660bad13287b39e1e33 Mon Sep 17 00:00:00 2001 From: David Howells Date: Fri, 27 Feb 2009 14:03:03 -0800 Subject: Fix recursive lock in free_uid()/free_user_ns() free_uid() and free_user_ns() are corecursive when CONFIG_USER_SCHED=n, but free_user_ns() is called from free_uid() by way of uid_hash_remove(), which requires uidhash_lock to be held. free_user_ns() then calls free_uid() to complete the destruction. Fix this by deferring the destruction of the user_namespace. Signed-off-by: David Howells Acked-by: Serge Hallyn Signed-off-by: Andrew Morton Signed-off-by: Linus Torvalds --- include/linux/user_namespace.h | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/user_namespace.h b/include/linux/user_namespace.h index 315bcd37522..cc4f45361db 100644 --- a/include/linux/user_namespace.h +++ b/include/linux/user_namespace.h @@ -13,6 +13,7 @@ struct user_namespace { struct kref kref; struct hlist_head uidhash_table[UIDHASH_SZ]; struct user_struct *creator; + struct work_struct destroyer; }; extern struct user_namespace init_user_ns; -- cgit v1.2.3 From 5c2522218059ca1f4174a568923b988aad3ddfda Mon Sep 17 00:00:00 2001 From: Chris Leech Date: Fri, 27 Feb 2009 10:01:36 +0000 Subject: net headers: cleanup dcbnl.h 1) add an include for 2) change dcbmsg.dcb_family from unsigned char to __u8 to be more consistent with use of kernel types Signed-off-by: Chris Leech Acked-by: Sam Ravnborg Signed-off-by: David S. Miller --- include/linux/dcbnl.h | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/dcbnl.h b/include/linux/dcbnl.h index b0ef274e003..7d2e1000618 100644 --- a/include/linux/dcbnl.h +++ b/include/linux/dcbnl.h @@ -20,10 +20,12 @@ #ifndef __LINUX_DCBNL_H__ #define __LINUX_DCBNL_H__ +#include + #define DCB_PROTO_VERSION 1 struct dcbmsg { - unsigned char dcb_family; + __u8 dcb_family; __u8 cmd; __u16 dcb_pad; }; -- cgit v1.2.3 From 709ab3261e3ed789c0bb31c6ab53c9eccb276522 Mon Sep 17 00:00:00 2001 From: Chris Leech Date: Fri, 27 Feb 2009 10:01:42 +0000 Subject: net headers: export dcbnl.h The DCB netlink interface is required for building the userspace tools available at e1000.sourceforge.net Signed-off-by: Chris Leech Signed-off-by: David S. Miller --- include/linux/Kbuild | 1 + 1 file changed, 1 insertion(+) (limited to 'include/linux') diff --git a/include/linux/Kbuild b/include/linux/Kbuild index b97cdc516a8..106c3ba5084 100644 --- a/include/linux/Kbuild +++ b/include/linux/Kbuild @@ -52,6 +52,7 @@ header-y += const.h header-y += cgroupstats.h header-y += cramfs_fs.h header-y += cycx_cfm.h +header-y += dcbnl.h header-y += dlmconstants.h header-y += dlm_device.h header-y += dlm_netlink.h -- cgit v1.2.3 From f5c1aa1537be39d8b9bb5279b5881d81898fd3cd Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Sun, 1 Mar 2009 12:32:08 +0100 Subject: Revert "gpu/drm, x86, PAT: PAT support for io_mapping_*" This reverts commit 17581ad812a9abb0182260374ef2e52d4a808a64. Sitsofe Wheeler reported that /dev/dri/card0 is MIA on his EeePC 900 and bisected it to this commit. Graphics card is an i915 in an EeePC 900: 00:02.0 VGA compatible controller [0300]: Intel Corporation Mobile 915GM/GMS/910GML Express Graphics Controller [8086:2592] (rev 04) ( Most likely the ioremap() of the driver failed and hence the card did not initialize. ) Reported-by: Sitsofe Wheeler Bisected-by: Sitsofe Wheeler Cc: Venkatesh Pallipadi Cc: Suresh Siddha Signed-off-by: Ingo Molnar --- include/linux/io-mapping.h | 6 ++---- 1 file changed, 2 insertions(+), 4 deletions(-) (limited to 'include/linux') diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h index f1ed66c4378..cbc2f0cd631 100644 --- a/include/linux/io-mapping.h +++ b/include/linux/io-mapping.h @@ -49,9 +49,8 @@ static inline struct io_mapping * io_mapping_create_wc(resource_size_t base, unsigned long size) { struct io_mapping *iomap; - pgprot_t prot; - if (!reserve_io_memtype_wc(base, size, &prot)) + if (!is_io_mapping_possible(base, size)) return NULL; iomap = kmalloc(sizeof(*iomap), GFP_KERNEL); @@ -60,14 +59,13 @@ io_mapping_create_wc(resource_size_t base, unsigned long size) iomap->base = base; iomap->size = size; - iomap->prot = prot; + iomap->prot = pgprot_writecombine(__pgprot(__PAGE_KERNEL)); return iomap; } static inline void io_mapping_free(struct io_mapping *mapping) { - free_io_memtype(mapping->base, mapping->size); kfree(mapping); } -- cgit v1.2.3 From 5ce04e3de8c36ba37c56e94e3c4dc7973c7f546c Mon Sep 17 00:00:00 2001 From: "Pallipadi, Venkatesh" Date: Sun, 1 Mar 2009 08:53:27 -0800 Subject: fix warning in io_mapping_map_wc() Signed-off-by: Venkatesh Pallipadi Signed-off-by: Ingo Molnar --- include/linux/io-mapping.h | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-) (limited to 'include/linux') diff --git a/include/linux/io-mapping.h b/include/linux/io-mapping.h index cbc2f0cd631..0adb0f91568 100644 --- a/include/linux/io-mapping.h +++ b/include/linux/io-mapping.h @@ -91,8 +91,11 @@ io_mapping_unmap_atomic(void *vaddr) static inline void * io_mapping_map_wc(struct io_mapping *mapping, unsigned long offset) { + resource_size_t phys_addr; + BUG_ON(offset >= mapping->size); - resource_size_t phys_addr = mapping->base + offset; + phys_addr = mapping->base + offset; + return ioremap_wc(phys_addr, PAGE_SIZE); } -- cgit v1.2.3 From f180053694b43d5714bf56cb95499a3c32ff155c Mon Sep 17 00:00:00 2001 From: Ingo Molnar Date: Mon, 2 Mar 2009 11:00:57 +0100 Subject: x86, mm: dont use non-temporal stores in pagecache accesses Impact: standardize IO on cached ops On modern CPUs it is almost always a bad idea to use non-temporal stores, as the regression in this commit has shown it: 30d697f: x86: fix performance regression in write() syscall The kernel simply has no good information about whether using non-temporal stores is a good idea or not - and trying to add heuristics only increases complexity and inserts fragility. The regression on cached write()s took very long to be found - over two years. So dont take any chances and let the hardware decide how it makes use of its caches. The only exception is drivers/gpu/drm/i915/i915_gem.c: there were we are absolutely sure that another entity (the GPU) will pick up the dirty data immediately and that the CPU will not touch that data before the GPU will. Also, keep the _nocache() primitives to make it easier for people to experiment with these details. There may be more clear-cut cases where non-cached copies can be used, outside of filemap.c. Cc: Salman Qazi Cc: Nick Piggin Cc: Linus Torvalds Signed-off-by: Ingo Molnar --- include/linux/uaccess.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-) (limited to 'include/linux') diff --git a/include/linux/uaccess.h b/include/linux/uaccess.h index 6f3c603b0d6..6b58367d145 100644 --- a/include/linux/uaccess.h +++ b/include/linux/uaccess.h @@ -41,13 +41,13 @@ static inline void pagefault_enable(void) #ifndef ARCH_HAS_NOCACHE_UACCESS static inline unsigned long __copy_from_user_inatomic_nocache(void *to, - const void __user *from, unsigned long n, unsigned long total) + const void __user *from, unsigned long n) { return __copy_from_user_inatomic(to, from, n); } static inline unsigned long __copy_from_user_nocache(void *to, - const void __user *from, unsigned long n, unsigned long total) + const void __user *from, unsigned long n) { return __copy_from_user(to, from, n); } -- cgit v1.2.3