search for: bitmap_set

Use bitmap_set and bitmap_clear rather than modifying individual bits in a memory region. Signed-off-by: Akinobu Mita <akinobu.mita at gmail.com> Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge at citrix.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk at oracle.com> Cc: xen-devel at lists.xensource.co...

Use bitmap_set and bitmap_clear rather than modifying individual bits in a memory region. Signed-off-by: Akinobu Mita <akinobu.mita at gmail.com> Cc: Jeremy Fitzhardinge <jeremy.fitzhardinge at citrix.com> Cc: Konrad Rzeszutek Wilk <konrad.wilk at oracle.com> Cc: xen-devel at lists.xensource.co...

...ne->free_area[order].free_list[t]) { > + pfn = page_to_pfn(list_entry(curr, struct page, lru)); > + if (pfn >= start_pfn && pfn <= end_pfn) { > + page_num = 1UL << order; > + if (pfn + page_num > end_pfn) > + page_num = end_pfn - pfn; > + bitmap_set(bitmap, pfn - start_pfn, page_num); > + } > + } > + } Nit: The 'page_num' nomenclature really confused me here. It is the number of bits being set in the bitmap. Seems like calling it nr_pages or num_pages would be more appropriate. Isn't this bitmap out of date by the...

...ne->free_area[order].free_list[t]) { > + pfn = page_to_pfn(list_entry(curr, struct page, lru)); > + if (pfn >= start_pfn && pfn <= end_pfn) { > + page_num = 1UL << order; > + if (pfn + page_num > end_pfn) > + page_num = end_pfn - pfn; > + bitmap_set(bitmap, pfn - start_pfn, page_num); > + } > + } > + } Nit: The 'page_num' nomenclature really confused me here. It is the number of bits being set in the bitmap. Seems like calling it nr_pages or num_pages would be more appropriate. Isn't this bitmap out of date by the...

Hi Rich, There's an unchecked calloc in hivex.c's hive_open: h->bitmap = calloc (1 + h->size / 32, 1); ... This subsequent deref could cause a segfault: BITMAP_SET (h->bitmap, blkoff);

...> + pfn = page_to_pfn(list_entry(curr, struct page, lru)); >>> + if (pfn >= start_pfn && pfn <= end_pfn) { >>> + page_num = 1UL << order; >>> + if (pfn + page_num > end_pfn) >>> + page_num = end_pfn - pfn; >>> + bitmap_set(bitmap, pfn - start_pfn, page_num); >>> + } >>> + } >>> + } >> >> Nit: The 'page_num' nomenclature really confused me here. It is the >> number of bits being set in the bitmap. Seems like calling it nr_pages >> or num_pages would be m...

...> + pfn = page_to_pfn(list_entry(curr, struct page, lru)); >>> + if (pfn >= start_pfn && pfn <= end_pfn) { >>> + page_num = 1UL << order; >>> + if (pfn + page_num > end_pfn) >>> + page_num = end_pfn - pfn; >>> + bitmap_set(bitmap, pfn - start_pfn, page_num); >>> + } >>> + } >>> + } >> >> Nit: The 'page_num' nomenclature really confused me here. It is the >> number of bits being set in the bitmap. Seems like calling it nr_pages >> or num_pages would be m...

...;< and >> instead of / and % aid the compiler, since we know we have powers of 2 based on initialization, but the compiler might not see it locally? > +/* Set the bit(s) associated with the given block. > + * If out of range, it is ignored. > + */ > +static inline void > +bitmap_set_blk (const struct bitmap *bm, uint64_t blk, unsigned v) > +{ > + uint64_t blk_offset = blk / (8 / bm->bpb); > + unsigned blk_bit = bm->bpb * (blk % (8 / bm->bpb)); > + > + if (blk_offset >= bm->size) { > + nbdkit_debug ("bitmap_set: block number is out of...

This is exactly the same as v1: https://www.redhat.com/archives/libguestfs/2018-December/msg00004.html except that it now frees the bitmap on unload (which the old code did not - there was always a memory leak). Rich.

...32 bits of the file + * (because blocks are 4-byte aligned). We found that the average + * block size in a registry file is ~50 bytes. So roughly 1 in 12 + * bits in the bitmap will be set, making it likely a more efficient + * structure than a hash table. + */ + char *bitmap; +#define BITMAP_SET(bitmap,off) (bitmap[(off)>>5] |= 1 << (((off)>>2)&7)) +#define BITMAP_CLR(bitmap,off) (bitmap[(off)>>5] &= ~ (1 << (((off)>>2)&7))) +#define BITMAP_TST(bitmap,off) (bitmap[(off)>>5] & (1 << (((off)>>2)&7))) +#define IS_VALID_...

...ist[t]) { > > + pfn = page_to_pfn(list_entry(curr, struct page, lru)); > > + if (pfn >= start_pfn && pfn <= end_pfn) { > > + page_num = 1UL << order; > > + if (pfn + page_num > end_pfn) > > + page_num = end_pfn - pfn; > > + bitmap_set(bitmap, pfn - start_pfn, page_num); > > + } > > + } > > + } > > Nit: The 'page_num' nomenclature really confused me here. It is the > number of bits being set in the bitmap. Seems like calling it nr_pages > or num_pages would be more appropriate. > &...

...+ list_for_each(curr, &zone->free_area[order].free_list[t]) { + pfn = page_to_pfn(list_entry(curr, struct page, lru)); + if (pfn >= start_pfn && pfn <= end_pfn) { + page_num = 1UL << order; + if (pfn + page_num > end_pfn) + page_num = end_pfn - pfn; + bitmap_set(bitmap, pfn - start_pfn, page_num); + } + } + } + + spin_unlock_irqrestore(&zone->lock, flags); +} + +int get_free_pages(unsigned long start_pfn, unsigned long end_pfn, + unsigned long *bitmap, unsigned long len) +{ + struct zone *zone; + int ret = 0; + + if (bitmap == NULL || start_pfn...

...ist[t]) { > > + pfn = page_to_pfn(list_entry(curr, struct page, lru)); > > + if (pfn >= start_pfn && pfn <= end_pfn) { > > + page_num = 1UL << order; > > + if (pfn + page_num > end_pfn) > > + page_num = end_pfn - pfn; > > + bitmap_set(bitmap, pfn - start_pfn, > page_num); > > + } > > + } > > + } > > Nit: The 'page_num' nomenclature really confused me here. It is the > number of bits being set in the bitmap. Seems like calling it nr_pages or > num_pages would be more appropriate. &g...

...n(list_entry(curr, struct page, lru)); > >>> + if (pfn >= start_pfn && pfn <= end_pfn) { > >>> + page_num = 1UL << order; > >>> + if (pfn + page_num > end_pfn) > >>> + page_num = end_pfn - pfn; > >>> + bitmap_set(bitmap, pfn - start_pfn, > page_num); > >>> + } > >>> + } > >>> + } > >> > >> Nit: The 'page_num' nomenclature really confused me here. It is the > >> number of bits being set in the bitmap. Seems like calling it > &...

...n(list_entry(curr, struct page, lru)); > >>> + if (pfn >= start_pfn && pfn <= end_pfn) { > >>> + page_num = 1UL << order; > >>> + if (pfn + page_num > end_pfn) > >>> + page_num = end_pfn - pfn; > >>> + bitmap_set(bitmap, pfn - start_pfn, page_num); > >>> + } > >>> + } > >>> + } > >> > >> Nit: The 'page_num' nomenclature really confused me here. It is the > >> number of bits being set in the bitmap. Seems like calling it nr_pages &g...

Should we use attribute((nonnull)) at all? There's a very interesting history of this in libvirt -- try looking at commit eefb881 plus the commits referencing eefb881 -- but it does seem to work for me using recent GCC and Clang. I only did a few functions because annotating them gets old quickly... Rich.

...ne->free_area[order].free_list[t]) { > + pfn = page_to_pfn(list_entry(curr, struct page, lru)); > + if (pfn >= start_pfn && pfn <= end_pfn) { > + page_num = 1UL << order; > + if (pfn + page_num > end_pfn) > + page_num = end_pfn - pfn; > + bitmap_set(bitmap, pfn - start_pfn, page_num); > + } > + } > + } > + > + spin_unlock_irqrestore(&zone->lock, flags); > +} > + > +int get_free_pages(unsigned long start_pfn, unsigned long end_pfn, > + unsigned long *bitmap, unsigned long len) > +{ > + struct zone *zo...

...ne->free_area[order].free_list[t]) { > + pfn = page_to_pfn(list_entry(curr, struct page, lru)); > + if (pfn >= start_pfn && pfn <= end_pfn) { > + page_num = 1UL << order; > + if (pfn + page_num > end_pfn) > + page_num = end_pfn - pfn; > + bitmap_set(bitmap, pfn - start_pfn, page_num); > + } > + } > + } > + > + spin_unlock_irqrestore(&zone->lock, flags); > +} > + > +int get_free_pages(unsigned long start_pfn, unsigned long end_pfn, > + unsigned long *bitmap, unsigned long len) > +{ > + struct zone *zo...

...in bytes. */ +static inline unsigned +bitmap_get (const struct bitmap *bm, uint64_t offset, unsigned default_) +{ + return bitmap_get_blk (bm, offset / bm->blksize, default_); +} + +/* Set the bit(s) associated with the given block. + * If out of range, it is ignored. + */ +static inline void +bitmap_set_blk (const struct bitmap *bm, uint64_t blk, unsigned v) +{ + uint64_t blk_offset = blk / (8 / bm->bpb); + unsigned blk_bit = bm->bpb * (blk % (8 / bm->bpb)); + + if (blk_offset >= bm->size) { + nbdkit_debug ("bitmap_set: block number is out of range"); + return; +...

...in bytes. */ +static inline unsigned +bitmap_get (const struct bitmap *bm, uint64_t offset, unsigned default_) +{ + return bitmap_get_blk (bm, offset / bm->blksize, default_); +} + +/* Set the bit(s) associated with the given block. + * If out of range, it is ignored. + */ +static inline void +bitmap_set_blk (const struct bitmap *bm, uint64_t blk, unsigned v) +{ + uint64_t blk_offset = blk / (8 / bm->bpb); + unsigned blk_bit = bm->bpb * (blk % (8 / bm->bpb)); + + if (blk_offset >= bm->size) { + nbdkit_debug ("bitmap_set: block number is out of range"); + return; +...