search for: pgoff

...+ * to the page while we read it in the page + * could not be made volatile. Do it now. + */ + page_make_volatile(page, 2); goto page_ok; } @@ -1576,13 +1582,13 @@ retry_all: * Do we have something in the page cache already? */ retry_find: - page = find_get_page(mapping, pgoff); + page = find_get_page_nodiscard(mapping, pgoff); if (prefer_adaptive_readahead() && VM_SequentialReadHint(area)) { if (!page) { page_cache_readahead_adaptive(mapping, ra, file, NULL, NULL, pgoff, pgoff, pgoff + 1); - page = find_get_page(mapping, pgoff); + page...

...+ * to the page while we read it in the page + * could not be made volatile. Do it now. + */ + page_make_volatile(page, 2); goto page_ok; } @@ -1576,13 +1582,13 @@ retry_all: * Do we have something in the page cache already? */ retry_find: - page = find_get_page(mapping, pgoff); + page = find_get_page_nodiscard(mapping, pgoff); if (prefer_adaptive_readahead() && VM_SequentialReadHint(area)) { if (!page) { page_cache_readahead_adaptive(mapping, ra, file, NULL, NULL, pgoff, pgoff, pgoff + 1); - page = find_get_page(mapping, pgoff); + page...

...picked the commits from upstream. You should consider pushing it to the archive. root at ibmx3650m2-210-147:/usr/src/blktap-2.0.91# diff -Naru ring.c /tmp/ring.c --- ring.c 2013-09-17 11:47:27.468200000 -0400 +++ /tmp/ring.c 2013-09-17 11:47:56.836200000 -0400 @@ -210,7 +210,7 @@ pgoff = 1 + request->usr_idx * BLKTAP_SEGMENT_MAX; - addr = do_mmap_pgoff(filp, addr, len, prot, flags, pgoff); + addr = vm_mmap(filp, addr, len, prot, flags, pgoff << PAGE_SHIFT); return IS_ERR_VALUE(addr) ? addr : 0; } @@ -230,8 +230,7 @@ addr = MMAP_VADDR(rin...

...include/linux/pagemap.h --- linux-2.6/include/linux/pagemap.h 2006-09-01 12:50:23.000000000 +0200 +++ linux-2.6-patched/include/linux/pagemap.h 2006-09-01 12:50:23.000000000 +0200 @@ -85,6 +85,13 @@ unsigned find_get_pages_contig(struct ad unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index, int tag, unsigned int nr_pages, struct page **pages); +#if defined(CONFIG_PAGE_STATES) +extern struct page * find_get_page_nodiscard(struct address_space *mapping, + unsigned long index); +#else +#define find_get_page_nodiscard(mapping, index) find_get_page(mapping, index)...

...include/linux/pagemap.h --- linux-2.6/include/linux/pagemap.h 2006-09-01 12:50:23.000000000 +0200 +++ linux-2.6-patched/include/linux/pagemap.h 2006-09-01 12:50:23.000000000 +0200 @@ -85,6 +85,13 @@ unsigned find_get_pages_contig(struct ad unsigned find_get_pages_tag(struct address_space *mapping, pgoff_t *index, int tag, unsigned int nr_pages, struct page **pages); +#if defined(CONFIG_PAGE_STATES) +extern struct page * find_get_page_nodiscard(struct address_space *mapping, + unsigned long index); +#else +#define find_get_page_nodiscard(mapping, index) find_get_page(mapping, index)...

...WAP_AGAIN; + unsigned long address; unsigned long cursor; unsigned long max_nl_cursor = 0; unsigned long max_nl_size = 0; @@ -826,7 +827,10 @@ static int try_to_unmap_file(struct page spin_lock(&mapping->i_mmap_lock); vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff) { - ret = try_to_unmap_one(page, vma, migration); + address = vma_address(page, vma); + if (address == -EFAULT) + continue; + ret = try_to_unmap_one(page, vma, address, migration); if (ret == SWAP_FAIL || !page_mapped(page)) goto out; } @@ -927,3 +931,51 @@ int try_to_unmap(s...

...WAP_AGAIN; + unsigned long address; unsigned long cursor; unsigned long max_nl_cursor = 0; unsigned long max_nl_size = 0; @@ -826,7 +827,10 @@ static int try_to_unmap_file(struct page spin_lock(&mapping->i_mmap_lock); vma_prio_tree_foreach(vma, &iter, &mapping->i_mmap, pgoff, pgoff) { - ret = try_to_unmap_one(page, vma, migration); + address = vma_address(page, vma); + if (address == -EFAULT) + continue; + ret = try_to_unmap_one(page, vma, address, migration); if (ret == SWAP_FAIL || !page_mapped(page)) goto out; } @@ -927,3 +931,51 @@ int try_to_unmap(s...

...e doorbell layout (usually > virtio-pci), this patch starts from a doorbell per page > model. Vhost-vdpa only support the hardware doorbell that sit at the > boundary of a page and does not share the page with other registers. > > Doorbell of each virtqueue must be mapped separately, pgoff is the > index of the virtqueue. This allows userspace to map a subset of the > doorbell which may be useful for the implementation of software > assisted virtqueue (control vq) in the future. > > Signed-off-by: Jason Wang <jasowang at redhat.com> > --- > drivers/vhost/v...

Greetings, after Carsten pitched CMM2 on the kvm mini summit here is a repost of version 5 of the guest page hinting patches. The code is still the same but has been adapted to the latest git level. -- blue skies, Martin. "Reality continues to ruin my life." - Calvin.

Greetings, after Carsten pitched CMM2 on the kvm mini summit here is a repost of version 5 of the guest page hinting patches. The code is still the same but has been adapted to the latest git level. -- blue skies, Martin. "Reality continues to ruin my life." - Calvin.

After way to many months here is the fifth version of the guest page hinting patches. Compared to version four a few improvements have been added: - Avoid page_host_discards() calls outside of page-states.h - The discard list is now implemented via the page_free_discarded hook and architecture specific code. - PG_state_change page flag has been replaced with architecture specficic

After way to many months here is the fifth version of the guest page hinting patches. Compared to version four a few improvements have been added: - Avoid page_host_discards() calls outside of page-states.h - The discard list is now implemented via the page_free_discarded hook and architecture specific code. - PG_state_change page flag has been replaced with architecture specficic

...11-05-18 14:38:26 24651 1108842816] WARNING (XendDomainInfo:2270) Domain has crashed: name=322-2.vhd id=2645. [2011-05-18 14:38:26 24651 1201162560] INFO (XendDomainInfo:2481) Preserving dead domain 322-2.vhd (2645). in dmesg, I saw privcmd_fault: vma=ffff88007b13ef18 7f8cf182e000-7f8cf192e000, pgoff=96, uv=00007f8cf18c4000 There are two VHD file I have, VHDA & VHDB, all have windows 2003 x86 64 rc2 installed, but has some different softwares. Since all crash happened short after VM running, so I did extra test, snapshot VHDA and start VM from snapshot, to see whether VM crash. In 814...

...m_gem_object *obj, int flags); struct drm_gem_object *omap_gem_prime_import(struct drm_device *dev, struct dma_buf *buffer); -vm_fault_t omap_gem_fault(struct vm_fault *vmf); int omap_gem_roll(struct drm_gem_object *obj, u32 roll); void omap_gem_cpu_sync_page(struct drm_gem_object *obj, int pgoff); void omap_gem_dma_sync_buffer(struct drm_gem_object *obj, -- 2.28.0

...userspace modeling of the doorbell layout (usually virtio-pci), this patch starts from a doorbell per page model. Vhost-vdpa only support the hardware doorbell that sit at the boundary of a page and does not share the page with other registers. Doorbell of each virtqueue must be mapped separately, pgoff is the index of the virtqueue. This allows userspace to map a subset of the doorbell which may be useful for the implementation of software assisted virtqueue (control vq) in the future. Signed-off-by: Jason Wang <jasowang at redhat.com> --- drivers/vhost/vdpa.c | 59 +++++++++++++++++++++++...

...sually >> virtio-pci), this patch starts from a doorbell per page >> model. Vhost-vdpa only support the hardware doorbell that sit at the >> boundary of a page and does not share the page with other registers. >> >> Doorbell of each virtqueue must be mapped separately, pgoff is the >> index of the virtqueue. This allows userspace to map a subset of the >> doorbell which may be useful for the implementation of software >> assisted virtqueue (control vq) in the future. >> >> Signed-off-by: Jason Wang <jasowang at redhat.com> >> -...

...el/events/core.c +++ b/kernel/events/core.c @@ -3466,6 +3466,7 @@ static int perf_mmap_fault(struct vm_area_struct *vma, struct vm_fault *vmf) struct ring_buffer *rb; int ret = VM_FAULT_SIGBUS; + file_update_time(vma->vm_file); if (vmf->flags & FAULT_FLAG_MKWRITE) { if (vmf->pgoff == 0) ret = 0; diff --git a/mm/memory.c b/mm/memory.c index a56e3ba..8e4686f 100644 --- a/mm/memory.c +++ b/mm/memory.c @@ -2619,10 +2619,6 @@ reuse: } } - /* file_update_time outside page_lock */ - if (vma->vm_file) - file_update_time(vma->vm_file); - return ret; } @@...

...ro ?BLKTAP_RING_SIZE? struct blktap_request *pending[BLKTAP_RING_SIZE]; ^ /var/lib/dkms/blktap/2.0.91/build/ring.c: In function ?blktap_ring_map_request?: /var/lib/dkms/blktap/2.0.91/build/ring.c:206:2: error: too few arguments to function ?do_mmap_pgoff? addr = do_mmap_pgoff(filp, addr, len, prot, flags, pgoff); ^ In file included from /usr/src/linux-headers-3.16-2-common/include/linux/pagemap.h:7:0, from /usr/src/linux-headers-3.16-2-common/include/linux/blkdev.h:14, from /var/lib/dkms/blktap/2.0.91/build/rin...

Hi, I have this bug when I activate java-6-jdk on Tomcat 5.5.26 in a Lenny Guest : Jul 31 21:24:15 tomcat01 kernel: 1 multicall(s) failed: cpu 3 Jul 31 21:24:15 tomcat01 kernel: call 1/1: op=14 arg=[b7f1a000] result=-22 Jul 31 21:24:15 tomcat01 kernel: ------------[ cut here ]------------ Jul 31 21:24:15 tomcat01 kernel: kernel BUG at arch/x86/xen/multicalls.c:103! Jul 31 21:24:15 tomcat01

Hi all: This series introduce basic functionality of doorbell mapping support for vhost-vDPA. Userspace program may use mmap() to map a the doorbell of a specific virtqueue into its address space. This is help to reudce the syscall or vmexit overhead. A new vdpa_config_ops was introduced to report the location of the doorbell, vhost_vdpa may then choose to map the doorbell when: - The doorbell