search for: kmemleak_ignor

...the tests in tools/virtio. The following patches are build fixes for those changes, as well as a fix for a typo that would have never built. The changes were tested on my amd64 system against 3.14-rc2. Joel Stanley (3): tools/virtio: update internal copies of headers tools/virtio: fix missing kmemleak_ignore symbol tools/virtio: add a missing ) drivers/virtio/virtio_ring.c | 1 + tools/virtio/linux/kmemleak.h | 3 +++ tools/virtio/linux/virtio.h | 4 ++-- tools/virtio/virtio_test.c | 2 +- 4 files changed, 7 insertions(+), 3 deletions(-) create mode 100644 tools/virtio/linux/kmemleak.h --...

...the tests in tools/virtio. The following patches are build fixes for those changes, as well as a fix for a typo that would have never built. The changes were tested on my amd64 system against 3.14-rc2. Joel Stanley (3): tools/virtio: update internal copies of headers tools/virtio: fix missing kmemleak_ignore symbol tools/virtio: add a missing ) drivers/virtio/virtio_ring.c | 1 + tools/virtio/linux/kmemleak.h | 3 +++ tools/virtio/linux/virtio.h | 4 ++-- tools/virtio/virtio_test.c | 2 +- 4 files changed, 7 insertions(+), 3 deletions(-) create mode 100644 tools/virtio/linux/kmemleak.h --...

...ff81502645>] igmp6_timer_handler+0x95/0xa0 [<ffffffff8104b57c>] call_timer_fn+0x2c/0x90 [<ffffffff8104b7ba>] run_timer_softirq+0x1da/0x1f0 [<ffffffff81045721>] __do_softirq+0xd1/0x1b0 May it be that the allocated memory within vring_add_indirect should be marked as kmemleak_ignore(), because it is mapped from a virtual to a physical address and thus kmemleak cannot detect that the memory is actually still being referenced. Thanks for your help, Christoph

...ff81502645>] igmp6_timer_handler+0x95/0xa0 [<ffffffff8104b57c>] call_timer_fn+0x2c/0x90 [<ffffffff8104b7ba>] run_timer_softirq+0x1da/0x1f0 [<ffffffff81045721>] __do_softirq+0xd1/0x1b0 May it be that the allocated memory within vring_add_indirect should be marked as kmemleak_ignore(), because it is mapped from a virtual to a physical address and thus kmemleak cannot detect that the memory is actually still being referenced. Thanks for your help, Christoph

...andler+0x95/0xa0 > [<ffffffff8104b57c>] call_timer_fn+0x2c/0x90 > [<ffffffff8104b7ba>] run_timer_softirq+0x1da/0x1f0 > [<ffffffff81045721>] __do_softirq+0xd1/0x1b0 > > May it be that the allocated memory within vring_add_indirect should be marked > as kmemleak_ignore(), because it is mapped from a virtual to a physical > address and thus kmemleak cannot detect that the memory is actually still > being referenced. > > > Thanks for your help, > Christoph Thanks! Does this work? virtio_ring: plug kmemleak false positive. unreferenced object...

...se a single buffer which doesn't continue */ > + vq->vring.desc[head].flags = VRING_DESC_F_INDIRECT; > + vq->vring.desc[head].addr = virt_to_phys(desc); > + /* avoid kmemleak false positive (hidden by virt_to_phys) */ > + kmemleak_ignore(desc); > + vq->vring.desc[head].len = total_sg * sizeof(struct vring_desc); > + > + /* Set up rest to use this indirect table. */ > + i = 0; > + total_sg = 1; This is a little too magical for me. Would it make sense to add...

...se a single buffer which doesn't continue */ > + vq->vring.desc[head].flags = VRING_DESC_F_INDIRECT; > + vq->vring.desc[head].addr = virt_to_phys(desc); > + /* avoid kmemleak false positive (hidden by virt_to_phys) */ > + kmemleak_ignore(desc); > + vq->vring.desc[head].len = total_sg * sizeof(struct vring_desc); > + > + /* Set up rest to use this indirect table. */ > + i = 0; > + total_sg = 1; This is a little too magical for me. Would it make sense to add...

...eue length though. Cheers, Rusty. ---------------------------------------------------------------- Alexander Gordeev (1): virtio: Use pci_enable_msix_exact() instead of pci_enable_msix() Joel Stanley (3): tools/virtio: update internal copies of headers tools/virtio: fix missing kmemleak_ignore symbol tools/virtio: add a missing ) Randy Dunlap (1): MAINTAINERS: virtio-dev is subscribers only Rusty Russell (8): virtio_balloon: don't softlockup on huge balloon changes. virtio_net: don't crash if virtqueue is broken. virtio_blk: don't crash, repor...

...eue length though. Cheers, Rusty. ---------------------------------------------------------------- Alexander Gordeev (1): virtio: Use pci_enable_msix_exact() instead of pci_enable_msix() Joel Stanley (3): tools/virtio: update internal copies of headers tools/virtio: fix missing kmemleak_ignore symbol tools/virtio: add a missing ) Randy Dunlap (1): MAINTAINERS: virtio-dev is subscribers only Rusty Russell (8): virtio_balloon: don't softlockup on huge balloon changes. virtio_net: don't crash if virtqueue is broken. virtio_blk: don't crash, repor...

...ring.desc[head].flags = > > + cpu_to_virtio16(vq->vq.vdev, VRING_DESC_F_INDIRECT); > > + vq->vring.desc[head].addr = > > + cpu_to_virtio64(vq->vq.vdev, virt_to_phys(desc)); > > /* kmemleak gives a false positive, as it's hidden by virt_to_phys */ > > kmemleak_ignore(desc); > > - vq->vring.desc[head].len = i * sizeof(struct vring_desc); > > + vq->vring.desc[head].len = > > + cpu_to_virtio32(vq->vq.vdev, i * sizeof(struct vring_desc)); > > > > - /* Update free pointer */ > > + BUG_ON(i != total_sg); > > + &g...

...ring.desc[head].flags = > > + cpu_to_virtio16(vq->vq.vdev, VRING_DESC_F_INDIRECT); > > + vq->vring.desc[head].addr = > > + cpu_to_virtio64(vq->vq.vdev, virt_to_phys(desc)); > > /* kmemleak gives a false positive, as it's hidden by virt_to_phys */ > > kmemleak_ignore(desc); > > - vq->vring.desc[head].len = i * sizeof(struct vring_desc); > > + vq->vring.desc[head].len = > > + cpu_to_virtio32(vq->vq.vdev, i * sizeof(struct vring_desc)); > > > > - /* Update free pointer */ > > + BUG_ON(i != total_sg); > > + &g...

...*/ - vq->vq.num_free--; - - /* Use a single buffer which doesn't continue */ - head = vq->free_head; - vq->vring.desc[head].flags = VRING_DESC_F_INDIRECT; - vq->vring.desc[head].addr = virt_to_phys(desc); - /* kmemleak gives a false positive, as it's hidden by virt_to_phys */ - kmemleak_ignore(desc); - vq->vring.desc[head].len = i * sizeof(struct vring_desc); - - /* Update free pointer */ - vq->free_head = vq->vring.desc[head].next; + desc = kmalloc(total_sg * sizeof(struct vring_desc), gfp); + if (!desc) + return NULL; - return head; + for (i = 0; i < total_sg; i++) +...

I resurrected these patches after prompting from Andy Lutomirski's recent patches. I put them on the back-burner because vring_bench had a 15% slowdown on my laptop: pktgen testing revealed a speedup, if anything, so I've cleaned them up. Rusty Russell (3): virtio_net: pass well-formed sgs to virtqueue_add_*() virtio_ring: assume sgs are always well-formed. virtio_ring: unify

I resurrected these patches after prompting from Andy Lutomirski's recent patches. I put them on the back-burner because vring_bench had a 15% slowdown on my laptop: pktgen testing revealed a speedup, if anything, so I've cleaned them up. Rusty Russell (3): virtio_net: pass well-formed sgs to virtqueue_add_*() virtio_ring: assume sgs are always well-formed. virtio_ring: unify

...t(struct vring_virtqueue *vq, > head = vq->free_head; > vq->vring.desc[head].flags = VRING_DESC_F_INDIRECT; > vq->vring.desc[head].addr = virt_to_phys(desc); > + /* kmemleak gives a false positive, as it's hidden by virt_to_phys */ > + kmemleak_ignore(desc); > vq->vring.desc[head].len = i * sizeof(struct vring_desc); > > /* Update free pointer */

...Ah, thanks a lot for posting this. I had this fixed in my tree but forgot to send. For the series: Acked-by: Michael S. Tsirkin <mst at redhat.com> Rusty, will you pick this up? > Joel Stanley (3): > tools/virtio: update internal copies of headers > tools/virtio: fix missing kmemleak_ignore symbol > tools/virtio: add a missing ) > > drivers/virtio/virtio_ring.c | 1 + > tools/virtio/linux/kmemleak.h | 3 +++ > tools/virtio/linux/virtio.h | 4 ++-- > tools/virtio/virtio_test.c | 2 +- > 4 files changed, 7 insertions(+), 3 deletions(-) > create mode 1...

...gt; ---------------------------------------------------------------- > Alexander Gordeev (1): > virtio: Use pci_enable_msix_exact() instead of pci_enable_msix() > > Joel Stanley (3): > tools/virtio: update internal copies of headers > tools/virtio: fix missing kmemleak_ignore symbol > tools/virtio: add a missing ) > > Randy Dunlap (1): > MAINTAINERS: virtio-dev is subscribers only > > Rusty Russell (8): > virtio_balloon: don't softlockup on huge balloon changes. > virtio_net: don't crash if virtqueue is broken...

...t(struct vring_virtqueue *vq, > head = vq->free_head; > vq->vring.desc[head].flags = VRING_DESC_F_INDIRECT; > vq->vring.desc[head].addr = virt_to_phys(desc); > + /* kmemleak gives a false positive, as it's hidden by virt_to_phys */ > + kmemleak_ignore(desc); > vq->vring.desc[head].len = i * sizeof(struct vring_desc); > > /* Update free pointer */

...sc[head].addr = virt_to_phys(desc); + vq->vring.desc[head].flags = + cpu_to_virtio_u16(vq->vq.vdev, VRING_DESC_F_INDIRECT); + vq->vring.desc[head].addr = + cpu_to_virtio_u64(vq->vq.vdev, virt_to_phys(desc)); /* kmemleak gives a false positive, as it's hidden by virt_to_phys */ kmemleak_ignore(desc); - vq->vring.desc[head].len = i * sizeof(struct vring_desc); + vq->vring.desc[head].len = + cpu_to_virtio_u32(vq->vq.vdev, i * sizeof(struct vring_desc)); - /* Update free pointer */ + BUG_ON(i != total_sg); + + /* Update free pointer (we store this in native endian) */ vq->...

...sc[head].addr = virt_to_phys(desc); + vq->vring.desc[head].flags = + cpu_to_virtio_u16(vq->vq.vdev, VRING_DESC_F_INDIRECT); + vq->vring.desc[head].addr = + cpu_to_virtio_u64(vq->vq.vdev, virt_to_phys(desc)); /* kmemleak gives a false positive, as it's hidden by virt_to_phys */ kmemleak_ignore(desc); - vq->vring.desc[head].len = i * sizeof(struct vring_desc); + vq->vring.desc[head].len = + cpu_to_virtio_u32(vq->vq.vdev, i * sizeof(struct vring_desc)); - /* Update free pointer */ + BUG_ON(i != total_sg); + + /* Update free pointer (we store this in native endian) */ vq->...