search for: unsafe_get_user

...This idea has been used for high speed userspace backend for years, e.g packet socket or recent AF_XDP. The only difference is the page was remap to from kernel to userspace. > I don't > like the idea I have to say. As a first step, why don't we switch to > unsafe_put_user/unsafe_get_user etc? Several reasons: - They only have x86 variant, it won't have any difference for the rest of architecture. - unsafe_put_user/unsafe_get_user is not sufficient for accessing structures (e.g accessing descriptor) or arrays (batching). - Unless we can batch at least the accessing of two...

...This idea has been used for high speed userspace backend for years, e.g packet socket or recent AF_XDP. The only difference is the page was remap to from kernel to userspace. > I don't > like the idea I have to say. As a first step, why don't we switch to > unsafe_put_user/unsafe_get_user etc? Several reasons: - They only have x86 variant, it won't have any difference for the rest of architecture. - unsafe_put_user/unsafe_get_user is not sufficient for accessing structures (e.g accessing descriptor) or arrays (batching). - Unless we can batch at least the accessing of two...

...ds the g.u.p mess. I'm still not very clear at the point. We only pin 2 or 4 pages, they're several other cases that will pin much more. > >>> I don't >>> like the idea I have to say. As a first step, why don't we switch to >>> unsafe_put_user/unsafe_get_user etc? >> >> Several reasons: >> >> - They only have x86 variant, it won't have any difference for the rest of >> architecture. > Is there an issue on other architectures? If yes they can be extended > there. Consider the unexpected amount of work and in the...

...ds the g.u.p mess. I'm still not very clear at the point. We only pin 2 or 4 pages, they're several other cases that will pin much more. > >>> I don't >>> like the idea I have to say. As a first step, why don't we switch to >>> unsafe_put_user/unsafe_get_user etc? >> >> Several reasons: >> >> - They only have x86 variant, it won't have any difference for the rest of >> architecture. > Is there an issue on other architectures? If yes they can be extended > there. Consider the unexpected amount of work and in the...

...y designed for priveledged userspace. > The only difference is the page was remap to > from kernel to userspace. At least that avoids the g.u.p mess. > > > I don't > > like the idea I have to say. As a first step, why don't we switch to > > unsafe_put_user/unsafe_get_user etc? > > > Several reasons: > > - They only have x86 variant, it won't have any difference for the rest of > architecture. Is there an issue on other architectures? If yes they can be extended there. > - unsafe_put_user/unsafe_get_user is not sufficient for accessing s...

...ry clear at the point. We only pin 2 or 4 pages, they're > several other cases that will pin much more. > > > > > > > > I don't > > > > like the idea I have to say. As a first step, why don't we switch to > > > > unsafe_put_user/unsafe_get_user etc? > > > > > > Several reasons: > > > > > > - They only have x86 variant, it won't have any difference for the rest of > > > architecture. > > Is there an issue on other architectures? If yes they can be extended > > there. > >...

...if it does just bounce things out to > a thread. I'm not sure what context you meant here. Is this for TX path of TUN? But a fundamental difference is my series is targeted for extreme heavy load not light one, 100% cpu for vhost is expected. > > 2. Switch to unsafe_put_user/unsafe_get_user, > and batch up multiple accesses. As I said, unless we can batch accessing of two difference places of three of avail, descriptor and used. It won't help for batching the accessing of a single place like used. I'm even not sure this can be done consider the case of packed virtq...

...if it does just bounce things out to > a thread. I'm not sure what context you meant here. Is this for TX path of TUN? But a fundamental difference is my series is targeted for extreme heavy load not light one, 100% cpu for vhost is expected. > > 2. Switch to unsafe_put_user/unsafe_get_user, > and batch up multiple accesses. As I said, unless we can batch accessing of two difference places of three of avail, descriptor and used. It won't help for batching the accessing of a single place like used. I'm even not sure this can be done consider the case of packed virtq...

Hi: This series tries to access virtqueue metadata through kernel virtual address instead of copy_user() friends since they had too much overheads like checks, spec barriers or even hardware feature toggling. Test shows about 24% improvement on TX PPS. It should benefit other cases as well. Please review Jason Wang (3): vhost: generalize adding used elem vhost: fine grain userspace memory

Hi: This series tries to access virtqueue metadata through kernel virtual address instead of copy_user() friends since they had too much overheads like checks, spec barriers or even hardware feature toggling. Test shows about 24% improvement on TX PPS. It should benefit other cases as well. Please review Jason Wang (3): vhost: generalize adding used elem vhost: fine grain userspace memory

...ks, spec barriers or even hardware feature > toggling. Userspace accesses through remapping tricks and next time there's a need for a new barrier we are left to figure it out by ourselves. I don't like the idea I have to say. As a first step, why don't we switch to unsafe_put_user/unsafe_get_user etc? That would be more of an apples to apples comparison, would it not? > Test shows about 24% improvement on TX PPS. It should benefit other > cases as well. > > Please review > > Jason Wang (3): > vhost: generalize adding used elem > vhost: fine grain userspace me...

...directly in an atomic context. This should cut latency down significantly, the tricky part is to only do it on a light load and disable this for the streaming case otherwise it's unfair. This might fail, if it does just bounce things out to a thread. 2. Switch to unsafe_put_user/unsafe_get_user, and batch up multiple accesses. 3. Allow adding a fixup point manually, such that multiple independent get_user accesses can get a single fixup (will allow better compiler optimizations). > Jason Wang (3): > vhost: generalize adding used elem > vhost: fine grain user...

...id of the double-underscore version. It's basically > always a mis-optimization due to entirely historical reasons. I can > pretty much guarantee that it's not visible in profiles. > > Linus OK. So maybe we should focus on switching to user_access_begin/end + unsafe_get_user/unsafe_put_user in a loop which does seem to be measureable. That moves the barrier out of the loop, which seems to be consistent with what you would expect. -- MST

...id of the double-underscore version. It's basically > always a mis-optimization due to entirely historical reasons. I can > pretty much guarantee that it's not visible in profiles. > > Linus OK. So maybe we should focus on switching to user_access_begin/end + unsafe_get_user/unsafe_put_user in a loop which does seem to be measureable. That moves the barrier out of the loop, which seems to be consistent with what you would expect. -- MST

On Fri, Nov 02, 2018 at 11:46:36AM +0000, Mark Rutland wrote: > On Thu, Nov 01, 2018 at 04:06:19PM -0700, Linus Torvalds wrote: > > On Thu, Nov 1, 2018 at 4:00 PM Kees Cook <keescook at chromium.org> wrote: > > > > > > + memset(&rsp, 0, sizeof(rsp)); > > > + rsp.response = VIRTIO_SCSI_S_FUNCTION_REJECTED; > > > + resp =

On Fri, Nov 02, 2018 at 11:46:36AM +0000, Mark Rutland wrote: > On Thu, Nov 01, 2018 at 04:06:19PM -0700, Linus Torvalds wrote: > > On Thu, Nov 1, 2018 at 4:00 PM Kees Cook <keescook at chromium.org> wrote: > > > > > > + memset(&rsp, 0, sizeof(rsp)); > > > + rsp.response = VIRTIO_SCSI_S_FUNCTION_REJECTED; > > > + resp =

On Thu, Nov 1, 2018 at 4:00 PM Kees Cook <keescook at chromium.org> wrote: > > + memset(&rsp, 0, sizeof(rsp)); > + rsp.response = VIRTIO_SCSI_S_FUNCTION_REJECTED; > + resp = vq->iov[out].iov_base; > + ret = __copy_to_user(resp, &rsp, sizeof(rsp)); > > Is it actually safe to trust that iov_base has passed an earlier > access_ok()

On Thu, Nov 1, 2018 at 4:00 PM Kees Cook <keescook at chromium.org> wrote: > > + memset(&rsp, 0, sizeof(rsp)); > + rsp.response = VIRTIO_SCSI_S_FUNCTION_REJECTED; > + resp = vq->iov[out].iov_base; > + ret = __copy_to_user(resp, &rsp, sizeof(rsp)); > > Is it actually safe to trust that iov_base has passed an earlier > access_ok()