lkml.org 
[lkml]   [2020]   [Jan]   [21]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
 
Messages in this thread
    /
    SubjectRe: [Patch v3 1/3] iommu: avoid unnecessary magazine allocations
    From
    Date
    On 18/12/2019 4:39 am, Cong Wang wrote:
    > The IOVA cache algorithm implemented in IOMMU code does not
    > exactly match the original algorithm described in the paper
    > "Magazines and Vmem: Extending the Slab Allocator to Many
    > CPUs and Arbitrary Resources".
    >
    > Particularly, it doesn't need to free the loaded empty magazine
    > when trying to put it back to global depot. To make it work, we
    > have to pre-allocate magazines in the depot and only recycle them
    > when all of them are full.
    >
    > Before this patch, rcache->depot[] contains either full or
    > freed entries, after this patch, it contains either full or
    > empty (but allocated) entries.

    How much additional memory overhead does this impose (particularly on
    systems that may have many domains mostly used for large, long-term
    mappings)? I'm wary that trying to micro-optimise for the "churn network
    packets as fast as possible" case may penalise every other case,
    potentially quite badly. Lower-end embedded systems are using IOMMUs in
    front of their GPUs, video codecs, etc. precisely because they *don't*
    have much memory to spare (and thus need to scrape together large
    buffers out of whatever pages they can find).

    But on the other hand, if we were to go down this route, then why is
    there any dynamic allocation/freeing left at all? Once both the depot
    and the rcaches are preallocated, then AFAICS it would make more sense
    to rework the overflow case in __iova_rcache_insert() to just free the
    IOVAs and swap the empty mag around rather than destroying and
    recreating it entirely.

    Perhaps there's a reasonable compromise wherein we don't preallocate,
    but still 'free' empty magazines back to the depot, such that busy
    domains will quickly reach a steady-state. In fact, having now dug up
    the paper at this point of writing this reply, that appears to be what
    fig. 3.1b describes anyway - I don't see any mention of preallocating
    the depot.

    Robin.

    >
    > Together with a few other changes to make it exactly match
    > the pseudo code in the paper.
    >
    > Cc: Joerg Roedel <joro@8bytes.org>
    > Cc: John Garry <john.garry@huawei.com>
    > Signed-off-by: Cong Wang <xiyou.wangcong@gmail.com>
    > ---
    > drivers/iommu/iova.c | 45 +++++++++++++++++++++++++++-----------------
    > 1 file changed, 28 insertions(+), 17 deletions(-)
    >
    > diff --git a/drivers/iommu/iova.c b/drivers/iommu/iova.c
    > index 41c605b0058f..cb473ddce4cf 100644
    > --- a/drivers/iommu/iova.c
    > +++ b/drivers/iommu/iova.c
    > @@ -862,12 +862,16 @@ static void init_iova_rcaches(struct iova_domain *iovad)
    > struct iova_cpu_rcache *cpu_rcache;
    > struct iova_rcache *rcache;
    > unsigned int cpu;
    > - int i;
    > + int i, j;
    >
    > for (i = 0; i < IOVA_RANGE_CACHE_MAX_SIZE; ++i) {
    > rcache = &iovad->rcaches[i];
    > spin_lock_init(&rcache->lock);
    > rcache->depot_size = 0;
    > + for (j = 0; j < MAX_GLOBAL_MAGS; ++j) {
    > + rcache->depot[j] = iova_magazine_alloc(GFP_KERNEL);
    > + WARN_ON(!rcache->depot[j]);
    > + }
    > rcache->cpu_rcaches = __alloc_percpu(sizeof(*cpu_rcache), cache_line_size());
    > if (WARN_ON(!rcache->cpu_rcaches))
    > continue;
    > @@ -900,24 +904,30 @@ static bool __iova_rcache_insert(struct iova_domain *iovad,
    >
    > if (!iova_magazine_full(cpu_rcache->loaded)) {
    > can_insert = true;
    > - } else if (!iova_magazine_full(cpu_rcache->prev)) {
    > + } else if (iova_magazine_empty(cpu_rcache->prev)) {
    > swap(cpu_rcache->prev, cpu_rcache->loaded);
    > can_insert = true;
    > } else {
    > - struct iova_magazine *new_mag = iova_magazine_alloc(GFP_ATOMIC);
    > + spin_lock(&rcache->lock);
    > + if (rcache->depot_size < MAX_GLOBAL_MAGS) {
    > + swap(rcache->depot[rcache->depot_size], cpu_rcache->prev);
    > + swap(cpu_rcache->prev, cpu_rcache->loaded);
    > + rcache->depot_size++;
    > + can_insert = true;
    > + } else {
    > + mag_to_free = cpu_rcache->loaded;
    > + }
    > + spin_unlock(&rcache->lock);
    > +
    > + if (mag_to_free) {
    > + struct iova_magazine *new_mag = iova_magazine_alloc(GFP_ATOMIC);
    >
    > - if (new_mag) {
    > - spin_lock(&rcache->lock);
    > - if (rcache->depot_size < MAX_GLOBAL_MAGS) {
    > - rcache->depot[rcache->depot_size++] =
    > - cpu_rcache->loaded;
    > + if (new_mag) {
    > + cpu_rcache->loaded = new_mag;
    > + can_insert = true;
    > } else {
    > - mag_to_free = cpu_rcache->loaded;
    > + mag_to_free = NULL;
    > }
    > - spin_unlock(&rcache->lock);
    > -
    > - cpu_rcache->loaded = new_mag;
    > - can_insert = true;
    > }
    > }
    >
    > @@ -963,14 +973,15 @@ static unsigned long __iova_rcache_get(struct iova_rcache *rcache,
    >
    > if (!iova_magazine_empty(cpu_rcache->loaded)) {
    > has_pfn = true;
    > - } else if (!iova_magazine_empty(cpu_rcache->prev)) {
    > + } else if (iova_magazine_full(cpu_rcache->prev)) {
    > swap(cpu_rcache->prev, cpu_rcache->loaded);
    > has_pfn = true;
    > } else {
    > spin_lock(&rcache->lock);
    > if (rcache->depot_size > 0) {
    > - iova_magazine_free(cpu_rcache->loaded);
    > - cpu_rcache->loaded = rcache->depot[--rcache->depot_size];
    > + swap(rcache->depot[rcache->depot_size - 1], cpu_rcache->prev);
    > + swap(cpu_rcache->prev, cpu_rcache->loaded);
    > + rcache->depot_size--;
    > has_pfn = true;
    > }
    > spin_unlock(&rcache->lock);
    > @@ -1019,7 +1030,7 @@ static void free_iova_rcaches(struct iova_domain *iovad)
    > iova_magazine_free(cpu_rcache->prev);
    > }
    > free_percpu(rcache->cpu_rcaches);
    > - for (j = 0; j < rcache->depot_size; ++j)
    > + for (j = 0; j < MAX_GLOBAL_MAGS; ++j)
    > iova_magazine_free(rcache->depot[j]);
    > }
    > }
    >

    \
     
     \ /
      Last update: 2020-01-21 12:11    [W:4.601 / U:0.300 seconds]
    ©2003-2020 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site