* [PATCH v2] iommu: Do physical merging in iommu_map_sg()
@ 2018-10-11 15:56 Robin Murphy
[not found] ` <90d22e679e64e9830ebe993665e2c04a612fdaf9.1539018455.git.robin.murphy-5wv7dgnIgG8@public.gmane.org>
0 siblings, 1 reply; 2+ messages in thread
From: Robin Murphy @ 2018-10-11 15:56 UTC (permalink / raw)
To: joro-zLv9SwRftAIdnm+yROfE0A
Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA, hch-jcswGhMUV9g
The original motivation for iommu_map_sg() was to give IOMMU drivers the
chance to map an IOVA-contiguous scatterlist as efficiently as they
could. It turns out that there isn't really much driver-specific
business involved there, so now that the default implementation is
mandatory let's just improve that - the main thing we're after is to use
larger pages wherever possible, and as long as domain->pgsize_bitmap
reflects reality, iommu_map() can already do that in a generic way. All
we need to do is detect physically-contiguous segments and batch them
into a single map operation, since whatever we do here is transparent to
our caller and not bound by any segment-length restrictions on the list
itself.
Speaking of efficiency, there's really very little point in duplicating
the checks that iommu_map() is going to do anyway, so those get cleared
up in the process.
Signed-off-by: Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org>
---
One final attempt at a clean-ish loop without descending into real
code-golf madness...
drivers/iommu/iommu.c | 41 ++++++++++++++++++++---------------------
1 file changed, 20 insertions(+), 21 deletions(-)
diff --git a/drivers/iommu/iommu.c b/drivers/iommu/iommu.c
index 8c15c5980299..ea8ce6322714 100644
--- a/drivers/iommu/iommu.c
+++ b/drivers/iommu/iommu.c
@@ -1676,33 +1676,32 @@ EXPORT_SYMBOL_GPL(iommu_unmap_fast);
size_t iommu_map_sg(struct iommu_domain *domain, unsigned long iova,
struct scatterlist *sg, unsigned int nents, int prot)
{
- struct scatterlist *s;
- size_t mapped = 0;
- unsigned int i, min_pagesz;
+ size_t len = 0, mapped = 0;
+ phys_addr_t start;
+ unsigned int i = 0;
int ret;
- if (unlikely(domain->pgsize_bitmap == 0UL))
- return 0;
+ while (i <= nents) {
+ phys_addr_t s_phys = sg_phys(sg);
- min_pagesz = 1 << __ffs(domain->pgsize_bitmap);
+ if (len && s_phys != start + len) {
+ ret = iommu_map(domain, iova + mapped, start, len, prot);
+ if (ret)
+ goto out_err;
- for_each_sg(sg, s, nents, i) {
- phys_addr_t phys = page_to_phys(sg_page(s)) + s->offset;
+ mapped += len;
+ len = 0;
+ }
- /*
- * We are mapping on IOMMU page boundaries, so offset within
- * the page must be 0. However, the IOMMU may support pages
- * smaller than PAGE_SIZE, so s->offset may still represent
- * an offset of that boundary within the CPU page.
- */
- if (!IS_ALIGNED(s->offset, min_pagesz))
- goto out_err;
+ if (len) {
+ len += sg->length;
+ } else {
+ len = sg->length;
+ start = s_phys;
+ }
- ret = iommu_map(domain, iova + mapped, phys, s->length, prot);
- if (ret)
- goto out_err;
-
- mapped += s->length;
+ if (++i < nents)
+ sg = sg_next(sg);
}
return mapped;
--
2.19.0.dirty
^ permalink raw reply related [flat|nested] 2+ messages in thread
* Re: [PATCH v2] iommu: Do physical merging in iommu_map_sg()
[not found] ` <90d22e679e64e9830ebe993665e2c04a612fdaf9.1539018455.git.robin.murphy-5wv7dgnIgG8@public.gmane.org>
@ 2018-11-06 15:38 ` Joerg Roedel
0 siblings, 0 replies; 2+ messages in thread
From: Joerg Roedel @ 2018-11-06 15:38 UTC (permalink / raw)
To: Robin Murphy
Cc: iommu-cunTk1MwBs9QetFLy7KEm3xJsTq8ys+cHZ5vskTnxNA, hch-jcswGhMUV9g
On Thu, Oct 11, 2018 at 04:56:42PM +0100, Robin Murphy wrote:
> The original motivation for iommu_map_sg() was to give IOMMU drivers the
> chance to map an IOVA-contiguous scatterlist as efficiently as they
> could. It turns out that there isn't really much driver-specific
> business involved there, so now that the default implementation is
> mandatory let's just improve that - the main thing we're after is to use
> larger pages wherever possible, and as long as domain->pgsize_bitmap
> reflects reality, iommu_map() can already do that in a generic way. All
> we need to do is detect physically-contiguous segments and batch them
> into a single map operation, since whatever we do here is transparent to
> our caller and not bound by any segment-length restrictions on the list
> itself.
>
> Speaking of efficiency, there's really very little point in duplicating
> the checks that iommu_map() is going to do anyway, so those get cleared
> up in the process.
>
> Signed-off-by: Robin Murphy <robin.murphy-5wv7dgnIgG8@public.gmane.org>
Looks correct, applied, thanks.
^ permalink raw reply [flat|nested] 2+ messages in thread
end of thread, other threads:[~2018-11-06 15:38 UTC | newest]
Thread overview: 2+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2018-10-11 15:56 [PATCH v2] iommu: Do physical merging in iommu_map_sg() Robin Murphy
[not found] ` <90d22e679e64e9830ebe993665e2c04a612fdaf9.1539018455.git.robin.murphy-5wv7dgnIgG8@public.gmane.org>
2018-11-06 15:38 ` Joerg Roedel
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).