* + i915-fix-remap_io_sg-to-verify-the-pgprot.patch added to -mm tree
@ 2021-03-28 19:11 akpm
0 siblings, 0 replies; only message in thread
From: akpm @ 2021-03-28 19:11 UTC (permalink / raw)
To: chris, daniel.vetter, hch, jani.nikula, joonas.lahtinen,
mm-commits, peterz, rodrigo.vivi
The patch titled
Subject: i915: fix remap_io_sg to verify the pgprot
has been added to the -mm tree. Its filename is
i915-fix-remap_io_sg-to-verify-the-pgprot.patch
This patch should soon appear at
https://ozlabs.org/~akpm/mmots/broken-out/i915-fix-remap_io_sg-to-verify-the-pgprot.patch
and later at
https://ozlabs.org/~akpm/mmotm/broken-out/i915-fix-remap_io_sg-to-verify-the-pgprot.patch
Before you just go and hit "reply", please:
a) Consider who else should be cc'ed
b) Prefer to cc a suitable mailing list as well
c) Ideally: find the original patch on the mailing list and do a
reply-to-all to that, adding suitable additional cc's
*** Remember to use Documentation/process/submit-checklist.rst when testing your code ***
The -mm tree is included into linux-next and is updated
there every 3-4 working days
------------------------------------------------------
From: Christoph Hellwig <hch@lst.de>
Subject: i915: fix remap_io_sg to verify the pgprot
remap_io_sg claims that that the pgprot is pre-verified using an
io_mapping, but actually does not get passed an io_mapping and just uses
the pgprot in the VMA. Remove the apply_to_page_range abuse and just loop
over remap_pfn_range for each segment.
Note: this could use io_mapping_map_user by passing an iomap to
remap_io_sg if the maintainers can verify that the pgprot in the iomap in
the only caller is indeed the desired one here.
Link: https://lkml.kernel.org/r/20210326055505.1424432-5-hch@lst.de
Signed-off-by: Christoph Hellwig <hch@lst.de>
Cc: Chris Wilson <chris@chris-wilson.co.uk>
Cc: Daniel Vetter <daniel.vetter@ffwll.ch>
Cc: Jani Nikula <jani.nikula@linux.intel.com>
Cc: Joonas Lahtinen <joonas.lahtinen@linux.intel.com>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Rodrigo Vivi <rodrigo.vivi@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
---
drivers/gpu/drm/i915/i915_mm.c | 73 +++++++++----------------------
1 file changed, 23 insertions(+), 50 deletions(-)
--- a/drivers/gpu/drm/i915/i915_mm.c~i915-fix-remap_io_sg-to-verify-the-pgprot
+++ a/drivers/gpu/drm/i915/i915_mm.c
@@ -28,46 +28,10 @@
#include "i915_drv.h"
-struct remap_pfn {
- struct mm_struct *mm;
- unsigned long pfn;
- pgprot_t prot;
-
- struct sgt_iter sgt;
- resource_size_t iobase;
-};
+#define EXPECTED_FLAGS (VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP)
#define use_dma(io) ((io) != -1)
-static inline unsigned long sgt_pfn(const struct remap_pfn *r)
-{
- if (use_dma(r->iobase))
- return (r->sgt.dma + r->sgt.curr + r->iobase) >> PAGE_SHIFT;
- else
- return r->sgt.pfn + (r->sgt.curr >> PAGE_SHIFT);
-}
-
-static int remap_sg(pte_t *pte, unsigned long addr, void *data)
-{
- struct remap_pfn *r = data;
-
- if (GEM_WARN_ON(!r->sgt.sgp))
- return -EINVAL;
-
- /* Special PTE are not associated with any struct page */
- set_pte_at(r->mm, addr, pte,
- pte_mkspecial(pfn_pte(sgt_pfn(r), r->prot)));
- r->pfn++; /* track insertions in case we need to unwind later */
-
- r->sgt.curr += PAGE_SIZE;
- if (r->sgt.curr >= r->sgt.max)
- r->sgt = __sgt_iter(__sg_next(r->sgt.sgp), use_dma(r->iobase));
-
- return 0;
-}
-
-#define EXPECTED_FLAGS (VM_PFNMAP | VM_DONTEXPAND | VM_DONTDUMP)
-
/**
* remap_io_sg - remap an IO mapping to userspace
* @vma: user vma to map to
@@ -82,12 +46,7 @@ int remap_io_sg(struct vm_area_struct *v
unsigned long addr, unsigned long size,
struct scatterlist *sgl, resource_size_t iobase)
{
- struct remap_pfn r = {
- .mm = vma->vm_mm,
- .prot = vma->vm_page_prot,
- .sgt = __sgt_iter(sgl, use_dma(iobase)),
- .iobase = iobase,
- };
+ unsigned long pfn, len, remapped = 0;
int err;
/* We rely on prevalidation of the io-mapping to skip track_pfn(). */
@@ -96,11 +55,25 @@ int remap_io_sg(struct vm_area_struct *v
if (!use_dma(iobase))
flush_cache_range(vma, addr, size);
- err = apply_to_page_range(r.mm, addr, size, remap_sg, &r);
- if (unlikely(err)) {
- zap_vma_ptes(vma, addr, r.pfn << PAGE_SHIFT);
- return err;
- }
-
- return 0;
+ do {
+ if (use_dma(iobase)) {
+ if (!sg_dma_len(sgl))
+ break;
+ pfn = (sg_dma_address(sgl) + iobase) >> PAGE_SHIFT;
+ len = sg_dma_len(sgl);
+ } else {
+ pfn = page_to_pfn(sg_page(sgl));
+ len = sgl->length;
+ }
+
+ err = remap_pfn_range(vma, addr + remapped, pfn, len,
+ vma->vm_page_prot);
+ if (err)
+ break;
+ remapped += len;
+ } while ((sgl = __sg_next(sgl)));
+
+ if (err)
+ zap_vma_ptes(vma, addr, remapped);
+ return err;
}
_
Patches currently in -mm which might be from hch@lst.de are
mm-add-remap_pfn_range_notrack.patch
mm-add-a-io_mapping_map_user-helper.patch
i915-use-io_mapping_map_user.patch
i915-fix-remap_io_sg-to-verify-the-pgprot.patch
samples-vfio-mdev-mdpy-use-remap_vmalloc_range.patch
mm-unexport-remap_vmalloc_range_partial.patch
^ permalink raw reply [flat|nested] only message in thread
only message in thread, other threads:[~2021-03-28 19:12 UTC | newest]
Thread overview: (only message) (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2021-03-28 19:11 + i915-fix-remap_io_sg-to-verify-the-pgprot.patch added to -mm tree akpm
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.