From: Jan Beulich <jbeulich@suse.com>
To: Juergen Gross <jgross@suse.com>,
Boris Ostrovsky <boris.ostrovsky@oracle.com>
Cc: Stefano Stabellini <sstabellini@kernel.org>,
lkml <linux-kernel@vger.kernel.org>,
"xen-devel@lists.xenproject.org" <xen-devel@lists.xenproject.org>
Subject: [PATCH 03/12] swiotlb-xen: maintain slab count properly
Date: Tue, 7 Sep 2021 14:05:12 +0200 [thread overview]
Message-ID: <dc054cb0-bec4-4db0-fc06-c9fc957b6e66@suse.com> (raw)
In-Reply-To: <588b3e6d-2682-160c-468e-44ca4867a570@suse.com>
Generic swiotlb code makes sure to keep the slab count a multiple of the
number of slabs per segment. Yet even without checking whether any such
assumption is made elsewhere, it is easy to see that xen_swiotlb_fixup()
might alter unrelated memory when calling xen_create_contiguous_region()
for the last segment, when that's not a full one - the function acts on
full order-N regions, not individual pages.
Align the slab count suitably when halving it for a retry. Add a build
time check and a runtime one. Replace the no longer useful local
variable "slabs" by an "order" one calculated just once, outside of the
loop. Re-use "order" for calculating "dma_bits", and change the type of
the latter as well as the one of "i" while touching this anyway.
Signed-off-by: Jan Beulich <jbeulich@suse.com>
--- a/drivers/xen/swiotlb-xen.c
+++ b/drivers/xen/swiotlb-xen.c
@@ -106,27 +106,26 @@ static int is_xen_swiotlb_buffer(struct
static int xen_swiotlb_fixup(void *buf, unsigned long nslabs)
{
- int i, rc;
- int dma_bits;
+ int rc;
+ unsigned int order = get_order(IO_TLB_SEGSIZE << IO_TLB_SHIFT);
+ unsigned int i, dma_bits = order + PAGE_SHIFT;
dma_addr_t dma_handle;
phys_addr_t p = virt_to_phys(buf);
- dma_bits = get_order(IO_TLB_SEGSIZE << IO_TLB_SHIFT) + PAGE_SHIFT;
+ BUILD_BUG_ON(IO_TLB_SEGSIZE & (IO_TLB_SEGSIZE - 1));
+ BUG_ON(nslabs % IO_TLB_SEGSIZE);
i = 0;
do {
- int slabs = min(nslabs - i, (unsigned long)IO_TLB_SEGSIZE);
-
do {
rc = xen_create_contiguous_region(
- p + (i << IO_TLB_SHIFT),
- get_order(slabs << IO_TLB_SHIFT),
+ p + (i << IO_TLB_SHIFT), order,
dma_bits, &dma_handle);
} while (rc && dma_bits++ < MAX_DMA_BITS);
if (rc)
return rc;
- i += slabs;
+ i += IO_TLB_SEGSIZE;
} while (i < nslabs);
return 0;
}
@@ -210,7 +209,7 @@ retry:
error:
if (repeat--) {
/* Min is 2MB */
- nslabs = max(1024UL, (nslabs >> 1));
+ nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));
bytes = nslabs << IO_TLB_SHIFT;
pr_info("Lowering to %luMB\n", bytes >> 20);
goto retry;
@@ -245,7 +244,7 @@ retry:
memblock_free(__pa(start), PAGE_ALIGN(bytes));
if (repeat--) {
/* Min is 2MB */
- nslabs = max(1024UL, (nslabs >> 1));
+ nslabs = max(1024UL, ALIGN(nslabs >> 1, IO_TLB_SEGSIZE));
bytes = nslabs << IO_TLB_SHIFT;
pr_info("Lowering to %luMB\n", bytes >> 20);
goto retry;
next prev parent reply other threads:[~2021-09-07 12:05 UTC|newest]
Thread overview: 45+ messages / expand[flat|nested] mbox.gz Atom feed top
2021-09-07 12:03 [PATCH 00/12] swiotlb-xen: fixes and adjustments Jan Beulich
2021-09-07 12:04 ` [PATCH 01/12] swiotlb-xen: avoid double free Jan Beulich
2021-09-08 6:48 ` Christoph Hellwig
2021-09-10 22:55 ` Stefano Stabellini
2021-09-07 12:04 ` [PATCH 02/12] swiotlb-xen: fix late init retry Jan Beulich
2021-09-08 6:50 ` Christoph Hellwig
2021-09-07 12:05 ` Jan Beulich [this message]
2021-09-08 6:53 ` [PATCH 03/12] swiotlb-xen: maintain slab count properly Christoph Hellwig
2021-09-10 23:10 ` Stefano Stabellini
2021-09-07 12:05 ` [PATCH 04/12] swiotlb-xen: ensure to issue well-formed XENMEM_exchange requests Jan Beulich
2021-09-08 6:57 ` Christoph Hellwig
2021-09-08 9:16 ` Jan Beulich
2021-09-10 23:14 ` Stefano Stabellini
2021-09-13 7:17 ` Jan Beulich
2021-09-13 20:31 ` Stefano Stabellini
2021-09-14 9:11 ` Jan Beulich
2021-09-15 1:22 ` Stefano Stabellini
2021-09-15 1:54 ` Stefano Stabellini
2021-09-15 8:21 ` Jan Beulich
2021-09-07 12:05 ` [PATCH 05/12] swiotlb-xen: suppress certain init retries Jan Beulich
2021-09-08 6:58 ` Christoph Hellwig
2021-09-10 23:18 ` Stefano Stabellini
2021-09-07 12:06 ` [PATCH 06/12] swiotlb-xen: limit " Jan Beulich
2021-09-08 7:00 ` Christoph Hellwig
2021-09-10 23:23 ` Stefano Stabellini
2021-09-07 12:06 ` [PATCH 07/12] swiotlb-xen: drop leftover __ref Jan Beulich
2021-09-08 7:02 ` Christoph Hellwig
2021-09-10 23:24 ` Stefano Stabellini
2021-09-07 12:07 ` [PATCH 08/12] swiotlb-xen: arrange to have buffer info logged Jan Beulich
2021-09-08 7:04 ` Christoph Hellwig
2021-09-10 23:26 ` Stefano Stabellini
2021-09-07 12:07 ` [PATCH 09/12] swiotlb-xen: drop DEFAULT_NSLABS Jan Beulich
2021-09-08 7:06 ` Christoph Hellwig
2021-09-10 23:27 ` Stefano Stabellini
2021-09-07 12:11 ` [PATCH 11/12] xen/pci-swiotlb: reduce visibility of symbols Jan Beulich
2021-09-08 7:08 ` Christoph Hellwig
2021-09-07 12:13 ` [PATCH 12/12] swiotlb-xen: this is PV-only on x86 Jan Beulich
2021-09-08 7:14 ` Christoph Hellwig
2021-09-10 23:48 ` Stefano Stabellini
2021-09-13 7:28 ` Jan Beulich
2021-09-13 20:54 ` Stefano Stabellini
2021-09-13 7:51 ` Jan Beulich
2021-09-13 20:49 ` Stefano Stabellini
2021-09-08 15:18 ` [PATCH 00/12] swiotlb-xen: fixes and adjustments Juergen Gross
2021-09-14 8:30 ` Juergen Gross
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=dc054cb0-bec4-4db0-fc06-c9fc957b6e66@suse.com \
--to=jbeulich@suse.com \
--cc=boris.ostrovsky@oracle.com \
--cc=jgross@suse.com \
--cc=linux-kernel@vger.kernel.org \
--cc=sstabellini@kernel.org \
--cc=xen-devel@lists.xenproject.org \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
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).