From: Stephen Rothwell <sfr@canb.auug.org.au>
To: Thomas Gleixner <tglx@linutronix.de>, Ingo Molnar <mingo@elte.hu>,
"H. Peter Anvin" <hpa@zytor.com>,
Peter Zijlstra <peterz@infradead.org>
Cc: linux-next@vger.kernel.org, linux-kernel@vger.kernel.org,
Andrew Morton <akpm@linux-foundation.org>,
Gerald Schaefer <gerald.schaefer@de.ibm.com>,
Xiao Guangrong <xiaoguangrong@linux.vnet.ibm.com>
Subject: linux-next: manual merge of the tip tree with Linus' tree
Date: Wed, 10 Oct 2012 13:45:27 +1100 [thread overview]
Message-ID: <20121010134527.a2c21165e0c7b1219d2b847c@canb.auug.org.au> (raw)
[-- Attachment #1: Type: text/plain, Size: 6211 bytes --]
Hi all,
Today's linux-next merge of the tip tree got a conflict in
mm/huge_memory.c between commits d516904bd239 ("thp: merge page pre-alloc
in khugepaged_loop into khugepaged_do_scan"), e3ebcf643811 ("thp: remove
assumptions on pgtable_t type") and 46dcde735c9d ("thp: introduce
pmdp_invalidate()") from Linus' tree and commit 93c9d633bd9e ("mm/thp:
Preserve pgprot across huge page split") from the tip tree.
I fixed it up (I think - see below) and can carry the fix as necessary
(no action is required).
--
Cheers,
Stephen Rothwell sfr@canb.auug.org.au
diff --cc mm/huge_memory.c
index a863af2,5b9ab25..0000000
--- a/mm/huge_memory.c
+++ b/mm/huge_memory.c
@@@ -1346,59 -1428,55 +1417,54 @@@ static int __split_huge_page_map(struc
spin_lock(&mm->page_table_lock);
pmd = page_check_address_pmd(page, mm, address,
PAGE_CHECK_ADDRESS_PMD_SPLITTING_FLAG);
- if (pmd) {
- pgtable = pgtable_trans_huge_withdraw(mm);
- pmd_populate(mm, &_pmd, pgtable);
-
- haddr = address;
- for (i = 0; i < HPAGE_PMD_NR; i++, haddr += PAGE_SIZE) {
- pte_t *pte, entry;
- BUG_ON(PageCompound(page+i));
- entry = mk_pte(page + i, vma->vm_page_prot);
- entry = maybe_mkwrite(pte_mkdirty(entry), vma);
- if (!pmd_write(*pmd))
- entry = pte_wrprotect(entry);
- else
- BUG_ON(page_mapcount(page) != 1);
- if (!pmd_young(*pmd))
- entry = pte_mkold(entry);
- pte = pte_offset_map(&_pmd, haddr);
- BUG_ON(!pte_none(*pte));
- set_pte_at(mm, haddr, pte, entry);
- pte_unmap(pte);
- }
+ if (!pmd)
+ goto unlock;
- smp_wmb(); /* make pte visible before pmd */
- /*
- * Up to this point the pmd is present and huge and
- * userland has the whole access to the hugepage
- * during the split (which happens in place). If we
- * overwrite the pmd with the not-huge version
- * pointing to the pte here (which of course we could
- * if all CPUs were bug free), userland could trigger
- * a small page size TLB miss on the small sized TLB
- * while the hugepage TLB entry is still established
- * in the huge TLB. Some CPU doesn't like that. See
- * http://support.amd.com/us/Processor_TechDocs/41322.pdf,
- * Erratum 383 on page 93. Intel should be safe but is
- * also warns that it's only safe if the permission
- * and cache attributes of the two entries loaded in
- * the two TLB is identical (which should be the case
- * here). But it is generally safer to never allow
- * small and huge TLB entries for the same virtual
- * address to be loaded simultaneously. So instead of
- * doing "pmd_populate(); flush_tlb_range();" we first
- * mark the current pmd notpresent (atomically because
- * here the pmd_trans_huge and pmd_trans_splitting
- * must remain set at all times on the pmd until the
- * split is complete for this pmd), then we flush the
- * SMP TLB and finally we write the non-huge version
- * of the pmd entry with pmd_populate.
- */
- pmdp_invalidate(vma, address, pmd);
- pmd_populate(mm, pmd, pgtable);
- ret = 1;
+ prot = pmd_pgprot(*pmd);
- pgtable = get_pmd_huge_pte(mm);
++ pgtable = pgtable_trans_huge_withdraw(mm);
+ pmd_populate(mm, &_pmd, pgtable);
+
+ for (i = 0, haddr = address; i < HPAGE_PMD_NR; i++, haddr += PAGE_SIZE) {
+ pte_t *pte, entry;
+
+ BUG_ON(PageCompound(page+i));
+ entry = mk_pte(page + i, prot);
+ entry = pte_mkdirty(entry);
+ if (!pmd_young(*pmd))
+ entry = pte_mkold(entry);
+ pte = pte_offset_map(&_pmd, haddr);
+ BUG_ON(!pte_none(*pte));
+ set_pte_at(mm, haddr, pte, entry);
+ pte_unmap(pte);
}
+
+ smp_wmb(); /* make ptes visible before pmd, see __pte_alloc */
+ /*
+ * Up to this point the pmd is present and huge.
+ *
+ * If we overwrite the pmd with the not-huge version, we could trigger
+ * a small page size TLB miss on the small sized TLB while the hugepage
+ * TLB entry is still established in the huge TLB.
+ *
+ * Some CPUs don't like that. See
+ * http://support.amd.com/us/Processor_TechDocs/41322.pdf, Erratum 383
+ * on page 93.
+ *
+ * Thus it is generally safer to never allow small and huge TLB entries
+ * for overlapping virtual addresses to be loaded. So we first mark the
+ * current pmd not present, then we flush the TLB and finally we write
+ * the non-huge version of the pmd entry with pmd_populate.
+ *
+ * The above needs to be done under the ptl because pmd_trans_huge and
+ * pmd_trans_splitting must remain set on the pmd until the split is
+ * complete. The ptl also protects against concurrent faults due to
+ * making the pmd not-present.
+ */
- set_pmd_at(mm, address, pmd, pmd_mknotpresent(*pmd));
- flush_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
++ pmdp_invalidate(vma, address, pmd);
+ pmd_populate(mm, pmd, pgtable);
+ ret = 1;
+
+ unlock:
spin_unlock(&mm->page_table_lock);
return ret;
@@@ -2279,23 -2300,30 +2345,21 @@@ static int khugepaged_has_work(void
static int khugepaged_wait_event(void)
{
return !list_empty(&khugepaged_scan.mm_head) ||
- !khugepaged_enabled();
+ kthread_should_stop();
}
-static void khugepaged_do_scan(struct page **hpage)
+static void khugepaged_do_scan(void)
{
+ struct page *hpage = NULL;
unsigned int progress = 0, pass_through_head = 0;
- unsigned int pages = khugepaged_pages_to_scan;
+ unsigned int pages = ACCESS_ONCE(khugepaged_pages_to_scan);
+ bool wait = true;
- barrier(); /* write khugepaged_pages_to_scan to local stack */
-
while (progress < pages) {
- cond_resched();
-
-#ifndef CONFIG_NUMA
- if (!*hpage) {
- *hpage = alloc_hugepage(khugepaged_defrag());
- if (unlikely(!*hpage)) {
- count_vm_event(THP_COLLAPSE_ALLOC_FAILED);
- break;
- }
- count_vm_event(THP_COLLAPSE_ALLOC);
- }
-#else
- if (IS_ERR(*hpage))
+ if (!khugepaged_prealloc_page(&hpage, &wait))
break;
-#endif
+
+ cond_resched();
if (unlikely(kthread_should_stop() || freezing(current)))
break;
[-- Attachment #2: Type: application/pgp-signature, Size: 836 bytes --]
next reply other threads:[~2012-10-10 2:45 UTC|newest]
Thread overview: 122+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-10-10 2:45 Stephen Rothwell [this message]
-- strict thread matches above, loose matches on Subject: below --
2024-02-08 1:30 linux-next: manual merge of the tip tree with Linus' tree Stephen Rothwell
2023-10-16 1:38 Stephen Rothwell
2023-08-09 2:15 Stephen Rothwell
2023-02-15 1:08 Stephen Rothwell
2023-02-15 11:15 ` Ingo Molnar
2022-07-13 5:08 Stephen Rothwell
2022-05-09 3:59 Stephen Rothwell
2022-01-04 2:45 Stephen Rothwell
2021-10-11 2:21 Stephen Rothwell
2021-10-11 20:48 ` Borislav Petkov
2020-11-06 2:39 Stephen Rothwell
2020-11-06 2:46 ` Steven Rostedt
2020-08-12 1:21 Stephen Rothwell
2020-06-11 1:00 Stephen Rothwell
2020-06-11 0:52 Stephen Rothwell
2020-06-11 10:43 ` Thomas Gleixner
2020-06-11 11:03 ` Stephen Rothwell
2020-06-10 1:49 Stephen Rothwell
2020-06-10 1:42 Stephen Rothwell
2020-05-18 5:10 Stephen Rothwell
2020-05-18 7:12 ` Borislav Petkov
2020-01-06 2:17 Stephen Rothwell
2020-01-06 6:53 ` Ingo Molnar
2019-12-20 1:53 Stephen Rothwell
2019-12-16 2:43 Stephen Rothwell
2019-12-03 2:10 Stephen Rothwell
2019-12-03 6:57 ` Ingo Molnar
2019-12-03 8:56 ` Stephen Rothwell
2019-09-02 7:31 Stephen Rothwell
2019-09-02 18:07 ` Ingo Molnar
2019-07-29 2:00 Stephen Rothwell
2019-03-06 5:53 Stephen Rothwell
2018-10-12 2:14 Stephen Rothwell
2018-10-12 2:18 ` Kees Cook
2018-10-12 10:47 ` Ingo Molnar
2018-10-12 18:56 ` Kees Cook
2018-06-07 1:53 Stephen Rothwell
2018-02-12 1:27 Stephen Rothwell
2018-02-12 7:19 ` Ingo Molnar
2018-02-06 0:54 Stephen Rothwell
2018-02-06 9:14 ` Peter Zijlstra
2018-02-06 0:44 Stephen Rothwell
2018-02-06 0:40 Stephen Rothwell
2018-02-06 12:52 ` Mathieu Desnoyers
2018-02-06 13:55 ` Will Deacon
2018-02-06 14:06 ` Mathieu Desnoyers
2018-02-06 14:11 ` Will Deacon
2018-02-06 17:05 ` Mathieu Desnoyers
2018-02-06 17:21 ` Will Deacon
2018-02-08 7:03 ` Ingo Molnar
2018-02-08 18:56 ` Will Deacon
2018-02-08 19:04 ` Mathieu Desnoyers
2018-02-09 18:25 ` Will Deacon
2018-02-09 20:35 ` Mathieu Desnoyers
2017-11-10 1:27 Stephen Rothwell
2017-05-22 3:27 Stephen Rothwell
2017-05-22 8:32 ` Mark Rutland
2017-05-22 21:44 ` Stephen Rothwell
2017-05-23 8:37 ` Mark Rutland
2016-02-03 0:09 Stephen Rothwell
2016-02-03 0:32 ` Toshi Kani
2015-09-18 1:12 Stephen Rothwell
2015-08-17 5:57 Stephen Rothwell
2015-07-06 0:08 Stephen Rothwell
2015-07-06 7:49 ` Paolo Bonzini
2015-07-06 21:34 ` Stephen Rothwell
2014-05-21 4:12 Stephen Rothwell
2014-05-21 4:24 ` H. Peter Anvin
2014-05-21 6:01 ` Ingo Molnar
2014-05-21 6:02 ` H. Peter Anvin
2014-05-21 6:05 ` Ingo Molnar
2014-05-21 6:10 ` Ingo Molnar
2014-05-21 4:12 Stephen Rothwell
2014-03-21 4:23 Stephen Rothwell
2014-03-21 8:47 ` Srivatsa S. Bhat
2014-02-18 3:09 Stephen Rothwell
2014-02-18 8:06 ` Hans-Christian Egtvedt
2014-02-18 14:15 ` Chen Gang
2014-01-30 1:42 Stephen Rothwell
2014-01-30 1:49 ` Andi Kleen
2014-01-30 1:58 ` Stephen Rothwell
2014-01-30 2:05 ` Andi Kleen
2014-01-28 0:43 Stephen Rothwell
2014-01-17 3:30 Stephen Rothwell
2014-01-17 17:27 ` Sören Brinkmann
2013-12-16 3:18 Stephen Rothwell
2013-09-04 4:24 Stephen Rothwell
2013-07-09 4:36 Stephen Rothwell
2012-12-03 4:19 Stephen Rothwell
2012-10-16 1:00 Stephen Rothwell
2012-10-21 16:29 ` Ingo Molnar
2012-10-15 0:32 Stephen Rothwell
2012-10-10 2:51 Stephen Rothwell
2012-06-26 4:36 Stephen Rothwell
2012-06-22 5:01 Stephen Rothwell
2012-06-04 3:52 Stephen Rothwell
2012-05-22 5:44 Stephen Rothwell
2012-01-03 5:16 Stephen Rothwell
2012-01-03 5:08 Stephen Rothwell
2012-01-03 8:06 ` Ingo Molnar
2012-01-03 8:36 ` Stephen Rothwell
2011-04-14 3:14 Stephen Rothwell
2011-04-14 9:02 ` Peter Zijlstra
2011-04-14 9:25 ` Ingo Molnar
2011-04-08 5:02 Stephen Rothwell
2011-03-16 5:08 Stephen Rothwell
2011-03-16 8:50 ` Sebastian Andrzej Siewior
2011-03-16 11:02 ` Stephen Rothwell
2010-10-22 2:08 Stephen Rothwell
2010-10-06 2:47 Stephen Rothwell
2010-09-23 3:32 Stephen Rothwell
2010-07-26 3:39 Stephen Rothwell
2010-01-18 7:08 Stephen Rothwell
2010-01-18 7:50 ` Ingo Molnar
2010-01-18 8:59 ` Stephen Rothwell
2009-12-09 4:06 Stephen Rothwell
2009-09-24 4:07 Stephen Rothwell
2009-09-24 8:33 ` Ingo Molnar
2009-09-24 9:23 ` Stephen Rothwell
2009-08-19 6:07 Stephen Rothwell
2009-08-19 10:37 ` Ingo Molnar
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=20121010134527.a2c21165e0c7b1219d2b847c@canb.auug.org.au \
--to=sfr@canb.auug.org.au \
--cc=akpm@linux-foundation.org \
--cc=gerald.schaefer@de.ibm.com \
--cc=hpa@zytor.com \
--cc=linux-kernel@vger.kernel.org \
--cc=linux-next@vger.kernel.org \
--cc=mingo@elte.hu \
--cc=peterz@infradead.org \
--cc=tglx@linutronix.de \
--cc=xiaoguangrong@linux.vnet.ibm.com \
/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).