public inbox for linux-kernel@vger.kernel.org
 help / color / mirror / Atom feed
From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Toshi Kani <toshi.kani@hp.com>,
	Borislav Petkov <bp@alien8.de>, "H. Peter Anvin" <hpa@zytor.com>,
	Andy Lutomirski <luto@kernel.org>,
	Denys Vlasenko <dvlasenk@redhat.com>,
	Brian Gerst <brgerst@gmail.com>,
	x86@kernel.org, mcgrof@suse.com,
	Dan Williams <dan.j.williams@intel.com>,
	Ingo Molnar <mingo@kernel.org>,
	Thomas Gleixner <tglx@linutronix.de>,
	Dave Airlie <airlied@redhat.com>,
	Ben Hutchings <ben.hutchings@codethink.co.uk>
Subject: [PATCH 4.4 46/47] x86/io: add interface to reserve io memtype for a resource range. (v1.1)
Date: Fri,  7 Sep 2018 23:10:42 +0200	[thread overview]
Message-ID: <20180907210909.311384712@linuxfoundation.org> (raw)
In-Reply-To: <20180907210904.607934824@linuxfoundation.org>

4.4-stable review patch.  If anyone has any objections, please let me know.

------------------

From: Dave Airlie <airlied@redhat.com>

commit 8ef4227615e158faa4ee85a1d6466782f7e22f2f upstream.

A recent change to the mm code in:
87744ab3832b mm: fix cache mode tracking in vm_insert_mixed()

started enforcing checking the memory type against the registered list for
amixed pfn insertion mappings. It happens that the drm drivers for a number
of gpus relied on this being broken. Currently the driver only inserted
VRAM mappings into the tracking table when they came from the kernel,
and userspace mappings never landed in the table. This led to a regression
where all the mapping end up as UC instead of WC now.

I've considered a number of solutions but since this needs to be fixed
in fixes and not next, and some of the solutions were going to introduce
overhead that hadn't been there before I didn't consider them viable at
this stage. These mainly concerned hooking into the TTM io reserve APIs,
but these API have a bunch of fast paths I didn't want to unwind to add
this to.

The solution I've decided on is to add a new API like the arch_phys_wc
APIs (these would have worked but wc_del didn't take a range), and
use them from the drivers to add a WC compatible mapping to the table
for all VRAM on those GPUs. This means we can then create userspace
mapping that won't get degraded to UC.

v1.1: use CONFIG_X86_PAT + add some comments in io.h

Cc: Toshi Kani <toshi.kani@hp.com>
Cc: Borislav Petkov <bp@alien8.de>
Cc: H. Peter Anvin <hpa@zytor.com>
Cc: Andy Lutomirski <luto@kernel.org>
Cc: Denys Vlasenko <dvlasenk@redhat.com>
Cc: Brian Gerst <brgerst@gmail.com>
Cc: x86@kernel.org
Cc: mcgrof@suse.com
Cc: Dan Williams <dan.j.williams@intel.com>
Acked-by: Ingo Molnar <mingo@kernel.org>
Reviewed-by: Thomas Gleixner <tglx@linutronix.de>
Signed-off-by: Dave Airlie <airlied@redhat.com>
Cc: Ben Hutchings <ben.hutchings@codethink.co.uk>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/x86/include/asm/io.h |    6 ++++++
 arch/x86/mm/pat.c         |   14 ++++++++++++++
 include/linux/io.h        |   22 ++++++++++++++++++++++
 3 files changed, 42 insertions(+)

--- a/arch/x86/include/asm/io.h
+++ b/arch/x86/include/asm/io.h
@@ -351,4 +351,10 @@ extern void arch_phys_wc_del(int handle)
 #define arch_phys_wc_add arch_phys_wc_add
 #endif
 
+#ifdef CONFIG_X86_PAT
+extern int arch_io_reserve_memtype_wc(resource_size_t start, resource_size_t size);
+extern void arch_io_free_memtype_wc(resource_size_t start, resource_size_t size);
+#define arch_io_reserve_memtype_wc arch_io_reserve_memtype_wc
+#endif
+
 #endif /* _ASM_X86_IO_H */
--- a/arch/x86/mm/pat.c
+++ b/arch/x86/mm/pat.c
@@ -726,6 +726,20 @@ void io_free_memtype(resource_size_t sta
 	free_memtype(start, end);
 }
 
+int arch_io_reserve_memtype_wc(resource_size_t start, resource_size_t size)
+{
+	enum page_cache_mode type = _PAGE_CACHE_MODE_WC;
+
+	return io_reserve_memtype(start, start + size, &type);
+}
+EXPORT_SYMBOL(arch_io_reserve_memtype_wc);
+
+void arch_io_free_memtype_wc(resource_size_t start, resource_size_t size)
+{
+	io_free_memtype(start, start + size);
+}
+EXPORT_SYMBOL(arch_io_free_memtype_wc);
+
 pgprot_t phys_mem_access_prot(struct file *file, unsigned long pfn,
 				unsigned long size, pgprot_t vma_prot)
 {
--- a/include/linux/io.h
+++ b/include/linux/io.h
@@ -154,4 +154,26 @@ enum {
 void *memremap(resource_size_t offset, size_t size, unsigned long flags);
 void memunmap(void *addr);
 
+/*
+ * On x86 PAT systems we have memory tracking that keeps track of
+ * the allowed mappings on memory ranges. This tracking works for
+ * all the in-kernel mapping APIs (ioremap*), but where the user
+ * wishes to map a range from a physical device into user memory
+ * the tracking won't be updated. This API is to be used by
+ * drivers which remap physical device pages into userspace,
+ * and wants to make sure they are mapped WC and not UC.
+ */
+#ifndef arch_io_reserve_memtype_wc
+static inline int arch_io_reserve_memtype_wc(resource_size_t base,
+					     resource_size_t size)
+{
+	return 0;
+}
+
+static inline void arch_io_free_memtype_wc(resource_size_t base,
+					   resource_size_t size)
+{
+}
+#endif
+
 #endif /* _LINUX_IO_H */



  parent reply	other threads:[~2018-09-07 21:42 UTC|newest]

Thread overview: 59+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2018-09-07 21:09 [PATCH 4.4 00/47] 4.4.155-stable review Greg Kroah-Hartman
2018-09-07 21:09 ` [PATCH 4.4 01/47] net: 6lowpan: fix reserved space for single frames Greg Kroah-Hartman
2018-09-07 21:09 ` [PATCH 4.4 02/47] net: mac802154: tx: expand tailroom if necessary Greg Kroah-Hartman
2018-09-07 21:09 ` [PATCH 4.4 03/47] 9p/net: Fix zero-copy path in the 9p virtio transport Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 04/47] net: lan78xx: Fix misplaced tasklet_schedule() call Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 05/47] spi: davinci: fix a NULL pointer dereference Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 06/47] drm/i915/userptr: reject zero user_size Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 07/47] powerpc/fadump: handle crash memory ranges array index overflow Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 08/47] powerpc/pseries: Fix endianness while restoring of r3 in MCE handler Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 09/47] fs/9p/xattr.c: catch the error of p9_client_clunk when setting xattr failed Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 10/47] 9p/virtio: fix off-by-one error in sg list bounds check Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 11/47] net/9p/client.c: version pointer uninitialized Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 12/47] net/9p/trans_fd.c: fix race-condition by flushing workqueue before the kfree() Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 13/47] x86/mm/pat: Fix L1TF stable backport for CPA, 2nd call Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 14/47] dm cache metadata: save in-core policy_hint_size to on-disk superblock Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 15/47] iio: ad9523: Fix displayed phase Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 16/47] iio: ad9523: Fix return value for ad952x_store() Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 17/47] vmw_balloon: fix inflation of 64-bit GFNs Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 18/47] vmw_balloon: do not use 2MB without batching Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 19/47] vmw_balloon: VMCI_DOORBELL_SET does not check status Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 20/47] vmw_balloon: fix VMCI use when balloon built into kernel Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 21/47] tracing: Do not call start/stop() functions when tracing_on does not change Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 22/47] tracing/blktrace: Fix to allow setting same value Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 23/47] kthread, tracing: Dont expose half-written comm when creating kthreads Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 24/47] uprobes: Use synchronize_rcu() not synchronize_sched() Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 25/47] 9p: fix multiple NULL-pointer-dereferences Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 26/47] PM / sleep: wakeup: Fix build error caused by missing SRCU support Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 27/47] pnfs/blocklayout: off by one in bl_map_stripe() Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 28/47] ARM: tegra: Fix Tegra30 Cardhu PCA954x reset Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 29/47] mm/tlb: Remove tlb_remove_table() non-concurrent condition Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 30/47] iommu/vt-d: Add definitions for PFSID Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 31/47] iommu/vt-d: Fix dev iotlb pfsid use Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 32/47] osf_getdomainname(): use copy_to_user() Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 33/47] sys: dont hold uts_sem while accessing userspace memory Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 34/47] userns: move user access out of the mutex Greg Kroah-Hartman
2018-09-09  3:56   ` Rafael David Tinoco
2018-09-09  9:03     ` Greg KH
2018-09-09  9:16     ` Greg KH
2018-09-09  9:29       ` Greg KH
2018-09-07 21:10 ` [PATCH 4.4 35/47] ubifs: Fix memory leak in lprobs self-check Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 36/47] Revert "UBIFS: Fix potential integer overflow in allocation" Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 37/47] ubifs: Check data node size before truncate Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 38/47] ubifs: Fix synced_i_size calculation for xattr inodes Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 39/47] pwm: tiehrpwm: Fix disabling of output of PWMs Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 40/47] fb: fix lost console when the user unplugs a USB adapter Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 41/47] udlfb: set optimal write delay Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 42/47] getxattr: use correct xattr length Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 43/47] bcache: release dc->writeback_lock properly in bch_writeback_thread() Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 44/47] perf auxtrace: Fix queue resize Greg Kroah-Hartman
2018-09-07 21:10 ` [PATCH 4.4 45/47] fs/quota: Fix spectre gadget in do_quotactl Greg Kroah-Hartman
2018-09-07 21:10 ` Greg Kroah-Hartman [this message]
2018-09-07 22:39 ` [PATCH 4.4 00/47] 4.4.155-stable review Nathan Chancellor
2018-09-08 21:13 ` Guenter Roeck
2018-09-09  9:01   ` Greg Kroah-Hartman
2018-09-09 15:52     ` Guenter Roeck
2018-09-09  4:52 ` Naresh Kamboju
2018-09-09  9:17   ` Greg Kroah-Hartman
2018-09-10  1:27     ` Dan Rue
2018-09-10 15:01 ` Shuah Khan

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=20180907210909.311384712@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=airlied@redhat.com \
    --cc=ben.hutchings@codethink.co.uk \
    --cc=bp@alien8.de \
    --cc=brgerst@gmail.com \
    --cc=dan.j.williams@intel.com \
    --cc=dvlasenk@redhat.com \
    --cc=hpa@zytor.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=luto@kernel.org \
    --cc=mcgrof@suse.com \
    --cc=mingo@kernel.org \
    --cc=stable@vger.kernel.org \
    --cc=tglx@linutronix.de \
    --cc=toshi.kani@hp.com \
    --cc=x86@kernel.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