linux-kernel.vger.kernel.org archive mirror
 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, Paul Burton <paulburton@kernel.org>,
	Dmitry Korotin <dkorotin@wavecomp.com>,
	linux-mips@vger.kernel.org
Subject: [PATCH 4.9 30/49] MIPS: tlbex: Fix build_restore_pagemask KScratch restore
Date: Sun, 27 Oct 2019 22:01:08 +0100	[thread overview]
Message-ID: <20191027203143.757870099@linuxfoundation.org> (raw)
In-Reply-To: <20191027203119.468466356@linuxfoundation.org>

From: Paul Burton <paulburton@kernel.org>

commit b42aa3fd5957e4daf4b69129e5ce752a2a53e7d6 upstream.

build_restore_pagemask() will restore the value of register $1/$at when
its restore_scratch argument is non-zero, and aims to do so by filling a
branch delay slot. Commit 0b24cae4d535 ("MIPS: Add missing EHB in mtc0
-> mfc0 sequence.") added an EHB instruction (Execution Hazard Barrier)
prior to restoring $1 from a KScratch register, in order to resolve a
hazard that can result in stale values of the KScratch register being
observed. In particular, P-class CPUs from MIPS with out of order
execution pipelines such as the P5600 & P6600 are affected.

Unfortunately this EHB instruction was inserted in the branch delay slot
causing the MFC0 instruction which performs the restoration to no longer
execute along with the branch. The result is that the $1 register isn't
actually restored, ie. the TLB refill exception handler clobbers it -
which is exactly the problem the EHB is meant to avoid for the P-class
CPUs.

Similarly build_get_pgd_vmalloc() will restore the value of $1/$at when
its mode argument equals refill_scratch, and suffers from the same
problem.

Fix this by in both cases moving the EHB earlier in the emitted code.
There's no reason it needs to immediately precede the MFC0 - it simply
needs to be between the MTC0 & MFC0.

This bug only affects Cavium Octeon systems which use
build_fast_tlb_refill_handler().

Signed-off-by: Paul Burton <paulburton@kernel.org>
Fixes: 0b24cae4d535 ("MIPS: Add missing EHB in mtc0 -> mfc0 sequence.")
Cc: Dmitry Korotin <dkorotin@wavecomp.com>
Cc: stable@vger.kernel.org # v3.15+
Cc: linux-mips@vger.kernel.org
Cc: linux-kernel@vger.kernel.org
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

---
 arch/mips/mm/tlbex.c |   23 +++++++++++++++--------
 1 file changed, 15 insertions(+), 8 deletions(-)

--- a/arch/mips/mm/tlbex.c
+++ b/arch/mips/mm/tlbex.c
@@ -661,6 +661,13 @@ static void build_restore_pagemask(u32 *
 				   int restore_scratch)
 {
 	if (restore_scratch) {
+		/*
+		 * Ensure the MFC0 below observes the value written to the
+		 * KScratch register by the prior MTC0.
+		 */
+		if (scratch_reg >= 0)
+			uasm_i_ehb(p);
+
 		/* Reset default page size */
 		if (PM_DEFAULT_MASK >> 16) {
 			uasm_i_lui(p, tmp, PM_DEFAULT_MASK >> 16);
@@ -675,12 +682,10 @@ static void build_restore_pagemask(u32 *
 			uasm_i_mtc0(p, 0, C0_PAGEMASK);
 			uasm_il_b(p, r, lid);
 		}
-		if (scratch_reg >= 0) {
-			uasm_i_ehb(p);
+		if (scratch_reg >= 0)
 			UASM_i_MFC0(p, 1, c0_kscratch(), scratch_reg);
-		} else {
+		else
 			UASM_i_LW(p, 1, scratchpad_offset(0), 0);
-		}
 	} else {
 		/* Reset default page size */
 		if (PM_DEFAULT_MASK >> 16) {
@@ -922,6 +927,10 @@ build_get_pgd_vmalloc64(u32 **p, struct
 	}
 	if (mode != not_refill && check_for_high_segbits) {
 		uasm_l_large_segbits_fault(l, *p);
+
+		if (mode == refill_scratch && scratch_reg >= 0)
+			uasm_i_ehb(p);
+
 		/*
 		 * We get here if we are an xsseg address, or if we are
 		 * an xuseg address above (PGDIR_SHIFT+PGDIR_BITS) boundary.
@@ -938,12 +947,10 @@ build_get_pgd_vmalloc64(u32 **p, struct
 		uasm_i_jr(p, ptr);
 
 		if (mode == refill_scratch) {
-			if (scratch_reg >= 0) {
-				uasm_i_ehb(p);
+			if (scratch_reg >= 0)
 				UASM_i_MFC0(p, 1, c0_kscratch(), scratch_reg);
-			} else {
+			else
 				UASM_i_LW(p, 1, scratchpad_offset(0), 0);
-			}
 		} else {
 			uasm_i_nop(p);
 		}



  parent reply	other threads:[~2019-10-27 21:37 UTC|newest]

Thread overview: 55+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2019-10-27 21:00 [PATCH 4.9 00/49] 4.9.198-stable review Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 01/49] scsi: ufs: skip shutdown if hba is not powered Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 02/49] scsi: megaraid: disable device when probe failed after enabled device Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 03/49] scsi: qla2xxx: Fix unbound sleep in fcport delete path Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 04/49] ARM: OMAP2+: Fix missing reset done flag for am3 and am43 Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 05/49] ARM: dts: am4372: Set memory bandwidth limit for DISPC Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 06/49] MIPS: dts: ar9331: fix interrupt-controller size Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 07/49] nl80211: fix null pointer dereference Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 08/49] mac80211: fix txq " Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 09/49] mips: Loongson: Fix the link time qualifier of serial_exit() Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 10/49] net: hisilicon: Fix usage of uninitialized variable in function mdio_sc_cfg_reg_write() Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 11/49] namespace: fix namespace.pl script to support relative paths Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 12/49] Revert "drm/radeon: Fix EEH during kexec" Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 13/49] ocfs2: fix panic due to ocfs2_wq is null Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 14/49] MIPS: Treat Loongson Extensions as ASEs Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 15/49] MIPS: elf_hwcap: Export userspace ASEs Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 16/49] loop: Add LOOP_SET_DIRECT_IO to compat ioctl Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 17/49] net: bcmgenet: Fix RGMII_MODE_EN value for GENET v1/2/3 Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 18/49] net: bcmgenet: Set phydev->dev_flags only for internal PHYs Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 19/49] sctp: change sctp_prot .no_autobind with true Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 20/49] net: avoid potential infinite loop in tc_ctl_action() Greg Kroah-Hartman
2019-10-27 21:00 ` [PATCH 4.9 21/49] ipv4: Return -ENETUNREACH if we cant create route but saddr is valid Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 22/49] memfd: Fix locking when tagging pins Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 23/49] USB: legousbtower: fix memleak on disconnect Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 24/49] ALSA: hda/realtek - Add support for ALC711 Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 25/49] usb: udc: lpc32xx: fix bad bit shift operation Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 26/49] USB: serial: ti_usb_3410_5052: fix port-close races Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 27/49] USB: ldusb: fix memleak on disconnect Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 28/49] USB: usblp: fix use-after-free " Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 29/49] USB: ldusb: fix read info leaks Greg Kroah-Hartman
2019-10-27 21:01 ` Greg Kroah-Hartman [this message]
2019-10-27 21:01 ` [PATCH 4.9 31/49] staging: wlan-ng: fix exit return when sme->key_idx >= NUM_WEPKEYS Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 32/49] scsi: core: try to get module before removing device Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 33/49] Input: da9063 - fix capability and drop KEY_SLEEP Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 34/49] ASoC: rsnd: Reinitialize bit clock inversion flag for every format setting Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 35/49] cfg80211: wext: avoid copying malformed SSIDs Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 36/49] mac80211: Reject malformed SSID elements Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 37/49] drm/edid: Add 6 bpc quirk for SDC panel in Lenovo G50 Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 38/49] scsi: zfcp: fix reaction on bit error threshold notification Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 39/49] mm/slub: fix a deadlock in show_slab_objects() Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 40/49] xtensa: drop EXPORT_SYMBOL for outs*/ins* Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 41/49] parisc: Fix vmap memory leak in ioremap()/iounmap() Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 42/49] CIFS: avoid using MID 0xFFFF Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 43/49] btrfs: block-group: Fix a memory leak due to missing btrfs_put_block_group() Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 44/49] memstick: jmb38x_ms: Fix an error handling path in jmb38x_ms_probe() Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 45/49] cpufreq: Avoid cpufreq_suspend() deadlock on system shutdown Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 46/49] xen/netback: fix error path of xenvif_connect_data() Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 47/49] PCI: PM: Fix pci_power_up() Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 48/49] Revert "net: sit: fix memory leak in sit_init_net()" Greg Kroah-Hartman
2019-10-27 21:01 ` [PATCH 4.9 49/49] RDMA/cxgb4: Do not dma memory off of the stack Greg Kroah-Hartman
2019-10-28  5:35 ` [PATCH 4.9 00/49] 4.9.198-stable review kernelci.org bot
2019-10-28  9:05 ` Didik Setiawan
2019-10-28 13:34 ` Guenter Roeck
2019-10-28 21:37 ` Jon Hunter
2019-10-29 15:17 ` Naresh Kamboju

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=20191027203143.757870099@linuxfoundation.org \
    --to=gregkh@linuxfoundation.org \
    --cc=dkorotin@wavecomp.com \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mips@vger.kernel.org \
    --cc=paulburton@kernel.org \
    --cc=stable@vger.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;
as well as URLs for NNTP newsgroup(s).