Skip to contentSkip to conversation

5.10.47-rc1 review

PATCH In review Sasha Levin 11 participants 117 messages Jun 28, 2021 – Jul 5, 2021

Subsystems (3)
lore
Share & tools
lore

Sasha Levin

author maintainer
#0 lore raw

cover 5.10.47-rc1 review

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed 30 Jun 2021 02:25:36 PM UTC. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

        https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/patch/?id=linux-5.10.y&id2=v5.10.46

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.

Thanks,
Sasha

-------------
Pseudo-Shortlog of commits:

Alex Shi (1):
mm: add VM_WARN_ON_ONCE_PAGE() macro

Alper Gun (1):
KVM: SVM: Call SEV Guest Decommission if ASID binding fails

Arnd Bergmann (1):
ARM: 9081/1: fix gcc-10 thumb2-kernel regression

Austin Kim (1):
net: ethtool: clear heap allocations for ethtool function

Bumyong Lee (1):
swiotlb: manipulate orig_addr when tlb_addr has offset

Catalin Marinas (2):
arm64: Ignore any DMA offsets in the max_zone_phys() calculation
arm64: Force NO_BLOCK_MAPPINGS if crashkernel reservation is required

Christian König (3):
drm/nouveau: wait for moving fence after pinning v2 drm/radeon: wait for moving fence after pinning drm/amdgpu: wait for moving fence after pinning

Christoph Hellwig (1):
scsi: sd: Call sd_revalidate_disk() for ioctl(BLKRRPART)

Daniel Vetter (1):
Revert "drm: add a locked version of drm_is_current_master"

Desmond Cheong Zhi Xi (1):
drm: add a locked version of drm_is_current_master

Du Cheng (1):
cfg80211: call cfg80211_leave_ocb when switching away from OCB

Eric Dumazet (6):
inet: annotate data race in inet_send_prepare() and
inet_dgram_connect()
net: annotate data race in sock_error()
inet: annotate date races around sk->sk_txhash net/packet: annotate data race in packet_sendmsg() net/packet: annotate accesses to po->bind net/packet: annotate accesses to po->ifindex

Eric Snowberg (4):
certs: Add EFI_CERT_X509_GUID support for dbx entries
certs: Move load_system_certificate_list to a common function
certs: Add ability to preload revocation certs
integrity: Load mokx variables into the blacklist keyring

Esben Haabendal (2):
net: ll_temac: Add memory-barriers for TX BD access
net: ll_temac: Avoid ndo_start_xmit returning NETDEV_TX_BUSY

Fabien Dessenne (1):
pinctrl: stm32: fix the reported number of GPIO lines per bank

Fuad Tabba (1):
KVM: selftests: Fix kvm_check_cap() assertion

Gabriel Knezek (1):
gpiolib: cdev: zero padding during conversion to gpioline_info_changed

Guillaume Ranquet (3):
dmaengine: mediatek: free the proper desc in desc_free handler
dmaengine: mediatek: do not issue a new desc if one is still current
dmaengine: mediatek: use GFP_NOWAIT instead of GFP_ATOMIC in prep_dma

Haibo Chen (1):
spi: spi-nxp-fspi: move the register operation after the clock enable

Heiko Carstens (1):
s390/stack: fix possible register corruption with stack switch helper

Heiner Kallweit (1):
i2c: i801: Ensure that SMBHSTSTS_INUSE_STS is cleared when leaving

    i801_access

Hugh Dickins (16):
mm/thp: fix __split_huge_pmd_locked() on shmem migration entry mm/thp: make is_huge_zero_pmd() safe and quicker mm/thp: try_to_unmap() use TTU_SYNC for safe splitting mm/thp: fix vma_address() if virtual address below file offset mm/thp: unmap_mapping_page() to fix THP truncate_cleanup_page()
mm: page_vma_mapped_walk(): use page for pvmw->page
mm: page_vma_mapped_walk(): settle PageHuge on entry
mm: page_vma_mapped_walk(): use pmde for *pvmw->pmd
mm: page_vma_mapped_walk(): prettify PVMW_MIGRATION block
mm: page_vma_mapped_walk(): crossing page table boundary
mm: page_vma_mapped_walk(): add a level of indentation
mm: page_vma_mapped_walk(): use goto instead of while (1)
mm: page_vma_mapped_walk(): get vma_address_end() earlier mm/thp: fix page_vma_mapped_walk() if THP mapped by ptes mm/thp: another PVMW_SYNC fix in page_vma_mapped_walk() mm, futex: fix shared futex pgoff on shmem huge page

Jeff Layton (2):
ceph: must hold snap_rwsem when filling inode for async create
netfs: fix test for whether we can skip read when writing beyond EOF

Johan Hovold (1):
i2c: robotfuzz-osif: fix control-request directions

Johannes Berg (5):
mac80211: remove warning in ieee80211_get_sband()
mac80211_hwsim: drop pending frames on stop
mac80211: drop multicast fragments
mac80211: reset profile_periodicity/ema_ap
mac80211: handle various extensible elements correctly

Jue Wang (1):
mm/thp: fix page_address_in_vma() on file THP tails

Kan Liang (1):
perf/x86: Track pmu in per-CPU cpu_hw_events

Kees Cook (4):
r8152: Avoid memcpy() over-reading of ETH_SS_STATS
sh_eth: Avoid memcpy() over-reading of ETH_SS_STATS
r8169: Avoid memcpy() over-reading of ETH_SS_STATS
net: qed: Fix memcpy() overflow of qed_dcbx_params()

Khem Raj (1):
riscv32: Use medany C model for modules

Laurent Pinchart (2):
dmaengine: xilinx: dpdma: Add missing dependencies to Kconfig
dmaengine: xilinx: dpdma: Limit descriptor IDs to 16 bits

Like Xu (1):
perf/x86/lbr: Remove cpuc->lbr_xsave allocation from atomic context

Maxime Ripard (2):
drm/vc4: hdmi: Move the HSM clock enable to runtime_pm drm/vc4: hdmi: Make sure the controller is powered in detect

Miaohe Lin (2):
mm/rmap: remove unneeded semicolon in page_not_mapped() mm/rmap: use page_not_mapped in try_to_unmap()

Mikel Rychliski (1):
PCI: Add AMD RS690 quirk to enable 64-bit DMA

Mimi Zohar (1):
module: limit enabling module.sig_enforce

Nathan Chancellor (1):
MIPS: generic: Update node names to avoid unit addresses

Neil Armstrong (1):
mmc: meson-gx: use memcpy_to/fromio for dram-access-quirk

Nicholas Piggin (1):
KVM: do not allow mapping valid but non-reference-counted pages

Pavel Skripkin (2):
net: caif: fix memory leak in ldisc_open
nilfs2: fix memory leak in nilfs_sysfs_delete_device_group

Peter Zijlstra (4):
x86/entry: Fix noinstr fail in __do_fast_syscall_32() x86/xen: Fix noinstr fail in exc_xen_unknown_trap() locking/lockdep: Improve noinstr vs errors
recordmcount: Correct st_shndx handling

Petr Mladek (2):
kthread_worker: split code for canceling the delayed work timer
kthread: prevent deadlock when kthread_mod_delayed_work() races with
kthread_cancel_delayed_work_sync()

Praneeth Bajjuri (1):
net: phy: dp83867: perform soft reset and retain established link

Rafael J. Wysocki (1):
Revert "PCI: PM: Do not read power state in pci_enable_device_flags()"

Sasha Levin (1):
Linux 5.10.47-rc1

Thomas Gleixner (3):
perf/x86/intel/lbr: Zero the xstate buffer on allocation x86/fpu: Preserve supervisor states in sanitize_restored_user_xstate() x86/fpu: Make init_fpstate correct with optimized XSAVE

Xu Yu (1):
mm, thp: use head page in __migration_entry_wait()

Yang Shi (1):
mm: thp: replace DEBUG_VM BUG with VM_WARN when unmap fails for split

Yifan Zhang (2):
Revert "drm/amdgpu/gfx9: fix the doorbell missing when in CGPG issue." Revert "drm/amdgpu/gfx10: enlarge CP_MEC_DOORBELL_RANGE_UPPER to cover
full doorbell."

Yu Kuai (2):
dmaengine: zynqmp_dma: Fix PM reference leak in
zynqmp_dma_alloc_chan_resourc()
dmaengine: stm32-mdma: fix PM reference leak in

    stm32_mdma_alloc_chan_resourc()

Zheng Yongjun (2):
net: ipv4: Remove unneed BUG() function
ping: Check return value of function 'ping_queue_rcv_skb'

Zou Wei (1):
dmaengine: rcar-dmac: Fix PM reference leak in rcar_dmac_probe()

 Makefile                                      |   4 +-
 arch/arm/kernel/setup.c                       |  16 +-
 arch/arm64/mm/init.c                          |  17 +-
 arch/arm64/mm/mmu.c                           |  37 ++--
 arch/mips/generic/board-boston.its.S          |  10 +-
 arch/mips/generic/board-ni169445.its.S        |  10 +-
 arch/mips/generic/board-ocelot.its.S          |  20 +--
 arch/mips/generic/board-xilfpga.its.S         |  10 +-
 arch/mips/generic/vmlinux.its.S               |  10 +-
 arch/riscv/Makefile                           |   2 +-
 arch/s390/include/asm/stacktrace.h            |  18 +-
 arch/x86/entry/common.c                       |   2 +-
 arch/x86/events/core.c                        |  23 ++-
 arch/x86/events/intel/core.c                  |   2 +-
 arch/x86/events/intel/ds.c                    |   4 +-
 arch/x86/events/intel/lbr.c                   |  36 ++--
 arch/x86/events/perf_event.h                  |  10 +-
 arch/x86/include/asm/fpu/internal.h           |  30 +---
 arch/x86/kernel/fpu/signal.c                  |  26 +--
 arch/x86/kernel/fpu/xstate.c                  |  41 ++++-
 arch/x86/kvm/svm/sev.c                        |  32 ++--
 arch/x86/pci/fixup.c                          |  44 +++++
 arch/x86/xen/enlighten_pv.c                   |   2 +
 certs/Kconfig                                 |  17 ++
 certs/Makefile                                |  21 ++-
 certs/blacklist.c                             |  64 +++++++
 certs/blacklist.h                             |   2 +
 certs/common.c                                |  57 +++++++
 certs/common.h                                |   9 +
 certs/revocation_certificates.S               |  21 +++
 certs/system_keyring.c                        |  55 +-----
 drivers/dma/Kconfig                           |   1 +
 drivers/dma/mediatek/mtk-uart-apdma.c         |  27 +--
 drivers/dma/sh/rcar-dmac.c                    |   2 +-
 drivers/dma/stm32-mdma.c                      |   4 +-
 drivers/dma/xilinx/xilinx_dpdma.c             |   7 +-
 drivers/dma/xilinx/zynqmp_dma.c               |   2 +-
 drivers/gpio/gpiolib-cdev.c                   |   1 +
 drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c   |  14 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c        |   6 +-
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c         |   6 +-
 drivers/gpu/drm/nouveau/nouveau_prime.c       |  17 +-
 drivers/gpu/drm/radeon/radeon_prime.c         |  16 +-
 drivers/gpu/drm/vc4/vc4_hdmi.c                |  44 +++--
 drivers/i2c/busses/i2c-i801.c                 |   3 +
 drivers/i2c/busses/i2c-robotfuzz-osif.c       |   4 +-
 drivers/mmc/host/meson-gx-mmc.c               |  50 +++++-
 drivers/net/caif/caif_serial.c                |   1 +
 drivers/net/ethernet/qlogic/qed/qed_dcbx.c    |   4 +-
 drivers/net/ethernet/realtek/r8169_main.c     |   2 +-
 drivers/net/ethernet/renesas/sh_eth.c         |   2 +-
 drivers/net/ethernet/xilinx/ll_temac_main.c   |  19 ++-
 drivers/net/phy/dp83867.c                     |   6 +-
 drivers/net/usb/r8152.c                       |   2 +-
 drivers/net/wireless/mac80211_hwsim.c         |   5 +
 drivers/pci/pci.c                             |  16 +-
 drivers/pinctrl/stm32/pinctrl-stm32.c         |   9 +-
 drivers/scsi/sd.c                             |  22 ++-
 drivers/spi/spi-nxp-fspi.c                    |  11 +-
 fs/ceph/addr.c                                |  54 ++++--
 fs/ceph/file.c                                |   3 +
 fs/ceph/inode.c                               |   2 +
 fs/nilfs2/sysfs.c                             |   1 +
 include/keys/system_keyring.h                 |  15 ++
 include/linux/debug_locks.h                   |   2 +
 include/linux/huge_mm.h                       |   8 +-
 include/linux/hugetlb.h                       |  16 --
 include/linux/mm.h                            |   3 +
 include/linux/mmdebug.h                       |  13 ++
 include/linux/pagemap.h                       |  13 +-
 include/linux/rmap.h                          |   1 +
 include/net/sock.h                            |  17 +-
 kernel/dma/swiotlb.c                          |   3 +
 kernel/futex.c                                |   3 +-
 kernel/kthread.c                              |  77 ++++++---
 kernel/locking/lockdep.c                      |   4 +-
 kernel/module.c                               |  14 +-
 lib/debug_locks.c                             |   2 +-
 mm/huge_memory.c                              |  56 +++---
 mm/hugetlb.c                                  |   5 +-
 mm/internal.h                                 |  53 ++++--
 mm/memory.c                                   |  41 +++++
 mm/migrate.c                                  |   1 +
 mm/page_vma_mapped.c                          | 160 +++++++++++-------
 mm/pgtable-generic.c                          |   4 +-
 mm/rmap.c                                     |  50 +++---
 mm/truncate.c                                 |  43 +++--
 net/ethtool/ioctl.c                           |  10 +-
 net/ipv4/af_inet.c                            |   4 +-
 net/ipv4/devinet.c                            |   2 +-
 net/ipv4/ping.c                               |  12 +-
 net/ipv6/addrconf.c                           |   2 +-
 net/mac80211/ieee80211_i.h                    |   2 +-
 net/mac80211/mlme.c                           |   8 +
 net/mac80211/rx.c                             |   9 +-
 net/mac80211/util.c                           |  22 +--
 net/packet/af_packet.c                        |  41 +++--
 net/wireless/util.c                           |   3 +
 scripts/Makefile                              |   1 +
 scripts/recordmcount.h                        |  15 +-
 .../platform_certs/keyring_handler.c          |  11 ++
 security/integrity/platform_certs/load_uefi.c |  20 ++-
 tools/testing/selftests/kvm/lib/kvm_util.c    |   2 +-
 virt/kvm/kvm_main.c                           |  19 ++-
 104 files changed, 1242 insertions(+), 560 deletions(-)
 create mode 100644 certs/common.c
 create mode 100644 certs/common.h
 create mode 100644 certs/revocation_certificates.S
Signature · 1 line
2.30.2
author maintainer
#1 lore raw

PATCH 1/101 module: limit enabling module.sig_enforce

From: Mimi Zohar <[email protected]>

[ Upstream commit 0c18f29aae7ce3dadd26d8ee3505d07cc982df75 ]

Irrespective as to whether CONFIG_MODULE_SIG is configured, specifying "module.sig_enforce=1" on the boot command line sets "sig_enforce". Only allow "sig_enforce" to be set when CONFIG_MODULE_SIG is configured.

This patch makes the presence of /sys/module/module/parameters/sig_enforce dependent on CONFIG_MODULE_SIG=y.

Fixes fda784e50aac ("module: export module signature enforcement status")
---
 kernel/module.c | 14 +++++++++-----
 1 file changed, 9 insertions(+), 5 deletions(-)
kernel/module.c +9 -5
index 908d46abe165..185b2655bc20 100644
@@ -272,9 +272,18 @@ static void module_assert_mutex_or_preempt(void)
 #endif
 }
 
+#ifdef CONFIG_MODULE_SIG
 static bool sig_enforce = IS_ENABLED(CONFIG_MODULE_SIG_FORCE);
 module_param(sig_enforce, bool_enable_only, 0644);
 
+void set_module_sig_enforced(void)
+{
+	sig_enforce = true;
+}
+#else
+#define sig_enforce false
+#endif
+
 /*
  * Export sig_enforce kernel cmdline parameter to allow other subsystems rely
  * on that instead of directly to CONFIG_MODULE_SIG_FORCE config.
@@ -285,11 +294,6 @@ bool is_module_sig_enforced(void)
 }
 EXPORT_SYMBOL(is_module_sig_enforced);
 
-void set_module_sig_enforced(void)
-{
-	sig_enforce = true;
-}
-
 /* Block module loading/unloading? */
 int modules_disabled = 0;
 core_param(nomodule, modules_disabled, bint, 0);
Signature · 1 line
2.30.2
author maintainer
#3 lore raw

PATCH 3/101 Revert "drm/amdgpu/gfx10: enlarge CP_MEC_DOORBELL_RANGE_UPPER to cover full doorbell."

In reply to Sasha Levin

From: Yifan Zhang <[email protected]>

commit baacf52a473b24e10322b67757ddb92ab8d86717 upstream.

This reverts commit 1c0b0efd148d5b24c4932ddb3fa03c8edd6097b3.

Reason for revert: Side effect of enlarging CP_MEC_DOORBELL_RANGE may cause some APUs fail to enter gfxoff in certain user cases.

---
 drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)
drivers/gpu/drm/amd/amdgpu/gfx_v10_0.c +1 -5
index 3c92dacbc24a..fc8da5fed779 100644
@@ -6590,12 +6590,8 @@ static int gfx_v10_0_kiq_init_register(struct amdgpu_ring *ring)
 	if (ring->use_doorbell) {
 		WREG32_SOC15(GC, 0, mmCP_MEC_DOORBELL_RANGE_LOWER,
 			(adev->doorbell_index.kiq * 2) << 2);
-		/* If GC has entered CGPG, ringing doorbell > first page doesn't
-		 * wakeup GC. Enlarge CP_MEC_DOORBELL_RANGE_UPPER to workaround
-		 * this issue.
-		 */
 		WREG32_SOC15(GC, 0, mmCP_MEC_DOORBELL_RANGE_UPPER,
-			(adev->doorbell.size - 4));
+			(adev->doorbell_index.userqueue_end * 2) << 2);
 	}
 
 	WREG32_SOC15(GC, 0, mmCP_HQD_PQ_DOORBELL_CONTROL,
Signature · 1 line
2.30.2
author maintainer
#4 lore raw

PATCH 4/101 drm: add a locked version of drm_is_current_master

In reply to Sasha Levin

From: Desmond Cheong Zhi Xi <[email protected]>

commit 1815d9c86e3090477fbde066ff314a7e9721ee0f upstream.

While checking the master status of the DRM file in drm_is_current_master(), the device's master mutex should be held. Without the mutex, the pointer fpriv->master may be freed concurrently by another process calling drm_setmaster_ioctl(). This could lead to use-after-free errors when the pointer is subsequently dereferenced in drm_lease_owner().

The callers of drm_is_current_master() from drm_auth.c hold the device's master mutex, but external callers do not. Hence, we implement drm_is_current_master_locked() to be used within drm_auth.c, and modify drm_is_current_master() to grab the device's master mutex before checking the master status.

---
 drivers/gpu/drm/drm_auth.c | 51 ++++++++++++++++++++++++--------------
 1 file changed, 32 insertions(+), 19 deletions(-)
drivers/gpu/drm/drm_auth.c +32 -19
index 232abbba3686..86d4b72e95cb 100644
@@ -61,6 +61,35 @@
  * trusted clients.
  */
 
+static bool drm_is_current_master_locked(struct drm_file *fpriv)
+{
+	lockdep_assert_held_once(&fpriv->master->dev->master_mutex);
+
+	return fpriv->is_master && drm_lease_owner(fpriv->master) == fpriv->minor->dev->master;
+}
+
+/**
+ * drm_is_current_master - checks whether @priv is the current master
+ * @fpriv: DRM file private
+ *
+ * Checks whether @fpriv is current master on its device. This decides whether a
+ * client is allowed to run DRM_MASTER IOCTLs.
+ *
+ * Most of the modern IOCTL which require DRM_MASTER are for kernel modesetting
+ * - the current master is assumed to own the non-shareable display hardware.
+ */
+bool drm_is_current_master(struct drm_file *fpriv)
+{
+	bool ret;
+
+	mutex_lock(&fpriv->master->dev->master_mutex);
+	ret = drm_is_current_master_locked(fpriv);
+	mutex_unlock(&fpriv->master->dev->master_mutex);
+
+	return ret;
+}
+EXPORT_SYMBOL(drm_is_current_master);
+
 int drm_getmagic(struct drm_device *dev, void *data, struct drm_file *file_priv)
 {
 	struct drm_auth *auth = data;
@@ -223,7 +252,7 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		goto out_unlock;
 
-	if (drm_is_current_master(file_priv))
+	if (drm_is_current_master_locked(file_priv))
 		goto out_unlock;
 
 	if (dev->master) {
@@ -272,7 +301,7 @@ int drm_dropmaster_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		goto out_unlock;
 
-	if (!drm_is_current_master(file_priv)) {
+	if (!drm_is_current_master_locked(file_priv)) {
 		ret = -EINVAL;
 		goto out_unlock;
 	}
@@ -321,7 +350,7 @@ void drm_master_release(struct drm_file *file_priv)
 	if (file_priv->magic)
 		idr_remove(&file_priv->master->magic_map, file_priv->magic);
 
-	if (!drm_is_current_master(file_priv))
+	if (!drm_is_current_master_locked(file_priv))
 		goto out;
 
 	drm_legacy_lock_master_cleanup(dev, master);
@@ -342,22 +371,6 @@ void drm_master_release(struct drm_file *file_priv)
 	mutex_unlock(&dev->master_mutex);
 }
 
-/**
- * drm_is_current_master - checks whether @priv is the current master
- * @fpriv: DRM file private
- *
- * Checks whether @fpriv is current master on its device. This decides whether a
- * client is allowed to run DRM_MASTER IOCTLs.
- *
- * Most of the modern IOCTL which require DRM_MASTER are for kernel modesetting
- * - the current master is assumed to own the non-shareable display hardware.
- */
-bool drm_is_current_master(struct drm_file *fpriv)
-{
-	return fpriv->is_master && drm_lease_owner(fpriv->master) == fpriv->minor->dev->master;
-}
-EXPORT_SYMBOL(drm_is_current_master);
-
 /**
  * drm_master_get - reference a master pointer
  * @master: &struct drm_master
Signature · 1 line
2.30.2
author maintainer
#6 lore raw

PATCH 6/101 drm/radeon: wait for moving fence after pinning

In reply to Sasha Levin

From: Christian König <[email protected]>

commit 4b41726aae563273bb4b4a9462ba51ce4d372f78 upstream.

We actually need to wait for the moving fence after pinning the BO to make sure that the pin is completed.

References: https://lore.kernel.org/dri-devel/[email protected]/

---
 drivers/gpu/drm/radeon/radeon_prime.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/gpu/drm/radeon/radeon_prime.c +13 -3
index b9de0e51c0be..cbad81578190 100644
@@ -94,9 +94,19 @@ int radeon_gem_prime_pin(struct drm_gem_object *obj)
 
 	/* pin buffer into GTT */
 	ret = radeon_bo_pin(bo, RADEON_GEM_DOMAIN_GTT, NULL);
-	if (likely(ret == 0))
-		bo->prime_shared_count++;
-
+	if (unlikely(ret))
+		goto error;
+
+	if (bo->tbo.moving) {
+		ret = dma_fence_wait(bo->tbo.moving, false);
+		if (unlikely(ret)) {
+			radeon_bo_unpin(bo);
+			goto error;
+		}
+	}
+
+	bo->prime_shared_count++;
+error:
 	radeon_bo_unreserve(bo);
 	return ret;
 }
Signature · 1 line
2.30.2
author maintainer
#5 lore raw

PATCH 5/101 drm/nouveau: wait for moving fence after pinning v2

In reply to Sasha Levin

From: Christian König <[email protected]>

commit 17b11f71795abdce46f62a808f906857e525cea8 upstream.

We actually need to wait for the moving fence after pinning the BO to make sure that the pin is completed.

v2: grab the lock while waiting

References: https://lore.kernel.org/dri-devel/[email protected]/

---
 drivers/gpu/drm/nouveau/nouveau_prime.c | 17 ++++++++++++++++-
 1 file changed, 16 insertions(+), 1 deletion(-)
drivers/gpu/drm/nouveau/nouveau_prime.c +16 -1
index b2ecb91f8ddc..5f5b87f99546 100644
@@ -111,7 +111,22 @@ int nouveau_gem_prime_pin(struct drm_gem_object *obj)
 	if (ret)
 		return -EINVAL;
 
-	return 0;
+	ret = ttm_bo_reserve(&nvbo->bo, false, false, NULL);
+	if (ret)
+		goto error;
+
+	if (nvbo->bo.moving)
+		ret = dma_fence_wait(nvbo->bo.moving, true);
+
+	ttm_bo_unreserve(&nvbo->bo);
+	if (ret)
+		goto error;
+
+	return ret;
+
+error:
+	nouveau_bo_unpin(nvbo);
+	return ret;
 }
 
 void nouveau_gem_prime_unpin(struct drm_gem_object *obj)
Signature · 1 line
2.30.2
author maintainer
#8 lore raw

PATCH 8/101 ARM: 9081/1: fix gcc-10 thumb2-kernel regression

In reply to Sasha Levin

From: Arnd Bergmann <[email protected]>

commit dad7b9896a5dbac5da8275d5a6147c65c81fb5f2 upstream.

When building the kernel wtih gcc-10 or higher using the CONFIG_CC_OPTIMIZE_FOR_PERFORMANCE=y flag, the compiler picks a slightly different set of registers for the inline assembly in cpu_init() that subsequently results in a corrupt kernel stack as well as remaining in FIQ mode. If a banked register is used for the last argument, the wrong version of that register gets loaded into CPSR_c. When building in Arm mode, the arguments are passed as immediate values and the bug cannot happen.

This got introduced when Daniel reworked the FIQ handling and was technically always broken, but happened to work with both clang and gcc before gcc-10 as long as they picked one of the lower registers. This is probably an indication that still very few people build the kernel in Thumb2 mode.

Marek pointed out the problem on IRC, Arnd narrowed it down to this inline assembly and Russell pinpointed the exact bug.

Change the constraints to force the final mode switch to use a non-banked register for the argument to ensure that the correct constant gets loaded. Another alternative would be to always use registers for the constant arguments to avoid the #ifdef that has now become more complex.

Cc <[email protected]> # v3.18+
Cc Daniel Thompson <[email protected]>
Fixes c0e7f7ee717e ("ARM: 8150/3: fiq: Replace default FIQ handler")
---
 arch/arm/kernel/setup.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)
arch/arm/kernel/setup.c +9 -7
index f90479d8b50c..b06602cea99c 100644
@@ -544,9 +544,11 @@ void notrace cpu_init(void)
 	 * In Thumb-2, msr with an immediate value is not allowed.
 	 */
 #ifdef CONFIG_THUMB2_KERNEL
-#define PLC	"r"
+#define PLC_l	"l"
+#define PLC_r	"r"
 #else
-#define PLC	"I"
+#define PLC_l	"I"
+#define PLC_r	"I"
 #endif
 
 	/*
@@ -568,15 +570,15 @@ void notrace cpu_init(void)
 	"msr	cpsr_c, %9"
 	    :
 	    : "r" (stk),
-	      PLC (PSR_F_BIT | PSR_I_BIT | IRQ_MODE),
+	      PLC_r (PSR_F_BIT | PSR_I_BIT | IRQ_MODE),
 	      "I" (offsetof(struct stack, irq[0])),
-	      PLC (PSR_F_BIT | PSR_I_BIT | ABT_MODE),
+	      PLC_r (PSR_F_BIT | PSR_I_BIT | ABT_MODE),
 	      "I" (offsetof(struct stack, abt[0])),
-	      PLC (PSR_F_BIT | PSR_I_BIT | UND_MODE),
+	      PLC_r (PSR_F_BIT | PSR_I_BIT | UND_MODE),
 	      "I" (offsetof(struct stack, und[0])),
-	      PLC (PSR_F_BIT | PSR_I_BIT | FIQ_MODE),
+	      PLC_r (PSR_F_BIT | PSR_I_BIT | FIQ_MODE),
 	      "I" (offsetof(struct stack, fiq[0])),
-	      PLC (PSR_F_BIT | PSR_I_BIT | SVC_MODE)
+	      PLC_l (PSR_F_BIT | PSR_I_BIT | SVC_MODE)
 	    : "r14");
 #endif
 }
Signature · 1 line
2.30.2
author maintainer
#9 lore raw

PATCH 9/101 mmc: meson-gx: use memcpy_to/fromio for dram-access-quirk

In reply to Sasha Levin

From: Neil Armstrong <[email protected]>

commit 103a5348c22c3fca8b96c735a9e353b8a0801842 upstream.

It has been reported that usage of memcpy() to/from an iomem mapping is invalid, and a recent arm64 memcpy update [1] triggers a memory abort when dram-access-quirk is used on the G12A/G12B platforms.

This adds a local sg_copy_to_buffer which makes usage of io versions of memcpy when dram-access-quirk is enabled.

[1] 285133040e6c ("arm64: Import latest memcpy()/memmove() implementation")

---
 drivers/mmc/host/meson-gx-mmc.c | 50 +++++++++++++++++++++++++++++----
 1 file changed, 45 insertions(+), 5 deletions(-)
drivers/mmc/host/meson-gx-mmc.c +45 -5
index 4ec41579940a..d3f40c9a8c6c 100644
@@ -165,6 +165,7 @@ struct meson_host {
 
 	unsigned int bounce_buf_size;
 	void *bounce_buf;
+	void __iomem *bounce_iomem_buf;
 	dma_addr_t bounce_dma_addr;
 	struct sd_emmc_desc *descs;
 	dma_addr_t descs_dma_addr;
@@ -734,6 +735,47 @@ static void meson_mmc_desc_chain_transfer(struct mmc_host *mmc, u32 cmd_cfg)
 	writel(start, host->regs + SD_EMMC_START);
 }
 
+/* local sg copy to buffer version with _to/fromio usage for dram_access_quirk */
+static void meson_mmc_copy_buffer(struct meson_host *host, struct mmc_data *data,
+				  size_t buflen, bool to_buffer)
+{
+	unsigned int sg_flags = SG_MITER_ATOMIC;
+	struct scatterlist *sgl = data->sg;
+	unsigned int nents = data->sg_len;
+	struct sg_mapping_iter miter;
+	unsigned int offset = 0;
+
+	if (to_buffer)
+		sg_flags |= SG_MITER_FROM_SG;
+	else
+		sg_flags |= SG_MITER_TO_SG;
+
+	sg_miter_start(&miter, sgl, nents, sg_flags);
+
+	while ((offset < buflen) && sg_miter_next(&miter)) {
+		unsigned int len;
+
+		len = min(miter.length, buflen - offset);
+
+		/* When dram_access_quirk, the bounce buffer is a iomem mapping */
+		if (host->dram_access_quirk) {
+			if (to_buffer)
+				memcpy_toio(host->bounce_iomem_buf + offset, miter.addr, len);
+			else
+				memcpy_fromio(miter.addr, host->bounce_iomem_buf + offset, len);
+		} else {
+			if (to_buffer)
+				memcpy(host->bounce_buf + offset, miter.addr, len);
+			else
+				memcpy(miter.addr, host->bounce_buf + offset, len);
+		}
+
+		offset += len;
+	}
+
+	sg_miter_stop(&miter);
+}
+
 static void meson_mmc_start_cmd(struct mmc_host *mmc, struct mmc_command *cmd)
 {
 	struct meson_host *host = mmc_priv(mmc);
@@ -777,8 +819,7 @@ static void meson_mmc_start_cmd(struct mmc_host *mmc, struct mmc_command *cmd)
 		if (data->flags & MMC_DATA_WRITE) {
 			cmd_cfg |= CMD_CFG_DATA_WR;
 			WARN_ON(xfer_bytes > host->bounce_buf_size);
-			sg_copy_to_buffer(data->sg, data->sg_len,
-					  host->bounce_buf, xfer_bytes);
+			meson_mmc_copy_buffer(host, data, xfer_bytes, true);
 			dma_wmb();
 		}
 
@@ -947,8 +988,7 @@ static irqreturn_t meson_mmc_irq_thread(int irq, void *dev_id)
 	if (meson_mmc_bounce_buf_read(data)) {
 		xfer_bytes = data->blksz * data->blocks;
 		WARN_ON(xfer_bytes > host->bounce_buf_size);
-		sg_copy_from_buffer(data->sg, data->sg_len,
-				    host->bounce_buf, xfer_bytes);
+		meson_mmc_copy_buffer(host, data, xfer_bytes, false);
 	}
 
 	next_cmd = meson_mmc_get_next_command(cmd);
@@ -1168,7 +1208,7 @@ static int meson_mmc_probe(struct platform_device *pdev)
 		 * instead of the DDR memory
 		 */
 		host->bounce_buf_size = SD_EMMC_SRAM_DATA_BUF_LEN;
-		host->bounce_buf = host->regs + SD_EMMC_SRAM_DATA_BUF_OFF;
+		host->bounce_iomem_buf = host->regs + SD_EMMC_SRAM_DATA_BUF_OFF;
 		host->bounce_dma_addr = res->start + SD_EMMC_SRAM_DATA_BUF_OFF;
 	} else {
 		/* data bounce buffer */
Signature · 1 line
2.30.2
author maintainer
#12 lore raw

PATCH 12/101 arm64: Force NO_BLOCK_MAPPINGS if crashkernel reservation is required

In reply to Sasha Levin

From: Catalin Marinas <[email protected]>

commit 2687275a5843d1089687f08fc64eb3f3b026a169 upstream.

mem_init() currently relies on knowing the boundaries of the crashkernel reservation to map such region with page granularity for later unmapping via set_memory_valid(..., 0). If the crashkernel reservation is deferred, such boundaries are not known when the linear mapping is created. Simply parse the command line for "crashkernel" and, if found, create the linear map with NO_BLOCK_MAPPINGS.

---
 arch/arm64/mm/mmu.c | 37 ++++++++++++++++---------------------
 1 file changed, 16 insertions(+), 21 deletions(-)
arch/arm64/mm/mmu.c +16 -21
index afdad7607850..58dc93e56617 100644
@@ -469,6 +469,21 @@ void __init mark_linear_text_alias_ro(void)
 			    PAGE_KERNEL_RO);
 }
 
+static bool crash_mem_map __initdata;
+
+static int __init enable_crash_mem_map(char *arg)
+{
+	/*
+	 * Proper parameter parsing is done by reserve_crashkernel(). We only
+	 * need to know if the linear map has to avoid block mappings so that
+	 * the crashkernel reservations can be unmapped later.
+	 */
+	crash_mem_map = true;
+
+	return 0;
+}
+early_param("crashkernel", enable_crash_mem_map);
+
 static void __init map_mem(pgd_t *pgdp)
 {
 	phys_addr_t kernel_start = __pa_symbol(_text);
@@ -477,7 +492,7 @@ static void __init map_mem(pgd_t *pgdp)
 	int flags = 0;
 	u64 i;
 
-	if (rodata_full || debug_pagealloc_enabled())
+	if (rodata_full || crash_mem_map || debug_pagealloc_enabled())
 		flags = NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS;
 
 	/*
@@ -487,11 +502,6 @@ static void __init map_mem(pgd_t *pgdp)
 	 * the following for-loop
 	 */
 	memblock_mark_nomap(kernel_start, kernel_end - kernel_start);
-#ifdef CONFIG_KEXEC_CORE
-	if (crashk_res.end)
-		memblock_mark_nomap(crashk_res.start,
-				    resource_size(&crashk_res));
-#endif
 
 	/* map all the memory banks */
 	for_each_mem_range(i, &start, &end) {
@@ -519,21 +529,6 @@ static void __init map_mem(pgd_t *pgdp)
 	__map_memblock(pgdp, kernel_start, kernel_end,
 		       PAGE_KERNEL, NO_CONT_MAPPINGS);
 	memblock_clear_nomap(kernel_start, kernel_end - kernel_start);
-
-#ifdef CONFIG_KEXEC_CORE
-	/*
-	 * Use page-level mappings here so that we can shrink the region
-	 * in page granularity and put back unused memory to buddy system
-	 * through /sys/kernel/kexec_crash_size interface.
-	 */
-	if (crashk_res.end) {
-		__map_memblock(pgdp, crashk_res.start, crashk_res.end + 1,
-			       PAGE_KERNEL,
-			       NO_BLOCK_MAPPINGS | NO_CONT_MAPPINGS);
-		memblock_clear_nomap(crashk_res.start,
-				     resource_size(&crashk_res));
-	}
-#endif
 }
 
 void mark_rodata_ro(void)
Signature · 1 line
2.30.2
author maintainer
#13 lore raw

PATCH 13/101 spi: spi-nxp-fspi: move the register operation after the clock enable

In reply to Sasha Levin

From: Haibo Chen <[email protected]>

[ Upstream commit f422316c8e9d3c4aff3c56549dfb44a677d02f14 ]

Move the register operation after the clock enable, otherwise system will stuck when this driver probe.

Fixes 71d80563b076 ("spi: spi-nxp-fspi: fix fspi panic by unexpected interrupts")
---
 drivers/spi/spi-nxp-fspi.c | 11 +++++------
 1 file changed, 5 insertions(+), 6 deletions(-)
drivers/spi/spi-nxp-fspi.c +5 -6
index ab9035662717..bcc0b5a3a459 100644
@@ -1033,12 +1033,6 @@ static int nxp_fspi_probe(struct platform_device *pdev)
 		goto err_put_ctrl;
 	}
 
-	/* Clear potential interrupts */
-	reg = fspi_readl(f, f->iobase + FSPI_INTR);
-	if (reg)
-		fspi_writel(f, reg, f->iobase + FSPI_INTR);
-
-
 	/* find the resources - controller memory mapped space */
 	if (is_acpi_node(f->dev->fwnode))
 		res = platform_get_resource(pdev, IORESOURCE_MEM, 1);
@@ -1076,6 +1070,11 @@ static int nxp_fspi_probe(struct platform_device *pdev)
 		}
 	}
 
+	/* Clear potential interrupts */
+	reg = fspi_readl(f, f->iobase + FSPI_INTR);
+	if (reg)
+		fspi_writel(f, reg, f->iobase + FSPI_INTR);
+
 	/* find the irq */
 	ret = platform_get_irq(pdev, 0);
 	if (ret < 0)
Signature · 1 line
2.30.2
author maintainer
#14 lore raw

PATCH 14/101 Revert "PCI: PM: Do not read power state in pci_enable_device_flags()"

In reply to Sasha Levin

From: "Rafael J. Wysocki" <[email protected]>

[ Upstream commit 4d6035f9bf4ea12776322746a216e856dfe46698 ]

Revert commit 4514d991d992 ("PCI: PM: Do not read power state in pci_enable_device_flags()") that is reported to cause PCI device initialization issues on some systems.

---
 drivers/pci/pci.c | 16 +++++++++++++---
 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/pci/pci.c +13 -3
index d5d9ea864fe6..9e971fffeb6a 100644
@@ -1874,11 +1874,21 @@ static int pci_enable_device_flags(struct pci_dev *dev, unsigned long flags)
 	int err;
 	int i, bars = 0;
 
-	if (atomic_inc_return(&dev->enable_cnt) > 1) {
-		pci_update_current_state(dev, dev->current_state);
-		return 0;		/* already enabled */
+	/*
+	 * Power state could be unknown at this point, either due to a fresh
+	 * boot or a device removal call.  So get the current power state
+	 * so that things like MSI message writing will behave as expected
+	 * (e.g. if the device really is in D0 at enable time).
+	 */
+	if (dev->pm_cap) {
+		u16 pmcsr;
+		pci_read_config_word(dev, dev->pm_cap + PCI_PM_CTRL, &pmcsr);
+		dev->current_state = (pmcsr & PCI_PM_CTRL_STATE_MASK);
 	}
 
+	if (atomic_inc_return(&dev->enable_cnt) > 1)
+		return 0;		/* already enabled */
+
 	bridge = pci_upstream_bridge(dev);
 	if (bridge)
 		pci_enable_bridge(bridge);
Signature · 1 line
2.30.2
author maintainer
#15 lore raw

PATCH 15/101 drm/vc4: hdmi: Move the HSM clock enable to runtime_pm

In reply to Sasha Levin

From: Maxime Ripard <[email protected]>

[ Upstream commit 411efa18e4b03840553ff58ad9b4621b82a30c04 ]

In order to access the HDMI controller, we need to make sure the HSM clock is enabled. If we were to access it with the clock disabled, the CPU would completely hang, resulting in an hard crash.

Since we have different code path that would require it, let's move that clock enable / disable to runtime_pm that will take care of the reference counting for us.

Fixes 4f6e3d66ac52 ("drm/vc4: Add runtime PM support to the HDMI encoder driver")
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 40 +++++++++++++++++++++++++---------
 1 file changed, 30 insertions(+), 10 deletions(-)
drivers/gpu/drm/vc4/vc4_hdmi.c +30 -10
index af5f01eff872..5978e99a0e85 100644
@@ -415,7 +415,6 @@ static void vc4_hdmi_encoder_post_crtc_powerdown(struct drm_encoder *encoder)
 		   HDMI_READ(HDMI_VID_CTL) & ~VC4_HD_VID_CTL_ENABLE);
 
 	clk_disable_unprepare(vc4_hdmi->pixel_bvb_clock);
-	clk_disable_unprepare(vc4_hdmi->hsm_clock);
 	clk_disable_unprepare(vc4_hdmi->pixel_clock);
 
 	ret = pm_runtime_put(&vc4_hdmi->pdev->dev);
@@ -666,13 +665,6 @@ static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder)
 		return;
 	}
 
-	ret = clk_prepare_enable(vc4_hdmi->hsm_clock);
-	if (ret) {
-		DRM_ERROR("Failed to turn on HSM clock: %d\n", ret);
-		clk_disable_unprepare(vc4_hdmi->pixel_clock);
-		return;
-	}
-
 	vc4_hdmi_cec_update_clk_div(vc4_hdmi);
 
 	/*
@@ -683,7 +675,6 @@ static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder)
 			       (hsm_rate > VC4_HSM_MID_CLOCK ? 150000000 : 75000000));
 	if (ret) {
 		DRM_ERROR("Failed to set pixel bvb clock rate: %d\n", ret);
-		clk_disable_unprepare(vc4_hdmi->hsm_clock);
 		clk_disable_unprepare(vc4_hdmi->pixel_clock);
 		return;
 	}
@@ -691,7 +682,6 @@ static void vc4_hdmi_encoder_pre_crtc_configure(struct drm_encoder *encoder)
 	ret = clk_prepare_enable(vc4_hdmi->pixel_bvb_clock);
 	if (ret) {
 		DRM_ERROR("Failed to turn on pixel bvb clock: %d\n", ret);
-		clk_disable_unprepare(vc4_hdmi->hsm_clock);
 		clk_disable_unprepare(vc4_hdmi->pixel_clock);
 		return;
 	}
@@ -1724,6 +1714,29 @@ static int vc5_hdmi_init_resources(struct vc4_hdmi *vc4_hdmi)
 	return 0;
 }
 
+#ifdef CONFIG_PM
+static int vc4_hdmi_runtime_suspend(struct device *dev)
+{
+	struct vc4_hdmi *vc4_hdmi = dev_get_drvdata(dev);
+
+	clk_disable_unprepare(vc4_hdmi->hsm_clock);
+
+	return 0;
+}
+
+static int vc4_hdmi_runtime_resume(struct device *dev)
+{
+	struct vc4_hdmi *vc4_hdmi = dev_get_drvdata(dev);
+	int ret;
+
+	ret = clk_prepare_enable(vc4_hdmi->hsm_clock);
+	if (ret)
+		return ret;
+
+	return 0;
+}
+#endif
+
 static int vc4_hdmi_bind(struct device *dev, struct device *master, void *data)
 {
 	const struct vc4_hdmi_variant *variant = of_device_get_match_data(dev);
@@ -1959,11 +1972,18 @@ static const struct of_device_id vc4_hdmi_dt_match[] = {
 	{}
 };
 
+static const struct dev_pm_ops vc4_hdmi_pm_ops = {
+	SET_RUNTIME_PM_OPS(vc4_hdmi_runtime_suspend,
+			   vc4_hdmi_runtime_resume,
+			   NULL)
+};
+
 struct platform_driver vc4_hdmi_driver = {
 	.probe = vc4_hdmi_dev_probe,
 	.remove = vc4_hdmi_dev_remove,
 	.driver = {
 		.name = "vc4_hdmi",
 		.of_match_table = vc4_hdmi_dt_match,
+		.pm = &vc4_hdmi_pm_ops,
 	},
 };
Signature · 1 line
2.30.2
author maintainer
#16 lore raw

PATCH 16/101 drm/vc4: hdmi: Make sure the controller is powered in detect

In reply to Sasha Levin

From: Maxime Ripard <[email protected]>

[ Upstream commit 9984d6664ce9dcbbc713962539eaf7636ea246c2 ]

If the HPD GPIO is not available and drm_probe_ddc fails, we end up reading the HDMI_HOTPLUG register, but the controller might be powered off resulting in a CPU hang. Make sure we have the power domain and the HSM clock powered during the detect cycle to prevent the hang from happening.

Fixes 4f6e3d66ac52 ("drm/vc4: Add runtime PM support to the HDMI encoder driver")
---
 drivers/gpu/drm/vc4/vc4_hdmi.c | 4 ++++
 1 file changed, 4 insertions(+)
drivers/gpu/drm/vc4/vc4_hdmi.c +4 -0
index 5978e99a0e85..88a8cb840cd5 100644
@@ -146,6 +146,8 @@ vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
 	struct vc4_hdmi *vc4_hdmi = connector_to_vc4_hdmi(connector);
 	bool connected = false;
 
+	WARN_ON(pm_runtime_resume_and_get(&vc4_hdmi->pdev->dev));
+
 	if (vc4_hdmi->hpd_gpio) {
 		if (gpio_get_value_cansleep(vc4_hdmi->hpd_gpio) ^
 		    vc4_hdmi->hpd_active_low)
@@ -167,10 +169,12 @@ vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
 			}
 		}
 
+		pm_runtime_put(&vc4_hdmi->pdev->dev);
 		return connector_status_connected;
 	}
 
 	cec_phys_addr_invalidate(vc4_hdmi->cec_adap);
+	pm_runtime_put(&vc4_hdmi->pdev->dev);
 	return connector_status_disconnected;
 }
 
Signature · 1 line
2.30.2
author maintainer
#18 lore raw

PATCH 17/101 x86/entry: Fix noinstr fail in __do_fast_syscall_32()

In reply to Sasha Levin

From: Peter Zijlstra <[email protected]>

[ Upstream commit 240001d4e3041832e8a2654adc3ccf1683132b92 ]

Fix:

vmlinux.o: warning: objtool: __do_fast_syscall_32()+0xf5: call to trace_hardirqs_off() leaves .noinstr.text section

Fixes 5d5675df792f ("x86/entry: Fix entry/exit mismatch on failed fast 32-bit syscalls")
---
 arch/x86/entry/common.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/entry/common.c +1 -1
index 2e4d91f3feea..93a3122cd15f 100644
@@ -127,8 +127,8 @@ static noinstr bool __do_fast_syscall_32(struct pt_regs *regs)
 		/* User code screwed up. */
 		regs->ax = -EFAULT;
 
-		instrumentation_end();
 		local_irq_disable();
+		instrumentation_end();
 		irqentry_exit_to_user_mode(regs);
 		return false;
 	}
Signature · 1 line
2.30.2
author maintainer
#19 lore raw

PATCH 18/101 x86/xen: Fix noinstr fail in exc_xen_unknown_trap()

In reply to Sasha Levin

From: Peter Zijlstra <[email protected]>

[ Upstream commit 4c9c26f1e67648f41f28f8c997c5c9467a3dbbe4 ]

Fix:

vmlinux.o: warning: objtool: exc_xen_unknown_trap()+0x7: call to printk() leaves .noinstr.text section

Fixes 2e92493637a0 ("x86/xen: avoid warning in Xen pv guest with CONFIG_AMD_MEM_ENCRYPT enabled")
---
 arch/x86/xen/enlighten_pv.c | 2 ++
 1 file changed, 2 insertions(+)
arch/x86/xen/enlighten_pv.c +2 -0
index 8064df638222..d3cdf467d91f 100644
@@ -586,8 +586,10 @@ DEFINE_IDTENTRY_RAW(xenpv_exc_debug)
 DEFINE_IDTENTRY_RAW(exc_xen_unknown_trap)
 {
 	/* This should never happen and there is no way to handle it. */
+	instrumentation_begin();
 	pr_err("Unknown trap in Xen PV mode.");
 	BUG();
+	instrumentation_end();
 }
 
 struct trap_array_entry {
Signature · 1 line
2.30.2
author maintainer
#20 lore raw

PATCH 19/101 locking/lockdep: Improve noinstr vs errors

In reply to Sasha Levin

From: Peter Zijlstra <[email protected]>

[ Upstream commit 49faa77759b211fff344898edc23bb780707fff5 ]

Better handle the failure paths.

vmlinux.o: warning: objtool: debug_locks_off()+0x23: call to console_verbose() leaves .noinstr.text section vmlinux.o: warning: objtool: debug_locks_off()+0x19: call to __kasan_check_write() leaves .noinstr.text section

  debug_locks_off+0x19/0x40:
  instrument_atomic_write at include/linux/instrumented.h:86
  (inlined by) __debug_locks_off at include/linux/debug_locks.h:17
  (inlined by) debug_locks_off at lib/debug_locks.c:41
Fixes 6eebad1ad303 ("lockdep: __always_inline more for noinstr")
---
 include/linux/debug_locks.h | 2 ++
 kernel/locking/lockdep.c    | 4 +++-
 lib/debug_locks.c           | 2 +-
 3 files changed, 6 insertions(+), 2 deletions(-)
include/linux/debug_locks.h +2 -0
index 2915f56ad421..edb5c186b0b7 100644
@@ -27,8 +27,10 @@ extern int debug_locks_off(void);
 	int __ret = 0;							\
 									\
 	if (!oops_in_progress && unlikely(c)) {				\
+		instrumentation_begin();				\
 		if (debug_locks_off() && !debug_locks_silent)		\
 			WARN(1, "DEBUG_LOCKS_WARN_ON(%s)", #c);		\
+		instrumentation_end();					\
 		__ret = 1;						\
 	}								\
 	__ret;								\
kernel/locking/lockdep.c +3 -1
index 858b96b438ce..cdca007551e7 100644
@@ -842,7 +842,7 @@ static int count_matching_names(struct lock_class *new_class)
 }
 
 /* used from NMI context -- must be lockless */
-static __always_inline struct lock_class *
+static noinstr struct lock_class *
 look_up_lock_class(const struct lockdep_map *lock, unsigned int subclass)
 {
 	struct lockdep_subclass_key *key;
@@ -850,12 +850,14 @@ look_up_lock_class(const struct lockdep_map *lock, unsigned int subclass)
 	struct lock_class *class;
 
 	if (unlikely(subclass >= MAX_LOCKDEP_SUBCLASSES)) {
+		instrumentation_begin();
 		debug_locks_off();
 		printk(KERN_ERR
 			"BUG: looking up invalid subclass: %u\n", subclass);
 		printk(KERN_ERR
 			"turning off the locking correctness validator.\n");
 		dump_stack();
+		instrumentation_end();
 		return NULL;
 	}
 
lib/debug_locks.c +1 -1
index 06d3135bd184..a75ee30b77cb 100644
@@ -36,7 +36,7 @@ EXPORT_SYMBOL_GPL(debug_locks_silent);
 /*
  * Generic 'turn off all lock debugging' function:
  */
-noinstr int debug_locks_off(void)
+int debug_locks_off(void)
 {
 	if (debug_locks && __debug_locks_off()) {
 		if (!debug_locks_silent) {
Signature · 1 line
2.30.2
author maintainer
#21 lore raw

PATCH 20/101 perf/x86/lbr: Remove cpuc->lbr_xsave allocation from atomic context

In reply to Sasha Levin

From: Like Xu <[email protected]>

[ Upstream commit 488e13a489e9707a7e81e1991fdd1f20c0f04689 ]

If the kernel is compiled with the CONFIG_LOCKDEP option, the conditional might_sleep_if() deep in kmem_cache_alloc() will generate the following trace, and potentially cause a deadlock when another LBR event is added:

[] BUG: sleeping function called from invalid context at include/linux/sched/mm.h:196 [] Call Trace:

  []  kmem_cache_alloc+0x36/0x250
  []  intel_pmu_lbr_add+0x152/0x170
  []  x86_pmu_add+0x83/0xd0

Make it symmetric with the release_lbr_buffers() call and mirror the existing DS buffers.

Fixes c085fb8774 ("perf/x86/intel/lbr: Support XSAVES for arch LBR read")
[peterz: simplified]
---
 arch/x86/events/core.c       |  6 ++++--
 arch/x86/events/intel/lbr.c  | 26 ++++++++++++++++++++------
 arch/x86/events/perf_event.h |  6 ++++++
 3 files changed, 30 insertions(+), 8 deletions(-)
arch/x86/events/core.c +4 -2
index a88c94d65693..b7f8ed87bfbc 100644
@@ -372,10 +372,12 @@ int x86_reserve_hardware(void)
 	if (!atomic_inc_not_zero(&pmc_refcount)) {
 		mutex_lock(&pmc_reserve_mutex);
 		if (atomic_read(&pmc_refcount) == 0) {
-			if (!reserve_pmc_hardware())
+			if (!reserve_pmc_hardware()) {
 				err = -EBUSY;
-			else
+			} else {
 				reserve_ds_buffers();
+				reserve_lbr_buffers();
+			}
 		}
 		if (!err)
 			atomic_inc(&pmc_refcount);
arch/x86/events/intel/lbr.c +20 -6
index e2b0efcba101..6c1231837382 100644
@@ -658,7 +658,6 @@ static inline bool branch_user_callstack(unsigned br_sel)
 
 void intel_pmu_lbr_add(struct perf_event *event)
 {
-	struct kmem_cache *kmem_cache = event->pmu->task_ctx_cache;
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
 
 	if (!x86_pmu.lbr_nr)
@@ -696,11 +695,6 @@ void intel_pmu_lbr_add(struct perf_event *event)
 	perf_sched_cb_inc(event->ctx->pmu);
 	if (!cpuc->lbr_users++ && !event->total_time_running)
 		intel_pmu_lbr_reset();
-
-	if (static_cpu_has(X86_FEATURE_ARCH_LBR) &&
-	    kmem_cache && !cpuc->lbr_xsave &&
-	    (cpuc->lbr_users != cpuc->lbr_pebs_users))
-		cpuc->lbr_xsave = kmem_cache_alloc(kmem_cache, GFP_KERNEL);
 }
 
 void release_lbr_buffers(void)
@@ -721,6 +715,26 @@ void release_lbr_buffers(void)
 	}
 }
 
+void reserve_lbr_buffers(void)
+{
+	struct kmem_cache *kmem_cache;
+	struct cpu_hw_events *cpuc;
+	int cpu;
+
+	if (!static_cpu_has(X86_FEATURE_ARCH_LBR))
+		return;
+
+	for_each_possible_cpu(cpu) {
+		cpuc = per_cpu_ptr(&cpu_hw_events, cpu);
+		kmem_cache = x86_get_pmu(cpu)->task_ctx_cache;
+		if (!kmem_cache || cpuc->lbr_xsave)
+			continue;
+
+		cpuc->lbr_xsave = kmem_cache_alloc_node(kmem_cache, GFP_KERNEL,
+							cpu_to_node(cpu));
+	}
+}
+
 void intel_pmu_lbr_del(struct perf_event *event)
 {
 	struct cpu_hw_events *cpuc = this_cpu_ptr(&cpu_hw_events);
arch/x86/events/perf_event.h +6 -0
index 6a8edfe59b09..d4f2ea2d9a9e 100644
@@ -1122,6 +1122,8 @@ void reserve_ds_buffers(void);
 
 void release_lbr_buffers(void);
 
+void reserve_lbr_buffers(void);
+
 extern struct event_constraint bts_constraint;
 extern struct event_constraint vlbr_constraint;
 
@@ -1267,6 +1269,10 @@ static inline void release_lbr_buffers(void)
 {
 }
 
+static inline void reserve_lbr_buffers(void)
+{
+}
+
 static inline int intel_pmu_init(void)
 {
 	return 0;
Signature · 1 line
2.30.2
author maintainer
#28 lore raw

PATCH 27/101 mac80211_hwsim: drop pending frames on stop

In reply to Sasha Levin

From: Johannes Berg <[email protected]>

[ Upstream commit bd18de517923903a177508fc8813f44e717b1c00 ]

Syzbot reports that we may be able to get into a situation where mac80211 has pending ACK frames on shutdown with hwsim. It appears that the reason for this is that syzbot uses the wmediumd hooks to intercept/injection frames, and may shut down hwsim, removing the radio(s), while frames are pending in the air simulation.

Clean out the pending queue when the interface is stopped, after this the frames can't be reported back to mac80211 properly anyway.

---
 drivers/net/wireless/mac80211_hwsim.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/net/wireless/mac80211_hwsim.c +5 -0
index 3b3fc7c9c91d..f147d4feedb9 100644
@@ -1623,8 +1623,13 @@ static int mac80211_hwsim_start(struct ieee80211_hw *hw)
 static void mac80211_hwsim_stop(struct ieee80211_hw *hw)
 {
 	struct mac80211_hwsim_data *data = hw->priv;
+
 	data->started = false;
 	hrtimer_cancel(&data->beacon_timer);
+
+	while (!skb_queue_empty(&data->pending))
+		ieee80211_free_txskb(hw, skb_dequeue(&data->pending));
+
 	wiphy_dbg(hw->wiphy, "%s\n", __func__);
 }
 
Signature · 1 line
2.30.2
author maintainer
#29 lore raw

PATCH 28/101 cfg80211: call cfg80211_leave_ocb when switching away from OCB

In reply to Sasha Levin

From: Du Cheng <[email protected]>

[ Upstream commit a64b6a25dd9f984ed05fade603a00e2eae787d2f ]

If the userland switches back-and-forth between NL80211_IFTYPE_OCB and NL80211_IFTYPE_ADHOC via send_msg(NL80211_CMD_SET_INTERFACE), there is a chance where the cleanup cfg80211_leave_ocb() is not called. This leads to initialization of in-use memory (e.g. init u.ibss while in-use by u.ocb) due to a shared struct/union within ieee80211_sub_if_data:

struct ieee80211_sub_if_data {
    ...
    union {
        struct ieee80211_if_ap ap;
        struct ieee80211_if_vlan vlan;
        struct ieee80211_if_managed mgd;
        struct ieee80211_if_ibss ibss; // <- shares address
        struct ieee80211_if_mesh mesh;
        struct ieee80211_if_ocb ocb; // <- shares address
        struct ieee80211_if_mntr mntr;
        struct ieee80211_if_nan nan;
    } u;
    ...
}

Therefore add handling of otype == NL80211_IFTYPE_OCB, during cfg80211_change_iface() to perform cleanup when leaving OCB mode.

link to syzkaller bug:
https://syzkaller.appspot.com/bug?id=0612dbfa595bf4b9b680ff7b4948257b8e3732d5

---
 net/wireless/util.c | 3 +++
 1 file changed, 3 insertions(+)
net/wireless/util.c +3 -0
index 2731267fd0f9..4fb8d1b14e76 100644
@@ -1059,6 +1059,9 @@ int cfg80211_change_iface(struct cfg80211_registered_device *rdev,
 		case NL80211_IFTYPE_MESH_POINT:
 			/* mesh should be handled? */
 			break;
+		case NL80211_IFTYPE_OCB:
+			cfg80211_leave_ocb(rdev, dev);
+			break;
 		default:
 			break;
 		}
Signature · 1 line
2.30.2
author maintainer
#26 lore raw

PATCH 25/101 dmaengine: xilinx: dpdma: Limit descriptor IDs to 16 bits

In reply to Sasha Levin

From: Laurent Pinchart <[email protected]>

[ Upstream commit 9f007e7b6643799e2a6538a5fe04f51c371c6657 ]

While the descriptor ID is stored in a 32-bit field in the hardware descriptor, only 16 bits are used by the hardware and are reported through the XILINX_DPDMA_CH_DESC_ID register. Failure to handle the wrap-around results in a descriptor ID mismatch after 65536 frames. Fix it.

---
 drivers/dma/xilinx/xilinx_dpdma.c | 7 +++++--
 1 file changed, 5 insertions(+), 2 deletions(-)
drivers/dma/xilinx/xilinx_dpdma.c +5 -2
index ff7dfb3fdeb4..6c709803203a 100644
@@ -113,6 +113,7 @@
 #define XILINX_DPDMA_CH_VDO				0x020
 #define XILINX_DPDMA_CH_PYLD_SZ				0x024
 #define XILINX_DPDMA_CH_DESC_ID				0x028
+#define XILINX_DPDMA_CH_DESC_ID_MASK			GENMASK(15, 0)
 
 /* DPDMA descriptor fields */
 #define XILINX_DPDMA_DESC_CONTROL_PREEMBLE		0xa5
@@ -866,7 +867,8 @@ static void xilinx_dpdma_chan_queue_transfer(struct xilinx_dpdma_chan *chan)
 	 * will be used, but it should be enough.
 	 */
 	list_for_each_entry(sw_desc, &desc->descriptors, node)
-		sw_desc->hw.desc_id = desc->vdesc.tx.cookie;
+		sw_desc->hw.desc_id = desc->vdesc.tx.cookie
+				    & XILINX_DPDMA_CH_DESC_ID_MASK;
 
 	sw_desc = list_first_entry(&desc->descriptors,
 				   struct xilinx_dpdma_sw_desc, node);
@@ -1086,7 +1088,8 @@ static void xilinx_dpdma_chan_vsync_irq(struct  xilinx_dpdma_chan *chan)
 	if (!chan->running || !pending)
 		goto out;
 
-	desc_id = dpdma_read(chan->reg, XILINX_DPDMA_CH_DESC_ID);
+	desc_id = dpdma_read(chan->reg, XILINX_DPDMA_CH_DESC_ID)
+		& XILINX_DPDMA_CH_DESC_ID_MASK;
 
 	/* If the retrigger raced with vsync, retry at the next frame. */
 	sw_desc = list_first_entry(&pending->descriptors,
Signature · 1 line
2.30.2
author maintainer
#25 lore raw

PATCH 24/101 dmaengine: xilinx: dpdma: Add missing dependencies to Kconfig

In reply to Sasha Levin

From: Laurent Pinchart <[email protected]>

[ Upstream commit 32828b82fb875b06511918b139d3a3cd93d34262 ]

The driver depends on both OF and IOMEM support, express those dependencies in Kconfig. This fixes a build failure on S390 reported by the 0day bot.

---
 drivers/dma/Kconfig | 1 +
 1 file changed, 1 insertion(+)
drivers/dma/Kconfig +1 -0
index f2db761ee548..f28bb2334e74 100644
@@ -693,6 +693,7 @@ config XILINX_ZYNQMP_DMA
 
 config XILINX_ZYNQMP_DPDMA
 	tristate "Xilinx DPDMA Engine"
+	depends on HAS_IOMEM && OF
 	select DMA_ENGINE
 	select DMA_VIRTUAL_CHANNELS
 	help
Signature · 1 line
2.30.2
author maintainer
#27 lore raw

PATCH 26/101 mac80211: remove warning in ieee80211_get_sband()

In reply to Sasha Levin

From: Johannes Berg <[email protected]>

[ Upstream commit 0ee4d55534f82a0624701d0bb9fc2304d4529086 ]

Syzbot reports that it's possible to hit this from userspace, by trying to add a station before any other connection setup has been done. Instead of trying to catch this in some other way simply remove the warning, that will appropriately reject the call from userspace.

---
 net/mac80211/ieee80211_i.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
net/mac80211/ieee80211_i.h +1 -1
index be40f6b16199..a83f0c2fcdf7 100644
@@ -1445,7 +1445,7 @@ ieee80211_get_sband(struct ieee80211_sub_if_data *sdata)
 	rcu_read_lock();
 	chanctx_conf = rcu_dereference(sdata->vif.chanctx_conf);
 
-	if (WARN_ON_ONCE(!chanctx_conf)) {
+	if (!chanctx_conf) {
 		rcu_read_unlock();
 		return NULL;
 	}
Signature · 1 line
2.30.2
author maintainer
#33 lore raw

PATCH 32/101 dmaengine: mediatek: use GFP_NOWAIT instead of GFP_ATOMIC in prep_dma

In reply to Sasha Levin

From: Guillaume Ranquet <[email protected]>

[ Upstream commit 9041575348b21ade1fb74d790f1aac85d68198c7 ]

As recommended by the doc in:
Documentation/drivers-api/dmaengine/provider.rst

Use GFP_NOWAIT to not deplete the emergency pool.

---
 drivers/dma/mediatek/mtk-uart-apdma.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/dma/mediatek/mtk-uart-apdma.c +1 -1
index a09ab2dd3b46..375e7e647df6 100644
@@ -349,7 +349,7 @@ static struct dma_async_tx_descriptor *mtk_uart_apdma_prep_slave_sg
 		return NULL;
 
 	/* Now allocate and setup the descriptor */
-	d = kzalloc(sizeof(*d), GFP_ATOMIC);
+	d = kzalloc(sizeof(*d), GFP_NOWAIT);
 	if (!d)
 		return NULL;
 
Signature · 1 line
2.30.2
author maintainer
#32 lore raw

PATCH 31/101 dmaengine: mediatek: do not issue a new desc if one is still current

In reply to Sasha Levin

From: Guillaume Ranquet <[email protected]>

[ Upstream commit 2537b40b0a4f61d2c83900744fe89b09076be9c6 ]

Avoid issuing a new desc if one is still being processed as this can lead to some desc never being marked as completed.

---
 drivers/dma/mediatek/mtk-uart-apdma.c | 20 ++++++++++++--------
 1 file changed, 12 insertions(+), 8 deletions(-)
drivers/dma/mediatek/mtk-uart-apdma.c +12 -8
index e38b67fc0c0c..a09ab2dd3b46 100644
@@ -204,14 +204,9 @@ static void mtk_uart_apdma_start_rx(struct mtk_chan *c)
 
 static void mtk_uart_apdma_tx_handler(struct mtk_chan *c)
 {
-	struct mtk_uart_apdma_desc *d = c->desc;
-
 	mtk_uart_apdma_write(c, VFF_INT_FLAG, VFF_TX_INT_CLR_B);
 	mtk_uart_apdma_write(c, VFF_INT_EN, VFF_INT_EN_CLR_B);
 	mtk_uart_apdma_write(c, VFF_EN, VFF_EN_CLR_B);
-
-	list_del(&d->vd.node);
-	vchan_cookie_complete(&d->vd);
 }
 
 static void mtk_uart_apdma_rx_handler(struct mtk_chan *c)
@@ -242,9 +237,17 @@ static void mtk_uart_apdma_rx_handler(struct mtk_chan *c)
 
 	c->rx_status = d->avail_len - cnt;
 	mtk_uart_apdma_write(c, VFF_RPT, wg);
+}
 
-	list_del(&d->vd.node);
-	vchan_cookie_complete(&d->vd);
+static void mtk_uart_apdma_chan_complete_handler(struct mtk_chan *c)
+{
+	struct mtk_uart_apdma_desc *d = c->desc;
+
+	if (d) {
+		list_del(&d->vd.node);
+		vchan_cookie_complete(&d->vd);
+		c->desc = NULL;
+	}
 }
 
 static irqreturn_t mtk_uart_apdma_irq_handler(int irq, void *dev_id)
@@ -258,6 +261,7 @@ static irqreturn_t mtk_uart_apdma_irq_handler(int irq, void *dev_id)
 		mtk_uart_apdma_rx_handler(c);
 	else if (c->dir == DMA_MEM_TO_DEV)
 		mtk_uart_apdma_tx_handler(c);
+	mtk_uart_apdma_chan_complete_handler(c);
 	spin_unlock_irqrestore(&c->vc.lock, flags);
 
 	return IRQ_HANDLED;
@@ -363,7 +367,7 @@ static void mtk_uart_apdma_issue_pending(struct dma_chan *chan)
 	unsigned long flags;
 
 	spin_lock_irqsave(&c->vc.lock, flags);
-	if (vchan_issue_pending(&c->vc)) {
+	if (vchan_issue_pending(&c->vc) && !c->desc) {
 		vd = vchan_next_desc(&c->vc);
 		c->desc = to_mtk_uart_apdma_desc(&vd->tx);
 
Signature · 1 line
2.30.2
author maintainer
#35 lore raw

PATCH 34/101 mac80211: drop multicast fragments

In reply to Sasha Levin

From: Johannes Berg <[email protected]>

[ Upstream commit a9799541ca34652d9996e45f80e8e03144c12949 ]

These are not permitted by the spec, just drop them.

---
 net/mac80211/rx.c | 9 +++------
 1 file changed, 3 insertions(+), 6 deletions(-)
net/mac80211/rx.c +3 -6
index ef8ff0bc66f1..38b5695c2a0c 100644
@@ -2250,17 +2250,15 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx)
 	sc = le16_to_cpu(hdr->seq_ctrl);
 	frag = sc & IEEE80211_SCTL_FRAG;
 
-	if (is_multicast_ether_addr(hdr->addr1)) {
-		I802_DEBUG_INC(rx->local->dot11MulticastReceivedFrameCount);
-		goto out_no_led;
-	}
-
 	if (rx->sta)
 		cache = &rx->sta->frags;
 
 	if (likely(!ieee80211_has_morefrags(fc) && frag == 0))
 		goto out;
 
+	if (is_multicast_ether_addr(hdr->addr1))
+		return RX_DROP_MONITOR;
+
 	I802_DEBUG_INC(rx->local->rx_handlers_fragments);
 
 	if (skb_linearize(rx->skb))
@@ -2386,7 +2384,6 @@ ieee80211_rx_h_defragment(struct ieee80211_rx_data *rx)
 
  out:
 	ieee80211_led_rx(rx->local);
- out_no_led:
 	if (rx->sta)
 		rx->sta->rx_stats.packets++;
 	return RX_CONTINUE;
Signature · 1 line
2.30.2
author maintainer
#36 lore raw

PATCH 35/101 net: ethtool: clear heap allocations for ethtool function

In reply to Sasha Levin

From: Austin Kim <[email protected]>

[ Upstream commit 80ec82e3d2c1fab42eeb730aaa7985494a963d3f ]

Several ethtool functions leave heap uncleared (potentially) by drivers. This will leave the unused portion of heap unchanged and might copy the full contents back to userspace.

---
 net/ethtool/ioctl.c | 10 +++++-----
 1 file changed, 5 insertions(+), 5 deletions(-)
net/ethtool/ioctl.c +5 -5
index 2917af3f5ac1..68ff19af195c 100644
@@ -1421,7 +1421,7 @@ static int ethtool_get_any_eeprom(struct net_device *dev, void __user *useraddr,
 	if (eeprom.offset + eeprom.len > total_len)
 		return -EINVAL;
 
-	data = kmalloc(PAGE_SIZE, GFP_USER);
+	data = kzalloc(PAGE_SIZE, GFP_USER);
 	if (!data)
 		return -ENOMEM;
 
@@ -1486,7 +1486,7 @@ static int ethtool_set_eeprom(struct net_device *dev, void __user *useraddr)
 	if (eeprom.offset + eeprom.len > ops->get_eeprom_len(dev))
 		return -EINVAL;
 
-	data = kmalloc(PAGE_SIZE, GFP_USER);
+	data = kzalloc(PAGE_SIZE, GFP_USER);
 	if (!data)
 		return -ENOMEM;
 
@@ -1765,7 +1765,7 @@ static int ethtool_self_test(struct net_device *dev, char __user *useraddr)
 		return -EFAULT;
 
 	test.len = test_len;
-	data = kmalloc_array(test_len, sizeof(u64), GFP_USER);
+	data = kcalloc(test_len, sizeof(u64), GFP_USER);
 	if (!data)
 		return -ENOMEM;
 
@@ -2281,7 +2281,7 @@ static int ethtool_get_tunable(struct net_device *dev, void __user *useraddr)
 	ret = ethtool_tunable_valid(&tuna);
 	if (ret)
 		return ret;
-	data = kmalloc(tuna.len, GFP_USER);
+	data = kzalloc(tuna.len, GFP_USER);
 	if (!data)
 		return -ENOMEM;
 	ret = ops->get_tunable(dev, &tuna, data);
@@ -2473,7 +2473,7 @@ static int get_phy_tunable(struct net_device *dev, void __user *useraddr)
 	ret = ethtool_phy_tunable_valid(&tuna);
 	if (ret)
 		return ret;
-	data = kmalloc(tuna.len, GFP_USER);
+	data = kzalloc(tuna.len, GFP_USER);
 	if (!data)
 		return -ENOMEM;
 	if (phy_drv_tunable) {
Signature · 1 line
2.30.2
author maintainer
#39 lore raw

PATCH 38/101 net: annotate data race in sock_error()

In reply to Sasha Levin

From: Eric Dumazet <[email protected]>

[ Upstream commit f13ef10059ccf5f4ed201cd050176df62ec25bb8 ]

sock_error() is known to be racy. The code avoids an atomic operation is sk_err is zero, and this field could be changed under us, this is fine.

Sysbot reported:

BUG: KCSAN: data-race in sock_alloc_send_pskb / unix_release_sock

write to 0xffff888131855630 of 4 bytes by task 9365 on cpu 1:

 unix_release_sock+0x2e9/0x6e0 net/unix/af_unix.c:550
 unix_release+0x2f/0x50 net/unix/af_unix.c:859
 __sock_release net/socket.c:599 [inline]
 sock_close+0x6c/0x150 net/socket.c:1258
 __fput+0x25b/0x4e0 fs/file_table.c:280
 ____fput+0x11/0x20 fs/file_table.c:313
 task_work_run+0xae/0x130 kernel/task_work.c:164
 tracehook_notify_resume include/linux/tracehook.h:189 [inline]
 exit_to_user_mode_loop kernel/entry/common.c:174 [inline]
 exit_to_user_mode_prepare+0x156/0x190 kernel/entry/common.c:208
 __syscall_exit_to_user_mode_work kernel/entry/common.c:290 [inline]
 syscall_exit_to_user_mode+0x20/0x40 kernel/entry/common.c:301
 do_syscall_64+0x56/0x90 arch/x86/entry/common.c:57
 entry_SYSCALL_64_after_hwframe+0x44/0xae

read to 0xffff888131855630 of 4 bytes by task 9385 on cpu 0:

 sock_error include/net/sock.h:2269 [inline]
 sock_alloc_send_pskb+0xe4/0x4e0 net/core/sock.c:2336
 unix_dgram_sendmsg+0x478/0x1610 net/unix/af_unix.c:1671
 unix_seqpacket_sendmsg+0xc2/0x100 net/unix/af_unix.c:2055
 sock_sendmsg_nosec net/socket.c:654 [inline]
 sock_sendmsg net/socket.c:674 [inline]
 ____sys_sendmsg+0x360/0x4d0 net/socket.c:2350
 __sys_sendmsg_sock+0x25/0x30 net/socket.c:2416
 io_sendmsg fs/io_uring.c:4367 [inline]
 io_issue_sqe+0x231a/0x6750 fs/io_uring.c:6135
 __io_queue_sqe+0xe9/0x360 fs/io_uring.c:6414
 __io_req_task_submit fs/io_uring.c:2039 [inline]
 io_async_task_func+0x312/0x590 fs/io_uring.c:5074
 __tctx_task_work fs/io_uring.c:1910 [inline]
 tctx_task_work+0x1d4/0x3d0 fs/io_uring.c:1924
 task_work_run+0xae/0x130 kernel/task_work.c:164
 tracehook_notify_signal include/linux/tracehook.h:212 [inline]
 handle_signal_work kernel/entry/common.c:145 [inline]
 exit_to_user_mode_loop kernel/entry/common.c:171 [inline]
 exit_to_user_mode_prepare+0xf8/0x190 kernel/entry/common.c:208
 __syscall_exit_to_user_mode_work kernel/entry/common.c:290 [inline]
 syscall_exit_to_user_mode+0x20/0x40 kernel/entry/common.c:301
 do_syscall_64+0x56/0x90 arch/x86/entry/common.c:57
 entry_SYSCALL_64_after_hwframe+0x44/0xae

value changed: 0x00000000 -> 0x00000068

Reported by Kernel Concurrency Sanitizer on:

CPU: 0 PID: 9385 Comm: syz-executor.3 Not tainted 5.13.0-rc4-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
---
 include/net/sock.h | 7 ++++++-
 1 file changed, 6 insertions(+), 1 deletion(-)
include/net/sock.h +6 -1
index f68184b8c0aa..9199ab7ab2da 100644
@@ -2232,8 +2232,13 @@ struct sk_buff *sock_dequeue_err_skb(struct sock *sk);
 static inline int sock_error(struct sock *sk)
 {
 	int err;
-	if (likely(!sk->sk_err))
+
+	/* Avoid an atomic operation for the common case.
+	 * This is racy since another cpu/thread can change sk_err under us.
+	 */
+	if (likely(data_race(!sk->sk_err)))
 		return 0;
+
 	err = xchg(&sk->sk_err, 0);
 	return -err;
 }
Signature · 1 line
2.30.2
author maintainer
#37 lore raw

PATCH 36/101 inet: annotate data race in inet_send_prepare() and inet_dgram_connect()

In reply to Sasha Levin

From: Eric Dumazet <[email protected]>

[ Upstream commit dcd01eeac14486b56a790f5cce9b823440ba5b34 ]

Both functions are known to be racy when reading inet_num as we do not want to grab locks for the common case the socket has been bound already. The race is resolved in inet_autobind() by reading again inet_num under the socket lock.

syzbot reported:

BUG: KCSAN: data-race in inet_send_prepare / udp_lib_get_port

write to 0xffff88812cba150e of 2 bytes by task 24135 on cpu 0:

 udp_lib_get_port+0x4b2/0xe20 net/ipv4/udp.c:308
 udp_v6_get_port+0x5e/0x70 net/ipv6/udp.c:89
 inet_autobind net/ipv4/af_inet.c:183 [inline]
 inet_send_prepare+0xd0/0x210 net/ipv4/af_inet.c:807
 inet6_sendmsg+0x29/0x80 net/ipv6/af_inet6.c:639
 sock_sendmsg_nosec net/socket.c:654 [inline]
 sock_sendmsg net/socket.c:674 [inline]
 ____sys_sendmsg+0x360/0x4d0 net/socket.c:2350
 ___sys_sendmsg net/socket.c:2404 [inline]
 __sys_sendmmsg+0x315/0x4b0 net/socket.c:2490
 __do_sys_sendmmsg net/socket.c:2519 [inline]
 __se_sys_sendmmsg net/socket.c:2516 [inline]
 __x64_sys_sendmmsg+0x53/0x60 net/socket.c:2516
 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47
 entry_SYSCALL_64_after_hwframe+0x44/0xae

read to 0xffff88812cba150e of 2 bytes by task 24132 on cpu 1:

 inet_send_prepare+0x21/0x210 net/ipv4/af_inet.c:806
 inet6_sendmsg+0x29/0x80 net/ipv6/af_inet6.c:639
 sock_sendmsg_nosec net/socket.c:654 [inline]
 sock_sendmsg net/socket.c:674 [inline]
 ____sys_sendmsg+0x360/0x4d0 net/socket.c:2350
 ___sys_sendmsg net/socket.c:2404 [inline]
 __sys_sendmmsg+0x315/0x4b0 net/socket.c:2490
 __do_sys_sendmmsg net/socket.c:2519 [inline]
 __se_sys_sendmmsg net/socket.c:2516 [inline]
 __x64_sys_sendmmsg+0x53/0x60 net/socket.c:2516
 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47
 entry_SYSCALL_64_after_hwframe+0x44/0xae

value changed: 0x0000 -> 0x9db4

Reported by Kernel Concurrency Sanitizer on:

CPU: 1 PID: 24132 Comm: syz-executor.2 Not tainted 5.13.0-rc4-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
---
 net/ipv4/af_inet.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
net/ipv4/af_inet.c +2 -2
index b7260c8cef2e..8267349afe23 100644
@@ -572,7 +572,7 @@ int inet_dgram_connect(struct socket *sock, struct sockaddr *uaddr,
 			return err;
 	}
 
-	if (!inet_sk(sk)->inet_num && inet_autobind(sk))
+	if (data_race(!inet_sk(sk)->inet_num) && inet_autobind(sk))
 		return -EAGAIN;
 	return sk->sk_prot->connect(sk, uaddr, addr_len);
 }
@@ -799,7 +799,7 @@ int inet_send_prepare(struct sock *sk)
 	sock_rps_record_flow(sk);
 
 	/* We may need to bind the socket. */
-	if (!inet_sk(sk)->inet_num && !sk->sk_prot->no_autobind &&
+	if (data_race(!inet_sk(sk)->inet_num) && !sk->sk_prot->no_autobind &&
 	    inet_autobind(sk))
 		return -EAGAIN;
 
Signature · 1 line
2.30.2
author maintainer
#38 lore raw

PATCH 37/101 ping: Check return value of function 'ping_queue_rcv_skb'

In reply to Sasha Levin

From: Zheng Yongjun <[email protected]>

[ Upstream commit 9d44fa3e50cc91691896934d106c86e4027e61ca ]

Function 'ping_queue_rcv_skb' not always return success, which will also return fail. If not check the wrong return value of it, lead to function ping_rcv return success.

---
 net/ipv4/ping.c | 12 +++++++-----
 1 file changed, 7 insertions(+), 5 deletions(-)
net/ipv4/ping.c +7 -5
index 248856b301c4..8ce8b7300b9d 100644
@@ -952,6 +952,7 @@ bool ping_rcv(struct sk_buff *skb)
 	struct sock *sk;
 	struct net *net = dev_net(skb->dev);
 	struct icmphdr *icmph = icmp_hdr(skb);
+	bool rc = false;
 
 	/* We assume the packet has already been checked by icmp_rcv */
 
@@ -966,14 +967,15 @@ bool ping_rcv(struct sk_buff *skb)
 		struct sk_buff *skb2 = skb_clone(skb, GFP_ATOMIC);
 
 		pr_debug("rcv on socket %p\n", sk);
-		if (skb2)
-			ping_queue_rcv_skb(sk, skb2);
+		if (skb2 && !ping_queue_rcv_skb(sk, skb2))
+			rc = true;
 		sock_put(sk);
-		return true;
 	}
-	pr_debug("no socket, dropping\n");
 
-	return false;
+	if (!rc)
+		pr_debug("no socket, dropping\n");
+
+	return rc;
 }
 EXPORT_SYMBOL_GPL(ping_rcv);
 
Signature · 1 line
2.30.2
author maintainer
#40 lore raw

PATCH 39/101 inet: annotate date races around sk->sk_txhash

In reply to Sasha Levin

From: Eric Dumazet <[email protected]>

[ Upstream commit b71eaed8c04f72a919a9c44e83e4ee254e69e7f3 ]

UDP sendmsg() path can be lockless, it is possible for another thread to re-connect an change sk->sk_txhash under us.

There is no serious impact, but we can use READ_ONCE()/WRITE_ONCE() pair to document the race.

BUG: KCSAN: data-race in __ip4_datagram_connect / skb_set_owner_w

write to 0xffff88813397920c of 4 bytes by task 30997 on cpu 1:

 sk_set_txhash include/net/sock.h:1937 [inline]
 __ip4_datagram_connect+0x69e/0x710 net/ipv4/datagram.c:75
 __ip6_datagram_connect+0x551/0x840 net/ipv6/datagram.c:189
 ip6_datagram_connect+0x2a/0x40 net/ipv6/datagram.c:272
 inet_dgram_connect+0xfd/0x180 net/ipv4/af_inet.c:580
 __sys_connect_file net/socket.c:1837 [inline]
 __sys_connect+0x245/0x280 net/socket.c:1854
 __do_sys_connect net/socket.c:1864 [inline]
 __se_sys_connect net/socket.c:1861 [inline]
 __x64_sys_connect+0x3d/0x50 net/socket.c:1861
 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47
 entry_SYSCALL_64_after_hwframe+0x44/0xae

read to 0xffff88813397920c of 4 bytes by task 31039 on cpu 0:

 skb_set_hash_from_sk include/net/sock.h:2211 [inline]
 skb_set_owner_w+0x118/0x220 net/core/sock.c:2101
 sock_alloc_send_pskb+0x452/0x4e0 net/core/sock.c:2359
 sock_alloc_send_skb+0x2d/0x40 net/core/sock.c:2373
 __ip6_append_data+0x1743/0x21a0 net/ipv6/ip6_output.c:1621
 ip6_make_skb+0x258/0x420 net/ipv6/ip6_output.c:1983
 udpv6_sendmsg+0x160a/0x16b0 net/ipv6/udp.c:1527
 inet6_sendmsg+0x5f/0x80 net/ipv6/af_inet6.c:642
 sock_sendmsg_nosec net/socket.c:654 [inline]
 sock_sendmsg net/socket.c:674 [inline]
 ____sys_sendmsg+0x360/0x4d0 net/socket.c:2350
 ___sys_sendmsg net/socket.c:2404 [inline]
 __sys_sendmmsg+0x315/0x4b0 net/socket.c:2490
 __do_sys_sendmmsg net/socket.c:2519 [inline]
 __se_sys_sendmmsg net/socket.c:2516 [inline]
 __x64_sys_sendmmsg+0x53/0x60 net/socket.c:2516
 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47
 entry_SYSCALL_64_after_hwframe+0x44/0xae

value changed: 0xbca3c43d -> 0xfdb309e0

Reported by Kernel Concurrency Sanitizer on:

CPU: 0 PID: 31039 Comm: syz-executor.2 Not tainted 5.13.0-rc3-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
---
 include/net/sock.h | 10 +++++++---
 1 file changed, 7 insertions(+), 3 deletions(-)
include/net/sock.h +7 -3
index 9199ab7ab2da..3c7addf95150 100644
@@ -1900,7 +1900,8 @@ static inline u32 net_tx_rndhash(void)
 
 static inline void sk_set_txhash(struct sock *sk)
 {
-	sk->sk_txhash = net_tx_rndhash();
+	/* This pairs with READ_ONCE() in skb_set_hash_from_sk() */
+	WRITE_ONCE(sk->sk_txhash, net_tx_rndhash());
 }
 
 static inline bool sk_rethink_txhash(struct sock *sk)
@@ -2172,9 +2173,12 @@ static inline void sock_poll_wait(struct file *filp, struct socket *sock,
 
 static inline void skb_set_hash_from_sk(struct sk_buff *skb, struct sock *sk)
 {
-	if (sk->sk_txhash) {
+	/* This pairs with WRITE_ONCE() in sk_set_txhash() */
+	u32 txhash = READ_ONCE(sk->sk_txhash);
+
+	if (txhash) {
 		skb->l4_hash = 1;
-		skb->hash = sk->sk_txhash;
+		skb->hash = txhash;
 	}
 }
 
Signature · 1 line
2.30.2
author maintainer
#41 lore raw

PATCH 40/101 net/packet: annotate data race in packet_sendmsg()

In reply to Sasha Levin

From: Eric Dumazet <[email protected]>

[ Upstream commit d1b5bee4c8be01585033be9b3a8878789285285f ]

There is a known race in packet_sendmsg(), addressed in commit 32d3182cd2cd ("net/packet: fix race in tpacket_snd()")

Now we have data_race(), we can use it to avoid a future KCSAN warning, as syzbot loves stressing af_packet sockets :)

---
 net/packet/af_packet.c | 9 ++++++---
 1 file changed, 6 insertions(+), 3 deletions(-)
net/packet/af_packet.c +6 -3
index ddb68aa836f7..7892d7074847 100644
@@ -3033,10 +3033,13 @@ static int packet_sendmsg(struct socket *sock, struct msghdr *msg, size_t len)
 	struct sock *sk = sock->sk;
 	struct packet_sock *po = pkt_sk(sk);
 
-	if (po->tx_ring.pg_vec)
+	/* Reading tx_ring.pg_vec without holding pg_vec_lock is racy.
+	 * tpacket_snd() will redo the check safely.
+	 */
+	if (data_race(po->tx_ring.pg_vec))
 		return tpacket_snd(po, msg);
-	else
-		return packet_snd(sock, msg, len);
+
+	return packet_snd(sock, msg, len);
 }
 
 /*
Signature · 1 line
2.30.2
author maintainer
#42 lore raw

PATCH 41/101 net: phy: dp83867: perform soft reset and retain established link

In reply to Sasha Levin

From: Praneeth Bajjuri <[email protected]>

[ Upstream commit da9ef50f545f86ffe6ff786174d26500c4db737a ]

Current logic is performing hard reset and causing the programmed registers to be wiped out.

as per datasheet: https://www.ti.com/lit/ds/symlink/dp83867cr.pdf
8.6.26 Control Register (CTRL)

do SW_RESTART to perform a reset not including the registers, If performed when link is already present,
it will drop the link and trigger re-auto negotiation.

---
 drivers/net/phy/dp83867.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)
drivers/net/phy/dp83867.c +1 -5
index 69d3eacc2b96..c716074fdef0 100644
@@ -792,16 +792,12 @@ static int dp83867_phy_reset(struct phy_device *phydev)
 {
 	int err;
 
-	err = phy_write(phydev, DP83867_CTRL, DP83867_SW_RESET);
+	err = phy_write(phydev, DP83867_CTRL, DP83867_SW_RESTART);
 	if (err < 0)
 		return err;
 
 	usleep_range(10, 20);
 
-	/* After reset FORCE_LINK_GOOD bit is set. Although the
-	 * default value should be unset. Disable FORCE_LINK_GOOD
-	 * for the phy to work properly.
-	 */
 	return phy_modify(phydev, MII_DP83867_PHYCTRL,
 			 DP83867_PHYCR_FORCE_LINK_GOOD, 0);
 }
Signature · 1 line
2.30.2
author maintainer
#43 lore raw

PATCH 42/101 riscv32: Use medany C model for modules

In reply to Sasha Levin

From: Khem Raj <[email protected]>

[ Upstream commit 5d2388dbf84adebeb6d9742164be8d32728e4269 ]

When CONFIG_CMODEL_MEDLOW is used it ends up generating riscv_hi20_rela relocations in modules which are not resolved during runtime and following errors would be seen

[    4.802714] virtio_input: target 00000000c1539090 can not be addressed by the 32-bit offset from PC = 39148b7b
[    4.854800] virtio_input: target 00000000c1539090 can not be addressed by the 32-bit offset from PC = 9774456d
---
 arch/riscv/Makefile | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
arch/riscv/Makefile +1 -1
index e241e0e85ac8..226c366072da 100644
@@ -14,7 +14,7 @@ ifeq ($(CONFIG_DYNAMIC_FTRACE),y)
 	LDFLAGS_vmlinux := --no-relax
 endif
 
-ifeq ($(CONFIG_64BIT)$(CONFIG_CMODEL_MEDLOW),yy)
+ifeq ($(CONFIG_CMODEL_MEDLOW),y)
 KBUILD_CFLAGS_MODULE += -mcmodel=medany
 endif
 
Signature · 1 line
2.30.2
author maintainer
#44 lore raw

PATCH 43/101 net: caif: fix memory leak in ldisc_open

In reply to Sasha Levin

From: Pavel Skripkin <[email protected]>

[ Upstream commit 58af3d3d54e87bfc1f936e16c04ade3369d34011 ]

Syzbot reported memory leak in tty_init_dev().
The problem was in unputted tty in ldisc_open()

static int ldisc_open(struct tty_struct *tty)
{
...

	ser->tty = tty_kref_get(tty);
...
	result = register_netdevice(dev);
	if (result) {
		rtnl_unlock();
		free_netdev(dev);
		return -ENODEV;
	}

...
}

Ser pointer is netdev private_data, so after free_netdev() this pointer goes away with unputted tty reference. So, fix it by adding tty_kref_put() before freeing netdev.

---
 drivers/net/caif/caif_serial.c | 1 +
 1 file changed, 1 insertion(+)
drivers/net/caif/caif_serial.c +1 -0
index d025ea434933..39fbd0be179c 100644
@@ -351,6 +351,7 @@ static int ldisc_open(struct tty_struct *tty)
 	rtnl_lock();
 	result = register_netdevice(dev);
 	if (result) {
+		tty_kref_put(tty);
 		rtnl_unlock();
 		free_netdev(dev);
 		return -ENODEV;
Signature · 1 line
2.30.2
author maintainer
#45 lore raw

PATCH 44/101 net/packet: annotate accesses to po->bind

In reply to Sasha Levin

From: Eric Dumazet <[email protected]>

[ Upstream commit c7d2ef5dd4b03ed0ee1d13bc0c55f9cf62d49bd6 ]

tpacket_snd(), packet_snd(), packet_getname() and packet_seq_show() can read po->num without holding a lock. This means other threads can change po->num at the same time.

KCSAN complained about this known fact [1] Add READ_ONCE()/WRITE_ONCE() to address the issue.

[1] BUG: KCSAN: data-race in packet_do_bind / packet_sendmsg

write to 0xffff888131a0dcc0 of 2 bytes by task 24714 on cpu 0:

 packet_do_bind+0x3ab/0x7e0 net/packet/af_packet.c:3181
 packet_bind+0xc3/0xd0 net/packet/af_packet.c:3255
 __sys_bind+0x200/0x290 net/socket.c:1637
 __do_sys_bind net/socket.c:1648 [inline]
 __se_sys_bind net/socket.c:1646 [inline]
 __x64_sys_bind+0x3d/0x50 net/socket.c:1646
 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47
 entry_SYSCALL_64_after_hwframe+0x44/0xae

read to 0xffff888131a0dcc0 of 2 bytes by task 24719 on cpu 1:

 packet_snd net/packet/af_packet.c:2899 [inline]
 packet_sendmsg+0x317/0x3570 net/packet/af_packet.c:3040
 sock_sendmsg_nosec net/socket.c:654 [inline]
 sock_sendmsg net/socket.c:674 [inline]
 ____sys_sendmsg+0x360/0x4d0 net/socket.c:2350
 ___sys_sendmsg net/socket.c:2404 [inline]
 __sys_sendmsg+0x1ed/0x270 net/socket.c:2433
 __do_sys_sendmsg net/socket.c:2442 [inline]
 __se_sys_sendmsg net/socket.c:2440 [inline]
 __x64_sys_sendmsg+0x42/0x50 net/socket.c:2440
 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47
 entry_SYSCALL_64_after_hwframe+0x44/0xae

value changed: 0x0000 -> 0x1200

Reported by Kernel Concurrency Sanitizer on:

CPU: 1 PID: 24719 Comm: syz-executor.5 Not tainted 5.13.0-rc4-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
---
 net/packet/af_packet.c | 16 ++++++++--------
 1 file changed, 8 insertions(+), 8 deletions(-)
net/packet/af_packet.c +8 -8
index 7892d7074847..c4eb26f0f1a7 100644
@@ -2682,7 +2682,7 @@ static int tpacket_snd(struct packet_sock *po, struct msghdr *msg)
 	}
 	if (likely(saddr == NULL)) {
 		dev	= packet_cached_dev_get(po);
-		proto	= po->num;
+		proto	= READ_ONCE(po->num);
 	} else {
 		err = -EINVAL;
 		if (msg->msg_namelen < sizeof(struct sockaddr_ll))
@@ -2895,7 +2895,7 @@ static int packet_snd(struct socket *sock, struct msghdr *msg, size_t len)
 
 	if (likely(saddr == NULL)) {
 		dev	= packet_cached_dev_get(po);
-		proto	= po->num;
+		proto	= READ_ONCE(po->num);
 	} else {
 		err = -EINVAL;
 		if (msg->msg_namelen < sizeof(struct sockaddr_ll))
@@ -3170,7 +3170,7 @@ static int packet_do_bind(struct sock *sk, const char *name, int ifindex,
 			/* prevents packet_notifier() from calling
 			 * register_prot_hook()
 			 */
-			po->num = 0;
+			WRITE_ONCE(po->num, 0);
 			__unregister_prot_hook(sk, true);
 			rcu_read_lock();
 			dev_curr = po->prot_hook.dev;
@@ -3180,7 +3180,7 @@ static int packet_do_bind(struct sock *sk, const char *name, int ifindex,
 		}
 
 		BUG_ON(po->running);
-		po->num = proto;
+		WRITE_ONCE(po->num, proto);
 		po->prot_hook.type = proto;
 
 		if (unlikely(unlisted)) {
@@ -3525,7 +3525,7 @@ static int packet_getname(struct socket *sock, struct sockaddr *uaddr,
 
 	sll->sll_family = AF_PACKET;
 	sll->sll_ifindex = po->ifindex;
-	sll->sll_protocol = po->num;
+	sll->sll_protocol = READ_ONCE(po->num);
 	sll->sll_pkttype = 0;
 	rcu_read_lock();
 	dev = dev_get_by_index_rcu(sock_net(sk), po->ifindex);
@@ -4413,7 +4413,7 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u,
 	was_running = po->running;
 	num = po->num;
 	if (was_running) {
-		po->num = 0;
+		WRITE_ONCE(po->num, 0);
 		__unregister_prot_hook(sk, false);
 	}
 	spin_unlock(&po->bind_lock);
@@ -4448,7 +4448,7 @@ static int packet_set_ring(struct sock *sk, union tpacket_req_u *req_u,
 
 	spin_lock(&po->bind_lock);
 	if (was_running) {
-		po->num = num;
+		WRITE_ONCE(po->num, num);
 		register_prot_hook(sk);
 	}
 	spin_unlock(&po->bind_lock);
@@ -4616,7 +4616,7 @@ static int packet_seq_show(struct seq_file *seq, void *v)
 			   s,
 			   refcount_read(&s->sk_refcnt),
 			   s->sk_type,
-			   ntohs(po->num),
+			   ntohs(READ_ONCE(po->num)),
 			   po->ifindex,
 			   po->running,
 			   atomic_read(&s->sk_rmem_alloc),
Signature · 1 line
2.30.2
author maintainer
#49 lore raw

PATCH 48/101 r8169: Avoid memcpy() over-reading of ETH_SS_STATS

In reply to Sasha Levin

From: Kees Cook <[email protected]>

[ Upstream commit da5ac772cfe2a03058b0accfac03fad60c46c24d ]

In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally reading across neighboring array fields.

The memcpy() is copying the entire structure, not just the first array. Adjust the source argument so the compiler can do appropriate bounds checking.

---
 drivers/net/ethernet/realtek/r8169_main.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/realtek/r8169_main.c +1 -1
index 3bb36f4a984e..a6bf80b52967 100644
@@ -1673,7 +1673,7 @@ static void rtl8169_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 {
 	switch(stringset) {
 	case ETH_SS_STATS:
-		memcpy(data, *rtl8169_gstrings, sizeof(rtl8169_gstrings));
+		memcpy(data, rtl8169_gstrings, sizeof(rtl8169_gstrings));
 		break;
 	}
 }
Signature · 1 line
2.30.2
author maintainer
#47 lore raw

PATCH 46/101 r8152: Avoid memcpy() over-reading of ETH_SS_STATS

In reply to Sasha Levin

From: Kees Cook <[email protected]>

[ Upstream commit 99718abdc00e86e4f286dd836408e2834886c16e ]

In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally reading across neighboring array fields.

The memcpy() is copying the entire structure, not just the first array. Adjust the source argument so the compiler can do appropriate bounds checking.

---
 drivers/net/usb/r8152.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/usb/r8152.c +1 -1
index f5010f8ac1ec..95e27fb7d2c1 100644
@@ -6054,7 +6054,7 @@ static void rtl8152_get_strings(struct net_device *dev, u32 stringset, u8 *data)
 {
 	switch (stringset) {
 	case ETH_SS_STATS:
-		memcpy(data, *rtl8152_gstrings, sizeof(rtl8152_gstrings));
+		memcpy(data, rtl8152_gstrings, sizeof(rtl8152_gstrings));
 		break;
 	}
 }
Signature · 1 line
2.30.2
author maintainer
#48 lore raw

PATCH 47/101 sh_eth: Avoid memcpy() over-reading of ETH_SS_STATS

In reply to Sasha Levin

From: Kees Cook <[email protected]>

[ Upstream commit 224004fbb033600715dbd626bceec10bfd9c58bc ]

In preparation for FORTIFY_SOURCE performing compile-time and run-time field bounds checking for memcpy(), memmove(), and memset(), avoid intentionally reading across neighboring array fields.

The memcpy() is copying the entire structure, not just the first array. Adjust the source argument so the compiler can do appropriate bounds checking.

---
 drivers/net/ethernet/renesas/sh_eth.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/ethernet/renesas/sh_eth.c +1 -1
index 6d84266c03ca..5cab2d3c0023 100644
@@ -2287,7 +2287,7 @@ static void sh_eth_get_strings(struct net_device *ndev, u32 stringset, u8 *data)
 {
 	switch (stringset) {
 	case ETH_SS_STATS:
-		memcpy(data, *sh_eth_gstrings_stats,
+		memcpy(data, sh_eth_gstrings_stats,
 		       sizeof(sh_eth_gstrings_stats));
 		break;
 	}
Signature · 1 line
2.30.2
author maintainer
#46 lore raw

PATCH 45/101 net/packet: annotate accesses to po->ifindex

In reply to Sasha Levin

From: Eric Dumazet <[email protected]>

[ Upstream commit e032f7c9c7cefffcfb79b9fc16c53011d2d9d11f ]

Like prior patch, we need to annotate lockless accesses to po->ifindex For instance, packet_getname() is reading po->ifindex (twice) while another thread is able to change po->ifindex.

KCSAN reported:

BUG: KCSAN: data-race in packet_do_bind / packet_getname

write to 0xffff888143ce3cbc of 4 bytes by task 25573 on cpu 1:

 packet_do_bind+0x420/0x7e0 net/packet/af_packet.c:3191
 packet_bind+0xc3/0xd0 net/packet/af_packet.c:3255
 __sys_bind+0x200/0x290 net/socket.c:1637
 __do_sys_bind net/socket.c:1648 [inline]
 __se_sys_bind net/socket.c:1646 [inline]
 __x64_sys_bind+0x3d/0x50 net/socket.c:1646
 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47
 entry_SYSCALL_64_after_hwframe+0x44/0xae

read to 0xffff888143ce3cbc of 4 bytes by task 25578 on cpu 0:

 packet_getname+0x5b/0x1a0 net/packet/af_packet.c:3525
 __sys_getsockname+0x10e/0x1a0 net/socket.c:1887
 __do_sys_getsockname net/socket.c:1902 [inline]
 __se_sys_getsockname net/socket.c:1899 [inline]
 __x64_sys_getsockname+0x3e/0x50 net/socket.c:1899
 do_syscall_64+0x4a/0x90 arch/x86/entry/common.c:47
 entry_SYSCALL_64_after_hwframe+0x44/0xae

value changed: 0x00000000 -> 0x00000001

Reported by Kernel Concurrency Sanitizer on:

CPU: 0 PID: 25578 Comm: syz-executor.5 Not tainted 5.13.0-rc6-syzkaller #0
Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 01/01/2011
---
 net/packet/af_packet.c | 16 +++++++++-------
 1 file changed, 9 insertions(+), 7 deletions(-)
net/packet/af_packet.c +9 -7
index c4eb26f0f1a7..08144559eed5 100644
@@ -3186,11 +3186,11 @@ static int packet_do_bind(struct sock *sk, const char *name, int ifindex,
 		if (unlikely(unlisted)) {
 			dev_put(dev);
 			po->prot_hook.dev = NULL;
-			po->ifindex = -1;
+			WRITE_ONCE(po->ifindex, -1);
 			packet_cached_dev_reset(po);
 		} else {
 			po->prot_hook.dev = dev;
-			po->ifindex = dev ? dev->ifindex : 0;
+			WRITE_ONCE(po->ifindex, dev ? dev->ifindex : 0);
 			packet_cached_dev_assign(po, dev);
 		}
 	}
@@ -3504,7 +3504,7 @@ static int packet_getname_spkt(struct socket *sock, struct sockaddr *uaddr,
 	uaddr->sa_family = AF_PACKET;
 	memset(uaddr->sa_data, 0, sizeof(uaddr->sa_data));
 	rcu_read_lock();
-	dev = dev_get_by_index_rcu(sock_net(sk), pkt_sk(sk)->ifindex);
+	dev = dev_get_by_index_rcu(sock_net(sk), READ_ONCE(pkt_sk(sk)->ifindex));
 	if (dev)
 		strlcpy(uaddr->sa_data, dev->name, sizeof(uaddr->sa_data));
 	rcu_read_unlock();
@@ -3519,16 +3519,18 @@ static int packet_getname(struct socket *sock, struct sockaddr *uaddr,
 	struct sock *sk = sock->sk;
 	struct packet_sock *po = pkt_sk(sk);
 	DECLARE_SOCKADDR(struct sockaddr_ll *, sll, uaddr);
+	int ifindex;
 
 	if (peer)
 		return -EOPNOTSUPP;
 
+	ifindex = READ_ONCE(po->ifindex);
 	sll->sll_family = AF_PACKET;
-	sll->sll_ifindex = po->ifindex;
+	sll->sll_ifindex = ifindex;
 	sll->sll_protocol = READ_ONCE(po->num);
 	sll->sll_pkttype = 0;
 	rcu_read_lock();
-	dev = dev_get_by_index_rcu(sock_net(sk), po->ifindex);
+	dev = dev_get_by_index_rcu(sock_net(sk), ifindex);
 	if (dev) {
 		sll->sll_hatype = dev->type;
 		sll->sll_halen = dev->addr_len;
@@ -4107,7 +4109,7 @@ static int packet_notifier(struct notifier_block *this,
 				}
 				if (msg == NETDEV_UNREGISTER) {
 					packet_cached_dev_reset(po);
-					po->ifindex = -1;
+					WRITE_ONCE(po->ifindex, -1);
 					if (po->prot_hook.dev)
 						dev_put(po->prot_hook.dev);
 					po->prot_hook.dev = NULL;
@@ -4617,7 +4619,7 @@ static int packet_seq_show(struct seq_file *seq, void *v)
 			   refcount_read(&s->sk_refcnt),
 			   s->sk_type,
 			   ntohs(READ_ONCE(po->num)),
-			   po->ifindex,
+			   READ_ONCE(po->ifindex),
 			   po->running,
 			   atomic_read(&s->sk_rmem_alloc),
 			   from_kuid_munged(seq_user_ns(seq), sock_i_uid(s)),
Signature · 1 line
2.30.2
author maintainer
#50 lore raw

PATCH 49/101 KVM: selftests: Fix kvm_check_cap() assertion

In reply to Sasha Levin

From: Fuad Tabba <[email protected]>

[ Upstream commit d8ac05ea13d789d5491a5920d70a05659015441d ]

KVM_CHECK_EXTENSION ioctl can return any negative value on error, and not necessarily -1. Change the assertion to reflect that.

Message-Id: <[email protected]>

---
 tools/testing/selftests/kvm/lib/kvm_util.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
tools/testing/selftests/kvm/lib/kvm_util.c +1 -1
index 126c6727a6b0..49805fd16fdf 100644
@@ -55,7 +55,7 @@ int kvm_check_cap(long cap)
 		exit(KSFT_SKIP);
 
 	ret = ioctl(kvm_fd, KVM_CHECK_EXTENSION, cap);
-	TEST_ASSERT(ret != -1, "KVM_CHECK_EXTENSION IOCTL failed,\n"
+	TEST_ASSERT(ret >= 0, "KVM_CHECK_EXTENSION IOCTL failed,\n"
 		"  rc: %i errno: %i", ret, errno);
 
 	close(kvm_fd);
Signature · 1 line
2.30.2
author maintainer
#54 lore raw

PATCH 50/101 net: qed: Fix memcpy() overflow of qed_dcbx_params()

In reply to Sasha Levin

From: Kees Cook <[email protected]>

[ Upstream commit 1c200f832e14420fa770193f9871f4ce2df00d07 ]

The source (&dcbx_info->operational.params) and dest (&p_hwfn->p_dcbx_info->set.config.params) are both struct qed_dcbx_params (560 bytes), not struct qed_dcbx_admin_params (564 bytes), which is used as the memcpy() size.

However it seems that struct qed_dcbx_operational_params (dcbx_info->operational)'s layout matches struct qed_dcbx_admin_params (p_hwfn->p_dcbx_info->set.config)'s 4 byte difference (3 padding, 1 byte for "valid").

On the assumption that the size is wrong (rather than the source structure type), adjust the memcpy() size argument to be 4 bytes smaller and add a BUILD_BUG_ON() to validate any changes to the structure sizes.

---
 drivers/net/ethernet/qlogic/qed/qed_dcbx.c | 4 +++-
 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/net/ethernet/qlogic/qed/qed_dcbx.c +3 -1
index 17d5b649eb36..e81dd34a3cac 100644
@@ -1266,9 +1266,11 @@ int qed_dcbx_get_config_params(struct qed_hwfn *p_hwfn,
 		p_hwfn->p_dcbx_info->set.ver_num |= DCBX_CONFIG_VERSION_STATIC;
 
 	p_hwfn->p_dcbx_info->set.enabled = dcbx_info->operational.enabled;
+	BUILD_BUG_ON(sizeof(dcbx_info->operational.params) !=
+		     sizeof(p_hwfn->p_dcbx_info->set.config.params));
 	memcpy(&p_hwfn->p_dcbx_info->set.config.params,
 	       &dcbx_info->operational.params,
-	       sizeof(struct qed_dcbx_admin_params));
+	       sizeof(p_hwfn->p_dcbx_info->set.config.params));
 	p_hwfn->p_dcbx_info->set.config.valid = true;
 
 	memcpy(params, &p_hwfn->p_dcbx_info->set, sizeof(struct qed_dcbx_set));
Signature · 1 line
2.30.2
author maintainer
#55 lore raw

PATCH 51/101 mac80211: reset profile_periodicity/ema_ap

In reply to Sasha Levin

From: Johannes Berg <[email protected]>

[ Upstream commit bbc6f03ff26e7b71d6135a7b78ce40e7dee3d86a ]

Apparently we never clear these values, so they'll remain set since the setting of them is conditional. Clear the values in the relevant other cases.

---
 net/mac80211/mlme.c | 8 ++++++++
 1 file changed, 8 insertions(+)
net/mac80211/mlme.c +8 -0
index 6d3220c66931..fbe26e912300 100644
@@ -4019,10 +4019,14 @@ static void ieee80211_rx_mgmt_beacon(struct ieee80211_sub_if_data *sdata,
 		if (elems.mbssid_config_ie)
 			bss_conf->profile_periodicity =
 				elems.mbssid_config_ie->profile_periodicity;
+		else
+			bss_conf->profile_periodicity = 0;
 
 		if (elems.ext_capab_len >= 11 &&
 		    (elems.ext_capab[10] & WLAN_EXT_CAPA11_EMA_SUPPORT))
 			bss_conf->ema_ap = true;
+		else
+			bss_conf->ema_ap = false;
 
 		/* continue assoc process */
 		ifmgd->assoc_data->timeout = jiffies;
@@ -5749,12 +5753,16 @@ int ieee80211_mgd_assoc(struct ieee80211_sub_if_data *sdata,
 					      beacon_ies->data, beacon_ies->len);
 		if (elem && elem->datalen >= 3)
 			sdata->vif.bss_conf.profile_periodicity = elem->data[2];
+		else
+			sdata->vif.bss_conf.profile_periodicity = 0;
 
 		elem = cfg80211_find_elem(WLAN_EID_EXT_CAPABILITY,
 					  beacon_ies->data, beacon_ies->len);
 		if (elem && elem->datalen >= 11 &&
 		    (elem->data[10] & WLAN_EXT_CAPA11_EMA_SUPPORT))
 			sdata->vif.bss_conf.ema_ap = true;
+		else
+			sdata->vif.bss_conf.ema_ap = false;
 	} else {
 		assoc_data->timeout = jiffies;
 		assoc_data->timeout_started = true;
Signature · 1 line
2.30.2
author maintainer
#56 lore raw

PATCH 52/101 mac80211: handle various extensible elements correctly

In reply to Sasha Levin

From: Johannes Berg <[email protected]>

[ Upstream commit 652e8363bbc7d149fa194a5cbf30b1001c0274b0 ]

Various elements are parsed with a requirement to have an exact size, when really we should only check that they have the minimum size that we need. Check only that and therefore ignore any additional data that they might carry.

---
 net/mac80211/util.c | 22 +++++++++++-----------
 1 file changed, 11 insertions(+), 11 deletions(-)
net/mac80211/util.c +11 -11
index d8f9fb0646a4..fbf56a203c0e 100644
@@ -954,7 +954,7 @@ static void ieee80211_parse_extension_element(u32 *crc,
 
 	switch (elem->data[0]) {
 	case WLAN_EID_EXT_HE_MU_EDCA:
-		if (len == sizeof(*elems->mu_edca_param_set)) {
+		if (len >= sizeof(*elems->mu_edca_param_set)) {
 			elems->mu_edca_param_set = data;
 			if (crc)
 				*crc = crc32_be(*crc, (void *)elem,
@@ -975,7 +975,7 @@ static void ieee80211_parse_extension_element(u32 *crc,
 		}
 		break;
 	case WLAN_EID_EXT_UORA:
-		if (len == 1)
+		if (len >= 1)
 			elems->uora_element = data;
 		break;
 	case WLAN_EID_EXT_MAX_CHANNEL_SWITCH_TIME:
@@ -983,7 +983,7 @@ static void ieee80211_parse_extension_element(u32 *crc,
 			elems->max_channel_switch_time = data;
 		break;
 	case WLAN_EID_EXT_MULTIPLE_BSSID_CONFIGURATION:
-		if (len == sizeof(*elems->mbssid_config_ie))
+		if (len >= sizeof(*elems->mbssid_config_ie))
 			elems->mbssid_config_ie = data;
 		break;
 	case WLAN_EID_EXT_HE_SPR:
@@ -992,7 +992,7 @@ static void ieee80211_parse_extension_element(u32 *crc,
 			elems->he_spr = data;
 		break;
 	case WLAN_EID_EXT_HE_6GHZ_CAPA:
-		if (len == sizeof(*elems->he_6ghz_capa))
+		if (len >= sizeof(*elems->he_6ghz_capa))
 			elems->he_6ghz_capa = data;
 		break;
 	}
@@ -1081,14 +1081,14 @@ _ieee802_11_parse_elems_crc(const u8 *start, size_t len, bool action,
 
 		switch (id) {
 		case WLAN_EID_LINK_ID:
-			if (elen + 2 != sizeof(struct ieee80211_tdls_lnkie)) {
+			if (elen + 2 < sizeof(struct ieee80211_tdls_lnkie)) {
 				elem_parse_failed = true;
 				break;
 			}
 			elems->lnk_id = (void *)(pos - 2);
 			break;
 		case WLAN_EID_CHAN_SWITCH_TIMING:
-			if (elen != sizeof(struct ieee80211_ch_switch_timing)) {
+			if (elen < sizeof(struct ieee80211_ch_switch_timing)) {
 				elem_parse_failed = true;
 				break;
 			}
@@ -1251,7 +1251,7 @@ _ieee802_11_parse_elems_crc(const u8 *start, size_t len, bool action,
 			elems->sec_chan_offs = (void *)pos;
 			break;
 		case WLAN_EID_CHAN_SWITCH_PARAM:
-			if (elen !=
+			if (elen <
 			    sizeof(*elems->mesh_chansw_params_ie)) {
 				elem_parse_failed = true;
 				break;
@@ -1260,7 +1260,7 @@ _ieee802_11_parse_elems_crc(const u8 *start, size_t len, bool action,
 			break;
 		case WLAN_EID_WIDE_BW_CHANNEL_SWITCH:
 			if (!action ||
-			    elen != sizeof(*elems->wide_bw_chansw_ie)) {
+			    elen < sizeof(*elems->wide_bw_chansw_ie)) {
 				elem_parse_failed = true;
 				break;
 			}
@@ -1279,7 +1279,7 @@ _ieee802_11_parse_elems_crc(const u8 *start, size_t len, bool action,
 			ie = cfg80211_find_ie(WLAN_EID_WIDE_BW_CHANNEL_SWITCH,
 					      pos, elen);
 			if (ie) {
-				if (ie[1] == sizeof(*elems->wide_bw_chansw_ie))
+				if (ie[1] >= sizeof(*elems->wide_bw_chansw_ie))
 					elems->wide_bw_chansw_ie =
 						(void *)(ie + 2);
 				else
@@ -1323,7 +1323,7 @@ _ieee802_11_parse_elems_crc(const u8 *start, size_t len, bool action,
 			elems->cisco_dtpc_elem = pos;
 			break;
 		case WLAN_EID_ADDBA_EXT:
-			if (elen != sizeof(struct ieee80211_addba_ext_ie)) {
+			if (elen < sizeof(struct ieee80211_addba_ext_ie)) {
 				elem_parse_failed = true;
 				break;
 			}
@@ -1349,7 +1349,7 @@ _ieee802_11_parse_elems_crc(const u8 *start, size_t len, bool action,
 							  elem, elems);
 			break;
 		case WLAN_EID_S1G_CAPABILITIES:
-			if (elen == sizeof(*elems->s1g_capab))
+			if (elen >= sizeof(*elems->s1g_capab))
 				elems->s1g_capab = (void *)pos;
 			else
 				elem_parse_failed = true;
Signature · 1 line
2.30.2
author maintainer
#57 lore raw

PATCH 53/101 recordmcount: Correct st_shndx handling

In reply to Sasha Levin

From: Peter Zijlstra <[email protected]>

[ Upstream commit fb780761e7bd9f2e94f5b9a296ead6b35b944206 ]

One should only use st_shndx when >SHN_UNDEF and <SHN_LORESERVE. When SHN_XINDEX, then use .symtab_shndx. Otherwise use 0.

This handles the case: st_shndx >= SHN_LORESERVE && st_shndx != SHN_XINDEX.

---
 scripts/recordmcount.h | 15 ++++++++++-----
 1 file changed, 10 insertions(+), 5 deletions(-)
scripts/recordmcount.h +10 -5
index f9b19524da11..1e9baa5c4fc6 100644
@@ -192,15 +192,20 @@ static unsigned int get_symindex(Elf_Sym const *sym, Elf32_Word const *symtab,
 				 Elf32_Word const *symtab_shndx)
 {
 	unsigned long offset;
+	unsigned short shndx = w2(sym->st_shndx);
 	int index;
 
-	if (sym->st_shndx != SHN_XINDEX)
-		return w2(sym->st_shndx);
+	if (shndx > SHN_UNDEF && shndx < SHN_LORESERVE)
+		return shndx;
 
-	offset = (unsigned long)sym - (unsigned long)symtab;
-	index = offset / sizeof(*sym);
+	if (shndx == SHN_XINDEX) {
+		offset = (unsigned long)sym - (unsigned long)symtab;
+		index = offset / sizeof(*sym);
 
-	return w(symtab_shndx[index]);
+		return w(symtab_shndx[index]);
+	}
+
+	return 0;
 }
 
 static unsigned int get_shnum(Elf_Ehdr const *ehdr, Elf_Shdr const *shdr0)
Signature · 1 line
2.30.2
author maintainer
#58 lore raw

PATCH 54/101 PCI: Add AMD RS690 quirk to enable 64-bit DMA

In reply to Sasha Levin

From: Mikel Rychliski <[email protected]>

[ Upstream commit cacf994a91d3a55c0c2f853d6429cd7b86113915 ]

Although the AMD RS690 chipset has 64-bit DMA support, BIOS implementations sometimes fail to configure the memory limit registers correctly.

The Acer F690GVM mainboard uses this chipset and a Marvell 88E8056 NIC. The sky2 driver programs the NIC to use 64-bit DMA, which will not work:

sky2 0000:02:00.0: error interrupt status=0x8
sky2 0000:02:00.0 eth0: tx timeout
sky2 0000:02:00.0 eth0: transmit ring 0 .. 22 report=0 done=0

Other drivers required by this mainboard either don't support 64-bit DMA, or have it disabled using driver specific quirks. For example, the ahci driver has quirks to enable or disable 64-bit DMA depending on the BIOS version (see ahci_sb600_enable_64bit() in ahci.c). This ahci quirk matches against the SB600 SATA controller, but the real issue is almost certainly with the RS690 PCI host that it was commonly attached to.

To avoid this issue in all drivers with 64-bit DMA support, fix the configuration of the PCI host. If the kernel is aware of physical memory above 4GB, but the BIOS never configured the PCI host with this information, update the registers with our values.

---
 arch/x86/pci/fixup.c | 44 ++++++++++++++++++++++++++++++++++++++++++++
 1 file changed, 44 insertions(+)
arch/x86/pci/fixup.c +44 -0
index 0a0e168be1cb..9b0e771302ce 100644
@@ -779,4 +779,48 @@ DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_AMD, 0x1571, pci_amd_enable_64bit_bar);
 DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_AMD, 0x15b1, pci_amd_enable_64bit_bar);
 DECLARE_PCI_FIXUP_RESUME(PCI_VENDOR_ID_AMD, 0x1601, pci_amd_enable_64bit_bar);
 
+#define RS690_LOWER_TOP_OF_DRAM2	0x30
+#define RS690_LOWER_TOP_OF_DRAM2_VALID	0x1
+#define RS690_UPPER_TOP_OF_DRAM2	0x31
+#define RS690_HTIU_NB_INDEX		0xA8
+#define RS690_HTIU_NB_INDEX_WR_ENABLE	0x100
+#define RS690_HTIU_NB_DATA		0xAC
+
+/*
+ * Some BIOS implementations support RAM above 4GB, but do not configure the
+ * PCI host to respond to bus master accesses for these addresses. These
+ * implementations set the TOP_OF_DRAM_SLOT1 register correctly, so PCI DMA
+ * works as expected for addresses below 4GB.
+ *
+ * Reference: "AMD RS690 ASIC Family Register Reference Guide" (pg. 2-57)
+ * https://www.amd.com/system/files/TechDocs/43372_rs690_rrg_3.00o.pdf
+ */
+static void rs690_fix_64bit_dma(struct pci_dev *pdev)
+{
+	u32 val = 0;
+	phys_addr_t top_of_dram = __pa(high_memory - 1) + 1;
+
+	if (top_of_dram <= (1ULL << 32))
+		return;
+
+	pci_write_config_dword(pdev, RS690_HTIU_NB_INDEX,
+				RS690_LOWER_TOP_OF_DRAM2);
+	pci_read_config_dword(pdev, RS690_HTIU_NB_DATA, &val);
+
+	if (val)
+		return;
+
+	pci_info(pdev, "Adjusting top of DRAM to %pa for 64-bit DMA support\n", &top_of_dram);
+
+	pci_write_config_dword(pdev, RS690_HTIU_NB_INDEX,
+		RS690_UPPER_TOP_OF_DRAM2 | RS690_HTIU_NB_INDEX_WR_ENABLE);
+	pci_write_config_dword(pdev, RS690_HTIU_NB_DATA, top_of_dram >> 32);
+
+	pci_write_config_dword(pdev, RS690_HTIU_NB_INDEX,
+		RS690_LOWER_TOP_OF_DRAM2 | RS690_HTIU_NB_INDEX_WR_ENABLE);
+	pci_write_config_dword(pdev, RS690_HTIU_NB_DATA,
+		top_of_dram | RS690_LOWER_TOP_OF_DRAM2_VALID);
+}
+DECLARE_PCI_FIXUP_FINAL(PCI_VENDOR_ID_ATI, 0x7910, rs690_fix_64bit_dma);
+
 #endif
Signature · 1 line
2.30.2
author maintainer
#59 lore raw

PATCH 55/101 net: ll_temac: Add memory-barriers for TX BD access

In reply to Sasha Levin

From: Esben Haabendal <[email protected]>

[ Upstream commit 28d9fab458b16bcd83f9dd07ede3d585c3e1a69e ]

Add a couple of memory-barriers to ensure correct ordering of read/write access to TX BDs.

In xmit_done, we should ensure that reading the additional BD fields are only done after STS_CTRL_APP0_CMPLT bit is set.

When xmit_done marks the BD as free by setting APP0=0, we need to ensure that the other BD fields are reset first, so we avoid racing with the xmit path, which writes to the same fields.

Finally, making sure to read APP0 of next BD after the current BD, ensures that we see all available buffers.

---
 drivers/net/ethernet/xilinx/ll_temac_main.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)
drivers/net/ethernet/xilinx/ll_temac_main.c +13 -1
index 01bb36e7cff0..b105e1d35d15 100644
@@ -774,12 +774,15 @@ static void temac_start_xmit_done(struct net_device *ndev)
 	stat = be32_to_cpu(cur_p->app0);
 
 	while (stat & STS_CTRL_APP0_CMPLT) {
+		/* Make sure that the other fields are read after bd is
+		 * released by dma
+		 */
+		rmb();
 		dma_unmap_single(ndev->dev.parent, be32_to_cpu(cur_p->phys),
 				 be32_to_cpu(cur_p->len), DMA_TO_DEVICE);
 		skb = (struct sk_buff *)ptr_from_txbd(cur_p);
 		if (skb)
 			dev_consume_skb_irq(skb);
-		cur_p->app0 = 0;
 		cur_p->app1 = 0;
 		cur_p->app2 = 0;
 		cur_p->app3 = 0;
@@ -788,6 +791,12 @@ static void temac_start_xmit_done(struct net_device *ndev)
 		ndev->stats.tx_packets++;
 		ndev->stats.tx_bytes += be32_to_cpu(cur_p->len);
 
+		/* app0 must be visible last, as it is used to flag
+		 * availability of the bd
+		 */
+		smp_mb();
+		cur_p->app0 = 0;
+
 		lp->tx_bd_ci++;
 		if (lp->tx_bd_ci >= lp->tx_bd_num)
 			lp->tx_bd_ci = 0;
@@ -814,6 +823,9 @@ static inline int temac_check_tx_bd_space(struct temac_local *lp, int num_frag)
 		if (cur_p->app0)
 			return NETDEV_TX_BUSY;
 
+		/* Make sure to read next bd app0 after this one */
+		rmb();
+
 		tail++;
 		if (tail >= lp->tx_bd_num)
 			tail = 0;
Signature · 1 line
2.30.2
author maintainer
#7 lore raw

PATCH 7/101 drm/amdgpu: wait for moving fence after pinning

In reply to Sasha Levin

From: Christian König <[email protected]>

commit 8ddf5b9bb479570a3825d70fecfb9399bc15700c upstream.

We actually need to wait for the moving fence after pinning the BO to make sure that the pin is completed.

References: https://lore.kernel.org/dri-devel/[email protected]/

---
 drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c | 14 +++++++++++++-
 1 file changed, 13 insertions(+), 1 deletion(-)
drivers/gpu/drm/amd/amdgpu/amdgpu_dma_buf.c +13 -1
index 1b56dbc1f304..e93ccdc5faf4 100644
@@ -238,9 +238,21 @@ static int amdgpu_dma_buf_pin(struct dma_buf_attachment *attach)
 {
 	struct drm_gem_object *obj = attach->dmabuf->priv;
 	struct amdgpu_bo *bo = gem_to_amdgpu_bo(obj);
+	int r;
 
 	/* pin buffer into GTT */
-	return amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT);
+	r = amdgpu_bo_pin(bo, AMDGPU_GEM_DOMAIN_GTT);
+	if (r)
+		return r;
+
+	if (bo->tbo.moving) {
+		r = dma_fence_wait(bo->tbo.moving, true);
+		if (r) {
+			amdgpu_bo_unpin(bo);
+			return r;
+		}
+	}
+	return 0;
 }
 
 /**
Signature · 1 line
2.30.2
author maintainer
#11 lore raw

PATCH 11/101 arm64: Ignore any DMA offsets in the max_zone_phys() calculation

In reply to Sasha Levin

From: Catalin Marinas <[email protected]>

commit 791ab8b2e3db0c6e4295467d10398800ec29144c upstream.

Currently, the kernel assumes that if RAM starts above 32-bit (or zone_bits), there is still a ZONE_DMA/DMA32 at the bottom of the RAM and such constrained devices have a hardwired DMA offset. In practice, we haven't noticed any such hardware so let's assume that we can expand ZONE_DMA32 to the available memory if no RAM below 4GB. Similarly, ZONE_DMA is expanded to the 4GB limit if no RAM addressable by zone_bits.

---
 arch/arm64/mm/init.c | 17 ++++++++++++-----
 1 file changed, 12 insertions(+), 5 deletions(-)
arch/arm64/mm/init.c +12 -5
index a985d292e820..c0a7f0d90b39 100644
@@ -174,14 +174,21 @@ static void __init reserve_elfcorehdr(void)
 #endif /* CONFIG_CRASH_DUMP */
 
 /*
- * Return the maximum physical address for a zone with a given address size
- * limit. It currently assumes that for memory starting above 4G, 32-bit
- * devices will use a DMA offset.
+ * Return the maximum physical address for a zone accessible by the given bits
+ * limit. If DRAM starts above 32-bit, expand the zone to the maximum
+ * available memory, otherwise cap it at 32-bit.
  */
 static phys_addr_t __init max_zone_phys(unsigned int zone_bits)
 {
-	phys_addr_t offset = memblock_start_of_DRAM() & GENMASK_ULL(63, zone_bits);
-	return min(offset + (1ULL << zone_bits), memblock_end_of_DRAM());
+	phys_addr_t zone_mask = DMA_BIT_MASK(zone_bits);
+	phys_addr_t phys_start = memblock_start_of_DRAM();
+
+	if (phys_start > U32_MAX)
+		zone_mask = PHYS_ADDR_MAX;
+	else if (phys_start > zone_mask)
+		zone_mask = U32_MAX;
+
+	return min(zone_mask, memblock_end_of_DRAM() - 1) + 1;
 }
 
 static void __init zone_sizes_init(unsigned long min, unsigned long max)
Signature · 1 line
2.30.2
author maintainer
#2 lore raw

PATCH 2/101 Revert "drm/amdgpu/gfx9: fix the doorbell missing when in CGPG issue."

In reply to Sasha Levin

From: Yifan Zhang <[email protected]>

commit ee5468b9f1d3bf48082eed351dace14598e8ca39 upstream.

This reverts commit 4cbbe34807938e6e494e535a68d5ff64edac3f20.

Reason for revert: side effect of enlarging CP_MEC_DOORBELL_RANGE may cause some APUs fail to enter gfxoff in certain user cases.

---
 drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c | 6 +-----
 1 file changed, 1 insertion(+), 5 deletions(-)
drivers/gpu/drm/amd/amdgpu/gfx_v9_0.c +1 -5
index 1859d293ef71..fb15e8b5af32 100644
@@ -3619,12 +3619,8 @@ static int gfx_v9_0_kiq_init_register(struct amdgpu_ring *ring)
 	if (ring->use_doorbell) {
 		WREG32_SOC15(GC, 0, mmCP_MEC_DOORBELL_RANGE_LOWER,
 					(adev->doorbell_index.kiq * 2) << 2);
-		/* If GC has entered CGPG, ringing doorbell > first page doesn't
-		 * wakeup GC. Enlarge CP_MEC_DOORBELL_RANGE_UPPER to workaround
-		 * this issue.
-		 */
 		WREG32_SOC15(GC, 0, mmCP_MEC_DOORBELL_RANGE_UPPER,
-					(adev->doorbell.size - 4));
+					(adev->doorbell_index.userqueue_end * 2) << 2);
 	}
 
 	WREG32_SOC15_RLC(GC, 0, mmCP_HQD_PQ_DOORBELL_CONTROL,
Signature · 1 line
2.30.2
author maintainer
#10 lore raw

PATCH 10/101 MIPS: generic: Update node names to avoid unit addresses

In reply to Sasha Levin

From: Nathan Chancellor <[email protected]>

commit e607ff630c6053ecc67502677c0e50053d7892d4 upstream.

With the latest mkimage from U-Boot 2021.04, the generic defconfigs no longer build, failing with:

/usr/bin/mkimage: verify_header failed for FIT Image support with exit code 1

This is expected after the linked U-Boot commits because '@' is forbidden in the node names due to the way that libfdt treats nodes with the same prefix but different unit addresses.

Switch the '@' in the node name to '-'. Drop the unit addresses from the hash and kernel child nodes because there is only one node so they do not need to have a number to differentiate them.

---
 arch/mips/generic/board-boston.its.S   | 10 +++++-----
 arch/mips/generic/board-ni169445.its.S | 10 +++++-----
 arch/mips/generic/board-ocelot.its.S   | 20 ++++++++++----------
 arch/mips/generic/board-xilfpga.its.S  | 10 +++++-----
 arch/mips/generic/vmlinux.its.S        | 10 +++++-----
 5 files changed, 30 insertions(+), 30 deletions(-)
arch/mips/generic/board-boston.its.S +5 -5
index a7f51f97b910..c45ad2759421 100644
@@ -1,22 +1,22 @@
 / {
 	images {
-		fdt@boston {
+		fdt-boston {
 			description = "img,boston Device Tree";
 			data = /incbin/("boot/dts/img/boston.dtb");
 			type = "flat_dt";
 			arch = "mips";
 			compression = "none";
-			hash@0 {
+			hash {
 				algo = "sha1";
 			};
 		};
 	};
 
 	configurations {
-		conf@boston {
+		conf-boston {
 			description = "Boston Linux kernel";
-			kernel = "kernel@0";
-			fdt = "fdt@boston";
+			kernel = "kernel";
+			fdt = "fdt-boston";
 		};
 	};
 };
arch/mips/generic/board-ni169445.its.S +5 -5
index e4cb4f95a8cc..0a2e8f7a8526 100644
@@ -1,22 +1,22 @@
 / {
 	images {
-		fdt@ni169445 {
+		fdt-ni169445 {
 			description = "NI 169445 device tree";
 			data = /incbin/("boot/dts/ni/169445.dtb");
 			type = "flat_dt";
 			arch = "mips";
 			compression = "none";
-			hash@0 {
+			hash {
 				algo = "sha1";
 			};
 		};
 	};
 
 	configurations {
-		conf@ni169445 {
+		conf-ni169445 {
 			description = "NI 169445 Linux Kernel";
-			kernel = "kernel@0";
-			fdt = "fdt@ni169445";
+			kernel = "kernel";
+			fdt = "fdt-ni169445";
 		};
 	};
 };
arch/mips/generic/board-ocelot.its.S +10 -10
index 3da23988149a..8c7e3a1b68d3 100644
@@ -1,40 +1,40 @@
 /* SPDX-License-Identifier: (GPL-2.0 OR MIT) */
 / {
 	images {
-		fdt@ocelot_pcb123 {
+		fdt-ocelot_pcb123 {
 			description = "MSCC Ocelot PCB123 Device Tree";
 			data = /incbin/("boot/dts/mscc/ocelot_pcb123.dtb");
 			type = "flat_dt";
 			arch = "mips";
 			compression = "none";
-			hash@0 {
+			hash {
 				algo = "sha1";
 			};
 		};
 
-		fdt@ocelot_pcb120 {
+		fdt-ocelot_pcb120 {
 			description = "MSCC Ocelot PCB120 Device Tree";
 			data = /incbin/("boot/dts/mscc/ocelot_pcb120.dtb");
 			type = "flat_dt";
 			arch = "mips";
 			compression = "none";
-			hash@0 {
+			hash {
 				algo = "sha1";
 			};
 		};
 	};
 
 	configurations {
-		conf@ocelot_pcb123 {
+		conf-ocelot_pcb123 {
 			description = "Ocelot Linux kernel";
-			kernel = "kernel@0";
-			fdt = "fdt@ocelot_pcb123";
+			kernel = "kernel";
+			fdt = "fdt-ocelot_pcb123";
 		};
 
-		conf@ocelot_pcb120 {
+		conf-ocelot_pcb120 {
 			description = "Ocelot Linux kernel";
-			kernel = "kernel@0";
-			fdt = "fdt@ocelot_pcb120";
+			kernel = "kernel";
+			fdt = "fdt-ocelot_pcb120";
 		};
 	};
 };
arch/mips/generic/board-xilfpga.its.S +5 -5
index a2e773d3f14f..08c1e900eb4e 100644
@@ -1,22 +1,22 @@
 / {
 	images {
-		fdt@xilfpga {
+		fdt-xilfpga {
 			description = "MIPSfpga (xilfpga) Device Tree";
 			data = /incbin/("boot/dts/xilfpga/nexys4ddr.dtb");
 			type = "flat_dt";
 			arch = "mips";
 			compression = "none";
-			hash@0 {
+			hash {
 				algo = "sha1";
 			};
 		};
 	};
 
 	configurations {
-		conf@xilfpga {
+		conf-xilfpga {
 			description = "MIPSfpga Linux kernel";
-			kernel = "kernel@0";
-			fdt = "fdt@xilfpga";
+			kernel = "kernel";
+			fdt = "fdt-xilfpga";
 		};
 	};
 };
arch/mips/generic/vmlinux.its.S +5 -5
index 1a08438fd893..3e254676540f 100644
@@ -6,7 +6,7 @@
 	#address-cells = <ADDR_CELLS>;
 
 	images {
-		kernel@0 {
+		kernel {
 			description = KERNEL_NAME;
 			data = /incbin/(VMLINUX_BINARY);
 			type = "kernel";
@@ -15,18 +15,18 @@
 			compression = VMLINUX_COMPRESSION;
 			load = /bits/ ADDR_BITS <VMLINUX_LOAD_ADDRESS>;
 			entry = /bits/ ADDR_BITS <VMLINUX_ENTRY_ADDRESS>;
-			hash@0 {
+			hash {
 				algo = "sha1";
 			};
 		};
 	};
 
 	configurations {
-		default = "conf@default";
+		default = "conf-default";
 
-		conf@default {
+		conf-default {
 			description = "Generic Linux kernel";
-			kernel = "kernel@0";
+			kernel = "kernel";
 		};
 	};
 };
Signature · 1 line
2.30.2
author maintainer
#24 lore raw

PATCH 23/101 dmaengine: stm32-mdma: fix PM reference leak in stm32_mdma_alloc_chan_resourc()

In reply to Sasha Levin

From: Yu Kuai <[email protected]>

[ Upstream commit 83eb4868d325b86e18509d0874e911497667cb54 ]

pm_runtime_get_sync will increment pm usage counter even it failed. Forgetting to putting operation will result in reference leak here. Fix it by replacing it with pm_runtime_resume_and_get to keep usage counter balanced.

---
 drivers/dma/stm32-mdma.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/dma/stm32-mdma.c +2 -2
index 08cfbfab837b..9d473923712a 100644
@@ -1448,7 +1448,7 @@ static int stm32_mdma_alloc_chan_resources(struct dma_chan *c)
 		return -ENOMEM;
 	}
 
-	ret = pm_runtime_get_sync(dmadev->ddev.dev);
+	ret = pm_runtime_resume_and_get(dmadev->ddev.dev);
 	if (ret < 0)
 		return ret;
 
@@ -1714,7 +1714,7 @@ static int stm32_mdma_pm_suspend(struct device *dev)
 	u32 ccr, id;
 	int ret;
 
-	ret = pm_runtime_get_sync(dev);
+	ret = pm_runtime_resume_and_get(dev);
 	if (ret < 0)
 		return ret;
 
Signature · 1 line
2.30.2
author maintainer
#22 lore raw

PATCH 21/101 perf/x86/intel/lbr: Zero the xstate buffer on allocation

In reply to Sasha Levin

From: Thomas Gleixner <[email protected]>

[ Upstream commit 7f049fbdd57f6ea71dc741d903c19c73b2f70950 ]

XRSTORS requires a valid xstate buffer to work correctly. XSAVES does not guarantee to write a fully valid buffer according to the SDM:

"XSAVES does not write to any parts of the XSAVE header other than the
XSTATE_BV and XCOMP_BV fields."

XRSTORS triggers a #GP:

"If bytes 63:16 of the XSAVE header are not all zero."

It's dubious at best how this can work at all when the buffer is not zeroed before use.

Allocate the buffers with __GFP_ZERO to prevent XRSTORS failure.

Fixes ce711ea3cab9 ("perf/x86/intel/lbr: Support XSAVES/XRSTORS for LBR context switch")
---
 arch/x86/events/intel/lbr.c | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)
arch/x86/events/intel/lbr.c +2 -1
index 6c1231837382..29ec4fe48507 100644
@@ -730,7 +730,8 @@ void reserve_lbr_buffers(void)
 		if (!kmem_cache || cpuc->lbr_xsave)
 			continue;
 
-		cpuc->lbr_xsave = kmem_cache_alloc_node(kmem_cache, GFP_KERNEL,
+		cpuc->lbr_xsave = kmem_cache_alloc_node(kmem_cache,
+							GFP_KERNEL | __GFP_ZERO,
 							cpu_to_node(cpu));
 	}
 }
Signature · 1 line
2.30.2
author maintainer
#23 lore raw

PATCH 22/101 dmaengine: zynqmp_dma: Fix PM reference leak in zynqmp_dma_alloc_chan_resourc()

In reply to Sasha Levin

From: Yu Kuai <[email protected]>

[ Upstream commit 8982d48af36d2562c0f904736b0fc80efc9f2532 ]

pm_runtime_get_sync will increment pm usage counter even it failed. Forgetting to putting operation will result in reference leak here. Fix it by replacing it with pm_runtime_resume_and_get to keep usage counter balanced.

---
 drivers/dma/xilinx/zynqmp_dma.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/dma/xilinx/zynqmp_dma.c +1 -1
index d8419565b92c..5fecf5aa6e85 100644
@@ -468,7 +468,7 @@ static int zynqmp_dma_alloc_chan_resources(struct dma_chan *dchan)
 	struct zynqmp_dma_desc_sw *desc;
 	int i, ret;
 
-	ret = pm_runtime_get_sync(chan->dev);
+	ret = pm_runtime_resume_and_get(chan->dev);
 	if (ret < 0)
 		return ret;
 
Signature · 1 line
2.30.2
author maintainer
#31 lore raw

PATCH 30/101 dmaengine: mediatek: free the proper desc in desc_free handler

In reply to Sasha Levin

From: Guillaume Ranquet <[email protected]>

[ Upstream commit 0a2ff58f9f8f95526ecb0ccd7517fefceb96f661 ]

The desc_free handler assumed that the desc we want to free was always
the current one associated with the channel.

This is seldom the case and this is causing use after free crashes in
multiple places (tx/rx/terminate...).

  BUG: KASAN: use-after-free in mtk_uart_apdma_rx_handler+0x120/0x304

  Call trace:
   dump_backtrace+0x0/0x1b0
   show_stack+0x24/0x34
   dump_stack+0xe0/0x150
   print_address_description+0x8c/0x55c
   __kasan_report+0x1b8/0x218
   kasan_report+0x14/0x20
   __asan_load4+0x98/0x9c
   mtk_uart_apdma_rx_handler+0x120/0x304
   mtk_uart_apdma_irq_handler+0x50/0x80
   __handle_irq_event_percpu+0xe0/0x210
   handle_irq_event+0x8c/0x184
   handle_fasteoi_irq+0x1d8/0x3ac
   __handle_domain_irq+0xb0/0x110
   gic_handle_irq+0x50/0xb8
   el0_irq_naked+0x60/0x6c

  Allocated by task 3541:
   __kasan_kmalloc+0xf0/0x1b0
   kasan_kmalloc+0x10/0x1c
   kmem_cache_alloc_trace+0x90/0x2dc
   mtk_uart_apdma_prep_slave_sg+0x6c/0x1a0
   mtk8250_dma_rx_complete+0x220/0x2e4
   vchan_complete+0x290/0x340
   tasklet_action_common+0x220/0x298
   tasklet_action+0x28/0x34
   __do_softirq+0x158/0x35c

  Freed by task 3541:
   __kasan_slab_free+0x154/0x224
   kasan_slab_free+0x14/0x24
   slab_free_freelist_hook+0xf8/0x15c
   kfree+0xb4/0x278
   mtk_uart_apdma_desc_free+0x34/0x44
   vchan_complete+0x1bc/0x340
   tasklet_action_common+0x220/0x298
   tasklet_action+0x28/0x34
   __do_softirq+0x158/0x35c

  The buggy address belongs to the object at ffff000063606800
   which belongs to the cache kmalloc-256 of size 256
  The buggy address is located 176 bytes inside of
   256-byte region [ffff000063606800, ffff000063606900)
  The buggy address belongs to the page:
  page:fffffe00016d8180 refcount:1 mapcount:0 mapping:ffff00000302f600 index:0x0 compound_mapcount: 0
  flags: 0xffff00000010200(slab|head)
  raw: 0ffff00000010200 dead000000000100 dead000000000122 ffff00000302f600
  raw: 0000000000000000 0000000080100010 00000001ffffffff 0000000000000000
  page dumped because: kasan: bad access detected
---
 drivers/dma/mediatek/mtk-uart-apdma.c | 5 +----
 1 file changed, 1 insertion(+), 4 deletions(-)
drivers/dma/mediatek/mtk-uart-apdma.c +1 -4
index 27c07350971d..e38b67fc0c0c 100644
@@ -131,10 +131,7 @@ static unsigned int mtk_uart_apdma_read(struct mtk_chan *c, unsigned int reg)
 
 static void mtk_uart_apdma_desc_free(struct virt_dma_desc *vd)
 {
-	struct dma_chan *chan = vd->tx.chan;
-	struct mtk_chan *c = to_mtk_uart_apdma_chan(chan);
-
-	kfree(c->desc);
+	kfree(container_of(vd, struct mtk_uart_apdma_desc, vd));
 }
 
 static void mtk_uart_apdma_start_tx(struct mtk_chan *c)
Signature · 1 line
2.30.2
author maintainer
#30 lore raw

PATCH 29/101 dmaengine: rcar-dmac: Fix PM reference leak in rcar_dmac_probe()

In reply to Sasha Levin

From: Zou Wei <[email protected]>

[ Upstream commit dea8464ddf553803382efb753b6727dbf3931d06 ]

pm_runtime_get_sync will increment pm usage counter even it failed. Forgetting to putting operation will result in reference leak here. Fix it by replacing it with pm_runtime_resume_and_get to keep usage counter balanced.

---
 drivers/dma/sh/rcar-dmac.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/dma/sh/rcar-dmac.c +1 -1
index a57705356e8b..991a7b5da29f 100644
@@ -1874,7 +1874,7 @@ static int rcar_dmac_probe(struct platform_device *pdev)
 
 	/* Enable runtime PM and initialize the device. */
 	pm_runtime_enable(&pdev->dev);
-	ret = pm_runtime_get_sync(&pdev->dev);
+	ret = pm_runtime_resume_and_get(&pdev->dev);
 	if (ret < 0) {
 		dev_err(&pdev->dev, "runtime PM get sync failed (%d)\n", ret);
 		return ret;
Signature · 1 line
2.30.2
author maintainer
#34 lore raw

PATCH 33/101 net: ipv4: Remove unneed BUG() function

In reply to Sasha Levin

From: Zheng Yongjun <[email protected]>

[ Upstream commit 5ac6b198d7e312bd10ebe7d58c64690dc59cc49a ]

When 'nla_parse_nested_deprecated' failed, it's no need to BUG() here, return -EINVAL is ok.

---
 net/ipv4/devinet.c  | 2 +-
 net/ipv6/addrconf.c | 2 +-
 2 files changed, 2 insertions(+), 2 deletions(-)
net/ipv4/devinet.c +1 -1
index 123a6d39438f..7c1859777429 100644
@@ -1989,7 +1989,7 @@ static int inet_set_link_af(struct net_device *dev, const struct nlattr *nla)
 		return -EAFNOSUPPORT;
 
 	if (nla_parse_nested_deprecated(tb, IFLA_INET_MAX, nla, NULL, NULL) < 0)
-		BUG();
+		return -EINVAL;
 
 	if (tb[IFLA_INET_CONF]) {
 		nla_for_each_nested(a, tb[IFLA_INET_CONF], rem)
net/ipv6/addrconf.c +1 -1
index 4c881f5d9080..884d430e23cb 100644
@@ -5799,7 +5799,7 @@ static int inet6_set_link_af(struct net_device *dev, const struct nlattr *nla)
 		return -EAFNOSUPPORT;
 
 	if (nla_parse_nested_deprecated(tb, IFLA_INET6_MAX, nla, NULL, NULL) < 0)
-		BUG();
+		return -EINVAL;
 
 	if (tb[IFLA_INET6_TOKEN]) {
 		err = inet6_set_iftoken(idev, nla_data(tb[IFLA_INET6_TOKEN]));
Signature · 1 line
2.30.2
author maintainer
#62 lore raw

PATCH 56/101 net: ll_temac: Avoid ndo_start_xmit returning NETDEV_TX_BUSY

In reply to Sasha Levin

From: Esben Haabendal <[email protected]>

[ Upstream commit f6396341194234e9b01cd7538bc2c6ac4501ab14 ]

As documented in Documentation/networking/driver.rst, the ndo_start_xmit method must not return NETDEV_TX_BUSY under any normal circumstances, and as recommended, we simply stop the tx queue in advance, when there is a risk that the next xmit would cause a NETDEV_TX_BUSY return.

---
 drivers/net/ethernet/xilinx/ll_temac_main.c | 5 +++++
 1 file changed, 5 insertions(+)
drivers/net/ethernet/xilinx/ll_temac_main.c +5 -0
index b105e1d35d15..6bd3a389d389 100644
@@ -942,6 +942,11 @@ temac_start_xmit(struct sk_buff *skb, struct net_device *ndev)
 	wmb();
 	lp->dma_out(lp, TX_TAILDESC_PTR, tail_p); /* DMA start */
 
+	if (temac_check_tx_bd_space(lp, MAX_SKB_FRAGS + 1)) {
+		netdev_info(ndev, "%s -> netif_stop_queue\n", __func__);
+		netif_stop_queue(ndev);
+	}
+
 	return NETDEV_TX_OK;
 }
 
Signature · 1 line
2.30.2
author maintainer
#63 lore raw

PATCH 57/101 perf/x86: Track pmu in per-CPU cpu_hw_events

In reply to Sasha Levin

From: Kan Liang <[email protected]>

[ Upstream commit 61e76d53c39bb768ad264d379837cfc56b9e35b4 ]

Some platforms, e.g. Alder Lake, have hybrid architecture. In the same package, there may be more than one type of CPU. The PMU capabilities are different among different types of CPU. Perf will register a dedicated PMU for each type of CPU.

Add a 'pmu' variable in the struct cpu_hw_events to track the dedicated PMU of the current CPU.

Current x86_get_pmu() use the global 'pmu', which will be broken on a hybrid platform. Modify it to apply the 'pmu' of the specific CPU.

Initialize the per-CPU 'pmu' variable with the global 'pmu'. There is nothing changed for the non-hybrid platforms.

The is_x86_event() will be updated in the later patch ("perf/x86:
Register hybrid PMUs") for hybrid platforms. For the non-hybrid platforms, nothing is changed here.

---
 arch/x86/events/core.c       | 17 +++++++++++++----
 arch/x86/events/intel/core.c |  2 +-
 arch/x86/events/intel/ds.c   |  4 ++--
 arch/x86/events/intel/lbr.c  |  9 +++++----
 arch/x86/events/perf_event.h |  4 +++-
 5 files changed, 24 insertions(+), 12 deletions(-)
arch/x86/events/core.c +13 -4
index b7f8ed87bfbc..e6db1a1f22d7 100644
@@ -45,9 +45,11 @@
 #include "perf_event.h"
 
 struct x86_pmu x86_pmu __read_mostly;
+static struct pmu pmu;
 
 DEFINE_PER_CPU(struct cpu_hw_events, cpu_hw_events) = {
 	.enabled = 1,
+	.pmu = &pmu,
 };
 
 DEFINE_STATIC_KEY_FALSE(rdpmc_never_available_key);
@@ -712,16 +714,23 @@ void x86_pmu_enable_all(int added)
 	}
 }
 
-static struct pmu pmu;
-
 static inline int is_x86_event(struct perf_event *event)
 {
 	return event->pmu == &pmu;
 }
 
-struct pmu *x86_get_pmu(void)
+struct pmu *x86_get_pmu(unsigned int cpu)
 {
-	return &pmu;
+	struct cpu_hw_events *cpuc = &per_cpu(cpu_hw_events, cpu);
+
+	/*
+	 * All CPUs of the hybrid type have been offline.
+	 * The x86_get_pmu() should not be invoked.
+	 */
+	if (WARN_ON_ONCE(!cpuc->pmu))
+		return &pmu;
+
+	return cpuc->pmu;
 }
 /*
  * Event scheduler state:
arch/x86/events/intel/core.c +1 -1
index ee659b5faf71..3b8b8eede1a8 100644
@@ -4747,7 +4747,7 @@ static void update_tfa_sched(void *ignored)
 	 * and if so force schedule out for all event types all contexts
 	 */
 	if (test_bit(3, cpuc->active_mask))
-		perf_pmu_resched(x86_get_pmu());
+		perf_pmu_resched(x86_get_pmu(smp_processor_id()));
 }
 
 static ssize_t show_sysctl_tfa(struct device *cdev,
arch/x86/events/intel/ds.c +2 -2
index 31a7a6566d07..945d470f62d0 100644
@@ -2076,7 +2076,7 @@ void __init intel_ds_init(void)
 					PERF_SAMPLE_TIME;
 				x86_pmu.flags |= PMU_FL_PEBS_ALL;
 				pebs_qual = "-baseline";
-				x86_get_pmu()->capabilities |= PERF_PMU_CAP_EXTENDED_REGS;
+				x86_get_pmu(smp_processor_id())->capabilities |= PERF_PMU_CAP_EXTENDED_REGS;
 			} else {
 				/* Only basic record supported */
 				x86_pmu.large_pebs_flags &=
@@ -2091,7 +2091,7 @@ void __init intel_ds_init(void)
 
 			if (x86_pmu.intel_cap.pebs_output_pt_available) {
 				pr_cont("PEBS-via-PT, ");
-				x86_get_pmu()->capabilities |= PERF_PMU_CAP_AUX_OUTPUT;
+				x86_get_pmu(smp_processor_id())->capabilities |= PERF_PMU_CAP_AUX_OUTPUT;
 			}
 
 			break;
arch/x86/events/intel/lbr.c +5 -4
index 29ec4fe48507..9c1a013d5682 100644
@@ -699,7 +699,7 @@ void intel_pmu_lbr_add(struct perf_event *event)
 
 void release_lbr_buffers(void)
 {
-	struct kmem_cache *kmem_cache = x86_get_pmu()->task_ctx_cache;
+	struct kmem_cache *kmem_cache;
 	struct cpu_hw_events *cpuc;
 	int cpu;
 
@@ -708,6 +708,7 @@ void release_lbr_buffers(void)
 
 	for_each_possible_cpu(cpu) {
 		cpuc = per_cpu_ptr(&cpu_hw_events, cpu);
+		kmem_cache = x86_get_pmu(cpu)->task_ctx_cache;
 		if (kmem_cache && cpuc->lbr_xsave) {
 			kmem_cache_free(kmem_cache, cpuc->lbr_xsave);
 			cpuc->lbr_xsave = NULL;
@@ -1624,7 +1625,7 @@ void intel_pmu_lbr_init_hsw(void)
 	x86_pmu.lbr_sel_mask = LBR_SEL_MASK;
 	x86_pmu.lbr_sel_map  = hsw_lbr_sel_map;
 
-	x86_get_pmu()->task_ctx_cache = create_lbr_kmem_cache(size, 0);
+	x86_get_pmu(smp_processor_id())->task_ctx_cache = create_lbr_kmem_cache(size, 0);
 
 	if (lbr_from_signext_quirk_needed())
 		static_branch_enable(&lbr_from_quirk_key);
@@ -1644,7 +1645,7 @@ __init void intel_pmu_lbr_init_skl(void)
 	x86_pmu.lbr_sel_mask = LBR_SEL_MASK;
 	x86_pmu.lbr_sel_map  = hsw_lbr_sel_map;
 
-	x86_get_pmu()->task_ctx_cache = create_lbr_kmem_cache(size, 0);
+	x86_get_pmu(smp_processor_id())->task_ctx_cache = create_lbr_kmem_cache(size, 0);
 
 	/*
 	 * SW branch filter usage:
@@ -1741,7 +1742,7 @@ static bool is_arch_lbr_xsave_available(void)
 
 void __init intel_pmu_arch_lbr_init(void)
 {
-	struct pmu *pmu = x86_get_pmu();
+	struct pmu *pmu = x86_get_pmu(smp_processor_id());
 	union cpuid28_eax eax;
 	union cpuid28_ebx ebx;
 	union cpuid28_ecx ecx;
arch/x86/events/perf_event.h +3 -1
index d4f2ea2d9a9e..f07d77cffb3c 100644
@@ -326,6 +326,8 @@ struct cpu_hw_events {
 	int				n_pair; /* Large increment events */
 
 	void				*kfree_on_online[X86_PERF_KFREE_MAX];
+
+	struct pmu			*pmu;
 };
 
 #define __EVENT_CONSTRAINT_RANGE(c, e, n, m, w, o, f) {	\
@@ -897,7 +899,7 @@ static struct perf_pmu_events_ht_attr event_attr_##v = {		\
 	.event_str_ht	= ht,						\
 }
 
-struct pmu *x86_get_pmu(void);
+struct pmu *x86_get_pmu(unsigned int cpu);
 extern struct x86_pmu x86_pmu __read_mostly;
 
 static __always_inline struct x86_perf_task_context_opt *task_context_opt(void *ctx)
Signature · 1 line
2.30.2
author maintainer
#64 lore raw

PATCH 58/101 pinctrl: stm32: fix the reported number of GPIO lines per bank

In reply to Sasha Levin

From: Fabien Dessenne <[email protected]>

[ Upstream commit 67e2996f72c71ebe4ac2fcbcf77e54479bb7aa11 ]

Each GPIO bank supports a variable number of lines which is usually 16, but is less in some cases : this is specified by the last argument of the "gpio-ranges" bank node property.
Report to the framework, the actual number of lines, so the libgpiod gpioinfo command lists the actually existing GPIO lines.

Fixes 1dc9d289154b ("pinctrl: stm32: add possibility to use gpio-ranges to declare bank range")
---
 drivers/pinctrl/stm32/pinctrl-stm32.c | 9 +++++++--
 1 file changed, 7 insertions(+), 2 deletions(-)
drivers/pinctrl/stm32/pinctrl-stm32.c +7 -2
index 7d9bdedcd71b..3af4430543dc 100644
@@ -1229,7 +1229,7 @@ static int stm32_gpiolib_register_bank(struct stm32_pinctrl *pctl,
 	struct device *dev = pctl->dev;
 	struct resource res;
 	int npins = STM32_GPIO_PINS_PER_BANK;
-	int bank_nr, err;
+	int bank_nr, err, i = 0;
 
 	if (!IS_ERR(bank->rstc))
 		reset_control_deassert(bank->rstc);
@@ -1251,9 +1251,14 @@ static int stm32_gpiolib_register_bank(struct stm32_pinctrl *pctl,
 
 	of_property_read_string(np, "st,bank-name", &bank->gpio_chip.label);
 
-	if (!of_parse_phandle_with_fixed_args(np, "gpio-ranges", 3, 0, &args)) {
+	if (!of_parse_phandle_with_fixed_args(np, "gpio-ranges", 3, i, &args)) {
 		bank_nr = args.args[1] / STM32_GPIO_PINS_PER_BANK;
 		bank->gpio_chip.base = args.args[1];
+
+		npins = args.args[2];
+		while (!of_parse_phandle_with_fixed_args(np, "gpio-ranges", 3,
+							 ++i, &args))
+			npins += args.args[2];
 	} else {
 		bank_nr = pctl->nbanks;
 		bank->gpio_chip.base = bank_nr * STM32_GPIO_PINS_PER_BANK;
Signature · 1 line
2.30.2
author maintainer
#65 lore raw

PATCH 59/101 i2c: i801: Ensure that SMBHSTSTS_INUSE_STS is cleared when leaving i801_access

In reply to Sasha Levin

From: Heiner Kallweit <[email protected]>

[ Upstream commit 065b6211a87746e196b56759a70c7851418dd741 ]

As explained in [0] currently we may leave SMBHSTSTS_INUSE_STS set, thus potentially breaking ACPI/BIOS usage of the SMBUS device.

Seems patch [0] needs a little bit more of review effort, therefore I'd suggest to apply a part of it as quick win. Just clearing SMBHSTSTS_INUSE_STS when leaving i801_access() should fix the referenced issue and leaves more time for discussing a more sophisticated locking handling.

[0] https://www.spinics.net/lists/linux-i2c/msg51558.html

Fixes 01590f361e94 ("i2c: i801: Instantiate SPD EEPROMs automatically")
---
 drivers/i2c/busses/i2c-i801.c | 3 +++
 1 file changed, 3 insertions(+)
drivers/i2c/busses/i2c-i801.c +3 -0
index e42b87e96f74..eab6fd6b890e 100644
@@ -974,6 +974,9 @@ static s32 i801_access(struct i2c_adapter *adap, u16 addr,
 	}
 
 out:
+	/* Unlock the SMBus device for use by BIOS/ACPI */
+	outb_p(SMBHSTSTS_INUSE_STS, SMBHSTSTS(priv));
+
 	pm_runtime_mark_last_busy(&priv->pci_dev->dev);
 	pm_runtime_put_autosuspend(&priv->pci_dev->dev);
 	mutex_unlock(&priv->acpi_lock);
Signature · 1 line
2.30.2
author maintainer
#66 lore raw

PATCH 60/101 gpiolib: cdev: zero padding during conversion to gpioline_info_changed

In reply to Sasha Levin

From: Gabriel Knezek <[email protected]>

[ Upstream commit cb8f63b8cbf39845244f3ccae43bb7e63bd70543 ]

When userspace requests a GPIO v1 line info changed event, lineinfo_watch_read() populates and returns the gpioline_info_changed structure. It contains 5 words of padding at the end which are not initialized before being returned to userspace.

Zero the structure in gpio_v2_line_info_change_to_v1() before populating its contents.

Fixes aad955842d1c ("gpiolib: cdev: support GPIO_V2_GET_LINEINFO_IOCTL and GPIO_V2_GET_LINEINFO_WATCH_IOCTL")
---
 drivers/gpio/gpiolib-cdev.c | 1 +
 1 file changed, 1 insertion(+)
drivers/gpio/gpiolib-cdev.c +1 -0
index ade3ecf2ee49..2613881a66e6 100644
@@ -1865,6 +1865,7 @@ static void gpio_v2_line_info_changed_to_v1(
 		struct gpio_v2_line_info_changed *lic_v2,
 		struct gpioline_info_changed *lic_v1)
 {
+	memset(lic_v1, 0, sizeof(*lic_v1));
 	gpio_v2_line_info_to_v1(&lic_v2->info, &lic_v1->info);
 	lic_v1->timestamp = lic_v2->timestamp_ns;
 	lic_v1->event_type = lic_v2->event_type;
Signature · 1 line
2.30.2
author maintainer
#67 lore raw

PATCH 61/101 scsi: sd: Call sd_revalidate_disk() for ioctl(BLKRRPART)

In reply to Sasha Levin

From: Christoph Hellwig <[email protected]>

[ Upstream commit d1b7f92035c6fb42529ada531e2cbf3534544c82 ]

While the disk state has nothing to do with partitions, BLKRRPART is used to force a full revalidate after things like a disk format for historical reasons. Restore that behavior.

---
 drivers/scsi/sd.c | 22 ++++++++++++++++++----
 1 file changed, 18 insertions(+), 4 deletions(-)
drivers/scsi/sd.c +18 -4
index 20a6564f87d9..01f87bcab3dd 100644
@@ -1389,6 +1389,22 @@ static void sd_uninit_command(struct scsi_cmnd *SCpnt)
 	}
 }
 
+static bool sd_need_revalidate(struct block_device *bdev,
+		struct scsi_disk *sdkp)
+{
+	if (sdkp->device->removable || sdkp->write_prot) {
+		if (bdev_check_media_change(bdev))
+			return true;
+	}
+
+	/*
+	 * Force a full rescan after ioctl(BLKRRPART).  While the disk state has
+	 * nothing to do with partitions, BLKRRPART is used to force a full
+	 * revalidate after things like a format for historical reasons.
+	 */
+	return test_bit(GD_NEED_PART_SCAN, &bdev->bd_disk->state);
+}
+
 /**
  *	sd_open - open a scsi disk device
  *	@bdev: Block device of the scsi disk to open
@@ -1425,10 +1441,8 @@ static int sd_open(struct block_device *bdev, fmode_t mode)
 	if (!scsi_block_when_processing_errors(sdev))
 		goto error_out;
 
-	if (sdev->removable || sdkp->write_prot) {
-		if (bdev_check_media_change(bdev))
-			sd_revalidate_disk(bdev->bd_disk);
-	}
+	if (sd_need_revalidate(bdev, sdkp))
+		sd_revalidate_disk(bdev->bd_disk);
 
 	/*
 	 * If the drive is empty, just let the open fail.
Signature · 1 line
2.30.2
author maintainer
#68 lore raw

PATCH 62/101 nilfs2: fix memory leak in nilfs_sysfs_delete_device_group

In reply to Sasha Levin

From: Pavel Skripkin <[email protected]>

[ Upstream commit 8fd0c1b0647a6bda4067ee0cd61e8395954b6f28 ]

My local syzbot instance hit memory leak in nilfs2. The problem was in missing kobject_put() in nilfs_sysfs_delete_device_group().

kobject_del() does not call kobject_cleanup() for passed kobject and it leads to leaking duped kobject name if kobject_put() was not called.

Fail log:

  BUG: memory leak
  unreferenced object 0xffff8880596171e0 (size 8):
  comm "syz-executor379", pid 8381, jiffies 4294980258 (age 21.100s)
  hex dump (first 8 bytes):
    6c 6f 6f 70 30 00 00 00                          loop0...
  backtrace:
     kstrdup+0x36/0x70 mm/util.c:60
     kstrdup_const+0x53/0x80 mm/util.c:83
     kvasprintf_const+0x108/0x190 lib/kasprintf.c:48
     kobject_set_name_vargs+0x56/0x150 lib/kobject.c:289
     kobject_add_varg lib/kobject.c:384 [inline]
     kobject_init_and_add+0xc9/0x160 lib/kobject.c:473
     nilfs_sysfs_create_device_group+0x150/0x800 fs/nilfs2/sysfs.c:999
     init_nilfs+0xe26/0x12b0 fs/nilfs2/the_nilfs.c:637
---
 fs/nilfs2/sysfs.c | 1 +
 1 file changed, 1 insertion(+)
fs/nilfs2/sysfs.c +1 -0
index 303d71430bdd..9c6c0e2e5880 100644
@@ -1053,6 +1053,7 @@ void nilfs_sysfs_delete_device_group(struct the_nilfs *nilfs)
 	nilfs_sysfs_delete_superblock_group(nilfs);
 	nilfs_sysfs_delete_segctor_group(nilfs);
 	kobject_del(&nilfs->ns_dev_kobj);
+	kobject_put(&nilfs->ns_dev_kobj);
 	kfree(nilfs->ns_dev_subgroups);
 }
 
Signature · 1 line
2.30.2
author maintainer
#69 lore raw

PATCH 63/101 s390/stack: fix possible register corruption with stack switch helper

In reply to Sasha Levin

From: Heiko Carstens <[email protected]>

commit 67147e96a332b56c7206238162771d82467f86c0 upstream.

The CALL_ON_STACK macro is used to call a C function from inline assembly, and therefore must consider the C ABI, which says that only registers 6-13, and 15 are non-volatile (restored by the called function).

The inline assembly incorrectly marks all registers used to pass parameters to the called function as read-only input operands, instead of operands that are read and written to. This might result in register corruption depending on usage, compiler, and compile options.

Fix this by marking all operands used to pass parameters as read/write operands. To keep the code simple even register 6, if used, is marked as read-write operand.

Fixes ff340d2472ec ("s390: add stack switch helper")
Cc <[email protected]> # 4.20
---
 arch/s390/include/asm/stacktrace.h | 18 +++++++++++-------
 1 file changed, 11 insertions(+), 7 deletions(-)
arch/s390/include/asm/stacktrace.h +11 -7
index ee056f4a4fa3..ee582896b6a3 100644
@@ -90,12 +90,16 @@ struct stack_frame {
 	CALL_ARGS_4(arg1, arg2, arg3, arg4);				\
 	register unsigned long r4 asm("6") = (unsigned long)(arg5)
 
-#define CALL_FMT_0 "=&d" (r2) :
-#define CALL_FMT_1 "+&d" (r2) :
-#define CALL_FMT_2 CALL_FMT_1 "d" (r3),
-#define CALL_FMT_3 CALL_FMT_2 "d" (r4),
-#define CALL_FMT_4 CALL_FMT_3 "d" (r5),
-#define CALL_FMT_5 CALL_FMT_4 "d" (r6),
+/*
+ * To keep this simple mark register 2-6 as being changed (volatile)
+ * by the called function, even though register 6 is saved/nonvolatile.
+ */
+#define CALL_FMT_0 "=&d" (r2)
+#define CALL_FMT_1 "+&d" (r2)
+#define CALL_FMT_2 CALL_FMT_1, "+&d" (r3)
+#define CALL_FMT_3 CALL_FMT_2, "+&d" (r4)
+#define CALL_FMT_4 CALL_FMT_3, "+&d" (r5)
+#define CALL_FMT_5 CALL_FMT_4, "+&d" (r6)
 
 #define CALL_CLOBBER_5 "0", "1", "14", "cc", "memory"
 #define CALL_CLOBBER_4 CALL_CLOBBER_5
@@ -117,7 +121,7 @@ struct stack_frame {
 		"	brasl	14,%[_fn]\n"				\
 		"	la	15,0(%[_prev])\n"			\
 		: [_prev] "=&a" (prev), CALL_FMT_##nr			\
-		  [_stack] "R" (stack),					\
+		: [_stack] "R" (stack),					\
 		  [_bc] "i" (offsetof(struct stack_frame, back_chain)),	\
 		  [_frame] "d" (frame),					\
 		  [_fn] "X" (fn) : CALL_CLOBBER_##nr);			\
Signature · 1 line
2.30.2
author maintainer
#70 lore raw

PATCH 64/101 KVM: do not allow mapping valid but non-reference-counted pages

In reply to Sasha Levin

From: Nicholas Piggin <[email protected]>

commit f8be156be163a052a067306417cd0ff679068c97 upstream.

It's possible to create a region which maps valid but non-refcounted pages (e.g., tail pages of non-compound higher order allocations). These host pages can then be returned by gfn_to_page, gfn_to_pfn, etc., family of APIs, which take a reference to the page, which takes it from 0 to 1. When the reference is dropped, this will free the page incorrectly.

Fix this by only taking a reference on valid pages if it was non-zero, which indicates it is participating in normal refcounting (and can be released with put_page).

This addresses CVE-2021-22543.

---
 virt/kvm/kvm_main.c | 19 +++++++++++++++++--
 1 file changed, 17 insertions(+), 2 deletions(-)
virt/kvm/kvm_main.c +17 -2
index f446c36f5800..1353439691cf 100644
@@ -1883,6 +1883,13 @@ static bool vma_is_valid(struct vm_area_struct *vma, bool write_fault)
 	return true;
 }
 
+static int kvm_try_get_pfn(kvm_pfn_t pfn)
+{
+	if (kvm_is_reserved_pfn(pfn))
+		return 1;
+	return get_page_unless_zero(pfn_to_page(pfn));
+}
+
 static int hva_to_pfn_remapped(struct vm_area_struct *vma,
 			       unsigned long addr, bool *async,
 			       bool write_fault, bool *writable,
@@ -1932,13 +1939,21 @@ static int hva_to_pfn_remapped(struct vm_area_struct *vma,
 	 * Whoever called remap_pfn_range is also going to call e.g.
 	 * unmap_mapping_range before the underlying pages are freed,
 	 * causing a call to our MMU notifier.
+	 *
+	 * Certain IO or PFNMAP mappings can be backed with valid
+	 * struct pages, but be allocated without refcounting e.g.,
+	 * tail pages of non-compound higher order allocations, which
+	 * would then underflow the refcount when the caller does the
+	 * required put_page. Don't allow those pages here.
 	 */ 
-	kvm_get_pfn(pfn);
+	if (!kvm_try_get_pfn(pfn))
+		r = -EFAULT;
 
 out:
 	pte_unmap_unlock(ptep, ptl);
 	*p_pfn = pfn;
-	return 0;
+
+	return r;
 }
 
 /*
Signature · 1 line
2.30.2
author maintainer
#71 lore raw

PATCH 65/101 i2c: robotfuzz-osif: fix control-request directions

In reply to Sasha Levin

From: Johan Hovold <[email protected]>

commit 4ca070ef0dd885616ef294d269a9bf8e3b258e1a upstream.

The direction of the pipe argument must match the request-type direction bit or control requests may fail depending on the host-controller-driver implementation.

Control transfers without a data stage are treated as OUT requests by the USB stack and should be using usb_sndctrlpipe(). Failing to do so will now trigger a warning.

Fix the OSIFI2C_SET_BIT_RATE and OSIFI2C_STOP requests which erroneously used the osif_usb_read() helper and set the IN direction bit.

Reported-by [email protected]
Fixes 83e53a8f120f ("i2c: Add bus driver for for OSIF USB i2c device.")
---
 drivers/i2c/busses/i2c-robotfuzz-osif.c | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/i2c/busses/i2c-robotfuzz-osif.c +2 -2
index a39f7d092797..66dfa211e736 100644
@@ -83,7 +83,7 @@ static int osif_xfer(struct i2c_adapter *adapter, struct i2c_msg *msgs,
 			}
 		}
 
-		ret = osif_usb_read(adapter, OSIFI2C_STOP, 0, 0, NULL, 0);
+		ret = osif_usb_write(adapter, OSIFI2C_STOP, 0, 0, NULL, 0);
 		if (ret) {
 			dev_err(&adapter->dev, "failure sending STOP\n");
 			return -EREMOTEIO;
@@ -153,7 +153,7 @@ static int osif_probe(struct usb_interface *interface,
 	 * Set bus frequency. The frequency is:
 	 * 120,000,000 / ( 16 + 2 * div * 4^prescale).
 	 * Using dev = 52, prescale = 0 give 100KHz */
-	ret = osif_usb_read(&priv->adapter, OSIFI2C_SET_BIT_RATE, 52, 0,
+	ret = osif_usb_write(&priv->adapter, OSIFI2C_SET_BIT_RATE, 52, 0,
 			    NULL, 0);
 	if (ret) {
 		dev_err(&interface->dev, "failure sending bit rate");
Signature · 1 line
2.30.2
author maintainer
#72 lore raw

PATCH 66/101 ceph: must hold snap_rwsem when filling inode for async create

In reply to Sasha Levin

From: Jeff Layton <[email protected]>

commit 27171ae6a0fdc75571e5bf3d0961631a1e4fb765 upstream.

...and add a lockdep assertion for it to ceph_fill_inode().

Fixes 9a8d03ca2e2c3 ("ceph: attempt to do async create when possible")
---
 fs/ceph/file.c  | 3 +++
 fs/ceph/inode.c | 2 ++
 2 files changed, 5 insertions(+)
fs/ceph/file.c +3 -0
index 209535d5b8d3..3d2e3dd4ee01 100644
@@ -578,6 +578,7 @@ static int ceph_finish_async_create(struct inode *dir, struct dentry *dentry,
 	struct ceph_inode_info *ci = ceph_inode(dir);
 	struct inode *inode;
 	struct timespec64 now;
+	struct ceph_mds_client *mdsc = ceph_sb_to_mdsc(dir->i_sb);
 	struct ceph_vino vino = { .ino = req->r_deleg_ino,
 				  .snap = CEPH_NOSNAP };
 
@@ -615,8 +616,10 @@ static int ceph_finish_async_create(struct inode *dir, struct dentry *dentry,
 
 	ceph_file_layout_to_legacy(lo, &in.layout);
 
+	down_read(&mdsc->snap_rwsem);
 	ret = ceph_fill_inode(inode, NULL, &iinfo, NULL, req->r_session,
 			      req->r_fmode, NULL);
+	up_read(&mdsc->snap_rwsem);
 	if (ret) {
 		dout("%s failed to fill inode: %d\n", __func__, ret);
 		ceph_dir_clear_complete(dir);
fs/ceph/inode.c +2 -0
index 346fcdfcd3e9..57cd78e942c0 100644
@@ -762,6 +762,8 @@ int ceph_fill_inode(struct inode *inode, struct page *locked_page,
 	bool new_version = false;
 	bool fill_inline = false;
 
+	lockdep_assert_held(&mdsc->snap_rwsem);
+
 	dout("%s %p ino %llx.%llx v %llu had %llu\n", __func__,
 	     inode, ceph_vinop(inode), le64_to_cpu(info->version),
 	     ci->i_version);
Signature · 1 line
2.30.2
author maintainer
#73 lore raw

PATCH 67/101 kthread_worker: split code for canceling the delayed work timer

In reply to Sasha Levin

From: Petr Mladek <[email protected]>

commit 34b3d5344719d14fd2185b2d9459b3abcb8cf9d8 upstream.

Patch series "kthread_worker: Fix race between kthread_mod_delayed_work() and kthread_cancel_delayed_work_sync()".

This patchset fixes the race between kthread_mod_delayed_work() and kthread_cancel_delayed_work_sync() including proper return value handling.

This patch (of 2):

Simple code refactoring as a preparation step for fixing a race between kthread_mod_delayed_work() and kthread_cancel_delayed_work_sync().

It does not modify the existing behavior.

---
 kernel/kthread.c | 46 +++++++++++++++++++++++++++++-----------------
 1 file changed, 29 insertions(+), 17 deletions(-)
kernel/kthread.c +29 -17
index 5edf7e19ab26..415417b76bfb 100644
@@ -1043,6 +1043,33 @@ void kthread_flush_work(struct kthread_work *work)
 }
 EXPORT_SYMBOL_GPL(kthread_flush_work);
 
+/*
+ * Make sure that the timer is neither set nor running and could
+ * not manipulate the work list_head any longer.
+ *
+ * The function is called under worker->lock. The lock is temporary
+ * released but the timer can't be set again in the meantime.
+ */
+static void kthread_cancel_delayed_work_timer(struct kthread_work *work,
+					      unsigned long *flags)
+{
+	struct kthread_delayed_work *dwork =
+		container_of(work, struct kthread_delayed_work, work);
+	struct kthread_worker *worker = work->worker;
+
+	/*
+	 * del_timer_sync() must be called to make sure that the timer
+	 * callback is not running. The lock must be temporary released
+	 * to avoid a deadlock with the callback. In the meantime,
+	 * any queuing is blocked by setting the canceling counter.
+	 */
+	work->canceling++;
+	raw_spin_unlock_irqrestore(&worker->lock, *flags);
+	del_timer_sync(&dwork->timer);
+	raw_spin_lock_irqsave(&worker->lock, *flags);
+	work->canceling--;
+}
+
 /*
  * This function removes the work from the worker queue. Also it makes sure
  * that it won't get queued later via the delayed work's timer.
@@ -1057,23 +1084,8 @@ static bool __kthread_cancel_work(struct kthread_work *work, bool is_dwork,
 				  unsigned long *flags)
 {
 	/* Try to cancel the timer if exists. */
-	if (is_dwork) {
-		struct kthread_delayed_work *dwork =
-			container_of(work, struct kthread_delayed_work, work);
-		struct kthread_worker *worker = work->worker;
-
-		/*
-		 * del_timer_sync() must be called to make sure that the timer
-		 * callback is not running. The lock must be temporary released
-		 * to avoid a deadlock with the callback. In the meantime,
-		 * any queuing is blocked by setting the canceling counter.
-		 */
-		work->canceling++;
-		raw_spin_unlock_irqrestore(&worker->lock, *flags);
-		del_timer_sync(&dwork->timer);
-		raw_spin_lock_irqsave(&worker->lock, *flags);
-		work->canceling--;
-	}
+	if (is_dwork)
+		kthread_cancel_delayed_work_timer(work, flags);
 
 	/*
 	 * Try to remove the work from a worker list. It might either
Signature · 1 line
2.30.2
author maintainer
#74 lore raw

PATCH 68/101 kthread: prevent deadlock when kthread_mod_delayed_work() races with kthread_cancel_delayed_work_sync()

In reply to Sasha Levin

From: Petr Mladek <[email protected]>

commit 5fa54346caf67b4b1b10b1f390316ae466da4d53 upstream.

The system might hang with the following backtrace:

	schedule+0x80/0x100
	schedule_timeout+0x48/0x138
	wait_for_common+0xa4/0x134
	wait_for_completion+0x1c/0x2c
	kthread_flush_work+0x114/0x1cc
	kthread_cancel_work_sync.llvm.16514401384283632983+0xe8/0x144
	kthread_cancel_delayed_work_sync+0x18/0x2c
	xxxx_pm_notify+0xb0/0xd8
	blocking_notifier_call_chain_robust+0x80/0x194
	pm_notifier_call_chain_robust+0x28/0x4c
	suspend_prepare+0x40/0x260
	enter_state+0x80/0x3f4
	pm_suspend+0x60/0xdc
	state_store+0x108/0x144
	kobj_attr_store+0x38/0x88
	sysfs_kf_write+0x64/0xc0
	kernfs_fop_write_iter+0x108/0x1d0
	vfs_write+0x2f4/0x368
	ksys_write+0x7c/0xec

It is caused by the following race between kthread_mod_delayed_work() and kthread_cancel_delayed_work_sync():

CPU0 CPU1

Context: Thread A Context: Thread B

kthread_mod_delayed_work()
spin_lock()
__kthread_cancel_work()

     spin_unlock()
     del_timer_sync()
				kthread_cancel_delayed_work_sync()
				  spin_lock()
				  __kthread_cancel_work()
				    spin_unlock()
				    del_timer_sync()
				    spin_lock()

				  work->canceling++
				  spin_unlock
     spin_lock()
   queue_delayed_work()
     // dwork is put into the worker->delayed_work_list

spin_unlock()

				  kthread_flush_work()
     // flush_work is put at the tail of the dwork

				    wait_for_completion()

Context: IRQ

  kthread_delayed_work_timer_fn()
    spin_lock()
    list_del_init(&work->node);
    spin_unlock()

BANG: flush_work is not longer linked and will never get proceed.

The problem is that kthread_mod_delayed_work() checks work->canceling flag before canceling the timer.

A simple solution is to (re)check work->canceling after __kthread_cancel_work(). But then it is not clear what should be returned when __kthread_cancel_work() removed the work from the queue (list) and it can't queue it again with the new @delay.

The return value might be used for reference counting. The caller has to know whether a new work has been queued or an existing one was replaced.

The proper solution is that kthread_mod_delayed_work() will remove the work from the queue (list) _only_ when work->canceling is not set. The flag must be checked after the timer is stopped and the remaining operations can be done under worker->lock.

Note that kthread_mod_delayed_work() could remove the timer and then bail out. It is fine. The other canceling caller needs to cancel the timer as well. The important thing is that the queue (list) manipulation is done atomically under worker->lock.

Fixes 9a6b06c8d9a220860468a ("kthread: allow to modify delayed kthread work")
Cc Minchan Kim <[email protected]>
Cc Nathan Chancellor <[email protected]>
Cc Nick Desaulniers <[email protected]>
Cc Oleg Nesterov <[email protected]>
Cc Tejun Heo <[email protected]>
---
 kernel/kthread.c | 35 ++++++++++++++++++++++++-----------
 1 file changed, 24 insertions(+), 11 deletions(-)
kernel/kthread.c +24 -11
index 415417b76bfb..36be4364b313 100644
@@ -1071,8 +1071,11 @@ static void kthread_cancel_delayed_work_timer(struct kthread_work *work,
 }
 
 /*
- * This function removes the work from the worker queue. Also it makes sure
- * that it won't get queued later via the delayed work's timer.
+ * This function removes the work from the worker queue.
+ *
+ * It is called under worker->lock. The caller must make sure that
+ * the timer used by delayed work is not running, e.g. by calling
+ * kthread_cancel_delayed_work_timer().
  *
  * The work might still be in use when this function finishes. See the
  * current_work proceed by the worker.
@@ -1080,13 +1083,8 @@ static void kthread_cancel_delayed_work_timer(struct kthread_work *work,
  * Return: %true if @work was pending and successfully canceled,
  *	%false if @work was not pending
  */
-static bool __kthread_cancel_work(struct kthread_work *work, bool is_dwork,
-				  unsigned long *flags)
+static bool __kthread_cancel_work(struct kthread_work *work)
 {
-	/* Try to cancel the timer if exists. */
-	if (is_dwork)
-		kthread_cancel_delayed_work_timer(work, flags);
-
 	/*
 	 * Try to remove the work from a worker list. It might either
 	 * be from worker->work_list or from worker->delayed_work_list.
@@ -1139,11 +1137,23 @@ bool kthread_mod_delayed_work(struct kthread_worker *worker,
 	/* Work must not be used with >1 worker, see kthread_queue_work() */
 	WARN_ON_ONCE(work->worker != worker);
 
-	/* Do not fight with another command that is canceling this work. */
+	/*
+	 * Temporary cancel the work but do not fight with another command
+	 * that is canceling the work as well.
+	 *
+	 * It is a bit tricky because of possible races with another
+	 * mod_delayed_work() and cancel_delayed_work() callers.
+	 *
+	 * The timer must be canceled first because worker->lock is released
+	 * when doing so. But the work can be removed from the queue (list)
+	 * only when it can be queued again so that the return value can
+	 * be used for reference counting.
+	 */
+	kthread_cancel_delayed_work_timer(work, &flags);
 	if (work->canceling)
 		goto out;
+	ret = __kthread_cancel_work(work);
 
-	ret = __kthread_cancel_work(work, true, &flags);
 fast_queue:
 	__kthread_queue_delayed_work(worker, dwork, delay);
 out:
@@ -1165,7 +1175,10 @@ static bool __kthread_cancel_work_sync(struct kthread_work *work, bool is_dwork)
 	/* Work must not be used with >1 worker, see kthread_queue_work(). */
 	WARN_ON_ONCE(work->worker != worker);
 
-	ret = __kthread_cancel_work(work, is_dwork, &flags);
+	if (is_dwork)
+		kthread_cancel_delayed_work_timer(work, &flags);
+
+	ret = __kthread_cancel_work(work);
 
 	if (worker->current_work != work)
 		goto out_fast;
Signature · 1 line
2.30.2
author maintainer
#78 lore raw

PATCH 72/101 mm/rmap: remove unneeded semicolon in page_not_mapped()

In reply to Sasha Levin

From: Miaohe Lin <[email protected]>

[ Upstream commit e0af87ff7afcde2660be44302836d2d5618185af ]

Remove extra semicolon without any functional change intended.

---
 mm/rmap.c | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)
mm/rmap.c +1 -1
index 6657000b18d4..5858639443b4 100644
@@ -1763,7 +1763,7 @@ bool try_to_unmap(struct page *page, enum ttu_flags flags)
 static int page_not_mapped(struct page *page)
 {
 	return !page_mapped(page);
-};
+}
 
 /**
  * try_to_munlock - try to munlock a page
Signature · 1 line
2.30.2
author maintainer
#77 lore raw

PATCH 71/101 mm: add VM_WARN_ON_ONCE_PAGE() macro

In reply to Sasha Levin

From: Alex Shi <[email protected]>

[ Upstream commit a4055888629bc0467d12d912cd7c90acdf3d9b12 part ]

Add VM_WARN_ON_ONCE_PAGE() macro.

Note on stable backport: original commit was titled mm/memcg: warning on !memcg after readahead page charged which included uses of this macro in mm/memcontrol.c: here omitted.

---
 include/linux/mmdebug.h | 13 +++++++++++++
 1 file changed, 13 insertions(+)
include/linux/mmdebug.h +13 -0
index 2ad72d2c8cc5..5d0767cb424a 100644
@@ -37,6 +37,18 @@ void dump_mm(const struct mm_struct *mm);
 			BUG();						\
 		}							\
 	} while (0)
+#define VM_WARN_ON_ONCE_PAGE(cond, page)	({			\
+	static bool __section(".data.once") __warned;			\
+	int __ret_warn_once = !!(cond);					\
+									\
+	if (unlikely(__ret_warn_once && !__warned)) {			\
+		dump_page(page, "VM_WARN_ON_ONCE_PAGE(" __stringify(cond)")");\
+		__warned = true;					\
+		WARN_ON(1);						\
+	}								\
+	unlikely(__ret_warn_once);					\
+})
+
 #define VM_WARN_ON(cond) (void)WARN_ON(cond)
 #define VM_WARN_ON_ONCE(cond) (void)WARN_ON_ONCE(cond)
 #define VM_WARN_ONCE(cond, format...) (void)WARN_ONCE(cond, format)
@@ -48,6 +60,7 @@ void dump_mm(const struct mm_struct *mm);
 #define VM_BUG_ON_MM(cond, mm) VM_BUG_ON(cond)
 #define VM_WARN_ON(cond) BUILD_BUG_ON_INVALID(cond)
 #define VM_WARN_ON_ONCE(cond) BUILD_BUG_ON_INVALID(cond)
+#define VM_WARN_ON_ONCE_PAGE(cond, page)  BUILD_BUG_ON_INVALID(cond)
 #define VM_WARN_ONCE(cond, format...) BUILD_BUG_ON_INVALID(cond)
 #define VM_WARN(cond, format...) BUILD_BUG_ON_INVALID(cond)
 #endif
Signature · 1 line
2.30.2
author maintainer
#79 lore raw

PATCH 73/101 mm/rmap: use page_not_mapped in try_to_unmap()

In reply to Sasha Levin

From: Miaohe Lin <[email protected]>

[ Upstream commit b7e188ec98b1644ff70a6d3624ea16aadc39f5e0 ]

page_mapcount_is_zero() calculates accurately how many mappings a hugepage has in order to check against 0 only. This is a waste of cpu time. We can do this via page_not_mapped() to save some possible atomic_read cycles. Remove the function page_mapcount_is_zero() as it's not used anymore and move page_not_mapped() above try_to_unmap() to avoid identifier undeclared compilation error.

---
 mm/rmap.c | 11 +++--------
 1 file changed, 3 insertions(+), 8 deletions(-)
mm/rmap.c +3 -8
index 5858639443b4..38573cb93578 100644
@@ -1716,9 +1716,9 @@ static bool invalid_migration_vma(struct vm_area_struct *vma, void *arg)
 	return vma_is_temporary_stack(vma);
 }
 
-static int page_mapcount_is_zero(struct page *page)
+static int page_not_mapped(struct page *page)
 {
-	return !total_mapcount(page);
+	return !page_mapped(page);
 }
 
 /**
@@ -1736,7 +1736,7 @@ bool try_to_unmap(struct page *page, enum ttu_flags flags)
 	struct rmap_walk_control rwc = {
 		.rmap_one = try_to_unmap_one,
 		.arg = (void *)flags,
-		.done = page_mapcount_is_zero,
+		.done = page_not_mapped,
 		.anon_lock = page_lock_anon_vma_read,
 	};
 
@@ -1760,11 +1760,6 @@ bool try_to_unmap(struct page *page, enum ttu_flags flags)
 	return !page_mapcount(page) ? true : false;
 }
 
-static int page_not_mapped(struct page *page)
-{
-	return !page_mapped(page);
-}
-
 /**
  * try_to_munlock - try to munlock a page
  * @page: the page to be munlocked
Signature · 1 line
2.30.2
author maintainer
#81 lore raw

PATCH 75/101 mm/thp: fix __split_huge_pmd_locked() on shmem migration entry

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

[ Upstream commit 99fa8a48203d62b3743d866fc48ef6abaee682be ]

Patch series "mm/thp: fix THP splitting unmap BUGs and related", v10.

Here is v2 batch of long-standing THP bug fixes that I had not got around to sending before, but prompted now by Wang Yugui's report
https://lore.kernel.org/linux-mm/[email protected]/

Wang Yugui has tested a rollup of these fixes applied to 5.10.39, and they have done no harm, but have *not* fixed that issue: something more is needed and I have no idea of what.

This patch (of 7):

Stressing huge tmpfs page migration racing hole punch often crashed on the VM_BUG_ON(!pmd_present) in pmdp_huge_clear_flush(), with DEBUG_VM=y kernel; or shortly afterwards, on a bad dereference in __split_huge_pmd_locked() when DEBUG_VM=n. They forgot to allow for pmd migration entries in the non-anonymous case.

Full disclosure: those particular experiments were on a kernel with more relaxed mmap_lock and i_mmap_rwsem locking, and were not repeated on the vanilla kernel: it is conceivable that stricter locking happens to avoid those cases, or makes them less likely; but __split_huge_pmd_locked() already allowed for pmd migration entries when handling anonymous THPs, so this commit brings the shmem and file THP handling into line.

And while there: use old_pmd rather than _pmd, as in the following blocks; and make it clearer to the eye that the !vma_is_anonymous() block is self-contained, making an early return after accounting for unmapping.

Fixes e71769ae5260 ("mm: enable thp migration for shmem thp")
Cc Kirill A. Shutemov <[email protected]>
Cc Yang Shi <[email protected]>
Cc Wang Yugui <[email protected]>
Cc "Matthew Wilcox (Oracle)" <[email protected]>
Cc Naoya Horiguchi <[email protected]>
Cc Alistair Popple <[email protected]>
Cc Ralph Campbell <[email protected]>
Cc Zi Yan <[email protected]>
Cc Miaohe Lin <[email protected]>
Cc Minchan Kim <[email protected]>
Cc Jue Wang <[email protected]>
Cc Peter Xu <[email protected]>
Cc Jan Kara <[email protected]>
Cc Shakeel Butt <[email protected]>
Cc Oscar Salvador <[email protected]>

Note on stable backport: this commit made intervening cleanups in pmdp_huge_clear_flush() redundant: here it's rediffed to skip them.

---
 mm/huge_memory.c     | 27 ++++++++++++++++++---------
 mm/pgtable-generic.c |  4 ++--
 2 files changed, 20 insertions(+), 11 deletions(-)
mm/huge_memory.c +18 -9
index d9ade23ac2b2..be0ad6f7981e 100644
@@ -2031,7 +2031,7 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 	count_vm_event(THP_SPLIT_PMD);
 
 	if (!vma_is_anonymous(vma)) {
-		_pmd = pmdp_huge_clear_flush_notify(vma, haddr, pmd);
+		old_pmd = pmdp_huge_clear_flush_notify(vma, haddr, pmd);
 		/*
 		 * We are going to unmap this huge page. So
 		 * just go ahead and zap it
@@ -2040,16 +2040,25 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 			zap_deposited_table(mm, pmd);
 		if (vma_is_special_huge(vma))
 			return;
-		page = pmd_page(_pmd);
-		if (!PageDirty(page) && pmd_dirty(_pmd))
-			set_page_dirty(page);
-		if (!PageReferenced(page) && pmd_young(_pmd))
-			SetPageReferenced(page);
-		page_remove_rmap(page, true);
-		put_page(page);
+		if (unlikely(is_pmd_migration_entry(old_pmd))) {
+			swp_entry_t entry;
+
+			entry = pmd_to_swp_entry(old_pmd);
+			page = migration_entry_to_page(entry);
+		} else {
+			page = pmd_page(old_pmd);
+			if (!PageDirty(page) && pmd_dirty(old_pmd))
+				set_page_dirty(page);
+			if (!PageReferenced(page) && pmd_young(old_pmd))
+				SetPageReferenced(page);
+			page_remove_rmap(page, true);
+			put_page(page);
+		}
 		add_mm_counter(mm, mm_counter_file(page), -HPAGE_PMD_NR);
 		return;
-	} else if (pmd_trans_huge(*pmd) && is_huge_zero_pmd(*pmd)) {
+	}
+
+	if (pmd_trans_huge(*pmd) && is_huge_zero_pmd(*pmd)) {
 		/*
 		 * FIXME: Do we want to invalidate secondary mmu by calling
 		 * mmu_notifier_invalidate_range() see comments below inside
mm/pgtable-generic.c +2 -2
index 9578db83e312..4e640baf9794 100644
@@ -135,8 +135,8 @@ pmd_t pmdp_huge_clear_flush(struct vm_area_struct *vma, unsigned long address,
 {
 	pmd_t pmd;
 	VM_BUG_ON(address & ~HPAGE_PMD_MASK);
-	VM_BUG_ON((pmd_present(*pmdp) && !pmd_trans_huge(*pmdp) &&
-			   !pmd_devmap(*pmdp)) || !pmd_present(*pmdp));
+	VM_BUG_ON(pmd_present(*pmdp) && !pmd_trans_huge(*pmdp) &&
+			   !pmd_devmap(*pmdp));
 	pmd = pmdp_huge_get_and_clear(vma->vm_mm, address, pmdp);
 	flush_pmd_tlb_range(vma, address, address + HPAGE_PMD_SIZE);
 	return pmd;
Signature · 1 line
2.30.2
author maintainer
#80 lore raw

PATCH 74/101 mm, thp: use head page in __migration_entry_wait()

In reply to Sasha Levin

From: Xu Yu <[email protected]>

commit ffc90cbb2970ab88b66ea51dd580469eede57b67 upstream.

We notice that hung task happens in a corner but practical scenario when CONFIG_PREEMPT_NONE is enabled, as follows.

Process 0 Process 1 Process 2..Inf
split_huge_page_to_list
unmap_page
split_huge_pmd_address
__migration_entry_wait(head)
__migration_entry_wait(tail)
remap_page (roll back)
remove_migration_ptes
rmap_walk_anon
cond_resched

Where __migration_entry_wait(tail) is occurred in kernel space, e.g., copy_to_user in fstat, which will immediately fault again without rescheduling, and thus occupy the cpu fully.

When there are too many processes performing __migration_entry_wait on tail page, remap_page will never be done after cond_resched.

This makes __migration_entry_wait operate on the compound head page, thus waits for remap_page to complete, whether the THP is split successfully or roll back.

Note that put_and_wait_on_page_locked helps to drop the page reference acquired with get_page_unless_zero, as soon as the page is on the wait queue, before actually waiting. So splitting the THP is only prevented for a brief interval.

---
 mm/migrate.c | 1 +
 1 file changed, 1 insertion(+)
mm/migrate.c +1 -0
index 7982256a5125..278e6f3fa62c 100644
@@ -326,6 +326,7 @@ void __migration_entry_wait(struct mm_struct *mm, pte_t *ptep,
 		goto out;
 
 	page = migration_entry_to_page(entry);
+	page = compound_head(page);
 
 	/*
 	 * Once page cache replacement of page migration started, page_count
Signature · 1 line
2.30.2
author maintainer
#75 lore raw

PATCH 69/101 x86/fpu: Preserve supervisor states in sanitize_restored_user_xstate()

In reply to Sasha Levin

From: Thomas Gleixner <[email protected]>

commit 9301982c424a003c0095bf157154a85bf5322bd0 upstream.

sanitize_restored_user_xstate() preserves the supervisor states only when the fx_only argument is zero, which allows unprivileged user space to put supervisor states back into init state.

Preserve them unconditionally.

[ bp: Fix a typo or two in the text. ]

Fixes 5d6b6a6f9b5c ("x86/fpu/xstate: Update sanitize_restored_xstate() for supervisor xstates")
---
 arch/x86/kernel/fpu/signal.c | 26 ++++++++------------------
 1 file changed, 8 insertions(+), 18 deletions(-)
arch/x86/kernel/fpu/signal.c +8 -18
index ec3ae3054792..b7b92cdf3add 100644
@@ -221,28 +221,18 @@ sanitize_restored_user_xstate(union fpregs_state *state,
 
 	if (use_xsave()) {
 		/*
-		 * Note: we don't need to zero the reserved bits in the
-		 * xstate_header here because we either didn't copy them at all,
-		 * or we checked earlier that they aren't set.
+		 * Clear all feature bits which are not set in
+		 * user_xfeatures and clear all extended features
+		 * for fx_only mode.
 		 */
+		u64 mask = fx_only ? XFEATURE_MASK_FPSSE : user_xfeatures;
 
 		/*
-		 * 'user_xfeatures' might have bits clear which are
-		 * set in header->xfeatures. This represents features that
-		 * were in init state prior to a signal delivery, and need
-		 * to be reset back to the init state.  Clear any user
-		 * feature bits which are set in the kernel buffer to get
-		 * them back to the init state.
-		 *
-		 * Supervisor state is unchanged by input from userspace.
-		 * Ensure supervisor state bits stay set and supervisor
-		 * state is not modified.
+		 * Supervisor state has to be preserved. The sigframe
+		 * restore can only modify user features, i.e. @mask
+		 * cannot contain them.
 		 */
-		if (fx_only)
-			header->xfeatures = XFEATURE_MASK_FPSSE;
-		else
-			header->xfeatures &= user_xfeatures |
-					     xfeatures_mask_supervisor();
+		header->xfeatures &= mask | xfeatures_mask_supervisor();
 	}
 
 	if (use_fxsr()) {
Signature · 1 line
2.30.2
author maintainer
#76 lore raw

PATCH 70/101 x86/fpu: Make init_fpstate correct with optimized XSAVE

In reply to Sasha Levin

From: Thomas Gleixner <[email protected]>

commit f9dfb5e390fab2df9f7944bb91e7705aba14cd26 upstream.

The XSAVE init code initializes all enabled and supported components with XRSTOR(S) to init state. Then it XSAVEs the state of the components back into init_fpstate which is used in several places to fill in the init state of components.

This works correctly with XSAVE, but not with XSAVEOPT and XSAVES because those use the init optimization and skip writing state of components which are in init state. So init_fpstate.xsave still contains all zeroes after this operation.

There are two ways to solve that:

1) Use XSAVE unconditionally, but that requires to reshuffle the buffer when XSAVES is enabled because XSAVES uses compacted format.

2) Save the components which are known to have a non-zero init state by other means.

Looking deeper, #2 is the right thing to do because all components the kernel supports have all-zeroes init state except the legacy features (FP, SSE). Those cannot be hard coded because the states are not identical on all CPUs, but they can be saved with FXSAVE which avoids all conditionals.

Use FXSAVE to save the legacy FP/SSE components in init_fpstate along with a BUILD_BUG_ON() which reminds developers to validate that a newly added component has all zeroes init state. As a bonus remove the now unused copy_xregs_to_kernel_booting() crutch.

The XSAVE and reshuffle method can still be implemented in the unlikely case that components are added which have a non-zero init state and no other means to save them. For now, FXSAVE is just simple and good enough.

[ bp: Fix a typo or two in the text. ]

Fixes 6bad06b76892 ("x86, xsave: Use xsaveopt in context-switch path when supported")
---
 arch/x86/include/asm/fpu/internal.h | 30 ++++++---------------
 arch/x86/kernel/fpu/xstate.c        | 41 ++++++++++++++++++++++++++---
 2 files changed, 46 insertions(+), 25 deletions(-)
arch/x86/include/asm/fpu/internal.h +8 -22
index fdee23ea4e17..16bf4d4a8159 100644
@@ -204,6 +204,14 @@ static inline void copy_fxregs_to_kernel(struct fpu *fpu)
 		asm volatile("fxsaveq %[fx]" : [fx] "=m" (fpu->state.fxsave));
 }
 
+static inline void fxsave(struct fxregs_state *fx)
+{
+	if (IS_ENABLED(CONFIG_X86_32))
+		asm volatile( "fxsave %[fx]" : [fx] "=m" (*fx));
+	else
+		asm volatile("fxsaveq %[fx]" : [fx] "=m" (*fx));
+}
+
 /* These macros all use (%edi)/(%rdi) as the single memory argument. */
 #define XSAVE		".byte " REX_PREFIX "0x0f,0xae,0x27"
 #define XSAVEOPT	".byte " REX_PREFIX "0x0f,0xae,0x37"
@@ -268,28 +276,6 @@ static inline void copy_fxregs_to_kernel(struct fpu *fpu)
 		     : "D" (st), "m" (*st), "a" (lmask), "d" (hmask)	\
 		     : "memory")
 
-/*
- * This function is called only during boot time when x86 caps are not set
- * up and alternative can not be used yet.
- */
-static inline void copy_xregs_to_kernel_booting(struct xregs_state *xstate)
-{
-	u64 mask = xfeatures_mask_all;
-	u32 lmask = mask;
-	u32 hmask = mask >> 32;
-	int err;
-
-	WARN_ON(system_state != SYSTEM_BOOTING);
-
-	if (boot_cpu_has(X86_FEATURE_XSAVES))
-		XSTATE_OP(XSAVES, xstate, lmask, hmask, err);
-	else
-		XSTATE_OP(XSAVE, xstate, lmask, hmask, err);
-
-	/* We should never fault when copying to a kernel buffer: */
-	WARN_ON_FPU(err);
-}
-
 /*
  * This function is called only during boot time when x86 caps are not set
  * up and alternative can not be used yet.
arch/x86/kernel/fpu/xstate.c +38 -3

60 lines not on this page · raw message

Signature · 1 line
2.30.2
author maintainer
#82 lore raw

PATCH 76/101 mm/thp: make is_huge_zero_pmd() safe and quicker

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit 3b77e8c8cde581dadab9a0f1543a347e24315f11 upstream.

Most callers of is_huge_zero_pmd() supply a pmd already verified present; but a few (notably zap_huge_pmd()) do not - it might be a pmd migration entry, in which the pfn is encoded differently from a present
pmd: which might pass the is_huge_zero_pmd() test (though not on x86, since L1TF forced us to protect against that); or perhaps even crash in pmd_page() applied to a swap-like entry.

Make it safe by adding pmd_present() check into is_huge_zero_pmd() itself; and make it quicker by saving huge_zero_pfn, so that is_huge_zero_pmd() will not need to do that pmd_page() lookup each time.

__split_huge_pmd_locked() checked pmd_trans_huge() before: that worked, but is unnecessary now that is_huge_zero_pmd() checks present.

Fixes e71769ae5260 ("mm: enable thp migration for shmem thp")
Cc Alistair Popple <[email protected]>
Cc Jan Kara <[email protected]>
Cc Jue Wang <[email protected]>
Cc "Matthew Wilcox (Oracle)" <[email protected]>
Cc Miaohe Lin <[email protected]>
Cc Minchan Kim <[email protected]>
Cc Naoya Horiguchi <[email protected]>
Cc Oscar Salvador <[email protected]>
Cc Peter Xu <[email protected]>
Cc Ralph Campbell <[email protected]>
Cc Shakeel Butt <[email protected]>
Cc Wang Yugui <[email protected]>
Cc Zi Yan <[email protected]>
---
 include/linux/huge_mm.h | 8 +++++++-
 mm/huge_memory.c        | 5 ++++-
 2 files changed, 11 insertions(+), 2 deletions(-)
include/linux/huge_mm.h +7 -1
index 0365aa97f8e7..ff55be011739 100644
@@ -297,6 +297,7 @@ struct page *follow_devmap_pud(struct vm_area_struct *vma, unsigned long addr,
 extern vm_fault_t do_huge_pmd_numa_page(struct vm_fault *vmf, pmd_t orig_pmd);
 
 extern struct page *huge_zero_page;
+extern unsigned long huge_zero_pfn;
 
 static inline bool is_huge_zero_page(struct page *page)
 {
@@ -305,7 +306,7 @@ static inline bool is_huge_zero_page(struct page *page)
 
 static inline bool is_huge_zero_pmd(pmd_t pmd)
 {
-	return is_huge_zero_page(pmd_page(pmd));
+	return READ_ONCE(huge_zero_pfn) == pmd_pfn(pmd) && pmd_present(pmd);
 }
 
 static inline bool is_huge_zero_pud(pud_t pud)
@@ -451,6 +452,11 @@ static inline bool is_huge_zero_page(struct page *page)
 	return false;
 }
 
+static inline bool is_huge_zero_pmd(pmd_t pmd)
+{
+	return false;
+}
+
 static inline bool is_huge_zero_pud(pud_t pud)
 {
 	return false;
mm/huge_memory.c +4 -1
index be0ad6f7981e..7d8159be4736 100644
@@ -61,6 +61,7 @@ static struct shrinker deferred_split_shrinker;
 
 static atomic_t huge_zero_refcount;
 struct page *huge_zero_page __read_mostly;
+unsigned long huge_zero_pfn __read_mostly = ~0UL;
 
 bool transparent_hugepage_enabled(struct vm_area_struct *vma)
 {
@@ -97,6 +98,7 @@ static struct page *get_huge_zero_page(void)
 		__free_pages(zero_page, compound_order(zero_page));
 		goto retry;
 	}
+	WRITE_ONCE(huge_zero_pfn, page_to_pfn(zero_page));
 
 	/* We take additional reference here. It will be put back by shrinker */
 	atomic_set(&huge_zero_refcount, 2);
@@ -146,6 +148,7 @@ static unsigned long shrink_huge_zero_page_scan(struct shrinker *shrink,
 	if (atomic_cmpxchg(&huge_zero_refcount, 1, 0) == 1) {
 		struct page *zero_page = xchg(&huge_zero_page, NULL);
 		BUG_ON(zero_page == NULL);
+		WRITE_ONCE(huge_zero_pfn, ~0UL);
 		__free_pages(zero_page, compound_order(zero_page));
 		return HPAGE_PMD_NR;
 	}
@@ -2058,7 +2061,7 @@ static void __split_huge_pmd_locked(struct vm_area_struct *vma, pmd_t *pmd,
 		return;
 	}
 
-	if (pmd_trans_huge(*pmd) && is_huge_zero_pmd(*pmd)) {
+	if (is_huge_zero_pmd(*pmd)) {
 		/*
 		 * FIXME: Do we want to invalidate secondary mmu by calling
 		 * mmu_notifier_invalidate_range() see comments below inside
Signature · 1 line
2.30.2
author maintainer
#83 lore raw

PATCH 77/101 mm/thp: try_to_unmap() use TTU_SYNC for safe splitting

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit 732ed55823fc3ad998d43b86bf771887bcc5ec67 upstream.

Stressing huge tmpfs often crashed on unmap_page()'s VM_BUG_ON_PAGE (!unmap_success): with dump_page() showing mapcount:1, but then its raw struct page output showing _mapcount ffffffff i.e. mapcount 0.

And even if that particular VM_BUG_ON_PAGE(!unmap_success) is removed, it is immediately followed by a VM_BUG_ON_PAGE(compound_mapcount(head)), and further down an IS_ENABLED(CONFIG_DEBUG_VM) total_mapcount BUG():
all indicative of some mapcount difficulty in development here perhaps. But the !CONFIG_DEBUG_VM path handles the failures correctly and silently.

I believe the problem is that once a racing unmap has cleared pte or pmd, try_to_unmap_one() may skip taking the page table lock, and emerge from try_to_unmap() before the racing task has reached decrementing mapcount.

Instead of abandoning the unsafe VM_BUG_ON_PAGE(), and the ones that follow, use PVMW_SYNC in try_to_unmap_one() in this case: adding TTU_SYNC to the options, and passing that from unmap_page().

When CONFIG_DEBUG_VM, or for non-debug too? Consensus is to do the same for both: the slight overhead added should rarely matter, except perhaps if splitting sparsely-populated multiply-mapped shmem. Once confident that bugs are fixed, TTU_SYNC here can be removed, and the race tolerated.

Fixes fec89c109f3a ("thp: rewrite freeze_page()/unfreeze_page() with generic rmap walkers")
Cc Alistair Popple <[email protected]>
Cc Jan Kara <[email protected]>
Cc Jue Wang <[email protected]>
Cc Kirill A. Shutemov <[email protected]>
Cc "Matthew Wilcox (Oracle)" <[email protected]>
Cc Miaohe Lin <[email protected]>
Cc Minchan Kim <[email protected]>
Cc Naoya Horiguchi <[email protected]>
Cc Oscar Salvador <[email protected]>
Cc Peter Xu <[email protected]>
Cc Ralph Campbell <[email protected]>
Cc Shakeel Butt <[email protected]>
Cc Wang Yugui <[email protected]>
Cc Yang Shi <[email protected]>
Cc Zi Yan <[email protected]>
---
 include/linux/rmap.h |  1 +
 mm/huge_memory.c     |  2 +-
 mm/page_vma_mapped.c | 11 +++++++++++
 mm/rmap.c            | 17 ++++++++++++++++-
 4 files changed, 29 insertions(+), 2 deletions(-)
include/linux/rmap.h +1 -0
index def5c62c93b3..8d04e7deedc6 100644
@@ -91,6 +91,7 @@ enum ttu_flags {
 
 	TTU_SPLIT_HUGE_PMD	= 0x4,	/* split huge PMD if any */
 	TTU_IGNORE_MLOCK	= 0x8,	/* ignore mlock */
+	TTU_SYNC		= 0x10,	/* avoid racy checks with PVMW_SYNC */
 	TTU_IGNORE_HWPOISON	= 0x20,	/* corrupted page is recoverable */
 	TTU_BATCH_FLUSH		= 0x40,	/* Batch TLB flushes where possible
 					 * and caller guarantees they will
mm/huge_memory.c +1 -1
index 7d8159be4736..14136390d641 100644
@@ -2342,7 +2342,7 @@ void vma_adjust_trans_huge(struct vm_area_struct *vma,
 
 static void unmap_page(struct page *page)
 {
-	enum ttu_flags ttu_flags = TTU_IGNORE_MLOCK |
+	enum ttu_flags ttu_flags = TTU_IGNORE_MLOCK | TTU_SYNC |
 		TTU_RMAP_LOCKED | TTU_SPLIT_HUGE_PMD;
 	bool unmap_success;
 
mm/page_vma_mapped.c +11 -0
index 5e77b269c330..3ba2bedc5794 100644
@@ -211,6 +211,17 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			pvmw->ptl = NULL;
 		}
 	} else if (!pmd_present(pmde)) {
+		/*
+		 * If PVMW_SYNC, take and drop THP pmd lock so that we
+		 * cannot return prematurely, while zap_huge_pmd() has
+		 * cleared *pmd but not decremented compound_mapcount().
+		 */
+		if ((pvmw->flags & PVMW_SYNC) &&
+		    PageTransCompound(pvmw->page)) {
+			spinlock_t *ptl = pmd_lock(mm, pvmw->pmd);
+
+			spin_unlock(ptl);
+		}
 		return false;
 	}
 	if (!map_pte(pvmw))
mm/rmap.c +16 -1
index 38573cb93578..2b721412bd7f 100644
@@ -1385,6 +1385,15 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 	struct mmu_notifier_range range;
 	enum ttu_flags flags = (enum ttu_flags)(long)arg;
 
+	/*
+	 * When racing against e.g. zap_pte_range() on another cpu,
+	 * in between its ptep_get_and_clear_full() and page_remove_rmap(),
+	 * try_to_unmap() may return false when it is about to become true,
+	 * if page table locking is skipped: use TTU_SYNC to wait for that.
+	 */
+	if (flags & TTU_SYNC)
+		pvmw.flags = PVMW_SYNC;
+
 	/* munlock has nothing to gain from examining un-locked vmas */
 	if ((flags & TTU_MUNLOCK) && !(vma->vm_flags & VM_LOCKED))
 		return true;
@@ -1757,7 +1766,13 @@ bool try_to_unmap(struct page *page, enum ttu_flags flags)
 	else
 		rmap_walk(page, &rwc);
 
-	return !page_mapcount(page) ? true : false;
+	/*
+	 * When racing against e.g. zap_pte_range() on another cpu,
+	 * in between its ptep_get_and_clear_full() and page_remove_rmap(),
+	 * try_to_unmap() may return false when it is about to become true,
+	 * if page table locking is skipped: use TTU_SYNC to wait for that.
+	 */
+	return !page_mapcount(page);
 }
 
 /**
Signature · 1 line
2.30.2
author maintainer
#84 lore raw

PATCH 78/101 mm/thp: fix vma_address() if virtual address below file offset

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit 494334e43c16d63b878536a26505397fce6ff3a2 upstream.

Running certain tests with a DEBUG_VM kernel would crash within hours, on the total_mapcount BUG() in split_huge_page_to_list(), while trying to free up some memory by punching a hole in a shmem huge page: split's try_to_unmap() was unable to find all the mappings of the page (which, on a !DEBUG_VM kernel, would then keep the huge page pinned in memory).

When that BUG() was changed to a WARN(), it would later crash on the VM_BUG_ON_VMA(end < vma->vm_start || start >= vma->vm_end, vma) in mm/internal.h:vma_address(), used by rmap_walk_file() for try_to_unmap().

vma_address() is usually correct, but there's a wraparound case when the vm_start address is unusually low, but vm_pgoff not so low:
vma_address() chooses max(start, vma->vm_start), but that decides on the wrong address, because start has become almost ULONG_MAX.

Rewrite vma_address() to be more careful about vm_pgoff; move the VM_BUG_ON_VMA() out of it, returning -EFAULT for errors, so that it can be safely used from page_mapped_in_vma() and page_address_in_vma() too.

Add vma_address_end() to apply similar care to end address calculation, in page_vma_mapped_walk() and page_mkclean_one() and try_to_unmap_one(); though it raises a question of whether callers would do better to supply pvmw->end to page_vma_mapped_walk() - I chose not, for a smaller patch.

An irritation is that their apparent generality breaks down on KSM pages, which cannot be located by the page->index that page_to_pgoff()
uses: as commit 4b0ece6fa016 ("mm: migrate: fix remove_migration_pte() for ksm pages") once discovered. I dithered over the best thing to do about that, and have ended up with a VM_BUG_ON_PAGE(PageKsm) in both vma_address() and vma_address_end(); though the only place in danger of using it on them was try_to_unmap_one().

Sidenote: vma_address() and vma_address_end() now use compound_nr() on a head page, instead of thp_size(): to make the right calculation on a hugetlbfs page, whether or not THPs are configured. try_to_unmap() is used on hugetlbfs pages, but perhaps the wrong calculation never mattered.

Fixes a8fa41ad2f6f ("mm, rmap: check all VMAs that PTE-mapped THP can be part of")
Cc Alistair Popple <[email protected]>
Cc Jan Kara <[email protected]>
Cc Jue Wang <[email protected]>
Cc "Matthew Wilcox (Oracle)" <[email protected]>
Cc Miaohe Lin <[email protected]>
Cc Minchan Kim <[email protected]>
Cc Naoya Horiguchi <[email protected]>
Cc Oscar Salvador <[email protected]>
Cc Peter Xu <[email protected]>
Cc Ralph Campbell <[email protected]>
Cc Shakeel Butt <[email protected]>
Cc Wang Yugui <[email protected]>
Cc Yang Shi <[email protected]>
Cc Zi Yan <[email protected]>
---
 mm/internal.h        | 53 ++++++++++++++++++++++++++++++++------------
 mm/page_vma_mapped.c | 16 +++++--------
 mm/rmap.c            | 16 ++++++-------
 3 files changed, 53 insertions(+), 32 deletions(-)
mm/internal.h +39 -14
index c43ccdddb0f6..840b8a330b9a 100644
@@ -379,27 +379,52 @@ static inline void mlock_migrate_page(struct page *newpage, struct page *page)
 extern pmd_t maybe_pmd_mkwrite(pmd_t pmd, struct vm_area_struct *vma);
 
 /*
- * At what user virtual address is page expected in @vma?
+ * At what user virtual address is page expected in vma?
+ * Returns -EFAULT if all of the page is outside the range of vma.
+ * If page is a compound head, the entire compound page is considered.
  */
 static inline unsigned long
-__vma_address(struct page *page, struct vm_area_struct *vma)
+vma_address(struct page *page, struct vm_area_struct *vma)
 {
-	pgoff_t pgoff = page_to_pgoff(page);
-	return vma->vm_start + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
+	pgoff_t pgoff;
+	unsigned long address;
+
+	VM_BUG_ON_PAGE(PageKsm(page), page);	/* KSM page->index unusable */
+	pgoff = page_to_pgoff(page);
+	if (pgoff >= vma->vm_pgoff) {
+		address = vma->vm_start +
+			((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
+		/* Check for address beyond vma (or wrapped through 0?) */
+		if (address < vma->vm_start || address >= vma->vm_end)
+			address = -EFAULT;
+	} else if (PageHead(page) &&
+		   pgoff + compound_nr(page) - 1 >= vma->vm_pgoff) {
+		/* Test above avoids possibility of wrap to 0 on 32-bit */
+		address = vma->vm_start;
+	} else {
+		address = -EFAULT;
+	}
+	return address;
 }
 
+/*
+ * Then at what user virtual address will none of the page be found in vma?
+ * Assumes that vma_address() already returned a good starting address.
+ * If page is a compound head, the entire compound page is considered.
+ */
 static inline unsigned long
-vma_address(struct page *page, struct vm_area_struct *vma)
+vma_address_end(struct page *page, struct vm_area_struct *vma)
 {
-	unsigned long start, end;
-
-	start = __vma_address(page, vma);
-	end = start + thp_size(page) - PAGE_SIZE;
-
-	/* page should be within @vma mapping range */
-	VM_BUG_ON_VMA(end < vma->vm_start || start >= vma->vm_end, vma);
-
-	return max(start, vma->vm_start);
+	pgoff_t pgoff;
+	unsigned long address;
+
+	VM_BUG_ON_PAGE(PageKsm(page), page);	/* KSM page->index unusable */
+	pgoff = page_to_pgoff(page) + compound_nr(page);
+	address = vma->vm_start + ((pgoff - vma->vm_pgoff) << PAGE_SHIFT);
+	/* Check for address beyond vma (or wrapped through 0?) */
+	if (address < vma->vm_start || address > vma->vm_end)
+		address = vma->vm_end;
+	return address;
 }
 
 static inline struct file *maybe_unlock_mmap_for_io(struct vm_fault *vmf,
mm/page_vma_mapped.c +6 -10
index 3ba2bedc5794..a540af346f88 100644
@@ -227,18 +227,18 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 	if (!map_pte(pvmw))
 		goto next_pte;
 	while (1) {
+		unsigned long end;
+
 		if (check_pte(pvmw))
 			return true;
 next_pte:
 		/* Seek to next pte only makes sense for THP */
 		if (!PageTransHuge(pvmw->page) || PageHuge(pvmw->page))
 			return not_found(pvmw);
+		end = vma_address_end(pvmw->page, pvmw->vma);
 		do {
 			pvmw->address += PAGE_SIZE;
-			if (pvmw->address >= pvmw->vma->vm_end ||
-			    pvmw->address >=
-					__vma_address(pvmw->page, pvmw->vma) +
-					thp_size(pvmw->page))
+			if (pvmw->address >= end)
 				return not_found(pvmw);
 			/* Did we cross page table boundary? */
 			if (pvmw->address % PMD_SIZE == 0) {
@@ -276,14 +276,10 @@ int page_mapped_in_vma(struct page *page, struct vm_area_struct *vma)
 		.vma = vma,
 		.flags = PVMW_SYNC,
 	};
-	unsigned long start, end;
-
-	start = __vma_address(page, vma);
-	end = start + thp_size(page) - PAGE_SIZE;
 
-	if (unlikely(end < vma->vm_start || start >= vma->vm_end))
+	pvmw.address = vma_address(page, vma);
+	if (pvmw.address == -EFAULT)
 		return 0;
-	pvmw.address = max(start, vma->vm_start);
 	if (!page_vma_mapped_walk(&pvmw))
 		return 0;
 	page_vma_mapped_walk_done(&pvmw);
mm/rmap.c +8 -8
index 2b721412bd7f..62cd108bb2bd 100644
@@ -700,7 +700,6 @@ static bool should_defer_flush(struct mm_struct *mm, enum ttu_flags flags)
  */
 unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma)
 {
-	unsigned long address;
 	if (PageAnon(page)) {
 		struct anon_vma *page__anon_vma = page_anon_vma(page);
 		/*
@@ -715,10 +714,8 @@ unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma)
 			return -EFAULT;
 	} else
 		return -EFAULT;
-	address = __vma_address(page, vma);
-	if (unlikely(address < vma->vm_start || address >= vma->vm_end))
-		return -EFAULT;
-	return address;
+
+	return vma_address(page, vma);
 }
 
 pmd_t *mm_find_pmd(struct mm_struct *mm, unsigned long address)
@@ -912,7 +909,7 @@ static bool page_mkclean_one(struct page *page, struct vm_area_struct *vma,
 	 */
 	mmu_notifier_range_init(&range, MMU_NOTIFY_PROTECTION_PAGE,
 				0, vma, vma->vm_mm, address,
-				min(vma->vm_end, address + page_size(page)));
+				vma_address_end(page, vma));
 	mmu_notifier_invalidate_range_start(&range);
 
 	while (page_vma_mapped_walk(&pvmw)) {
@@ -1415,9 +1412,10 @@ static bool try_to_unmap_one(struct page *page, struct vm_area_struct *vma,
 	 * Note that the page can not be free in this function as call of
 	 * try_to_unmap() must hold a reference on the page.
 	 */
+	range.end = PageKsm(page) ?
+			address + PAGE_SIZE : vma_address_end(page, vma);
 	mmu_notifier_range_init(&range, MMU_NOTIFY_CLEAR, 0, vma, vma->vm_mm,
-				address,
-				min(vma->vm_end, address + page_size(page)));
+				address, range.end);
 	if (PageHuge(page)) {
 		/*
 		 * If sharing is possible, start and end will be adjusted
@@ -1869,6 +1867,7 @@ static void rmap_walk_anon(struct page *page, struct rmap_walk_control *rwc,
 		struct vm_area_struct *vma = avc->vma;
 		unsigned long address = vma_address(page, vma);
 
+		VM_BUG_ON_VMA(address == -EFAULT, vma);
 		cond_resched();
 
 		if (rwc->invalid_vma && rwc->invalid_vma(vma, rwc->arg))
@@ -1923,6 +1922,7 @@ static void rmap_walk_file(struct page *page, struct rmap_walk_control *rwc,
 			pgoff_start, pgoff_end) {
 		unsigned long address = vma_address(page, vma);
 
+		VM_BUG_ON_VMA(address == -EFAULT, vma);
 		cond_resched();
 
 		if (rwc->invalid_vma && rwc->invalid_vma(vma, rwc->arg))
Signature · 1 line
2.30.2
author maintainer
#85 lore raw

PATCH 79/101 mm/thp: fix page_address_in_vma() on file THP tails

In reply to Sasha Levin

From: Jue Wang <[email protected]>

commit 31657170deaf1d8d2f6a1955fbc6fa9d228be036 upstream.

Anon THP tails were already supported, but memory-failure may need to use page_address_in_vma() on file THP tails, which its page->mapping check did not permit: fix it.

hughd adds: no current usage is known to hit the issue, but this does fix a subtle trap in a general helper: best fixed in stable sooner than later.

---
 mm/rmap.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
mm/rmap.c +4 -4
index 62cd108bb2bd..14f84f70c557 100644
@@ -709,11 +709,11 @@ unsigned long page_address_in_vma(struct page *page, struct vm_area_struct *vma)
 		if (!vma->anon_vma || !page__anon_vma ||
 		    vma->anon_vma->root != page__anon_vma->root)
 			return -EFAULT;
-	} else if (page->mapping) {
-		if (!vma->vm_file || vma->vm_file->f_mapping != page->mapping)
-			return -EFAULT;
-	} else
+	} else if (!vma->vm_file) {
+		return -EFAULT;
+	} else if (vma->vm_file->f_mapping != compound_head(page)->mapping) {
 		return -EFAULT;
+	}
 
 	return vma_address(page, vma);
 }
Signature · 1 line
2.30.2
author maintainer
#86 lore raw

PATCH 80/101 mm/thp: unmap_mapping_page() to fix THP truncate_cleanup_page()

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

[ Upstream commit 22061a1ffabdb9c3385de159c5db7aac3a4df1cc ]

There is a race between THP unmapping and truncation, when truncate sees pmd_none() and skips the entry, after munmap's zap_huge_pmd() cleared it, but before its page_remove_rmap() gets to decrement
compound_mapcount: generating false "BUG: Bad page cache" reports that the page is still mapped when deleted. This commit fixes that, but not in the way I hoped.

The first attempt used try_to_unmap(page, TTU_SYNC|TTU_IGNORE_MLOCK) instead of unmap_mapping_range() in truncate_cleanup_page(): it has often been an annoyance that we usually call unmap_mapping_range() with no pages locked, but there apply it to a single locked page. try_to_unmap() looks more suitable for a single locked page.

However, try_to_unmap_one() contains a VM_BUG_ON_PAGE(!pvmw.pte,page):
it is used to insert THP migration entries, but not used to unmap THPs. Copy zap_huge_pmd() and add THP handling now? Perhaps, but their TLB needs are different, I'm too ignorant of the DAX cases, and couldn't decide how far to go for anon+swap. Set that aside.

The second attempt took a different tack: make no change in truncate.c, but modify zap_huge_pmd() to insert an invalidated huge pmd instead of clearing it initially, then pmd_clear() between page_remove_rmap() and unlocking at the end. Nice. But powerpc blows that approach out of the water, with its serialize_against_pte_lookup(), and interesting pgtable usage. It would need serious help to get working on powerpc (with a minor optimization issue on s390 too). Set that aside.

Just add an "if (page_mapped(page)) synchronize_rcu();" or other such delay, after unmapping in truncate_cleanup_page()? Perhaps, but though that's likely to reduce or eliminate the number of incidents, it would give less assurance of whether we had identified the problem correctly.

This successful iteration introduces "unmap_mapping_page(page)" instead of try_to_unmap(), and goes the usual unmap_mapping_range_tree() route, with an addition to details. Then zap_pmd_range() watches for this case, and does spin_unlock(pmd_lock) if so - just like page_vma_mapped_walk() now does in the PVMW_SYNC case. Not pretty, but safe.

Note that unmap_mapping_page() is doing a VM_BUG_ON(!PageLocked) to assert its interface; but currently that's only used to make sure that page->mapping is stable, and zap_pmd_range() doesn't care if the page is locked or not. Along these lines, in invalidate_inode_pages2_range() move the initial unmap_mapping_range() out from under page lock, before then calling unmap_mapping_page() under page lock if still mapped.

Fixes fc127da085c2 ("truncate: handle file thp")
Cc Alistair Popple <[email protected]>
Cc Jan Kara <[email protected]>
Cc Jue Wang <[email protected]>
Cc "Matthew Wilcox (Oracle)" <[email protected]>
Cc Miaohe Lin <[email protected]>
Cc Minchan Kim <[email protected]>
Cc Naoya Horiguchi <[email protected]>
Cc Oscar Salvador <[email protected]>
Cc Peter Xu <[email protected]>
Cc Ralph Campbell <[email protected]>
Cc Shakeel Butt <[email protected]>
Cc Wang Yugui <[email protected]>
Cc Zi Yan <[email protected]>

Note on stable backport: fixed up call to truncate_cleanup_page() in truncate_inode_pages_range().

---
 include/linux/mm.h |  3 +++
 mm/memory.c        | 41 +++++++++++++++++++++++++++++++++++++++++
 mm/truncate.c      | 43 +++++++++++++++++++------------------------
 3 files changed, 63 insertions(+), 24 deletions(-)
include/linux/mm.h +3 -0
index 5106db3ad1ce..289c26f055cd 100644
@@ -1648,6 +1648,7 @@ struct zap_details {
 	struct address_space *check_mapping;	/* Check page->mapping if set */
 	pgoff_t	first_index;			/* Lowest page->index to unmap */
 	pgoff_t last_index;			/* Highest page->index to unmap */
+	struct page *single_page;		/* Locked page to be unmapped */
 };
 
 struct page *vm_normal_page(struct vm_area_struct *vma, unsigned long addr,
@@ -1695,6 +1696,7 @@ extern vm_fault_t handle_mm_fault(struct vm_area_struct *vma,
 extern int fixup_user_fault(struct mm_struct *mm,
 			    unsigned long address, unsigned int fault_flags,
 			    bool *unlocked);
+void unmap_mapping_page(struct page *page);
 void unmap_mapping_pages(struct address_space *mapping,
 		pgoff_t start, pgoff_t nr, bool even_cows);
 void unmap_mapping_range(struct address_space *mapping,
@@ -1715,6 +1717,7 @@ static inline int fixup_user_fault(struct mm_struct *mm, unsigned long address,
 	BUG();
 	return -EFAULT;
 }
+static inline void unmap_mapping_page(struct page *page) { }
 static inline void unmap_mapping_pages(struct address_space *mapping,
 		pgoff_t start, pgoff_t nr, bool even_cows) { }
 static inline void unmap_mapping_range(struct address_space *mapping,
mm/memory.c +41 -0

54 lines not on this page · raw message

mm/truncate.c +19 -24

75 lines not on this page · raw message

Signature · 1 line
2.30.2
author maintainer
#87 lore raw

PATCH 81/101 mm: thp: replace DEBUG_VM BUG with VM_WARN when unmap fails for split

In reply to Sasha Levin

From: Yang Shi <[email protected]>

[ Upstream commit 504e070dc08f757bccaed6d05c0f53ecbfac8a23 ]

When debugging the bug reported by Wang Yugui [1], try_to_unmap() may fail, but the first VM_BUG_ON_PAGE() just checks page_mapcount() however it may miss the failure when head page is unmapped but other subpage is mapped. Then the second DEBUG_VM BUG() that check total mapcount would catch it. This may incur some confusion.

As this is not a fatal issue, so consolidate the two DEBUG_VM checks into one VM_WARN_ON_ONCE_PAGE().

[1] https://lore.kernel.org/linux-mm/[email protected]/

Note on stable backport: fixed up variables in split_huge_page_to_list().

---
 mm/huge_memory.c | 24 +++++++-----------------
 1 file changed, 7 insertions(+), 17 deletions(-)
mm/huge_memory.c +7 -17
index 14136390d641..6301ecc1f679 100644
@@ -2344,15 +2344,15 @@ static void unmap_page(struct page *page)
 {
 	enum ttu_flags ttu_flags = TTU_IGNORE_MLOCK | TTU_SYNC |
 		TTU_RMAP_LOCKED | TTU_SPLIT_HUGE_PMD;
-	bool unmap_success;
 
 	VM_BUG_ON_PAGE(!PageHead(page), page);
 
 	if (PageAnon(page))
 		ttu_flags |= TTU_SPLIT_FREEZE;
 
-	unmap_success = try_to_unmap(page, ttu_flags);
-	VM_BUG_ON_PAGE(!unmap_success, page);
+	try_to_unmap(page, ttu_flags);
+
+	VM_WARN_ON_ONCE_PAGE(page_mapped(page), page);
 }
 
 static void remap_page(struct page *page, unsigned int nr)
@@ -2642,7 +2642,7 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 	struct deferred_split *ds_queue = get_deferred_split_queue(head);
 	struct anon_vma *anon_vma = NULL;
 	struct address_space *mapping = NULL;
-	int count, mapcount, extra_pins, ret;
+	int extra_pins, ret;
 	unsigned long flags;
 	pgoff_t end;
 
@@ -2702,7 +2702,6 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 	}
 
 	unmap_page(head);
-	VM_BUG_ON_PAGE(compound_mapcount(head), head);
 
 	/* prevent PageLRU to go away from under us, and freeze lru stats */
 	spin_lock_irqsave(&pgdata->lru_lock, flags);
@@ -2721,9 +2720,7 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 
 	/* Prevent deferred_split_scan() touching ->_refcount */
 	spin_lock(&ds_queue->split_queue_lock);
-	count = page_count(head);
-	mapcount = total_mapcount(head);
-	if (!mapcount && page_ref_freeze(head, 1 + extra_pins)) {
+	if (page_ref_freeze(head, 1 + extra_pins)) {
 		if (!list_empty(page_deferred_list(head))) {
 			ds_queue->split_queue_len--;
 			list_del(page_deferred_list(head));
@@ -2739,16 +2736,9 @@ int split_huge_page_to_list(struct page *page, struct list_head *list)
 		__split_huge_page(page, list, end, flags);
 		ret = 0;
 	} else {
-		if (IS_ENABLED(CONFIG_DEBUG_VM) && mapcount) {
-			pr_alert("total_mapcount: %u, page_count(): %u\n",
-					mapcount, count);
-			if (PageTail(page))
-				dump_page(head, NULL);
-			dump_page(page, "total_mapcount(head) > 0");
-			BUG();
-		}
 		spin_unlock(&ds_queue->split_queue_lock);
-fail:		if (mapping)
+fail:
+		if (mapping)
 			xa_unlock(&mapping->i_pages);
 		spin_unlock_irqrestore(&pgdata->lru_lock, flags);
 		remap_page(head, thp_nr_pages(head));
Signature · 1 line
2.30.2
author maintainer
#89 lore raw

PATCH 83/101 mm: page_vma_mapped_walk(): settle PageHuge on entry

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit 6d0fd5987657cb0c9756ce684e3a74c0f6351728 upstream.

page_vma_mapped_walk() cleanup: get the hugetlbfs PageHuge case out of the way at the start, so no need to worry about it later.

---
 mm/page_vma_mapped.c | 12 ++++++++----
 1 file changed, 8 insertions(+), 4 deletions(-)
mm/page_vma_mapped.c +8 -4
index 3cd41168e802..4f04472effb7 100644
@@ -152,10 +152,11 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 	if (pvmw->pmd && !pvmw->pte)
 		return not_found(pvmw);
 
-	if (pvmw->pte)
-		goto next_pte;
-
 	if (unlikely(PageHuge(page))) {
+		/* The only possible mapping was handled on last iteration */
+		if (pvmw->pte)
+			return not_found(pvmw);
+
 		/* when pud is not present, pte will be NULL */
 		pvmw->pte = huge_pte_offset(mm, pvmw->address, page_size(page));
 		if (!pvmw->pte)
@@ -167,6 +168,9 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			return not_found(pvmw);
 		return true;
 	}
+
+	if (pvmw->pte)
+		goto next_pte;
 restart:
 	pgd = pgd_offset(mm, pvmw->address);
 	if (!pgd_present(*pgd))
@@ -232,7 +236,7 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			return true;
 next_pte:
 		/* Seek to next pte only makes sense for THP */
-		if (!PageTransHuge(page) || PageHuge(page))
+		if (!PageTransHuge(page))
 			return not_found(pvmw);
 		end = vma_address_end(page, pvmw->vma);
 		do {
Signature · 1 line
2.30.2
author maintainer
#88 lore raw

PATCH 82/101 mm: page_vma_mapped_walk(): use page for pvmw->page

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit f003c03bd29e6f46fef1b9a8e8d636ac732286d5 upstream.

Patch series "mm: page_vma_mapped_walk() cleanup and THP fixes".

I've marked all of these for stable: many are merely cleanups, but I think they are much better before the main fix than after.

This patch (of 11):

page_vma_mapped_walk() cleanup: sometimes the local copy of pvwm->page was used, sometimes pvmw->page itself: use the local copy "page" throughout.

---
 mm/page_vma_mapped.c | 9 ++++-----
 1 file changed, 4 insertions(+), 5 deletions(-)
mm/page_vma_mapped.c +4 -5
index a540af346f88..3cd41168e802 100644
@@ -155,7 +155,7 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 	if (pvmw->pte)
 		goto next_pte;
 
-	if (unlikely(PageHuge(pvmw->page))) {
+	if (unlikely(PageHuge(page))) {
 		/* when pud is not present, pte will be NULL */
 		pvmw->pte = huge_pte_offset(mm, pvmw->address, page_size(page));
 		if (!pvmw->pte)
@@ -216,8 +216,7 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 		 * cannot return prematurely, while zap_huge_pmd() has
 		 * cleared *pmd but not decremented compound_mapcount().
 		 */
-		if ((pvmw->flags & PVMW_SYNC) &&
-		    PageTransCompound(pvmw->page)) {
+		if ((pvmw->flags & PVMW_SYNC) && PageTransCompound(page)) {
 			spinlock_t *ptl = pmd_lock(mm, pvmw->pmd);
 
 			spin_unlock(ptl);
@@ -233,9 +232,9 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			return true;
 next_pte:
 		/* Seek to next pte only makes sense for THP */
-		if (!PageTransHuge(pvmw->page) || PageHuge(pvmw->page))
+		if (!PageTransHuge(page) || PageHuge(page))
 			return not_found(pvmw);
-		end = vma_address_end(pvmw->page, pvmw->vma);
+		end = vma_address_end(page, pvmw->vma);
 		do {
 			pvmw->address += PAGE_SIZE;
 			if (pvmw->address >= end)
Signature · 1 line
2.30.2
author maintainer
#91 lore raw

PATCH 85/101 mm: page_vma_mapped_walk(): prettify PVMW_MIGRATION block

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit e2e1d4076c77b3671cf8ce702535ae7dee3acf89 upstream.

page_vma_mapped_walk() cleanup: rearrange the !pmd_present() block to follow the same "return not_found, return not_found, return true" pattern as the block above it (note: returning not_found there is never premature, since existence or prior existence of huge pmd guarantees good alignment).

---
 mm/page_vma_mapped.c | 30 ++++++++++++++----------------
 1 file changed, 14 insertions(+), 16 deletions(-)
mm/page_vma_mapped.c +14 -16
index 821147eca0e1..a9cd4e4b8d95 100644
@@ -197,24 +197,22 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			if (pmd_page(pmde) != page)
 				return not_found(pvmw);
 			return true;
-		} else if (!pmd_present(pmde)) {
-			if (thp_migration_supported()) {
-				if (!(pvmw->flags & PVMW_MIGRATION))
-					return not_found(pvmw);
-				if (is_migration_entry(pmd_to_swp_entry(pmde))) {
-					swp_entry_t entry = pmd_to_swp_entry(pmde);
+		}
+		if (!pmd_present(pmde)) {
+			swp_entry_t entry;
 
-					if (migration_entry_to_page(entry) != page)
-						return not_found(pvmw);
-					return true;
-				}
-			}
-			return not_found(pvmw);
-		} else {
-			/* THP pmd was split under us: handle on pte level */
-			spin_unlock(pvmw->ptl);
-			pvmw->ptl = NULL;
+			if (!thp_migration_supported() ||
+			    !(pvmw->flags & PVMW_MIGRATION))
+				return not_found(pvmw);
+			entry = pmd_to_swp_entry(pmde);
+			if (!is_migration_entry(entry) ||
+			    migration_entry_to_page(entry) != page)
+				return not_found(pvmw);
+			return true;
 		}
+		/* THP pmd was split under us: handle on pte level */
+		spin_unlock(pvmw->ptl);
+		pvmw->ptl = NULL;
 	} else if (!pmd_present(pmde)) {
 		/*
 		 * If PVMW_SYNC, take and drop THP pmd lock so that we
Signature · 1 line
2.30.2
author maintainer
#90 lore raw

PATCH 84/101 mm: page_vma_mapped_walk(): use pmde for *pvmw->pmd

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit 3306d3119ceacc43ea8b141a73e21fea68eec30c upstream.

page_vma_mapped_walk() cleanup: re-evaluate pmde after taking lock, then use it in subsequent tests, instead of repeatedly dereferencing pointer.

---
 mm/page_vma_mapped.c | 11 ++++++-----
 1 file changed, 6 insertions(+), 5 deletions(-)
mm/page_vma_mapped.c +6 -5
index 4f04472effb7..821147eca0e1 100644
@@ -190,18 +190,19 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 	pmde = READ_ONCE(*pvmw->pmd);
 	if (pmd_trans_huge(pmde) || is_pmd_migration_entry(pmde)) {
 		pvmw->ptl = pmd_lock(mm, pvmw->pmd);
-		if (likely(pmd_trans_huge(*pvmw->pmd))) {
+		pmde = *pvmw->pmd;
+		if (likely(pmd_trans_huge(pmde))) {
 			if (pvmw->flags & PVMW_MIGRATION)
 				return not_found(pvmw);
-			if (pmd_page(*pvmw->pmd) != page)
+			if (pmd_page(pmde) != page)
 				return not_found(pvmw);
 			return true;
-		} else if (!pmd_present(*pvmw->pmd)) {
+		} else if (!pmd_present(pmde)) {
 			if (thp_migration_supported()) {
 				if (!(pvmw->flags & PVMW_MIGRATION))
 					return not_found(pvmw);
-				if (is_migration_entry(pmd_to_swp_entry(*pvmw->pmd))) {
-					swp_entry_t entry = pmd_to_swp_entry(*pvmw->pmd);
+				if (is_migration_entry(pmd_to_swp_entry(pmde))) {
+					swp_entry_t entry = pmd_to_swp_entry(pmde);
 
 					if (migration_entry_to_page(entry) != page)
 						return not_found(pvmw);
Signature · 1 line
2.30.2
author maintainer
#93 lore raw

PATCH 87/101 mm: page_vma_mapped_walk(): add a level of indentation

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit b3807a91aca7d21c05d5790612e49969117a72b9 upstream.

page_vma_mapped_walk() cleanup: add a level of indentation to much of the body, making no functional change in this commit, but reducing the later diff when this is all converted to a loop.

[[email protected]: : page_vma_mapped_walk(): add a level of indentation fix]
---
 mm/page_vma_mapped.c | 105 ++++++++++++++++++++++---------------------
 1 file changed, 55 insertions(+), 50 deletions(-)
mm/page_vma_mapped.c +55 -50
index 36b40fd23f94..3c3a7cb2de85 100644
@@ -172,62 +172,67 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 	if (pvmw->pte)
 		goto next_pte;
 restart:
-	pgd = pgd_offset(mm, pvmw->address);
-	if (!pgd_present(*pgd))
-		return false;
-	p4d = p4d_offset(pgd, pvmw->address);
-	if (!p4d_present(*p4d))
-		return false;
-	pud = pud_offset(p4d, pvmw->address);
-	if (!pud_present(*pud))
-		return false;
-	pvmw->pmd = pmd_offset(pud, pvmw->address);
-	/*
-	 * Make sure the pmd value isn't cached in a register by the
-	 * compiler and used as a stale value after we've observed a
-	 * subsequent update.
-	 */
-	pmde = READ_ONCE(*pvmw->pmd);
-	if (pmd_trans_huge(pmde) || is_pmd_migration_entry(pmde)) {
-		pvmw->ptl = pmd_lock(mm, pvmw->pmd);
-		pmde = *pvmw->pmd;
-		if (likely(pmd_trans_huge(pmde))) {
-			if (pvmw->flags & PVMW_MIGRATION)
-				return not_found(pvmw);
-			if (pmd_page(pmde) != page)
-				return not_found(pvmw);
-			return true;
-		}
-		if (!pmd_present(pmde)) {
-			swp_entry_t entry;
+	{
+		pgd = pgd_offset(mm, pvmw->address);
+		if (!pgd_present(*pgd))
+			return false;
+		p4d = p4d_offset(pgd, pvmw->address);
+		if (!p4d_present(*p4d))
+			return false;
+		pud = pud_offset(p4d, pvmw->address);
+		if (!pud_present(*pud))
+			return false;
 
-			if (!thp_migration_supported() ||
-			    !(pvmw->flags & PVMW_MIGRATION))
-				return not_found(pvmw);
-			entry = pmd_to_swp_entry(pmde);
-			if (!is_migration_entry(entry) ||
-			    migration_entry_to_page(entry) != page)
-				return not_found(pvmw);
-			return true;
-		}
-		/* THP pmd was split under us: handle on pte level */
-		spin_unlock(pvmw->ptl);
-		pvmw->ptl = NULL;
-	} else if (!pmd_present(pmde)) {
+		pvmw->pmd = pmd_offset(pud, pvmw->address);
 		/*
-		 * If PVMW_SYNC, take and drop THP pmd lock so that we
-		 * cannot return prematurely, while zap_huge_pmd() has
-		 * cleared *pmd but not decremented compound_mapcount().
+		 * Make sure the pmd value isn't cached in a register by the
+		 * compiler and used as a stale value after we've observed a
+		 * subsequent update.
 		 */
-		if ((pvmw->flags & PVMW_SYNC) && PageTransCompound(page)) {
-			spinlock_t *ptl = pmd_lock(mm, pvmw->pmd);
+		pmde = READ_ONCE(*pvmw->pmd);
 
-			spin_unlock(ptl);
+		if (pmd_trans_huge(pmde) || is_pmd_migration_entry(pmde)) {
+			pvmw->ptl = pmd_lock(mm, pvmw->pmd);
+			pmde = *pvmw->pmd;
+			if (likely(pmd_trans_huge(pmde))) {
+				if (pvmw->flags & PVMW_MIGRATION)
+					return not_found(pvmw);
+				if (pmd_page(pmde) != page)
+					return not_found(pvmw);
+				return true;
+			}
+			if (!pmd_present(pmde)) {
+				swp_entry_t entry;
+
+				if (!thp_migration_supported() ||
+				    !(pvmw->flags & PVMW_MIGRATION))
+					return not_found(pvmw);
+				entry = pmd_to_swp_entry(pmde);
+				if (!is_migration_entry(entry) ||
+				    migration_entry_to_page(entry) != page)
+					return not_found(pvmw);
+				return true;
+			}
+			/* THP pmd was split under us: handle on pte level */
+			spin_unlock(pvmw->ptl);
+			pvmw->ptl = NULL;
+		} else if (!pmd_present(pmde)) {
+			/*
+			 * If PVMW_SYNC, take and drop THP pmd lock so that we
+			 * cannot return prematurely, while zap_huge_pmd() has
+			 * cleared *pmd but not decremented compound_mapcount().
+			 */
+			if ((pvmw->flags & PVMW_SYNC) &&
+			    PageTransCompound(page)) {
+				spinlock_t *ptl = pmd_lock(mm, pvmw->pmd);
+
+				spin_unlock(ptl);
+			}
+			return false;
 		}
-		return false;
+		if (!map_pte(pvmw))
+			goto next_pte;
 	}
-	if (!map_pte(pvmw))
-		goto next_pte;
 	while (1) {
 		unsigned long end;
 
Signature · 1 line
2.30.2
author maintainer
#94 lore raw

PATCH 88/101 mm: page_vma_mapped_walk(): use goto instead of while (1)

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit 474466301dfd8b39a10c01db740645f3f7ae9a28 upstream.

page_vma_mapped_walk() cleanup: add a label this_pte, matching next_pte, and use "goto this_pte", in place of the "while (1)" loop at the end.

---
 mm/page_vma_mapped.c | 7 +++----
 1 file changed, 3 insertions(+), 4 deletions(-)
mm/page_vma_mapped.c +3 -4
index 3c3a7cb2de85..cb43affe6c76 100644
@@ -143,6 +143,7 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 {
 	struct mm_struct *mm = pvmw->vma->vm_mm;
 	struct page *page = pvmw->page;
+	unsigned long end;
 	pgd_t *pgd;
 	p4d_t *p4d;
 	pud_t *pud;
@@ -232,10 +233,7 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 		}
 		if (!map_pte(pvmw))
 			goto next_pte;
-	}
-	while (1) {
-		unsigned long end;
-
+this_pte:
 		if (check_pte(pvmw))
 			return true;
 next_pte:
@@ -264,6 +262,7 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			pvmw->ptl = pte_lockptr(mm, pvmw->pmd);
 			spin_lock(pvmw->ptl);
 		}
+		goto this_pte;
 	}
 }
 
Signature · 1 line
2.30.2
author maintainer
#92 lore raw

PATCH 86/101 mm: page_vma_mapped_walk(): crossing page table boundary

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit 448282487483d6fa5b2eeeafaa0acc681e544a9c upstream.

page_vma_mapped_walk() cleanup: adjust the test for crossing page table boundary - I believe pvmw->address is always page-aligned, but nothing else here assumed that; and remember to reset pvmw->pte to NULL after unmapping the page table, though I never saw any bug from that.

---
 mm/page_vma_mapped.c | 8 ++++----
 1 file changed, 4 insertions(+), 4 deletions(-)
mm/page_vma_mapped.c +4 -4
index a9cd4e4b8d95..36b40fd23f94 100644
@@ -243,16 +243,16 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			if (pvmw->address >= end)
 				return not_found(pvmw);
 			/* Did we cross page table boundary? */
-			if (pvmw->address % PMD_SIZE == 0) {
-				pte_unmap(pvmw->pte);
+			if ((pvmw->address & (PMD_SIZE - PAGE_SIZE)) == 0) {
 				if (pvmw->ptl) {
 					spin_unlock(pvmw->ptl);
 					pvmw->ptl = NULL;
 				}
+				pte_unmap(pvmw->pte);
+				pvmw->pte = NULL;
 				goto restart;
-			} else {
-				pvmw->pte++;
 			}
+			pvmw->pte++;
 		} while (pte_none(*pvmw->pte));
 
 		if (!pvmw->ptl) {
Signature · 1 line
2.30.2
author maintainer
#96 lore raw

PATCH 90/101 mm/thp: fix page_vma_mapped_walk() if THP mapped by ptes

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit a9a7504d9beaf395481faa91e70e2fd08f7a3dde upstream.

Running certain tests with a DEBUG_VM kernel would crash within hours, on the total_mapcount BUG() in split_huge_page_to_list(), while trying to free up some memory by punching a hole in a shmem huge page: split's try_to_unmap() was unable to find all the mappings of the page (which, on a !DEBUG_VM kernel, would then keep the huge page pinned in memory).

Crash dumps showed two tail pages of a shmem huge page remained mapped by pte: ptes in a non-huge-aligned vma of a gVisor process, at the end of a long unmapped range; and no page table had yet been allocated for the head of the huge page to be mapped into.

Although designed to handle these odd misaligned huge-page-mapped-by-pte cases, page_vma_mapped_walk() falls short by returning false prematurely when !pmd_present or !pud_present or !p4d_present or !pgd_present: there are cases when a huge page may span the boundary, with ptes present in the next.

Restructure page_vma_mapped_walk() as a loop to continue in these cases, while keeping its layout much as before. Add a step_forward() helper to advance pvmw->address across those boundaries: originally I tried to use mm's standard p?d_addr_end() macros, but hit the same crash 512 times less often: because of the way redundant levels are folded together, but folded differently in different configurations, it was just too difficult to use them correctly; and step_forward() is simpler anyway.

Fixes ace71a19cec5 ("mm: introduce page_vma_mapped_walk()")
Cc Alistair Popple <[email protected]>
Cc Matthew Wilcox <[email protected]>
Cc Peter Xu <[email protected]>
Cc Ralph Campbell <[email protected]>
Cc Wang Yugui <[email protected]>
Cc Will Deacon <[email protected]>
Cc Yang Shi <[email protected]>
Cc Zi Yan <[email protected]>
---
 mm/page_vma_mapped.c | 34 +++++++++++++++++++++++++---------
 1 file changed, 25 insertions(+), 9 deletions(-)
mm/page_vma_mapped.c +25 -9
index 5df88f3b454c..2ad76a3d871d 100644
@@ -115,6 +115,13 @@ static bool check_pte(struct page_vma_mapped_walk *pvmw)
 	return pfn_is_match(pvmw->page, pfn);
 }
 
+static void step_forward(struct page_vma_mapped_walk *pvmw, unsigned long size)
+{
+	pvmw->address = (pvmw->address + size) & ~(size - 1);
+	if (!pvmw->address)
+		pvmw->address = ULONG_MAX;
+}
+
 /**
  * page_vma_mapped_walk - check if @pvmw->page is mapped in @pvmw->vma at
  * @pvmw->address
@@ -182,16 +189,22 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 	if (pvmw->pte)
 		goto next_pte;
 restart:
-	{
+	do {
 		pgd = pgd_offset(mm, pvmw->address);
-		if (!pgd_present(*pgd))
-			return false;
+		if (!pgd_present(*pgd)) {
+			step_forward(pvmw, PGDIR_SIZE);
+			continue;
+		}
 		p4d = p4d_offset(pgd, pvmw->address);
-		if (!p4d_present(*p4d))
-			return false;
+		if (!p4d_present(*p4d)) {
+			step_forward(pvmw, P4D_SIZE);
+			continue;
+		}
 		pud = pud_offset(p4d, pvmw->address);
-		if (!pud_present(*pud))
-			return false;
+		if (!pud_present(*pud)) {
+			step_forward(pvmw, PUD_SIZE);
+			continue;
+		}
 
 		pvmw->pmd = pmd_offset(pud, pvmw->address);
 		/*
@@ -238,7 +251,8 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 
 				spin_unlock(ptl);
 			}
-			return false;
+			step_forward(pvmw, PMD_SIZE);
+			continue;
 		}
 		if (!map_pte(pvmw))
 			goto next_pte;
@@ -268,7 +282,9 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 			spin_lock(pvmw->ptl);
 		}
 		goto this_pte;
-	}
+	} while (pvmw->address < end);
+
+	return false;
 }
 
 /**
Signature · 1 line
2.30.2
author maintainer
#95 lore raw

PATCH 89/101 mm: page_vma_mapped_walk(): get vma_address_end() earlier

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit a765c417d876cc635f628365ec9aa6f09470069a upstream.

page_vma_mapped_walk() cleanup: get THP's vma_address_end() at the start, rather than later at next_pte.

It's a little unnecessary overhead on the first call, but makes for a simpler loop in the following commit.

---
 mm/page_vma_mapped.c | 13 +++++++++----
 1 file changed, 9 insertions(+), 4 deletions(-)
mm/page_vma_mapped.c +9 -4
index cb43affe6c76..5df88f3b454c 100644
@@ -170,6 +170,15 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 		return true;
 	}
 
+	/*
+	 * Seek to next pte only makes sense for THP.
+	 * But more important than that optimization, is to filter out
+	 * any PageKsm page: whose page->index misleads vma_address()
+	 * and vma_address_end() to disaster.
+	 */
+	end = PageTransCompound(page) ?
+		vma_address_end(page, pvmw->vma) :
+		pvmw->address + PAGE_SIZE;
 	if (pvmw->pte)
 		goto next_pte;
 restart:
@@ -237,10 +246,6 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 		if (check_pte(pvmw))
 			return true;
 next_pte:
-		/* Seek to next pte only makes sense for THP */
-		if (!PageTransHuge(page))
-			return not_found(pvmw);
-		end = vma_address_end(page, pvmw->vma);
 		do {
 			pvmw->address += PAGE_SIZE;
 			if (pvmw->address >= end)
Signature · 1 line
2.30.2
author maintainer
#97 lore raw

PATCH 91/101 mm/thp: another PVMW_SYNC fix in page_vma_mapped_walk()

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit a7a69d8ba88d8dcee7ef00e91d413a4bd003a814 upstream.

Aha! Shouldn't that quick scan over pte_none()s make sure that it holds ptlock in the PVMW_SYNC case? That too might have been responsible for BUGs or WARNs in split_huge_page_to_list() or its unmap_page(), though I've never seen any.

---
 mm/page_vma_mapped.c | 4 ++++
 1 file changed, 4 insertions(+)
mm/page_vma_mapped.c +4 -0
index 2ad76a3d871d..610ebbee787c 100644
@@ -275,6 +275,10 @@ bool page_vma_mapped_walk(struct page_vma_mapped_walk *pvmw)
 				goto restart;
 			}
 			pvmw->pte++;
+			if ((pvmw->flags & PVMW_SYNC) && !pvmw->ptl) {
+				pvmw->ptl = pte_lockptr(mm, pvmw->pmd);
+				spin_lock(pvmw->ptl);
+			}
 		} while (pte_none(*pvmw->pte));
 
 		if (!pvmw->ptl) {
Signature · 1 line
2.30.2
author maintainer
#98 lore raw

PATCH 92/101 mm, futex: fix shared futex pgoff on shmem huge page

In reply to Sasha Levin

From: Hugh Dickins <[email protected]>

commit fe19bd3dae3d15d2fbfdb3de8839a6ea0fe94264 upstream.

If more than one futex is placed on a shmem huge page, it can happen that waking the second wakes the first instead, and leaves the second
waiting: the key's shared.pgoff is wrong.

When 3.11 commit 13d60f4b6ab5 ("futex: Take hugepages into account when generating futex_key"), the only shared huge pages came from hugetlbfs, and the code added to deal with its exceptional page->index was put into hugetlb source. Then that was missed when 4.8 added shmem huge pages.

page_to_pgoff() is what others use for this nowadays: except that, as currently written, it gives the right answer on hugetlbfs head, but nonsense on hugetlbfs tails. Fix that by calling hugetlbfs-specific hugetlb_basepage_index() on PageHuge tails as well as on head.

Yes, it's unconventional to declare hugetlb_basepage_index() there in pagemap.h, rather than in hugetlb.h; but I do not expect anything but page_to_pgoff() ever to need it.

[[email protected]: give hugetlb_basepage_index() prototype the correct scope]

---
 include/linux/hugetlb.h | 16 ----------------
 include/linux/pagemap.h | 13 +++++++------
 kernel/futex.c          |  3 +--
 mm/hugetlb.c            |  5 +----
 4 files changed, 9 insertions(+), 28 deletions(-)
include/linux/hugetlb.h +0 -16
index b5807f23caf8..5b68c9787f7c 100644
@@ -628,17 +628,6 @@ static inline int hstate_index(struct hstate *h)
 	return h - hstates;
 }
 
-pgoff_t __basepage_index(struct page *page);
-
-/* Return page->index in PAGE_SIZE units */
-static inline pgoff_t basepage_index(struct page *page)
-{
-	if (!PageCompound(page))
-		return page->index;
-
-	return __basepage_index(page);
-}
-
 extern int dissolve_free_huge_page(struct page *page);
 extern int dissolve_free_huge_pages(unsigned long start_pfn,
 				    unsigned long end_pfn);
@@ -871,11 +860,6 @@ static inline int hstate_index(struct hstate *h)
 	return 0;
 }
 
-static inline pgoff_t basepage_index(struct page *page)
-{
-	return page->index;
-}
-
 static inline int dissolve_free_huge_page(struct page *page)
 {
 	return 0;
include/linux/pagemap.h +7 -6
index b032f094a782..fcb3f040102a 100644
@@ -501,7 +501,7 @@ static inline struct page *read_mapping_page(struct address_space *mapping,
 }
 
 /*
- * Get index of the page with in radix-tree
+ * Get index of the page within radix-tree (but not for hugetlb pages).
  * (TODO: remove once hugetlb pages will have ->index in PAGE_SIZE)
  */
 static inline pgoff_t page_to_index(struct page *page)
@@ -520,15 +520,16 @@ static inline pgoff_t page_to_index(struct page *page)
 	return pgoff;
 }
 
+extern pgoff_t hugetlb_basepage_index(struct page *page);
+
 /*
- * Get the offset in PAGE_SIZE.
- * (TODO: hugepage should have ->index in PAGE_SIZE)
+ * Get the offset in PAGE_SIZE (even for hugetlb pages).
+ * (TODO: hugetlb pages should have ->index in PAGE_SIZE)
  */
 static inline pgoff_t page_to_pgoff(struct page *page)
 {
-	if (unlikely(PageHeadHuge(page)))
-		return page->index << compound_order(page);
-
+	if (unlikely(PageHuge(page)))
+		return hugetlb_basepage_index(page);
 	return page_to_index(page);
 }
 
kernel/futex.c +1 -2
index 3136aba17772..98a6e1b80bfe 100644
@@ -35,7 +35,6 @@
 #include <linux/jhash.h>
 #include <linux/pagemap.h>
 #include <linux/syscalls.h>
-#include <linux/hugetlb.h>
 #include <linux/freezer.h>
 #include <linux/memblock.h>
 #include <linux/fault-inject.h>
@@ -652,7 +651,7 @@ static int get_futex_key(u32 __user *uaddr, bool fshared, union futex_key *key,
 
 		key->both.offset |= FUT_OFF_INODE; /* inode-based key */
 		key->shared.i_seq = get_inode_sequence_number(inode);
-		key->shared.pgoff = basepage_index(tail);
+		key->shared.pgoff = page_to_pgoff(tail);
 		rcu_read_unlock();
 	}
 
mm/hugetlb.c +1 -4
index bc1006a32733..d4f89c2f9544 100644
@@ -1635,15 +1635,12 @@ struct address_space *hugetlb_page_mapping_lock_write(struct page *hpage)
 	return NULL;
 }
 
-pgoff_t __basepage_index(struct page *page)
+pgoff_t hugetlb_basepage_index(struct page *page)
 {
 	struct page *page_head = compound_head(page);
 	pgoff_t index = page_index(page_head);
 	unsigned long compound_idx;
 
-	if (!PageHuge(page_head))
-		return page_index(page);
-
 	if (compound_order(page_head) >= MAX_ORDER)
 		compound_idx = page_to_pfn(page) - page_to_pfn(page_head);
 	else
Signature · 1 line
2.30.2
author maintainer
#101 lore raw

PATCH 95/101 netfs: fix test for whether we can skip read when writing beyond EOF

In reply to Sasha Levin

From: Jeff Layton <[email protected]>

commit 827a746f405d25f79560c7868474aec5aee174e1 upstream.

It's not sufficient to skip reading when the pos is beyond the EOF. There may be data at the head of the page that we need to fill in before the write.

Add a new helper function that corrects and clarifies the logic of when we can skip reads, and have it only zero out the part of the page that won't have data copied in for the write.

Finally, don't set the page Uptodate after zeroing. It's not up to date since the write data won't have been copied in yet.

[DH made the following changes:

- Prefixed the new function with "netfs_".

- Don't call zero_user_segments() for a full-page write.

- Altered the beyond-last-page check to avoid a DIV instruction and got rid of then-redundant zero-length file check.
]

[ Note: this fix is commit 827a746f405d in mainline kernels. The
original bug was in ceph, but got lifted into the fs/netfs library for v5.13. This backport should apply to stable kernels v5.10 though v5.12. ]

---
 fs/ceph/addr.c | 54 ++++++++++++++++++++++++++++++++++++++------------
 1 file changed, 41 insertions(+), 13 deletions(-)
fs/ceph/addr.c +41 -13
index 35c83f65475b..8b0507f69c15 100644
@@ -1302,6 +1302,45 @@ ceph_find_incompatible(struct page *page)
 	return NULL;
 }
 
+/**
+ * prep_noread_page - prep a page for writing without reading first
+ * @page: page being prepared
+ * @pos: starting position for the write
+ * @len: length of write
+ *
+ * In some cases, write_begin doesn't need to read at all:
+ * - full page write
+ * - file is currently zero-length
+ * - write that lies in a page that is completely beyond EOF
+ * - write that covers the the page from start to EOF or beyond it
+ *
+ * If any of these criteria are met, then zero out the unwritten parts
+ * of the page and return true. Otherwise, return false.
+ */
+static bool skip_page_read(struct page *page, loff_t pos, size_t len)
+{
+	struct inode *inode = page->mapping->host;
+	loff_t i_size = i_size_read(inode);
+	size_t offset = offset_in_page(pos);
+
+	/* Full page write */
+	if (offset == 0 && len >= PAGE_SIZE)
+		return true;
+
+	/* pos beyond last page in the file */
+	if (pos - offset >= i_size)
+		goto zero_out;
+
+	/* write that covers the whole page from start to EOF or beyond it */
+	if (offset == 0 && (pos + len) >= i_size)
+		goto zero_out;
+
+	return false;
+zero_out:
+	zero_user_segments(page, 0, offset, offset + len, PAGE_SIZE);
+	return true;
+}
+
 /*
  * We are only allowed to write into/dirty the page if the page is
  * clean, or already dirty within the same snap context.
@@ -1315,7 +1354,6 @@ static int ceph_write_begin(struct file *file, struct address_space *mapping,
 	struct ceph_snap_context *snapc;
 	struct page *page = NULL;
 	pgoff_t index = pos >> PAGE_SHIFT;
-	int pos_in_page = pos & ~PAGE_MASK;
 	int r = 0;
 
 	dout("write_begin file %p inode %p page %p %d~%d\n", file, inode, page, (int)pos, (int)len);
@@ -1350,19 +1388,9 @@ static int ceph_write_begin(struct file *file, struct address_space *mapping,
 			break;
 		}
 
-		/*
-		 * In some cases we don't need to read at all:
-		 * - full page write
-		 * - write that lies completely beyond EOF
-		 * - write that covers the the page from start to EOF or beyond it
-		 */
-		if ((pos_in_page == 0 && len == PAGE_SIZE) ||
-		    (pos >= i_size_read(inode)) ||
-		    (pos_in_page == 0 && (pos + len) >= i_size_read(inode))) {
-			zero_user_segments(page, 0, pos_in_page,
-					   pos_in_page + len, PAGE_SIZE);
+		/* No need to read in some cases */
+		if (skip_page_read(page, pos, len))
 			break;
-		}
 
 		/*
 		 * We need to read it. If we get back -EINPROGRESS, then the page was
Signature · 1 line
2.30.2
author maintainer
#99 lore raw

PATCH 93/101 KVM: SVM: Call SEV Guest Decommission if ASID binding fails

In reply to Sasha Levin

From: Alper Gun <[email protected]>

commit 934002cd660b035b926438244b4294e647507e13 upstream.

Send SEV_CMD_DECOMMISSION command to PSP firmware if ASID binding fails. If a failure happens after a successful LAUNCH_START command, a decommission command should be executed. Otherwise, guest context will be unfreed inside the AMD SP. After the firmware will not have memory to allocate more SEV guest context, LAUNCH_START command will begin to fail with SEV_RET_RESOURCE_LIMIT error.

The existing code calls decommission inside sev_unbind_asid, but it is not called if a failure happens before guest activation succeeds. If sev_bind_asid fails, decommission is never called. PSP firmware has a limit for the number of guests. If sev_asid_binding fails many times, PSP firmware will not have resources to create another guest context.

Fixes 59414c989220 ("KVM: SVM: Add support for KVM_SEV_LAUNCH_START command")

Message-Id: <[email protected]>

---
 arch/x86/kvm/svm/sev.c | 32 +++++++++++++++++++++-----------
 1 file changed, 21 insertions(+), 11 deletions(-)
arch/x86/kvm/svm/sev.c +21 -11
index 16b10b9436dc..01547bdbfb06 100644
@@ -130,9 +130,25 @@ static void sev_asid_free(int asid)
 	mutex_unlock(&sev_bitmap_lock);
 }
 
-static void sev_unbind_asid(struct kvm *kvm, unsigned int handle)
+static void sev_decommission(unsigned int handle)
 {
 	struct sev_data_decommission *decommission;
+
+	if (!handle)
+		return;
+
+	decommission = kzalloc(sizeof(*decommission), GFP_KERNEL);
+	if (!decommission)
+		return;
+
+	decommission->handle = handle;
+	sev_guest_decommission(decommission, NULL);
+
+	kfree(decommission);
+}
+
+static void sev_unbind_asid(struct kvm *kvm, unsigned int handle)
+{
 	struct sev_data_deactivate *data;
 
 	if (!handle)
@@ -152,15 +168,7 @@ static void sev_unbind_asid(struct kvm *kvm, unsigned int handle)
 
 	kfree(data);
 
-	decommission = kzalloc(sizeof(*decommission), GFP_KERNEL);
-	if (!decommission)
-		return;
-
-	/* decommission handle */
-	decommission->handle = handle;
-	sev_guest_decommission(decommission, NULL);
-
-	kfree(decommission);
+	sev_decommission(handle);
 }
 
 static int sev_guest_init(struct kvm *kvm, struct kvm_sev_cmd *argp)
@@ -288,8 +296,10 @@ static int sev_launch_start(struct kvm *kvm, struct kvm_sev_cmd *argp)
 
 	/* Bind ASID to this guest */
 	ret = sev_bind_asid(kvm, start->handle, error);
-	if (ret)
+	if (ret) {
+		sev_decommission(start->handle);
 		goto e_free_session;
+	}
 
 	/* return handle to userspace */
 	params.handle = start->handle;
Signature · 1 line
2.30.2
author maintainer
#100 lore raw

PATCH 94/101 swiotlb: manipulate orig_addr when tlb_addr has offset

In reply to Sasha Levin

From: Bumyong Lee <[email protected]>

commit 5f89468e2f060031cd89fd4287298e0eaf246bf6 upstream.

in case of driver wants to sync part of ranges with offset, swiotlb_tbl_sync_single() copies from orig_addr base to tlb_addr with offset and ends up with data mismatch.

It was removed from
"swiotlb: don't modify orig_addr in swiotlb_tbl_sync_single", but said logic has to be added back in.

From Linus's email:
"That commit which the removed the offset calculation entirely, because the old

(unsigned long)tlb_addr & (IO_TLB_SIZE - 1)

was wrong, but instead of removing it, I think it should have just fixed it to be

        (tlb_addr - mem->start) & (IO_TLB_SIZE - 1);

instead. That way the slot offset always matches the slot index calculation."

(Unfortunatly that broke NVMe).

The use-case that drivers are hitting is as follow:

1. Get dma_addr_t from dma_map_single()

dma_addr_t tlb_addr = dma_map_single(dev, vaddr, vsize, DMA_TO_DEVICE);

    |<---------------vsize------------->|
    +-----------------------------------+
    |                                   | original buffer
    +-----------------------------------+

vaddr

swiotlb_align_offset

     |<----->|<---------------vsize------------->|
     +-------+-----------------------------------+
     |       |                                   | swiotlb buffer
     +-------+-----------------------------------+

tlb_addr

2. Do something
3. Sync dma_addr_t through dma_sync_single_for_device(..)

dma_sync_single_for_device(dev, tlb_addr + offset, size, DMA_TO_DEVICE);

Error case.
Copy data to original buffer but it is from base addr (instead of
base addr + offset) in original buffer:

swiotlb_align_offset

     |<----->|<- offset ->|<- size ->|
     +-------+-----------------------------------+
     |       |            |##########|           | swiotlb buffer
     +-------+-----------------------------------+
          tlb_addr

    |<- size ->|
    +-----------------------------------+
    |##########|                        | original buffer
    +-----------------------------------+

vaddr

The fix is to copy the data to the original buffer and take into account the offset, like so:

swiotlb_align_offset

     |<----->|<- offset ->|<- size ->|
     +-------+-----------------------------------+
     |       |            |##########|           | swiotlb buffer
     +-------+-----------------------------------+
          tlb_addr

    |<- offset ->|<- size ->|
    +-----------------------------------+
    |            |##########|           | original buffer
    +-----------------------------------+

vaddr

[One fix which was Linus's that made more sense to as it created a symmetry would break NVMe. The reason for that is the:
unsigned int offset = (tlb_addr - mem->start) & (IO_TLB_SIZE - 1);

would come up with the proper offset, but it would lose the alignment (which this patch contains).]

---
 kernel/dma/swiotlb.c | 3 +++
 1 file changed, 3 insertions(+)
kernel/dma/swiotlb.c +3 -0
index 0f61b14b0099..0ed0e1f215c7 100644
@@ -667,6 +667,9 @@ void swiotlb_tbl_sync_single(struct device *hwdev, phys_addr_t tlb_addr,
 	if (orig_addr == INVALID_PHYS_ADDR)
 		return;
 
+	orig_addr += (tlb_addr & (IO_TLB_SIZE - 1)) -
+		swiotlb_align_offset(hwdev, orig_addr);
+
 	switch (target) {
 	case SYNC_FOR_CPU:
 		if (likely(dir == DMA_FROM_DEVICE || dir == DMA_BIDIRECTIONAL))
Signature · 1 line
2.30.2
author maintainer
#102 lore raw

PATCH 96/101 Revert "drm: add a locked version of drm_is_current_master"

In reply to Sasha Levin

From: Daniel Vetter <[email protected]>

commit f54b3ca7ea1e5e02f481cf4ca54568e57bd66086 upstream.

This reverts commit 1815d9c86e3090477fbde066ff314a7e9721ee0f.

Unfortunately this inverts the locking hierarchy, so back to the drawing board. Full lockdep splat below:

======================================================
WARNING: possible circular locking dependency detected 5.13.0-rc7-CI-CI_DRM_10254+ #1 Not tainted
------------------------------------------------------
kms_frontbuffer/1087 is trying to acquire lock:
ffff88810dcd01a8 (&dev->master_mutex){+.+.}-{3:3}, at: drm_is_current_master+0x1b/0x40 but task is already holding lock:
ffff88810dcd0488 (&dev->mode_config.mutex){+.+.}-{3:3}, at: drm_mode_getconnector+0x1c6/0x4a0 which lock already depends on the new lock.
the existing dependency chain (in reverse order) is:
-> #2 (&dev->mode_config.mutex){+.+.}-{3:3}:

       __mutex_lock+0xab/0x970
       drm_client_modeset_probe+0x22e/0xca0
       __drm_fb_helper_initial_config_and_unlock+0x42/0x540
       intel_fbdev_initial_config+0xf/0x20 [i915]
       async_run_entry_fn+0x28/0x130
       process_one_work+0x26d/0x5c0
       worker_thread+0x37/0x380
       kthread+0x144/0x170
       ret_from_fork+0x1f/0x30
-> #1 (&client->modeset_mutex){+.+.}-{3:3}:
       __mutex_lock+0xab/0x970
       drm_client_modeset_commit_locked+0x1c/0x180
       drm_client_modeset_commit+0x1c/0x40
       __drm_fb_helper_restore_fbdev_mode_unlocked+0x88/0xb0
       drm_fb_helper_set_par+0x34/0x40
       intel_fbdev_set_par+0x11/0x40 [i915]
       fbcon_init+0x270/0x4f0
       visual_init+0xc6/0x130
       do_bind_con_driver+0x1e5/0x2d0
       do_take_over_console+0x10e/0x180
       do_fbcon_takeover+0x53/0xb0
       register_framebuffer+0x22d/0x310
       __drm_fb_helper_initial_config_and_unlock+0x36c/0x540
       intel_fbdev_initial_config+0xf/0x20 [i915]
       async_run_entry_fn+0x28/0x130
       process_one_work+0x26d/0x5c0
       worker_thread+0x37/0x380
       kthread+0x144/0x170
       ret_from_fork+0x1f/0x30
-> #0 (&dev->master_mutex){+.+.}-{3:3}:
       __lock_acquire+0x151e/0x2590
       lock_acquire+0xd1/0x3d0
       __mutex_lock+0xab/0x970
       drm_is_current_master+0x1b/0x40
       drm_mode_getconnector+0x37e/0x4a0
       drm_ioctl_kernel+0xa8/0xf0
       drm_ioctl+0x1e8/0x390
       __x64_sys_ioctl+0x6a/0xa0
       do_syscall_64+0x39/0xb0
       entry_SYSCALL_64_after_hwframe+0x44/0xae
other info that might help us debug this:
Chain exists of: &dev->master_mutex --> &client->modeset_mutex --> &dev->mode_config.mutex
 Possible unsafe locking scenario:
       CPU0                    CPU1
       ----                    ----
  lock(&dev->mode_config.mutex);
                               lock(&client->modeset_mutex);
                               lock(&dev->mode_config.mutex);
  lock(&dev->master_mutex);
---
 drivers/gpu/drm/drm_auth.c | 51 ++++++++++++++------------------------
 1 file changed, 19 insertions(+), 32 deletions(-)
drivers/gpu/drm/drm_auth.c +19 -32
index 86d4b72e95cb..232abbba3686 100644
@@ -61,35 +61,6 @@
  * trusted clients.
  */
 
-static bool drm_is_current_master_locked(struct drm_file *fpriv)
-{
-	lockdep_assert_held_once(&fpriv->master->dev->master_mutex);
-
-	return fpriv->is_master && drm_lease_owner(fpriv->master) == fpriv->minor->dev->master;
-}
-
-/**
- * drm_is_current_master - checks whether @priv is the current master
- * @fpriv: DRM file private
- *
- * Checks whether @fpriv is current master on its device. This decides whether a
- * client is allowed to run DRM_MASTER IOCTLs.
- *
- * Most of the modern IOCTL which require DRM_MASTER are for kernel modesetting
- * - the current master is assumed to own the non-shareable display hardware.
- */
-bool drm_is_current_master(struct drm_file *fpriv)
-{
-	bool ret;
-
-	mutex_lock(&fpriv->master->dev->master_mutex);
-	ret = drm_is_current_master_locked(fpriv);
-	mutex_unlock(&fpriv->master->dev->master_mutex);
-
-	return ret;
-}
-EXPORT_SYMBOL(drm_is_current_master);
-
 int drm_getmagic(struct drm_device *dev, void *data, struct drm_file *file_priv)
 {
 	struct drm_auth *auth = data;
@@ -252,7 +223,7 @@ int drm_setmaster_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		goto out_unlock;
 
-	if (drm_is_current_master_locked(file_priv))
+	if (drm_is_current_master(file_priv))
 		goto out_unlock;
 
 	if (dev->master) {
@@ -301,7 +272,7 @@ int drm_dropmaster_ioctl(struct drm_device *dev, void *data,
 	if (ret)
 		goto out_unlock;
 
-	if (!drm_is_current_master_locked(file_priv)) {
+	if (!drm_is_current_master(file_priv)) {
 		ret = -EINVAL;
 		goto out_unlock;
 	}
@@ -350,7 +321,7 @@ void drm_master_release(struct drm_file *file_priv)
 	if (file_priv->magic)
 		idr_remove(&file_priv->master->magic_map, file_priv->magic);
 
-	if (!drm_is_current_master_locked(file_priv))
+	if (!drm_is_current_master(file_priv))
 		goto out;
 
 	drm_legacy_lock_master_cleanup(dev, master);
@@ -371,6 +342,22 @@ void drm_master_release(struct drm_file *file_priv)
 	mutex_unlock(&dev->master_mutex);
 }
 
+/**
+ * drm_is_current_master - checks whether @priv is the current master
+ * @fpriv: DRM file private
+ *
+ * Checks whether @fpriv is current master on its device. This decides whether a
+ * client is allowed to run DRM_MASTER IOCTLs.
+ *
+ * Most of the modern IOCTL which require DRM_MASTER are for kernel modesetting
+ * - the current master is assumed to own the non-shareable display hardware.
+ */
+bool drm_is_current_master(struct drm_file *fpriv)
+{
+	return fpriv->is_master && drm_lease_owner(fpriv->master) == fpriv->minor->dev->master;
+}
+EXPORT_SYMBOL(drm_is_current_master);
+
 /**
  * drm_master_get - reference a master pointer
  * @master: &struct drm_master
Signature · 1 line
2.30.2
author maintainer
#104 lore raw

PATCH 98/101 certs: Move load_system_certificate_list to a common function

In reply to Sasha Levin

From: Eric Snowberg <[email protected]>

[ Upstream commit 2565ca7f5ec1a98d51eea8860c4ab923f1ca2c85 ]

Move functionality within load_system_certificate_list to a common function, so it can be reused in the future.

DH Changes:
- Added inclusion of common.h to common.c (Eric [1]).

---
 certs/Makefile         |  2 +-
 certs/common.c         | 57 ++++++++++++++++++++++++++++++++++++++++++
 certs/common.h         |  9 +++++++
 certs/system_keyring.c | 49 +++---------------------------------
 4 files changed, 70 insertions(+), 47 deletions(-)
 create mode 100644 certs/common.c
 create mode 100644 certs/common.h
certs/Makefile +1 -1
index f4c25b67aad9..f4b90bad8690 100644
@@ -3,7 +3,7 @@
 # Makefile for the linux kernel signature checking certificates.
 #
 
-obj-$(CONFIG_SYSTEM_TRUSTED_KEYRING) += system_keyring.o system_certificates.o
+obj-$(CONFIG_SYSTEM_TRUSTED_KEYRING) += system_keyring.o system_certificates.o common.o
 obj-$(CONFIG_SYSTEM_BLACKLIST_KEYRING) += blacklist.o
 ifneq ($(CONFIG_SYSTEM_BLACKLIST_HASH_LIST),"")
 obj-$(CONFIG_SYSTEM_BLACKLIST_KEYRING) += blacklist_hashes.o
certs/common.c new file +57 -0

57 lines not on this page · raw message

certs/common.h new file +9 -0
new file mode 100644
index 000000000000..abdb5795936b
@@ -0,0 +1,9 @@
+/* SPDX-License-Identifier: GPL-2.0-or-later */
+
+#ifndef _CERT_COMMON_H
+#define _CERT_COMMON_H
+
+int load_certificate_list(const u8 cert_list[], const unsigned long list_size,
+			  const struct key *keyring);
+
+#endif
certs/system_keyring.c +3 -46

63 lines not on this page · raw message

Signature · 1 line
2.30.2
author maintainer
#103 lore raw

PATCH 97/101 certs: Add EFI_CERT_X509_GUID support for dbx entries

In reply to Sasha Levin

From: Eric Snowberg <[email protected]>

[ Upstream commit 56c5812623f95313f6a46fbf0beee7fa17c68bbf ]

This fixes CVE-2020-26541.

The Secure Boot Forbidden Signature Database, dbx, contains a list of now revoked signatures and keys previously approved to boot with UEFI Secure Boot enabled. The dbx is capable of containing any number of EFI_CERT_X509_SHA256_GUID, EFI_CERT_SHA256_GUID, and EFI_CERT_X509_GUID entries.

Currently when EFI_CERT_X509_GUID are contained in the dbx, the entries are skipped.

Add support for EFI_CERT_X509_GUID dbx entries. When a EFI_CERT_X509_GUID is found, it is added as an asymmetrical key to the .blacklist keyring. Anytime the .platform keyring is used, the keys in the .blacklist keyring are referenced, if a matching key is found, the key will be rejected.

[DH: Made the following changes:
- Added to have a config option to enable the facility. This allows a Kconfig solution to make sure that pkcs7_validate_trust() is enabled.[1][2]
- Moved the functions out from the middle of the blacklist functions.
- Added kerneldoc comments.]

---
 certs/Kconfig                                 |  9 ++++
 certs/blacklist.c                             | 43 +++++++++++++++++++
 certs/blacklist.h                             |  2 +
 certs/system_keyring.c                        |  6 +++
 include/keys/system_keyring.h                 | 15 +++++++
 .../platform_certs/keyring_handler.c          | 11 +++++
 6 files changed, 86 insertions(+)
certs/Kconfig +9 -0
index c94e93d8bccf..76e469b56a77 100644
@@ -83,4 +83,13 @@ config SYSTEM_BLACKLIST_HASH_LIST
 	  wrapper to incorporate the list into the kernel.  Each <hash> should
 	  be a string of hex digits.
 
+config SYSTEM_REVOCATION_LIST
+	bool "Provide system-wide ring of revocation certificates"
+	depends on SYSTEM_BLACKLIST_KEYRING
+	depends on PKCS7_MESSAGE_PARSER=y
+	help
+	  If set, this allows revocation certificates to be stored in the
+	  blacklist keyring and implements a hook whereby a PKCS#7 message can
+	  be checked to see if it matches such a certificate.
+
 endmenu
certs/blacklist.c +43 -0

49 lines not on this page · raw message

certs/blacklist.h +2 -0
index 1efd6fa0dc60..51b320cf8574 100644
@@ -1,3 +1,5 @@
 #include <linux/kernel.h>
+#include <linux/errno.h>
+#include <crypto/pkcs7.h>
 
 extern const char __initconst *const blacklist_hashes[];
certs/system_keyring.c +6 -0
index 798291177186..cc165b359ea3 100644
@@ -241,6 +241,12 @@ int verify_pkcs7_message_sig(const void *data, size_t len,
 			pr_devel("PKCS#7 platform keyring is not available\n");
 			goto error;
 		}
+
+		ret = is_key_on_revocation_list(pkcs7);
+		if (ret != -ENOKEY) {
+			pr_devel("PKCS#7 platform key is on revocation list\n");
+			goto error;
+		}
 	}
 	ret = pkcs7_validate_trust(pkcs7, trusted_keys);
 	if (ret < 0) {
include/keys/system_keyring.h +15 -0
index fb8b07daa9d1..875e002a4180 100644
@@ -31,6 +31,7 @@ extern int restrict_link_by_builtin_and_secondary_trusted(
 #define restrict_link_by_builtin_and_secondary_trusted restrict_link_by_builtin_trusted
 #endif
 
+extern struct pkcs7_message *pkcs7;
 #ifdef CONFIG_SYSTEM_BLACKLIST_KEYRING
 extern int mark_hash_blacklisted(const char *hash);
 extern int is_hash_blacklisted(const u8 *hash, size_t hash_len,
@@ -49,6 +50,20 @@ static inline int is_binary_blacklisted(const u8 *hash, size_t hash_len)
 }
 #endif
 
+#ifdef CONFIG_SYSTEM_REVOCATION_LIST
+extern int add_key_to_revocation_list(const char *data, size_t size);
+extern int is_key_on_revocation_list(struct pkcs7_message *pkcs7);
+#else
+static inline int add_key_to_revocation_list(const char *data, size_t size)
+{
+	return 0;
+}
+static inline int is_key_on_revocation_list(struct pkcs7_message *pkcs7)
+{
+	return -ENOKEY;
+}
+#endif
+
 #ifdef CONFIG_IMA_BLACKLIST_KEYRING
 extern struct key *ima_blacklist_keyring;
 
security/integrity/platform_certs/keyring_handler.c +11 -0
index c5ba695c10e3..5604bd57c990 100644
@@ -55,6 +55,15 @@ static __init void uefi_blacklist_binary(const char *source,
 	uefi_blacklist_hash(source, data, len, "bin:", 4);
 }
 
+/*
+ * Add an X509 cert to the revocation list.
+ */
+static __init void uefi_revocation_list_x509(const char *source,
+					     const void *data, size_t len)
+{
+	add_key_to_revocation_list(data, len);
+}
+
 /*
  * Return the appropriate handler for particular signature list types found in
  * the UEFI db and MokListRT tables.
@@ -76,5 +85,7 @@ __init efi_element_handler_t get_handler_for_dbx(const efi_guid_t *sig_type)
 		return uefi_blacklist_x509_tbs;
 	if (efi_guidcmp(*sig_type, efi_cert_sha256_guid) == 0)
 		return uefi_blacklist_binary;
+	if (efi_guidcmp(*sig_type, efi_cert_x509_guid) == 0)
+		return uefi_revocation_list_x509;
 	return 0;
 }
Signature · 1 line
2.30.2
author maintainer
#105 lore raw

PATCH 99/101 certs: Add ability to preload revocation certs

In reply to Sasha Levin

From: Eric Snowberg <[email protected]>

[ Upstream commit d1f044103dad70c1cec0a8f3abdf00834fec8b98 ]

Add a new Kconfig option called SYSTEM_REVOCATION_KEYS. If set, this option should be the filename of a PEM-formated file containing X.509 certificates to be included in the default blacklist keyring.

DH Changes:
- Make the new Kconfig option depend on SYSTEM_REVOCATION_LIST.
- Fix SYSTEM_REVOCATION_KEYS=n, but CONFIG_SYSTEM_REVOCATION_LIST=y[1][2].
- Use CONFIG_SYSTEM_REVOCATION_LIST for extract-cert[3].
- Use CONFIG_SYSTEM_REVOCATION_LIST for revocation_certificates.o[3].

---
 certs/Kconfig                   |  8 ++++++++
 certs/Makefile                  | 19 +++++++++++++++++--
 certs/blacklist.c               | 21 +++++++++++++++++++++
 certs/revocation_certificates.S | 21 +++++++++++++++++++++
 scripts/Makefile                |  1 +
 5 files changed, 68 insertions(+), 2 deletions(-)
 create mode 100644 certs/revocation_certificates.S
certs/Kconfig +8 -0
index 76e469b56a77..ab88d2a7f3c7 100644
@@ -92,4 +92,12 @@ config SYSTEM_REVOCATION_LIST
 	  blacklist keyring and implements a hook whereby a PKCS#7 message can
 	  be checked to see if it matches such a certificate.
 
+config SYSTEM_REVOCATION_KEYS
+	string "X.509 certificates to be preloaded into the system blacklist keyring"
+	depends on SYSTEM_REVOCATION_LIST
+	help
+	  If set, this option should be the filename of a PEM-formatted file
+	  containing X.509 certificates to be included in the default blacklist
+	  keyring.
+
 endmenu
certs/Makefile +17 -2
index f4b90bad8690..b6db52ebf0be 100644
@@ -4,7 +4,8 @@
 #
 
 obj-$(CONFIG_SYSTEM_TRUSTED_KEYRING) += system_keyring.o system_certificates.o common.o
-obj-$(CONFIG_SYSTEM_BLACKLIST_KEYRING) += blacklist.o
+obj-$(CONFIG_SYSTEM_BLACKLIST_KEYRING) += blacklist.o common.o
+obj-$(CONFIG_SYSTEM_REVOCATION_LIST) += revocation_certificates.o
 ifneq ($(CONFIG_SYSTEM_BLACKLIST_HASH_LIST),"")
 obj-$(CONFIG_SYSTEM_BLACKLIST_KEYRING) += blacklist_hashes.o
 else
@@ -29,7 +30,7 @@ $(obj)/x509_certificate_list: scripts/extract-cert $(SYSTEM_TRUSTED_KEYS_SRCPREF
 	$(call if_changed,extract_certs,$(SYSTEM_TRUSTED_KEYS_SRCPREFIX)$(CONFIG_SYSTEM_TRUSTED_KEYS))
 endif # CONFIG_SYSTEM_TRUSTED_KEYRING
 
-clean-files := x509_certificate_list .x509.list
+clean-files := x509_certificate_list .x509.list x509_revocation_list
 
 ifeq ($(CONFIG_MODULE_SIG),y)
 ###############################################################################
@@ -104,3 +105,17 @@ targets += signing_key.x509
 $(obj)/signing_key.x509: scripts/extract-cert $(X509_DEP) FORCE
 	$(call if_changed,extract_certs,$(MODULE_SIG_KEY_SRCPREFIX)$(CONFIG_MODULE_SIG_KEY))
 endif # CONFIG_MODULE_SIG
+
+ifeq ($(CONFIG_SYSTEM_REVOCATION_LIST),y)
+
+$(eval $(call config_filename,SYSTEM_REVOCATION_KEYS))
+
+$(obj)/revocation_certificates.o: $(obj)/x509_revocation_list
+
+quiet_cmd_extract_certs  = EXTRACT_CERTS   $(patsubst "%",%,$(2))
+      cmd_extract_certs  = scripts/extract-cert $(2) $@
+
+targets += x509_revocation_list
+$(obj)/x509_revocation_list: scripts/extract-cert $(SYSTEM_REVOCATION_KEYS_SRCPREFIX)$(SYSTEM_REVOCATION_KEYS_FILENAME) FORCE
+	$(call if_changed,extract_certs,$(SYSTEM_REVOCATION_KEYS_SRCPREFIX)$(CONFIG_SYSTEM_REVOCATION_KEYS))
+endif
certs/blacklist.c +21 -0
index 59b2f106b294..c973de883cf0 100644
@@ -16,9 +16,15 @@
 #include <linux/seq_file.h>
 #include <keys/system_keyring.h>
 #include "blacklist.h"
+#include "common.h"
 
 static struct key *blacklist_keyring;
 
+#ifdef CONFIG_SYSTEM_REVOCATION_LIST
+extern __initconst const u8 revocation_certificate_list[];
+extern __initconst const unsigned long revocation_certificate_list_size;
+#endif
+
 /*
  * The description must be a type prefix, a colon and then an even number of
  * hex digits.  The hash is kept in the description.
@@ -220,3 +226,18 @@ static int __init blacklist_init(void)
  * Must be initialised before we try and load the keys into the keyring.
  */
 device_initcall(blacklist_init);
+
+#ifdef CONFIG_SYSTEM_REVOCATION_LIST
+/*
+ * Load the compiled-in list of revocation X.509 certificates.
+ */
+static __init int load_revocation_certificate_list(void)
+{
+	if (revocation_certificate_list_size)
+		pr_notice("Loading compiled-in revocation X.509 certificates\n");
+
+	return load_certificate_list(revocation_certificate_list, revocation_certificate_list_size,
+				     blacklist_keyring);
+}
+late_initcall(load_revocation_certificate_list);
+#endif
certs/revocation_certificates.S new file +21 -0
new file mode 100644
index 000000000000..f21aae8a8f0e
@@ -0,0 +1,21 @@
+/* SPDX-License-Identifier: GPL-2.0 */
+#include <linux/export.h>
+#include <linux/init.h>
+
+	__INITRODATA
+
+	.align 8
+	.globl revocation_certificate_list
+revocation_certificate_list:
+__revocation_list_start:
+	.incbin "certs/x509_revocation_list"
+__revocation_list_end:
+
+	.align 8
+	.globl revocation_certificate_list_size
+revocation_certificate_list_size:
+#ifdef CONFIG_64BIT
+	.quad __revocation_list_end - __revocation_list_start
+#else
+	.long __revocation_list_end - __revocation_list_start
+#endif
scripts/Makefile +1 -0
index c36106bce80e..9adb6d247818 100644
@@ -14,6 +14,7 @@ hostprogs-always-$(CONFIG_ASN1)				+= asn1_compiler
 hostprogs-always-$(CONFIG_MODULE_SIG_FORMAT)		+= sign-file
 hostprogs-always-$(CONFIG_SYSTEM_TRUSTED_KEYRING)	+= extract-cert
 hostprogs-always-$(CONFIG_SYSTEM_EXTRA_CERTIFICATE)	+= insert-sys-cert
+hostprogs-always-$(CONFIG_SYSTEM_REVOCATION_LIST)	+= extract-cert
 
 HOSTCFLAGS_sorttable.o = -I$(srctree)/tools/include
 HOSTCFLAGS_asn1_compiler.o = -I$(srctree)/include
Signature · 1 line
2.30.2
author maintainer
#106 lore raw

PATCH 100/101 integrity: Load mokx variables into the blacklist keyring

In reply to Sasha Levin

From: Eric Snowberg <[email protected]>

[ Upstream commit ebd9c2ae369a45bdd9f8615484db09be58fc242b ]

During boot the Secure Boot Forbidden Signature Database, dbx, is loaded into the blacklist keyring. Systems booted with shim have an equivalent Forbidden Signature Database called mokx. Currently mokx is only used by shim and grub, the contents are ignored by the kernel.

Add the ability to load mokx into the blacklist keyring during boot.

---
 security/integrity/platform_certs/load_uefi.c | 20 +++++++++++++++++--
 1 file changed, 18 insertions(+), 2 deletions(-)
security/integrity/platform_certs/load_uefi.c +18 -2
index ee4b4c666854..f290f78c3f30 100644
@@ -132,8 +132,9 @@ static int __init load_moklist_certs(void)
 static int __init load_uefi_certs(void)
 {
 	efi_guid_t secure_var = EFI_IMAGE_SECURITY_DATABASE_GUID;
-	void *db = NULL, *dbx = NULL;
-	unsigned long dbsize = 0, dbxsize = 0;
+	efi_guid_t mok_var = EFI_SHIM_LOCK_GUID;
+	void *db = NULL, *dbx = NULL, *mokx = NULL;
+	unsigned long dbsize = 0, dbxsize = 0, mokxsize = 0;
 	efi_status_t status;
 	int rc = 0;
 
@@ -175,6 +176,21 @@ static int __init load_uefi_certs(void)
 		kfree(dbx);
 	}
 
+	mokx = get_cert_list(L"MokListXRT", &mok_var, &mokxsize, &status);
+	if (!mokx) {
+		if (status == EFI_NOT_FOUND)
+			pr_debug("mokx variable wasn't found\n");
+		else
+			pr_info("Couldn't get mokx list\n");
+	} else {
+		rc = parse_efi_signature_list("UEFI:MokListXRT",
+					      mokx, mokxsize,
+					      get_handler_for_dbx);
+		if (rc)
+			pr_err("Couldn't parse mokx signatures %d\n", rc);
+		kfree(mokx);
+	}
+
 	/* Load the MokListRT certs */
 	rc = load_moklist_certs();
 
Signature · 1 line
2.30.2
author maintainer
#107 lore raw

PATCH 101/101 Linux 5.10.47-rc1

In reply to Sasha Levin
---
 Makefile | 4 ++--
 1 file changed, 2 insertions(+), 2 deletions(-)
Makefile +2 -2
index 7ab22f105a03..97c664d872b6 100644
@@ -1,8 +1,8 @@
 # SPDX-License-Identifier: GPL-2.0
 VERSION = 5
 PATCHLEVEL = 10
-SUBLEVEL = 46
-EXTRAVERSION =
+SUBLEVEL = 47
+EXTRAVERSION = -rc1
 NAME = Dare mighty things
 
 # *DOCUMENTATION*
Signature · 1 line
2.30.2
Tested-by on the v1 series
#108 lore raw
In reply to Sasha Levin
Quoted 9 lines · Sasha Levin

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed 30 Jun 2021 02:25:36 PM UTC. Anything received after that time might be too late.

Quoted 9 lines

The whole patch series can be found in one patch at:

        https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/patch/?id=linux-5.10.y&id2=v5.10.46

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.

Thanks,
Sasha

5.10.47-rc1 Successfully Compiled and booted on my Raspberry PI 4b (8g) (bcm2711)

Tested-by Linux Kernel Functional Testing on the v1 series
#109 lore raw
In reply to Sasha Levin
Quoted 10 lines · Sasha Levin

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed 30 Jun 2021 02:25:36 PM UTC. Anything received after that time might be too late.

Quoted 8 lines

The whole patch series can be found in one patch at:

        https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/patch/?id=linux-5.10.y&id2=v5.10.46

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.

Thanks,
Sasha

Results from Linaro’s test farm.
No regressions on arm64, arm, x86_64, and i386.

## Build
* kernel: 5.10.47-rc1
* git: https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
* git branch: linux-5.10.y
* git commit: a41d5119dc1eec2f0ffaa0a1777b0218a76b9f0e
* git describe: v5.10.46-101-ga41d5119dc1e
* test details:
https://qa-reports.linaro.org/lkft/linux-stable-rc-linux-5.10.y/build/v5.10.46-101-ga41d5119dc1e

## No regressions (compared to v5.10.46-13-g88b257611f2a)

## No fixes (compared to v5.10.46-13-g88b257611f2a)

## Test result summary
total: 80209, pass: 65767, fail: 2329, skip: 11358, xfail: 755,

## Build Summary
* arc: 10 total, 10 passed, 0 failed
* arm: 193 total, 193 passed, 0 failed
* arm64: 27 total, 27 passed, 0 failed
* dragonboard-410c: 1 total, 1 passed, 0 failed
* hi6220-hikey: 1 total, 1 passed, 0 failed
* i386: 26 total, 26 passed, 0 failed
* juno-r2: 1 total, 1 passed, 0 failed
* mips: 45 total, 45 passed, 0 failed
* parisc: 9 total, 9 passed, 0 failed
* powerpc: 27 total, 27 passed, 0 failed
* riscv: 21 total, 21 passed, 0 failed
* s390: 18 total, 18 passed, 0 failed
* sh: 18 total, 18 passed, 0 failed
* sparc: 9 total, 9 passed, 0 failed
* x15: 1 total, 1 passed, 0 failed
* x86: 1 total, 1 passed, 0 failed
* x86_64: 27 total, 27 passed, 0 failed

## Test suites summary
* fwts
* igt-gpu-tools
* install-android-platform-tools-r2600
* kselftest-
* kselftest-android
* kselftest-bpf
* kselftest-breakpoints
* kselftest-capabilities
* kselftest-cgroup
* kselftest-clone3
* kselftest-core
* kselftest-cpu-hotplug
* kselftest-cpufreq
* kselftest-drivers
* kselftest-efivarfs
* kselftest-filesystems
* kselftest-firmware
* kselftest-fpu
* kselftest-futex
* kselftest-gpio
* kselftest-intel_pstate
* kselftest-ipc
* kselftest-ir
* kselftest-kcmp
* kselftest-kexec
* kselftest-kvm
* kselftest-lib
* kselftest-livepatch
* kselftest-lkdtm
* kselftest-membarrier
* kselftest-memfd
* kselftest-memory-hotplug
* kselftest-mincore
* kselftest-mount
* kselftest-mqueue
* kselftest-net
* kselftest-netfilter
* kselftest-nsfs
* kselftest-openat2
* kselftest-pid_namespace
* kselftest-pidfd
* kselftest-proc
* kselftest-pstore
* kselftest-ptrace
* kselftest-rseq
* kselftest-rtc
* kselftest-seccomp
* kselftest-sigaltstack
* kselftest-size
* kselftest-splice
* kselftest-static_keys
* kselftest-sync
* kselftest-sysctl
* kselftest-tc-testing
* kselftest-timens
* kselftest-timers
* kselftest-tmpfs
* kselftest-tpm2
* kselftest-user
* kselftest-vm
* kselftest-vsyscall-mode-native-
* kselftest-vsyscall-mode-none-
* kselftest-x86
* kselftest-zram
* kunit
* kvm-unit-tests
* libgpiod
* libhugetlbfs
* linux-log-parser
* ltp-cap_bounds-tests
* ltp-commands-tests
* ltp-containers-tests
* ltp-controllers-tests
* ltp-cpuhotplug-tests
* ltp-crypto-tests
* ltp-cve-tests
* ltp-dio-tests
* ltp-fcntl-locktests-tests
* ltp-filecaps-tests
* ltp-fs-tests
* ltp-fs_bind-tests
* ltp-fs_perms_simple-tests
* ltp-fsx-tests
* ltp-hugetlb-tests
* ltp-io-tests
* ltp-ipc-tests
* ltp-math-tests
* ltp-mm-tests
* ltp-nptl-tests
* ltp-open-posix-tests
* ltp-pty-tests
* ltp-sched-tests
* ltp-securebits-tests
* ltp-syscalls-tests
* ltp-tracing-tests
* network-basic-tests
* packetdrill
* perf
* rcutorture
* ssuite
* v4l2-compliance

--
Linaro LKFT
https://lkft.linaro.org

maintainer Tested-by on the v1 series
#110 lore raw
In reply to Sasha Levin
Quoted 9 lines · Sasha Levin

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed 30 Jun 2021 02:25:36 PM UTC. Anything received after that time might be too late.

Quoted 8 lines

The whole patch series can be found in one patch at:

        https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/patch/?id=linux-5.10.y&id2=v5.10.46

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.

Thanks,
Sasha

All tests passing for Tegra ...

Test results for stable-v5.10:

    12 builds:	12 pass, 0 fail
    28 boots:	28 pass, 0 fail
    70 tests:	70 pass, 0 fail

Linux version: 5.10.47-rc1-ga41d5119dc1e
Boards tested: tegra124-jetson-tk1, tegra186-p2771-0000,

                tegra194-p2972-0000, tegra194-p3509-0000+p3668-0000,
                tegra20-ventana, tegra210-p2371-2180,
                tegra210-p3450-0000, tegra30-cardhu-a04

Jon

Tested-by on the v1 series
#111 lore raw
In reply to Sasha Levin

Hi Sasha,

Quoted 6 lines · Sasha Levin

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Quoted 8 lines

Responses should be made by Wed 30 Jun 2021 02:25:36 PM UTC. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

        https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/patch/?id=linux-5.10.y&id2=v5.10.46

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.

Build test:
mips (gcc version 11.1.1 20210615): 63 configs -> no failure arm (gcc version 11.1.1 20210615): 105 configs -> no new failure arm64 (gcc version 11.1.1 20210615): 3 configs -> no failure x86_64 (gcc version 10.2.1 20210110): 2 configs -> no failure

Boot test:
x86_64: Booted on my test laptop. No regression.
x86_64: Booted on qemu. No regression.
arm64: Booted on rpi4b (4GB model). No regression.

--
Regards
Sudip

Tested-by on the v1 series
#113 lore raw
In reply to Sasha Levin
Quoted 9 lines · Sasha Levin

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed 30 Jun 2021 02:25:36 PM UTC. Anything received after that time might be too late.

Build results:

	total: 156 pass: 156 fail: 0
Qemu test results:
	total: 455 pass: 455 fail: 0

Guenter

Tested-by Hulk Robot on the v1 series
#114 lore raw
In reply to Sasha Levin
Quoted 9 lines · Sasha Levin

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed 30 Jun 2021 02:25:36 PM UTC. Anything received after that time might be too late.

Quoted 9 lines

The whole patch series can be found in one patch at:

         https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/patch/?id=linux-5.10.y&id2=v5.10.46

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.

Thanks,
Sasha

Tested on arm64 and x86 for 5.10.47-rc1,

Kernel repo:
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git
Branch: linux-5.10.y
Version: 5.10.47-rc1
Commit: a41d5119dc1eec2f0ffaa0a1777b0218a76b9f0e
Compiler: gcc version 7.3.0 (GCC)

arm64:
--------------------------------------------------------------------
Testcase Result Summary:
total: 8906
passed: 8906
failed: 0
timeout: 0
--------------------------------------------------------------------

x86:
--------------------------------------------------------------------
Testcase Result Summary:
total: 8906
passed: 8906
failed: 0
timeout: 0
--------------------------------------------------------------------

author maintainer
#112 lore raw
In reply to Sudip Mukherjee
Quoted 24 lines · Sudip Mukherjee

Hi Sasha,

Quoted 6 lines · Sasha Levin

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Quoted 8 lines

Responses should be made by Wed 30 Jun 2021 02:25:36 PM UTC. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

        https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/patch/?id=linux-5.10.y&id2=v5.10.46

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.

Build test:
mips (gcc version 11.1.1 20210615): 63 configs -> no failure arm (gcc version 11.1.1 20210615): 105 configs -> no new failure arm64 (gcc version 11.1.1 20210615): 3 configs -> no failure x86_64 (gcc version 10.2.1 20210110): 2 configs -> no failure

Quoted 6 lines

Boot test:
x86_64: Booted on my test laptop. No regression.
x86_64: Booted on qemu. No regression.
arm64: Booted on rpi4b (4GB model). No regression.

Thanks for testing Sudip!

Signature · 2 lines
Thanks,
Sasha
author maintainer
#115 lore raw
In reply to Samuel Zou
Quoted 18 lines · Samuel Zou
Quoted 15 lines · Sasha Levin

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Responses should be made by Wed 30 Jun 2021 02:25:36 PM UTC. Anything received after that time might be too late.

The whole patch series can be found in one patch at:

        https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/patch/?id=linux-5.10.y&id2=v5.10.46

or in the git tree and branch at:
git://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git linux-5.10.y
and the diffstat can be found below.

Quoted 5 lines
Quoted 3 lines

Thanks,
Sasha

Tested on arm64 and x86 for 5.10.47-rc1,

Thanks for testing Samuel!

Signature · 2 lines
Thanks,
Sasha
#17 lore raw

Re: [PATCH 5.10 016/101] drm/vc4: hdmi: Make sure the controller is powered in detect

In reply to Sasha Levin

Hi!

Quoted 7 lines

[ Upstream commit 9984d6664ce9dcbbc713962539eaf7636ea246c2 ]

If the HPD GPIO is not available and drm_probe_ddc fails, we end up reading the HDMI_HOTPLUG register, but the controller might be powered off resulting in a CPU hang. Make sure we have the power domain and the HSM clock powered during the detect cycle to prevent the hang from happening.

If the WARN_ON ever triggers, we'll get unbalance on pm usage counts. This should use pm_runtime_get_sync().

Plus, we don't really need to duplicate code at function exit.

Quoted 13 lines

+++ b/drivers/gpu/drm/vc4/vc4_hdmi.c

(quoted diff) +2 -0
@@ -146,6 +146,8 @@ vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
 	struct vc4_hdmi *vc4_hdmi = connector_to_vc4_hdmi(connector);
 	bool connected = false;
 
+	WARN_ON(pm_runtime_resume_and_get(&vc4_hdmi->pdev->dev));
+
 	if (vc4_hdmi->hpd_gpio) {
 		if (gpio_get_value_cansleep(vc4_hdmi->hpd_gpio) ^
 		    vc4_hdmi->hpd_active_low)
@@ -167,10 +169,12 @@ vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
 			}
 		}

Review vc4_hdmi_connector_detect(struct drm_connector *connector, bool force) · line 180

Quoted 10 lines
(quoted diff) +2 -0
 
+		pm_runtime_put(&vc4_hdmi->pdev->dev);
 		return connector_status_connected;
 	}
 
 	cec_phys_addr_invalidate(vc4_hdmi->cec_adap);
+	pm_runtime_put(&vc4_hdmi->pdev->dev);
 	return connector_status_disconnected;
 }
 
drivers/gpu/drm/vc4/vc4_hdmi.c +3 -6
index 88a8cb840cd5..3b145f9f6c87 100644
@@ -146,7 +146,7 @@ vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
 	struct vc4_hdmi *vc4_hdmi = connector_to_vc4_hdmi(connector);
 	bool connected = false;
 
-	WARN_ON(pm_runtime_resume_and_get(&vc4_hdmi->pdev->dev));
+	WARN_ON(pm_runtime_get_sync(&vc4_hdmi->pdev->dev));
 
 	if (vc4_hdmi->hpd_gpio) {
 		if (gpio_get_value_cansleep(vc4_hdmi->hpd_gpio) ^
@@ -168,12 +168,9 @@ vc4_hdmi_connector_detect(struct drm_connector *connector, bool force)
 				kfree(edid);
 			}
 		}
+	} else
+		cec_phys_addr_invalidate(vc4_hdmi->cec_adap);
 
-		pm_runtime_put(&vc4_hdmi->pdev->dev);
-		return connector_status_connected;
-	}
-
-	cec_phys_addr_invalidate(vc4_hdmi->cec_adap);
 	pm_runtime_put(&vc4_hdmi->pdev->dev);
 	return connector_status_disconnected;
 }
Signature · 2 lines
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
Tested-by on the v1 series
#116 lore raw
In reply to Sasha Levin

Hi!

Quoted 4 lines

This is the start of the stable review cycle for the 5.10.47 release. There are 101 patches in this series, all will be posted as a response to this one. If anyone has any issues with these being applied, please let me know.

Could I get Cc-ed on these mails for 5.10-stable releases? You'll get more timely test results in exchange.

CIP testing did not find any kernel problems there (siemens board has problems with network connectivity in bootloader?)

https://gitlab.com/cip-project/cip-testing/linux-stable-rc-ci/-/tree/linux-5.10.y

Best regards,
Pavel

Signature · 2 lines
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
#51 lore raw

Re: [PATCH 5.10 049/101] KVM: selftests: Fix kvm_check_cap() assertion

In reply to Sasha Levin

Hi!

Quoted 8 lines

From: Fuad Tabba <[email protected]>

[ Upstream commit d8ac05ea13d789d5491a5920d70a05659015441d ]

KVM_CHECK_EXTENSION ioctl can return any negative value on error, and not necessarily -1. Change the assertion to reflect that.

This is userland code, right?

tools/testing/selftests/kvm/lib/kvm_util.c int kvm_check_cap(long cap) · line 59

Quoted 8 lines

+++ b/tools/testing/selftests/kvm/lib/kvm_util.c

tools/testing/selftests/kvm/lib/kvm_util.c +1 -1
@@ -55,7 +55,7 @@ int kvm_check_cap(long cap)
 		exit(KSFT_SKIP);
 
 	ret = ioctl(kvm_fd, KVM_CHECK_EXTENSION, cap);
-	TEST_ASSERT(ret != -1, "KVM_CHECK_EXTENSION IOCTL failed,\n"
+	TEST_ASSERT(ret >= 0, "KVM_CHECK_EXTENSION IOCTL failed,\n"
 		"  rc: %i errno: %i", ret, errno);

And syscalls return -1 on error in userland, not anything else. So this should not be needed.

Best regards,
Pavel
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

#60 lore raw

Re: [PATCH 5.10 055/101] net: ll_temac: Add memory-barriers for TX BD access

In reply to Sasha Levin

Hi!

Quoted 2 lines

Add a couple of memory-barriers to ensure correct ordering of read/write access to TX BDs.

So... this is dealing with CPU-to-device consistency, not CPU-to-CPU, right?

drivers/net/ethernet/xilinx/ll_temac_main.c static void temac_start_xmit_done(struct net_device *ndev) · line 781

Quoted 10 lines

+++ b/drivers/net/ethernet/xilinx/ll_temac_main.c

drivers/net/ethernet/xilinx/ll_temac_main.c +4 -0
@@ -774,12 +774,15 @@ static void temac_start_xmit_done(struct net_device *ndev)
 	stat = be32_to_cpu(cur_p->app0);
 
 	while (stat & STS_CTRL_APP0_CMPLT) {
+		/* Make sure that the other fields are read after bd is
+		 * released by dma
+		 */
+		rmb();
 		dma_unmap_single(ndev->dev.parent,

Full barrier, as expected.

drivers/net/ethernet/xilinx/ll_temac_main.c static void temac_start_xmit_done(struct net_device *ndev) · line 797

Quoted 8 lines
drivers/net/ethernet/xilinx/ll_temac_main.c +4 -0
@@ -788,6 +791,12 @@ static void temac_start_xmit_done(struct net_device *ndev)
 		ndev->stats.tx_packets++;
 		ndev->stats.tx_bytes += be32_to_cpu(cur_p->len);
 
+		/* app0 must be visible last, as it is used to flag
+		 * availability of the bd
+		 */
+		smp_mb();

SMP-only barrier, but full barrier is needed here AFAICT.

Pavel

Signature · 2 lines
DENX Software Engineering GmbH,      Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany
#52 lore raw

Re: [PATCH 5.10 049/101] KVM: selftests: Fix kvm_check_cap() assertion

In reply to Pavel Machek

Hi Pavel,

Quoted 14 lines · Pavel Machek

Hi!

Quoted 8 lines

From: Fuad Tabba <[email protected]>

[ Upstream commit d8ac05ea13d789d5491a5920d70a05659015441d ]

KVM_CHECK_EXTENSION ioctl can return any negative value on error, and not necessarily -1. Change the assertion to reflect that.

This is userland code, right?

Quoted 8 lines

tools/testing/selftests/kvm/lib/kvm_util.c int kvm_check_cap(long cap) · line 59

Quoted 8 lines

+++ b/tools/testing/selftests/kvm/lib/kvm_util.c

tools/testing/selftests/kvm/lib/kvm_util.c +1 -1
@@ -55,7 +55,7 @@ int kvm_check_cap(long cap)
              exit(KSFT_SKIP);
 
      ret = ioctl(kvm_fd, KVM_CHECK_EXTENSION, cap);
-     TEST_ASSERT(ret != -1, "KVM_CHECK_EXTENSION IOCTL failed,\n"
+     TEST_ASSERT(ret >= 0, "KVM_CHECK_EXTENSION IOCTL failed,\n"
              "  rc: %i errno: %i", ret, errno);

There's at least one case that I am aware of that potentially would return a value other than -1 on error, which is a check for KVM_CAP_MSI_DEVID (-EINVAL, -22):

https://elixir.bootlin.com/linux/latest/source/arch/arm64/kvm/arm.c#L229

Also, considering that this is test code, it might be good to have the check be as strict as possible.

Cheers,
/fuad

Quoted 8 lines

And syscalls return -1 on error in userland, not anything else. So this should not be needed.

Best regards,
Pavel
--
DENX Software Engineering GmbH, Managing Director: Wolfgang Denk
HRB 165235 Munich, Office: Kirchenstr.5, D-82194 Groebenzell, Germany

#61 lore raw

Re: [PATCH 5.10 055/101] net: ll_temac: Add memory-barriers for TX BD access

In reply to Pavel Machek
Quoted 7 lines · Pavel Machek

Hi!

Quoted 2 lines

Add a couple of memory-barriers to ensure correct ordering of read/write access to TX BDs.

So... this is dealing with CPU-to-device consistency, not CPU-to-CPU, right?

Actually, One of both. When looping over the buffers, looking for CMPLT bit in APP0 (in temac_start_xmit_done()), the challenge is CPU-to-device consistency, as the CMPLT bit is set by device, and read by CPU.

But when we clear APP0 (and the other fields) in the same loop, it is CPU-to-CPU, as APP0 is cleared by CPU and read by CPU.

Quoted 18 lines

drivers/net/ethernet/xilinx/ll_temac_main.c static void temac_start_xmit_done(struct net_device *ndev) · line 781

Quoted 11 lines

+++ b/drivers/net/ethernet/xilinx/ll_temac_main.c

drivers/net/ethernet/xilinx/ll_temac_main.c +4 -0
@@ -774,12 +774,15 @@ static void temac_start_xmit_done(struct net_device *ndev)
 	stat = be32_to_cpu(cur_p->app0);
 
 	while (stat & STS_CTRL_APP0_CMPLT) {
+		/* Make sure that the other fields are read after bd is
+		 * released by dma
+		 */
+		rmb();
 		dma_unmap_single(ndev->dev.parent,

Full barrier, as expected.

Quoted 4 lines
drivers/net/ethernet/xilinx/ll_temac_main.c +0 -0
@@ -788,6 +791,12 @@ static void temac_start_xmit_done(struct net_device *ndev)
 		ndev->stats.tx_packets++;
 		ndev->stats.tx_bytes += be32_to_cpu(cur_p->len);
 
Quoted 6 lines

drivers/net/ethernet/xilinx/ll_temac_main.c static void temac_start_xmit_done(struct net_device *ndev) · line 797

Quoted 4 lines
drivers/net/ethernet/xilinx/ll_temac_main.c +4 -0
+		/* app0 must be visible last, as it is used to flag
+		 * availability of the bd
+		 */
+		smp_mb();

SMP-only barrier, but full barrier is needed here AFAICT.

I don't think that is needed. See above.

/Esben

#53 lore raw

Re: [PATCH 5.10 049/101] KVM: selftests: Fix kvm_check_cap() assertion

In reply to Fuad Tabba
Quoted 10 lines · Fuad Tabba
Quoted 5 lines

tools/testing/selftests/kvm/lib/kvm_util.c int kvm_check_cap(long cap) · line 59

Quoted 5 lines
       ret = ioctl(kvm_fd, KVM_CHECK_EXTENSION, cap);
tools/testing/selftests/kvm/lib/kvm_util.c +1 -1
-     TEST_ASSERT(ret != -1, "KVM_CHECK_EXTENSION IOCTL failed,\n"
+     TEST_ASSERT(ret >= 0, "KVM_CHECK_EXTENSION IOCTL failed,\n"
               "  rc: %i errno: %i", ret, errno);

There's at least one case that I am aware of that potentially would return a value other than -1 on error, which is a check for KVM_CAP_MSI_DEVID (-EINVAL, -22):

https://elixir.bootlin.com/linux/latest/source/arch/arm64/kvm/arm.c#L229

In userspace that becomes -1, errno == EINVAL. I probably just misread the "ret != -1" as "ret == 0" when applying this patch; it doesn't hurt but it is certainly not needed for stable.

Paolo