ARM: vexpress/spc: Avoid negative array index when !SMP
Subsystems (12)
Sasha Levin
maintainerPATCH 6/27 drm/msm/disp: check the return value of kzalloc()
In reply to Sasha LevinFrom: Xiaoke Wang <[email protected]>
[ Upstream commit f75e582b0c3ee8f0bddc2248cc8b9175f29c5937 ]
kzalloc() is a memory allocation function which can return NULL when some internal memory errors happen. So it is better to check it to prevent potential wrong memory access.
--- drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/gpu/drm/msm/disp/msm_disp_snapshot_util.c +2 -0
@@ -169,6 +169,8 @@ void msm_disp_snapshot_add_block(struct msm_disp_state *disp_state, u32 len,va_list va; new_blk =kzalloc (sizeof (struct msm_disp_state_block ),GFP_KERNEL ); +if (!new_blk) +return ;va_start (va, fmt);
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 14/27 drm/msm/mdp5: check the return of kzalloc()
In reply to Sasha LevinFrom: Xiaoke Wang <[email protected]>
[ Upstream commit 047ae665577776b7feb11bd4f81f46627cff95e7 ]
kzalloc() is a memory allocation function which can return NULL when some internal memory errors happen. So it is better to check it to prevent potential wrong memory access.
Besides, since mdp5_plane_reset() is void type, so we should better set plane-state to NULL after releasing it.
Patchwork: https://patchwork.freedesktop.org/patch/481055/
--- drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/gpu/drm/msm/disp/mdp5/mdp5_plane.c +3 -0
@@ -90,7 +90,10 @@ static void mdp5_plane_reset(struct drm_plane *plane)__drm_atomic_helper_plane_destroy_state (plane->state);kfree (to_mdp5_plane_state (plane->state)); + plane->state =NULL ; mdp5_state =kzalloc (sizeof (*mdp5_state),GFP_KERNEL ); +if (!mdp5_state) +return ;if (plane->type ==DRM_PLANE_TYPE_PRIMARY ) mdp5_state->base.zpos =STAGE_BASE ;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 15/27 drm/msm: Stop using iommu_present()
In reply to Sasha LevinFrom: Robin Murphy <[email protected]>
[ Upstream commit e2a88eabb02410267519b838fb9b79f5206769be ]
Even if some IOMMU has registered itself on the platform "bus", that doesn't necessarily mean it provides translation for the device we care about. Replace iommu_present() with a more appropriate check.
Patchwork: https://patchwork.freedesktop.org/patch/480707/
--- drivers/gpu/drm/msm/msm_drv.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/msm/msm_drv.c +1 -1
@@ -403,7 +403,7 @@ bool msm_use_mmu(struct drm_device *dev)struct msm_drm_private *priv = dev->dev_private;/* a2xx comes with its own MMU */ -return priv->is_a2xx ||iommu_present (&platform_bus_type); +return priv->is_a2xx ||device_iommu_mapped (dev->dev); }static int msm_init_vram (struct drm_device *dev)
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 5/27 ALSA: usb-audio: Fix undefined behavior due to shift overflowing the constant
In reply to Sasha LevinFrom: Borislav Petkov <[email protected]>
[ Upstream commit 1ef8715975de8bd481abbd0839ed4f49d9e5b0ff ]
Fix:
sound/usb/midi.c: In function ‘snd_usbmidi_out_endpoint_create’:
sound/usb/midi.c:1389:2: error: case label does not reduce to an integer constant
case USB_ID(0xfc08, 0x0101): /* Unknown vendor Cable */
^~~~See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.
[ A slight correction with parentheses around the argument by tiwai ]
--- sound/usb/usbaudio.h | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
sound/usb/usbaudio.h +1 -1
@@ -8,7 +8,7 @@*/ /* handling of USB vendor/product ID pairs as 32-bit numbers */ -#define USB_ID (vendor, product) (((vendor) << 16) | (product)) +#define USB_ID (vendor, product) (((unsigned int)(vendor) << 16) | (product))#define USB_ID_VENDOR (id) ((id) >> 16)#define USB_ID_PRODUCT (id) ((u16)(id))
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 1/27 ARM: vexpress/spc: Avoid negative array index when !SMP
From: Kees Cook <[email protected]>
[ Upstream commit b3f1dd52c991d79118f35e6d1bf4d7cb09882e38 ]
When building multi_v7_defconfig+CONFIG_SMP=n, -Warray-bounds exposes a couple negative array index accesses:
arch/arm/mach-vexpress/spc.c: In function 've_spc_clk_init':arch/arm/mach-vexpress/spc.c:583:21: warning: array subscript -1 is below array bounds of 'bool[2]' {aka '_Bool[2]'} [-Warray-bounds]
583 | if (init_opp_table[cluster])
| ~~~~~~~~~~~~~~^~~~~~~~~
arch/arm/mach-vexpress/spc.c:556:7: note: while referencing 'init_opp_table'
556 | bool init_opp_table[MAX_CLUSTERS] = { false };
| ^~~~~~~~~~~~~~
arch/arm/mach-vexpress/spc.c:592:18: warning: array subscript -1 is below array bounds of 'bool[2]' {aka '_Bool[2]'} [-Warray-bounds]
592 | init_opp_table[cluster] = true;
| ~~~~~~~~~~~~~~^~~~~~~~~
arch/arm/mach-vexpress/spc.c:556:7: note: while referencing 'init_opp_table'
556 | bool init_opp_table[MAX_CLUSTERS] = { false };
| ^~~~~~~~~~~~~~Skip this logic when built !SMP.
--- arch/arm/mach-vexpress/spc.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/arm/mach-vexpress/spc.c +1 -1
@@ -580,7 +580,7 @@ static int __init ve_spc_clk_init(void) } cluster =topology_physical_package_id (cpu_dev->id); -if (init_opp_table[cluster]) +if (cluster <0 || init_opp_table[cluster])continue ;if (ve_init_opp_table (cpu_dev))
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 2/27 reset: renesas: Check return value of reset_control_deassert()
From: Heiner Kallweit <[email protected]>
[ Upstream commit da18980a855edf44270f05455e0ec3f2472f64cc ]
Deasserting the reset is vital, therefore bail out in case of error.
--- drivers/reset/reset-rzg2l-usbphy-ctrl.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/reset/reset-rzg2l-usbphy-ctrl.c +3 -1
@@ -121,7 +121,9 @@ static int rzg2l_usbphy_ctrl_probe(struct platform_device *pdev)return dev_err_probe (dev,PTR_ERR (priv->rstc),"failed to get reset\n" ); -reset_control_deassert (priv->rstc); + error =reset_control_deassert (priv->rstc); +if (error) +return error; priv->rcdev.ops = &rzg2l_usbphy_ctrl_reset_ops; priv->rcdev.of_reset_n_cells =1 ;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 3/27 reset: tegra-bpmp: Restore Handle errors in BPMP response
In reply to Sasha LevinFrom: Sameer Pujar <[email protected]>
[ Upstream commit d1da1052ffad63aa5181b69f20a6952e31f339c2 ]
This reverts following commit 69125b4b9440 ("reset: tegra-bpmp: Revert Handle errors in BPMP response").
The Tegra194 HDA reset failure is fixed by commit d278dc9151a0 ("ALSA:
hda/tegra: Fix Tegra194 HDA reset failure"). The temporary revert of original commit c045ceb5a145 ("reset: tegra-bpmp: Handle errors in BPMP response") can be removed now.
--- drivers/reset/tegra/reset-bpmp.c | 9 ++++++++- 1 file changed, 8 insertions(+), 1 deletion(-)
drivers/reset/tegra/reset-bpmp.c +8 -1
@@ -20,6 +20,7 @@ static int tegra_bpmp_reset_common(struct reset_controller_dev *rstc,struct tegra_bpmp *bpmp =to_tegra_bpmp (rstc);struct mrq_reset_request request;struct tegra_bpmp_message msg; +int err;memset (&request,0 ,sizeof (request)); request.cmd = command;
@@ -30,7 +31,13 @@ static int tegra_bpmp_reset_common(struct reset_controller_dev *rstc, msg.tx.data = &request; msg.tx.size =sizeof (request); -return tegra_bpmp_transfer (bpmp, &msg); + err =tegra_bpmp_transfer (bpmp, &msg); +if (err) +return err; +if (msg.rx.ret) +return -EINVAL ; + +return 0 ; } static int tegra_bpmp_reset_module(struct reset_controller_dev *rstc,
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 4/27 platform/x86: samsung-laptop: Fix an unsigned comparison which can never be negative
In reply to Sasha LevinFrom: Jiapeng Chong <[email protected]>
[ Upstream commit 0284d4d1be753f648f28b77bdfbe6a959212af5c ]
Eliminate the follow smatch warnings:
drivers/platform/x86/samsung-laptop.c:1124 kbd_led_set() warn: unsigned 'value' is never less than zero.
--- drivers/platform/x86/samsung-laptop.c | 2 -- 1 file changed, 2 deletions(-)
drivers/platform/x86/samsung-laptop.c +0 -2
@@ -1121,8 +1121,6 @@ static void kbd_led_set(struct led_classdev *led_cdev,if (value > samsung->kbd_led.max_brightness) value = samsung->kbd_led.max_brightness; -else if (value <0 ) - value =0 ; samsung->kbd_led_wk = value;queue_work (samsung->led_workqueue, &samsung->kbd_led_work);
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 7/27 arm64: dts: imx: Fix imx8*-var-som touchscreen property sizes
In reply to Sasha LevinFrom: Rob Herring <[email protected]>
[ Upstream commit 1bc12d301594eafde0a8529d28d459af81053b3a ]
The common touchscreen properties are all 32-bit, not 16-bit. These properties must not be too important as they are all ignored in case of an error reading them.
--- arch/arm64/boot/dts/freescale/imx8mm-var-som.dtsi | 8 ++++---- arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi | 8 ++++---- 2 files changed, 8 insertions(+), 8 deletions(-)
arch/arm64/boot/dts/freescale/imx8mm-var-som.dtsi +4 -4
@@ -89,12 +89,12 @@ touchscreen@0 { pendown-gpio = <&gpio1 3 GPIO_ACTIVE_LOW>; ti,x-min = /bits/ 16 <125>; - touchscreen-size-x = /bits/ 16 <4008>; + touchscreen-size-x = <4008>; ti,y-min = /bits/ 16 <282>; - touchscreen-size-y = /bits/ 16 <3864>; + touchscreen-size-y = <3864>; ti,x-plate-ohms = /bits/ 16 <180>; - touchscreen-max-pressure = /bits/ 16 <255>; - touchscreen-average-samples = /bits/ 16 <10>; + touchscreen-max-pressure = <255>; + touchscreen-average-samples = <10>; ti,debounce-tol = /bits/ 16 <3>; ti,debounce-rep = /bits/ 16 <1>; ti,settle-delay-usec = /bits/ 16 <150>;
arch/arm64/boot/dts/freescale/imx8mn-var-som.dtsi +4 -4
@@ -70,12 +70,12 @@ touchscreen@0 { pendown-gpio = <&gpio1 3 GPIO_ACTIVE_LOW>; ti,x-min = /bits/ 16 <125>; - touchscreen-size-x = /bits/ 16 <4008>; + touchscreen-size-x = <4008>; ti,y-min = /bits/ 16 <282>; - touchscreen-size-y = /bits/ 16 <3864>; + touchscreen-size-y = <3864>; ti,x-plate-ohms = /bits/ 16 <180>; - touchscreen-max-pressure = /bits/ 16 <255>; - touchscreen-average-samples = /bits/ 16 <10>; + touchscreen-max-pressure = <255>; + touchscreen-average-samples = <10>; ti,debounce-tol = /bits/ 16 <3>; ti,debounce-rep = /bits/ 16 <1>; ti,settle-delay-usec = /bits/ 16 <150>;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 8/27 vxlan: fix error return code in vxlan_fdb_append
In reply to Sasha LevinFrom: Hongbin Wang <[email protected]>
[ Upstream commit 7cea5560bf656b84f9ed01c0cc829d4eecd0640b ]
When kmalloc and dst_cache_init failed, should return ENOMEM rather than ENOBUFS.
--- drivers/net/vxlan.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/vxlan.c +2 -2
@@ -711,11 +711,11 @@ static int vxlan_fdb_append(struct vxlan_fdb *f, rd =kmalloc (sizeof (*rd),GFP_ATOMIC );if (rd ==NULL ) -return -ENOBUFS ; +return -ENOMEM ;if (dst_cache_init (&rd->dst_cache,GFP_ATOMIC )) {kfree (rd); -return -ENOBUFS ; +return -ENOMEM ; } rd->remote_ip = *ip;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 9/27 cifs: Check the IOCB_DIRECT flag, not O_DIRECT
In reply to Sasha LevinFrom: David Howells <[email protected]>
[ Upstream commit 994fd530a512597ffcd713b0f6d5bc916c5698f0 ]
Use the IOCB_DIRECT indicator flag on the I/O context rather than checking to see if the file was opened O_DIRECT.
--- fs/cifs/cifsfs.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
fs/cifs/cifsfs.c +1 -1
@@ -934,7 +934,7 @@ cifs_loose_read_iter(struct kiocb *iocb, struct iov_iter *iter)ssize_t rc;struct inode *inode =file_inode (iocb->ki_filp); -if (iocb->ki_filp->f_flags &O_DIRECT ) +if (iocb->ki_flags &IOCB_DIRECT )return cifs_user_readv (iocb, iter); rc =cifs_revalidate_mapping (inode);
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 10/27 net: atlantic: Avoid out-of-bounds indexing
In reply to Sasha LevinFrom: Kai-Heng Feng <[email protected]>
[ Upstream commit 8d3a6c37d50d5a0504c126c932cc749e6dd9c78f ]
UBSAN warnings are observed on atlantic driver:
[ 294.432996] UBSAN: array-index-out-of-bounds in /build/linux-Qow4fL/linux-5.15.0/drivers/net/ethernet/aquantia/atlantic/aq_nic.c:484:48 [ 294.433695] index 8 is out of range for type 'aq_vec_s *[8]'
The ring is dereferenced right before breaking out the loop, to prevent that from happening, only use the index in the loop to fix the issue.
--- .../net/ethernet/aquantia/atlantic/aq_nic.c | 8 +++---- .../net/ethernet/aquantia/atlantic/aq_vec.c | 24 +++++++++---------- 2 files changed, 16 insertions(+), 16 deletions(-)
drivers/net/ethernet/aquantia/atlantic/aq_nic.c +4 -4
@@ -480,8 +480,8 @@ int aq_nic_start(struct aq_nic_s *self)if (err <0 ) goto err_exit; -for (i =0U , aq_vec = self->aq_vec[0 ]; - self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i]) { +for (i =0U ; self->aq_vecs > i; ++i) { + aq_vec = self->aq_vec[i]; err =aq_vec_start (aq_vec);if (err <0 ) goto err_exit;
@@ -511,8 +511,8 @@ int aq_nic_start(struct aq_nic_s *self)mod_timer (&self->polling_timer, jiffies +AQ_CFG_POLLING_TIMER_INTERVAL ); }else { -for (i =0U , aq_vec = self->aq_vec[0 ]; - self->aq_vecs > i; ++i, aq_vec = self->aq_vec[i]) { +for (i =0U ; self->aq_vecs > i; ++i) { + aq_vec = self->aq_vec[i]; err =aq_pci_func_alloc_irq (self, i, self->ndev->name, aq_vec_isr, aq_vec,aq_vec_get_affinity_mask (aq_vec));
drivers/net/ethernet/aquantia/atlantic/aq_vec.c +12 -12
@@ -43,8 +43,8 @@ static int aq_vec_poll(struct napi_struct *napi, int budget)if (!self) { err = -EINVAL ; }else { -for (i =0U , ring = self->ring[0 ]; - self->tx_rings > i; ++i, ring = self->ring[i]) { +for (i =0U ; self->tx_rings > i; ++i) { + ring = self->ring[i];u64_stats_update_begin (&ring[AQ_VEC_RX_ID ].stats.rx.syncp); ring[AQ_VEC_RX_ID ].stats.rx.polls++;u64_stats_update_end (&ring[AQ_VEC_RX_ID ].stats.rx.syncp);
@@ -182,8 +182,8 @@ int aq_vec_init(struct aq_vec_s *self, const struct aq_hw_ops *aq_hw_ops, self->aq_hw_ops = aq_hw_ops; self->aq_hw = aq_hw; -for (i =0U , ring = self->ring[0 ]; - self->tx_rings > i; ++i, ring = self->ring[i]) { +for (i =0U ; self->tx_rings > i; ++i) { + ring = self->ring[i]; err =aq_ring_init (&ring[AQ_VEC_TX_ID ],ATL_RING_TX );if (err <0 ) goto err_exit;
@@ -224,8 +224,8 @@ int aq_vec_start(struct aq_vec_s *self)unsigned int i =0U ;int err =0 ; -for (i =0U , ring = self->ring[0 ]; - self->tx_rings > i; ++i, ring = self->ring[i]) { +for (i =0U ; self->tx_rings > i; ++i) { + ring = self->ring[i]; err = self->aq_hw_ops->hw_ring_tx_start (self->aq_hw, &ring[AQ_VEC_TX_ID ]);if (err <0 )
@@ -248,8 +248,8 @@ void aq_vec_stop(struct aq_vec_s *self)struct aq_ring_s *ring =NULL ;unsigned int i =0U ; -for (i =0U , ring = self->ring[0 ]; - self->tx_rings > i; ++i, ring = self->ring[i]) { +for (i =0U ; self->tx_rings > i; ++i) { + ring = self->ring[i]; self->aq_hw_ops->hw_ring_tx_stop (self->aq_hw, &ring[AQ_VEC_TX_ID ]);
@@ -268,8 +268,8 @@ void aq_vec_deinit(struct aq_vec_s *self)if (!self) goto err_exit; -for (i =0U , ring = self->ring[0 ]; - self->tx_rings > i; ++i, ring = self->ring[i]) { +for (i =0U ; self->tx_rings > i; ++i) { + ring = self->ring[i];aq_ring_tx_clean (&ring[AQ_VEC_TX_ID ]);aq_ring_rx_deinit (&ring[AQ_VEC_RX_ID ]); }
@@ -297,8 +297,8 @@ void aq_vec_ring_free(struct aq_vec_s *self)if (!self) goto err_exit; -for (i =0U , ring = self->ring[0 ]; - self->tx_rings > i; ++i, ring = self->ring[i]) { +for (i =0U ; self->tx_rings > i; ++i) { + ring = self->ring[i];aq_ring_free (&ring[AQ_VEC_TX_ID ]);if (i < self->rx_rings)aq_ring_free (&ring[AQ_VEC_RX_ID ]);
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 11/27 mt76: Fix undefined behavior due to shift overflowing the constant
In reply to Sasha LevinFrom: Borislav Petkov <[email protected]>
[ Upstream commit dbc2b1764734857d68425468ffa8486e97ab89df ]
Fix:
drivers/net/wireless/mediatek/mt76/mt76x2/pci.c: In function ‘mt76x2e_probe’:
././include/linux/compiler_types.h:352:38: error: call to ‘__compiletime_assert_946’ \
declared with attribute error: FIELD_PREP: mask is not constant
_compiletime_assert(condition, msg, __compiletime_assert_, __COUNTER__)
See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.
--- drivers/net/wireless/mediatek/mt76/mt76x2/pci.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/mediatek/mt76/mt76x2/pci.c +1 -1
@@ -80,7 +80,7 @@ mt76x2e_probe(struct pci_dev *pdev, const struct pci_device_id *id)mt76_rmw_field (dev,0x15a10 ,0x1f <<16 ,0x9 );/* RG_SSUSB_G1_CDR_BIC_LTR = 0xf */ -mt76_rmw_field (dev,0x15a0c ,0xf <<28 ,0xf ); +mt76_rmw_field (dev,0x15a0c ,0xfU <<28 ,0xf );/* RG_SSUSB_CDR_BR_PE1D = 0x3 */ mt76_rmw_field (dev,0x15c58 ,0x3 <<6 ,0x3 );
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 12/27 brcmfmac: sdio: Fix undefined behavior due to shift overflowing the constant
In reply to Sasha LevinFrom: Borislav Petkov <[email protected]>
[ Upstream commit 6fb3a5868b2117611f41e421e10e6a8c2a13039a ]
Fix:
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c: In function ‘brcmf_sdio_drivestrengthinit’:
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3798:2: error: case label does not reduce to an integer constant
case SDIOD_DRVSTR_KEY(BRCM_CC_43143_CHIP_ID, 17):
^~~~
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c:3809:2: error: case label does not reduce to an integer constant
case SDIOD_DRVSTR_KEY(BRCM_CC_43362_CHIP_ID, 13):
^~~~See https://lore.kernel.org/r/YkwQ6%[email protected] for the gory details as to why it triggers with older gccs only.
--- drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/broadcom/brcm80211/brcmfmac/sdio.c +1 -1
@@ -557,7 +557,7 @@ enum brcmf_sdio_frmtype { BRCMF_SDIO_FT_SUB, }; -#define SDIOD_DRVSTR_KEY (chip, pmu) (((chip) << 16) | (pmu)) +#define SDIOD_DRVSTR_KEY (chip, pmu) (((unsigned int)(chip) << 16) | (pmu))/* SDIO Pad drive strength to select value mappings */ struct sdiod_drive_str {
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 13/27 dpaa_eth: Fix missing of_node_put in dpaa_get_ts_info()
In reply to Sasha LevinFrom: Lv Ruyi <[email protected]>
[ Upstream commit 1a7eb80d170c28be2928433702256fe2a0bd1e0f ]
Both of of_get_parent() and of_parse_phandle() return node pointer with refcount incremented, use of_node_put() on it to decrease refcount when done.
--- drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/net/ethernet/freescale/dpaa/dpaa_ethtool.c +6 -2
@@ -489,11 +489,15 @@ static int dpaa_get_ts_info(struct net_device *net_dev, info->phc_index =-1 ; fman_node =of_get_parent (mac_node); -if (fman_node) +if (fman_node) { ptp_node =of_parse_phandle (fman_node,"ptimer-handle" ,0 ); +of_node_put (fman_node); + } -if (ptp_node) +if (ptp_node) { ptp_dev =of_find_device_by_node (ptp_node); +of_node_put (ptp_node); + }if (ptp_dev) ptp =platform_get_drvdata (ptp_dev);
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 17/27 scsi: iscsi: Move iscsi_ep_disconnect()
In reply to Sasha LevinFrom: Mike Christie <[email protected]>
[ Upstream commit c34f95e98d8fb750eefd4f3fe58b4f8b5e89253b ]
This patch moves iscsi_ep_disconnect() so it can be called earlier in the next patch.
--- drivers/scsi/scsi_transport_iscsi.c | 38 ++++++++++++++--------------- 1 file changed, 19 insertions(+), 19 deletions(-)
drivers/scsi/scsi_transport_iscsi.c +19 -19
@@ -2236,6 +2236,25 @@ static void iscsi_stop_conn(struct iscsi_cls_conn *conn, int flag)ISCSI_DBG_TRANS_CONN (conn,"Stopping conn done.\n" ); } +static void iscsi_ep_disconnect (struct iscsi_cls_conn *conn,bool is_active) +{ +struct iscsi_cls_session *session =iscsi_conn_to_session (conn); +struct iscsi_endpoint *ep; + +ISCSI_DBG_TRANS_CONN (conn,"disconnect ep.\n" ); + conn->state =ISCSI_CONN_FAILED ; + +if (!conn->ep || !session->transport->ep_disconnect) +return ; + + ep = conn->ep; + conn->ep =NULL ; + + session->transport->unbind_conn (conn, is_active); + session->transport->ep_disconnect (ep); +ISCSI_DBG_TRANS_CONN (conn,"disconnect ep done.\n" ); +} +static int iscsi_if_stop_conn (struct iscsi_transport *transport,struct iscsi_uevent *ev) {
@@ -2276,25 +2295,6 @@ static int iscsi_if_stop_conn(struct iscsi_transport *transport,return 0 ; } -static void iscsi_ep_disconnect (struct iscsi_cls_conn *conn,bool is_active) -{ -struct iscsi_cls_session *session =iscsi_conn_to_session (conn); -struct iscsi_endpoint *ep; - -ISCSI_DBG_TRANS_CONN (conn,"disconnect ep.\n" ); - conn->state =ISCSI_CONN_FAILED ; - -if (!conn->ep || !session->transport->ep_disconnect) -return ; - - ep = conn->ep; - conn->ep =NULL ; - - session->transport->unbind_conn (conn, is_active); - session->transport->ep_disconnect (ep); -ISCSI_DBG_TRANS_CONN (conn,"disconnect ep done.\n" ); -} -static void iscsi_cleanup_conn_work_fn (struct work_struct *work) { struct iscsi_cls_conn *conn = container_of(work, struct iscsi_cls_conn,
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 16/27 net: macb: Restart tx only if queue pointer is lagging
In reply to Sasha LevinFrom: Tomas Melin <[email protected]>
[ Upstream commit 5ad7f18cd82cee8e773d40cc7a1465a526f2615c ]
commit 4298388574da ("net: macb: restart tx after tx used bit read") added support for restarting transmission. Restarting tx does not work in case controller asserts TXUBR interrupt and TQBP is already at the end of the tx queue. In that situation, restarting tx will immediately cause assertion of another TXUBR interrupt. The driver will end up in an infinite interrupt loop which it cannot break out of.
For cases where TQBP is at the end of the tx queue, instead only clear TX_USED interrupt. As more data gets pushed to the queue, transmission will resume.
This issue was observed on a Xilinx Zynq-7000 based board. During stress test of the network interface,
driver would get stuck on interrupt loop within seconds or minutes causing CPU to stall.
--- drivers/net/ethernet/cadence/macb_main.c | 8 ++++++++ 1 file changed, 8 insertions(+)
drivers/net/ethernet/cadence/macb_main.c +8 -0
@@ -1689,6 +1689,7 @@ static void macb_tx_restart(struct macb_queue *queue)unsigned int head = queue->tx_head;unsigned int tail = queue->tx_tail;struct macb *bp = queue->bp; +unsigned int head_idx, tbqp;if (bp->caps &MACB_CAPS_ISR_CLEAR_ON_WRITE )queue_writel (queue,ISR ,MACB_BIT (TXUBR ));
@@ -1696,6 +1697,13 @@ static void macb_tx_restart(struct macb_queue *queue)if (head == tail)return ; + tbqp =queue_readl (queue,TBQP ) /macb_dma_desc_get_size (bp); + tbqp =macb_adj_dma_desc_idx (bp,macb_tx_ring_wrap (bp, tbqp)); + head_idx =macb_adj_dma_desc_idx (bp,macb_tx_ring_wrap (bp, head)); + +if (tbqp == head_idx) +return ; +macb_writel (bp,NCR ,macb_readl (bp,NCR ) |MACB_BIT (TSTART )); }
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 18/27 scsi: iscsi: Fix offload conn cleanup when iscsid restarts
In reply to Sasha LevinFrom: Mike Christie <[email protected]>
[ Upstream commit cbd2283aaf47fef4ded4b29124b1ef3beb515f3a ]
When userspace restarts during boot or upgrades it won't know about the offload driver's endpoint and connection mappings. iscsid will start by cleaning up the old session by doing a stop_conn call. Later, if we are able to create a new connection, we clean up the old endpoint during the binding stage. The problem is that if we do stop_conn before doing the ep_disconnect call offload, drivers can still be executing I/O. We then might free tasks from the under the card/driver.
This moves the ep_disconnect call to before we do the stop_conn call for this case. It will then work and look like a normal recovery/cleanup procedure from the driver's point of view.
--- drivers/scsi/scsi_transport_iscsi.c | 48 +++++++++++++++++------------ 1 file changed, 28 insertions(+), 20 deletions(-)
drivers/scsi/scsi_transport_iscsi.c +28 -20
@@ -2255,6 +2255,23 @@ static void iscsi_ep_disconnect(struct iscsi_cls_conn *conn, bool is_active)ISCSI_DBG_TRANS_CONN (conn,"disconnect ep done.\n" ); } +static void iscsi_if_disconnect_bound_ep (struct iscsi_cls_conn *conn, +struct iscsi_endpoint *ep, +bool is_active) +{ +/* Check if this was a conn error and the kernel took ownership */ +if (!test_bit (ISCSI_CLS_CONN_BIT_CLEANUP , &conn->flags)) { +iscsi_ep_disconnect (conn, is_active); + }else { +ISCSI_DBG_TRANS_CONN (conn,"flush kernel conn cleanup.\n" ); +mutex_unlock (&conn->ep_mutex); + +flush_work (&conn->cleanup_work); + +mutex_lock (&conn->ep_mutex); + } +} +static int iscsi_if_stop_conn (struct iscsi_transport *transport,struct iscsi_uevent *ev) {
@@ -2275,6 +2292,16 @@ static int iscsi_if_stop_conn(struct iscsi_transport *transport,cancel_work_sync (&conn->cleanup_work);iscsi_stop_conn (conn, flag); }else { +/* +* For offload, when iscsid is restarted it won't know about +* existing endpoints so it can't do a ep_disconnect. We clean +* it up here for userspace. +*/ +mutex_lock (&conn->ep_mutex); +if (conn->ep) +iscsi_if_disconnect_bound_ep (conn, conn->ep, true); +mutex_unlock (&conn->ep_mutex); +/* * Figure out if it was the kernel or userspace initiating this. */
@@ -3003,16 +3030,7 @@ static int iscsi_if_ep_disconnect(struct iscsi_transport *transport, }mutex_lock (&conn->ep_mutex); -/* Check if this was a conn error and the kernel took ownership */ -if (test_bit (ISCSI_CLS_CONN_BIT_CLEANUP , &conn->flags)) { -ISCSI_DBG_TRANS_CONN (conn,"flush kernel conn cleanup.\n" ); -mutex_unlock (&conn->ep_mutex); - -flush_work (&conn->cleanup_work); - goto put_ep; - } - -iscsi_ep_disconnect (conn, false); +iscsi_if_disconnect_bound_ep (conn, ep, false);mutex_unlock (&conn->ep_mutex); put_ep:iscsi_put_endpoint (ep);
@@ -3723,16 +3741,6 @@ static int iscsi_if_transport_conn(struct iscsi_transport *transport,switch (nlh->nlmsg_type) {case ISCSI_UEVENT_BIND_CONN : -if (conn->ep) { -/* -* For offload boot support where iscsid is restarted -* during the pivot root stage, the ep will be intact -* here when the new iscsid instance starts up and -* reconnects. -*/ -iscsi_ep_disconnect (conn, true); - } - session =iscsi_session_lookup (ev->u.b_conn.sid);if (!session) { err = -EINVAL ;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 19/27 scsi: iscsi: Release endpoint ID when its freed
In reply to Sasha LevinFrom: Mike Christie <[email protected]>
[ Upstream commit 3c6ae371b8a1ffba1fc415989fd581ebf841ed0a ]
We can't release the endpoint ID until all references to the endpoint have been dropped or it could be allocated while in use. This has us use an idr instead of looping over all conns to find a free ID and then free the ID when all references have been dropped instead of when the device is only deleted.
--- drivers/scsi/scsi_transport_iscsi.c | 71 ++++++++++++++--------------- include/scsi/scsi_transport_iscsi.h | 2 +- 2 files changed, 36 insertions(+), 37 deletions(-)
drivers/scsi/scsi_transport_iscsi.c +35 -36
@@ -86,6 +86,9 @@ struct iscsi_internal {struct transport_container session_cont; }; +static DEFINE_IDR (iscsi_ep_idr); +static DEFINE_MUTEX (iscsi_ep_idr_mutex); +static atomic_t iscsi_session_nr;/* sysfs session id for next new session */ static struct workqueue_struct *iscsi_eh_timer_workq;
@@ -169,6 +172,11 @@ struct device_attribute dev_attr_##_prefix##_##_name = \static void iscsi_endpoint_release (struct device *dev) {struct iscsi_endpoint *ep =iscsi_dev_to_endpoint (dev); + +mutex_lock (&iscsi_ep_idr_mutex); +idr_remove (&iscsi_ep_idr, ep->id); +mutex_unlock (&iscsi_ep_idr_mutex); +kfree (ep); }
@@ -181,7 +189,7 @@ static ssize_t show_ep_handle(struct device *dev, struct device_attribute *attr, char *buf) { struct iscsi_endpoint *ep = iscsi_dev_to_endpoint(dev); -return sysfs_emit (buf,"%llu\n" , (unsigned long long ) ep->id); +return sysfs_emit (buf,"%d\n" , ep->id); } static ISCSI_ATTR(ep, handle, S_IRUGO, show_ep_handle, NULL);
@@ -194,48 +202,32 @@ static struct attribute_group iscsi_endpoint_group = { .attrs = iscsi_endpoint_attrs, }; -#define ISCSI_MAX_EPID -1 - -static int iscsi_match_epid (struct device *dev,const void *data) -{ -struct iscsi_endpoint *ep =iscsi_dev_to_endpoint (dev); -const uint64_t *epid = data; - -return *epid == ep->id; -} -struct iscsi_endpoint *iscsi_create_endpoint (int dd_size) { -struct device *dev;struct iscsi_endpoint *ep; -uint64_t id; -int err; - -for (id =1 ; id <ISCSI_MAX_EPID ; id++) { - dev =class_find_device (&iscsi_endpoint_class,NULL , &id, - iscsi_match_epid); -if (!dev) -break ; -else -put_device (dev); - } -if (id ==ISCSI_MAX_EPID ) { -printk (KERN_ERR "Too many connections. Max supported %u\n" , -ISCSI_MAX_EPID -1 ); -return NULL ; - } +int err, id; ep =kzalloc (sizeof (*ep) + dd_size,GFP_KERNEL );if (!ep)return NULL ; +mutex_lock (&iscsi_ep_idr_mutex); + id =idr_alloc (&iscsi_ep_idr, ep,0 ,-1 ,GFP_NOIO ); +if (id <0 ) { +mutex_unlock (&iscsi_ep_idr_mutex); +printk (KERN_ERR "Could not allocate endpoint ID. Error %d.\n" , + id); + goto free_ep; + } +mutex_unlock (&iscsi_ep_idr_mutex); + ep->id = id; ep->dev.class = &iscsi_endpoint_class; -dev_set_name (&ep->dev,"ep-%llu" , (unsigned long long ) id); +dev_set_name (&ep->dev,"ep-%d" , id); err =device_register (&ep->dev);if (err) - goto free_ep; + goto free_id; err =sysfs_create_group (&ep->dev.kobj, &iscsi_endpoint_group);if (err)
@@ -249,6 +241,10 @@ iscsi_create_endpoint(int dd_size)device_unregister (&ep->dev);return NULL ; +free_id: +mutex_lock (&iscsi_ep_idr_mutex); +idr_remove (&iscsi_ep_idr, id); +mutex_unlock (&iscsi_ep_idr_mutex); free_ep:kfree (ep);return NULL ;
@@ -276,14 +272,17 @@ EXPORT_SYMBOL_GPL(iscsi_put_endpoint);*/ struct iscsi_endpoint *iscsi_lookup_endpoint (u64 handle) { -struct device *dev; +struct iscsi_endpoint *ep; - dev =class_find_device (&iscsi_endpoint_class,NULL , &handle, - iscsi_match_epid); -if (!dev) -return NULL ; +mutex_lock (&iscsi_ep_idr_mutex); + ep =idr_find (&iscsi_ep_idr, handle); +if (!ep) + goto unlock; -return iscsi_dev_to_endpoint (dev); +get_device (&ep->dev); +unlock: +mutex_unlock (&iscsi_ep_idr_mutex); +return ep; }EXPORT_SYMBOL_GPL (iscsi_lookup_endpoint);
include/scsi/scsi_transport_iscsi.h +1 -1
@@ -294,7 +294,7 @@ extern void iscsi_host_for_each_session(struct Scsi_Host *shost,struct iscsi_endpoint {void *dd_data;/* LLD private data */ struct device dev; -uint64_t id; +int id;struct iscsi_cls_conn *conn; };
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 20/27 scsi: iscsi: Merge suspend fields
In reply to Sasha LevinFrom: Mike Christie <[email protected]>
[ Upstream commit 5bd856256f8c03e329f8ff36d8c8efcb111fe6df ]
Move the tx and rx suspend fields into one flags field.
--- drivers/scsi/bnx2i/bnx2i_hwi.c | 2 +- drivers/scsi/bnx2i/bnx2i_iscsi.c | 2 +- drivers/scsi/cxgbi/libcxgbi.c | 6 +++--- drivers/scsi/libiscsi.c | 20 ++++++++++---------- drivers/scsi/libiscsi_tcp.c | 2 +- include/scsi/libiscsi.h | 9 +++++---- 6 files changed, 21 insertions(+), 20 deletions(-)
drivers/scsi/bnx2i/bnx2i_hwi.c +1 -1
@@ -1977,7 +1977,7 @@ static int bnx2i_process_new_cqes(struct bnx2i_conn *bnx2i_conn)if (nopin->cq_req_sn != qp->cqe_exp_seq_sn)break ; -if (unlikely (test_bit (ISCSI_SUSPEND_BIT , &conn->suspend_rx))) { +if (unlikely (test_bit (ISCSI_CONN_FLAG_SUSPEND_RX , &conn->flags))) {if (nopin->op_code ==ISCSI_OP_NOOP_IN && nopin->itt == (u16 )RESERVED_ITT ) { printk(KERN_ALERT "bnx2i: Unsolicited "
drivers/scsi/bnx2i/bnx2i_iscsi.c +1 -1
@@ -1721,7 +1721,7 @@ static int bnx2i_tear_down_conn(struct bnx2i_hba *hba,struct iscsi_conn *conn = ep->conn->cls_conn->dd_data;/* Must suspend all rx queue activity for this ep */ -set_bit (ISCSI_SUSPEND_BIT , &conn->suspend_rx); +set_bit (ISCSI_CONN_FLAG_SUSPEND_RX , &conn->flags); }/* CONN_DISCONNECT timeout may or may not be an issue depending * on what transcribed in TCP layer, different targets behave
drivers/scsi/cxgbi/libcxgbi.c +3 -3
@@ -1634,11 +1634,11 @@ void cxgbi_conn_pdu_ready(struct cxgbi_sock *csk)log_debug (1 <<CXGBI_DBG_PDU_RX ,"csk 0x%p, conn 0x%p.\n" , csk, conn); -if (unlikely (!conn || conn->suspend_rx)) { +if (unlikely (!conn ||test_bit (ISCSI_CONN_FLAG_SUSPEND_RX , &conn->flags))) {log_debug (1 <<CXGBI_DBG_PDU_RX , -"csk 0x%p, conn 0x%p, id %d, suspend_rx %lu!\n" , +"csk 0x%p, conn 0x%p, id %d, conn flags 0x%lx!\n" , csk, conn, conn ? conn->id :0xFF , - conn ? conn->suspend_rx :0xFF ); + conn ? conn->flags :0xFF );return ; }
drivers/scsi/libiscsi.c +10 -10
@@ -1392,8 +1392,8 @@ static bool iscsi_set_conn_failed(struct iscsi_conn *conn)if (conn->stop_stage ==0 ) session->state =ISCSI_STATE_FAILED ; -set_bit (ISCSI_SUSPEND_BIT , &conn->suspend_tx); -set_bit (ISCSI_SUSPEND_BIT , &conn->suspend_rx); +set_bit (ISCSI_CONN_FLAG_SUSPEND_TX , &conn->flags); +set_bit (ISCSI_CONN_FLAG_SUSPEND_RX , &conn->flags);return true; }
@@ -1454,7 +1454,7 @@ static int iscsi_xmit_task(struct iscsi_conn *conn, struct iscsi_task *task,* Do this after dropping the extra ref because if this was a requeue * it's removed from that list and cleanup_queued_task would miss it. */ -if (test_bit (ISCSI_SUSPEND_BIT , &conn->suspend_tx)) { +if (test_bit (ISCSI_CONN_FLAG_SUSPEND_TX , &conn->flags)) {/* * Save the task and ref in case we weren't cleaning up this * task and get woken up again.
@@ -1532,7 +1532,7 @@ static int iscsi_data_xmit(struct iscsi_conn *conn)int rc =0 ;spin_lock_bh (&conn->session->frwd_lock); -if (test_bit (ISCSI_SUSPEND_BIT , &conn->suspend_tx)) { +if (test_bit (ISCSI_CONN_FLAG_SUSPEND_TX , &conn->flags)) {ISCSI_DBG_SESSION (conn->session,"Tx suspended!\n" );spin_unlock_bh (&conn->session->frwd_lock);return -ENODATA ;
@@ -1746,7 +1746,7 @@ int iscsi_queuecommand(struct Scsi_Host *host, struct scsi_cmnd *sc) goto fault; } -if (test_bit (ISCSI_SUSPEND_BIT , &conn->suspend_tx)) { +if (test_bit (ISCSI_CONN_FLAG_SUSPEND_TX , &conn->flags)) { reason =FAILURE_SESSION_IN_RECOVERY ; sc->result =DID_REQUEUE <<16 ; goto fault;
@@ -1935,7 +1935,7 @@ static void fail_scsi_tasks(struct iscsi_conn *conn, u64 lun, int error)void iscsi_suspend_queue (struct iscsi_conn *conn) {spin_lock_bh (&conn->session->frwd_lock); -set_bit (ISCSI_SUSPEND_BIT , &conn->suspend_tx); +set_bit (ISCSI_CONN_FLAG_SUSPEND_TX , &conn->flags);spin_unlock_bh (&conn->session->frwd_lock); }EXPORT_SYMBOL_GPL (iscsi_suspend_queue);
@@ -1953,7 +1953,7 @@ void iscsi_suspend_tx(struct iscsi_conn *conn)struct Scsi_Host *shost = conn->session->host;struct iscsi_host *ihost =shost_priv (shost); -set_bit (ISCSI_SUSPEND_BIT , &conn->suspend_tx); +set_bit (ISCSI_CONN_FLAG_SUSPEND_TX , &conn->flags);if (ihost->workq)flush_workqueue (ihost->workq); }
@@ -1961,7 +1961,7 @@ EXPORT_SYMBOL_GPL(iscsi_suspend_tx);static void iscsi_start_tx (struct iscsi_conn *conn) { -clear_bit (ISCSI_SUSPEND_BIT , &conn->suspend_tx); +clear_bit (ISCSI_CONN_FLAG_SUSPEND_TX , &conn->flags);iscsi_conn_queue_work (conn); }
@@ -3324,8 +3324,8 @@ int iscsi_conn_bind(struct iscsi_cls_session *cls_session,/* * Unblock xmitworker(), Login Phase will pass through. */ -clear_bit (ISCSI_SUSPEND_BIT , &conn->suspend_rx); -clear_bit (ISCSI_SUSPEND_BIT , &conn->suspend_tx); +clear_bit (ISCSI_CONN_FLAG_SUSPEND_RX , &conn->flags); +clear_bit (ISCSI_CONN_FLAG_SUSPEND_TX , &conn->flags);return 0 ; }EXPORT_SYMBOL_GPL (iscsi_conn_bind);
drivers/scsi/libiscsi_tcp.c +1 -1
@@ -927,7 +927,7 @@ int iscsi_tcp_recv_skb(struct iscsi_conn *conn, struct sk_buff *skb,*/ conn->last_recv = jiffies; -if (unlikely (conn->suspend_rx)) { +if (unlikely (test_bit (ISCSI_CONN_FLAG_SUSPEND_RX , &conn->flags))) {ISCSI_DBG_TCP (conn,"Rx suspended!\n" ); *status =ISCSI_TCP_SUSPENDED ;return 0 ;
include/scsi/libiscsi.h +5 -4
@@ -52,8 +52,10 @@ enum {#define ISID_SIZE 6 -/* Connection suspend "bit" */ -#define ISCSI_SUSPEND_BIT 1 +/* Connection flags */ +#define ISCSI_CONN_FLAG_SUSPEND_TX BIT(0) +#define ISCSI_CONN_FLAG_SUSPEND_RX BIT(1) +#define ISCSI_ITT_MASK 0x1fff#define ISCSI_TOTAL_CMDS_MAX 4096
@@ -199,8 +201,7 @@ struct iscsi_conn {struct list_head cmdqueue;/* data-path cmd queue */ struct list_head requeue;/* tasks needing another run */ struct work_struct xmitwork;/* per-conn. xmit workqueue */ -unsigned long suspend_tx;/* suspend Tx */ -unsigned long suspend_rx;/* suspend Rx */ +unsigned long flags;/* ISCSI_CONN_FLAGs */ /* negotiated params */ unsigned max_recv_dlength;/* initiator_max_recv_dsl*/
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 21/27 scsi: iscsi: Fix NOP handling during conn recovery
In reply to Sasha LevinFrom: Mike Christie <[email protected]>
[ Upstream commit 44ac97109e42f87b1a34954704b81b6c8eca80c4 ]
If a offload driver doesn't use the xmit workqueue, then when we are doing ep_disconnect libiscsi can still inject PDUs to the driver. This adds a check for if the connection is bound before trying to inject PDUs.
--- drivers/scsi/libiscsi.c | 7 ++++++- include/scsi/libiscsi.h | 2 +- 2 files changed, 7 insertions(+), 2 deletions(-)
drivers/scsi/libiscsi.c +6 -1
@@ -678,7 +678,8 @@ __iscsi_conn_send_pdu(struct iscsi_conn *conn, struct iscsi_hdr *hdr,struct iscsi_task *task;itt_t itt; -if (session->state ==ISCSI_STATE_TERMINATE ) +if (session->state ==ISCSI_STATE_TERMINATE || + !test_bit (ISCSI_CONN_FLAG_BOUND , &conn->flags))return NULL ;if (opcode ==ISCSI_OP_LOGIN || opcode ==ISCSI_OP_TEXT ) {
@@ -2214,6 +2215,8 @@ void iscsi_conn_unbind(struct iscsi_cls_conn *cls_conn, bool is_active)iscsi_suspend_tx (conn);spin_lock_bh (&session->frwd_lock); +clear_bit (ISCSI_CONN_FLAG_BOUND , &conn->flags); +if (!is_active) {/* * if logout timed out before userspace could even send a PDU
@@ -3312,6 +3315,8 @@ int iscsi_conn_bind(struct iscsi_cls_session *cls_session,spin_lock_bh (&session->frwd_lock);if (is_leading) session->leadconn = conn; + +set_bit (ISCSI_CONN_FLAG_BOUND , &conn->flags);spin_unlock_bh (&session->frwd_lock);/*
include/scsi/libiscsi.h +1 -1
@@ -55,7 +55,7 @@ enum {/* Connection flags */ #define ISCSI_CONN_FLAG_SUSPEND_TX BIT(0)#define ISCSI_CONN_FLAG_SUSPEND_RX BIT(1) - +#define ISCSI_CONN_FLAG_BOUND BIT(2)#define ISCSI_ITT_MASK 0x1fff#define ISCSI_TOTAL_CMDS_MAX 4096
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 26/27 nvme-pci: disable namespace identifiers for the MAXIO MAP1002/1202
In reply to Sasha LevinFrom: Christoph Hellwig <[email protected]>
[ Upstream commit a98a945b80f8684121d477ae68ebc01da953da1f ]
The MAXIO MAP1002/1202 controllers reports completely bogus Namespace identifiers that even change after suspend cycles. Disable using the Identifiers entirely.
--- drivers/nvme/host/pci.c | 4 ++++ 1 file changed, 4 insertions(+)
drivers/nvme/host/pci.c +4 -0
@@ -3352,6 +3352,10 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, { PCI_DEVICE(0x2646, 0x2263), /* KINGSTON A2000 NVMe SSD */ .driver_data = NVME_QUIRK_NO_DEEPEST_PS, }, + { PCI_DEVICE(0x1e4B, 0x1002), /* MAXIO MAP1002 */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, + { PCI_DEVICE(0x1e4B, 0x1202), /* MAXIO MAP1202 */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(PCI_VENDOR_ID_AMAZON, 0x0061), .driver_data = NVME_QUIRK_DMA_ADDRESS_BITS_48, }, { PCI_DEVICE(PCI_VENDOR_ID_AMAZON, 0x0065),
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 22/27 scsi: qedi: Fix failed disconnect handling
In reply to Sasha LevinFrom: Mike Christie <[email protected]>
[ Upstream commit 857b06527f707f5df634b854898a191b5c1d0272 ]
We set the qedi_ep state to EP_STATE_OFLDCONN_START when the ep is created. Then in qedi_set_path we kick off the offload work. If userspace times out the connection and calls ep_disconnect, qedi will only flush the offload work if the qedi_ep state has transitioned away from EP_STATE_OFLDCONN_START. If we can't connect we will not have transitioned state and will leave the offload work running, and we will free the qedi_ep from under it.
This patch just has us init the work when we create the ep, then always flush it.
--- drivers/scsi/qedi/qedi_iscsi.c | 69 +++++++++++++++++----------------- 1 file changed, 34 insertions(+), 35 deletions(-)
drivers/scsi/qedi/qedi_iscsi.c +34 -35
@@ -859,6 +859,37 @@ static int qedi_task_xmit(struct iscsi_task *task)return qedi_iscsi_send_ioreq (task); } +static void qedi_offload_work (struct work_struct *work) +{ +struct qedi_endpoint *qedi_ep = + container_of(work, struct qedi_endpoint, offload_work); +struct qedi_ctx *qedi; +int wait_delay =5 *HZ ; +int ret; + + qedi = qedi_ep->qedi; + + ret =qedi_iscsi_offload_conn (qedi_ep); +if (ret) { +QEDI_ERR (&qedi->dbg_ctx, +"offload error: iscsi_cid=%u, qedi_ep=%p, ret=%d\n" , + qedi_ep->iscsi_cid, qedi_ep, ret); + qedi_ep->state =EP_STATE_OFLDCONN_FAILED ; +return ; + } + + ret =wait_event_interruptible_timeout (qedi_ep->tcp_ofld_wait, + (qedi_ep->state == +EP_STATE_OFLDCONN_COMPL ), + wait_delay); +if (ret <=0 || qedi_ep->state !=EP_STATE_OFLDCONN_COMPL ) { + qedi_ep->state =EP_STATE_OFLDCONN_FAILED ; +QEDI_ERR (&qedi->dbg_ctx, +"Offload conn TIMEOUT iscsi_cid=%u, qedi_ep=%p\n" , + qedi_ep->iscsi_cid, qedi_ep); + } +} +static struct iscsi_endpoint *qedi_ep_connect (struct Scsi_Host *shost,struct sockaddr *dst_addr,int non_blocking)
@@ -907,6 +938,7 @@ qedi_ep_connect(struct Scsi_Host *shost, struct sockaddr *dst_addr, } qedi_ep = ep->dd_data;memset (qedi_ep,0 ,sizeof (struct qedi_endpoint )); +INIT_WORK (&qedi_ep->offload_work, qedi_offload_work); qedi_ep->state =EP_STATE_IDLE ; qedi_ep->iscsi_cid = (u32)-1 ; qedi_ep->qedi = qedi;
@@ -1055,12 +1087,11 @@ static void qedi_ep_disconnect(struct iscsi_endpoint *ep) qedi_ep = ep->dd_data; qedi = qedi_ep->qedi; +flush_work (&qedi_ep->offload_work); +if (qedi_ep->state ==EP_STATE_OFLDCONN_START ) goto ep_exit_recover; -if (qedi_ep->state !=EP_STATE_OFLDCONN_NONE ) -flush_work (&qedi_ep->offload_work); -if (qedi_ep->conn) { qedi_conn = qedi_ep->conn; abrt_conn = qedi_conn->abrt_conn;
@@ -1234,37 +1265,6 @@ static int qedi_data_avail(struct qedi_ctx *qedi, u16 vlanid)return rc; } -static void qedi_offload_work (struct work_struct *work) -{ -struct qedi_endpoint *qedi_ep = - container_of(work, struct qedi_endpoint, offload_work); -struct qedi_ctx *qedi; -int wait_delay =5 *HZ ; -int ret; - - qedi = qedi_ep->qedi; - - ret =qedi_iscsi_offload_conn (qedi_ep); -if (ret) { -QEDI_ERR (&qedi->dbg_ctx, -"offload error: iscsi_cid=%u, qedi_ep=%p, ret=%d\n" , - qedi_ep->iscsi_cid, qedi_ep, ret); - qedi_ep->state =EP_STATE_OFLDCONN_FAILED ; -return ; - } - - ret =wait_event_interruptible_timeout (qedi_ep->tcp_ofld_wait, - (qedi_ep->state == -EP_STATE_OFLDCONN_COMPL ), - wait_delay); -if ((ret <=0 ) || (qedi_ep->state !=EP_STATE_OFLDCONN_COMPL )) { - qedi_ep->state =EP_STATE_OFLDCONN_FAILED ; -QEDI_ERR (&qedi->dbg_ctx, -"Offload conn TIMEOUT iscsi_cid=%u, qedi_ep=%p\n" , - qedi_ep->iscsi_cid, qedi_ep); - } -} -static int qedi_set_path (struct Scsi_Host *shost,struct iscsi_path *path_data) {struct qedi_ctx *qedi;
@@ -1380,7 +1380,6 @@ static int qedi_set_path(struct Scsi_Host *shost, struct iscsi_path *path_data) qedi_ep->dst_addr, qedi_ep->dst_port); } -INIT_WORK (&qedi_ep->offload_work, qedi_offload_work);queue_work (qedi->offload_thread, &qedi_ep->offload_work); ret =0 ;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 24/27 VFS: filename_create(): fix incorrect intent.
In reply to Sasha LevinFrom: NeilBrown <[email protected]>
[ Upstream commit b3d4650d82c71b9c9a8184de9e8bb656012b289e ]
When asked to create a path ending '/', but which is not to be a directory (LOOKUP_DIRECTORY not set), filename_create() will never try to create the file. If it doesn't exist, -ENOENT is reported.
However, it still passes LOOKUP_CREATE|LOOKUP_EXCL to the filesystems ->lookup() function, even though there is no intent to create. This is misleading and can cause incorrect behaviour.
If you try
ln -s foo /path/dir/
where 'dir' is a directory on an NFS filesystem which is not currently known in the dcache, this will fail with ENOENT.
But as the name is not in the dcache, nfs_lookup gets called with LOOKUP_CREATE|LOOKUP_EXCL and so it returns NULL without performing any lookup, with the expectation that a subsequent call to create the target will be made, and the lookup can be combined with the creation. In the case with a trailing '/' and no LOOKUP_DIRECTORY, that call is never made. Instead filename_create() sees that the dentry is not (yet) positive and returns -ENOENT - even though the directory actually exists.
So only set LOOKUP_CREATE|LOOKUP_EXCL if there really is an intent to create, and use the absence of these flags to decide if -ENOENT should be returned.
Note that filename_parentat() is only interested in LOOKUP_REVAL, so we split that out and store it in 'reval_flag'. __lookup_hash() then gets reval_flag combined with whatever create flags were determined to be needed.
--- fs/namei.c | 22 ++++++++++------------ 1 file changed, 10 insertions(+), 12 deletions(-)
fs/namei.c +10 -12
@@ -3625,18 +3625,14 @@ static struct dentry *filename_create(int dfd, struct filename *name, {struct dentry *dentry =ERR_PTR (-EEXIST );struct qstr last; +bool want_dir = lookup_flags &LOOKUP_DIRECTORY ; +unsigned int reval_flag = lookup_flags &LOOKUP_REVAL ; +unsigned int create_flags =LOOKUP_CREATE |LOOKUP_EXCL ;int type;int err2;int error; -bool is_dir = (lookup_flags &LOOKUP_DIRECTORY ); -/* -* Note that only LOOKUP_REVAL and LOOKUP_DIRECTORY matter here. Any -* other flags passed in are ignored! -*/ - lookup_flags &=LOOKUP_REVAL ; - - error =filename_parentat (dfd, name, lookup_flags, path, &last, &type); + error =filename_parentat (dfd, name, reval_flag, path, &last, &type);if (error)return ERR_PTR (error);
@@ -3650,11 +3646,13 @@ static struct dentry *filename_create(int dfd, struct filename *name,/* don't fail immediately if it's r/o, at least try to report other errors */ err2 =mnt_want_write (path->mnt);/* -* Do the final lookup. +* Do the final lookup. Suppress 'create' if there is a trailing +* '/', and a directory wasn't requested. */ - lookup_flags |=LOOKUP_CREATE |LOOKUP_EXCL ; +if (last.name[last.len] && !want_dir) + create_flags =0 ;inode_lock_nested (path->dentry->d_inode,I_MUTEX_PARENT ); - dentry =__lookup_hash (&last, path->dentry, lookup_flags); + dentry =__lookup_hash (&last, path->dentry, reval_flag | create_flags);if (IS_ERR (dentry)) goto unlock;
@@ -3668,7 +3666,7 @@ static struct dentry *filename_create(int dfd, struct filename *name,* all is fine. Let's be bastards - you had / on the end, you've * been asking for (non-existent) directory. -ENOENT for you. */ -if (unlikely (!is_dir && last.name[last.len])) { +if (unlikely (!create_flags)) { error = -ENOENT ; goto fail; }
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 23/27 stat: fix inconsistency between struct stat and struct compat_stat
In reply to Sasha LevinFrom: Mikulas Patocka <[email protected]>
[ Upstream commit 932aba1e169090357a77af18850a10c256b50819 ]
struct stat (defined in arch/x86/include/uapi/asm/stat.h) has 32-bit st_dev and st_rdev; struct compat_stat (defined in arch/x86/include/asm/compat.h) has 16-bit st_dev and st_rdev followed by a 16-bit padding.
This patch fixes struct compat_stat to match struct stat.
[ Historical note: the old x86 'struct stat' did have that 16-bit field
that the compat layer had kept around, but it was changes back in 2003 by "struct stat - support larger dev_t":
https://git.kernel.org/pub/scm/linux/kernel/git/tglx/history.git/commit/?id=e95b2065677fe32512a597a79db94b77b90c968d
and back in those days, the x86_64 port was still new, and separate from the i386 code, and had already picked up the old version with a 16-bit st_dev field ]
Note that we can't change compat_dev_t because it is used by compat_loop_info.
Also, if the st_dev and st_rdev values are 32-bit, we don't have to use old_valid_dev to test if the value fits into them. This fixes -EOVERFLOW on filesystems that are on NVMe because NVMe uses the major number 259.
--- arch/x86/include/asm/compat.h | 6 ++---- fs/stat.c | 19 ++++++++++--------- 2 files changed, 12 insertions(+), 13 deletions(-)
arch/x86/include/asm/compat.h +2 -4
@@ -28,15 +28,13 @@ typedef u16 compat_ipc_pid_t;typedef __kernel_fsid_t compat_fsid_t ;struct compat_stat { -compat_dev_t st_dev; -u16 __pad1; +u32 st_dev;compat_ino_t st_ino;compat_mode_t st_mode;compat_nlink_t st_nlink;__compat_uid_t st_uid;__compat_gid_t st_gid; -compat_dev_t st_rdev; -u16 __pad2; +u32 st_rdev;u32 st_size;u32 st_blksize;u32 st_blocks;
fs/stat.c +10 -9
@@ -334,9 +334,6 @@ SYSCALL_DEFINE2(fstat, unsigned int, fd, struct __old_kernel_stat __user *, stat# define choose_32_64 (a,b) b#endif -#define valid_dev (x) choose_32_64(old_valid_dev(x),true) -#define encode_dev (x) choose_32_64(old_encode_dev,new_encode_dev)(x) -#ifndef INIT_STRUCT_STAT_PADDING # define INIT_STRUCT_STAT_PADDING (st) memset(&st, 0, sizeof(st))#endif
@@ -345,7 +342,9 @@ static int cp_new_stat(struct kstat *stat, struct stat __user *statbuf) {struct stat tmp; -if (!valid_dev (stat->dev) || !valid_dev (stat->rdev)) +if (sizeof (tmp.st_dev) <4 && !old_valid_dev (stat->dev)) +return -EOVERFLOW ; +if (sizeof (tmp.st_rdev) <4 && !old_valid_dev (stat->rdev))return -EOVERFLOW ;#if BITS_PER_LONG ==32 if (stat->size >MAX_NON_LFS )
@@ -353,7 +352,7 @@ static int cp_new_stat(struct kstat *stat, struct stat __user *statbuf)#endif INIT_STRUCT_STAT_PADDING (tmp); - tmp.st_dev =encode_dev (stat->dev); + tmp.st_dev =new_encode_dev (stat->dev); tmp.st_ino = stat->ino;if (sizeof (tmp.st_ino) <sizeof (stat->ino) && tmp.st_ino != stat->ino)return -EOVERFLOW ;
@@ -363,7 +362,7 @@ static int cp_new_stat(struct kstat *stat, struct stat __user *statbuf)return -EOVERFLOW ;SET_UID (tmp.st_uid,from_kuid_munged (current_user_ns (), stat->uid));SET_GID (tmp.st_gid,from_kgid_munged (current_user_ns (), stat->gid)); - tmp.st_rdev =encode_dev (stat->rdev); + tmp.st_rdev =new_encode_dev (stat->rdev); tmp.st_size = stat->size; tmp.st_atime = stat->atime.tv_sec; tmp.st_mtime = stat->mtime.tv_sec;
@@ -644,11 +643,13 @@ static int cp_compat_stat(struct kstat *stat, struct compat_stat __user *ubuf) {struct compat_stat tmp; -if (!old_valid_dev (stat->dev) || !old_valid_dev (stat->rdev)) +if (sizeof (tmp.st_dev) <4 && !old_valid_dev (stat->dev)) +return -EOVERFLOW ; +if (sizeof (tmp.st_rdev) <4 && !old_valid_dev (stat->rdev))return -EOVERFLOW ;memset (&tmp,0 ,sizeof (tmp)); - tmp.st_dev =old_encode_dev (stat->dev); + tmp.st_dev =new_encode_dev (stat->dev); tmp.st_ino = stat->ino;if (sizeof (tmp.st_ino) <sizeof (stat->ino) && tmp.st_ino != stat->ino)return -EOVERFLOW ;
@@ -658,7 +659,7 @@ static int cp_compat_stat(struct kstat *stat, struct compat_stat __user *ubuf)return -EOVERFLOW ;SET_UID (tmp.st_uid,from_kuid_munged (current_user_ns (), stat->uid));SET_GID (tmp.st_gid,from_kgid_munged (current_user_ns (), stat->gid)); - tmp.st_rdev =old_encode_dev (stat->rdev); + tmp.st_rdev =new_encode_dev (stat->rdev);if ((u64 ) stat->size >MAX_NON_LFS )return -EOVERFLOW ; tmp.st_size = stat->size;
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 25/27 nvme: add a quirk to disable namespace identifiers
In reply to Sasha LevinFrom: Christoph Hellwig <[email protected]>
[ Upstream commit 00ff400e6deee00f7b15e200205b2708b63b8cf6 ]
Add a quirk to disable using and exporting namespace identifiers for controllers where they are broken beyond repair.
The most directly visible problem with non-unique namespace identifiers is that they break the /dev/disk/by-id/ links, with the link for a supposedly unique identifier now pointing to one of multiple possible namespaces that share the same ID, and a somewhat random selection of which one actually shows up.
--- drivers/nvme/host/core.c | 24 ++++++++++++++++++------ drivers/nvme/host/nvme.h | 5 +++++ 2 files changed, 23 insertions(+), 6 deletions(-)
drivers/nvme/host/core.c +18 -6
@@ -1354,6 +1354,8 @@ static int nvme_process_ns_desc(struct nvme_ctrl *ctrl, struct nvme_ns_ids *ids, warn_str, cur->nidl);return -1 ; } +if (ctrl->quirks &NVME_QUIRK_BOGUS_NID ) +return NVME_NIDT_EUI64_LEN ;memcpy (ids->eui64, data +sizeof (*cur),NVME_NIDT_EUI64_LEN );return NVME_NIDT_EUI64_LEN ;case NVME_NIDT_NGUID :
@@ -1362,6 +1364,8 @@ static int nvme_process_ns_desc(struct nvme_ctrl *ctrl, struct nvme_ns_ids *ids, warn_str, cur->nidl);return -1 ; } +if (ctrl->quirks &NVME_QUIRK_BOGUS_NID ) +return NVME_NIDT_NGUID_LEN ;memcpy (ids->nguid, data +sizeof (*cur),NVME_NIDT_NGUID_LEN );return NVME_NIDT_NGUID_LEN ;case NVME_NIDT_UUID :
@@ -1370,6 +1374,8 @@ static int nvme_process_ns_desc(struct nvme_ctrl *ctrl, struct nvme_ns_ids *ids, warn_str, cur->nidl);return -1 ; } +if (ctrl->quirks &NVME_QUIRK_BOGUS_NID ) +return NVME_NIDT_UUID_LEN ;uuid_copy (&ids->uuid, data +sizeof (*cur));return NVME_NIDT_UUID_LEN ;case NVME_NIDT_CSI :
@@ -1466,12 +1472,18 @@ static int nvme_identify_ns(struct nvme_ctrl *ctrl, unsigned nsid,if ((*id)->ncap ==0 )/* namespace not allocated or attached */ goto out_free_id; -if (ctrl->vs >=NVME_VS (1 ,1 ,0 ) && - !memchr_inv (ids->eui64,0 ,sizeof (ids->eui64))) -memcpy (ids->eui64, (*id)->eui64,sizeof (ids->eui64)); -if (ctrl->vs >=NVME_VS (1 ,2 ,0 ) && - !memchr_inv (ids->nguid,0 ,sizeof (ids->nguid))) -memcpy (ids->nguid, (*id)->nguid,sizeof (ids->nguid)); + +if (ctrl->quirks &NVME_QUIRK_BOGUS_NID ) { +dev_info (ctrl->device, +"Ignoring bogus Namespace Identifiers\n" ); + }else { +if (ctrl->vs >=NVME_VS (1 ,1 ,0 ) && + !memchr_inv (ids->eui64,0 ,sizeof (ids->eui64))) +memcpy (ids->eui64, (*id)->eui64,sizeof (ids->eui64)); +if (ctrl->vs >=NVME_VS (1 ,2 ,0 ) && + !memchr_inv (ids->nguid,0 ,sizeof (ids->nguid))) +memcpy (ids->nguid, (*id)->nguid,sizeof (ids->nguid)); + }return 0 ;
drivers/nvme/host/nvme.h +5 -0
@@ -144,6 +144,11 @@ enum nvme_quirks {* encoding the generation sequence number. */ NVME_QUIRK_SKIP_CID_GEN = (1 <<17 ), + +/* +* Reports garbage in the namespace identifiers (eui64, nguid, uuid). +*/ + NVME_QUIRK_BOGUS_NID = (1 << 18), };/*
Signature · 1 line
2.35.1
Sasha Levin
maintainerPATCH 27/27 nvme-pci: disable namespace identifiers for Qemu controllers
In reply to Sasha LevinFrom: Christoph Hellwig <[email protected]>
[ Upstream commit 66dd346b84d79fde20832ed691a54f4881eac20d ]
Qemu unconditionally reports a UUID, which depending on the qemu version is either all-null (which is incorrect but harmless) or contains a single bit set for all controllers. In addition it can also optionally report a eui64 which needs to be manually set. Disable namespace identifiers for Qemu controlles entirely even if in some cases they could be set correctly through manual intervention.
--- drivers/nvme/host/pci.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
drivers/nvme/host/pci.c +4 -1
@@ -3314,7 +3314,10 @@ static const struct pci_device_id nvme_id_table[] = { .driver_data = NVME_QUIRK_IGNORE_DEV_SUBNQN, }, { PCI_VDEVICE(INTEL, 0x5845), /* Qemu emulated controller */ .driver_data =NVME_QUIRK_IDENTIFY_CNS | - NVME_QUIRK_DISABLE_WRITE_ZEROES, }, +NVME_QUIRK_DISABLE_WRITE_ZEROES | + NVME_QUIRK_BOGUS_NID, }, + { PCI_VDEVICE(REDHAT, 0x0010), /* Qemu emulated controller */ + .driver_data = NVME_QUIRK_BOGUS_NID, }, { PCI_DEVICE(0x126f, 0x2263), /* Silicon Motion unidentified */ .driver_data = NVME_QUIRK_NO_NS_DESC_LIST, }, { PCI_DEVICE(0x1bb1, 0x0100), /* Seagate Nytro Flash Storage */
Signature · 1 line
2.35.1