6.15.4-rc1 review
Subsystems (5)
Greg Kroah-Hartman
maintainerPATCH 419/592 wifi: ath12k: Fix memory leak due to multiple rx_stats allocation
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sidhanta Sahu <[email protected]>
[ Upstream commit c426497fa2055c8005196922e7d29c41d7e0948a ]
rx_stats for each arsta is allocated when adding a station. arsta->rx_stats will be freed when a station is removed.
Redundant allocations are occurring when the same station is added multiple times. This causes ath12k_mac_station_add() to be called multiple times, and rx_stats is allocated each time. As a result there is memory leaks.
Prevent multiple allocations of rx_stats when ath12k_mac_station_add() is called repeatedly by checking if rx_stats is already allocated before allocating again. Allocate arsta->rx_stats if arsta->rx_stats is NULL respectively.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.3.1-00173-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
--- drivers/net/wireless/ath/ath12k/mac.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/net/wireless/ath/ath12k/mac.c +6 -3
@@ -5560,10 +5560,13 @@ static int ath12k_mac_station_add(struct ath12k *ar, ar->max_num_stations); goto exit; } - arsta->rx_stats =kzalloc (sizeof (*arsta->rx_stats),GFP_KERNEL ); +if (!arsta->rx_stats) { - ret = -ENOMEM ; - goto dec_num_station; + arsta->rx_stats =kzalloc (sizeof (*arsta->rx_stats),GFP_KERNEL ); +if (!arsta->rx_stats) { + ret = -ENOMEM ; + goto dec_num_station; + } } peer_param.vdev_id = arvif->vdev_id;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 420/592 wifi: ath11k: determine PM policy based on machine model
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Baochen Qiang <[email protected]>
[ Upstream commit ce8669a27016354dfa8bf3c954255cb9f3583bae ]
To handle the Lenovo unexpected wakeup issue [1], previously we revert commit 166a490f59ac ("wifi: ath11k: support hibernation"). So currently WLAN target is put into WoWLAN mode during suspend. This is a temporary solution as it does not work on machines where WLAN power is cut off.
The thought here is that we do WoWLAN suspend on Lenovo machines while do non-WoWLAN suspend (which is done in the reverted commit) on other machines. This requires us to identify Lenovo machines from others. For that purpose, read board vendor and product name from DMI interface, match it against all known affected machines. If there is a match, choose WoWLAN suspend mode, else choose non-WoWLAN mode. Save the mode in ab for later reference.
[1] https://bugzilla.kernel.org/show_bug.cgi?id=219196
Tested-on: WCN6855 hw2.0 PCI WLAN.HSP.1.1-03125-QCAHSPSWPL_V1_V2_SILICONZ_LITE-3.6510.30
--- drivers/net/wireless/ath/ath11k/core.c | 55 ++++++++++++++++++++++++++ drivers/net/wireless/ath/ath11k/core.h | 7 ++++ 2 files changed, 62 insertions(+)
drivers/net/wireless/ath/ath11k/core.c +55 -0
@@ -907,6 +907,52 @@ static const struct ath11k_hw_params ath11k_hw_params[] = { }, }; +static const struct dmi_system_id ath11k_pm_quirk_table[] = { + { + .driver_data = (void *)ATH11K_PM_WOW , + .matches = { +DMI_MATCH (DMI_BOARD_VENDOR ,"LENOVO" ), +DMI_MATCH (DMI_PRODUCT_NAME ,"21J4" ), + }, + }, + { + .driver_data = (void *)ATH11K_PM_WOW , + .matches = { +DMI_MATCH (DMI_BOARD_VENDOR ,"LENOVO" ), +DMI_MATCH (DMI_PRODUCT_NAME ,"21K4" ), + }, + }, + { + .driver_data = (void *)ATH11K_PM_WOW , + .matches = { +DMI_MATCH (DMI_BOARD_VENDOR ,"LENOVO" ), +DMI_MATCH (DMI_PRODUCT_NAME ,"21K6" ), + }, + }, + { + .driver_data = (void *)ATH11K_PM_WOW , + .matches = { +DMI_MATCH (DMI_BOARD_VENDOR ,"LENOVO" ), +DMI_MATCH (DMI_PRODUCT_NAME ,"21K8" ), + }, + }, + { + .driver_data = (void *)ATH11K_PM_WOW , + .matches = { +DMI_MATCH (DMI_BOARD_VENDOR ,"LENOVO" ), +DMI_MATCH (DMI_PRODUCT_NAME ,"21KA" ), + }, + }, + { + .driver_data = (void *)ATH11K_PM_WOW , + .matches = { +DMI_MATCH (DMI_BOARD_VENDOR ,"LENOVO" ), +DMI_MATCH (DMI_PRODUCT_NAME ,"21F9" ), + }, + }, + {} +}; +static inline struct ath11k_pdev *ath11k_core_get_single_pdev (struct ath11k_base *ab) {WARN_ON (!ab->hw_params.single_pdev_only);
@@ -2334,8 +2380,17 @@ EXPORT_SYMBOL(ath11k_core_pre_init);int ath11k_core_init (struct ath11k_base *ab) { +const struct dmi_system_id *dmi_id;int ret; + dmi_id =dmi_first_match (ath11k_pm_quirk_table); +if (dmi_id) + ab->pm_policy = (kernel_ulong_t )dmi_id->driver_data; +else + ab->pm_policy =ATH11K_PM_DEFAULT ; + +ath11k_dbg (ab,ATH11K_DBG_BOOT ,"pm policy %u\n" , ab->pm_policy); + ret =ath11k_core_soc_create (ab);if (ret) {ath11k_err (ab,"failed to create soc core: %d\n" , ret);
drivers/net/wireless/ath/ath11k/core.h +7 -0
@@ -894,6 +894,11 @@ struct ath11k_msi_config {u16 hw_rev; }; +enum ath11k_pm_policy { +ATH11K_PM_DEFAULT , +ATH11K_PM_WOW , +}; +/* Master structure to hold the hw data which may be used in core module */ struct ath11k_base {enum ath11k_hw_rev hw_rev;
@@ -1060,6 +1065,8 @@ struct ath11k_base { } testmode;#endif +enum ath11k_pm_policy pm_policy; +/* must be last */ u8 drv_priv[]__aligned(sizeof(void *)) ; };
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 421/592 wifi: ath12k: fix link valid field initialization in the monitor Rx
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hari Chandrakanthan <[email protected]>
[ Upstream commit 2826139f9295821fe2b049318a1cc057ec003131 ]
Currently, the link_valid field is not initialized in the monitor Rx path. This can result in random values for the link_valid and link_id leads to undefined behaviour in mac80211. Therefore, initialize the link_valid field in the monitor Rx path.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.3.1-00173-QCAHKSWPL_SILICONZ-1
Tested-on: WCN7850 hw2.0 PCI WLAN.HMT.1.0.c5-00481-QCAHMTSWPL_V1.0_V2.0_SILICONZ-3
--- drivers/net/wireless/ath/ath12k/dp_mon.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/net/wireless/ath/ath12k/dp_mon.c +2 -0
@@ -2097,6 +2097,8 @@ static void ath12k_dp_mon_rx_deliver_msdu(struct ath12k *ar, struct napi_structbool is_mcbc = rxcb->is_mcbc;bool is_eapol_tkip = rxcb->is_eapol; + status->link_valid =0 ; +if ((status->encoding ==RX_ENC_HE ) && !(status->flag &RX_FLAG_RADIOTAP_HE ) && !(status->flag &RX_FLAG_SKIP_MONITOR )) { he =skb_push (msdu,sizeof (known));
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 422/592 wifi: ath12k: fix incorrect CE addresses
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Balamurugan S <[email protected]>
[ Upstream commit 60031d9c3589c7983fd1deb4a4c0bebf0929890e ]
In the current ath12k implementation, the CE addresses CE_HOST_IE_ADDRESS and CE_HOST_IE_2_ADDRESS are incorrect. These values were inherited from ath11k, but ath12k does not currently use them.
However, the Ath12k AHB support relies on these addresses. Therefore, correct the CE addresses for ath12k.
Tested-on: IPQ5332 hw1.0 AHB WLAN.WBE.1.3.1-00130-QCAHKSWPL_SILICONZ-1
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.1.1-00210-QCAHKSWPL_SILICONZ-1
--- drivers/net/wireless/ath/ath12k/ce.h | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/net/wireless/ath/ath12k/ce.h +3 -3
@@ -1,7 +1,7 @@/* SPDX-License-Identifier: BSD-3-Clause-Clear */ /* * Copyright (c) 2018-2021 The Linux Foundation. All rights reserved. -* Copyright (c) 2021-2022, 2024 Qualcomm Innovation Center, Inc. All rights reserved. +* Copyright (c) 2021-2022, 2024-2025 Qualcomm Innovation Center, Inc. All rights reserved. */ #ifndef ATH12K_CE_H
@@ -39,8 +39,8 @@#define PIPEDIR_INOUT_H2H 4/* bidirectional, host to host */ /* CE address/mask */ -#define CE_HOST_IE_ADDRESS 0x00A1803C -#define CE_HOST_IE_2_ADDRESS 0x00A18040 +#define CE_HOST_IE_ADDRESS 0x75804C +#define CE_HOST_IE_2_ADDRESS 0x758050#define CE_HOST_IE_3_ADDRESS CE_HOST_IE_ADDRESS#define CE_HOST_IE_3_SHIFT 0xC
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 423/592 wifi: ath12k: Pass correct values of center freq1 and center freq2 for 160 MHz
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Suraj P Kizhakkethil <[email protected]>
[ Upstream commit b1b01e46a3db5ad44d1e4691ba37c1e0832cd5cf ]
Currently, for 160 MHz bandwidth, center frequency1 and center frequency2 are not passed correctly to the firmware. Set center frequency1 as the center frequency
of the primary 80 MHz channel segment and center frequency2 as the center frequency of the 160 MHz channel and pass the values to the firmware.
Tested-on: QCN9274 hw2.0 PCI WLAN.WBE.1.3.1-00173-QCAHKSWPL_SILICONZ-1
--- drivers/net/wireless/ath/ath12k/wmi.c | 16 +++++++++++++--- 1 file changed, 13 insertions(+), 3 deletions(-)
drivers/net/wireless/ath/ath12k/wmi.c +13 -3
@@ -1037,14 +1037,24 @@ int ath12k_wmi_vdev_down(struct ath12k *ar, u8 vdev_id)static void ath12k_wmi_put_wmi_channel (struct ath12k_wmi_channel_params *chan,struct wmi_vdev_start_req_arg *arg) { +u32 center_freq1 = arg->band_center_freq1; +memset (chan,0 ,sizeof (*chan)); chan->mhz =cpu_to_le32 (arg->freq); - chan->band_center_freq1 =cpu_to_le32 (arg->band_center_freq1); -if (arg->mode ==MODE_11AC_VHT80_80 ) + chan->band_center_freq1 =cpu_to_le32 (center_freq1); +if (arg->mode ==MODE_11BE_EHT160 ) { +if (arg->freq > center_freq1) + chan->band_center_freq1 =cpu_to_le32 (center_freq1 +40 ); +else + chan->band_center_freq1 =cpu_to_le32 (center_freq1 -40 ); + + chan->band_center_freq2 =cpu_to_le32 (center_freq1); + }else if (arg->mode ==MODE_11BE_EHT80_80 ) { chan->band_center_freq2 =cpu_to_le32 (arg->band_center_freq2); -else + }else { chan->band_center_freq2 =0 ; + } chan->info |=le32_encode_bits (arg->mode,WMI_CHAN_INFO_MODE );if (arg->passive)
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 424/592 net/mlx5: HWS, Harden IP version definer checks
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vlad Dogaru <[email protected]>
[ Upstream commit 6991a975e416154576b0f5f06256aec13e23b0a7 ]
Replicate some sanity checks that firmware does, since hardware steering does not go through firmware.
When creating a definer, disallow matching on IP addresses without also matching on IP version. The latter can be satisfied by matching either on the version field in the IP header, or on the ethertype field.
Also refuse to match IPv4 IHL alongside IPv6.
--- .../mellanox/mlx5/core/steering/hws/definer.c | 44 ++++++++++++++++++- 1 file changed, 42 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlx5/core/steering/hws/definer.c +42 -2
@@ -509,9 +509,9 @@ static int hws_definer_conv_outer(struct mlx5hws_definer_conv_data *cd, u32 *match_param) { + bool is_ipv6, smac_set, dmac_set, ip_addr_set, ip_ver_set;struct mlx5hws_definer_fc *fc = cd->fc;struct mlx5hws_definer_fc *curr_fc; -bool is_ipv6, smac_set, dmac_set;u32 *s_ipv6, *d_ipv6; if (HWS_IS_FLD_SET_SZ(match_param, outer_headers.l4_type, 0x2) ||
@@ -521,6 +521,20 @@ hws_definer_conv_outer(struct mlx5hws_definer_conv_data *cd,return -EINVAL ; } + ip_addr_set =HWS_IS_FLD_SET_SZ (match_param, + outer_headers.src_ipv4_src_ipv6, +0x80 ) || +HWS_IS_FLD_SET_SZ (match_param, + outer_headers.dst_ipv4_dst_ipv6,0x80 ); + ip_ver_set =HWS_IS_FLD_SET (match_param, outer_headers.ip_version) || +HWS_IS_FLD_SET (match_param, outer_headers.ethertype); + +if (ip_addr_set && !ip_ver_set) { +mlx5hws_err (cd->ctx, +"Unsupported match on IP address without version or ethertype\n" ); +return -EINVAL ; + } +/* L2 Check ethertype */ HWS_SET_HDR(fc, match_param, ETH_TYPE_O, outer_headers.ethertype,
@@ -576,6 +590,12 @@ hws_definer_conv_outer(struct mlx5hws_definer_conv_data *cd, is_ipv6 = s_ipv6[0 ] || s_ipv6[1 ] || s_ipv6[2 ] || d_ipv6[0 ] || d_ipv6[1 ] || d_ipv6[2 ]; +/* IHL is an IPv4-specific field. */ +if (is_ipv6 &&HWS_IS_FLD_SET (match_param, outer_headers.ipv4_ihl)) { +mlx5hws_err (cd->ctx,"Unsupported match on IPv6 address and IPv4 IHL\n" ); +return -EINVAL ; + } +if (is_ipv6) {/* Handle IPv6 source address */ HWS_SET_HDR(fc, match_param, IPV6_SRC_127_96_O,
@@ -665,9 +685,9 @@ static int hws_definer_conv_inner(struct mlx5hws_definer_conv_data *cd, u32 *match_param) { + bool is_ipv6, smac_set, dmac_set, ip_addr_set, ip_ver_set;struct mlx5hws_definer_fc *fc = cd->fc;struct mlx5hws_definer_fc *curr_fc; -bool is_ipv6, smac_set, dmac_set;u32 *s_ipv6, *d_ipv6; if (HWS_IS_FLD_SET_SZ(match_param, inner_headers.l4_type, 0x2) ||
@@ -677,6 +697,20 @@ hws_definer_conv_inner(struct mlx5hws_definer_conv_data *cd,return -EINVAL ; } + ip_addr_set =HWS_IS_FLD_SET_SZ (match_param, + inner_headers.src_ipv4_src_ipv6, +0x80 ) || +HWS_IS_FLD_SET_SZ (match_param, + inner_headers.dst_ipv4_dst_ipv6,0x80 ); + ip_ver_set =HWS_IS_FLD_SET (match_param, inner_headers.ip_version) || +HWS_IS_FLD_SET (match_param, inner_headers.ethertype); + +if (ip_addr_set && !ip_ver_set) { +mlx5hws_err (cd->ctx, +"Unsupported match on IP address without version or ethertype\n" ); +return -EINVAL ; + } +/* L2 Check ethertype */ HWS_SET_HDR(fc, match_param, ETH_TYPE_I, inner_headers.ethertype,
@@ -731,6 +765,12 @@ hws_definer_conv_inner(struct mlx5hws_definer_conv_data *cd, is_ipv6 = s_ipv6[0 ] || s_ipv6[1 ] || s_ipv6[2 ] || d_ipv6[0 ] || d_ipv6[1 ] || d_ipv6[2 ]; +/* IHL is an IPv4-specific field. */ +if (is_ipv6 &&HWS_IS_FLD_SET (match_param, inner_headers.ipv4_ihl)) { +mlx5hws_err (cd->ctx,"Unsupported match on IPv6 address and IPv4 IHL\n" ); +return -EINVAL ; + } +if (is_ipv6) {/* Handle IPv6 source address */ HWS_SET_HDR(fc, match_param, IPV6_SRC_127_96_I,
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 425/592 fbcon: Make sure modelist not set on unregistered console
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kees Cook <[email protected]>
[ Upstream commit cedc1b63394a866bf8663a3e40f4546f1d28c8d8 ]
It looks like attempting to write to the "store_modes" sysfs node will run afoul of unregistered consoles:
UBSAN: array-index-out-of-bounds in drivers/video/fbdev/core/fbcon.c:122:28 index -1 is out of range for type 'fb_info *[32]' ... fbcon_info_from_console+0x192/0x1a0 drivers/video/fbdev/core/fbcon.c:122 fbcon_new_modelist+0xbf/0x2d0 drivers/video/fbdev/core/fbcon.c:3048 fb_new_modelist+0x328/0x440 drivers/video/fbdev/core/fbmem.c:673 store_modes+0x1c9/0x3e0 drivers/video/fbdev/core/fbsysfs.c:113 dev_attr_store+0x55/0x80 drivers/base/core.c:2439
static struct fb_info *fbcon_registered_fb[FB_MAX];
...
static signed char con2fb_map[MAX_NR_CONSOLES];
...
static struct fb_info *fbcon_info_from_console(int console)
...
return fbcon_registered_fb[con2fb_map[console]];
If con2fb_map contains a -1 things go wrong here. Instead, return NULL, as callers of fbcon_info_from_console() are trying to compare against existing "info" pointers, so error handling should kick in correctly.
--- drivers/video/fbdev/core/fbcon.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/video/fbdev/core/fbcon.c +6 -1
@@ -117,9 +117,14 @@ static signed char con2fb_map_boot[MAX_NR_CONSOLES];static struct fb_info *fbcon_info_from_console (int console) { +signed char fb;WARN_CONSOLE_UNLOCKED (); -return fbcon_registered_fb[con2fb_map[console]]; + fb = con2fb_map[console]; +if (fb <0 || fb >=ARRAY_SIZE (fbcon_registered_fb)) +return NULL ; + +return fbcon_registered_fb[fb]; }static int logo_lines;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 426/592 wifi: iwlwifi: mld: Work around Clang loop unrolling bug
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kees Cook <[email protected]>
[ Upstream commit 368556dd234dc4a506a35a0c99c0eee2ab475c77 ]
The nested loop in iwl_mld_send_proto_offload() confuses Clang into thinking there could be a final loop iteration past the end of the "nsc" array (which is only 4 entries). The FORTIFY checking in memcmp() (via ipv6_addr_cmp()) notices this (due to the available bytes in the out-of-bounds position of &nsc[4] being 0), and errors out, failing the build. For some reason (likely due to architectural loop unrolling configurations), this is only exposed on ARM builds currently. Due to Clang's lack of inline tracking[1], the warning is not very helpful:
include/linux/fortify-string.h:719:4: error: call to '__read_overflow' declared with 'error' attribute: detected read beyond size of object (1st parameter)
719 | __read_overflow();
| ^1 error generated.
But this was tracked down to iwl_mld_send_proto_offload()'s ipv6_addr_cmp() call.
An upstream Clang bug has been filed[2] to track this. For now fix the build by explicitly bounding the inner loop by "n_nsc", which is what "c" is already limited to.
--- drivers/net/wireless/intel/iwlwifi/mld/d3.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/net/wireless/intel/iwlwifi/mld/d3.c +1 -1
@@ -1757,7 +1757,7 @@ iwl_mld_send_proto_offload(struct iwl_mld *mld,addrconf_addr_solict_mult (&wowlan_data->target_ipv6_addrs[i], &solicited_addr); -for (j =0 ; j < c; j++) +for (j =0 ; j < n_nsc && j < c; j++)if (ipv6_addr_cmp (&nsc[j].dest_ipv6_addr, &solicited_addr) ==0 )break ;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 427/592 watchdog: da9052_wdt: respect TWDMIN
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marcus Folkesson <[email protected]>
[ Upstream commit 325f510fcd9cda5a44bcb662b74ba4e3dabaca10 ]
We have to wait at least the minimium time for the watchdog window (TWDMIN) before writings to the wdt register after the watchdog is activated.
Otherwise the chip will assert TWD_ERROR and power down to reset mode.
--- drivers/watchdog/da9052_wdt.c | 1 + 1 file changed, 1 insertion(+)
drivers/watchdog/da9052_wdt.c +1 -0
@@ -168,6 +168,7 @@ static int da9052_wdt_probe(struct platform_device *pdev) da9052_wdt = &driver_data->wdt; da9052_wdt->timeout =DA9052_DEF_TIMEOUT ; + da9052_wdt->min_hw_heartbeat_ms =DA9052_TWDMIN ; da9052_wdt->info = &da9052_wdt_info; da9052_wdt->ops = &da9052_wdt_ops; da9052_wdt->parent = dev;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 428/592 watchdog: stm32: Fix wakeup source leaks on device unbind
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Krzysztof Kozlowski <[email protected]>
[ Upstream commit b6f8a417e17f1929bb8e7e6ba9f4677f1f3ce364 ]
Device can be unbound or probe can fail, so driver must also release memory for the wakeup source.
--- drivers/watchdog/stm32_iwdg.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/watchdog/stm32_iwdg.c +1 -1
@@ -291,7 +291,7 @@ static int stm32_iwdg_irq_init(struct platform_device *pdev,return 0 ;if (of_property_read_bool (np,"wakeup-source" )) { - ret =device_init_wakeup (dev, true); + ret =devm_device_init_wakeup (dev);if (ret)return ret;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 429/592 i3c: mipi-i3c-hci: Fix handling status of i3c_hci_irq_handler()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jarkko Nikula <[email protected]>
[ Upstream commit 279c24021b838e76ca8441e9446e0ab45271153a ]
Return IRQ_HANDLED from the i3c_hci_irq_handler() only if some INTR_STATUS bit was set or if DMA/PIO handler handled it.
Currently it returns IRQ_HANDLED in case INTR_STATUS is zero and IO handler returns false. Which could be the case if interrupt comes from other device or is spurious.
--- drivers/i3c/master/mipi-i3c-hci/core.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
drivers/i3c/master/mipi-i3c-hci/core.c +3 -3
@@ -594,6 +594,7 @@ static irqreturn_t i3c_hci_irq_handler(int irq, void *dev_id)if (val) {reg_write (INTR_STATUS , val); + result =IRQ_HANDLED ; }if (val &INTR_HC_RESET_CANCEL ) {
@@ -605,12 +606,11 @@ static irqreturn_t i3c_hci_irq_handler(int irq, void *dev_id) val &= ~INTR_HC_INTERNAL_ERR ; } - hci->io->irq_handler (hci); +if (hci->io->irq_handler (hci)) + result =IRQ_HANDLED ;if (val)dev_err (&hci->master.dev,"unexpected INTR_STATUS %#x\n" , val); -else - result =IRQ_HANDLED ;return result; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 430/592 bus: fsl-mc: increase MC_CMD_COMPLETION_TIMEOUT_MS value
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Laurentiu Tudor <[email protected]>
[ Upstream commit 23d060136841c58c2f9ee8c08ad945d1879ead4b ]
In case the MC firmware runs in debug mode with extensive prints pushed to the console, the current timeout of 500ms is not enough. Increase the timeout value so that we don't have any chance of wrongly assuming that the firmware is not responding when it's just taking more time.
--- drivers/bus/fsl-mc/mc-sys.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/bus/fsl-mc/mc-sys.c +1 -1
@@ -19,7 +19,7 @@/* * Timeout in milliseconds to wait for the completion of an MC command */ -#define MC_CMD_COMPLETION_TIMEOUT_MS 500 +#define MC_CMD_COMPLETION_TIMEOUT_MS 15000/* * usleep_range() min and max values used to throttle down polling
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 431/592 ARM: OMAP2+: Fix l4ls clk domain handling in STANDBY
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Sukrut Bellary <[email protected]>
[ Upstream commit 47fe74098f3dadba2f9cc1e507d813a4aa93f5f3 ]
Don't put the l4ls clk domain to sleep in case of standby. Since CM3 PM FW[1](ti-v4.1.y) doesn't wake-up/enable the l4ls clk domain upon wake-up, CM3 PM FW fails to wake-up the MPU.
[1] https://git.ti.com/cgit/processor-firmware/ti-amx3-cm3-pm-firmware/
--- arch/arm/mach-omap2/clockdomain.h | 1 + arch/arm/mach-omap2/clockdomains33xx_data.c | 2 +- arch/arm/mach-omap2/cm33xx.c | 14 +++++++++++++- 3 files changed, 15 insertions(+), 2 deletions(-)
arch/arm/mach-omap2/clockdomain.h +1 -0
@@ -48,6 +48,7 @@#define CLKDM_NO_AUTODEPS (1 << 4)#define CLKDM_ACTIVE_WITH_MPU (1 << 5)#define CLKDM_MISSING_IDLE_REPORTING (1 << 6) +#define CLKDM_STANDBY_FORCE_WAKEUP BIT(7)#define CLKDM_CAN_HWSUP (CLKDM_CAN_ENABLE_AUTO | CLKDM_CAN_DISABLE_AUTO)#define CLKDM_CAN_SWSUP (CLKDM_CAN_FORCE_SLEEP | CLKDM_CAN_FORCE_WAKEUP)
arch/arm/mach-omap2/clockdomains33xx_data.c +1 -1
@@ -19,7 +19,7 @@ static struct clockdomain l4ls_am33xx_clkdm = { .pwrdm = { .name = "per_pwrdm" }, .cm_inst = AM33XX_CM_PER_MOD, .clkdm_offs = AM33XX_CM_PER_L4LS_CLKSTCTRL_OFFSET, - .flags = CLKDM_CAN_SWSUP, + .flags = CLKDM_CAN_SWSUP | CLKDM_STANDBY_FORCE_WAKEUP, };static struct clockdomain l3s_am33xx_clkdm = {
arch/arm/mach-omap2/cm33xx.c +13 -1
@@ -20,6 +20,9 @@#include "cm-regbits-34xx.h" #include "cm-regbits-33xx.h" #include "prm33xx.h" +#if IS_ENABLED (CONFIG_SUSPEND ) +#include <linux/suspend.h> +#endif /* * CLKCTRL_IDLEST_*: possible values for the CM_*_CLKCTRL.IDLEST bitfield:
@@ -328,8 +331,17 @@ static int am33xx_clkdm_clk_disable(struct clockdomain *clkdm) {bool hwsup = false; +#if IS_ENABLED (CONFIG_SUSPEND ) +/* +* In case of standby, Don't put the l4ls clk domain to sleep. +* Since CM3 PM FW doesn't wake-up/enable the l4ls clk domain +* upon wake-up, CM3 PM FW fails to wake-up th MPU. +*/ +if (pm_suspend_target_state ==PM_SUSPEND_STANDBY && + (clkdm->flags &CLKDM_STANDBY_FORCE_WAKEUP )) +return 0 ; +#endif hwsup =am33xx_cm_is_clkdm_in_hwsup (clkdm->cm_inst, clkdm->clkdm_offs); -if (!hwsup && (clkdm->flags &CLKDM_CAN_FORCE_SLEEP ))am33xx_clkdm_sleep (clkdm);
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 432/592 tee: Prevent size calculation wraparound on 32-bit kernels
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jann Horn <[email protected]>
[ Upstream commit 39bb67edcc582b3b386a9ec983da67fa8a10ec03 ]
The current code around TEE_IOCTL_PARAM_SIZE() is a bit wrong on 32-bit kernels: Multiplying a user-provided 32-bit value with the size of a structure can wrap around on such platforms.
Fix it by using saturating arithmetic for the size calculation.
This has no security consequences because, in all users of TEE_IOCTL_PARAM_SIZE(), the subsequent kcalloc() implicitly checks for wrapping.
--- drivers/tee/tee_core.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
drivers/tee/tee_core.c +6 -5
@@ -10,6 +10,7 @@#include <linux/fs.h> #include <linux/idr.h> #include <linux/module.h> +#include <linux/overflow.h> #include <linux/slab.h> #include <linux/tee_core.h> #include <linux/uaccess.h>
@@ -19,7 +20,7 @@#define TEE_NUM_DEVICES 32 -#define TEE_IOCTL_PARAM_SIZE (x) (sizeof(struct tee_param) * (x)) +#define TEE_IOCTL_PARAM_SIZE (x) (size_mul(sizeof(struct tee_param), (x)))#define TEE_UUID_NS_NAME_SIZE 128
@@ -487,7 +488,7 @@ static int tee_ioctl_open_session(struct tee_context *ctx,if (copy_from_user (&arg, uarg,sizeof (arg)))return -EFAULT ; -if (sizeof (arg) +TEE_IOCTL_PARAM_SIZE (arg.num_params) != buf.buf_len) +if (size_add (sizeof (arg),TEE_IOCTL_PARAM_SIZE (arg.num_params)) != buf.buf_len)return -EINVAL ;if (arg.num_params) {
@@ -565,7 +566,7 @@ static int tee_ioctl_invoke(struct tee_context *ctx,if (copy_from_user (&arg, uarg,sizeof (arg)))return -EFAULT ; -if (sizeof (arg) +TEE_IOCTL_PARAM_SIZE (arg.num_params) != buf.buf_len) +if (size_add (sizeof (arg),TEE_IOCTL_PARAM_SIZE (arg.num_params)) != buf.buf_len)return -EINVAL ;if (arg.num_params) {
@@ -699,7 +700,7 @@ static int tee_ioctl_supp_recv(struct tee_context *ctx,if (get_user (num_params, &uarg->num_params))return -EFAULT ; -if (sizeof (*uarg) +TEE_IOCTL_PARAM_SIZE (num_params) != buf.buf_len) +if (size_add (sizeof (*uarg),TEE_IOCTL_PARAM_SIZE (num_params)) != buf.buf_len)return -EINVAL ; params =kcalloc (num_params,sizeof (struct tee_param ),GFP_KERNEL );
@@ -798,7 +799,7 @@ static int tee_ioctl_supp_send(struct tee_context *ctx, get_user(num_params, &uarg->num_params))return -EFAULT ; -if (sizeof (*uarg) +TEE_IOCTL_PARAM_SIZE (num_params) > buf.buf_len) +if (size_add (sizeof (*uarg),TEE_IOCTL_PARAM_SIZE (num_params)) > buf.buf_len)return -EINVAL ; params =kcalloc (num_params,sizeof (struct tee_param ),GFP_KERNEL );
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 433/592 Revert "bus: ti-sysc: Probe for l4_wkup and l4_cfg interconnect devices first"
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alexander Sverdlin <[email protected]>
[ Upstream commit 36305857b1ead8f6ca033a913162ebc09bee0b43 ]
This reverts commit 4700a00755fb5a4bb5109128297d6fd2d1272ee6.
It breaks target-module@2b300050 ("ti,sysc-omap2") probe on AM62x in a case when minimally-configured system tries to network-boot:
[ 6.888776] probe of 2b300050.target-module returned 517 after 258 usecs [ 17.129637] probe of 2b300050.target-module returned 517 after 708 usecs [ 17.137397] platform 2b300050.target-module: deferred probe pending: (reason unknown) [ 26.878471] Waiting up to 100 more seconds for network.
There are minimal configurations possible when the deferred device is not being probed any more (because everything else has been successfully probed) and deferral lists are not processed any more.
Stable mmc enumeration can be achieved by filling /aliases node properly (4700a00755fb commit's rationale).
After revert:
[ 9.006816] IP-Config: Complete: [ 9.010058] device=lan0, ...
--- drivers/bus/ti-sysc.c | 49 ------------------------------------------- 1 file changed, 49 deletions(-)
drivers/bus/ti-sysc.c +0 -49
@@ -677,51 +677,6 @@ static int sysc_parse_and_check_child_range(struct sysc *ddata)return 0 ; } -/* Interconnect instances to probe before l4_per instances */ -static struct resource early_bus_ranges[] = { -/* am3/4 l4_wkup */ - { .start =0x44c00000 , .end =0x44c00000 +0x300000 , }, -/* omap4/5 and dra7 l4_cfg */ - { .start =0x4a000000 , .end =0x4a000000 +0x300000 , }, -/* omap4 l4_wkup */ - { .start =0x4a300000 , .end =0x4a300000 +0x30000 , }, -/* omap5 and dra7 l4_wkup without dra7 dcan segment */ - { .start =0x4ae00000 , .end =0x4ae00000 +0x30000 , }, -}; - -static atomic_t sysc_defer =ATOMIC_INIT (10 ); - -/** -* sysc_defer_non_critical - defer non_critical interconnect probing -* @ddata: device driver data -* -* We want to probe l4_cfg and l4_wkup interconnect instances before any -* l4_per instances as l4_per instances depend on resources on l4_cfg and -* l4_wkup interconnects. -*/ -static int sysc_defer_non_critical (struct sysc *ddata) -{ -struct resource *res; -int i; - -if (!atomic_read (&sysc_defer)) -return 0 ; - -for (i =0 ; i <ARRAY_SIZE (early_bus_ranges); i++) { - res = &early_bus_ranges[i]; -if (ddata->module_pa >= res->start && - ddata->module_pa <= res->end) { -atomic_set (&sysc_defer,0 ); - -return 0 ; - } - } - -atomic_dec_if_positive (&sysc_defer); - -return -EPROBE_DEFER ; -} -static struct device_node *stdout_path;static void sysc_init_stdout_path (struct sysc *ddata)
@@ -947,10 +902,6 @@ static int sysc_map_and_check_registers(struct sysc *ddata)if (error)return error; - error =sysc_defer_non_critical (ddata); -if (error) -return error; -sysc_check_children (ddata);if (!of_property_present (np,"reg" ))
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 434/592 fs/xattr.c: fix simple_xattr_list()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stephen Smalley <[email protected]>
[ Upstream commit 800d0b9b6a8b1b354637b4194cc167ad1ce2bdd3 ]
commit 8b0ba61df5a1 ("fs/xattr.c: fix simple_xattr_list to always include security.* xattrs") failed to reset err after the call to security_inode_listsecurity(), which returns the length of the returned xattr name. This results in simple_xattr_list() incorrectly returning this length even if a POSIX acl is also set on the inode.
--- fs/xattr.c | 1 + 1 file changed, 1 insertion(+)
fs/xattr.c +1 -0
@@ -1479,6 +1479,7 @@ ssize_t simple_xattr_list(struct inode *inode, struct simple_xattrs *xattrs, buffer += err; } remaining_size -= err; + err =0 ;read_lock (&xattrs->lock);for (rbp =rb_first (&xattrs->rb_root); rbp; rbp =rb_next (rbp)) {
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 435/592 platform/x86/amd: pmc: Clear metrics table at start of cycle
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mario Limonciello <[email protected]>
[ Upstream commit 4dbd11796f3a8eb95647507befc41995458a4023 ]
The area of memory that contains the metrics table may contain garbage when the cycle starts. This normally doesn't matter because the cycle itself will populate it with valid data, however commit 9f5595d5f03fd ("platform/x86/amd: pmc: Require at least 2.5 seconds between HW sleep cycles") started to use it during the check() phase. Depending upon what garbage is in the table it's possible that the system will wait 2.5 seconds for even the first cycle, which will be visible to a user.
To prevent this from happening explicitly clear the table when logging is started.
--- drivers/platform/x86/amd/pmc/pmc.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/platform/x86/amd/pmc/pmc.c +2 -0
@@ -157,6 +157,8 @@ static int amd_pmc_setup_smu_logging(struct amd_pmc_dev *dev)return -ENOMEM ; } +memset_io (dev->smu_virt_addr,0 ,sizeof (struct smu_metrics )); +/* Start the logging */ amd_pmc_send_cmd (dev,0 ,NULL ,SMU_MSG_LOG_RESET , false);amd_pmc_send_cmd (dev,0 ,NULL ,SMU_MSG_LOG_START , false);
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 436/592 platform/x86/amd: pmf: Use device managed allocations
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mario Limonciello <[email protected]>
[ Upstream commit d9db3a941270d92bbd1a6a6b54a10324484f2f2d ]
If setting up smart PC fails for any reason then this can lead to a double free when unloading amd-pmf. This is because dev->buf was freed but never set to NULL and is again freed in amd_pmf_remove().
To avoid subtle allocation bugs in failures leading to a double free change all allocations into device managed allocations.
--- drivers/platform/x86/amd/pmf/core.c | 3 +- drivers/platform/x86/amd/pmf/tee-if.c | 56 ++++++++++----------------- 2 files changed, 22 insertions(+), 37 deletions(-)
drivers/platform/x86/amd/pmf/core.c +1 -2
@@ -280,7 +280,7 @@ int amd_pmf_set_dram_addr(struct amd_pmf_dev *dev, bool alloc_buffer)dev_err (dev->dev,"Invalid CPU id: 0x%x" , dev->cpu_id); } - dev->buf =kzalloc (dev->mtable_size,GFP_KERNEL ); + dev->buf =devm_kzalloc (dev->dev, dev->mtable_size,GFP_KERNEL );if (!dev->buf)return -ENOMEM ; }
@@ -493,7 +493,6 @@ static void amd_pmf_remove(struct platform_device *pdev)mutex_destroy (&dev->lock);mutex_destroy (&dev->update_mutex);mutex_destroy (&dev->cb_mutex); -kfree (dev->buf); }static const struct attribute_group *amd_pmf_driver_groups[] = {
drivers/platform/x86/amd/pmf/tee-if.c +21 -35
122 lines not on this page · raw message
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 437/592 platform/x86/amd: pmf: Prevent amd_pmf_tee_deinit() from running twice
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mario Limonciello <[email protected]>
[ Upstream commit 93103d56650d7a38ed37ba4041578310f82776ae ]
If any of the tee init fails, pass up the errors and clear the tee_ctx pointer. This will prevent cleaning up multiple times.
--- drivers/platform/x86/amd/pmf/tee-if.c | 11 ++++++++--- 1 file changed, 8 insertions(+), 3 deletions(-)
drivers/platform/x86/amd/pmf/tee-if.c +8 -3
@@ -420,12 +420,12 @@ static int amd_pmf_ta_open_session(struct tee_context *ctx, u32 *id, const uuid_ rc =tee_client_open_session (ctx, &sess_arg,NULL );if (rc <0 || sess_arg.ret !=0 ) {pr_err ("Failed to open TEE session err:%#x, rc:%d\n" , sess_arg.ret, rc); -return rc; +return rc ?: -EINVAL ; } *id = sess_arg.session; -return rc; +return 0 ; }static int amd_pmf_register_input_device (struct amd_pmf_dev *dev)
@@ -460,7 +460,9 @@ static int amd_pmf_tee_init(struct amd_pmf_dev *dev, const uuid_t *uuid) dev->tee_ctx =tee_client_open_context (NULL , amd_pmf_amdtee_ta_match,NULL ,NULL );if (IS_ERR (dev->tee_ctx)) {dev_err (dev->dev,"Failed to open TEE context\n" ); -return PTR_ERR (dev->tee_ctx); + ret =PTR_ERR (dev->tee_ctx); + dev->tee_ctx =NULL ; +return ret; } ret =amd_pmf_ta_open_session (dev->tee_ctx, &dev->session_id, uuid);
@@ -500,9 +502,12 @@ static int amd_pmf_tee_init(struct amd_pmf_dev *dev, const uuid_t *uuid)static void amd_pmf_tee_deinit (struct amd_pmf_dev *dev) { +if (!dev->tee_ctx) +return ;tee_shm_free (dev->fw_shm_pool);tee_client_close_session (dev->tee_ctx, dev->session_id);tee_client_close_context (dev->tee_ctx); + dev->tee_ctx =NULL ; }int amd_pmf_init_smart_pc (struct amd_pmf_dev *dev)
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 438/592 platform/x86: dell_rbu: Fix list usage
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stuart Hayes <[email protected]>
[ Upstream commit 61ce04601e0d8265ec6d2ffa6df5a7e1bce64854 ]
Pass the correct list head to list_for_each_entry*() when looping through the packet list.
Without this patch, reading the packet data via sysfs will show the data incorrectly (because it starts at the wrong packet), and clearing the packet list will result in a NULL pointer dereference.
--- drivers/platform/x86/dell/dell_rbu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/platform/x86/dell/dell_rbu.c +2 -2
@@ -292,7 +292,7 @@ static int packet_read_list(char *data, size_t * pread_length) remaining_bytes = *pread_length; bytes_read = rbu_data.packet_read_count; -list_for_each_entry (newpacket, (&packet_data_head.list)->next, list) { +list_for_each_entry (newpacket, &packet_data_head.list, list) { bytes_copied =do_packet_read (pdest, newpacket, remaining_bytes, bytes_read, &temp_count); remaining_bytes -= bytes_copied;
@@ -315,7 +315,7 @@ static void packet_empty_list(void) {struct packet_data *newpacket, *tmp; -list_for_each_entry_safe (newpacket, tmp, (&packet_data_head.list)->next, list) { +list_for_each_entry_safe (newpacket, tmp, &packet_data_head.list, list) {list_del (&newpacket->list);/*
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 439/592 platform/x86: dell_rbu: Stop overwriting data buffer
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stuart Hayes <[email protected]>
[ Upstream commit f4b0fa38d5fefe9aed6ed831f3bd3538c168ee19 ]
The dell_rbu driver will use memset() to clear the data held by each packet when it is no longer needed (when the driver is unloaded, the packet size is changed, etc).
The amount of memory that is cleared (before this patch) is the normal packet size. However, the last packet in the list may be smaller.
Fix this to only clear the memory actually used by each packet, to prevent it from writing past the end of data buffer.
Because the packet data buffers are allocated with __get_free_pages() (in page-sized increments), this bug could only result in a buffer being overwritten when a packet size larger than one page is used. The only user of the dell_rbu module should be the Dell BIOS update program, which uses a packet size of 4096, so no issues should be seen without the patch, it just blocks the possiblity.
--- drivers/platform/x86/dell/dell_rbu.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/x86/dell/dell_rbu.c +1 -1
@@ -322,7 +322,7 @@ static void packet_empty_list(void)* zero out the RBU packet memory before freeing * to make sure there are no stale RBU packets left in memory */ -memset (newpacket->data,0 , rbu_data.packetsize); +memset (newpacket->data,0 , newpacket->length);set_memory_wb ((unsigned long )newpacket->data,1 << newpacket->ordernum); free_pages((unsigned long) newpacket->data,
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 440/592 ovl: fix debug print in case of mkdir error
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Amir Goldstein <[email protected]>
[ Upstream commit 527c88d8390d6c0358dea4d71696795c05328925 ]
We want to print the name in case of mkdir failure and now we will get a cryptic (efault) as name.
--- fs/overlayfs/overlayfs.h | 8 +++++--- 1 file changed, 5 insertions(+), 3 deletions(-)
fs/overlayfs/overlayfs.h +5 -3
@@ -246,9 +246,11 @@ static inline struct dentry *ovl_do_mkdir(struct ovl_fs *ofs, struct dentry *dentry, umode_t mode) { - dentry =vfs_mkdir (ovl_upper_mnt_idmap (ofs), dir, dentry, mode); -pr_debug ("mkdir(%pd2, 0%o) = %i\n" , dentry, mode,PTR_ERR_OR_ZERO (dentry)); -return dentry; +struct dentry *ret; + + ret =vfs_mkdir (ovl_upper_mnt_idmap (ofs), dir, dentry, mode); +pr_debug ("mkdir(%pd2, 0%o) = %i\n" , dentry, mode,PTR_ERR_OR_ZERO (ret)); +return ret; } static inline int ovl_do_mknod(struct ovl_fs *ofs,
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 441/592 powerpc/vdso: Fix build of VDSO32 with pcrel
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Christophe Leroy <[email protected]>
[ Upstream commit b93755f408325170edb2156c6a894ed1cae5f4f6 ]
Building vdso32 on power10 with pcrel leads to following errors:
VDSO32A arch/powerpc/kernel/vdso/gettimeofday-32.o
arch/powerpc/kernel/vdso/gettimeofday.S: Assembler messages:
arch/powerpc/kernel/vdso/gettimeofday.S:40: Error: syntax error; found @', expected ,' arch/powerpc/kernel/vdso/gettimeofday.S:71: Info: macro invoked from here arch/powerpc/kernel/vdso/gettimeofday.S:40: Error: junk at end of line: `@notoc' arch/powerpc/kernel/vdso/gettimeofday.S:71: Info: macro invoked from here
...
make[2]: *** [arch/powerpc/kernel/vdso/Makefile:85: arch/powerpc/kernel/vdso/gettimeofday-32.o] Error 1 make[1]: *** [arch/powerpc/Makefile:388: vdso_prepare] Error 2
Once the above is fixed, the following happens:
VDSO32C arch/powerpc/kernel/vdso/vgettimeofday-32.o cc1: error: '-mpcrel' requires '-mcmodel=medium' make[2]: *** [arch/powerpc/kernel/vdso/Makefile:89: arch/powerpc/kernel/vdso/vgettimeofday-32.o] Error 1 make[1]: *** [arch/powerpc/Makefile:388: vdso_prepare] Error 2 make: *** [Makefile:251: __sub-make] Error 2
Make sure pcrel version of CFUNC() macro is used only for powerpc64 builds and remove -mpcrel for powerpc32 builds.
--- arch/powerpc/include/asm/ppc_asm.h | 2 +- arch/powerpc/kernel/vdso/Makefile | 2 +- 2 files changed, 2 insertions(+), 2 deletions(-)
arch/powerpc/include/asm/ppc_asm.h +1 -1
@@ -183,7 +183,7 @@/* * Used to name C functions called from asm */ -#ifdef CONFIG_PPC_KERNEL_PCREL +#if defined(__powerpc64__) && defined(CONFIG_PPC_KERNEL_PCREL )#define CFUNC (name) name@notoc#else #define CFUNC (name) name
arch/powerpc/kernel/vdso/Makefile +1 -1
@@ -53,7 +53,7 @@ ldflags-$(CONFIG_LD_ORPHAN_WARN) += -Wl,--orphan-handling=$(CONFIG_LD_ORPHAN_WAR ldflags-y += $(filter-out $(CC_AUTO_VAR_INIT_ZERO_ENABLER) $(CC_FLAGS_FTRACE) -Wa$(comma)%, $(KBUILD_CPPFLAGS) $(KBUILD_CFLAGS)) CC32FLAGS := -m32 -CC32FLAGSREMOVE := -mcmodel=medium -mabi=elfv1 -mabi=elfv2 -mcall-aixdesc +CC32FLAGSREMOVE := -mcmodel=medium -mabi=elfv1 -mabi=elfv2 -mcall-aixdesc -mpcrel ifdef CONFIG_CC_IS_CLANG # This flag is supported by clang for 64-bit but not 32-bit so it will cause # an unused command line flag warning for this file.
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 442/592 powerpc/eeh: Fix missing PE bridge reconfiguration during VFIO EEH recovery
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Narayana Murty N <[email protected]>
[ Upstream commit 33bc69cf6655cf60829a803a45275f11a74899e5 ]
VFIO EEH recovery for PCI passthrough devices fails on PowerNV and pseries platforms due to missing host-side PE bridge reconfiguration. In the current implementation, eeh_pe_configure() only performs RTAS or OPAL-based bridge reconfiguration for native host devices, but skips it entirely for PEs managed through VFIO in guest passthrough scenarios.
This leads to incomplete EEH recovery when a PCI error affects a passthrough device assigned to a QEMU/KVM guest. Although VFIO triggers the EEH recovery flow through VFIO_EEH_PE_ENABLE ioctl, the platform-specific bridge reconfiguration step is silently bypassed. As a result, the PE's config space is not fully restored, causing subsequent config space access failures or EEH freeze-on-access errors inside the guest.
This patch fixes the issue by ensuring that eeh_pe_configure() always invokes the platform's configure_bridge() callback (e.g., pseries_eeh_phb_configure_bridge) even for VFIO-managed PEs. This ensures that RTAS or OPAL calls to reconfigure the PE bridge are correctly issued on the host side, restoring the PE's configuration space after an EEH event.
This fix is essential for reliable EEH recovery in QEMU/KVM guests using VFIO PCI passthrough on PowerNV and pseries systems.
Tested with:
- QEMU/KVM guest using VFIO passthrough (IBM Power9,(lpar)Power11 host)
- Injected EEH errors with pseries EEH errinjct tool on host, recovery verified on qemu guest.
- Verified successful config space access and CAP_EXP DevCtl restoration after recovery
--- arch/powerpc/kernel/eeh.c | 2 ++ 1 file changed, 2 insertions(+)
arch/powerpc/kernel/eeh.c +2 -0
@@ -1509,6 +1509,8 @@ int eeh_pe_configure(struct eeh_pe *pe)/* Invalid PE ? */ if (!pe)return -ENODEV ; +else + ret = eeh_ops->configure_bridge (pe);return ret; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 443/592 fs: drop assert in file_seek_cur_needs_f_lock
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Luis Henriques <[email protected]>
[ Upstream commit dd2d6b7f6f519d078a866a36a625b0297d81c5bc ]
The assert in function file_seek_cur_needs_f_lock() can be triggered very easily because there are many users of vfs_llseek() (such as overlayfs) that do their custom locking around llseek instead of relying on fdget_pos(). Just drop the overzealous assertion.
--- fs/file.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
fs/file.c +6 -2
@@ -1198,8 +1198,12 @@ bool file_seek_cur_needs_f_lock(struct file *file)if (!(file->f_mode &FMODE_ATOMIC_POS ) && !file->f_op->iterate_shared)return false; -VFS_WARN_ON_ONCE ((file_count (file) >1 ) && - !mutex_is_locked (&file->f_pos_lock)); +/* +* Note that we are not guaranteed to be called after fdget_pos() on +* this file obj, in which case the caller is expected to provide the +* appropriate locking. +*/ +return true; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 444/592 io_uring/kbuf: dont truncate end buffer for multiple buffer peeks
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Axboe <[email protected]>
commit 26ec15e4b0c1d7b25214d9c0be1d50492e2f006c upstream.
If peeking a bunch of buffers, normally io_ring_buffers_peek() will truncate the end buffer. This isn't optimal as presumably more data will be arriving later, and hence it's better to stop with the last full buffer rather than truncate the end buffer.
--- io_uring/kbuf.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
io_uring/kbuf.c +4 -1
@@ -270,8 +270,11 @@ static int io_ring_buffers_peek(struct i/* truncate end piece, if needed, for non partial buffers */ if (len > arg->max_len) { len = arg->max_len; -if (!(bl->flags &IOBL_INC )) +if (!(bl->flags &IOBL_INC )) { +if (iov != arg->iovs) +break ; buf->len = len; + } } iov->iov_base =u64_to_user_ptr (buf->addr);
Greg Kroah-Hartman
maintainerPATCH 445/592 io_uring/rsrc: validate buffer count with offset for cloning
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Axboe <[email protected]>
commit 1d27f11bf02b38c431e49a17dee5c10a2b4c2e28 upstream.
syzbot reports that it can trigger a WARN_ON() for kmalloc() attempt that's too big:
WARNING: CPU: 0 PID: 6488 at mm/slub.c:5024 __kvmalloc_node_noprof+0x520/0x640 mm/slub.c:5024 Modules linked in: CPU: 0 UID: 0 PID: 6488 Comm: syz-executor312 Not tainted 6.15.0-rc7-syzkaller-gd7fa1af5b33e #0 PREEMPT Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025 pstate: 20400005 (nzCv daif +PAN -UAO -TCO -DIT -SSBS BTYPE=--) pc : __kvmalloc_node_noprof+0x520/0x640 mm/slub.c:5024 lr : __do_kmalloc_node mm/slub.c:-1 [inline] lr : __kvmalloc_node_noprof+0x3b4/0x640 mm/slub.c:5012
sp : ffff80009cfd7a90
x29: ffff80009cfd7ac0 x28: ffff0000dd52a120 x27: 0000000000412dc0
x26: 0000000000000178 x25: ffff7000139faf70 x24: 0000000000000000
x23: ffff800082f4cea8 x22: 00000000ffffffff x21: 000000010cd004a8
x20: ffff0000d75816c0 x19: ffff0000dd52a000 x18: 00000000ffffffff
x17: ffff800092f39000 x16: ffff80008adbe9e4 x15: 0000000000000005
x14: 1ffff000139faf1c x13: 0000000000000000 x12: 0000000000000000
x11: ffff7000139faf21 x10: 0000000000000003 x9 : ffff80008f27b938 x8 : 0000000000000002 x7 : 0000000000000000 x6 : 0000000000000000 x5 : 00000000ffffffff x4 : 0000000000400dc0 x3 : 0000000200000000 x2 : 000000010cd004a8 x1 : ffff80008b3ebc40 x0 : 0000000000000001
Call trace: __kvmalloc_node_noprof+0x520/0x640 mm/slub.c:5024 (P) kvmalloc_array_node_noprof include/linux/slab.h:1065 [inline] io_rsrc_data_alloc io_uring/rsrc.c:206 [inline] io_clone_buffers io_uring/rsrc.c:1178 [inline] io_register_clone_buffers+0x484/0xa14 io_uring/rsrc.c:1287 __io_uring_register io_uring/register.c:815 [inline] __do_sys_io_uring_register io_uring/register.c:926 [inline] __se_sys_io_uring_register io_uring/register.c:903 [inline] __arm64_sys_io_uring_register+0x42c/0xea8 io_uring/register.c:903 __invoke_syscall arch/arm64/kernel/syscall.c:35 [inline] invoke_syscall+0x98/0x2b8 arch/arm64/kernel/syscall.c:49 el0_svc_common+0x130/0x23c arch/arm64/kernel/syscall.c:132 do_el0_svc+0x48/0x58 arch/arm64/kernel/syscall.c:151 el0_svc+0x58/0x17c arch/arm64/kernel/entry-common.c:767 el0t_64_sync_handler+0x78/0x108 arch/arm64/kernel/entry-common.c:786 el0t_64_sync+0x198/0x19c arch/arm64/kernel/entry.S:600
which is due to offset + buffer_count being too large. The registration code checks only the total count of buffers, but given that the indexing is an array, it should also check offset + count. That can't exceed IORING_MAX_REG_BUFFERS either, as there's no way to reach buffers beyond that limit.
There's no issue with registrering a table this large, outside of the fact that it's pointless to register buffers that cannot be reached, and that it can trigger this kmalloc() warning for attempting an allocation that is too large.
--- io_uring/rsrc.c | 2 ++ 1 file changed, 2 insertions(+)
io_uring/rsrc.c +2 -0
@@ -1174,6 +1174,8 @@ static int io_clone_buffers(struct io_rireturn -EINVAL ;if (check_add_overflow (arg->nr, arg->dst_off, &nbufs))return -EOVERFLOW ; +if (nbufs >IORING_MAX_REG_BUFFERS ) +return -EINVAL ; ret =io_rsrc_data_alloc (&data,max (nbufs, ctx->buf_table.nr));if (ret)
Greg Kroah-Hartman
maintainerPATCH 446/592 io_uring: fix task leak issue in io_wq_create()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Penglei Jiang <[email protected]>
commit 89465d923bda180299e69ee2800aab84ad0ba689 upstream.
Add missing put_task_struct() in the error path
--- io_uring/io-wq.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
io_uring/io-wq.c +3 -1
@@ -1236,8 +1236,10 @@ struct io_wq *io_wq_create(unsigned bounatomic_set (&wq->worker_refs,1 );init_completion (&wq->worker_done); ret =cpuhp_state_add_instance_nocalls (io_wq_online, &wq->cpuhp_node); -if (ret) +if (ret) { +put_task_struct (wq->task); goto err; + }return wq; err:
Greg Kroah-Hartman
maintainerPATCH 447/592 drivers/rapidio/rio_cm.c: prevent possible heap overwrite
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Andrew Morton <[email protected]>
commit 50695153d7ddde3b1696dbf0085be0033bf3ddb3 upstream.
In
riocm_cdev_ioctl(RIO_CM_CHAN_SEND)
-> cm_chan_msg_send()
-> riocm_ch_send()
cm_chan_msg_send() checks that userspace didn't send too much data but riocm_ch_send() failed to check that userspace sent sufficient data. The result is that riocm_ch_send() can write to fields in the rio_ch_chan_hdr which were outside the bounds of the space which cm_chan_msg_send() allocated.
Address this by teaching riocm_ch_send() to check that the entire rio_ch_chan_hdr was copied in from userspace.
--- drivers/rapidio/rio_cm.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/rapidio/rio_cm.c +3 -0
@@ -789,6 +789,9 @@ static int riocm_ch_send(u16 ch_id, voidif (buf ==NULL || ch_id ==0 || len ==0 || len >RIO_MAX_MSG_SIZE )return -EINVAL ; +if (len <sizeof (struct rio_ch_chan_hdr )) +return -EINVAL ;/* insufficient data from user */ + ch =riocm_get_channel (ch_id);if (!ch) { riocm_error("%s(%d) ch_%d not found", current->comm,
Greg Kroah-Hartman
maintainerPATCH 448/592 platform/loongarch: laptop: Get brightness setting from EC on probe
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Yao Zi <[email protected]>
commit 1205088fd0393bd9eae96b62bf1e4b9eb1b73edf upstream.
Previously during driver probe, 1 is unconditionally taken as current brightness value and set to props.brightness, which will be considered as the brightness before suspend and restored to EC on resume. Since a brightness value of 1 almost never matches EC's state on coldboot (my laptop's EC defaults to 80), this causes surprising changes of screen brightness on the first time of resume after coldboot.
Let's get brightness from EC and take it as the current brightness on probe of the laptop driver to avoid the surprising behavior. Tested on TongFang L860-T2 Loongson-3A5000 laptop.
--- drivers/platform/loongarch/loongson-laptop.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/loongarch/loongson-laptop.c +1 -1
@@ -392,8 +392,8 @@ static int laptop_backlight_register(voiif (!acpi_evalf (hotkey_handle, &status,"ECLL" ,"d" ))return -EIO ; - props.brightness =1 ; props.max_brightness = status; + props.brightness =ec_get_brightness (); props.type =BACKLIGHT_PLATFORM ; backlight_device_register("loongson_laptop",
Greg Kroah-Hartman
maintainerPATCH 449/592 platform/loongarch: laptop: Unregister generic_sub_drivers on exit
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Yao Zi <[email protected]>
commit f78fb2576f22b0ba5297412a9aa7691920666c41 upstream.
Without correct unregisteration, ACPI notify handlers and the platform drivers installed by generic_subdriver_init() will become dangling references after removing the loongson_laptop module, triggering various kernel faults when a hotkey is sent or at kernel shutdown.
--- drivers/platform/loongarch/loongson-laptop.c | 12 +++++++++--- 1 file changed, 9 insertions(+), 3 deletions(-)
drivers/platform/loongarch/loongson-laptop.c +9 -3
@@ -611,11 +611,17 @@ static int __init generic_acpi_laptop_in static void __exit generic_acpi_laptop_exit(void) { +int i; +if (generic_inputdev) { -if (input_device_registered) -input_unregister_device (generic_inputdev); -else +if (!input_device_registered) {input_free_device (generic_inputdev); + }else { +input_unregister_device (generic_inputdev); + +for (i =0 ; i <ARRAY_SIZE (generic_sub_drivers); i++) +generic_subdriver_exit (&generic_sub_drivers[i]); + } } }
Greg Kroah-Hartman
maintainerPATCH 450/592 platform/loongarch: laptop: Add backlight power control support
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Yao Zi <[email protected]>
commit 53c762b47f726e4079a1f06f684bce2fc0d56fba upstream.
loongson_laptop_turn_{on,off}_backlight() are designed for controlling the power of the backlight, but they aren't really used in the driver previously.
Unify these two functions since they only differ in arguments passed to ACPI method, and wire up loongson_laptop_backlight_update() to update the power state of the backlight as well. Tested on the TongFang L860-T2 Loongson-3A5000 laptop.
--- drivers/platform/loongarch/loongson-laptop.c | 73 +++++++++++++-------------- 1 file changed, 37 insertions(+), 36 deletions(-)
drivers/platform/loongarch/loongson-laptop.c +37 -36
@@ -56,8 +56,7 @@ static struct input_dev *generic_inputdestatic acpi_handle hotkey_handle;static struct key_entry hotkey_keycode_map[GENERIC_HOTKEY_MAP_MAX ]; -int loongson_laptop_turn_on_backlight (void ); -int loongson_laptop_turn_off_backlight (void ); +static bool bl_powered;static int loongson_laptop_backlight_update (struct backlight_device *bd);/* 2. ACPI Helpers and device model */
@@ -354,16 +353,42 @@ static int ec_backlight_level(u8 level)return level; } +static int ec_backlight_set_power (bool state) +{ +int status; +union acpi_object arg0 = {ACPI_TYPE_INTEGER }; +struct acpi_object_list args = {1 , &arg0 }; + + arg0.integer.value = state; + status =acpi_evaluate_object (NULL ,"\\BLSW" , &args,NULL ); +if (ACPI_FAILURE (status)) { +pr_info ("Loongson lvds error: 0x%x\n" , status); +return -EIO ; + } + +return 0 ; +} +static int loongson_laptop_backlight_update (struct backlight_device *bd) { -int lvl =ec_backlight_level (bd->props.brightness); +bool target_powered = !backlight_is_blank (bd); +int ret =0 , lvl =ec_backlight_level (bd->props.brightness);if (lvl <0 )return -EIO ; +if (ec_set_brightness (lvl))return -EIO ; -return 0 ; +if (target_powered != bl_powered) { + ret =ec_backlight_set_power (target_powered); +if (ret <0 ) +return ret; + + bl_powered = target_powered; + } + +return ret; }static int loongson_laptop_get_brightness (struct backlight_device *bd)
@@ -384,7 +409,7 @@ static const struct backlight_ops backlistatic int laptop_backlight_register (void ) { -int status =0 ; +int status =0 , ret;struct backlight_properties props;memset (&props,0 ,sizeof (props));
@@ -392,44 +417,20 @@ static int laptop_backlight_register(voiif (!acpi_evalf (hotkey_handle, &status,"ECLL" ,"d" ))return -EIO ; + ret =ec_backlight_set_power (true); +if (ret) +return ret; + + bl_powered = true; + props.max_brightness = status; props.brightness =ec_get_brightness (); + props.power =BACKLIGHT_POWER_ON ; props.type =BACKLIGHT_PLATFORM ;backlight_device_register ("loongson_laptop" ,NULL ,NULL , &backlight_laptop_ops, &props); -return 0 ; -} - -int loongson_laptop_turn_on_backlight (void ) -{ -int status; -union acpi_object arg0 = {ACPI_TYPE_INTEGER }; -struct acpi_object_list args = {1 , &arg0 }; - - arg0.integer.value =1 ; - status =acpi_evaluate_object (NULL ,"\\BLSW" , &args,NULL ); -if (ACPI_FAILURE (status)) { -pr_info ("Loongson lvds error: 0x%x\n" , status); -return -ENODEV ; - } - -return 0 ; -} - -int loongson_laptop_turn_off_backlight (void ) -{ -int status; -union acpi_object arg0 = {ACPI_TYPE_INTEGER }; -struct acpi_object_list args = {1 , &arg0 }; - - arg0.integer.value =0 ; - status =acpi_evaluate_object (NULL ,"\\BLSW" , &args,NULL ); -if (ACPI_FAILURE (status)) { -pr_info ("Loongson lvds error: 0x%x\n" , status); -return -ENODEV ; - }return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 451/592 LoongArch: vDSO: Correctly use asm parameters in syscall wrappers
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Thomas Weißschuh <[email protected]>
commit e242bbbb6d7ac7556aa1e358294dc7e3c82cc902 upstream.
The syscall wrappers use the "a0" register for two different register variables, both the first argument and the return value. Here the "ret" variable is used as both input and output while the argument register is only used as input. Clang treats the conflicting input parameters as an undefined behaviour and optimizes away the argument assignment.
The code seems to work by chance for the most part today but that may change in the future. Specifically clock_gettime_fallback() fails with clockids from 16 to 23, as implemented by the upcoming auxiliary clocks.
Switch the "ret" register variable to a pure output, similar to the other architectures' vDSO code. This works in both clang and GCC.
--- arch/loongarch/include/asm/vdso/getrandom.h | 2 +- arch/loongarch/include/asm/vdso/gettimeofday.h | 6 +++--- 2 files changed, 4 insertions(+), 4 deletions(-)
arch/loongarch/include/asm/vdso/getrandom.h +1 -1
@@ -20,7 +20,7 @@ static __always_inline ssize_t getrandom asmvolatile (" syscall 0\n" - :"+r" (ret) + :"=r" (ret) :"r" (nr),"r" (buffer),"r" (len),"r" (flags) :"$t0" ,"$t1" ,"$t2" ,"$t3" ,"$t4" ,"$t5" ,"$t6" ,"$t7" ,"$t8" ,"memory" );
arch/loongarch/include/asm/vdso/gettimeofday.h +3 -3
@@ -25,7 +25,7 @@ static __always_inline long gettimeofday asmvolatile (" syscall 0\n" - :"+r" (ret) + :"=r" (ret) :"r" (nr),"r" (tv),"r" (tz) :"$t0" ,"$t1" ,"$t2" ,"$t3" ,"$t4" ,"$t5" ,"$t6" ,"$t7" ,"$t8" ,"memory" );
@@ -44,7 +44,7 @@ static __always_inline long clock_gettim asmvolatile (" syscall 0\n" - :"+r" (ret) + :"=r" (ret) :"r" (nr),"r" (clkid),"r" (ts) :"$t0" ,"$t1" ,"$t2" ,"$t3" ,"$t4" ,"$t5" ,"$t6" ,"$t7" ,"$t8" ,"memory" );
@@ -63,7 +63,7 @@ static __always_inline int clock_getres_ asmvolatile (" syscall 0\n" - :"+r" (ret) + :"=r" (ret) :"r" (nr),"r" (clkid),"r" (ts) :"$t0" ,"$t1" ,"$t2" ,"$t3" ,"$t4" ,"$t5" ,"$t6" ,"$t7" ,"$t8" ,"memory" );
Greg Kroah-Hartman
maintainerPATCH 452/592 LoongArch: Avoid using $r0/$r1 as "mask" for csrxchg
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Huacai Chen <[email protected]>
commit 52c22661c79a7b6af7fad9f77200738fc6c51878 upstream.
When building kernel with LLVM there are occasionally such errors:
In file included from ./include/linux/spinlock.h:59:
In file included from ./include/linux/irqflags.h:17:arch/loongarch/include/asm/irqflags.h:38:3: error: must not be $r0 or $r1
38 | "csrxchg %[val], %[mask], %[reg]\n\t"
| ^
<inline asm>:1:16: note: instantiated into assembly here
1 | csrxchg $a1, $ra, 0
| ^To prevent the compiler from allocating $r0 or $r1 for the "mask" of the csrxchg instruction, the 'q' constraint must be used but Clang < 21 does not support it. So force to use $t0 in the inline asm, in order to avoid using $r0/$r1 while keeping the backward compatibility.
--- arch/loongarch/include/asm/irqflags.h | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-)
arch/loongarch/include/asm/irqflags.h +12 -4
@@ -14,40 +14,48 @@static inline void arch_local_irq_enable (void ) {u32 flags =CSR_CRMD_IE ; + register u32 mask asm("t0") = CSR_CRMD_IE; + __asm__ __volatile__("csrxchg %[val], %[mask], %[reg]\n\t" : [val]"+r" (flags) - : [mask]"r" (CSR_CRMD_IE ), [reg]"i" (LOONGARCH_CSR_CRMD ) + : [mask]"r" (mask), [reg]"i" (LOONGARCH_CSR_CRMD ) :"memory" ); }static inline void arch_local_irq_disable (void ) {u32 flags =0 ; + register u32 mask asm("t0") = CSR_CRMD_IE; + __asm__ __volatile__("csrxchg %[val], %[mask], %[reg]\n\t" : [val]"+r" (flags) - : [mask]"r" (CSR_CRMD_IE ), [reg]"i" (LOONGARCH_CSR_CRMD ) + : [mask]"r" (mask), [reg]"i" (LOONGARCH_CSR_CRMD ) :"memory" ); }static inline unsigned long arch_local_irq_save (void ) {u32 flags =0 ; + register u32 mask asm("t0") = CSR_CRMD_IE; + __asm__ __volatile__("csrxchg %[val], %[mask], %[reg]\n\t" : [val]"+r" (flags) - : [mask]"r" (CSR_CRMD_IE ), [reg]"i" (LOONGARCH_CSR_CRMD ) + : [mask]"r" (mask), [reg]"i" (LOONGARCH_CSR_CRMD ) :"memory" );return flags; }static inline void arch_local_irq_restore (unsigned long flags) { + register u32 mask asm("t0") = CSR_CRMD_IE; + __asm__ __volatile__("csrxchg %[val], %[mask], %[reg]\n\t" : [val]"+r" (flags) - : [mask]"r" (CSR_CRMD_IE ), [reg]"i" (LOONGARCH_CSR_CRMD ) + : [mask]"r" (mask), [reg]"i" (LOONGARCH_CSR_CRMD ) :"memory" ); }
Greg Kroah-Hartman
maintainerPATCH 453/592 LoongArch: Fix panic caused by NULL-PMD in huge_pte_offset()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tianyang Zhang <[email protected]>
commit ee084fa96123ede8b0563a1b5a9b23adc43cd50d upstream.
ERROR INFO:
CPU 25 Unable to handle kernel paging request at virtual address 0x0
... Call Trace: [<900000000023c30c>] huge_pte_offset+0x3c/0x58 [<900000000057fd4c>] hugetlb_follow_page_mask+0x74/0x438 [<900000000051fee8>] __get_user_pages+0xe0/0x4c8 [<9000000000522414>] faultin_page_range+0x84/0x380 [<9000000000564e8c>] madvise_vma_behavior+0x534/0xa48 [<900000000056689c>] do_madvise+0x1bc/0x3e8 [<9000000000566df4>] sys_madvise+0x24/0x38 [<90000000015b9e88>] do_syscall+0x78/0x98 [<9000000000221f18>] handle_syscall+0xb8/0x158
In some cases, pmd may be NULL and rely on NULL as the return value for processing, so it is necessary to determine this situation here.
--- arch/loongarch/mm/hugetlbpage.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
arch/loongarch/mm/hugetlbpage.c +2 -1
@@ -47,7 +47,8 @@ pte_t *huge_pte_offset(struct mm_struct pmd =pmd_offset (pud, addr); } } -return pmd_none (pmdp_get (pmd)) ?NULL : (pte_t *) pmd; + +return (!pmd ||pmd_none (pmdp_get (pmd))) ?NULL : (pte_t *) pmd; }uint64_t pmd_to_entrylo (unsigned long pmd_val)
Greg Kroah-Hartman
maintainerPATCH 454/592 firmware: cs_dsp: Fix OOB memory read access in KUnit test (ctl cache)
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jaroslav Kysela <[email protected]>
commit f4ba2ea57da51d616b689c4b8826c517ff5a8523 upstream.
KASAN reported out of bounds access - cs_dsp_ctl_cache_init_multiple_offsets(). The code uses mock_coeff_template.length_bytes (4 bytes) for register value allocations. But later, this length is set to 8 bytes which causes test code failures.
As fix, just remove the lenght override, keeping the original value 4 for all operations.
--- drivers/firmware/cirrus/test/cs_dsp_test_control_cache.c | 1 - 1 file changed, 1 deletion(-)
drivers/firmware/cirrus/test/cs_dsp_test_control_cache.c +0 -1
@@ -776,7 +776,6 @@ static void cs_dsp_ctl_cache_init_multip "dummyalg", NULL);/* Create controls identical except for offset */ - def.length_bytes =8 ; def.offset_dsp_words =0 ; def.shortname ="CtlA" ;cs_dsp_mock_wmfw_add_coeff_desc (local->wmfw_builder, &def);
Greg Kroah-Hartman
maintainerPATCH 455/592 firmware: cs_dsp: Fix OOB memory read access in KUnit test (wmfw info)
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jaroslav Kysela <[email protected]>
commit d979b783d61f7f1f95664031b71a33afc74627b2 upstream.
KASAN reported out of bounds access - cs_dsp_mock_wmfw_add_info(), because the source string length was rounded up to the allocation size.
--- drivers/firmware/cirrus/test/cs_dsp_mock_wmfw.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/firmware/cirrus/test/cs_dsp_mock_wmfw.c +2 -1
@@ -133,10 +133,11 @@ void cs_dsp_mock_wmfw_add_info(struct csif (info_len %4 ) {/* Create a padded string with length a multiple of 4 */ +size_t copy_len = info_len; info_len =round_up (info_len,4 ); tmp =kunit_kzalloc (builder->test_priv->test, info_len,GFP_KERNEL );KUNIT_ASSERT_NOT_ERR_OR_NULL (builder->test_priv->test, tmp); -memcpy (tmp, info, info_len); +memcpy (tmp, info, copy_len); info = tmp; }
Greg Kroah-Hartman
maintainerPATCH 456/592 jffs2: check that raw node were preallocated before writing summary
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Artem Sadovnikov <[email protected]>
commit ec9e6f22bce433b260ea226de127ec68042849b0 upstream.
Syzkaller detected a kernel bug in jffs2_link_node_ref, caused by fault injection in jffs2_prealloc_raw_node_refs. jffs2_sum_write_sumnode doesn't check return value of jffs2_prealloc_raw_node_refs and simply lets any error propagate into jffs2_sum_write_data, which eventually calls jffs2_link_node_ref in order to link the summary to an expectedly allocated node.
kernel BUG at fs/jffs2/nodelist.c:592!
invalid opcode: 0000 [#1] PREEMPT SMP KASAN NOPTI
CPU: 1 PID: 31277 Comm: syz-executor.7 Not tainted 6.1.128-syzkaller-00139-ge10f83ca10a1 #0 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014 RIP: 0010:jffs2_link_node_ref+0x570/0x690 fs/jffs2/nodelist.c:592 Call Trace: <TASK> jffs2_sum_write_data fs/jffs2/summary.c:841 [inline] jffs2_sum_write_sumnode+0xd1a/0x1da0 fs/jffs2/summary.c:874 jffs2_do_reserve_space+0xa18/0xd60 fs/jffs2/nodemgmt.c:388 jffs2_reserve_space+0x55f/0xaa0 fs/jffs2/nodemgmt.c:197 jffs2_write_inode_range+0x246/0xb50 fs/jffs2/write.c:362 jffs2_write_end+0x726/0x15d0 fs/jffs2/file.c:301 generic_perform_write+0x314/0x5d0 mm/filemap.c:3856 __generic_file_write_iter+0x2ae/0x4d0 mm/filemap.c:3973 generic_file_write_iter+0xe3/0x350 mm/filemap.c:4005 call_write_iter include/linux/fs.h:2265 [inline] do_iter_readv_writev+0x20f/0x3c0 fs/read_write.c:735 do_iter_write+0x186/0x710 fs/read_write.c:861 vfs_iter_write+0x70/0xa0 fs/read_write.c:902 iter_file_splice_write+0x73b/0xc90 fs/splice.c:685 do_splice_from fs/splice.c:763 [inline] direct_splice_actor+0x10c/0x170 fs/splice.c:950 splice_direct_to_actor+0x337/0xa10 fs/splice.c:896 do_splice_direct+0x1a9/0x280 fs/splice.c:1002 do_sendfile+0xb13/0x12c0 fs/read_write.c:1255 __do_sys_sendfile64 fs/read_write.c:1323 [inline] __se_sys_sendfile64 fs/read_write.c:1309 [inline] __x64_sys_sendfile64+0x1cf/0x210 fs/read_write.c:1309 do_syscall_x64 arch/x86/entry/common.c:51 [inline] do_syscall_64+0x35/0x80 arch/x86/entry/common.c:81 entry_SYSCALL_64_after_hwframe+0x6e/0xd8
Fix this issue by checking return value of jffs2_prealloc_raw_node_refs before calling jffs2_sum_write_data.
Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
--- fs/jffs2/summary.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
fs/jffs2/summary.c +6 -1
@@ -858,7 +858,10 @@ int jffs2_sum_write_sumnode(struct jffs2spin_unlock (&c->erase_completion_lock); jeb = c->nextblock; -jffs2_prealloc_raw_node_refs (c, jeb,1 ); + ret =jffs2_prealloc_raw_node_refs (c, jeb,1 ); + +if (ret) + goto out;if (!c->summary->sum_num || !c->summary->sum_list_head) {JFFS2_WARNING ("Empty summary info!!!\n" );
@@ -872,6 +875,8 @@ int jffs2_sum_write_sumnode(struct jffs2 datasize += padsize; ret =jffs2_sum_write_data (c, jeb, infosize, datasize, padsize); + +out:spin_lock (&c->erase_completion_lock);return ret; }
Greg Kroah-Hartman
maintainerPATCH 457/592 jffs2: check jffs2_prealloc_raw_node_refs() result in few other places
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Fedor Pchelkin <[email protected]>
commit 2b6d96503255a3ed676cd70f8368870c6d6a25c6 upstream.
Fuzzing hit another invalid pointer dereference due to the lack of checking whether jffs2_prealloc_raw_node_refs() completed successfully. Subsequent logic implies that the node refs have been allocated.
Handle that. The code is ready for propagating the error upwards.
KASAN: null-ptr-deref in range [0x0000000000000008-0x000000000000000f] CPU: 1 PID: 5835 Comm: syz-executor145 Not tainted 5.10.234-syzkaller #0 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.12.0-1 04/01/2014 RIP: 0010:jffs2_link_node_ref+0xac/0x690 fs/jffs2/nodelist.c:600 Call Trace: jffs2_mark_erased_block fs/jffs2/erase.c:460 [inline] jffs2_erase_pending_blocks+0x688/0x1860 fs/jffs2/erase.c:118 jffs2_garbage_collect_pass+0x638/0x1a00 fs/jffs2/gc.c:253 jffs2_reserve_space+0x3f4/0xad0 fs/jffs2/nodemgmt.c:167 jffs2_write_inode_range+0x246/0xb50 fs/jffs2/write.c:362 jffs2_write_end+0x712/0x1110 fs/jffs2/file.c:302 generic_perform_write+0x2c2/0x500 mm/filemap.c:3347 __generic_file_write_iter+0x252/0x610 mm/filemap.c:3465 generic_file_write_iter+0xdb/0x230 mm/filemap.c:3497 call_write_iter include/linux/fs.h:2039 [inline] do_iter_readv_writev+0x46d/0x750 fs/read_write.c:740 do_iter_write+0x18c/0x710 fs/read_write.c:866 vfs_writev+0x1db/0x6a0 fs/read_write.c:939 do_pwritev fs/read_write.c:1036 [inline] __do_sys_pwritev fs/read_write.c:1083 [inline] __se_sys_pwritev fs/read_write.c:1078 [inline] __x64_sys_pwritev+0x235/0x310 fs/read_write.c:1078 do_syscall_64+0x30/0x40 arch/x86/entry/common.c:46 entry_SYSCALL_64_after_hwframe+0x67/0xd1
Found by Linux Verification Center (linuxtesting.org) with Syzkaller.
--- fs/jffs2/erase.c | 4 +++- fs/jffs2/scan.c | 4 +++- 2 files changed, 6 insertions(+), 2 deletions(-)
fs/jffs2/erase.c +3 -1
@@ -425,7 +425,9 @@ static void jffs2_mark_erased_block(stru .totlen = cpu_to_je32(c->cleanmarker_size) }; -jffs2_prealloc_raw_node_refs (c, jeb,1 ); + ret =jffs2_prealloc_raw_node_refs (c, jeb,1 ); +if (ret) + goto filebad; marker.hdr_crc =cpu_to_je32 (crc32 (0 , &marker,sizeof (struct jffs2_unknown_node )-4 ));
fs/jffs2/scan.c +3 -1
@@ -256,7 +256,9 @@ int jffs2_scan_medium(struct jffs2_sb_injffs2_dbg (1 ,"%s(): Skipping %d bytes in nextblock to ensure page alignment\n" , __func__, skip); -jffs2_prealloc_raw_node_refs (c, c->nextblock,1 ); + ret =jffs2_prealloc_raw_node_refs (c, c->nextblock,1 ); +if (ret) + goto out; jffs2_scan_dirty_space(c, c->nextblock, skip); }#endif
Greg Kroah-Hartman
maintainerPATCH 458/592 cifs: deal with the channel loading lag while picking channels
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Shyam Prasad N <[email protected]>
commit 66d590b828b1fd9fa337047ae58fe1c4c6f43609 upstream.
Our current approach to select a channel for sending requests is this:
1. iterate all channels to find the min and max queue depth
2. if min and max are not the same, pick the channel with min depth
3. if min and max are same, round robin, as all channels are equally loaded
The problem with this approach is that there's a lag between selecting a channel and sending the request (that increases the queue depth on the channel). While these numbers will eventually catch up, there could be a skew in the channel usage, depending on the application's I/O parallelism and the server's speed of handling requests.
With sufficient parallelism, this lag can artificially increase the queue depth, thereby impacting the performance negatively.
This change will change the step 1 above to start the iteration from the last selected channel. This is to reduce the skew in channel usage even in the presence of this lag.
--- fs/smb/client/transport.c | 14 +++++++------- 1 file changed, 7 insertions(+), 7 deletions(-)
fs/smb/client/transport.c +7 -7
@@ -1018,14 +1018,16 @@ struct TCP_Server_Info *cifs_pick_channeuint index =0 ;unsigned int min_in_flight =UINT_MAX , max_in_flight =0 ;struct TCP_Server_Info *server =NULL ; -int i; +int i, start, cur;if (!ses)return NULL ;spin_lock (&ses->chan_lock); + start =atomic_inc_return (&ses->chan_seq);for (i =0 ; i < ses->chan_count; i++) { - server = ses->chans[i].server; + cur = (start + i) % ses->chan_count; + server = ses->chans[cur].server;if (!server || server->terminate)continue ;
@@ -1042,17 +1044,15 @@ struct TCP_Server_Info *cifs_pick_channe*/ if (server->in_flight < min_in_flight) { min_in_flight = server->in_flight; - index = i; + index = cur; }if (server->in_flight > max_in_flight) max_in_flight = server->in_flight; }/* if all channels are equally loaded, fall back to round-robin */ -if (min_in_flight == max_in_flight) { - index = (uint )atomic_inc_return (&ses->chan_seq); - index %= ses->chan_count; - } +if (min_in_flight == max_in_flight) + index = (uint )start % ses->chan_count; server = ses->chans[index].server;spin_unlock (&ses->chan_lock);
Greg Kroah-Hartman
maintainerPATCH 459/592 cifs: serialize other channels when query server interfaces is pending
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Shyam Prasad N <[email protected]>
commit b5e3e6e28cf3853566ba5d816f79aba5be579158 upstream.
Today, during smb2_reconnect, session_mutex is released as soon as the tcon is reconnected and is in a good state. However, in case multichannel is enabled, there is also a query of server interfaces that follows. We've seen that this query can race with reconnects of other channels, causing them to step on each other with reconnects.
This change extends the hold of session_mutex till after the query of server interfaces is complete. In order to avoid recursive smb2_reconnect checks during query ioctl, this change also introduces a session flag for sessions where such a query is in progress.
--- fs/smb/client/cifsglob.h | 1 + fs/smb/client/smb2pdu.c | 24 ++++++++++++++++++------ 2 files changed, 19 insertions(+), 6 deletions(-)
fs/smb/client/cifsglob.h +1 -0
@@ -1084,6 +1084,7 @@ struct cifs_chan { };#define CIFS_SES_FLAG_SCALE_CHANNELS (0x1) +#define CIFS_SES_FLAGS_PENDING_QUERY_INTERFACES (0x2)/* * Session structure. One of these for each uid session with a particular host
fs/smb/client/smb2pdu.c +18 -6
@@ -411,14 +411,19 @@ skip_sess_setup:if (!rc && (server->capabilities &SMB2_GLOBAL_CAP_MULTI_CHANNEL ) && server->ops->query_server_interfaces) { -mutex_unlock (&ses->session_mutex); -/* -* query server network interfaces, in case they change +* query server network interfaces, in case they change. +* Also mark the session as pending this update while the query +* is in progress. This will be used to avoid calling +* smb2_reconnect recursively. */ + ses->flags |=CIFS_SES_FLAGS_PENDING_QUERY_INTERFACES ; xid =get_xid (); rc = server->ops->query_server_interfaces (xid, tcon, false);free_xid (xid); + ses->flags &= ~CIFS_SES_FLAGS_PENDING_QUERY_INTERFACES ; + +mutex_unlock (&ses->session_mutex);if (rc == -EOPNOTSUPP && ses->chan_count >1 ) {/*
@@ -560,11 +565,18 @@ static int smb2_ioctl_req_init(u32 opcod struct TCP_Server_Info *server, void **request_buf, unsigned int *total_len) { - /* Skip reconnect only for FSCTL_VALIDATE_NEGOTIATE_INFO IOCTLs */ - if (opcode == FSCTL_VALIDATE_NEGOTIATE_INFO) { +/* +* Skip reconnect in one of the following cases: +* 1. For FSCTL_VALIDATE_NEGOTIATE_INFO IOCTLs +* 2. For FSCTL_QUERY_NETWORK_INTERFACE_INFO IOCTL when called from +* smb2_reconnect (indicated by CIFS_SES_FLAG_SCALE_CHANNELS ses flag) +*/ +if (opcode ==FSCTL_VALIDATE_NEGOTIATE_INFO || + (opcode ==FSCTL_QUERY_NETWORK_INTERFACE_INFO && + (tcon->ses->flags &CIFS_SES_FLAGS_PENDING_QUERY_INTERFACES )))return __smb2_plain_req_init (SMB2_IOCTL , tcon, server, request_buf, total_len); - } +return smb2_plain_req_init (SMB2_IOCTL , tcon, server, request_buf, total_len); }
Greg Kroah-Hartman
maintainerPATCH 460/592 cifs: do not disable interface polling on failure
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Shyam Prasad N <[email protected]>
commit 42ca547b13a20e7cbb04fbdf8d5f089ac4bb35b7 upstream.
When a server has multichannel enabled, we keep polling the server for interfaces periodically. However, when this query fails, we disable the polling. This can be problematic as it takes away the chance for the server to start advertizing again.
This change reschedules the delayed work, even if the current call failed. That way, multichannel sessions can recover.
--- fs/smb/client/connect.c | 6 +----- fs/smb/client/smb2pdu.c | 9 +++++---- 2 files changed, 6 insertions(+), 9 deletions(-)
fs/smb/client/connect.c +1 -5
@@ -116,13 +116,9 @@ static void smb2_query_server_interfaces rc = server->ops->query_server_interfaces (xid, tcon, false);free_xid (xid); -if (rc) { -if (rc == -EOPNOTSUPP ) -return ; - +if (rc)cifs_dbg (FYI ,"%s: failed to query server interfaces: %d\n" , __func__, rc); - }queue_delayed_work (cifsiod_wq, &tcon->query_interfaces, (SMB_INTERFACE_POLL_INTERVAL *HZ ));
fs/smb/client/smb2pdu.c +5 -4
@@ -423,6 +423,10 @@ skip_sess_setup:free_xid (xid); ses->flags &= ~CIFS_SES_FLAGS_PENDING_QUERY_INTERFACES ; +/* regardless of rc value, setup polling */ +queue_delayed_work (cifsiod_wq, &tcon->query_interfaces, + (SMB_INTERFACE_POLL_INTERVAL *HZ )); +mutex_unlock (&ses->session_mutex);if (rc == -EOPNOTSUPP && ses->chan_count >1 ) {
@@ -443,11 +447,8 @@ skip_sess_setup:if (ses->chan_max > ses->chan_count && ses->iface_count && !SERVER_IS_CHAN (server)) { -if (ses->chan_count ==1 ) { +if (ses->chan_count ==1 )cifs_server_dbg (VFS ,"supports multichannel now\n" ); -queue_delayed_work (cifsiod_wq, &tcon->query_interfaces, - (SMB_INTERFACE_POLL_INTERVAL *HZ )); - }cifs_try_adding_channels (ses); }
Greg Kroah-Hartman
maintainerPATCH 461/592 tracing: Fix regression of filter waiting a long time on RCU synchronization
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
commit a9d0aab5eb33a44792a66b7af13ff50d7b3e7022 upstream.
When faultable trace events were added, a trace event may no longer use normal RCU to synchronize but instead used synchronize_rcu_tasks_trace(). This synchronization takes a much longer time to synchronize.
The filter logic would free the filters by calling tracepoint_synchronize_unregister() after it unhooked the filter strings and before freeing them. With this function now calling synchronize_rcu_tasks_trace() this increased the time to free a filter tremendously. On a PREEMPT_RT system, it was even more noticeable.
# time trace-cmd record -p function sleep 1 [..] real 2m29.052s user 0m0.244s sys 0m20.136s
As trace-cmd would clear out all the filters before recording, it could take up to 2 minutes to do a recording of "sleep 1".
To find out where the issues was:
~# trace-cmd sqlhist -e -n sched_stack select start.prev_state as state, end.next_comm as comm, TIMESTAMP_DELTA_USECS as delta, start.STACKTRACE as stack from sched_switch as start join sched_switch as end on start.prev_pid = end.next_pid
Which will produce the following commands (and -e will also execute them):
echo 's:sched_stack s64 state; char comm[16]; u64 delta; unsigned long stack[];' >> /sys/kernel/tracing/dynamic_events
echo 'hist:keys=prev_pid:__arg_18057_2=prev_state,__arg_18057_4=common_timestamp.usecs,__arg_18057_7=common_stacktrace' >> /sys/kernel/tracing/events/sched/sched_switch/trigger
echo 'hist:keys=next_pid:__state_18057_1=$__arg_18057_2,__comm_18057_3=next_comm,__delta_18057_5=common_timestamp.usecs-$__arg_18057_4,__stack_18057_6=$__arg_18057_7:onmatch(sched.sched_switch).trace(sched_stack,$__state_18057_1,$__comm_18057_3,$__delta_18057_5,$__stack_18057_6)' >> /sys/kernel/tracing/events/sched/sched_switch/trigger
The above creates a synthetic event that creates a stack trace when a task schedules out and records it with the time it scheduled back in. Basically the time a task is off the CPU. It also records the state of the task when it left the CPU (running, blocked, sleeping, etc). It also saves the comm of the task as "comm" (needed for the next command).
~# echo 'hist:keys=state,stack.stacktrace:vals=delta:sort=state,delta if comm == "trace-cmd" && state & 3' > /sys/kernel/tracing/events/synthetic/sched_stack/trigger
The above creates a histogram with buckets per state, per stack, and the value of the total time it was off the CPU for that stack trace. It filters on tasks with "comm == trace-cmd" and only the sleeping and blocked states (1 - sleeping, 2 - blocked).
~# trace-cmd record -p function sleep 1
~# cat /sys/kernel/tracing/events/synthetic/sched_stack/hist | tail -18
{ state: 2, stack.stacktrace __schedule+0x1545/0x3700
schedule+0xe2/0x390
schedule_timeout+0x175/0x200
wait_for_completion_state+0x294/0x440
__wait_rcu_gp+0x247/0x4f0
synchronize_rcu_tasks_generic+0x151/0x230
apply_subsystem_event_filter+0xa2b/0x1300
subsystem_filter_write+0x67/0xc0
vfs_write+0x1e2/0xeb0
ksys_write+0xff/0x1d0
do_syscall_64+0x7b/0x420
entry_SYSCALL_64_after_hwframe+0x76/0x7e} hitcount: 237 delta: 99756288 <<--------------- Delta is 99 seconds!
Totals:
Hits: 525
Entries: 21
Dropped: 0This shows that this particular trace waited for 99 seconds on synchronize_rcu_tasks() in apply_subsystem_event_filter().
In fact, there's a lot of places in the filter code that spends a lot of time waiting for synchronize_rcu_tasks_trace() in order to free the filters.
Add helper functions that will use call_rcu*() variants to asynchronously free the filters. This brings the timings back to normal:
# time trace-cmd record -p function sleep 1 [..] real 0m14.681s user 0m0.335s sys 0m28.616s
And the histogram also shows this:
~# cat /sys/kernel/tracing/events/synthetic/sched_stack/hist | tail -21
{ state: 2, stack.stacktrace __schedule+0x1545/0x3700
schedule+0xe2/0x390
schedule_timeout+0x175/0x200
wait_for_completion_state+0x294/0x440
__wait_rcu_gp+0x247/0x4f0
synchronize_rcu_normal+0x3db/0x5c0
tracing_reset_online_cpus+0x8f/0x1e0
tracing_open+0x335/0x440
do_dentry_open+0x4c6/0x17a0
vfs_open+0x82/0x360
path_openat+0x1a36/0x2990
do_filp_open+0x1c5/0x420
do_sys_openat2+0xed/0x180
__x64_sys_openat+0x108/0x1d0
do_syscall_64+0x7b/0x420} hitcount: 2 delta: 77044
Totals:
Hits: 55
Entries: 28
Dropped: 0Where the total waiting time of synchronize_rcu_tasks_trace() is 77 milliseconds.
--- kernel/trace/trace_events_filter.c | 186 +++++++++++++++++++++++++++---------- 1 file changed, 138 insertions(+), 48 deletions(-)
kernel/trace/trace_events_filter.c +138 -48
@@ -1335,22 +1335,139 @@ static void filter_free_subsystem_preds( } } +struct filter_list { +struct list_head list; +struct event_filter *filter; +}; + +struct filter_head { +struct list_head list; +struct rcu_head rcu; +}; + + +static void free_filter_list (struct rcu_head *rhp) +{ + struct filter_head *filter_list = container_of(rhp, struct filter_head, rcu); +struct filter_list *filter_item, *tmp; + +list_for_each_entry_safe (filter_item, tmp, &filter_list->list, list) { +__free_filter (filter_item->filter); +list_del (&filter_item->list); +kfree (filter_item); + } +kfree (filter_list); +} + +static void free_filter_list_tasks (struct rcu_head *rhp) +{ +call_rcu (rhp, free_filter_list); +} + +/* +* The tracepoint_synchronize_unregister() is a double rcu call. +* It calls synchronize_rcu_tasks_trace() followed by synchronize_rcu(). +* Instead of waiting for it, simply call these via the call_rcu*() +* variants. +*/ +static void delay_free_filter (struct filter_head *head) +{ +call_rcu_tasks_trace (&head->rcu, free_filter_list_tasks); +} + +static void try_delay_free_filter (struct event_filter *filter) +{ +struct filter_head *head; +struct filter_list *item; + + head =kmalloc (sizeof (*head),GFP_KERNEL ); +if (!head) + goto free_now; + +INIT_LIST_HEAD (&head->list); + + item =kmalloc (sizeof (*item),GFP_KERNEL ); +if (!item) { +kfree (head); + goto free_now; + } + + item->filter = filter; +list_add_tail (&item->list, &head->list); +delay_free_filter (head); +return ; + + free_now: +/* Make sure the filter is not being used */ +tracepoint_synchronize_unregister (); +__free_filter (filter); +} +static inline void __free_subsystem_filter (struct trace_event_file *file) {__free_filter (file->filter); file->filter =NULL ; } +static inline void event_set_filter (struct trace_event_file *file, +struct event_filter *filter) +{ +rcu_assign_pointer (file->filter, filter); +} + +static inline void event_clear_filter (struct trace_event_file *file) +{ +RCU_INIT_POINTER (file->filter,NULL ); +} +static void filter_free_subsystem_filters (struct trace_subsystem_dir *dir, -struct trace_array *tr) +struct trace_array *tr, +struct event_filter *filter) {struct trace_event_file *file; +struct filter_head *head; +struct filter_list *item; + + head =kmalloc (sizeof (*head),GFP_KERNEL ); +if (!head) + goto free_now; + +INIT_LIST_HEAD (&head->list); + + item =kmalloc (sizeof (*item),GFP_KERNEL ); +if (!item) { +kfree (head); + goto free_now; + } + + item->filter = filter; +list_add_tail (&item->list, &head->list);list_for_each_entry (file, &tr->events, list) {if (file->system != dir)continue ; -__free_subsystem_filter (file); + item =kmalloc (sizeof (*item),GFP_KERNEL ); +if (!item) + goto free_now; + item->filter =event_filter (file); +list_add_tail (&item->list, &head->list); +event_clear_filter (file); + } + +delay_free_filter (head); +return ; + free_now: +tracepoint_synchronize_unregister (); + +if (head) +free_filter_list (&head->rcu); + +list_for_each_entry (file, &tr->events, list) { +if (file->system != dir || !file->filter) +continue ; +__free_filter (file->filter); } +__free_filter (filter); }int filter_assign_type (const char *type)
@@ -2120,22 +2237,6 @@ static inline void event_set_filtered_fltrace_buffered_event_enable (); } -static inline void event_set_filter (struct trace_event_file *file, -struct event_filter *filter) -{ -rcu_assign_pointer (file->filter, filter); -} - -static inline void event_clear_filter (struct trace_event_file *file) -{ -RCU_INIT_POINTER (file->filter,NULL ); -} - -struct filter_list { -struct list_head list; -struct event_filter *filter; -}; - static int process_system_preds(struct trace_subsystem_dir *dir, struct trace_array *tr, struct filter_parse_error *pe,
@@ -2144,11 +2245,16 @@ static int process_system_preds(struct tstruct trace_event_file *file;struct filter_list *filter_item;struct event_filter *filter =NULL ; -struct filter_list *tmp; -LIST_HEAD (filter_list); +struct filter_head *filter_list;bool fail = true;int err; + filter_list =kmalloc (sizeof (*filter_list),GFP_KERNEL ); +if (!filter_list) +return -ENOMEM ; + +INIT_LIST_HEAD (&filter_list->list); +list_for_each_entry (file, &tr->events, list) {if (file->system != dir)
@@ -2175,7 +2281,7 @@ static int process_system_preds(struct tif (!filter_item) goto fail_mem; -list_add_tail (&filter_item->list, &filter_list); +list_add_tail (&filter_item->list, &filter_list->list);/* * Regardless of if this returned an error, we still * replace the filter for the call.
@@ -2195,31 +2301,22 @@ static int process_system_preds(struct t* Do a synchronize_rcu() and to ensure all calls are * done with them before we free them. */ -tracepoint_synchronize_unregister (); -list_for_each_entry_safe (filter_item, tmp, &filter_list, list) { -__free_filter (filter_item->filter); -list_del (&filter_item->list); -kfree (filter_item); - } +delay_free_filter (filter_list);return 0 ; fail:/* No call succeeded */ -list_for_each_entry_safe (filter_item, tmp, &filter_list, list) { -list_del (&filter_item->list); -kfree (filter_item); - } +free_filter_list (&filter_list->rcu);parse_error (pe,FILT_ERR_BAD_SUBSYS_FILTER ,0 );return -EINVAL ; fail_mem:__free_filter (filter); +/* If any call succeeded, we still need to sync */ if (!fail) -tracepoint_synchronize_unregister (); -list_for_each_entry_safe (filter_item, tmp, &filter_list, list) { -__free_filter (filter_item->filter); -list_del (&filter_item->list); -kfree (filter_item); - } +delay_free_filter (filter_list); +else +free_filter_list (&filter_list->rcu); +return -ENOMEM ; }
@@ -2361,9 +2458,7 @@ int apply_event_filter(struct trace_evenevent_clear_filter (file); -/* Make sure the filter is not being used */ -tracepoint_synchronize_unregister (); -__free_filter (filter); +try_delay_free_filter (filter);return 0 ; }
@@ -2387,11 +2482,8 @@ int apply_event_filter(struct trace_evenevent_set_filter (file, filter); -if (tmp) { -/* Make sure the call is done with the filter */ -tracepoint_synchronize_unregister (); -__free_filter (tmp); - } +if (tmp) +try_delay_free_filter (tmp); }return err;
@@ -2417,9 +2509,7 @@ int apply_subsystem_event_filter(struct filter = system->filter; system->filter =NULL ;/* Ensure all filters are no longer used */ -tracepoint_synchronize_unregister (); -filter_free_subsystem_filters (dir, tr); -__free_filter (filter); +filter_free_subsystem_filters (dir, tr, filter);return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 462/592 smb: improve directory cache reuse for readdir operations
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Bharath SM <[email protected]>
commit 72dd7961a4bb4fa1fc456169a61dd12e68e50645 upstream.
Currently, cached directory contents were not reused across subsequent 'ls' operations because the cache validity check relied on comparing the ctx pointer, which changes with each readdir invocation. As a result, the cached dir entries was not marked as valid and the cache was not utilized for subsequent 'ls' operations.
This change uses the file pointer, which remains consistent across all readdir calls for a given directory instance, to associate and validate the cache. As a result, cached directory contents can now be correctly reused, improving performance for repeated directory listings.
Performance gains with local windows SMB server:
Without the patch and default actimeo=1:
1000 directory enumeration operations on dir with 10k files took 135.0s
With this patch and actimeo=0:
1000 directory enumeration operations on dir with 10k files took just 5.1s
--- fs/smb/client/cached_dir.h | 8 ++++---- fs/smb/client/readdir.c | 28 +++++++++++++++------------- 2 files changed, 19 insertions(+), 17 deletions(-)
fs/smb/client/cached_dir.h +4 -4
@@ -21,10 +21,10 @@ struct cached_dirent {struct cached_dirents {bool is_valid:1 ;bool is_failed:1 ; -struct dir_context *ctx;/* -* Only used to make sure we only take entries -* from a single context. Never dereferenced. -*/ +struct file *file;/* +* Used to associate the cache with a single +* open file instance. +*/ struct mutex de_mutex;int pos;/* Expected ctx->pos */ struct list_head entries;
fs/smb/client/readdir.c +15 -13
@@ -850,9 +850,9 @@ static bool emit_cached_dirents(struct c }static void update_cached_dirents_count (struct cached_dirents *cde, -struct dir_context *ctx) +struct file *file) { -if (cde->ctx != ctx) +if (cde->file != file)return ;if (cde->is_valid || cde->is_failed)return ;
@@ -861,9 +861,9 @@ static void update_cached_dirents_count( }static void finished_cached_dirents_count (struct cached_dirents *cde, -struct dir_context *ctx) +struct dir_context *ctx,struct file *file) { -if (cde->ctx != ctx) +if (cde->file != file)return ;if (cde->is_valid || cde->is_failed)return ;
@@ -876,11 +876,12 @@ static void finished_cached_dirents_counstatic void add_cached_dirent (struct cached_dirents *cde,struct dir_context *ctx,const char *name,int namelen, -struct cifs_fattr *fattr) +struct cifs_fattr *fattr, +struct file *file) {struct cached_dirent *de; -if (cde->ctx != ctx) +if (cde->file != file)return ;if (cde->is_valid || cde->is_failed)return ;
@@ -910,7 +911,8 @@ static void add_cached_dirent(struct cacstatic bool cifs_dir_emit (struct dir_context *ctx,const char *name,int namelen,struct cifs_fattr *fattr, -struct cached_fid *cfid) +struct cached_fid *cfid, +struct file *file) {bool rc;ino_t ino =cifs_uniqueid_to_ino_t (fattr->cf_uniqueid);
@@ -922,7 +924,7 @@ static bool cifs_dir_emit(struct dir_conif (cfid) {mutex_lock (&cfid->dirents.de_mutex);add_cached_dirent (&cfid->dirents, ctx, name, namelen, - fattr); + fattr, file);mutex_unlock (&cfid->dirents.de_mutex); }
@@ -1022,7 +1024,7 @@ static int cifs_filldir(char *find_entrycifs_prime_dcache (file_dentry (file), &name, &fattr);return !cifs_dir_emit (ctx, name.name, name.len, - &fattr, cfid); + &fattr, cfid, file); }
@@ -1073,8 +1075,8 @@ int cifs_readdir(struct file *file, stru* we need to initialize scanning and storing the * directory content. */ -if (ctx->pos ==0 && cfid->dirents.ctx ==NULL ) { - cfid->dirents.ctx = ctx; +if (ctx->pos ==0 && cfid->dirents.file ==NULL ) { + cfid->dirents.file = file; cfid->dirents.pos =2 ; }/*
@@ -1142,7 +1144,7 @@ int cifs_readdir(struct file *file, stru }else {if (cfid) {mutex_lock (&cfid->dirents.de_mutex); -finished_cached_dirents_count (&cfid->dirents, ctx); +finished_cached_dirents_count (&cfid->dirents, ctx, file);mutex_unlock (&cfid->dirents.de_mutex); }cifs_dbg (FYI ,"Could not find entry\n" );
@@ -1183,7 +1185,7 @@ int cifs_readdir(struct file *file, stru ctx->pos++;if (cfid) {mutex_lock (&cfid->dirents.de_mutex); -update_cached_dirents_count (&cfid->dirents, ctx); +update_cached_dirents_count (&cfid->dirents, file);mutex_unlock (&cfid->dirents.de_mutex); }
Greg Kroah-Hartman
maintainerPATCH 463/592 scsi: storvsc: Increase the timeouts to storvsc_timeout
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dexuan Cui <[email protected]>
commit b2f966568faaad326de97481096d0f3dc0971c43 upstream.
Currently storvsc_timeout is only used in storvsc_sdev_configure(), and 5s and 10s are used elsewhere. It turns out that rarely the 5s is not enough on Azure, so let's use storvsc_timeout everywhere.
In case a timeout happens and storvsc_channel_init() returns an error, close the VMBus channel so that any host-to-guest messages in the channel's ringbuffer, which might come late, can be safely ignored.
Add a "const" to storvsc_timeout.
--- drivers/scsi/storvsc_drv.c | 10 ++++++---- 1 file changed, 6 insertions(+), 4 deletions(-)
drivers/scsi/storvsc_drv.c +6 -4
@@ -362,7 +362,7 @@ MODULE_PARM_DESC(ring_avail_percent_lowa/* * Timeout in seconds for all devices managed by this driver. */ -static int storvsc_timeout =180 ; +static const int storvsc_timeout =180 ;#if IS_ENABLED (CONFIG_SCSI_FC_ATTRS )static struct scsi_transport_template *fc_transport_template;
@@ -768,7 +768,7 @@ static void handle_multichannel_storagereturn ; } - t =wait_for_completion_timeout (&request->wait_event,10 *HZ ); + t =wait_for_completion_timeout (&request->wait_event, storvsc_timeout *HZ );if (t ==0 ) {dev_err (dev,"Failed to create sub-channel: timed out\n" );return ;
@@ -833,7 +833,7 @@ static int storvsc_execute_vstor_op(struif (ret !=0 )return ret; - t =wait_for_completion_timeout (&request->wait_event,5 *HZ ); + t =wait_for_completion_timeout (&request->wait_event, storvsc_timeout *HZ );if (t ==0 )return -ETIMEDOUT ;
@@ -1350,6 +1350,8 @@ static int storvsc_connect_to_vsp(structreturn ret; ret =storvsc_channel_init (device, is_fc); +if (ret) +vmbus_close (device->channel);return ret; }
@@ -1668,7 +1670,7 @@ static int storvsc_host_reset_handler(stif (ret !=0 )return FAILED ; - t =wait_for_completion_timeout (&request->wait_event,5 *HZ ); + t =wait_for_completion_timeout (&request->wait_event, storvsc_timeout *HZ );if (t ==0 )return TIMEOUT_ERROR ;
Greg Kroah-Hartman
maintainerPATCH 464/592 scsi: s390: zfcp: Ensure synchronous unit_add
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Peter Oberparleiter <[email protected]>
commit 9697ca0d53e3db357be26d2414276143c4a2cd49 upstream.
Improve the usability of the unit_add sysfs attribute by ensuring that the associated FCP LUN scan processing is completed synchronously. This enables configuration tooling to consistently determine the end of the scan process to allow for serialization of follow-on actions.
While the scan process associated with unit_add typically completes synchronously, it is deferred to an asynchronous background process if unit_add is used before initial remote port scanning has completed. This occurs when unit_add is used immediately after setting the associated FCP device online.
To ensure synchronous unit_add processing, wait for remote port scanning to complete before initiating the FCP LUN scan.
--- drivers/s390/scsi/zfcp_sysfs.c | 2 ++ 1 file changed, 2 insertions(+)
drivers/s390/scsi/zfcp_sysfs.c +2 -0
@@ -449,6 +449,8 @@ static ssize_t zfcp_sysfs_unit_add_storeif (kstrtoull (buf,0 , (unsigned long long *) &fcp_lun))return -EINVAL ; +flush_work (&port->rport_work); + retval =zfcp_unit_add (port, fcp_lun);if (retval)return retval;
Greg Kroah-Hartman
maintainerPATCH 465/592 nvme: always punt polled uring_cmd end_io work to task_work
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Axboe <[email protected]>
commit 9ce6c9875f3e995be5fd720b65835291f8a609b1 upstream.
Currently NVMe uring_cmd completions will complete locally, if they are polled. This is done because those completions are always invoked from task context. And while that is true, there's no guarantee that it's invoked under the right ring context, or even task. If someone does NVMe passthrough via multiple threads and with a limited number of poll queues, then ringA may find completions from ringB. For that case, completing the request may not be sound.
Always just punt the passthrough completions via task_work, which will redirect the completion, if needed.
--- drivers/nvme/host/ioctl.c | 21 +++++++-------------- 1 file changed, 7 insertions(+), 14 deletions(-)
drivers/nvme/host/ioctl.c +7 -14
@@ -429,21 +429,14 @@ static enum rq_end_io_ret nvme_uring_cmd pdu->result =le64_to_cpu (nvme_req (req)->result.u64);/* -* For iopoll, complete it directly. Note that using the uring_cmd -* helper for this is safe only because we check blk_rq_is_poll(). -* As that returns false if we're NOT on a polled queue, then it's -* safe to use the polled completion helper. -* -* Otherwise, move the completion to task work. +* IOPOLL could potentially complete this request directly, but +* if multiple rings are polling on the same queue, then it's possible +* for one ring to find completions for another ring. Punting the +* completion via task_work will always direct it to the right +* location, rather than potentially complete requests for ringA +* under iopoll invocations from ringB. */ -if (blk_rq_is_poll (req)) { -if (pdu->bio) -blk_rq_unmap_user (pdu->bio); -io_uring_cmd_iopoll_done (ioucmd, pdu->result, pdu->status); - }else { -io_uring_cmd_do_in_task_lazy (ioucmd, nvme_uring_task_cb); - } - +io_uring_cmd_do_in_task_lazy (ioucmd, nvme_uring_task_cb);return RQ_END_IO_FREE ; }
Greg Kroah-Hartman
maintainerPATCH 466/592 net_sched: sch_sfq: reject invalid perturb period
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
commit 7ca52541c05c832d32b112274f81a985101f9ba8 upstream.
Gerrard Tai reported that SFQ perturb_period has no range check yet, and this can be used to trigger a race condition fixed in a separate patch.
We want to make sure ctl->perturb_period * HZ will not overflow and is positive.
Tested:
tc qd add dev lo root sfq perturb -10 # negative value : error
Error: sch_sfq: invalid perturb period.
tc qd add dev lo root sfq perturb 1000000000 # too big : error
Error: sch_sfq: invalid perturb period.
tc qd add dev lo root sfq perturb 2000000 # acceptable value
tc -s -d qd sh dev lo
qdisc sfq 8005: root refcnt 2 limit 127p quantum 64Kb depth 127 flows 128 divisor 1024 perturb 2000000sec
Sent 0 bytes 0 pkt (dropped 0, overlimits 0 requeues 0)
backlog 0b 0p requeues 0
--- net/sched/sch_sfq.c | 10 ++++++++-- 1 file changed, 8 insertions(+), 2 deletions(-)
net/sched/sch_sfq.c +8 -2
@@ -656,6 +656,14 @@ static int sfq_change(struct Qdisc *sch,NL_SET_ERR_MSG_MOD (extack,"invalid quantum" );return -EINVAL ; } + +if (ctl->perturb_period <0 || + ctl->perturb_period >INT_MAX /HZ ) { +NL_SET_ERR_MSG_MOD (extack,"invalid perturb period" ); +return -EINVAL ; + } + perturb_period = ctl->perturb_period *HZ ; +if (ctl_v1 && !red_check_params (ctl_v1->qth_min, ctl_v1->qth_max, ctl_v1->Wlog, ctl_v1->Scell_log,NULL ))return -EINVAL ;
@@ -672,14 +680,12 @@ static int sfq_change(struct Qdisc *sch, headdrop = q->headdrop; maxdepth = q->maxdepth; maxflows = q->maxflows; - perturb_period = q->perturb_period; quantum = q->quantum; flags = q->flags;/* update and validate configuration */ if (ctl->quantum) quantum = ctl->quantum; - perturb_period = ctl->perturb_period *HZ ;if (ctl->flows) maxflows =min_t (u32, ctl->flows,SFQ_MAX_FLOWS );if (ctl->divisor) {
Greg Kroah-Hartman
maintainerPATCH 467/592 net: clear the dst when changing skb protocol
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jakub Kicinski <[email protected]>
commit ba9db6f907ac02215e30128770f85fbd7db2fcf9 upstream.
A not-so-careful NAT46 BPF program can crash the kernel if it indiscriminately flips ingress packets from v4 to v6:
BUG: kernel NULL pointer dereference, address: 0000000000000000
ip6_rcv_core (net/ipv6/ip6_input.c:190:20)
ipv6_rcv (net/ipv6/ip6_input.c:306:8)
process_backlog (net/core/dev.c:6186:4)
napi_poll (net/core/dev.c:6906:9)
net_rx_action (net/core/dev.c:7028:13)
do_softirq (kernel/softirq.c:462:3)
netif_rx (net/core/dev.c:5326:3)
dev_loopback_xmit (net/core/dev.c:4015:2)
ip_mc_finish_output (net/ipv4/ip_output.c:363:8)
NF_HOOK (./include/linux/netfilter.h:314:9)
ip_mc_output (net/ipv4/ip_output.c:400:5)
dst_output (./include/net/dst.h:459:9)
ip_local_out (net/ipv4/ip_output.c:130:9)
ip_send_skb (net/ipv4/ip_output.c:1496:8)
udp_send_skb (net/ipv4/udp.c:1040:8)
udp_sendmsg (net/ipv4/udp.c:1328:10)The output interface has a 4->6 program attached at ingress. We try to loop the multicast skb back to the sending socket. Ingress BPF runs as part of netif_rx(), pushes a valid v6 hdr and changes skb->protocol to v6. We enter ip6_rcv_core which tries to use skb_dst(). But the dst is still an IPv4 one left after IPv4 mcast output.
Clear the dst in all BPF helpers which change the protocol. Try to preserve metadata dsts, those may carry non-routing metadata.
--- net/core/filter.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-)
net/core/filter.c +13 -6
@@ -3233,6 +3233,13 @@ static const struct bpf_func_proto bpf_s .arg1_type = ARG_PTR_TO_CTX, }; +static void bpf_skb_change_protocol (struct sk_buff *skb,u16 proto) +{ + skb->protocol =htons (proto); +if (skb_valid_dst (skb)) +skb_dst_drop (skb); +} +static int bpf_skb_generic_push (struct sk_buff *skb,u32 off,u32 len) {/* Caller already did skb_cow() with len as headroom,
@@ -3329,7 +3336,7 @@ static int bpf_skb_proto_4_to_6(struct s } } - skb->protocol =htons (ETH_P_IPV6 ); +bpf_skb_change_protocol (skb,ETH_P_IPV6 );skb_clear_hash (skb);return 0 ;
@@ -3359,7 +3366,7 @@ static int bpf_skb_proto_6_to_4(struct s } } - skb->protocol =htons (ETH_P_IP ); +bpf_skb_change_protocol (skb,ETH_P_IP );skb_clear_hash (skb);return 0 ;
@@ -3550,10 +3557,10 @@ static int bpf_skb_net_grow(struct sk_bu/* Match skb->protocol to new outer l3 protocol */ if (skb->protocol ==htons (ETH_P_IP ) && flags &BPF_F_ADJ_ROOM_ENCAP_L3_IPV6 ) - skb->protocol =htons (ETH_P_IPV6 ); +bpf_skb_change_protocol (skb,ETH_P_IPV6 );else if (skb->protocol ==htons (ETH_P_IPV6 ) && flags &BPF_F_ADJ_ROOM_ENCAP_L3_IPV4 ) - skb->protocol =htons (ETH_P_IP ); +bpf_skb_change_protocol (skb,ETH_P_IP ); }if (skb_is_gso (skb)) {
@@ -3606,10 +3613,10 @@ static int bpf_skb_net_shrink(struct sk_/* Match skb->protocol to new outer l3 protocol */ if (skb->protocol ==htons (ETH_P_IP ) && flags &BPF_F_ADJ_ROOM_DECAP_L3_IPV6 ) - skb->protocol =htons (ETH_P_IPV6 ); +bpf_skb_change_protocol (skb,ETH_P_IPV6 );else if (skb->protocol ==htons (ETH_P_IPV6 ) && flags &BPF_F_ADJ_ROOM_DECAP_L3_IPV4 ) - skb->protocol =htons (ETH_P_IP ); +bpf_skb_change_protocol (skb,ETH_P_IP );if (skb_is_gso (skb)) {struct skb_shared_info *shinfo =skb_shinfo (skb);
Greg Kroah-Hartman
maintainerPATCH 468/592 mm: close theoretical race where stale TLB entries could linger
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ryan Roberts <[email protected]>
commit 383c4613c67c26e90e8eebb72e3083457d02033f upstream.
Commit 3ea277194daa ("mm, mprotect: flush TLB if potentially racing with a parallel reclaim leaving stale TLB entries") described a theoretical race as such:
"""
Nadav Amit identified a theoretical race between page reclaim and mprotect due to TLB flushes being batched outside of the PTL being held.
He described the race as follows:
CPU0 CPU1 ---- ---- user accesses memory using RW PTE [PTE now cached in TLB] try_to_unmap_one() ==> ptep_get_and_clear() ==> set_tlb_ubc_flush_pending() mprotect(addr, PROT_READ) ==> change_pte_range() ==> [ PTE non-present - no flush ] user writes using cached RW PTE ... try_to_unmap_flush()
The same type of race exists for reads when protecting for PROT_NONE and also exists for operations that can leave an old TLB entry behind such as munmap, mremap and madvise.
"""
The solution was to introduce flush_tlb_batched_pending() and call it under the PTL from mprotect/madvise/munmap/mremap to complete any pending tlb flushes.
However, while madvise_free_pte_range() and madvise_cold_or_pageout_pte_range() were both retro-fitted to call flush_tlb_batched_pending() immediately after initially acquiring the PTL, they both temporarily release the PTL to split a large folio if they stumble upon one. In this case, where re-acquiring the PTL flush_tlb_batched_pending() must be called again, but it previously was not. Let's fix that.
There are 2 Fixes: tags here: the first is the commit that fixed madvise_free_pte_range(). The second is the commit that added madvise_cold_or_pageout_pte_range(), which looks like it copy/pasted the faulty pattern from madvise_free_pte_range().
This is a theoretical bug discovered during code review.
--- mm/madvise.c | 2 ++ 1 file changed, 2 insertions(+)
mm/madvise.c +2 -0
@@ -503,6 +503,7 @@ restart:pte_offset_map_lock (mm, pmd, addr, &ptl);if (!start_pte)break ; +flush_tlb_batched_pending (mm);arch_enter_lazy_mmu_mode ();if (!err) nr =0 ;
@@ -736,6 +737,7 @@ static int madvise_free_pte_range(pmd_t start_pte = pte;if (!start_pte)break ; +flush_tlb_batched_pending (mm);arch_enter_lazy_mmu_mode ();if (!err) nr =0 ;
Greg Kroah-Hartman
maintainerPATCH 469/592 udmabuf: use sgtable-based scatterlist wrappers
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Marek Szyprowski <[email protected]>
commit afe382843717d44b24ef5014d57dcbaab75a4052 upstream.
Use common wrappers operating directly on the struct sg_table objects to fix incorrect use of scatterlists sync calls. dma_sync_sg_for_*() functions have to be called with the number of elements originally passed to dma_map_sg_*() function, not the one returned in sgtable's nents.
--- drivers/dma-buf/udmabuf.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
drivers/dma-buf/udmabuf.c +2 -3
@@ -264,8 +264,7 @@ static int begin_cpu_udmabuf(struct dma_ ubuf->sg =NULL ; } }else { -dma_sync_sg_for_cpu (dev, ubuf->sg->sgl, ubuf->sg->nents, - direction); +dma_sync_sgtable_for_cpu (dev, ubuf->sg, direction); }return ret;
@@ -280,7 +279,7 @@ static int end_cpu_udmabuf(struct dma_buif (!ubuf->sg)return -EINVAL ; -dma_sync_sg_for_device (dev, ubuf->sg->sgl, ubuf->sg->nents, direction); +dma_sync_sgtable_for_device (dev, ubuf->sg, direction);return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 470/592 mm/vma: reset VMA iterator on commit_merge() OOM failure
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Lorenzo Stoakes <[email protected]>
commit 0cf4b1687a187ba9247c71721d8b064634eda1f7 upstream.
While an OOM failure in commit_merge() isn't really feasible due to the allocation which might fail (a maple tree pre-allocation) being 'too small to fail', we do need to handle this case correctly regardless.
In vma_merge_existing_range(), we can theoretically encounter failures which result in an OOM error in two ways - firstly dup_anon_vma() might fail with an OOM error, and secondly commit_merge() failing, ultimately, to pre-allocate a maple tree node.
The abort logic for dup_anon_vma() resets the VMA iterator to the initial range, ensuring that any logic looping on this iterator will correctly proceed to the next VMA.
However the commit_merge() abort logic does not do the same thing. This resulted in a syzbot report occurring because mlockall() iterates through VMAs, is tolerant of errors, but ended up with an incorrect previous VMA being specified due to incorrect iterator state.
While making this change, it became apparent we are duplicating logic - the logic introduced in commit 41e6ddcaa0f1 ("mm/vma: add give_up_on_oom option on modify/merge, use in uffd release") duplicates the vmg->give_up_on_oom check in both abort branches.
Additionally, we observe that we can perform the anon_dup check safely on dup_anon_vma() failure, as this will not be modified should this call fail.
Finally, we need to reset the iterator in both cases, so now we can simply use the exact same code to abort for both.
We remove the VM_WARN_ON(err != -ENOMEM) as it would be silly for this to be otherwise and it allows us to implement the abort check more neatly.
--- mm/vma.c | 22 ++++------------------ 1 file changed, 4 insertions(+), 18 deletions(-)
mm/vma.c +4 -18
@@ -927,26 +927,9 @@ static __must_check struct vm_area_struc err =dup_anon_vma (next, middle, &anon_dup); } -if (err) +if (err ||commit_merge (vmg)) goto abort; - err =commit_merge (vmg); -if (err) { -VM_WARN_ON (err != -ENOMEM ); - -if (anon_dup) -unlink_anon_vmas (anon_dup); - -/* -* We've cleaned up any cloned anon_vma's, no VMAs have been -* modified, no harm no foul if the user requests that we not -* report this and just give up, leaving the VMAs unmerged. -*/ -if (!vmg->give_up_on_oom) - vmg->state =VMA_MERGE_ERROR_NOMEM ; -return NULL ; - } -khugepaged_enter_vma (vmg->target, vmg->flags); vmg->state =VMA_MERGE_SUCCESS ;return vmg->target;
@@ -955,6 +938,9 @@ abort:vma_iter_set (vmg->vmi, start);vma_iter_load (vmg->vmi); +if (anon_dup) +unlink_anon_vmas (anon_dup); +/* * This means we have failed to clone anon_vma's correctly, but no * actual changes to VMAs have occurred, so no harm no foul - if the
Greg Kroah-Hartman
maintainerPATCH 471/592 x86/mm/pat: dont collapse pages without PSE set
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Juergen Gross <[email protected]>
commit 1dbf30fdb5e57fb2c39f17f35f2b544d5de34397 upstream.
Collapsing pages to a leaf PMD or PUD should be done only if X86_FEATURE_PSE is available, which is not the case when running e.g. as a Xen PV guest.
--- arch/x86/mm/pat/set_memory.c | 3 +++ 1 file changed, 3 insertions(+)
arch/x86/mm/pat/set_memory.c +3 -0
@@ -1257,6 +1257,9 @@ static int collapse_pmd_page(pmd_t *pmd, unsigned long addr,pgprot_t pgprot;int i =0 ; +if (!cpu_feature_enabled (X86_FEATURE_PSE )) +return 0 ; + addr &=PMD_MASK ; pte =pte_offset_kernel (pmd, addr); first = *pte;
Signature · 1 line
2.50.0
Greg Kroah-Hartman
maintainerPATCH 472/592 x86/Kconfig: only enable ROX cache in execmem when STRICT_MODULE_RWX is set
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mike Rapoport (Microsoft) <[email protected]>
commit 47410d839fcda6890cb82828f874f97710982f24 upstream.
Currently ROX cache in execmem is enabled regardless of STRICT_MODULE_RWX setting. This breaks an assumption that module memory is writable when STRICT_MODULE_RWX is disabled, for instance for kernel debuggin.
Only enable ROX cache in execmem when STRICT_MODULE_RWX is set to restore the original behaviour of module text permissions.
--- arch/x86/Kconfig | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
arch/x86/Kconfig +1 -1
@@ -88,7 +88,7 @@ config X86 select ARCH_HAS_DMA_OPS if GART_IOMMU || XEN select ARCH_HAS_EARLY_DEBUG if KGDB select ARCH_HAS_ELF_RANDOMIZE - select ARCH_HAS_EXECMEM_ROX if X86_64 + select ARCH_HAS_EXECMEM_ROX if X86_64 && STRICT_MODULE_RWX select ARCH_HAS_FAST_MULTIPLIER select ARCH_HAS_FORTIFY_SOURCE select ARCH_HAS_GCOV_PROFILE_ALL
Greg Kroah-Hartman
maintainerPATCH 473/592 x86/its: move its_pages array to struct mod_arch_specific
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mike Rapoport (Microsoft) <[email protected]>
commit 0b0cae7119a0ec9449d7261b5e672a5fed765068 upstream.
The of pages with ITS thunks allocated for modules are tracked by an array in 'struct module'.
Since this is very architecture specific data structure, move it to 'struct mod_arch_specific'.
No functional changes.
--- arch/x86/include/asm/module.h | 8 ++++++++ arch/x86/kernel/alternative.c | 19 ++++++++++--------- include/linux/module.h | 5 ----- 3 files changed, 18 insertions(+), 14 deletions(-)
arch/x86/include/asm/module.h +8 -0
@@ -5,12 +5,20 @@#include <asm-generic/module.h> #include <asm/orc_types.h> +struct its_array { +#ifdef CONFIG_MITIGATION_ITS +void **pages; +int num; +#endif +}; +struct mod_arch_specific {#ifdef CONFIG_UNWINDER_ORC unsigned int num_orcs;int *orc_unwind_ip;struct orc_entry *orc_unwind;#endif +struct its_array its_pages; };#endif /* _ASM_X86_MODULE_H */
arch/x86/kernel/alternative.c +10 -9
@@ -195,8 +195,8 @@ void its_fini_mod(struct module *mod) its_page =NULL ;mutex_unlock (&text_mutex); -for (int i =0 ; i < mod->its_num_pages; i++) { -void *page = mod->its_page_array[i]; +for (int i =0 ; i < mod->arch.its_pages.num; i++) { +void *page = mod->arch.its_pages.pages[i];execmem_restore_rox (page,PAGE_SIZE ); } }
@@ -206,11 +206,11 @@ void its_free_mod(struct module *mod)if (!cpu_feature_enabled (X86_FEATURE_INDIRECT_THUNK_ITS ))return ; -for (int i =0 ; i < mod->its_num_pages; i++) { -void *page = mod->its_page_array[i]; +for (int i =0 ; i < mod->arch.its_pages.num; i++) { +void *page = mod->arch.its_pages.pages[i];execmem_free (page); } - kfree(mod->its_page_array); + kfree(mod->arch.its_pages.pages); }#endif /* CONFIG_MODULES */
@@ -223,14 +223,15 @@ static void *its_alloc(void)#ifdef CONFIG_MODULES if (its_mod) { -void *tmp =krealloc (its_mod->its_page_array, - (its_mod->its_num_pages+1 ) *sizeof (void *), +struct its_array *pages = &its_mod->arch.its_pages; +void *tmp =krealloc (pages->pages, + (pages->num+1 ) *sizeof (void *),GFP_KERNEL );if (!tmp)return NULL ; - its_mod->its_page_array = tmp; - its_mod->its_page_array[its_mod->its_num_pages++] = page; + pages->pages = tmp; + pages->pages[pages->num++] = page;execmem_make_temp_rw (page,PAGE_SIZE ); }
include/linux/module.h +0 -5
@@ -586,11 +586,6 @@ struct module {atomic_t refcnt;#endif -#ifdef CONFIG_MITIGATION_ITS -int its_num_pages; -void **its_page_array; -#endif -#ifdef CONFIG_CONSTRUCTORS /* Constructor functions. */ ctor_fn_t *ctors;
Greg Kroah-Hartman
maintainerPATCH 474/592 x86/its: explicitly manage permissions for ITS pages
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Peter Zijlstra (Intel) <[email protected]>
commit a82b26451de126a5ae130361081986bc459afe9b upstream.
execmem_alloc() sets permissions differently depending on the kernel configuration, CPU support for PSE and whether a page is allocated before or after mark_rodata_ro().
Add tracking for pages allocated for ITS when patching the core kernel and make sure the permissions for ITS pages are explicitly managed for both kernel and module allocations.
--- arch/x86/kernel/alternative.c | 74 +++++++++++++++++++++++++++++------------- 1 file changed, 52 insertions(+), 22 deletions(-)
arch/x86/kernel/alternative.c +52 -22
@@ -138,6 +138,24 @@ static struct module *its_mod;#endif static void *its_page;static unsigned int its_offset; +struct its_array its_pages; + +static void *__its_alloc (struct its_array *pages) +{ + void *page __free(execmem) = execmem_alloc(EXECMEM_MODULE_TEXT, PAGE_SIZE); +if (!page) +return NULL ; + +void *tmp =krealloc (pages->pages, (pages->num+1 ) *sizeof (void *), +GFP_KERNEL ); +if (!tmp) +return NULL ; + + pages->pages = tmp; + pages->pages[pages->num++] = page; + +return no_free_ptr (page); +}/* Initialize a thunk with the "jmp *reg; int3" instructions. */ static void *its_init_thunk (void *thunk,int reg)
@@ -173,6 +191,21 @@ static void *its_init_thunk(void *thunk,return thunk + offset; } +static void its_pages_protect (struct its_array *pages) +{ +for (int i =0 ; i < pages->num; i++) { +void *page = pages->pages[i]; +execmem_restore_rox (page,PAGE_SIZE ); + } +} + +static void its_fini_core (void ) +{ +if (IS_ENABLED (CONFIG_STRICT_KERNEL_RWX )) +its_pages_protect (&its_pages); +kfree (its_pages.pages); +} +#ifdef CONFIG_MODULES void its_init_mod (struct module *mod) {
@@ -195,10 +228,8 @@ void its_fini_mod(struct module *mod) its_page =NULL ;mutex_unlock (&text_mutex); -for (int i =0 ; i < mod->arch.its_pages.num; i++) { -void *page = mod->arch.its_pages.pages[i]; -execmem_restore_rox (page,PAGE_SIZE ); - } +if (IS_ENABLED (CONFIG_STRICT_MODULE_RWX )) +its_pages_protect (&mod->arch.its_pages); }void its_free_mod (struct module *mod)
@@ -216,28 +247,23 @@ void its_free_mod(struct module *mod)static void *its_alloc (void ) { - void *page __free(execmem) = execmem_alloc(EXECMEM_MODULE_TEXT, PAGE_SIZE); +struct its_array *pages = &its_pages; +void *page; +#ifdef CONFIG_MODULE +if (its_mod) + pages = &its_mod->arch.its_pages; +#endif + + page =__its_alloc (pages);if (!page)return NULL ; -#ifdef CONFIG_MODULES -if (its_mod) { -struct its_array *pages = &its_mod->arch.its_pages; -void *tmp =krealloc (pages->pages, - (pages->num+1 ) *sizeof (void *), -GFP_KERNEL ); -if (!tmp) -return NULL ; - - pages->pages = tmp; - pages->pages[pages->num++] = page; +execmem_make_temp_rw (page,PAGE_SIZE ); +if (pages == &its_pages) +set_memory_x ((unsigned long )page,1 ); -execmem_make_temp_rw (page,PAGE_SIZE ); - } -#endif /* CONFIG_MODULES */ - -return no_free_ptr (page); +return page; }static void *its_allocate_thunk (int reg)
@@ -291,7 +317,9 @@ u8 *its_static_thunk(int reg) return thunk; } -#endif +#else +static inline void its_fini_core (void ) {} +#endif /* CONFIG_MITIGATION_ITS */ /* * Nomenclature for variable names to simplify and clarify this code and ease
@@ -2368,6 +2396,8 @@ void __init alternative_instructions(voiapply_retpolines (__retpoline_sites, __retpoline_sites_end);apply_returns (__return_sites, __return_sites_end); +its_fini_core (); +/* * Adjust all CALL instructions to point to func()-10, including * those in .altinstr_replacement.
Greg Kroah-Hartman
maintainerPATCH 475/592 Revert "mm/execmem: Unify early execmem_cache behaviour"
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mike Rapoport (Microsoft) <[email protected]>
commit 7cd9a11dd0c3d1dd225795ed1b5b53132888e7b5 upstream.
The commit d6d1e3e6580c ("mm/execmem: Unify early execmem_cache behaviour") changed early behaviour of execemem ROX cache to allow its usage in early x86 code that allocates text pages when CONFIG_MITGATION_ITS is enabled.
The permission management of the pages allocated from execmem for ITS mitigation is now completely contained in arch/x86/kernel/alternatives.c and therefore there is no need to special case early allocations in execmem.
This reverts commit d6d1e3e6580ca35071ad474381f053cbf1fb6414.
--- arch/x86/mm/init_32.c | 3 --- arch/x86/mm/init_64.c | 3 --- include/linux/execmem.h | 8 +------- mm/execmem.c | 40 +++------------------------------------- 4 files changed, 4 insertions(+), 50 deletions(-)
arch/x86/mm/init_32.c +0 -3
@@ -30,7 +30,6 @@#include <linux/initrd.h> #include <linux/cpumask.h> #include <linux/gfp.h> -#include <linux/execmem.h> #include <asm/asm.h> #include <asm/bios_ebda.h>
@@ -756,8 +755,6 @@ void mark_rodata_ro(void)pr_info ("Write protecting kernel text and read-only data: %luk\n" , size >>10 ); -execmem_cache_make_ro (); - kernel_set_to_readonly =1 ;#ifdef CONFIG_CPA_DEBUG
arch/x86/mm/init_64.c +0 -3
@@ -34,7 +34,6 @@#include <linux/gfp.h> #include <linux/kcore.h> #include <linux/bootmem_info.h> -#include <linux/execmem.h> #include <asm/processor.h> #include <asm/bios_ebda.h>
@@ -1392,8 +1391,6 @@ void mark_rodata_ro(void) (end - start) >> 10);set_memory_ro (start, (end - start) >>PAGE_SHIFT ); -execmem_cache_make_ro (); - kernel_set_to_readonly =1 ;/*
include/linux/execmem.h +1 -7
@@ -54,7 +54,7 @@ enum execmem_range_flags { EXECMEM_ROX_CACHE = (1 << 1), }; -#if defined(CONFIG_ARCH_HAS_EXECMEM_ROX ) && defined(CONFIG_EXECMEM ) +#ifdef CONFIG_ARCH_HAS_EXECMEM_ROX /** * execmem_fill_trapping_insns - set memory to contain instructions that * will trap
@@ -94,15 +94,9 @@ int execmem_make_temp_rw(void *ptr, size* Return: 0 on success or negative error code on failure. */ int execmem_restore_rox (void *ptr,size_t size); - -/* -* Called from mark_readonly(), where the system transitions to ROX. -*/ -void execmem_cache_make_ro (void );#else static inline int execmem_make_temp_rw (void *ptr,size_t size) {return 0 ; }static inline int execmem_restore_rox (void *ptr,size_t size) {return 0 ; } -static inline void execmem_cache_make_ro (void ) { }#endif /**
mm/execmem.c +3 -37
@@ -254,34 +254,6 @@ out_unlock:return ptr; } -static bool execmem_cache_rox = false; - -void execmem_cache_make_ro (void ) -{ -struct maple_tree *free_areas = &execmem_cache.free_areas; -struct maple_tree *busy_areas = &execmem_cache.busy_areas; -MA_STATE (mas_free, free_areas,0 ,ULONG_MAX ); -MA_STATE (mas_busy, busy_areas,0 ,ULONG_MAX ); -struct mutex *mutex = &execmem_cache.mutex; -void *area; - - execmem_cache_rox = true; - -mutex_lock (mutex); - -mas_for_each (&mas_free, area,ULONG_MAX ) { -unsigned long pages =mas_range_len (&mas_free) >>PAGE_SHIFT ; -set_memory_ro (mas_free.index, pages); - } - -mas_for_each (&mas_busy, area,ULONG_MAX ) { -unsigned long pages =mas_range_len (&mas_busy) >>PAGE_SHIFT ; -set_memory_ro (mas_busy.index, pages); - } - -mutex_unlock (mutex); -} -static int execmem_cache_populate (struct execmem_range *range,size_t size) {unsigned long vm_flags =VM_ALLOW_HUGE_VMAP ;
@@ -302,15 +274,9 @@ static int execmem_cache_populate(struct/* fill memory with instructions that will trap */ execmem_fill_trapping_insns (p, alloc_size,/* writable = */ true); -if (execmem_cache_rox) { - err =set_memory_rox ((unsigned long )p, vm->nr_pages); -if (err) - goto err_free_mem; - }else { - err =set_memory_x ((unsigned long )p, vm->nr_pages); -if (err) - goto err_free_mem; - } + err =set_memory_rox ((unsigned long )p, vm->nr_pages); +if (err) + goto err_free_mem; err =execmem_cache_add (p, alloc_size);if (err)
Greg Kroah-Hartman
maintainerPATCH 476/592 x86/virt/tdx: Avoid indirect calls to TDX assembly functions
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kai Huang <[email protected]>
commit 0b3bc018e86afdc0cbfef61328c63d5c08f8b370 upstream.
Two 'static inline' TDX helper functions (sc_retry() and sc_retry_prerr()) take function pointer arguments which refer to assembly functions. Normally, the compiler inlines the TDX helper, realizes that the function pointer targets are completely static -- thus can be resolved at compile time -- and generates direct call instructions.
But, other times (like when CONFIG_CC_OPTIMIZE_FOR_SIZE=y), the compiler declines to inline the helpers and will instead generate indirect call instructions.
Indirect calls to assembly functions require special annotation (for various Control Flow Integrity mechanisms). But TDX assembly functions lack the special annotations and can only be called directly.
Annotate both the helpers as '__always_inline' to prod the compiler into maintaining the direct calls. There is no guarantee here, but Peter has volunteered to report the compiler bug if this assumption ever breaks[1].
--- arch/x86/include/asm/tdx.h | 2 +- arch/x86/virt/vmx/tdx/tdx.c | 5 +++-- 2 files changed, 4 insertions(+), 3 deletions(-)
arch/x86/include/asm/tdx.h +1 -1
@@ -100,7 +100,7 @@ void tdx_init(void);typedef u64 (*sc_func_t )(u64 fn,struct tdx_module_args *args); -static inline u64 sc_retry (sc_func_t func,u64 fn, +static __always_inline u64 sc_retry(sc_func_t func, u64 fn,struct tdx_module_args *args) {int retry =RDRAND_RETRY_LOOPS ;
arch/x86/virt/vmx/tdx/tdx.c +3 -2
@@ -69,8 +69,9 @@ static inline void seamcall_err_ret(u64 args->r9, args->r10, args->r11); } -static inline int sc_retry_prerr (sc_func_t func,sc_err_func_t err_func, -u64 fn,struct tdx_module_args *args) +static __always_inline int sc_retry_prerr(sc_func_t func, +sc_err_func_t err_func, +u64 fn,struct tdx_module_args *args) {u64 sret =sc_retry (func, fn, args);
Greg Kroah-Hartman
maintainerPATCH 477/592 selftests/x86: Add a test to detect infinite SIGTRAP handler loop
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Xin Li (Intel) <[email protected]>
commit f287822688eeb44ae1cf6ac45701d965efc33218 upstream.
When FRED is enabled, if the Trap Flag (TF) is set without an external debugger attached, it can lead to an infinite loop in the SIGTRAP handler. To avoid this, the software event flag in the augmented SS must be cleared, ensuring that no single-step trap remains pending when ERETU completes.
This test checks for that specific scenario—verifying whether the kernel correctly prevents an infinite SIGTRAP loop in this edge case when FRED is enabled.
The test should _always_ pass with IDT event delivery, thus no need to disable the test even when FRED is not enabled.
--- tools/testing/selftests/x86/Makefile | 2 tools/testing/selftests/x86/sigtrap_loop.c | 101 +++++++++++++++++++++++++++++ 2 files changed, 102 insertions(+), 1 deletion(-) create mode 100644 tools/testing/selftests/x86/sigtrap_loop.c
tools/testing/selftests/x86/Makefile +1 -1
@@ -12,7 +12,7 @@ CAN_BUILD_WITH_NOPIE := $(shell ./check_ TARGETS_C_BOTHBITS := single_step_syscall sysret_ss_attrs syscall_nt test_mremap_vdso \ check_initial_reg_state sigreturn iopl ioperm \ - test_vsyscall mov_ss_trap \ + test_vsyscall mov_ss_trap sigtrap_loop \ syscall_arg_fault fsgsbase_restore sigaltstack TARGETS_C_BOTHBITS += nx_stack TARGETS_C_32BIT_ONLY := entry_from_vm86 test_syscall_vdso unwind_vdso \
tools/testing/selftests/x86/sigtrap_loop.c new file +101 -0
101 lines not on this page · raw message
Greg Kroah-Hartman
maintainerPATCH 478/592 ksmbd: fix null pointer dereference in destroy_previous_session
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Namjae Jeon <[email protected]>
commit 7ac5b66acafcc9292fb935d7e03790f2b8b2dc0e upstream.
If client set ->PreviousSessionId on kerberos session setup stage, NULL pointer dereference error will happen. Since sess->user is not set yet, It can pass the user argument as NULL to destroy_previous_session. sess->user will be set in ksmbd_krb5_authenticate(). So this patch move calling destroy_previous_session() after ksmbd_krb5_authenticate().
--- fs/smb/server/smb2pdu.c | 11 ++++++----- 1 file changed, 6 insertions(+), 5 deletions(-)
fs/smb/server/smb2pdu.c +6 -5
@@ -1607,17 +1607,18 @@ static int krb5_authenticate(struct ksmb out_len = work->response_sz - (le16_to_cpu (rsp->SecurityBufferOffset) +4 ); -/* Check previous session */ - prev_sess_id =le64_to_cpu (req->PreviousSessionId); -if (prev_sess_id && prev_sess_id != sess->id) -destroy_previous_session (conn, sess->user, prev_sess_id); - retval =ksmbd_krb5_authenticate (sess, in_blob, in_len, out_blob, &out_len);if (retval) {ksmbd_debug (SMB ,"krb5 authentication failed\n" );return -EINVAL ; } + +/* Check previous session */ + prev_sess_id =le64_to_cpu (req->PreviousSessionId); +if (prev_sess_id && prev_sess_id != sess->id) +destroy_previous_session (conn, sess->user, prev_sess_id); + rsp->SecurityBufferLength =cpu_to_le16 (out_len); if ((conn->sign || server_conf.enforced_signing) ||
Greg Kroah-Hartman
maintainerPATCH 479/592 fgraph: Do not enable function_graph tracer when setting funcgraph-args
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Steven Rostedt <[email protected]>
commit 327e28664307d49ce3fa71ba30dcc0007c270974 upstream.
When setting the funcgraph-args option when function graph tracer is net enabled, it incorrectly enables it. Worse, it unregisters itself when it was never registered. Then when it gets enabled again, it will register itself a second time causing a WARNing.
~# echo 1 > /sys/kernel/tracing/options/funcgraph-args ~# head -20 /sys/kernel/tracing/trace
# tracer: nop
#
# entries-in-buffer/entries-written: 813/26317372 #P:8
#
# _-----=> irqs-off/BH-disabled
# / _----=> need-resched
# | / _---=> hardirq/softirq
# || / _--=> preempt-depth
# ||| / _-=> migrate-disable
# |||| / delay
# TASK-PID CPU# ||||| TIMESTAMP FUNCTION
# | | | ||||| | |
<idle>-0 [007] d..4. 358.966010: 7) 1.692 us | fetch_next_timer_interrupt(basej=4294981640, basem=357956000000, base_local=0xffff88823c3ae040, base_global=0xffff88823c3af300, tevt=0xffff888100e47cb8);
<idle>-0 [007] d..4. 358.966012: 7) | tmigr_cpu_deactivate(nextexp=357988000000) {
<idle>-0 [007] d..4. 358.966013: 7) | _raw_spin_lock(lock=0xffff88823c3b2320) {
<idle>-0 [007] d..4. 358.966014: 7) 0.981 us | preempt_count_add(val=1);
<idle>-0 [007] d..5. 358.966017: 7) 1.058 us | do_raw_spin_lock(lock=0xffff88823c3b2320);
<idle>-0 [007] d..4. 358.966019: 7) 5.824 us | }
<idle>-0 [007] d..5. 358.966021: 7) | tmigr_inactive_up(group=0xffff888100cb9000, child=0x0, data=0xffff888100e47bc0) {
<idle>-0 [007] d..5. 358.966022: 7) | tmigr_update_events(group=0xffff888100cb9000, child=0x0, data=0xffff888100e47bc0) {Notice the "tracer: nop" at the top there. The current tracer is the "nop" tracer, but the content is obviously the function graph tracer.
Enabling function graph tracing will cause it to register again and trigger a warning in the accounting:
~# echo function_graph > /sys/kernel/tracing/current_tracer -bash: echo: write error: Device or resource busy
With the dmesg of:
------------[ cut here ]------------ WARNING: CPU: 7 PID: 1095 at kernel/trace/ftrace.c:3509 ftrace_startup_subops+0xc1e/0x1000 Modules linked in: kvm_intel kvm irqbypass CPU: 7 UID: 0 PID: 1095 Comm: bash Not tainted 6.16.0-rc2-test-00006-gea03de4105d3 #24 PREEMPT Hardware name: QEMU Standard PC (Q35 + ICH9, 2009), BIOS 1.16.3-debian-1.16.3-2 04/01/2014 RIP: 0010:ftrace_startup_subops+0xc1e/0x1000 Code: 48 b8 22 01 00 00 00 00 ad de 49 89 84 24 88 01 00 00 8b 44 24 08 89 04 24 e9 c3 f7 ff ff c7 04 24 ed ff ff ff e9 b7 f7 ff ff <0f> 0b c7 04 24 f0 ff ff ff e9 a9 f7 ff ff c7 04 24 f4 ff ff ff e9 RSP: 0018:ffff888133cff948 EFLAGS: 00010202 RAX: 0000000000000001 RBX: 1ffff1102679ff31 RCX: 0000000000000000 RDX: 1ffffffff0b27a60 RSI: ffffffff8593d2f0 RDI: ffffffff85941140 RBP: 00000000000c2041 R08: ffffffffffffffff R09: ffffed1020240221 R10: ffff88810120110f R11: ffffed1020240214 R12: ffffffff8593d2f0 R13: ffffffff8593d300 R14: ffffffff85941140 R15: ffffffff85631100 FS: 00007f7ec6f28740(0000) GS:ffff8882b5251000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f7ec6f181c0 CR3: 000000012f1d0005 CR4: 0000000000172ef0 Call Trace: <TASK> ? __pfx_ftrace_startup_subops+0x10/0x10 ? find_held_lock+0x2b/0x80 ? ftrace_stub_direct_tramp+0x10/0x10 ? ftrace_stub_direct_tramp+0x10/0x10 ? trace_preempt_on+0xd0/0x110 ? __pfx_trace_graph_entry_args+0x10/0x10 register_ftrace_graph+0x4d2/0x1020 ? tracing_reset_online_cpus+0x14b/0x1e0 ? __pfx_register_ftrace_graph+0x10/0x10 ? ring_buffer_record_enable+0x16/0x20 ? tracing_reset_online_cpus+0x153/0x1e0 ? __pfx_tracing_reset_online_cpus+0x10/0x10 ? __pfx_trace_graph_return+0x10/0x10 graph_trace_init+0xfd/0x160 tracing_set_tracer+0x500/0xa80 ? __pfx_tracing_set_tracer+0x10/0x10 ? lock_release+0x181/0x2d0 ? _copy_from_user+0x26/0xa0 tracing_set_trace_write+0x132/0x1e0 ? __pfx_tracing_set_trace_write+0x10/0x10 ? ftrace_graph_func+0xcc/0x140 ? ftrace_stub_direct_tramp+0x10/0x10 ? ftrace_stub_direct_tramp+0x10/0x10 ? ftrace_stub_direct_tramp+0x10/0x10 vfs_write+0x1d0/0xe90 ? __pfx_vfs_write+0x10/0x10
Have the setting of the funcgraph-args check if function_graph tracer is the current tracer of the instance, and if not, do nothing, as there's nothing to do (the option is checked when function_graph tracing starts).
--- kernel/trace/trace_functions_graph.c | 6 ++++++ 1 file changed, 6 insertions(+)
kernel/trace/trace_functions_graph.c +6 -0
@@ -475,10 +475,16 @@ static int graph_trace_init(struct tracereturn 0 ; } +static struct tracer graph_trace; +static int ftrace_graph_trace_args (struct trace_array *tr,int set) {trace_func_graph_ent_t entry; +/* Do nothing if the current tracer is not this tracer */ +if (tr->current_trace != &graph_trace) +return 0 ; +if (set) entry = trace_graph_entry_args;else
Greg Kroah-Hartman
maintainerPATCH 480/592 platform/x86: ideapad-laptop: use usleep_range() for EC polling
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Rong Zhang <[email protected]>
commit 5808c34216954cd832bd4b8bc52dfa287049122b upstream.
It was reported that ideapad-laptop sometimes causes some recent (since 2024) Lenovo ThinkBook models shut down when:
- suspending/resuming
- closing/opening the lid
- (dis)connecting a charger
- reading/writing some sysfs properties, e.g., fan_mode, touchpad
- pressing down some Fn keys, e.g., Brightness Up/Down (Fn+F5/F6)
- (seldom) loading the kmod
The issue has existed since the launch day of such models, and there have been some out-of-tree workarounds (see Link:) for the issue. One disables some functionalities, while another one simply shortens IDEAPAD_EC_TIMEOUT. The disabled functionalities have read_ec_data() in their call chains, which calls schedule() between each poll.
It turns out that these models suffer from the indeterminacy of schedule() because of their low tolerance for being polled too frequently. Sometimes schedule() returns too soon due to the lack of ready tasks, causing the margin between two polls to be too short. In this case, the command is somehow aborted, and too many subsequent polls (they poll for "nothing!") may eventually break the state machine in the EC, resulting in a hard shutdown. This explains why shortening IDEAPAD_EC_TIMEOUT works around the issue - it reduces the total number of polls sent to the EC.
Even when it doesn't lead to a shutdown, frequent polls may also disturb the ongoing operation and notably delay (+ 10-20ms) the availability of EC response. This phenomenon is unlikely to be exclusive to the models mentioned above, so dropping the schedule() manner should also slightly improve the responsiveness of various models.
Fix these issues by migrating to usleep_range(150, 300). The interval is chosen to add some margin to the minimal 50us and considering EC responses are usually available after 150-2500us based on my test. It should be enough to fix these issues on all models subject to the EC bug without introducing latency on other models.
Tested on ThinkBook 14 G7+ ASP and solved both issues. No regression was introduced in the test on a model without the EC bug (ThinkBook X IMH, thanks Eric).
--- drivers/platform/x86/ideapad-laptop.c | 19 +++++++++++++++++-- 1 file changed, 17 insertions(+), 2 deletions(-)
drivers/platform/x86/ideapad-laptop.c +17 -2
@@ -15,6 +15,7 @@#include <linux/bug.h> #include <linux/cleanup.h> #include <linux/debugfs.h> +#include <linux/delay.h> #include <linux/device.h> #include <linux/dmi.h> #include <linux/i8042.h>
@@ -267,6 +268,20 @@ static void ideapad_shared_exit(struct i*/ #define IDEAPAD_EC_TIMEOUT 200/* in ms */ +/* +* Some models (e.g., ThinkBook since 2024) have a low tolerance for being +* polled too frequently. Doing so may break the state machine in the EC, +* resulting in a hard shutdown. +* +* It is also observed that frequent polls may disturb the ongoing operation +* and notably delay the availability of EC response. +* +* These values are used as the delay before the first poll and the interval +* between subsequent polls to solve the above issues. +*/ +#define IDEAPAD_EC_POLL_MIN_US 150 +#define IDEAPAD_EC_POLL_MAX_US 300 +static int eval_int (acpi_handle handle,const char *name,unsigned long *res) {unsigned long long result;
@@ -383,7 +398,7 @@ static int read_ec_data(acpi_handle hand end_jiffies = jiffies +msecs_to_jiffies (IDEAPAD_EC_TIMEOUT ) +1 ;while (time_before (jiffies, end_jiffies)) { -schedule (); +usleep_range (IDEAPAD_EC_POLL_MIN_US ,IDEAPAD_EC_POLL_MAX_US ); err =eval_vpcr (handle,1 , &val);if (err)
@@ -414,7 +429,7 @@ static int write_ec_cmd(acpi_handle hand end_jiffies = jiffies +msecs_to_jiffies (IDEAPAD_EC_TIMEOUT ) +1 ;while (time_before (jiffies, end_jiffies)) { -schedule (); +usleep_range (IDEAPAD_EC_POLL_MIN_US ,IDEAPAD_EC_POLL_MAX_US ); err =eval_vpcr (handle,1 , &val);if (err)
Greg Kroah-Hartman
maintainerPATCH 481/592 Revert "platform/x86: alienware-wmi-wmax: Add G-Mode support to Alienware m16 R1"
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kurt Borja <[email protected]>
commit e2468dc700743683e1d1793bbd855e2536fd3de2 upstream.
This reverts commit 5ff79cabb23a2f14d2ed29e9596aec908905a0e6.
Although the Alienware m16 R1 AMD model supports G-Mode, it actually has a lower power ceiling than plain "performance" profile, which results in lower performance.
--- drivers/platform/x86/dell/alienware-wmi-wmax.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/platform/x86/dell/alienware-wmi-wmax.c +1 -1
@@ -91,7 +91,7 @@ static const struct dmi_system_id awcc_dDMI_MATCH (DMI_SYS_VENDOR ,"Alienware" ), DMI_MATCH(DMI_PRODUCT_NAME, "Alienware m16 R1 AMD"), }, - .driver_data = &g_series_quirks, + .driver_data = &generic_quirks, }, { .ident = "Alienware m16 R2",
Greg Kroah-Hartman
maintainerPATCH 482/592 selinux: fix selinux_xfrm_alloc_user() to set correct ctx_len
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Stephen Smalley <[email protected]>
commit 86c8db86af43f52f682e53a0f2f0828683be1e52 upstream.
We should count the terminating NUL byte as part of the ctx_len. Otherwise, UBSAN logs a warning:
UBSAN: array-index-out-of-bounds in security/selinux/xfrm.c:99:14 index 60 is out of range for type 'char [*]'
The allocation itself is correct so there is no actual out of bounds indexing, just a warning.
--- security/selinux/xfrm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
security/selinux/xfrm.c +1 -1
@@ -94,7 +94,7 @@ static int selinux_xfrm_alloc_user(struc ctx->ctx_doi =XFRM_SC_DOI_LSM ; ctx->ctx_alg =XFRM_SC_ALG_SELINUX ; - ctx->ctx_len = str_len; + ctx->ctx_len = str_len +1 ;memcpy (ctx->ctx_str, &uctx[1 ], str_len); ctx->ctx_str[str_len] ='\0' ; rc = security_context_to_sid(ctx->ctx_str, str_len,
Greg Kroah-Hartman
maintainerPATCH 483/592 platform/x86/intel-uncore-freq: Fail module load when plat_info is NULL
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Srinivas Pandruvada <[email protected]>
commit 685f88c72a0c4d12d3bd2ff50286938f14486f85 upstream.
Address a Smatch static checker warning regarding an unchecked dereference in the function call:set_cdie_id(i, cluster_info, plat_info)
when plat_info is NULL.
Instead of addressing this one case, in general if plat_info is NULL then it can cause other issues. For example in a two package system it will give warning for duplicate sysfs entry as package ID will be always zero for both packages when creating string for attribute group name.
plat_info is derived from TPMI ID TPMI_BUS_INFO, which is integral to the core TPMI design. Therefore, it should not be NULL on a production platform. Consequently, the module should fail to load if plat_info is NULL.
--- drivers/platform/x86/intel/uncore-frequency/uncore-frequency-tpmi.c | 9 ++++++--- 1 file changed, 6 insertions(+), 3 deletions(-)
drivers/platform/x86/intel/uncore-frequency/uncore-frequency-tpmi.c +6 -3
@@ -467,10 +467,13 @@ static int uncore_probe(struct auxiliary/* Get the package ID from the TPMI core */ plat_info =tpmi_get_platform_data (auxdev); -if (plat_info) - pkg = plat_info->package_id; -else +if (unlikely (!plat_info)) {dev_info (&auxdev->dev,"Platform information is NULL\n" ); + ret = -ENODEV ; + goto err_rem_common; + } + + pkg = plat_info->package_id;for (i =0 ; i < num_resources; ++i) {struct tpmi_uncore_power_domain_info *pd_info;
Greg Kroah-Hartman
maintainerPATCH 484/592 sched_ext, sched/core: Dont call scx_group_set_weight() prematurely from sched_create_group()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tejun Heo <[email protected]>
commit 33796b91871ad4010c8188372dd1faf97cf0f1c0 upstream.
During task_group creation, sched_create_group() calls scx_group_set_weight() with CGROUP_WEIGHT_DFL to initialize the sched_ext portion. This is premature and ends up calling ops.cgroup_set_weight() with an incorrect @cgrp before ops.cgroup_init() is called.
sched_create_group() should just initialize SCX related fields in the new task_group. Fix it by factoring out scx_tg_init() from sched_init() and making sched_create_group() call that function instead of scx_group_set_weight().
v2: Retain CONFIG_EXT_GROUP_SCHED ifdef in sched_init() as removing it leads
to build failures on !CONFIG_GROUP_SCHED configs.
--- kernel/sched/core.c | 4 ++-- kernel/sched/ext.c | 5 +++++ kernel/sched/ext.h | 2 ++ 3 files changed, 9 insertions(+), 2 deletions(-)
kernel/sched/core.c +2 -2
@@ -8571,7 +8571,7 @@ void __init sched_init(void)init_cfs_bandwidth (&root_task_group.cfs_bandwidth,NULL );#endif /* CONFIG_FAIR_GROUP_SCHED */ #ifdef CONFIG_EXT_GROUP_SCHED - root_task_group.scx_weight =CGROUP_WEIGHT_DFL ; +scx_tg_init (&root_task_group);#endif /* CONFIG_EXT_GROUP_SCHED */ #ifdef CONFIG_RT_GROUP_SCHED root_task_group.rt_se = (struct sched_rt_entity **)ptr;
@@ -9011,7 +9011,7 @@ struct task_group *sched_create_group(stif (!alloc_rt_sched_group (tg, parent)) goto err; -scx_group_set_weight (tg,CGROUP_WEIGHT_DFL ); +scx_tg_init (tg);alloc_uclamp_sched_group (tg, parent);return tg;
kernel/sched/ext.c +5 -0
@@ -3936,6 +3936,11 @@ bool scx_can_stop_tick(struct rq *rq)DEFINE_STATIC_PERCPU_RWSEM (scx_cgroup_rwsem);static bool scx_cgroup_enabled; +void scx_tg_init (struct task_group *tg) +{ + tg->scx_weight =CGROUP_WEIGHT_DFL ; +} +int scx_tg_online (struct task_group *tg) {int ret =0 ;
kernel/sched/ext.h +2 -0
@@ -80,6 +80,7 @@ static inline void scx_update_idle(struc#ifdef CONFIG_CGROUP_SCHED #ifdef CONFIG_EXT_GROUP_SCHED +void scx_tg_init (struct task_group *tg);int scx_tg_online (struct task_group *tg);void scx_tg_offline (struct task_group *tg);int scx_cgroup_can_attach (struct cgroup_taskset *tset);
@@ -89,6 +90,7 @@ void scx_cgroup_cancel_attach(struct cgrvoid scx_group_set_weight (struct task_group *tg,unsigned long cgrp_weight);void scx_group_set_idle (struct task_group *tg,bool idle);#else /* CONFIG_EXT_GROUP_SCHED */ +static inline void scx_tg_init (struct task_group *tg) {}static inline int scx_tg_online (struct task_group *tg) {return 0 ; }static inline void scx_tg_offline (struct task_group *tg) {}static inline int scx_cgroup_can_attach (struct cgroup_taskset *tset) {return 0 ; }
Greg Kroah-Hartman
maintainerPATCH 485/592 atm: Revert atm_account_tx() if copy_from_iter_full() fails.
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kuniyuki Iwashima <[email protected]>
commit 7851263998d4269125fd6cb3fdbfc7c6db853859 upstream.
In vcc_sendmsg(), we account skb->truesize to sk->sk_wmem_alloc by atm_account_tx().
It is expected to be reverted by atm_pop_raw() later called by vcc->dev->ops->send(vcc, skb).
However, vcc_sendmsg() misses the same revert when copy_from_iter_full() fails, and then we will leak a socket.
Let's factorise the revert part as atm_return_tx() and call it in the failure path.
Note that the corresponding sk_wmem_alloc operation can be found in alloc_tx() as of the blamed commit.
$ git blame -L:alloc_tx net/atm/common.c c55fa3cccbc2c~
--- include/linux/atmdev.h | 6 ++++++ net/atm/common.c | 1 + net/atm/raw.c | 2 +- 3 files changed, 8 insertions(+), 1 deletion(-)
include/linux/atmdev.h +6 -0
@@ -249,6 +249,12 @@ static inline void atm_account_tx(structATM_SKB (skb)->atm_options = vcc->atm_options; } +static inline void atm_return_tx (struct atm_vcc *vcc,struct sk_buff *skb) +{ +WARN_ON_ONCE (refcount_sub_and_test (ATM_SKB (skb)->acct_truesize, + &sk_atm (vcc)->sk_wmem_alloc)); +} +static inline void atm_force_charge (struct atm_vcc *vcc,int truesize) {atomic_add (truesize, &sk_atm (vcc)->sk_rmem_alloc);
net/atm/common.c +1 -0
@@ -635,6 +635,7 @@ int vcc_sendmsg(struct socket *sock, str skb->dev =NULL ;/* for paths shared with net_device interfaces */ if (!copy_from_iter_full (skb_put (skb, size), size, &m->msg_iter)) { +atm_return_tx (vcc, skb);kfree_skb (skb); error = -EFAULT ; goto out;
net/atm/raw.c +1 -1
@@ -36,7 +36,7 @@ static void atm_pop_raw(struct atm_vcc *pr_debug ("(%d) %d -= %d\n" , vcc->vci,sk_wmem_alloc_get (sk),ATM_SKB (skb)->acct_truesize); -WARN_ON (refcount_sub_and_test (ATM_SKB (skb)->acct_truesize, &sk->sk_wmem_alloc)); +atm_return_tx (vcc, skb);dev_kfree_skb_any (skb); sk->sk_write_space (sk); }
Greg Kroah-Hartman
maintainerPATCH 486/592 drm/nouveau/nvkm: factor out current GSP RPC command policies
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Zhi Wang <[email protected]>
commit 4570355f8eaa476164cfb7ca959fdbf0cebbc9eb upstream.
There can be multiple cases of handling the GSP RPC messages, which are the reply of GSP RPC commands according to the requirement of the callers and the nature of the GSP RPC commands.
The current supported reply policies are "callers don't care" and "receive the entire message" according to the requirement of the callers. To introduce a new policy, factor out the current RPC command reply polices. Also, centralize the handling of the reply in a single function.
Factor out NVKM_GSP_RPC_REPLY_NOWAIT as "callers don't care" and NVKM_GSP_RPC_REPLY_RECV as "receive the entire message". Introduce a kernel doc to document the policies. Factor out r535_gsp_rpc_handle_reply().
No functional change is intended for small GSP RPC commands. For large GSP commands, the caller decides the policy of how to handle the returned GSP RPC message.
--- Documentation/gpu/nouveau.rst | 3 drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h | 34 +++++++-- drivers/gpu/drm/nouveau/nvkm/subdev/bar/r535.c | 2 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 75 ++++++++++----------- drivers/gpu/drm/nouveau/nvkm/subdev/instmem/r535.c | 2 5 files changed, 72 insertions(+), 44 deletions(-)
Documentation/gpu/nouveau.rst +3 -0
@@ -27,3 +27,6 @@ GSP Support .. kernel-doc:: drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c :doc: GSP message queue element + +.. kernel-doc:: drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h + :doc: GSP message handling policy
drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h +28 -6
@@ -31,6 +31,25 @@ typedef int (*nvkm_gsp_msg_ntfy_func)(vostruct nvkm_gsp_event ;typedef void (*nvkm_gsp_event_func )(struct nvkm_gsp_event *,void *repv,u32 repc); +/** +* DOC: GSP message handling policy +* +* When sending a GSP RPC command, there can be multiple cases of handling +* the GSP RPC messages, which are the reply of GSP RPC commands, according +* to the requirement of the callers and the nature of the GSP RPC commands. +* +* NVKM_GSP_RPC_REPLY_NOWAIT - If specified, immediately return to the +* caller after the GSP RPC command is issued. +* +* NVKM_GSP_RPC_REPLY_RECV - If specified, wait and receive the entire GSP +* RPC message after the GSP RPC command is issued. +* +*/ +enum nvkm_gsp_rpc_reply_policy { +NVKM_GSP_RPC_REPLY_NOWAIT =0 , +NVKM_GSP_RPC_REPLY_RECV , +}; +struct nvkm_gsp {const struct nvkm_gsp_func *func;struct nvkm_subdev subdev;
@@ -188,7 +207,8 @@ struct nvkm_gsp { const struct nvkm_gsp_rm { void *(*rpc_get)(struct nvkm_gsp *, u32 fn, u32 argc); - void *(*rpc_push)(struct nvkm_gsp *, void *argv, bool wait, u32 repc); + void *(*rpc_push)(struct nvkm_gsp *gsp, void *argv, + enum nvkm_gsp_rpc_reply_policy policy, u32 repc); void (*rpc_done)(struct nvkm_gsp *gsp, void *repv); void *(*rm_ctrl_get)(struct nvkm_gsp_object *, u32 cmd, u32 argc);
@@ -255,9 +275,10 @@ nvkm_gsp_rpc_get(struct nvkm_gsp *gsp, u }static inline void * -nvkm_gsp_rpc_push (struct nvkm_gsp *gsp,void *argv,bool wait,u32 repc) +nvkm_gsp_rpc_push (struct nvkm_gsp *gsp,void *argv, +enum nvkm_gsp_rpc_reply_policy policy,u32 repc) { -return gsp->rm->rpc_push (gsp, argv, wait, repc); +return gsp->rm->rpc_push (gsp, argv, policy, repc); } static inline void *
@@ -268,13 +289,14 @@ nvkm_gsp_rpc_rd(struct nvkm_gsp *gsp, u3if (IS_ERR_OR_NULL (argv))return argv; -return nvkm_gsp_rpc_push (gsp, argv, true, argc); +return nvkm_gsp_rpc_push (gsp, argv,NVKM_GSP_RPC_REPLY_RECV , argc); }static inline int -nvkm_gsp_rpc_wr (struct nvkm_gsp *gsp,void *argv,bool wait) +nvkm_gsp_rpc_wr (struct nvkm_gsp *gsp,void *argv, +enum nvkm_gsp_rpc_reply_policy policy) { -void *repv =nvkm_gsp_rpc_push (gsp, argv, wait,0 ); +void *repv =nvkm_gsp_rpc_push (gsp, argv, policy,0 );if (IS_ERR (repv))return PTR_ERR (repv);
drivers/gpu/drm/nouveau/nvkm/subdev/bar/r535.c +1 -1
@@ -56,7 +56,7 @@ r535_bar_bar2_update_pde(struct nvkm_gsp rpc->info.entryValue = addr ? ((addr >>4 ) |2 ) :0 ;/* PD3 entry format! */ rpc->info.entryLevelShift =47 ;//XXX: probably fetch this from mmu! -return nvkm_gsp_rpc_wr (gsp, rpc, true); +return nvkm_gsp_rpc_wr (gsp, rpc,NVKM_GSP_RPC_REPLY_RECV ); } static void
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +39 -36
151 lines not on this page · raw message
drivers/gpu/drm/nouveau/nvkm/subdev/instmem/r535.c +1 -1
@@ -105,7 +105,7 @@ fbsr_memlist(struct nvkm_gsp_device *dev rpc->pteDesc.pte_pde[i].pte = (phys >>12 ) + i; } - ret =nvkm_gsp_rpc_wr (gsp, rpc, true); + ret =nvkm_gsp_rpc_wr (gsp, rpc,NVKM_GSP_RPC_REPLY_RECV );if (ret)return ret;
Greg Kroah-Hartman
maintainerPATCH 487/592 drm/nouveau/nvkm: introduce new GSP reply policy NVKM_GSP_RPC_REPLY_POLL
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Zhi Wang <[email protected]>
commit a738fa9105ac2897701ba4067c33e85faa27d1e2 upstream.
Some GSP RPC commands need a new reply policy: "caller don't care about the message content but want to make sure a reply is received". To support this case, a new reply policy is introduced.
NV_VGPU_MSG_FUNCTION_ALLOC_MEMORY is a large GSP RPC command. The actual required policy is NVKM_GSP_RPC_REPLY_POLL. This can be observed from the dump of the GSP message queue. After the large GSP RPC command is issued, GSP will write only an empty RPC header in the queue as the reply.
Without this change, the policy "receiving the entire message" is used for NV_VGPU_MSG_FUNCTION_ALLOC_MEMORY. This causes the timeout of receiving the returned GSP message in the suspend/resume path.
Introduce the new reply policy NVKM_GSP_RPC_REPLY_POLL, which waits for the returned GSP message but discards it for the caller. Use the new policy NVKM_GSP_RPC_REPLY_POLL on the GSP RPC command NV_VGPU_MSG_FUNCTION_ALLOC_MEMORY.
--- drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h | 4 ++++ drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 3 +++ drivers/gpu/drm/nouveau/nvkm/subdev/instmem/r535.c | 2 +- 3 files changed, 8 insertions(+), 1 deletion(-)
drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h +4 -0
@@ -44,10 +44,14 @@ typedef void (*nvkm_gsp_event_func)(stru* NVKM_GSP_RPC_REPLY_RECV - If specified, wait and receive the entire GSP * RPC message after the GSP RPC command is issued. * +* NVKM_GSP_RPC_REPLY_POLL - If specified, wait for the specific reply and +* discard the reply before returning to the caller. +* */ enum nvkm_gsp_rpc_reply_policy {NVKM_GSP_RPC_REPLY_NOWAIT =0 ,NVKM_GSP_RPC_REPLY_RECV , +NVKM_GSP_RPC_REPLY_POLL , };struct nvkm_gsp {
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +3 -0
@@ -602,6 +602,9 @@ r535_gsp_rpc_handle_reply(struct nvkm_gselse repv = reply;break ; +case NVKM_GSP_RPC_REPLY_POLL : + repv =r535_gsp_msg_recv (gsp, fn,0 ); +break ; }return repv;
drivers/gpu/drm/nouveau/nvkm/subdev/instmem/r535.c +1 -1
@@ -105,7 +105,7 @@ fbsr_memlist(struct nvkm_gsp_device *dev rpc->pteDesc.pte_pde[i].pte = (phys >>12 ) + i; } - ret =nvkm_gsp_rpc_wr (gsp, rpc,NVKM_GSP_RPC_REPLY_RECV ); + ret =nvkm_gsp_rpc_wr (gsp, rpc,NVKM_GSP_RPC_REPLY_POLL );if (ret)return ret;
Greg Kroah-Hartman
maintainerPATCH 488/592 gpio: loongson-64bit: Correct Loongson-7A2000 ACPI GPIO access mode
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Binbin Zhou <[email protected]>
commit 72f37957007d25f4290e3ba40d40aaec1dd0b0cf upstream.
According to the description of the Loongson-7A2000 ACPI GPIO register in the manual, its access mode should be BIT_CTRL_MODE, otherwise there maybe some unpredictable behavior.
--- drivers/gpio/gpio-loongson-64bit.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpio/gpio-loongson-64bit.c +1 -1
@@ -266,7 +266,7 @@ static const struct loongson_gpio_chip_d/* LS7A2000 ACPI GPIO */ static const struct loongson_gpio_chip_data loongson_gpio_ls7a2000_data1 = { .label ="ls7a2000_gpio" , - .mode =BYTE_CTRL_MODE , + .mode =BIT_CTRL_MODE , .conf_offset =0x4 , .in_offset =0x8 , .out_offset =0x0 ,
Greg Kroah-Hartman
maintainerPATCH 489/592 arm64: Restrict pagetable teardown to avoid false warning
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dev Jain <[email protected]>
commit 650768c512faba8070bf4cfbb28c95eb5cd203f3 upstream.
Commit 9c006972c3fe ("arm64: mmu: drop pXd_present() checks from pXd_free_pYd_table()") removes the pxd_present() checks because the caller checks pxd_present(). But, in case of vmap_try_huge_pud(), the caller only checks pud_present(); pud_free_pmd_page() recurses on each pmd through pmd_free_pte_page(), wherein the pmd may be none. Thus it is possible to hit a warning in the latter, since pmd_none => !pmd_table(). Thus, add a pmd_present() check in pud_free_pmd_page().
This problem was found by code inspection.
--- arch/arm64/mm/mmu.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
arch/arm64/mm/mmu.c +2 -1
@@ -1286,7 +1286,8 @@ int pud_free_pmd_page(pud_t *pudp, unsig next = addr; end = addr +PUD_SIZE ;do { -pmd_free_pte_page (pmdp, next); +if (pmd_present (pmdp_get (pmdp))) +pmd_free_pte_page (pmdp, next); }while (pmdp++, next +=PMD_SIZE , next != end);pud_clear (pudp);
Greg Kroah-Hartman
maintainerPATCH 490/592 ALSA: usb-audio: Rename ALSA kcontrol PCM and PCM1 for the KTMicro sound card
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: wangdicheng <[email protected]>
commit 93adf20ff4d6e865e0b974110d3cf2f07c057177 upstream.
PCM1 not in Pulseaudio's control list; standardize control to "Speaker" and "Headphone".
--- sound/usb/mixer_maps.c | 12 ++++++++++++ 1 file changed, 12 insertions(+)
sound/usb/mixer_maps.c +12 -0
@@ -383,6 +383,13 @@ static const struct usbmix_name_map ms_u { 0 } /* terminator */ }; +/* KTMicro USB */ +static struct usbmix_name_map s31b2_0022_map[] = { + {23 ,"Speaker Playback" }, + {18 ,"Headphone Playback" }, + {0 } +}; +/* ASUS ROG Zenith II with Realtek ALC1220-VB */ static const struct usbmix_name_map asus_zenith_ii_map[] = { {19 ,NULL ,12 },/* FU, Input Gain Pad - broken response, disabled */
@@ -692,6 +699,11 @@ static const struct usbmix_ctl_map usbmi .id = USB_ID(0x045e, 0x083c), .map = ms_usb_link_map, }, + { +/* KTMicro USB */ + .id = USB_ID(0X31b2, 0x0022), + .map = s31b2_0022_map, + }, { 0 } /* terminator */ };
Greg Kroah-Hartman
maintainerPATCH 491/592 ALSA: hda/intel: Add Thinkpad E15 to PM deny list
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Takashi Iwai <[email protected]>
commit c987a390f1b3b8bdac11031d7004e3410fe259bd upstream.
Lenovo Thinkpad E15 with Conexant CX8070 codec seems causing ugly noises after runtime-PM suspend. Disable the codec runtime PM as a workaround.
--- sound/pci/hda/hda_intel.c | 2 ++ 1 file changed, 2 insertions(+)
sound/pci/hda/hda_intel.c +2 -0
@@ -2276,6 +2276,8 @@ static const struct snd_pci_quirk power_SND_PCI_QUIRK (0x1734 ,0x1232 ,"KONTRON SinglePC" ,0 ),/* Dell ALC3271 */ SND_PCI_QUIRK (0x1028 ,0x0962 ,"Dell ALC3271" ,0 ), +/* https://bugzilla.kernel.org/show_bug.cgi?id=220210 */ + SND_PCI_QUIRK(0x17aa, 0x5079, "Lenovo Thinkpad E15", 0), {} };
Greg Kroah-Hartman
maintainerPATCH 492/592 ALSA: hda/realtek - Add mute LED support for HP Victus 16-s1xxx and HP Victus 15-fa1xxx
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Edip Hazuri <[email protected]>
commit a0914bf56e26d2cf457690602883f9cd2ec2c646 upstream.
The mute led on those laptops is using ALC245 but requires a quirk to work This patch enables the existing quirk for the devices.
Tested on my Victus 16-s1011nt Laptop and my friend's Victus 15-fa1xxx. The LED behaviour works as intended.
--- sound/pci/hda/patch_realtek.c | 2 ++ 1 file changed, 2 insertions(+)
sound/pci/hda/patch_realtek.c +2 -0
@@ -10787,6 +10787,7 @@ static const struct hda_quirk alc269_fixSND_PCI_QUIRK (0x103c ,0x8b97 ,"HP" ,ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF ),SND_PCI_QUIRK (0x103c ,0x8bb3 ,"HP Slim OMEN" ,ALC287_FIXUP_CS35L41_I2C_2 ),SND_PCI_QUIRK (0x103c ,0x8bb4 ,"HP Slim OMEN" ,ALC287_FIXUP_CS35L41_I2C_2 ), +SND_PCI_QUIRK (0x103c ,0x8bc8 ,"HP Victus 15-fa1xxx" ,ALC245_FIXUP_HP_MUTE_LED_COEFBIT ),SND_PCI_QUIRK (0x103c ,0x8bcd ,"HP Omen 16-xd0xxx" ,ALC245_FIXUP_HP_MUTE_LED_V1_COEFBIT ),SND_PCI_QUIRK (0x103c ,0x8bdd ,"HP Envy 17" ,ALC287_FIXUP_CS35L41_I2C_2 ), SND_PCI_QUIRK(0x103c, 0x8bde, "HP Envy 17", ALC287_FIXUP_CS35L41_I2C_2),
@@ -10840,6 +10841,7 @@ static const struct hda_quirk alc269_fixSND_PCI_QUIRK (0x103c ,0x8c91 ,"HP EliteBook 660" ,ALC236_FIXUP_HP_GPIO_LED ),SND_PCI_QUIRK (0x103c ,0x8c96 ,"HP" ,ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF ),SND_PCI_QUIRK (0x103c ,0x8c97 ,"HP ZBook" ,ALC236_FIXUP_HP_MUTE_LED_MICMUTE_VREF ), +SND_PCI_QUIRK (0x103c ,0x8c9c ,"HP Victus 16-s1xxx (MB 8C9C)" ,ALC245_FIXUP_HP_MUTE_LED_COEFBIT ),SND_PCI_QUIRK (0x103c ,0x8ca1 ,"HP ZBook Power" ,ALC236_FIXUP_HP_GPIO_LED ),SND_PCI_QUIRK (0x103c ,0x8ca2 ,"HP ZBook Power" ,ALC236_FIXUP_HP_GPIO_LED ), SND_PCI_QUIRK(0x103c, 0x8ca4, "HP ZBook Fury", ALC245_FIXUP_CS35L41_SPI_2_HP_GPIO_LED),
Greg Kroah-Hartman
maintainerPATCH 493/592 ALSA: hda/realtek: enable headset mic on Latitude 5420 Rugged
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jonathan Lane <[email protected]>
commit efa6bdf1bc75e26cafaa5f1d775e8bb7c5b0c431 upstream.
Like many Dell laptops, the 3.5mm port by default can not detect a combined headphones+mic headset or even a pure microphone. This change enables the port's functionality.
--- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -10509,6 +10509,7 @@ static const struct hda_quirk alc269_fixSND_PCI_QUIRK (0x1028 ,0x0871 ,"Dell Precision 3630" ,ALC255_FIXUP_DELL_HEADSET_MIC ),SND_PCI_QUIRK (0x1028 ,0x0872 ,"Dell Precision 3630" ,ALC255_FIXUP_DELL_HEADSET_MIC ),SND_PCI_QUIRK (0x1028 ,0x0873 ,"Dell Precision 3930" ,ALC255_FIXUP_DUMMY_LINEOUT_VERB ), +SND_PCI_QUIRK (0x1028 ,0x0879 ,"Dell Latitude 5420 Rugged" ,ALC269_FIXUP_DELL4_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x1028 ,0x08ad ,"Dell WYSE AIO" ,ALC225_FIXUP_DELL_WYSE_AIO_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x1028 ,0x08ae ,"Dell WYSE NB" ,ALC225_FIXUP_DELL1_MIC_NO_PRESENCE ), SND_PCI_QUIRK(0x1028, 0x0935, "Dell", ALC274_FIXUP_DELL_AIO_LINEOUT_VERB),
Greg Kroah-Hartman
maintainerPATCH 494/592 ALSA: hda/realtek: Fix built-in mic on ASUS VivoBook X513EA
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chris Chiu <[email protected]>
commit c6451a7325874c119def1d4094f6815c0c8fdc23 upstream.
The built-in mic of ASUS VivoBook X513EA is broken recently by the fix of the pin sort. The fixup ALC256_FIXUP_ASUS_MIC_NO_PRESENCE is working for addressing the regression, too.
--- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -10907,6 +10907,7 @@ static const struct hda_quirk alc269_fixSND_PCI_QUIRK (0x103c ,0x8e60 ,"HP Trekker " ,ALC287_FIXUP_CS35L41_I2C_2 ),SND_PCI_QUIRK (0x103c ,0x8e61 ,"HP Trekker " ,ALC287_FIXUP_CS35L41_I2C_2 ),SND_PCI_QUIRK (0x103c ,0x8e62 ,"HP Trekker " ,ALC287_FIXUP_CS35L41_I2C_2 ), +SND_PCI_QUIRK (0x1043 ,0x1032 ,"ASUS VivoBook X513EA" ,ALC256_FIXUP_ASUS_MIC_NO_PRESENCE ),SND_PCI_QUIRK (0x1043 ,0x103e ,"ASUS X540SA" ,ALC256_FIXUP_ASUS_MIC ),SND_PCI_QUIRK (0x1043 ,0x103f ,"ASUS TX300" ,ALC282_FIXUP_ASUS_TX300 ), SND_PCI_QUIRK(0x1043, 0x1054, "ASUS G614FH/FM/FP", ALC287_FIXUP_CS35L41_I2C_2),
Greg Kroah-Hartman
maintainerPATCH 495/592 ALSA: hda/realtek: Add quirk for Asus GU605C
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Richard Fitzgerald <[email protected]>
commit 7b23887a0c70d15459f02c51651a111e9e5cab86 upstream.
The GU605C has similar audio hardware to the GU605M so apply the same quirk.
Note that in the linked bugzilla there are two separate problems with the GU605C. This patch fixes one of the problems, so I haven't added a Closes: tag.
--- sound/pci/hda/patch_realtek.c | 1 + 1 file changed, 1 insertion(+)
sound/pci/hda/patch_realtek.c +1 -0
@@ -10908,6 +10908,7 @@ static const struct hda_quirk alc269_fixSND_PCI_QUIRK (0x103c ,0x8e61 ,"HP Trekker " ,ALC287_FIXUP_CS35L41_I2C_2 ),SND_PCI_QUIRK (0x103c ,0x8e62 ,"HP Trekker " ,ALC287_FIXUP_CS35L41_I2C_2 ),SND_PCI_QUIRK (0x1043 ,0x1032 ,"ASUS VivoBook X513EA" ,ALC256_FIXUP_ASUS_MIC_NO_PRESENCE ), +SND_PCI_QUIRK (0x1043 ,0x1034 ,"ASUS GU605C" ,ALC285_FIXUP_ASUS_GU605_SPI_SPEAKER2_TO_DAC1 ),SND_PCI_QUIRK (0x1043 ,0x103e ,"ASUS X540SA" ,ALC256_FIXUP_ASUS_MIC ),SND_PCI_QUIRK (0x1043 ,0x103f ,"ASUS TX300" ,ALC282_FIXUP_ASUS_TX300 ), SND_PCI_QUIRK(0x1043, 0x1054, "ASUS G614FH/FM/FP", ALC287_FIXUP_CS35L41_I2C_2),
Greg Kroah-Hartman
maintainerPATCH 496/592 drm/appletbdrm: Make appletbdrm depend on X86
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Aditya Garg <[email protected]>
commit de5fbbe1531f645c8b56098be8d1faf31e46f7f0 upstream.
The appletbdrm driver is exclusively for Touch Bars on x86 Intel Macs. The M1 Macs have a separate driver. So, lets avoid compiling it for other architectures.
--- drivers/gpu/drm/tiny/Kconfig | 1 + 1 file changed, 1 insertion(+)
drivers/gpu/drm/tiny/Kconfig +1 -0
@@ -3,6 +3,7 @@ config DRM_APPLETBDRM tristate "DRM support for Apple Touch Bars" depends on DRM && USB && MMU + depends on X86 || COMPILE_TEST select DRM_GEM_SHMEM_HELPER select DRM_KMS_HELPER help
Greg Kroah-Hartman
maintainerPATCH 497/592 mm/madvise: handle madvise_lock() failure during race unwinding
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: SeongJae Park <[email protected]>
commit 9c49e5d09f076001e05537734d7df002162eb2b5 upstream.
When unwinding race on -ERESTARTNOINTR handling of process_madvise(), madvise_lock() failure is ignored. Check the failure and abort remaining works in the case.
--- mm/madvise.c | 5 ++++- 1 file changed, 4 insertions(+), 1 deletion(-)
mm/madvise.c +4 -1
@@ -1832,7 +1832,9 @@ static ssize_t vector_madvise(struct mm_/* Drop and reacquire lock to unwind race. */ madvise_unlock (mm, behavior); -madvise_lock (mm, behavior); + ret =madvise_lock (mm, behavior); +if (ret) + goto out;continue ; }if (ret <0 )
@@ -1841,6 +1843,7 @@ static ssize_t vector_madvise(struct mm_ }madvise_unlock (mm, behavior); +out: ret = (total_len -iov_iter_count (iter)) ? : ret;return ret;
Greg Kroah-Hartman
maintainerPATCH 498/592 erofs: remove unused trace event erofs_destroy_inode
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Gao Xiang <[email protected]>
commit 30b58444807c93bffeaba7d776110f2a909d2f9a upstream.
The trace event erofs_destroy_inode was added but remains unused. This unused event contributes approximately 5KB to the kernel module size.
--- include/trace/events/erofs.h | 18 ------------------ 1 file changed, 18 deletions(-)
include/trace/events/erofs.h +0 -18
@@ -211,24 +211,6 @@ TRACE_EVENT(erofs_map_blocks_exit, show_mflags(__entry->mflags), __entry->ret) ); -TRACE_EVENT (erofs_destroy_inode, - TP_PROTO(struct inode *inode), - -TP_ARGS (inode), - -TP_STRUCT__entry ( - __field( dev_t, dev ) -__field ( erofs_nid_t, nid ) - ), - -TP_fast_assign ( - __entry->dev = inode->i_sb->s_dev; - __entry->nid = EROFS_I(inode)->nid; - ), - -TP_printk ("dev = (%d,%d), nid = %llu" ,show_dev_nid (__entry)) -); -#endif /* _TRACE_EROFS_H */ /* This part must be outside protection */
Greg Kroah-Hartman
maintainerPATCH 499/592 nfsd: use threads array as-is in netlink interface
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jeff Layton <[email protected]>
commit 8ea688a3372e8369dc04395b39b4e71a6d91d4d5 upstream.
The old nfsdfs interface for starting a server with multiple pools handles the special case of a single entry array passed down from userland by distributing the threads over every NUMA node.
The netlink control interface however constructs an array of length nfsd_nrpools() and fills any unprovided slots with 0's. This behavior defeats the special casing that the old interface relies on.
Change nfsd_nl_threads_set_doit() to pass down the array from userland as-is.
--- fs/nfsd/nfsctl.c | 5 ++--- 1 file changed, 2 insertions(+), 3 deletions(-)
fs/nfsd/nfsctl.c +2 -3
@@ -1611,7 +1611,7 @@ out_unlock:*/ int nfsd_nl_threads_set_doit (struct sk_buff *skb,struct genl_info *info) { -int *nthreads, count =0 , nrpools, i, ret = -EOPNOTSUPP , rem; +int *nthreads, nrpools =0 , i, ret = -EOPNOTSUPP , rem;struct net *net =genl_info_net (info);struct nfsd_net *nn =net_generic (net, nfsd_net_id);const struct nlattr *attr;
@@ -1623,12 +1623,11 @@ int nfsd_nl_threads_set_doit(struct sk_b/* count number of SERVER_THREADS values */ nlmsg_for_each_attr (attr, info->nlhdr,GENL_HDRLEN , rem) {if (nla_type (attr) ==NFSD_A_SERVER_THREADS ) - count++; + nrpools++; }mutex_lock (&nfsd_mutex); - nrpools =max (count,nfsd_nrpools (net)); nthreads =kcalloc (nrpools,sizeof (int ),GFP_KERNEL );if (!nthreads) { ret = -ENOMEM ;
Greg Kroah-Hartman
maintainerPATCH 500/592 sunrpc: handle SVC_GARBAGE during svc auth processing as auth error
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jeff Layton <[email protected]>
commit 94d10a4dba0bc482f2b01e39f06d5513d0f75742 upstream.
tianshuo han reported a remotely-triggerable crash if the client sends a kernel RPC server a specially crafted packet. If decoding the RPC reply fails in such a way that SVC_GARBAGE is returned without setting the rq_accept_statp pointer, then that pointer can be dereferenced and a value stored there.
If it's the first time the thread has processed an RPC, then that pointer will be set to NULL and the kernel will crash. In other cases, it could create a memory scribble.
The server sunrpc code treats a SVC_GARBAGE return from svc_authenticate or pg_authenticate as if it should send a GARBAGE_ARGS reply. RFC 5531 says that if authentication fails that the RPC should be rejected instead with a status of AUTH_ERR.
Handle a SVC_GARBAGE return as an AUTH_ERROR, with a reason of AUTH_BADCRED instead of returning GARBAGE_ARGS in that case. This sidesteps the whole problem of touching the rpc_accept_statp pointer in this situation and avoids the crash.
--- net/sunrpc/svc.c | 11 ++--------- 1 file changed, 2 insertions(+), 9 deletions(-)
net/sunrpc/svc.c +2 -9
@@ -1369,7 +1369,8 @@ svc_process_common(struct svc_rqst *rqstcase SVC_OK :break ;case SVC_GARBAGE : - goto err_garbage_args; + rqstp->rq_auth_stat = rpc_autherr_badcred; + goto err_bad_auth;case SVC_SYSERR : goto err_system_err;case SVC_DENIED :
@@ -1510,14 +1511,6 @@ err_bad_proc: *rqstp->rq_accept_statp = rpc_proc_unavail; goto sendit; -err_garbage_args: -svc_printk (rqstp,"failed to decode RPC header\n" ); - -if (serv->sv_stats) - serv->sv_stats->rpcbadfmt++; - *rqstp->rq_accept_statp = rpc_garbage_args; - goto sendit; - err_system_err:if (serv->sv_stats) serv->sv_stats->rpcbadfmt++;
Greg Kroah-Hartman
maintainerPATCH 501/592 io_uring/net: always use current transfer count for buffer put
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Axboe <[email protected]>
commit 51a4598ad5d9eb6be4ec9ba65bbfdf0ac302eb2e upstream.
A previous fix corrected the retry condition for when to continue a current bundle, but it missed that the current (not the total) transfer count also applies to the buffer put. If not, then for incrementally consumed buffer rings repeated completions on the same request may end up over consuming.
--- io_uring/net.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
io_uring/net.c +1 -1
@@ -829,7 +829,7 @@ static inline bool io_recv_finish(structif (sr->flags &IORING_RECVSEND_BUNDLE ) {size_t this_ret = *ret - sr->done_io; - cflags |=io_put_kbufs (req, *ret,io_bundle_nbufs (kmsg, this_ret), + cflags |=io_put_kbufs (req, this_ret,io_bundle_nbufs (kmsg, this_ret), issue_flags);if (sr->retry) cflags = req->cqe.flags | (cflags &CQE_F_MASK );
Greg Kroah-Hartman
maintainerPATCH 502/592 drm/xe/svm: Fix regression disallowing 64K SVM migration
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Maarten Lankhorst <[email protected]>
commit d6fb4f01736a1d18cc981eb04fa2907a7121fc27 upstream.
When changing the condition from >= SZ_64K, it was changed to <= SZ_64K. This disallows migration of 64K, which is the exact minimum allowed.
--- drivers/gpu/drm/xe/xe_svm.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/xe/xe_svm.c +1 -1
@@ -750,7 +750,7 @@ static bool xe_svm_range_needs_migrate_treturn false; } -if (range_size <=SZ_64K && !supports_4K_migration (vm->xe)) { +if (range_size <SZ_64K && !supports_4K_migration (vm->xe)) {drm_dbg (&vm->xe->drm,"Platform doesn't support SZ_4K range migration\n" );return false; }
Greg Kroah-Hartman
maintainerPATCH 503/592 drm/v3d: Avoid NULL pointer dereference in `v3d_job_update_stats()`
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Maíra Canal <[email protected]>
commit e1bc3a13bd775791cca0bb144d977b00f3598042 upstream.
The following kernel Oops was recently reported by Mesa CI:
[ 800.139824] Unable to handle kernel NULL pointer dereference at virtual address 0000000000000588 [ 800.148619] Mem abort info: [ 800.151402] ESR = 0x0000000096000005 [ 800.155141] EC = 0x25: DABT (current EL), IL = 32 bits [ 800.160444] SET = 0, FnV = 0 [ 800.163488] EA = 0, S1PTW = 0 [ 800.166619] FSC = 0x05: level 1 translation fault [ 800.171487] Data abort info: [ 800.174357] ISV = 0, ISS = 0x00000005, ISS2 = 0x00000000 [ 800.179832] CM = 0, WnR = 0, TnD = 0, TagAccess = 0 [ 800.184873] GCS = 0, Overlay = 0, DirtyBit = 0, Xs = 0 [ 800.190176] user pgtable: 4k pages, 39-bit VAs, pgdp=00000001014c2000 [ 800.196607] [0000000000000588] pgd=0000000000000000, p4d=0000000000000000, pud=0000000000000000 [ 800.205305] Internal error: Oops: 0000000096000005 [#1] PREEMPT SMP [ 800.211564] Modules linked in: vc4 snd_soc_hdmi_codec drm_display_helper v3d cec gpu_sched drm_dma_helper drm_shmem_helper drm_kms_helper drm drm_panel_orientation_quirks snd_soc_core snd_compress snd_pcm_dmaengine snd_pcm i2c_brcmstb snd_timer snd backlight [ 800.234448] CPU: 0 UID: 0 PID: 0 Comm: swapper/0 Not tainted 6.12.25+rpt-rpi-v8 #1 Debian 1:6.12.25-1+rpt1 [ 800.244182] Hardware name: Raspberry Pi 4 Model B Rev 1.4 (DT) [ 800.250005] pstate: 600000c5 (nZCv daIF -PAN -UAO -TCO -DIT -SSBS BTYPE=--) [ 800.256959] pc : v3d_job_update_stats+0x60/0x130 [v3d] [ 800.262112] lr : v3d_job_update_stats+0x48/0x130 [v3d] [ 800.267251] sp : ffffffc080003e60 [ 800.270555] x29: ffffffc080003e60 x28: ffffffd842784980 x27: 0224012000000000 [ 800.277687] x26: ffffffd84277f630 x25: ffffff81012fd800 x24: 0000000000000020 [ 800.284818] x23: ffffff8040238b08 x22: 0000000000000570 x21: 0000000000000158 [ 800.291948] x20: 0000000000000000 x19: ffffff8040238000 x18: 0000000000000000 [ 800.299078] x17: ffffffa8c1bd2000 x16: ffffffc080000000 x15: 0000000000000000 [ 800.306208] x14: 0000000000000000 x13: 0000000000000000 x12: 0000000000000000 [ 800.313338] x11: 0000000000000040 x10: 0000000000001a40 x9 : ffffffd83b39757c [ 800.320468] x8 : ffffffd842786420 x7 : 7fffffffffffffff x6 : 0000000000ef32b0 [ 800.327598] x5 : 00ffffffffffffff x4 : 0000000000000015 x3 : ffffffd842784980 [ 800.334728] x2 : 0000000000000004 x1 : 0000000000010002 x0 : 000000ba4c0ca382 [ 800.341859] Call trace: [ 800.344294] v3d_job_update_stats+0x60/0x130 [v3d] [ 800.349086] v3d_irq+0x124/0x2e0 [v3d] [ 800.352835] __handle_irq_event_percpu+0x58/0x218 [ 800.357539] handle_irq_event+0x54/0xb8 [ 800.361369] handle_fasteoi_irq+0xac/0x240 [ 800.365458] handle_irq_desc+0x48/0x68 [ 800.369200] generic_handle_domain_irq+0x24/0x38 [ 800.373810] gic_handle_irq+0x48/0xd8 [ 800.377464] call_on_irq_stack+0x24/0x58 [ 800.381379] do_interrupt_handler+0x88/0x98 [ 800.385554] el1_interrupt+0x34/0x68 [ 800.389123] el1h_64_irq_handler+0x18/0x28 [ 800.393211] el1h_64_irq+0x64/0x68 [ 800.396603] default_idle_call+0x3c/0x168 [ 800.400606] do_idle+0x1fc/0x230 [ 800.403827] cpu_startup_entry+0x40/0x50 [ 800.407742] rest_init+0xe4/0xf0 [ 800.410962] start_kernel+0x5e8/0x790 [ 800.414616] __primary_switched+0x80/0x90 [ 800.418622] Code: 8b170277 8b160296 11000421 b9000861 (b9401ac1) [ 800.424707] ---[ end trace 0000000000000000 ]--- [ 800.457313] ---[ end Kernel panic - not syncing: Oops: Fatal exception in interrupt ]---
This issue happens when the file descriptor is closed before the jobs submitted by it are completed. When the job completes, we update the global GPU stats and the per-fd GPU stats, which are exposed through fdinfo. If the file descriptor was closed, then the struct v3d_file_priv and its stats were already freed and we can't update the per-fd stats.
Therefore, if the file descriptor was already closed, don't update the per-fd GPU stats, only update the global ones.
--- drivers/gpu/drm/v3d/v3d_sched.c | 8 ++++++-- 1 file changed, 6 insertions(+), 2 deletions(-)
drivers/gpu/drm/v3d/v3d_sched.c +6 -2
@@ -199,7 +199,6 @@ v3d_job_update_stats(struct v3d_job *jobstruct v3d_dev *v3d = job->v3d;struct v3d_file_priv *file = job->file->driver_priv;struct v3d_stats *global_stats = &v3d->queue[queue].stats; -struct v3d_stats *local_stats = &file->stats[queue];u64 now =local_clock ();unsigned long flags;
@@ -209,7 +208,12 @@ v3d_job_update_stats(struct v3d_job *jobelse preempt_disable (); -v3d_stats_update (local_stats, now); +/* Don't update the local stats if the file context has already closed */ +if (file) +v3d_stats_update (&file->stats[queue], now); +else +drm_dbg (&v3d->drm,"The file descriptor was closed before job completion\n" ); +v3d_stats_update (global_stats, now);if (IS_ENABLED (CONFIG_LOCKDEP ))
Greg Kroah-Hartman
maintainerPATCH 504/592 smb: fix secondary channel creation issue with kerberos by populating hostname when adding channels
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Bharath SM <[email protected]>
commit 306cb65bb0cb243389fcbd0a66907d5bdea07d1e upstream.
When mounting a share with kerberos authentication with multichannel support, share mounts correctly, but fails to create secondary channels. This occurs because the hostname is not populated when adding the channels. The hostname is necessary for the userspace cifs.upcall program to retrieve the required credentials and pass it back to kernel, without hostname secondary channels fails establish.
--- fs/smb/client/sess.c | 3 +-- 1 file changed, 1 insertion(+), 2 deletions(-)
fs/smb/client/sess.c +1 -2
@@ -498,8 +498,7 @@ cifs_ses_add_channel(struct cifs_ses *se ctx->domainauto = ses->domainAuto; ctx->domainname = ses->domainName; -/* no hostname for extra channels */ - ctx->server_hostname ="" ; + ctx->server_hostname = ses->server->hostname; ctx->username = ses->user_name; ctx->password = ses->password;
Greg Kroah-Hartman
maintainerPATCH 505/592 drm/msm/dp: Disable wide bus support for SDM845
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: James A. MacInnes <[email protected]>
[ Upstream commit 83c4c67076c209787515e06fffd41dd0bdab09b9 ]
When widebus was enabled for DisplayPort in commit c7c412202623 ("drm/msm/dp: enable widebus on all relevant chipsets") it was clarified that it is only supported on DPU 5.0.0 onwards which includes SC7180 on DPU revision 6.2. However, this patch missed that the description structure for SC7180 is also reused for SDM845 (because of identical io_start address) which is only DPU 4.0.0, leading to a wrongly enbled widebus feature and corruption on that platform.
Create a separate msm_dp_desc_sdm845 structure for this SoC compatible, with the wide_bus_supported flag turned off.
Patchwork: https://patchwork.freedesktop.org/patch/636944/
--- drivers/gpu/drm/msm/dp/dp_display.c | 7 ++++++- 1 file changed, 6 insertions(+), 1 deletion(-)
drivers/gpu/drm/msm/dp/dp_display.c +6 -1
@@ -127,6 +127,11 @@ static const struct msm_dp_desc msm_dp_desc_sa8775p[] = { {} }; +static const struct msm_dp_desc msm_dp_desc_sdm845[] = { + { .io_start =0x0ae90000 , .id =MSM_DP_CONTROLLER_0 }, + {} +}; +static const struct msm_dp_desc msm_dp_desc_sc7180[] = { { .io_start =0x0ae90000 , .id =MSM_DP_CONTROLLER_0 , .wide_bus_supported = true }, {}
@@ -179,7 +184,7 @@ static const struct of_device_id msm_dp_dt_match[] = { { .compatible = "qcom,sc8180x-edp", .data = &msm_dp_desc_sc8180x }, { .compatible = "qcom,sc8280xp-dp", .data = &msm_dp_desc_sc8280xp }, { .compatible = "qcom,sc8280xp-edp", .data = &msm_dp_desc_sc8280xp }, - { .compatible = "qcom,sdm845-dp", .data = &msm_dp_desc_sc7180 }, + { .compatible = "qcom,sdm845-dp", .data = &msm_dp_desc_sdm845 }, { .compatible = "qcom,sm8350-dp", .data = &msm_dp_desc_sc7180 }, { .compatible = "qcom,sm8650-dp", .data = &msm_dp_desc_sm8650 }, { .compatible = "qcom,x1e80100-dp", .data = &msm_dp_desc_x1e80100 },
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 506/592 drm/msm/disp: Correct porch timing for SDM845
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: James A. MacInnes <[email protected]>
[ Upstream commit 146e87f3e11de0dfa091ff87e34b4bc6eec761a4 ]
Type-C DisplayPort inoperable due to incorrect porch settings.
- Re-used wide_bus_en as flag to prevent porch shifting
Patchwork: https://patchwork.freedesktop.org/patch/636945/
--- .../gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c | 14 +++++++++----- 1 file changed, 9 insertions(+), 5 deletions(-)
drivers/gpu/drm/msm/disp/dpu1/dpu_encoder_phys_vid.c +9 -5
@@ -94,17 +94,21 @@ static void drm_mode_to_intf_timing_params( timing->vsync_polarity =0 ; } -/* for DP/EDP, Shift timings to align it to bottom right */ -if (phys_enc->hw_intf->cap->type ==INTF_DP ) { + timing->wide_bus_en =dpu_encoder_is_widebus_enabled (phys_enc->parent); + timing->compression_en =dpu_encoder_is_dsc_enabled (phys_enc->parent); + +/* +* For DP/EDP, Shift timings to align it to bottom right. +* wide_bus_en is set for everything excluding SDM845 & +* porch changes cause DisplayPort failure and HDMI tearing. +*/ +if (phys_enc->hw_intf->cap->type ==INTF_DP && timing->wide_bus_en) { timing->h_back_porch += timing->h_front_porch; timing->h_front_porch =0 ; timing->v_back_porch += timing->v_front_porch; timing->v_front_porch =0 ; } - timing->wide_bus_en =dpu_encoder_is_widebus_enabled (phys_enc->parent); - timing->compression_en =dpu_encoder_is_dsc_enabled (phys_enc->parent); -/* * for DP, divide the horizonal parameters by 2 when * widebus is enabled
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 507/592 drm/msm/dsi/dsi_phy_10nm: Fix missing initial VCO rate
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Krzysztof Kozlowski <[email protected]>
[ Upstream commit 8a48e35becb214743214f5504e726c3ec131cd6d ]
Driver unconditionally saves current state on first init in dsi_pll_10nm_init(), but does not save the VCO rate, only some of the divider registers. The state is then restored during probe/enable via msm_dsi_phy_enable() -> msm_dsi_phy_pll_restore_state() -> dsi_10nm_pll_restore_state().
Restoring calls dsi_pll_10nm_vco_set_rate() with pll_10nm->vco_current_rate=0, which basically overwrites existing rate of VCO and messes with clock hierarchy, by setting frequency to 0 to clock tree. This makes anyway little sense - VCO rate was not saved, so should not be restored.
If PLL was not configured configure it to minimum rate to avoid glitches and configuring entire in clock hierarchy to 0 Hz.
Patchwork: https://patchwork.freedesktop.org/patch/654796/
--- drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c | 7 +++++++ 1 file changed, 7 insertions(+)
drivers/gpu/drm/msm/dsi/phy/dsi_phy_10nm.c +7 -0
@@ -704,6 +704,13 @@ static int dsi_pll_10nm_init(struct msm_dsi_phy *phy)/* TODO: Remove this when we have proper display handover support */ msm_dsi_phy_pll_save_state (phy); +/* +* Store also proper vco_current_rate, because its value will be used in +* dsi_10nm_pll_restore_state(). +*/ +if (!dsi_pll_10nm_vco_recalc_rate (&pll_10nm->clk_hw,VCO_REF_CLK_RATE )) + pll_10nm->vco_current_rate = pll_10nm->phy->cfg->min_pll_rate; +return 0 ; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 508/592 drm/msm: Fix CP_RESET_CONTEXT_STATE bitfield names
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Connor Abbott <[email protected]>
[ Upstream commit b1c9e797ad37d188675505b66a3a4bbeea5d9560 ]
Based on kgsl.
Patchwork: https://patchwork.freedesktop.org/patch/654922/
--- drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/gpu/drm/msm/registers/adreno/adreno_pm4.xml +2 -1
@@ -2255,7 +2255,8 @@ opcode: CP_LOAD_STATE4 (30) (4 dwords) <reg32 offset="0" name="0"> <bitfield name="CLEAR_ON_CHIP_TS" pos="0" type="boolean"/> <bitfield name="CLEAR_RESOURCE_TABLE" pos="1" type="boolean"/> - <bitfield name="CLEAR_GLOBAL_LOCAL_TS" pos="2" type="boolean"/> + <bitfield name="CLEAR_BV_BR_COUNTER" pos="2" type="boolean"/> + <bitfield name="RESET_GLOBAL_LOCAL_TS" pos="3" type="boolean"/> </reg32> </domain>
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 509/592 drm/msm/a7xx: Call CP_RESET_CONTEXT_STATE
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Connor Abbott <[email protected]>
[ Upstream commit 2b520c6104f34e3a548525173c38ebca4402cac3 ]
Calling this packet is necessary when we switch contexts because there are various pieces of state used by userspace to synchronize between BR and BV that are persistent across submits and we need to make sure that they are in a "safe" state when switching contexts. Otherwise a userspace submission in one context could cause another context to function incorrectly and hang, effectively a denial of service (although without leaking data). This was missed during initial a7xx bringup.
Patchwork: https://patchwork.freedesktop.org/patch/654924/
--- drivers/gpu/drm/msm/adreno/a6xx_gpu.c | 14 ++++++++++++++ 1 file changed, 14 insertions(+)
drivers/gpu/drm/msm/adreno/a6xx_gpu.c +14 -0
@@ -130,6 +130,20 @@ static void a6xx_set_pagetable(struct a6xx_gpu *a6xx_gpu,OUT_RING (ring,lower_32_bits (rbmemptr (ring, fence)));OUT_RING (ring,upper_32_bits (rbmemptr (ring, fence)));OUT_RING (ring, submit->seqno -1 ); + +OUT_PKT7 (ring,CP_THREAD_CONTROL ,1 ); +OUT_RING (ring,CP_SET_THREAD_BOTH ); + +/* Reset state used to synchronize BR and BV */ +OUT_PKT7 (ring,CP_RESET_CONTEXT_STATE ,1 ); +OUT_RING (ring, +CP_RESET_CONTEXT_STATE_0_CLEAR_ON_CHIP_TS | +CP_RESET_CONTEXT_STATE_0_CLEAR_RESOURCE_TABLE | +CP_RESET_CONTEXT_STATE_0_CLEAR_BV_BR_COUNTER | +CP_RESET_CONTEXT_STATE_0_RESET_GLOBAL_LOCAL_TS ); + +OUT_PKT7 (ring,CP_THREAD_CONTROL ,1 ); +OUT_RING (ring,CP_SET_THREAD_BR ); }if (!sysprof) {
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 510/592 drm/ssd130x: fix ssd132x_clear_screen() columns
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: John Keeping <[email protected]>
[ Upstream commit e479da4054875c4cc53a7fb956ebff03d2dac939 ]
The number of columns relates to the width, not the height. Use the correct variable.
--- drivers/gpu/drm/solomon/ssd130x.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/solomon/ssd130x.c +1 -1
@@ -974,7 +974,7 @@ static void ssd130x_clear_screen(struct ssd130x_device *ssd130x, u8 *data_array)static void ssd132x_clear_screen (struct ssd130x_device *ssd130x,u8 *data_array) { -unsigned int columns =DIV_ROUND_UP (ssd130x->height,SSD132X_SEGMENT_WIDTH ); +unsigned int columns =DIV_ROUND_UP (ssd130x->width,SSD132X_SEGMENT_WIDTH );unsigned int height = ssd130x->height;memset (data_array,0 , columns * height);
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 511/592 ionic: Prevent driver/fw getting out of sync on devcmd(s)
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Brett Creeley <[email protected]>
[ Upstream commit 5466491c9e3309ed5c7adbb8fad6e93fcc9a8fe9 ]
Some stress/negative firmware testing around devcmd(s) returning EAGAIN found that the done bit could get out of sync in the firmware when it wasn't cleared in a retry case.
While here, change the type of the local done variable to a bool to match the return type from ionic_dev_cmd_done().
--- drivers/net/ethernet/pensando/ionic/ionic_main.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/net/ethernet/pensando/ionic/ionic_main.c +2 -1
@@ -516,9 +516,9 @@ static int __ionic_dev_cmd_wait(struct ionic *ionic, unsigned long max_seconds,unsigned long start_time;unsigned long max_wait;unsigned long duration; -int done =0 ;bool fw_up;int opcode; +bool done;int err;/* Wait for dev cmd to complete, retrying if we get EAGAIN,
@@ -526,6 +526,7 @@ static int __ionic_dev_cmd_wait(struct ionic *ionic, unsigned long max_seconds,*/ max_wait = jiffies + (max_seconds *HZ ); try_again: + done = false; opcode = idev->opcode; start_time = jiffies; for (fw_up = ionic_is_fw_running(idev);
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 512/592 drm/nouveau/gsp: split rpc handling out on its own
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ben Skeggs <[email protected]>
[ Upstream commit 8a8b1ec5261f20d86c76c8fb235ee2441744bc10 ]
Later patches in the series add HALs around various RM APIs in order to support a newer version of GSP-RM firmware. In order to do this, begin by splitting the code up into "modules" that roughly represent RM's API boundaries so they can be more easily managed.
Aside from moving the RPC function pointers, no code change is indended.
Stable-dep-of: 9802f0a63b64 ("drm/nouveau: fix a use-after-free in r535_gsp_rpc_push()")
--- drivers/gpu/drm/nouveau/Kbuild | 1 + .../gpu/drm/nouveau/include/nvkm/subdev/gsp.h | 13 +- .../gpu/drm/nouveau/nvkm/subdev/gsp/Kbuild | 2 + .../gpu/drm/nouveau/nvkm/subdev/gsp/r535.c | 665 +---------------- .../gpu/drm/nouveau/nvkm/subdev/gsp/rm/Kbuild | 5 + .../nouveau/nvkm/subdev/gsp/rm/r535/Kbuild | 6 + .../drm/nouveau/nvkm/subdev/gsp/rm/r535/rm.c | 10 + .../drm/nouveau/nvkm/subdev/gsp/rm/r535/rpc.c | 692 ++++++++++++++++++ .../gpu/drm/nouveau/nvkm/subdev/gsp/rm/rm.h | 20 + .../gpu/drm/nouveau/nvkm/subdev/gsp/rm/rpc.h | 18 + 10 files changed, 762 insertions(+), 670 deletions(-) create mode 100644 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/Kbuild create mode 100644 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/Kbuild create mode 100644 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rm.c create mode 100644 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rpc.c create mode 100644 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rm.h create mode 100644 drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rpc.h
drivers/gpu/drm/nouveau/Kbuild +1 -0
@@ -2,6 +2,7 @@ ccflags-y += -I $(src)/include ccflags-y += -I $(src)/include/nvkm ccflags-y += -I $(src)/nvkm +ccflags-y += -I $(src)/nvkm/subdev/gsp ccflags-y += -I $(src) # NVKM - HW resource manager
drivers/gpu/drm/nouveau/include/nvkm/subdev/gsp.h +6 -7
@@ -210,10 +210,7 @@ struct nvkm_gsp { } gr; const struct nvkm_gsp_rm { - void *(*rpc_get)(struct nvkm_gsp *, u32 fn, u32 argc); - void *(*rpc_push)(struct nvkm_gsp *gsp, void *argv, - enum nvkm_gsp_rpc_reply_policy policy, u32 repc); - void (*rpc_done)(struct nvkm_gsp *gsp, void *repv); + const struct nvkm_rm_api *api; void *(*rm_ctrl_get)(struct nvkm_gsp_object *, u32 cmd, u32 argc); int (*rm_ctrl_push)(struct nvkm_gsp_object *, void **argv, u32 repc);
@@ -272,17 +269,19 @@ nvkm_gsp_rm(struct nvkm_gsp *gsp)return gsp && (gsp->fws.rm || gsp->fw.img); } +#include <rm/rm.h> +static inline void *nvkm_gsp_rpc_get (struct nvkm_gsp *gsp,u32 fn,u32 argc) { -return gsp->rm->rpc_get (gsp, fn, argc); +return gsp->rm->api->rpc->get (gsp, fn, argc); }static inline void *nvkm_gsp_rpc_push (struct nvkm_gsp *gsp,void *argv,enum nvkm_gsp_rpc_reply_policy policy,u32 repc) { -return gsp->rm->rpc_push (gsp, argv, policy, repc); +return gsp->rm->api->rpc->push (gsp, argv, policy, repc); } static inline void *
@@ -311,7 +310,7 @@ nvkm_gsp_rpc_wr(struct nvkm_gsp *gsp, void *argv,static inline void nvkm_gsp_rpc_done (struct nvkm_gsp *gsp,void *repv) { - gsp->rm->rpc_done (gsp, repv); + gsp->rm->api->rpc->done (gsp, repv); } static inline void *
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/Kbuild +2 -0
@@ -10,3 +10,5 @@ nvkm-y += nvkm/subdev/gsp/ga102.o nvkm-y += nvkm/subdev/gsp/ad102.o nvkm-y += nvkm/subdev/gsp/r535.o + +include $(src)/nvkm/subdev/gsp/rm/Kbuild
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/r535.c +2 -663
685 lines not on this page · raw message
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/Kbuild new file +5 -0
@@ -0,0 +1,5 @@ +# SPDX-License-Identifier: MIT +# +# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. + +include $(src)/nvkm/subdev/gsp/rm/r535/Kbuild
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/Kbuild new file +6 -0
@@ -0,0 +1,6 @@ +# SPDX-License-Identifier: MIT +# +# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. + +nvkm-y += nvkm/subdev/gsp/rm/r535/rm.o +nvkm-y += nvkm/subdev/gsp/rm/r535/rpc.o
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rm.c new file +10 -0
@@ -0,0 +1,10 @@ +/* SPDX-License-Identifier: MIT +* +* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. +*/ +#include <rm/rm.h> + +const struct nvkm_rm_api +r535_rm = { + .rpc = &r535_rpc, +};
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rpc.c new file +692 -0
692 lines not on this page · raw message
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rm.h new file +20 -0
@@ -0,0 +1,20 @@ +/* SPDX-License-Identifier: MIT +* +* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. +*/ +#include <subdev/gsp.h> +#ifndef __NVKM_RM_H__ +#define __NVKM_RM_H__ + +struct nvkm_rm_api { +const struct nvkm_rm_api_rpc { +void *(*get)(struct nvkm_gsp *,u32 fn,u32 argc); +void *(*push)(struct nvkm_gsp *gsp,void *argv, +enum nvkm_gsp_rpc_reply_policy policy,u32 repc); +void (*done)(struct nvkm_gsp *gsp,void *repv); + } *rpc; +}; + +extern const struct nvkm_rm_api r535_rm; +extern const struct nvkm_rm_api_rpc r535_rpc; +#endif
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/rpc.h new file +18 -0
@@ -0,0 +1,18 @@ +/* SPDX-License-Identifier: MIT +* +* Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. +*/ +#ifndef __NVKM_RM_RPC_H__ +#define __NVKM_RM_RPC_H__ +#include "rm.h" + +#define to_payload_hdr (p, header) \ + container_of((void *)p, typeof(*header), params) + +int r535_gsp_rpc_poll (struct nvkm_gsp *,u32 fn); + +struct nvfw_gsp_rpc *r535_gsp_msg_recv (struct nvkm_gsp *,int fn,u32 gsp_rpc_len); +int r535_gsp_msg_ntfy_add (struct nvkm_gsp *,u32 fn,nvkm_gsp_msg_ntfy_func ,void *priv); + +int r535_rpc_status_to_errno (uint32_t rpc_status); +#endif
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 513/592 drm/nouveau: fix a use-after-free in r535_gsp_rpc_push()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Zhi Wang <[email protected]>
[ Upstream commit 9802f0a63b641f4cddb2139c814c2e95cb825099 ]
The RPC container is released after being passed to r535_gsp_rpc_send().
When sending the initial fragment of a large RPC and passing the caller's RPC container, the container will be freed prematurely. Subsequent attempts to send remaining fragments will therefore result in a use-after-free.
Allocate a temporary RPC container for holding the initial fragment of a large RPC when sending. Free the caller's container when all fragments are successfully sent.
--- .../drm/nouveau/nvkm/subdev/gsp/rm/r535/rpc.c | 17 ++++++++++++----- 1 file changed, 12 insertions(+), 5 deletions(-)
drivers/gpu/drm/nouveau/nvkm/subdev/gsp/rm/r535/rpc.c +12 -5
@@ -638,12 +638,18 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload,if (payload_size > max_payload_size) {const u32 fn = rpc->function;u32 remain_payload_size = payload_size; +void *next; -/* Adjust length, and send initial RPC. */ - rpc->length =sizeof (*rpc) + max_payload_size; - msg->checksum = rpc->length; +/* Send initial RPC. */ + next =r535_gsp_rpc_get (gsp, fn, max_payload_size); +if (IS_ERR (next)) { + repv = next; + goto done; + } - repv =r535_gsp_rpc_send (gsp, payload,NVKM_GSP_RPC_REPLY_NOWAIT ,0 ); +memcpy (next, payload, max_payload_size); + + repv =r535_gsp_rpc_send (gsp, next,NVKM_GSP_RPC_REPLY_NOWAIT ,0 );if (IS_ERR (repv)) goto done;
@@ -654,7 +660,6 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload,while (remain_payload_size) {u32 size =min (remain_payload_size, max_payload_size); -void *next; next =r535_gsp_rpc_get (gsp,NV_VGPU_MSG_FUNCTION_CONTINUATION_RECORD , size);if (IS_ERR (next)) {
@@ -675,6 +680,8 @@ r535_gsp_rpc_push(struct nvkm_gsp *gsp, void *payload,/* Wait for reply. */ repv =r535_gsp_rpc_handle_reply (gsp, fn, policy, payload_size +sizeof (*rpc)); +if (!IS_ERR (repv)) +kvfree (msg); }else { repv =r535_gsp_rpc_send (gsp, payload, policy, gsp_rpc_len); }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 514/592 drm/nouveau/bl: increase buffer size to avoid truncate warning
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jacob Keller <[email protected]>
[ Upstream commit 61b2b3737499f1fb361a54a16828db24a8345e85 ]
The nouveau_get_backlight_name() function generates a unique name for the backlight interface, appending an id from 1 to 99 for all backlight devices after the first.
GCC 15 (and likely other compilers) produce the following -Wformat-truncation warning:
nouveau_backlight.c: In function ‘nouveau_backlight_init’:
nouveau_backlight.c:56:69: error: ‘%d’ directive output may be truncated writing between 1 and 10 bytes into a region of size 3 [-Werror=format-truncation=]
56 | snprintf(backlight_name, BL_NAME_SIZE, "nv_backlight%d", nb);
| ^~
In function ‘nouveau_get_backlight_name’,
inlined from ‘nouveau_backlight_init’ at nouveau_backlight.c:351:7:
nouveau_backlight.c:56:56: note: directive argument in the range [1, 2147483647]
56 | snprintf(backlight_name, BL_NAME_SIZE, "nv_backlight%d", nb);
| ^~~~~~~~~~~~~~~~
nouveau_backlight.c:56:17: note: ‘snprintf’ output between 14 and 23 bytes into a destination of size 15
56 | snprintf(backlight_name, BL_NAME_SIZE, "nv_backlight%d", nb);
| ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~The warning started appearing after commit ab244be47a8f ("drm/nouveau:
Fix a potential theorical leak in nouveau_get_backlight_name()") This fix for the ida usage removed the explicit value check for ids larger than 99. The compiler is unable to intuit that the ida_alloc_max() limits the returned value range between 0 and 99.
Because the compiler can no longer infer that the number ranges from 0 to 99, it thinks that it could use as many as 11 digits (10 + the potential - sign for negative numbers).
The warning has gone unfixed for some time, with at least one kernel test robot report. The code breaks W=1 builds, which is especially frustrating with the introduction of CONFIG_WERROR.
The string is stored temporarily on the stack and then copied into the device name. Its not a big deal to use 11 more bytes of stack rounding out to an even 24 bytes. Increase BL_NAME_SIZE to 24 to avoid the truncation warning. This fixes the W=1 builds that include this driver.
Compile tested only.
--- drivers/gpu/drm/nouveau/nouveau_backlight.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/nouveau/nouveau_backlight.c +1 -1
@@ -42,7 +42,7 @@#include "nouveau_acpi.h" static struct ida bl_ida; -#define BL_NAME_SIZE 15 // 12 for name + 2 for digits + 1 for '\0' +#define BL_NAME_SIZE 24 // 12 for name + 11 for digits + 1 for '\0' static bool nouveau_get_backlight_name(char backlight_name[BL_NAME_SIZE],
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 515/592 rust: devres: fix race in Devres::drop()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Danilo Krummrich <[email protected]>
[ Upstream commit f744201c6159fc7323c40936fd079525f7063598 ]
In Devres::drop() we first remove the devres action and then drop the wrapped device resource.
The design goal is to give the owner of a Devres object control over when the device resource is dropped, but limit the overall scope to the corresponding device being bound to a driver.
However, there's a race that was introduced with commit 8ff656643d30 ("rust: devres: remove action in Devres::drop"), but also has been (partially) present from the initial version on.
In Devres::drop(), the devres action is removed successfully and subsequently the destructor of the wrapped device resource runs. However, there is no guarantee that the destructor of the wrapped device resource completes before the driver core is done unbinding the corresponding device.
If in Devres::drop(), the devres action can't be removed, it means that the devres callback has been executed already, or is still running concurrently. In case of the latter, either Devres::drop() wins revoking the Revocable or the devres callback wins revoking the Revocable. If Devres::drop() wins, we (again) have no guarantee that the destructor of the wrapped device resource completes before the driver core is done unbinding the corresponding device.
CPU0 CPU1
------------------------------------------------------------------------Devres::drop() { Devres::devres_callback() {
self.data.revoke() { this.data.revoke() {
is_available.swap() == true
is_available.swap == false
}
}
// [...]
// device fully unbound
drop_in_place() {
// release device resource
} }
}
Depending on the specific device resource, this can potentially lead to user-after-free bugs.
In order to fix this, implement the following logic.
In the devres callback, we're always good when we get to revoke the device resource ourselves, i.e. Revocable::revoke() returns true.
If Revocable::revoke() returns false, it means that Devres::drop(), concurrently, already drops the device resource and we have to wait for Devres::drop() to signal that it finished dropping the device resource.
Note that if we hit the case where we need to wait for the completion of Devres::drop() in the devres callback, it means that we're actually racing with a concurrent Devres::drop() call, which already started revoking the device resource for us. This is rather unlikely and means that the concurrent Devres::drop() already started doing our work and we just need to wait for it to complete it for us. Hence, there should not be any additional overhead from that.
(Actually, for now it's even better if Devres::drop() does the work for us, since it can bypass the synchronize_rcu() call implied by Revocable::revoke(), but this goes away anyways once I get to implement the split devres callback approach, which allows us to first flip the atomics of all registered Devres objects of a certain device, execute a single synchronize_rcu() and then drop all revocable objects.)
In Devres::drop() we try to revoke the device resource. If that is *not* successful, it means that the devres callback already did and we're good.
Otherwise, we try to remove the devres action, which, if successful, means that we're good, since the device resource has just been revoked by us *before* we removed the devres action successfully.
If the devres action could not be removed, it means that the devres callback must be running concurrently, hence we signal that the device resource has been revoked by us, using the completion.
This makes it safe to drop a Devres object from any task and at any point of time, which is one of the design goals.
--- rust/kernel/devres.rs | 37 +++++++++++++++++++++++++++++-------- 1 file changed, 29 insertions(+), 8 deletions(-)
rust/kernel/devres.rs +29 -8
@@ -13,7 +13,7 @@ ffi::c_void, prelude::*, revocable::Revocable, - sync::Arc, + sync::{Arc, Completion}, types::ARef, };
@@ -25,13 +25,17 @@ struct DevresInner<T> { callback: unsafe extern "C" fn(*mut c_void),#[pin] data: Revocable<T>, +#[pin] + revoke: Completion, }/// This abstraction is meant to be used by subsystems to containerize [`Device`] bound resources to /// manage their lifetime. /// /// [`Device`] bound resources should be freed when either the resource goes out of scope or the -/// [`Device`] is unbound respectively, depending on what happens first. +/// [`Device`] is unbound respectively, depending on what happens first. In any case, it is always +/// guaranteed that revoking the device resource is completed before the corresponding [`Device`] +/// is unbound. /// /// To achieve that [`Devres`] registers a devres callback on creation, which is called once the /// [`Device`] is unbound, revoking access to the encapsulated resource (see also [`Revocable`]).
@@ -105,6 +109,7 @@ fn new(dev: &Device, data: T, flags: Flags) -> Result<Arc<DevresInner<T>>> { dev: dev.into(), callback: Self::devres_callback, data <- Revocable::new(data), + revoke <- Completion::new(), }), flags, )?;
@@ -133,26 +138,28 @@ fn as_ptr(&self) -> *const Self { selfas _ } -fn remove_action (this: &Arc <Self >) { +fn remove_action (this: &Arc <Self >) ->bool {// SAFETY: // - `self.inner.dev` is a valid `Device`, // - the `action` and `data` pointers are the exact same ones as given to devm_add_action() // previously, // - `self` is always valid, even if the action has been released already. -let ret =unsafe { +let success =unsafe { bindings::devm_remove_action_nowarn ( this.dev.as_raw (),Some (this.callback), this.as_ptr ()as _ , ) - }; + } ==0 ; -if ret ==0 { +if success {// SAFETY: We leaked an `Arc` reference to devm_add_action() in `DevresInner::new`; if // devm_remove_action_nowarn() was successful we can (and have to) claim back ownership // of this reference. let _ =unsafe {Arc ::from_raw (this.as_ptr ()) }; } + + success }#[allow(clippy::missing_safety_doc)]
@@ -164,7 +171,12 @@ fn remove_action(this: &Arc<Self>) {// `DevresInner::new`. let inner =unsafe {Arc ::from_raw (ptr) }; - inner.data.revoke (); +if !inner.data.revoke () { +// If `revoke()` returns false, it means that `Devres::drop` already started revoking +// `inner.data` for us. Hence we have to wait until `Devres::drop()` signals that it +// completed revoking `inner.data`. + inner.revoke.wait_for_completion (); + } } }
@@ -196,6 +208,15 @@ fn deref(&self) -> &Self::Target {impl <T >Drop for Devres <T > {fn drop (&mut self) { -DevresInner ::remove_action (&self.0 ); +// SAFETY: When `drop` runs, it is guaranteed that nobody is accessing the revocable data +// anymore, hence it is safe not to wait for the grace period to finish. +if unsafe { self.revoke_nosync () } { +// We revoked `self.0.data` before the devres action did, hence try to remove it. +if !DevresInner ::remove_action (&self.0 ) { +// We could not remove the devres action, which means that it now runs concurrently, +// hence signal that `self.0.data` has been revoked successfully. + self.0 .revoke.complete_all (); + } + } } }
Signature · 1 line
2.39.5
Re: [PATCH 6.15 515/592] rust: devres: fix race in Devres::drop()
Quoted 98 lines · Greg Kroah-Hartman
6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Danilo Krummrich <[email protected]>
[ Upstream commit
f744201c6159fc7323c40936fd079525f7063598]In
Devres::drop()we first remove the devres action and then drop the wrapped device resource.The design goal is to give the owner of a Devres object control over when the device resource is dropped, but limit the overall scope to the corresponding device being bound to a driver.
However, there's a race that was introduced with commit
8ff656643d30("rust: devres: remove action inDevres::drop"), but also has been (partially) present from the initial version on.In
Devres::drop(), the devres action is removed successfully and subsequently the destructor of the wrapped device resource runs. However, there is no guarantee that the destructor of the wrapped device resource completes before the driver core is done unbinding the corresponding device.If in
Devres::drop(), the devres action can't be removed, it means that the devres callback has been executed already, or is still running concurrently. In case of the latter, eitherDevres::drop()wins revoking the Revocable or the devres callback wins revoking the Revocable. IfDevres::drop()wins, we (again) have no guarantee that the destructor of the wrapped device resource completes before the driver core is done unbinding the corresponding device.CPU0 CPU1
------------------------------------------------------------------------Devres::drop() { Devres::devres_callback() { self.data.revoke() { this.data.revoke() { is_available.swap() == true is_available.swap == false } } // [...] // device fully unbound drop_in_place() { // release device resource } } }Depending on the specific device resource, this can potentially lead to user-after-free bugs.
In order to fix this, implement the following logic.
In the devres callback, we're always good when we get to revoke the device resource ourselves, i.e.
Revocable::revoke()returns true.If
Revocable::revoke()returns false, it means thatDevres::drop(), concurrently, already drops the device resource and we have to wait forDevres::drop()to signal that it finished dropping the device resource.Note that if we hit the case where we need to wait for the completion of
Devres::drop()in the devres callback, it means that we're actually racing with a concurrentDevres::drop()call, which already started revoking the device resource for us. This is rather unlikely and means that the concurrentDevres::drop()already started doing our work and we just need to wait for it to complete it for us. Hence, there should not be any additional overhead from that.(Actually, for now it's even better if
Devres::drop()does the work for us, since it can bypass thesynchronize_rcu()call implied byRevocable::revoke(), but this goes away anyways once I get to implement the split devres callback approach, which allows us to first flip the atomics of all registered Devres objects of a certain device, execute a singlesynchronize_rcu()and then drop all revocable objects.)In
Devres::drop()we try to revoke the device resource. If that is *not* successful, it means that the devres callback already did and we're good.Otherwise, we try to remove the devres action, which, if successful, means that we're good, since the device resource has just been revoked by us *before* we removed the devres action successfully.
If the devres action could not be removed, it means that the devres callback must be running concurrently, hence we signal that the device resource has been revoked by us, using the completion.
This makes it safe to drop a Devres object from any task and at any point of time, which is one of the design goals.
Fixes 76c01ded724b ("rust: add devres abstraction")Reported-by Alice Ryhl <[email protected]>Reviewed-by Benno Lossin <[email protected]>Signed-off-by Danilo Krummrich <[email protected]>Signed-off-by Sasha Levin <[email protected]>
Quoted 3 lines
--- rust/kernel/devres.rs | 37 +++++++++++++++++++++++++++++-------- 1 file changed, 29 insertions(+), 8 deletions(-)
This is missing the prerequisite patch #1 from
https://lore.kernel.org/all/[email protected]
---
Cheers,
Benno
Re: [PATCH 6.15 515/592] rust: devres: fix race in Devres::drop()
[CCing Miguel (JFYI) as well as Christian, who reported the build error[1] with 6.15.4-rc1 (which I'm seeing as well[2]) caused by the patch this mail is about according to Benno.]
[1] https://lore.kernel.org/all/[email protected]/
Quoted 12 lines · Benno Lossin
Quoted 11 lines · Greg Kroah-Hartman
6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Danilo Krummrich <[email protected]>
[ Upstream commit
f744201c6159fc7323c40936fd079525f7063598]In
Devres::drop()we first remove the devres action and then drop the wrapped device resource.
Quoted 7 lines
Quoted 3 lines
The design goal is to give the owner of a Devres object control over when the device resource is dropped, but limit the overall scope to the corresponding device being bound to a driver.
[...]
This is missing the prerequisite patch #1 fromhttps://lore.kernel.org/all/[email protected]
You afaics mean 1b56e765bf8990 ("rust: completion: implement initial abstraction") [v6.16-rc3] – which did not cleanly apply to 6.15.4-rc1 in a quick test; it was also not possible to revert this patch ("rust:
devres: fix race in Devres::drop()") cleanly; reverting worked after reverting "rust: devres: do not dereference to the internal Revocable" First, but that lead to another build error:
https://download.copr.fedorainfracloud.org/results/@kernel-vanilla/fedora-rc/fedora-42-x86_64/09202837-stablerc-fedorarc-releases/builder-live.log.gz
Ciao, Thorsten
Re: [PATCH 6.15 515/592] rust: devres: fix race in Devres::drop()
Quoted 3 lines · Thorsten Leemhuis
[CCing Miguel (JFYI) as well as Christian, who reported the build error[1] with 6.15.4-rc1 (which I'm seeing as well[2]) caused by the patch this mail is about according to Benno.]
Thanks!
Quoted 20 lines
[1] https://lore.kernel.org/all/[email protected]/
Quoted 15 lines · Benno Lossin
Quoted 14 lines · Greg Kroah-Hartman
6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Danilo Krummrich <[email protected]>
[ Upstream commit
f744201c6159fc7323c40936fd079525f7063598]In
Devres::drop()we first remove the devres action and then drop the wrapped device resource.The design goal is to give the owner of a Devres object control over when the device resource is dropped, but limit the overall scope to the corresponding device being bound to a driver.
Quoted 7 lines
Quoted 4 lines
[...]
This is missing the prerequisite patch #1 fromhttps://lore.kernel.org/all/[email protected]You afaics mean
1b56e765bf8990("rust: completion: implement initial abstraction") [v6.16-rc3]
Yes that is the prerequisite.
Quoted 1 line
– which did not cleanly apply to 6.15.4-rc1 in
In which repository is that tag? I didn't find it in the stable tree.
I tried applying it on top of v6.15.3 and that also results in a conflict, but only in bindgen_helpers.h and helpers.c, so we can simply provide a fixed patch.
@Danilo, I think this should be backported, how do you want to proceed?
---
Cheers,
Benno
Quoted 7 lines
a quick test; it was also not possible to revert this patch ("rust:
devres: fix race inDevres::drop()") cleanly; reverting worked after reverting "rust: devres: do not dereference to the internal Revocable" First, but that lead to another build error:
https://download.copr.fedorainfracloud.org/results/@kernel-vanilla/fedora-rc/fedora-42-x86_64/09202837-stablerc-fedorarc-releases/builder-live.log.gzCiao, Thorsten
Re: [PATCH 6.15 515/592] rust: devres: fix race in Devres::drop()
Quoted 27 lines · Benno Lossin
Quoted 3 lines · Thorsten Leemhuis
[CCing Miguel (JFYI) as well as Christian, who reported the build error[1] with 6.15.4-rc1 (which I'm seeing as well[2]) caused by the patch this mail is about according to Benno.]
Thanks!
Quoted 21 lines · Benno Lossin
Quoted 19 lines · Greg Kroah-Hartman
Quoted 6 lines
6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Danilo Krummrich <[email protected]>
Quoted 8 lines
[ Upstream commit
f744201c6159fc7323c40936fd079525f7063598]In
Devres::drop()we first remove the devres action and then drop the wrapped device resource.The design goal is to give the owner of a Devres object control over when the device resource is dropped, but limit the overall scope to the corresponding device being bound to a driver.
[...]
This is missing the prerequisite patch #1 fromhttps://lore.kernel.org/all/[email protected]
Quoted 8 lines
Quoted 2 lines
You afaics mean
1b56e765bf8990("rust: completion: implement initial abstraction") [v6.16-rc3]Yes that is the prerequisite.
Quoted 1 line
– which did not cleanly apply to 6.15.4-rc1 in
In which repository is that tag? I didn't find it in the stable tree.
Use this tree and branch:
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/log/?h=linux-6.15.y
There sadly is no tag ("send a patch to Greg's rc-release scripts to make them tag -rc releases" is somewhere near the end of my todo list for a while already...).
Quoted 3 lines
I tried applying it on top of v6.15.3 and that also results in a conflict, but only in
bindgen_helpers.handhelpers.c, so we can simply provide a fixed patch.
Yeah, that likely is needed to make Greg happy here.
Quoted 1 line
@Danilo, I think this should be backported, how do you want to proceed?
Ciao, Thorsten
Greg Kroah-Hartman
maintainerRe: [PATCH 6.15 515/592] rust: devres: fix race in Devres::drop()
Quoted 45 lines · Thorsten Leemhuis
Quoted 27 lines · Benno Lossin
Quoted 3 lines · Thorsten Leemhuis
[CCing Miguel (JFYI) as well as Christian, who reported the build error[1] with 6.15.4-rc1 (which I'm seeing as well[2]) caused by the patch this mail is about according to Benno.]
Thanks!
Quoted 21 lines · Benno Lossin
Quoted 19 lines · Greg Kroah-Hartman
Quoted 6 lines
6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Danilo Krummrich <[email protected]>
Quoted 8 lines
[ Upstream commit
f744201c6159fc7323c40936fd079525f7063598]In
Devres::drop()we first remove the devres action and then drop the wrapped device resource.The design goal is to give the owner of a Devres object control over when the device resource is dropped, but limit the overall scope to the corresponding device being bound to a driver.
[...]
This is missing the prerequisite patch #1 fromhttps://lore.kernel.org/all/[email protected]Quoted 8 lines
Quoted 2 lines
You afaics mean
1b56e765bf8990("rust: completion: implement initial abstraction") [v6.16-rc3]Yes that is the prerequisite.
Quoted 1 line
– which did not cleanly apply to 6.15.4-rc1 in
In which repository is that tag? I didn't find it in the stable tree.
Use this tree and branch:
https://git.kernel.org/pub/scm/linux/kernel/git/stable/linux-stable-rc.git/log/?h=linux-6.15.y
There sadly is no tag ("send a patch to Greg's rc-release scripts to make them tag -rc releases" is somewhere near the end of my todo list for a while already...).
Quoted 5 lines
Quoted 3 lines
I tried applying it on top of v6.15.3 and that also results in a conflict, but only in
bindgen_helpers.handhelpers.c, so we can simply provide a fixed patch.Yeah, that likely is needed to make Greg happy here.
For now I'll just go drop this commit and wait for it to be submitted in a series that actually builds.
And I need to go fix my test-server to actually build the rust code, I didn't realize that wasn't happening and is a huge gap in my test builds...
thanks,
greg k-h
Re: [PATCH 6.15 515/592] rust: devres: fix race in Devres::drop()
Quoted 10 lines · Greg Kroah-Hartman
Quoted 6 lines · Thorsten Leemhuis
Quoted 3 lines · Benno Lossin
I tried applying it on top of v6.15.3 and that also results in a conflict, but only in
bindgen_helpers.handhelpers.c, so we can simply provide a fixed patch.Yeah, that likely is needed to make Greg happy here.
For now I'll just go drop this commit and wait for it to be submitted in a series that actually builds.
I'll send a series based on v6.15.3.
The dependencies you already dropped, i.e.
queue-6.15/rust-devres-fix-doctest-build-under-config_pci.patch queue-6.15/rust-devres-implement-devres-access.patch
won't be needed any more.
Greg Kroah-Hartman
maintainerRe: [PATCH 6.15 515/592] rust: devres: fix race in Devres::drop()
Quoted 12 lines · Danilo Krummrich
Quoted 10 lines · Greg Kroah-Hartman
Quoted 6 lines · Thorsten Leemhuis
Quoted 3 lines · Benno Lossin
I tried applying it on top of v6.15.3 and that also results in a conflict, but only in
bindgen_helpers.handhelpers.c, so we can simply provide a fixed patch.Yeah, that likely is needed to make Greg happy here.
For now I'll just go drop this commit and wait for it to be submitted in a series that actually builds.
Quoted 9 lines
I'll send a series based on v6.15.3.
The dependencies you already dropped, i.e.
queue-6.15/rust-devres-fix-doctest-build-under-config_pci.patchqueue-6.15/rust-devres-implement-devres-access.patchwon't be needed any more.
Great, thanks for doing that.
greg k-h
Greg Kroah-Hartman
maintainerPATCH 516/592 rust: devres: implement Devres::access()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Danilo Krummrich <[email protected]>
[ Upstream commit f301cb978c068faa8fcd630be2cb317a2d0ec063 ]
Implement a direct accessor for the data stored within the Devres for cases where we can prove that we own a reference to a Device<Bound> (i.e. a bound device) of the same device that was used to create the corresponding Devres container.
Usually, when accessing the data stored within a Devres container, it is not clear whether the data has been revoked already due to the device being unbound and, hence, we have to try whether the access is possible and subsequently keep holding the RCU read lock for the duration of the access.
However, when we can prove that we hold a reference to Device<Bound> matching the device the Devres container has been created with, we can guarantee that the device is not unbound for the duration of the lifetime of the Device<Bound> reference and, hence, it is not possible for the data within the Devres container to be revoked.
Therefore, in this case, we can bypass the atomic check and the RCU read lock, which is a great optimization and simplification for drivers.
Stable-dep-of: 20c96ed278e3 ("rust: devres: do not dereference to the internal Revocable")
--- rust/kernel/devres.rs | 38 ++++++++++++++++++++++++++++++++++++++ 1 file changed, 38 insertions(+)
rust/kernel/devres.rs +38 -0
@@ -196,6 +196,44 @@ pub fn new_foreign_owned(dev: &Device, data: T, flags: Flags) -> Result {Ok (()) } + +/// Obtain `&'a T`, bypassing the [`Revocable`]. +/// +/// This method allows to directly obtain a `&'a T`, bypassing the [`Revocable`], by presenting +/// a `&'a Device<Bound>` of the same [`Device`] this [`Devres`] instance has been created with. +/// +/// # Errors +/// +/// An error is returned if `dev` does not match the same [`Device`] this [`Devres`] instance +/// has been created with. +/// +/// # Example +/// +/// ```no_run +/// # use kernel::{device::Core, devres::Devres, pci}; +/// +/// fn from_core(dev: &pci::Device<Core>, devres: Devres<pci::Bar<0x4>>) -> Result { +/// let bar = devres.access(dev.as_ref())?; +/// +/// let _ = bar.read32(0x0); +/// +/// // might_sleep() +/// +/// bar.write32(0x42, 0x0); +/// +/// Ok(()) +/// } +/// ``` +pub fn access <'a>(&'a self, dev: &'aDevice <Bound >) ->Result <&'aT > { +if self.0 .dev.as_raw () != dev.as_raw () { +return Err (EINVAL ); + } + +// SAFETY: `dev` being the same device as the device this `Devres` has been created for +// proves that `self.0.data` hasn't been revoked and is guaranteed to not be revoked as +// long as `dev` lives; `dev` lives at least as long as `self`. +Ok (unsafe { self.deref ().access () }) + } }impl <T >Deref for Devres <T > {
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 517/592 rust: devres: do not dereference to the internal Revocable
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Danilo Krummrich <[email protected]>
[ Upstream commit 20c96ed278e362ae4e324ed7d8c69fb48c508d3c ]
We can't expose direct access to the internal Revocable, since this allows users to directly revoke the internal Revocable without Devres having the chance to synchronize with the devres callback -- we have to guarantee that the internal Revocable has been fully revoked before the device is fully unbound.
Hence, remove the corresponding Deref implementation and, instead, provide indirect accessors for the internal Revocable.
Note that we can still support Devres::revoke() by implementing the required synchronization (which would be almost identical to the synchronization in Devres::drop()).
--- rust/kernel/devres.rs | 27 ++++++++++++++++----------- 1 file changed, 16 insertions(+), 11 deletions(-)
rust/kernel/devres.rs +16 -11
@@ -12,13 +12,11 @@ error::{Error, Result}, ffi::c_void, prelude::*, - revocable::Revocable, - sync::{Arc, Completion}, + revocable::{Revocable, RevocableGuard}, + sync::{rcu, Arc, Completion}, types::ARef, }; -use core::ops::Deref ; -#[pin_data] struct DevresInner <T > { dev:ARef <Device >,
@@ -232,15 +230,22 @@ pub fn access<'a>(&'a self, dev: &'a Device<Bound>) -> Result<&'a T> {// SAFETY: `dev` being the same device as the device this `Devres` has been created for // proves that `self.0.data` hasn't been revoked and is guaranteed to not be revoked as // long as `dev` lives; `dev` lives at least as long as `self`. -Ok (unsafe { self.deref ().access () }) +Ok (unsafe { self.0 .data.access () }) } -} -impl <T >Deref for Devres <T > { -type Target =Revocable <T >; +/// [`Devres`] accessor for [`Revocable::try_access`]. +pub fn try_access (&self) ->Option <RevocableGuard <'_,T >> { + self.0 .data.try_access () + } + +/// [`Devres`] accessor for [`Revocable::try_access_with`]. +pub fn try_access_with <R ,F :FnOnce (&T ) ->R >(&self, f:F ) ->Option <R > { + self.0 .data.try_access_with (f) + } -fn deref (&self) -> &Self ::Target { - &self.0 .data +/// [`Devres`] accessor for [`Revocable::try_access_with_guard`]. +pub fn try_access_with_guard <'a>(&'a self, guard: &'a rcu::Guard ) ->Option <&'aT > { + self.0 .data.try_access_with_guard (guard) } }
@@ -248,7 +253,7 @@ impl<T> Drop for Devres<T> {fn drop (&mut self) {// SAFETY: When `drop` runs, it is guaranteed that nobody is accessing the revocable data // anymore, hence it is safe not to wait for the grace period to finish. -if unsafe { self.revoke_nosync () } { +if unsafe { self.0 .data.revoke_nosync () } {// We revoked `self.0.data` before the devres action did, hence try to remove it. if !DevresInner ::remove_action (&self.0 ) {// We could not remove the devres action, which means that it now runs concurrently,
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 518/592 drm/i915/pmu: Fix build error with GCOV and AutoFDO enabled
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Tzung-Bi Shih <[email protected]>
[ Upstream commit a7137b1825b535eb7258b25beeb0d5425e0037d2 ]
i915_pmu.c may fail to build with GCOV and AutoFDO enabled.
../drivers/gpu/drm/i915/i915_pmu.c:116:3: error: call to '__compiletime_assert_487' declared with 'error' attribute: BUILD_BUG_ON failed: bit > BITS_PER_TYPE(typeof_member(struct i915_pmu, enable)) - 1
116 | BUILD_BUG_ON(bit >
| ^Here is a way to reproduce the issue:
$ git checkout v6.15 $ mkdir build $ ./scripts/kconfig/merge_config.sh -O build -n -m <(cat <<EOF CONFIG_DRM=y CONFIG_PCI=y CONFIG_DRM_I915=y
CONFIG_PERF_EVENTS=y
CONFIG_DEBUG_FS=yCONFIG_GCOV_KERNEL=yCONFIG_GCOV_PROFILE_ALL=y
CONFIG_AUTOFDO_CLANG=y
EOF
)
$ PATH=${PATH}:${HOME}/llvm-20.1.5-x86_64/bin make LLVM=1 O=build \
olddefconfig
$ PATH=${PATH}:${HOME}/llvm-20.1.5-x86_64/bin make LLVM=1 O=build \
CLANG_AUTOFDO_PROFILE=...PATH_TO_SOME_AFDO_PROFILE... \
drivers/gpu/drm/i915/i915_pmu.oAlthough not super sure what happened, by reviewing the code, it should depend on __builtin_constant_p(bit) directly instead of assuming __builtin_constant_p(config) makes bit a builtin constant.
Also fix a nit, to reuse the bit local variable.
--- drivers/gpu/drm/i915/i915_pmu.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/gpu/drm/i915/i915_pmu.c +2 -2
@@ -112,7 +112,7 @@ static u32 config_mask(const u64 config) {unsigned int bit =config_bit (config); -if (__builtin_constant_p (config)) +if (__builtin_constant_p (bit))BUILD_BUG_ON (bit > BITS_PER_TYPE(typeof_member(struct i915_pmu, enable)) -1 );
@@ -121,7 +121,7 @@ static u32 config_mask(const u64 config)BITS_PER_TYPE (typeof_member (struct i915_pmu , enable)) - 1); -return BIT (config_bit (config)); +return BIT (bit); }static bool is_engine_event (struct perf_event *event)
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 519/592 hwmon: (occ) Rework attribute registration for stack usage
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit 744c2fe950e936c4d62430de899d6253424200ed ]
clang produces an output with excessive stack usage when building the occ_setup_sensor_attrs() function, apparently the result of having a lot of struct literals and building with the -fno-strict-overflow option that leads clang to skip some optimization in case the 'attr' pointer overruns:
drivers/hwmon/occ/common.c:775:12: error: stack frame size (1392) exceeds limit (1280) in 'occ_setup_sensor_attrs' [-Werror,-Wframe-larger-than]
Replace the custom macros for initializing the attributes with a simpler function call that does not run into this corner case.
--- drivers/hwmon/occ/common.c | 212 +++++++++++++++---------------------- 1 file changed, 85 insertions(+), 127 deletions(-)
drivers/hwmon/occ/common.c +85 -127
@@ -747,29 +747,30 @@ static ssize_t occ_show_extended(struct device *dev, }/* -* Some helper macros to make it easier to define an occ_attribute. Since these -* are dynamically allocated, we shouldn't use the existing kernel macros which +* A helper to make it easier to define an occ_attribute. Since these +* are dynamically allocated, we cannot use the existing kernel macros which * stringify the name argument. */ -#define ATTR_OCC (_name, _mode, _show, _store) { \ - .attr = { \ - .name = _name, \ - .mode = VERIFY_OCTAL_PERMISSIONS(_mode), \ - }, \ - .show = _show, \ - .store = _store, \ -} - -#define SENSOR_ATTR_OCC (_name, _mode, _show, _store, _nr, _index) { \ - .dev_attr = ATTR_OCC(_name, _mode, _show, _store), \ - .index = _index, \ - .nr = _nr, \ +static void occ_init_attribute (struct occ_attribute *attr,int mode, +ssize_t (*show)(struct device *dev,struct device_attribute *attr,char *buf), +ssize_t (*store)(struct device *dev,struct device_attribute *attr, +const char *buf,size_t count), +int nr,int index,const char *fmt, ...) +{ +va_list args; + +va_start (args, fmt); +vsnprintf (attr->name,sizeof (attr->name), fmt, args); +va_end (args); + + attr->sensor.dev_attr.attr.name = attr->name; + attr->sensor.dev_attr.attr.mode = mode; + attr->sensor.dev_attr.show = show; + attr->sensor.dev_attr.store = store; + attr->sensor.index = index; + attr->sensor.nr = nr; } -#define OCC_INIT_ATTR (_name, _mode, _show, _store, _nr, _index) \ - ((struct sensor_device_attribute_2) \ - SENSOR_ATTR_OCC(_name, _mode, _show, _store, _nr, _index)) -/* * Allocate and instatiate sensor_device_attribute_2s. It's most efficient to * use our own instead of the built-in hwmon attribute types.
@@ -855,14 +856,15 @@ static int occ_setup_sensor_attrs(struct occ *occ) sensors->extended.num_sensors =0 ; } - occ->attrs =devm_kzalloc (dev,sizeof (*occ->attrs) * num_attrs, + occ->attrs =devm_kcalloc (dev, num_attrs,sizeof (*occ->attrs),GFP_KERNEL );if (!occ->attrs)return -ENOMEM ;/* null-terminated list */ - occ->group.attrs =devm_kzalloc (dev,sizeof (*occ->group.attrs) * - num_attrs +1 ,GFP_KERNEL ); + occ->group.attrs =devm_kcalloc (dev, num_attrs +1 , +sizeof (*occ->group.attrs), +GFP_KERNEL );if (!occ->group.attrs)return -ENOMEM ;
@@ -872,43 +874,33 @@ static int occ_setup_sensor_attrs(struct occ *occ) s = i +1 ; temp = ((struct temp_sensor_2 *)sensors->temp.data) + i; -snprintf (attr->name,sizeof (attr->name),"temp%d_label" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_temp,NULL , -0 , i); +occ_init_attribute (attr,0444 , show_temp,NULL , +0 , i,"temp%d_label" , s); attr++;if (sensors->temp.version ==2 && temp->fru_type ==OCC_FRU_TYPE_VRM ) { -snprintf (attr->name,sizeof (attr->name), -"temp%d_alarm" , s); +occ_init_attribute (attr,0444 , show_temp,NULL , +1 , i,"temp%d_alarm" , s); }else { -snprintf (attr->name,sizeof (attr->name), -"temp%d_input" , s); +occ_init_attribute (attr,0444 , show_temp,NULL , +1 , i,"temp%d_input" , s); } - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_temp,NULL , -1 , i); attr++;if (sensors->temp.version >1 ) { -snprintf (attr->name,sizeof (attr->name), -"temp%d_fru_type" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_temp,NULL ,2 , i); +occ_init_attribute (attr,0444 , show_temp,NULL , +2 , i,"temp%d_fru_type" , s); attr++; -snprintf (attr->name,sizeof (attr->name), -"temp%d_fault" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_temp,NULL ,3 , i); +occ_init_attribute (attr,0444 , show_temp,NULL , +3 , i,"temp%d_fault" , s); attr++;if (sensors->temp.version ==0x10 ) { -snprintf (attr->name,sizeof (attr->name), -"temp%d_max" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_temp,NULL , -4 , i); +occ_init_attribute (attr,0444 , show_temp,NULL , +4 , i,"temp%d_max" , s); attr++; } }
@@ -917,14 +909,12 @@ static int occ_setup_sensor_attrs(struct occ *occ)for (i =0 ; i < sensors->freq.num_sensors; ++i) { s = i +1 ; -snprintf (attr->name,sizeof (attr->name),"freq%d_label" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_freq,NULL , -0 , i); +occ_init_attribute (attr,0444 , show_freq,NULL , +0 , i,"freq%d_label" , s); attr++; -snprintf (attr->name,sizeof (attr->name),"freq%d_input" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_freq,NULL , -1 , i); +occ_init_attribute (attr,0444 , show_freq,NULL , +1 , i,"freq%d_input" , s); attr++; }
@@ -940,32 +930,24 @@ static int occ_setup_sensor_attrs(struct occ *occ) s = (i *4 ) +1 ;for (j =0 ; j <4 ; ++j) { -snprintf (attr->name,sizeof (attr->name), -"power%d_label" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_power,NULL , - nr++, i); +occ_init_attribute (attr,0444 , show_power, +NULL , nr++, i, +"power%d_label" , s); attr++; -snprintf (attr->name,sizeof (attr->name), -"power%d_average" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_power,NULL , - nr++, i); +occ_init_attribute (attr,0444 , show_power, +NULL , nr++, i, +"power%d_average" , s); attr++; -snprintf (attr->name,sizeof (attr->name), -"power%d_average_interval" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_power,NULL , - nr++, i); +occ_init_attribute (attr,0444 , show_power, +NULL , nr++, i, +"power%d_average_interval" , s); attr++; -snprintf (attr->name,sizeof (attr->name), -"power%d_input" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_power,NULL , - nr++, i); +occ_init_attribute (attr,0444 , show_power, +NULL , nr++, i, +"power%d_input" , s); attr++; s++;
@@ -977,28 +959,20 @@ static int occ_setup_sensor_attrs(struct occ *occ)for (i =0 ; i < sensors->power.num_sensors; ++i) { s = i +1 ; -snprintf (attr->name,sizeof (attr->name), -"power%d_label" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_power,NULL ,0 , i); +occ_init_attribute (attr,0444 , show_power,NULL , +0 , i,"power%d_label" , s); attr++; -snprintf (attr->name,sizeof (attr->name), -"power%d_average" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_power,NULL ,1 , i); +occ_init_attribute (attr,0444 , show_power,NULL , +1 , i,"power%d_average" , s); attr++; -snprintf (attr->name,sizeof (attr->name), -"power%d_average_interval" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_power,NULL ,2 , i); +occ_init_attribute (attr,0444 , show_power,NULL , +2 , i,"power%d_average_interval" , s); attr++; -snprintf (attr->name,sizeof (attr->name), -"power%d_input" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_power,NULL ,3 , i); +occ_init_attribute (attr,0444 , show_power,NULL , +3 , i,"power%d_input" , s); attr++; }
@@ -1006,56 +980,43 @@ static int occ_setup_sensor_attrs(struct occ *occ) }if (sensors->caps.num_sensors >=1 ) { -snprintf (attr->name,sizeof (attr->name),"power%d_label" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_caps,NULL , -0 ,0 ); +occ_init_attribute (attr,0444 , show_caps,NULL , +0 ,0 ,"power%d_label" , s); attr++; -snprintf (attr->name,sizeof (attr->name),"power%d_cap" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_caps,NULL , -1 ,0 ); +occ_init_attribute (attr,0444 , show_caps,NULL , +1 ,0 ,"power%d_cap" , s); attr++; -snprintf (attr->name,sizeof (attr->name),"power%d_input" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_caps,NULL , -2 ,0 ); +occ_init_attribute (attr,0444 , show_caps,NULL , +2 ,0 ,"power%d_input" , s); attr++; -snprintf (attr->name,sizeof (attr->name), -"power%d_cap_not_redundant" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_caps,NULL , -3 ,0 ); +occ_init_attribute (attr,0444 , show_caps,NULL , +3 ,0 ,"power%d_cap_not_redundant" , s); attr++; -snprintf (attr->name,sizeof (attr->name),"power%d_cap_max" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_caps,NULL , -4 ,0 ); +occ_init_attribute (attr,0444 , show_caps,NULL , +4 ,0 ,"power%d_cap_max" , s); attr++; -snprintf (attr->name,sizeof (attr->name),"power%d_cap_min" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , show_caps,NULL , -5 ,0 ); +occ_init_attribute (attr,0444 , show_caps,NULL , +5 ,0 ,"power%d_cap_min" , s); attr++; -snprintf (attr->name,sizeof (attr->name),"power%d_cap_user" , - s); - attr->sensor =OCC_INIT_ATTR (attr->name,0644 , show_caps, - occ_store_caps_user,6 ,0 ); +occ_init_attribute (attr,0644 , show_caps, occ_store_caps_user, +6 ,0 ,"power%d_cap_user" , s); attr++;if (sensors->caps.version >1 ) { -snprintf (attr->name,sizeof (attr->name), -"power%d_cap_user_source" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_caps,NULL ,7 ,0 ); +occ_init_attribute (attr,0444 , show_caps,NULL , +7 ,0 ,"power%d_cap_user_source" , s); attr++;if (sensors->caps.version >2 ) { -snprintf (attr->name,sizeof (attr->name), -"power%d_cap_min_soft" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - show_caps,NULL , -8 ,0 ); +occ_init_attribute (attr,0444 , show_caps,NULL , +8 ,0 , +"power%d_cap_min_soft" , s); attr++; } }
@@ -1064,19 +1025,16 @@ static int occ_setup_sensor_attrs(struct occ *occ)for (i =0 ; i < sensors->extended.num_sensors; ++i) { s = i +1 ; -snprintf (attr->name,sizeof (attr->name),"extn%d_label" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - occ_show_extended,NULL ,0 , i); +occ_init_attribute (attr,0444 , occ_show_extended,NULL , +0 , i,"extn%d_label" , s); attr++; -snprintf (attr->name,sizeof (attr->name),"extn%d_flags" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - occ_show_extended,NULL ,1 , i); +occ_init_attribute (attr,0444 , occ_show_extended,NULL , +1 , i,"extn%d_flags" , s); attr++; -snprintf (attr->name,sizeof (attr->name),"extn%d_input" , s); - attr->sensor =OCC_INIT_ATTR (attr->name,0444 , - occ_show_extended,NULL ,2 , i); +occ_init_attribute (attr,0444 , occ_show_extended,NULL , +2 , i,"extn%d_input" , s); attr++; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 520/592 hwmon: (occ) fix unaligned accesses
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Arnd Bergmann <[email protected]>
[ Upstream commit 2c021b45c154958566aad0cae9f74ab26a2d5732 ]
Passing a pointer to an unaligned integer as a function argument is undefined behavior:
drivers/hwmon/occ/common.c:492:27: warning: taking address of packed member 'accumulator' of class or structure 'power_sensor_2' may result in an unaligned pointer value [-Waddress-of-packed-member]
492 | val = occ_get_powr_avg(&power->accumulator,
| ^~~~~~~~~~~~~~~~~~
drivers/hwmon/occ/common.c:493:13: warning: taking address of packed member 'update_tag' of class or structure 'power_sensor_2' may result in an unaligned pointer value [-Waddress-of-packed-member]
493 | &power->update_tag);
| ^~~~~~~~~~~~~~~~~Move the get_unaligned() calls out of the function and pass these through argument registers instead.
--- drivers/hwmon/occ/common.c | 28 +++++++++++++--------------- 1 file changed, 13 insertions(+), 15 deletions(-)
drivers/hwmon/occ/common.c +13 -15
@@ -459,12 +459,10 @@ static ssize_t occ_show_power_1(struct device *dev,return sysfs_emit (buf,"%llu\n" , val); } -static u64 occ_get_powr_avg (u64 *accum,u32 *samples) +static u64 occ_get_powr_avg (u64 accum,u32 samples) { -u64 divisor =get_unaligned_be32 (samples); - -return (divisor ==0 ) ?0 : -div64_u64 (get_unaligned_be64 (accum) *1000000ULL , divisor); +return (samples ==0 ) ?0 : +mul_u64_u32_div (accum,1000000UL , samples); } static ssize_t occ_show_power_2(struct device *dev,
@@ -489,8 +487,8 @@ static ssize_t occ_show_power_2(struct device *dev,get_unaligned_be32 (&power->sensor_id), power->function_id, power->apss_channel);case 1 : - val =occ_get_powr_avg (&power->accumulator, - &power->update_tag); + val =occ_get_powr_avg (get_unaligned_be64 (&power->accumulator), +get_unaligned_be32 (&power->update_tag));break ;case 2 : val = (u64)get_unaligned_be32(&power->update_tag) *
@@ -527,8 +525,8 @@ static ssize_t occ_show_power_a0(struct device *dev,return sysfs_emit (buf,"%u_system\n" ,get_unaligned_be32 (&power->sensor_id));case 1 : - val =occ_get_powr_avg (&power->system.accumulator, - &power->system.update_tag); + val =occ_get_powr_avg (get_unaligned_be64 (&power->system.accumulator), +get_unaligned_be32 (&power->system.update_tag));break ;case 2 : val = (u64)get_unaligned_be32(&power->system.update_tag) *
@@ -541,8 +539,8 @@ static ssize_t occ_show_power_a0(struct device *dev,return sysfs_emit (buf,"%u_proc\n" ,get_unaligned_be32 (&power->sensor_id));case 5 : - val =occ_get_powr_avg (&power->proc.accumulator, - &power->proc.update_tag); + val =occ_get_powr_avg (get_unaligned_be64 (&power->proc.accumulator), +get_unaligned_be32 (&power->proc.update_tag));break ;case 6 : val = (u64)get_unaligned_be32(&power->proc.update_tag) *
@@ -555,8 +553,8 @@ static ssize_t occ_show_power_a0(struct device *dev,return sysfs_emit (buf,"%u_vdd\n" ,get_unaligned_be32 (&power->sensor_id));case 9 : - val =occ_get_powr_avg (&power->vdd.accumulator, - &power->vdd.update_tag); + val =occ_get_powr_avg (get_unaligned_be64 (&power->vdd.accumulator), +get_unaligned_be32 (&power->vdd.update_tag));break ;case 10 : val = (u64)get_unaligned_be32(&power->vdd.update_tag) *
@@ -569,8 +567,8 @@ static ssize_t occ_show_power_a0(struct device *dev,return sysfs_emit (buf,"%u_vdn\n" ,get_unaligned_be32 (&power->sensor_id));case 13 : - val =occ_get_powr_avg (&power->vdn.accumulator, - &power->vdn.update_tag); + val =occ_get_powr_avg (get_unaligned_be64 (&power->vdn.accumulator), +get_unaligned_be32 (&power->vdn.update_tag));break ;case 14 : val = (u64)get_unaligned_be32(&power->vdn.update_tag) *
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 521/592 hwmon: (ltc4282) avoid repeated register write
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Nuno Sá <[email protected]>
[ Upstream commit c25892b7a1744355e16281cd24a9b59ec15ec974 ]
The fault enabled bits were being mistankenly enabled twice in case the FW property is present. Remove one of the writes.
--- drivers/hwmon/ltc4282.c | 7 ------- 1 file changed, 7 deletions(-)
drivers/hwmon/ltc4282.c +0 -7
@@ -1511,13 +1511,6 @@ static int ltc4282_setup(struct ltc4282_state *st, struct device *dev)return ret; } -if (device_property_read_bool (dev,"adi,fault-log-enable" )) { - ret =regmap_set_bits (st->map,LTC4282_ADC_CTRL , -LTC4282_FAULT_LOG_EN_MASK ); -if (ret) -return ret; - } -if (device_property_read_bool (dev,"adi,fault-log-enable" )) { ret =regmap_set_bits (st->map,LTC4282_ADC_CTRL ,LTC4282_FAULT_LOG_EN_MASK );if (ret)
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 522/592 pldmfw: Select CRC32 when PLDMFW is selected
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Simon Horman <[email protected]>
[ Upstream commit 1224b218a4b9203656ecc932152f4c81a97b4fcc ]
pldmfw calls crc32 code and depends on it being enabled, else there is a link error as follows. So PLDMFW should select CRC32.
lib/pldmfw/pldmfw.o: In function `pldmfw_flash_image':
pldmfw.c:(.text+0x70f): undefined reference to `crc32_le_base'
This problem was introduced by commit b8265621f488 ("Add pldmfw library for PLDM firmware update").
It manifests as of commit d69ea414c9b4 ("ice: implement device flash update via devlink").
And is more likely to occur as of commit 9ad19171b6d6 ("lib/crc: remove unnecessary prompt for CONFIG_CRC32 and drop 'default y'").
Found by chance while exercising builds based on tinyconfig.
--- lib/Kconfig | 1 + 1 file changed, 1 insertion(+)
lib/Kconfig +1 -0
@@ -716,6 +716,7 @@ config GENERIC_LIB_DEVMEM_IS_ALLOWED config PLDMFW bool + select CRC32 default n config ASN1_ENCODER
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 523/592 aoe: clean device rq_list in aoedev_downdev()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Justin Sanders <[email protected]>
[ Upstream commit 7f90d45e57cb2ef1f0adcaf925ddffdfc5e680ca ]
An aoe device's rq_list contains accepted block requests that are waiting to be transmitted to the aoe target. This queue was added as part of the conversion to blk_mq. However, the queue was not cleaned out when an aoe device is downed which caused blk_mq_freeze_queue() to sleep indefinitely waiting for those requests to complete, causing a hang. This fix cleans out the queue before calling blk_mq_freeze_queue().
--- drivers/block/aoe/aoedev.c | 8 ++++++++ 1 file changed, 8 insertions(+)
drivers/block/aoe/aoedev.c +8 -0
@@ -198,6 +198,7 @@ aoedev_downdev(struct aoedev *d) {struct aoetgt *t, **tt, **te;struct list_head *head, *pos, *nx; +struct request *rq, *rqnext;int i; d->flags &= ~DEVFL_UP ;
@@ -223,6 +224,13 @@ aoedev_downdev(struct aoedev *d)/* clean out the in-process request (if any) */ aoe_failip (d); +/* clean out any queued block requests */ +list_for_each_entry_safe (rq, rqnext, &d->rq_list, queuelist) { +list_del_init (&rq->queuelist); +blk_mq_start_request (rq); +blk_mq_end_request (rq,BLK_STS_IOERR ); + } +/* fast fail all pending I/O */ if (d->blkq) {/* UP is cleared, freeze+quiesce to insure all are errored */
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 524/592 io_uring/sqpoll: dont put task_struct on tctx setup failure
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jens Axboe <[email protected]>
[ Upstream commit f2320f1dd6f6f82cb2c7aff23a12bab537bdea89 ]
A recent commit moved the error handling of sqpoll thread and tctx failures into the thread itself, as part of fixing an issue. However, it missed that tctx allocation may also fail, and that io_sq_offload_create() does its own error handling for the task_struct in that case.
Remove the manual task putting in io_sq_offload_create(), as io_sq_thread() will notice that the tctx did not get setup and hence it should put itself and exit.
--- io_uring/sqpoll.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-)
io_uring/sqpoll.c +1 -4
@@ -419,7 +419,6 @@ void io_sqpoll_wait_sq(struct io_ring_ctx *ctx)__cold int io_sq_offload_create (struct io_ring_ctx *ctx,struct io_uring_params *p) { -struct task_struct *task_to_put =NULL ;int ret;/* Retain compatibility with failing for an invalid attach attempt */
@@ -498,7 +497,7 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx,rcu_assign_pointer (sqd->thread, tsk);mutex_unlock (&sqd->lock); - task_to_put =get_task_struct (tsk); +get_task_struct (tsk); ret =io_uring_alloc_task_context (tsk, ctx);wake_up_new_task (tsk);if (ret)
@@ -513,8 +512,6 @@ __cold int io_sq_offload_create(struct io_ring_ctx *ctx,complete (&ctx->sq_data->exited); err:io_sq_thread_finish (ctx); -if (task_to_put) -put_task_struct (task_to_put);return ret; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 525/592 net: ice: Perform accurate aRFS flow match
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Krishna Kumar <[email protected]>
[ Upstream commit 5d3bc9e5e725aa36cca9b794e340057feb6880b4 ]
This patch fixes an issue seen in a large-scale deployment under heavy incoming pkts where the aRFS flow wrongly matches a flow and reprograms the NIC with wrong settings. That mis-steering causes RX-path latency spikes and noisy neighbor effects when many connections collide on the same hash (some of our production servers have 20-30K connections).
set_rps_cpu() calls ndo_rx_flow_steer() with flow_id that is calculated by hashing the skb sized by the per rx-queue table size. This results in multiple connections (even across different rx-queues) getting the same hash value. The driver steer function modifies the wrong flow to use this rx-queue, e.g.: Flow#1 is first added:
Flow#1: <ip1, port1, ip2, port2>, Hash 'h', q#10
Later when a new flow needs to be added:
Flow#2: <ip3, port3, ip4, port4>, Hash 'h', q#20
The driver finds the hash 'h' from Flow#1 and updates it to use q#20. This results in both flows getting un-optimized - packets for Flow#1 goes to q#20, and then reprogrammed back to q#10 later and so on; and Flow #2 programming is never done as Flow#1 is matched first for all misses. Many flows may wrongly share the same hash and reprogram rules of the original flow each with their own q#.
Tested on two 144-core servers with 16K netperf sessions for 180s. Netperf clients are pinned to cores 0-71 sequentially (so that wrong packets on q#s 72-143 can be measured). IRQs are set 1:1 for queues -> CPUs, enable XPS, enable aRFS (global value is 144 * rps_flow_cnt).
Test notes about results from ice_rx_flow_steer():
---------------------------------------------------
1. "Skip:" counter increments here:
if (fltr_info->q_index == rxq_idx ||
arfs_entry->fltr_state != ICE_ARFS_ACTIVE)
goto out;
2. "Add:" counter increments here:
ret = arfs_entry->fltr_info.fltr_id;
INIT_HLIST_NODE(&arfs_entry->list_entry);
3. "Update:" counter increments here:
/* update the queue to forward to on an already existing flow */Runtime comparison: original code vs with the patch for different rps_flow_cnt values.
+-------------------------------+--------------+--------------+ | rps_flow_cnt | 512 | 2048 | +-------------------------------+--------------+--------------+ | Ratio of Pkts on Good:Bad q's | 214 vs 822K | 1.1M vs 980K | | Avoid wrong aRFS programming | 0 vs 310K | 0 vs 30K | | CPU User | 216 vs 183 | 216 vs 206 | | CPU System | 1441 vs 1171 | 1447 vs 1320 | | CPU Softirq | 1245 vs 920 | 1238 vs 961 | | CPU Total | 29 vs 22.7 | 29 vs 24.9 | | aRFS Update | 533K vs 59 | 521K vs 32 | | aRFS Skip | 82M vs 77M | 7.2M vs 4.5M | +-------------------------------+--------------+--------------+
A separate TCP_STREAM and TCP_RR with 1,4,8,16,64,128,256,512 connections showed no performance degradation.
Some points on the patch/aRFS behavior:
1. Enabling full tuple matching ensures flows are always correctly matched, even with smaller hash sizes.
2. 5-6% drop in CPU utilization as the packets arrive at the correct CPUs and fewer calls to driver for programming on misses.
3. Larger hash tables reduces mis-steering due to more unique flow hashes, but still has clashes. However, with larger per-device rps_flow_cnt, old flows take more time to expire and new aRFS flows cannot be added if h/w limits are reached (rps_may_expire_flow() succeeds when 10*rps_flow_cnt pkts have been processed by this cpu that are not part of the flow).
--- drivers/net/ethernet/intel/ice/ice_arfs.c | 48 +++++++++++++++++++++++ 1 file changed, 48 insertions(+)
drivers/net/ethernet/intel/ice/ice_arfs.c +48 -0
@@ -377,6 +377,50 @@ ice_arfs_is_perfect_flow_set(struct ice_hw *hw, __be16 l3_proto, u8 l4_proto)return false; } +/** +* ice_arfs_cmp - Check if aRFS filter matches this flow. +* @fltr_info: filter info of the saved ARFS entry. +* @fk: flow dissector keys. +* @n_proto: One of htons(ETH_P_IP) or htons(ETH_P_IPV6). +* @ip_proto: One of IPPROTO_TCP or IPPROTO_UDP. +* +* Since this function assumes limited values for n_proto and ip_proto, it +* is meant to be called only from ice_rx_flow_steer(). +* +* Return: +* * true - fltr_info refers to the same flow as fk. +* * false - fltr_info and fk refer to different flows. +*/ +static bool +ice_arfs_cmp (const struct ice_fdir_fltr *fltr_info,const struct flow_keys *fk, +__be16 n_proto,u8 ip_proto) +{ +/* Determine if the filter is for IPv4 or IPv6 based on flow_type, +* which is one of ICE_FLTR_PTYPE_NONF_IPV{4,6}_{TCP,UDP}. +*/ +bool is_v4 = fltr_info->flow_type ==ICE_FLTR_PTYPE_NONF_IPV4_TCP || + fltr_info->flow_type ==ICE_FLTR_PTYPE_NONF_IPV4_UDP ; + +/* Following checks are arranged in the quickest and most discriminative +* fields first for early failure. +*/ +if (is_v4) +return n_proto ==htons (ETH_P_IP ) && + fltr_info->ip.v4.src_port == fk->ports.src && + fltr_info->ip.v4.dst_port == fk->ports.dst && + fltr_info->ip.v4.src_ip == fk->addrs.v4addrs.src && + fltr_info->ip.v4.dst_ip == fk->addrs.v4addrs.dst && + fltr_info->ip.v4.proto == ip_proto; + +return fltr_info->ip.v6.src_port == fk->ports.src && + fltr_info->ip.v6.dst_port == fk->ports.dst && + fltr_info->ip.v6.proto == ip_proto && + !memcmp (&fltr_info->ip.v6.src_ip, &fk->addrs.v6addrs.src, +sizeof (struct in6_addr )) && + !memcmp (&fltr_info->ip.v6.dst_ip, &fk->addrs.v6addrs.dst, +sizeof (struct in6_addr )); +} +/** * ice_rx_flow_steer - steer the Rx flow to where application is being run * @netdev: ptr to the netdev being adjusted
@@ -448,6 +492,10 @@ ice_rx_flow_steer(struct net_device *netdev, const struct sk_buff *skb,continue ; fltr_info = &arfs_entry->fltr_info; + +if (!ice_arfs_cmp (fltr_info, &fk, n_proto, ip_proto)) +continue ; + ret = fltr_info->fltr_id; if (fltr_info->q_index == rxq_idx ||
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 526/592 ice: fix eswitch code memory leak in reset scenario
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Grzegorz Nitka <[email protected]>
[ Upstream commit 48c8b214974dc55283bd5f12e3a483b27c403bbc ]
Add simple eswitch mode checker in attaching VF procedure and allocate required port representor memory structures only in switchdev mode. The reset flows triggers VF (if present) detach/attach procedure. It might involve VF port representor(s) re-creation if the device is configured is switchdev mode (not legacy one).
The memory was blindly allocated in current implementation, regardless of the mode and not freed if in legacy mode.
Kmemeleak trace:
unreferenced object (percpu) 0x7e3bce5b888458 (size 40):
comm "bash", pid 1784, jiffies 4295743894
hex dump (first 32 bytes on cpu 45):
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 00 ................
backtrace (crc 0):
pcpu_alloc_noprof+0x4c4/0x7c0
ice_repr_create+0x66/0x130 [ice]
ice_repr_create_vf+0x22/0x70 [ice]
ice_eswitch_attach_vf+0x1b/0xa0 [ice]
ice_reset_all_vfs+0x1dd/0x2f0 [ice]
ice_pci_err_resume+0x3b/0xb0 [ice]
pci_reset_function+0x8f/0x120
reset_store+0x56/0xa0
kernfs_fop_write_iter+0x120/0x1b0
vfs_write+0x31c/0x430
ksys_write+0x61/0xd0
do_syscall_64+0x5b/0x180
entry_SYSCALL_64_after_hwframe+0x76/0x7eTesting hints (ethX is PF netdev):
- create at least one VF
echo 1 > /sys/class/net/ethX/device/sriov_numvfs
- trigger the reset
echo 1 > /sys/class/net/ethX/device/reset
--- drivers/net/ethernet/intel/ice/ice_eswitch.c | 6 +++++- 1 file changed, 5 insertions(+), 1 deletion(-)
drivers/net/ethernet/intel/ice/ice_eswitch.c +5 -1
@@ -502,10 +502,14 @@ ice_eswitch_attach(struct ice_pf *pf, struct ice_repr *repr, unsigned long *id)*/ int ice_eswitch_attach_vf (struct ice_pf *pf,struct ice_vf *vf) { -struct ice_repr *repr =ice_repr_create_vf (vf);struct devlink *devlink =priv_to_devlink (pf); +struct ice_repr *repr;int err; +if (!ice_is_eswitch_mode_switchdev (pf)) +return 0 ; + + repr =ice_repr_create_vf (vf);if (IS_ERR (repr))return PTR_ERR (repr);
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 527/592 e1000e: set fixed clock frequency indication for Nahum 11 and Nahum 13
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vitaly Lifshits <[email protected]>
[ Upstream commit 688a0d61b2d7427189c4eb036ce485d8fc957cbb ]
On some systems with Nahum 11 and Nahum 13 the value of the XTAL clock in the software STRAP is incorrect. This causes the PTP timer to run at the wrong rate and can lead to synchronization issues.
The STRAP value is configured by the system firmware, and a firmware update is not always possible. Since the XTAL clock on these systems always runs at 38.4MHz, the driver may ignore the STRAP and just set the correct value.
--- drivers/net/ethernet/intel/e1000e/netdev.c | 14 +++++++++++--- drivers/net/ethernet/intel/e1000e/ptp.c | 8 +++++--- 2 files changed, 16 insertions(+), 6 deletions(-)
drivers/net/ethernet/intel/e1000e/netdev.c +11 -3
@@ -3534,9 +3534,6 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca)case e1000_pch_cnp:case e1000_pch_tgp:case e1000_pch_adp: -case e1000_pch_mtp: -case e1000_pch_lnp: -case e1000_pch_ptp:case e1000_pch_nvp:if (er32 (TSYNCRXCTL ) &E1000_TSYNCRXCTL_SYSCFI ) {/* Stable 24MHz frequency */
@@ -3552,6 +3549,17 @@ s32 e1000e_get_base_timinca(struct e1000_adapter *adapter, u32 *timinca) adapter->cc.shift = shift; }break ; +case e1000_pch_mtp: +case e1000_pch_lnp: +case e1000_pch_ptp: +/* System firmware can misreport this value, so set it to a +* stable 38400KHz frequency. +*/ + incperiod =INCPERIOD_38400KHZ ; + incvalue =INCVALUE_38400KHZ ; + shift =INCVALUE_SHIFT_38400KHZ ; + adapter->cc.shift = shift; +break ;case e1000_82574:case e1000_82583:/* Stable 25MHz frequency */
drivers/net/ethernet/intel/e1000e/ptp.c +5 -3
@@ -295,15 +295,17 @@ void e1000e_ptp_init(struct e1000_adapter *adapter)case e1000_pch_cnp:case e1000_pch_tgp:case e1000_pch_adp: -case e1000_pch_mtp: -case e1000_pch_lnp: -case e1000_pch_ptp:case e1000_pch_nvp:if (er32 (TSYNCRXCTL ) &E1000_TSYNCRXCTL_SYSCFI ) adapter->ptp_clock_info.max_adj =MAX_PPB_24MHZ ;else adapter->ptp_clock_info.max_adj =MAX_PPB_38400KHZ ;break ; +case e1000_pch_mtp: +case e1000_pch_lnp: +case e1000_pch_ptp: + adapter->ptp_clock_info.max_adj =MAX_PPB_38400KHZ ; +break ;case e1000_82574:case e1000_82583: adapter->ptp_clock_info.max_adj =MAX_PPB_25MHZ ;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 528/592 workqueue: Initialize wq_isolated_cpumask in workqueue_init_early()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Chuyi Zhou <[email protected]>
[ Upstream commit 261dce3d64021e7ec828a17b4975ce9182e54ceb ]
Now when isolcpus is enabled via the cmdline, wq_isolated_cpumask does not include these isolated CPUs, even wq_unbound_cpumask has already excluded them. It is only when we successfully configure an isolate cpuset partition that wq_isolated_cpumask gets overwritten by workqueue_unbound_exclude_cpumask(), including both the cmdline-specified isolated CPUs and the isolated CPUs within the cpuset partitions.
Fix this issue by initializing wq_isolated_cpumask properly in workqueue_init_early().
--- kernel/workqueue.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
kernel/workqueue.c +2 -1
@@ -7756,7 +7756,8 @@ void __init workqueue_init_early(void)restrict_unbound_cpumask ("workqueue.unbound_cpus" , &wq_cmdline_cpumask);cpumask_copy (wq_requested_unbound_cpumask, wq_unbound_cpumask); - +cpumask_andnot (wq_isolated_cpumask, cpu_possible_mask, +housekeeping_cpumask (HK_TYPE_DOMAIN )); pwq_cache =KMEM_CACHE (pool_workqueue,SLAB_PANIC ); unbound_wq_update_pwq_attrs_buf =alloc_workqueue_attrs ();
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 529/592 ksmbd: add free_transport ops in ksmbd connection
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Namjae Jeon <[email protected]>
[ Upstream commit a89f5fae998bdc4d0505306f93844c9ae059d50c ]
free_transport function for tcp connection can be called from smbdirect. It will cause kernel oops. This patch add free_transport ops in ksmbd connection, and add each free_transports for tcp and smbdirect.
--- fs/smb/server/connection.c | 2 +- fs/smb/server/connection.h | 1 + fs/smb/server/transport_rdma.c | 10 ++++++++-- fs/smb/server/transport_tcp.c | 3 ++- 4 files changed, 12 insertions(+), 4 deletions(-)
fs/smb/server/connection.c +1 -1
@@ -40,7 +40,7 @@ void ksmbd_conn_free(struct ksmbd_conn *conn)kvfree (conn->request_buf);kfree (conn->preauth_info);if (atomic_dec_and_test (&conn->refcnt)) { -ksmbd_free_transport (conn->transport); + conn->transport->ops->free_transport (conn->transport);kfree (conn); } }
fs/smb/server/connection.h +1 -0
@@ -132,6 +132,7 @@ struct ksmbd_transport_ops { void *buf, unsigned int len, struct smb2_buffer_desc_v1 *desc, unsigned int desc_len); +void (*free_transport)(struct ksmbd_transport *kt); };struct ksmbd_transport {
fs/smb/server/transport_rdma.c +8 -2
@@ -159,7 +159,8 @@ struct smb_direct_transport { };#define KSMBD_TRANS (t) ((struct ksmbd_transport *)&((t)->transport)) - +#define SMBD_TRANS (t) ((struct smb_direct_transport *)container_of(t, \ + struct smb_direct_transport, transport))enum {SMB_DIRECT_MSG_NEGOTIATE_REQ =0 ,SMB_DIRECT_MSG_DATA_TRANSFER
@@ -410,6 +411,11 @@ static struct smb_direct_transport *alloc_transport(struct rdma_cm_id *cm_id)return NULL ; } +static void smb_direct_free_transport (struct ksmbd_transport *kt) +{ +kfree (SMBD_TRANS (kt)); +} +static void free_transport (struct smb_direct_transport *t) {struct smb_direct_recvmsg *recvmsg;
@@ -455,7 +461,6 @@ static void free_transport(struct smb_direct_transport *t)smb_direct_destroy_pools (t);ksmbd_conn_free (KSMBD_TRANS (t)->conn); -kfree (t); } static struct smb_direct_sendmsg
@@ -2281,4 +2286,5 @@ static const struct ksmbd_transport_ops ksmbd_smb_direct_transport_ops = { .read = smb_direct_read, .rdma_read = smb_direct_rdma_read, .rdma_write = smb_direct_rdma_write, + .free_transport = smb_direct_free_transport, };
fs/smb/server/transport_tcp.c +2 -1
@@ -93,7 +93,7 @@ static struct tcp_transport *alloc_transport(struct socket *client_sk)return t; } -void ksmbd_free_transport (struct ksmbd_transport *kt) +static void ksmbd_tcp_free_transport (struct ksmbd_transport *kt) {struct tcp_transport *t =TCP_TRANS (kt);
@@ -656,4 +656,5 @@ static const struct ksmbd_transport_ops ksmbd_tcp_transport_ops = { .read = ksmbd_tcp_read, .writev = ksmbd_tcp_writev, .disconnect = ksmbd_tcp_disconnect, + .free_transport = ksmbd_tcp_free_transport, };
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 530/592 net: ti: icssg-prueth: Fix packet handling for XDP_TX
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Meghana Malladi <[email protected]>
[ Upstream commit 60524f1d2bdf222db6dc3f680e0272441f697fe4 ]
While transmitting XDP frames for XDP_TX, page_pool is used to get the DMA buffers (already mapped to the pages) and need to be freed/reycled once the transmission is complete. This need not be explicitly done by the driver as this is handled more gracefully by the xdp driver while returning the xdp frame. __xdp_return() frees the XDP memory based on its memory type, under which page_pool memory is also handled. This change fixes the transmit queue timeout while running XDP_TX.
logs:
[ 309.069682] icssg-prueth icssg1-eth eth2: NETDEV WATCHDOG: CPU: 0: transmit queue 0 timed out 45860 ms [ 313.933780] icssg-prueth icssg1-eth eth2: NETDEV WATCHDOG: CPU: 0: transmit queue 0 timed out 50724 ms [ 319.053656] icssg-prueth icssg1-eth eth2: NETDEV WATCHDOG: CPU: 0: transmit queue 0 timed out 55844 ms
...
--- drivers/net/ethernet/ti/icssg/icssg_common.c | 19 ++----------------- 1 file changed, 2 insertions(+), 17 deletions(-)
drivers/net/ethernet/ti/icssg/icssg_common.c +2 -17
@@ -98,20 +98,11 @@ void prueth_xmit_free(struct prueth_tx_chn *tx_chn, {struct cppi5_host_desc_t *first_desc, *next_desc;dma_addr_t buf_dma, next_desc_dma; -struct prueth_swdata *swdata; -struct page *page;u32 buf_dma_len; first_desc = desc; next_desc = first_desc; - swdata =cppi5_hdesc_get_swdata (desc); -if (swdata->type ==PRUETH_SWDATA_PAGE ) { - page = swdata->data.page; -page_pool_recycle_direct (page->pp, swdata->data.page); - goto free_desc; - } -cppi5_hdesc_get_obuf (first_desc, &buf_dma, &buf_dma_len);k3_udma_glue_tx_cppi5_to_dma_addr (tx_chn->tx_chn, &buf_dma);
@@ -135,7 +126,6 @@ void prueth_xmit_free(struct prueth_tx_chn *tx_chn,k3_cppi_desc_pool_free (tx_chn->desc_pool, next_desc); } -free_desc:k3_cppi_desc_pool_free (tx_chn->desc_pool, first_desc); }EXPORT_SYMBOL_GPL (prueth_xmit_free);
@@ -612,13 +602,8 @@ u32 emac_xmit_xdp_frame(struct prueth_emac *emac,k3_udma_glue_tx_dma_to_cppi5_addr (tx_chn->tx_chn, &buf_dma);cppi5_hdesc_attach_buf (first_desc, buf_dma, xdpf->len, buf_dma, xdpf->len); swdata =cppi5_hdesc_get_swdata (first_desc); -if (page) { - swdata->type =PRUETH_SWDATA_PAGE ; - swdata->data.page = page; - }else { - swdata->type =PRUETH_SWDATA_XDPF ; - swdata->data.xdpf = xdpf; - } + swdata->type =PRUETH_SWDATA_XDPF ; + swdata->data.xdpf = xdpf;/* Report BQL before sending the packet */ netif_txq =netdev_get_tx_queue (ndev, tx_chn->id);
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 531/592 net: netmem: fix skb_ensure_writable with unreadable skbs
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Mina Almasry <[email protected]>
[ Upstream commit 6f793a1d053775f8324b8dba1e7ed224f8b0166f ]
skb_ensure_writable should succeed when it's trying to write to the header of the unreadable skbs, so it doesn't need an unconditional skb_frags_readable check. The preceding pskb_may_pull() call will succeed if write_len is within the head and fail if we're trying to write to the unreadable payload, so we don't need an additional check.
Removing this check restores DSCP functionality with unreadable skbs as it's called from dscp_tg.
--- net/core/skbuff.c | 3 --- 1 file changed, 3 deletions(-)
net/core/skbuff.c +0 -3
@@ -6220,9 +6220,6 @@ int skb_ensure_writable(struct sk_buff *skb, unsigned int write_len)if (!pskb_may_pull (skb, write_len))return -ENOMEM ; -if (!skb_frags_readable (skb)) -return -EFAULT ; -if (!skb_cloned (skb) ||skb_clone_writable (skb, write_len))return 0 ;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 532/592 bnxt_en: Fix double invocation of bnxt_ulp_stop()/bnxt_ulp_start()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kalesh AP <[email protected]>
[ Upstream commit 1e9ac33fa271be0d2480fd732f9642d81542500b ]
Before the commit under the Fixes tag below, bnxt_ulp_stop() and bnxt_ulp_start() were always invoked in pairs. After that commit, the new bnxt_ulp_restart() can be invoked after bnxt_ulp_stop() has been called. This may result in the RoCE driver's aux driver .suspend() method being invoked twice. The 2nd bnxt_re_suspend() call will crash when it dereferences a NULL pointer:
(NULL ib_device): Handle device suspend call
BUG: kernel NULL pointer dereference, address: 0000000000000b78 PGD 0 P4D 0 Oops: Oops: 0000 [#1] SMP PTI CPU: 20 UID: 0 PID: 181 Comm: kworker/u96:5 Tainted: G S 6.15.0-rc1 #4 PREEMPT(voluntary) Tainted: [S]=CPU_OUT_OF_SPEC Hardware name: Dell Inc. PowerEdge R730/072T6D, BIOS 2.4.3 01/17/2017 Workqueue: bnxt_pf_wq bnxt_sp_task [bnxt_en] RIP: 0010:bnxt_re_suspend+0x45/0x1f0 [bnxt_re] Code: 8b 05 a7 3c 5b f5 48 89 44 24 18 31 c0 49 8b 5c 24 08 4d 8b 2c 24 e8 ea 06 0a f4 48 c7 c6 04 60 52 c0 48 89 df e8 1b ce f9 ff <48> 8b 83 78 0b 00 00 48 8b 80 38 03 00 00 a8 40 0f 85 b5 00 00 00 RSP: 0018:ffffa2e84084fd88 EFLAGS: 00010246 RAX: 0000000000000000 RBX: 0000000000000000 RCX: 0000000000000001 RDX: 0000000000000000 RSI: ffffffffb4b6b934 RDI: 00000000ffffffff RBP: ffffa1760954c9c0 R08: 0000000000000000 R09: c0000000ffffdfff R10: 0000000000000001 R11: ffffa2e84084fb50 R12: ffffa176031ef070 R13: ffffa17609775000 R14: ffffa17603adc180 R15: 0000000000000000 FS: 0000000000000000(0000) GS:ffffa17daa397000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 0000000000000b78 CR3: 00000004aaa30003 CR4: 00000000003706f0 DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000 DR3: 0000000000000000 DR6: 00000000fffe0ff0 DR7: 0000000000000400 Call Trace: <TASK> bnxt_ulp_stop+0x69/0x90 [bnxt_en] bnxt_sp_task+0x678/0x920 [bnxt_en] ? __schedule+0x514/0xf50 process_scheduled_works+0x9d/0x400 worker_thread+0x11c/0x260 ? __pfx_worker_thread+0x10/0x10 kthread+0xfe/0x1e0 ? __pfx_kthread+0x10/0x10 ret_from_fork+0x2b/0x40 ? __pfx_kthread+0x10/0x10 ret_from_fork_asm+0x1a/0x30
Check the BNXT_EN_FLAG_ULP_STOPPED flag and do not proceed if the flag is already set. This will preserve the original symmetrical bnxt_ulp_stop() and bnxt_ulp_start().
Also, inside bnxt_ulp_start(), clear the BNXT_EN_FLAG_ULP_STOPPED flag after taking the mutex to avoid any race condition. And for symmetry, only proceed in bnxt_ulp_start() if the BNXT_EN_FLAG_ULP_STOPPED is set.
--- drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c | 24 ++++++++----------- 1 file changed, 10 insertions(+), 14 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt_ulp.c +10 -14
@@ -231,10 +231,9 @@ void bnxt_ulp_stop(struct bnxt *bp)return ;mutex_lock (&edev->en_dev_lock); -if (!bnxt_ulp_registered (edev)) { -mutex_unlock (&edev->en_dev_lock); -return ; - } +if (!bnxt_ulp_registered (edev) || + (edev->flags &BNXT_EN_FLAG_ULP_STOPPED )) + goto ulp_stop_exit; edev->flags |=BNXT_EN_FLAG_ULP_STOPPED ;if (aux_priv) {
@@ -250,6 +249,7 @@ void bnxt_ulp_stop(struct bnxt *bp) adrv->suspend (adev, pm); } } +ulp_stop_exit:mutex_unlock (&edev->en_dev_lock); }
@@ -258,19 +258,13 @@ void bnxt_ulp_start(struct bnxt *bp, int err)struct bnxt_aux_priv *aux_priv = bp->aux_priv;struct bnxt_en_dev *edev = bp->edev; -if (!edev) -return ; - - edev->flags &= ~BNXT_EN_FLAG_ULP_STOPPED ; - -if (err) +if (!edev || err)return ;mutex_lock (&edev->en_dev_lock); -if (!bnxt_ulp_registered (edev)) { -mutex_unlock (&edev->en_dev_lock); -return ; - } +if (!bnxt_ulp_registered (edev) || + !(edev->flags &BNXT_EN_FLAG_ULP_STOPPED )) + goto ulp_start_exit;if (edev->ulp_tbl->msix_requested)bnxt_fill_msix_vecs (bp, edev->msix_entries);
@@ -287,6 +281,8 @@ void bnxt_ulp_start(struct bnxt *bp, int err) adrv->resume (adev); } } +ulp_start_exit: + edev->flags &= ~BNXT_EN_FLAG_ULP_STOPPED ;mutex_unlock (&edev->en_dev_lock); }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 533/592 bnxt_en: Add a helper function to configure MRU and RSS
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Pavan Chebbi <[email protected]>
[ Upstream commit e11baaea94e2923739a98abeee85eb0667c04fd3 ]
Add a new helper function that will configure MRU and RSS table of a VNIC. This will be useful when we configure both on a VNIC when resetting an RX ring. This function will be used again in the next bug fix patch where we have to reconfigure VNICs for RSS contexts.
Stable-dep-of: 5dacc94c6fe6 ("bnxt_en: Update MRU and RSS table of RSS contexts on queue reset")
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 37 ++++++++++++++++------- 1 file changed, 26 insertions(+), 11 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +26 -11
@@ -10738,6 +10738,26 @@ void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx, bp->num_rss_ctx--; } +static int bnxt_set_vnic_mru_p5 (struct bnxt *bp,struct bnxt_vnic_info *vnic, +u16 mru) +{ +int rc; + +if (mru) { + rc =bnxt_hwrm_vnic_set_rss_p5 (bp, vnic, true); +if (rc) { +netdev_err (bp->dev,"hwrm vnic %d set rss failure rc: %d\n" , + vnic->vnic_id, rc); +return rc; + } + } + vnic->mru = mru; +bnxt_hwrm_vnic_update (bp, vnic, +VNIC_UPDATE_REQ_ENABLES_MRU_VALID ); + +return 0 ; +} +static void bnxt_hwrm_realloc_rss_ctx_vnic (struct bnxt *bp) {bool set_tpa = !!(bp->flags &BNXT_FLAG_TPA );
@@ -15884,6 +15904,7 @@ static int bnxt_queue_start(struct net_device *dev, void *qmem, int idx)struct bnxt_vnic_info *vnic;struct bnxt_napi *bnapi;int i, rc; +u16 mru; rxr = &bp->rx_ring[idx]; clone = qmem;
@@ -15933,18 +15954,13 @@ static int bnxt_queue_start(struct net_device *dev, void *qmem, int idx)napi_enable_locked (&bnapi->napi);bnxt_db_nq_arm (bp, &cpr->cp_db, cpr->cp_raw_cons); + mru = bp->dev->mtu +ETH_HLEN +VLAN_HLEN ;for (i =0 ; i < bp->nr_vnics; i++) { vnic = &bp->vnic_info[i]; - rc =bnxt_hwrm_vnic_set_rss_p5 (bp, vnic, true); -if (rc) { -netdev_err (bp->dev,"hwrm vnic %d set rss failure rc: %d\n" , - vnic->vnic_id, rc); + rc =bnxt_set_vnic_mru_p5 (bp, vnic, mru); +if (rc)return rc; - } - vnic->mru = bp->dev->mtu +ETH_HLEN +VLAN_HLEN ; -bnxt_hwrm_vnic_update (bp, vnic, -VNIC_UPDATE_REQ_ENABLES_MRU_VALID ); }return 0 ;
@@ -15969,9 +15985,8 @@ static int bnxt_queue_stop(struct net_device *dev, void *qmem, int idx)for (i =0 ; i < bp->nr_vnics; i++) { vnic = &bp->vnic_info[i]; - vnic->mru =0 ; -bnxt_hwrm_vnic_update (bp, vnic, -VNIC_UPDATE_REQ_ENABLES_MRU_VALID ); + +bnxt_set_vnic_mru_p5 (bp, vnic,0 ); }/* Make sure NAPI sees that the VNIC is disabled */ synchronize_net ();
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 534/592 bnxt_en: Update MRU and RSS table of RSS contexts on queue reset
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Pavan Chebbi <[email protected]>
[ Upstream commit 5dacc94c6fe61cde6f700e95cf35af9944b022c4 ]
The commit under the Fixes tag below which updates the VNICs' RSS and MRU during .ndo_queue_start(), needs to be extended to cover any non-default RSS contexts which have their own VNICs. Without this step, packets that are destined to a non-default RSS context may be dropped after .ndo_queue_start().
We further optimize this scheme by updating the VNIC only if the RX ring being restarted is in the RSS table of the VNIC. Updating the VNIC (in particular setting the MRU to 0) will momentarily stop all traffic to all rings in the RSS table. Any VNIC that has the RX ring excluded from the RSS table can skip this step and avoid the traffic disruption.
Note that this scheme is just an improvement. A VNIC with multiple rings in the RSS table will still see traffic disruptions to all rings in the RSS table when one of the rings is being restarted. We are working on a FW scheme that will improve upon this further.
--- drivers/net/ethernet/broadcom/bnxt/bnxt.c | 56 +++++++++++++++++++++-- 1 file changed, 51 insertions(+), 5 deletions(-)
drivers/net/ethernet/broadcom/bnxt/bnxt.c +51 -5
@@ -10738,11 +10738,39 @@ void bnxt_del_one_rss_ctx(struct bnxt *bp, struct bnxt_rss_ctx *rss_ctx, bp->num_rss_ctx--; } +static bool bnxt_vnic_has_rx_ring (struct bnxt *bp,struct bnxt_vnic_info *vnic, +int rxr_id) +{ +u16 tbl_size =bnxt_get_rxfh_indir_size (bp->dev); +int i, vnic_rx; + +/* Ntuple VNIC always has all the rx rings. Any change of ring id +* must be updated because a future filter may use it. +*/ +if (vnic->flags &BNXT_VNIC_NTUPLE_FLAG ) +return true; + +for (i =0 ; i < tbl_size; i++) { +if (vnic->flags &BNXT_VNIC_RSSCTX_FLAG ) + vnic_rx =ethtool_rxfh_context_indir (vnic->rss_ctx)[i]; +else + vnic_rx = bp->rss_indir_tbl[i]; + +if (rxr_id == vnic_rx) +return true; + } + +return false; +} +static int bnxt_set_vnic_mru_p5 (struct bnxt *bp,struct bnxt_vnic_info *vnic, -u16 mru) +u16 mru,int rxr_id) {int rc; +if (!bnxt_vnic_has_rx_ring (bp, vnic, rxr_id)) +return 0 ; +if (mru) { rc =bnxt_hwrm_vnic_set_rss_p5 (bp, vnic, true);if (rc) {
@@ -10758,6 +10786,24 @@ static int bnxt_set_vnic_mru_p5(struct bnxt *bp, struct bnxt_vnic_info *vnic,return 0 ; } +static int bnxt_set_rss_ctx_vnic_mru (struct bnxt *bp,u16 mru,int rxr_id) +{ +struct ethtool_rxfh_context *ctx; +unsigned long context; +int rc; + +xa_for_each (&bp->dev->ethtool->rss_ctx, context, ctx) { +struct bnxt_rss_ctx *rss_ctx =ethtool_rxfh_context_priv (ctx); +struct bnxt_vnic_info *vnic = &rss_ctx->vnic; + + rc =bnxt_set_vnic_mru_p5 (bp, vnic, mru, rxr_id); +if (rc) +return rc; + } + +return 0 ; +} +static void bnxt_hwrm_realloc_rss_ctx_vnic (struct bnxt *bp) {bool set_tpa = !!(bp->flags &BNXT_FLAG_TPA );
@@ -15958,12 +16004,11 @@ static int bnxt_queue_start(struct net_device *dev, void *qmem, int idx)for (i =0 ; i < bp->nr_vnics; i++) { vnic = &bp->vnic_info[i]; - rc =bnxt_set_vnic_mru_p5 (bp, vnic, mru); + rc =bnxt_set_vnic_mru_p5 (bp, vnic, mru, idx);if (rc)return rc; } - -return 0 ; +return bnxt_set_rss_ctx_vnic_mru (bp, mru, idx); err_reset: netdev_err(bp->dev, "Unexpected HWRM error during queue start rc: %d\n",
@@ -15986,8 +16031,9 @@ static int bnxt_queue_stop(struct net_device *dev, void *qmem, int idx)for (i =0 ; i < bp->nr_vnics; i++) { vnic = &bp->vnic_info[i]; -bnxt_set_vnic_mru_p5 (bp, vnic,0 ); +bnxt_set_vnic_mru_p5 (bp, vnic,0 , idx); } +bnxt_set_rss_ctx_vnic_mru (bp,0 , idx);/* Make sure NAPI sees that the VNIC is disabled */ synchronize_net (); rxr = &bp->rx_ring[idx];
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 535/592 ptp: fix breakage after ptp_vclock_in_use() rework
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vladimir Oltean <[email protected]>
[ Upstream commit 5ab73b010cad294851e558f1d4714a85c6f206c7 ]
What is broken
--------------
ptp4l, and any other application which calls clock_adjtime() on a physical clock, is greeted with error -EBUSY after commit 87f7ce260a3c ("ptp: remove ptp->n_vclocks check logic in ptp_vclock_in_use()").
Explanation for the breakage
----------------------------
The blamed commit was based on the false assumption that ptp_vclock_in_use() callers already test for n_vclocks prior to calling this function.
This is notably incorrect for the code path below, in which there is, in fact, no n_vclocks test:
ptp_clock_adjtime()
-> ptp_clock_freerun()
-> ptp_vclock_in_use()
The result is that any clock adjustment on any physical clock is now impossible. This is _despite_ there not being any vclock over this physical clock.
$ ptp4l -i eno0 -2 -P -m ptp4l[58.425]: selected /dev/ptp0 as PTP clock [ 58.429749] ptp: physical clock is free running ptp4l[58.431]: Failed to open /dev/ptp0: Device or resource busy failed to create a clock $ cat /sys/class/ptp/ptp0/n_vclocks 0
The patch makes the ptp_vclock_in_use() function say "if it's not a virtual clock, then this physical clock does have virtual clocks on top".
Then ptp_clock_freerun() uses this information to say "this physical clock has virtual clocks on top, so it must stay free-running".
Then ptp_clock_adjtime() uses this information to say "well, if this physical clock has to be free-running, I can't do it, return -EBUSY".
Simply put, ptp_vclock_in_use() cannot be simplified so as to remove the test whether vclocks are in use.
What did the blamed commit intend to fix
----------------------------------------
The blamed commit presents a lockdep warning stating "possible recursive locking detected", with the n_vclocks_store() and ptp_clock_unregister() functions involved.
The recursive locking seems this:n_vclocks_store()
-> mutex_lock_interruptible(&ptp->n_vclocks_mux) // 1 -> device_for_each_child_reverse(..., unregister_vclock)
-> unregister_vclock()
-> ptp_vclock_unregister()
-> ptp_clock_unregister()
-> ptp_vclock_in_use()
-> mutex_lock_interruptible(&ptp->n_vclocks_mux) // 2The issue can be triggered by creating and then deleting vclocks:
$ echo 2 > /sys/class/ptp/ptp0/n_vclocks $ echo 0 > /sys/class/ptp/ptp0/n_vclocks
But note that in the original stack trace, the address of the first lock is different from the address of the second lock. This is because at step 1 marked above, &ptp->n_vclocks_mux is the lock of the parent (physical) PTP clock, and at step 2, the lock is of the child (virtual) PTP clock. They are different locks of different devices.
In this situation there is no real deadlock, the lockdep warning is caused by the fact that the mutexes have the same lock class on both the parent and the child. Functionally it is fine.
Proposed alternative solution
-----------------------------
We must reintroduce the body of ptp_vclock_in_use() mostly as it was structured prior to the blamed commit, but avoid the lockdep warning.
Based on the fact that vclocks cannot be nested on top of one another (ptp_is_attribute_visible() hides n_vclocks for virtual clocks), we already know that ptp->n_vclocks is zero for a virtual clock. And ptp->is_virtual_clock is a runtime invariant, established at ptp_clock_register() time and never changed. There is no need to serialize on any mutex in order to read ptp->is_virtual_clock, and we take advantage of that by moving it outside the lock.
Thus, virtual clocks do not need to acquire &ptp->n_vclocks_mux at all, and step 2 in the code walkthrough above can simply go away. We can simply return false to the question "ptp_vclock_in_use(a virtual clock)".
Other notes
-----------
Releasing &ptp->n_vclocks_mux before ptp_vclock_in_use() returns execution seems racy, because the returned value can become stale as soon as the function returns and before the return value is used (i.e. n_vclocks_store() can run any time). The locking requirement should somehow be transferred to the caller, to ensure a longer life time for the returned value, but this seems out of scope for this severe bug fix.
Because we are also fixing up the logic from the original commit, there is another Fixes: tag for that.
--- drivers/ptp/ptp_private.h | 22 +++++++++++++++++++++- 1 file changed, 21 insertions(+), 1 deletion(-)
drivers/ptp/ptp_private.h +21 -1
@@ -98,7 +98,27 @@ static inline int queue_cnt(const struct timestamp_event_queue *q)/* Check if ptp virtual clock is in use */ static inline bool ptp_vclock_in_use (struct ptp_clock *ptp) { -return !ptp->is_virtual_clock; +bool in_use = false; + +/* Virtual clocks can't be stacked on top of virtual clocks. +* Avoid acquiring the n_vclocks_mux on virtual clocks, to allow this +* function to be called from code paths where the n_vclocks_mux of the +* parent physical clock is already held. Functionally that's not an +* issue, but lockdep would complain, because they have the same lock +* class. +*/ +if (ptp->is_virtual_clock) +return false; + +if (mutex_lock_interruptible (&ptp->n_vclocks_mux)) +return true; + +if (ptp->n_vclocks) + in_use = true; + +mutex_unlock (&ptp->n_vclocks_mux); + +return in_use; }/* Check if ptp clock shall be free running */
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 536/592 ptp: allow reading of currently dialed frequency to succeed on free-running clocks
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vladimir Oltean <[email protected]>
[ Upstream commit aa112cbc5f0ac6f3b44d829005bf34005d9fe9bb ]
There is a bug in ptp_clock_adjtime() which makes it refuse the operation even if we just want to read the current clock dialed frequency, not modify anything (tx->modes == 0). That should be possible even if the clock is free-running. For context, the kernel UAPI is the same for getting and setting the frequency of a POSIX clock.
For example, ptp4l errors out at clock_create() -> clockadj_get_freq() -> clock_adjtime() time, when it should logically only have failed on actual adjustments to the clock, aka if the clock was configured as slave. But in master mode it should work.
This was discovered when examining the issue described in the previous commit, where ptp_clock_freerun() returned true despite n_vclocks being zero.
--- drivers/ptp/ptp_clock.c | 3 ++- 1 file changed, 2 insertions(+), 1 deletion(-)
drivers/ptp/ptp_clock.c +2 -1
@@ -121,7 +121,8 @@ static int ptp_clock_adjtime(struct posix_clock *pc, struct __kernel_timex *tx)struct ptp_clock_info *ops;int err = -EOPNOTSUPP ; -if (ptp_clock_freerun (ptp)) { +if (tx->modes & (ADJ_SETOFFSET |ADJ_FREQUENCY |ADJ_OFFSET ) && +ptp_clock_freerun (ptp)) {pr_err ("ptp: physical clock is free running\n" );return -EBUSY ; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 537/592 wifi: carl9170: do not ping device which has failed to load firmware
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Dmitry Antipov <[email protected]>
[ Upstream commit 15d25307692312cec4b57052da73387f91a2e870 ]
Syzkaller reports [1, 2] crashes caused by an attempts to ping the device which has failed to load firmware. Since such a device doesn't pass 'ieee80211_register_hw()', an internal workqueue managed by 'ieee80211_queue_work()' is not yet created and an attempt to queue work on it causes null-ptr-deref.
[1] https://syzkaller.appspot.com/bug?extid=9a4aec827829942045ff
[2] https://syzkaller.appspot.com/bug?extid=0d8afba53e8fb2633217
--- drivers/net/wireless/ath/carl9170/usb.c | 19 +++++++++++++------ 1 file changed, 13 insertions(+), 6 deletions(-)
drivers/net/wireless/ath/carl9170/usb.c +13 -6
@@ -438,14 +438,21 @@ static void carl9170_usb_rx_complete(struct urb *urb)if (atomic_read (&ar->rx_anch_urbs) ==0 ) {/* -* The system is too slow to cope with -* the enormous workload. We have simply -* run out of active rx urbs and this -* unfortunately leads to an unpredictable -* device. +* At this point, either the system is too slow to +* cope with the enormous workload (so we have simply +* run out of active rx urbs and this unfortunately +* leads to an unpredictable device), or the device +* is not fully functional after an unsuccessful +* firmware loading attempts (so it doesn't pass +* ieee80211_register_hw() and there is no internal +* workqueue at all). */ -ieee80211_queue_work (ar->hw, &ar->ping_work); +if (ar->registered) +ieee80211_queue_work (ar->hw, &ar->ping_work); +else +pr_warn_once ("device %s is not registered\n" , +dev_name (&ar->udev->dev)); } }else {/*
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 538/592 mpls: Use rcu_dereference_rtnl() in mpls_route_input_rcu().
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 6dbb0d97c5096072c78a6abffe393584e57ae945 ]
As syzbot reported [0], mpls_route_input_rcu() can be called from mpls_getroute(), where is under RTNL.
net->mpls.platform_label is only updated under RTNL.
Let's use rcu_dereference_rtnl() in mpls_route_input_rcu() to silence the splat.
[0]:
WARNING: suspicious RCU usage 6.15.0-rc7-syzkaller-00082-g5cdb2c77c4c3 #0 Not tainted
----------------------------net/mpls/af_mpls.c:84 suspicious rcu_dereference_check() usage!
other info that might help us debug this:
rcu_scheduler_active = 2, debug_locks = 1
1 lock held by syz.2.4451/17730:
#0: ffffffff9012a3e8 (rtnl_mutex){+.+.}-{4:4}, at: rtnl_lock net/core/rtnetlink.c:80 [inline] #0: ffffffff9012a3e8 (rtnl_mutex){+.+.}-{4:4}, at: rtnetlink_rcv_msg+0x371/0xe90 net/core/rtnetlink.c:6961
stack backtrace: CPU: 1 UID: 0 PID: 17730 Comm: syz.2.4451 Not tainted 6.15.0-rc7-syzkaller-00082-g5cdb2c77c4c3 #0 PREEMPT(full) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025 Call Trace: <TASK> __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x16c/0x1f0 lib/dump_stack.c:120 lockdep_rcu_suspicious+0x166/0x260 kernel/locking/lockdep.c:6865 mpls_route_input_rcu+0x1d4/0x200 net/mpls/af_mpls.c:84 mpls_getroute+0x621/0x1ea0 net/mpls/af_mpls.c:2381 rtnetlink_rcv_msg+0x3c9/0xe90 net/core/rtnetlink.c:6964 netlink_rcv_skb+0x16d/0x440 net/netlink/af_netlink.c:2534 netlink_unicast_kernel net/netlink/af_netlink.c:1313 [inline] netlink_unicast+0x53a/0x7f0 net/netlink/af_netlink.c:1339 netlink_sendmsg+0x8d1/0xdd0 net/netlink/af_netlink.c:1883 sock_sendmsg_nosec net/socket.c:712 [inline] __sock_sendmsg net/socket.c:727 [inline] ____sys_sendmsg+0xa98/0xc70 net/socket.c:2566 ___sys_sendmsg+0x134/0x1d0 net/socket.c:2620 __sys_sendmmsg+0x200/0x420 net/socket.c:2709 __do_sys_sendmmsg net/socket.c:2736 [inline] __se_sys_sendmmsg net/socket.c:2733 [inline] __x64_sys_sendmmsg+0x9c/0x100 net/socket.c:2733 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0xcd/0x230 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f0a2818e969 Code: ff ff c3 66 2e 0f 1f 84 00 00 00 00 00 0f 1f 40 00 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f0a28f52038 EFLAGS: 00000246 ORIG_RAX: 0000000000000133 RAX: ffffffffffffffda RBX: 00007f0a283b5fa0 RCX: 00007f0a2818e969 RDX: 0000000000000003 RSI: 0000200000000080 RDI: 0000000000000003 RBP: 00007f0a28210ab1 R08: 0000000000000000 R09: 0000000000000000 R10: 0000000000000000 R11: 0000000000000246 R12: 0000000000000000 R13: 0000000000000000 R14: 00007f0a283b5fa0 R15: 00007ffce5e9f268 </TASK>
--- net/mpls/af_mpls.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/mpls/af_mpls.c +2 -2
@@ -81,8 +81,8 @@ static struct mpls_route *mpls_route_input_rcu(struct net *net, unsigned index)if (index < net->mpls.platform_labels) {struct mpls_route __rcu **platform_label = -rcu_dereference (net->mpls.platform_label); - rt =rcu_dereference (platform_label[index]); +rcu_dereference_rtnl (net->mpls.platform_label); + rt =rcu_dereference_rtnl (platform_label[index]); }return rt; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 539/592 atm: atmtcp: Free invalid length skb in atmtcp_c_send().
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 2f370ae1fb6317985f3497b1bb80d457508ca2f7 ]
syzbot reported the splat below. [0]
vcc_sendmsg() copies data passed from userspace to skb and passes it to vcc->dev->ops->send().
atmtcp_c_send() accesses skb->data as struct atmtcp_hdr after checking if skb->len is 0, but it's not enough.
Also, when skb->len == 0, skb and sk (vcc) were leaked because dev_kfree_skb() is not called and sk_wmem_alloc adjustment is missing to revert atm_account_tx() in vcc_sendmsg(), which is expected to be done in atm_pop_raw().
Let's properly free skb with an invalid length in atmtcp_c_send().
[0]:
BUG: KMSAN: uninit-value in atmtcp_c_send+0x255/0xed0 drivers/atm/atmtcp.c:294 atmtcp_c_send+0x255/0xed0 drivers/atm/atmtcp.c:294 vcc_sendmsg+0xd7c/0xff0 net/atm/common.c:644 sock_sendmsg_nosec net/socket.c:712 [inline] __sock_sendmsg+0x330/0x3d0 net/socket.c:727 ____sys_sendmsg+0x7e0/0xd80 net/socket.c:2566 ___sys_sendmsg+0x271/0x3b0 net/socket.c:2620 __sys_sendmsg net/socket.c:2652 [inline] __do_sys_sendmsg net/socket.c:2657 [inline] __se_sys_sendmsg net/socket.c:2655 [inline] __x64_sys_sendmsg+0x211/0x3e0 net/socket.c:2655 x64_sys_call+0x32fb/0x3db0 arch/x86/include/generated/asm/syscalls_64.h:47 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0xd9/0x210 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f
Uninit was created at:
slab_post_alloc_hook mm/slub.c:4154 [inline] slab_alloc_node mm/slub.c:4197 [inline] kmem_cache_alloc_node_noprof+0x818/0xf00 mm/slub.c:4249 kmalloc_reserve+0x13c/0x4b0 net/core/skbuff.c:579 __alloc_skb+0x347/0x7d0 net/core/skbuff.c:670 alloc_skb include/linux/skbuff.h:1336 [inline] vcc_sendmsg+0xb40/0xff0 net/atm/common.c:628 sock_sendmsg_nosec net/socket.c:712 [inline] __sock_sendmsg+0x330/0x3d0 net/socket.c:727 ____sys_sendmsg+0x7e0/0xd80 net/socket.c:2566 ___sys_sendmsg+0x271/0x3b0 net/socket.c:2620 __sys_sendmsg net/socket.c:2652 [inline] __do_sys_sendmsg net/socket.c:2657 [inline] __se_sys_sendmsg net/socket.c:2655 [inline] __x64_sys_sendmsg+0x211/0x3e0 net/socket.c:2655 x64_sys_call+0x32fb/0x3db0 arch/x86/include/generated/asm/syscalls_64.h:47 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0xd9/0x210 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f CPU: 1 UID: 0 PID: 5798 Comm: syz-executor192 Not tainted 6.16.0-rc1-syzkaller-00010-g2c4a1f3fe03e #0 PREEMPT(undef) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025
--- drivers/atm/atmtcp.c | 4 +++- 1 file changed, 3 insertions(+), 1 deletion(-)
drivers/atm/atmtcp.c +3 -1
@@ -288,7 +288,9 @@ static int atmtcp_c_send(struct atm_vcc *vcc,struct sk_buff *skb)struct sk_buff *new_skb;int result =0 ; -if (!skb->len)return 0 ; +if (skb->len <sizeof (struct atmtcp_hdr )) + goto done; + dev = vcc->dev_data; hdr = (struct atmtcp_hdr *) skb->data;if (hdr->length ==ATMTCP_HDR_MAGIC ) {
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 540/592 tcp: fix tcp_packet_delayed() for tcp_is_non_sack_preventing_reopen() behavior
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Neal Cardwell <[email protected]>
[ Upstream commit d0fa59897e049e84432600e86df82aab3dce7aa5 ]
After the following commit from 2024:
commit e37ab7373696 ("tcp: fix to allow timestamp undo if no retransmits were sent")
...there was buggy behavior where TCP connections without SACK support could easily see erroneous undo events at the end of fast recovery or RTO recovery episodes. The erroneous undo events could cause those connections to suffer repeated loss recovery episodes and high retransmit rates.
The problem was an interaction between the non-SACK behavior on these connections and the undo logic. The problem is that, for non-SACK connections at the end of a loss recovery episode, if snd_una == high_seq, then tcp_is_non_sack_preventing_reopen() holds steady in CA_Recovery or CA_Loss, but clears tp->retrans_stamp to 0. Then upon the next ACK the "tcp: fix to allow timestamp undo if no retransmits were sent" logic saw the tp->retrans_stamp at 0 and erroneously concluded that no data was retransmitted, and erroneously performed an undo of the cwnd reduction, restoring cwnd immediately to the value it had before loss recovery. This caused an immediate burst of traffic and build-up of queues and likely another immediate loss recovery episode.
This commit fixes tcp_packet_delayed() to ignore zero retrans_stamp values for non-SACK connections when snd_una is at or above high_seq, because tcp_is_non_sack_preventing_reopen() clears retrans_stamp in this case, so it's not a valid signal that we can undo.
Note that the commit named in the Fixes footer restored long-present behavior from roughly 2005-2019, so apparently this bug was present for a while during that era, and this was simply not caught.
--- net/ipv4/tcp_input.c | 37 +++++++++++++++++++++++++------------ 1 file changed, 25 insertions(+), 12 deletions(-)
net/ipv4/tcp_input.c +25 -12
@@ -2480,20 +2480,33 @@ static inline bool tcp_packet_delayed(const struct tcp_sock *tp) {const struct sock *sk = (const struct sock *)tp; -if (tp->retrans_stamp && -tcp_tsopt_ecr_before (tp, tp->retrans_stamp)) -return true;/* got echoed TS before first retransmission */ - -/* Check if nothing was retransmitted (retrans_stamp==0), which may -* happen in fast recovery due to TSQ. But we ignore zero retrans_stamp -* in TCP_SYN_SENT, since when we set FLAG_SYN_ACKED we also clear -* retrans_stamp even if we had retransmitted the SYN. +/* Received an echoed timestamp before the first retransmission? */ +if (tp->retrans_stamp) +return tcp_tsopt_ecr_before (tp, tp->retrans_stamp); + +/* We set tp->retrans_stamp upon the first retransmission of a loss +* recovery episode, so normally if tp->retrans_stamp is 0 then no +* retransmission has happened yet (likely due to TSQ, which can cause +* fast retransmits to be delayed). So if snd_una advanced while +* (tp->retrans_stamp is 0 then apparently a packet was merely delayed, +* not lost. But there are exceptions where we retransmit but then +* clear tp->retrans_stamp, so we check for those exceptions. */ -if (!tp->retrans_stamp &&/* no record of a retransmit/SYN? */ - sk->sk_state !=TCP_SYN_SENT )/* not the FLAG_SYN_ACKED case? */ -return true;/* nothing was retransmitted */ -return false; +/* (1) For non-SACK connections, tcp_is_non_sack_preventing_reopen() +* clears tp->retrans_stamp when snd_una == high_seq. +*/ +if (!tcp_is_sack (tp) && !before (tp->snd_una, tp->high_seq)) +return false; + +/* (2) In TCP_SYN_SENT tcp_clean_rtx_queue() clears tp->retrans_stamp +* when setting FLAG_SYN_ACKED is set, even if the SYN was +* retransmitted. +*/ +if (sk->sk_state ==TCP_SYN_SENT ) +return false; + +return true;/* tp->retrans_stamp is zero; no retransmit yet */ }/* Undo procedures. */
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 541/592 io_uring: fix potential page leak in io_sqe_buffer_register()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Penglei Jiang <[email protected]>
[ Upstream commit e1c75831f682eef0f68b35723437146ed86070b1 ]
If allocation of the 'imu' fails, then the existing pages aren't unpinned in the error path. This is mostly a theoretical issue, requiring fault injection to hit.
Move unpin_user_pages() to unified error handling to fix the page leak issue.
--- io_uring/rsrc.c | 6 +++--- 1 file changed, 3 insertions(+), 3 deletions(-)
io_uring/rsrc.c +3 -3
@@ -810,10 +810,8 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx, imu->nr_bvecs = nr_pages; ret =io_buffer_account_pin (ctx, pages, nr_pages, imu, last_hpage); -if (ret) { -unpin_user_pages (pages, nr_pages); +if (ret) goto done; - } size = iov->iov_len;/* store original address for later verification */
@@ -843,6 +841,8 @@ static struct io_rsrc_node *io_sqe_buffer_register(struct io_ring_ctx *ctx,if (ret) {if (imu)io_free_imu (ctx, imu); +if (pages) +unpin_user_pages (pages, nr_pages);io_cache_free (&ctx->node_cache, node); node =ERR_PTR (ret); }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 542/592 drm/amdkfd: move SDMA queue reset capability check to node_show
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jesse.Zhang <[email protected]>
[ Upstream commit a55737dab6ba63eb4241e9c6547629058af31e12 ]
Relocate the per-SDMA queue reset capability check from kfd_topology_set_capabilities() to node_show() to ensure we read the latest value of sdma.supported_reset after all IP blocks are initialized.
--- drivers/gpu/drm/amd/amdkfd/kfd_topology.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/gpu/drm/amd/amdkfd/kfd_topology.c +4 -2
@@ -510,6 +510,10 @@ static ssize_t node_show(struct kobject *kobj, struct attribute *attr, dev->node_props.capability |=HSA_CAP_AQL_QUEUE_DOUBLE_MAP ; +if (KFD_GC_VERSION (dev->gpu) <IP_VERSION (10 ,0 ,0 ) && + (dev->gpu->adev->sdma.supported_reset &AMDGPU_RESET_TYPE_PER_QUEUE )) + dev->node_props.capability2 |=HSA_CAP2_PER_SDMA_QUEUE_RESET_SUPPORTED ; +sysfs_show_32bit_prop (buffer, offs,"max_engine_clk_fcompute" , dev->node_props.max_engine_clk_fcompute);
@@ -2001,8 +2005,6 @@ static void kfd_topology_set_capabilities(struct kfd_topology_device *dev)if (!amdgpu_sriov_vf (dev->gpu->adev)) dev->node_props.capability |=HSA_CAP_PER_QUEUE_RESET_SUPPORTED ; -if (dev->gpu->adev->sdma.supported_reset &AMDGPU_RESET_TYPE_PER_QUEUE ) - dev->node_props.capability2 |=HSA_CAP2_PER_SDMA_QUEUE_RESET_SUPPORTED ; }else { dev->node_props.debug_prop |=HSA_DBG_WATCH_ADDR_MASK_LO_BIT_GFX10 |HSA_DBG_WATCH_ADDR_MASK_HI_BIT ;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 543/592 Octeontx2-pf: Fix Backpresure configuration
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Hariprasad Kelam <[email protected]>
[ Upstream commit 9ac8d0c640a161486eaf71d1999ee990fad62947 ]
NIX block can receive packets from multiple links such as MAC (RPM), LBK and CPT.
-----------------
RPM --| NIX |
-----------------
|
|
LBKEach link supports multiple channels for example RPM link supports 16 channels. In case of link oversubsribe, NIX will assert backpressure on receive channels.
The previous patch considered a single channel per link, resulting in backpressure not being enabled on the remaining channels
--- drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/marvell/octeontx2/nic/otx2_common.c +2 -2
@@ -1822,7 +1822,7 @@ int otx2_nix_config_bp(struct otx2_nic *pfvf, bool enable) req->chan_cnt =IEEE_8021QAZ_MAX_TCS ; req->bpid_per_chan =1 ; }else { - req->chan_cnt =1 ; + req->chan_cnt = pfvf->hw.rx_chan_cnt; req->bpid_per_chan =0 ; }
@@ -1847,7 +1847,7 @@ int otx2_nix_cpt_config_bp(struct otx2_nic *pfvf, bool enable) req->chan_cnt =IEEE_8021QAZ_MAX_TCS ; req->bpid_per_chan =1 ; }else { - req->chan_cnt =1 ; + req->chan_cnt = pfvf->hw.rx_chan_cnt; req->bpid_per_chan =0 ; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 544/592 tipc: fix null-ptr-deref when acquiring remote ip of ethernet bearer
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Haixia Qu <[email protected]>
[ Upstream commit f82727adcf2992822e12198792af450a76ebd5ef ]
The reproduction steps:
1. create a tun interface
2. enable l2 bearer
3. TIPC_NL_UDP_GET_REMOTEIP with media name set to tun
tipc: Started in network mode
tipc: Node identity 8af312d38a21, cluster identity 4711
tipc: Enabled bearer <eth:syz_tun>, priority 1
Oops: general protection fault KASAN: null-ptr-deref in range CPU: 1 UID: 1000 PID: 559 Comm: poc Not tainted 6.16.0-rc1+ #117 PREEMPT Hardware name: QEMU Ubuntu 24.04 PC RIP: 0010:tipc_udp_nl_dump_remoteip+0x4a4/0x8f0
the ub was in fact a struct dev.
when bid != 0 && skip_cnt != 0, bearer_list[bid] may be NULL or other media when other thread changes it.
fix this by checking media_id.
--- net/tipc/udp_media.c | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
net/tipc/udp_media.c +2 -2
@@ -489,7 +489,7 @@ int tipc_udp_nl_dump_remoteip(struct sk_buff *skb, struct netlink_callback *cb)rtnl_lock (); b =tipc_bearer_find (net, bname); -if (!b) { +if (!b || b->bcast_addr.media_id !=TIPC_MEDIA_TYPE_UDP ) {rtnl_unlock ();return -EINVAL ; }
@@ -500,7 +500,7 @@ int tipc_udp_nl_dump_remoteip(struct sk_buff *skb, struct netlink_callback *cb)rtnl_lock (); b =rtnl_dereference (tn->bearer_list[bid]); -if (!b) { +if (!b || b->bcast_addr.media_id !=TIPC_MEDIA_TYPE_UDP ) {rtnl_unlock ();return -EINVAL ; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 545/592 tcp: fix passive TFO socket having invalid NAPI ID
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Wei <[email protected]>
[ Upstream commit dbe0ca8da1f62b6252e7be6337209f4d86d4a914 ]
There is a bug with passive TFO sockets returning an invalid NAPI ID 0 from SO_INCOMING_NAPI_ID. Normally this is not an issue, but zero copy receive relies on a correct NAPI ID to process sockets on the right queue.
Fix by adding a sk_mark_napi_id_set().
--- net/ipv4/tcp_fastopen.c | 3 +++ 1 file changed, 3 insertions(+)
net/ipv4/tcp_fastopen.c +3 -0
@@ -3,6 +3,7 @@#include <linux/tcp.h> #include <linux/rcupdate.h> #include <net/tcp.h> +#include <net/busy_poll.h> void tcp_fastopen_init_key_once (struct net *net) {
@@ -279,6 +280,8 @@ static struct sock *tcp_fastopen_create_child(struct sock *sk,refcount_set (&req->rsk_refcnt,2 ); +sk_mark_napi_id_set (child, skb); +/* Now finish processing the fastopen child socket. */ tcp_init_transfer (child,BPF_SOCK_OPS_PASSIVE_ESTABLISHED_CB , skb);
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 546/592 eth: fbnic: avoid double free when failing to DMA-map FW msg
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Jakub Kicinski <[email protected]>
[ Upstream commit 5bd1bafd4474ee26f504b41aba11f3e2a1175b88 ]
The semantics are that caller of fbnic_mbx_map_msg() retains the ownership of the message on error. All existing callers dutifully free the page.
--- drivers/net/ethernet/meta/fbnic/fbnic_fw.c | 5 +---- 1 file changed, 1 insertion(+), 4 deletions(-)
drivers/net/ethernet/meta/fbnic/fbnic_fw.c +1 -4
@@ -127,11 +127,8 @@ static int fbnic_mbx_map_msg(struct fbnic_dev *fbd, int mbx_idx,return -EBUSY ; addr =dma_map_single (fbd->dev, msg,PAGE_SIZE , direction); -if (dma_mapping_error (fbd->dev, addr)) { -free_page ((unsigned long )msg); - +if (dma_mapping_error (fbd->dev, addr))return -ENOSPC ; - } mbx->buf_info[tail].msg = msg; mbx->buf_info[tail].addr = addr;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 547/592 net: lan743x: fix potential out-of-bounds write in lan743x_ptp_io_event_clock_get()
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Alexey Kodanev <[email protected]>
[ Upstream commit e353b0854d3a1a31cb061df8d022fbfea53a0f24 ]
Before calling lan743x_ptp_io_event_clock_get(), the 'channel' value is checked against the maximum value of PCI11X1X_PTP_IO_MAX_CHANNELS(8). This seems correct and aligns with the PTP interrupt status register (PTP_INT_STS) specifications.
However, lan743x_ptp_io_event_clock_get() writes to ptp->extts[] with only LAN743X_PTP_N_EXTTS(4) elements, using channel as an index:
lan743x_ptp_io_event_clock_get(..., u8 channel,...)
{
...
/* Update Local timestamp */
extts = &ptp->extts[channel];
extts->ts.tv_sec = sec;
...
}To avoid an out-of-bounds write and utilize all the supported GPIO inputs, set LAN743X_PTP_N_EXTTS to 8.
Detected using the static analysis tool - Svace.
--- drivers/net/ethernet/microchip/lan743x_ptp.h | 4 ++-- 1 file changed, 2 insertions(+), 2 deletions(-)
drivers/net/ethernet/microchip/lan743x_ptp.h +2 -2
@@ -18,9 +18,9 @@*/ #define LAN743X_PTP_N_EVENT_CHAN 2#define LAN743X_PTP_N_PEROUT LAN743X_PTP_N_EVENT_CHAN -#define LAN743X_PTP_N_EXTTS 4 -#define LAN743X_PTP_N_PPS 0#define PCI11X1X_PTP_IO_MAX_CHANNELS 8 +#define LAN743X_PTP_N_EXTTS PCI11X1X_PTP_IO_MAX_CHANNELS +#define LAN743X_PTP_N_PPS 0#define PTP_CMD_CTL_TIMEOUT_CNT 50struct lan743x_adapter ;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 548/592 ublk: santizize the arguments from userspace when adding a device
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Ronnie Sahlberg <[email protected]>
[ Upstream commit 8c8472855884355caf3d8e0c50adf825f83454b2 ]
Sanity check the values for queue depth and number of queues we get from userspace when adding a device.
--- drivers/block/ublk_drv.c | 3 +++ 1 file changed, 3 insertions(+)
drivers/block/ublk_drv.c +3 -0
@@ -2710,6 +2710,9 @@ static int ublk_ctrl_add_dev(const struct ublksrv_ctrl_cmd *header)if (copy_from_user (&info, argp,sizeof (info)))return -EFAULT ; +if (info.queue_depth >UBLK_MAX_QUEUE_DEPTH || info.nr_hw_queues >UBLK_MAX_NR_QUEUES ) +return -EINVAL ; +if (capable (CAP_SYS_ADMIN )) info.flags &= ~UBLK_F_UNPRIVILEGED_DEV ;else if (!(info.flags &UBLK_F_UNPRIVILEGED_DEV ))
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 549/592 drm/xe/bmg: Update Wa_16023588340
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Vinay Belgaumkar <[email protected]>
[ Upstream commit 16c1241b08751a67cd7a0221ea9f82b0b02806f4 ]
This allows for additional L2 caching modes.
--- drivers/gpu/drm/xe/xe_gt.c | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-)
drivers/gpu/drm/xe/xe_gt.c +1 -1
@@ -116,7 +116,7 @@ static void xe_gt_enable_host_l2_vram(struct xe_gt *gt)xe_gt_mcr_multicast_write (gt,XE2_GAMREQSTRM_CTRL , reg); } -xe_gt_mcr_multicast_write (gt,XEHPC_L3CLOS_MASK (3 ),0x3 ); +xe_gt_mcr_multicast_write (gt,XEHPC_L3CLOS_MASK (3 ),0xF );xe_force_wake_put (gt_to_fw (gt), fw_ref); }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 550/592 calipso: Fix null-ptr-deref in calipso_req_{set,del}attr().
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Kuniyuki Iwashima <[email protected]>
[ Upstream commit 10876da918fa1aec0227fb4c67647513447f53a9 ]
syzkaller reported a null-ptr-deref in sock_omalloc() while allocating a CALIPSO option. [0]
The NULL is of struct sock, which was fetched by sk_to_full_sk() in calipso_req_setattr().
Since commit a1a5344ddbe8 ("tcp: avoid two atomic ops for syncookies"), reqsk->rsk_listener could be NULL when SYN Cookie is returned to its client, as hinted by the leading SYN Cookie log.
Here are 3 options to fix the bug:
1) Return 0 in calipso_req_setattr()
2) Return an error in calipso_req_setattr()
3) Alaways set rsk_listener
1) is no go as it bypasses LSM, but 2) effectively disables SYN Cookie for CALIPSO. 3) is also no go as there have been many efforts to reduce atomic ops and make TCP robust against DDoS. See also commit 3b24d854cb35 ("tcp/dccp: do not touch listener sk_refcnt under synflood").
As of the blamed commit, SYN Cookie already did not need refcounting, and no one has stumbled on the bug for 9 years, so no CALIPSO user will care about SYN Cookie.
Let's return an error in calipso_req_setattr() and calipso_req_delattr() in the SYN Cookie case.
This can be reproduced by [1] on Fedora and now connect() of nc times out.
[0]:
TCP: request_sock_TCPv6: Possible SYN flooding on port [::]:20002. Sending cookies.
Oops: general protection fault, probably for non-canonical address 0xdffffc0000000006: 0000 [#1] PREEMPT SMP KASAN NOPTI KASAN: null-ptr-deref in range [0x0000000000000030-0x0000000000000037] CPU: 3 UID: 0 PID: 12262 Comm: syz.1.2611 Not tainted 6.14.0 #2 Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS rel-1.16.3-0-ga6ed6b701f0a-prebuilt.qemu.org 04/01/2014 RIP: 0010:read_pnet include/net/net_namespace.h:406 [inline] RIP: 0010:sock_net include/net/sock.h:655 [inline] RIP: 0010:sock_kmalloc+0x35/0x170 net/core/sock.c:2806 Code: 89 d5 41 54 55 89 f5 53 48 89 fb e8 25 e3 c6 fd e8 f0 91 e3 00 48 8d 7b 30 48 b8 00 00 00 00 00 fc ff df 48 89 fa 48 c1 ea 03 <80> 3c 02 00 0f 85 26 01 00 00 48 b8 00 00 00 00 00 fc ff df 4c 8b RSP: 0018:ffff88811af89038 EFLAGS: 00010216 RAX: dffffc0000000000 RBX: 0000000000000000 RCX: ffff888105266400 RDX: 0000000000000006 RSI: ffff88800c890000 RDI: 0000000000000030 RBP: 0000000000000050 R08: 0000000000000000 R09: ffff88810526640e R10: ffffed1020a4cc81 R11: ffff88810526640f R12: 0000000000000000 R13: 0000000000000820 R14: ffff888105266400 R15: 0000000000000050 FS: 00007f0653a07640(0000) GS:ffff88811af80000(0000) knlGS:0000000000000000 CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 CR2: 00007f863ba096f4 CR3: 00000000163c0005 CR4: 0000000000770ef0 PKRU: 80000000 Call Trace: <IRQ> ipv6_renew_options+0x279/0x950 net/ipv6/exthdrs.c:1288 calipso_req_setattr+0x181/0x340 net/ipv6/calipso.c:1204 calipso_req_setattr+0x56/0x80 net/netlabel/netlabel_calipso.c:597 netlbl_req_setattr+0x18a/0x440 net/netlabel/netlabel_kapi.c:1249 selinux_netlbl_inet_conn_request+0x1fb/0x320 security/selinux/netlabel.c:342 selinux_inet_conn_request+0x1eb/0x2c0 security/selinux/hooks.c:5551 security_inet_conn_request+0x50/0xa0 security/security.c:4945 tcp_v6_route_req+0x22c/0x550 net/ipv6/tcp_ipv6.c:825 tcp_conn_request+0xec8/0x2b70 net/ipv4/tcp_input.c:7275 tcp_v6_conn_request+0x1e3/0x440 net/ipv6/tcp_ipv6.c:1328 tcp_rcv_state_process+0xafa/0x52b0 net/ipv4/tcp_input.c:6781 tcp_v6_do_rcv+0x8a6/0x1a40 net/ipv6/tcp_ipv6.c:1667 tcp_v6_rcv+0x505e/0x5b50 net/ipv6/tcp_ipv6.c:1904 ip6_protocol_deliver_rcu+0x17c/0x1da0 net/ipv6/ip6_input.c:436 ip6_input_finish+0x103/0x180 net/ipv6/ip6_input.c:480 NF_HOOK include/linux/netfilter.h:314 [inline] NF_HOOK include/linux/netfilter.h:308 [inline] ip6_input+0x13c/0x6b0 net/ipv6/ip6_input.c:491 dst_input include/net/dst.h:469 [inline] ip6_rcv_finish net/ipv6/ip6_input.c:79 [inline] ip6_rcv_finish+0xb6/0x490 net/ipv6/ip6_input.c:69 NF_HOOK include/linux/netfilter.h:314 [inline] NF_HOOK include/linux/netfilter.h:308 [inline] ipv6_rcv+0xf9/0x490 net/ipv6/ip6_input.c:309 __netif_receive_skb_one_core+0x12e/0x1f0 net/core/dev.c:5896 __netif_receive_skb+0x1d/0x170 net/core/dev.c:6009 process_backlog+0x41e/0x13b0 net/core/dev.c:6357 __napi_poll+0xbd/0x710 net/core/dev.c:7191 napi_poll net/core/dev.c:7260 [inline] net_rx_action+0x9de/0xde0 net/core/dev.c:7382 handle_softirqs+0x19a/0x770 kernel/softirq.c:561 do_softirq.part.0+0x36/0x70 kernel/softirq.c:462 </IRQ> <TASK> do_softirq arch/x86/include/asm/preempt.h:26 [inline] __local_bh_enable_ip+0xf1/0x110 kernel/softirq.c:389 local_bh_enable include/linux/bottom_half.h:33 [inline] rcu_read_unlock_bh include/linux/rcupdate.h:919 [inline] __dev_queue_xmit+0xc2a/0x3c40 net/core/dev.c:4679 dev_queue_xmit include/linux/netdevice.h:3313 [inline] neigh_hh_output include/net/neighbour.h:523 [inline] neigh_output include/net/neighbour.h:537 [inline] ip6_finish_output2+0xd69/0x1f80 net/ipv6/ip6_output.c:141 __ip6_finish_output net/ipv6/ip6_output.c:215 [inline] ip6_finish_output+0x5dc/0xd60 net/ipv6/ip6_output.c:226 NF_HOOK_COND include/linux/netfilter.h:303 [inline] ip6_output+0x24b/0x8d0 net/ipv6/ip6_output.c:247 dst_output include/net/dst.h:459 [inline] NF_HOOK include/linux/netfilter.h:314 [inline] NF_HOOK include/linux/netfilter.h:308 [inline] ip6_xmit+0xbbc/0x20d0 net/ipv6/ip6_output.c:366 inet6_csk_xmit+0x39a/0x720 net/ipv6/inet6_connection_sock.c:135 __tcp_transmit_skb+0x1a7b/0x3b40 net/ipv4/tcp_output.c:1471 tcp_transmit_skb net/ipv4/tcp_output.c:1489 [inline] tcp_send_syn_data net/ipv4/tcp_output.c:4059 [inline] tcp_connect+0x1c0c/0x4510 net/ipv4/tcp_output.c:4148 tcp_v6_connect+0x156c/0x2080 net/ipv6/tcp_ipv6.c:333 __inet_stream_connect+0x3a7/0xed0 net/ipv4/af_inet.c:677 tcp_sendmsg_fastopen+0x3e2/0x710 net/ipv4/tcp.c:1039 tcp_sendmsg_locked+0x1e82/0x3570 net/ipv4/tcp.c:1091 tcp_sendmsg+0x2f/0x50 net/ipv4/tcp.c:1358 inet6_sendmsg+0xb9/0x150 net/ipv6/af_inet6.c:659 sock_sendmsg_nosec net/socket.c:718 [inline] __sock_sendmsg+0xf4/0x2a0 net/socket.c:733 __sys_sendto+0x29a/0x390 net/socket.c:2187 __do_sys_sendto net/socket.c:2194 [inline] __se_sys_sendto net/socket.c:2190 [inline] __x64_sys_sendto+0xe1/0x1c0 net/socket.c:2190 do_syscall_x64 arch/x86/entry/common.c:52 [inline] do_syscall_64+0xc3/0x1d0 arch/x86/entry/common.c:83 entry_SYSCALL_64_after_hwframe+0x77/0x7f RIP: 0033:0x7f06553c47ed Code: 02 b8 ff ff ff ff c3 66 0f 1f 44 00 00 f3 0f 1e fa 48 89 f8 48 89 f7 48 89 d6 48 89 ca 4d 89 c2 4d 89 c8 4c 8b 4c 24 08 0f 05 <48> 3d 01 f0 ff ff 73 01 c3 48 c7 c1 a8 ff ff ff f7 d8 64 89 01 48 RSP: 002b:00007f0653a06fc8 EFLAGS: 00000246 ORIG_RAX: 000000000000002c RAX: ffffffffffffffda RBX: 00007f0655605fa0 RCX: 00007f06553c47ed RDX: 0000000000000000 RSI: 0000000000000000 RDI: 000000000000000b RBP: 00007f065545db38 R08: 0000200000000140 R09: 000000000000001c R10: f7384d4ea84b01bd R11: 0000000000000246 R12: 0000000000000000 R13: 00007f0655605fac R14: 00007f0655606038 R15: 00007f06539e7000 </TASK> Modules linked in:
[1]:
dnf install -y selinux-policy-targeted policycoreutils netlabel_tools procps-ng nmap-ncat mount -t selinuxfs none /sys/fs/selinuxload_policy
netlabelctl calipso add pass doi:1
netlabelctl map del default
netlabelctl map add default address:::1 protocol:calipso,1
sysctl net.ipv4.tcp_syncookies=2
nc -l ::1 80 &
nc ::1 80
--- net/ipv6/calipso.c | 8 ++++++++ 1 file changed, 8 insertions(+)
net/ipv6/calipso.c +8 -0
@@ -1207,6 +1207,10 @@ static int calipso_req_setattr(struct request_sock *req,struct ipv6_opt_hdr *old, *new;struct sock *sk =sk_to_full_sk (req_to_sk (req)); +/* sk is NULL for SYN+ACK w/ SYN Cookie */ +if (!sk) +return -ENOMEM ; +if (req_inet->ipv6_opt && req_inet->ipv6_opt->hopopt) old = req_inet->ipv6_opt->hopopt;else
@@ -1247,6 +1251,10 @@ static void calipso_req_delattr(struct request_sock *req)struct ipv6_txoptions *txopts;struct sock *sk =sk_to_full_sk (req_to_sk (req)); +/* sk is NULL for SYN+ACK w/ SYN Cookie */ +if (!sk) +return ; +if (!req_inet->ipv6_opt || !req_inet->ipv6_opt->hopopt)return ;
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 551/592 mlxbf_gige: return EPROBE_DEFER if PHY IRQ is not available
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: David Thompson <[email protected]>
[ Upstream commit e7ea5f5b1858ddb96b152584d5fe06e6fc623e89 ]
The message "Error getting PHY irq. Use polling instead" is emitted when the mlxbf_gige driver is loaded by the kernel before the associated gpio-mlxbf driver, and thus the call to get the PHY IRQ fails since it is not yet available. The driver probe() must return -EPROBE_DEFER if acpi_dev_gpio_irq_get_by() returns the same.
--- drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_main.c | 6 ++++-- 1 file changed, 4 insertions(+), 2 deletions(-)
drivers/net/ethernet/mellanox/mlxbf_gige/mlxbf_gige_main.c +4 -2
@@ -447,8 +447,10 @@ static int mlxbf_gige_probe(struct platform_device *pdev) priv->llu_plu_irq =platform_get_irq (pdev,MLXBF_GIGE_LLU_PLU_INTR_IDX ); phy_irq =acpi_dev_gpio_irq_get_by (ACPI_COMPANION (&pdev->dev),"phy" ,0 ); -if (phy_irq <0 ) { -dev_err (&pdev->dev,"Error getting PHY irq. Use polling instead" ); +if (phy_irq == -EPROBE_DEFER ) { + err = -EPROBE_DEFER ; + goto out; + }else if (phy_irq <0 ) { phy_irq =PHY_POLL ; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 552/592 net: atm: add lec_mutex
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit d13a3824bfd2b4774b671a75cf766a16637a0e67 ]
syzbot found its way in net/atm/lec.c, and found an error path in lecd_attach() could leave a dangling pointer in dev_lec[].
Add a mutex to protect dev_lecp[] uses from lecd_attach(), lec_vcc_attach() and lec_mcast_attach().
Following patch will use this mutex for /proc/net/atm/lec.
BUG: KASAN: slab-use-after-free in lecd_attach net/atm/lec.c:751 [inline] BUG: KASAN: slab-use-after-free in lane_ioctl+0x2224/0x23e0 net/atm/lec.c:1008 Read of size 8 at addr ffff88807c7b8e68 by task syz.1.17/6142 CPU: 1 UID: 0 PID: 6142 Comm: syz.1.17 Not tainted 6.16.0-rc1-syzkaller-00239-g08215f5486ec #0 PREEMPT(full) Hardware name: Google Google Compute Engine/Google Compute Engine, BIOS Google 05/07/2025 Call Trace: <TASK> __dump_stack lib/dump_stack.c:94 [inline] dump_stack_lvl+0x116/0x1f0 lib/dump_stack.c:120 print_address_description mm/kasan/report.c:408 [inline] print_report+0xcd/0x680 mm/kasan/report.c:521 kasan_report+0xe0/0x110 mm/kasan/report.c:634 lecd_attach net/atm/lec.c:751 [inline] lane_ioctl+0x2224/0x23e0 net/atm/lec.c:1008 do_vcc_ioctl+0x12c/0x930 net/atm/ioctl.c:159 sock_do_ioctl+0x118/0x280 net/socket.c:1190 sock_ioctl+0x227/0x6b0 net/socket.c:1311 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:907 [inline] __se_sys_ioctl fs/ioctl.c:893 [inline] __x64_sys_ioctl+0x18e/0x210 fs/ioctl.c:893 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0xcd/0x4c0 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f </TASK> Allocated by task 6132: kasan_save_stack+0x33/0x60 mm/kasan/common.c:47 kasan_save_track+0x14/0x30 mm/kasan/common.c:68 poison_kmalloc_redzone mm/kasan/common.c:377 [inline] __kasan_kmalloc+0xaa/0xb0 mm/kasan/common.c:394 kasan_kmalloc include/linux/kasan.h:260 [inline] __do_kmalloc_node mm/slub.c:4328 [inline] __kvmalloc_node_noprof+0x27b/0x620 mm/slub.c:5015 alloc_netdev_mqs+0xd2/0x1570 net/core/dev.c:11711 lecd_attach net/atm/lec.c:737 [inline] lane_ioctl+0x17db/0x23e0 net/atm/lec.c:1008 do_vcc_ioctl+0x12c/0x930 net/atm/ioctl.c:159 sock_do_ioctl+0x118/0x280 net/socket.c:1190 sock_ioctl+0x227/0x6b0 net/socket.c:1311 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:907 [inline] __se_sys_ioctl fs/ioctl.c:893 [inline] __x64_sys_ioctl+0x18e/0x210 fs/ioctl.c:893 do_syscall_x64 arch/x86/entry/syscall_64.c:63 [inline] do_syscall_64+0xcd/0x4c0 arch/x86/entry/syscall_64.c:94 entry_SYSCALL_64_after_hwframe+0x77/0x7f Freed by task 6132: kasan_save_stack+0x33/0x60 mm/kasan/common.c:47 kasan_save_track+0x14/0x30 mm/kasan/common.c:68 kasan_save_free_info+0x3b/0x60 mm/kasan/generic.c:576 poison_slab_object mm/kasan/common.c:247 [inline] __kasan_slab_free+0x51/0x70 mm/kasan/common.c:264 kasan_slab_free include/linux/kasan.h:233 [inline] slab_free_hook mm/slub.c:2381 [inline] slab_free mm/slub.c:4643 [inline] kfree+0x2b4/0x4d0 mm/slub.c:4842 free_netdev+0x6c5/0x910 net/core/dev.c:11892 lecd_attach net/atm/lec.c:744 [inline] lane_ioctl+0x1ce8/0x23e0 net/atm/lec.c:1008 do_vcc_ioctl+0x12c/0x930 net/atm/ioctl.c:159 sock_do_ioctl+0x118/0x280 net/socket.c:1190 sock_ioctl+0x227/0x6b0 net/socket.c:1311 vfs_ioctl fs/ioctl.c:51 [inline] __do_sys_ioctl fs/ioctl.c:907 [inline] __se_sys_ioctl fs/ioctl.c:893 [inline] __x64_sys_ioctl+0x18e/0x210 fs/ioctl.c:893
--- net/atm/lec.c | 7 +++++++ 1 file changed, 7 insertions(+)
net/atm/lec.c +7 -0
@@ -124,6 +124,7 @@ static unsigned char bus_mac[ETH_ALEN] = { 0xff, 0xff, 0xff, 0xff, 0xff, 0xff };/* Device structures */ static struct net_device *dev_lec[MAX_LEC_ITF ]; +static DEFINE_MUTEX (lec_mutex);#if IS_ENABLED (CONFIG_BRIDGE )static void lec_handle_bridge (struct sk_buff *skb,struct net_device *dev)
@@ -685,6 +686,7 @@ static int lec_vcc_attach(struct atm_vcc *vcc, void __user *arg)int bytes_left;struct atmlec_ioc ioc_data; +lockdep_assert_held (&lec_mutex);/* Lecd must be up in this case */ bytes_left =copy_from_user (&ioc_data, arg,sizeof (struct atmlec_ioc ));if (bytes_left !=0 )
@@ -710,6 +712,7 @@ static int lec_vcc_attach(struct atm_vcc *vcc, void __user *arg)static int lec_mcast_attach (struct atm_vcc *vcc,int arg) { +lockdep_assert_held (&lec_mutex);if (arg <0 || arg >=MAX_LEC_ITF )return -EINVAL ; arg =array_index_nospec (arg,MAX_LEC_ITF );
@@ -725,6 +728,7 @@ static int lecd_attach(struct atm_vcc *vcc, int arg)int i;struct lec_priv *priv; +lockdep_assert_held (&lec_mutex);if (arg <0 ) arg =0 ;if (arg >=MAX_LEC_ITF )
@@ -742,6 +746,7 @@ static int lecd_attach(struct atm_vcc *vcc, int arg)snprintf (dev_lec[i]->name,IFNAMSIZ ,"lec%d" , i);if (register_netdev (dev_lec[i])) {free_netdev (dev_lec[i]); + dev_lec[i] =NULL ;return -EINVAL ; }
@@ -1003,6 +1008,7 @@ static int lane_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)return -ENOIOCTLCMD ; } +mutex_lock (&lec_mutex);switch (cmd) {case ATMLEC_CTRL : err =lecd_attach (vcc, (int )arg);
@@ -1017,6 +1023,7 @@ static int lane_ioctl(struct socket *sock, unsigned int cmd, unsigned long arg)break ; } +mutex_unlock (&lec_mutex);return err; }
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 553/592 net: atm: fix /proc/net/atm/lec handling
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Eric Dumazet <[email protected]>
[ Upstream commit d03b79f459c7935cff830d98373474f440bd03ae ]
/proc/net/atm/lec must ensure safety against dev_lec[] changes.
It appears it had dev_put() calls without prior dev_hold(), leading to imbalance and UAF.
--- net/atm/lec.c | 5 +++-- 1 file changed, 3 insertions(+), 2 deletions(-)
net/atm/lec.c +3 -2
@@ -909,7 +909,6 @@ static void *lec_itf_walk(struct lec_state *state, loff_t *l) v = (dev &&netdev_priv (dev)) ?lec_priv_walk (state, l,netdev_priv (dev)) :NULL ;if (!v && dev) { -dev_put (dev);/* Partial state reset for the next time we get called */ dev =NULL ; }
@@ -933,6 +932,7 @@ static void *lec_seq_start(struct seq_file *seq, loff_t *pos) {struct lec_state *state = seq->private; +mutex_lock (&lec_mutex); state->itf =0 ; state->dev =NULL ; state->locked =NULL ;
@@ -950,8 +950,9 @@ static void lec_seq_stop(struct seq_file *seq, void *v)if (state->dev) {spin_unlock_irqrestore (&state->locked->lec_arp_lock, state->flags); -dev_put (state->dev); + state->dev =NULL ; } +mutex_unlock (&lec_mutex); }static void *lec_seq_next (struct seq_file *seq,void *v,loff_t *pos)
Signature · 1 line
2.39.5
Greg Kroah-Hartman
maintainerPATCH 554/592 tools: ynl: parse extack for sub-messages
In reply to Greg Kroah-Hartman6.15-stable review patch. If anyone has any objections, please let me know.
------------------
From: Donald Hunter <[email protected]>
[ Upstream commit 09d7ff0694ea133c50ad905fd6e548c13f8af458 ]
Extend the Python YNL extack decoding to handle sub-messages in the same way that YNL C does. This involves retaining the input values so that they are available during extack decoding.
./tools/net/ynl/pyynl/cli.py --family rt-link --do newlink --create \
--json '{
"linkinfo": {"kind": "netkit", "data": {"policy": 10} }
}'
Netlink error: Invalid argument
nl_len = 92 (76) nl_flags = 0x300 nl_type = 2
error: -22
extack: {'msg': 'Provided default xmit policy not supported', 'bad-attr': '.linkinfo.data(netkit).policy'}Stable-dep-of: 9738280aae59 ("tools: ynl: fix mixing ops and notifications on one socket")
--- tools/net/ynl/pyynl/lib/ynl.py | 39 ++++++++++++++++++++++------------ 1 file changed, 25 insertions(+), 14 deletions(-)
tools/net/ynl/pyynl/lib/ynl.py +25 -14
@@ -594,7 +594,7 @@ class YnlFamily(SpecFamily): scalar_selector = self._get_scalar(attr, value["selector"]) attr_payload = struct.pack("II", scalar_value, scalar_selector) elif attr['type'] == 'sub-message': - msg_format = self._resolve_selector(attr, search_attrs) + msg_format, _ = self._resolve_selector(attr, search_attrs) attr_payload = b'' if msg_format.fixed_header: attr_payload += self._encode_struct(msg_format.fixed_header, value)
@@ -712,10 +712,10 @@ class YnlFamily(SpecFamily): raise Exception(f"No message format for '{value}' in sub-message spec '{sub_msg}'") spec = sub_msg_spec.formats[value] - return spec + return spec, value def _decode_sub_msg(self, attr, attr_spec, search_attrs): - msg_format = self._resolve_selector(attr_spec, search_attrs) + msg_format, _ = self._resolve_selector(attr_spec, search_attrs) decoded = {} offset = 0 if msg_format.fixed_header:
@@ -787,7 +787,7 @@ class YnlFamily(SpecFamily): return rsp - def _decode_extack_path(self, attrs, attr_set, offset, target): + def _decode_extack_path(self, attrs, attr_set, offset, target, search_attrs): for attr in attrs: try: attr_spec = attr_set.attrs_by_val[attr.type]
@@ -801,26 +801,37 @@ class YnlFamily(SpecFamily): if offset + attr.full_len <= target: offset += attr.full_len continue - if attr_spec['type'] != 'nest': + + pathname = attr_spec.name + if attr_spec['type'] == 'nest': + sub_attrs = self.attr_sets[attr_spec['nested-attributes']] + search_attrs = SpaceAttrs(sub_attrs, search_attrs.lookup(attr_spec['name'])) + elif attr_spec['type'] == 'sub-message': + msg_format, value = self._resolve_selector(attr_spec, search_attrs) + if msg_format is None: + raise Exception(f"Can't resolve sub-message of {attr_spec['name']} for extack") + sub_attrs = self.attr_sets[msg_format.attr_set] + pathname += f"({value})" + else: raise Exception(f"Can't dive into {attr.type} ({attr_spec['name']}) for extack") offset += 4 - subpath = self._decode_extack_path(NlAttrs(attr.raw), - self.attr_sets[attr_spec['nested-attributes']], - offset, target) + subpath = self._decode_extack_path(NlAttrs(attr.raw), sub_attrs, + offset, target, search_attrs) if subpath is None: return None - return '.' + attr_spec.name + subpath + return '.' + pathname + subpath return None - def _decode_extack(self, request, op, extack): + def _decode_extack(self, request, op, extack, vals): if 'bad-attr-offs' not in extack: return msg = self.nlproto.decode(self, NlMsg(request, 0, op.attr_set), op) offset = self.nlproto.msghdr_size() + self._struct_size(op.fixed_header) + search_attrs = SpaceAttrs(op.attr_set, vals) path = self._decode_extack_path(msg.raw_attrs, op.attr_set, offset, - extack['bad-attr-offs']) + extack['bad-attr-offs'], search_attrs) if path: del extack['bad-attr-offs'] extack['bad-attr'] = path
@@ -1012,7 +1023,7 @@ class YnlFamily(SpecFamily): for (method, vals, flags) in ops: op = self.ops[method] msg = self._encode_message(op, vals, flags, req_seq) - reqs_by_seq[req_seq] = (op, msg, flags) + reqs_by_seq[req_seq] = (op, vals, msg, flags) payload += msg req_seq += 1
@@ -1027,9 +1038,9 @@ class YnlFamily(SpecFamily): self._recv_dbg_print(reply, nms) for nl_msg in nms: if nl_msg.nl_seq in reqs_by_seq: - (op, req_msg, req_flags) = reqs_by_seq[nl_msg.nl_seq] + (op, vals, req_msg, req_flags) = reqs_by_seq[nl_msg.nl_seq] if nl_msg.extack: - self._decode_extack(req_msg, op, nl_msg.extack) + self._decode_extack(req_msg, op, nl_msg.extack, vals) else: op = None req_flags = []
Signature · 1 line
2.39.5