summaryrefslogtreecommitdiff
AgeCommit message (Collapse)Author
2022-11-10net: mv643xx_eth: disable napi when init rxq or txq failed in mv643xx_eth_open()Zhengchao Shao
When failed to init rxq or txq in mv643xx_eth_open() for opening device, napi isn't disabled. When open mv643xx_eth device next time, it will trigger a BUG_ON() in napi_enable(). Compile tested only. Fixes: 2257e05c1705 ("mv643xx_eth: get rid of receive-side locking") Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> Link: https://lore.kernel.org/r/20221109025432.80900-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10ethernet: s2io: disable napi when start nic failed in s2io_card_up()Zhengchao Shao
When failed to start nic or add interrupt service routine in s2io_card_up() for opening device, napi isn't disabled. When open s2io device next time, it will trigger a BUG_ON()in napi_enable(). Compile tested only. Fixes: 5f490c968056 ("S2io: Fixed synchronization between scheduling of napi with card reset and close") Signed-off-by: Zhengchao Shao <shaozhengchao@huawei.com> Link: https://lore.kernel.org/r/20221109023741.131552-1-shaozhengchao@huawei.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10drm/vc4: kms: Fix IS_ERR() vs NULL check for vc4_kmsGaosheng Cui
The drm_atomic_get_new_private_obj_state() function returns NULL on error path, drm_atomic_get_old_private_obj_state() function returns NULL on error path, too, they does not return error pointers. By the way, vc4_hvs_get_new/old_global_state() should return ERR_PTR(-EINVAL), otherwise there will be null-ptr-defer issue, such as follows: In function vc4_atomic_commit_tail(): |-- old_hvs_state = vc4_hvs_get_old_global_state(state); <-- return NULL |-- if (WARN_ON(IS_ERR(old_hvs_state))) <-- no return |-- unsigned long state_rate = max(old_hvs_state->core_clock_rate, new_hvs_state->core_clock_rate); <-- null-ptr-defer Fixes: 9ec03d7f1ed3 ("drm/vc4: kms: Wait on previous FIFO users before a commit") Signed-off-by: Gaosheng Cui <cuigaosheng1@huawei.com> Signed-off-by: Maxime Ripard <maxime@cerno.tech> Link: https://patchwork.freedesktop.org/patch/msgid/20221110094445.2930509-6-cuigaosheng1@huawei.com
2022-11-10net: mana: Fix return type of mana_start_xmit()Nathan Huckleberry
The ndo_start_xmit field in net_device_ops is expected to be of type netdev_tx_t (*ndo_start_xmit)(struct sk_buff *skb, struct net_device *dev). The mismatched return type breaks forward edge kCFI since the underlying function definition does not match the function hook definition. A new warning in clang will catch this at compile time: drivers/net/ethernet/microsoft/mana/mana_en.c:382:21: error: incompatible function pointer types initializing 'netdev_tx_t (*)(struct sk_buff *, struct net_device *)' (aka 'enum netdev_tx (*)(struct sk_buff *, struct net_device *)') with an expression of type 'int (struct sk_buff *, struct net_device *)' [-Werror,-Wincompatible-function-pointer-types-strict] .ndo_start_xmit = mana_start_xmit, ^~~~~~~~~~~~~~~ 1 error generated. The return type of mana_start_xmit should be changed from int to netdev_tx_t. Reported-by: Dan Carpenter <error27@gmail.com> Link: https://github.com/ClangBuiltLinux/linux/issues/1703 Link: https://github.com/ClangBuiltLinux/linux/issues/1750 Signed-off-by: Nathan Huckleberry <nhuck@google.com> Reviewed-by: Dexuan Cui <decui@microsoft.com> [nathan: Rebase on net-next and resolve conflicts Add note about new clang warning] Signed-off-by: Nathan Chancellor <nathan@kernel.org> Link: https://lore.kernel.org/r/20221109002629.1446680-1-nathan@kernel.org Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10wifi: iwlwifi: mei: wait for the mac to stop on suspendAvraham Stern
When iwlmei driver is removed, it removes all the interfaces since CSME will take over the NIC while there is no SAP connection. When the iwlmei driver is removed due to the host being suspended, this results in the interface being down after resume since the interface was removed on suspend so mac80211 will not try to bring it up on resume. Since on suspend the mac is stopped anyway by mac80211, there is no need to remove the interface in this case. Wait for the mac to stop instead. Once the mac is stopped, the driver will not access the NIC anymore so it is safe for CSME to take over. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.371102f04586.I813a964607b3e92ea275a0cf56df57227c11ae88@changeid
2022-11-10wifi: iwlwifi: iwlmei: report disconnection as temporaryAvraham Stern
Disconnections are reported to CSME as long link down. As a result, CSME will take ownership right away so the host doesn't have a chance to reconnect, although in many cases it could quickly reconnect and preserve host connectivity. Report disconnections as temporary so CSME will give the host some grace time to reconnect. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.401665457652.Ie5de9eacc5a51cf0965a08c202caf3fbc3a91ec4@changeid
2022-11-10wifi: iwlwifi: mei: use wait_event_timeout() return valueAvraham Stern
wait_event_timeout() return value indicates whether the condition evaluated to true or not, so no need to re-take the lock and check the got_ownership flag. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.33159054626f.Ief9e2dc42f96f8044c197d32172003a5ead0f8d3@changeid
2022-11-10wifi: iwlwifi: mei: implement PLDR flowAvraham Stern
If the FW needs to do OTP re-read, the driver must notify CSME before loading the FW so CSME will not try to access the NIC during the re-read. Once the alive notification is received, CSME is notified that NIC access is allowed again. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.49eb8c6d455f.I7f0a5debb2d3d662a4151199bbec24613f324c13@changeid
2022-11-10wifi: iwlwifi: mvm: Fix getting the lowest rateIlan Peer
When setting the rate for Tx, the code tried to get the lowest allowed rate but without considering the BSS basic rates. Fix this by considering the basic rates. In addition, declare support for configuring beacon Tx rate and when configured use the configured Tx beacon rate to set beacon tx command rate. Signed-off-by: Ilan Peer <ilan.peer@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.be8403f0d5c0.I7d141646746b96310efd75fc77ca9aebc61aefcc@changeid
2022-11-10wifi: iwlwifi: mvm: support new key APIJohannes Berg
In order to support MLD, the key API is also changing to have station masks instead of just the station ID etc. Change the driver to support this, and add the new code in a new file so it's more clearly separated. For now this isn't separated at the mac80211 ops level, which we wanted to do, but we're calling these functions in a place when pre-start keys are installed in iwl_mvm_start_ap_ibss(), and the function has some glue logic to mac80211. We may want to change that later. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.ed9ccd814abc.Iacc7360de68807fbac19e5b67c86504b39cc15df@changeid
2022-11-10wifi: iwlwifi: dbg: use bit of DRAM alloc ID to store failed allocsRotem Saado
The failed_alloc variable is used as a bitmask in the loop where we check DRAM allocations. But erroneously, we were clearing the DRAM alloc IDs we removed as an integer. Fix that by clearing them as bits instead. Signed-off-by: Rotem Saado <rotem.saado@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.688dec28b1d9.I470b8d29c28d16f25f4192773f075940de7ed33c@changeid
2022-11-10wifi: iwlwifi: mvm: print an error instead of a warning on invalid rateLuca Coelho
In some rare occasions, the firmware may let some frames with invalid rates, such as CCK rates on the high band, come through. This causes the driver to issue a warning, but since this is a possible issue and it's not really a bug in the driver, convert the warning into an error. Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.3d3673c70556.I13464b11d405fd6021618b0a32404cecb7e9ac51@changeid
2022-11-10wifi: iwlwifi: cfg: disable STBC for BL step A devicesLuca Coelho
The A step of BL devices doesn't support STBC, so we need to disable it. Create new HT params and a macro to use specifically with BL A-step devices where STBC is disabled. Signed-off-by: Luca Coelho <luciano.coelho@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.806dbfa53f40.I4b14a8be30795bdcd73686bd1f9222e7ab93b505@changeid
2022-11-10wifi: iwlwifi: dbg: add support for DBGC4 on BZ family and aboveRotem Saado
Update the for loop that iterates over all the DBGCs to include DBGC4. DBGC4 is only supported staring from the BZ family of devices, so check the family before accepting it. Signed-off-by: Rotem Saado <rotem.saado@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.0b6050b41060.Id662c716c2e3c2c850740d1f71e2801e02aeddaf@changeid
2022-11-10wifi: iwlwifi: mvm: use old checksum for Bz A-stepJohannes Berg
For Bz A-step hardware, the checksum offload is broken and we need to use the old way, which is still there. Do that, which requires taking the checksum capability bits out of the IWL_DEVICE_BZ macro and listing them individually. Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221102165239.6bc379f1b0b1.I204223f1b1c2fe26f414aea6679ef7fce681c33a@changeid
2022-11-10wifi: iwlwifi: mei: fix potential NULL-ptr deref after cloneJohannes Berg
If cloning the SKB fails, don't try to use it, but rather return as if we should pass it. Coverity CID: 1503456 Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME") Signed-off-by: Johannes Berg <johannes.berg@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221030191011.0ce03ba99601.I87960b7cb0a3d16b9fd8d9144027e7e2587f5a58@changeid
2022-11-10wifi: iwlwifi: mei: avoid blocking sap messages handling due to rtnl lockAvraham Stern
The AMT_STATE sap message handler tries to take the rtnl lock. This means that in case the rtnl lock is already taken, sap messages will not be processed. When an interface is brought up, the host requests ownership from csme. However, since the rtnl lock is already held, if there is a pending amt state message, the host will not be able to read the ownership confirm message because the amt state message handler is pending. As a result, the host fails to get ownership although csme granted it. Fix it by moving the part that needs the rtnl lock into a dedicated worker, so handling sap messages can continue. Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME") Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221030191011.8599f2b4e9dd.I518f79e9099bf815c5f8d90235b4ce3250f59970@changeid
2022-11-10pinctrl: devicetree: fix null pointer dereferencing in pinctrl_dt_to_mapZeng Heng
Here is the BUG report by KASAN about null pointer dereference: BUG: KASAN: null-ptr-deref in strcmp+0x2e/0x50 Read of size 1 at addr 0000000000000000 by task python3/2640 Call Trace: strcmp __of_find_property of_find_property pinctrl_dt_to_map kasprintf() would return NULL pointer when kmalloc() fail to allocate. So directly return ENOMEM, if kasprintf() return NULL pointer. Fixes: 57291ce295c0 ("pinctrl: core device tree mapping table parsing support") Signed-off-by: Zeng Heng <zengheng4@huawei.com> Link: https://lore.kernel.org/r/20221110082056.2014898-1-zengheng4@huawei.com Signed-off-by: Linus Walleij <linus.walleij@linaro.org>
2022-11-10wifi: iwlwifi: mei: fix tx DHCP packet for devices with new Tx APIEmmanuel Grumbach
Devices with new Tx API have the IV introduced by the HW and it is not present in the skb at all. Hence we don't need to tell iwl_mvm_mei_tx_copy_to_csme to jump over 8 bytes to get to the ethernet header. Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME") Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221030191011.12dc42133502.Idd744ffeeb84b880eb497963ee02563cbb959a42@changeid
2022-11-10wifi: iwlwifi: mei: don't send SAP commands if AMT is disabledEmmanuel Grumbach
We should not send any SAP command to CSME if AMT is disabled. Reported-by: Toke Høiland-Jørgensen <toke@toke.dk> Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME") Signed-off-by: Emmanuel Grumbach <emmanuel.grumbach@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221030191011.ea222d41c781.Ifc90ddc3e35187683ff7f59371d792b61c8854c8@changeid
2022-11-10wifi: iwlwifi: mei: make sure ownership confirmed message is sentAvraham Stern
It is possible that CSME will try to take ownership while the driver is stopping. In this case, if the CSME takes ownership message arrives after the driver started unregistering, the iwl_mei_cache->ops is already invalid, so the host will not answer with the ownership confirmed message. Similarly, if the take ownership message arrived after the mac was stopped or when iwl_mvm_up() failed, setting rfkill will not trigger sending the confirm message. As a result, CSME will not take ownership, which will result in a disconnection. Fix it by sending the ownership confirmed message immediately in such cases. Fixes: 2da4366f9e2c ("iwlwifi: mei: add the driver to allow cooperation with CSME") Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221030191011.b2a4c009e3e6.I7f931b7ee8b168e8ac88b11f23bff98b7ed3cb19@changeid
2022-11-10wifi: iwlwifi: mvm: send TKIP connection status to csmeAvraham Stern
Notify csme when associated with TKIP cipher. TKIP is supported by csme. Signed-off-by: Avraham Stern <avraham.stern@intel.com> Signed-off-by: Gregory Greenman <gregory.greenman@intel.com> Link: https://lore.kernel.org/r/20221030191011.cea63e4a355c.If6cdfa52529a79b923191c89dad7ed871d2ad2c6@changeid
2022-11-10Merge branch 'macsec-clear-encryption-keys-in-h-w-drivers'Paolo Abeni
Antoine Tenart says: ==================== macsec: clear encryption keys in h/w drivers Commit aaab73f8fba4 ("macsec: clear encryption keys from the stack after setting up offload") made sure to clean encryption keys from the stack after setting up offloading but some h/w drivers did a copy of the key which need to be zeroed as well. The MSCC PHY driver can actually be converted not to copy the encryption key at all, but such patch would be quite difficult to backport. I'll send a following up patch doing this in net-next once this series lands. Tested on the MSCC PHY but not on the atlantic NIC. ==================== Link: https://lore.kernel.org/r/20221108153459.811293-1-atenart@kernel.org Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10net: atlantic: macsec: clear encryption keys from the stackAntoine Tenart
Commit aaab73f8fba4 ("macsec: clear encryption keys from the stack after setting up offload") made sure to clean encryption keys from the stack after setting up offloading, but the atlantic driver made a copy and did not clear it. Fix this. [4 Fixes tags below, all part of the same series, no need to split this] Fixes: 9ff40a751a6f ("net: atlantic: MACSec ingress offload implementation") Fixes: b8f8a0b7b5cb ("net: atlantic: MACSec ingress offload HW bindings") Fixes: 27736563ce32 ("net: atlantic: MACSec egress offload implementation") Fixes: 9d106c6dd81b ("net: atlantic: MACSec egress offload HW bindings") Signed-off-by: Antoine Tenart <atenart@kernel.org> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10net: phy: mscc: macsec: clear encryption keys when freeing a flowAntoine Tenart
Commit aaab73f8fba4 ("macsec: clear encryption keys from the stack after setting up offload") made sure to clean encryption keys from the stack after setting up offloading, but the MSCC PHY driver made a copy, kept it in the flow data and did not clear it when freeing a flow. Fix this. Fixes: 28c5107aa904 ("net: phy: mscc: macsec support") Signed-off-by: Antoine Tenart <atenart@kernel.org> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10gro: avoid checking for a failed searchRichard Gobert
After searching for a protocol handler in dev_gro_receive, checking for failure is redundant. Skip the failure code after finding the corresponding handler. Suggested-by: Eric Dumazet <edumazet@google.com> Signed-off-by: Richard Gobert <richardbgobert@gmail.com> Reviewed-by: Eric Dumazet <edumazet@google.com> Link: https://lore.kernel.org/r/20221108123320.GA59373@debian Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10spi: mediatek: Fix DEVAPC Violation at KO RemoveZhichao Liu
A DEVAPC violation occurs when removing the module due to accessing HW registers without base clock. To fix this bug, the correct method is: 1. Call the runtime resume function to enable the clock; 2. Operate the registers to reset the HW; 3. Turn off the clocks and disable the device RPM mechanism. Signed-off-by: Zhichao Liu <zhichao.liu@mediatek.com> Reviewed-by: AngeloGioacchino Del Regno <angelogioacchino.delregno@collabora.com> Link: https://lore.kernel.org/r/20221110072839.30961-1-zhichao.liu@mediatek.com Signed-off-by: Mark Brown <broonie@kernel.org>
2022-11-10Merge branch 'stmmac-dwmac-loongson-fixes-three-leaks'Paolo Abeni
Yang Yingliang says: ==================== stmmac: dwmac-loongson: fixes three leaks patch #2 fixes missing pci_disable_device() in the error path in probe() patch #1 and pach #3 fix missing pci_disable_msi() and of_node_put() in error and remove() path. ==================== Link: https://lore.kernel.org/r/20221108114647.4144952-1-yangyingliang@huawei.com Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10stmmac: dwmac-loongson: fix missing of_node_put() while module exitingYang Yingliang
The node returned by of_get_child_by_name() with refcount decremented, of_node_put() needs be called when finish using it. So add it in the error path in loongson_dwmac_probe() and in loongson_dwmac_remove(). Fixes: 2ae34111fe4e ("stmmac: dwmac-loongson: fix invalid mdio_node") Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10stmmac: dwmac-loongson: fix missing pci_disable_device() in ↵Yang Yingliang
loongson_dwmac_probe() Add missing pci_disable_device() in the error path in loongson_dwmac_probe(). Fixes: 30bba69d7db4 ("stmmac: pci: Add dwmac support for Loongson") Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10stmmac: dwmac-loongson: fix missing pci_disable_msi() while module exitingYang Yingliang
pci_enable_msi() has been called in loongson_dwmac_probe(), so pci_disable_msi() needs be called in remove path and error path of probe(). Fixes: 30bba69d7db4 ("stmmac: pci: Add dwmac support for Loongson") Signed-off-by: Yang Yingliang <yangyingliang@huawei.com> Signed-off-by: Paolo Abeni <pabeni@redhat.com>
2022-11-10ALSA: usb-audio: add quirk to fix Hamedal C20 disconnect issueAi Chao
For Hamedal C20, the current rate is different from the runtime rate, snd_usb_endpoint stop and close endpoint to resetting rate. if snd_usb_endpoint close the endpoint, sometimes usb will disconnect the device. Signed-off-by: Ai Chao <aichao@kylinos.cn> Link: https://lore.kernel.org/r/20221110063452.295110-1-aichao@kylinos.cn Signed-off-by: Takashi Iwai <tiwai@suse.de>
2022-11-10Merge tag 'amd-drm-fixes-6.1-2022-11-09' of ↵Dave Airlie
https://gitlab.freedesktop.org/agd5f/linux into drm-fixes amd-drm-fixes-6.1-2022-11-09: amdgpu: - SMU 13.0.4 update - GPUVM TLB race fix - DCN 3.1.4 fixes - DCN 3.2.x fixes - Vega10 fan fix - BACO fix for Beige Goby board - PSR fix - GPU VM PT locking fixes amdkfd: - CRIU fixes Signed-off-by: Dave Airlie <airlied@redhat.com> From: Alex Deucher <alexander.deucher@amd.com> Link: https://patchwork.freedesktop.org/patch/msgid/20221109234554.6028-1-alexander.deucher@amd.com
2022-11-10ALSA: hda/realtek: Add Positivo C6300 model quirkEdson Juliano Drosdeck
Positivo Master C6300 (1849:a233) require quirk for anabling headset-mic Signed-off-by: Edson Juliano Drosdeck <edson.drosdeck@gmail.com> Cc: <stable@vger.kernel.org> Link: https://lore.kernel.org/r/20221109171732.5417-1-edson.drosdeck@gmail.com Signed-off-by: Takashi Iwai <tiwai@suse.de>
2022-11-10net: mana: Define data structures for protection domain and memory registrationAjay Sharma
The MANA hardware support protection domain and memory registration for use in RDMA environment. Add those definitions and expose them for use by the RDMA driver. Signed-off-by: Ajay Sharma <sharmaajay@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-12-git-send-email-longli@linuxonhyperv.com Reviewed-by: Dexuan Cui <decui@microsoft.com> Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Define data structures for allocating doorbell page from GDMALong Li
The RDMA device needs to allocate doorbell pages for each user context. Define the GDMA data structures for use by the RDMA driver. Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-11-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Define and process GDMA response code GDMA_STATUS_MORE_ENTRIESAjay Sharma
When doing memory registration, the PF may respond with GDMA_STATUS_MORE_ENTRIES to indicate a follow request is needed. This is not an error and should be processed as expected. Signed-off-by: Ajay Sharma <sharmaajay@microsoft.com> Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-10-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Define max values for SGL entriesLong Li
The number of maximum SGl entries should be computed from the maximum WQE size for the intended queue type and the corresponding OOB data size. This guarantees the hardware queue can successfully queue requests up to the queue depth exposed to the upper layer. Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-9-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Move header files to a common locationLong Li
In preparation to add MANA RDMA driver, move all the required header files to a common location for use by both Ethernet and RDMA drivers. Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-8-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Record port number in netdevLong Li
The port number is useful for user-mode application to identify this net device based on port index. Set to the correct value in ndev. Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-7-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Export Work Queue functions for use by RDMA driverLong Li
RDMA device may need to create Ethernet device queues for use by Queue Pair type RAW. This allows a user-mode context accesses Ethernet hardware queues. Export the supporting functions for use by the RDMA driver. Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-6-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Set the DMA device max segment sizeAjay Sharma
MANA hardware doesn't have any restrictions on the DMA segment size, set it to the max allowed value. Signed-off-by: Ajay Sharma <sharmaajay@microsoft.com> Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-5-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Handle vport sharing between devicesLong Li
For outgoing packets, the PF requires the VF to configure the vport with corresponding protection domain and doorbell ID for the kernel or user context. The vport can't be shared between different contexts. Implement the logic to exclusively take over the vport by either the Ethernet device or RDMA device. Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-4-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Record the physical address for doorbell page regionLong Li
For supporting RDMA device with multiple user contexts with their individual doorbell pages, record the start address of doorbell page region for use by the RDMA driver to allocate user context doorbell IDs. Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-3-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-10net: mana: Add support for auxiliary deviceLong Li
In preparation for supporting MANA RDMA driver, add support for auxiliary device in the Ethernet driver. The RDMA device is modeled as an auxiliary device to the Ethernet device. Reviewed-by: Dexuan Cui <decui@microsoft.com> Signed-off-by: Long Li <longli@microsoft.com> Link: https://lore.kernel.org/r/1667502990-2559-2-git-send-email-longli@linuxonhyperv.com Acked-by: Haiyang Zhang <haiyangz@microsoft.com> Signed-off-by: Leon Romanovsky <leonro@nvidia.com>
2022-11-09Merge branch 'fix panic bringing up veth with xdp progs'Martin KaFai Lau
John Fastabend says: ==================== Not sure if folks want to take this through BPF tree or networking tree. I took a quick look and didn't see any pending fixes so seems no one has noticed the panic yet. It reproducible and easy to repro. I put bpf in the title thinking it woudl be great to run through the BPF selftests given its XDP triggering the panic. Sorry maintainers resent with CC'ing actual lists. Had a scripting issue. Also dropped henqqi has they are bouncing. Thanks! ==================== Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
2022-11-09bpf: veth driver panics when xdp prog attached before veth_openJohn Fastabend
The following panic is observed when bringing up (veth_open) a veth device that has an XDP program attached. [ 61.519185] kernel BUG at net/core/dev.c:6442! [ 61.519456] invalid opcode: 0000 [#1] PREEMPT SMP PTI [ 61.519752] CPU: 0 PID: 408 Comm: ip Tainted: G W 6.1.0-rc2-185930-gd9095f92950b-dirty #26 [ 61.520288] Hardware name: QEMU Standard PC (i440FX + PIIX, 1996), BIOS 1.15.0-1 04/01/2014 [ 61.520806] RIP: 0010:napi_enable+0x3d/0x40 [ 61.521077] Code: f6 f6 80 61 08 00 00 02 74 0d 48 83 bf 88 01 00 00 00 74 03 80 cd 01 48 89 d0 f0 48 0f b1 4f 10 48 39 c2 75 c8 c3 cc cc cc cc <0f> 0b 90 48 8b 87 b0 00 00 00 48 81 c7 b0 00 00 00 45 31 c0 48 39 [ 61.522226] RSP: 0018:ffffbc9800cc36f8 EFLAGS: 00010246 [ 61.522557] RAX: 0000000000000001 RBX: 0000000000000300 RCX: 0000000000000001 [ 61.523004] RDX: 0000000000000010 RSI: ffffffff8b0de852 RDI: ffff9f03848e5000 [ 61.523452] RBP: 0000000000000000 R08: 0000000000000800 R09: 0000000000000000 [ 61.523899] R10: ffff9f0384a96800 R11: ffffffffffa48061 R12: ffff9f03849c3000 [ 61.524345] R13: 0000000000000300 R14: ffff9f03848e5000 R15: 0000001000000100 [ 61.524792] FS: 00007f58cb64d2c0(0000) GS:ffff9f03bbc00000(0000) knlGS:0000000000000000 [ 61.525301] CS: 0010 DS: 0000 ES: 0000 CR0: 0000000080050033 [ 61.525673] CR2: 00007f6cc629b498 CR3: 000000010498c000 CR4: 00000000000006f0 [ 61.526121] Call Trace: [ 61.526284] <TASK> [ 61.526425] __veth_napi_enable_range+0xd6/0x230 [ 61.526723] veth_enable_xdp+0xd0/0x160 [ 61.526969] veth_open+0x2e/0xc0 [ 61.527180] __dev_open+0xe2/0x1b0 [ 61.527405] __dev_change_flags+0x1a1/0x210 [ 61.527673] dev_change_flags+0x1c/0x60 This happens because we are calling veth_napi_enable() on already enabled queues. The root cause is in commit 2e0de6366ac16 changed the control logic dropping this case, if (priv->_xdp_prog) { err = veth_enable_xdp(dev); if (err) return err; - } else if (veth_gro_requested(dev)) { + /* refer to the logic in veth_xdp_set() */ + if (!rtnl_dereference(peer_rq->napi)) { + err = veth_napi_enable(peer); + if (err) + return err; + } so that now veth_napi_enable is called if the peer has not yet initialiazed its peer_rq->napi. The issue is this will happen even if the NIC is not up. Then in veth_enable_xdp just above we have similar path, veth_enable_xdp napi_already_on = (dev->flags & IFF_UP) && rcu_access_pointer(rq->napi) err = veth_enable_xdp_range(dev, 0, dev->real_num_rx_queues, napi_already_on); The trouble is an xdp prog is assigned before bringing the device up each of the veth_open path will enable the peers xdp napi structs. But then when we bring the peer up it will similar try to enable again because from veth_open the IFF_UP flag is not set until after the op in __dev_open so we believe napi_alread_on = false. To fix this just drop the IFF_UP test and rely on checking if the napi struct is enabled. This also matches the peer check in veth_xdp for disabling. To reproduce run ./test_xdp_meta.sh I found adding Cilium/Tetragon tests for XDP. Fixes: 2e0de6366ac16 ("veth: Avoid drop packets when xdp_redirect performs") Signed-off-by: John Fastabend <john.fastabend@gmail.com> Link: https://lore.kernel.org/r/20221108221650.808950-2-john.fastabend@gmail.com Acked-by: Jakub Kicinski <kuba@kernel.org> Signed-off-by: Martin KaFai Lau <martin.lau@kernel.org>
2022-11-09selftests: Fix test group SKIPPED resultDomenico Cerasuolo
When showing the result of a test group, if one of the subtests was skipped, while still having passing subtests, the group result was marked as SKIP. E.g.: 223/1 usdt/basic:SKIP 223/2 usdt/multispec:OK 223/3 usdt/urand_auto_attach:OK 223/4 usdt/urand_pid_attach:OK 223 usdt:SKIP The test result of usdt in the example above should be OK instead of SKIP, because the test group did have passing tests and it would be considered in "normal" state. With this change, only if all of the subtests were skipped, the group test is marked as SKIP. When only some of the subtests are skipped, a more detailed result is given, stating how many of the subtests were skipped. E.g: 223/1 usdt/basic:SKIP 223/2 usdt/multispec:OK 223/3 usdt/urand_auto_attach:OK 223/4 usdt/urand_pid_attach:OK 223 usdt:OK (SKIP: 1/4) Signed-off-by: Domenico Cerasuolo <dceras@meta.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Acked-by: Yonghong Song <yhs@fb.com> Link: https://lore.kernel.org/bpf/20221109184039.3514033-1-cerasuolodomenico@gmail.com
2022-11-09Merge branch 'libbpf: Resolve unambigous forward declarations'Andrii Nakryiko
Eduard Zingerman says: ==================== The patch-set is consists of the following parts: - An update for libbpf's hashmap interface from void* -> void* to a polymorphic one, allowing to use both long and void* keys and values w/o additional casts. Interface functions are hidden behind auxiliary macro that add casts as necessary. This simplifies many use cases in libbpf as hashmaps there are mostly integer to integer and required awkward looking incantations like `(void *)(long)off` previously. Also includes updates for perf as it copies map implementation from libbpf. - A change to `lib/bpf/btf.c:btf__dedup` that adds a new pass named "Resolve unambiguous forward declaration". This pass builds a hashmap `name_off -> uniquely named struct or union` and uses it to replace FWD types by structs or unions. This is necessary for corner cases when FWD is not used as a part of some struct or union definition de-duplicated by `btf_dedup_struct_types`. The goal of the patch-set is to resolve forward declarations that don't take part in type graphs comparisons if declaration name is unambiguous. Example: CU #1: struct foo; // standalone forward declaration struct foo *some_global; CU #2: struct foo { int x; }; struct foo *another_global; Currently the de-duplicated BTF for this example looks as follows: [1] STRUCT 'foo' size=4 vlen=1 ... [2] INT 'int' size=4 ... [3] PTR '(anon)' type_id=1 [4] FWD 'foo' fwd_kind=struct [5] PTR '(anon)' type_id=4 The goal of this patch-set is to simplify it as follows: [1] STRUCT 'foo' size=4 vlen=1 'x' type_id=2 bits_offset=0 [2] INT 'int' size=4 bits_offset=0 nr_bits=32 encoding=SIGNED [3] PTR '(anon)' type_id=1 For defconfig kernel with BTF enabled this removes 63 forward declarations. For allmodconfig kernel with BTF enabled this removes ~5K out of ~21K forward declarations in ko objects. This unlocks some additional de-duplication in ko objects, but impact is tiny: ~13K less BTF ids out of ~2M. Changelog: v3 -> v4 Changes suggested by Andrii: - hashmap interface rework to allow use of integer and pointer keys an values w/o casts as suggested in [1]. v2 -> v3 Changes suggested by Andrii: - perf's util/hashtable.{c,h} are synchronized with libbpf implementation, perf's source code updated accordingly; - changes to libbpf, bpf selftests and perf are combined in a single patch to simplify bisecting; - hashtable interface updated to be long -> long instead of uintptr_t -> uintptr_t; - btf_dedup_resolve_fwds updated to correctly use IS_ERR / PTR_ERR macro; - test cases for btf_dedup_resolve_fwds are updated for better clarity. v1 -> v2 - Style fixes in btf_dedup_resolve_fwd and btf_dedup_resolve_fwds as suggested by Alan. [v1] https://lore.kernel.org/bpf/20221102110905.2433622-1-eddyz87@gmail.com/ [v2] https://lore.kernel.org/bpf/20221103033430.2611623-1-eddyz87@gmail.com/ [v3] https://lore.kernel.org/bpf/20221106202910.4193104-1-eddyz87@gmail.com/ [1] https://lore.kernel.org/bpf/CAEf4BzZ8KFneEJxFAaNCCFPGqp20hSpS2aCj76uRk3-qZUH5xg@mail.gmail.com/ ==================== Signed-off-by: Andrii Nakryiko <andrii@kernel.org>
2022-11-09selftests/bpf: Tests for btf_dedup_resolve_fwdsEduard Zingerman
Tests to verify the following behavior of `btf_dedup_resolve_fwds`: - remapping for struct forward declarations; - remapping for union forward declarations; - no remapping if forward declaration kind does not match similarly named struct or union declaration; - no remapping if forward declaration name is ambiguous; - base ids are considered for fwd resolution in split btf scenario. Signed-off-by: Eduard Zingerman <eddyz87@gmail.com> Signed-off-by: Andrii Nakryiko <andrii@kernel.org> Reviewed-by: Alan Maguire <alan.maguire@oracle.com> Link: https://lore.kernel.org/bpf/20221109142611.879983-4-eddyz87@gmail.com
generated by cgit (git 2.34.1) at 2025-09-17 04:58:52 +0000