- 24 Jun, 2020 17 commits
-
-
David S. Miller authored
Jisheng Zhang says: ==================== net: phy: call phy_disable_interrupts() in phy_init_hw() We face an issue with rtl8211f, a pin is shared between INTB and PMEB, and the PHY Register Accessible Interrupt is enabled by default, so the INTB/PMEB pin is always active in polling mode case. As Heiner pointed out "I was thinking about calling phy_disable_interrupts() in phy_init_hw(), to have a defined init state as we don't know in which state the PHY is if the PHY driver is loaded. We shouldn't assume that it's the chip power-on defaults, BIOS or boot loader could have changed this. Or in case of dual-boot systems the other OS could leave the PHY in whatever state." patch1 makes phy_disable_interrupts() non-static so that it could be used in phy_init_hw() to have a defined init state. patch2 calls phy_disable_interrupts() in phy_init_hw() to have a defined init state. Since v3: - call phy_disable_interrupts() have interrupts disabled first then config_init, thank Florian Since v2: - Don't export phy_disable_interrupts() but just make it non-static Since v1: - EXPORT the correct symbol ==================== Signed-off-by:David S. Miller <davem@davemloft.net>
-
Jisheng Zhang authored
Call phy_disable_interrupts() in phy_init_hw() to "have a defined init state as we don't know in which state the PHY is if the PHY driver is loaded. We shouldn't assume that it's the chip power-on defaults, BIOS or boot loader could have changed this. Or in case of dual-boot systems the other OS could leave the PHY in whatever state." as pointed out by Heiner. Suggested-by:
Heiner Kallweit <hkallweit1@gmail.com> Signed-off-by:
Jisheng Zhang <Jisheng.Zhang@synaptics.com> Signed-off-by:
-
Jisheng Zhang authored
We face an issue with rtl8211f, a pin is shared between INTB and PMEB, and the PHY Register Accessible Interrupt is enabled by default, so the INTB/PMEB pin is always active in polling mode case. As Heiner pointed out "I was thinking about calling phy_disable_interrupts() in phy_init_hw(), to have a defined init state as we don't know in which state the PHY is if the PHY driver is loaded. We shouldn't assume that it's the chip power-on defaults, BIOS or boot loader could have changed this. Or in case of dual-boot systems the other OS could leave the PHY in whatever state." Make phy_disable_interrupts() non-static so that it could be used in phy_init_hw() to have a defined init state. Suggested-by:
-
Sascha Hauer authored
When writing the serdes configuration register was moved to mvneta_config_interface() the whole code block was removed from mvneta_port_power_up() in the assumption that its only purpose was to write the serdes configuration register. As mentioned by Russell King its purpose was also to check for valid interface modes early so that later in the driver we do not have to care for unexpected interface modes. Add back the test to let the driver bail out early on unhandled interface modes. Fixes: b4748553 ("net: ethernet: mvneta: Fix Serdes configuration for SoCs without comphy") Signed-off-by:
Sascha Hauer <s.hauer@pengutronix.de> Reviewed-by:
Russell King <rmk+kernel@armlinux.org.uk> Signed-off-by:
-
Sascha Hauer authored
In mvneta_config_interface() the RGMII modes are catched by the default case which is an error return. The RGMII modes are valid modes for the driver, so instead of returning an error add a break statement to return successfully. This avoids this warning for non comphy SoCs which use RGMII, like SolidRun Clearfog: WARNING: CPU: 0 PID: 268 at drivers/net/ethernet/marvell/mvneta.c:3512 mvneta_start_dev+0x220/0x23c Fixes: b4748553 ("net: ethernet: mvneta: Fix Serdes configuration for SoCs without comphy") Signed-off-by:
-
Daniel Mack authored
The driver for Marvell switches puts all ports in IGMP snooping mode which results in all IGMP/MLD frames that ingress on the ports to be forwarded to the CPU only. The bridge code in the kernel can then interpret these frames and act upon them, for instance by updating the mdb in the switch to reflect multicast memberships of stations connected to the ports. However, the IGMP/MLD frames must then also be forwarded to other ports of the bridge so external IGMP queriers can track membership reports, and external multicast clients can receive query reports from foreign IGMP queriers. Currently, this is impossible as the EDSA tagger sets offload_fwd_mark on the skb when it unwraps the tagged frames, and that will make the switchdev layer prevent the skb from egressing on any other port of the same switch. To fix that, look at the To_CPU code in the DSA header and make forwarding of the frame possible for trapped IGMP packets. Introduce some #defines for the frame types to make the code a bit more comprehensive. This was tested on a Marvell 88E6352 variant. Signed-off-by:
Daniel Mack <daniel@zonque.org> Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Tested-by:
-
Lorenzo Bianconi authored
ovs connection tracking module performs de-fragmentation on incoming fragmented traffic. Take info account if traffic has been de-fragmented in execute_check_pkt_len action otherwise we will perform the wrong nested action considering the original packet size. This issue typically occurs if ovs-vswitchd adds a rule in the pipeline that requires connection tracking (e.g. OVN stateful ACLs) before execute_check_pkt_len action. Moreover take into account GSO fragment size for GSO packet in execute_check_pkt_len routine Fixes: 4d5ec89f ("net: openvswitch: Add a new action check_pkt_len") Signed-off-by:
Lorenzo Bianconi <lorenzo@kernel.org> Signed-off-by:
-
David S. Miller authored
Russell King says: ==================== Two phylink pause fixes While testing, I discovered two issues with ethtool -A with phylink. First, if there is a PHY bound to the network device, we hit a deadlock when phylib tries to notify us of the link changing as a result of triggering a renegotiation. Second, when we are manually forcing the pause settings, and there is no renegotiation triggered, we do not update the MAC via the new mac_link_up approach. These two patches solve both problems, and will need to be backported to v5.7; they do not apply cleanly there due to the introduction of PCS in the v5.8 merge window. ==================== Signed-off-by: -
Russell King authored
We have been relying on link events and mac_config() when the manual pause modes are changed. With recent developments, such as moving the programming of link state to mac_link_up(), this no longer works. To ensure that we update the MAC, we must generate a link-down followed by a link-up event; we can do that by setting mac_link_dropped and triggering a resolve. Fixes: 91a208f2 ("net: phylink: propagate resolved link config via mac_link_up()") Signed-off-by:
-
Russell King authored
Fix a phylink's ethtool set_pauseparam support deadlock caused by phylib interacting with phylink: we must not hold the state lock while calling phylib functions that may call into phylink_phy_change(). Fixes: f904f15e ("net: phylink: allow ethtool -A to change flow control advertisement") Signed-off-by:
-
Tariq Toukan authored
Clearing the sock TX queue in sk_set_socket() might cause unexpected out-of-order transmit when called from sock_orphan(), as outstanding packets can pick a different TX queue and bypass the ones already queued. This is undesired in general. More specifically, it breaks the in-order scheduling property guarantee for device-offloaded TLS sockets. Remove the call to sk_tx_queue_clear() in sk_set_socket(), and add it explicitly only where needed. Fixes: e022f0b4 ("net: Introduce sk_tx_queue_mapping") Signed-off-by:
Tariq Toukan <tariqt@mellanox.com> Reviewed-by:
Boris Pismenny <borisp@mellanox.com> Signed-off-by:
-
Andrew Lunn authored
A last minute change put the TDR cable test parameters into a nest. The validation is not sufficient, resulting in an oops if the nest is missing. Set default values first, then update them if the nest is provided. Fixes: f2bc8ad3 ("net: ethtool: Allow PHY cable test TDR data to configured") Signed-off-by:
-
David S. Miller authored
Michael Chan says: ==================== bnxt_en: Bug fixes. The first patch stores the firmware version code which is needed by the next 2 patches to determine some worarounds based on the firmware version. The workarounds are to disable legacy TX push mode and to clear the hardware statistics during ifdown. The last patch checks that it is a PF before reading the VPD. Please also queue these for -stable. Thanks. ==================== Reviewed-by:
Jakub Kicinski <kuba@kernel.org> Signed-off-by:
-
Vasundhara Volam authored
Virtual functions does not have VPD information. This patch modifies calling bnxt_read_vpd_info() only for PFs and avoids an unnecessary error log. Fixes: a0d0fd70 ("bnxt_en: Read partno and serialno of the board from VPD") Signed-off-by:
Vasundhara Volam <vasundhara-v.volam@broadcom.com> Signed-off-by:
Michael Chan <michael.chan@broadcom.com> Signed-off-by:
-
Michael Chan authored
On older firmware, the hardware statistics are not cleared when the driver frees the hardware stats contexts during ifdown. The driver expects these stats to be cleared and saves a copy before freeing the stats contexts. During the next ifup, the driver will likely allocate the same hardware stats contexts and this will cause a big increase in the counters as the old counters are added back to the saved counters. We fix it by making an additional firmware call to clear the counters before freeing the hw stats contexts when the firmware is the older 20.x firmware. Fixes: b8875ca3 ("bnxt_en: Save ring statistics before reset.") Reported-by:
Jakub Kicinski <kicinski@fb.com> Reviewed-by:
-
Michael Chan authored
Older firmware may not support legacy TX push properly and may not be disabling it. So we check certain firmware versions that may have this problem and disable legacy TX push unconditionally. Fixes: c0c050c5 ("bnxt_en: New Broadcom ethernet driver.") Reviewed-by:
Edwin Peer <edwin.peer@broadcom.com> Signed-off-by:
-
Michael Chan authored
We currently only store the firmware version as a string for ethtool and devlink info. Store it also as a version code. The next 2 patches will need to check the firmware major version to determine some workarounds. We also use the 16-bit firmware version fields if the firmware is newer and provides the 16-bit fields. Reviewed-by:
-
- 23 Jun, 2020 23 commits
-
-
David S. Miller authored
Rahul Lakkireddy says: ==================== cxgb4/cxgb4vf: fix warnings reported by sparse This series of patches fix various warnings reported by the sparse tool. Patches 1 and 2 fix lock context imbalance warnings. Patch 3 fixes cast to restricted __be64 warning when fetching timestamp in PTP path. Patch 4 fixes several cast to restricted __be32 warnings in TC-U32 offload parser. Patch 5 fixes several cast from restricted __be16 warnings in parsing L4 ports for filters. Patch 6 fixes several restricted __be32 degrades to integer warnings when comparing IP address masks for exact-match filters. Patch 7 fixes cast to restricted __be64 warning when fetching SGE queue contexts in device dump collection. Patch 8 fixes cast from restricted __sum16 warning when saving IPv4 partial checksum. Patch 9 fixes issue with string array scope in DCB path. Patch 10 fixes a set but unused variable warning when DCB is disabled. Patch 11 fixes several kernel-doc comment warnings in cxgb4 driver. Patch 12 fixes several kernel-doc comment warnings in cxgb4vf driver. ==================== Signed-off-by: -
Rahul Lakkireddy authored
Update several kernel-doc line comments to fix warnings reported by make W=1. Fixes following class of warnings reported by make W=1 in several places: cxgb4vf_main.c:275: warning: Function parameter or member 'persistent' not described in 'cxgb4vf_change_mac' cxgb4vf_main.c:275: warning: Excess function parameter 'persist' description in 'cxgb4vf_change_mac' Fixes: 16f8bd4b ("cxgb4vf: Add core T4 PCI-E SR-IOV Virtual Function hardware definitions and device communication code") Fixes: c6e0d914 ("cxgb4vf: Add T4 Virtual Function Scatter-Gather Engine DMA code") Fixes: e0a8b34a ("cxgb4vf: Add and initialize some sge params for VF driver") Fixes: c3168cab ("cxgb4/cxgbvf: Handle 32-bit fw port capabilities") Fixes: 0e23daeb ("drivers/net: chelsio/cxgb*: Convert timers to use timer_setup()") Fixes: 3f8cfd0d ("cxgb4/cxgb4vf: Program hash region for {t4/t4vf}_change_mac()") Signed-off-by:
Rahul Lakkireddy <rahul.lakkireddy@chelsio.com> Signed-off-by:
-
Rahul Lakkireddy authored
Update several kernel-doc line comments to fix warnings reported by make W=1. Fixes following class of warnings reported by make W=1 in several places: l2t.c:616: warning: Cannot understand * @dev: net_device pointer t4_hw.c:3175: warning: Function parameter or member 'adap' not described in 't4_get_exprom_version' t4_hw.c:3175: warning: Excess function parameter 'adapter' description in 't4_get_exprom_version' Fixes: 56d36be4 ("cxgb4: Add HW and FW support code") Fixes: fd3a4790 ("cxgb4: Add packet queues and packet DMA code") Fixes: 26f7cbc0 ("cxgb4: Don't attempt to upgrade T4 firmware when cxgb4 will end up as a slave") Fixes: 793dad94 ("RDMA/cxgb4: Fix bug for active and passive LE hash collision path") Fixes: ba3f8cd5 ("cxgb4: Add support in cxgb4 to get expansion rom version via ethtool") Fixes: f7502659 ("cxgb4: Add API to alloc l2t entry; also update existing ones") Fixes: ddc7740d ("cxgb4: Decode link down reason code obtained from firmware") Fixes: 193c4c28 ("cxgb4: Update T6 Buffer Group and Channel Mappings") Fixes: 8f46d467 ("cxgb4: Use Firmware params to get buffer-group map") Fixes: a4569504 ("cxgb4: time stamping interface for PTP") Fixes: 9c33e420 ("cxgb4: Add PTP Hardware Clock (PHC) support") Fixes: c3168cab ("cxgb4/cxgbvf: Handle 32-bit fw port capabilities") Fixes: 5ccf9d04 ("cxgb4: update API for TP indirect register access") Fixes: 3bdb376e ("cxgb4: introduce SMT ops to prepare for SMAC rewrite support") Fixes: 736c3b94 ("cxgb4: collect egress and ingress SGE queue contexts") Fixes: f56ec676 ("cxgb4: Add support for ethtool i2c dump") Fixes: 9d5fd927 ("cxgb4/cxgb4vf: add support for ndo_set_vf_vlan") Fixes: 98f3697f ("cxgb4: add tc flower match support for tunnel VNI") Fixes: 02d805dc ("cxgb4: use new fw interface to get the VIN and smt index") Fixes: 3f8cfd0d ("cxgb4/cxgb4vf: Program hash region for {t4/t4vf}_change_mac()") Fixes: d429005f ("cxgb4/cxgb4vf: Add support for SGE doorbell queue timer") Fixes: 0e395b3c ("cxgb4: add FLOWC based QoS offload") Signed-off-by:
-
Rahul Lakkireddy authored
Remove the set but unused variable when DCB is disabled. Instead, do the calculation directly inline. Fixes following warning in make W=1: cxgb4_main.c: In function 'cfg_queues': cxgb4_main.c:5380:29: warning: variable 'n1g' set but not used [-Wunused-but-set-variable] u32 i, n10g = 0, qidx = 0, n1g = 0; ^ Fixes: 116ca924 ("cxgb4: fix checks for max queues to allocate") Signed-off-by: -
Rahul Lakkireddy authored
Move the DCB version string array extern to header file. Fixes following sparse warning: cxgb4_dcb.c:13:12: warning: symbol 'dcb_ver_array' was not declared. Should it be static? Fixes: ebddd97a ("cxgb4: add support to display DCB info") Signed-off-by:
-
Rahul Lakkireddy authored
The checksum field in IPv4 header is in __sum16 and ip_fast_csum() also returns __sum16. So, no need to cast it to u16. Fixes following sparse warning: sge.c:1539:47: warning: cast from restricted __sum16 sge.c:1539:44: warning: incorrect type in assignment (different base types) sge.c:1539:44: expected restricted __sum16 [usertype] check sge.c:1539:44: got unsigned short [usertype] Fixes: d0a1299c ("cxgb4: add support for vxlan segmentation offload") Signed-off-by:
-
Rahul Lakkireddy authored
The data in destination buffer is expected to be be parsed in big endian. So, use the right context. Fixes following sparse warning: cudbg_lib.c:2041:44: warning: incorrect type in assignment (different base types) cudbg_lib.c:2041:44: expected unsigned long long [usertype] cudbg_lib.c:2041:44: got restricted __be64 [usertype] Fixes: 736c3b94 ("cxgb4: collect egress and ingress SGE queue contexts") Signed-off-by:
-
Rahul Lakkireddy authored
Use correct type to check for all-mask exact match IP addresses. Fixes following sparse warnings due to big endian value checks against 0xffffffff in is_addr_all_mask(): cxgb4_filter.c:977:25: warning: restricted __be32 degrades to integer cxgb4_filter.c:983:37: warning: restricted __be32 degrades to integer cxgb4_filter.c:984:37: warning: restricted __be32 degrades to integer cxgb4_filter.c:985:37: warning: restricted __be32 degrades to integer cxgb4_filter.c:986:37: warning: restricted __be32 degrades to integer Fixes: 3eb8b62d ("cxgb4: add support to create hash-filters via tc-flower offload") Signed-off-by:
-
Rahul Lakkireddy authored
The source and destination L4 ports in filter offload need to be in CPU endian. They will finally be converted to Big Endian after all operations are done and before giving them to hardware. The L4 ports for NAT are expected to be passed as a byte stream TCB. So, treat them as such. Fixes following sparse warnings in several places: cxgb4_tc_flower.c:159:33: warning: cast from restricted __be16 cxgb4_tc_flower.c:159:33: warning: incorrect type in argument 1 (different base types) cxgb4_tc_flower.c:159:33: expected unsigned short [usertype] val cxgb4_tc_flower.c:159:33: got restricted __be16 [usertype] dst Fixes: dca4faeb ("cxgb4: Add LE hash collision bug fix path in LLD driver") Fixes: 62488e4b ("cxgb4: add basic tc flower offload support") Fixes: 557ccbf9 ("cxgb4: add tc flower support for L3/L4 rewrite") Signed-off-by:
-
Rahul Lakkireddy authored
TC-U32 passes all keys values and masks in __be32 format. The parser already expects this and hence pass the value and masks in __be32 natively to the parser. Fixes following sparse warnings in several places: cxgb4_tc_u32.c:57:21: warning: incorrect type in assignment (different base types) cxgb4_tc_u32.c:57:21: expected unsigned int [usertype] val cxgb4_tc_u32.c:57:21: got restricted __be32 [usertype] val cxgb4_tc_u32_parse.h:48:24: warning: cast to restricted __be32 Fixes: 2e8aad7b ("cxgb4: add parser to translate u32 filters to internal spec") Signed-off-by:
-
Rahul Lakkireddy authored
Use get_unaligned_be64() to fetch the timestamp needed for ns_to_ktime() conversion. Fixes following sparse warning: sge.c:3282:43: warning: cast to restricted __be64 Fixes: a4569504 ("cxgb4: time stamping interface for PTP") Signed-off-by:
-
Rahul Lakkireddy authored
Check for whether PTP is enabled or not at the caller and perform locking/unlocking at the caller. Fixes following sparse warning: sge.c:1641:26: warning: context imbalance in 'cxgb4_eth_xmit' - different lock contexts for basic block Fixes: a4569504 ("cxgb4: time stamping interface for PTP") Signed-off-by:
-
Rahul Lakkireddy authored
Move code handling L2T ARP failures to the only caller. Fixes following sparse warning: skbuff.h:2091:29: warning: context imbalance in 'handle_failed_resolution' - unexpected unlock Fixes: 749cb5fe ("cxgb4: Replace arpq_head/arpq_tail with SKB double link-list code") Signed-off-by:
-
David S. Miller authored
Alexander Lobakin says: ==================== net: qed/qede: various stability fixes This set addresses several near-critical issues that were observed and reproduced on different test and production configurations. v2: - don't split the "Fixes:" tag across several lines in patch 9; - no functional changes. ==================== Signed-off-by: -
Alexander Lobakin authored
Variable 'abs_ppfid' in qed_dev.c:qed_llh_add_mac_filter() always gets printed, but is initialized only under 'ref_cnt == 1' condition. This results in: In file included from ./include/linux/kernel.h:15:0, from ./include/asm-generic/bug.h:19, from ./arch/x86/include/asm/bug.h:86, from ./include/linux/bug.h:5, from ./include/linux/io.h:11, from drivers/net/ethernet/qlogic/qed/qed_dev.c:35: drivers/net/ethernet/qlogic/qed/qed_dev.c: In function 'qed_llh_add_mac_filter': ./include/linux/printk.h:358:2: warning: 'abs_ppfid' may be used uninitialized in this function [-Wmaybe-uninitialized] printk(KERN_NOTICE pr_fmt(fmt), ##__VA_ARGS__) ^~~~~~ drivers/net/ethernet/qlogic/qed/qed_dev.c:983:17: note: 'abs_ppfid' was declared here u8 filter_idx, abs_ppfid; ^~~~~~~~~ ...under W=1+. Fix this by initializing it with zero. Fixes: 79284ade ("qed: Add llh ppfid interface and 100g support for offload protocols") Signed-off-by:Alexander Lobakin <alobakin@marvell.com> Signed-off-by:
Igor Russkikh <irusskikh@marvell.com> Signed-off-by:
Michal Kalderon <michal.kalderon@marvell.com> Signed-off-by:
-
Alexander Lobakin authored
Sizes of all ILT blocks must be reset before ILT recomputing when disabling clients, or memory allocation may exceed ILT shadow array and provoke system crashes. Fixes: 1408cc1f ("qed: Introduce VFs") Signed-off-by:
-
Alexander Lobakin authored
Set edev->cdev pointer to NULL after calling remove() callback to avoid using of already freed object. Fixes: ccc67ef5 ("qede: Error recovery process") Signed-off-by:
-
Alexander Lobakin authored
Currently PTP cyclecounter and timecounter are initialized only on the first probing and are cleaned up during removal. This means that PTP becomes non-functional after device recovery. Fix this by unconditional PTP initialization on probing and clearing Tx pending bit on exiting. Fixes: ccc67ef5 ("qede: Error recovery process") Signed-off-by:
-
Alexander Lobakin authored
This is likely a copy'n'paste mistake. The amount of ILT lines to reserve for a single VF was being multiplied by the total VFs count. This led to a huge redundancy in reservation and potential lines drainouts. Fixes: 1408cc1f ("qed: Introduce VFs") Signed-off-by:
-
Alexander Lobakin authored
25ms sleep cycles in waiting for PF response are excessive and may lead to different timeout failures. Start to wait with short udelays, and in most cases polling will end here. If the time was not sufficient, switch to msleeps. usleep_range() may go far beyond 100us depending on platform and tick configuration, hence atomic udelays for consistency. Also add explicit DMA barriers since 'done' always comes from a shared request-response DMA pool, and note that in the comment nearby. Fixes: 1408cc1f ("qed: Introduce VFs") Signed-off-by:
-
Alexander Lobakin authored
Set rdma_wq pointer to NULL after destroying the workqueue and check for it when adding new events to fix crashes on driver unload. Fixes: cee9fbd8 ("qede: Add qedr framework") Signed-off-by:
-
Alexander Lobakin authored
qed_spq_unregister_async_cb() should be called before qed_rdma_info_free() to avoid crash-spawning uses-after-free. Instead of calling it from each subsystem exit code, do it in one place on PF down. Fixes: 291d57f6 ("qed: Fix rdma_info structure allocation") Signed-off-by:
-
Alexander Lobakin authored
qed_chain_get_element_left{,_u32} returned 0 when the difference between producer and consumer page count was equal to the total page count. Fix this by conditional expanding of producer value (vs unconditional). This allowed to eliminate normalizaton against total page count, which was the cause of this bug. Misc: replace open-coded constants with common defines. Fixes: a91eb52a ("qed: Revisit chain implementation") Signed-off-by:
-
