[5.13,058/151] net: mvvp2: fix short frame size on s390

Message ID	20210816125445.979349376@linuxfoundation.org
State	Superseded
Headers	show Return-Path: <stable-owner@kernel.org> From: Greg Kroah-Hartman <gregkh@linuxfoundation.org> To: linux-kernel@vger.kernel.org Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>, stable@vger.kernel.org, Sven Auhagen <sven.auhagen@voleatech.de>, Matteo Croce <mcroce@microsoft.com>, "David S. Miller" <davem@davemloft.net>, John Hubbard <jhubbard@nvidia.com>, Sasha Levin <sashal@kernel.org> Subject: [PATCH 5.13 058/151] net: mvvp2: fix short frame size on s390 Date: Mon, 16 Aug 2021 15:01:28 +0200 Message-Id: <20210816125445.979349376@linuxfoundation.org> In-Reply-To: <20210816125444.082226187@linuxfoundation.org> References: <20210816125444.082226187@linuxfoundation.org> User-Agent: quilt/0.66 MIME-Version: 1.0 Content-Type: text/plain; charset=UTF-8 Content-Transfer-Encoding: 8bit Precedence: bulk
Series	None \| expand [5.13,002/151] Revert "usb: dwc3: gadget: Use list_replace_init() before traversing lists" [5.13,003/151] iio: adc: ti-ads7950: Ensure CS is deasserted after reading channels [5.13,004/151] iio: adis: set GPIO reset pin direction [5.13,005/151] iio: humidity: hdc100x: Add margin to the conversion time [5.13,006/151] iio: adc: Fix incorrect exit of for-loop [5.13,007/151] ASoC: amd: Fix reference to PCM buffer address [5.13,008/151] ASoC: xilinx: Fix reference to PCM buffer address [5.13,009/151] ASoC: uniphier: Fix reference to PCM buffer address [5.13,010/151] ASoC: tlv320aic31xx: Fix jack detection after suspend [5.13,011/151] ASoC: kirkwood: Fix reference to PCM buffer address [5.13,012/151] ASoC: intel: atom: Fix reference to PCM buffer address [5.13,013/151] i2c: dev: zero out array used for i2c reads from userspace [5.13,014/151] cifs: Handle race conditions during rename [5.13,015/151] cifs: create sd context must be a multiple of 8 [5.13,016/151] cifs: Call close synchronously during unlink/rename/lease break. [5.13,017/151] cifs: use the correct max-length for dentry_path_raw() [5.13,018/151] io_uring: drop ctx->uring_lock before flushing work item [5.13,019/151] io_uring: fix ctx-exit io_rsrc_put_work() deadlock [5.13,020/151] scsi: lpfc: Move initialization of phba->poll_list earlier to avoid crash [5.13,021/151] cgroup: rstat: fix A-A deadlock on 32bit around u64_stats_sync [5.13,022/151] seccomp: Fix setting loaded filter count during TSYNC [5.13,023/151] net: wwan: mhi_wwan_ctrl: Fix possible deadlock [5.13,024/151] net: ethernet: ti: cpsw: fix min eth packet size for non-switch use-cases [5.13,025/151] ARC: fp: set FPU_STATUS.FWE to enable FPU_STATUS update on context switch [5.13,026/151] ceph: reduce contention in ceph_check_delayed_caps() [5.13,027/151] pinctrl: k210: Fix k210_fpioa_probe() [5.13,028/151] ACPI: NFIT: Fix support for virtual SPA ranges [5.13,029/151] libnvdimm/region: Fix label activation vs errors [5.13,030/151] riscv: kexec: do not add -mno-relax flag if compiler doesnt support it [5.13,031/151] vmlinux.lds.h: Handle clangs module.{c,d}tor sections [5.13,032/151] drm/i915/gvt: Fix cached atomics setting for Windows VM [5.13,033/151] drm/i915/display: Fix the 12 BPC bits for PIPE_MISC reg [5.13,034/151] drm/amd/display: Remove invalid assert for ODM + MPC case [5.13,035/151] drm/amd/display: use GFP_ATOMIC in amdgpu_dm_irq_schedule_work [5.13,036/151] drm/amdgpu: Add preferred mode in modeset when freesync video modes enabled. [5.13,037/151] drm/amdgpu: dont enable baco on boco platforms in runpm [5.13,038/151] drm/amdgpu: handle VCN instances when harvesting (v2) [5.13,039/151] ieee802154: hwsim: fix GPF in hwsim_set_edge_lqi [5.13,040/151] ieee802154: hwsim: fix GPF in hwsim_new_edge_nl [5.13,041/151] drm/mediatek: Fix cursor plane no update [5.13,042/151] pinctrl: mediatek: Fix fallback behavior for bias_set_combo [5.13,043/151] ASoC: cs42l42: Correct definition of ADC Volume control [5.13,044/151] ASoC: cs42l42: Dont allow SND_SOC_DAIFMT_LEFT_J [5.13,045/151] ASoC: cs42l42: Fix bclk calculation for mono [5.13,046/151] interconnect: qcom: icc-rpmh: Add BCMs to commit list in pre_aggregate [5.13,047/151] selftests/sgx: Fix Q1 and Q2 calculation in sigstruct.c [5.13,048/151] ASoC: SOF: Intel: Kconfig: fix SoundWire dependencies [5.13,049/151] ASoC: SOF: Intel: hda-ipc: fix reply size checking [5.13,050/151] ASoC: cs42l42: Fix inversion of ADC Notch Switch control [5.13,051/151] ASoC: cs42l42: Remove duplicate control for WNF filter frequency [5.13,052/151] netfilter: nf_conntrack_bridge: Fix memory leak when error [5.13,053/151] pinctrl: tigerlake: Fix GPIO mapping for newer version of software [5.13,054/151] ASoC: cs42l42: PLL must be running when changing MCLK_SRC_SEL [5.13,055/151] ASoC: cs42l42: Fix LRCLK frame start edge [5.13,056/151] ASoC: cs42l42: Fix mono playback [5.13,057/151] net: dsa: mt7530: add the missing RxUnicast MIB counter [5.13,058/151] net: mvvp2: fix short frame size on s390 [5.13,059/151] platform/x86: pcengines-apuv2: Add missing terminating entries to gpio-lookup tables [5.13,060/151] perf/x86/intel: Apply mid ACK for small core [5.13,061/151] drm/amd/pm: Fix a memory leak in an error handling path in vangogh_tables_init() [5.13,062/151] libbpf: Fix probe for BPF_PROG_TYPE_CGROUP_SOCKOPT [5.13,063/151] libbpf: Do not close un-owned FD 0 on errors [5.13,064/151] bpf: Fix integer overflow involving bucket_size [5.13,065/151] net: dsa: qca: ar9331: make proper initial port defaults [5.13,066/151] net: phy: micrel: Fix link detection on ksz87xx switch" [5.13,067/151] ppp: Fix generating ifname when empty IFLA_IFNAME is specified [5.13,068/151] io_uring: clear TIF_NOTIFY_SIGNAL when running task work [5.13,069/151] net/smc: fix wait on already cleared link [5.13,070/151] net/smc: Correct smc link connection counter in case of smc client [5.13,071/151] net: sched: act_mirred: Reset ct info when mirror/redirect skb [5.13,072/151] ice: Prevent probing virtual functions [5.13,073/151] ice: Stop processing VF messages during teardown [5.13,074/151] ice: dont remove netdev->dev_addr from uc sync list [5.13,075/151] iavf: Set RSS LUT and key in reset handle path [5.13,076/151] psample: Add a fwd declaration for skbuff [5.13,077/151] bareudp: Fix invalid read beyond skbs linear data [5.13,078/151] io-wq: fix bug of creating io-wokers unconditionally [5.13,079/151] io-wq: fix IO_WORKER_F_FIXED issue in create_io_worker() [5.13,080/151] net/mlx5: Dont skip subfunction cleanup in case of error in module init [5.13,081/151] net/mlx5: DR, Add fail on error check on decap [5.13,082/151] net/mlx5e: Avoid creating tunnel headers for local route [5.13,083/151] net/mlx5e: Destroy page pool after XDP SQ to fix use-after-free [5.13,084/151] net/mlx5: Block switchdev mode while devlink traps are active [5.13,085/151] net/mlx5e: TC, Fix error handling memory leak [5.13,086/151] net/mlx5: Synchronize correct IRQ when destroying CQ [5.13,087/151] net/mlx5: Fix return value from tracer initialization [5.13,088/151] drm/meson: fix colour distortion from HDR set during vendor u-boot [5.13,089/151] ovl: fix deadlock in splice write [5.13,090/151] bpf: Fix potentially incorrect results with bpf_get_local_storage() [5.13,091/151] net: dsa: microchip: Fix ksz_read64() [5.13,092/151] net: dsa: microchip: ksz8795: Fix PVID tag insertion [5.13,093/151] net: dsa: microchip: ksz8795: Reject unsupported VLAN configuration [5.13,094/151] net: dsa: microchip: ksz8795: Fix VLAN untagged flag change on deletion [5.13,095/151] net: dsa: microchip: ksz8795: Use software untagging on CPU port [5.13,096/151] net: dsa: microchip: ksz8795: Fix VLAN filtering [5.13,097/151] net: dsa: microchip: ksz8795: Dont use phy_port_cnt in VLAN table lookup [5.13,098/151] net: Fix memory leak in ieee802154_raw_deliver [5.13,099/151] net: igmp: fix data-race in igmp_ifc_timer_expire() [5.13,100/151] net: dsa: hellcreek: fix broken backpressure in .port_fdb_dump [5.13,101/151] net: dsa: lan9303: fix broken backpressure in .port_fdb_dump [5.13,102/151] net: dsa: lantiq: fix broken backpressure in .port_fdb_dump [5.13,103/151] net: dsa: sja1105: fix broken backpressure in .port_fdb_dump [5.13,104/151] pinctrl: sunxi: Dont underestimate number of functions [5.13,105/151] net: bridge: fix flags interpretation for extern learn fdb entries [5.13,106/151] net: bridge: fix memleak in br_add_if() [5.13,107/151] net: linkwatch: fix failure to restore device state across suspend/resume [5.13,108/151] tcp_bbr: fix u32 wrap bug in round logic if bbr_init() called after 2B packets [5.13,109/151] net: igmp: increase size of mr_ifc_count [5.13,110/151] drm/i915: Only access SFC_DONE when media domain is not fused off [5.13,111/151] xen/events: Fix race in set_evtchn_to_irq [5.13,112/151] vsock/virtio: avoid potential deadlock when vsock device remove [5.13,113/151] nbd: Aovid double completion of a request [5.13,114/151] arm64: efi: kaslr: Fix occasional random alloc (and boot) failure [5.13,115/151] KVM: arm64: Fix off-by-one in range_is_memory [5.13,116/151] efi/libstub: arm64: Force Image reallocation if BSS was not reserved [5.13,117/151] efi/libstub: arm64: Relax 2M alignment again for relocatable kernels [5.13,118/151] powerpc/kprobes: Fix kprobe Oops happens in booke [5.13,119/151] i2c: iproc: fix race between client unreg and tasklet [5.13,120/151] x86/tools: Fix objdump version check again [5.13,121/151] genirq: Provide IRQCHIP_AFFINITY_PRE_STARTUP [5.13,122/151] x86/msi: Force affinity setup before startup [5.13,123/151] x86/ioapic: Force affinity setup before startup [5.13,124/151] x86/resctrl: Fix default monitoring groups reporting [5.13,125/151] genirq/msi: Ensure deactivation on teardown [5.13,126/151] genirq/timings: Prevent potential array overflow in __irq_timings_store() [5.13,127/151] powerpc/interrupt: Fix OOPS by not calling do_IRQ() from timer_interrupt() [5.13,128/151] PCI/MSI: Enable and mask MSI-X early [5.13,129/151] PCI/MSI: Mask all unused MSI-X entries [5.13,130/151] PCI/MSI: Enforce that MSI-X table entry is masked for update [5.13,131/151] PCI/MSI: Enforce MSI[X] entry updates to be visible [5.13,132/151] PCI/MSI: Do not set invalid bits in MSI mask [5.13,133/151] PCI/MSI: Correct misleading comments [5.13,134/151] PCI/MSI: Use msi_mask_irq() in pci_msi_shutdown() [5.13,135/151] PCI/MSI: Protect msi_desc::masked for multi-MSI [5.13,136/151] powerpc/interrupt: Do not call single_step_exception() from other exceptions [5.13,137/151] powerpc/pseries: Fix update of LPAR security flavor after LPM [5.13,138/151] powerpc/32s: Fix napping restore in data storage interrupt (DSI) [5.13,139/151] powerpc/smp: Fix OOPS in topology_init() [5.13,140/151] powerpc/xive: Do not skip CPU-less nodes when creating the IPIs [5.13,141/151] powerpc/32: Fix critical and debug interrupts on BOOKE [5.13,142/151] efi/libstub: arm64: Double check image alignment at entry [5.13,143/151] locking/rtmutex: Use the correct rtmutex debugging config option [5.13,144/151] KVM: VMX: Use current VMCS to query WAITPKG support for MSR emulation [5.13,145/151] KVM: nVMX: Use vmx_need_pf_intercept() when deciding if L0 wants a #PF [5.13,146/151] KVM: x86/mmu: Dont leak non-leaf SPTEs when zapping all SPTEs [5.13,147/151] KVM: x86/mmu: Protect marking SPs unsync when using TDP MMU with spinlock [5.13,148/151] ceph: add some lockdep assertions around snaprealm handling [5.13,149/151] ceph: clean up locking annotation for ceph_get_snap_realm and __lookup_snap_realm [5.13,150/151] ceph: take snap_empty_lock atomically with snaprealm refcount change [5.13,151/151] kasan, slub: reset tag when printing address

Message ID

20210816125445.979349376@linuxfoundation.org

State

Superseded

Headers

From: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
To: linux-kernel@vger.kernel.org
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>,
	stable@vger.kernel.org, Sven Auhagen <sven.auhagen@voleatech.de>,
	Matteo Croce <mcroce@microsoft.com>,
	"David S. Miller" <davem@davemloft.net>,
	John Hubbard <jhubbard@nvidia.com>, Sasha Levin <sashal@kernel.org>
Subject: [PATCH 5.13 058/151] net: mvvp2: fix short frame size on s390
Date: Mon, 16 Aug 2021 15:01:28 +0200
Message-Id: <20210816125445.979349376@linuxfoundation.org>
In-Reply-To: <20210816125444.082226187@linuxfoundation.org>
References: <20210816125444.082226187@linuxfoundation.org>
User-Agent: quilt/0.66
MIME-Version: 1.0
Content-Type: text/plain; charset=UTF-8
Content-Transfer-Encoding: 8bit
Precedence: bulk

Series

None | expand

Commit Message

Greg KH Aug. 16, 2021, 1:01 p.m. UTC

From: John Hubbard <jhubbard@nvidia.com>

[ Upstream commit 704e624f7b3e8a4fc1ce43fb564746d1d07b20c0 ]

On s390, the following build warning occurs:

drivers/net/ethernet/marvell/mvpp2/mvpp2.h:844:2: warning: overflow in
conversion from 'long unsigned int' to 'int' changes value from
'18446744073709551584' to '-32' [-Woverflow]
844 |  ((total_size) - MVPP2_SKB_HEADROOM - MVPP2_SKB_SHINFO_SIZE)

This happens because MVPP2_SKB_SHINFO_SIZE, which is 320 bytes (which is
already 64-byte aligned) on some architectures, actually gets ALIGN'd up
to 512 bytes in the s390 case.

So then, when this is invoked:

    MVPP2_RX_MAX_PKT_SIZE(MVPP2_BM_SHORT_FRAME_SIZE)

...that turns into:

     704 - 224 - 512 == -32

...which is not a good frame size to end up with! The warning above is a
bit lucky: it notices a signed/unsigned bad behavior here, which leads
to the real problem of a frame that is too short for its contents.

Increase MVPP2_BM_SHORT_FRAME_SIZE by 32 (from 704 to 736), which is
just exactly big enough. (The other values can't readily be changed
without causing a lot of other problems.)

Fixes: 07dd0a7aae7f ("mvpp2: add basic XDP support")
Cc: Sven Auhagen <sven.auhagen@voleatech.de>
Cc: Matteo Croce <mcroce@microsoft.com>
Cc: David S. Miller <davem@davemloft.net>
Signed-off-by: John Hubbard <jhubbard@nvidia.com>
Signed-off-by: David S. Miller <davem@davemloft.net>
Signed-off-by: Sasha Levin <sashal@kernel.org>
---
 drivers/net/ethernet/marvell/mvpp2/mvpp2.h | 2 +-
 1 file changed, 1 insertion(+), 1 deletion(-)

diff --git a/drivers/net/ethernet/marvell/mvpp2/mvpp2.h b/drivers/net/ethernet/marvell/mvpp2/mvpp2.h
index 4a61c90003b5..722209a14f53 100644
--- a/drivers/net/ethernet/marvell/mvpp2/mvpp2.h
+++ b/drivers/net/ethernet/marvell/mvpp2/mvpp2.h
@@ -938,7 +938,7 @@  enum mvpp22_ptp_packet_format {
 #define MVPP2_BM_COOKIE_POOL_OFFS	8
 #define MVPP2_BM_COOKIE_CPU_OFFS	24
 
-#define MVPP2_BM_SHORT_FRAME_SIZE	704	/* frame size 128 */
+#define MVPP2_BM_SHORT_FRAME_SIZE	736	/* frame size 128 */
 #define MVPP2_BM_LONG_FRAME_SIZE	2240	/* frame size 1664 */
 #define MVPP2_BM_JUMBO_FRAME_SIZE	10432	/* frame size 9856 */
 /* BM short pool packet size

[5.13,058/151] net: mvvp2: fix short frame size on s390

Commit Message

Patch