- 01 Jun, 2017 1 commit
-
-
Woojung Huh authored
Adding support for the Microchip KSZ switch family tail tagging. Reviewed-by:
Andrew Lunn <andrew@lunn.ch> Reviewed-by:
Florian Fainelli <f.fainelli@gmail.com> Signed-off-by:
Woojung Huh <Woojung.Huh@microchip.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
- 31 May, 2017 39 commits
-
-
David S. Miller authored
Alexei Starovoitov says: ==================== bpf: stack depth tracking Introduce tracking of bpf program stack depth in the verifier and use that info to reduce bpf program stack consumption in the interpreter and x64 JIT. Other JITs can take advantage of it as well in the future. Most of the programs consume very little stack, so it's good optimization in general and it's the first step toward bpf to bpf function calls. Also use internal opcode for bpf_tail_call() marking to make clear that jmp|call|x opcode is not uapi and may be used for actual indirect call opcode in the future. ==================== Signed-off-by: -
Alexei Starovoitov authored
Take advantage of stack_depth tracking in x64 JIT. Round up allocated stack by 8 bytes to make sure it stays aligned for functions called from JITed bpf program. Signed-off-by:
Alexei Starovoitov <ast@kernel.org> Acked-by:
Daniel Borkmann <daniel@iogearbox.net> Signed-off-by:
-
Alexei Starovoitov authored
in order to JIT programs with different stack sizes we need to make epilogue and exception path to be stack size independent, hence move auxiliary stack space from the bottom of the stack to the top of the stack. Nice side effect is that JITed function prologue becomes shorter due to imm8 offset encoding vs imm32. Signed-off-by:
-
Alexei Starovoitov authored
16 __bpf_prog_run() interpreters for various stack sizes add .text but not a lot comparing to run-time stack savings text data bss dec hex filename 26350 10328 624 37302 91b6 kernel/bpf/core.o.before_split 25777 10328 624 36729 8f79 kernel/bpf/core.o.after_split 26970 10328 624 37922 9422 kernel/bpf/core.o.now Signed-off-by:
-
Alexei Starovoitov authored
test_bpf.ko doesn't call verifier before selecting interpreter or JITing, hence the tests need to manually specify the amount of stack they consume. Signed-off-by:
-
Alexei Starovoitov authored
To track stack depth of classic bpf programs we only need to analyze ST|STX instructions, since check_load_and_stores() verifies that programs can load from stack only after write. We also need to change the way cBPF stack slots map to eBPF stack, since typical classic programs are using slots 0 and 1, so they need to map to stack offsets -4 and -8 respectively in order to take advantage of small stack interpreter and JITs. Signed-off-by:
-
Alexei Starovoitov authored
The next set of patches will take advantage of stack_depth tracking, so make sure that the program that does bpf_tail_call() has stack depth large enough for the callee. We could have tracked the stack depth of the prog_array owner program and only allow insertion of the programs with stack depth less than the owner, but it will break existing applications. Some of them have trivial root bpf program that only does multiple bpf_tail_calls and at init time the prog array is empty. In the future we may add a flag to do such tracking optionally, but for now play simple and safe. Signed-off-by:
-
Alexei Starovoitov authored
teach verifier to track bpf program stack depth Signed-off-by:
-
Alexei Starovoitov authored
split __bpf_prog_run() interpreter into stack allocation and execution parts. The code section shrinks which helps interpreter performance in some cases. text data bss dec hex filename 26350 10328 624 37302 91b6 kernel/bpf/core.o.before 25777 10328 624 36729 8f79 kernel/bpf/core.o.after Very short programs got slower (due to extra function call): Before: test_bpf: #89 ALU64_ADD_K: 1 + 2 = 3 jited:0 7 PASS test_bpf: #90 ALU64_ADD_K: 3 + 0 = 3 jited:0 8 PASS test_bpf: #91 ALU64_ADD_K: 1 + 2147483646 = 2147483647 jited:0 7 PASS test_bpf: #92 ALU64_ADD_K: 4294967294 + 2 = 4294967296 jited:0 11 PASS test_bpf: #93 ALU64_ADD_K: 2147483646 + -2147483647 = -1 jited:0 7 PASS After: test_bpf: #89 ALU64_ADD_K: 1 + 2 = 3 jited:0 11 PASS test_bpf: #90 ALU64_ADD_K: 3 + 0 = 3 jited:0 11 PASS test_bpf: #91 ALU64_ADD_K: 1 + 2147483646 = 2147483647 jited:0 11 PASS test_bpf: #92 ALU64_ADD_K: 4294967294 + 2 = 4294967296 jited:0 14 PASS test_bpf: #93 ALU64_ADD_K: 2147483646 + -2147483647 = -1 jited:0 10 PASS Longer programs got faster: Before: test_bpf: #266 BPF_MAXINSNS: Ctx heavy transformations jited:0 20286 20513 PASS test_bpf: #267 BPF_MAXINSNS: Call heavy transformations jited:0 31853 31768 PASS test_bpf: #268 BPF_MAXINSNS: Jump heavy test jited:0 9815 PASS test_bpf: #269 BPF_MAXINSNS: Very long jump backwards jited:0 6 PASS test_bpf: #270 BPF_MAXINSNS: Edge hopping nuthouse jited:0 13959 PASS test_bpf: #271 BPF_MAXINSNS: Jump, gap, jump, ... jited:0 210 PASS test_bpf: #272 BPF_MAXINSNS: ld_abs+get_processor_id jited:0 21724 PASS test_bpf: #273 BPF_MAXINSNS: ld_abs+vlan_push/pop jited:0 19118 PASS After: test_bpf: #266 BPF_MAXINSNS: Ctx heavy transformations jited:0 19008 18827 PASS test_bpf: #267 BPF_MAXINSNS: Call heavy transformations jited:0 29238 28450 PASS test_bpf: #268 BPF_MAXINSNS: Jump heavy test jited:0 9485 PASS test_bpf: #269 BPF_MAXINSNS: Very long jump backwards jited:0 12 PASS test_bpf: #270 BPF_MAXINSNS: Edge hopping nuthouse jited:0 13257 PASS test_bpf: #271 BPF_MAXINSNS: Jump, gap, jump, ... jited:0 213 PASS test_bpf: #272 BPF_MAXINSNS: ld_abs+get_processor_id jited:0 19389 PASS test_bpf: #273 BPF_MAXINSNS: ld_abs+vlan_push/pop jited:0 19583 PASS For real world production programs the difference is noise. This patch is first step towards reducing interpreter stack consumption. Signed-off-by:
-
Alexei Starovoitov authored
free up BPF_JMP | BPF_CALL | BPF_X opcode to be used by actual indirect call by register and use kernel internal opcode to mark call instruction into bpf_tail_call() helper. Signed-off-by:
-
David S. Miller authored
Jakub Kicinski says: ==================== nfp: move BPF offload code into app This series moves the eBPF offload code out of netdev/vNIC handling and starts building the nfp_app. Port init is moved into the apps as well because various apps associate vNICs, representors with ports differently. First patch adds a helper for updating tc stats which has been waiting in my tree to be included in any moderately related series. Next series will bring communicating with FW using control messages, then representors, BPF maps, tc flower... :) ==================== Signed-off-by: -
Jakub Kicinski authored
Free management FW info when app FW load failed. Fixes: eefbde7e ("nfp: add hwmon support") Signed-off-by:
Jakub Kicinski <jakub.kicinski@netronome.com> Signed-off-by:
-
Jakub Kicinski authored
Allow apps to associate private data with vNICs and move BPF-specific fields of nfp_net to such structure. Signed-off-by:
-
Jakub Kicinski authored
Move bulk of the eBPF offload code out of common vNIC code into app-specific callbacks. Signed-off-by:
-
Jakub Kicinski authored
Pure move of eBPF offload files to BPF app directory, only change the names and relative header location. nfp_asm.h stays in the main dir and it doesn't really have to include nfp_bpf.h. Signed-off-by:
-
Jakub Kicinski authored
Let the app print its name in ethtool -i output. Signed-off-by:
-
Jakub Kicinski authored
Start fleshing out the apps by turning the vNIC init code to a per-app callback. The two initial apps we have are NIC and eBPF. Signed-off-by:
-
Jakub Kicinski authored
Turn the function to read number of ports into a generic helper. While at it make sure we propagate all errors other than -ENOENT. Signed-off-by:
-
Jakub Kicinski authored
GCC 7 checks for fall through comments, add the two missing ones. Signed-off-by:
-
Jakub Kicinski authored
Forgetting to disable preemption around tcf_action_stats_update() seems to be a common mistake. Add a helper function for updating stats on all actions of a filter. Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queueDavid S. Miller authored
Jeff Kirsher says: ==================== 10GbE Intel Wired LAN Driver Updates 2017-05-31 This series contains updates to ixgbe and ixgbevf only. Scott enables support for TSO & GSO for MPLS encapsulated packets for both ixgbe and ixgbevf. Liwei Song fixes an issue where seqcount/seqlock in ixgbe_get_stats64() are not initialized in time, so move the initialization into probe routine after the transmit and receive rings are initialized. Paul cleans up led_[on|off] for X550EM_X, since the firmware configures the PHY & MAC and we have no PHY access so LED on/off is not supported with this device. Emil provides several fixes, starting with enabling RSS on VF to VF traffic on the same PF. Fixed PHY identification, where the previous method was unreliable, so use a different register to ensure proper identification. Cleaned up the logic which could cause us to skip the link configuration, this skipping over the link configuration was leaving SFP+ PHY's in an unstable state, so always call setup_mac_link(). Added RS1 (rate select 1) support for ixgbe. Lastly, fixed incorrect logic in the setting up of SFP+ link speed. Mark fixes the thermal sensor event logic, where it was being executed when there really was no thermal event. So simplify the logic to only execute when there is a thermal event. Tony adds additional error checks and reporting when setting a VF MAC address to ensure that the MAC filter was successfully added. Also fixed possible truncation warnings, as well as implicit fallthrough warnings. ==================== Signed-off-by: -
git://git.kernel.org/pub/scm/linux/kernel/git/jkirsher/next-queueDavid S. Miller authored
Jeff Kirsher says: ==================== 40GbE Intel Wired LAN Driver Updates 2017-05-31 This series contains updates to i40e and i40evf only. Jesse provides a couple of fixes, starting with cleaning up duplicate lines of code. Fixed a missing line which enables RSS as a negotiated feature. Since the VF does not have any way of reporting FCoE enabled, so just force the code to always report FCoE as disabled. Jake provides several fixes and changes, starting with fixing a race condition in i40e. The hardware has a limitation on transmit PTP packets, which requires us to limit the driver to timestamping a single packet at once. This is done using a state bitlock which enforces that only one timestamp request is honored at a time, unfortunately this suffers from a race condition. Fixed a corner case where we failed to cleanup the bit lock after a failed transmit, and resulted in a state bit being locked forever. Added a new statistic which tracks when a transmit timestamp request is skipped/ignored, since the driver can only handle one transmit timestamp request at a time. Christophe Jaillet fixes a NULL pointer dereference if kzalloc fails. ==================== Signed-off-by: -
Michael Kelley authored
Add direct #include statements for declarations of csum_tcpudp_magic() and csum_ipv6_magic(). While the needed #include's are picked up indirectly for the x86 architecture, they aren't on other architectures, resulting in compile errors. Signed-off-by:
Michael Kelley <mikelley@microsoft.com> Signed-off-by:
-
Vlad Yasevich authored
Small clean-up to rtmsg_ifinfo() to use the rtnl_get_event() interface instead of using 'internal' values directly. Signed-off-by:
Vladislav Yasevich <vyasevic@redhat.com> Acked-by:
David Ahern <dsahern@gmail.com> Signed-off-by:
-
Jesper Dangaard Brouer authored
An eBPF ELF file generated with LLVM can contain several program section, which can be used for bpf tail calls. The bpf prog file descriptors are accessible via array prog_fd[]. At-least XDP samples assume ordering, and uses prog_fd[0] is the main XDP program to attach. The actual order of array prog_fd[] depend on whether or not a bpf program section is referencing any maps or not. Not using a map result in being loaded/processed after all other prog section. Thus, this can lead to some very strange and hard to debug situation, as the user can only see a FD and cannot correlated that with the ELF section name. The fix is rather simple, and even removes duplicate memcmp code. Simply load program sections as the last step, instead of load_and_attach while processing the relocation section. When working with tail calls, it become even more essential that the order of prog_fd[] is consistant, like the current dependency of the map_fd[] order. Signed-off-by:
Jesper Dangaard Brouer <brouer@redhat.com> Acked-by:
-
Arnd Bergmann authored
The new mlxfw code fails to build without the xz library: drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2.o: In function `mlxfw_mfa2_xz_dec_run': :(.text.mlxfw_mfa2_xz_dec_run+0x8): undefined reference to `xz_dec_run' drivers/net/ethernet/mellanox/mlxfw/mlxfw_mfa2.o: In function `mlxfw_mfa2_file_component_get': :(.text.mlxfw_mfa2_file_component_get+0x218): undefined reference to `xz_dec_init' :(.text.mlxfw_mfa2_file_component_get+0x2c0): undefined reference to `xz_dec_end' This adds a Kconfig 'select' statement for it, which is also what the other user of that library has. Fixes: 410ed13c ("Add the mlxfw module for Mellanox firmware flash process") Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Acked-by:
Yotam Gigi <yotamg@mellanox.com> Signed-off-by:
-
Vivien Didelot authored
The current dsa_register_switch function takes a useless struct device pointer argument, which always equals ds->dev. Drivers either call it with ds->dev, or with the same device pointer passed to dsa_switch_alloc, which ends up being assigned to ds->dev. This patch removes the second argument of the dsa_register_switch and _dsa_register_switch functions. Signed-off-by:
Vivien Didelot <vivien.didelot@savoirfairelinux.com> Signed-off-by:
-
David S. Miller authored
Vivien Didelot says: ==================== net: dsa: mv88e6xxx: PHY nitpicks This patchset isolates more PPU code into phy.c and makes distinction between PHY Registers read and write implementations vs. generic PHY routines. ==================== Signed-off-by: -
Vivien Didelot authored
Respect the implicit naming convention used in all register sets specific files, by renaming the mv88e6xxx_ppu_* functions with the mv88e6xxx_phy_* prefix. This is simply a s/xxx_ppu/xxx_phy_ppu/ substitution. Signed-off-by:
-
Vivien Didelot authored
Make it clear that mv88e6xxx_phy_ppu_{read,write} are an implementation of the .phy_{read,write} operations, by renaming them with the mv88e6185 prefix, since 88E6185 it is the reference switch model supported in an upstream board (ZII Dev Rev B), which makes use of them. Distinguish the signatures of implementation specific and generic PHY functions in the phy.h header. Signed-off-by: -
Vivien Didelot authored
Similarly to the VTU, PVT and ATU setup, provide a mv88e6xxx_phy_setup helper which wraps mv88e6xxx_ppu_enable, so that no more PPU-related functions are exposed outside of phy.c. Thus make mv88e6xxx_ppu_enable static. Signed-off-by:
-
Emil Tantilov authored
Check for ret_val instead of !ret_val to allow the rest of the code to execute and configure the speed properly. Signed-off-by:
Emil Tantilov <emil.s.tantilov@intel.com> Tested-by:
Krishneil Singh <krishneil.k.singh@intel.com> Signed-off-by:
Jeff Kirsher <jeffrey.t.kirsher@intel.com>
-
Emil Tantilov authored
Add RS1 configuration to ixgbe_set_soft_rate_select_speed() Signed-off-by:
Andrew Bowers <andrewx.bowers@intel.com> Signed-off-by:
-
Emil Tantilov authored
Remove the logic which would previously skip the link configuration in the case where we are already at the requested speed in ixgbe_setup_mac_link_multispeed_fiber(). By exiting early we are skipping the link configuration and as such the driver may not always configure the PHY correctly for SFP+. Signed-off-by:
-
Emil Tantilov authored
Make sure the writes are processed immediately. Without the flush it is possible for operations on one port to spill over the other as the resource is shared. Signed-off-by:
-
Emil Tantilov authored
Previous method was unreliable. Use a different register to differentiate between the SKUs. Signed-off-by:
-
Tony Nguyen authored
Additions to gcc 7 now warn whenever a switch statement falls through implicitly. This patch adds explicit fall through comments to address the following warnings: drivers/net/ethernet/intel/ixgbevf/vf.c: In function ‘ixgbevf_get_reta_locked’: drivers/net/ethernet/intel/ixgbevf/vf.c:336:6: warning: this statement may fall through [-Wimplicit-fallthrough=] if (hw->mac.type < ixgbe_mac_X550_vf) ^ drivers/net/ethernet/intel/ixgbevf/vf.c:338:2: note: here default: ^~~~~~~ drivers/net/ethernet/intel/ixgbevf/vf.c: In function ‘ixgbevf_get_rss_key_locked’: drivers/net/ethernet/intel/ixgbevf/vf.c:402:6: warning: this statement may fall through [-Wimplicit-fallthrough=] if (hw->mac.type < ixgbe_mac_X550_vf) ^ drivers/net/ethernet/intel/ixgbevf/vf.c:404:2: note: here default: ^~~~~~~ Signed-off-by:Tony Nguyen <anthony.l.nguyen@intel.com> Tested-by:
-
Tony Nguyen authored
The following warning is now shown as a result of new checks added for gcc 7: drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c: In function ‘ixgbevf_open’: drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c:1363:13: warning: ‘%d’ directive output may be truncated writing between 1 and 10 bytes into a region of size between 3 and 18 [-Wformat-truncation=] "%s-%s-%d", netdev->name, "TxRx", ri++); ^~ drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c:1363:6: note: directive argument in the range [0, 2147483647] "%s-%s-%d", netdev->name, "TxRx", ri++); ^~~~~~~~~~ drivers/net/ethernet/intel/ixgbevf/ixgbevf_main.c:1362:4: note: ‘snprintf’ output between 8 and 32 bytes into a destination of size 24 snprintf(q_vector->name, sizeof(q_vector->name) - 1, ^~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ "%s-%s-%d", netdev->name, "TxRx", ri++); ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ Resolve this warning by making a couple of changes. - Don't reserve space for the null terminator. Since snprintf adds the null terminator automatically, there is no need for us to reserve a byte for it. - Change a couple variables that can never be negative from int to unsigned int. While we're making changes to the format string, move the constant strings into the format string instead of providing them as specifiers. Signed-off-by: -
Tony Nguyen authored
This patch adds/changes fall through comments to address new warnings produced by gcc 7. Fixed formatting on a couple of comments in the function. Signed-off-by:
-
