- 24 Jul, 2013 3 commits
-
-
Dan Carpenter authored
We can't dereference "ent" after passing it to free_cmd(). Signed-off-by:
Dan Carpenter <dan.carpenter@oracle.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Lekensteyn authored
The work queue is initialised in rtl_open (when the interface goes up), but canceled in rtl_remove_one (when the PCI device gets removed). If the network interface is not brought up, then the work queue struct is not initialised. When the device is removed, the attempt to cancel the uninitialised work queue causes a lockdep warning. This patch fixes the issue by moving cancel_work_sync to rtl_close (to match rtl_open). (Note that rtl_close is also called via unregister_netdev in rtl_remove_one.) Signed-off-by:
Peter Wu <lekensteyn@gmail.com> Acked-by:
Francois Romieu <romieu@fr.zoreil.com> Signed-off-by:
-
Fabio Estevam authored
Add MODULE_ALIAS, so that auto module loading can work. Signed-off-by:
Fabio Estevam <fabio.estevam@freescale.com> Signed-off-by:
-
- 23 Jul, 2013 1 commit
-
-
Ben Hutchings authored
Received packets are only scattered if this is enabled in both the matching filter and the receiving queue. This was not being done for filters inserted for RFS, so any packet requiring more than a single descriptor was dropped. Signed-off-by:
Ben Hutchings <bhutchings@solarflare.com> Signed-off-by:
-
- 22 Jul, 2013 2 commits
-
-
Jiri Slaby authored
Since 4bb16672 (build some drivers only when compile-testing), PTP_1588_CLOCK_PCH depends on (X86 || COMPILE_TEST). But PCH_GBE selects PTP_1588_CLOCK_PCH without depending on x86. Fix this by adding the same dependency here. Signed-off-by:
Jiri Slaby <jslaby@suse.cz> Reported-by: <fengguang.wu@intel.com> [intel's build test robot] Cc: "David S. Miller" <davem@davemloft.net> Cc: Ben Hutchings <ben@decadent.org.uk> Cc: netdev@vger.kernel.org Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org> Cc: kbuild-all@01.org Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/pablo/nfDavid S. Miller authored
Pablo Neira Ayuso says: ==================== The following patchset contains Netfilter fixes for your net tree, they are: * Fix potential NULL dereference in the socket match if revision 0 is used, from Eric Dumazet. * Fix missing expectation NAT initialization that results in dumping the NAT part via ctnetlink, thus leading to problems in expectation synchronization through conntrackd, from myself. ==================== Signed-off-by:
-
- 21 Jul, 2013 2 commits
-
-
git://git.kernel.org/pub/scm/linux/kernel/git/linville/wirelessDavid S. Miller authored
John W. Linville says: ==================== Please accept this batch of fixes intended for the 3.11 tree... Alexey Khoroshilov fixes a suspend-related race in ath9k_htc. Arnd Bergmann corrects the alignment of a structure in the ssb code to be compatible with ARM devices. Bob Copeland provides an ath5k fix that corrects a mistaken variable initialization. Felix Fietkau corrects some frame accounting for dropped frames in ath9k. Geert Uytterhoeven brings a Kconfig fix to indicate the DMA requirements for rt2x00. Larry Finger offers two rtlwifi fixes: one that properly initializes a callback; and, a scattered collection of Kconfig, Makefile, and EXPORT_SYMBOL changes that correct some build problems. Finally, Sujith Manoharan provides an ath9k fix to disable a feature on a specific hardware device. ==================== Signed-off-by: -
git://gitorious.org/linux-can/linux-canDavid S. Miller authored
Marc Kleine-Budde says: ==================== here are two fixes for the v3.11 release cycle: Maximilian Schneider contributes a patch for the esd_usb2 CAN driver. It adds sanity checking to the data coming from the USB CAN adapter before using it. Alexey Khoroshilov from the Linux Driver Verification project fixes an urb leak in the error handling of the USB 8dev's usb_8dev_start() function. ==================== Signed-off-by:
-
- 20 Jul, 2013 15 commits
-
-
Eric Dumazet authored
commit 9f00b2e7 ("bridge: only expire the mdb entry when query is received") added a nasty bug as an active timer can be reinitialized. setup_timer() must be done once, no matter how many time mod_timer() is called. br_multicast_new_group() is the right place to do this. Reported-by:
Srivatsa S. Bhat <srivatsa.bhat@linux.vnet.ibm.com> Diagnosed-by:
Thomas Gleixner <tglx@linutronix.de> Signed-off-by:
Eric Dumazet <edumazet@google.com> Tested-by:
-
Michal Tesar authored
Limit the min/max value passed to the /proc/sys/net/ipv4/tcp_syn_retries. Signed-off-by:
Michal Tesar <mtesar@redhat.com> Signed-off-by:
-
Dan Carpenter authored
This patch doesn't change the compiled code because ARC_HDR_SIZE is 4 and sizeof(int) is 4, but the intent was to use the header size and not the sizeof the header size. Signed-off-by:
-
Flavio Leitner authored
The veth device doesn't provide the vlan features, so TSO for example is disabled and that causes performance issues when using tagged traffic. The test topology looks like this: br0 br1 / \ / \ vnet veth0.10 ----- veth1.10 vnet VM VM The netperf results with current veth driver: MIGRATED TCP STREAM TEST from 192.168.1.1 () port 0 AF_INET to 192.168.1.2 () port 0 AF_INET Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 87380 16384 16384 10.01 2210.22 Now after applying the proposed patch: MIGRATED TCP STREAM TEST from 192.168.1.1 () port 0 AF_INET to 192.168.1.2 () port 0 AF_INET Recv Send Send Socket Socket Message Elapsed Size Size Size Time Throughput bytes bytes bytes secs. 10^6bits/sec 87380 16384 16384 10.00 13067.47 Signed-off-by:Flavio Leitner <fbl@redhat.com> Signed-off-by:
-
Manish Chopra authored
o Clear cached vport vlan variable(vp->vlan) in PF on PCI FLR and back-channel termination which will allow to configure guest VLAN on VF after force off/shut down the guest VM. Signed-off-by:
Manish Chopra <manish.chopra@qlogic.com> Signed-off-by:
Sucheta Chakraborty <sucheta.chakraborty@qlogic.com> Signed-off-by:
-
Jitendra Kalsaria authored
o Driver was freeing Tx frag which was never mapped before which result into panic as kernel was unable to handle paging request. BUG: unable to handle kernel paging request at ffffc9002552a000 IP: [<ffffffffa05ed762>] qlcnic_release_tx_buffers+0x72/0x170 [qlcnic] PGD 87fc15067 PUD 47febf067 PMD 4758c5067 PTE 0 Oops: 0000 [#1] SMP crash> bt PID: 27343 TASK: ffff8802a5de8080 CPU: 27 COMMAND: "ifconfig" [ffff8802a34b3850] machine_kexec at ffffffff81035b7b [ffff8802a34b38b0] crash_kexec at ffffffff810c0db2 [ffff8802a34b3980] oops_end at ffffffff815111d0 [ffff8802a34b39b0] no_context at ffffffff81046bfb [ffff8802a34b3a00] __bad_area_nosemaphore at ffffffff81046e85 [ffff8802a34b3a50] bad_area_nosemaphore at ffffffff81046f53 [ffff8802a34b3a60] __do_page_fault at ffffffff810476b1 [ffff8802a34b3b80] do_page_fault at ffffffff8151311e [ffff8802a34b3bb0] page_fault at ffffffff815104d5 [exception RIP: qlcnic_release_tx_buffers+114] RIP: ffffffffa05ed762 RSP: ffff8802a34b3c68 RFLAGS: 00010246 RAX: ffff88087989c000 RBX: ffffc90025529ff8 RCX: 0000000000000001 RDX: 0000000000000013 RSI: 0000000000000013 RDI: 0000000000000000 RBP: ffff8802a34b3ca8 R8: 0000000000000000 R9: 0000000000000000 R10: 000000000000000c R11: 0000000000000000 R12: 0000000000000012 R13: ffffc90025529ec0 R14: ffff880761e876e0 R15: 00000000000003ff ORIG_RAX: ffffffffffffffff CS: 0010 SS: 0018 [ffff8802a34b3cb0] __qlcnic_down at ffffffffa05e8b15 [qlcnic] [ffff8802a34b3d00] qlcnic_close at ffffffffa05e8b78 [qlcnic] [ffff8802a34b3d10] dev_close at ffffffff81449d81 [ffff8802a34b3d30] dev_change_flags at ffffffff814495c1 Signed-off-by:Jitendra Kalsaria <jitendra.kalsaria@qlogic.com> Signed-off-by:
-
Shahed Shaikh authored
Driver was using wrong mask for template version. Signed-off-by:
Shahed Shaikh <shahed.shaikh@qlogic.com> Signed-off-by:
-
Himanshu Madhani authored
o Commit b938662d ("qlcnic: Fix ethtool supported port status for 83xx") introduced regression for display of link status for 83xx adapter while refactoring port status display. This patch is to fix the link status display for 83xx adapter. Signed-off-by:
Himanshu Madhani <himanshu.madhani@qlogic.com> Signed-off-by:
-
Manish Chopra authored
o "qlcnic_sriov" structure pointer should be accessed only when SR-IOV is enabled. Access this pointer after SR-IOV PF check. Signed-off-by:
-
Sucheta Chakraborty authored
o Multicast MAC was not getting programmed due to which multicast packets were being dropped by FW. This patch fixes commit 168e4fb54c11865668ad50eff81b5f2729e0e0f4 ("qlcnic: Secondary unicast MAC address support.") which introduced bug in handling multicast packets. Signed-off-by: -
Sucheta Chakraborty authored
o Check for non-NULL set_mac_filter_count function pointer before calling it fixes the panic. This patch fixes regression introduced by patch "qlcnic: Secondary unicast MAC address support." with commit id 168e4fb54c11865668ad50eff81b5f2729e0e0f4. Signed-off-by:
-
Pratik Pujar authored
NAPI poll function does not re-enable the interrupt, if __QLCNIC_DEV_UP is not set in adapter state. This was preventing driver from receiving any packet. Signed-off-by:
Pratik Pujar <pratik.pujar@qlogic.com> Signed-off-by:
-
Shahed Shaikh authored
83xx adapter specific code was accessing 82xx register which resulted in invalid register offset. This patch uses proper register access method. Signed-off-by:
-
stephen hemminger authored
There are two race conditions in existing code for doing IGMP management in workqueue in vxlan. First, the vxlan_group_used function checks the list of vxlan's without any protection, and it is possible for open followed by close to occur before the igmp work queue runs. To solve these move the check into vxlan_open/stop so it is protected by RTNL. And split into two work structures so that there is no racy reference to underlying device state. Signed-off-by:
Stephen Hemminger <stephen@networkplumber.org> Signed-off-by:
-
stephen hemminger authored
Fix memory leaks and other badness from VXLAN network namespace teardown. When network namespace is removed, all the vxlan devices should be unregistered (not closed). Signed-off-by:
Pravin B Shelar <pshelar@nicira.com> Signed-off-by:
-
- 19 Jul, 2013 6 commits
-
-
John W. Linville authored
Merge branch 'master' of git://git.kernel.org/pub/scm/linux/kernel/git/linville/wireless into for-davem
-
Alexey Khoroshilov authored
If usb_8dev_start() fails to submit urb, it unanchors the urb but forgets to free it. Found by Linux Driver Verification project (linuxtesting.org). Signed-off-by:
Alexey Khoroshilov <khoroshilov@ispras.ru> Signed-off-by:
Marc Kleine-Budde <mkl@pengutronix.de>
-
Maximilian Schneider authored
The esd_usb2_read_bulk_callback() function is parsing the data that comes from the USB CAN adapter. One datum is used as an index to access the dev->nets[] array. This patch adds the missing bounds checking. Acked-by:
Matthias Fuchs <matthias.fuchs@esd.eu> Signed-off-by:
Maximilian Schneider <max@schneidersoft.net> Signed-off-by:
-
git://git.kernel.org/pub/scm/linux/kernel/git/davem/netLinus Torvalds authored
Pull networking fixes from David Miller: "A couple interesting SKB fragment handling fixes, plus the usual small bits here and there: 1) Fix 64-bit divide build failure on 32-bit platforms in mlx5, from Tim Gardner. 2) Get rid of a stupid reimplementation on "%*phC" in our sysfs MAC address printing helper. 3) Fix NETIF_F_SG capability advertisement in hyperv driver, if the device can't do checksumming offloads then it shouldn't say it can do SG either. From Haiyang Zhang. 4) bgmac needs to depend on PHYLIB, from Hauke Mehrtens. 5) Don't leak DMA mappings on mapping failures, from Neil Horman. 6) We need to reset the transport header of SKBs in ipv4 before we attempt to perform early socket demux, just like ipv6 does. From Eric Dumazet. 7) Add missing locking on vxlan device removal, from Stephen Hemminger. 8) xen-netfront has to make two passes over an SKB to prepare it for transfer. One pass calculates the number of slots needed, the second massages the SKB and fills the slots. Unfortunately, the first pass doesn't calculate the number of slots properly so we can end up trying to build a MAX_SKB_FRAGS + 1 SKB which doesn't work out so well. Fix from Jan Beulich with help and discussion with several others. 9) Fix a similar problem in tun and macvtap, which have to split up scatter-gather elements at PAGE_SIZE boundaries. Don't do zerocopy if it would result in a > MAX_SKB_FRAGS skb. Fixes from Jason Wang. 10) On receive, once we've decoded the VLAN state completely, clear skb->vlan_tci. Otherwise demuxed tunnels underneath can trigger the VLAN code again, corrupting the packet. Fix from Eric Dumazet" * git://git.kernel.org/pub/scm/linux/kernel/git/davem/net: vlan: fix a race in egress prio management vlan: mask vlan prio bits macvtap: do not zerocopy if iov needs more pages than MAX_SKB_FRAGS tuntap: do not zerocopy if iov needs more pages than MAX_SKB_FRAGS pkt_sched: sch_qfq: remove a source of high packet delay/jitter xen-netfront: pull on receive skb may need to happen earlier vxlan: add necessary locking on device removal hyperv: Fix the NETIF_F_SG flag setting in netvsc net: Fix sysfs_format_mac() code duplication. be2net: Fix to avoid hardware workaround when not needed macvtap: do not assume 802.1Q when send vlan packets macvtap: fix the missing ret value of TUNSETQUEUE ipv4: set transport header earlier mlx5 core: Fix __udivdi3 when compiling for 32 bit arches bgmac: add dependency to phylib net/irda: fixed style issues in irlan_eth ethtool: fixed trailing statements in ethtool ndisc: bool initializations should use true and false atl1e: unmap partially mapped skb on dma error and free skb -
git://git.kernel.org/pub/scm/linux/kernel/git/tip/tipLinus Torvalds authored
Pull x86 fixes from Peter Anvin: "Trying again to get the fixes queue, including the fixed IDT alignment patch. The UEFI patch is by far the biggest issue at hand: it is currently causing quite a few machines to boot. Which is sad, because the only reason they would is because their BIOSes touch memory that has already been freed. The other major issue is that we finally have tracked down the root cause of a significant number of machines failing to suspend/resume" * 'x86-urgent-for-linus' of git://git.kernel.org/pub/scm/linux/kernel/git/tip/tip: x86: Make sure IDT is page aligned x86, suspend: Handle CPUs which fail to #GP on RDMSR x86/platform/ce4100: Add header file for reboot type Revert "UEFI: Don't pass boot services regions to SetVirtualAddressMap()" efivars: check for EFI_RUNTIME_SERVICES
-
git://neil.brown.name/mdLinus Torvalds authored
Pull md bug fixes from NeilBrown: "Sorry boss, back at work now boss. Here's them nice shiny patches ya wanted. All nicely tagged and justified for -stable and everyfing: Three bug fixes for md in 3.10 3.10 wasn't a good release for md. The bio changes left a couple of bugs, and an md "fix" created another one. These three patches appear to fix the issues and have been tagged for -stable" * tag 'md-3.11-fixes' of git://neil.brown.name/md: md/raid1: fix bio handling problems in process_checks() md: Remove recent change which allows devices to skip recovery. md/raid10: fix two problems with RAID10 resync.
-
- 18 Jul, 2013 11 commits
-
-
git://people.freedesktop.org/~airlied/linuxLinus Torvalds authored
Pull drm fixes from Dave Airlie: "You'll be terribly disappointed in this, I'm not trying to sneak any features in or anything, its mostly radeon and intel fixes, a couple of ARM driver fixes" * 'drm-fixes' of git://people.freedesktop.org/~airlied/linux: (34 commits) drm/radeon/dpm: add debugfs support for RS780/RS880 (v3) drm/radeon/dpm/atom: fix broken gcc harder drm/radeon/dpm/atom: restructure logic to work around a compiler bug drm/radeon/dpm: fix atom vram table parsing drm/radeon: fix an endian bug in atom table parsing drm/radeon: add a module parameter to disable aspm drm/rcar-du: Use the GEM PRIME helpers drm/shmobile: Use the GEM PRIME helpers uvesafb: Really allow mtrr being 0, as documented and warn()ed radeon kms: do not flush uninitialized hotplug work drm/radeon/dpm/sumo: handle boost states properly when forcing a perf level drm/radeon: align VM PTBs (Page Table Blocks) to 32K drm/radeon: allow selection of alignment in the sub-allocator drm/radeon: never unpin UVD bo v3 drm/radeon: fix UVD fence emit drm/radeon: add fault decode function for CIK drm/radeon: add fault decode function for SI (v2) drm/radeon: add fault decode function for cayman/TN (v2) drm/radeon: use radeon device for request firmware drm/radeon: add missing ttm_eu_backoff_reservation to radeon_bo_list_validate ...
-
Sujith Manoharan authored
The multicast search bit is disabled for the AR9003 family, but this is required for AR9002 too. Fix this in the INI override routine. Signed-off-by:
Sujith Manoharan <c_manoha@qca.qualcomm.com> Signed-off-by:
John W. Linville <linville@tuxdriver.com>
-
Larry Finger authored
The present build configuration for the rtlwifi family of drivers will fail under two known conditions: (1) If rtlwifi is selected without selecting any of the dependent drivers, there are errors in the build. (2) If the PCI drivers are built into the kernel and the USB drivers are modules, or vice versa, there are missing globals. The first condition is fixed by never building rtlwifi unless at least one of the device drivers is selected. The second failure is fixed by splitting the PCI and USB codes out of rtlwifi, and creating their own mini drivers. If the drivers that use them are modules, they will also be modules. Although a number of files are touched by this patch, only Makefile and Kconfig have undergone significant changes. The only modifications to the other files were to export entry points needed by the new rtl_pci and rtl_usb units, or to rename two variables that had names that were likely to cause namespace collisions. Reported-by: Fengguang Wu <fengguang.wu@intel.com> [Condition 1] Reported-by: Ben Hutchings <bhutchings@solarflare.com> [Condition 2] Signed-off-by:Larry Finger <Larry.Finger@lwfinger.net> Cc: Ben Hutchings <bhutchings@solarflare.com> Cc: Fengguang Wu <fengguang.wu@intel.com> Signed-off-by:
-
Bob Copeland authored
Bit 32 was always set which looks to have been accidental, according to git history. Signed-off-by:
Bob Copeland <me@bobcopeland.com> Signed-off-by:
-
Geert Uytterhoeven authored
If NO_DMA=y: drivers/built-in.o: In function `rt2x00queue_unmap_skb': drivers/net/wireless/rt2x00/rt2x00queue.c:129: undefined reference to `dma_unmap_single' drivers/net/wireless/rt2x00/rt2x00queue.c:133: undefined reference to `dma_unmap_single' drivers/built-in.o: In function `rt2x00queue_map_txskb': drivers/net/wireless/rt2x00/rt2x00queue.c:112: undefined reference to `dma_map_single' drivers/net/wireless/rt2x00/rt2x00queue.c:115: undefined reference to `dma_mapping_error' drivers/built-in.o: In function `rt2x00queue_alloc_rxskb': drivers/net/wireless/rt2x00/rt2x00queue.c:93: undefined reference to `dma_map_single' drivers/net/wireless/rt2x00/rt2x00queue.c:95: undefined reference to `dma_mapping_error' Signed-off-by:
Geert Uytterhoeven <geert@linux-m68k.org> Cc: linux-wireless@vger.kernel.org Signed-off-by:
-
Arnd Bergmann authored
The ARM OABI and EABI disagree on the alignment of structures with small members, so module init tools may interpret the ssb device table incorrectly, as shown by this warning when building the b43 device driver in an OABI kernel: FATAL: drivers/net/wireless/b43/b43: sizeof(struct ssb_device_id)=6 is not a modulo of the size of section __mod_ssb_device_table=88. Forcing the default (EABI) alignment on the structure makes this problem go away. Since the ssb_device_id may have the same problem, better fix both structures. Signed-off-by:
Arnd Bergmann <arnd@arndb.de> Cc: Russell King <linux@arm.linux.org.uk> Cc: John W. Linville <linville@tuxdriver.com> Cc: Michael Buesch <mb@bu3sch.de> Cc: Larry Finger <Larry.Finger@lwfinger.net> Signed-off-by:
-
Eric Dumazet authored
egress_priority_map[] hash table updates are protected by rtnl, and we never remove elements until device is dismantled. We have to make sure that before inserting an new element in hash table, all its fields are committed to memory or else another cpu could find corrupt values and crash. Signed-off-by:
-
Eric Dumazet authored
In commit 48cc32d3 ("vlan: don't deliver frames for unknown vlans to protocols") Florian made sure we set pkt_type to PACKET_OTHERHOST if the vlan id is set and we could find a vlan device for this particular id. But we also have a problem if prio bits are set. Steinar reported an issue on a router receiving IPv6 frames with a vlan tag of 4000 (id 0, prio 2), and tunneled into a sit device, because skb->vlan_tci is set. Forwarded frame is completely corrupted : We can see (8100:4000) being inserted in the middle of IPv6 source address : 16:48:00.780413 IP6 2001:16d8:8100:4000:ee1c:0:9d9:bc87 > 9f94:4d95:2001:67c:29f4::: ICMP6, unknown icmp6 type (0), length 64 0x0000: 0000 0029 8000 c7c3 7103 0001 a0ae e651 0x0010: 0000 0000 ccce 0b00 0000 0000 1011 1213 0x0020: 1415 1617 1819 1a1b 1c1d 1e1f 2021 2223 0x0030: 2425 2627 2829 2a2b 2c2d 2e2f 3031 3233 It seems we are not really ready to properly cope with this right now. We can probably do better in future kernels : vlan_get_ingress_priority() should be a netdev property instead of a per vlan_dev one. For stable kernels, lets clear vlan_tci to fix the bugs. Reported-by:
Steinar H. Gunderson <sesse@google.com> Signed-off-by:
-
Jason Wang authored
We try to linearize part of the skb when the number of iov is greater than MAX_SKB_FRAGS. This is not enough since each single vector may occupy more than one pages, so zerocopy_sg_fromiovec() may still fail and may break the guest network. Solve this problem by calculate the pages needed for iov before trying to do zerocopy and switch to use copy instead of zerocopy if it needs more than MAX_SKB_FRAGS. This is done through introducing a new helper to count the pages for iov, and call uarg->callback() manually when switching from zerocopy to copy to notify vhost. We can do further optimization on top. This bug were introduced from b92946e2 (macvtap: zerocopy: validate vectors before building skb). Cc: Michael S. Tsirkin <mst@redhat.com> Signed-off-by:
Jason Wang <jasowang@redhat.com> Signed-off-by:
-
Jason Wang authored
We try to linearize part of the skb when the number of iov is greater than MAX_SKB_FRAGS. This is not enough since each single vector may occupy more than one pages, so zerocopy_sg_fromiovec() may still fail and may break the guest network. Solve this problem by calculate the pages needed for iov before trying to do zerocopy and switch to use copy instead of zerocopy if it needs more than MAX_SKB_FRAGS. This is done through introducing a new helper to count the pages for iov, and call uarg->callback() manually when switching from zerocopy to copy to notify vhost. We can do further optimization on top. The bug were introduced from commit 0690899b (tun: experimental zero copy tx support) Cc: Michael S. Tsirkin <mst@redhat.com> Signed-off-by:
-
Paolo Valente authored
QFQ+ inherits from QFQ a design choice that may cause a high packet delay/jitter and a severe short-term unfairness. As QFQ, QFQ+ uses a special quantity, the system virtual time, to track the service provided by the ideal system it approximates. When a packet is dequeued, this quantity must be incremented by the size of the packet, divided by the sum of the weights of the aggregates waiting to be served. Tracking this sum correctly is a non-trivial task, because, to preserve tight service guarantees, the decrement of this sum must be delayed in a special way [1]: this sum can be decremented only after that its value would decrease also in the ideal system approximated by QFQ+. For efficiency, QFQ+ keeps track only of the 'instantaneous' weight sum, increased and decreased immediately as the weight of an aggregate changes, and as an aggregate is created or destroyed (which, in its turn, happens as a consequence of some class being created/destroyed/changed). However, to avoid the problems caused to service guarantees by these immediate decreases, QFQ+ increments the system virtual time using the maximum value allowed for the weight sum, 2^10, in place of the dynamic, instantaneous value. The instantaneous value of the weight sum is used only to check whether a request of weight increase or a class creation can be satisfied. Unfortunately, the problems caused by this choice are worse than the temporary degradation of the service guarantees that may occur, when a class is changed or destroyed, if the instantaneous value of the weight sum was used to update the system virtual time. In fact, the fraction of the link bandwidth guaranteed by QFQ+ to each aggregate is equal to the ratio between the weight of the aggregate and the sum of the weights of the competing aggregates. The packet delay guaranteed to the aggregate is instead inversely proportional to the guaranteed bandwidth. By using the maximum possible value, and not the actual value of the weight sum, QFQ+ provides each aggregate with the worst possible service guarantees, and not with service guarantees related to the actual set of competing aggregates. To see the consequences of this fact, consider the following simple example. Suppose that only the following aggregates are backlogged, i.e., that only the classes in the following aggregates have packets to transmit: one aggregate with weight 10, say A, and ten aggregates with weight 1, say B1, B2, ..., B10. In particular, suppose that these aggregates are always backlogged. Given the weight distribution, the smoothest and fairest service order would be: A B1 A B2 A B3 A B4 A B5 A B6 A B7 A B8 A B9 A B10 A B1 A B2 ... QFQ+ would provide exactly this optimal service if it used the actual value for the weight sum instead of the maximum possible value, i.e., 11 instead of 2^10. In contrast, since QFQ+ uses the latter value, it serves aggregates as follows (easy to prove and to reproduce experimentally): A B1 B2 B3 B4 B5 B6 B7 B8 B9 B10 A A A A A A A A A A B1 B2 ... B10 A A ... By replacing 10 with N in the above example, and by increasing N, one can increase at will the maximum packet delay and the jitter experienced by the classes in aggregate A. This patch addresses this issue by just using the above 'instantaneous' value of the weight sum, instead of the maximum possible value, when updating the system virtual time. After the instantaneous weight sum is decreased, QFQ+ may deviate from the ideal service for a time interval in the order of the time to serve one maximum-size packet for each backlogged class. The worst-case extent of the deviation exhibited by QFQ+ during this time interval [1] is basically the same as of the deviation described above (but, without this patch, QFQ+ suffers from such a deviation all the time). Finally, this patch modifies the comment to the function qfq_slot_insert, to make it coherent with the fact that the weight sum used by QFQ+ can now be lower than the maximum possible value. [1] P. Valente, "Extending WF2Q+ to support a dynamic traffic mix", Proceedings of AAA-IDEA'05, June 2005. Signed-off-by:
Paolo Valente <paolo.valente@unimore.it> Signed-off-by:
-
