- 27 Feb, 2009 40 commits
-
-
Andy Grover authored
Signed-off-by:
Andy Grover <andy.grover@oracle.com> Signed-off-by:
David S. Miller <davem@davemloft.net>
-
Andy Grover authored
Signed-off-by:
-
Andy Grover authored
Registers as an RDS transport and an IB client, and uses IB CM API to allocate ids, queue pairs, and the rest of that fun stuff. Signed-off-by:
-
Andy Grover authored
Some transports may support RDMA features. This handles the non-transport-specific parts, like pinning user pages and tracking mapped regions. Signed-off-by:
-
Andy Grover authored
Upon receiving a datagram from the transport, RDS parses the headers and potentially queues an ACK. Signed-off-by:
-
Andy Grover authored
This is the code to send an RDS datagram. Signed-off-by:
-
Andy Grover authored
Parsing of newly-received RDS message headers (including ext. headers) and copy-to/from-user routines. page.c implements a per-cpu page remainder cache, to reduce the number of allocations needed for small datagrams. Signed-off-by:
-
Andy Grover authored
RDS exposes a few tunable parameters via sysctls. Signed-off-by:
-
Andy Grover authored
A simple rds transport to handle loopback connections. Signed-off-by:
-
Andy Grover authored
While arguably the fact that the underlying transport needs a connection to convey RDS's datagrame reliably is not important to rds proper, the transports implemented so far (IB and TCP) have both been connection-oriented, and so the connection state machine-related code is in the common rds code. This patch also includes several work items, to handle connecting, sending, receiving, and shutdown. Signed-off-by:
-
Andy Grover authored
RDS currently generates a lot of stats that are accessible via the rds-info utility. This code implements the support for this. Signed-off-by:
-
Andy Grover authored
RDS supports multiple transports. While this initial submission only supports Infiniband transport, this abstraction allows others to be added. We're working on an iWARP transport, and also see UDP over DCB as another possibility. This code handles transport registration. Signed-off-by:
-
Andy Grover authored
RDS handles per-socket congestion by updating peers with a complete congestion map (8KB). This code keeps track of these maps for itself and ones received from peers. Signed-off-by:
-
Andy Grover authored
RDS's main data structure definitions and exported functions. Signed-off-by:
-
Andy Grover authored
Implement the RDS (Reliable Datagram Sockets) interface. Signed-off-by:
-
Ben Hutchings authored
Based on a patch by Jaswinder Singh <jaswinder@infradead.org>. Compile-tested only. Signed-off-by:
Ben Hutchings <ben@decadent.org.uk> Signed-off-by:
-
Ben Hutchings authored
Firmware blob is little endian. Compile-tested only. Signed-off-by:
-
Matt Carlson authored
This patch injects myself into the TG3 MAINTAINERS entry. Signed-off-by:
Matt Carlson <mcarlson@broadcom.com> Signed-off-by:
Michael Chan <mchan@broadcom.com> Signed-off-by:
-
Matt Carlson authored
This patch updates the copyright to 2009 and updates the version number to 3.98. Signed-off-by:
-
Matt Carlson authored
This patch adds version reporting for hardware selfboot. Signed-off-by:
Benjamin Li <benli@broadcom.com> Signed-off-by:
-
Matt Carlson authored
This patch adds code to report the DASH firmware version. Signed-off-by:
-
Matt Carlson authored
This patch adds code to obtain the bootcode version for versions of bootcode that do not use the new method. Signed-off-by:
-
Matt Carlson authored
This patch reorganizes the firmware version routines in preparation for the following patches. Signed-off-by:
-
Matt Carlson authored
The remaining uses of tg3_nvram_read_swab() either intended to read the data from NVRAM exactly as tg3_nvram_read_be32() did or hide deeper interpretations of the data. For the former case, a direct replacement of tg3_nvram_read_swab() with tg3_nvram_read_be32() is in order. For the latter case, we remove tg3_nvram_read_swab() and document what the code is really doing. Signed-off-by:
-
Matt Carlson authored
Any software requesting NVRAM data as it exists on NVRAM is necessarily requesting that the results be returned as a bytestream. A bytestream data read in from the device can also be thought to be in big endian format. Therefore, all the LE notations in the driver are mislabeled. This patch converts all LE notations to BE notations, carefully evaluating the surrounding code in the process. Signed-off-by:
-
Matt Carlson authored
This patch removes the blind byteswap of NVRAM data as it is read in. To preserve the logic at the call sites, this patch also inverts every call to tg3_nvram_read() and tg3_nvram_read_swab(). The call swap gets confusing within tg3_nvram_read_le() (LE is a misnomer), but the reader should be able to convince himself / herself that the resulting behavior is still unchanged. Signed-off-by:
-
Matt Carlson authored
This patch moves all NVRAM routines earlier in the source file to eliminate forward declarations. Signed-off-by:
-
Matt Carlson authored
tg3_vlan_rx_register() touches the hardware if netif_running() returns false. This patch fixes the problem. Signed-off-by:
-
Matt Carlson authored
The ethtool interface has acquired some new enumerations since the tg3 driver's tg3_set_settings() error checking code was written. The error checking code is no longer complete. This patch rewrites the error checking so that it is future-proofed. Signed-off-by:
-
Matt Carlson authored
The hardware's MTU is configured so that it will accept packets whose VLAN tag is left inline. This means that the hardware will also accept other packets larger than the desired MTU size. This patch checks for and discards packets that are larger than the requested MTU but are not VLAN tagged frames. Signed-off-by:
-
Matt Carlson authored
Commit 6833c043 introduced the phy auto-powerdown capability. While the APD feature only works for 5761 and 5784 asic revisions, the (harmless portion of the) code was applied to all 5705 and newer devices. However, the 5906 phy departs from the usual design. This commit was interfering with the 5906's ability to negotiate link against some switches. This patch corrects the problem. Signed-off-by:
-
Hannes Eder authored
Impact: Attribute functions with __acquires(...) resp. __releases(...). Fix this sparse warnings: net/wanrouter/wanproc.c:82:13: warning: context imbalance in 'r_start' - wrong count at exit net/wanrouter/wanproc.c:103:13: warning: context imbalance in 'r_stop' - unexpected unlock net/wanrouter/wanmain.c:765:13: warning: context imbalance in 'lock_adapter_irq' - wrong count at exit net/wanrouter/wanmain.c:771:13: warning: context imbalance in 'unlock_adapter_irq' - unexpected unlock Signed-off-by:
Hannes Eder <hannes@hanneseder.net> Signed-off-by:
-
Hannes Eder authored
Impact: Attribute function with __releases(...) Fix this sparse warning: net/ipv4/inet_fragment.c:276:35: warning: context imbalance in 'inet_frag_find' - unexpected unlock Signed-off-by:
-
Hannes Eder authored
Impact: Remove redundant variable declarations, resp. rename inner scope variable. Fix this sparse warnings: net/decnet/af_decnet.c:1252:40: warning: symbol 'skb' shadows an earlier one net/decnet/af_decnet.c:1223:24: originally declared here net/decnet/af_decnet.c:1582:29: warning: symbol 'val' shadows an earlier one net/decnet/af_decnet.c:1527:22: originally declared here net/decnet/dn_dev.c:687:21: warning: symbol 'err' shadows an earlier one net/decnet/dn_dev.c:670:13: originally declared here net/decnet/sysctl_net_decnet.c:182:21: warning: symbol 'len' shadows an earlier one net/decnet/sysctl_net_decnet.c:173:16: originally declared here Signed-off-by:
-
Hannes Eder authored
Impact: Attribute functions with __acquires(...) resp. __releases(...). Fix this sparse warnings: net/decnet/dn_dev.c:1324:13: warning: context imbalance in 'dn_dev_seq_start' - wrong count at exit net/decnet/dn_dev.c:1366:13: warning: context imbalance in 'dn_dev_seq_stop' - unexpected unlock Signed-off-by:
-
Hannes Eder authored
Impact: Include header file. Fix this sparse warning: net/core/sysctl_net_core.c:123:32: warning: symbol 'net_core_path' was not declared. Should it be static? Signed-off-by:
-
Hannes Eder authored
Impact: Use 'static const char[]' instead of 'static char[]', and since the data is const now it can be placed in __initconst. Fix this warning: net/appletalk/ddp.c: In function 'atalk_init': net/appletalk/ddp.c:1894: warning: format not a string literal and no format arguments Signed-off-by:
-
Hannes Eder authored
Impact: Trust in the comment and add '__force' to the cast. Fix this sparse warning: net/9p/trans_fd.c:420:34: warning: cast adds address space to expression (<asn:1>) Signed-off-by:
-
Hannes Eder authored
Impact: Attribute function with __acquires(...) resp. __releases(...). Fix this sparse warnings: net/802/tr.c:492:21: warning: context imbalance in 'rif_seq_start' - wrong count at exit net/802/tr.c:519:13: warning: context imbalance in 'rif_seq_stop' - unexpected unlock Signed-off-by:
-
Wei Yongjun authored
Remove some pointless conditionals before kfree_skb(). Signed-off-by:
Wei Yongjun <yjwei@cn.fujitsu.com> Signed-off-by:
-
