| Age | Commit message (Collapse) | Author |
|---|
|
192c4a80c broke the build on at least Debian/Hurd and Debian/kFreeBSD,
because it's not linux but still using glibc. To fix it, test for
__GLIBC__ in addition to __linux__.
|
|
This reverts commit 192c4a80c7d6fe9949aecb864901c4a3d9549f36.
This breaks on Debian/Hurd, Debian/kFreeBSD, and possibly other
platforms as well. Will be solved differently in a follow-up commit.
|
|
|
|
|
|
* maint:
ssh: clean test specs
|
|
|
|
* maint:
ssh: add econnaborted to disconnect msgs in test suite
ssh: fix the check that open-ssh supports certain pubkeys in a test suite
|
|
* hans/ssh/cuddle_tests:
ssh: add econnaborted to disconnect msgs in test suite
ssh: fix the check that open-ssh supports certain pubkeys in a test suite
|
|
* maint:
ssh: fix error (wrong suite) in test/ssh.spec
|
|
|
|
Each service process maintains a dictionary of peers, mapping an
application alias to a {pid(), #diameter_caps{}} list of connected
peers. These lists are potentially large, peers were appended to the end
of the list for no particular reason, and these long lists were
constructed/deconstructed when filtering them for pick_peer callbacks.
Many simultaneous outgoing request could then slow the VM to a crawl,
with many scheduled processes mired in list manipulation.
The pseudo-dicts are now replaced by plain ets tables. The reason for
them was (once upon a time) to have an interface interchangeable with a
plain dict for debugging purposes, but strict swapablity hasn't been the
case for some time now, and in practice a swap has never taken place.
Additional tables mapping Origin-Host/Realm have also been introduced,
to minimize the size of the peers lists when peers are filtered on
host/realm. For example, a filter like
{any, [{all, [realm, host]}, realm]}
is probably a very common case: preferring a Destination-Realm/Host
match before falling back on Destination-Realm alone. This is now more
efficiently (but not equivalently) expressed as
{first, [{all, [realm, host]}, realm]}
to stop the search when the best match is made, and extracts peers from
host/realm tables instead of searching through the list of all peers
supporting the application in question. The code to try and start with a
lookup isn't exhaustive, and the 'any' filter is still as inefficient as
previously.
|
|
See commit 862af31d.
|
|
|
|
OTP-13164 more efficient peer lists
One module. Downgrade not supported.
|
|
* maint:
ssh: benchmark all common kex and cipher algorithms
New structure of the report
ssh: ssh_benchmark_SUITE re-organized
ssh: Add first version of ssh_benchmark_SUITE
|
|
|
|
|
|
|
|
|
|
|
|
* anders/diameter/17.5.6.7/OTP-13211:
vsn -> 1.9.2.2
Update/fix appup for 17.5.6.7
Be resilient to diameter_service state upgrades
|
|
* anders/diameter/request_leak/OTP-13137:
Fix request table leak at retransmission
Fix request table leak at exit signal
|
|
* anders/diameter/17/watchdog/OTP-12969:
Fix watchdog function_clause
|
|
* anders/diameter/M-bit/OTP-12947:
Add service_opt() strict_mbit
|
|
|
|
OTP-12947 strict_mbit
OTP-12969 watchdog function_clause
OTP-13137 request leak
diameter_config (that allows the new option) should be loaded after the
others.
Anchor was missing from one regexp. Patches did not accumulate through
older versions.
|
|
By not failing in code that looks up state: pick_peer and service_info.
|
|
There is no reason for the socket on the erts side of the proxy to
accept connections from other hosts, so let's bind it to the loopback
interface.
Also change {ip, {127,0,0,1}} to {ip, loopback} for the erts side of
the socket for outgoing connections, to avoid hardcoding IPv4.
|
|
|
|
|
|
* bjorn/deprecate-random/OTP-12502:
test_server tests: Update test cases to cope with use of 'rand'
|
|
OTP-13207
* sverk/crypto/aes-ecb-192-bit:
crypto: Support 192-bit keys for AES ECB
|
|
|
|
OTP-13206
* sverk/crypto/aes-cbc-192-bit:
crypto: Support 192-bit keys for AES CBC
|
|
OTP-13205
* sverk/crypto/evp-aes-gcm:
crypto: Fix potential memory leak in error case for block cipher
crypto: Optimize AES-GCM cipher to not use dynamic allocation
Use EVP for AES-GCM
|
|
cover:compile_beam and cover:compile_beam_directory crashed when
trying to compile a beam file without a 'file' attribute. This has
been corrected, so an error is returned instead.
|
|
Functions in this module are never called, and some functions are
outdated. In order to avoid confusion (and need for update), the
module is now reduced to a simple dummy module.
|
|
In 80757f9, test_server was updated use the new 'rand' module instead
of 'random', but the the shuffle test cases were not updated.
|
|
* bjorn/erts/clean-up-preloaded/OTP-13112:
Update preloaded modules
erl_prim_loader doc: Remove mention of user supplied loader
erl_prim_loader: Clean up string_match()
erl_prim_loader: Avoid making absolute paths
erl_prim_loader: Clean up splitting of filenames
init: Eliminate the concat/1 function
Clean up start of erl_prim_loader
Remove erl_prim_loader:get_files/2
erl_prim_loader: Remove code for handling OSE
erl_prim_loader: Break loop/3 into two functions for readability
prim_file: Suppress a dialyzer warning
Reduce the ludicrous number of arguments for eval_script()
Clean up handling of boot_vars
init_SUITE: Correct the test for a "real system"
Remove useless 'catch' in start_it/1
Simplify get_flag() and get_flag_list()
Clean up parsing and lookup of flags
|
|
* bjorn/kernel/remove-code-path-cache/OTP-13191:
Remove the code path cache in the code server
|
|
The code generated by the HiPE compiler for pattern matching with
UTF binaries was such that sometimes THE_NON_VALUE was stored in
the roots followed by the garbage collector. This was not an issue
for the vanilla native code compiler, but was problematic for the
ErLLVM back-end.
Fix the issue by not storing THE_NON_VALUE in the live roots. An
alternative fix would be to change the code of the garbage collector.
With this fix, there are no more (known) failing test cases for the
ErLLVM back-end (at least on x86_64 with LLVM 3.5, which is the
configuation regularly tested). Thanks to @margnus1 for the fix.
|
|
|
|
|
|
|
|
|
|
|
|
- A test for proper handling of negative numbers in binary search
key tables
- A test for HiPE's ICode range analysis
|
|
|
|
|
|
Take out extraneous spaces at the end of lines.
|
