summaryrefslogtreecommitdiff
path: root/src
AgeCommit message (Collapse)AuthorFilesLines
2023-06-03Remove "All rights reserved" from Oracle copyright noticesAlan Coopersmith5-5/+5
Oracle no longer includes this term in our copyright & license notices. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2022-12-11AddResponseToBuffer: Copy respose data as well to avoid use after scopeJeremy Huddleston Sequoia1-2/+6
==88383==ERROR: AddressSanitizer: stack-use-after-scope on address 0x00016f431d70 at pc 0x000100e08d14 bp 0x00016f429670 sp 0x00016f428e28 READ of size 32 at 0x00016f431d70 thread T0 #0 0x100e08d10 in __asan_memcpy+0x1a4 (libclang_rt.asan_osx_dynamic.dylib:arm64e+0x3cd10) #1 0x1009e2660 in HandleClientResponses server.c:372 #2 0x1009e2214 in HandleClientRequest server.c:685 #3 0x1009df314 in XhivRunServer server.c:715 #4 0x1009ded40 in XhivOpenServer server.c:822 #5 0x1009cfb3c in XhivOpenDisplay xlib_client.c:55 #6 0x1009cf614 in testOverflowFields XQueryFont.c:101 #7 0x1009ce2b0 in main XQueryFont.c:133 #8 0x197de3e4c (<unknown module>) Address 0x00016f431d70 is located in stack of thread T0 at offset 560 in frame #0 0x1009dfd44 in HandleClientRequest server.c:461 This frame has 11 object(s): [32, 36) 'req' (line 483) [48, 52) 'length' (line 485) [64, 96) 'extension' (line 534) [128, 184) 'default_qext_response' (line 551) [224, 256) 'getp_reply' (line 597) [288, 344) 'getp_response' (line 605) [384, 416) 'getif_reply' (line 619) [448, 504) 'getif_response' (line 625) [544, 548) 'newseq' (line 638) [560, 592) 'bigreq_reply' (line 653) <== Memory access at offset 560 is inside this variable [624, 680) 'bigreq_response' (line 658) HINT: this may be a false positive if your program uses some custom stack unwind mechanism, swapcontext or vfork (longjmp and C++ exceptions *are* supported) SUMMARY: AddressSanitizer: stack-use-after-scope (libclang_rt.asan_osx_dynamic.dylib:arm64e+0x3cd10) in __asan_memcpy+0x1a4 Fixes: https://gitlab.freedesktop.org/xorg/test/xhiv/-/issues/1 Signed-off-by: Jeremy Huddleston Sequoia <jeremyhu@apple.com>
2013-05-17Avoid double free of client.conn if HandleClientRequest detects disconnectAlan Coopersmith1-1/+3
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-05-10Use strsignal(3) which is more standard than sig2str()Matthieu Herrb1-3/+4
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-05-05Make query extension replies static, instead of pointers to stack buffersAlan Coopersmith1-5/+13
Make sure the data is still valid when its time to send to client Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-05-01Send SIGHUP to server at end of xcb testAlan Coopersmith1-0/+4
Since xcb doesn't actually shut down the socket when an error occurs, we'd hang until our SIGALRM goes off if we did create an xcb connection error in our test, so we make sure the server exits when desired. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-05-01Allow xcb tests to end with connection errorsAlan Coopersmith1-2/+4
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-04-28Allow tests to repeat a data packet a specified number of timesAlan Coopersmith1-4/+10
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-04-27Allow test programs to provide responses to selected QueryExtension callsAlan Coopersmith1-7/+21
Allows them to provide their own psuedo-implementations of Extensions. If the sequence in a response is set to XHIV_SEQ_MATCHDATA, then the new match_data field has a pointer to additional matching constraints. So far only X_QueryExtension will check those, and it expects a null terminated extension name string to match against the queried extension. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-04-27Install custom XIOError handler to abort on IO errorAlan Coopersmith1-0/+15
Allows debugging the core file to see what went wrong Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-03-30Allow requests to return a chain of multiple repliesAlan Coopersmith1-12/+24
Needed for XListFontsWithInfo test
2013-03-30Expose sequence number setting "extension" to test programsAlan Coopersmith4-20/+54
Allows setting specific sequeunce numbers for matching request patterns to avoid having to match exact sequences of calls Xlib may make behind the scenes to check for/use various extensions for you. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-03-30Add alarm() call to kill processes after 5 minutesAlan Coopersmith1-0/+4
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-03-30Switch xtrans from X11_t to XSERV_t to avoid conflicts with pre-xcb XlibAlan Coopersmith4-17/+38
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-03-30Allow building without xcb for testing on older systemsAlan Coopersmith1-1/+4
Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
2013-03-30Create initial frameworkAlan Coopersmith6-0/+1016
This provides a local static libXhiv.a library that forks and runs a fake X server process that simply responds with a canned set of replies, and which test cases can be written that check to see how libraries handle those replies. Signed-off-by: Alan Coopersmith <alan.coopersmith@oracle.com>
generated by cgit v1.2.3 (git 2.39.1) at 2024-11-21 13:56:39 +0000