- 18 Oct, 2019 17 commits
-
-
Federico Vaga authored
-
Federico Vaga authored
Signed-off-by:Federico Vaga <federico.vaga@cern.ch>
-
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Thomas Oulevey authored
Signed-off-by:
Thomas Oulevey <thomas.oulevey@cern.ch> Signed-off-by:
-
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by:
-
- 16 Oct, 2019 4 commits
-
-
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Redirect everything to software directory Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by:
-
- 14 Oct, 2019 8 commits
-
-
Federico Vaga authored
-
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
The main reason it prevent a successful calloc that does not allocate anything which results in: 'errno = 0' and 'return -1' (which is confusing). Providing invalid inputs is not acceptable Signed-off-by: -
Federico Vaga authored
BUG reported on gcc-9, the buffer path[] may not be enough to contain the string that we want to generate from sprintf(). This patch allocates properly the buffer Reported-by:
Dimitris Lampridis <dimitris.lampridis@cern.ch>
Signed-off-by:
-
- 13 Sep, 2019 4 commits
-
-
Federico Vaga authored
-
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
the fd_hmq_sync file descriptor was never closed, which is not a problem when also the process that how the FD dies, but it is a problem when the process does ``trtl_close()`` and then ``trtl_open()`` again many times. Reported-by:
Dimitris Lampridis <dimitris.lampridis@cern.ch>
Signed-off-by: -
Federico Vaga authored
The funciton ``trtl_close()`` was closing all possible file descriptors based on the maximum number of CPUs and HMQs configurable on Mock-Turtle. Consequentially ``trtl_close()`` may close random file descriptors because fd_cpu and fd_hmq may contain random value for unused entries. Instead, the ``trtl_close()`` should close **only** those devices really existing according to the configuration ROM. Reported-by:
Dimitris Lampridis <dimitris.lampridis@cern.ch>
Signed-off-by:
-
- 14 May, 2019 2 commits
-
-
Federico Vaga authored
-
Federico Vaga authored
Signed-off-by:
-
- 10 May, 2019 5 commits
-
-
Federico Vaga authored
-
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by: -
Federico Vaga authored
Signed-off-by:
-
