Con Kolivas
074b8193b2
Flush output.
2011-08-16 11:21:26 +10:00
Con Kolivas
53a4245a43
Log summary at the end as per any other output.
2011-08-16 11:15:04 +10:00
Con Kolivas
939a0e02b7
Only show cpu algo in summary if cpu mining.
2011-08-16 10:46:25 +10:00
Con Kolivas
cdece1f248
Simplify code and tidy output.
2011-08-16 10:45:02 +10:00
Con Kolivas
8e38a149d8
Simplify output code and remove HW errors from CPU stats.
2011-08-16 10:37:34 +10:00
Con Kolivas
426f72dec7
As we have keepalives now, blaming network flakiness on timeouts appears to have been wrong.
...
Set a timeout for longpoll to 1 hour, and most other network connectivity to 1 minute.
2011-08-16 10:27:19 +10:00
Con Kolivas
c53a7b0d2d
Only display the CPU algo when we're CPU mining.
2011-08-16 10:26:42 +10:00
Con Kolivas
657812ada4
Probe for slightly longer for when network conditions are lagging.
2011-08-16 10:15:47 +10:00
Con Kolivas
8febf57af0
Keep options in alphabetical order.
2011-08-16 09:45:16 +10:00
Con Kolivas
845af9113e
Add --submit-stale option which submits all shares, regardless of whether they would normally be considered stale.
2011-08-16 09:42:26 +10:00
Znort 987
eed634a3ab
Merge branch 'ckolivas-master' into merge-ckolivas-master
2011-08-16 00:04:50 +02:00
Con Kolivas
4e48561a13
Rework the last-initialised time displayed.
2011-08-15 23:23:53 +10:00
Con Kolivas
cfe8534c39
Partial restart of threads has proven to be unsuccessful so reinstate device re-initialisation to restart GPUs.
...
Do this by having a reinit thread that is told via a queue which device to restart.
If this thread fails to return from opencl code, it should not interrupt the workings of other devices.
2011-08-15 23:23:52 +10:00
Con Kolivas
42d49ffdc7
Revert "Restart threads by abstracting out the clcontext initialisation and using that instead of probing all cards."
...
This reverts commit 8f186e61e2
.
2011-08-15 23:23:45 +10:00
Con Kolivas
cf543507c6
Revert "Preinitialise the devices only once on startup."
...
This reverts commit 071a0ad2f1
.
2011-08-15 23:23:39 +10:00
Con Kolivas
b1289a0159
Revert "Move the non cl_ variables into the cgpu info struct to allow creating a new cl state on reinit, preserving known GPU variables."
...
This reverts commit 28880d0dc7
.
2011-08-15 23:23:29 +10:00
Con Kolivas
7c50bb02ab
Revert "Display last initialised time in gpu management info."
...
This reverts commit 177e07aed3
.
2011-08-15 23:23:20 +10:00
Con Kolivas
cac54f30b8
Revert "When pinging a sick cpu, flush finish and then ping it in a separate thread in the hope it recovers, but without blocking code elsewhere."
...
This reverts commit a466942fd8
.
2011-08-15 23:23:09 +10:00
Znort 987
92af1925a6
Remove benchmark data from main.c
2011-08-15 15:08:13 +02:00
Znort 987
8256402a63
Remove harmless warning on win32
2011-08-15 14:58:36 +02:00
Znort 987
593022dfd7
Added somme comments
2011-08-15 14:00:09 +02:00
Znort 987
e923ac78d5
Add CPU algo used to summary
2011-08-15 13:59:44 +02:00
Znort 987
4e3f38d047
Re-indent code for fork_monitor
2011-08-15 13:59:07 +02:00
Znort 987
39bdec8bbe
Various minor changes
...
- Fixed a type bug on 32bit platform in algo bench
- Some re-indenting
2011-08-15 13:58:30 +02:00
Znort 987
75639da1e3
Add WIN32 safe CPU benchmark method
...
Method for WIN32
- Add a hidden '--bench-algo' option to cgminer that only
benchmarks the specified algorithm.
- Create a chunk of shared mem
- Launch a new instance of cgminer with --bench-algo, and
act as a debugger for it to silently catch crashes.
- Once it completes, either successfully or because of a
crash, read benchmarked value from shared mem
2011-08-15 13:55:00 +02:00
Znort 987
84b0467cb2
Fix broken assert in algo bencher
2011-08-15 13:53:15 +02:00
Znort 987
6fd06387b3
On unix, make algo benchmark more generic
2011-08-15 13:51:48 +02:00
Znort 987
01886493cf
Add set_non_blocking routing
...
In preparation for upcoming generic unix automated
algorithm benchmark method (as opposed to linux specific),
add a utility method.
2011-08-15 13:49:26 +02:00
Znort 987
29f0f9bc18
Properly detect number of cores on WIN32
2011-08-15 13:48:17 +02:00
Znort 987
7158f9f9ed
Add automated benchmark of the CPU hashers
...
The --algo switch now accepts the "auto" argument.
When "auto" is passed to --algo, cgminer starts by benchmarking
all the CPU algorithms it nows about and picks the fastest.
This is useful for benchmarking, but also for folks who run
cgminer on a large number of heterogeneous computers because
it saves them from having to configure each instance optimally.
Caveat emptor: depending on the platform, some algorithms will
fail with "illegal instruction" (e.g. via padlock code on non via
platforms, or SSE4 code on non SSE4 platforms).
To protect against this, cgminer runs the benchmarks in a child
process. The crash, if any occurs in the child, and the parent
marks this algo as "fails" and continues benchmarking the next
algorithm.
Unfortunately, this safeguard is only implemented for Linux
because I have no clue how to do this on win32 (nor access to
a win32 box).
2011-08-15 13:47:50 +02:00
Con Kolivas
b73d93cd29
Bump version number.
2011-08-14 14:20:55 +10:00
Con Kolivas
1cca876b7a
Update NEWS.
2011-08-14 14:19:53 +10:00
Con Kolivas
50504ef9e0
If a work item has had any mining done on it, don't consider it discarded work.
2011-08-14 13:57:58 +10:00
Con Kolivas
656067a44e
Stale work test should respect age of rolled work as well.
2011-08-14 12:30:50 +10:00
Con Kolivas
232408050c
Stick to rolling only work from the current pool unless we're in load balance mode or lagging to avoid aggressive rolling imitating load balancing.
2011-08-14 11:35:53 +10:00
Con Kolivas
01aadda25c
Simply set the time of day when the work is decoded.
2011-08-14 11:24:38 +10:00
Con Kolivas
639d38fbd4
Revert "Make sure to give work items a starting time only once when they're staged."
...
This reverts commit 626ae10d7c
.
Failed to address all work items.
2011-08-14 07:54:45 +10:00
Con Kolivas
626ae10d7c
Make sure to give work items a starting time only once when they're staged.
2011-08-14 02:33:55 +10:00
Con Kolivas
d6dd5d700b
Test each work item to see if it can be rolled instead of per-pool and roll whenever possible, adhering to the 60 second timeout.
2011-08-14 01:54:47 +10:00
Con Kolivas
95f878294f
The extra shift in the output actually appears detrimental in cgminer and there is a miniscule chance of missing the actual result if it ends up in the same spot as MAXBUFFERS.
2011-08-13 23:59:58 +10:00
Con Kolivas
e9b8344a37
Free all work in one place allowing us to perform actions on it in the future.
2011-08-13 20:56:41 +10:00
Con Kolivas
6da13fb87f
Only put the message that a pool is down if not pinging it every minute.
2011-08-13 20:56:34 +10:00
Con Kolivas
510be5c9a4
Summary on exit is sometimes getting lost without a flush of stdout.
2011-08-13 20:56:16 +10:00
Con Kolivas
42c22025d8
The can_roll function fails inappropriately in stale_work.
2011-08-13 20:56:09 +10:00
Con Kolivas
d39e5abfba
We can't hit this path since the earlier check for stale work will prevent us ever finding work it can roll here.
2011-08-13 20:56:04 +10:00
Con Kolivas
17e5e17e88
Only consider a pool lagging if we actually need the work and we have none staged despite queue requests stacking up.
2011-08-13 20:55:53 +10:00
Con Kolivas
a466942fd8
When pinging a sick cpu, flush finish and then ping it in a separate thread in the hope it recovers, but without blocking code elsewhere.
2011-08-13 20:55:43 +10:00
Con Kolivas
177e07aed3
Display last initialised time in gpu management info.
2011-08-13 20:55:30 +10:00
Con Kolivas
28880d0dc7
Move the non cl_ variables into the cgpu info struct to allow creating a new cl state on reinit, preserving known GPU variables.
...
Create a new context from scratch in initCQ in case something was corrupted to maximise our chance of succesfully creating a new worker thread.
2011-08-13 20:54:20 +10:00
Con Kolivas
071a0ad2f1
Preinitialise the devices only once on startup.
2011-08-13 20:50:00 +10:00