Reserve allocation should be displayed when erroring due to lack of memory on startup #11282
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
abellina
reviewed
Jul 31, 2024
| s"the minimum allocation of ${toMB(minAllocation)} (calculated from " + | ||
| s"${RapidsConf.RMM_ALLOC_MIN_FRACTION} (=${conf.rmmAllocMinFraction}) " + | ||
| s"and ${toMB(info.total)} MB total memory)") | ||
| s"${toMB(poolAllocation)} MiB (gpu.free: ${toMB(info.free)}," + |
There was a problem hiding this comment.
should we rename toMB to toMiB (nit pick)
There was a problem hiding this comment.
I noticed but left it out to be conservative. Will update
| s"${RapidsConf.RMM_ALLOC_FRACTION}: (=${conf.rmmAllocFraction}," + | ||
| s"${RapidsConf.RMM_ALLOC_RESERVE}: ${reserveAmount} => " + | ||
| s"(gpu.free - reserve) * allocFraction = ${toMB(poolAllocation)})" + | ||
| s"was less than allocation of ${toMB(minAllocation)} MiB (gpu.total: " + |
There was a problem hiding this comment.
I think we can bring this message into two: the part before "was less than" and the part after. That way we can reuse it in the second case (it is almost verbatim except for this text, as far as I can tell).
|
@abellina I think this is pretty close, whenever you have a chance to take a look. Thank u! |
|
build |
abellina
reviewed
Oct 9, 2024
| s"less than allocation of ${toMiB(minAllocation)} MiB (gpu.total: " + | ||
| s"${toMiB(info.total)} MiB, ${RapidsConf.RMM_ALLOC_MIN_FRACTION}: " + | ||
| s"${conf.rmmAllocMinFraction} => gpu.total *" + | ||
| s"minAllocFraction = ${toMiB(minAllocation)} MiB). Please ensure that the GPU has" + |
There was a problem hiding this comment.
Suggested change
| s"minAllocFraction = ${toMiB(minAllocation)} MiB). Please ensure that the GPU has" + | |
| s"minAllocFraction = ${toMiB(minAllocation)} MiB). Please ensure that the GPU has " + |
abellina
reviewed
Oct 9, 2024
sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuDeviceManager.scala
Outdated
Show resolved
Hide resolved
abellina
reviewed
Oct 9, 2024
sql-plugin/src/main/scala/com/nvidia/spark/rapids/GpuDeviceManager.scala
Outdated
Show resolved
Hide resolved
abellina
reviewed
Oct 9, 2024
revans2
previously approved these changes
Oct 10, 2024
…emory on startup Signed-off-by: Kuhu Shukla <kuhus@nvidia.com>
Signed-off-by: Kuhu Shukla <kuhus@nvidia.com>
kuhushukla
force-pushed
the
issue_11168
branch
from
0f8aa7e
to
2e41c75
Compare
Signed-off-by: Kuhu Shukla <kuhus@nvidia.com>
kuhushukla
force-pushed
the
issue_11168
branch
from
2e41c75
to
9e84b60
Compare
abellina
approved these changes
Oct 31, 2024
|
build |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Fixes #11168
I think using
is the right decision here but I would like for someone to correct or confirm.
I don't have a test for this. Can look into it if preferred for the review.