diff options
| author | Christian Krinitsin <mail@krinitsin.com> | 2025-06-30 12:34:26 +0000 |
|---|---|---|
| committer | Christian Krinitsin <mail@krinitsin.com> | 2025-06-30 12:35:44 +0000 |
| commit | 25f8033d556aa17afaea4a5196ea7a69fe248320 (patch) | |
| tree | 0f056db167683be54ea1e5e72d29d6069af55e7d /results/classifier/deepseek-2-tmp/output/performance/1070 | |
| parent | 8e6da29e4ee5fc14bc1cc816a24f21271f14090d (diff) | |
| download | qemu-analysis-25f8033d556aa17afaea4a5196ea7a69fe248320.tar.gz qemu-analysis-25f8033d556aa17afaea4a5196ea7a69fe248320.zip | |
add new temporary deepseek-r1:14b results
Diffstat (limited to 'results/classifier/deepseek-2-tmp/output/performance/1070')
| -rw-r--r-- | results/classifier/deepseek-2-tmp/output/performance/1070 | 11 |
1 files changed, 11 insertions, 0 deletions
diff --git a/results/classifier/deepseek-2-tmp/output/performance/1070 b/results/classifier/deepseek-2-tmp/output/performance/1070 new file mode 100644 index 000000000..db6f4fda8 --- /dev/null +++ b/results/classifier/deepseek-2-tmp/output/performance/1070 @@ -0,0 +1,11 @@ + +gdbstub XML generation for ARM is done for every CPU +Description of problem: +- As arm_cpu_register_gdb_regs_for_features is called from the device + realize stage for each vCPU in user mode we end up uselessly + regenerating the XML for every new thread. Once you get up to 100 + threads this starts exceeding the large maps done for QHT and PageDesc +Steps to reproduce: +See above command line, valgrind picks it up +Additional information: +See also #866, #967 |