summary refs log tree commit diff stats
path: root/results/classifier/118/performance/1473451
diff options
context:
space:
mode:
Diffstat (limited to 'results/classifier/118/performance/1473451')
-rw-r--r--results/classifier/118/performance/147345149
1 files changed, 49 insertions, 0 deletions
diff --git a/results/classifier/118/performance/1473451 b/results/classifier/118/performance/1473451
new file mode 100644
index 000000000..a275469c4
--- /dev/null
+++ b/results/classifier/118/performance/1473451
@@ -0,0 +1,49 @@
+performance: 0.936
+mistranslation: 0.906
+semantic: 0.764
+virtual: 0.739
+files: 0.715
+graphic: 0.692
+user-level: 0.684
+device: 0.682
+architecture: 0.677
+network: 0.584
+ppc: 0.545
+permissions: 0.544
+kernel: 0.519
+vnc: 0.463
+arm: 0.462
+boot: 0.433
+hypervisor: 0.419
+peripherals: 0.414
+register: 0.408
+i386: 0.401
+debug: 0.399
+socket: 0.394
+risc-v: 0.392
+x86: 0.366
+PID: 0.362
+VMM: 0.315
+assembly: 0.300
+TCG: 0.217
+KVM: 0.199
+
+Please support the native bios format for dec alpha
+
+Currently qemu-system-alpha -bios parameter takes an ELF image.
+However HP maintains firmware updates for those systems.
+
+Some example rom files can be found here ftp://ftp.hp.com/pub/alphaserver/firmware/current_platforms/v7.3_release/DS20_DS20e/
+
+It might allow things like using the SRM firmware.
+The ARC(nt) firmware would allow to build and test windows applications for that platforms without having the relevant hardware
+
+QEMU does not really implement a "true" ev67.
+
+We cheat and implement something that is significantly faster to emulate.
+E.g. doing all TLB refill within qemu, rather than in the PALcode.
+
+So, no, there's no chance of running true SRM or ARC firmware.
+
+But In that case it’s impossible to emulate or even compile Windows for Dec Alpha.
+