Conversation
sinceforYy
requested review from
Anzooooo,
cebarobot and
good-circle
as code owners
|
Is there any context about these changes? I don't get the reason. |
|
plz see this issue |
|
Alright, regardless, perhaps from now on we should follow OpenXiangShan is example and standardize the descriptions for NEMU PR to facilitate better review by reviewers. |
|
Alright, I've added a detailed description of the PR. |
cebarobot
left a comment
cebarobot
left a comment
There was a problem hiding this comment.
What's the different between these pairs?
vmode_get_sieandget_vsievmode_set_sieandset_vsievmode_get_sipandget_vsipvmode_set_sipandset_vsip
I have diffed these pairs and found that the key difference seems to be about hideleg and hedeleg?
Also, I don't get the reason that causes this bug. Is it a misunderstanding of the RISC-V priv spec?
By the way, please rebase to the latest master and remove the PMEM modification. This has been fixed before.
|
Yes, this bug is caused by a misunderstanding of the RISC-V priv spec. Additionally, regarding PMEM, I will rebase it. |
sinceforYy
force-pushed
the
fix-csr-xie-xip
branch
from
cc40209 to
55295c4
Compare
4 participants
Guest read/write operations on sie/sip in VS mode are functionally equivalent to host read/write operations on vsie/vsip in M mode.
vmode_set_sip and set_vsip, as well as vmode_set_sie and set_vsie, are not entirely equivalent. Therefore, replace them with the correct set_vsip and set_vsie methods.