On 12/01/17 09:42, Ard Biesheuvel wrote:
> On 30 November 2017 at 19:00, Jordan Justen wrote:
>> As I recall, this enables a modest (non-essential) improvement in
>> fragmentation, at the cost of more ways to configure OVMF to hang
>> during boot. (With info available via debugcon.)
>>
>> I would
On 11/30/17 20:00, Jordan Justen wrote:
> As I recall, this enables a modest (non-essential) improvement in
> fragmentation,
My testing from 2 minutes ago shows that the UEFI memmap (as dumped by
the Linux guest) goes from 82 entries to 56 entries. I consider the
defragmentation a QoS question, an
On 30 November 2017 at 19:00, Jordan Justen wrote:
> As I recall, this enables a modest (non-essential) improvement in
> fragmentation, at the cost of more ways to configure OVMF to hang
> during boot. (With info available via debugcon.)
>
> I would prefer if were able to always fallback and conti
As I recall, this enables a modest (non-essential) improvement in
fragmentation, at the cost of more ways to configure OVMF to hang
during boot. (With info available via debugcon.)
I would prefer if were able to always fallback and continue boot.
Later, after GOP has started we could visibly warn
Repo: https://github.com/lersek/edk2.git
Branch: memmap_defrag_v2
This version is identical to the v1 seres I posted in March. Some
patches of the v1 series have been committed. I've been rebasing and
using the rest since, and am now posting it as v2. References:
* https://bugzilla.tianocore.or
5 matches
Mail list logo