linux-kernel - Re: [PATCH 3/6] KVM: selftests: memslot_perf

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Message-ID: <b720dbad-4b8b-a617-f782-7f95bcdb3d54@redhat.com>
Date:   Tue, 18 Oct 2022 06:18:39 +0800
From:   Gavin Shan <gshan@...hat.com>
To:     "Maciej S. Szmigiero" <mail@...iej.szmigiero.name>
Cc:     kvmarm@...ts.cs.columbia.edu, kvmarm@...ts.linux.dev,
        kvm@...r.kernel.org, linux-kernel@...r.kernel.org,
        ajones@...tanamicro.com, pbonzini@...hat.com, maz@...nel.org,
        shuah@...nel.org, oliver.upton@...ux.dev, seanjc@...gle.com,
        peterx@...hat.com, ricarkol@...gle.com, zhenyzha@...hat.com,
        shan.gavin@...il.com
Subject: Re: [PATCH 3/6] KVM: selftests: memslot_perf_test: Probe memory slots
 for once

On 10/18/22 1:34 AM, Maciej S. Szmigiero wrote:
> On 14.10.2022 09:19, Gavin Shan wrote:
>> prepare_vm() is called in every iteration and run. The allowed memory
>> slots (KVM_CAP_NR_MEMSLOTS) are probed for multiple times. It's not
>> free and unnecessary.
>>
>> Move the probing logic for the allowed memory slots to parse_args()
>> for once, which is upper layer of prepare_vm().
>>
>> No functional change intended.
>>
>> Signed-off-by: Gavin Shan <gshan@...hat.com>
>> ---
>>   .../testing/selftests/kvm/memslot_perf_test.c | 29 ++++++++++---------
>>   1 file changed, 16 insertions(+), 13 deletions(-)
>>
>> diff --git a/tools/testing/selftests/kvm/memslot_perf_test.c b/tools/testing/selftests/kvm/memslot_perf_test.c
>> index dcb492b3f27b..d5aa9148f96f 100644
>> --- a/tools/testing/selftests/kvm/memslot_perf_test.c
>> +++ b/tools/testing/selftests/kvm/memslot_perf_test.c
>> @@ -245,27 +245,17 @@ static bool prepare_vm(struct vm_data *data, int nslots, uint64_t *maxslots,
>>                  void *guest_code, uint64_t mempages,
>>                  struct timespec *slot_runtime)
>>   {
>> -    uint32_t max_mem_slots;
>>       uint64_t rempages;
>>       uint64_t guest_addr;
>>       uint32_t slot;
>>       struct timespec tstart;
>>       struct sync_area *sync;
>> -    max_mem_slots = kvm_check_cap(KVM_CAP_NR_MEMSLOTS);
>> -    TEST_ASSERT(max_mem_slots > 1,
>> -            "KVM_CAP_NR_MEMSLOTS should be greater than 1");
>> -    TEST_ASSERT(nslots > 1 || nslots == -1,
>> -            "Slot count cap should be greater than 1");
>> -    if (nslots != -1)
>> -        max_mem_slots = min(max_mem_slots, (uint32_t)nslots);
>> -    pr_info_v("Allowed number of memory slots: %"PRIu32"\n", max_mem_slots);
>> -
>>       TEST_ASSERT(mempages > 1,
>>               "Can't test without any memory");
>>       data->npages = mempages;
>> -    data->nslots = max_mem_slots - 1;
>> +    data->nslots = nslots;
>>       data->pages_per_slot = mempages / data->nslots;
>>       if (!data->pages_per_slot) {
>>           *maxslots = mempages + 1;
>> @@ -885,8 +875,8 @@ static bool parse_args(int argc, char *argv[],
>>               break;
>>           case 's':
>>               targs->nslots = atoi(optarg);
>> -            if (targs->nslots <= 0 && targs->nslots != -1) {
>> -                pr_info("Slot count cap has to be positive or -1 for no cap\n");
>> +            if (targs->nslots <= 1 && targs->nslots != -1) {
>> +                pr_info("Slot count cap must be larger than 1 or -1 for no cap\n");
>>                   return false;
>>               }
>>               break;
>> @@ -932,6 +922,19 @@ static bool parse_args(int argc, char *argv[],
>>           return false;
>>       }
>> +    /* Memory slot 0 is reserved */
>> +    if (targs->nslots == -1) {
>> +        targs->nslots = kvm_check_cap(KVM_CAP_NR_MEMSLOTS) - 1;
>> +        if (targs->nslots < 1) {
>> +            pr_info("KVM_CAP_NR_MEMSLOTS should be greater than 1\n");
>> +            return false;
>> +        }
>> +    } else {
>> +        targs->nslots--;
>> +    }
>> +
>> +    pr_info_v("Number of memory slots: %d\n", targs->nslots);
>> +
> 
> Can't see any capping of the command line provided slot count to
> KVM_CAP_NR_MEMSLOTS value, like the old code did.
> 

Indeed. I wanted to avoid extra variable @max_mem_slots and the
capping is missed. I will fix it up in next revision.

>>       return true;
>>   }

Thanks,
Gavin