From: Brijesh Singh <brijesh.singh@amd.com>
To: Borislav Petkov <bp@suse.de>
Cc: brijesh.singh@amd.com, "Paolo Bonzini" <pbonzini@redhat.com>,
"Radim Krčmář" <rkrcmar@redhat.com>,
"Herbert Xu" <herbert@gondor.apana.org.au>,
"Gary Hook" <gary.hook@amd.com>,
"Tom Lendacky" <thomas.lendacky@amd.com>,
linux-crypto@vger.kernel.org, kvm@vger.kernel.org,
linux-kernel@vger.kernel.org
Subject: Re: [Part2 PATCH v5.1 12.7/31] crypto: ccp: Implement SEV_PEK_CSR ioctl command
Date: Thu, 12 Oct 2017 23:13:44 -0500 [thread overview]
Message-ID: <df846578-247a-978f-703e-ed6a6d23ee01@amd.com> (raw)
In-Reply-To: <d83b3e27-1195-817e-e976-36982e275588@amd.com>
On 10/12/17 9:24 PM, Brijesh Singh wrote:
>
> On 10/12/17 2:53 PM, Borislav Petkov wrote:
> ...
>
>> Ok, a couple of things here:
>>
>> * Move the checks first and the allocations second so that you allocate
>> memory only after all checks have been passed and you don't allocate
>> pointlessly.
>
> I assume you mean performing the SEV state check before allocating the
> memory for the CSR blob, right ? In my patches, I typically perform all
> the SW specific checks and allocation before invoking the HW routines.
> Handling the PSP commands will take longer compare to kmalloc() or
> access_ok() etc. If its not a big deal then I would prefer to keep that
> way.
>
>
>> * That:
>>
>> if (state == SEV_STATE_WORKING) {
>> ret = -EBUSY;
>> goto e_free_blob;
>> } else if (state == SEV_STATE_UNINIT) {
>> ret = sev_firmware_init(&argp->error);
>> if (ret)
>> goto e_free_blob;
>> do_shutdown = 1;
>> }
>>
>> is a repeating pattern. Perhaps it should be called
>> sev_firmware_reinit() and called by other functions.
>
>> * The rest is simplifications and streamlining.
>>
>> ---
>> diff --git a/drivers/crypto/ccp/psp-dev.c b/drivers/crypto/ccp/psp-dev.c
>> index e3ee68afd068..d41f5448a25b 100644
>> --- a/drivers/crypto/ccp/psp-dev.c
>> +++ b/drivers/crypto/ccp/psp-dev.c
>> @@ -302,33 +302,30 @@ static int sev_ioctl_pek_csr(struct sev_issue_cmd *argp)
>> int ret, state;
>> void *blob;
>>
>> - if (copy_from_user(&input, (void __user *)(uintptr_t)argp->data,
>> - sizeof(struct sev_user_data_pek_csr)))
>> + if (copy_from_user(&input, (void __user *)argp->data, sizeof(input)))
>> + return -EFAULT;
>> +
>> + if (!input.address)
>> + return -EINVAL;
>> +
As per the spec, its perfectly legal to pass input.address = 0x0 and
input.length = 0x0. If userspace wants to query CSR length then it will
fill all the fields with. In response, FW will return error
(LENGTH_TO_SMALL) and input.length will be filled with the expected
length. Several command work very similar (e.g PEK_CSR,
PEK_CERT_EXPORT). A typical usage from userspace will be:
- query the length of the blob (call command with all fields set to zero)
- SEV FW will response with expected length
- userspace allocate the blob and retries the command.
>> + /* allocate a physically contiguous buffer to store the CSR blob */
>> + if (!access_ok(VERIFY_WRITE, input.address, input.length) ||
>> + input.length > SEV_FW_BLOB_MAX_SIZE)
>> return -EFAULT;
>>
>> data = kzalloc(sizeof(*data), GFP_KERNEL);
>> if (!data)
>> return -ENOMEM;
>>
>> - /* allocate a temporary physical contigous buffer to store the CSR blob */
>> - blob = NULL;
>> - if (input.address) {
>> - if (!access_ok(VERIFY_WRITE, input.address, input.length) ||
>> - input.length > SEV_FW_BLOB_MAX_SIZE) {
>> - ret = -EFAULT;
>> - goto e_free;
>> - }
>> -
>> - blob = kmalloc(input.length, GFP_KERNEL);
>> - if (!blob) {
>> - ret = -ENOMEM;
>> - goto e_free;
>> - }
>> -
>> - data->address = __psp_pa(blob);
>> - data->len = input.length;
>> + blob = kmalloc(input.length, GFP_KERNEL);
>> + if (!blob) {
>> + ret = -ENOMEM;
>> + goto e_free;
>> }
>>
>> + data->address = __psp_pa(blob);
>> + data->len = input.length;
>> +
>> ret = sev_platform_get_state(&state, &argp->error);
>> if (ret)
>> goto e_free_blob;
>> @@ -349,25 +346,23 @@ static int sev_ioctl_pek_csr(struct sev_issue_cmd *argp)
>> do_shutdown = 1;
>> }
>>
>> - ret = sev_handle_cmd(SEV_CMD_PEK_CSR, data, &argp->error);
>> + ret = sev_do_cmd(SEV_CMD_PEK_CSR, data, &argp->error);
>>
>> input.length = data->len;
>>
>> /* copy blob to userspace */
>> - if (blob &&
>> - copy_to_user((void __user *)(uintptr_t)input.address,
>> - blob, input.length)) {
>> + if (copy_to_user((void __user *)input.address, blob, input.length)) {
>> ret = -EFAULT;
>> goto e_shutdown;
>> }
>>
>> - if (copy_to_user((void __user *)(uintptr_t)argp->data, &input,
>> - sizeof(struct sev_user_data_pek_csr)))
>> + if (copy_to_user((void __user *)argp->data, &input, sizeof(input)))
>> ret = -EFAULT;
>>
>> e_shutdown:
>> if (do_shutdown)
>> - sev_handle_cmd(SEV_CMD_SHUTDOWN, 0, NULL);
>> + ret = sev_do_cmd(SEV_CMD_SHUTDOWN, 0, NULL);
>> +
>> e_free_blob:
>> kfree(blob);
>> e_free:
>> @@ -408,10 +403,10 @@ static long sev_ioctl(struct file *file, unsigned int ioctl, unsigned long arg)
>> ret = sev_ioctl_pdh_gen(&input);
>> break;
>>
>> - case SEV_PEK_CSR: {
>> + case SEV_PEK_CSR:
>> ret = sev_ioctl_pek_csr(&input);
>> break;
>> - }
>> +
>> default:
>> ret = -EINVAL;
>> goto out;
>>
next prev parent reply other threads:[~2017-10-13 4:13 UTC|newest]
Thread overview: 74+ messages / expand[flat|nested] mbox.gz Atom feed top
2017-10-04 13:13 [Part2 PATCH v5 00/31] x86: Secure Encrypted Virtualization (AMD) Brijesh Singh
2017-10-04 13:13 ` [Part2 PATCH v5 09/31] crypto: ccp: Build the AMD secure processor driver only with AMD CPU support Brijesh Singh
2017-10-04 21:47 ` Borislav Petkov
2017-10-04 23:06 ` Brijesh Singh
2017-10-04 13:13 ` [Part2 PATCH v5 10/31] crypto: ccp: Add Platform Security Processor (PSP) device support Brijesh Singh
2017-10-05 9:56 ` Borislav Petkov
2017-10-06 23:09 ` [Part2 PATCH v5.1 " Brijesh Singh
2017-10-04 13:13 ` [Part2 PATCH v5 11/31] crypto: ccp: Define SEV key management command id Brijesh Singh
2017-10-05 20:56 ` Borislav Petkov
2017-10-08 21:14 ` Brijesh Singh
2017-10-04 13:13 ` [Part2 PATCH v5 12/31] crypto: ccp: Add Secure Encrypted Virtualization (SEV) command support Brijesh Singh
2017-10-06 18:49 ` Borislav Petkov
2017-10-06 19:48 ` Brijesh Singh
2017-10-07 18:13 ` Brijesh Singh
2017-10-07 1:05 ` [Part2 PATCH v5.1 12.1/31] " Brijesh Singh
2017-10-07 1:06 ` [Part2 PATCH v5.1 12.2/31] crypto: ccp: Define SEV userspace ioctl and command id Brijesh Singh
2017-10-07 14:20 ` Borislav Petkov
2017-10-08 21:18 ` Brijesh Singh
2017-10-11 16:46 ` [Part2 PATCH v5.2 12.1/31] " Brijesh Singh
2017-10-12 13:27 ` Borislav Petkov
2017-10-12 14:18 ` Brijesh Singh
2017-10-07 1:06 ` [Part2 PATCH v5.1 12.3/31] crypto: ccp: Implement SEV_FACTORY_RESET ioctl command Brijesh Singh
2017-10-11 14:32 ` Borislav Petkov
2017-10-11 16:55 ` [Part2 PATCH v5.2 " Brijesh Singh
2017-10-12 14:13 ` Borislav Petkov
2017-10-07 1:06 ` [Part2 PATCH v5.1 12.4/31] crypto: ccp: Implement SEV_PLATFORM_STATUS " Brijesh Singh
2017-10-11 17:01 ` [Part2 PATCH v5.2 " Brijesh Singh
2017-10-11 17:02 ` [Part2 PATCH v5.1 " Borislav Petkov
2017-10-11 19:49 ` Brijesh Singh
2017-10-11 20:04 ` Borislav Petkov
2017-10-11 20:10 ` Borislav Petkov
2017-10-11 20:10 ` Brijesh Singh
2017-10-11 20:28 ` Borislav Petkov
2017-10-11 20:45 ` Brijesh Singh
2017-10-11 20:53 ` Brijesh Singh
2017-10-11 20:54 ` Borislav Petkov
2017-10-07 1:06 ` [Part2 PATCH v5.1 12.5/31] crypto: ccp: Implement SEV_PEK_GEN " Brijesh Singh
2017-10-12 18:28 ` Borislav Petkov
2017-10-12 20:11 ` Brijesh Singh
2017-10-12 20:21 ` Borislav Petkov
2017-10-12 20:34 ` Brijesh Singh
2017-10-07 1:06 ` [Part2 PATCH v5.1 12.6/31] crypto: ccp: Implement SEV_PDH_GEN " Brijesh Singh
2017-10-12 18:48 ` Borislav Petkov
2017-10-12 20:21 ` Brijesh Singh
2017-10-12 20:23 ` Borislav Petkov
2017-10-07 1:06 ` [Part2 PATCH v5.1 12.7/31] crypto: ccp: Implement SEV_PEK_CSR " Brijesh Singh
2017-10-12 19:53 ` Borislav Petkov
2017-10-13 2:24 ` Brijesh Singh
2017-10-13 4:13 ` Brijesh Singh [this message]
2017-10-13 10:20 ` Borislav Petkov
2017-10-13 9:14 ` Borislav Petkov
2017-10-07 1:06 ` [Part2 PATCH v5.1 12.8/31] crypto: ccp: Implement SEV_PEK_CERT_IMPORT " Brijesh Singh
2017-10-13 14:53 ` Borislav Petkov
2017-10-13 16:09 ` Brijesh Singh
2017-10-07 1:06 ` [Part2 PATCH v5.1 12.9/31] crypto: ccp: Implement SEV_PDH_CERT_EXPORT " Brijesh Singh
2017-10-13 15:01 ` Borislav Petkov
2017-10-07 18:40 ` [Part2 PATCH v5.1 12.1/31] crypto: ccp: Add Secure Encrypted Virtualization (SEV) command support Borislav Petkov
2017-10-08 13:30 ` Brijesh Singh
2017-10-08 14:00 ` Borislav Petkov
2017-10-09 0:11 ` Brijesh Singh
2017-10-09 15:21 ` Borislav Petkov
2017-10-10 15:00 ` Brijesh Singh
2017-10-10 18:43 ` Tom Lendacky
2017-10-10 20:04 ` Borislav Petkov
2017-10-11 14:19 ` Borislav Petkov
2017-10-11 14:23 ` Brijesh Singh
2017-10-11 16:50 ` [Part2 PATCH v5.2 12.2/31] " Brijesh Singh
2017-10-12 14:08 ` Borislav Petkov
2017-10-12 21:11 ` Brijesh Singh
2017-10-12 21:41 ` Borislav Petkov
2017-10-12 21:52 ` Brijesh Singh
2017-10-12 22:22 ` Borislav Petkov
2017-10-12 18:21 ` Borislav Petkov
2017-10-12 20:05 ` Brijesh Singh
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=df846578-247a-978f-703e-ed6a6d23ee01@amd.com \
--to=brijesh.singh@amd.com \
--cc=bp@suse.de \
--cc=gary.hook@amd.com \
--cc=herbert@gondor.apana.org.au \
--cc=kvm@vger.kernel.org \
--cc=linux-crypto@vger.kernel.org \
--cc=linux-kernel@vger.kernel.org \
--cc=pbonzini@redhat.com \
--cc=rkrcmar@redhat.com \
--cc=thomas.lendacky@amd.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox