From: Tom Lendacky <thomas.lendacky@amd.com>
To: Dan Williams <dan.j.williams@intel.com>,
linux-kernel@vger.kernel.org, x86@kernel.org,
linux-coco@lists.linux.dev, svsm-devel@coconut-svsm.dev
Cc: Thomas Gleixner <tglx@linutronix.de>,
Ingo Molnar <mingo@redhat.com>, Borislav Petkov <bp@alien8.de>,
Dave Hansen <dave.hansen@linux.intel.com>,
"H. Peter Anvin" <hpa@zytor.com>,
Andy Lutomirski <luto@kernel.org>,
Peter Zijlstra <peterz@infradead.org>,
Michael Roth <michael.roth@amd.com>,
Ashish Kalra <ashish.kalra@amd.com>,
Joel Becker <jlbec@evilplan.org>, Christoph Hellwig <hch@lst.de>
Subject: Re: [PATCH v3 12/14] fs/configfs: Add a callback to determine attribute visibility
Date: Tue, 16 Apr 2024 14:54:08 -0500 [thread overview]
Message-ID: <3cb02504-4854-8bc6-7962-e590e656856d@amd.com> (raw)
In-Reply-To: <661ec2968fa55_4d56129470@dwillia2-mobl3.amr.corp.intel.com.notmuch>
On 4/16/24 13:25, Dan Williams wrote:
> Tom Lendacky wrote:
>> On 4/16/24 00:46, Dan Williams wrote:
>>> Tom Lendacky wrote:
>>>> In order to support dynamic decisions as to whether an attribute should be
>>>> created, add a callback that returns a bool to indicate whether the
>>>> attribute should be display. If no callback is registered, the attribute
> [..]
>>>> Cc: Joel Becker <jlbec@evilplan.org>
>>>> Cc: Christoph Hellwig <hch@lst.de>
>>>> Signed-off-by: Tom Lendacky <thomas.lendacky@amd.com>
>>>> ---
>>>> fs/configfs/file.c | 7 +++
>>>> include/linux/configfs.h | 111 +++++++++++++++++++++++++++------------
>>>> 2 files changed, 84 insertions(+), 34 deletions(-)
>>>>
> [..]
>>>> diff --git a/include/linux/configfs.h b/include/linux/configfs.h
>>>> index 2606711adb18..c836d7bc7c9e 100644
>>>> --- a/include/linux/configfs.h
>>>> +++ b/include/linux/configfs.h
>>>> @@ -116,35 +116,57 @@ struct configfs_attribute {
>>>> const char *ca_name;
>>>> struct module *ca_owner;
>>>> umode_t ca_mode;
>>>> + bool (*is_visible)(const struct config_item *, const struct configfs_attribute *);
>>>> ssize_t (*show)(struct config_item *, char *);
>>>> ssize_t (*store)(struct config_item *, const char *, size_t);
>>>> };
>>>>
>>>> -#define CONFIGFS_ATTR(_pfx, _name) \
>>>> +#define __CONFIGFS_ATTR(_pfx, _name, _vis) \
>>>> static struct configfs_attribute _pfx##attr_##_name = { \
>>>> .ca_name = __stringify(_name), \
>>>> .ca_mode = S_IRUGO | S_IWUSR, \
>>>> .ca_owner = THIS_MODULE, \
>>>> + .is_visible = _vis, \
>>>> .show = _pfx##_name##_show, \
>>>> .store = _pfx##_name##_store, \
>>>
>>> Shouldn't this operation live in configfs_group_operations? That would
>>> mirror the sysfs organization, and likely saves some memory.
>>
>> I suppose it can, but then you lose the grouping of attributes within
>> the same directory, right? A configfs group will result in moving the
>> entries into a subdirectory, right? If we go with the group level, then
>> we will be moving the existing TSM extra attributes and the new TSM SVSM
>> attributes into new, separate sub-directories.
>
> I am not following the concern about "losing the grouping"? Here is what
> I was thinking with having the visibility routines in group operations.
> This is just the broard strokes, it compiles, but still needs the finer
> detail work to make tdx-guest skip all the attributes that do not apply
> to it. Might need to be broken up a bit more, but hopefully conveys the
> idea. Does this address your grouping concern?
Yes and no. Basically the is_visible()/is_bin_visible() callback will
have to check every index value for a "group" against the passed in
value. I was trying to group the values together using an enum in order
to make it a bit easier and more readable in the callback. Adding
another attribute to the group requires updates in multiple places. But
thats just how I was looking at it. I can also see where you might want
to selectively hide/show entries and this method works well for that.
I'll follow this approach (add you as Co-developed-by: or Suggested-by:,
whichever you prefer) and submit a v4.
Thanks,
Tom
>
> diff --git a/drivers/virt/coco/sev-guest/sev-guest.c b/drivers/virt/coco/sev-guest/sev-guest.c
> index 87f241825bc3..39b8455f0ba5 100644
> --- a/drivers/virt/coco/sev-guest/sev-guest.c
> +++ b/drivers/virt/coco/sev-guest/sev-guest.c
> @@ -968,7 +968,7 @@ static int __init sev_guest_probe(struct platform_device *pdev)
> snp_dev->input.resp_gpa = __pa(snp_dev->response);
> snp_dev->input.data_gpa = __pa(snp_dev->certs_data);
>
> - ret = tsm_register(&sev_tsm_ops, snp_dev, &tsm_report_extra_type);
> + ret = tsm_register(&sev_tsm_ops, snp_dev);
> if (ret)
> goto e_free_cert_data;
>
> diff --git a/drivers/virt/coco/tdx-guest/tdx-guest.c b/drivers/virt/coco/tdx-guest/tdx-guest.c
> index 1253bf76b570..654d20ea524a 100644
> --- a/drivers/virt/coco/tdx-guest/tdx-guest.c
> +++ b/drivers/virt/coco/tdx-guest/tdx-guest.c
> @@ -301,7 +301,7 @@ static int __init tdx_guest_init(void)
> goto free_misc;
> }
>
> - ret = tsm_register(&tdx_tsm_ops, NULL, NULL);
> + ret = tsm_register(&tdx_tsm_ops, NULL);
> if (ret)
> goto free_quote;
>
> diff --git a/drivers/virt/coco/tsm.c b/drivers/virt/coco/tsm.c
> index d1c2db83a8ca..b31be0e61728 100644
> --- a/drivers/virt/coco/tsm.c
> +++ b/drivers/virt/coco/tsm.c
> @@ -14,7 +14,6 @@
>
> static struct tsm_provider {
> const struct tsm_ops *ops;
> - const struct config_item_type *type;
> void *data;
> } provider;
> static DECLARE_RWSEM(tsm_rwsem);
> @@ -252,34 +251,18 @@ static ssize_t tsm_report_auxblob_read(struct config_item *cfg, void *buf,
> }
> CONFIGFS_BIN_ATTR_RO(tsm_report_, auxblob, NULL, TSM_OUTBLOB_MAX);
>
> -#define TSM_DEFAULT_ATTRS() \
> - &tsm_report_attr_generation, \
> - &tsm_report_attr_provider
> -
> static struct configfs_attribute *tsm_report_attrs[] = {
> - TSM_DEFAULT_ATTRS(),
> - NULL,
> -};
> -
> -static struct configfs_attribute *tsm_report_extra_attrs[] = {
> - TSM_DEFAULT_ATTRS(),
> - &tsm_report_attr_privlevel,
> - &tsm_report_attr_privlevel_floor,
> + [TSM_REPORT_GENERATION] = &tsm_report_attr_generation,
> + [TSM_REPORT_PROVIDER] = &tsm_report_attr_provider,
> + [TSM_REPORT_PRIVLEVEL] = &tsm_report_attr_privlevel,
> + [TSM_REPORT_PRIVLEVEL_FLOOR] = &tsm_report_attr_privlevel_floor,
> NULL,
> };
>
> -#define TSM_DEFAULT_BIN_ATTRS() \
> - &tsm_report_attr_inblob, \
> - &tsm_report_attr_outblob
> -
> static struct configfs_bin_attribute *tsm_report_bin_attrs[] = {
> - TSM_DEFAULT_BIN_ATTRS(),
> - NULL,
> -};
> -
> -static struct configfs_bin_attribute *tsm_report_bin_extra_attrs[] = {
> - TSM_DEFAULT_BIN_ATTRS(),
> - &tsm_report_attr_auxblob,
> + [TSM_REPORT_INBLOB] = &tsm_report_attr_inblob,
> + [TSM_REPORT_OUTBLOB] = &tsm_report_attr_outblob,
> + [TSM_REPORT_AUXBLOB] = &tsm_report_attr_auxblob,
> NULL,
> };
>
> @@ -297,21 +280,12 @@ static struct configfs_item_operations tsm_report_item_ops = {
> .release = tsm_report_item_release,
> };
>
> -const struct config_item_type tsm_report_default_type = {
> +static const struct config_item_type tsm_report_type = {
> .ct_owner = THIS_MODULE,
> .ct_bin_attrs = tsm_report_bin_attrs,
> .ct_attrs = tsm_report_attrs,
> .ct_item_ops = &tsm_report_item_ops,
> };
> -EXPORT_SYMBOL_GPL(tsm_report_default_type);
> -
> -const struct config_item_type tsm_report_extra_type = {
> - .ct_owner = THIS_MODULE,
> - .ct_bin_attrs = tsm_report_bin_extra_attrs,
> - .ct_attrs = tsm_report_extra_attrs,
> - .ct_item_ops = &tsm_report_item_ops,
> -};
> -EXPORT_SYMBOL_GPL(tsm_report_extra_type);
>
> static struct config_item *tsm_report_make_item(struct config_group *group,
> const char *name)
> @@ -326,12 +300,38 @@ static struct config_item *tsm_report_make_item(struct config_group *group,
> if (!state)
> return ERR_PTR(-ENOMEM);
>
> - config_item_init_type_name(&state->cfg, name, provider.type);
> + config_item_init_type_name(&state->cfg, name, &tsm_report_type);
> return &state->cfg;
> }
>
> +static bool tsm_report_attr_visible(struct configfs_attribute *attr, int n)
> +{
> + guard(rwsem_read)(&tsm_rwsem);
> + if (!provider.ops)
> + return false;
> +
> + if (!provider.ops->is_visible)
> + return true;
> +
> + return provider.ops->is_visible(n);
> +}
> +
> +static bool tsm_report_bin_attr_visible(struct configfs_bin_attribute *attr,
> + int n)
> +{
> + if (!provider.ops)
> + return false;
> +
> + if (!provider.ops->is_bin_visible)
> + return true;
> +
> + return provider.ops->is_bin_visible(n);
> +}
> +
> static struct configfs_group_operations tsm_report_group_ops = {
> .make_item = tsm_report_make_item,
> + .is_visible = tsm_report_attr_visible,
> + .is_bin_visible = tsm_report_bin_attr_visible,
> };
>
> static const struct config_item_type tsm_reports_type = {
> @@ -353,16 +353,10 @@ static struct configfs_subsystem tsm_configfs = {
> .su_mutex = __MUTEX_INITIALIZER(tsm_configfs.su_mutex),
> };
>
> -int tsm_register(const struct tsm_ops *ops, void *priv,
> - const struct config_item_type *type)
> +int tsm_register(const struct tsm_ops *ops, void *priv)
> {
> const struct tsm_ops *conflict;
>
> - if (!type)
> - type = &tsm_report_default_type;
> - if (!(type == &tsm_report_default_type || type == &tsm_report_extra_type))
> - return -EINVAL;
> -
> guard(rwsem_write)(&tsm_rwsem);
> conflict = provider.ops;
> if (conflict) {
> @@ -372,7 +366,6 @@ int tsm_register(const struct tsm_ops *ops, void *priv,
>
> provider.ops = ops;
> provider.data = priv;
> - provider.type = type;
> return 0;
> }
> EXPORT_SYMBOL_GPL(tsm_register);
> @@ -384,7 +377,6 @@ int tsm_unregister(const struct tsm_ops *ops)
> return -EBUSY;
> provider.ops = NULL;
> provider.data = NULL;
> - provider.type = NULL;
> return 0;
> }
> EXPORT_SYMBOL_GPL(tsm_unregister);
> diff --git a/fs/configfs/dir.c b/fs/configfs/dir.c
> index 18677cd4e62f..213e88f4cec2 100644
> --- a/fs/configfs/dir.c
> +++ b/fs/configfs/dir.c
> @@ -580,6 +580,7 @@ static void detach_attrs(struct config_item * item)
> static int populate_attrs(struct config_item *item)
> {
> const struct config_item_type *t = item->ci_type;
> + struct configfs_group_operations *ops = t->ct_group_ops;
> struct configfs_attribute *attr;
> struct configfs_bin_attribute *bin_attr;
> int error = 0;
> @@ -589,12 +590,17 @@ static int populate_attrs(struct config_item *item)
> return -EINVAL;
> if (t->ct_attrs) {
> for (i = 0; (attr = t->ct_attrs[i]) != NULL; i++) {
> + if (ops && ops->is_visible && !ops->is_visible(attr, i))
> + continue;
> if ((error = configfs_create_file(item, attr)))
> break;
> }
> }
> if (t->ct_bin_attrs) {
> for (i = 0; (bin_attr = t->ct_bin_attrs[i]) != NULL; i++) {
> + if (ops && ops->is_bin_visible &&
> + !ops->is_bin_visible(bin_attr, i))
> + continue;
> error = configfs_create_bin_file(item, bin_attr);
> if (error)
> break;
> diff --git a/fs/configfs/file.c b/fs/configfs/file.c
> index 0ad32150611e..356d23b6b9cf 100644
> --- a/fs/configfs/file.c
> +++ b/fs/configfs/file.c
> @@ -444,7 +444,8 @@ const struct file_operations configfs_bin_file_operations = {
> * @attr: atrribute descriptor.
> */
>
> -int configfs_create_file(struct config_item * item, const struct configfs_attribute * attr)
> +int configfs_create_file(struct config_item *item,
> + const struct configfs_attribute *attr)
> {
> struct dentry *dir = item->ci_dentry;
> struct configfs_dirent *parent_sd = dir->d_fsdata;
> diff --git a/include/linux/configfs.h b/include/linux/configfs.h
> index 2606711adb18..31553f12db7c 100644
> --- a/include/linux/configfs.h
> +++ b/include/linux/configfs.h
> @@ -216,6 +216,8 @@ struct configfs_group_operations {
> struct config_group *(*make_group)(struct config_group *group, const char *name);
> void (*disconnect_notify)(struct config_group *group, struct config_item *item);
> void (*drop_item)(struct config_group *group, struct config_item *item);
> + bool (*is_visible)(struct configfs_attribute *attr, int n);
> + bool (*is_bin_visible)(struct configfs_bin_attribute *attr, int n);
> };
>
> struct configfs_subsystem {
> diff --git a/include/linux/tsm.h b/include/linux/tsm.h
> index de8324a2223c..a45b12943223 100644
> --- a/include/linux/tsm.h
> +++ b/include/linux/tsm.h
> @@ -42,6 +42,19 @@ struct tsm_report {
> u8 *auxblob;
> };
>
> +enum tsm_attr_index {
> + TSM_REPORT_GENERATION,
> + TSM_REPORT_PROVIDER,
> + TSM_REPORT_PRIVLEVEL,
> + TSM_REPORT_PRIVLEVEL_FLOOR,
> +};
> +
> +enum tsm_bin_attr_index {
> + TSM_REPORT_INBLOB,
> + TSM_REPORT_OUTBLOB,
> + TSM_REPORT_AUXBLOB,
> +};
> +
> /**
> * struct tsm_ops - attributes and operations for tsm instances
> * @name: tsm id reflected in /sys/kernel/config/tsm/report/$report/provider
> @@ -55,15 +68,11 @@ struct tsm_report {
> struct tsm_ops {
> const char *name;
> const unsigned int privlevel_floor;
> + bool (*is_visible)(enum tsm_attr_index index);
> + bool (*is_bin_visible)(enum tsm_bin_attr_index index);
> int (*report_new)(struct tsm_report *report, void *data);
> };
>
> -extern const struct config_item_type tsm_report_default_type;
> -
> -/* publish @privlevel, @privlevel_floor, and @auxblob attributes */
> -extern const struct config_item_type tsm_report_extra_type;
> -
> -int tsm_register(const struct tsm_ops *ops, void *priv,
> - const struct config_item_type *type);
> +int tsm_register(const struct tsm_ops *ops, void *priv);
> int tsm_unregister(const struct tsm_ops *ops);
> #endif /* __TSM_H */
next prev parent reply other threads:[~2024-04-16 19:54 UTC|newest]
Thread overview: 61+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-25 22:26 [PATCH v3 00/14] Provide SEV-SNP support for running under an SVSM Tom Lendacky
2024-03-25 22:26 ` [PATCH v3 01/14] x86/sev: Rename snp_init() in the boot/compressed/sev.c file Tom Lendacky
2024-04-09 17:09 ` Borislav Petkov
2024-04-09 17:44 ` Tom Lendacky
2024-04-09 17:57 ` Borislav Petkov
2024-04-12 16:19 ` Gupta, Pankaj
2024-03-25 22:26 ` [PATCH v3 02/14] x86/sev: Make the VMPL0 checking function more generic Tom Lendacky
2024-04-12 16:41 ` Gupta, Pankaj
2024-04-17 11:46 ` Borislav Petkov
2024-04-17 20:35 ` Tom Lendacky
2024-04-17 20:50 ` Borislav Petkov
2024-04-18 18:38 ` Tom Lendacky
2024-04-21 7:12 ` Borislav Petkov
2024-03-25 22:26 ` [PATCH v3 03/14] x86/sev: Check for the presence of an SVSM in the SNP Secrets page Tom Lendacky
2024-04-12 17:03 ` Gupta, Pankaj
2024-04-17 20:40 ` Borislav Petkov
2024-04-18 21:17 ` Tom Lendacky
2024-04-22 22:07 ` Borislav Petkov
2024-03-25 22:26 ` [PATCH v3 04/14] x86/sev: Use kernel provided SVSM Calling Areas Tom Lendacky
2024-04-12 16:04 ` Gupta, Pankaj
2024-03-25 22:26 ` [PATCH v3 05/14] x86/sev: Perform PVALIDATE using the SVSM when not at VMPL0 Tom Lendacky
2024-03-25 22:26 ` [PATCH v3 06/14] x86/sev: Use the SVSM to create a vCPU when not in VMPL0 Tom Lendacky
2024-04-12 15:28 ` Gupta, Pankaj
2024-03-25 22:26 ` [PATCH v3 07/14] x86/sev: Provide SVSM discovery support Tom Lendacky
2024-04-15 16:12 ` Gupta, Pankaj
2024-03-25 22:26 ` [PATCH v3 08/14] x86/sev: Provide guest VMPL level to userspace Tom Lendacky
2024-03-25 22:26 ` [PATCH v3 09/14] virt: sev-guest: Choose the VMPCK key based on executing VMPL Tom Lendacky
2024-04-16 4:54 ` Dan Williams
2024-04-16 15:17 ` Tom Lendacky
2024-04-16 15:47 ` Dan Williams
2024-03-25 22:26 ` [PATCH v3 10/14] configfs-tsm: Allow the privlevel_floor attribute to be updated Tom Lendacky
2024-04-16 4:55 ` Dan Williams
2024-04-16 15:23 ` Tom Lendacky
2024-04-16 15:57 ` Dan Williams
2024-04-16 16:17 ` Tom Lendacky
2024-03-25 22:26 ` [PATCH v3 11/14] x86/sev: Extend the config-fs attestation support for an SVSM Tom Lendacky
2024-04-16 5:37 ` Dan Williams
2024-04-16 15:53 ` Tom Lendacky
2024-04-16 16:19 ` Dan Williams
2024-03-25 22:26 ` [PATCH v3 12/14] fs/configfs: Add a callback to determine attribute visibility Tom Lendacky
2024-04-16 5:46 ` Dan Williams
2024-04-16 16:01 ` Tom Lendacky
2024-04-16 18:25 ` Dan Williams
2024-04-16 19:54 ` Tom Lendacky [this message]
2024-04-16 20:03 ` Dan Williams
2024-03-25 22:26 ` [PATCH v3 13/14] x86/sev: Hide SVSM attestation entries if not running under an SVSM Tom Lendacky
2024-04-09 18:12 ` Kuppuswamy Sathyanarayanan
2024-04-12 15:52 ` Tom Lendacky
2024-04-15 19:16 ` Tom Lendacky
2024-04-15 19:48 ` Kuppuswamy Sathyanarayanan
2024-04-15 20:13 ` Tom Lendacky
2024-04-15 21:50 ` Kuppuswamy Sathyanarayanan
2024-04-15 22:03 ` Tom Lendacky
2024-04-16 6:09 ` Dan Williams
2024-04-16 6:08 ` Dan Williams
2024-04-16 6:05 ` Dan Williams
2024-04-16 5:47 ` Dan Williams
2024-04-16 16:07 ` Tom Lendacky
2024-04-16 6:03 ` Dan Williams
2024-04-16 16:10 ` Tom Lendacky
2024-03-25 22:26 ` [PATCH v3 14/14] x86/sev: Allow non-VMPL0 execution when an SVSM is present Tom Lendacky
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=3cb02504-4854-8bc6-7962-e590e656856d@amd.com \
--to=thomas.lendacky@amd.com \
--cc=ashish.kalra@amd.com \
--cc=bp@alien8.de \
--cc=dan.j.williams@intel.com \
--cc=dave.hansen@linux.intel.com \
--cc=hch@lst.de \
--cc=hpa@zytor.com \
--cc=jlbec@evilplan.org \
--cc=linux-coco@lists.linux.dev \
--cc=linux-kernel@vger.kernel.org \
--cc=luto@kernel.org \
--cc=michael.roth@amd.com \
--cc=mingo@redhat.com \
--cc=peterz@infradead.org \
--cc=svsm-devel@coconut-svsm.dev \
--cc=tglx@linutronix.de \
--cc=x86@kernel.org \
/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
Powered by JetHome