| 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
| ||
|
Message-ID: <F89ECF87BC754A49BB74E794F6A0EC2FF68267@GSjpTKYDCembx31.service.hitachi.net>
Date: Wed, 5 Oct 2016 04:13:14 +0000
From: 岩松信洋 / IWAMATSU，NOBUHIRO
<nobuhiro.iwamatsu.kw@...achi.com>
To: Kees Cook <keescook@...omium.org>
CC: Anton Vorontsov <anton@...msg.org>,
Colin Cross <ccross@...roid.com>,
"Tony Luck" <tony.luck@...el.com>,
LKML <linux-kernel@...r.kernel.org>,
Mark Salyzyn <salyzyn@...roid.com>,
阿口誠司 / AGUCHI，SEIJI
<seiji.aguchi.tr@...achi.com>
Subject: RE: [PATCH v2 4/5] ramoops: support multiple pmsg instances
Hi,
> -----Original Message-----
> From: keescook@...gle.com [mailto:keescook@...gle.com] On Behalf Of Kees
> Cook
> Sent: Friday, September 09, 2016 6:53 AM
> To: 岩松信洋 / IWAMATSU，NOBUHIRO
> Cc: Anton Vorontsov; Colin Cross; Tony Luck; LKML; Hiraku Toyooka; Mark
> Salyzyn; 阿口誠司 / AGUCHI，SEIJI
> Subject: Re: [PATCH v2 4/5] ramoops: support multiple pmsg instances
>
> On Sun, Jul 24, 2016 at 8:56 PM, Nobuhiro Iwamatsu
> <nobuhiro.iwamatsu.kw@...achi.com> wrote:
> > From: Hiraku Toyooka <hiraku.toyooka.gu@...achi.com>
> >
> > This enables ramoops to deal with multiple pmsg instances.
> > A User can configure the size of each buffers by its module parameter
> > as follows.
> >
> > pmsg_size=0x1000,0x2000,...
> >
> > Then, the ramoops allocates multiple buffers and tells the number of
> > buffers to pstore to create multiple /dev/pmsg[ID].
> >
> > Signed-off-by: Hiraku Toyooka <hiraku.toyooka.gu@...achi.com>
> > Signed-off-by: Nobuhiro Iwamatsu <nobuhiro.iwamatsu.kw@...achi.com>
> > Cc: Mark Salyzyn <salyzyn@...roid.com>
> > Cc: Seiji Aguchi <seiji.aguchi.tr@...achi.com>
> > ---
> > Documentation/ramoops.txt | 22 +++++++
> > fs/pstore/ram.c | 146
> ++++++++++++++++++++++++++++++++++++++-------
> > include/linux/pstore_ram.h | 8 ++-
> > 3 files changed, 153 insertions(+), 23 deletions(-)
> >
> > diff --git a/Documentation/ramoops.txt b/Documentation/ramoops.txt
> > index 5d86756..cff6ac7 100644
> > --- a/Documentation/ramoops.txt
> > +++ b/Documentation/ramoops.txt
> > @@ -126,3 +126,25 @@ file. Here is an example of usage:
> > 0 ffffffff811d9c54 ffffffff8101a7a0 __const_udelay <-
> native_machine_emergency_restart+0x110/0x1e0
> > 0 ffffffff811d9c34 ffffffff811d9c80 __delay <-
> __const_udelay+0x30/0x40
> > 0 ffffffff811d9d14 ffffffff811d9c3f delay_tsc <- __delay+0xf/0x20
> > +
> > +6. Pmsg support
> > +
> > +Ramoops supports pmsg - logging userspace mesages in persistent
> > +store. By default, one pmsg area becomes available in ramoops. You
> > +can write data into /dev/pmsg0, e.g.:
> > +
> > + # echo foo > /dev/pmsg0
> > +
> > +After reboot, the stored data can be read from pmsg-ramoops-0 on the
> > +pstore filesystem.
> > +
> > +You can specify size of the pmsg area by additional kernel command line,
> e.g.:
> > +
> > + "ramoops.pmsg_size=0x1000"
> > +
> > +You can also use multiple pmsg areas, e.g.:
> > +
> > + "ramoops.pmsg_size=0x1000,0x2000,..."
> > +
> > +Then, pmsg0, pmsg1, ... will appear on /dev. This is useful to
> > +control individual content aging or priority.
>
> The documentation for the DT bindings will need updating too, in:
> Documentation/devicetree/bindings/reserved-memory/ramoops.txt
OK, I will add DT bindings to documentation.
>
> > diff --git a/fs/pstore/ram.c b/fs/pstore/ram.c index 288c5d0..2bf893f
> > 100644
> > --- a/fs/pstore/ram.c
> > +++ b/fs/pstore/ram.c
> > @@ -51,8 +51,8 @@ static ulong ramoops_ftrace_size = MIN_MEM_SIZE;
> > module_param_named(ftrace_size, ramoops_ftrace_size, ulong, 0400);
> > MODULE_PARM_DESC(ftrace_size, "size of ftrace log");
> >
> > -static ulong ramoops_pmsg_size = MIN_MEM_SIZE;
> > -module_param_named(pmsg_size, ramoops_pmsg_size, ulong, 0400);
> > +static char *ramoops_pmsg_size_str;
> > +module_param_named(pmsg_size, ramoops_pmsg_size_str, charp, 0400);
> > MODULE_PARM_DESC(pmsg_size, "size of user space message log");
> >
> > static unsigned long long mem_address; @@ -86,14 +86,14 @@ struct
> > ramoops_context {
> > struct persistent_ram_zone **dprzs;
> > struct persistent_ram_zone *cprz;
> > struct persistent_ram_zone *fprz;
> > - struct persistent_ram_zone *mprz;
> > + struct persistent_ram_zone **mprzs;
> > phys_addr_t phys_addr;
> > unsigned long size;
> > unsigned int memtype;
> > size_t record_size;
> > size_t console_size;
> > size_t ftrace_size;
> > - size_t pmsg_size;
> > + size_t *pmsg_size;
> > int dump_oops;
> > struct persistent_ram_ecc_info ecc_info;
> > unsigned int max_dump_cnt;
> > @@ -103,6 +103,7 @@ struct ramoops_context {
> > unsigned int console_read_cnt;
> > unsigned int ftrace_read_cnt;
> > unsigned int pmsg_read_cnt;
> > + unsigned int num_pmsg;
> > struct pstore_info pstore;
> > };
> >
> > @@ -220,9 +221,10 @@ static ssize_t ramoops_pstore_read(u64 *id, enum
> pstore_type_id *type,
> > if (!prz_ok(prz))
> > prz = ramoops_get_next_prz(&cxt->fprz,
> &cxt->ftrace_read_cnt,
> > 1, id, type,
> PSTORE_TYPE_FTRACE, 0);
> > - if (!prz_ok(prz))
> > - prz = ramoops_get_next_prz(&cxt->mprz,
> &cxt->pmsg_read_cnt,
> > - 1, id, type,
> PSTORE_TYPE_PMSG, 0);
> > + while (cxt->pmsg_read_cnt < cxt->num_pmsg && !prz)
> > + prz = ramoops_get_next_prz(cxt->mprzs,
> &cxt->pmsg_read_cnt,
> > + cxt->num_pmsg, id, type,
> > + PSTORE_TYPE_PMSG, 0);
> > if (!prz_ok(prz))
> > return 0;
> >
> > @@ -286,9 +288,11 @@ static int notrace ramoops_pstore_write_buf(enum
> pstore_type_id type,
> > persistent_ram_write(cxt->fprz, buf, size);
> > return 0;
> > } else if (type == PSTORE_TYPE_PMSG) {
> > - if (!cxt->mprz)
> > + if (part >= cxt->num_pmsg)
> > + return -EINVAL;
> > + if (!cxt->mprzs || !cxt->mprzs[part])
> > return -ENOMEM;
> > - persistent_ram_write(cxt->mprz, buf, size);
> > + persistent_ram_write(cxt->mprzs[part], buf, size);
> > return 0;
> > }
> >
> > @@ -348,7 +352,9 @@ static int ramoops_pstore_erase(enum pstore_type_id
> type, u64 id, int count,
> > prz = cxt->fprz;
> > break;
> > case PSTORE_TYPE_PMSG:
> > - prz = cxt->mprz;
> > + if (id >= cxt->num_pmsg)
> > + return -EINVAL;
> > + prz = cxt->mprzs[id];
> > break;
> > default:
> > return -EINVAL;
> > @@ -426,6 +432,37 @@ fail_prz:
> > return err;
> > }
> >
> > +static int ramoops_init_mprzs(struct device *dev, struct
> ramoops_context *cxt,
> > + phys_addr_t *paddr, unsigned long total)
> > +{
> > + int err = -ENOMEM;
> > + int i;
> > +
> > + if (!total)
> > + return 0;
> > +
> > + if (*paddr + total - cxt->phys_addr > cxt->size) {
> > + dev_err(dev, "no room for pmsg\n");
> > + return -ENOMEM;
> > + }
> > +
> > + cxt->mprzs = kcalloc(cxt->num_pmsg, sizeof(*cxt->mprzs),
> GFP_KERNEL);
> > + if (!cxt->mprzs)
> > + goto fail_prz;
> > +
> > + for (i = 0; i < cxt->num_pmsg; i++) {
> > + err = __ramoops_init_prz(dev, cxt, &cxt->mprzs[i],
> paddr,
> > + cxt->pmsg_size[i], 0, true);
> > + if (err)
> > + goto fail_prz;
> > + }
> > +
> > + return 0;
> > +fail_prz:
> > + ramoops_free_przs(cxt->mprzs);
> > + return err;
> > +}
> > +
> > static int ramoops_init_prz(struct device *dev, struct ramoops_context
> *cxt,
> > struct persistent_ram_zone **prz,
> > phys_addr_t *paddr, size_t sz, u32 sig) @@
> > -472,6 +509,8 @@ static int ramoops_probe(struct platform_device *pdev)
> > size_t dump_mem_sz;
> > phys_addr_t paddr;
> > int err = -EINVAL;
> > + unsigned long pmsg_size_total = 0;
> > + unsigned int num_pmsg = 0;
> >
> > /* Only a single ramoops area allowed at a time, so fail extra
> > * probes.
> > @@ -492,8 +531,18 @@ static int ramoops_probe(struct platform_device
> *pdev)
> > pdata->console_size =
> rounddown_pow_of_two(pdata->console_size);
> > if (pdata->ftrace_size && !is_power_of_2(pdata->ftrace_size))
> > pdata->ftrace_size =
> rounddown_pow_of_two(pdata->ftrace_size);
> > - if (pdata->pmsg_size && !is_power_of_2(pdata->pmsg_size))
> > - pdata->pmsg_size =
> rounddown_pow_of_two(pdata->pmsg_size);
> > + if (pdata->pmsg_size) {
> > + for (;; num_pmsg++) {
> > + unsigned long size =
> > + pdata->pmsg_size[num_pmsg];
> > +
> > + if (!size)
> > + break;
> > + if (!is_power_of_2(size))
> > + pdata->pmsg_size[num_pmsg]
> > + = rounddown_pow_of_two(size);
> > + pmsg_size_total += size;
> > + }
> > + }
> >
> > cxt->size = pdata->mem_size;
> > cxt->phys_addr = pdata->mem_address; @@ -501,17 +550,26 @@
> > static int ramoops_probe(struct platform_device *pdev)
> > cxt->record_size = pdata->record_size;
> > cxt->console_size = pdata->console_size;
> > cxt->ftrace_size = pdata->ftrace_size;
> > - cxt->pmsg_size = pdata->pmsg_size;
> > cxt->dump_oops = pdata->dump_oops;
> > cxt->ecc_info = pdata->ecc_info;
> > + cxt->num_pmsg = num_pmsg;
> > + if (num_pmsg) {
> > + cxt->pmsg_size = kcalloc(num_pmsg, sizeof(size_t),
> GFP_KERNEL);
> > + if (!cxt->pmsg_size) {
> > + err = -ENOMEM;
> > + goto fail_out;
> > + }
> > + memcpy(cxt->pmsg_size, pdata->pmsg_size,
> > + sizeof(size_t) * num_pmsg);
> > + }
> >
> > paddr = cxt->phys_addr;
> >
> > dump_mem_sz = cxt->size - cxt->console_size - cxt->ftrace_size
> > - - cxt->pmsg_size;
> > + - pmsg_size_total;
> > err = ramoops_init_dprzs(dev, cxt, &paddr, dump_mem_sz);
> > if (err)
> > - goto fail_out;
> > + goto fail_init_dprzs;
> >
> > err = ramoops_init_prz(dev, cxt, &cxt->cprz, &paddr,
> > cxt->console_size, 0); @@ -523,11
> > +581,12 @@ static int ramoops_probe(struct platform_device *pdev)
> > if (err)
> > goto fail_init_fprz;
> >
> > - err = ramoops_init_prz(dev, cxt, &cxt->mprz, &paddr,
> cxt->pmsg_size, 0);
> > + err = ramoops_init_mprzs(dev, cxt, &paddr, pmsg_size_total);
> > if (err)
> > - goto fail_init_mprz;
> > + goto fail_init_mprzs;
> >
> > cxt->pstore.data = cxt;
> > + cxt->pstore.num_pmsg = num_pmsg;
> > /*
> > * Console can handle any buffer size, so prefer LOG_LINE_MAX.
> If we
> > * have to handle dumps, we must have at least record_size
> > buffer. And @@ -560,7 +619,6 @@ static int ramoops_probe(struct
> platform_device *pdev)
> > record_size = pdata->record_size;
> > dump_oops = pdata->dump_oops;
> > ramoops_console_size = pdata->console_size;
> > - ramoops_pmsg_size = pdata->pmsg_size;
>
> Doesn't this mean the module parameters for the pmsg_size list won't be
> visible to sysfs?
I see. I will think other solution.
>
> > ramoops_ftrace_size = pdata->ftrace_size;
> >
> > pr_info("attached 0x%lx@...llx, ecc: %d/%d\n", @@ -573,14
> > +631,16 @@ fail_buf:
> > kfree(cxt->pstore.buf);
> > fail_clear:
> > cxt->pstore.bufsize = 0;
> > - persistent_ram_free(cxt->mprz);
> > -fail_init_mprz:
> > + ramoops_free_przs(cxt->mprzs);
> > +fail_init_mprzs:
> > persistent_ram_free(cxt->fprz);
> > fail_init_fprz:
> > persistent_ram_free(cxt->cprz);
> > fail_init_cprz:
> > cxt->max_dump_cnt = 0;
> > ramoops_free_przs(cxt->dprzs);
> > +fail_init_dprzs:
> > + kfree(cxt->pmsg_size);
> > fail_out:
> > return err;
> > }
> > @@ -594,8 +654,9 @@ static int ramoops_remove(struct platform_device
> > *pdev)
> >
> > kfree(cxt->pstore.buf);
> > cxt->pstore.bufsize = 0;
> > + kfree(cxt->pmsg_size);
> >
> > - persistent_ram_free(cxt->mprz);
> > + ramoops_free_przs(cxt->mprzs);
> > persistent_ram_free(cxt->fprz);
> > persistent_ram_free(cxt->cprz);
> > ramoops_free_przs(cxt->dprzs); @@ -611,6 +672,38 @@ static
> > struct platform_driver ramoops_driver = {
> > },
> > };
> >
> > +static unsigned long *parse_size_str(char *size_str) {
> > + int i, ret;
> > + unsigned long *size_array, count = 1;
> > +
> > + if (size_str) {
> > + char *s = size_str;
> > + /* Necessary array size is the number of commas + 1 */
> > + for (; (s = strchr(s, ',')); s++)
> > + count++;
> > + }
> > +
> > + size_array = kcalloc(count + 1, sizeof(unsigned long),
> GFP_KERNEL);
> > + if (!size_array)
> > + goto out;
> > +
> > + if (size_str) {
> > + for (i = 0; i < count; i++) {
> > + ret = get_option(&size_str, (int
> *)&size_array[i]);
> > + if (ret == 1)
> > + break;
> > + else if (ret != 2) {
> > + size_array[i] = 0;
> > + break;
> > + }
> > + }
> > + } else
> > + size_array[0] = MIN_MEM_SIZE;
> > +out:
> > + return size_array;
> > +}
> > +
> > static void ramoops_register_dummy(void) {
> > if (!mem_size)
> > @@ -630,7 +723,9 @@ static void ramoops_register_dummy(void)
> > dummy_data->record_size = record_size;
> > dummy_data->console_size = ramoops_console_size;
> > dummy_data->ftrace_size = ramoops_ftrace_size;
> > - dummy_data->pmsg_size = ramoops_pmsg_size;
> > + dummy_data->pmsg_size = parse_size_str(ramoops_pmsg_size_str);
>
> I think this function is wrong, since it looks like the num_pmsg count a
> bit above expects to find a NULL-terminated list. I think
> parse_size_str() needs to include an extra empty array allocation to make
> sure the list is always NULL-terminated.
OK, I will fix this.
>
> > + if (!dummy_data->pmsg_size)
> > + goto fail_pmsg_size;
> > dummy_data->dump_oops = dump_oops;
> > /*
> > * For backwards compatibility ramoops.ecc=1 means 16 bytes
> > ECC @@ -643,7 +738,13 @@ static void ramoops_register_dummy(void)
> > if (IS_ERR(dummy)) {
> > pr_info("could not create platform device: %ld\n",
> > PTR_ERR(dummy));
> > + goto fail_pdev;
> > }
> > + return;
> > +fail_pdev:
> > + kfree(dummy_data->pmsg_size);
> > +fail_pmsg_size:
> > + kfree(dummy_data);
> > }
> >
> > static int __init ramoops_init(void)
> > @@ -657,6 +758,7 @@ static void __exit ramoops_exit(void) {
> > platform_driver_unregister(&ramoops_driver);
> > platform_device_unregister(dummy);
> > + kfree(dummy_data->pmsg_size);
> > kfree(dummy_data);
> > }
> > module_exit(ramoops_exit);
> > diff --git a/include/linux/pstore_ram.h b/include/linux/pstore_ram.h
> > index 4660aaa..af7bbdd 100644
> > --- a/include/linux/pstore_ram.h
> > +++ b/include/linux/pstore_ram.h
> > @@ -72,6 +72,12 @@ ssize_t persistent_ram_ecc_string(struct
> persistent_ram_zone *prz,
> > * Ramoops platform data
> > * @mem_size memory size for ramoops
> > * @mem_address physical memory address to contain ramoops
> > + * @pmsg_size array containing size of each pmsg area. 0 value in the
> array
> > + * indicates the end of the content. For example, if the
> following
> > + * array is given,
> > + * .pmsg_size = { 0x1000, 0x2000, 0x0, 0x3000, };
> > + * then, pmsg areas are allocated for the first two size
> values
> > + * and '0x3000' is just ignored.
> > */
> >
> > struct ramoops_platform_data {
> > @@ -81,7 +87,7 @@ struct ramoops_platform_data {
> > unsigned long record_size;
> > unsigned long console_size;
> > unsigned long ftrace_size;
> > - unsigned long pmsg_size;
> > + unsigned long *pmsg_size;
> > int dump_oops;
> > struct persistent_ram_ecc_info ecc_info; };
> > --
> > 2.8.1
> >
> >
>
> -Kees
>
> --
> Kees Cook
> Nexus Security
Best regards,
Nobuhiro
Powered by blists - more mailing lists