From: Rodrigo Vivi <rodrigo.vivi@intel.com>
To: "Ghimiray, Himal Prasad" <himal.prasad.ghimiray@intel.com>
Cc: <intel-xe@lists.freedesktop.org>
Subject: Re: [PATCH 2/3] drm/xe: declare busted upon GuC load failure
Date: Fri, 15 Mar 2024 09:54:47 -0400 [thread overview]
Message-ID: <ZfRTJyRVlXbtocIq@intel.com> (raw)
In-Reply-To: <6961d753-007f-48ad-b70d-1add093ae61a@intel.com>
On Fri, Mar 15, 2024 at 04:46:16PM +0530, Ghimiray, Himal Prasad wrote:
>
> On 15-03-2024 06:33, Rodrigo Vivi wrote:
> > Let's block the device upon any GuC load failure.
> > But let's continue with the probe so guc logs can be read
> > from the debugfs.
> >
> > v2: - s/wedged/busted
> > - do not block probe or we lose guc_logs in debugfs (Matt)
> >
> > Cc: Matthew Brost <matthew.brost@intel.com>
> > Signed-off-by: Rodrigo Vivi <rodrigo.vivi@intel.com>
> > ---
> > drivers/gpu/drm/xe/xe_guc.c | 41 ++++++++++++++++---------------------
> > 1 file changed, 18 insertions(+), 23 deletions(-)
> >
> > diff --git a/drivers/gpu/drm/xe/xe_guc.c b/drivers/gpu/drm/xe/xe_guc.c
> > index caa86ccbe9e7..85edaaa734a5 100644
> > --- a/drivers/gpu/drm/xe/xe_guc.c
> > +++ b/drivers/gpu/drm/xe/xe_guc.c
> > @@ -461,7 +461,7 @@ static int guc_xfer_rsa(struct xe_guc *guc)
> > return 0;
> > }
> > -static int guc_wait_ucode(struct xe_guc *guc)
> > +static void guc_wait_ucode(struct xe_guc *guc)
> > {
> > struct xe_device *xe = guc_to_xe(guc);
> > u32 status;
> > @@ -491,31 +491,28 @@ static int guc_wait_ucode(struct xe_guc *guc)
> > if (ret) {
> > struct drm_device *drm = &xe->drm;
> > - drm_info(drm, "GuC load failed: status = 0x%08X\n", status);
> > - drm_info(drm, "GuC load failed: status: Reset = %d, BootROM = 0x%02X, UKernel = 0x%02X, MIA = 0x%02X, Auth = 0x%02X\n",
> > - REG_FIELD_GET(GS_MIA_IN_RESET, status),
> > - REG_FIELD_GET(GS_BOOTROM_MASK, status),
> > - REG_FIELD_GET(GS_UKERNEL_MASK, status),
> > - REG_FIELD_GET(GS_MIA_MASK, status),
> > - REG_FIELD_GET(GS_AUTH_STATUS_MASK, status));
> > -
> > - if ((status & GS_BOOTROM_MASK) == GS_BOOTROM_RSA_FAILED) {
> > - drm_info(drm, "GuC firmware signature verification failed\n");
> > - ret = -ENOEXEC;
> > - }
> > + drm_err(drm, "GuC load failed: status = 0x%08X\n", status);
> > + drm_err(drm, "GuC load failed: status: Reset = %d, BootROM = 0x%02X, UKernel = 0x%02X, MIA = 0x%02X, Auth = 0x%02X\n",
> > + REG_FIELD_GET(GS_MIA_IN_RESET, status),
> > + REG_FIELD_GET(GS_BOOTROM_MASK, status),
> > + REG_FIELD_GET(GS_UKERNEL_MASK, status),
> > + REG_FIELD_GET(GS_MIA_MASK, status),
> > + REG_FIELD_GET(GS_AUTH_STATUS_MASK, status));
> > +
> > + if ((status & GS_BOOTROM_MASK) == GS_BOOTROM_RSA_FAILED)
> > + drm_err(drm, "GuC firmware signature verification failed\n");
> > if (REG_FIELD_GET(GS_UKERNEL_MASK, status) ==
> > XE_GUC_LOAD_STATUS_EXCEPTION) {
> > - drm_info(drm, "GuC firmware exception. EIP: %#x\n",
> > - xe_mmio_read32(guc_to_gt(guc),
> > - SOFT_SCRATCH(13)));
> > - ret = -ENXIO;
> > + drm_err(drm, "GuC firmware exception. EIP: %#x\n",
> > + xe_mmio_read32(guc_to_gt(guc),
> > + SOFT_SCRATCH(13)));
> > }
> > +
> > + xe_device_declare_busted(xe);
> > } else {
> > drm_dbg(&xe->drm, "GuC successfully loaded");
> > }
> > -
> > - return ret;
> > }
> > static int __xe_guc_upload(struct xe_guc *guc)
> > @@ -545,16 +542,14 @@ static int __xe_guc_upload(struct xe_guc *guc)
> > goto out;
> > /* Wait for authentication */
> > - ret = guc_wait_ucode(guc);
> > - if (ret)
> > - goto out;
> > + guc_wait_ucode(guc);
> > xe_uc_fw_change_status(&guc->fw, XE_UC_FIRMWARE_RUNNING);
> > return 0;
> > out:
> > xe_uc_fw_change_status(&guc->fw, XE_UC_FIRMWARE_LOAD_FAIL);
> > - return 0 /* FIXME: ret, don't want to stop load currently */;
>
> This label is used for other failures too apart from guc_wait_ucode.
I'm afraid I didn't find those here... perhaps I missunderstood the 'label'
you mentioned.
my grep for returs with this comment failed
>
> Dont we want to ignore ret and return 0 those failures too ?
anyway, perhaps that could be a follow up clean up?
>
>
> BR
>
> Himal Ghimiray
>
> > + return ret;
> > }
> > /**
next prev parent reply other threads:[~2024-03-15 13:55 UTC|newest]
Thread overview: 23+ messages / expand[flat|nested] mbox.gz Atom feed top
2024-03-15 1:03 [PATCH 1/3] drm/xe: Introduce a simple busted state Rodrigo Vivi
2024-03-15 1:03 ` [PATCH 2/3] drm/xe: declare busted upon GuC load failure Rodrigo Vivi
2024-03-15 11:16 ` Ghimiray, Himal Prasad
2024-03-15 13:54 ` Rodrigo Vivi [this message]
2024-03-15 14:23 ` Ghimiray, Himal Prasad
2024-03-15 1:03 ` [PATCH 3/3] drm/xe: Force busted state and block GT reset upon any GPU hang Rodrigo Vivi
2024-03-15 4:31 ` Ghimiray, Himal Prasad
2024-03-15 13:59 ` Rodrigo Vivi
2024-03-15 1:08 ` ✓ CI.Patch_applied: success for series starting with [1/3] drm/xe: Introduce a simple busted state Patchwork
2024-03-15 1:08 ` ✓ CI.checkpatch: " Patchwork
2024-03-15 1:09 ` ✓ CI.KUnit: " Patchwork
2024-03-15 1:20 ` ✓ CI.Build: " Patchwork
2024-03-15 1:22 ` ✓ CI.Hooks: " Patchwork
2024-03-15 1:23 ` [PATCH 1/3] " Dixit, Ashutosh
2024-03-15 3:08 ` Lucas De Marchi
2024-03-15 3:09 ` Rodrigo Vivi
2024-03-15 14:14 ` Tvrtko Ursulin
2024-03-18 23:23 ` Dixit, Ashutosh
2024-03-15 1:24 ` ✓ CI.checksparse: success for series starting with [1/3] " Patchwork
2024-03-15 1:48 ` ✓ CI.BAT: " Patchwork
2024-03-15 7:13 ` [PATCH 1/3] " Aravind Iddamsetty
2024-03-15 13:50 ` Rodrigo Vivi
2024-03-15 11:52 ` Ghimiray, Himal Prasad
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=ZfRTJyRVlXbtocIq@intel.com \
--to=rodrigo.vivi@intel.com \
--cc=himal.prasad.ghimiray@intel.com \
--cc=intel-xe@lists.freedesktop.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